Mayo Clinic Cardiology: Concise Textbook

Fourth Edition
MAYO CLINIC SCIENTIFIC PRESS

Mayo Clinic Atlas of Regional Anesthesia and Ultrasound-Guided Nerve Blockade

Edited by James R. Hebl, MD, and Robert L. Lennon, DO
Mayo Clinic Preventive Medicine and Public Health Board Review
Edited by Prathibha Varkey, MBBS, MPH, MHPE
Mayo Clinic Internal Medicine Board Review, Ninth Edition
Edited by Amit K. Ghosh, MD
Mayo Clinic Challenging Images for Pulmonary Board Review
Edited by Edward C. Rosenow III, MD
Mayo Clinic Gastroenterology and Hepatology Board Review, Fourth Edition
Edited by Stephen C. Hauser, MD
Mayo Clinic Infectious Diseases Board Review
Edited by Zelalem Temesgen, MD
Mayo Clinic Antimicrobial Handbook: Quick Guide, Second Edition
Edited by John W. Wilson, MD, and Lynn L. Estes, PharmD
Just Enough Physiology
By James R. Munis, MD, PhD
 Mayo Clinic Cardiology:
Concise Textbook
FOURTH EDITION

Editors-in-Chief

Joseph G. Murphy, MD
Margaret A. Lloyd, MD
Associate Editors

Peter A. Brady, MB, ChB, MD

Lyle J. Olson, MD

Raymond C. Shields, MD

MAYO CLINIC SCIENTIFIC PRESS OXFORD UNIVERSITY PRESS

The triple-shield Mayo logo and the words MAYO, MAYO CLINIC, and MAYO CLINIC
SCIENTIFIC PRESS are marks of Mayo Foundation for Medical Education and Research.

3
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide.

Oxford New York

Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto

With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam

Oxford is a registered trademark of Oxford University Press in the UK and

certain other countries.

Published by Oxford University Press, Inc.

198 Madison Avenue, New York, New York 10016
www.oup.com

© Mayo Foundation for Medical Education and Research 2013

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, without the prior permission of Mayo Foundation for Medical Education and Research. Inquiries should be
addressed to Scientific Publications, Plummer 10, Mayo Clinic, 200 First Street SW, Rochester, MN 55905.

You must not circulate this work in any other form

and you must impose this same condition on any acquirer.

Library of Congress Cataloging-in-Publication Data

Mayo Clinic cardiology : concise textbook / editors-in-chief, Joseph G. Murphy, Margaret A. Lloyd ; associate editors, Peter A.
Brady, Lyle J. Olson, Raymond C. Shields. — 4th ed.
p. ; cm.
Cardiology
Includes bibliographical references and index.
Summary: “Organized to present a comprehensive overview of the field of cardiology in an accessible, reader-friendly format
that can be covered in about 12 months, this new edition contains roughly 50% new material, the cardiac pharmacology
section has been completely reworked, cardiovascular trials have been included, and the entire book has been updated to reflect
current practice guidelines and recent developments. The book is peppered throughout with numerous tables and clinical
pearls that aid the student, as well as the teacher, to remain focused”—Provided by publisher.
ISBN 978-0-19-991571-2 (alk. paper)
I. Murphy, Joseph G. II. Lloyd, Margaret A. III. Mayo Clinic. IV. Title: Cardiology.
[DNLM: 1. Heart Diseases—Examination Questions. 2. Heart Diseases—Outlines. WG 18.2]

616.1'20076—dc23
2012025452

Mayo Foundation does not endorse any particular products or services, and the reference to any products or services in this
book is for informational purposes only and should not be taken as an endorsement by the authors or Mayo Foundation. Care
has been taken to confi rm the accuracy of the information presented and to describe generally accepted practices. However,
the authors, editors, and publisher are not responsible for errors or omissions or for any consequences from application of the
information in this book and make no warranty, express or implied, with respect to the contents of the publication. This book
should not be relied on apart from the advice of a qualified health care provider.

The authors, editors, and publisher have exerted efforts to ensure that drug selection and dosage set forth in this text are in
accordance with current recommendations and practice at the time of publication. However, in view of ongoing research,
changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, readers
are urged to check the package insert for each drug for any change in indications and dosage and for added wordings and
precautions. This is particularly important when the recommended agent is a new or infrequently employed drug.

Some drugs and medical devices presented in this publication have US Food and Drug Administration (FDA) clearance for
limited use in restricted research settings. It is the responsibility of the health care providers to ascertain the FDA status of
each drug or device planned for use in their clinical practice.

9 8 7 6 5 4 3 2 1
Printed in China
on acid-free paper
Dedication

This book is dedicated to our friend and colleague, Mark Callahan,

MD, who tragically died shortly before completion of this edition. Mark
was a dedicated Mayo Clinic clinician and educator who inspired all
of his colleagues and a generation of Mayo Medical School students,
cardiology residents, and fellows. His passions were the clinical care of
patients, medical teaching, and health care in underserved countries,
as evidenced by his multiple trips to Haiti. He contributed to multiple
editions of this textbook. We will always miss Mark for his integrity,
dedication to serving others, and endless good humor.

Joseph G. Murphy, MD
Margaret A. Lloyd, MD
This page intentionally left blank
 Preface
It has been a great honor to serve as editors-in-chief of this, the
fourth, edition of Mayo Clinic Cardiology: Concise Textbook.
Large textbooks are never the work of 1 or 2 individuals but
rather the product of a group of dedicated authors and a team
of publications professionals, as has been the case for this book.
This textbook is unique in several regards in that it comes almost
entirely from a single institution but was written by a diverse
faculty of more than 100 physicians, many with an international
background. Mayo Clinic contributors practice at multiple Mayo
Clinic locations—Minnesota (Rochester, Austin, and Mankato);
Scottsdale, Arizona; and Jacksonville, Florida.
This textbook is primarily a teaching and learning textbook
of cardiology rather than a comprehensive reference textbook
of cardiology. Our concise textbook specifically addresses the
learning needs of cardiology fellows-in-training seeking initial
cardiovascular board certification and those of busy clinicians in
cardiology practice seeking cardiovascular board recertification.
It will also be useful for international physicians studying for
examinations of the Royal Colleges of Physicians, anesthesiolo-
gists, critical care physicians, internists and general physicians
with a special interest in cardiology, and coronary care and crit-
ical care nurses.
In response to welcome feedback from readers of the 3 previ-
ous textbook editions, we have reduced the size of the textbook
by several hundred pages while striving to produce a readable
textbook with features that facilitate its use as a review tool,
including highlighted entries of key points. Newer electronic
search methods have made textbook references less timely; thus,
we deleted most chapter references but have included selected
readings when appropriate.
This textbook is designed to present the field of cardiology in
a reader-friendly format that can be studied over about 12 months.
Many small cardiology textbooks are bare-bones compilations of
facts that do not explain the fundamental concepts of cardiovascu-
lar disease, and many large cardiology textbooks describe the field
of cardiology in great detail. Mayo Clinic Cardiology: Concise
Textbook is designed to be a bridge between these approaches.
We sought to present a solid framework of ideas with sufficient
depth to make the subjects interesting yet concise, aimed specif-
ically toward fellows-in-training or practicing clinicians want-
ing to update their knowledge. The book contains more than
1,000 figures, many of which are in color, to supplement the text.
Summary points in each chapter facilitate review of key facts.
We appreciate comments and suggestions from our readers
about how we might improve this textbook.
Joseph G. Murphy, MD
murphy.joseph@mayo.edu
Margaret A. Lloyd, MD
lloyd.margaret@mayo.edu
Mayo Clinic
This page intentionally left blank
 Acknowledgments
We thank all our colleagues in the Division of Cardiovascular
Diseases at Mayo Clinic in Rochester, Arizona, and Florida who
generously contributed to this work. We also thank William D.
Edwards, MD, for permission to use images from the Mayo Clinic
cardiology pathologic image database. Randall J. Fritz, DVM,
and LeAnn Stee contributed enormously through their editorial
guidance. Craig Panner at Oxford University Press patiently saw
this project through countless tribulations. We thank the produc-
tions teams at both Mayo Clinic and Oxford University Press:
at Mayo—Kenna Atherton (manager, Scientific Publications),
Kristin M. Nett and Jane Craig (editorial assistants), John
Hedlund (proofreader), Ann Ihrke and Alissa Baumgartner
(coding); at Oxford University Press—Kathryn Winder and
Karen Kwak (production and manufacturing).
The cover art is an image of the Heart Nebula, which lies
more than 7,000 light years from Earth in the Perseus Arm of
the Milky Way galaxy in the constellation Cassiopeia. We thank
Matthew T. Russell for granting permission for use of his strik-
ing image of the Heart Nebula as an overlay for the image of
Cassiopeia (NASA/JPL-Caltech/UCLA) found at http://www.
nasa.gov/multimedia/imagegallery/image_feature_1610.html.
We specifically acknowledge colleagues from outside North
America who contributed many ideas to this book and who trans-
lated previous editions of the book into several foreign languages.
This page intentionally left blank
 Contents
Contributors xv
Section I Fundamentals of Cardiovascular Disease 1 Thomas C. Gerber, MD, PhD, and
Chapter 1 Cardiovascular Examination 3
Clarence Shub, MD
Chapter 2 Applied Anatomy of the Heart and
Great Vessels 20
Joseph G. Murphy, MD, and R. Scott Wright, MD
Chapter 3 Evidence-Based Medicine and Statistics in
Cardiology 44
Charanjit S. Rihal, MD
Chapter 4 Noncardiac Surgery in Patients With
Heart Disease 49
J. Wells Askew III, MD, and Clarence Shub, MD
Chapter 5 Essential Molecular Biology of Cardiovascular
Diseases 58
Cindy W. Tom, MD, and
Robert D. Simari, MD
Section II Noninvasive Imaging 69
Chapter 6 Restrictions on Drivers and Aircraft Pilots With
Cardiac Disease 71
Stephen L. Kopecky, MD
Chapter 7 Principles of Echocardiography 77
Teresa S. M. Tsang, MD
Chapter 8 Stress Echocardiography 100
Patricia A. Pellikka, MD
Chapter 9 Transesophageal Echocardiography 106
Sarinya Puwanant, MD, Lawrence J. Sinak, MD,
and Krishnaswamy Chandrasekaran, MD
Chapter 10 Nuclear Imaging 120
J. Wells Askew III, MD, and Todd D. Miller, MD
Chapter 11 Positron Emission Tomography 130
Panithaya Chareonthaitawee, MD
Chapter 12 Cardiovascular Computed Tomography and
Magnetic Resonance Imaging 139
Eric M. Walser, MD
Chapter 13 Cardiac Radiography 159
Jerome F. Breen, MD, Mark J. Callahan, MD,
and Margaret A. Lloyd, MD
Chapter 14 Atlas of Congenital Heart Defects 177
Sabrina D. Phillips, MD, and Frank Cetta Jr, MD
Chapter 15 Cardiopulmonary Exercise Testing in Clinical
Medicine 185
Thomas G. Allison, PhD, MPH
Chapter 16 Stress Test Selection 197
Thomas G. Allison, PhD, MPH
Section III Electrophysiology 203
Chapter 17 Electrocardiographic Diagnoses: Criteria and
Definitions of Abnormalities 205
Stephen C. Hammill, MD
Chapter 18 Cardiac Cellular Electrophysiology 239
Hon-Chi Lee, MD, PhD
Chapter 19 Indications for Invasive and Noninvasive
Electrophysiologic Testing 250
Barry A. Boilson, MD, and
Peter A. Brady, MB, ChB, MD
Chapter 20 Heritable Cardiomyopathies and Channelopathies:
Clinical Presentations, Genetics, and Implications
of Genetic Testing 257
J. Martijn Bos, MD, PhD, and
Michael J. Ackerman, MD, PhD
Chapter 21 Pediatric Arrhythmias 266
Bryan C. Cannon, MD
Chapter 22 Atrial Fibrillation and Flutter 270
Peter A. Brady, MB, ChB, MD
xii Contents

Chapter 23 Supraventricular Tachycardia 281 Chapter 39 Carcinoid and Drug-Related Heart Disease 409
Joseph J. Gard, MD, Ammar Habib, MD, Heidi M. Connolly, MD, and
and Samuel J. Asirvatham, MD Patricia A. Pellikka, MD

Chapter 24 Ventricular Tachycardia and Ectopy 289 Chapter 40 Prosthetic Heart Valves 416
Thomas M. Munger, MD Martha A. Grogan, MD, and
Fletcher A. Miller Jr, MD
Chapter 25 Arrhythmias in Congenital Heart Disease 302
Christopher J. McLeod MB, ChB, PhD, and Chapter 41 Surgery for Cardiac Valve Disease 425
Peter A. Brady, MB, ChB, MD Rakesh M. Suri, MD, DPhil, and
Thomas A. Orszulak, MD
Chapter 26 Evaluation and Management of Pregnancy Associated
Arrhythmias 307 Chapter 42 Pathogenesis of Valvular Heart Disease 435
Peter A. Brady, MB, ChB, MD Nalini M. Rajamannan, MD, and
Joseph G. Murphy, MD
Chapter 27 Syncope: Diagnosis and Treatment 313
Win-Kuang Shen, MD Chapter 43 Epidemiology of Valvular Heart Diseases 442
Vuyisile T. Nkomo, MD, MPH
Chapter 28 Pacemakers 324
Yong-Mei Cha, MD, and David L. Hayes, MD Section V Aorta and Peripheral Vascular Disease 445

Chapter 44 Peripheral Vascular Disease 447

Chapter 29 Cardiac Resynchronization Therapy 334
Peter C. Spittell, MD
Grace Lin, MD, and David L. Hayes, MD
Chapter 45 Cerebrovascular Disease and Carotid
Chapter 30 Implantable Cardioverter-Defibrillator Artery Stenting 455
Troubleshooting 339
Peter C. Spittell, MD, and David R. Holmes Jr, MD
Barry A. Boilson, MD, and Margaret A. Lloyd, MD
Chapter 46 The Aorta 460
Chapter 31 Implantable Cardioverter-Defibrillator Trials and
Peter C. Spittell, MD
Prevention of Sudden Cardiac Death 349
Margaret A. Lloyd, MD, and
Chapter 47 Renal Artery Disease 470
Bernard J. Gersh, MB, ChB, DPhil
Verghese Mathew, MD
Chapter 32 Heart Disease in Athletes 357
Chapter 48 Pathophysiology of Arterial Thrombosis 475
Stephen C. Hammill, MD
Robert D. McBane, MD, and
Waldemar E. Wysokinski, MD, PhD
Chapter 33 Autonomic Function Testing 359
K. L. Venkatachalam, MD
Chapter 49 Treatment and Prevention of Arterial
Thrombosis 484
Chapter 34 Antiarrhythmic Drugs 363
Robert D. McBane, MD, and
Peter A. Brady, MB, ChB, MD Waldemar E. Wysokinski, MD, PhD

Chapter 35 Intracardiac Electrophysiology Tracings 372 Chapter 50 Venous and Lymphatic Disorders 500
Malini Madhavan, MBBS, Raymond C. Shields, MD
Christopher J. McLeod MB, ChB, PhD,
Douglas L. Packer, MD, and
Chapter 51 Vasculitis 506
Samuel J. Asirvatham, MD
Paul W. Wennberg, MD

Section IV Valvular Heart Disease 383 Chapter 52 Marfan Syndrome 514

Chapter 36 Valvular Stenosis 385 Naser M. Ammash, MD, and Heidi M. Connolly, MD
Rick A. Nishimura, MD
Section VI Coronary Artery Disease Risk Factors 523
Chapter 37 Valvular Regurgitation 393 Chapter 53 Coronary Artery Disease Epidemiology 525
Rick A. Nishimura, MD Thomas G. Allison, PhD, MPH

Chapter 38 Rheumatic Heart Disease 403 Chapter 54 Metabolic Syndrome 531

Andrew G. Moore, MD Thomas G. Allison, PhD, MPH
 Contents xiii

Chapter 55 Pathogenesis of Atherosclerosis 537 Chapter 70 Mechanical Complications of Acute Myocardial

Joseph L. Blackshear, MD, and Birgit Kantor, MD Infarction 657
Joseph G. Murphy, MD, and Margaret A. Lloyd, MD
Chapter 56 Dyslipidemia 551
Joseph G. Murphy, MD, Thomas Chapter 71 Cardiogenic Shock 665
G. Allison, PhD, MPH, and R. Scott Wright, MD Joseph G. Murphy, MD

Chapter 57 Lipid-Lowering Clinical Trials and Medications 558 Chapter 72 Reperfusion Strategies for ST-Elevation
Joseph G. Murphy, MD, R. Scott Wright, MD, Myocardial Infarction 672
and Thomas G. Allison, PhD, MPH Joseph G. Murphy, MD

Chapter 58 Novel Biomarkers of Coronary Artery Chapter 73 Cardiac Emergencies 682

Disease Risk 571 Joseph G. Murphy, MD, Barry A. Boilson, MD,
Iftikhar J. Kullo, MD, FACC, FAHA and Margaret A. Lloyd, MD

Chapter 59 Diabetes and Coronary Artery Disease 574 Chapter 74 Risk Stratification After Myocardial Infarction 690
Robert L. Frye, MD, and David R. Holmes Jr, MD Randal J. Thomas, MD, MS

Chapter 60 Hypertension 578 Chapter 75 Cardiac Rehabilitation 695

Adrian J. B. Brady, BSc, MD, R. Scott Wright, MD, Thomas G. Allison, PhD, MPH
and Joseph G. Murphy, MD
Chapter 76 Coronary Artery Bypass Surgery 701
Chapter 61 Stroke and Cardiovascular Disease 591 Thoralf M. Sundt III, MD
Joseph G. Murphy, MD, and
Margaret A. Lloyd, MD
Section VIII Diseases of the Heart, Pericardium, and
Pulmonary Circulation 707
Chapter 62 Heart Disease in Women 595
Patricia J. M. Best, MD, and
Sharonne N. Hayes, MD Chapter 77 Pericardial Diseases 709
Rowlens M. Melduni, MD
Chapter 63 Heart Disease in the Elderly 603
Imran S. Syed, MD, Joseph G. Murphy, MD, and Chapter 78 Pulmonary Embolism 725
R. Scott Wright, MD Vivek Iyer, MD

Section VII Myocardial Infarction 607 Chapter 79 Pulmonary Hypertension 735

Rachel J. Le, MD, and
Chapter 64 Cardiac Biomarkers 609 Garvan C. Kane, MD, PhD
Brian P. Shapiro, MD, Luciano Babuin, MD, PhD,
and Allan S. Jaffe, MD Chapter 80 Pregnancy and the Heart 752
Heidi M. Connolly, MD
Chapter 65 Acute Coronary Syndromes 615
R. Scott Wright, MD, Stephen L. Kopecky, MD, Chapter 81 Adult Congenital Heart Disease 761
and Joseph G. Murphy, MD Naser M. Ammash, MD, and
Carole A. Warnes, MD
Chapter 66 Chronic Stable Angina 627
Kalkidan G. Bishu, MD, and Chapter 82 HIV Infection and the Heart 773
Frank V. Brozovich, MD, PhD Joseph G. Murphy, MD, and
Zelalem Temesgen, MD
Chapter 67 Right Ventricular Infarction 637
Joseph G. Murphy, MD, and R. Scott Wright, MD Chapter 83 Infective Endocarditis 777
Nandan S. Anavekar, MB, BCh, M. Rizwan Sohail,
Chapter 68 Adjunctive Therapy in Acute Myocardial MD, and Joseph G. Murphy, MD
Infarction 641
R. Scott Wright, MD, Imran S. Syed, MD, and Chapter 84 Systemic Disease and the Heart 797
Joseph G. Murphy, MD Grace Lin, MD, Marian T. McEvoy, MD,
and Joseph G. Murphy, MD
Chapter 69 Arrhythmia Complications of Acute Myocardial
Infarction 652 Chapter 85 Cardiac Tumors 807
Joseph G. Murphy, MD, and Margaret A. Lloyd, MD Joseph G. Murphy, MD, and R. Scott Wright, MD
xiv Contents

Chapter 86 Sleep Apnea and Cardiac Disease 812
Tomas Kara, MD, PhD, Tomas Konecny, MD,
and Virend Somers, MD, PhD
Chapter 87 Cardiovascular Trauma 820
Joseph G. Murphy, MD, and R. Scott Wright, MD
Chapter 100 Hypertrophic Cardiomyopathy 898
Steve R. Ommen, MD
Chapter 101 Cardiac Transplantation 908
Barry A. Boilson, MD, Richard C. Daly, MD,
and Sudhir S. Kushwaha, MD

Chapter 88 Acute Brain Injury and the Heart 824 Section X Invasive/Interventional Cardiology 919
Nandan S. Anavekar, MB, BCh,
Chapter 102 Diagnostic Cardiac Angiography 921
Sarinya Puwanant, MD, and
Krishnaswamy Chandrasekaran, MD André C. Lapeyre III, MD

Chapter 89 Noncardiac Anesthesia in Patients With Chapter 103 Coronary Artery Physiology and Intracoronary
Cardiovascular Disease 830 Imaging 953
Laurence C. Torsher, MD Abhiram Prasad, MD

Section IX Cardiomyopathy and Heart Failure 837 Chapter 104 Principles of Myocardial Revascularization in
Interventional Cardiology 966
Joseph G. Murphy, MD, Gregory W. Barsness, MD,
Chapter 90 Reflex and Humoral Control of the Circulation 839
and Rajiv Gulati, MD
Alfredo L. Clavell, MD
Chapter 105 Percutaneous Treatment of Structural
Chapter 91 Systolic Heart Function 843 Heart Disease 971
Wayne L. Miller, MD, PhD, and Lyle J. Olson, MD Margaret A. Lloyd, MD, Joseph G. Murphy, MD,
and Charanjit S. Rihal, MD
Chapter 92 Heart Failure With Preserved Ejection Fraction 850
Barry A. Borlaug, MD Chapter 106 Catheter Closure of Intracardiac Shunts 977
Guy S. Reeder, MD
Chapter 93 Heart Failure: Diagnosis and Evaluation 858
Richard J. Rodeheffer, MD, and Chapter 107 Invasive Hemodynamics 984
Margaret M. Redfield, MD Rick A. Nishimura, MD

Chapter 94 Right Ventricular Failure and Cor Pulmonale 864 Chapter 108 Atlas of Hemodynamic Tracings 995
Robert P. Frantz, MD, and Joseph G. Murphy, MD Rick A. Nishimura, MD, and
Joseph G. Murphy, MD
Chapter 95 Pharmacologic Therapy of Systolic Ventricular
Dysfunction and Heart Failure 869
Section XI Appendix 1023
Richard J. Rodeheffer, MD, and
Margaret M. Redfield, MD Chapter 109 Preparing for Cardiology Examinations 1025
Joseph G. Murphy, MD, and
Chapter 96 Mechanical Ventricular Assist Devices 874 Margaret A. Lloyd, MD
John A. Schirger, MD, Soon J. Park, MD, and
Sudhir S. Kushwaha, MD Chapter 110 Principles of Coronary Stenting 1035
Joseph G. Murphy, MD, Gregory
Chapter 97 Myocarditis 877 W. Barsness, MD, and Rajiv Gulati, MD
Leslie T. Cooper Jr, MD, and Lori A. Blauwet, MD
Chapter 111 Endomyocardial Biopsy 1041
Chapter 98 Dilated Cardiomyopathy 888 Joseph G. Murphy, MD, Robert P. Frantz, MD,
Horng H. Chen, MD, BCH, and and Leslie T. Cooper Jr, MD
Sanjay Dandamudi, MD
Credit Lines 1047
Chapter 99 Restrictive Cardiomyopathy 892
Sudhir S. Kushwaha, MD Index 1059
 Contributors
Michael J. Ackerman, MD, PhD
Consultant, Divisions of Cardiovascular Diseases, Pediatric
Cardiology, and Pediatric Research Laboratories, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, of Pediatrics, and
of Pharmacology, College of Medicine, Mayo Clinic
Thomas G. Allison, PhD, MPH
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Associate Professor of Medicine,
College of Medicine, Mayo Clinic
Naser M. Ammash, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Nandan S. Anavekar, MB, BCh
Senior Associate Consultant, Division of Cardiovascular Diseases,
Mayo Clinic, Rochester, Minnesota; Assistant Professor of
Medicine, College of Medicine, Mayo Clinic
Samuel J. Asirvatham, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine and of Pediatrics,
College of Medicine, Mayo Clinic
J. Wells Askew III, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
Luciano Babuin, MD
Research Collaborator in Cardiovascular Diseases, Mayo School
of Graduate Medical Education, College of Medicine, Mayo
Clinic, Rochester, Minnesota; Present address: Azienda
Ospedaliera, Padova, Italy
Gregory W. Barsness, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Associate Professor of Medicine,
College of Medicine, Mayo Clinic
Patricia J. M. Best, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Associate Professor of Medicine,
College of Medicine, Mayo Clinic
Kalkidan G. Bishu, MD
Fellow in Cardiovascular Diseases, Mayo School of Graduate
Medical Education, College of Medicine, Mayo Clinic,
Rochester, Minnesota
Joseph L. Blackshear, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Jacksonville, Florida; Professor of Medicine, College of
Medicine, Mayo Clinic
Lori A. Blauwet, MD
Senior Associate Consultant, Division of Cardiovascular
Diseases, Mayo Clinic, Rochester, Minnesota;
Assistant Professor of Medicine, College of Medicine,
Mayo Clinic
Barry A. Boilson, MD
Senior Associate Consultant, Division of Cardiovascular Diseases,
Mayo Clinic, Rochester, Minnesota; Assistant Professor of
Medicine, College of Medicine, Mayo Clinic
Barry A. Borlaug, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Associate Professor of Medicine,
College of Medicine, Mayo Clinic
J. Martijn Bos, MD, PhD
Research Associate, Division of Cardiovascular Diseases, Mayo
Clinic, Rochester, Minnesota
Adrian J. B. Brady, BSc, MD
Consultant Cardiologist, Glasgow Royal Infirmary,
Glasgow, United Kingdom; Associate Professor,
University of Glasgow
Peter A. Brady, MB, ChB, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Associate Professor of Medicine,
College of Medicine, Mayo Clinic
Jerome F. Breen, MD
Consultant, Division of Cardiac Radiology, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Radiology,
College of Medicine, Mayo Clinic
Frank V. Brozovich, MD, PhD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine and of Physiology,
College of Medicine, Mayo Clinic
Mark J. Callahan, MD†
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
xvi Contributors

Bryan C. Cannon, MD Thomas C. Gerber, MD, PhD

Senior Associate Consultant, Division of Pediatric Cardiology,
Mayo Clinic, Rochester, Minnesota; Assistant Professor of
Pediatrics, College of Medicine, Mayo Clinic
Frank Cetta Jr, MD
Chair, Division of Pediatric Cardiology, Mayo Clinic, Rochester,
Minnesota; Professor of Medicine and of Pediatrics, College of
Medicine, Mayo Clinic
Yong-Mei Cha, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Krishnaswamy Chandrasekaran, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Scottsdale, Arizona; Professor of Medicine, College of
Medicine, Mayo Clinic
Panithaya Chareonthaitawee, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Associate Professor of Medicine,
College of Medicine, Mayo Clinic
Horng H. Chen, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Alfredo L. Clavell, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
Heidi M. Connolly, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Leslie T. Cooper Jr, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Richard C. Daly, MD
Consultant, Division of Cardiovascular Surgery, Mayo Clinic,
Rochester, Minnesota; Professor of Surgery, College of
Medicine, Mayo Clinic
Sanjay Dandamudi, MD
Fellow in Internal Medicine, Mayo School of Graduate Medical
Education, College of Medicine, Mayo Clinic, Rochester,
Minnesota
Robert P. Frantz, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Associate Professor of Medicine,
College of Medicine, Mayo Clinic
Robert L. Frye, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Joseph J. Gard, MD
Resident in Cardiovascular Diseases, Mayo School of Graduate
Medical Education, College of Medicine, Mayo Clinic,
Rochester, Minnesota
†
Deceased.
Consultant, Division of Cardiovascular Diseases and Department
of Radiology, Mayo Clinic, Rochester, Minnesota; Professor of
Medicine and Radiology, College of Medicine, Mayo Clinic
Bernard J. Gersh, MB, ChB, DPhil
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Martha A. Grogan, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
Rajiv Gulati, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Associate Professor of Medicine,
College of Medicine, Mayo Clinic
Ammar Habib, MD
Resident in Internal Medicine, Mayo School of Graduate Medical
Education, College of Medicine, Mayo Clinic, Rochester,
Minnesota; Present address: University of Kansas Medical
Center, Kansas City, Kansas
Stephen C. Hammill, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
David L. Hayes, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Sharonne N. Hayes, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Associate Professor of Medicine,
College of Medicine, Mayo Clinic
Michael J. Hogan, MD
Consultant, Division of Consultative Medicine, Mayo Clinic,
Scottsdale, Arizona; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
David R. Holmes Jr, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Vivek N. Iyer, MD
Senior Associate Consultant, Division of Pulmonary and Critical
Care Medicine, Mayo Clinic, Rochester, Minnesota; Assistant
Professor of Medicine, College of Medicine, Mayo Clinic
Allan S. Jaffe, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Garvan C. Kane, MD, PhD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
Birgit Kantor, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Associate Professor of Medicine,
College of Medicine, Mayo Clinic
 Contributors
Tomas Kara, MD, PhD
Research Associate, Division of Cardiovascular Diseases, Mayo
Clinic, Rochester, Minnesota; Associate Professor of Medicine,
College of Medicine, Mayo Clinic
Bijoy K. Khandheria, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Scottsdale, Arizona; Professor of Medicine, College of
Medicine, Mayo Clinic; Present address: Aurora Health Care,
Milwaukee, Wisconsin
Tomas Konecny, MD
Fellow in Cardiovascular Diseases, Mayo Clinic School of
Graduate Medical Education and Assistant Professor of
Medicine, College of Medicine, Mayo Clinic, Rochester,
Minnesota
Stephen L. Kopecky, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Iftikhar J. Kullo, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Sudhir S. Kushwaha, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
André C. Lapeyre III, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
Rachel J. Le, MD
Fellow in Cardiovascular Diseases and Internal Medicine, Mayo
School of Graduate Medical Education, College of Medicine,
Mayo Clinic, Rochester, Minnesota
Hon-Chi Lee, MD, PhD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Grace Lin, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
Margaret A. Lloyd, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
Malini Madhavan, MBBS
Resident in Cardiovascular Diseases, Mayo School of Graduate
Medical Education, College of Medicine, Mayo Clinic,
Rochester, Minnesota
Verghese Mathew, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Robert D. McBane, MD
Consultant, Divisions of Cardiovascular Diseases and Hematology,
Mayo Clinic, Rochester, Minnesota; Professor of Medicine,
College of Medicine, Mayo Clinic
xvii
Marian T. McEvoy, MD
Consultant, Department of Dermatology, Mayo Clinic, Rochester,
Minnesota; Professor of Dermatology, College of Medicine,
Mayo Clinic
Christopher J. McLeod, MB, ChB, PhD
Senior Associate Consultant, Division of Cardiovascular Diseases,
Mayo Clinic, Rochester, Minnesota; Assistant Professor of
Medicine, College of Medicine, Mayo Clinic
Rowlens M. Melduni, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
Fletcher A. Miller Jr, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Todd D. Miller, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Wayne L. Miller, MD, PhD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Andrew G. Moore, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic
Health System—Austin, Austin, Minnesota; Instructor in
Medicine, College of Medicine, Mayo Clinic, Rochester,
Minnesota
Thomas M. Munger, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
Joseph G. Murphy, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Ajay Nehra, MD
Consultant, Department of Urology, Mayo Clinic, Rochester,
Minnesota; Professor of Urology, College of Medicine, Mayo
Clinic; Present address: Rush University Medical Center,
Chicago, Illinois
Rick A. Nishimura, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Vuyisile T. Nkomo, MD, MPH
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
Lyle J. Olson, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Steve R. Ommen, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
xviii Contributors
Thomas A. Orszulak, MD
Emeritus Consultant, Division of Cardiovascular Surgery, Mayo
Clinic, Rochester, Minnesota; Emeritus Professor of Surgery,
College of Medicine, Mayo Clinic
Douglas L. Packer, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Soon J. Park, MD
Consultant, Division of Cardiovascular Surgery, Mayo Clinic,
Rochester, Minnesota; Professor of Surgery, College of
Medicine, Mayo Clinic
Patricia A. Pellikka, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Sabrina D. Phillips, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
Abhiram Prasad, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Sarinya Puwanant, MD
Research Fellow in Cardiovascular Diseases, College of Medicine,
Mayo Clinic, Rochester, Minnesota; Present address: King
Chulalongkorn Memorial Hospital, Chulalongkorn University,
Bangkok, Thailand
Nalini M. Rajamannan, MD
Research Collaborator in Biochemistry, Mayo School of Graduate
Medical Education, College of Medicine, Mayo Clinic,
Rochester, Minnesota
Margaret M. Redfield, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Guy S. Reeder, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Charanjit S. Rihal, MD
Chair, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Richard J. Rodeheffer, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
John A. Schirger, MD
Senior Associate Consultant, Division of Cardiovascular Diseases,
Mayo Clinic, Rochester, Minnesota; Assistant Professor of
Medicine, College of Medicine, Mayo Clinic
Win-Kuang Shen, MD
Chair, Division of Cardiovascular Diseases, Mayo Clinic,
Scottsdale, Arizona; Professor of Medicine, College of
Medicine, Mayo Clinic
Brian P. Shapiro, MD
Senior Associate Consultant, Division of Cardiovascular Diseases,
Mayo Clinic, Jacksonville, Florida; Assistant Professor of
Medicine, College of Medicine, Mayo Clinic
Raymond C. Shields, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
Clarence Shub, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Robert D. Simari, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Lawrence J. Sinak, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
M. Rizwan Sohail, MD
Consultant, Division of Infectious Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
Virend Somers, MD, PhD
Consultant, Divisions of Cardiovascular Diseases and Nephrology
& Hypertension, Mayo Clinic, Rochester, Minnesota; Professor
of Medicine, College of Medicine, Mayo Clinic
Peter C. Spittell, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
Thoralf M. Sundt III, MD
Consultant, Division of Cardiovascular Surgery, Mayo Clinic,
Rochester, Minnesota; Professor of Surgery, College of
Medicine, Mayo Clinic; Present address: Massachusetts
General Hospital, Boston, Massachusetts
Rakesh M. Suri, MD, DPhil
Consultant, Division of Cardiovascular Surgery, Mayo Clinic,
Rochester, Minnesota; Associate Professor of Surgery, College
of Medicine, Mayo Clinic
Imran S. Syed, MD
Consultant, Division of Cardiovascular Diseases, Mayo Health
System—Mankato, Mankato, Minnesota; Assistant Professor of
Medicine, College of Medicine, Mayo Clinic
Zelalem Temesgen, MD
Consultant, Division of Infectious Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Randal J. Thomas, MD, MS
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Associate Professor of Medicine,
College of Medicine, Mayo Clinic
Cindy W. Tom, MD
Fellow in Cardiovascular Diseases, Mayo School of Graduate
Medical Education, College of Medicine, Mayo Clinic,
Rochester, Minnesota; Present address: ViaCristi Clinic,
Wichita, Kansas
 Contributors
Laurence C. Torsher, MD
Consultant, Division of Anesthesia, Mayo Clinic, Rochester,
Minnesota; Assistant Professor of Anesthesiology and of
Medical Education, College of Medicine, Mayo Clinic
Teresa S. M. Tsang, MD
Research Collaborator, Division of Cardiovascular Diseases, Mayo
Clinic, Rochester, Minnesota
K. L. Venkatachalam, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Jacksonville, Florida; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
Eric M. Walser, MD
Consultant, Department of Radiology, Mayo Clinic, Jacksonville,
Florida; Professor of Radiology, College of Medicine, Mayo
Clinic; Present address: University of Texas Medical Branch,
Galveston, Texas
xix
Carole A. Warnes, MD
Consultant, Divisions of Cardiovascular Diseases and Pediatric
Cardiology, Mayo Clinic, Rochester, Minnesota; Professor of
Medicine, College of Medicine, Mayo Clinic
Paul W. Wennberg, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Assistant Professor of Medicine, College
of Medicine, Mayo Clinic
R. Scott Wright, MD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Professor of Medicine, College of
Medicine, Mayo Clinic
Waldemar E. Wysokinski, MD, PhD
Consultant, Division of Cardiovascular Diseases, Mayo Clinic,
Rochester, Minnesota; Associate Professor of Medicine,
College of Medicine, Mayo Clinic
This page intentionally left blank
 Section I

Fundamentals of Cardiovascular Disease

This page intentionally left blank

Cardiovascular Examinationa
CLARENCE SHUB, MD
General Appearance
The physical examination, including the general appearance of
the patient, is an extremely important component of cardiol-
ogy examinations. Important clues to a cardiac diagnosis can be
obtained from inspection of the patient (Table 1.1).
Blood Pressure
Blood pressure should always be determined in both arms and
in the legs if there is any suspicion of coarctation of the aorta.
If systolic blood pressure differs between arms by more than
10 mm Hg, it is abnormal (Box 1.1).
Abnormalities on Palpation of the Precordium
The patient should be examined in both the supine and the left
lateral decubitus position. Examining the apical impulse by the
posterior approach with the patient in the sitting position may
be the best method to appreciate subtle abnormalities of precor-
dial motion. The normal apical impulse occurs during early sys-
tole with an outward motion imparted to the chest wall. During
mid and late systole, the LV is diminishing in volume and the
apical impulse moves away from the chest wall. Thus, outward
precordial apical motion occurring in late systole is abnormal.
Remember that point of maximal impulse is not synonymous
with apical impulse.
a
Portions of this chapter have been previously published in Shub C.
Cardiac physical examination: clinical “pearls” and application. ACC
Curr J Rev. Sep/Oct 1999. Used with permission.
Abbreviations and acronyms are expanded at the end of this chapter.
1
Palpation of the Apex
Constrictive pericarditis or tricuspid regurgitation produces a
subtle systolic precordial retraction.
The apical impulse of LV enlargement is usually widened or
diffuse (>3 cm in diameter), can be palpated in 2 interspaces,
and is displaced leftward. A subtle presystolic ventricular
rapid filling wave (A wave)—frequently associated with LV
hypertrophy—may be better visualized than palpated by observ-
ing the motion of the stethoscope applied lightly on the chest
wall, with appropriate timing during simultaneous auscultation.
Likewise, a palpable A wave can be detected in this manner. The
apical impulse of LV hypertrophy without dilatation is sustained
and localized but should not be displaced.
Causes of a palpable A wave (presystolic impulse) include the
following:
1. Aortic stenosis
2. Hypertrophic obstructive cardiomyopathy
3. Systemic hypertension
• The apical impulse of LV hypertrophy without dilatation is sus-
tained and localized. It should not be displaced but may be accom-
panied by a palpable presystolic outward movement, the A wave.
• Outward precordial apical motion occurring in late systole is
abnormal.
• Multiple abnormal outward precordial movements may occur: pre-
systolic, systolic, or late systolic rebound.
Palpation of the Lower Sternal Area
Precordial motion in the lower sternal area usually reflects RV
motion. RV hypertrophy due to systolic overload (eg, in pulmo-
nary stenosis) causes a sustained outward lift. Diastolic overload
(eg, in ASD) causes a vigorous nonsustained motion. In severe
3
4 I Fundamentals of Cardiovascular Disease

Table 1.1. Clinical Clues to Specific Cardiac Abnormalities Detectable from the General Examination
Condition Appearance Associated Cardiac Abnormalities
Marfan syndrome Tall Aortic root dilatation
Long extremities Mitral valve prolapse
Acromegaly Large stature Cardiac hypertrophy
Coarse facial features
“Spade” hands
Turner syndrome Web neck Aortic coarctation
Hypertelorism Pulmonary stenosis
Short stature
Pickwickian syndrome Severe obesity Pulmonary hypertension
Somnolence
Friedreich ataxia Lurching gait Hypertrophic cardiomyopathy
Hammertoe
Pes cavus
Duchenne type muscular dystrophy Pseudohypertrophy of calves Cardiomyopathy
Ankylosing spondylitis Straight back syndrome Aortic regurgitation
Stiff (“poker”) spine Heart block (rare)
Jaundice Yellow skin or sclera Right-sided congestive heart failure
Prosthetic valve dysfunction (hemolysis)
Sickle cell anemia Cutaneous ulcers Pulmonary hypertension
Painful “crises” Secondary cardiomyopathy
Lentigines (LEOPARD syndromea) Brown skin macules that do not Hypertrophic obstructive cardiomyopathy
increase with sunlight Pulmonary stenosis
Hereditary hemorrhagic telangiectasia Small capillary hemangiomas on face Pulmonary arteriovenous fistula
(Osler-Weber-Rendu disease) or mouth, with or without cyanosis
Pheochromocytoma Pale, diaphoretic skin Catecholamine-induced secondary dilated
Neurofibromatosis—café-au-lait spots cardiomyopathy
Lupus Butterfly rash on face Verrucous endocarditis
Raynaud phenomenon—hands Myocarditis
Livedo reticularis Pericarditis
Sarcoidosis Cutaneous nodules Secondary cardiomyopathy
Erythema nodosum Heart block
Tuberous sclerosis Angiofibromas (face; adenoma sebaceum) Rhabdomyoma
Myxedema Coarse, dry skin Pericardial effusion
Thinning of lateral eyebrows Left ventricular dysfunction
Hoarseness
Right-to-left intracardiac shunt Cyanosis and clubbing of distal extremities Any of the lesions that cause Eisenmenger syndrome
Differential cyanosis and clubbing Reversed shunt through patent ductus arteriosus
Holt-Oram syndrome Rudimentary or absent thumb Atrial septal defect
Down syndrome Mental retardation Endocardial cushion defect
Simian crease of palm
Characteristic facies
Scleroderma Tight, shiny skin of fingers with contraction Pulmonary hypertension
Characteristic taut mouth and facies Myocardial, pericardial, or endocardial disease
Rheumatoid arthritis Typical hand deformity Myocardial, pericardial, or endocardial disease
Subcutaneous nodules (often subclinical)
Thoracic bony abnormality Pectus excavatum Pseudocardiomegaly
Straight back syndrome Mitral valve prolapse
Carcinoid syndrome Reddish cyanosis of face Right-sided cardiac valve stenosis or regurgitation
Periodic flushing
a
LEOPARD syndrome: lentigines, electrocardiographic changes, ocular hypertelorism, pulmonary stenosis, abnormal genitalia, retardation of
growth, deafness.
(Previously published. See “Credit Lines” section.)

mitral regurgitation, the LA expands in systole but is limited in

its posterior motion by the spine. The RV may then be pushed
forward, and the parasternal region is “lifted” indirectly.
Significant overlap of sites of maximal pulsation occurs in
LV and RV overload states. For example, in RV overload, the
abnormal impulse can overlap with the LV in the apical sternal
region (between the apex and the left lower sternal border). An
LV apical aneurysm may produce a delayed outward motion and
cause a “rocking” motion.
Palpation of the Left Upper Sternal Area
Abnormal pulsations at the left upper sternal border (pulmonic
area) can be due to a dilated pulmonary artery (eg, poststenotic
dilatation in pulmonary valve stenosis, idiopathic dilatation of
the pulmonary artery, or increased pulmonary flow related to
ASD or pulmonary hypertension). Pulsations of increased blood
flow are dynamic and quick, whereas pulsations due to pressure
overload cause a sustained impulse.
 1 Cardiovascular Examination
Box 1.1. Causes of Blood Pressure Discrepancy
Between Arms or Between Arms and Legs
Arterial occlusion or stenosis of any cause
Dissecting aortic aneurysm
Coarctation of the aorta
Supravalvular aortic stenosis
Thoracic outlet syndrome
• If the apical impulse is not palpable and the patient is hemodynam-
ically unstable, consider cardiac tamponade as the first diagnosis.
Palpation of the Right Upper Sternal Area
Abnormal pulsations at the right upper sternal border (aortic
area) should suggest an aortic aneurysm. An enlarged left lobe
of the liver associated with severe tricuspid regurgitation may
be appreciated in the epigastrium, and the epigastric site may
be the location of the maximal cardiac impulse in patients with
emphysema or an enlarged RV.
• RV hypertrophy due to systolic overload causes a sustained out-
ward lift. Diastolic overload (as in ASD) causes a vigorous non-
sustained motion.
• In severe mitral regurgitation, the LA expands in systole but is
limited in its posterior motion by the spine. The RV may then be
pushed forward, and the parasternal region is “lifted” indirectly.
• Significant overlap of sites of maximal pulsation occurs in LV and
RV overload states.
• Pulsations of increased blood flow are dynamic and quick, whereas
pulsations due to pressure overload cause a sustained impulse.
Jugular Veins
Abnormal waveforms in the jugular veins reflect abnormal
hemodynamics of the right side of the heart. In the presence of
normal sinus rhythm, there are 2 positive or outward moving
waves (a and v) and 2 visible negative or inward moving waves
(x and y) (Figure 1.1). The x descent is sometimes referred to as
the systolic collapse. Ordinarily, the c wave is not readily visi-
ble. The a wave can be identified by simultaneous auscultation
of the heart and inspection of the jugular veins. The a wave
occurs at about the time of the S1. The x descent follows. The
v wave, a slower, more undulating wave, occurs near S2. The y
descent follows. The a wave is normally larger than the v wave,
and the x descent is more marked than the y descent (Boxes 1.2
and 1.3).
Normal jugular venous pressure decreases with inspira-
tion and increases with expiration. Veins that fill at inspi-
ration (Kussmaul sign), however, are a clue to constrictive
pericarditis, pulmonary embolism, or RV infarction (Box 1.3
and Table 1.2).
• Jugular veins that fill at inspiration (Kussmaul sign) are a clue to
constrictive pericarditis, pulmonary embolism, or RV infarction.
“Hepatojugular” (Abdominojugular) Reflux Sign
The neck veins distend with steady (>10 seconds) upper abdom-
inal compression while the patient continues to breathe nor-
mally without straining. Straining may cause a false-positive
5
a
a
xc h
xl v Wave
y
xl Trough Ascending
limb
S1 S2
Figure 1.1. Normal Jugular Venous Pulse. The jugular v wave is built
up during systole, and its height reflects the rate of filling and the elastic-
ity of the right atrium. Between the bottom of the y descent (y trough)
and the beginning of the a wave is the period of relatively slow fill-
ing of the “atrioventricle” or diastasis period. The wave built up during
diastasis is the h wave. The h wave height also reflects the stiffness of
the right atrium. S1 indicates first heart sound; S2, second heart sound.
(Previously published. See “Credit Lines” section.)
“hepatojugular” reflux sign. The neck veins may collapse or
remain distended. Jugular venous pressure that remains increased
and then decreases abruptly (≥4 cm water) indicates an abnormal
response. It may occur in LV failure with secondary pulmonary
hypertension. In patients with chronic congestive heart failure,
a positive hepatojugular reflux sign (with or without increased
jugular venous pressure), an S3, and radiographic pulmonary
vascular redistribution are independent predictors of increased
pulmonary capillary wedge pressure. The abdominojugular
maneuver can also be useful for eliciting venous pulsations if
they are difficult to visualize.
• A positive “hepatojugular” (abdominojugular) reflux sign may be
present in LV failure with secondary pulmonary hypertension.
• If the jugular veins are engorged but not pulsatile, consider supe-
rior vena caval obstruction.
Arterial Pulse
Abnormalities of the Carotid Pulse
Hyperdynamic Carotid Pulse
A vigorous, hyperdynamic carotid pulse is consistent with AR.
It may also occur in other states of high cardiac output or be
caused by the wide pulse pressure associated with atherosclero-
sis, especially in the elderly.
Box 1.2. Timing of Jugular Venous Pulse Waves
a Wave—precedes the carotid arterial pulse and
is simultaneous with S4 , just before S1
x Descent—between S1 and S 2
v Wave—just after S 2
y Descent—after the v wave in early diastole
Abbreviations: S1, first heart sound; S2, second heart sound; S4, fourth
heart sound.
6 I Fundamentals of Cardiovascular Disease
Box 1.3. Abnormal Jugular Venous Pulse Waves
Increased a wave
Tricuspid stenosis
Decreased right ventricular compliance due to
right ventricular hypertrophy in severe pulmonary
hypertension
Pulmonary stenosis
Pulmonary vascular disease
Severe left ventricular hypertrophy due to pressure
by the hypertrophied septum on right ventricular
filling (Bernheim effect)
Hypertrophic obstructive cardiomyopathy
Rapid x descent
Cardiac tamponade
Increased v wave
Tricuspid regurgitation
Atrial septal defect
Rapid y descent (Friedreich sign)
Constrictive pericarditis
Dicrotic and Bisferious Pulses
A dicrotic carotid pulse occurs in myocardial failure, especially
in association with hypotension, decreased cardiac output, and
increased peripheral resistance. Dicrotic and bisferious are the
Greek and Latin terms, respectively, for twice beating, but in car-
diology they are not equivalent. The second impulse occurs in
early diastole with the dicrotic pulse and in late systole with the
bisferious pulse. The bisferious pulse usually occurs in combined
AR and aortic stenosis, but occasionally it occurs in pure AR.
Aortic Stenosis
Pulsus parvus (soft or weak pulse) classically occurs in aortic
stenosis, but it can also occur with severe stenosis of any car-
diac valve or with low cardiac output of any cause. Severe aortic
stenosis also produces a slowly increasing delayed pulse (pulsus
tardus). Because of the effects of aging on the carotid arteries,
the typical findings of pulsus parvus and pulsus tardus may be
less apparent or absent in the elderly, even with severe degrees of
aortic stenosis.
Hypertrophic Obstructive Cardiomyopathy
In hypertrophic obstructive cardiomyopathy, the ventricu-
lar obstruction begins in mid systole, increases as contraction
Table 1.2. Differentiation of Internal Jugular Vein Pulse and
Carotid Pulse
Jugular Vein Pulse Carotid Pulse
Double peak when in sinus rhythm Single peak
Obliterated by gentle pressure Unaffected by gentle pressure
Changes with position and inspiration Unaffected by position or
inspiration
proceeds, and decreases in late systole. The initial carotid impulse
is brisk. The pulse may be bisferious as well (Box 1.4).
Inequality of the carotid pulses can be due to carotid athero-
sclerosis, especially in elderly patients. In a young patient, con-
sider supravalvular aortic stenosis. (The right side then should
have the stronger pulse.) Aortic dissection and thoracic outlet
syndrome may also produce inequality of arterial pulses. A pul-
sating cervical mass, usually on the right, may be caused by ath-
erosclerotic “buckling” of the right common carotid artery and
give the false impression of a carotid aneurysm.
Transmitted Murmurs
Transmitted murmurs of aortic origin, most often due to aortic
stenosis (less often due to coarctation, PDA, pulmonary steno-
sis, or VSD), decrease in intensity as the stethoscope ascends
the neck, whereas a carotid bruit is usually louder higher in the
neck and decreases in intensity as the stethoscope is inched prox-
imally toward the chest. Both conditions may coexist, especially
in elderly patients. An abrupt change in the acoustic characteris-
tics (pitch) of the bruit as the stethoscope is inched upward may
be a clue to the presence of combined lesions.
Pulsus Paradoxus
Paradoxical pulse is an exaggeration of the normal (≤10 mm Hg)
inspiratory decrease in arterial pressure. It occurs classically
with cardiac tamponade but occurs occasionally with other
restrictive cardiac abnormalities, severe congestive heart fail-
ure, pulmonary embolism, and chronic obstructive pulmonary
disease (Box 1.5).
Pulsus Alternans
Pulsus alternans (alternation of stronger and weaker beats) rarely
occurs in healthy persons and then is transient after a premature
ventricular contraction. It usually is associated with severe myo-
cardial failure and is frequently accompanied by an S3, both of
which impart an ominous prognosis. Pulsus alternans may be
affected by alterations in venous return and may disappear as
congestive heart failure progresses. Electrical alternans (alter-
nating variation in the height of the QRS complex) is unrelated
to pulsus alternans (Box 1.6).
• A dicrotic carotid pulse occurs in myocardial failure, often in asso-
ciation with hypotension, decreased cardiac output, and increased
peripheral resistance.
Box 1.4. Causes of a Double-Impulse Carotid Arterial
Pulse
Dicrotic pulse (systolic and diastolic impulses)
Cardiomyopathy
Left ventricular failure
Bisferious pulse (2 systolic impulses)
Aortic regurgitation
Combined aortic valve stenosis and regurgitation
(dominant regurgitation)
Hypertrophic cardiomyopathy
 1 Cardiovascular Examination
Box 1.5. Causes of Pulsus Paradoxus
Constrictive pericarditis
Pericardial tamponade
Severe emphysema
Severe asthma
Severe heart failure
Pulmonary embolism
Morbid obesity
• Pulsus parvus (soft or weak pulse) classically occurs in aortic ste-
nosis, but it can also occur with severe stenosis of any cardiac valve
or with severely low cardiac output of any cause.
• Because of the effects of aging on the carotid arteries, the typical
findings of pulsus parvus and pulsus tardus may be less apparent or
absent in the elderly, even with severe degrees of aortic stenosis.
• Inequality of the carotid pulses can be due to carotid atheroscle-
rosis, especially in elderly patients. In a young patient, consider
supravalvular aortic stenosis. (The right side then should have the
stronger pulse.)
• Transmitted murmurs of aortic origin, most often due to aortic steno-
sis (less often due to coarctation, PDA, pulmonary stenosis, or VSD),
decrease in intensity as the stethoscope ascends the neck, whereas
a carotid bruit is usually louder higher in the neck and decreases in
intensity as the stethoscope is inched proximally toward the chest.
• Paradoxical pulse occurs classically with cardiac tamponade but
occurs occasionally with other restrictive cardiac abnormalities,
severe congestive heart failure, pulmonary embolism, and chronic
obstructive pulmonary disease.
• Pulsus alternans usually is associated with severe myocardial fail-
ure and is frequently accompanied by an S3, both of which impart
an ominous prognosis.
Abnormalities of the Femoral Pulse
In hypertension, simultaneous palpation of radial and femoral
pulses may reveal a delay or relative weakening of the femoral
pulses, suggesting aortic coarctation. The finding of a femo-
ral (or carotid) bruit in an adult suggests diffuse atherosclerosis.
Fibromuscular dysplasia is less common and occurs in younger
patients.
Heart Sounds
First Heart Sound
Only the M1 and T1 components of the S1 are normally audible.
The M1 occurs before the T1 and is the loudest component. Wide
splitting of the S1 occurs with right bundle branch block and
Ebstein anomaly.
Box 1.6. Pulsus Alternans and Electrical Alternans
Pulsus alternans
Severe heart failure
Electrical alternans
Pericardial tamponade
Large pericardial effusions
7
Factors Influencing the Intensity of the S1
PR Interval. The PR interval varies inversely with the loudness
of S1—with a long PR interval, the S1 is soft; conversely, with a
short PR interval, the S1 is loud.
Mitral Valve Disease. Mitral stenosis produces a loud S1 if the
valve is pliable. When the valve becomes calcified and immobile,
the intensity of the S1 decreases. The S1 may also be soft in severe
AR (related to early closure of the mitral valve) caused by LV
filling from the aorta.
The Rate of Increase of Systolic Pressure Within the LV. A
loud S1 can be produced by hypercontractile states, such as fever,
exercise, thyrotoxicosis, and pheochromocytoma. Conversely, a
soft S1 can occur in LV failure.
If the S1 seems louder at the lower left sternal border than at
the apex (implying a loud T1), suspect ASD or tricuspid stenosis.
Atrial fibrillation produces a variable S1 intensity. (The intensity
is inversely related to the previous R-R cycle length; a longer
cycle length produces a softer S1.) A variable S1 intensity during a
wide complex, regular tachycardia suggests atrioventricular dis-
sociation and ventricular tachycardia. The marked delay of the
T1 in Ebstein anomaly is related to the late billowing effect of
the deformed (sail-like) anterior leaflet of the tricuspid valve as it
closes in systole. Box 1.7 lists causes of an abnormal S1.
• If the S1 is louder at the base than at the apex, suspect an ejection
sound masquerading as the S1. If the S1 is louder at the lower left
sternal border than at the apex (implying a loud T1), suspect ASD
or tricuspid stenosis.
• A variable S1 intensity during a wide complex, regular tachycardia
suggests atrioventricular dissociation and ventricular tachycardia.
• The marked delay of the T1 in Ebstein anomaly is related to the late
billowing effect of the deformed (sail-like) anterior leaflet of the
tricuspid valve as it closes in systole.
Systolic Ejection Clicks
The ejection click (or sound) follows the S1 closely and can be
confused with a widely split S1 or, occasionally, with an early
Box 1.7. Abnormalities of the S1 and Their Causes
Loud S1
Short PR interval
Mitral stenosis
Left atrial myxoma
Hypercontractile states
Soft S1
Long PR interval
Depressed left ventricular function
Early closure of mitral valve in acute severe aortic
incompetence
Ruptured mitral valve leaflet or chordae
Left bundle branch block
Abbreviation: S1, first heart sound.
8 I Fundamentals of Cardiovascular Disease
nonejection click. Clicks can originate from the left or right side
of the heart.
The 3 possible mechanisms for production of the clicks are
as follows:
1. Intrinsic abnormality of the aortic or pulmonary valve, such as
congenital bicuspid aortic valve
2. Pulsatile distention of a dilated great artery, as occurs in increased
flow states such as truncus arteriosus (aortic click) or ASD (pulmo-
nary click) or in idiopathic dilatation of the pulmonary artery
3. Increased pressure in the great vessel, such as in aortic or pulmonary
hypertension
Because an aortic click is not usually heard with uncompli-
cated coarctation, its presence should suggest associated bicuspid
aortic valve. In the latter condition, the click diminishes in inten-
sity, becomes “buried” in the systolic murmur, and ultimately
disappears as the valve becomes heavily calcified and immo-
bile later in the course of the disease. Although a click implies
cusp mobility, its presence does not necessarily exclude severe
stenosis. A click would be expected to be absent in subvalvular
stenosis. The timing of the pulmonary click in relation to the S1
(reflecting the isovolumic contraction period of the RV) is asso-
ciated with hemodynamic severity in valvular pulmonary steno-
sis. With a higher systolic gradient and a lower pulmonary artery
systolic pressure, the isovolumic contraction period shortens and
thus the click occurs earlier in relation to the S1. A pulmonary
click can occur in idiopathic dilatation of the pulmonary artery,
and this condition may mimic ASD, especially in young adults.
The pulmonary click due to valvular pulmonary stenosis is the
only right-sided heart sound that decreases with inspiration. Most
other right-sided auscultatory events either increase in intensity
with inspiration (most commonly) or show minimal change. The
pulmonary click is best heard along the upper left sternal border,
but if it is loud enough or if the RV is markedly dilated, it may
be heard throughout the precordium. The aortic click radiates to
the aortic area and the apex and does not change with respiration.
The causes of ejection clicks are listed in Box 1.8.
• The presence, absence, or loudness of the ejection click does not
correlate with the degree of valvular stenosis.
• An aortic click is not usually heard with uncomplicated coarcta-
tion; its presence should suggest associated bicuspid aortic valve.
Box 1.8. Causes of Ejection Clicks
Aortic click
Congenital valvular aortic stenosis
Congenital bicuspid aortic valve
Truncus arteriosus
Aortic incompetence
Aortic root dilatation or aneurysm
Pulmonary click
Pulmonary valve stenosis
Atrial septal defect
Chronic pulmonary hypertension
Tetralogy of Fallot with pulmonary valve stenosis
(absent if there is only infundibular stenosis)
Idiopathic dilated pulmonary artery
• A click is absent in subvalvular or supravalvular aortic stenosis and
in hypertrophic obstructive cardiomyopathy.
• A pulmonary click can occur in idiopathic dilatation of the pulmo-
nary artery, a condition that may mimic ASD, especially in young
adults.
• The pulmonary click is best heard along the upper left sternal border.
• The aortic click radiates to the aortic area and the apex and does
not change with respiration.
Mid-to-Late Nonejection Clicks (Systolic Clicks)
Nonejection clicks are most commonly due to mitral valve pro-
lapse. Rarely, nonejection clicks can be caused by papillary
muscle dysfunction, rheumatic mitral valve disease, or hypertro-
phic obstructive cardiomyopathy. Other rare causes of nonejec-
tion clicks (which can masquerade as mitral prolapse) include
ventricular or atrial septal aneurysms, ventricular free wall
aneurysms, and ventricular and atrial mobile tumors, such as
myxoma. A nonejection click not due to mitral valve prolapse
does not have the typical responses to bedside maneuvers found
with mitral valve prolapse, as outlined below.
Mitral Valve Prolapse
Maneuvers that decrease LV volume, such as standing or the
Valsalva maneuver, move the click earlier in the cardiac cycle.
Conversely, maneuvers that increase LV volume, such as assum-
ing the supine position and elevating the legs, move the click later
in the cardiac cycle. With a decrease in LV volume, a systolic
murmur, if present, would become longer. Interventions that
increase systemic blood pressure make the murmur louder.
• Miscellaneous causes of nonejection clicks (which can masquerade
as mitral prolapse) include ventricular or atrial septal aneurysms,
ventricular free wall aneurysms, and ventricular and atrial mobile
tumors, such as myxoma.
• Maneuvers that decrease LV volume, such as standing or the Valsalva
maneuver, move the click earlier in the cardiac cycle. Conversely,
maneuvers that increase LV volume, such as assuming the supine posi-
tion and elevating the legs, move the click later in the cardiac cycle.
Second Heart Sound
The S2 is often best heard along the upper and middle left ster-
nal border. Splitting of the S2 (Figure 1.2) is best heard during
normal breathing with the patient in the sitting position.
Determinants of the S2 include the following:
1. Ventricular activation (bundle branch block delays closure of the
ventricle’s respective semilunar valve)
2. Ejection time
3. Valve gradient (increased gradient with low pressure in the great
vessel delays closure)
4. Elastic recoil of the great artery (decreased elastic recoil delays
closure, such as in idiopathic dilatation of the pulmonary artery)
Splitting of the S2
Wide but physiologic splitting of the S2 (Figure 1.3) may be due
to the following:
1. Delayed electrical activation of the RV, such as in right bundle
branch block or premature ventricular contraction originating in the
LV (which conducts with a right bundle branch block pattern)
2. Delay of RV contraction, such as in increased RV stroke volume and
RV failure
3. Pulmonary stenosis (prolonged ejection time)
 1 Cardiovascular Examination 9

Splitting?

Normal Narrow fixed Wide Reversed

Normal Pulm HT RBBB PDA (L to R shunt)

PS LBBB
ASD (“fixed”) AS
IHD

Pseudosplitting of S2:
S2 + OS
S2 + S3
S2 + pericardial
knock
S2 + tumor
plop

A2 vs P2?

A2 > P2 P2 > A2

Normal Pulm HT (any cause)

AS
Figure 1.2. Branching Logic Tree for S2 Splitting. A2 indicates aortic closure sound; AS, aortic stenosis; ASD, atrial septal defect; IHD, ischemic
heart disease; LBBB, left bundle branch block; L to R, left-to-right; OS, opening snap; P2, pulmonic closure sound; PDA, patent ductus arteriosus;
PS, pulmonary stenosis; Pulm HT, pulmonary hypertension; RBBB, right bundle branch block; S2, second heart sound; S3, third heart sound; VSD,
ventricular septal defect. (Previously published. See “Credit Lines” section.)

In ASD, there is only minimal respiratory variation in S2
splitting. This is referred to as fi xed splitting. Fixed splitting
should be verified with the patient in the sitting or standing posi-
tion because healthy persons occasionally appear to have fixed
splitting in the supine position. When the degree of splitting is
unusually wide, especially when the P2 is diminished, suspect
concomitant pulmonary stenosis. Indeed, this condition is the
cause of the most widely split S2 that can be recorded.
Wide fixed splitting, although considered typical of ASD,
occurs in only 70% of patients with ASD. However, persistent
expiratory splitting is audible in most. Normal respiratory var-
iation of the S2 occurs in up to 8% of patients with ASD. With
Eisenmenger physiology, the left-to-right shunting decreases and
the degree of splitting narrows. A pulmonary systolic ejection
murmur (increased flow) is common in patients with ASD, and
with a significant left-to-right shunt, a diastolic tricuspid flow
murmur can be heard also. Like aortic stenosis, as pulmonary
stenosis increases in severity, the P2 decreases in intensity, and
ultimately the S2 becomes single.
The wide splitting of the S2 in mitral regurgitation and VSD
is related to early aortic valve closure (in VSD, the P2 is delayed
also), which in turn is due to decreased LV ejection time, but
the loud pansystolic regurgitant murmur often obscures the wide
splitting of the S2 so that the S2 appears to be single.
Partial anomalous pulmonary venous connection may occur
alone or in combination with ASD (most often of the sinus veno-
sus type). Wide splitting of the S2 occurs in both conditions, but
it usually shows normal respiratory variation in isolated partial
anomalous pulmonary venous connection.
Pulmonary hypertension may cause wide splitting of the S2,
although the intensity of the P2 is usually increased and widely
transmitted throughout the precordium.
• Fixed splitting should be verified with the patient in the sitting or
standing position because healthy persons occasionally appear to
have fixed splitting in the supine position.
• Wide fixed splitting, although considered typical of ASD, occurs in
only 70% of patients with ASD.
• Wide splitting of the S2 occurs in both partial anomalous pulmonary
venous connection and ASD, but it usually shows normal respiratory
variation in isolated partial anomalous pulmonary venous connection.
• Pulmonary hypertension may cause wide splitting of the S2,
although the intensity of the P2 is usually increased and widely
transmitted throughout the precordium.
10 I Fundamentals of Cardiovascular Disease

obstructive cardiomyopathy implies a hemodynamically sig-

Insp Exp nificant resting LV outflow tract gradient. Transient paradoxi-
A P AP cal splitting of the S2 can occur with myocardial ischemia, such
as during an episode of angina, either alone or in combination
Normal with an apical systolic murmur of mitral regurgitation (papillary
muscle dysfunction) or prominent S4.
• When paradoxical splitting of the S2 occurs in association with aor-
A P A P
tic stenosis, usually in young adults (assuming left bundle branch
block is absent), severe aortic obstruction is suggested. Similarly,
ASD
paradoxical splitting in hypertrophic obstructive cardiomyopathy
implies a hemodynamically significant resting LV outflow tract
gradient.
• Transient paradoxical splitting of the S2 can occur with myocardial
A P A P ischemia, such as during an episode of angina, either alone or in
combination with an apical systolic murmur of mitral regurgitation
PS (papillary muscle dysfunction) or a prominent S4.

Intensity of the S2
A P AP Loud S2. Ordinarily, the intensity of the A2 exceeds that of the
P2. In adults, a P2 that is louder than the A2, especially if the P2 is
MI transmitted to the apex, implies either pulmonary hypertension
or marked RV dilatation with the RV occupying the apical zone.
The latter may occur in ASD (approximately 50% of patients).
Hearing 2 components of the S2 at the apex is abnormal in adults
PA P A because ordinarily only the A2 is heard at the apex. Thus, when
both components of the S2 are heard at the apex in adults, suspect
AS
ASD or pulmonary hypertension.
Soft S2. Decreased intensity of the A2 or P2, which may cause a
single S2, reflects stiffening and decreased mobility of the aortic
AP AP valve (aortic stenosis) or the pulmonary valve (pulmonary steno-
VSD sis). A single S2 may also be heard in older patients and in the
following situations:
1. Only 1 semilunar valve is functioning, such as in persistent truncus
arteriosus, pulmonary atresia, or tetralogy of Fallot
Figure 1.3. Normal and Abnormal Patterns in the Respiratory
2. One component of the S2 is enveloped in a long systolic murmur,
Variation of the Second Heart Sound. The heights of the bars are pro-
such as in VSD
portional to the sound intensity. Solid red bars represent the first heart
3. The great vessels have an abnormal relationship, such as in transpo-
sound. A indicates aortic component; AS, aortic stenosis; ASD, atrial
sition of the great arteries
septal defect; Exp, expiration; Insp, inspiration; MI, mitral incompe-
tence; P, pulmonary component; PS, pulmonary stenosis; VSD, ventric- • When both components of the S2 are heard at the apex in adults,
ular septal defect. (Previously published. See “Credit Lines” section.) implying an increased pulmonary component of the S2, suspect
ASD or pulmonary hypertension.

Paradoxical Splitting of the S2

Paradoxical (reversed) splitting of the S2 is usually caused by con-
ditions that delay aortic closure. Examples include the following:
1. Electrical delay of LV contraction, such as left bundle branch block
(most commonly)
2. Mechanical delay of LV ejection, such as aortic stenosis and hyper-
trophic obstructive cardiomyopathy
3. Severe LV systolic failure of any cause
4. PDA, AR, and systemic hypertension are other rare causes of para-
doxical splitting
Paradoxical splitting of the S2 (ie, with normal QRS duration)
may be an important bedside clue to significant LV dysfunction.
In severe aortic stenosis, the paradoxical splitting is only rarely
recognized because the late systolic ejection murmur obscures
the S2. However, when paradoxical splitting of the S2 occurs in
association with aortic stenosis, usually in young adults (assum-
ing left bundle branch block is absent), severe aortic obstruction
is suggested. Similarly, paradoxical splitting in hypertrophic
Opening Snap
A high-pitched snapping sound related to mitral or tricuspid
valve opening, when present, is abnormal and is referred to as
an OS. An OS arises from a doming stenotic tricuspid valve or,
more commonly, mitral valve. The intensity of an OS correlates
with valve mobility. Rarely, an OS occurs without atrioventric-
ular valve stenosis in conditions associated with increased flow
through the valve, such as hemodynamically significant mitral
regurgitation.
In mitral stenosis, the presence of an OS, often accompanied
by a loud S1, implies a pliable mitral valve. The OS is often well
transmitted to the left sternal border and even to the aortic area.
In mitral stenosis, the absence of an OS implies the following:
1. Severe valvular immobility and calcification (an OS can still be
heard in some of these cases)
2. Mitral regurgitation is the predominant lesion
• Significant mitral stenosis may be present without an OS if the
mitral valve leaflets are fixed and immobile.
 1 Cardiovascular Examination
S2 -OS Interval
The S2–mitral OS interval reflects the isovolumic relaxation
period of the LV. With increased severity of mitral stenosis and
greater increase in LA pressures, the S2-OS interval becomes
shorter and may be confused with a split S2. The S2-OS interval
should not vary with respiration. The S2-OS interval widens on
standing, whereas the split S2 either does not change or nar-
rows. Mild mitral stenosis is associated with an S2-OS interval
of more than 90 ms, and severe mitral stenosis with an interval
of less than 70 ms. However, the S2-OS interval is an unreli-
able predictor of the severity of mitral stenosis. Other factors
that increase LA pressures, such as mitral regurgitation or LV
failure, can also affect this interval. When the S2-OS interval is
more than 110 to 120 ms, the OS may be confused with an LV
S3. In comparison, the LV S3 is usually low-pitched and is local-
ized to the apex.
A tricuspid valve OS caused by tricuspid stenosis can be
recognized by its location along the left sternal border and its
increase with inspiration. In normal sinus rhythm, a prominent a
wave can be seen in the jugular venous pulse, along with slowing
of the y descent.
An LV S3, which implies that rapid LV filling can occur, is
rare in pure mitral stenosis. Also, an RV S3 can occur in mitral
stenosis with severe secondary pulmonary hypertension and
RV failure. An RV S3 is found along the left sternal border and
increases with inspiration. A tumor “plop” due to an atrial myx-
oma has the same early diastolic timing as an OS and can be
confused with it.
• In mitral stenosis, the presence of an OS, often accompanied by
a loud S1, implies a pliable mitral valve that is not heavily calci-
fied. (In such cases, the patient may be a candidate for mitral com-
missurotomy or balloon valvuloplasty rather than mitral valve
replacement.)
• In general, mild mitral stenosis is associated with an S2-OS inter-
val >90 ms, and severe mitral stenosis with an interval <70 ms.
• A tumor “plop” due to atrial myxoma has the same early diastolic
timing as an OS and can be confused with it.
Third Heart Sound
The exact mechanism of S3 production is controversial, but its
timing relates to the peak of rapid ventricular filling with rapid
flow deceleration. Factors related to S3 intensity include the
following:
1. Volume and velocity of blood flow across the atrioventricular valve
2. Ventricular relaxation and compliance
Although a physiologic S3 can be heard in young healthy
persons, it should not be audible after age 40. An RV S3 may
be augmented with inspiration. The physiologic S3 may disap-
pear in the standing position; the pathologic S3 persists. An S3
in a patient with mitral regurgitation implies severe regurgita-
tion or a failing LV or both. The presence of a diastolic flow
rumble (“relative” mitral stenosis) after the S3 suggests severe
mitral regurgitation. An S3 is less common in conditions that
cause thick, poorly compliant ventricles (eg, LV hypertrophy
that occurs with pressure overload states, such as aortic stenosis
or hypertension) until late in the disease. An S3 may occur in
hypertrophic obstructive cardiomyopathy with normal systolic
function.
The pericardial knock of constrictive pericarditis is similar to
an S3 and is associated with sudden arrest of ventricular expansion
in early diastole. The pericardial knock is of higher frequency
11
Box 1.9. Causes of the S3
Physiologic in young adults and children
Severe left ventricular dysfunction of any cause
Left ventricular dilatation without failure
Mitral regurgitation
Ventricular septal defect
Patent ductus arteriosus
Right ventricular S3 in right ventricular failure
and severe tricuspid regurgitation
Pericardial knock in constrictive pericarditis
S3 is augmented in intensity with increased
venous return
Leg elevation
Exercise
Release phase of Valsalva maneuver
S3 is augmented in intensity with increased
systemic peripheral resistance
Sustained handgrip
Abbreviation: S3, third heart sound.
than the S3, occurs slightly earlier in diastole, may vary with res-
piration, and is more widely transmitted. The causes of the S3 are
listed in Box 1.9.
• An S3 in a patient with mitral regurgitation implies severe regurgi-
tation or a failing LV or both.
• An S3 is less common in conditions that cause thick, poorly com-
pliant ventricles (eg, LV hypertrophy that occurs with pressure
overload states).
• The pericardial knock is of higher frequency than the S3, occurs
slightly earlier in diastole, may vary with respiration, and is more
widely transmitted.
Fourth Heart Sound
The S4, thought to originate within the ventricular cavity, results
from a forceful atrial contraction into a ventricle having limited
distensibility, such as in hypertrophy or fibrosis. It is not heard in
healthy young persons or in atrial fibrillation.
Common pathologic states in which an S4 is often present
include the following:
1. Aortic stenosis
2. Hypertension
3. Hypertrophic obstructive cardiomyopathy
4. Pulmonary stenosis
5. Ischemic heart disease
As the S4 becomes closer to the S1, the intensity of the S1 increases.
Sitting or standing may attenuate the S4. A loud S4 can be heard in
acute mitral regurgitation (eg, with ruptured chordae tendineae)
or regurgitation of recent onset (the LA has not yet significantly
dilated). With chronic mitral regurgitation due to rheumatic dis-
ease, the LA dilates, becomes more distensible, and generates a less
forceful contraction. Under these circumstances, an S4 is usually
absent. An S4 can still be heard in patients with LV hypertrophy or
ischemic heart disease, despite enlargement of the LA.
12 I Fundamentals of Cardiovascular Disease
Although an S4 can be heard in otherwise healthy elderly
patients, a palpable S4 (a wave) should not be present unless the
LV is abnormal. An S4 can originate from the RV. A right-sided
S4 is increased in intensity with inspiration, is often associated
with large jugular venous a waves, and is best heard along the
left sternal border rather than at the apex (this is the usual site
of an LV S4).
In patients with aortic stenosis who are younger than 40 years,
the presence of an S4 usually indicates significant obstruction.
Similarly, the presence of right-sided S4, in association with pul-
monary stenosis, indicates severe pulmonary valve obstruction.
An S4 is present in most patients with hypertrophic obstructive
cardiomyopathy, in patients with acute myocardial infarction,
and often in patients with systemic hypertension.
• A loud S4 can be heard in acute mitral regurgitation (eg, with rup-
tured chordae tendineae) and can be a clue that the regurgitation
is of recent onset.
• Although an S4 can be heard in otherwise healthy elderly patients,
a palpable S4 (a wave) should not be present unless the LV is
abnormal.
• An S4 is present in most patients with hypertrophic obstructive
cardiomyopathy, in patients with acute myocardial infarction, and
often in patients with systemic hypertension.
Cardiac Murmurs
Systolic Murmurs
Systolic murmurs (Figure 1.4) may be divided into 2 categories:
1. Ejection types, such as aortic or pulmonary stenosis
2. Pansystolic or regurgitant types, such as mitral regurgitation, tricus-
pid regurgitation, or VSD
Most, but not all, systolic murmurs fit into this simple classifi-
cation scheme. Factors that differentiate the various causes of LV
outflow tract obstruction are shown in Table 1.3. The effects of
various maneuvers on murmurs are shown in Figure 1.5.
Aortic and Pulmonary Stenosis
Stenosis of the aortic or pulmonary valve delays the peak intensity
of the systolic murmur by prolonging ejection. The magnitude of
the delay is proportional to the severity of the obstruction, but
the intensity (loudness) of an ejection systolic murmur may not
reflect the severity of the obstruction. Thus, for example, a patient
with mild aortic stenosis or a normal mechanical aortic prosthe-
sis and increased cardiac output may have a loud murmur (grade
3 or 4). Conversely, a patient with severe aortic stenosis and low
cardiac output may have only a grade 1 or 2 murmur. However,
the peak intensity may still be delayed. For valvular pulmonary
stenosis, early stenosis is suggested by an early ejection click, a
widely split S2, and delayed peak intensity of a systolic murmur.
Differentiation of Aortic Stenosis From Aortic Sclerosis
A frequent clinical problem is the differentiation of aortic stenosis
from benign aortic sclerosis. With aortic sclerosis, there should
be no other clinical, electrocardiographic, or radiographic evi-
dence of heart disease. The systolic murmur is generally grade
1 or 2 and peaks early. The carotid upstroke should be normal.
A normal S2 (ie, the A2 is preserved) supports a benign process,
but the S2 can appear single in healthy elderly subjects. The sys-
tolic murmur of aortic stenosis, in contrast, is delayed (peaking
late in systole) and is usually louder, and the carotid pulse is
weakened and delayed (pulsus parvus et tardus) (an exception
is elderly persons, who may have normal carotid pulses with sig-
nificant aortic stenosis). The apical impulse in aortic stenosis is
frequently abnormal also (see the “Abnormalities on Palpation of
the Precordium” section above).
Supravalvular Aortic Stenosis
The systolic murmur of supravalvular aortic stenosis is maxi-
mal in the first or second right intercostal space, and a carotid
pulse inequality may be present (see the “Abnormalities of the
Carotid Pulse” section above). Patients are usually young. (The
differential diagnosis of LV outflow tract obstruction is shown
in Table 1.3.)
Mitral Regurgitation
Although the murmur of mitral regurgitation is usually pansys-
tolic, the timing can be late systolic (which would suggest mitral
prolapse, papillary muscle dysfunction, or, less commonly, rheu-
matic disease). The systolic murmur of mitral regurgitation can
also be early systolic in acute severe mitral regurgitation with
markedly increased LA pressures, reducing the late systolic
LV-LA gradient. In these cases, the patients are usually hemo-
dynamically unstable and have evidence of significant pulmo-
nary congestion. The systolic murmur of severe chronic mitral
regurgitation is usually loud (grade 3 or 4 or louder). The systolic
murmur of severe acute mitral regurgitation can have variable
intensity, especially in low cardiac output states or shock (such
as acute myocardial infarction with LV dysfunction and papil-
lary muscle dysfunction). Under these circumstances, the sys-
tolic murmur may be unimpressive or even absent. The systolic
murmur of posterior mitral leaflet syndrome can be transmitted
to the aortic area and be confused with aortic stenosis. Except in
the elderly, palpation of the carotid pulse helps differentiate these
2 conditions. In about 15% of cases, pure aortic stenosis causes a
localized apical systolic murmur. Auscultation during inhalation
of amyl nitrite can help differentiate this murmur from mitral
regurgitation (Tables 1.4 and 1.5). The systolic murmur of ante-
rior mitral leaflet syndrome is transmitted posteriorly and can be
heard along the thoracic spine and even at the base of the skull.
Tricuspid Regurgitation
The systolic murmur of tricuspid regurgitation is usually best
heard at the lower left sternal border or over the xiphisternum, but
it may also be heard to the right of the sternum, over the apicos-
ternal area, or over the apex (if the RV is sufficiently dilated and
occupies the position usually taken by the LV). The systolic mur-
mur of significant tricuspid regurgitation may be subtle or even
inaudible clinically, but the jugular venous pulse may have large v
waves. Inspiration may accentuate the murmur of tricuspid regur-
gitation but not consistently so, and the absence of inspiratory
augmentation does not exclude tricuspid regurgitation (with severe
tricuspid regurgitation, the x descent becomes obliterated).
Hypertrophic Obstructive Cardiomyopathy
Patients with hypertrophic obstructive cardiomyopathy can have
3 different types and locations of systolic murmurs:
1. Mid to lower left sternal border (LV outflow tract obstruction)
2. Apex (associated mitral regurgitation)
3. Upper left sternal border (RV outflow tract obstruction)—uncommon
(a bedside clue is a prominent jugular venous a wave).
 1 Cardiovascular Examination 13

Figure 1.4. Sketches of Various Murmurs and Heart Sounds. A1, Short midsystolic murmur with normal aortic (A2) and pulmonic (P2) compo-
nents of the second heart sound (S2)—findings consistent with an innocent murmur. A2, Holosystolic murmur that decreases in the latter part of
systole—a configuration observed in acute mitral regurgitation. A3, An ejection sound, a short early systolic murmur, and an accentuated, closely
split S2—consistent with pulmonary hypertension, as with an Eisenmenger ventricular septal defect (VSD). B1, Early to midsystolic murmur with
vibratory component—typical of an innocent murmur. B2, An ejection sound followed by a diamond-shaped murmur and wide splitting of S2 that
may be present with atrial septal defect (ASD) or mild pulmonic stenosis; an ejection sound is more likely with valvular pulmonic stenosis. B3,
Crescendo-decrescendo systolic murmur, not holosystolic; the third heart sound (S3) and fourth heart sound (S4) are present—findings consistent
with mitral systolic murmur heard in congestive cardiomyopathy or coronary artery disease with papillary muscle dysfunction and cardiac decom-
pensation. C1, Longer, somewhat vibratory crescendo-decrescendo systolic murmur with wide splitting of S2. If S2 becomes fused with expiration,
ASD is less likely; if the remainder of the cardiovascular evaluation is normal, this finding is consistent with an innocent murmur. C2, Midsystolic
murmur and wide splitting of S2 that was “fixed”—findings typical of ASD. C3, Prolonged diamond-shaped systolic murmur masking A2 with P2, S4,
and ejection sound—findings typical of valvular pulmonic stenosis of moderate severity. D1, Late apical systolic murmur of prolapsing mitral valve
leaflet. D2, Systolic click—late apical systolic murmur of prolapsing mitral leaflet syndrome. D3, S4 and midsystolic murmur consistent with mitral
systolic murmur of cardiomyopathy or ischemic heart disease. E1, Early crescendo-decrescendo systolic murmur ending in midsystole consistent with
innocent murmur or VSD. E2 and E3, Holosystolic murmur consistent with mitral or tricuspid regurgitation or VSD with pulmonary hypertension.
(Previously published. See “Credit Lines” section.)
14 I Fundamentals of Cardiovascular Disease

Table 1.3. Factors That Differentiate Various Causes of Left Ventricular Outflow Tract Obstruction
Feature Valvular Supravalvular Discrete Subvalvular HOCM
Valve calcification Common after age 40 y Absent Absent Absent
Dilated ascending aorta Common Rare Rare Rare
PP after VPB Increased Increased Increased Decreased
Valsalva effect on SM Decreased Decreased Decreased Increased
Murmur of AR Common Rare Sometimes Absent
Fourth heart sound If severe Uncommon Uncommon Common
Paradoxical splitting Sometimes Absent Absent Common
Ejection click Most (unless valve calcified) Absent Absent Uncommon or absent
Maximal thrill and murmur RIS 2 RIS 1 RIS 2 LIS 4
Carotid pulse Normal to diminished and delayed Unequal Normal to diminished Brisk, jerky, systolic rebound, bifid
Abbreviations: AR, aortic regurgitation; HOCM, hypertrophic obstructive cardiomyopathy; LIS, left intercostal space; PP, pulse pressure; RIS, right intercostal
space; SM, systolic murmur; VPB, ventricular premature beat.
(Previously published. See “Credit Lines” section.)

Frequently, the louder systolic murmur at the mid left sternal
border, which can be widely transmitted, may merge with or
mask the others.
Ventricular Septal Defect
Depending on the size of the defect and the pressure gradient
between the LV and the RV, the systolic murmur of VSD is
typically pansystolic and associated with a thrill along the left
sternal border, but the murmur can be variable in contour and
the thrill absent. The murmur parallels the pressure difference
between the 2 ventricles (in turn related to pulmonary and sys-
temic vascular resistances). With significant pulmonary hyper-
tension, the murmur duration shortens and may resemble an early
systolic ejection-type murmur. If the maximal intensity of the
systolic murmur is in the first and second left intercostal spaces

Changes in Intensity of Systolic Murmur

Diagnosis Systolic Second Heart Amyl Phenyl-

Murmur Sound Erect Squatting Nitrite ephrine

Hypertrophic Variable (reversed,

obstructive partially reversed,
cardiomyopathy narrow, or normal)

Mitral incompetence Variable; Widely split, but A2

(pure and severe) see text masked by murmur

Papillary muscle Normal or

dysfunction partially reversed or

Billowing posterior Normal

leaflet or

Rheumatic disease Slightly wide

(moderate)

Narrow split or
Valvular Mild to moderate partially reversed
aortic
stenosis Marked Single or reversed

Ventricular septal Slightly wide

defect or

Innocent vibratory Normal

systolic murmur

Figure 1.5. Character of Systolic Murmur and Second Heart Sound in Various Conditions. The effects of posture (erect or squatting), amyl nitrite
inhalation, and phenylephrine injection on the intensity of the murmur are shown. Arrows indicate increase (pointing up), decrease (pointing down),
and degree of change (length of arrow). Dashes indicate no change from control. A2 indicates aortic valve component of the second heart sound.
(Previously published. See “Credit Lines” section.)
 1 Cardiovascular Examination 15

Table 1.4. Effect of Selected Physiologic Changes and Physical or Pharmacologic

Maneuvers on Common Cardiac Murmurs
Effect on Murmur

Feature Augmented Little or No Change Decreased

Amyl nitrite HOCM MR (SM)
AS, PS VSD (SM)
Innocent SM AR (DM)
MS Austin Flint (DM)
TS
Handgrip AR AS, TR HOCM
MR PR
VSD TS
MSa
Long cardiac cycle length (eg, atrial AS MR
fibrillation, premature ventricular PS AR
contraction) HOCM
Valsalva maneuver HOCM AS
MV prolapseb PS
Posture
Standing HOCM AS
MV prolapseb PS
Squatting AR, MR, VSD HOCM
MV prolapsec
Abbreviations: AR, aortic regurgitation; AS, aortic stenosis; DM, diastolic murmur; HOCM, hypertrophic
obstructive cardiomyopathy; MR, mitral regurgitation; MS, mitral stenosis; MV, mitral valve; PR, pulmonary
regurgitation; PS, pulmonary stenosis; SM, systolic murmur; TR, tricuspid regurgitation; TS, tricuspid
stenosis; VSD, ventricular septal defect.
a
Related to increased cardiac output.
b
Duration of systolic murmur increased (early onset); variable augmentation effect.
c
Duration of systolic murmur decreased (later onset); variable intensity effect.

with radiation to the left clavicle, suspect supracristal VSD or
PDA. The systolic murmur of multiple VSDs is indistinguish-
able from that of single defects. The same is true for LV–right
atrial shunts. The loud pansystolic murmur of VSD may mask
associated defects, such as PDA. A wide pulse pressure suggests
PDA or associated AR. The combination of VSD and AR may
suggest PDA, but the systolic murmur in PDA—not the murmur
in the VSD-AR combination—peaks at the S2. A systolic mur-
mur in the posterior thorax may be caused by the following:
1. Coarctation
2. Aortic dissection
3. Anterior mitral leaflet syndrome (with a posteriorly directed jet of
mitral regurgitation)
4. Peripheral pulmonary artery stenosis
5. Pulmonary arteriovenous fistula
• Stenosis of a semilunar valve delays the peak intensity of the sys-
tolic murmur by prolonging ejection. The magnitude of the delay
is proportional to the severity of the obstruction.
• For valvular pulmonary stenosis, severe stenosis is suggested by an
early ejection click, a widely split S2, and delayed peak intensity of
a systolic murmur.
• The systolic murmur of supravalvular aortic stenosis is maximal
in the first or second right intercostal space, and a carotid pulse
inequality may be present.
• Although the murmur of mitral regurgitation is usually pansystolic,
the timing can be late systolic (which would suggest mitral pro-
lapse, papillary muscle dysfunction, or, less commonly, rheumatic
disease).
• An early systolic murmur of mitral regurgitation can be heard in
acute severe cases with markedly increased LA pressures, reduc-
ing the late systolic LV-LA gradient.
• The systolic murmur of posterior mitral leaflet syndrome can
be transmitted to the aortic area and be confused with aortic
stenosis.
• The systolic murmur of anterior mitral leaflet syndrome is trans-
mitted posteriorly and can be heard along the thoracic spine and
even at the base of the skull.
• Inspiration may accentuate the murmur of tricuspid regurgitation
but not consistently so, and the absence of inspiratory augmenta-
tion does not exclude tricuspid regurgitation.
• The systolic murmur of VSD is typically pansystolic and associ-
ated with a thrill along the left sternal border, but the murmur can
be variable in contour.
• If the maximal intensity of a systolic murmur is in the first and
second left intercostal spaces with radiation to the left clavicle,
suspect supracristal VSD or PDA.
• A loud pansystolic murmur of VSD may mask associated
defects, such as PDA. A wide pulse pressure suggests PDA or
associated AR.
• The combination of VSD and AR may suggest PDA, but the mur-
mur in PDA—not the murmur in the VSD-AR combination—peaks
at the S2.
Innocent Systolic Murmurs
Innocent systolic murmurs are generally related to increased
blood flow or turbulence across a semilunar valve, especially
the aortic valve. These murmurs are common at all ages. In
young patients, they are apt to be heard over the pulmonary area.
Innocent systolic murmurs usually are soft (grade 2 or less) and
short (never pansystolic) and have no associated abnormal clini-
cal findings (eg, S2 is normal and there are no clicks). In older
16 I Fundamentals of Cardiovascular Disease

Table 1.5. Effect of Amyl Nitrite and Vasopressors on Various Murmurs

Diagnosis Amyl Nitrite Phenylephrine
Systolic Murmurs
Mitral insufficiency Decrease Increase
Ventricular septal defect Decrease Increase
Patent ductus arteriosus Decrease Increase
Tetralogy of Fallot Decrease Increase
Atrial septal defect Increase Increase or no change
Idiopathic hypertrophic subaortic stenosis Increase Decrease
Aortic stenosis (valvular) Increase No change
Pulmonary stenosis (valvular and muscular) Increase No change
Tricuspid insufficiency Increase No change
Systolic ejection murmur (innocent) Increase Decrease
Diastolic Murmurs
Aortic insufficiency Decrease Increase
Austin Flint murmur Decrease Increase
Mitral stenosis Increase Decrease
Pulmonary insufficiency Increase No change
Pulmonary insufficiency due to Decrease Increase
Eisenmenger syndrome
Tricuspid stenosis Increase No change
(Previously published. See “Credit Lines” section.)

patients, they generally emanate from a sclerotic aortic valve or
dilated aortic root and can be heard at the aortic area, left ster-
nal border, or apex. If heard at the apex, they may be confused
with the murmur of mitral regurgitation. In younger patients, an
innocent systolic murmur may originate from the RV outflow
tract or pulmonary artery. A PDA or VSD can masquerade as an
innocent murmur.
An innocent systolic murmur heard at the lower left sternal
border should be differentiated from the systolic murmur of
VSD, tricuspid regurgitation, infundibular pulmonary stenosis,
or hypertrophic obstructive cardiomyopathy. When uncertain
about the cause of a systolic murmur, a Valsalva maneuver should
be performed (Table 1.4). Findings that suggest that a systolic
murmur is pathologic are listed in Box 1.10.
• Innocent systolic murmurs usually are soft (grade 2 or less) and
short and have no associated abnormal clinical findings.
• In younger patients, an innocent systolic murmur often originates
from the RV outflow tract or pulmonary artery.
• A PDA or VSD can masquerade as an innocent murmur.
Box 1.10. Findings That Suggest That a Systolic
Murmur is Pathologic
Loud (grade 3 or more)
Long duration
Associated with ejection or nonejection click
Loud S1, A2 , or P2
Presence of an opening snap
Presence of left or right ventricular hypertrophy or
heave
Fixed or expiratory splitting of S 2
Abbreviations: A2, aortic component of second heart sound; P2,
pulmonary component of second heart sound; S1, first heart sound,
S2, second heart sound.
Diastolic Murmurs
In general, the loudness of a diastolic murmur correlates with the
severity of the underlying abnormality.
Aortic Regurgitation
The murmur of mild AR may be difficult to hear and may be
clinically “silent.” This murmur is best heard with the patient sit-
ting and leaning forward while holding expiration. Consider AR
when there is a wide arterial pulse pressure, especially in young
or middle-aged patients (older patients may have generalized
aortic atherosclerosis causing wide pulse pressure). The murmur
of AR is typically early diastolic (immediately after the S2) and
decrescendo. In mitral stenosis, an early diastolic murmur may be
caused by concomitant AR or pulmonary regurgitation (Graham
Steell murmur) but more often by concomitant AR. Severe AR,
especially if acute, may be associated with markedly increased
LV end-diastolic pressures. These pressures decrease the gradient
between the aorta and the LV in diastole, and the murmur tapers
rapidly. Thus, a short, early diastolic murmur does not exclude sig-
nificant acute AR, especially if the patient has evidence of acute
heart failure. A patient with severe AR due to infective endo-
carditis may present in this way. In mild AR, the LV end-diastolic
pressure remains normal, the gradient persists throughout most of
diastole, and the murmur may persist longer into diastole. With
severe, chronic AR, there is often a wide pulse pressure (with
hyperdynamic pulses), a systolic ejection murmur that usually
peaks early (related to increased aortic flow), decreased diastolic
blood pressure, and LV enlargement detected by palpation.
The anatomical location of the aortic valve is not under the
second right intercostal space (the “aortic area”) but is situated
lower in the thorax under the midsternum, although the “jet”
of aortic stenosis is often best heard in the aortic area. The
murmur of AR is often best heard along the left sternal border.
When it is primarily transmitted down the right sternal border,
one should suspect diseases of the aortic root, such as aortic
aneurysm or dissection. The combination of hypertension, chest
pain, and right sternal border transmission of the AR murmur
suggests proximal aortic dissection. When the AR is of valvular
 1 Cardiovascular Examination
origin, the murmur can be heard in the aortic area, but it is also
transmitted along the left sternal border and to the apex.
• Consider AR when there is a wide arterial pulse pressure, espe-
cially in young or middle-aged patients.
• In the presence of mitral stenosis, an associated early diastolic
murmur may be due to concomitant AR or pulmonary regurgita-
tion (Graham Steell murmur) but more often to concomitant AR.
• A short, early diastolic murmur does not exclude significant acute
AR, especially if the patient has evidence of acute heart failure.
• The murmur of AR is often best heard at the left sternal border.
When it is primarily transmitted down the right sternal border, one
should suspect diseases of the aortic root, such as aortic aneurysm
or dissection.
• The combination of hypertension, chest pain, and right sternal
border transmission of the AR murmur suggests proximal aortic
dissection.
Austin Flint Murmur
An Austin Flint murmur is related to mitral inflow turbu-
lence caused by the AR jet and implies a significant AR leak.
Because this may produce an apical diastolic rumble that is mid-
diastolic with presystolic accentuation, it may be confused with
mitral stenosis. The presence of radiographic LA enlargement
or atrial fibrillation favors mitral stenosis rather than isolated
AR. Administration of amyl nitrite can help differentiate these
murmurs (Table 1.5): the Austin Flint murmur decreases (as the
LV afterload decreases), whereas the mitral stenosis murmur
increases (as do all valvular stenotic murmurs in response to
amyl nitrite). Also, there should be no OS or other features of
mitral valve disease. Obviously, a patient with rheumatic heart
disease can have both AR and mitral stenosis. When AR has a
“honking” or “cooing” quality, consider a perforated, everted, or
ruptured aortic cusp, such as with infective endocarditis.
• With administration of amyl nitrite, the Austin Flint murmur
decreases (as the LV afterload decreases), whereas the murmur of
mitral stenosis increases (as do all valvular stenotic murmurs in
response to amyl nitrite).
• When AR has a “honking” or “cooing” quality, consider a per-
forated, everted, or ruptured aortic cusp, such as with infective
endocarditis.
Pulmonary Regurgitation
Although the murmur of pulmonary regurgitation may sound sim-
ilar to the murmur of AR, it is usually localized to the pulmonary
area and, like most right-sided events, gets louder with inspiration.
The murmur characteristics depend on the cause. The murmur of
pulmonary regurgitation due to pulmonary hypertension begins in
early diastole (immediately after P2) and is long and high-pitched.
In comparison, the murmur of pulmonary regurgitation due to
organic pulmonary valve disease is lower pitched, harsher, and
rumbling, beginning slightly later in diastole and often ending in
mid-diastole. Pulmonary regurgitation, especially when mild or
even moderate, is frequently inaudible. In the presence of mitral
stenosis, an early diastolic murmur heard at the left sternal border
is more likely to be AR than pulmonary regurgitation.
• The murmur of pulmonary regurgitation due to pulmonary hyper-
tension begins in early diastole and is long and high-pitched. In
comparison, the murmur of pulmonary regurgitation due to organic
pulmonary valve disease is lower pitched, harsher, and rumbling,
beginning slightly later in diastole and often ending in mid diastole.
17
Mitral Stenosis
The diastolic murmur of mitral stenosis is very localized (to the
apex), is low-pitched, and begins at the time of mitral valve open-
ing. The presence of a loud S1 or an OS should prompt a careful
search for this easily overlooked diastolic murmur. With the patient
in the left lateral decubitus position, the stethoscope may have to
be inched around the apical region to find the highly localized,
subtle, flow rumble of mitral stenosis. If the murmur is not audi-
ble, exercise (such as sit-ups) may augment mitral flow and bring
out the murmur. Other provocative maneuvers that increase flow
across the mitral valve, such as administration of amyl nitrite, also
augment the murmur of mitral stenosis (Table 1.4). The duration
of the diastolic murmur is related to the severity of mitral stenosis,
persisting as long as there is a significant pressure gradient across
the mitral valve. Therefore, a pandiastolic murmur implies severe
mitral stenosis. The murmur may crescendo in late diastole (pre-
systolic accentuation), even in atrial fibrillation, suggesting that
atrial contraction is not required for this phenomenon.
Rarely in mitral stenosis, the diastolic murmur is not heard
(“silent” mitral stenosis). The usual reasons for silent mitral
stenosis are the following:
1. Improper auscultation (most commonly)
2. Very mild mitral stenosis
3. A decrease in flow rates across the mitral valve, such as in severe
congestive heart failure or concomitant aortic or tricuspid stenosis
4. Abnormal chest wall configuration limiting auscultation, such as in
obesity or severe chronic obstructive pulmonary disease, in which
case all sounds should be indistinct or distant
Consider mitral stenosis and focus the cardiac examination
accordingly with new onset of atrial fibrillation or when atrial
fibrillation occurs with any of the following:
1. Stroke or other systemic or peripheral embolus (an atrial myxoma
may also manifest in this way)
2. “Unexplained” pulmonary hypertension
3. “Unexplained” congestive heart failure
4. “Unexplained” recurrent pleural effusions
• The duration of the diastolic murmur is related to the severity of
mitral stenosis, persisting as long as there is a significant pressure
gradient across the mitral valve.
• Even in the apparent absence of a murmur, important ausculta-
tory clues to the presence of mitral stenosis include a loud S1 or
an OS.
• Consider mitral stenosis when atrial fibrillation occurs with any
of the following: 1) stroke or other systemic or peripheral embo-
lus, 2) “unexplained” pulmonary hypertension, 3) “unexplained”
congestive heart failure, and 4) “unexplained” recurrent pleural
effusions.
Tricuspid Stenosis
The bedside differentiation between tricuspid stenosis and mitral
stenosis includes the following:
1. Response to inspiration—murmur of tricuspid stenosis increases
2. Location—the diastolic murmur of tricuspid stenosis is best heard
at the left sternal border, whereas the murmur of mitral stenosis is
localized to the apex. The associated OS, if present, augments with
inspiration
3. Frequency—tricuspid stenosis is higher in frequency and begins
earlier in diastole than mitral stenosis (these differences may be dif-
ficult to appreciate at the bedside)
4. Large jugular venous a wave with a slow y descent—suggestive of
tricuspid stenosis (other causes of large a waves, including pulmo-
nary stenosis and pulmonary hypertension, should not interfere with
RV filling and therefore are not associated with a slow y descent)
18 I Fundamentals of Cardiovascular Disease
Rarely, there may be a diastolic thrill palpable along the lower
left sternal border and hepatic (presystolic) pulsation. Other
causes of RV inflow obstruction, such as thrombus or extrinsic
RV compression, can masquerade as tricuspid stenosis.
Tricuspid stenosis usually occurs in patients with rheumatic
heart disease (although there are rarer causes such as carcinoid).
In patients with rheumatic heart disease, especially females,
concomitant mitral valve disease is almost always present. The
clinical finding of left-sided valve lesions often overshadows the
tricuspid involvement, and the murmur of tricuspid stenosis may
be mistaken for aortic or pulmonary regurgitation.
• A large jugular venous a wave with a slow y descent should suggest
tricuspid stenosis.
• The clinical finding of left-sided valve lesions often overshadows
the tricuspid involvement, and the murmur of tricuspid stenosis
may be mistaken for aortic or pulmonary regurgitation.
Mid-Diastolic Flow Murmurs
Almost any condition that increases flow across atrioventricular
valves (such as mitral regurgitation, PDA, intracardiac shunts, or
complete heart block) can also cause a short mid-diastolic flow
rumble (functional diastolic murmur) in the absence of organic
atrioventricular valve stenosis. (Actually, the rumble begins in
early rather than mid-diastole, but it is delayed in comparison
with the early diastolic murmur of semilunar valve regurgita-
tion.) The murmur may begin after a prominent S3 and does not
show presystolic accentuation.
• Almost any condition that increases flow across atrioventricular
valves (such as mitral regurgitation, PDA, intracardiac shunts, or
complete heart block) can also cause a short mid-diastolic flow
rumble (functional diastolic murmur) in the absence of organic
atrioventricular valve stenosis.
Continuous Murmurs
Continuous murmurs should be differentiated from to-and-fro
murmurs (such as occur in combined aortic stenosis and AR).
In AR, the systolic component decreases before S2, whereas the
continuous murmur of PDA, for example, typically peaks at S2.
Murmurs caused by coronary arteriovenous fistula, venous hum,
and ruptured sinus of Valsalva aneurysm peak later in diastole.
Murmurs due to dilated bronchial vessels, such as in pulmonary
atresia, can be heard anywhere in the chest, axillae, or back.
When a continuous murmur is loudest in the posterior thorax,
consider the following:
1. Coarctation
2. Pulmonary arteriovenous fistula
3. Peripheral pulmonary stenosis
• Continuous murmurs should be differentiated from to-and-fro
murmurs (such as occur in combined aortic stenosis and AR).
In AR, the systolic component decreases before S2, whereas the
continuous murmur of PDA typically peaks at S2.
Bedside Physiologic Maneuvers to Differentiate
Different Types of Murmurs
Several bedside physiologic maneuvers can be used to distin-
guish types of murmurs (Table 1.4).
Valsalva Maneuver
The Valsalva maneuver is useful for differentiating right-sided
murmurs from left-sided murmurs. During the active strain
phase, with decreased venous return, most murmurs decrease in
intensity. There are 2 important exceptions to this rule:
1. The murmur of hypertrophic obstructive cardiomyopathy typically
gets louder.
2. The murmur of mitral valve prolapse may get longer (and possibly
louder).
After the release of the Valsalva maneuver, with a sudden
increase in venous return, right-sided murmurs return immedi-
ately (within 1 or 2 cardiac cycles), whereas left-sided murmurs
gradually return after several cardiac cycles. Thus, it is possi-
ble to differentiate between aortic and pulmonary stenosis and
between aortic and pulmonary regurgitation.
Respiration
The effect of normal respiration is also useful for distinguishing
between right-sided murmurs and left-sided murmurs. In gen-
eral, right-sided murmurs are augmented with inspiration (fre-
quent exceptions occur with tricuspid regurgitation). In cases of
severe RV failure, the RV may be unable to augment its output
with inspiration, and pulmonary or tricuspid murmurs may fail
to become louder with inspiration.
R-R Cycle Length
Varying R-R cycle length (such as in atrial fibrillation or with
frequent premature ventricular contractions) affects murmurs in
specific ways that can be of diagnostic value at the bedside. In
general, systolic ejection murmurs (such as with aortic or pul-
monary stenosis) increase after a long cycle length, whereas
regurgitant murmurs (such as with mitral or tricuspid regurgita-
tion) do not. The systolic murmur of hypertrophic obstructive
cardiomyopathy is augmented with the increased contractility
of a post–premature ventricular contraction beat, but the periph-
eral arterial pulse volume decreases because LV outflow tract
obstruction worsens.
Handgrip
Isometric exercise (eg, handgrip), by increasing systemic blood
pressure (afterload), augments the murmurs of mitral regurgita-
tion, AR, and VSD but does not significantly alter the murmur of
aortic stenosis and tends to decrease the murmur of hypertrophic
obstructive cardiomyopathy.
Squatting
Prompt squatting causes a rapid transient increase in venous
return and a sustained increase in peripheral resistance. The lat-
ter may augment the murmurs of mitral regurgitation and AR.
Because LV volume and peripheral resistance increase, the mur-
mur of hypertrophic obstructive cardiomyopathy becomes softer.
Then, after the upright position is assumed, with decreased LV
volume and peripheral resistance, the murmur of hypertrophic
obstructive cardiomyopathy becomes louder.
Amyl Nitrite
Administration of amyl nitrite is simple, inexpensive, and, in
most patients, safe (exceptions are in acute myocardial infarc-
tion or critical carotid artery stenosis, in which even tran-
sient hypotension should be avoided if possible). Amyl nitrite
causes acute systemic vasodilation, resulting in a transient
 1 Cardiovascular Examination 19

(30–45 seconds) decrease in systemic blood pressure, followed Miscellaneous

by reflex tachycardia and an increase in venous return and car-
The mammary souffle can be continuous and can mimic PDA.
diac output. All stenotic murmurs, including the murmur of
It can be obliterated by pressure with the examining finger next
hypertrophic obstructive cardiomyopathy, become louder. The
to the stethoscope. Innocent venous hums are loudest in the neck
murmur of mitral regurgitation usually decreases because of
but can be transmitted to the precordium and be mistaken for
the decrease in LV afterload (during the vasodilation phase).
PDA or atrioventricular fistula. The venous hum is loudest in the
The diastolic murmur of AR diminishes, whereas the murmur
sitting or standing position. Motion of the neck or jugular vein
of mitral stenosis becomes louder because of the increased flow
compression affects the intensity of the murmur.
across the mitral valve, especially during the tachycardia phase.
The systolic murmur of mitral prolapse may become longer • The innocent venous hum is loudest in the neck but can be trans-
(as LV volume decreases initially) but not necessarily louder, mitted to the precordium and be mistaken for PDA or atrioven-
because LV pressures also are decreased. The main useful- tricular fistula.
ness of amyl nitrite is to differentiate within the following pairs • The venous hum is of variable quality, is loudest in the sitting or
(Tables 1.4 and 1.5): standing position, and decreases in the supine position.
1. A small VSD (decrease in murmur) and pulmonary stenosis (increase
in murmur) Abbreviations
2. Aortic stenosis (increase) and mitral regurgitation (decrease) A2 aortic valve component of the second heart sound
3. AR (decrease) and mitral stenosis (increase) AR aortic regurgitation
4. AR (decrease) and pulmonary regurgitation (increase) ASD atrial septal defect
5. Mitral regurgitation (decrease) and tricuspid regurgitation (increase) LA left atrium
• After release of the Valsalva maneuver, with a sudden increase in LV left ventricle
venous return, right-sided murmurs return immediately (within 1 M1 mitral valve component of the first heart sound
or 2 cardiac cycles), whereas left-sided murmurs gradually return OS opening snap
after several cardiac cycles. P2 pulmonic valve component of the second heart sound
PDA patent ductus arteriosus
• Systolic ejection murmurs (such as with aortic or pulmonary
RV right ventricle
stenosis) increase after a long cycle length, whereas regurgi-
S1 first heart sound
tant murmurs (such as with mitral or tricuspid regurgitation)
S2 second heart sound
do not.
S3 third heart sound
• Amyl nitrite causes all stenotic murmurs, including the mur- S4 fourth heart sound
mur of hypertrophic obstructive cardiomyopathy, to become T1 tricuspid valve component of the first heart sound
louder. VSD ventricular septal defect
 2
Applied Anatomy of the Heart
and Great Vesselsa
JOSEPH G. MURPHY, MD, and R. SCOTT WRIGHT, MD
Mediastinum
The mediastinum contains the heart, great vessels, distal portion
of the trachea, right and left bronchi, esophagus, thymus, auto-
nomic nerves (cardiac and splanchnic, left recurrent laryngeal,
and bilateral vagal and phrenic), various small arteries (such as
bronchial and esophageal) and veins (such as bronchial, azygos,
and hemiazygos), lymph nodes, cardiopulmonary lymphatics,
and thoracic duct.
Enlargement of a cardiac chamber or great vessel may dis-
place or compress an adjacent noncardiac structure. An enlarged
left atrium may displace the left bronchus superiorly and the
esophagus rightward. An aberrant retroesophageal right subcla-
vian artery indents the esophagus posteriorly and may cause dys-
phagia. Mediastinal neoplasms can compress the atria, superior
vena cava, or pulmonary veins.
Pericardium
The pericardium surrounds the heart and consists of fibrous and
serous portions. The fibrous pericardium forms a tough outer
sac, which envelops the heart and attaches to the great vessels.
The ascending aorta, pulmonary artery, terminal 2 to 4 cm of
superior vena cava, and short lengths of the pulmonary veins and
inferior vena cava are intrapericardial.
The fibrous pericardium is inelastic and limits the diastolic
distention of the heart during exercise. Cardiac enlargement or
a mL of serous fluid that allows the tissue surfaces to glide over
Adapted from Edwards WD. Applied anatomy of the heart. In: Giuliani
ER, Gersh BJ, McGoon MD, Hayes DL, Schaff HV, editors. Mayo Clinic
practice of cardiology. 3rd ed. St Louis (MO): Mosby; 1996. p. 422–89.
Used with permission of Mayo Foundation for Medical Education and
Research.
Abbreviations and acronyms are expanded at the end of this chapter.
20
chronic pericardial effusions that develop slowly will stretch
the fibrous pericardium. The fibrous pericardium cannot stretch
acutely, and the rapid accumulation of as little as 200 mL of fluid
may produce fatal cardiac tamponade. Hemopericardium results
from perforation of either the heart or the intrapericardial great
vessels.
The serous pericardium is a delicate mesothelial layer that
lines the inner aspect of the fibrous pericardium (parietal peri-
cardium) and the outer surface of the heart and intrapericardial
great vessels (visceral pericardium). The visceral pericardium,
or epicardium, contains the coronary arteries and veins, auto-
nomic nerves, lymphatic channels, and variable amounts of adi-
pose tissue.
In obese persons, excessive epicardial fat may encase the
heart, but because pericardial fat is liquid at body temperature,
cardiac motion is generally unhindered.
Between the great arteries (aorta and pulmonary artery) and
the atria is a tunnel-like transverse sinus. Posteriorly, the peri-
cardial reflection forms an inverted U-shaped cul-de-sac known
as the oblique sinus.
A sequential saphenous vein bypass graft to the left coronary
system may be positioned posteriorly through the transverse
sinus. A persistent left superior vena cava occupies the expected
site of the ligament of Marshall, along the junction between the
appendage and body of the left atrium.
Between the parietal and visceral layers of the serous pericar-
dium is the pericardial cavity, which normally contains 10 to 20
each other with minimal friction.
Thick and roughened surfaces associated with fibrinous peri-
carditis lead to an auscultatory friction rub, and organization
of such an exudate may result in fibrous adhesions between the
epicardium and the parietal pericardium. Focal adhesions are
 2 Applied Anatomy of the Heart and Great Vessels 21

usually unimportant, but occasionally they may allow the accu-

R int jugular
mulation of loculated fluid or, rarely, tamponade of an individual L int jugular
cardiac chamber, usually the right ventricle. After cardiac sur- R ext jugular
L ext jugular
gery, the opened pericardial cavity may become sealed again if L subclavian
the parietal pericardium adheres to the sternum; the raw pericar- R subclavian
dial surfaces, which are lined by fibrovascular granulation tissue, L innominate
R innominate
may ooze enough blood to cause cardiac tamponade. L int thoracic
R int thoracic
Densely fibrotic adhesions, with or without calcification, can
hinder cardiac motion and may restrict cardiac filling. The peri- L sup intercostal
Sup vena cava
cardium is thickened in persons with chronic constriction but
may be of normal thickness if constriction develops relatively
Accessory hemiazygos
rapidly. In constrictive pericarditis, surgical excision of only the
anterior pericardium (between the phrenic nerves) is often inad-
equate, because the remaining pericardium surrounds enough of Azygos
the heart to maintain constriction. Hemiazygos
Most postoperative pericardial adhesions are usually func- Hepatic
tionally unimportant, but they may obscure the location of the
coronary arteries at subsequent cardiac operation.
Inf vena cava
• The fibrous pericardium cannot stretch acutely, and the rapid accu- R adrenal L adrenal vein
mulation of as little as 200 mL of fluid may produce fatal cardiac
tamponade. L renal
R renal
• A sequential saphenous vein bypass graft to the left circumflex
coronary artery may be positioned posteriorly through the trans-
L gonadal
verse sinus.
R gonadal

Great Veins L common iliac

Bilaterally, the subclavian and internal jugular veins merge to form
bilateral innominate (or brachiocephalic) veins, which join to form
the superior vena cava (or superior caval vein) (Figure 2.1).
Superior Vena Cava
The right internal jugular vein, right innominate vein, and supe-
rior vena cava afford a relatively straight intravascular route to
the right atrium and tricuspid orifice. This route may be used
for passage of a stiff endomyocardial bioptome across the tri-
cuspid valve and into the right ventricular apex to obtain a car-
diac biopsy specimen. Similarly, both temporary and permanent
transvenous pacemaker leads are inserted into either the subcla-
vian or the internal jugular vein and are threaded into the right
ventricular apex.
Catheters and pacemakers within the innominate veins and
superior vena cava become partially coated with thrombus and
may be associated with thrombotic venous obstruction, pulmo-
nary thromboembolism, or secondary infection. Mediastinal
neoplasms, fibrosis, and aortic aneurysms may compress the thin-
walled veins and result in the superior vena cava syndrome.
Inferior Vena Cava
The inferior vena cava receives systemic venous drainage from
the legs and retroperitoneal viscera and, at the level of the liver,
from the intra-abdominal systemic venous drainage (portal cir-
culation) via the hepatic veins. If the inferior vena cava, which
is retroperitoneal, becomes trapped and compressed between the
vertebral column posteriorly and either an adjacent retroperito-
neal structure (eg, an abdominal aortic aneurysm) or an intra-
peritoneal structure (eg, a neoplasm), the result is inferior vena
cava syndrome.
Venous thrombi in the lower extremities may extend into the
inferior vena cava or may become dislodged and embolize to the
right heart and pulmonary circulation. Renal cell carcinomas
Int iliac
Ext iliac
Figure 2.1. Systemic Veins, Excluding the Portal Circulation. (See
end of chapter for abbreviations used in this figure.)
may extend intravascularly within the renal veins and inferior
vena cava and may even form tethered intracavitary right-sided
cardiac masses. Hepatocellular carcinomas often involve the
hepatic veins and occasionally may enter the suprahepatic infe-
rior vena cava or right atrium.
The superior and inferior pulmonary veins from each lung
enter the left atrium. The proximal 1 to 3 cm of the pulmonary
veins contain cardiac muscle within the media and may thereby
function like sphincters during atrial systole as well as when sig-
nificant mitral valve disease exists.
The thin-walled, low-pressure pulmonary veins may be
compressed extrinsically by mediastinal neoplasms or fibrosis.
Pulmonary vein stenosis is a rare complication of electrophysi-
ologic radiofrequency ablation.
Congenital Anomalies of the Venous System
Congenital anomalies of the systemic veins include a persis-
tent left superior vena cava (with or without a left innominate
vein) joining the coronary sinus or, rarely, the left atrium; an
unroofed or absent coronary sinus; a large right sinus venosus
valve (so-called cor triatriatum dexter); azygos continuity of the
inferior vena cava; and bilateral subrenal inferior venae cavae.
Anomalous Venous Connection
In total anomalous pulmonary venous connection, the conflu-
ence of pulmonary veins does not join the left atrium but rather
22 I Fundamentals of Cardiovascular Disease
maintains connection to derivatives of the cardinal or umbilico-
vitelline veins, such as the left innominate vein, coronary sinus,
or ductus venosus. An interatrial communication must also be
present.
In partial anomalous pulmonary venous connection, only
some veins (usually from the right lung) lack left atrial connec-
tions. Connection of the right pulmonary veins to the right atrium
commonly accompanies sinus venosus atrial septal defects,
whereas connection of these veins to the suprahepatic inferior
vena cava is usually part of the scimitar syndrome.
Cor Triatriatum
Cor triatriatum sinistrum results when the junction between the
common pulmonary vein and the left atrium is stenotic. A fenes-
trated membranous or muscular shelf subdivides the left atrium
into a posterosuperior chamber, which receives the pulmonary
veins, and an anteroinferior chamber, which contains the atrial
appendage and mitral orifice.
• Mediastinal neoplasms, fibrosis, and aortic aneurysms may com-
press the thin-walled veins and result in the superior vena cava
syndrome.
• If the inferior vena cava becomes trapped and compressed between
the vertebral column posteriorly and either an adjacent retroperito-
neal structure or an intraperitoneal structure, the result is inferior
vena cava syndrome.
• The thin-walled, low-pressure pulmonary veins may be com-
pressed extrinsically by mediastinal neoplasms or fibrosis.
• Connection of 1 (usually the upper) or both right pulmonary veins
to the right atrium commonly accompanies sinus venosus atrial
septal defects.
Cardiac Chambers
Atria
Right Atrium
The right atrium and the superior vena cava form the right lateral
border of the frontal chest radiographic cardiac silhouette. The
right atrium receives the systemic venous return from the superior
and inferior venae cavae and receives most of the coronary venous
return via the coronary sinus and numerous small thebesian veins.
The ostium of the inferior vena cava is bordered anteriorly by a
crescentic eustachian valve, which may be large and fenestrated
and form a Chiari network. The coronary sinus ostium is partly
shielded by a fenestrated thebesian valve. The right atrium free
wall has a smooth-walled posterior portion, which receives the
caval and coronary sinus blood flow, and a muscular anterolateral
portion, which contains ridge-like pectinate muscles and a large
pyramid-shaped appendage. Separating the 2 regions is a promi-
nent C-shaped muscle bundle, the crista terminalis (or terminal
crest). The right atrial appendage abuts the right aortic sinus and
overlies the proximal right coronary artery.
The thickness of the right atrial free wall varies considerably.
The atrial wall between the pectinate muscles is paper-thin and
can be perforated by a stiff catheter.
With atrial enlargement and blood flow stasis, mural thrombi
may form within the recesses between the pectinate muscles,
particularly in the atrial appendage. Indwelling cardiac catheters
or pacemaker wires tend to injure the endocardium at the cavo-
atrial junction and are often associated with shallow linear mural
thrombi. An atrial pacing lead can be inserted into the muscle
bundles within the appendage.
Atrial Septum
The atrial septum has interatrial and atrioventricular compo-
nents (Figure 2.2). The interatrial portion contains the fossa
ovalis (or oval fossa), which includes an arch-shaped outer
muscular rim (the limbus or limb) and a central fibrous mem-
brane (the valve). The foramen ovale (or oval foramen, which
is patent throughout fetal life) represents a potential interatrial
shunt, which courses between the anterosuperior limbic rim and
the valve of the fossa ovalis and then through the natural val-
vular perforation (ostium secundum or second ostium) into the
left atrium. In approximately two-thirds of patients, the foramen
ovale closes anatomically during the first year of life as the valve
of the fossa ovalis becomes permanently sealed to the limbus. In
the remaining third, this flap-valve closes functionally only when
left atrial pressure exceeds right atrial pressure; this constitutes
a so-called valvular-competent (or probe-patent) patent foramen
ovale (Figure 2.3). If right atrial pressure exceeds left atrial pres-
sure, as in right heart failure or during the Valsalva maneuver,
right-to-left shunting will occur.
Through a patent foramen ovale, systemic venous emboli may
enter the systemic arterial circulation. Such paradoxical emboli
may be thrombotic (eg, from the legs) or nonthrombotic (eg, air
emboli).
Pronounced atrial dilatation may so stretch the atrial septum
that the limbus no longer covers the ostium secundum in the
valve of the fossa ovalis. As a result, interatrial shunting may
occur across the valvular-incompetent patent foramen ovale. In
some patients, aneurysms of the valve of the fossa ovalis may
develop and may undulate during the cardiac cycle. Atrial dilata-
tion stimulates the release of natriuretic peptide.
The atrioventricular component of the atrial septum, which
separates the right atrium from the left ventricle, is primarily
muscular but also has a small fibrous component (the atrioven-
tricular portion of the membranous septum) (Figure 2.2).
The atrioventricular septum corresponds to the triangle
of Koch, an important anatomical landmark that contains the
R
L
P
V
LLPV
IAS
RA LA
AVS
AVS
IVS LV
IVS
Figure 2.2. Atrial Anatomy with Tricuspid and Mitral Valves in
Profile. The atrioventricular septum is anterior to the interatrial sep-
tum and posterior to the interventricular septum; note also the infolded
nature of the limbus (white arrows) and the relative thinness of the valve
of the fossa ovalis (red arrow). The origin of the normal tricuspid valve
is below that of the mitral valve, allowing the possibility of a right atrial–
to–left ventricular shunt. This 4-chamber view is from a 15-year-old
boy. (See end of chapter for abbreviations used in this figure.)
 2 Applied Anatomy of the Heart and Great Vessels
Figure 2.3. Thin Wall of Foramen Ovale (Transilluminated).
atrioventricular node (Figure 2.4); it is bound by the septal tricus-
pid annulus, the coronary sinus ostium, and the tendon of Todaro.
The tendon of Todaro is a subendocardial fibrous cord that
extends from the eustachian-thebesian valvular commissure to the
anteroseptal tricuspid commissure (at the membranous septum);
it corresponds approximately to the level of the mitral annulus.
The thickness of the atrial septum varies considerably. The
valve of the fossa ovalis is a paper-thin translucent membrane at
birth but becomes more fibrotic with time and may become 1 to 2
mm thick. The limbus of the fossa ovalis ranges from 4 to 8 mm in
thickness; however, lipomatous hypertrophy may produce a bulg-
ing mass more than 3 times this thickness. The muscular atrioven-
tricular septum forms the summit of the ventricular septum and
may be 5 to 10 mm thick; it may be much thicker in hypertrophic
cardiomyopathy or concentric left ventricular hypertrophy. The
membranous septum generally is less than 1 mm thick.
Left Atrium
The left atrium, a posterior midline chamber, receives pulmo-
nary venous blood and expels it across the mitral orifice and into
Figure 2.4. Position of Atrioventricular Node (Triangle of Koch)
(Arrow).
23
the left ventricle. The esophagus and descending thoracic aorta
abut the left atrial wall. Thus, the left atrium, atrial septum, and
mitral valve are particularly well visualized with transesopha-
geal echocardiography. The body of the left atrium does not
contribute to the frontal cardiac silhouette; however, the left
atrial appendage, when enlarged, may form the portion of the
left cardiac border between the left ventricle and the pulmonary
trunk. Normally the appendage, shaped like a wind sock, abuts
the pulmonary artery and overlies the bifurcation of the left main
coronary artery.
With chronic obstruction to left atrial emptying (eg, rheu-
matic mitral stenosis), the dilated left atrium may shift the atrial
septum rightward and in severe cases may actually form the right
cardiac border on chest radiographs. Moreover, the esophagus
can be shifted rightward, and the left bronchus may be elevated.
Mural thrombi often develop within the atrial appendage or, less
commonly, the atrial body, and in severe cases can virtually fill
the chamber except for small channels leading from the pulmo-
nary veins to the mitral orifice. In contrast to left atrial mural
thrombi, which tend to involve the free wall, most myxomas
arise from the left side of the atrial septum.
Comparison of Atria
The right atrial free wall contains a crista terminalis and pecti-
nate muscles, whereas the left atrial free wall has neither. The
right atrial appendage is large and pyramidal, in contrast to the
wind sock–like left atrial appendage. Finally, the atrial septum
is characterized by the fossa ovalis on the right side and by the
ostium secundum on the left (Figure 2.5).
Owing to hemodynamic streaming within the right atrium
during intrauterine life, superior vena caval blood is directed
toward the tricuspid orifice, and inferior vena caval blood, car-
rying well-oxygenated placental blood, is directed by the eusta-
chian valve toward the foramen ovale. As a result, the most
highly oxygenated blood in the fetal circulation is directed,
via the left heart, to the coronary arteries, the upper extremi-
ties, and the brain. Even postnatally, the superior vena cava
maintains its orientation toward the tricuspid annulus, and the
24 I Fundamentals of Cardiovascular Disease

Figure 2.5. Normal Atria.

inferior vena cava maintains its orientation toward the atrial
septum (Figure 2.6).
Consequently, an endomyocardial biopsy specimen of the
right ventricular apex is much more easily obtained through a
superior vena caval approach than through an inferior vena caval
approach. In contrast, passing a catheter from the right atrium
into the left atrium through the foramen ovale is much easier than
using an inferior vena caval approach. In patients in whom the
foramen ovale is anatomically sealed, the valve of the fossa ova-
lis may be intentionally perforated (transseptal approach); how-
ever, this membrane becomes thicker and more fibrotic with age.
Atrial Septal Defect
A secundum atrial septal defect involves the fossa ovalis region
of the interatrial septum. It is the most common form of atrial
septal defect and often is an isolated anomaly.
A primum atrial septal defect involves the atrioventricular
septum and represents a malformation of the endocardial cush-
ions; it is almost invariably associated with mitral and tricuspid
abnormalities, particularly a cleft in the anterior mitral leaflet.
A sinus venosus atrial septal defect involves the posterior
aspect of the atrial septum and is usually associated with an
anomalous right atrial connection of the right pulmonary veins.
A coronary sinus atrial septal defect is usually associated with
an absent (unroofed) coronary sinus and connection of the left
superior vena cava to the left atrium.
• Most myxomas arise from the left side of the atrial septum.
• A secundum atrial septal defect involves the fossa ovalis region of
the interatrial septum.
• A coronary sinus atrial septal defect is usually associated with an
absent coronary sinus and connection of the left superior vena cava
to the left atrium.

Figure 2.6. Right Atrial Hemodynamic Streaming. Superior vena caval blood is directed toward the tricuspid orifice, and inferior vena caval blood
is directed toward the fossa ovalis. This opened right atrium is from a 31-year-old man. (Previously published. See “Credit Lines” section.)
 2 Applied Anatomy of the Heart and Great Vessels
Ventricles
Right Ventricle
The right ventricle does not contribute to the borders of the fron-
tal cardiac silhouette radiographically. It is crescent-shaped in
the short-axis view and triangular-shaped in the long-axis view
(Figure 2.7).
Conditions that impose a pressure overload on the right ven-
tricle (eg, pulmonary hypertension) cause straightening of the
ventricular septum such that both ventricles are D shaped on
cross-section. In extreme cases, such as Ebstein anomaly or total
anomalous pulmonary venous connection, leftward bowing of
the ventricular septum may result not only in a circular right ven-
tricle and crescentic left ventricle but also in possible obstruction
of the left ventricular outflow tract.
The right ventricular chamber consists of 3 regions—inlet,
trabecular, and outlet. The inlet region receives the tricuspid
valve and its cordal and papillary muscle attachments. A complex
meshwork of muscle bundles characterizes the anteroapical tra-
becular region. In contrast, the outlet region is smoother-walled
and is also known as the infundibulum, conus, or right ventricu-
lar outflow tract. Along the outflow tract, an arch of muscle sepa-
rates the tricuspid and pulmonary valves. The arch consists of a
parietal band, outlet septum, and septal band, known collectively
as the crista supraventricularis (ie, the supraventricular crest).
During right ventricular endomyocardial biopsy, the bioptome
is directed septally, not only to avoid injury to the cardiac conduc-
tion system and tricuspid apparatus but also to prevent possible
perforation of the relatively thin free wall. Tissue is more often
procured from the meshwork of apical trabeculations than from
the septal surface per se. When permanent transvenous pacemaker
electrodes are inserted into the right ventricle, the apical trabecu-
lations trap the tined tip and thereby prevent dislodgment.
With trauma or vigorous cardiopulmonary resuscitation in
which ribs are fractured, the jagged bones may be forced through
the parietal pericardium anteriorly and may lacerate an epicar-
dial coronary artery or may perforate the right atrial or right
or left ventricular free wall. Furthermore, if the force of closed
chest cardiac massage during cardiopulmonary resuscitation is
at the midsternum rather than the xiphoid area, the right ven-
tricular outflow tract may be compressed; the resultant high right
ventricular pressure may cause apical rupture.
Figure 2.7. Right Ventricle, Showing Marked Trabeculation and Tricuspid Valve.
25
Left Ventricle
The left ventricle forms the left border of the frontal cardiac sil-
houette radiographically. It is circular in short-axis views and is
approximated in 3 dimensions by a truncated ellipsoid.
Conditions such as aortic stenosis and chronic hypertension,
which impose a pressure overload on the left ventricle, induce
concentric left ventricular hypertrophy without appreciable
dilatation. Although the short-axis chamber diameter does not
increase significantly, the wall thickness generally increases
25% to 75%, and the heart weight may double or triple.
Disorders that impose a volume overload on the left ven-
tricle, such as chronic aortic or mitral regurgitation or dilated
cardiomyopathy, are attended not only by hypertrophy but also
by chamber dilatation. They thereby produce a globoid heart
with increased base-apex and short-axis dimensions. Although
the heart weight may double or triple, the left ventricular wall
thickness generally remains within the normal range because
of the thinning effect of dilatation. Accordingly, when the left
ventricle is dilated, wall thickness cannot be used as a reli-
able indicator of hypertrophy. The term volume hypertrophy is
favored in this situation. Hypertrophy, with or without cham-
ber dilatation, decreases myocardial compliance and impairs
diastolic filling.
Like the right ventricle, the left ventricle can be divided into
inlet, apical, and outlet regions. The inlet receives the mitral
valve apparatus, the apex contains fine trabeculations, and the
outlet is angled away from the remainder of the chamber. Inflow
and outflow tracts are separated by the anterior mitral leaflet,
which forms an intracavitary curtain between the 2.
The anterior mitral leaflet is also in direct contact, at its annu-
lus, with the left and posterior aortic valve cusps. For compari-
son, the membranous septum (Figure 2.8) abuts the right and
posterior aortic cusps, and the outlet septum lies beneath the
right and left aortic cusps.
For practical purposes, the base-apex length of the left ven-
tricle is divided into thirds—basal (corresponding to the mitral
leaflets and tendinous cords), midventricular (corresponding to
the mitral papillary muscles), and apical levels. Each level is sub-
divided into segments, thus forming the basis for regional analy-
sis of the left ventricle (eg, the evaluation of regional wall motion
abnormalities) (Figure 2.9 and Table 2.1).
26 I Fundamentals of Cardiovascular Disease
Figure 2.8. Left Ventricle. Thin-walled membranous septum is transilluminated.
Hypertrophic cardiomyopathy is characterized by asym-
metric (nonconcentric) left ventricular hypertrophy that dispro-
portionately involves the septum. Cardiac amyloid may mimic
hypertrophic cardiomyopathy.
In the normal heart of an elderly person, left ventricular
geometry is altered (the septum is more sigmoid in shape), and
mild fibrosis and calcification of the aortic and mitral valves may
contribute to the low-grade systolic ejection murmurs that are
so common in the elderly. With advancing age, the aortic annu-
lus dilates appreciably and tilts rightward and less posteriorly,
thereby altering the shape and direction of the left ventricular out-
flow tract, which may simulate hypertrophic cardiomyopathy.
Because left ventricular trabeculae carneae are small, per-
manent apical entrapment of a tined transvenous pacemaker
electrode is difficult and may necessitate placing epicardial elec-
trodes (eg, in patients with corrected transposition of the great
arteries or with complete transposition of the great arteries and a
previous Mustard or Senning operation).
During left ventricular endomyocardial biopsy, care must be
taken not to injure the mitral apparatus or left bundle branch and
not to perforate the apex (Figure 2.10).
In some persons, apical or anteroseptal trabeculae carneae
may form a prominent spongy meshwork that may be misinter-
preted as apical mural thrombus on imaging studies.
AS A
RV RV
LVOT
PS AL
MV
P PL
Basal
Figure 2.9. Regional Analysis of the Left Ventricle. Short-axis views show the recommended 16-segment system. (See end of chapter for abbrevia-
tions used in this figure.)
Comparison of Ventricles
Normally, left ventricular wall thickness is 3 to 4 times that of
the right ventricle. In short-axis views, the left ventricle is circu-
lar and the right ventricle is crescentic. Whereas the tricuspid and
pulmonary valves are separated from one another, the mitral and
aortic valves are in direct continuity. The right ventricular apex
is much more heavily trabeculated than the left.
With 2-dimensional echocardiography, ventricular morphol-
ogy is best inferred from the morphology of the atrioventricular
valves, particularly by differences in the annular levels at the
cardiac crux.
Congenital Heart Disease
Ventricular Septal Defect. The most common ventricular
septal defect, either isolated or associated with other cardiac
anomalies, is the membranous (perimembranous) type, which
involves the membranous septum. An infundibular (also called
outlet, supracristal, or subarterial) ventricular septal defect is
commonly encountered in tetralogy of Fallot and truncus arte-
riosus. A malalignment ventricular septal defect occurs when 1
of the great arteries overrides the septum and attains biventricu-
lar origin, or both great arteries arise from 1 ventricle. Muscular
defects involve the muscular septum and can be solitary or
AS A
A
RV
IS LV AL S LV
L
I
I IL
Midventricular Apical
 2 Applied Anatomy of the Heart and Great Vessels
Table 2.1. Percentage of Regional Left Ventricular (LV)
Mass
% LV Volume
Level per Segment No. of Segments Total, %
Basal 7.2 6 43
Midventricular 6.0 6 36
Apical 5.3 4 21
multiple (so-called Swiss cheese septum). A defect of the atrio-
ventricular septum is considered to be an atrioventricular canal
defect, and straddling of an atrioventricular valve most com-
monly occurs across a defect of this type.
Tetralogy of Fallot. Within the spectrum of cyanotic con-
genital heart disease is a group of anomalies characterized by a
maldevelopment of the conotruncal septum. Tetralogy of Fallot,
the most common anomaly in this group, results from displace-
ment of the infundibular septum and is characterized by a large
malalignment ventricular septal defect, an overriding aorta,
and variable degrees of infundibular and valvular pulmonary
stenosis. When the pulmonary valve is atretic, pulmonary blood
flow may come from the ductus arteriosus or systemic collateral
arteries.
Transposition of the Great Arteries. Complete transposi-
tion of the great arteries is associated with abnormal conotruncal
septation and parallel rather than intertwined great arteries, such
that the aorta arises from the right ventricle and the pulmonary
artery emanates from the left ventricle; a ventricular septal defect
is present in about one-third of cases.
Truncus Arteriosus. Truncus arteriosus implies absent
conotruncal septation and is characterized by a single arterial
Figure 2.10. Left Ventricle (Free Wall [left] and Septum [right]). Mitral valve and papillary muscles are on the free wall.
27
trunk from which the aorta, pulmonary arteries, and coronary
arteries arise; the ventricular septal defect is of the membranous
or infundibular type.
Double-Outlet Right Ventricle. Double-outlet right ventricle
is characterized by the origin of both great arteries from the right
ventricle, a malalignment ventricular septal defect, and infundib-
ular septal displacement that differs from the type observed in
tetralogy of Fallot.
Myocyte Response to Injury
Myocardial cells are one-half contractile elements and one-third
mitochondria by volume. They are exquisitely sensitive to oxygen
deprivation, and ischemia represents the most common form of
myocardial injury. Other injurious agents include viruses, chemi-
cals, and excessive cardiac workload (volume or pressure).
The heart can respond to stress or injury in only limited ways.
Adaptive responses include hypertrophy and dilatation, whereas
sublethal cellular injury is characterized by various degenerative
changes. Necrosis is the histologic hallmark of lethal cellular
injury, and it elicits an inflammatory response with subsequent
healing by scar formation.
Hypertrophy of cardiac muscle cells is accompanied by
degenerative changes, an increase in interstitial collagen, and
a decrease in ventricular compliance. In dilated hearts, hyper-
trophied myocytes are also stretched, but with relatively normal
diameters. In dilated hearts, the best histologic indicators of
hypertrophy are nuclear alterations.
Acute myocardial ischemia is characterized by intense sarco-
plasmic staining with eosin dyes, prominent sarcoplasmic con-
traction bands, and, occasionally, stretched and wavy myocardial
cells. When ischemic cells are irreversibly injured, the changes
of coagulative necrosis appear. Nuclei fade away (karyolysis) or
28 I Fundamentals of Cardiovascular Disease

fragment (karyorrhexis), and the sarcoplasm develops a glassy

homogeneous appearance, although in many cases the cross-stri-
ations remain intact for several days. Necrotic myocardium elicits
an inflammatory infiltrate of neutrophils and macrophages, which CZ
is used histologically to differentiate acute infarction from acute
ischemia. Because myocardial cells have a very limited ability to
replicate, healing is by organization with scar formation.

Cardiac Valves
Atrioventricular Valves RZ
The right (tricuspid) and left (mitral) atrioventricular valves have
5 components: 3 form the valvular apparatus (annulus, leaflets,
and commissures) and 2 form the tensor apparatus (chordae A
tendineae and papillary muscles).

Valve Annulus
The annulus of each atrioventricular valve is saddle-shaped. As
part of the fibrous cardiac skeleton at the base of the heart, each
annulus electrically insulates the atrium from the ventricle. Since
the tricuspid annulus is an incomplete fibrous ring, loose connec- C
tive tissue maintains insulation at the points of fibrous disconti- Pap
ap M
Pap *
nuity. The mitral annulus, in contrast, constitutes a continuous
ring of fibrous tissue.

Valve Leaflets
The valve leaflets are delicate fibrous tissue flaps that close
the anatomical valvular orifice during ventricular systole
(Figure 2.11). The leading edge of each leaflet is its free edge, and
its serrated appearance results from direct cordal insertions into
this border. The closing edge, in contrast, represents a slightly
thickened nodular ridge several millimeters above the free edge.
When the valve closes, apposing leaflets contact one another
along their closing edges, and interdigitation of these nodular
ridges ensures a competent seal. Each leaflet comprises 2 major
layers: the fibrosa, which forms the strong structural backbone
of the valve, and the spongiosa, which acts as a shock absorber
along the atrial surface, particularly at the closing edge (rough
zone), where 1 leaflet coapts with an adjacent leaflet.
Chordae Tendineae
The chordae tendineae are strong, fibrous tendinous cords
that act as guidewires to anchor and support the leaflets. They
restrict excessive valvular excursion during ventricular systole
and thereby prevent valvular prolapse into the atria. Most ten-
dinous cords branch 1 or more times, so that generally more
than 100 cords insert into the free edge of each atrioventricular
valve. By virtue of these numerous cordal insertions, the force
of systolic ventricular blood is evenly distributed throughout the
undersurface of each leaflet.
Papillary Muscles
The papillary muscles, which may have multiple heads, are coni-
cal mounds of ventricular muscle that receive the majority of the
tendinous cords (Figure 2.12). Because of their position directly
beneath a commissure, each papillary muscle receives cords
from 2 adjacent leaflets. As a result, papillary muscle contrac-
tion tends to pull the 2 leaflets toward each other and thereby
facilitates valve closure.
B
Figure 2.11. Components of an Atrioventricular Valve. A, Each leaf-
let has a large clear zone (CZ) and a smaller rough zone (RZ) between
its free edge (arrow) and closing edge (dashed line). B, Each commis-
sure (C) separates 2 leaflets and overlies a papillary muscle (Pap M); a
fan-like commissural tendinous cord (*) connects the tip of the papil-
lary muscle to the commissure. Specimen is from the mitral valve of an
8-year-old girl.
In the elderly, mild mitral annular dilatation may be present,
with or without atrial dilatation. Leaflets become thicker, with
increasing nodularity of the rough zone and mild hooding defor-
mity of the entire leaflet. Contributing to the hooding deformity
is a decrease in ventricular base-apex length, which makes the
thickened cords appear relatively longer than necessary, thus
simulating mitral valve prolapse.
Tricuspid Valve
The plane of the tricuspid annulus faces toward the right ven-
tricular apex. Along the free wall, the annulus inserts into the
atrioventricular junction, whereas along the septum, it separates
the atrioventricular and interventricular portions of the septum.
In living persons, the area defined by the tricuspid annulus var-
ies with the cardiac cycle: it is maximal during ventricular diastole
(about 11 cm2) and decreases by about 30% during ventricular
systole. The reduction in area results from contraction of the
underlying basal right ventricular myocardium, since the incom-
plete tricuspid annulus cannot adequately constrict by itself.
The 3 tricuspid leaflets are not always well separated from
one another. The septal (medial) leaflet lies parallel to the ven-
tricular septum, and the posterior (inferior) leaflet lies parallel
to the diaphragmatic aspect of the right ventricular free wall. In
 2 Applied Anatomy of the Heart and Great Vessels 29

Figure 2.12. Mitral Valve Papillary Muscles (Short-Axis View).

contrast, the anterior (anterosuperior) tricuspid leaflet forms a
large sail-like intracavitary curtain that partially separates the
inflow tract from the outflow tract.
Because of differences in leaflet size and cordal length, the
excursion of the posterior and septal leaflets is less than that of
the anterior leaflet. In annular dilatation, leaflet excursion is inad-
equate to effect central coaptation, and valvular incompetence
results. Because the tricuspid annulus is incomplete, and because
the basal right ventricular myocardium forms a subjacent mus-
cular ring, dilatation of the right ventricle commonly produces
annular dilatation and tricuspid regurgitation. Right atrial dilata-
tion alone, as in constrictive pericarditis, usually does not cause
significant tricuspid insufficiency.
Valvular incompetence also may be observed in conditions
that limit leaflet and cordal excursion, such as rheumatic dis-
ease (fibrosis and scar retraction), carcinoid endocardial plaques
(thickening and retraction), and eosinophilic endomyocardial
diseases (thrombotic adherence to the underlying myocardium).
In normal hearts, mild tricuspid regurgitation is common.
Tricuspid stenosis involves commissural and cordal fusion
and may occur in rheumatic or carcinoid heart disease.
with the basal ventricular myocardium, dilatation of the ventricle
rarely increases annular circumference more than 25%.
Secondary left atrial dilatation may contribute to the progres-
sion of preexisting mitral incompetence by displacing the poste-
rior leaflet and its annulus and thereby hindering the excursion
of this taut leaflet.
Mitral Leaflets
The mitral leaflets form a continuous funnel-shaped veil with 2
prominent indentations, the anterolateral and posteromedial com-
missures. Although the 2 commissures do not extend entirely to
the annulus, they effectively separate the 2 leaflets. In contrast
to the other 3 cardiac valves, which each comprise 3 leaflets or
cusps, the mitral valve has only 2 leaflets. At midleaflet level, the
mitral orifice is elliptical or football-shaped, and its long axis
aligns with the 2 commissures and their papillary muscles.
Although the anterior leaflet occupies only about 35% of the
annular circumference, its leaflet area is almost identical to the area
of the posterior leaflet, about 5 cm2. The total mitral leaflet surface
area is 10 cm2, nearly twice that necessary to close the systolic

• Right atrial dilatation alone usually does not cause significant tri-
cuspid insufficiency.
• In normal hearts, mild tricuspid regurgitation is common.

Mitral Valve
Mitral Annulus
The plane of the mitral annulus faces toward the left ventricular
apex (Figure 2.13). The orifice changes shape during the cardiac
cycle, from elliptical during ventricular systole to more circular
during diastole. In living persons, the normal area defined by
the mitral annulus is maximal during ventricular diastole (about
7 cm2) and decreases 10% to 15% during systole.
Mitral annular calcification almost invariably involves only
the posterior mitral leaflet and forms a C-shaped ring of annu-
lar and subannular calcium which may impede basal ventricular
contraction and thereby produce mitral regurgitation. Similarly,
inadequate basal ventricular contraction may contribute to valvu-
lar incompetence in pronounced left ventricular dilatation; how-
ever, because only part of the mitral annulus is in direct contact Figure 2.13. Four Valves at Base of Heart.
30 I Fundamentals of Cardiovascular Disease
Figure 2.14. Mitral Valve Leaflets and Cords (Short-Axis View).
annular orifice, 5.2 cm2. However, some folding of leaflet tissue is
needed to ensure a competent seal, and the normal leaflets are not
as redundant as they might appear (Figures 2.14 and 2.15).
The myxomatous (or floppy) mitral valve is characterized
by annular dilatation, stretched tendinous cords, and redun-
dant hooded folds of leaflet tissue, which are prone to prolapse,
incomplete coaptation, cordal rupture, and mitral regurgitation.
In contrast, rheumatic mitral insufficiency results from scar
retraction of leaflets and cords. In infective endocarditis, viru-
lent organisms may perforate the leaflet tissue and produce acute
mitral regurgitation. In hypertrophic cardiomyopathy, the ante-
rior mitral leaflet contacts the ventricular septum during systole
and contributes both to left ventricular outflow tract obstruction
and to mitral incompetence.
In chronic aortic insufficiency, the regurgitant stream may
strike the anterior mitral leaflet and produce not only a fibrotic
jet lesion but also the leaflet flutter and premature valve closure
that are so characteristic echocardiographically.
Papillary Muscles
A fan-shaped cord emanates from the tip of each of the 2 papil-
lary muscles and inserts into its overlying commissure and into
Figure 2.15. Mitral Valve Leaflets and Annulus (Short-Axis View).
both adjacent leaflets (Figure 2.11B). Similarly, a smaller com-
missural cord inserts into each minor commissure between their
posterior scallops. Two particularly prominent cords insert along
each half of the ventricular surface of the anterior mitral leaflet,
and these so-called strut cords offer additional support for this
mid-cavitary leaflet that also forms part of the wall of the left
ventricular outflow tract. Cordal length is generally 1 to 2 cm.
Rheumatic mitral stenosis is characterized by cordal and com-
missural fusions, which obliterate the secondary intercordal ori-
fices and narrow the primary valve orifice (Figure 2.16). Cordal
rupture may occur in a myxomatous (floppy) valve, an infected
valve, or, rarely, an apparently normal valve and lead to acute
mitral regurgitation.
The mitral papillary muscles occupy the middle third of
the left ventricular base-apex length. Two prominent muscles
originate from the anterolateral and posteromedial (inferome-
dial) free wall, beneath their respective mitral commissures.
Trabeculations not only anchor the papillary muscles but also
may form a muscle bridge between the 2 papillary groups and
thereby contribute to valve closure.
The anterolateral muscle is a single structure with a midline
groove in 70% to 85% of cases, whereas the posteromedial mus-
cle is multiple or is bifid or trifid in 60% to 70%. The anterolateral
 2 Applied Anatomy of the Heart and Great Vessels
Figure 2.16. Valve Fibrosis in Rheumatic Mitral Stenosis.
muscle is generally larger and extends closer to the annulus than
the posteromedial muscle. Occasionally, an accessory papillary
muscle is interposed between the 2 major muscles along the free
wall. No papillary muscles or tendinous cords originate from the
septum and terminate on the mitral leaflets. However, in about
50% of persons, 1 or more cord-like structures, known as left
ventricular false tendons (or pseudotendons), arise from a papil-
lary muscle and insert either onto the septal surface or onto the
opposite papillary muscle.
Chronic postinfarction mitral regurgitation is associated with
papillary muscle atrophy and scarring, thinning and scarring of
the subjacent left ventricular free wall, and left ventricular dila-
tation. Acute postinfarction mitral regurgitation may be associ-
ated with rupture of a papillary muscle (almost invariably the
posteromedial) and can involve the entire muscle or only 1 of its
multiple heads.
Competent function of the mitral valve requires the harmo-
nious interaction of all valvular components, including the left
atrium and left ventricle.
• Secondary left atrial dilatation may contribute to the progression
of preexisting mitral incompetence.
• In hypertrophic cardiomyopathy, the anterior mitral leaflet may con-
tact the ventricular septum during systole and contribute both to left
ventricular outflow tract obstruction and to mitral incompetence.
• Chronic postinfarction mitral incompetence is associated with
papillary muscle atrophy and scarring.
Semilunar Valves
The right (pulmonary) and left (aortic) semilunar valves, in con-
trast to the atrioventricular valves, have no tensor apparatus and,
therefore, are structurally simpler valves (Figure 2.17). They
consist of an annulus, cusps, and commissures. Behind each cusp
is an outpouching of the arterial root, known as a sinus (ie, sinus
31
of Valsalva). There are 3 aortic sinuses and 3 pulmonary sinuses,
which impart a cloverleaf shape to the arterial roots.
The annuli of the semilunar valves are part of the fibrous
cardiac skeleton. They are nonplanar structures, shaped like a
triradiate crown.
The cusps are half-moon–shaped (semilunar), pocket-like
flaps of delicate fibrous tissue that close the valvular orifice
during ventricular diastole. The leading edge of each cusp is
its free edge. The closing edge, in contrast, represents a slightly
thickened ridge that lies a few millimeters below the free edge,
along the ventricular surface of the cusp. At the center of each
cusp, the closing edge meets the free edge and forms a small
fibrous mound, the nodule of Arantius. When the valve closes,
apposing cusps contact one another along the surfaces between
their free and closing edges (ie, the lunular areas), forming a
competent seal.
Like the atrioventricular valves, the semilunar valves con-
tain 2 major layers histologically. The fibrosa forms the struc-
tural backbone of the valve and is continuous with the annulus,
whereas the spongiosa acts more like a shock absorber along the
ventricular surface, especially at the closing edge. Cusps contain
little elastic tissue and, accordingly, have no appreciable elas-
tic recoil. The opening and closing of the semilunar valves is
a passive process that entails cusp excursion and annulocuspid
hinge-like motion.
In the elderly, degenerative changes in the aortic valve may
result in low-grade systolic ejection murmurs. The closing edges
become thickened and, along the nodules of Arantius, may form
whisker-like projections called Lambl excrescences. Lunular
fenestrations also tend to develop with increasing age.
Disease processes that can increase cusp rigidity (eg, fibro-
sis or calcification) or lead to commissural fusion (eg, rheumatic
valvulitis) tend to narrow the effective valvular orifice and pro-
duce stenosis. In contrast, processes that straighten the cuspid
32 I Fundamentals of Cardiovascular Disease
Figure 2.17. Aortic Valve and Pulmonary Valve (From Below).
line between commissures and thereby hold the commissures
open (eg, arterial root dilatation or rheumatic cuspid scar retrac-
tion) tend to produce regurgitation.
Pulmonary Valve
The plane of the pulmonary annulus faces toward the left mid-
scapula, with an area of about 3.5 cm2 (Figure 2.18). The cusps are
usually similar in size, although minor variations are common.
Pulmonary incompetence occurs in conditions that produce
dilatation of the pulmonary artery and annulus (eg, pulmonary
hypertension or heart failure). Combined pulmonary stenosis and
Figure 2.18. Pulmonary Valve.
incompetence are features of carcinoid heart disease, in which
the annulus becomes constricted and stenotic and in which the
cusps are also retracted and insufficient. Pure pulmonary steno-
sis is almost always congenital.
Aortic Valve
The plane of the aortic valve faces the right shoulder. In the liv-
ing person, the normal aortic annular area averages about 3 cm2
(Figure 2.19).
Unoperated symptomatic aortic stenosis has a worse progno-
sis than many malignancies. The vast majority of stenotic aortic
 2 Applied Anatomy of the Heart and Great Vessels
Figure 2.19. Normal Aortic Valve. Left, Closed. Right, Open.
valves are calcified. Most commonly, the valve has degenerative
(senile) calcification or is a calcified congenitally bicuspid valve
(Figure 2.20). Only rarely are heavily calcified valves the site of
active infective endocarditis.
Aortic root dilatation stretches the commissures open and
thereby produces aortic insufficiency in either a tricuspid or a
bicuspid aortic valve. Acute aortic regurgitation may be pro-
duced by infective aortic endocarditis with cuspid perforation or
by acute aortic dissection with commissural prolapse. Chronic
aortic regurgitation with coexistent aortic stenosis is most com-
monly associated with postrheumatic cuspid retraction, which
yields a fixed triangular orifice.
Among cases of infective endocarditis, perhaps none present
so varied a clinical spectrum as those associated with aortic annu-
lar abscesses. The possible clinical presentations depend greatly
on the particular cusp or cusps involved. Subvalvular extension
may involve the anterior mitral leaflet, left bundle branch, or ven-
tricular septal myocardium; involvement of the ventricular sep-
tal myocardium may produce a large abscess cavity or result in
rupture into a ventricular chamber with the formation of either
an aorto–right ventricular or an aorto–left ventricular fistula. An
aortic annular abscess may expand laterally and enter the peri-
cardial cavity and thereby produce purulent pericarditis or fatal
hemopericardium, or it may expand into adjacent cardiac cham-
bers or vessels and produce various fistulas (aorto–right atrial,
aorto–left atrial, or aortopulmonary).
Fibrous Cardiac Skeleton
At the base of the heart, the fibrous cardiac skeleton encircles
the 4 cardiac valves. It comprises not only the 4 valvular annuli
but also their intervalvular collagenous attachments (the right
and left fibrous trigones, the intervalvular fibrosa, and the conus
33
ligament) and the membranous septum and tendon of Todaro.
This fibrous scaffold is firmly anchored to the ventricles but is
rather loosely attached to the atria. Thus, the cardiac skeleton
not only electrically insulates the atria from the ventricles but
also supports the cardiac valves and provides a firm foundation
against which the ventricles may contract.
Because of the intervalvular attachments of the fibrous car-
diac skeleton, disease or surgery on 1 valve can affect the size,
shape, position, or relative angulation of its neighboring valves
and also can affect the adjacent coronary arteries or cardiac con-
duction system. Tricuspid annuloplasty or replacement may be
complicated by injury to the right coronary artery or atrioven-
tricular conduction tissues, whereas mitral valve replacement
may be attended by trauma to the circumflex coronary artery,
coronary sinus, or aortic valve. At aortic valve replacement, the
anterior mitral leaflet, left bundle branch, or coronary ostia may
be injured inadvertently.
Most congenital anomalies of the pulmonary valve are associ-
ated with stenosis. Isolated pulmonary stenosis is almost always
due to a dome-shaped acommissural valve, with congenital fusion
of all 3 commissures. However, forms of pulmonary stenosis
associated with other cardiac malformations, such as tetralogy
of Fallot, usually result from a bicuspid or unicommissural valve
(often with a hypoplastic annulus) or from a dysplastic valve with
3 thickened cusps.
Congenitally bicuspid aortic valves affect 1% to 2% of the
general population and constitute the most common form of con-
genital heart disease. Although they usually are neither stenotic
nor insufficient at birth, most bicuspid valves become stenotic
during adulthood as the cusps calcify, and some become insuffi-
cient as a result of infective endocarditis or aortic root dilatation.
In contrast, the congenitally unicommissural aortic valve is usu-
ally stenotic at birth and becomes progressively more obstructive
34 I Fundamentals of Cardiovascular Disease
Figure 2.20. Calcification of Aortic Valve in Degenerative Aortic Stenosis. Left, Gross specimen. Right, Matched radiograph of valve shows
calcium deposition.
as calcification develops in adulthood. Aortic atresia is associated
with the hypoplastic left heart syndrome and is usually fatal dur-
ing the first week of life. All congenital anomalies of the aortic
valve are much more common in males than in females.
In truncus arteriosus, the truncal valve most commonly com-
prises 3 cusps and resembles a normal aortic valve. However, it
may be quadricuspid, bicuspid, or, rarely, pentacuspid and may
contain 1 or more raphes; such nontricuspid valves are often
incompetent, particularly if the truncal root is dilated.
• Disease processes that can increase cusp rigidity tend to narrow
the effective valvular orifice and produce stenosis.
• Processes that straighten the cuspid line between commissures
tend to produce regurgitation.
• Pulmonary incompetence occurs in conditions that produce dila-
tation of the pulmonary trunk and annulus, such as pulmonary
hypertension or heart failure.
• Pure pulmonary stenosis is almost always congenital.
• An aortic annular abscess may expand laterally and enter the peri-
cardial cavity.
• Congenitally bicuspid aortic valves affect 1%–2% of the general
population.
Great Arteries
Pulmonary Arteries
The pulmonary artery arises anteriorly and to the left of the
ascending aorta and is directed toward the left shoulder. In
adults, it is slightly larger in diameter than the ascending aorta,
although its wall thickness is roughly half that of the aorta. At
the bifurcation, the right pulmonary artery travels horizontally
beneath the aortic arch and behind the superior vena cava, and
the left pulmonary artery courses over the left main bronchus
(Figure 2.21). The main and left pulmonary arteries contribute to
the left border of the frontal cardiac silhouette radiographically.
In pulmonary hypertension, especially in children with pli-
able tracheobronchial cartilage, the tense and dilated pulmonary
arteries can compress the left bronchus and the left upper and
right middle lobar bronchi and thereby contribute to recurrent
bronchopneumonia in those lobes. Furthermore, the dilated pul-
monary artery may displace the aortic arch rightward and second-
arily produce tracheal indentation and, occasionally, hoarseness
as a result of compression of the left recurrent laryngeal nerve.
Aorta
The aorta arises at the level of the aortic valve annulus and ter-
minates at the aortic bifurcation, approximately at the level of
the umbilicus and the fourth lumbar vertebra. The aorta has 4
major divisions: ascending aorta, aortic arch, descending tho-
racic aorta, and abdominal aorta (Figure 2.22).
Ascending Aorta
The ascending aorta lies almost entirely within the pericardial
sac and includes sinus and tubular portions, which are demar-
cated by the aortic sinotubular junction. The aortic valve leaflets
 2 Applied Anatomy of the Heart and Great Vessels 35

above the right cusp, and may produce coronary ostial steno-
sis. Among the causes of aortic root dilatation, perhaps aging,
mucoid medial degeneration (so-called cystic medial necrosis),
Trachea and chronic hypertension are the most common and may produce
an ascending aortic aneurysm, aortic valvular regurgitation, or
acute aortic dissection.
RUL LUL Aortic Arch
RPA LPA
The aortic arch travels over the right pulmonary artery and the
PT left bronchus. From its superior aspect emanate the innominate (or
brachiocephalic), left common carotid, and left subclavian arter-
Lingula ies, in that order. In 11% of persons, the innominate and left com-
RML mon carotid arteries form a common ostium, and in 5%, the left
vertebral artery arises directly from the aortic arch, between the
left common carotid and left subclavian arteries. The ligamentum
RLL arteriosum represents the obstructed fibrotic or fibrocalcific rem-
LLL
nant of the fetal ductus arteriosus (ductal artery), which joins the
Figure 2.21. Pulmonary Arteries and Bronchi. The right and left proximal left pulmonary artery to the undersurface of the aortic
pulmonary arteries do not exhibit mirror-image symmetry. (See end of arch. The aortic arch contributes to the left superior border of the
chapter for abbreviations used in this figure.) frontal cardiac silhouette and forms the radiographic aortic knob.
Aortic Dissection. When aortic dissections do not involve the
are related to the 3 sinuses, and the right and left coronary arter- ascending aorta (type III or type B), the intimal tear is commonly
ies arise from the right and left aortic sinuses, respectively. The near the ligamentum arteriosum or the ostium of the left subcla-
ascending aorta lies posterior and to the right of the pulmonary vian artery. By virtue of severe torsional and shear stresses placed
artery. on the heart and great vessels during nonpenetrating decelerative
With age or with the development of atherosclerosis, the aortic chest trauma (as can occur in motor vehicle accidents), the aorta
sinotubular junction can become heavily calcified, particularly may be transected at the junction between the aortic arch and the
descending thoracic aorta. When the tear is incomplete, a post-
traumatic pseudoaneurysm can develop with time. Aneurysms of
the aortic arch may be associated with hypertension, atheroscle-
R common carotid L common carotid
rosis, or aortitis, or they may be idiopathic.
Aortic arch

R subclavian L subclavian
Descending Thoracic Aorta
Innominate Ligamentum The descending thoracic aorta abuts the left anterior surface of
arteriosum the vertebral column and lies adjacent to the esophagus and the
left atrium. Its posterolateral branches are the bilateral intercostal
Ascending aorta

Tubular aorta
Descending thoracic aorta

Bronchial arteries, and its anterior branches include the bronchial, esopha-
Sinotubular jct geal, mediastinal, pericardial, and superior phrenic arteries. The
Aortic sinus Intercostal bronchial arteries, most commonly 2 left and 1 right, nourish
Coronary art the bronchial walls and the pulmonary arterial and venous walls.
Esophageal
Uncommonly, bronchial arteries may arise from intercostal or
subclavian arteries or, rarely, from a coronary artery. The bron-
Diaphragm chial veins drain not only into the azygos and hemiazygos veins
but also into the pulmonary veins.
Celiac L gastric If the bronchial circulation is adequate, pulmonary emboli
Hepatic
Splenic usually do not cause pulmonary infarction. In several forms of
Sup mes art
R adrenal
L adrenal pulmonary hypertension, the bronchial arteries become quite
L renal enlarged and tortuous.
Abdominal aorta

R renal
Aneurysms of the descending thoracic aorta may be associ-
R gonadal L gonadal
ated with aortic dissection, aortitis, atherosclerosis, hypertension,
or trauma. They may or may not extend below the diaphragm.
Middle sacral Inf mes art
L common iliac Abdominal Aorta

R ext iliac The abdominal aorta travels along the left anterior surface of
L ext iliac the vertebral column and lies adjacent to the inferior vena cava.
R int iliac L int iliac The major lateral (retroperitoneal) branches include the renal,
adrenal, right and left lumbar, and inferior phrenic arteries. The
Figure 2.22. Systemic Arteries. The aorta may be divided into the gonadal arteries arise somewhat more anteriorly but remain ret-
ascending aorta, the aortic arch, the descending thoracic aorta, and roperitoneal. The intraperitoneal branches arise anteriorly and
the abdominal aorta. (See end of chapter for abbreviations used in this include the celiac artery (with its left gastric, splenic, and hepatic
figure.) branches) and the superior and inferior mesenteric arteries. The
36 I Fundamentals of Cardiovascular Disease

distal aortic branches include the right and left common iliac
arteries and a small middle sacral artery.
Atherosclerotic abdominal aortic aneurysms are most com-
monly infrarenal. They tend to bulge anteriorly and thereby
stretch and compress the gonadal and inferior mesenteric arteries.
Such aneurysms are generally filled with laminated thrombus and
so their residual lumens often appear normal or even narrowed
rather than dilated. Rupture of an atherosclerotic abdominal aor-
tic aneurysm may be associated with extensive retroperitoneal
hemorrhage, with or without intraperitoneal hemorrhage.

Congenital Heart Disease

Aortopulmonary Window. An aortopulmonary septal
defect is a large opening between the ascending aorta and the
pulmonary trunk; hemodynamically, the defect resembles a pat-
ent ductus arteriosus. Rarely, 1 pulmonary artery originates from
the ascending aorta or ductus arteriosus, while the other arises
normally from the pulmonary trunk. Congenital stenosis of the
pulmonary arteries is usually associated with maternal rubella
during the first trimester. In pulmonary atresia with ventricular
septal defect, the pulmonary arteries may be derived from the
right or left ductus arteriosus and from bronchial or other sys-
temic collateral arteries (analogous to total anomalous pulmo-
nary venous connection).
Aortic Arch Congenital Abnormalities. Various anomalies
result from faulty development of the aortic arches. A right aortic
arch results from persistence of the right fourth aortic arch and
disappearance of its left counterpart; it most commonly accompa-
nies tetralogy of Fallot, pulmonary atresia with ventricular septal
defect, and truncus arteriosus. A double aortic arch results from
persistence of both fourth aortic arches. An aberrant retroesopha-
geal right subclavian artery is a relatively common anomaly, which
may cause dysphagia; it probably results from persistence of the
right dorsal aorta and resorption of the right fourth aortic arch.
Ductus Arteriosus. A patent ductus arteriosus may be iso-
lated or may accompany other cardiac malformations. A left
ductus arteriosus joins the proximal left pulmonary artery to the
aortic arch, whereas a right ductus arteriosus joins the proximal
right pulmonary artery to the right subclavian artery; in cases of
right aortic arch with mirror-image brachiocephalic branching,
the situation is opposite.
Coarctation of the Aorta. Coarctation of the aorta results
from an obstructive infolded ridge just distal to the left subclavian
artery and opposite the ductus arteriosus; it is associated with a
congenitally bicuspid aortic valve in at least half of the cases.
• Acute aortic dissection is commonly associated with an intimal
tear above the right aortic cusp and with eventual rupture into the
pericardial sac.
• When aortic dissections do not involve the ascending aorta (type
III or type B), the intimal tear is commonly near the ligamentum
arteriosum or the ostium of the left subclavian artery.
Coronary Circulation
Right Coronary Artery
The right coronary artery arises nearly perpendicularly from the
right aortic sinus (Figure 2.23). In 50% of persons, 1 or more
conus arteries also originate from the right aortic sinus, anterior
to the right coronary ostium. Rarely, the descending septal artery
or the sinus nodal artery originates directly from the aorta.
Figure 2.23. Coronary Ostia. The conus artery and right coronary
artery (black arrow) arise separately from the right cusp in this speci-
men. The red arrow points to the left coronary ostium.
In contrast, the left coronary artery arises from the left aortic
sinus and tends to arise at an acute angle and travel parallel to
the aortic sinus wall. When the left main artery is exceptionally
short, its ostium may assume a double-barrel appearance.
Among the various causes of coronary ostial stenosis, per-
haps the most common is degenerative calcification of the aortic
sinotubular junction, which often affects the right aortic sinus.
Stenosis of the right coronary ostium occurs 6 to 8 times more
often than that of the left. Aortitis associated with syphilis or
ankylosing spondylitis also may be complicated by coronary
ostial obstruction. Iatrogenic ostial injury may complicate coro-
nary arteriography, intraoperative coronary perfusion, or aortic
valve replacement.
The right coronary artery travels within the right atrioventric-
ular sulcus (or groove) (Figure 2.24). In 50% of persons, the first
anterior branch is the conus artery, which nourishes the right ven-
tricular outflow tract; in the remainder, this artery arises indepen-
dently from the right aortic sinus. The descending septal artery,
which arises from the proximal right coronary artery or, rarely,
from the conus artery or right aortic sinus, supplies blood to the
infundibular septum and, in some individuals, the distal atrioven-
tricular (His) bundle. Along the acute cardiac margin, from base
to apex, a prominent acute marginal branch courses, and between
this vessel and the conus artery, several smaller marginal branches
arise and travel parallel to the acute margin; these vessels nourish
the lateral two-thirds of the anterior right ventricular free wall.
Beyond the acute margin, along the inferior surface of the
heart, the length of the right coronary artery varies inversely
with that of the circumflex coronary artery. However, in 90%
of human hearts, the right coronary artery gives rise not only
to the posterior descending artery, which travels in the inferior
interventricular sulcus, but also to branches that supply blood to
the inferior left ventricular free wall. Accordingly, these arteries
nourish the inferior third of the ventricular septum (the inlet sep-
tum), including the right bundle branch and the posterior portion
of the left bundle branch, and the inferior left ventricular free
wall, including the posteromedial mitral papillary muscle.
Left Main Coronary Artery
The left main coronary artery travels between the pulmonary
artery and the left atrium and is covered in part by the left atrial
appendage. In two-thirds of persons, it bifurcates into left anterior
 2 Applied Anatomy of the Heart and Great Vessels 37

A
Arch of aorta
Superior vena cava
Pulmonary trunk

Sinuatrial nodal artery

L coronary artery (main stem)
R coronary artery L auricular appendage
L atrial rami
Outlines of
L conus artery
Anterior aortic sinus
R posterior aortic sinus
L posterior aortic sinus L (obtuse) marginal artery
Diagonal artery
R conus artery
Anterior interventricular artery
R anterior ventricular arteries
Interventricular anterior
Atrioventricular nodal artery septal rami

Posterior interventricular
arteries
R (acute) marginal artery

B C
Ascending aorta
R posterior
Pulmonary trunk atrial arteries
Superior vena cava

L pulmonary veins
R pulmonary veins
Stem of posterior
atrial arteries

Region of crux
of heart

Inferior vena cava

Anterior interventricular R coronary artery Posterior interventricular
artery (termination)
Posterior interventricular arteries
arteries

Figure 2.24. Coronary Arterial System. A, Anterior view. B and C, Posteroinferior views showing right dominance (B) and left dominance (C).
L indicates left; R, right. (Previously published. See “Credit Lines” section.)

descending and circumflex branches, and in the remaining third,
it trifurcates into the aforementioned branches and an intermedi-
ate artery (ramus intermedius), which follows a course similar to
that of either the first diagonal or the first marginal branch.
Left Anterior Descending Coronary Artery
The left anterior descending coronary artery travels within the
anterior interventricular sulcus (or groove) and, after wrapping
around the apex, may ascend a variable distance along the inferior
interventricular sulcus. Septal perforating branches nourish not
only the anterosuperior two-thirds and entire apical third of the
ventricular septum but also the atrioventricular (His) bundle
and the right and anterior left bundle branches (Figure 2.25).
The proximal septal perforators anastomose with the descend-
ing septal artery. Epicardial branches, called diagonals, nourish
the anterior left ventricular free wall and the medial third of the
anterior right ventricular free wall. Myocardial bridges may be
demonstrated angiographically in 12% of persons and almost
invariably involve the anterior descending artery; they produce
38 I Fundamentals of Cardiovascular Disease

Figure 2.25. Left Anterior Descending Coronary Artery With Septal Perforators.

critical systolic luminal narrowing in only 1% to 2% of hearts and
probably carry a benign prognosis in most cases (Figure 2.26).
Circumflex Coronary Artery
The circumflex coronary artery travels within the left atrioven-
tricular sulcus (or groove) and often terminates just beyond the
obtuse marginal branch. The circumflex artery nourishes the
lateral left ventricular free wall; however, in the 10% of persons
in whom the circumflex artery gives rise to the posterior descend-
ing branch, it also supplies blood to the inferior left ventricular
free wall and the inferior third of the ventricular septum. The
circumflex and anterior descending arteries nourish the ante-
rolateral mitral papillary muscles, and the circumflex and right
coronary arteries supply blood to the posteromedial mitral papil-
lary muscles.

Figure 2.26. Myocardial Bridge Over the Left Anterior Descending Coronary Artery.
 2 Applied Anatomy of the Heart and Great Vessels
The 4 major epicardial coronary arteries occupy only 2 planes
of the heart. The right and circumflex arteries delineate the plane
of the atrioventricular sulcus (cardiac base), and the left main
artery and anterior and posterior descending arteries delineate
the plane of the ventricular septum.
The origin of the posterior descending artery determines
the blood supply to the inferior portion of the left ventricle and
thereby defines coronary dominance. In 70% of hearts, the right
coronary artery crosses the crux and gives rise to this branch,
establishing right coronary dominance. In 10%, the circumflex
coronary artery terminates as the posterior descending branch
and thereby establishes left coronary dominance. Both the right
and circumflex arteries supply blood to the cardiac crux in the
remaining 20% and constitute so-called shared coronary domi-
nance. The dominant coronary artery, however, does not supply
blood to most of the left ventricular myocardium. In persons with
right coronary dominance, for example, the anterior descending
artery supplies blood to about 45% of the left ventricle, the cir-
cumflex about 20%, and the right coronary arteries 35%.
Blood Supply of the Cardiac Conduction
System
The sinus nodal artery arises from the right coronary artery in
60% of persons and from the circumflex artery in 40%, but its
artery of origin does not depend on patterns of coronary arterial
dominance. The atrioventricular nodal artery originates from the
dominant artery and, accordingly, arises from the right coronary
in 90% and the circumflex in 10%. The atrioventricular nodal
artery and the first septal perforator of the anterior descending
artery offer dual blood supply to the atrioventricular bundle
(ie, the bundle of His). Other septal perforating branches of the
Figure 2.27. Myocardial Arteriole.
39
anterior descending artery supply blood to the anterior aspect
of the left bundle branch, and septal perforators of the posterior
descending branch (an extension of the dominant artery) supply
blood to the posteroinferior portion of the left bundle branch. The
right bundle branch receives a dual blood supply from the septal
perforators of the anterior and posterior descending arteries.
Coronary Collateral Circulation
In the human heart, the major epicardial coronary arteries com-
municate with one another by means of anastomotic channels 50
to 200 μm in diameter (Figure 2.27). Normally, these small col-
lateral arteries carry very little blood flow. However, if arterial
obstruction induces a pressure gradient across such a channel,
with time the collateral vessel may dilate and provide an avenue
for significant blood flow beyond the stenotic lesion. Functional
collaterals may develop between the terminal branches of 2 coro-
nary arteries, between the side branches of 2 arteries, between
branches of the same artery, or within the same branch (via the
vasa vasorum). They are most numerous in the ventricular sep-
tum (between the septal perforators of the anterior and poste-
rior descending arteries), in the ventricular apex (between the
anterior descending septal perforators), in the anterior right
ventricular free wall (between the anterior descending and right
or conus arteries), in the anterolateral left ventricular free wall
(between the anterior descending diagonals and the circumflex
marginals), at the cardiac crux (between the right and circum-
flex arteries), and along the atria (the Kugel anastomotic artery
between the right and circumflex arteries). Smaller subendocar-
dial anastomoses also exist (Figure 2.28).
The most common sites for high-grade atherosclerotic lesions
are the proximal half of the anterior descending and circumflex
40 I Fundamentals of Cardiovascular Disease
Figure 2.28. Septal Perforators (Coronary Cast).
arteries and the origin and entire length of the right coronary
artery. The distribution and severity of atherosclerotic plaques
do not differ significantly among patients with angina pectoris,
acute myocardial infarction, end-stage ischemic heart disease, or
sudden death.
Congenital malformations of the coronary arteries include
anomalous ostial origin, anomalous arterial branching patterns,
and anomalous arterial anastomoses.
Coronary Veins
The venous circulation of the heart comprises a coronary sinus
system, an anterior cardiac venous system, and the thebesian
venous system (Figure 2.29). Small thebesian veins drain directly
into a cardiac chamber, particularly the right atrium or right ven-
tricle; the ostia of these veins are easily recognized along the
relatively smooth atrial walls but are difficult to identify in the
trabeculated ventricles.
During cardiac electrophysiologic studies among patients with
Wolff-Parkinson-White syndrome and left-sided bypass tracts, a
catheter electrode may be positioned within the coronary sinus
and great cardiac vein, adjacent to the mitral annulus, to localize
the aberrant conduction pathways. Left ventricular pacing leads
may also be placed in cardiac veins via the coronary sinus.
Cardiac Lymphatics
Myocardial lymphatics drain toward the epicardial surface,
where they are joined by lymphatic channels from the conduc-
tion system, atria, and valves. Larger epicardial lymphatics then
travel in a retrograde manner with the coronary arteries back
to the aortic root, where a confluence of right and left cardiac
lymphatics drains into a pretracheal lymph node and eventually
empties into the right lymphatic duct.
The coronary veins and cardiac lymphatics work in concert
to remove excess fluid from the myocardial interstitium and
pericardial sac. Accordingly, obstruction of either system or of
both systems may result in myocardial edema and pericardial
effusion.
Cardiac Conduction System
Sinus Node
The sinus node is the primary pacemaker of the heart. It is an
epicardial structure that measures approximately 15 × 5 × 2 mm
and is located in the sulcus terminalis (intercavarum) near the
superior cavoatrial junction (Figure 2.30). Through its center
passes a relatively large sinus nodal artery. Sinus nodal function
is greatly influenced by numerous sympathetic and parasympa-
thetic nerves that terminate within its boundaries.
Histologically, the sinus node consists of specialized cardiac
muscle cells embedded within a prominent collagenous stroma.
Its myocardial cells are smaller than ventricular muscle cells
and contain only scant contractile elements. Ultrastructurally,
the sinus node comprises transitional cells and variable numbers
of P cells centrally and atrial myocardial cells peripherally. The
P cells are thought to be the source of normal cardiac impulse
formation.
Because the sinus node occupies an epicardial position, its
function may be affected by pericarditis or metastatic neo-
plasms. In cardiac amyloidosis, extensive fibrosis or amyloid
deposition may involve the sinus node. Although the sinus node
is rarely infarcted, its function can be altered by adjacent atrial
infarction.
 2 Applied Anatomy of the Heart and Great Vessels
Pulmonary trunk
L pulmonary veins
Oblique vein of
L atrium
Great cardiac vein
L marginal vein
Posterior vein of
L ventricle
Middle cardiac vein
R marginal vein
Figure 2.29. Coronary Veins From Posterior Aspect of Heart. L indicates left; R, right. (Previously published. See “Credit Lines” section.)
Internodal Tracts
No morphologically distinct conduction pathways are seen
between the sinus and atrioventricular nodes with light micros-
copy, but electrophysiologic studies support the concept of 3
functional preferential conduction pathways. In ultrastructural
studies, some investigators have observed specialized cardiac
muscle cells in these internodal tracts.
Lipomatous hypertrophy of the atrial septum may interfere
with internodal conduction and induce various atrial arrhyth-
mias. Because the functional preferential pathways travel only
in the limbus and not in the valve of the fossa ovalis, internodal
conduction disturbances do not occur with intentional septal per-
foration at cardiac catheterization (transseptal approach), with
the Rashkind balloon atrial septostomy, or with the Blalock-
Hanlon partial (posterior) atrial septectomy. Severe disturbances
of internodal conduction may result from the Mustard operation
for complete transposition of the great arteries, in which the
entire atrial septum is resected; the surgical atriotomy may dis-
rupt the crista terminalis.
Atrioventricular Node
The atrioventricular node is a subendocardial right atrial
structure that measures approximately 6×4×1.5 mm. It is
located within the triangle of Koch (bordered by the tendon of
Todaro, septal tricuspid annulus, and coronary sinus ostium)
and abuts the right fibrous trigone (central fibrous body).
The atrioventricular nodal artery courses near the node but
not necessarily through it. Sympathetic and parasympathetic
nerves enter the atrioventricular node and greatly influence
its function.
Like the sinus node, the atrioventricular node histologically
consists of a complex interwoven pattern of small specialized
cardiac muscle cells within a fibrous stroma. With advanced age,
the atrioventricular node acquires progressively more fibrous tis-
sue, although not as extensively as the sinus node.
41
Aorta
Superior vena cava
R pulmonary veins
R atrium
Coronary sinus
Inferior vena cava
Small cardiac vein
The so-called mesothelioma of the atrioventricular node is a
small and rare primary neoplasm which, by virtue of its posi-
tion, produces various arrhythmias and may cause sudden death.
Metastatic neoplasms may rarely infiltrate the atrioventricular
node but do not necessarily alter its function. Sarcoid granulo-
mas tend to involve the basal ventricular myocardium and may
destroy the atrioventricular conduction system. Because of its
subendocardial position, the atrioventricular node may be ablated
nonsurgically at the time of electrophysiologic study.
Atrioventricular Bundle
The atrioventricular bundle (ie, the bundle of His) arises from the
distal portion of the atrioventricular node and courses through
the central fibrous body to the summit of the muscular ventricu-
lar septum, adjacent to the membranous septum. It provides
the only normal physiologic avenue for electrical conduction
between ventricles. The position of the atrioventricular bundle
within the central fibrous body (right fibrous trigone) is near the
annuli of the aortic, mitral, and tricuspid valves. The atrioven-
tricular bundle has a dual blood supply—from the atrioventricu-
lar nodal artery and from the first septal perforating branch of
the anterior descending artery. In some persons, a septal branch
of the proximal right coronary artery also nourishes the atrioven-
tricular bundle.
The atrioventricular bundle is made up of numerous paral-
lel bundles of specialized cardiac muscle cells, which are sepa-
rated by delicate fibrous septa. The entire atrioventricular bundle
is insulated by a collagenous sheath. With increasing age, the
fibrous septa become thicker, and the functional elements may
be partially replaced by adipose tissue. Ultrastructurally, the
atrioventricular bundle contains Purkinje cells and ventricular
myocardial cells in parallel arrangement.
In some persons, alternate conduction pathways exist between
the atria and the ventricles, either within the existing atrioven-
tricular conduction system or elsewhere along the fibrous car-
diac skeleton, and may produce various arrhythmias. Atrionodal
42 I Fundamentals of Cardiovascular Disease

Aorta

R auricular appendage
R pulmonary artery
Bachmann bundle Pulmonary valve
Superior vena cava
Sinuatrial node Torus aorticus
Pars membranacea septi
(atrioventricular part)
Fossa ovalis Radiation of L bundle branch
Tendon of Todaro R bundle branch
Atrioventricular node
Valve of inferior vena cava
Septal cusp of tricuspid valve
Coronary sinus
Septomarginal trabecula
(moderator band)
Inferior vena cava

Papillary muscle

B
L auricular appendage

Cut chordae tendineae Aorta

of mitral valve

Pulmonary trunk

R pulmonary veins
Anterior papillary muscle

Aortic valve

Radiation of L bundle branch

Inferior vena cava

Figure 2.30. Cardiac Conduction System. A, Right heart. The sinus (sinuatrial) and atrioventricular nodes are both right atrial structures. B,
left heart. The left bundle branch forms a broad sheet that does not divide into distinct anterior and posterior fascicles. L indicates left; R, right.
(Previously published. See “Credit Lines” section.)

bypass tracts (of James) connect the atria to the distal atrio-
ventricular node, and atriofascicular tracts (of Brechenmacher)
connect the atria to the atrioventricular bundle. Nodoventricular
and fasciculoventricular bypass fibers (of Mahaim) connect the
atrioventricular node and atrioventricular bundle, respectively, to
the underlying ventricular septal summit. These bypass fibers are
commonly observed histologically and are apparently nonfunc-
tional in most persons, although they may produce ventricular
preexcitation in some instances.
Ventricular preexcitation is usually associated with aberrant
atrioventricular bypass tracts that bridge the tricuspid or mitral
annuli. These tracts often travel within the adipose tissue of the
atrioventricular sulcus rather than through a defect in the valvu-
lar annuli. Such bypass tracts can be single or multiple and may
be identified by electrophysiologic mapping.
Acquired complete heart block may involve the atrioventric-
ular node and bundle or both bundle branches. That occurring
with acute myocardial infarction is usually transient and more
commonly complicates inferoseptal than anteroseptal infarction.
Usually the atrioventricular node and atrioventricular bundle are
edematous, or the bundle branches are focally infarcted. Acute
heart block also can complicate aortic infective endocarditis.
Chronic heart block may be associated with ischemic heart dis-
ease or with fibrocalcific disorders of the aortic or mitral valves,
but it most commonly results from idiopathic fibrosis of the atrio-
ventricular bundle and bilateral bundle branches. Heart block
may also complicate aortic or mitral valve replacement.
Congenital complete heart block manifests as persistent
bradycardia in utero and can represent an isolated anomaly or
may accompany other cardiac malformations. It results from
 2 Applied Anatomy of the Heart and Great Vessels 43

interruption of atrioventricular conduction pathways, either at plexus onto the heart and thereby innervate the coronary arter-
the junction between atrial muscle and the atrioventricular node ies, cardiac conduction system, and myocardium. Furthermore,
or at the junction between the atrioventricular node and the atrio- afferent nerves for pain and various reflexes ascend from the
ventricular bundle. The different embryologic origins of these heart toward the cardiac plexus.
3 regions account for the specific sites of disrupted conduction The transplanted human heart is completely denervated
tissue. and responds only to circulating (humoral) substances and not
to autonomic impulses. Similarly, afferent pathways are also
lost, including pain tracts and various reflexes. Consequently, if
Bundle Branches
chronic cardiac transplant rejection produces diffuse coronary
As an extension of the atrioventricular bundle, the right bundle arterial obstruction, subsequent myocardial ischemia and infarc-
branch forms a cordlike structure, approximately 50 mm in tion will be asymptomatic.
length and 1 mm in diameter, which courses along the septal and The asplenia syndrome is characterized by bilateral right-
moderator bands to the level of the anterior tricuspid papillary sided symmetry and is generally associated with right atrial
muscle. The left bundle branch forms a broad fenestrated sheet isomerism, right pulmonary isomerism, abdominal situs ambig-
of conduction fibers that spread along the septal subendocardium uus, and, in some instances, bilateral sinus nodes. In contrast, the
of the left ventricle and separates incompletely and variably sinus node may be congenitally absent or malpositioned in cases
into 2 or 3 indistinct fascicles. The fascicles travel toward the of polysplenia with left atrial isomerism.
left ventricular apex and both mitral papillary muscle groups.
The bundle branches are nourished by septal perforators aris- • The transplanted heart is completely denervated and responds
ing from the anterior and posterior descending coronary arteries. only to circulating (humoral) substances and not to autonomic
impulses.
Histologically, the bundle branches consist of parallel tracts of
specialized cardiac muscle cells that are insulated by a delicate
fibrous sheath. Ultrastructurally, Purkinje cells and ventricular
myocardial cells form the bundle branches.
Abbreviations Used in Figures
Right bundle branch block may be idiopathic or be associated A Anterior
with ischemic heart disease, chronic systemic hypertension, or Art Artery
pulmonary hypertension. Right ventriculotomy usually produces AL Anterolateral
the electrocardiographic features of right bundle branch block, AS Anteroseptal
AVS Atrioventricular septum
even though the bundle may not have been transected.
Ext External
Chronic left bundle branch block may be associated with fibro- I Inferior
calcific degeneration of the ventricular septal summit as a result IAS Interatrial septum
of chronic ischemia, left ventricular hypertension, calcification IL Inferolateral
of the aortic or mitral valves, or any form of cardiomyopathy. Inf Inferior
Int Internal
• The sinus node comprises transitional cells and variable numbers IS Inferoseptal
of P cells centrally and atrial myocardial cells peripherally. IVS Interventricular septum
• Severe disturbances of internodal conduction may result from the Jct Junction
Mustard operation for complete transposition of the great arteries, L Left
in which the entire atrial septum is resected. LA Left atrium
• The atrioventricular bundle has a dual blood supply—from the LLL Left lower lobe
atrioventricular nodal artery and from the first septal perforating LLPV Left lower pulmonary vein
branch of the anterior descending artery. LPA Left pulmonary artery
LUL Left upper lobe
• Acute heart block may complicate aortic infective endocarditis.
LV Left ventricle
• Congenital complete heart block manifests as persistent bradycar- LVOT Left ventricular outflow tract
dia in utero. Mes Mesenteric
MV Mitral valve
Cardiac Innervation P Posterior
PL Posterolateral
Because the embryonic heart tube first forms in the future neck PS Posteroseptal
region, its autonomic innervation also arises from this level. PT Pulmonary trunk
Three pairs of cervical sympathetic cardiac nerves originate R Right
from the cervical ganglia and intermingle as they join the cardiac RA Right atrium
plexus, between the great arteries and the tracheal bifurcation. RLL Right lower lobe
RLPV Right lower pulmonary vein
Several thoracic sympathetic cardiac nerves arise from the upper
RML Right middle lobe
thoracic ganglia and also join the cardiac plexus. From the para- RPA Right pulmonary artery
sympathetic vagus nerves emanate the superior and inferior cer- RUL Right upper lobe
vical vagal cardiac nerves and the thoracic vagal cardiac nerves, RV Right ventricle
which likewise interweave within the cardiac plexus. The various S Septal
sympathetic and parasympathetic nerves then descend from this Sup Superior
 3

Evidence-Based Medicine
and Statistics in Cardiology
CHARANJIT S. RIHAL, MD

Evidence-Based Medicine The results of 1 RCT (the HERS trial) reversed decades of
thought on the cardiovascular effects of female hormone therapy
Evidence-based medicine is the conscientious and explicit use
that had been based on an abundance of misinterpreted or biased
of the current best evidence from systematic research to guide
observational data. Judicious use of the evidence requires exhaus-
medical decision making in the care of individual patients. Its
tive searches for all relevant evidence and careful appraisal of its
practice takes into consideration 3 elements: clinical setting,
validity, results, and applicability.
scientific evidence, and physician-patient factors.

Clinical Setting Patient-Physician Factors

Evidence alone cannot direct management of an individual
The clinical setting includes specific information about the
patient. Physician expertise and patient preferences are crucial
patient (history, physical examination, imaging, and laboratory
in making the final medical decision, which is influenced by the
studies) and the location where the medical decision occurs,
prior experiences, expectations, ethics, and cultural beliefs of
including culture, societal values, characteristics of the practice
both parties.
site, and constraints due to time, reimbursement, and the avail-
ability of technology (eg, primary angioplasty may be better than
thrombolysis if access to a referral center is possible within a Statistics in Cardiology
reasonable time).
Statistics is the analysis of numerical data to infer proportions
in a population from those in a representative sample and to
Scientific Evidence measure the probability that observed results are chance find-
ings. Cardiology examinations test an understanding of the basic
Any observation about the relationship between events is con-
principles of biostatistics and their use in medical decision mak-
sidered evidence; however, a hierarchy of evidence exists. An
ing with statistical questions about 1) the probability of disease
RCT with a sample size of 1 provides definite evidence about
or therapeutic outcomes and 2) the evaluation of diagnostic test
the effects of treatment in an individual patient. Among studies
results.
of populations, the best evidence comes from meta-analyses of
methodologically sound RCTs that have consistent results. The
next best evidence is from single RCTs, which are better than Disease and Therapeutic Outcomes
meta-analyses of observational studies and single observational Prevalence and Incidence
studies. At the bottom of the hierarchy are unsystematic observa-
tions of clinical phenomena. The prevalence of a disease is the proportion of people in a popu-
lation who have the disease at a given time. The incidence of a
disease in a population is the proportion of people at risk of the
Abbreviations and acronyms are expanded at the end of this chapter. disease in whom the disease develops during a specific period.
44
 3 Evidence-Based Medicine and Statistics in Cardiology
Consider the following example: In a county of 100,000 peo-
ple, 2,000 have atrial fibrillation. Ten years later in the county,
300 people had died (of whom 25 had atrial fibrillation), 250
people had been born (none had atrial fibrillation), and 500 other
people had atrial fibrillation. The baseline prevalence of atrial
fibrillation in the population was 2% (2,000/100,000), the inci-
dence of atrial fibrillation in the baseline population during those
10 years was 0.5% (500/100,000), and the prevalence of atrial
fibrillation at the end of the 10 years was 2.5% (2,475/99,950).
Note that incidence must be used in the context of a specific time
period.
Measures of Effect
Measures of effect are summarized in Box 3.1 and defined as
follows:
1. Event rate is the number of people who have an event during follow-
up divided by the number of people at baseline who were at risk of
having the event. Event rates are usually described for experimental
and control groups in RCTs and observational studies.
2. Relative risk, which is synonymous with risk ratio, is the event rate
in the experimental group divided by the event rate in the control
group.
3. Odds ratio is a measure of the strength of association between a con-
dition or exposure and a disease or outcome. The odds of an event
is the probability of it occurring divided by the probability of it not
occurring. The OR is the odds of an event in 1 group (ie, an expo-
sure group) divided by the odds of the event in another group (ie, a
control group). It is the measure of effect used in cross-sectional and
case-control studies.
4. Absolute risk reduction is the event rate in the control group minus
the event rate in the experimental group—but only if the difference
is a positive number. If it is a negative number, its absolute value is
the absolute risk increase.
5. Relative risk reduction is the ARR divided by the event rate in
the control group. It can also be calculated as 1 minus the RR. If
the event rate in the experimental group is larger than the event
rate in the control group, the relative risk increase is calculated as
the ARI divided by the event rate in the control group, or the abso-
lute value of 1 minus the RR. The RRR is often the statistic used in
pharmaceutical promotional material because it shows the largest
numerical effect. When the event rate in the control group is small,
a large RRR may reflect a small ARR that does not have clinical
importance.
Box 3.1. Summary of Measures of Effect
RR = EER/CER
ARR = CER − EER
ARI = EER − CER
RRR = 1 − EER/CER
NNT = 1/ARR
NNH = 1/ARI
Abbreviations: ARI, absolute risk increase; ARR, absolute risk
reduction; CER, control event rate; EER, experimental event rate;
NNH, number needed to harm; NNT, number needed to treat; RR,
relative risk or risk ratio; RRR, relative risk reduction.
45
6. Number needed to treat better conceptualizes the risk relationships
described above. NNT is the number of patients who need to receive
an intervention to prevent 1 unfavorable outcome. This is calcu-
lated by rounding the reciprocal of the ARR to the next higher inte-
ger. NNT must always be accompanied by duration of treatment or
follow-up.
7. Number needed to harm is a concept similar to NNT, but it is the
number of patients who need to receive an intervention for 1 patient
to have an unfavorable outcome. This is calculated by rounding the
reciprocal of the ARI to the next higher integer. Like NNT, it must
be accompanied by a duration of treatment or follow-up.
• Incidence must be used in the context of a specific time period.
• OR is the measure of effect used in cross-sectional and case-
control studies.
• The RRR is often the statistic used in pharmaceutical promo-
tional material because it shows the largest numerical effect.
Null Hypothesis
The null hypothesis of a study states that any observed difference
(usually in treatment effects) between groups is due to chance
alone and that no true difference exists between groups. When
the probability that the study result is due to chance is less than
a specific value (α level), the null hypothesis is rejected and the
difference between groups is considered statistically significant.
α Level
The α level is the threshold value for statistical significance of
a test for differences between groups. The α level is the prob-
ability that a statistical test erroneously supports the conclusion
that a chance observed difference between groups is real (a type
I error). The α level is determined by the researcher (typically
.05 or .01). It is important to correctly interpret the P value in
the results of a study: If P is less than .05, there is less than a
5% chance that the observed difference between groups is due
to chance.
β Level
The β level is the probability that a statistical test erroneously
supports the conclusion that a real observed difference between
groups is due to chance (a type II error). The β level is determined
by the researcher (typically .2 or .1). The power of a statistical test
is the ability to detect a given difference between groups if one
exists, and it is calculated as 1 minus β. A β of .2 gives a statistical
power of .8 (ie, an 80% chance of observing a real difference).
Confidence Intervals
A confidence interval is a range of values in which the true
value for the total population (from which the study sample was
selected) is likely to exist. A 95% CI means that the true value
will be outside the CI 5% of the time. A CI that spans the line of
no difference (0 for ARR and RRR; 1 for RR and OR) represents
a result that is not statistically significant. In this case, compar-
ing the clinical significance of the lower and upper limits of the
CI might help identify a beneficial or harmful trend that could
be confirmed in larger trials. The CI is narrower (and the results
more precise) as the sample size increases. It is important to cor-
rectly interpret the CI in the results of a study: A 95% CI is a
range of values that 95% of the time will contain the real value
for the phenomenon being observed in the source population.
46 I Fundamentals of Cardiovascular Disease

An example of how the different measures of effect are calcu- Table 3.1. A 2 × 2 Table and Derivation of Diagnostic
lated and how these statistical concepts are interpreted is shown Test Features
below with the published data of the SHOCK Trial (N Engl J
Med. 1999 Aug 26;341[9]:625–34). This was a multicenter RCT Criterion Standard
Result
of 302 patients with cardiogenic shock complicating acute myo-
cardial infarction. Patients were randomly assigned to emer- Positive Negative
gency revascularization (experimental group, n = 152) or initial
medical stabilization (control group, n = 150). The primary end Diagnostic Test Result Positive a c
point was 30-day mortality. At 30 days, 71 patients had died in Negative b d
the revascularization group and 84 patients had died in the medi- a
cal group. Thus, the event rate in the revascularization group was Sensitivity =
a+b
47% (71/152), and the event rate in the medical therapy group
was 56% (84/150). The RR of death at 30 days for patients in d
the revascularization group compared with the medical therapy Specificity =
c+d
group was 84% (0.47/0.56). The ARR of death at 30 days for
a
patients in the revascularization group was 9.0% (0.56−0.47). PPV =
a+c
The RRR of death at 30 days for patients in the revasculariza-
tion group was 16% (1−0.84 or 0.09/0.56). The NNT to prevent d
NPV =
1 death at 30 days by performing emergent revascularization b+d
instead of using medical stabilization was 12 (1/0.09 = 11.1). The
P value for the ARR was .11, which means that there was an 11% a/( a + b ) Sensitivity
+ LR = =
chance that the observed difference was due to chance and not a c/(c + d ) 1 − Specificity
real difference between the groups. Thus, with a predetermined
b/( a + b ) 1 − Specificity
α level of .05, the null hypothesis was not rejected and the pri- −LR = =
mary study result was not statistically significant. The 95% CI d /(c + d ) Sensitivity
for the RR was 0.67 to 1.04, which means that 95% of the times Abbreviations: −LR, negative likelihood ratio; +LR, positive
that this protocol is reproduced in the same population, the real likelihood ratio; NPV, negative predictive value; PPV, positive
RR will be between 0.67 and 1.04. Since the CI spans the line of predictive value.
no difference (ie, 1), the result is not statistically significant.
positive, the disease is ruled in. A mnemonic for this concept
• The α level is the probability that a statistical test erroneously sup-
is SpPin (which refers to the phrase “specific test when positive
ports the conclusion that a chance observed difference between
groups is real (a type I error). rules in disease”).
The sensitivity and specificity of a test depend on the inherent
• The β level is the probability that a statistical test erroneously
quality of the test and the characteristics of the specific popula-
supports the conclusion that a real observed difference between
groups is due to chance (a type II error). tion in which it is tested. They might be most helpful when the
values are high and can be used to rule in or rule out disease;
• The CI is narrower (and the results more precise) as the sample
size increases.
however, they are not usually practical to the clinician since they
do not help revise the probability of disease in an individual
patient.
Diagnostic Tests
Medical decision making requires the appropriate understanding Predictive Values
and application of diagnostic test results, which include histori-
cal information, physical examination findings, and laboratory The PPV and NPV of a test provide the answer to the specific
evaluations. The performance of a diagnostic test is assessed by question a clinician might ask: What is the probability that the test
comparing the results of the diagnostic test with the results of a result (positive or negative) for this patient is true? It is important
criterion standard in a study population. Several terms are useful to note that the predictive values are affected by the prevalence
for describing the features of a diagnostic test (Table 3.1). of disease in the population in which the test is applied.
The PPV of a test is the probability that a patient with a
positive test result actually has the disease. This is calculated as
Sensitivity and Specificity
the number of true positives (a) divided by the total number of
The sensitivity of a test is the proportion of patients with the dis- positives (true positives [a] + false positives [c]).
ease who have a positive test. Sensitivity is calculated as the num- The NPV of a test is the probability that a patient with a nega-
ber of true positives (a) divided by the total number of patients tive test result actually does not have the disease. This is cal-
with the disease (ie, true positives [a] + false negatives [b]). If the culated as the number of true negatives (d) divided by the total
test has a high sensitivity and the result is negative, the disease is number of negatives (true negatives [d] + false negatives [b]).
ruled out. A mnemonic for this concept is SnNout (which refers
to the phrase “sensitive test when negative rules out disease”). • The sensitivity of a test is the proportion of patients with the dis-
ease who have a positive test.
The specificity of a test is the proportion of patients without
the disease who have a negative test. Specificity is calculated • The specificity of a test is the proportion of patients without the
as the number of true negatives (d) divided by the total num- disease who have a negative test.
ber of patients without the disease (ie, true negatives [d] + false • The PPV of a test is the probability that a patient with a positive test
positives [c]). If the test has a high specificity and the result is result actually has the disease.
 3 Evidence-Based Medicine and Statistics in Cardiology 47

• The NPV of a test is the probability that a patient with a negative The posttest probability of disease can be obtained in 2 ways.
test result actually does not have the disease. The first is by use of a nomogram (Figure 3.1). The second is by
converting the pretest probability to odds, multiplying by the LR
(or multiple LRs for serial tests), and then converting the posttest
Likelihood Ratios
odds to probability, as follows:
The +LR and −LR of a test are extremely useful measures
because they facilitate probabilistic medical decision making Odds
Probability =
(ie, the use of Bayes theorem). They can be used to combine the Odds + 1
results of multiple and serial diagnostic tests, and they can be
used directly to calculate posttest probability of disease. Also,
they are less affected by the prevalence of disease in a patient Odds = Probability
population. 1 − Probability
The +LR is calculated as the probability of having a positive
test in the presence of disease (a/a + b) divided by the probability
Pretest Odds × LR = Posttest Odds
of having a positive test in the absence of disease (c/c+ d). This is
equivalent to the sensitivity divided by the difference of 1 minus
The following formula simplifies the above conversions and
the specificity.
directly relates pretest probability to posttest probability without
The −LR is calculated as the probability of having a negative
converting to odds:
test in the presence of disease (a/a + b) divided by the probability
of having a negative test in the absence of disease (c/c + d). This Pretest Probability × LR
is equivalent to subtracting the sensitivity from 1 and dividing Posttest Probability =
that difference by the specificity. 1 + Pretest Probability × (LR − 1)

Use the +LR if a test is positive, and use the −LR if a test is neg-
0.1 99 ative. An LR greater than 1 increases the probability of disease,
and an LR less than 1 decreases the probability of disease. When
0.2 98 the pretest probability of disease is intermediate, an LR greater
than 10 essentially rules in the disease, whereas an LR less than
0.1 essentially rules out the disease.
0.5 95 An example of how to evaluate diagnostic tests is shown
2,000 below with the published data of the Breathing Not Properly
1 1,000 90 Multinational Study (Circulation. 2002 Jul 23;106[4]:416–22).
500
This was a prospective study of 1,538 patients presenting to
2 200 80 emergency departments with a chief complaint of dyspnea. The
100 accuracy of using a BNP concentration greater than 100 pg/mL
50 70
5 (a threshold derived from a receiver operating characteristic
20 60
curve in the same population) as a diagnostic test for heart failure
10 10 50 was compared with the criterion standard, which was consensus
5 40
20 2 30
1
30 0.5 20 Table 3.2. Evaluation of Brain Natriuretic Peptide (BNP)
40 0.2 Concentration as a Diagnostic Test for Heart Failure
50 0.1 10
Criterion Standard Result
60 0.05
5 (Cardiologists’ Review)
70 0.02
0.01 Positive Negative
80 0.005 2
Diagnostic Test Result Positive a = 650 c = 220
0.002 (BNP>100 pg/mL)
90 Negative b d
0.001 1
0.0005 a + b = 722
95 0.5 a + b + c + d = 1,538
Values for b and d are calculated by use of simple algebra. The following
are derived using the formulas in Table 3.1:
98 0.2
Sensitivity = 650/722 = 90%
99 0.1 Specificity = 596/816 = 73%
PPV = 650/870 = 75%
Pretest Likelihood Posttest NPV = 596/668 = 89%
Probability, % Ratio Probability, % +LR = 0.9/0.27 = 3.3
−LR = 0.1/0.73 = 0.14
Figure 3.1. Likelihood ratio nomogram for converting pretest prob-
ability to posttest probability. (Previously published. See “Credit Lines” Abbreviations: −LR, negative likelihood ratio; +LR, positive likelihood ratio;
section.) NPV, negative predictive value; PPV, positive predictive value.
48 I Fundamentals of Cardiovascular Disease

of 2 independent cardiologists who reviewed the entire medical Abbreviations

record and classified the cause of dyspnea as either heart failure ARI absolute risk increase
or noncardiac. The cardiologists’ review identified 722 patients ARR absolute risk reduction
with heart failure, of whom 650 had a BNP level greater than BNP brain natriuretic peptide
100 pg/mL. Another 220 patients also had a BNP level greater LR likelihood ratio
than 100 pg/mL. From these data, one can derive the sensitivity, −LR negative likelihood ratio
specificity, predictive values, and LRs for a BNP level greater +LR positive likelihood ratio
than 100 pg/mL for the diagnosis of heart failure. After first try- NNT number needed to treat
ing this yourself, review Table 3.2. Thus, for a patient with dys- NPV negative predictive value
pnea in the emergency department, with a history and physical OR odds ratio
PPV positive predictive value
examination findings that yield a pretest probability of 25% for
RCT randomized controlled trial
acute heart failure, a BNP level less than 100 pg/mL provides a RR relative risk
posttest probability of 4%, by using either the nomogram or the RRR relative risk reduction
posttest probability formula: 0.25 × 0.14/[1 + 0.25 × (0.14 − 1)].

Suggested Reading Names of Clinical Trials

The Evidence-Based Medicine Working Group. Users’ guides to the HERS Heart and Estrogen/Progestin Replacement Study
medical literature: essentials of evidence-based clinical practice. SHOCK Should We Emergently Revascularize Occluded Coronaries
Guyatt G, Rennie D, editors. Chicago (IL): AMA Press; c2002. for Cardiogenic Shock

Noncardiac Surgery in Patients
With Heart Disease
J. WELLS ASKEW III, MD, and CLARENCE SHUB, MD
Introduction
The ACCF and the AHA have published guidelines for the peri-
operative evaluation and management of patients undergoing
noncardiac surgery. The guidelines focus on patients undergo-
ing noncardiac surgery who are at risk of perioperative cardiac
morbidity or mortality. This at-risk group includes patients with
known CAD, patients with recent signs or symptoms suggestive
of CAD, and asymptomatic patients older than 50 years who
have the potential for coronary artery atherosclerosis. In general,
a conservative approach is recommended. Expensive, noninva-
sive or invasive preoperative testing (eg, coronary angiography
and revascularization) is rarely warranted just “to get the patient
through an operation.” Rather, the indications for extensive peri-
operative testing or revascularization are generally similar to
those in a nonoperative setting, and its use should be limited to
clinical scenarios in which the results will specifically influence
patient management.
• Focus on evaluating at-risk patients: patients with known CAD,
patients with recent signs or symptoms of CAD, or asympto-
matic patients older than 50 years who have the potential for
atherosclerosis.
• Testing or revascularization is not indicated just “to get the patient
through an operation.”
General and Stepwise Approach to
Preoperative Evaluation
A preoperative consultation should include the following: a
detailed clinical history with a focus on the patient’s cardio-
vascular status, physical examination with a comprehensive
Abbreviations and acronyms are expanded at the end of this chapter.
4
cardiovascular examination, determination of the patient’s func-
tional capacity, review of available pertinent data (laboratory
tests, ECG, and ancillary imaging studies), and consideration
of comorbid diseases, clinical risk, surgical risk, adjunct test-
ing if indicated, and perioperative medical therapy. A stepwise
approach for evaluating patients is outlined in the ACCF/AHA
2009 focused update of the 2007 perioperative guidelines and
consists of the following questions:
1. Does the patient need emergency cardiac surgery?
2. Does the patient have an active cardiac condition?
3. Is the surgery low risk?
4. Does the patient have satisfactory functional capacity without
symptoms?
5. Does the patient have clinical risk factors?
6. Is medical therapy (particularly β-blockers) indicated?
Does the Patient Need Emergency Surgery?
Extensive cardiac assessment, which may delay lifesaving sur-
gery, should not be performed in patient- or surgery-specific
instances when emergency surgical intervention is recommended.
The consultant should provide recommendations for periopera-
tive medical management and surveillance as well as selective
postoperative risk stratification if indicated.
Does the Patient Have an Active Cardiac
Condition?
Active cardiac conditions that could significantly increase peri-
operative risk include the following:
1. Unstable coronary syndromes
2. Decompensated heart failure (NYHA class IV: worsening or
new-onset heart failure)
49
50 I Fundamentals of Cardiovascular Disease
3. Clinically significant arrhythmias
4. Severe valvular disease
With the exception of emergency noncardiac surgery, these
active cardiac conditions merit further evaluation and manage-
ment before noncardiac surgery is performed.
Is the Surgery Low Risk?
Low-risk surgical procedures are procedures that have a reported
risk of less than 1% for cardiac death and nonfatal myocardial
infarction. Because of the exceedingly low cardiac event rate asso-
ciated with low-risk surgical procedures (even in high-risk patients),
patients do not generally require preoperative cardiac testing or new
pharmacologic therapy before surgery. Risk stratification based on
the noncardiac surgical procedure is important (Box 4.1).
High-risk operations include aortic or other major vascular sur-
gery and peripheral vascular surgery. Compared with other surgi-
cal procedures, high-risk operations have been associated with a
higher incidence of postoperative congestive heart failure and a
3-fold greater incidence of myocardial infarction. A major opera-
tion is often associated with large extravascular and intravascular
fluid shifts or blood loss and postoperative hypoxemia. The magni-
tude and anticipated duration of the procedure are also important.
More extensive procedures and long procedures are associated
with greater perioperative risk. Patients undergoing peripheral
vascular operations are at increased risk, primarily because of the
increased incidence of associated CAD. Emergency major opera-
tions, especially in the elderly, are also considered high risk.
Box 4.1. Cardiac Riska Stratification for Noncardiac
Surgical Procedures
High risk (reported cardiac risk often >5%)
Emergency major operations, particularly in the
elderly
Aortic or other major vascular surgery
Peripheral vascular surgery
Anticipated prolonged surgical procedures
associated with large fluid shifts or blood loss
Intermediate risk (reported cardiac risk
generally 1%–5%)
Carotid endarterectomy
Head and neck surgery
Intraperitoneal and intrathoracic surgery
Orthopedic surgery
Prostate surgery
Low riskb (reported cardiac risk generally <1%)
Endoscopic procedures
Superficial procedures
Cataract surgery
Breast surgery
a
Combined incidence of cardiac death and nonfatal myocardial
infarction.
b
Generally, further preoperative cardiac testing is not required.
(Previously published. See “Credit Lines” section.)
Intermediate-risk operations include carotid endarterectomy,
head and neck procedures, orthopedic operations, prostate oper-
ations, and intraperitoneal and intrathoracic procedures.
Low-risk operations include ophthalmologic procedures,
endoscopic surgery, breast surgery, uncomplicated herniorrhaphy,
and other ambulatory operations (eg, cutaneous procedures).
Does the Patient Have Satisfactory Functional
Capacity Without Symptoms?
Functional status is a reliable predictor for perioperative and
long-term cardiac events and is typically defined on the basis of
METs. A patient’s functional status or MET level can often be
estimated by a detailed clinical history and the patient’s abil-
ity to perform activities of daily living (Table 4.1). Activities
reflecting an exertional tolerance of 4 to 5 METs (eg, climbing 2
flights of stairs without stopping because of limiting symptoms)
indicate an acceptable FAC that approximates the physiologic
stress of most noncardiac surgical procedures. Patients with a
satisfactory FAC, as described above, can generally proceed with
noncardiac surgery without additional perioperative testing or
pharmacologic therapy.
Does the Patient Have Clinical Risk Factors?
The clinical risk factors used for preoperative cardiac evaluation
in the ACCF/AHA guideline algorithm include ischemic heart
disease, compensated or prior heart failure, diabetes mellitus,
renal insufficiency, and cerebrovascular disease. These clinical
risk factors have been associated with an increased risk of car-
diac events at noncardiac surgery and differ from the traditional
atherosclerotic risk factors (ie, hypertension, hyperlipidemia,
diabetes mellitus, tobacco use, and family history of premature
atherosclerosis) defined in the original Framingham Heart Study.
The presence of clinical risk factors should be determined if the
patient has a functional capacity less than 4 METs, an unknown
functional capacity, or cardiac symptoms.
The clinical cardiac risk factors defined in the ACCF/AHA
preoperative guidelines were derived from the Revised Cardiac
Risk Index, which consists of clinical predictors of major adverse
Table 4.1. Estimated Requirements for Various Activities
Requirement Activity
1 MET Can you take care of yourself?
Eat, dress, or use the toilet?
Walk indoors around the house?
Walk a block or 2 on level ground at 2–3 mph
(3.2–4.8 kph)?
Do light work around the house like dusting or
washing dishes?
4 METs Climb a flight of stairs or walk up a hill?
Walk on level ground at 4 mph (6.4 kph)?
Run a short distance?
Do heavy work around the house like scrubbing
floors or lifting or moving heavy furniture?
Participate in moderate recreational activities such
as golf, bowling, dancing, doubles tennis, or
throwing a baseball or football?
>10 METs Participate in strenuous sports such as swimming,
singles tennis, football, basketball, or skiing?
Abbreviations: kph, kilometers per hour; MET, metabolic equivalent task; mph,
miles per hour.
(Previously published. See “Credit Lines” section.)
 4 Noncardiac Surgery in Patients With Heart Disease
Box 4.2. Predictors in the Revised Cardiac Risk Index
High-risk noncardiac surgery
History of ischemic heart disease
History of congestive heart failure
History of cerebrovascular disease
Diabetes mellitus requiring preoperative insulin
Preoperative creatinine >2.0 mg/dL
(Previously published. See “Credit Lines” section.)
perioperative cardiac events that were retrospectively identified
and then prospectively validated. The Revised Cardiac Risk Index
identified 6 independent predictors of major cardiac complica-
tions (Box 4.2). One can easily calculate the risk of major cardiac
complications by using the number of predictors (Table 4.2). Not
surprisingly, as the number of clinical risk factors increases, the
perioperative risk increases.
Clinical risk factors can be stratified further into major,
intermediate, and low (or minor) risks (Box 4.3). Active car-
diac conditions (unstable coronary syndromes, decompensated
heart failure, significant arrhythmias, and severe valvular heart
disease) are more important than dormant ones. When present,
active cardiac conditions indicate major clinical risk. The pres-
ence of 1 or more major risk predictors warrants further evalua-
tion and (usually) treatment that may delay or cancel the elective
operation. Intermediate risk factors warrant careful clinical
assessment and, when appropriate, use of additional cardiac test-
ing. Minor predictors have relatively less clinical importance
and have not been recognized as independent perioperative risk
factors.
In general, if the patient has no clinical risk factors (see
above), noncardiac surgery can proceed as planned. If the patient
has 1 or more clinical risk factors, additional perioperative car-
diac assessment may be indicated. If a patient will undergo
intermediate-risk surgery and has at least 1 clinical risk factor,
the clinician must decide whether to proceed with the planned
surgery using pharmacologic control or consider noninvasive
cardiac testing (if testing will change management).
Additional Preoperative Cardiac Evaluation
Additional preoperative cardiac testing may be indicated on the
basis of the initial clinical assessment, the presence of active
cardiac conditions, the number of clinical risk factors, or poor
functional capacity (Box 4.4).
It is important to carefully review the results of noninvasive
stress tests when performed as part of a preoperative assessment
Table 4.2. Rates of Major Cardiac
Complications in the Revised Cardiac
Risk Index
No. of Predictors Cardiac Risk, %
0 0.4
1 0.9
2 7
≥3 11
(Previously published. See “Credit Lines” section.)
51
Box 4.3. Clinical Predictors of Increased Periopera-
tive Cardiovascular Risk (Myocardial Infarction,
Congestive Heart Failure, and Death)
Major risk
Unstable coronary syndromes
Recent myocardial infarctiona with evidence of important
ischemic risk by clinical symptoms or noninvasive study
Unstable or severeb angina (Canadian class III or IV)
Decompensated congestive heart failure
Significant arrhythmias
High-grade atrioventricular block
Symptomatic ventricular arrhythmias in the presence of
underlying heart disease
Supraventricular arrhythmias with uncontrolled
ventricular rate
Severe valvular disease
Intermediate risk
History of ischemic heart disease
History of cerebrovascular disease
Compensated or prior congestive heart failure
Diabetes mellitus
Renal insufficiency
Minor risk
Advanced age
Abnormal ECG (left ventricular hypertrophy, left
bundle branch block, ST-T abnormalities)
Rhythm other than sinus (eg, atrial fibrillation)
Low functional capacity (eg, inability to climb 1
flight of stairs with a bag of groceries)
Uncontrolled systemic hypertension
Abbreviation: ECG, electrocardiogram.
a
The American College of Cardiology National Database Library
defines recent myocardial infarction as myocardial infarction
occurring >7 days but ≤30 days previously.
b
May include “stable” angina in patients who are unusually
sedentary.
(Previously published. See “Credit Lines” section.)
of cardiac risk (Box 4.5). One of the important features of sup-
plemental cardiac testing is the ability to objectively measure a
patient’s FAC in addition to assessing the hemodynamic response
with stress, the presence of inducible myocardial ischemia, and
arrhythmias.
Perioperative cardiac events increase in patients who have an
abnormal exercise stress test at a low cardiac workload. Risk is
increased in patients who have inducible ischemia at a low level
of exercise (<4 METs), at a heart rate less than 100 beats per
minute, or at less than 70% of the age-predicted heart rate in con-
junction with 1 or more of the following: 0.1 mV or more of hori-
zontal or downsloping ST-segment depression, more than 0.1 mV
of ST-segment elevation in the absence of pathologic Q waves,
ST-segment deviation in multiple ECG leads, typical angina,
52 I Fundamentals of Cardiovascular Disease
Box 4.4. Additional Preoperative Cardiac Testing
Electrocardiogram
Recommended for patients who have at least 1
clinical risk factor and are undergoing a vascular
surgical procedure
Recommended for patients who have known
coronary artery disease, peripheral arterial disease,
or cerebrovascular disease and are undergoing
intermediate-risk surgical procedures
Not indicated for asymptomatic patients
undergoing low-risk surgical procedures
Noninvasive evaluation of left ventricular function
Not recommended as part of a routine
perioperative evaluation
Reasonable to do if patients have unexplained
dyspnea or if patients with current or prior heart
failure have worsening dyspnea
Stress Testing
Patients with active cardiac conditions should be
evaluated and treated according to ACCF/AHA
guidelines
Reasonable to do if patients have ≥3 clinical risk
factors and a low FAC (<4 METs) and they are
undergoing vascular surgery (assuming that results
may change management)
Consider for patients who have 1 or 2 clinical
risk factors and a low FAC and are undergoing
intermediate-risk or vascular surgery (assuming that
results may change management)
Not useful if patients will undergo low-risk surgery or
if patients with no clinical risk factors will undergo
intermediate-risk surgery
Abbreviations: ACCF, American College of Cardiology Foundation;
AHA, American Heart Association; FAC, functional aerobic capacity;
MET, metabolic equivalent task.
prolonged ischemic response after exercise, or exercise-induced
decrease in systolic blood pressure (>10 mm Hg).
Myocardial perfusion imaging has been extensively studied
as a risk stratification tool in patients before surgery. The nega-
tive predictive value of a normal scan is high (approximately
99% for nonfatal myocardial infarction or cardiac death). The
presence of ischemia and the extent of the perfusion abnormal-
ity can be helpful in perioperative risk stratification. Cardiac
events increase as a function of the extent of perfusion abnor-
mality, with the lowest risk among patients with normal results
or low-risk results (ie, mildly abnormal). A reversible perfusion
defect has a greater effect on perioperative cardiac events than a
fixed defect, but a fixed defect indicates a greater risk than a nor-
mal scan. Reversible defects accounting for at least 20% of the
left ventricular mass have been shown to significantly increase
the risk of perioperative myocardial infarction or cardiac death
(clinical risk increases with increasing extent of ischemia).
Similar findings have been demonstrated with stress echocar-
diography. Several studies suggest that the extent of stress-induced
Box 4.5. Prognostic Gradient of Ischemic Responses
During an ECG-Monitored Exercise Test for Patients
With Suspected or Proven CAD
High risk
Ischemia induced by low-level exercisea (<4 METs
or heart rate <100 bpm [or <70% of age-predicted
rate]), manifested by ≥1 of the following:
Horizontal or downsloping ST-segment depression >0.1 mV
ST-segment elevation >0.1 mV in noninfarct lead
Abnormal findings in ≥5 leads
Persistent ischemic response >3 min after exertion
Typical angina
Intermediate risk
Ischemia induced by moderate-level exercise (4–6
METs or heart rate 100–130 bpm [or 70%-85% of age-
predicted rate]), manifested by ≥1 of the following:
Horizontal or downsloping ST-segment depression >0.1 mV
Typical angina
Persistent ischemic response >1–3 min after exertion
Abnormal findings in 3 or 4 leads
Low risk
No ischemia or ischemia induced at high-level
exercise (>7 METs or heart rate >130 bpm [or >85% of
age-predicted rate]), manifested by the following:
Horizontal or downsloping ST-segment depression >0.1 mV
Typical angina
Abnormal findings in 1 or 2 leads
Inadequate test
Inability to reach adequate target workload or heart
rate response for age without an ischemic response
For patients undergoing a noncardiac operation,
ability to exercise to at least the intermediate-risk
level without ischemia should be considered a low
risk for perioperative ischemic events
Abbreviations: bpm, beats per minute; CAD, coronary artery disease;
ECG, electrocardiographically; MET, metabolic equivalent task.
a
Workload and heart rate estimates for risk severity require adjustment
for patient age. Maximal target heart rates for 40- and 80-year-olds
taking no cardioactive medication are 180 and 140 bpm, respectively.
(Previously published. See “Credit Lines” section.)
wall motion abnormalities and, perhaps more importantly, the
calculated ischemic threshold can be used to identify persons at
risk of perioperative events. Development of wall motion abnor-
malities during dobutamine stress echocardiography at low heart
rates (particularly at a heart rate <60% of the age-predicted
maximum) is a predictor of perioperative events (especially for
patients with clinical risk factors).
• Reversible myocardial perfusion defects (>20% of the left ventri-
cle) increase the risk of perioperative cardiac events.
• Abnormal dobutamine stress echocardiograms with an ischemic
threshold <60% indicate increased risk of perioperative cardiac
events.
 4 Noncardiac Surgery in Patients With Heart Disease
Disease-Specific Considerations
Coronary Artery Disease
When patients have known CAD, important information to con-
sider includes the extent and severity of jeopardized myocar-
dium, the ischemic threshold (as assessed by dobutamine stress
echocardiography) (ie, the cardiac workload required to induce
ischemia), underlying left ventricular systolic function, and cur-
rent cardiac medications.
Interventional procedures rarely are needed only to lower the risk
of a noncardiac operation. Thus, the strategy of performing coronary
revascularization only to avoid perioperative cardiac complications
should be reserved for a small subset of very high-risk patients.
According to the ACCF/AHA guidelines, class I indications
for preoperative coronary angiography for patients with sus-
pected or proven CAD include the following (Box 4.6):
1. Evidence for high risk of adverse outcome from results of noninva-
sive testing
2. Severe (class III or IV) angina unresponsive to medical therapy
3. Unstable angina
4. Nondiagnostic or equivocal noninvasive test results for a high-risk
patient
Coronary angiography is not indicated for low-risk patients or
for those who are asymptomatic after coronary revascularization
and have good exercise capacity.
Preoperative (noncardiac surgery) coronary revascularization
with coronary artery bypass graft surgery or percutaneous coro-
nary intervention is generally reserved for patients who have any
of the following:
1. Symptomatic left main or 3-vessel disease
2. Stable angina with 2-vessel disease including the proximal left ante-
rior descending artery and a left ventricular ejection fraction less
than 50% or significant ischemia on stress testing
3. High-risk unstable angina
4. Non–ST-segment elevation myocardial infarction
5. Acute ST-segment elevation myocardial infarction
The strategy of performing coronary angiography and percu-
taneous intervention (angioplasty or stenting) before a noncardiac
operation to reduce the risk of a noncardiac procedure depends
on individual circumstances and has not proved beneficial in
controlled clinical trials. Prolonged antiplatelet therapy with
clopidogrel after coronary stenting mandates delaying noncar-
diac surgery. The frequency of perioperative cardiac events in a
patient who has undergone percutaneous coronary intervention is
highest immediately after the operative procedure and decreases
with time. After balloon angioplasty only, it is reasonable to delay
surgery for 4 weeks to allow vessel healing. Delaying surgery for
more than 8 weeks after balloon angioplasty may theoretically
increase perioperative cardiac risk owing to elastic recoil and
restenosis of the treated coronary artery segment. In patients who
require noncardiac surgery after placement of bare metal stents,
noncardiac surgery should be delayed at least 6 weeks after stent
placement, at which time stents are generally endothelialized and
antiplatelet therapy can be temporarily discontinued. If drug-
eluting stents are used, delaying elective surgery for 12 months, if
possible, is preferable. The potential risk of stent thrombosis due
to prematurely stopping clopidogrel therapy should be considered
in the overall perioperative risk assessment.
• Generally, the perioperative risk with nonvascular, non–high-risk
operations is low, limiting the value of cardiac stress testing.
Standard clinical evaluation should suffice for most low-risk
patients.
53
Box 4.6. Recommendations for Coronary Angiog-
raphy in Perioperative Evaluation
Class I indications for patients with suspected
or known CAD
Evidence for high risk of adverse outcome based on
noninvasive test results
Angina unresponsive to adequate medical therapy
Unstable angina, particularly when facing
intermediate-risk or high-risk noncardiac surgery
Equivocal noninvasive test results for patients at
high clinical risk undergoing high-risk surgery
Class IIa indications
Multiple markers of intermediate clinical risk and
planned vascular surgery (noninvasive testing
should be considered first)
Moderate to large region of ischemia on
noninvasive testing but without high-risk features
and without lower LVEF
Nondiagnostic noninvasive test results for patients
of intermediate clinical risk undergoing high-risk
noncardiac surgery
Urgent noncardiac surgery while convalescing from
acute MI
Class IIb indications
Perioperative MI
Medically stabilized class III or IV angina and
planned low-risk or minor surgery
Class III indications
Low-risk noncardiac surgery with known CAD and
no high-risk results on noninvasive testing
Asymptomatic after coronary revascularization with
excellent exercise capacity (≥7 METs)
Mild stable angina with good left ventricular
function and no high-risk noninvasive test results
Noncandidate for coronary revascularization
owing to concomitant medical illness, severe left
ventricular dysfunction (eg, LVEF <20%), or refusal
Candidate for liver, lung, or renal transplant who is
older than 40 years as part of evaluation for transplant,
unless noninvasive testing indicates high risk
Abbreviations: CAD, coronary artery disease; LVEF, left ventricular
ejection fraction; MET, metabolic equivalent task; MI, myocardial
infarction.
(Previously published. See “Credit Lines” section.)
• Active patients who have chronic stable angina and can perform
activities of daily living (4–5 METs) probably can tolerate the
stress of most types of noncardiac operations.
• If a patient has undergone coronary revascularization within the
past 5 years and if the clinical status has remained stable without
recurrent symptoms or signs of ischemia, additional cardiac test-
ing generally is not needed.
54 I Fundamentals of Cardiovascular Disease
• For patients with ischemic heart disease who have not had coro-
nary revascularization but who have undergone stress testing in the
past 2 years (assuming adequate testing and a favorable outcome of
testing), it usually is unnecessary to repeat testing unless there has
been an acceleration of angina or new symptoms of ischemia have
appeared during the interim.
• In selected patients with known CAD and significant (class III or
IV) symptomatic limitation or accelerating angina, preoperative
coronary angiography is often indicated, as it would be even if a
noncardiac operation were not being contemplated.
• The strategy of performing coronary angiography and “preven-
tive” angioplasty or coronary stenting preoperatively to reduce the
risk of noncardiac surgery has not been proved in controlled clin-
ical trials.
• Perioperative medical treatment is an acceptable option for care-
fully screened patients who have stable CAD and are scheduled
for vascular surgery.
• Patients who have received a stent should wait at least 6 weeks after
placement of a bare metal stent and 12 months after placement of a
drug-eluting stent before they undergo noncardiac surgery.
Vascular Disease
Over time, perioperative cardiac event rates have decreased, espe-
cially for patients undergoing nonvascular operations, partly because
of improved patient selection, anesthetic techniques, and periopera-
tive management. Many recent studies have focused on patients
having vascular procedures, because they are at higher risk.
Major open vascular surgery carries a high risk of periopera-
tive cardiac events (>5%) and is also associated with shifts in
intravascular volume that result in labile systemic and intracar-
diac pressures. Stress testing is not routinely recommended as
a way to reduce cardiac risk in patients who are at intermediate
clinical risk (1 or 2 clinical risk factors as previously described),
provided that they are receiving β-blocker therapy with accept-
able heart rate control. A study that randomly assigned patients
undergoing a major vascular surgery to stress testing or no stress
testing did not find a significant difference in rates of periopera-
tive nonfatal myocardial infarction or cardiac death.
Results from routine coronary angiography performed before
a vascular operation have shown that more than half of patients
with clinically suspected CAD have severe multivessel or inop-
erable CAD. Even patients with peripheral vascular disease and
no previous history of heart disease may have severe CAD, espe-
cially those with diabetes mellitus.
• Even patients with peripheral vascular disease and no previous his-
tory of heart disease may have severe CAD, especially those with
diabetes mellitus.
Valvular Heart Disease
In patients with valvular heart disease, the risk of a noncardiac
operation depends on 1) the type, anatomical location, and sever-
ity of the valve lesion; 2) left ventricular systolic function; and 3)
NYHA functional class.
Patients with severe, symptomatic aortic stenosis have the
greatest risk and, ideally, should undergo a corrective aortic
valve operation or, in selected cases, balloon valvuloplasty before
having a noncardiac operation. However, aortic balloon valvulo-
plasty has inherent risks, including serious vascular access com-
plications and embolic events. A selected, small group of Mayo
Clinic patients who had severe aortic stenosis and who were not
candidates for (or who refused) an aortic valve operation or val-
vuloplasty had noncardiac operations with a low risk of having
major complications; nonetheless, most patients with aortic ste-
nosis should be considered high risk, and aortic valve replace-
ment is generally warranted. Patients with severe mitral stenosis
are at increased risk of perioperative heart failure, especially
if tachycardia occurs. Patients with milder degrees of aortic or
mitral stenosis have a lower risk.
Generally, patients with aortic or mitral regurgitation, espe-
cially if they have only mild symptoms, seem to be at lower risk
than those with stenotic lesions. Patients with advanced symptoms
of heart failure (NYHA class III or IV) and those with severe val-
vular regurgitation and left ventricular systolic dysfunction are
at greater risk (regardless of the mechanism) and should undergo
further evaluation and treatment before having a noncardiac oper-
ation, especially if it includes a high-risk surgical procedure.
• Patients with severe, symptomatic aortic stenosis have the greatest
risk and, ideally, should undergo a corrective aortic valve opera-
tion before having a noncardiac operation.
• Patients with severe mitral stenosis are at increased risk of periop-
erative heart failure, especially if tachycardia occurs.
• Patients with advanced symptoms of heart failure (NYHA class III
or IV) and those with severe valvular regurgitation and left ven-
tricular systolic dysfunction are at greater risk (regardless of the
mechanism) and should undergo further evaluation and treatment
before having a noncardiac operation, especially if it includes a
high-risk surgical procedure.
Hypertrophic Obstructive Cardiomyopathy
In general, patients with hypertrophic obstructive cardiomyop-
athy tolerate noncardiac operations reasonably well. However,
some of them are at increased risk. Patients with hypertrophic
cardiomyopathy generally require closer perioperative monitor-
ing. Hemodynamic changes associated with an anesthetic-related
decrease in peripheral resistance, hypovolemia, or adrenergic
stimulation may increase the left ventricular outflow tract gradi-
ent and lead to hemodynamic deterioration.
• In patients with hypertrophic obstructive cardiomyopathy, hemo-
dynamic changes associated with an anesthetic-related decrease
in peripheral resistance, hypovolemia, or adrenergic stimulation
may increase the left ventricular outflow tract gradient and lead to
hemodynamic deterioration.
Arrhythmias and Conduction Disturbances
Postoperative atrial tachyarrhythmias affect almost 1 million
patients annually. In contrast, bradyarrhythmias or ventricular
arrhythmias severe enough to require treatment affect less than
1% of patients undergoing noncardiac surgery. Clinical evalua-
tion should seek to uncover any underlying heart or pulmonary
disease, drug toxicity, and electrolyte or metabolic abnormality
that might be causing arrhythmias or conduction disturbances.
Symptomatic or hemodynamically significant arrhythmias
should be treated before the patient undergoes a noncardiac oper-
ation; the indications for treatment are similar to those in the
nonoperative setting. It is important to correct even mild degrees
of preoperative hypokalemia in patients taking digitalis. The res-
piratory alkalosis that usually occurs during general anesthesia
may cause a decrease in extracellular potassium concentration
and provoke arrhythmias. Asymptomatic conduction system
disease such as bundle branch block, bifascicular block, or even
trifascicular block does not predict high-grade or complete heart
block during a noncardiac operation and does not by itself man-
date prophylactic temporary pacing. Atrial tachyarrhythmias
are a common complication after thoracic surgery and can be
 4 Noncardiac Surgery in Patients With Heart Disease
associated with a longer hospital stay. Calcium channel blockers
and β-blockers reduce the risk of atrial tachyarrhythmias.
• Symptomatic or hemodynamically significant arrhythmias should
be treated before the patient undergoes a noncardiac operation.
• It is important to correct even mild degrees of preoperative hypo-
kalemia in patients taking digitalis.
• Asymptomatic conduction system disease such as bundle branch
block, bifascicular block, or even trifascicular block does not pre-
dict high-grade or complete heart block during a noncardiac opera-
tion and does not by itself mandate prophylactic temporary pacing.
Perioperative Myocardial Infarction
The risk of perioperative myocardial infarction in patients with-
out clinical evidence of heart disease is approximately 0.15%.
Although the mechanism of myocardial infarction may be due
to hemodynamic changes in the perioperative period (“demand
ischemia”), coronary plaque rupture has also been described in
autopsy series with thrombotic coronary artery occlusion. The
mortality rate associated with perioperative myocardial infarc-
tion is significantly higher than that with an infarct unrelated to
an operation. Previously, the risk of perioperative myocardial
infarction was less well recognized and the antemortem diag-
nosis was more difficult. Increased awareness of the problem,
better patient selection, improved anesthetic and operative tech-
niques, improved perioperative monitoring and management,
and improved diagnostic techniques (including biomarkers such
as serum troponin) have all contributed to a reduction in mortal-
ity from perioperative myocardial infarction.
The risk of perioperative reinfarction is increased in the
first 6 months after an index myocardial infarction. This risk
decreases with increasing time between the index infarction and
the planned operation. After percutaneous revascularization,
patients have a high-risk period of 6 weeks and then an interme-
diate-risk period of 3 to 6 months. Ideally, noncardiac surgery is
delayed at least 3 months. Sometimes clinical circumstances may
warrant proceeding with surgery sooner than recommended (eg,
rapidly spreading tumors, impending aortic aneurysm rupture,
major fractures, and infections requiring drainage).
According to the ACCF/AHA guidelines, an elective surgi-
cal procedure can be performed before 6 months has elapsed
provided that the patient undergoes postinfarction risk stratifica-
tion. Absence of postinfarction ischemia, a negative postinfarc-
tion stress test, and complete myocardial revascularization after
infarction suggest a reduced risk of reinfarction with an elective
operation. The ACCF/AHA guidelines suggest that it is prudent
to wait at least 4 to 6 weeks after infarction before proceeding
with an elective operation.
• Patients without clinical evidence of heart disease are at low risk
(about 0.15%) of perioperative myocardial infarction.
• Risk of perioperative reinfarction varies inversely with the time
between the index infarction and the operation.
• Patients with a negative postinfarction stress test or complete
postinfarction myocardial revascularization can proceed with an
elective operation at 4 to 6 weeks after infarction.
Preoperative Hemodynamic Assessment and
Intraoperative Hemodynamic Monitoring
If overt heart failure is present, medical therapy should be opti-
mized preoperatively, but dehydration and hypotension from
overly aggressive diuretic and vasodilator therapy must be avoided.
In patients with overt heart failure, preoperative or intraoperative
55
monitoring with Swan-Ganz catheters may be useful in selected
cases, especially with high-risk procedures, so that intravenous
fluid and drug therapy can be guided optimally. Monitoring
should be continued into the postoperative period, when major
extravascular fluid mobilization could precipitate pulmonary
edema in patients with severe valvular heart disease or left ven-
tricular dysfunction. The risks of invasive hemodynamic monitor-
ing must be balanced against the potential benefits. Many patients
can be managed adequately on a clinical basis without the need
for invasive hemodynamic monitoring. Randomized trials have
not proved a major benefit from invasive hemodynamic monitor-
ing for decreasing perioperative cardiac morbidity. Preoperative
serum brain natriuretic peptide determinations may be useful as a
predictor of perioperative outcome in selected patients.
• If overt heart failure is present, medical therapy should be opti-
mized preoperatively.
• In patients with overt heart failure, preoperative or intraoperative
monitoring with Swan-Ganz catheters may be useful, especially
with high-risk procedures, so that intravenous fluid and drug ther-
apy can be guided optimally.
• Randomized trials have not proved a major benefit from invasive
hemodynamic monitoring for decreasing perioperative cardiac
morbidity.
Preoperative Medications and Management
Cardiovascular medications should be taken up to the time of the
operation and then as soon after the operation as possible. Many
studies have suggested that β-blockers reduce perioperative
ischemia and may decrease adverse cardiac events. β-Blocker
therapy should be continued in patients undergoing surgery who
are already receiving β-blocker therapy. β-Blockers titrated to
heart rate and blood pressure are typically recommended for
patients undergoing vascular surgery who have been identified as
having more than 1 clinical risk factor (ischemic heart disease,
history of heart failure, history of cerebrovascular disease, dia-
betes mellitus, or renal insufficiency) or for patients undergoing
intermediate-risk surgery who have been identified as having
CAD or more than 1 clinical risk factor. The usefulness of
β-blocker therapy is uncertain for patients without clinical risk
factors undergoing vascular surgery and for patients undergoing
vascular surgery (or an intermediate-risk surgery) with only 1
clinical risk factor in the absence of CAD. The routine adminis-
tration of high-dose or fixed-dose β-blockers on the day of sur-
gery is not recommended on the basis of mixed study results and
recent findings of the POISE trial group. Adverse effects include
hypotension and bradycardia, which have been shown to corre-
late with increased events (perioperative death or stroke).
There is preliminary evidence that the use of statin therapy
preoperatively may decrease perioperative cardiac events, and in
small randomized trials clonidine has also been shown to reduce
perioperative ischemia and death.
Although mild or moderate hypertension usually does not
warrant delaying the operation, severe hypertension (ie, sys-
tolic pressure >180 mm Hg and diastolic pressure >110 mm Hg)
should be controlled before the operation is performed. Patients
with hypertension whose blood pressure is controlled with med-
ication usually tolerate anesthesia better than those with poorly
controlled blood pressure. The decision to delay the operation to
achieve improved blood pressure control should take into account
the urgency of the operation. Significant perioperative hyperten-
sion occurs in approximately 25% of patients with hypertension,
appears unrelated to preoperative control, and occurs frequently
in patients undergoing abdominal aortic aneurysm repair and
56 I Fundamentals of Cardiovascular Disease
other peripheral vascular procedures, including carotid endarter-
ectomy. If oral intake of antihypertensive medications must be
interrupted, parenteral therapy may be needed perioperatively.
Various antihypertensive medications can be used, including
intravenous β-blockers, vasodilators, calcium channel blockers,
and angiotensin-converting enzyme inhibitors. For patients tak-
ing clonidine orally, it may be helpful to switch to a long-act-
ing clonidine cutaneous patch preoperatively to avoid “rebound
hypertension” perioperatively.
Orthopedic hip operations in patients older than 65 are among
the most common types of noncardiac surgery. Although peri-
operative risk is increased among elderly patients undergoing
elective total hip arthroplasty, the risk is even higher among
elderly patients undergoing urgent operative repair of hip frac-
tures. Preoperative myocardial ischemia occurs in up to 35% of
elderly patients before hip fracture surgery. Early preoperative
administration of epidural analgesia may lessen overall periop-
erative risks. Appropriate analgesia is especially important in
this elderly group.
• Treatment with β-blockers should remain uninterrupted as long as
possible, especially in patients with CAD.
• β -Blocker therapy should be continued in patients undergoing sur-
gery who are already receiving β-blockers.
• β -Blockers titrated to heart rate and blood pressure are typically
recommended for patients undergoing vascular surgery or inter-
mediate-risk surgery who have been identified as having more than
1 clinical risk factor.
• The usefulness of β-blocker therapy is uncertain for patients
without clinical risk factors undergoing vascular surgery and for
patients undergoing vascular surgery (or an intermediate-risk sur-
gery) with only 1 clinical risk factor in the absence of CAD.
• Routine administration of high-dose or fixed-dose β-blockers on
the day of surgery is generally not recommended.
• Although mild or moderate hypertension usually does not warrant
delaying the operation, severe hypertension should be controlled
before the operation is performed.
Approach to Patients Requiring Long-term
Oral Anticoagulation
The issue of discontinuation of oral anticoagulation in the
perioperative setting requires balancing the thromboembolic
potential of the patient’s cardiovascular disease with the hemor-
rhagic risk of the operation. For most cardiovascular situations,
including patients with bioprosthetic valves, the acute throm-
boembolic potential is low. The thromboembolic potential is
high for patients with mechanical prosthetic valves, especially
those in the mitral or tricuspid position, and for patients with
recent embolic episodes from, for example, cardiomyopathies,
atrial fibrillation, ventricular aneurysms, or acute infarctions.
Unfortunately, there are no large randomized trials comparing
the risk of thromboembolism with the risk of hemorrhage in var-
ious conditions and types of operations.
Several small studies have suggested that in patients with low
or moderate thromboembolic risk, the warfarin dosage can be
temporarily decreased to achieve an INR less than 2.0 for 5 to 7
days with relative safety. A reasonable approach to these patients
would be to discontinue the use of warfarin several days before
the operation, which should be performed as soon as the INR
is 1.5 or less. Oral anticoagulation is resumed as soon as possi-
ble postoperatively, and heparin is reserved for patients whose
INR is less than 2.0 for 5 days or more. In patients at high risk
of thromboembolic complications (Box 4.7), intravenous heparin
Box 4.7. High-Risk Factors for Thromboembolism
Older-generation thrombogenic valve
(eg, Björk-Shiley valve)
Mechanical mitral valve replacement
Mechanical aortic valve replacement with any
risk factor or multiples of the following risk factors
in the absence of a prosthetic heart valve:
1. Atrial fibrillation
2. Left ventricular dysfunction
3. Previous thromboembolism
4. Hypercoagulable condition
coverage can be instituted until 6 hours before the operation and
then resumed as soon as possible postoperatively. The use of
heparin can be discontinued after the use of warfarin has been
resumed and the INR is in the therapeutic range.
Recent studies have shown that standardized periprocedural
use of subcutaneous LMWH is associated with a low risk of
thromboembolic and major bleeding complications. LMWH is
an alternative to unfractionated heparin for “bridging” therapy.
The use of LMWH is appealing because it can be given in an
outpatient setting. In patients who have mechanical valves and
require interruption of warfarin therapy for emergency noncar-
diac surgery, fresh frozen plasma is preferred over high-dose
vitamin K.
Five days of subtherapeutic INR for patients with low to mod-
erate thromboembolic risk is probably reasonable.
• Perioperative heparin coverage should be used in patients with a
high thromboembolic risk.
• Perioperative use of subcutaneous LMWH is an alternative to
standard anticoagulation with unfractionated heparin.
Postoperative Surveillance for
Myocardial Infarction
Postoperative surveillance for myocardial infarction in patients
is generally not recommended. Evaluation for perioperative
myocardial infarction (ECG or troponin measurements) is not
recommended for patients without CAD unless signs of cardi-
ovascular compromise develop. The use of troponin measure-
ments is not recommended for asymptomatic, hemodynamically
stable patients who have undergone low-risk surgery and is not
well established for patients having undergone vascular or inter-
mediate-risk surgery. Postoperative troponin measurements are
recommended for patients who have ECG changes or a typical
acute coronary syndrome presentation.
Examination Strategy
For the purposes of cardiology examinations, the most important
information to remember about evaluating patients with known
or suspected heart disease before noncardiac operations includes
the following:
1. Does the patient need emergency surgery?
2. Does the patient have an active cardiac condition?
3. Is the surgery low risk?
4. Does the patient have satisfactory functional capacity without
symptoms?
 4 Noncardiac Surgery in Patients With Heart Disease 57

5. Does the patient have clinical risk factors? AHA American Heart Association
6. Is medical therapy (particularly β-blockers) indicated? CAD coronary artery disease
ECG electrocardiogram
Unnecessary cardiac testing and the associated increase in FAC functional aerobic capacity
health care costs and delay in surgery may be avoided without INR international normalized ratio
increased perioperative cardiac risk to the patient by answering LMWH low-molecular-weight heparin
these questions and using a stepwise approach for each patient MET metabolic equivalent task
undergoing preoperative evaluation. NYHA New York Heart Association

Abbreviations Name of Clinical Trial

ACCF American College of Cardiology Foundation POISE Perioperative Ischemic Evaluation
 5
Essential Molecular Biology
of Cardiovascular Diseasesa
CINDY W. TOM, MD, and ROBERT D. SIMARI, MD
Basics of Genetic Structure
• Deoxyribonucleic acid specifies the amino acid sequence of
proteins.
• DNA consists of nucleotide chains, and each nucleotide consists of a
nitrogenous base (purine or pyrimidine), a pentose sugar (2-deoxyri-
bose in DNA and ribose in RNA), and a phosphate group.
• The nucleotide sequence determines the amino acids encoded. It is
the sequence-specific pairing of these nucleotides that is the basis for
inheritance of the genetic code.
transcription translation
• DNA RNA Protein.
Basics of Gene Structure
• A gene is a collection of adjacent nucleotides that specify the amino
acids of a unique polypeptide.
• A chromosome is a long thread of DNA that contains many genes.
• The human genome has 23 pairs of chromosomes (44 autosomal and
2 sex chromosomes) containing about 3 billion base pairs of DNA
and approximately 20,000 to 30,000 genes.
• The primary structure of DNA is determined by its base pair com-
position (Figure 5.1).
• Introns are DNA sequences whose RNA products are nonfunctional
and are removed from messenger RNA.
a
Portions previously published in Singh R, Pislaru SV, Simari RD.
ABCs of molecular cardiology and the impact of the Human Genome
Project on clinical cardiology. Cardiol Rev. 2002 Jan-Feb;10(1):24–33.
Used with permission.
Abbreviations and acronyms are expanded at the end of this chapter.
58
• Exons are DNA sequences whose fully functional RNA products
exit the nucleus and enter the cytoplasm, where they are translated
into specific proteins (Figure 5.2).
• All chromosomes are organized with 2 arms attached to a centromere.
• The centromere is the point where a chromosome is attached to a
spindle during cell division.
• A chromosome consists of both nucleic acids and proteins. DNA is
complexed with proteins, called histones (Figure 5.3), into nucleo-
protein fibers, called chromatin. It is chromatin that gives DNA its
beadlike appearance.
• DNA and RNA chains form in the 5' to 3' direction, and proteins
are formed in the same direction, which is from the amino to the
carboxy terminus of the polypeptide chain.
• Promoters are upstream regulatory elements that bind RNA poly-
merases and the complex of proteins that regulate transcription.
• Enhancers and suppressors are bidirectional DNA sequences that
modulate transcription of DNA by binding proteins known as tran-
scription factors.
Basics of Gene Function
• The DNA sequence within the exons of a gene and the amino acids
that it codes are colinear (Figure 5.4).
• The genetic code is a degenerate code with 64 possible codons speci-
fying only 20 amino acids.
• Transcription, the formation of RNA from DNA, is a complex pro-
cess that involves many proteins in a highly regulated fashion and is
dependent on RNA polymerases (Figure 5.4).
• A codon is 3 successive nucleotides of messenger RNA that code for
a single amino acid.
• Translation is the complex interaction between 3 types of RNA:
messenger RNA acts as a template, ribosomal RNA forms the ribo-
somes, and transfer RNA acts as a carrier of the amino acids to be
incorporated into the polypeptide.
 5 Essential Molecular Biology of Cardiovascular Diseases 59

Sugar-phosphate backbone
Phosphate
Pentose
5′

C T
A A G A Adenine base

G A
T T C Thymine base
T

3′

Figure 5.1. The Structure of DNA. C indicates cytosine; G, guanine. (Previously published. See “Credit Lines” section.)

Transcription
factors

RNA polymerase
Exon 1 Exon 2 Exon 3

DNA

Transcription
Transcription-initiation complex pre-
mRNA
5′ 3′
Transcript processing

RNA-clipping enzyme

AAUAAA

5′ cap

PolyA tail

AAAA . . .

Adenosine-adding enzyme
(terminal transferase)
Intron lariat

Splicing
AAAA . . .
Nucleus

Spliceosome

Cytoplasm Processed
transcript mRNA
AAAA . . .

Translation into protein

Figure 5.2. Gene Expression. A indicates adenine; mRNA, messenger RNA; U, uridine. (Previously published. See “Credit Lines” section.)
60 I Fundamentals of Cardiovascular Disease

• Transcription factors are specific proteins capable of binding spe-

cific DNA sequences and modulating the rate of transcription
(Figure 5.5).
H4 • Translational and posttranslational regulation varies with the
proteins studied. Some proteins (clotting factors) are made in a
H3 precursor form and activated by other proteins (posttranslational
regulation). Other gene products are regulated either solely or
partially at the transcriptional level.
H2A H2B
Tools of the Cardiovascular
Molecular Biologist
When heated, DNA can be denatured, that is, separated into 2
H3 H2A complementary strands. With cooling, these strands can be rean-
H2B
nealed. Complementary sequences of DNA from other sources,
such as short oligonucleotides, can bind in a species-specific
H4 manner to single-stranded DNA templates. This binding is
referred to as hybridization.

Figure 5.3. Relationship Between Histones 2A, 2B, 3, and 4 and
Associated DNA in a Nucleosome.
Basics of Gene Expression
• Transcriptional regulation modulates the production of
messenger RNA.
• Translational and posttranslational regulation can modulate the
production and activity of proteins that are expressed.
DNA Techniques
To facilitate handling of short segments of DNA, fragments are
often inserted into plasmids, which are circular double-stranded
DNA structures (Figure 5.6). Plasmids have their origin in bac-
teria and are often associated with the transfer of antibiotic
resistance. Plasmids can be used to carry DNA fragments for
manipulation and to express genes in vitro or in vivo.
Manipulating DNA Fragments
Portions of DNA, whether genomic or cellular, can be cut, ligated,
and extended using a set of powerful enzymes. Restriction

5′
3′

5′ 3′
U
A
C
U A mRNA
A A G A U U C
Transcribed strand T
G C
A A G Translation
T T C T A
G Codon
Nucleus U
C
A G C U A C U A C G U A
Cytoplasm
A U G A U G
U
A
Anticodon C

tRNA

Growing ALA CYS SER TYR TYR

polypeptide chain VAL

Figure 5.4. Transcription and Translation. A indicates adenine; ALA, alanine; C, cystosine; CYS, cysteine; G, guanine, SER, serine; T, Thymine;
TYR, tyrosine; U, uridine; VAL, valine. (Previously published. See “Credit Lines” section.)
 5 Essential Molecular Biology of Cardiovascular Diseases 61

A B
F

2 L
COOH

3
C H
1
Zn
C H Zn

NH2 Y COOH
NH2

C HOOC COOH
D HOOC COOH

L L Helix Helix
L L
L L Loop Loop
L L
Helix Helix

+ +
+ + + +
+ + +
+ +
+ + +
+ +
NH2 NH2 NH2 NH2

Figure 5.5. Representative Types of Transcription Factors. A, Helix-turn-helix homeodomain. B, C2H2 zinc finger. C, leucine zipper. D, Helix-
loop-helix. (Previously published. See “Credit Lines” section.)

endonucleases are bacterial proteins that are capable of cutting
DNA at specific DNA sequences (Figure 5.7). Restriction endo-
nucleases cut within DNA sequences and leave either a 3' or a 5'
overlap or blunt ends. Exonucleases are capable of cutting DNA
ends that are free in order to tailor recombination events.
Enzymes called DNA polymerases can be used to extend
DNA fragments. Ligases are capable of joining either blunt
or complementary (sticky) ends of DNA, creating a recom-
binant DNA molecule. This process is often referred to as
cloning.

Cleavage
by restriction
enzymes
Sticky ends
G AAT
TC Ligation
CT T A A
G
C
G
A AT T

CT
T A A

Genomic DNA Plasmid vector

G

Recombinant DNA
molecule

Incorporation of DNA fragment

into plasmid vector DNA molecules in bacteria

Figure 5.6. Amplification of DNA by Bacterial Cloning. A indicates adenine; C, cytosine; G, guanine, T, thymine. (Previously published. See
“Credit Lines” section.)
62 I Fundamentals of Cardiovascular Disease

EcoRI makes sticky ends RNA or DNA

–
32P-labeled
Migration
size markers
+

Electrophoresis
G A A T T C G A A T T C
C T T A A G C T T A A G
Solution passes through

Paper towels Sponge

G A A T T C
C T T A A G Gel

Salt
solution Nitrocellulose

Joining of fragments
using DNA ligase

Gel Filter
G A A T T C
C T T A A G

DNA
Figure 5.7. Restriction Endonucleases. A indicates adenine; C, cyto- transferred
sine; EcoRI, endonuclease enzyme isolated from strains of Escherichia Hybridize with unique
nucleic acid probe
coli; G, guanine; T, thymine.

Analyzing DNA Fragments

The base pair sequence of a fragment of DNA can be determined
Filter in
at a gross level or at an exact level. The use of restriction digests “Seal-a-Meal”
can be very helpful in determining the relative composition of bag
a DNA sequence. Because restriction endonucleases identify
specific DNA sequences and the distance between adjacent cuts
can be determined with agarose gel electrophoresis, nondetailed
maps of DNA fragments can be made.
Remove
unbound Probe hybridized
Sequencing probe to complementary
sequence
Sequencing is a powerful tool that enables molecular biologists
to create a detailed map containing each base within a portion
of DNA.

Southern Blotting
Hybridization of a radiolabeled probe (oligonucleotide) to a
template DNA can be used to identify specific DNA fragments Expose to
(Figure 5.8). Southern blotting is used to analyze DNA. Northern
blotting is used to analyze RNA. Western blotting is used to
analyze proteins.

Polymerase Chain Reaction

Autoradiogram
PCR takes advantage of the heat stability of the DNA polymer-
ase (Figure 5.9). The goal of PCR is to amplify a target DNA
fragment. PCR is a powerful tool for amplifying minute portions
of DNA for analysis. Because they can amplify minute quantities
of DNA, sensitive PCR tests have been developed for various Figure 5.8. Analyzing DNA and RNA by Gel Electrophoresis and
diseases that can be tested by extracting DNA from a small blood Blotting. (Previously published. See “Credit Lines” section.)
 5 Essential Molecular Biology of Cardiovascular Diseases
Cycle 1
Separation of strands
Primers
Sequence
to be
Primers Heat
Figure 5.9. Amplificaiton of DNA by the Polymerase Chain Reaction. (Previously published. See “Credit Lines” section.)
sample. Current tests for Chagas disease involve PCR methods
for the detection of parasitic DNA in collected blood samples.
PCR is also being developed for the detection of pathogens in
valve tissue samples from patients with infectious endocarditis
who have negative results of blood culture.
RNA Techniques
Unlike DNA, which is stable for relatively long periods even at
room temperature, RNA is an unstable molecule that is suscep-
tible to degradation from ubiquitous ribonucleases. Thus, great
care is required to obtain and analyze RNA from cells or tis-
sue. In addition, messenger RNA representing the genes that are
expressed in any cell makes up only a minority of the total RNA
within a cell.
Northern Blotting
As with DNA, RNA can be electrophoresed and transferred to
membranes and hybridized with labeled probes (Figure 5.8).
This process is called Northern blotting. It provides a means for
identifying and quantifying the amount of an RNA species in a
sample.
Reverse Transcriptase-Polymerase Chain Reaction
Amplification of messenger RNA can be performed using reverse
transcriptase-PCR.
Analysis of DNA-Protein Interactions
Techniques have been developed to analyze proteins that bind
in a sequence-specific manner to DNA. A popular technique to
63
Cycle 2 Cycle 3
Separation of strands
Separation of strands
Heat Heat
assess this protein-DNA interaction is the electrophoretic mobil-
ity shift assay. This assay is based on the fact that the electro-
phoretic migration of DNA fragments bound to protein is slower
(“shifted”) than that of unbound DNA. The bound protein can
be identified further by adding to the mixture antibodies spe-
cific for the proteins, resulting in further electrophoretic delay
(“supershift”).
Western Blotting
Western blotting is used to sequence proteins by gel electropho-
resis, using antibody probes analogous to Southern blotting for
DNA analysis.
Molecular Genetics: Gene Mapping
The identification of genes associated with disease has been the
focus of an enormous effort in biomedical science. Classically,
the identification of a disease-related protein led to the devel-
opment of probes, either nucleic acids or antibodies, with which
to screen DNA libraries. The screening of a genomic library
can identify the gene associated with the disease (Figure 5.10).
However, the number of diseases for which detailed biochemical
defects are known is few.
Without detailed knowledge of the disease protein, positional
cloning techniques (gene mapping) can be used to identify dis-
ease-related genes. Gene mapping is based on Mendel’s laws of
genetics, which state that genes sort independently. Thus, truly
independent genes should have a 50% chance of association after
meiotic sorting. Genes that are tightly linked on a chromosome
will have less chance to cross over during meiosis, creating an
increased chance of co-segregation (linkage) (Figure 5.11 and
Figure 5.12). Linkage analysis is based on the statistical chances
64 I Fundamentals of Cardiovascular Disease

Cleave DNA with restriction enzymes Loci Loci

far apart close together

Fragment 1 Fragment 2 Fragment 3 Fragment 4

A Aa a

A Aa a
Insert fragments into vectors B Bb b B Bb b

Introduce vectors into bacteria

A A a a

A A a a
B B b b B B b b

Plate bacteria on petri dish

Bacteria
generate
A A a a
visible
colonies
A a A a
Fix replicate library B b B b B b B b
on nylon filters

Meiosis I

Radioactive probe
A A a a
Probe hybridizes with
complementary sequence A a A a
in fragment B b B b B b B b

Recombinants
X-ray film detects Figure 5.11. Segregation of Chromosomes. Differences between
radioactive colonies
close and distant loci are depicted.
on filter

linkage (95% probability of linkage). After a linkage has been

determined, the chromosomal region can be analyzed further to
isolate the gene of interest.
Trace colonies
back to In certain situations, reasonable guesses can be made from
master plate the knowledge of a disease which would allow for a list of poten-
tial genes that might be responsible for the disease (candidate
genes). This candidate approach is improved if linkage data
Figure 5.10. Identifying a Gene in a DNA Library. (Previously pub-
are available to narrow the search to a limited list of reasonable
lished. See “Credit Lines” section.)

Functional Positional
that the genes will sort independently. For instance, genes in
disparate locations will sort independently, whereas 2 genes in
Disease Disease
proximity are less likely to sort independently.
An increasing number of chromosomal markers are available
for linkage analysis. Pedigree analysis of families is needed to Map Function Map Function
determine the patterns of coinheritance and the potential linkage
with known markers. A likelihood ratio is determined for the
Gene Gene
coinheritance. For instance, if the likelihood for cosegregation
is 1,000:1, the log of the odds score is 3 (log 1,000 = 3). A log of Figure 5.12. Functional and Positional Cloning of Disease-Related
the odds score of 3 is the lower limit for statistically significant Genes.
 5 Essential Molecular Biology of Cardiovascular Diseases
candidates. Some of the genes associated with hypertrophic
cardiomyopathy have been identified with the candidate gene
approach.
Genetic diseases can be associated with single-gene or mul-
tigene abnormalities. Single-gene defects provide a model for
understanding the mendelian inheritance of disease (Figure 5.13).
Autosomally related disorders can be dominant or recessive.
Autosomal dominant disorders require only 1 copy (of the 2)
of the mutant gene to cause the disease. In dominant disorders,
a child has a 50% chance of inheriting the gene and, thus, the
disease. Healthy siblings do not possess the mutation, but every
patient has an affected parent. Males and females are affected
equally. Penetrance and late or variable expression of the dis-
order may affect the course of the disease. Autosomal recessive
disorders require 2 mutant copies of the gene. A single copy of
A. Autosomal dominant inheritance
Multiple generations affected
Sexes affected equally frequently
In familial cases, only one parent need be affected
Male-to-male transmission occurs
Offspring of affected parent have a 50% chance
of being affected
C. Mitochondrial inheritance
Sexes equally frequently and severely affected
Transmission only through women; offspring of
affected men are unaffected
All offspring of affected women may be affected
Figure 5.13. A through D, Mendelian Inheritance Patterns. (Previously published. See “Credit Lines” section.)
65
the mutant gene results in the person being a carrier. The off-
spring of 2 carriers have a 1-in-4 chance of having the disease
and a 1-in-2 risk of being a carrier. All of the offspring of affected
individuals are carriers.
X-linked disorders result from mutant genes on the X chro-
mosome. Because males have 1 X chromosome, these disorders
act like dominant mutations. In females, they usually act like
recessive mutations. Women act as carriers, and transmission
of disease is only from females to sons. Transmission of carrier
status can come from females to daughters or from males who
give their diseased X chromosome exclusively to their daughters.
Mutations in the mitochondrial genome are passed from mother
to son. Multigenic disorders result from the interaction of mul-
tiple genes, and their inheritance pattern is difficult—if not
impossible—to discern.
B. Autosomal recessive inheritance
Single generation affected
Sexes affected equally frequently
In familial cases, only one parent need be affected
Offspring of 2 carriers: 25% affected, 50% carrier,
25% normal
Two-thirds of clinically normal offspring are
carriers
D. X-linked inheritance
No male-to-male transmission
All daughters of affected males are carriers
Sons of carrier mother have 50% chance
of being affected; daughters have 50% chance
of being carriers
66 I Fundamentals of Cardiovascular Disease

Molecular Basis of Cardiovascular Diseases Myocardial Diseases

Polymorphisms are differences in DNA sequence which are fre-
quently found in populations. Many polymorphisms are thought
to have arisen after selective pressure for alleles conferred adap-
tive advantages in different habitats. A classic example of gene
polymorphism among individuals is the variability in hepatic
metabolism of medications by cytochrome P450. Clopidogrel
is an antiplatelet agent widely used in the prevention of arte-
rial thrombosis, especially after percutaneous coronary inter-
ventions. It is a pro-drug metabolized to the active agent by the
CYP2C19 enzyme, a component of the hepatic cytochrome P450
system. It is recognized that some individuals exhibit “clopi-
dogrel resistance,” and current evidence suggests that single
nucleotide polymorphisms in the gene for CYP2C19 result in
reduced production of the active metabolite from the pro-drug.
Additionally, research into gene polymorphisms that are respon-
sible for sensitivity of adrenergic receptors to β-blockers and
lipid homeostasis has contributed to an understanding of opti-
mal medical therapy.
Mutations are changes in the DNA sequence associated with
disease. Mutations can arise from 1) point mutations, where sin-
gle nucleotide substitutions cause a single amino acid change;
2) deletion; or 3) insertion of extra DNA. Millions of single
nucleotide polymorphisms exist in the human genome, and thou-
sands of mutations have been identified for human disease.
Of the many types of diseases that affect the cardiovascu-
lar system, an increasing number have been identified as result-
ing from single-gene (monogenic) or polygenic mutations. The
cardiac diseases about which the genetic basis is known include
hypertrophic cardiomyopathy and long QT syndromes.
Hypertrophic cardiomyopathy and arrhythmogenic syndromes
are discussed in Chapter 20, HERITABLE CARDIOMYOPATHIES AND
CHANNELOPATHIES.
Dilated Cardiomyopathy
DCM is a primary disorder of the myocardium characterized by
increased left ventricular size, decreased ejection fraction, and, in
later stages, symptoms of congestive heart failure. In up to 20%
to 30% of cases, DCM may have a genetic cause. Some forms of
DCM are related to systemic diseases such as muscular dystrophy,
whereas others appear as distinct clinical entities (Table 5.1).
Duchenne muscular dystrophy and its milder form, Becker
muscular dystrophy, are associated with defects in the dystrophin
gene (Xp21); cardiac involvement is common and is manifested
as DCM and heart failure. This is an X-linked disorder. The
defective dystrophin gene results in altered expression of dys-
trophin in cardiac and skeletal muscles. In X-linked DCM, the
defect in dystrophin is limited to the myocardium and may be
associated with specific mutations within the gene. Other forms
of muscular dystrophy associated with DCM include limb-girdle
muscular dystrophy (inherited in an autosomal recessive pattern
and associated with defects in the dystrophin-associated glyco-
protein complex) and congenital muscular dystrophy (an autoso-
mal dominant disorder associated with defects in laminin).
Genetic defects in DCM not associated with muscular dys-
trophy have been identified. Two unrelated families with DCM
have been shown to have mutations in cardiac actin. Unlike
hypertrophic cardiomyopathy, which is associated with deficits

Table 5.1. Cardiomyopathy Mutants

Disease Phenotype Mutation Inheritance Chromosome Mechanism
Kearns-Sayre DCM, heart block, short stature, Mitochondrial DNA mt mt Deletions in mitochondrial
syndrome ataxia, cognitive impairment, ptosis, DNA result in impaired
weakness, deafness electron transport chain &
mitochondrial function
Duchenne muscular DCM, conduction abnormalities, distal Dystrophin XR Xp21 Loss of stabilizing
dystrophy muscle weakness, Gowers sign, glycoprotein complex
lordosis, mental retardation on muscle fiber’s plasma
membrane → muscle fiber
degradation
Fabry disease Late-onset LVH, CAD, conduction α-Galactosidase A XR Xq22 Abnormal lysosomal deposits
defects, aortic root dilatation, skin
lesion, valvulopathy, neuropathy,
TIA/CVA, proteinuria, & renal
disease
Marfan syndrome Dilatation of aorta & sinuses of Valsalva, Fibrillin 1 AD 15q21 Suspected altered calcium
aortic root dilatation/dissection, binding to growth-
scoliosis, arachnodactyly, ectopia factor–like domain
lentis, dural ectasia in protein needed
for fibrillogenesis in
connective tissues
Williams syndrome Supravalvular aortic stenosis, pulmonary Elastin ADa 7q11 Unclear; thought to be
artery stenosis, microcephaly, due to gene deletion of
micrognathi/“elfin facies,” mental the transcription factor
retardation, hypercalcemia, renal involved
artery stenosis, hypertension
Abbreviations: AD, autosomal dominant; CAD, coronary artery disease; CVA, cerebrovascular accident; DCM, dilated cardiomyopathy; LVH, left ventricular
hypertrophy; mt, mitochondrial; TIA, transient ischemic attack; XR, X-linked recessive.
a
Sporadic new mutations.
 5 Essential Molecular Biology of Cardiovascular Diseases 67

in proteins associated with force generation, it has been hypoth-
esized that defects in actin may be associated with deficits in
force transmission.
Mitochondria contain a small amount of DNA that codes for 13
genes involved in mitochondrial metabolism. The Kearns-Sayre
syndrome is an example of a mitochondrial cardiomyopathy due
to a mitochondrial DNA deletion with maternal inheritance (all
children of affected mothers will have the mutation).
Congenital Heart Disease
Two common examples of the known molecular basis for congen-
ital heart disease are Marfan syndrome and supravalvular aortic
stenosis. Marfan syndrome is an autosomal dominant disease
with high penetrance, characterized by skeletal, cardiovascu-
lar, and ocular abnormalities. Genetic linkage analysis led to the
identification of fibrillin as the abnormal protein responsible for
this disease. Thirty mutations have been found in the fibrillin
gene on chromosome 15 associated with Marfan syndrome.
Linkage analysis has determined that alterations in the elastin
gene on chromosome 7 are responsible for supravalvular aortic
stenosis. Like Marfan syndrome, this disorder is inherited in an
autosomal dominant fashion with high penetrance. Supravalvular
aortic stenosis in combination with mental retardation, connec-
tive tissue abnormalities, and hypercalcemia is called Williams
syndrome.
Several other single-gene defects are associated with congen-
ital heart disease, including familial atrial septal defects, Holt-
Oram syndrome on chromosome 12, and the atrioventricular
canal defects associated with Down syndrome. In general, the
genetic abnormalities associated with congenital heart disease
may result in several clinical presentations.
Risk Factors for Atherosclerosis
Atherosclerosis is a complex disease associated with acquired
and genetic factors. The genetic background is clearly polygenic
and associated with a complex interplay with environmental
factors. Of the known risk factors for atherosclerosis, 2 pro-
vide models for a genetic understanding: hyperlipidemia and
hyperhomocysteinemia.
Hyperlipidemia
One of the most widely understood cardiovascular genetic disor-
ders is associated with familial hypercholesterolemia (Table 5.2).
Familial hypercholesterolemia is a relatively common cause of
increased levels of LDL and is associated with a defective LDL
receptor gene on chromosome 19. Familial hypercholesterolemia
is an autosomal dominant disorder in which the heterozygotes
are affected to an intermediate degree. Homozygotes have LDL
cholesterol levels of 650 to 1,000 mg/dL and are often affected by
coronary atherosclerosis before the age of 10 years. The homo-
zygote phenotype is notable for the presence of planar cutaneous
xanthomas. The heterozygotes have LDL levels that are twice
normal, and they are at high risk of premature coronary artery
disease. At least 150 different mutations of the LDL receptor
gene are associated with familial hypercholesterolemia.
The genetic determinants for other abnormalities of lipo-
protein metabolism are being studied. The clinical clues to a
potential genetic cause for lipoprotein disorders include the
presence of premature atherosclerosis in the patient or a fi rst-
degree relative, the presence of xanthomas, and extremely
high levels of total cholesterol (>300 mg/dL) or triglycerides
(>500 mg/dL).
Hyperhomocysteinemia
Both genetic and nutritional deficiencies can lead to hyperhomo-
cysteinemia. Classic hyperhomocysteinemia-homocystinuria is
caused by defects in the cystathione β-synthetase gene (chromo-
some 21q22) and is inherited in an autosomal recessive pattern.
Homozygotes have a marfanoid appearance and develop prema-
ture vascular disease. Heterozygotes, who appear healthy, may

Table 5.2. Genetic Causes of Hypercholesterolemia and Hyperhomocysteinemia

Disease Phenotype Mutation Inheritance Chromosome Mechanism
Cholesterol mutants
FH Xanthomas, xanthelasma, LDLR AD 19p13 LDL receptor dysfunction
aortic stenosis in 50% of results in failure to take up
homozygotes cholesterol from plasma
Sitosterolemia Tendon xanthomas, premature ABCG5 or ABCG8 AR 2p21 ATP-binding cassette
atherosclerosis
Familial defective apo Similar to FH Apo B-100 (ligand AD 2p24–23 Impaired binding of LDL
B-100 on LDL particle) to the LDL (apo B/E)
receptor
Cerebrotendinous Ataxia, Achilles tendon Mitochondrial AR 2q33 Bile acid synthesis blocked;
xanthomatosis xanthomata, premature 27-hydroxylase ? Cholestanol causing
atherosclerosis, cataracts (CYP27) neurologic toxicity
Hyperhomocysteinemia Prothrombotic in ACS
Most common form Homozygotes (TT): marfanoid 5,10-methylene- AR 21q22 Enzymatic activity of
from T mutation appearance, premature tetrahydrofolate MTHFR reduced (T
CAD/ cerebrovascular reductase (MTHFR) mutation) from Ala-to-Val
disease substitution causing ↑
Heterozygotes (T): normal homocysteine, especially
appearance, may have with low folate
vascular disease
Abbreviations: ACS, acute coronary syndrome; AD, autosomal dominant; apo, apolipoprotein; AR, autosomal recessive; ATP, adenosine triphosphate; CAD,
coronary artery disease; FH, familial hypercholesterolemia; LDL, low-density lipoprotein.
68 I Fundamentals of Cardiovascular Disease
have an increased risk of vascular disease. Defects in the gene
for 5,10-methylenetetrahydrofolate reductase are also associated
with hyperhomocysteinemia and premature vascular disease.
Treatment with folate and vitamins B6 and B12 can decrease
homocysteine levels in these patients. Fortification of flour has
decreased the incidence of hyperhomocystinemia.
Potential Molecular-Based Therapies for
Cardiovascular Disease
Genetically based treatments can be divided into 2 categories:
those dependent on genetic technology for development and
those that use genetic material as drugs.
Recombinant Approaches to Drug Development
Recombinant DNA technology has played an important part in
the development of revolutionary drugs based on naturally occur-
ring enzymes. An important example of this is recombinant tis-
sue plasminogen activator (r-tPA) (alteplase). The use of alteplase
as a fibrinolytic agent to treat myocardial infarction helped usher
in a new age of rapid reperfusion therapy and has been shown in
large international trials to decrease mortality.
Recombinant DNA technology can also be used to gen-
erate antibody fragments that can bind to and block certain
molecular targets. Abciximab (ReoPro), which is made of the
antigen-binding fragment (Fab fragment) of a chimeric antibody
to the glycoprotein IIb/IIIa receptor on platelets, has proved
useful in treating high-risk patients undergoing percutaneous
transluminal coronary angioplasty. Other monoclonal antibodies
are also under clinical trial investigation for use as adjunctive
therapy for acute myocardial infarction, including the anti–C5
complement antibody pexelizumab (many monoclonal antibody
generic names have the suffix −mab).
Gene Therapies for Cardiovascular Disease
Gene transfer is the modulation of foreign or native gene expres-
sion by the introduction of new genetic material into a cell or
organism. Gene therapy strategies for cardiovascular disease
have been limited to somatic cells (non-germline cells). Genes
are introduced into cells through the use of vectors, which can
contain elements that are both viral and nonviral.
The initial demonstration of vascular gene transfer used
plasmid DNA and retroviral vectors to deliver reporter genes
to the vasculature. These studies were limited by the low effi-
ciency of gene transfer because retroviral vectors are capable of
infecting only dividing cells, which are rarely present in normal
arteries.
The development of adenoviral vectors led to the ability to
demonstrate potential therapeutic benefits from vascular gene
transfer. Adenoviruses are DNA viruses capable of infecting
dividing and nondividing cells. The ability to clone transgenes
into replication-deficient adenoviral vectors has resulted in more
efficient (yet transient) transgene expression in vascular tissue.
A prime target for gene transfer has been restenosis after per-
cutaneous transluminal coronary angioplasty. Restenosis occurs
as a result of cellular proliferation, matrix production, thrombo-
sis, and chronic renarrowing. Gene transfer approaches have tar-
geted cellular proliferation as a means to limit restenosis. Gene
transfer strategies aimed at killing proliferating cells include the
use of the herpesvirus thymidine kinase gene (tk) and the pro-
drug ganciclovir. The tk gene product sensitizes infected cells to
the killing effects of ganciclovir.
Another major target for cardiovascular gene transfer has
been the development of strategies to create angiogenesis within
ischemic tissue by delivering genes encoding for growth factors.
Genes for vascular endothelial growth factor and members of the
fibroblast growth factor family have been used for this purpose.
Vascular endothelial growth factor and fibroblast growth fac-
tor are highly potent secreted peptides; as such, their genes can
be delivered to relatively few cells within the vessel, resulting
in locally increased concentrations of protein. This angiogenic
strategy is currently being tested in clinical studies.
Abbreviations
DCM dilated cardiomyopathy
LDL low-density lipoprotein
PCR polymerase chain reaction
 Section II

Noninvasive Imaging
This page intentionally left blank
 6
Restrictions on Drivers and Aircraft
Pilots With Cardiac Diseasea
STEPHEN L. KOPECKY, MD
This chapter is divided into 3 sections:
1. Restrictions for patients operating a private motor vehicle:
These restrictions are from the National Highway Traffic Safety
Administration. There are no clear-cut statutory rules for who
can or cannot drive. The recommendations are designed to aid
the physician when performing a medical evaluation and making
recommendations.
2. Restrictions for patients operating a CMV: Drivers of these vehi-
cles must be certified as fit-to-drive by a medical practitioner,
although there are no statutory rules to follow. The guidelines are
recommendations.
3. Restrictions for patients operating an aircraft: The FAA requires all
pilots to obtain an airman medical certificate. The ultimate decision
to grant a license to an individual pilot is made by the FAA.
Noncommercial License
Table 6.1 provides a reference list of medical conditions that may
impair driving skills and the recommendations for each condition.
The corresponding recommendations are based on scientific evi-
dence whenever possible, but the use of these recommendations
has not been proven to reduce crash risk. (Although scientific evi-
dence links certain medical conditions and levels of functional
impairment with crash risk, more research is needed to estab-
lish that driving restrictions based on these medical conditions
a medically fit-to-drive by a medical practitioner. Although med-
Recommendations drawn from guidelines of the US Department of
Transportation Federal Motor Carrier Safety Administration (http://
www.fmcsa.dot.gov), the National Highway Traffic and Safety
Administration (http://www.nhtsa.gov), and the National Transportation
Safety Board (http://www.ntsb.gov).
Abbreviations and acronyms are expanded at the end of this chapter.
and levels of functional impairment significantly reduce crash
risk.) These recommendations are provided to assist physicians
in the decision-making process and are not intended to be for-
mal practice guidelines nor substitutes for the physician’s clinical
judgment.
In the inpatient setting, whenever appropriate, driving
should be addressed before the patient is discharged. Even for
patients whose symptoms clearly preclude driving, it should not
be assumed that they are aware that they should not drive. The
physician should counsel the patient about driving and discuss a
future plan (eg, resumption of driving after symptoms resolve,
driver rehabilitation after symptoms stabilize, permanent driving
cessation).
For patients with cardiac conditions, the main consideration
in determining medical fitness to drive is the risk of presyncope
or syncope due to a bradyarrhythmia or tachyarrhythmia. For the
patient with a known arrhythmia, the physician should identify
and treat the underlying cause of arrhythmia, if possible, and rec-
ommend temporary driving cessation until symptoms have been
controlled.
Certification of Commercial Drivers
A CMV is a motor vehicle used in commerce to transport pas-
sengers or property. Drivers of these vehicles must be certified as
ical examiners are not required to have specific training or to
demonstrate any special competence to medically certify CMV
drivers, examiners are expected to exercise good medical judg-
ment during the evaluation, and they could face litigation in the
case of an undesirable outcome.
71
72
Table 6.1. Driving Restrictions and Waiting Periods Before Patients With Cardiovascular Disease Can Resume Driving
Condition
Angina pectoris
Stable
Unstable
Angioplasty
Myocardial infarction (acute)
Uncomplicated
Complicated (eg, arrhythmia, congestive
heart failure, dizziness, recurrent
myocardial infarction)
Cardiac surgery involving median
sternotomy
Atrial fibrillation or flutter with
bradycardia or rapid ventricular
response
High-grade AV or intraventricular block
Isolated block
LBBB, bifascicular block, Mobitz type I AV
block, first-degree AV block
Mobitz type II AV block, trifascicular block,
acquired third-degree AV block
Congenital third-degree AV block
PSVT (including WPW syndrome),
brief nonsustained paroxysmal
VT, paroxysmal supraventricular
tachycardia, paroxysmal atrial
fibrillation or flutter
Sick sinus syndrome, sinus bradycardia,
sinus exit block, sinus arrest
Prolonged, nonsustained VT
Sustained VT, ventricular fibrillation,
cardiac arrest, ICDs
Cardiac arrest
Pacemaker (insertion or revision)
II Noninvasive Imaging
Restrictions and Waiting Periodsa
No additional restrictions; no waiting period
Patients should not drive if they experience symptoms at rest or at the wheel
Patients may usually resume driving 2–7 d after successful percutaneous coronary intervention if they are
clinically stable and asymptomatic
Waiting period of 2 wk before resuming driving
Waiting period of 1 mo and complete assessment by a cardiologist before resuming driving
Patients may usually resume driving 4 wk after coronary artery bypass graft or valve replacement surgery
and 8 wk after heart transplant, depending on the resolution of cardiac symptoms and the course of
recovery
In the absence of surgical and postsurgical complications, the main limitation to driving is the risk of
sternal disruption after median sternotomy
For patients undergoing a coronary artery graft operation with the minimally invasive surgery technique,
the waiting period may be considerably shorter
No further restrictions after heart rate and symptoms have been controlled
Patients should not drive after an acute episode that causes dizziness or syncope until the condition is
stabilized
For symptomatic block corrected without a pacemaker (eg, by stopping the use of medications that caused
the block), the patient may resume driving when asymptomatic for 4 wk and electrocardiographic
documentation shows resolution of the block
No restriction
No restrictions if there are no associated signs of cerebral ischemia
Should not drive unless satisfactorily treated
No restrictions if there are no associated signs of cerebral ischemia
No restrictions if the patient is asymptomatic during documented episodes
Patients with a history of symptomatic tachycardia may resume driving after they have been asymptomatic
for 6 mo during antiarrhythmic therapy
Patients who undergo radiofrequency ablation may resume driving after 6 mo if there is no recurrence of
symptoms or sooner if no preexcitation or arrhythmias are induced at subsequent EP testing
No restrictions with the following:
1. No associated signs of cerebral ischemia and no underlying heart disease
2. Ventricular preexcitation and no associated cerebral ischemia
3. Satisfactory control with resolved signs of cerebral ischemia
4. Satisfactory control with underlying heart disease
No restrictions if the patient is asymptomatic
Regular medical follow-up is recommended to monitor progression
No restrictions if the patient is asymptomatic during documented episodes
Patients with symptomatic VT may resume driving after 3 mo if they are receiving antiarrhythmic therapy
(with or without an ICD) guided by EP testing and if VT is noninducible at subsequent EP testing
Patients may resume driving after 6 mo if they are receiving empirical antiarrhymic therapy (with or
without an ICD) without arrhythmic events or if they have an ICD without additional antiarrhythmic
therapy
Patients may resume driving after 3 mo if they are receiving antiarrhythmic therapy (with or without an
ICD) guided by EP testing and if VT is noninducible at subsequent EP testing
Patients may resume driving after 6 mo without arrhythmic events if they are receiving empirical
antiarrhythmic therapy (with or without an ICD) or if they have an ICD without additional
antiarrhythmic therapy
For long-distance or sustained high-speed travel, patients should be encouraged to have an adult
companion drive
Patients should avoid the use of cruise control
After recovery, patients need a certificate from an appropriate specialist before they resume driving; the
underlying cause should have been treated and the other relevant criteria in this table met
The patient may resume driving after 1 wk if all the following are met:
1. The patient no longer experiences presyncope or syncope
2. The electrocardiogram shows normal sensing and capture
3. The pacemaker performs within the manufacturer’s specifications
(continued)
 6 Restrictions on Drivers and Aircraft Pilots With Cardiac Disease 73

Table 6.1. (Continued)

Condition Restrictions and Waiting Periodsa

Hypertrophic cardiomyopathy Patients who experience syncope or presyncope should not drive until they have been treated
Congenital heart disease Assessment should be based on the presence or absence of myocardial ischemia, left ventricular
dysfunction, valvular lesions, or disturbances of cardiac rhythm and on the relevant guidelines in this
table
Congestive heart failure Physicians should reassess patients for driving fitness every 6 mo to 2 y as needed, depending on clinical
course and control of symptoms
Patients with NYHA functional class III (marked limitation of activity but no symptoms at rest; working
capacity of 2–4 METs) should be reassessed at least every 6 mo
Patients have no restrictions if they meet the following criteria:
1. NYHA functional class I (no functional limitations and able to achieve 7 METs without symptoms or
objective evidence of cardiac dysfunction)
2. NYHA functional class II (mild functional limitations and able to achieve 7 METs)
3. NYHA functional class III (moderate limitations, working capacity <2 METS, and symptoms
at rest) if no signs of cerebral ischemia (eg, dizziness, palpitations, lightheadedness, and loss of
consciousness) or dyspnea
Patients should not drive if they are in NYHA functional class IV (severe impairment, working capacity
<2 METs, and symptoms at rest)
Heart transplant Waiting period of 2 mo before resuming driving
Waiting period may be shortened at the discretion of the specialist
Annual reassessment recommended
Hypotension Not a contraindication to driving unless it has caused episodes of loss of consciousness (syncope) or
confusion caused by cerebral ischemia or hypoperfusion
If cerebral ischemia or hypoperfusion has occurred, the patient should discontinue driving
If treatment can prevent further attacks, the patient may resume driving
Hypertension No driving restrictions for any type of hypertension other than uncontrolled malignant hypertension
Valvular heart disease
Medically treated or untreated No restrictions if there is no associated cerebral ischemia
Patients who experience syncope or presyncope should not drive until they have been treated (especially if
they have aortic stenosis)
Surgically treated (eg, mechanical prostheses, Waiting period of 6 wk
mitral bioprostheses, or valvuloplasty with No restrictions if no thromboembolic complications or symptoms of cerebral ischemia
nonsinus rhythm)
Abbreviations: AV, atrioventricular; EP, electrophysiologic; ICD, implantable cardioverter-defibrillator; LBBB, left bundle branch block; MET, metabolic
equivalent task; NYHA, New York Heart Association; PSVT, paroxysmal supraventricular tachycardia; VT, ventricular tachycardia; WPW, Wolff-Parkinson-
White syndrome.
a
The recommendations are general guidelines that all drivers with cardiovascular disease should satisfy. These are federal restrictions—private and commercial
drivers may have additional restrictions imposed by individual states.

With regard to cardiovascular disease, the current federal reg-
ulations state that a person is physically qualified to drive a CMV
if that person meets either of the following criteria:
1. Has no current clinical diagnosis of myocardial infarction, angina
pectoris, coronary insufficiency, or thrombosis.
2. Has no other cardiovascular disease of a variety known to be accom-
panied by syncope, dyspnea, collapse, or congestive heart failure.
Because the CMV license does not provide the opportunity
for the examiner to restrict work activity, the commercial driver
must be able to perform very heavy work in order to be certified.
For commercial drivers, an exercise treadmill test requires exer-
cising to a workload capacity of at least 6 metabolic equivalent
tasks (through stage II of a Bruce protocol), attaining a heart
rate greater than 85% of the predicted maximum (unless the
patient is receiving β-blocker therapy), and having an increase
in the systolic blood pressure of more than 20 mm Hg without
angina, without significant ST-segment depression or elevation,
and without significant ventricular dysrhythmia. Stress radio-
nuclide or echocardiographic imaging should be performed for
symptomatic individuals, individuals with an abnormal resting
electrocardiogram, patients receiving digoxin, and drivers who
do not meet the minimal requirements of the standard exercise
treadmill test.
Driving ability may be compromised with any cardiovascu-
lar condition associated with cerebral ischemia, such as cardiac
arrhythmias, cardiac pacemakers or ICDs, hypertrophic cardio-
myopathy, congestive heart failure, and valvular heart disease.
Table 6.2 is a compendium of conditions with specific guidelines
for medical approval or disapproval for CMV certification.
FAA Medical Certification
Heart disease is the number one cause of medical certificate
denial for aircraft pilots of all classes. The FAA requires all pilots
of fixed- or rotor-wing aircraft that exceed certain weight limi-
tations to obtain an airman medical certificate. There are more
than 5,000 private AMEs around the world who are designated
to provide medical applications, give forensic examinations, and
issue FAA medical certificates to qualified airmen. There are 3
standards of medical fitness (examinations for classes I, II, and
III), each designed for the type of flying:
1. Class I medical certificate: An examination is required for air-
line captains of scheduled air carriers and is valid for 6 calendar
months.
2. Class II medical certificate: Required for cocaptains (first officers)
and other professional pilots (eg, agricultural spray pilots).
3. Class III medical certificate: Required for recreational pilots.
74
Table 6.2. Compendium of Conditions and Guidelines for Medical Approval or Disapproval for Commercial Motor Vehicle (CMV)
Certification
Condition
Without known CAD
Essential hypertension
Stage 1 (systolic BP 140–159 mm Hg
or diastolic BP 90–99 mm Hg)
Stage 2 (BP 160–179/100–109 mm Hg)
Stage 3 (BP >180/110 mm Hg)
Angina pectoris
MI
PCI
CABG surgery
Supraventricular tachycardias
Atrial fibrillation or flutter
Multifocal atrial tachycardia
AVNRT, WPW syndrome, atrial
tachycardia, junctional tachycardia
Ventricular arrhythmias
II Noninvasive Imaging
Guidelines Medical Approvala,b
Asymptomatic, high-risk person (CAD risk-equivalent condition defined as presence of Yes
diabetes mellitus, peripheral vascular disease, or Framingham risk score predicting a
20% CAD event risk during the next 10 y)
Asymptomatic high-risk person older than 45 y with multiple risk factors for CAD with No
abnormal ETT
May be certified to drive for 1 y; annual examination BP ≤140/90 mm Hg Yes
(if <160/100 mm Hg, certification may be extended 1 time for 3 mo)
May have a 1-time certification for 3 mo; at recheck, if BP ≤140/90 mm Hg, certification Yes
may be extended 1 y from initial examination; annual BP ≤140/90 mm Hg
This is immediately disqualifying No
If at recheck, BP ≤140/90 mm Hg, and treatment is well tolerated, may be certified for Yes
6 mo from initial examination; evaluate every 6 mo; BP ≤140/90 mm Hg
Should have biennial ETT at minimum
Rest angina or change in anginal pattern within 3 mo of examination; abnormal ETT; No
ischemic changes on resting ECG; intolerance to cardiovascular therapy
Should have biennial ETT at minimum (if test is positive or inconclusive, imaging stress
test may be indicated)
At least 2 mo after MI; approved by cardiologist; no angina; post-MI LVEF ≥40% (by Yes
echocardiogram or ventriculogram); tolerance to current cardiovascular medications
Recurrent anginal symptoms; post-MI LVEF <40% (by echocardiogram or No
ventriculogram); abnormal ETT demonstrated before planned return to work;
ischemic changes on resting ECG; poor tolerance to current cardiovascular
medications
At least 1 wk after procedure; asymptomatic; no injury to the vascular access site; Yes
approved by cardiologist; tolerance to medications
Incomplete healing or complication at vascular access site; rest angina; ischemic ECG No
changes
After uncomplicated, elective PCI to treat stable angina, a commercial driver may return
to work as soon as 1 wk after the procedure
Criteria for continuing to work after PCI:
1. ETT 3–6 mo after PCI; then annual medical qualification examination
2. Commercial driver should have negative ETT at least every other year (criteria
above)
3. Tolerance of all cardiovascular medications
4. Driver should not experience orthostatic symptoms, including symptomatic light-
headedness, a resting systolic BP <95 mm Hg, or a systolic BP decrease >20 mm
Hg upon standing
Asymptomatic; ≥3 mo after CABG surgery; tolerance to medications without Yes
orthostatic symptoms; LVEF >40% on resting echocardiogram at the first qualifying
examination after CABG surgery (a documented report of an echocardiogram
performed in-hospital after CABG surgery is equally sufficient); approval by
cardiologist
LV dysfunction (LVEF <40%) No
After CABG surgery, a patient should have annual medical evaluation; after 5 y, an
annual ETT (with imaging if indicated by standard criteria)
Acceptable exercise capacity; the maximal heart rate achieved is >85% of the age-
predicted maximum (unless the patient is receiving β-blockers), no ischemic signs or
symptoms, a workload of ≥6 METs, appropriate systolic BP and heart rate response,
and no ventricular dysrhythmias
The examiner should have a low threshold for requiring stress imaging studies instead
of a standard ETT
May be certified if asymptomatic and anticoagulated adequately for ≥1 mo; Yes
anticoagulation monitored by at least monthly INR; and rate or rhythm control
deemed adequate
Symptomatic No
Symptomatic No
WPW with atrial fibrillation No
CAD and sustained VT with poor prognosis and high risk No
NSVT with LVEF ≥40% and symptoms of cerebral hypoperfusion No
Dilated cardiomyopathy with NSVT (LVEF ≤40%) or sustained VT (any LVEF) No
or syncope or near syncope (any LVEF)
(continued)
 6 Restrictions on Drivers and Aircraft Pilots With Cardiac Disease
Table 6.2. (Continued)
Condition
PPM
ICD
Hypertrophic cardiomyopathy
Idiopathic dilated cardiomyopathy
and congestive heart failure
Restrictive cardiomyopathy
Mitral stenosis
Repair or replacement of any valve
Mitral regurgitation
Aortic stenosis
Aortic regurgitation
Mechanical valves
Congenital heart disease
75
Guidelines Medical Approvala,b
Long QT syndrome, Brugada syndrome, or any cardiovascular diagnosis with No
symptomatic VT
Any indication for PPM with symptoms (without PPM implantation) No
Neurocardiogenic syncope and hypersensitive carotid sinus with syncope No
If 3 mo after PPM implantation with documented correct function by pacemaker center Yes
and absence of symptom recurrence
In the United States, >1 million people have PPMs; when assessing the risk for
sudden, unexpected incapacitation of a patient with a PPM, the underlying disease
responsible for the PPM indication must be considered
If the patient has no symptoms for 3 mo after implantation and correct function of the Yes
PPM has been documented
If patients have neurocardiogenic syncope with both vasodepressor and negative Yes
chronotropic components and maintaining the heart rate can attenuate the
vasodepressor aspect and if asymptomatic
ICDs do not prevent arrhythmias; patients who have had a cardiac arrest and who have No
received an ICD No
Patients who have received an ICD for primary prevention (at risk of an arrest due to
electrophysiologic testing or other evaluation), the risk of loss of consciousness is
considerable
No
Symptomatic congestive heart failure No
Ventricular arrhythmias, LVEF ≤50%, and asymptomatic No
Ventricular arrhythmias, LVEF <40%, and asymptomatic No
No ventricular arrhythmias, LVEF 40%-50%, and asymptomatic Yes
No
Moderate and asymptomatic with annual evaluation and echocardiography Yes
Severe (mitral valve area ≤1.0 cm 2, New York Heart Association class II or higher, atrial No
fibrillation, pulmonary artery pressure ≥50% of systemic pressure, and inability to
exercise for >6 METs on Bruce protocol (stage II)
≥4 wk after percutaneous balloon mitral valvotomy or ≥3 mo after surgical Yes
commissurotomy
Moderate and asymptomatic, with normal LV size and function and normal pulmonary Yes
artery pressure on annual echocardiogram
Severe and asymptomatic with echocardiogram every 6–12 mo Yes
Severe and symptomatic with inability to achieve >6 METs on Bruce No
protocol, ruptured chordae or flail leaflet, atrial fibrillation, LV dysfunction
(LVEF <60%), thromboembolism or pulmonary artery pressure >50% of systolic
arterial pressure
Moderate (aortic valve area ≥1.0–1.5 cm 2) and asymptomatic, with echocardiogram Yes
every 1–2 y
Moderate with angina, heart failure, syncope, atrial fibrillation, LV dysfunction with No
LVEF <50%, or thromboembolism
Severe (aortic valve area <1.0 cm 2) (irrespective of symptoms or LV function) No
Severe and asymptomatic, with normal LV function (LVEF ≥50%), mild LV dilatation Yes
(LVEDD <60 mm and LVESD <50 mm), and echocardiogram every 6–12 mo
Severe and symptomatic; unable to complete Bruce protocol stage II, with LVEF No
<50%, and LVEDD >70 mm or LVESD >55 mm
Symptomatic, LV dysfunction (LVEF <40%), thromboembolic complication after No
the procedure, pulmonary hypertension >50% of systemic, and unable to maintain
adequate anticoagulation (based on monthly INR checks)
Patients with milder forms of congenital heart disease or in whom the condition has Yes
spontaneously resolved (eg, spontaneous closure of a ventricular septal defect) or
who have had surgical repair of a malformation
Waiting period of ≥3 mo after cardiac surgery is recommended for clinical and
hemodynamic evaluation of results; evaluation by a cardiologist should be performed
every 1–2 y, owing to the risk of subsequent cardiovascular complications
Bicuspid aortic valve with aortic transverse diameter >5.5 cm No
Marfan syndrome with any aortic root enlargement, more than moderate aortic No
valve regurgitation, more than mild mitral valve regurgitation related to mitral
valve prolapse, or LV dysfunction (without associated valvular lesion) with
LVEF <40%
Atrial septal defect that is asymptomatic (ie, no history of paradoxical embolus) Yes
Atrial septal defect—ostium secundum with symptoms of dyspnea, palpitation, or No
paradoxical embolus; pulmonary artery pressure >50% systemic; right-to-left shunt;
or pulmonary-to-systemic flow ratio >1.5:1
(continued)
76 II Noninvasive Imaging

Table 6.2. (Continued)

Condition Guidelines Medical Approvala,b

Ventricular septal defect—small with symptoms of dyspnea, palpitations, or syncope; No

pulmonary artery pressure >50% systemic; right-to-left shunt; LV enlargement or
reduced function; or pulmonary-to-systemic flow ratio >1.5:1
Ventricular septal defect—moderate size (risk of sudden incapacitation from a No
paradoxical embolism or progressive pulmonary hypertension)
Certification 3 mo after surgical closure if absence of the disqualifying criteria outlined Yes
above, QRS complex <120 ms on ECG, and no serious dysrhythmia on 24-h
ambulatory ECG
If the right ventricular conduction is >120 ms on ECG, the driver may still be Yes
certified if invasive His bundle studies show no infra-His block or other serious
electrophysiologic disorders indicating a high risk of incapacitation, as determined
by a cardiologist with expertise in cardiac electrophysiology
Pulmonary hypertension that is >50% systemic from any cause No
Abbreviations: AVNRT, atrioventricular nodal reentrant tachycardia; BP, blood pressure; CABG, coronary artery bypass graft; CAD, coronary artery disease;
ECG, electrocardiogram; ETT, exercise treadmill test; ICD, implantable cardioverter-defibrillator; INR, international normalized ratio; LV, left ventricular;
LVEDD, left ventricular end-diastolic diameter; LVEF, left ventricular ejection fraction; LVESD, left ventricular end-systolic diameter; MI, myocardial
infarction, MET, metabolic equivalent task; NSVT, nonsustained ventricular tachycardia; PCI, percutaneous coronary intervention; PPM, permanent pacemaker;
VT, ventricular tachycardia; WPW, Wolff-Parkinson-White.
a
For medical approval for a CMV license (“Yes”), all criteria must be present. For disapproval (“No”), only 1 criterion must be present.
b
Restrictions are national. Individual states may impose additional restrictions on private and commercial drivers.

The cardiovascular evaluation of a pilot with underlying car- after a percutaneous coronary intervention; class I pilots must
diac conditions may be done by a cardiologist in conjunction undergo coronary angiography to show patency, but class II
with an AME, although ultimate certification involves review and class III pilots may undergo noninvasive imaging to show
by the FAA. Absolute disqualifications are persistent angina absence of ischemia. After insertion of a permanent pacemaker,
pectoris, more than 1 replaced heart valve (except for the Ross pilots must be in stable condition for 6 weeks before forensic
procedure), ICD implantation, and heart transplant. If the sys- testing may ensue. Heart valve replacement cases are considered
tolic blood pressure is greater than 155 mm Hg, or if the diastolic by the Aeromedical Certification Division of the FAA on a case-
blood pressure is greater than 95 mm Hg, a pilot is grounded by-case basis. Currently, without direct approval from the FAA,
until the blood pressure is under control and the FAA is satisfied AMEs cannot approve certification of cardiovascular conditions
that no serious underlying cardiovascular disease is the cause. that would be otherwise disqualifying.
Through a waiver system, the FAA may grant an Authorization
for Special Issuance of a Medical Certificate and allow a pilot to
Abbreviations
fly if medical stability can be established.
After a myocardial infarction or any invasive intervention, AME aviation medical examiner
pilots have a mandatory 6-month stand-down time to establish CMV commercial motor vehicle
condition stability before forensic testing may be performed for FAA Federal Aviation Administration
certification purposes. All pilots must undergo testing 6 months ICD implantable cardioverter-defibrillator

Principles of Echocardiography
TERESA S. M. TSANG, MD
This chapter summarizes the central role of echocardiography
in both the initial diagnosis and the quantification of the nature
and severity of specific cardiovascular diseases. It is important
to appreciate the relative strengths, weaknesses, and incremen-
tal value of information obtained by different echocardiographic
methods. The appropriateness criteria for use of transthoracic
echocardiography and TEE were published jointly in 2007 by the
American College of Cardiology Foundation Quality Strategic
Directions Committee Appropriateness Criteria Working Group,
American Society of Echocardiography, American College
of Emergency Physicians, American Society of Nuclear Car-
diology, Society for Cardiovascular Angiography and Inter-
ventions, Society of Cardiovascular Computed Tomography, and
the Society for Cardiovascular Magnetic Resonance. Box 7.1 lists
appropriate, uncertain, and inappropriate indications for the use
of echocardiography.
Transthoracic Echocardiography
Anatomical and functional assessment of the cardiac chambers,
valves, myocardium, pericardium, atrial septum, inferior vena
cava, and aorta are important aspects of the basic echocardio-
graphic examination.
M-Mode and 2-Dimensional Echocardiography
M-mode imaging, which dates from the early days of echocar-
diography, is still a useful part of a complete ultrasonographic
examination and can be acquired with 2D guidance. The typical
views obtained with the transducer placed at the left parasternal
Abbreviations and acronyms are expanded at the end of this chapter.
7
region, sweeping from the ventricular level to the mitral valve
level to the aortic valve level, are shown in Figure 7.1.
Measurements of the LV dimensions and wall thickness
can be readily derived from M-mode recordings and are usu-
ally made according to the recommendations of the American
Society of Echocardiography at end diastole (the onset of the
QRS complex) and end systole (the point of maximal upward
motion of the LV posterior wall endocardium). These measure-
ments are made from leading edge to leading edge. LV ejection
fraction can be readily calculated from measurements obtained
by M-mode or 2D assessments (see “Assessment of Ventricular
Function” section in this chapter).
Use of 2D imaging provides important structural and func-
tional information on cardiac disease. The American Society
of Echocardiography has recommended that cardiac imaging
be performed in 3 orthogonal planes: long-axis (from the aor-
tic root to the apex), short-axis (perpendicular to the long axis),
and 4-chamber (traversing both ventricles and atria through the
mitral and tricuspid valves). Long-axis and short-axis refer to
axes of the heart not the body. The 3 planes can be visualized
with 4 standard transducer positions: parasternal, apical, sub-
costal, and suprasternal. The views obtained are depicted in
Figure 7.2.
Doppler Echocardiography
Doppler echocardiography uses the Doppler effect, that is, the
change in the frequency of sound waves as the sound source
moves toward or away from the observer (Equation 7.1).
This Doppler frequency shift is detected and translated into
a blood flow velocity (Equation 7.2) by the Doppler transducer
and instrument.
77
78 II Noninvasive Imaging

Box 7.1. Indications for Use of Echocardiography

Appropriate indications
1. Symptoms potentially due to suspected cardiac etiology, including but not limited to dyspnea, shortness of breath, light-
headedness, syncope, transient ischemic attack, cerebrovascular events
2. Prior testing that is concerning for heart disease (ie, chest radiograph, baseline scout images for stress echocardiogram,
ECG, elevation of serum brain natriuretic peptide)
3. Assessment of known or suspected adult congenital heart disease, including anomalies of great vessels and cardiac
chambers and valves, or suspected intracardiac shunt (ASD, VSD, PDA) either in patient not operated on or in patient
after repair or operation
4. Patients who have sustained or nonsustained SVT or VT
5. Initial evaluation of LV function following acute MI
6. Reevaluation of LV function following MI during recovery phase when results will guide therapy
7. Evaluation of known or suspected pulmonary hypertension, including evaluation of right ventricular function and
estimated pulmonary artery pressure
8. Evaluation of hypotension or hemodynamic instability of uncertain or suspected cardiac etiology
9. Evaluation of acute chest pain with suspected myocardial ischemia in patients with nondiagnostic laboratory markers
and ECG and in whom a resting echocardiogram can be performed during pain
10. Evaluation of suspected complication of myocardial ischemia or MI, including but not limited to acute mitral
regurgitation, hypoxemia, abnormal chest radiograph, VSD, free-wall rupture or tamponade, shock, right ventricular
involvement, heart failure, or thrombus
11. Evaluation of respiratory failure with suspected cardiac etiology
12. Evaluation of a patient with known or suspected acute pulmonary embolism to guide therapy (ie, thrombectomy and
thrombolytics)
13. Initial evaluation of a murmur in a patient for whom there is a reasonable suspicion of valvular or structural heart disease
14. Initial evaluation of a patient with suspected mitral valve prolapse
15. Initial evaluation of known or suspected native valve stenosis
16. Routine (yearly) evaluation of an asymptomatic patient with severe native valve stenosis
17. Reevaluation of a patient with native valve stenosis who has had a change in clinical status
18. Initial evaluation of known or suspected native valve regurgitation
19. Routine (yearly) reevaluation of an asymptomatic patient with severe native valve regurgitation and no change in
clinical status
20. Reevaluation of native valve regurgitation in a patient with a change in clinical status
21. Initial evaluation of a prosthetic valve for establishing a baseline after placement
22. Reevaluation of a patient with a prosthetic valve and suspected dysfunction or thrombosis or a change in clinical status
23. Initial evaluation of suspected infective endocarditis (native or prosthetic valve) with positive blood cultures or a new
murmur
24. Reevaluation of infective endocarditis in patients with any of the following: virulent organism, severe hemodynamic
lesion, aortic involvement, persistent bacteremia, a change in clinical status, or symptomatic deterioration
25. Evaluation for a cardiovascular source of an embolic event (PFO or ASD, thrombus, neoplasm)
26. Evaluation of a cardiac mass (suspected tumor or thrombus)
27. Evaluation of pericardial conditions, including but not limited to pericardial mass, effusion, constrictive pericarditis,
effusive-constrictive conditions, patients after cardiac surgery, or suspected pericardial tamponade
28. In known or suspected Marfan disease, evaluation of proximal aortic root or mitral valve (or both)
29. Initial evaluation of suspected hypertensive heart disease
30. Initial evaluation of known or suspected heart failure (systolic or diastolic)
31. Reevaluation of known heart failure (systolic or diastolic) to guide therapy in a patient with a change in clinical status
32. Evaluation for dyssynchrony in a patient being considered for CRT
33. Patient with a known implanted pacing device and symptoms possibly due to suboptimal pacing device settings to
reevaluate for dyssynchrony or revision of pacing device settings (or both)
34. Initial evaluation of known or suspected hypertrophic cardiomyopathy
35. Reevaluation of known hypertrophic cardiomyopathy in a patient with a change in clinical status to guide or evaluate
therapy
36. Evaluation of suspected restrictive, infiltrative, or genetic cardiomyopathy
37. Screening study for structure and function in first-degree relatives of patients with inherited cardiomyopathy
38. Baseline and serial reevaluations of patients undergoing therapy with cardiotoxic agents
39. Evaluation of suspected acute aortic pathology, including dissection or transsection
40. Guidance during percutaneous noncoronary cardiac interventions, including but not limited to septal ablation
in patients with hypertrophic cardiomyopathy, mitral valvuloplasty, PFO or ASD closure, or radiofrequency ablation
41. To determine mechanism of regurgitation and determine suitability of valve repair
42. To diagnose or manage endocarditis with a moderate or high pre-test probability (eg, bacteremia, especially
staphylococcal bacteremia or fungemia)
43. Persistent fever in a patient with an intracardiac device
44. Evaluation of a patient with atrial fibrillation or atrial flutter to facilitate clinical decisions about anticoagulation,
cardioversion, and radiofrequency ablation

(continued)
 7 Principles of Echocardiography 79

Box 7.1. (Continued)

Uncertain indication for echocardiography
1. Evaluation for a cardiovascular source of an embolic event in a patient who has a normal TTE and normal ECG and no
history of atrial fibrillation or atrial flutter
Inappropriate indications for echocardiography
1. Routine (yearly) reevaluation of asymptomatic patients with corrected ASD, VSD, or PDA >1 y after successful correction
2. Patients who have isolated APC or PVC without other evidence of heart disease
3. Evaluation of LV function with prior ventricular function evaluation within the past year with normal function (eg, prior
echocardiogram, left ventriculogram, SPECT, cardiac MRI) in patients in whom there has been no change in clinical
status
4. Initial evaluation of a patient with suspected pulmonary embolism to establish a diagnosis
5. Routine (yearly) reevaluation of mitral valve prolapse in patients with no or mild mitral regurgitation and no change in
clinical status
6. Routine (yearly) reevaluation of an asymptomatic patient with mild native aortic stenosis or mild to moderate native
mitral stenosis and no change in clinical status
7. Routine (yearly) reevaluation of native valvular regurgitation in an asymptomatic patient with mild regurgitation, no
change in clinical status, and normal LV size
8. Routine (yearly) reevaluation of a patient with a prosthetic valve in whom there is no suspicion of valvular dysfunction
and no change in clinical status
9. Evaluation of native or prosthetic valves in patients with transient fever but without evidence of bacteremia or new
murmur
10. Routine evaluation of patients with systemic hypertension without suspected hypertensive heart disease
11. Reevaluation of a patient with known hypertensive heart disease without a change in clinical status
12. Routine (yearly) reevaluation of patients with heart failure (systolic or diastolic) in whom there is no change in clinical
status
13. Routine (yearly) reevaluation of hypertrophic cardiomyopathy in a patient with no change in clinical status
14. Evaluation of a patient with atrial fibrillation or atrial flutter for left atrial thrombus or spontaneous contrast when a
decision has been made to use anticoagulation and not to perform cardioversion

Abbreviations: APC, atrial premature contraction; ASD, atrial septal defect; CRT, cardiac resynchronization therapy; ECG, electrocardiogram; LV,
left ventricular; MI, myocardial infarction; MRI, magnetic resonance imaging; PDA, patent ductus arteriosus; PFO, patent foramen ovale; PVC,
premature ventricular contraction; SVT, supraventricular tachycardia; SPECT, single-photon emission computed tomography; TTE, transthoracic
echocardiography; VSD, ventricular septal defect; VT, ventricular tachycardia.
(Previously published. See “Credit Lines” section.)

A B

Figure 7.1. A, An M-mode cursor is placed along different levels (1, ventricular level; 2, mitral valve level; 3, aortic valve level) of the heart,
with parasternal long-axis 2-dimensional echocardiographic guidance. B through D, Representative normal M-mode echocardiograms at the mid-
ventricular, mitral valve, and aortic valve levels, respectively. Arrows in B indicate end-diastolic (EDd) and end-systolic (ESd) dimensions of the left
ventricle. C, The M-mode echocardiogram of the anterior mitral leaflet; A, peak of late opening with atrial systole; C, closure of the mitral valve; D,
end systole before mitral valve opening; E, peak of early opening; F, mid-diastolic closure. The double-headed arrow in D indicates the dimension
of the left atrium at end systole. Ao indicates aorta; AV, aortic valve; LA, left atrium; LV, left ventricle; PW, posterior wall; RVOT, right ventricular
outflow tract; VS, ventricular septum. (Previously published. See “Credit Lines” section.)
80 II Noninvasive Imaging

Figure 7.1. (Continued)

A B

Figure 7.2. A, Drawings of the longitudinal views from the 4 standard transthoracic transducer positions. Shown are the parasternal long-axis
view (1), parasternal right ventricular (RV) inflow view (2), apical 4-chamber view (3), apical 5-chamber view (4), apical 2-chamber view (5), subcos-
tal 4-chamber view (6), subcostal long-axis (5-chamber) view (7), and suprasternal notch view (8). B, Drawings of short-axis views. These views are
obtained by rotating the transducer 90° clockwise from the longitudinal position. Drawings 1 through 6 show parasternal short-axis views at different
levels by angulating the transducer from a superior medial position (for the imaging of the aortic and pulmonary valves) to an inferolateral position,
tilting toward the apex (from level 1 to level 6 short-axis views). Shown are short-axis views of the right ventricular outflow (RVO) tract and pulmo-
nary valve (1), aortic valve and left atrium (LA) (2), and RVO tract (3), and short-axis views at the left ventricular (LV) basal (mitral valve [MV])
level (4), the LV midlevel (papillary muscle) (5), and the LV apical level (6). A good view to visualize the RVO tract is the subcostal short-axis view
(7). Also shown is the suprasternal notch short-axis view of the aorta (Ao) (8). RA indicates right atrium; RPA, right pulmonary artery. (Previously
published. See “Credit Lines” section.)
 7 Principles of Echocardiography 81

Equation 7.1. Frequency Shift (Δf)

2 ft v cos θ
Δf =
c
Δf = Doppler frequency shift
ft = transmitted frequency
cos θ = (cosine of angle theta) cosine of angle between the
vector of the moving object and the interrogating beam
c = (constant), velocity of sound in tissue or water
(1,560 m/s)
v = velocity of the moving object

Equation 7.2. Velocity (v), m/s

Δfffc
v=
2 ft cos θ

(Definitions as in Equation 7.1)

The velocity of blood can be used to determine gradients,

intracardiac pressures, volumetric flow, and valve areas.
Pulsed wave Doppler echocardiography and continuous wave
Doppler echocardiography are the 2 most commonly used spec-
tral Doppler modalities (Box 7.2). Pulsed wave Doppler echocar-
diography is “site specific,” allowing the measurement of blood
velocities at a particular region of interest. The disadvantage is
aliasing of the signal when velocities reach one-half of the pulse
repetition frequency, or the Nyquist limit (Figure 7.3). This
property limits the maximal velocity that can be measured with
pulsed wave Doppler echocardiography.
Continuous wave Doppler echocardiography measures all
velocities in the path of the ultrasound beam, is not site specific,
and is not limited by aliasing. The disadvantage is that although
very high velocities can be recorded, the specific anatomical site
where the highest velocity is present cannot be accurately local-
ized (but it can be inferred). Continuous wave Doppler echocar-
diography is typically used to measure high-velocity jets and
gradients.
Color flow imaging is computer-enhanced pulsed wave
Doppler echocardiography that displays the velocity and direc-
tional information of blood flow. Red depicts blood flow toward
the transducer and blue, away from the transducer. Color flow
Box 7.2. Appropriate Use of Pulsed Wave and
Continuous Wave Doppler Echocardiography
Pulsed wave
Flow volume
Diastolic filling variables
Pulmonary or hepatic vein flow
Localization of site of flow disturbance
Continuous wave
Valvular and other stenotic gradients
Intracardiac pressure
Mitral regurgitant velocities
Pressure half-time measurements
Intracavitary gradients
Figure 7.3. Pulsed wave and continuous wave Doppler spectra
from a patient with aortic stenosis (AS) and aortic regurgitation (AR).
The pulsed wave sample volume is in the left ventricular outflow tract
(LVOT) and demonstrates aliasing and “wrapping around” the baseline
of the high-velocity AR signal. The continuous wave signal displays the
entire AS and AR signals. (Previously published. See “Credit Lines”
section.)
imaging, like pulsed wave Doppler imaging, has a Nyquist limit
and displays aliasing. Color flow imaging is used to detect, local-
ize, and semiquantify abnormal flow, such as that resulting from
valvular regurgitation, shunts, or intracavitary obstruction.
Tissue Doppler imaging is now an integral part of routine
electrocardiography and is most commonly used for assessment
of mitral annular motion, which is part of a comprehensive dias-
tolic function assessment (see “Diastolic Function Assessment”
section). The velocity of the mitral annulus motion represents
the velocity of changes in the LV long-axis dimensions. The
diastolic velocity has been considered a measure of the intrin-
sic speed of myocardial relaxation. Early diastolic velocity is
recorded at the septal or lateral mitral annulus with a pulse wave
technique and a 1.5-mm sample volume. The E/e′ ratio provides
an excellent assessment of LV diastolic filling pressures in sinus
rhythm and in atrial fibrillation. Cutoff values differ among ech-
ocardiographic laboratories, depending on specificities and sen-
sitivities chosen and on whether septal or lateral e′ is used. In
the laboratory at Mayo Clinic, septal e′ is preferentially used,
and E/e′ is considered elevated if it is more than 15 and low
or normal if it is less than 8. However, between 8 and 15, there
is considerable variability in filling pressures. Many laborato-
ries, however, prefer to use the average of septal e′ and lateral e′.
When the average is used, 8 or less is a good indicator for low or
normal filling pressure, and 13 or more suggests elevated filling
pressure.
82 II Noninvasive Imaging
Contrast Echocardiography
Identification of intracardiac shunts is one of the most frequent
indications for contrast echocardiography, and agitated saline
solution is the most commonly used contrast agent. Saline bub-
bles do not cross the pulmonary vascular bed, and this precludes
opacification of left-sided chambers without an intracardiac
shunt.
In recent years, stabilized solutions of microbubbles have been
developed that can traverse the pulmonary capillary bed in high
concentration after intravenous injection. These microbubble
agents produce high-intensity signals not only within the LV but
also within the myocardium. Contrast agent facilitates the iden-
tification of the endomyocardial border and is most often used in
stress echocardiography when visualization of the endocardium
is essential for assessment of ischemia.
Second harmonic imaging enhances the ultrasonic backscat-
ter from contrast microbubbles (which resonate in an ultrasonic
field) while decreasing the returning signal from myocardium
(which does not resonate).
Assessment of Chamber Size and Wall Thickness
The American Society of Echocardiography has recently pub-
lished updated guidelines for cardiac chamber quantification.
LV Size
It is recommended that LVEDD, end-systolic diameter, and
wall thicknesses be measured at the level of the LV minor axis,
approximately at the mitral valve leaflet tips (Figure 7.1A). These
linear measurements can be made directly from 2D images
(Figure 7.1A) or by using 2D-targeted M-mode echocardiography
(Figure 7.1B). It is not always possible to align the M-mode cursor
perpendicularly to the long axis of the ventricle, a requirement
that is critical for measurement of a true minor-axis dimension.
As an alternative, chamber dimension and wall thicknesses can
be acquired from the parasternal short-axis view with direct 2D
measurements or targeted M-mode echocardiography, provided
that the M-mode cursor can be positioned perpendicularly to the
septum and LV posterior wall. As a general guideline, the upper
limit of a normal LVEDD is approximately 5.5 cm, but it varies
according to body surface area. LV enlargement is an important
finding, especially in patients with valvular regurgitant lesions,
hypertension, cardiomyopathy, and LV remodeling after myocar-
dial infarction. Thus, accurate measurement of LV diameter with
serial echocardiographic monitoring is important for many clini-
cal diagnoses. LV size is often interpreted as LVEDD.
LV Wall Thickness
LV wall thickness is routinely measured in a standard echocar-
diographic study. LV SWT and PWT are measured at end-dias-
tole from 2D or M-mode recordings routinely. The measurements
of the septal and posterior walls are obtained at the same level of
the ventricle as the LV diameter (Figure 7.1B). LV mass can then
be calculated from Equation 7.3.
Equation 7.3. Left Ventricular (LV) Mass
LV mass = 0.8 × {1.04 [(LVEDD + PWTd + SWTd)3 −
(LVEDD)3]} + 0.6 g
LVEDD = left ventricular end-diastolic diameter
PWTd = end-diastolic posterior wall thickness
SWTd = end-diastolic septal wall thickness
Left Atrial Size
Traditionally, 1-dimensional M-mode echocardiography has
been used for assessment of LA size (Figure 7.1D). Recently, LA
volume indexed to body surface area has been shown to be more
accurate. The upper limit of the normal range for indexed LA vol-
ume is 28 mL/m2 (29–33 mL/m2 is considered mildly enlarged,
34–39 mL/m2 is moderately enlarged, and ≥40 mL/m2 is severely
enlarged). The biplane area-length method (Figure 7.4) and
biplane Simpson summation of discs method (Figure 7.5) have
been considered valid for assessment of LA volume.
Right Ventricular Size, Right Atrial Size, and Right
Ventricular Wall Thickness
Right ventricular and right atrial sizes are usually assessed
qualitatively in most clinical laboratories. This is particularly
important in patients with pulmonary hypertension, pulmonary
diseases, and tricuspid or pulmonary valve lesions. Abnormalities
may also reflect the severity of left heart disease. Some guidelines
with respect to assessment and interpretation of the right ventric-
ular and right atrial sizes have been included in the most recent
recommendations for chamber quantification by the American
Society of Echocardiography.
L4C L2C
A4C A2C
8 × A4C × A2C
LA VolumeAL =
3πL
0.85 × A4C × A2C
=
L
Figure 7.4. Biplane Area-Length Method. The difference between
the 4-chamber (4C) and 2-chamber (2C) lengths should be no more than
5 mm; otherwise, foreshortening should be considered. In the formula,
L is the average of the 2 lengths. (Note: the smaller of the 2 lengths has
been used for L, which is acceptable if the difference between the 2 is ≤5
mm). The arrow in the electrocardiographic tracing indicates the stage
of the cardiac cycle represented by the drawings. A indicates area; AL,
area-length; L, length; LA, left atrial.
 7 Principles of Echocardiography 83

Equation 7.5. Use of 2-Dimensional–Directed M-Mode

Echocardiography to Calculate Ejection Fraction (EF), %

LVEDD2 LVESD2
EF = × 100
LVEDD2

LVEDD = left ventricular end-diastolic diameter

LVESD = left ventricular end-systolic diameter

Equation 7.6. Fractional Shortening (FS), %

LVEDD LVESD
FS = × 100
LVEDD

LVEDD = left ventricular end-diastolic diameter

LVESD = left ventricular end-systolic diameter
Cardiac output can be derived from 2D volumes or from use
of Doppler echocardiographic techniques (Equations 7.7.10).
LA VolumeModified = π 20 L
Simpson
∑ ai × bi ×
4 i=1 20 Equation 7.7. Use of 2-Dimensional Volumes to Calculate
Figure 7.5. Biplane Simpson Summation of Discs Method. The
Stroke Volume (SV), mL
length (L) is the dimension perpendicular to the discs, from the plane
of the mitral annulus to the superior aspect of the left atrium (LA). The SV = LVEDV − LVESV
arrow in the electrocardiographic tracing indicates the stage of the car-
(Definitions as in Equation 7.4)
diac cycle represented by the drawings. ai indicates area by integration
along the chord in the 4-chamber view; bi, area by integration along the
chord in the 2-chamber view. Equation 7.8. Use of TVI to Calculate Stroke Volume (SV), mL

SV = Area × TVI
Right ventricular free wall thickness (normally <0.5 cm) is
measured with either M-mode or 2D imaging. Although right Area = πr (ie, the cross-sectional area [cm2] through which
2

ventricular free wall thickness can be assessed from the apical velocity is recorded)
and parasternal long-axis views, the subcostal view at the level
2
of the tricuspid valve chordae tendineae, measured at the peak ⎛ d⎞
r2 = π = 0 785d 2
of the R wave, provides less variation and closely correlates with ⎝ 2⎠
right ventricular peak systolic pressure.
d = diameter
Assessment of Ventricular Function r = radius
(Systolic, Diastolic, Global, and Regional)
TVI = time-velocity integral = stroke distance (cm), which is the
Systolic Function Assessment distance over which blood travels in 1 cardiac cycle (the cycle
LV global systolic function can be evaluated with several echocar- velocity [cm/s] divided by time [s])
diographic techniques. These include 2D volumes derived from
measurements of 2- and 4-chamber areas and lengths (area-
Equation 7.9. Cardiac Output (CO), L/min
length method) and the modified Simpson method (or summa-
tion of discs). Formulas for calculating LV ejection fraction from CO = Stroke Volume × Heart Rate
2D volumes (Equation 7.4) or 2D-directed M-mode echocar-
diography (Equation 7.5) and for calculating fractional shorten-
ing (Equation 7.6) are shown below. Equation 7.10. Cardiac Index (CI), L/min per m2

Cardiac Output
Equation 7.4. Use of 2-Dimensional Volumes to Calculate CI =
Ejection Fraction (EF), % Body Surfacee Area

LVEDV LVESV Regional LV function is based on the 2D assessment of the

EF = × 100
LVEDV
LVEDV = left ventricular end-diastolic volume
LVESV = left ventricular end-systolic volume
(This formula can be applied to any contracting cavity; volumes are
measured by the modified Simpson method with online software.)
contractility of 16 LV wall segments (6 segments at the base and
mid-ventricle and 4 at the apical level) (Figure 7.6). A numerical
score is given to each segment depending on contractility: 1, nor-
mal; 2, hypokinetic; 3, akinetic; 4, dyskinetic; and 5, aneurysm.
A wall motion score index can then be derived (Equation 7.11).
Studies have demonstrated adverse prognostic significance from
high wall motion scores.
84 II Noninvasive Imaging

Parasternal Short-Axis Apical 4-Chamber Parasternal Long-Axis

Basal

Mid

Apical Apical 2-Chamber Apical Long-Axis

Figure 7.6. Schema of the 16 left ventricular wall segments used to assess regional systolic function and wall motion score index. A indicates
anterior; AL, anterolateral; Ao, aorta; AS, anteroseptal; I, inferior; IL, inferolateral; IS, inferoseptal; L, lateral; LA, left atrium; LV, left ventricle; P,
posterior; PL, posterolateral; PS, posteroseptal; RA, right atrium; RV, right ventricle; S, septal.

Equation 7.11. Wall Motion Score Index (WMSI) Diastolic Function Assessment
Mitral inflow assessment is fundamental to the evaluation of dias-
Sum off Wall Scores
WMSI = tolic function. Mitral E and A velocities, deceleration time, and
No. of Segments Visualized IVRT are measured (Figure 7.7). In general, 3 abnormal patterns
are recognized: impaired relaxation (grade 1 diastolic dysfunc-
Scoring of segmental contraction tion), pseudonormal filling (grade 2 diastolic dysfunction), and
1 = normal (includes hyperdynamic walls) restrictive filling (grade 3 [reversible] and grade 4 [irreversible]
diastolic dysfunction) (Figure 7.8). At Mayo Clinic, abnormal
2 = hypokinetic relaxation with elevated filling pressures (grade 1A) is distin-
guished from that without elevated filling pressures (grade 1).
3 = akinetic Mitral inflow patterns change with loading conditions.
4 = dyskinetic Therefore, other assessments are also necessary to provide
a more comprehensive evaluation, especially to distinguish
5 = aneurysm pseudonormal from normal. The Valsalva maneuver can be

Relaxation Abnormality Restrictive

Figure 7.7. Schematic left ventricular (LV), aortic (Ao), and left atrial (LA) pressure tracings and corresponding mitral inflow Doppler spectra.
A indicates atrial contraction; DT, deceleration time; E, early filling phase; IVRT, isovolumic relation time. (Previously published. See “Credit Lines”
section.)
 7 Principles of Echocardiography 85

Abnormal
Relaxation Pseudonormal Restrictive
Normal (Grade 1) (Grade 2) (Grade 3-4)
QRS
T P
ECG
E wave
A wave

Mitral Inflow
Diastole
Systole

Pulmonary Veins
Atrial reversal
Tissue Doppler

Vp

Color M-mode

Best Good Bad Worst

Figure 7.8. Diastolic Function Continuum. Diastolic function assessment with mitral inflow, pulmonary venous flow, tissue Doppler imaging, and
color M-mode. In the color M-mode images, the white line indicates the slope of the first aliasing velocity during early filling, measured from the
mitral valve plane to 4 cm distally into the left ventricular cavity (flow propagation velocity [Vp]; Vp > 50 cm/s is considered normal). A indicates late
mitral inflow velocity; E, early mitral inflow velocity; ECG, electrocardiogram.

used to decrease preload and unmask the seemingly normal pat-
tern of pseudonormal filling to reveal a pattern characteristic
of relaxation abnormality. The pulmonary venous flow pattern,
tissue Doppler mitral annular velocity profile (Figure 7.9), LA
volume indexed to body surface area, and color M-mode prop-
agation velocity all contribute to the assessment and grading of
diastolic function and filling pressures (Figure 7.8). A compre-
hensive guideline for the grading of LV diastolic function and
assessment of filling pressure has recently been proposed by the
American Society of Echocardiography. The guidelines include
algorithms for diastolic function grading (Figure 7.10), estima-
tion of LV filling pressure in patients with depressed ejection
fraction (Figure 7.11), and estimation of LV filling pressure in
patients with normal ejection fraction (Figure 7.12).
Hemodynamic Assessment
The following is a list of commonly used echocardiographic
hemodynamic variables and their clinical usefulness:
1. Pressure gradients (maximal instantaneous and mean)—valvular
stenosis, prosthetic valve, left and right ventricular outflow tract
obstruction, and coarctation of aorta
2. Intracardiac pressures—right ventricular, pulmonary artery, and LV
systolic and end-diastolic pressures
3. Volumetric flow—stroke volume, cardiac output, regurgitant volume
and fraction, and, less commonly, shunt fraction (Qp/Qs)
4. Valve areas—continuity equation and pressure half-time
5. Diastolic filling variables
To make these measurements, it is essential to understand
and use the modified Bernoulli equation (Equation 7.12 and
Figure 7.13), in which the decrease in pressure across a stenosis
is equal to 4v2, and the concept of the TVI or “stroke distance”
(Figure 7.14).
Equation 7.12. Gradient (ΔP), mm Hg
ΔP = 4(v22 − v12)
or
ΔP = 4v2
P = pressure
v2 = accelerated velocity across a stenosis
v1 = velocity proximal to a stenosis
Note: Normally v1 is much smaller than v2 and can usually be
omitted. Therefore, the equation can be simplified to 4v2.
v = velocity across any vessel, chamber, or valve
When comparing Doppler-derived gradients with those mea-
sured invasively, it is important to remember that the maximal
instantaneous gradient measured by Doppler is not equal to the
peak-to-peak gradient measured at catheterization (Figure 7.15).
The maximal instantaneous gradient is always higher than the
“nonphysiologic” (ie, nonsimultaneous) peak-to-peak gradient.
Doppler- and catheter-derived mean pressure gradients are
comparable.
With the modified Bernoulli equation (Equation 7.12) and the
measured Doppler velocity of a regurgitant or restrictive flow
 Mitral
Flow

Mitral
Annulus
Velocity

Relaxation Pseudo- Restrictive

Normal Abnormality normalization Physiology

Figure 7.9. Patterns of Mitral Inflow and Mitral Annulus Velocity From Normal to Restrictive Physiology. The mitral annulus velocity was
obtained from the septal side of the mitral annulus with tissue Doppler imaging. Each calibration mark in the recording of mitral annulus velocity
represents 5 cm/s. Early diastolic annulus velocity (e′) is greater than late diastolic annulus velocity (a′) in a normal pattern. In all other patterns,
e′ is not greater than a′. In relaxation abnormality, e′ and a′ parallel early (E) and late (A) velocities of mitral inflow. However, when filling pressure is
increased (pseudonormalization and restrictive physiology), e′ remains decreased (ie, persistent underlying relaxation abnormality) while mitral inflow
E velocity increases. Hence, E/e′ may be useful in estimating left ventricular filling pressure. (Previously published. See “Credit Lines” section.)

Septal e´
Lateral e´
LA volume

Septal e´ ≥8 cm/s Septal e´ ≥8 cm/s Septal e´ <8 cm/s

Lateral e´ ≥10 cm/s Lateral e´ ≥10 cm/s Lateral e´ <10 cm/s
LA <34 mL/m2 LA ≥34 mL/m2 LA ≥34 mL/m2

E/A <0.8 E/A 0.8-1.5 E/A ≥2

DT >200 ms DT 160-200 ms DT <160 ms
Av E/e´ ≤8 Av E/e´ 9-12 Av E/e´ ≥13
Ar − A <0 ms Ar − A ≥30 ms Ar − A ≥30 ms
Val Δ E/A <0.5 Val Δ E/A ≥0.5 Val Δ E/A ≥0.5

Normal function,
Normal
athlete’s heart, Grade I Grade II Grade III
function
or constriction

Figure 7.10. Practical Approach to Grading Diastolic Dysfunction. A indicates late diastolic mitral inflow velocity; Ar, peak atrial reversal veloc-
ity; Av, average; DT, deceleration time; E, early diastolic mitral inflow velocity; e′, early diastolic velocity of the mitral annulus; LA, left atrial; Val Δ,
change with Valsalva maneuver. (Previously published. See “Credit Lines” section.)
86
 Mitral E/A

E/A <1 and E/A ≥1 to <2, or E/A ≥2,

E ≤50 cm/s E/A <1 and E >50 cm/s DT <150 ms

E/e´ (average e´) <8 E/e´ (average e´) ≥15

E/Vp <1.4 E/Vp ≥2.5
S/D >1 S/D <1
Ar − A <0 ms Ar − A ≥30 ms
Val Δ E/A <0.5 Val Δ E/A ≥0.5
PAS <30 mm Hg PAS ≥35 mm Hg
IVRT/TE − e´ >2 IVRT/TE − e´ <2

Normal LAP Normal LAP ↑ LAP ↑ LAP

Figure 7.11. Estimation of Left Ventricular Filling Pressure in Patients With Depressed Ejection Fraction. A indicates late diastolic mitral inflow veloc-
ity; Ar, peak atrial reversal velocity; Val Δ, change with Valsalva maneuver; D, peak anterograde diastolic velocity; DT, deceleration time; E, early diastolic
mitral inflow velocity; e′, early diastolic velocity of mitral annulus; IVRT, isovolumetric relaxation time; LAP, left atrial pressure; PAS, pulmonary artery
systolic pressure; S, peak systolic velocity; T, time interval; Vp, flow propagation velocity. (Previously published. See “Credit Lines” section.)

E/e´

Sep E/e´ ≥15 or

E/e´ ≤8
E/e´ 9-14 Lat E/e´ ≥12 or
(Sep, Lat, or Av)
Av E/e´ ≥13

LA volume <34 mL/m2 LA volume ≥34 mL/m2

Ar − A <0 ms Ar − A ≥30 ms
Val Δ E/A <0.5 Val Δ E/A ≥0.5
PAS <30 mm Hg PAS >35 mm Hg
IVRT/TE − e´ >2 IVRT/TE − e´ <2

Normal LAP Normal LAP ↑ LAP ↑ LAP

Figure 7.12. Estimation of Left Ventricular Filling Pressure in Patients With Normal Ejection Fraction. A indicates late diastolic mitral inflow
velocity; Ar, peak atrial reversal velocity; Av, average; Val Δ, change with Valsalva maneuver; E, early diastolic mitral inflow velocity; e′, early dia-
stolic velocity of the mitral annulus; IVRT, isovolumetric relaxation time; LA, left atrial; LAP, left atrial pressure; Lat, lateral; PAS, pulmonary artery
systolic pressure; Sep, septal; T, time interval. (Previously published. See “Credit Lines” section.)
87
88 II Noninvasive Imaging

P1 − P2= Maximal
instantaneous Peak-to-peak
gradient gradient
2
dν
½ ρ (ν2 − ν1 )
2 2
+ ρ ds + R (ν) Ao
dt
1

Convective Flow Viscous

acceleration acceleration friction
LV

4ν2

Figure 7.13. Derivation of the modified Bernoulli equation, which

measures the pressure difference (P1 − P2) across a restrictive orifice.
In most clinical situations, the viscous friction and flow acceleration
components are negligible and can be ignored. If the proximal velocity
(v1) is very small compared with the distal velocity (v2), as in severe aor- Doppler peak
tic stenosis, the proximal velocity term can be omitted, resulting in the velocity
simplified equation ΔP = 4v2. R indicates a constant of viscous friction;
ρ, density of blood (1.06 × 103 kg/m3). Figure 7.15. Schema of left ventricular (LV) and aortic (Ao) pressure
tracings and the corresponding Doppler velocity spectrum demonstrat-
ing the difference between peak-to-peak and maximal instantaneous
gradients. The mean gradient (blue area) is the area under the curve
jet, the pressure difference between the 2 chambers can be cal- of the Doppler spectrum and is closely correlated with the mean gradi-
culated. If the pressure in 1 of the chambers can be measured ent measured invasively (red area). (Previously published. See “Credit
accurately or estimated noninvasively, the pressure in the other Lines” section.)
chamber can be derived as shown in the following examples:
1. RV or PA Systolic Pressure = 4(TR Systolic Velocity)2 + RA Pressure
motion, or the clinical examination findings are consistent with
2. PA Diastolic Pressure = 4(PR End-Diastolic Velocity)2 + RA Pressure
3. LA Pressure = Systolic BP − 4(MR Systolic Velocity) 2
a marked increase in central venous pressure, 20 mm Hg or
4. RV Systolic Pressure = Systolic BP − 4(VSD Velocity) 2 more is added to the pressure difference measured by Doppler
echocardiography.
where BP = blood pressure, LA = left atrium, MR = mitral regur- Regurgitant volume (Equation 7.13), regurgitant fraction
gitation, PA = pulmonary artery, PR = pulmonary regurgitation, (Equation 7.14), and Qp/Qs (Equation 7.15) are obtained by com-
RA = right atrium, RV = right ventricle, TR = tricuspid regurgi- paring the flow through a nonregurgitant reference valve with the
tation, and VSD = ventricular septal defect. flow through the affected valve or chamber.
Right atrial pressure can be estimated by any 1 or a combina-
tion of techniques, including clinical estimate of central venous
pressure, nomograms derived from Doppler catheter correlation Equation 7.13. Regurgitant Volume, mL
studies, and echocardiographic estimates based on right atrial
and inferior vena caval size and inferior vena caval reactivity Regurgitant Volume = SVvalve − SVsystemic
to inspiratory effort. In practice, if the right atrium and infe- SV = stroke volume
rior vena cava appear normal, 5 mm Hg is used for right atrial
pressure estimates. If the inferior vena cava is mildly dilated or SVsystemic = systemic flow measured elsewhere in an unaffected
has blunted inspiratory collapse, 10 to 14 mm Hg is assumed. area of the heart (area × TVI)
If the inferior vena cava is plethoric, has little or no inspiratory SVvalve = flow volume (area × TVI) across the regurgitant valve
(forward plus regurgitant flow)
TVI = time-velocity integral
oke
Str ume Equation 7.14. Regurgitant Fraction, %
TVI Area vol L)
× = A (m Vvalve − SV
SV Vsystemic
(cm) (cm2) Regurgitant Fraction =
SV
Vvalve

(Definitions as in Equation 7.13)

Figure 7.14. The time-velocity integral (TVI) is the calculated area
under the Doppler spectrum over time. It is also known as “stroke dis-
tance” because it represents the distance in centimeters (cm) that blood Equation 7.15. Pulmonary-to-Systemic Flow Ratio (Qp/Qs)
travels with each stroke or beat. The stroke volume in milliliters (mL) is
the volume of the cylinder formed by the product of the cross-sectional Qp Area PV × TVI PV
area (A) of the blood vessel or orifice and the distance (TVI) that the =
blood moves in the specified time period (ie, systole or diastole). Qs Area LVOT × TVI LVOT
 7 Principles of Echocardiography 89

Qp = pulmonary stroke volume (usually measured at pulmonary The proximal isovelocity surface area (PISA) method
valve annulus [PV]) (Figure 7.16) is used most frequently in the context of quanti-
fying mitral and aortic regurgitation. This method is a variation
Qs = systemic stroke volume (usually measured at left ventricu-
of the continuity equation and uses the property of flow conver-
lar outflow tract [LVOT])
gence of fluid as it approaches a restrictive orifice. Blood forms
TVI = time-velocity integral multiple concentric “shells” or “hemispheres” of isovelocity. As
the surface area decreases, the velocity increases. The veloc-
The continuity equation, which is based on the principle of ity at a given distance from the orifice (vr) can be measured by
conservation of mass (“what goes in must come out”), states altering the aliasing velocity of the color flow Doppler signal.
that flow proximal and distal to an orifice must be equal in a The flow rate through the orifice can be calculated (Equation
closed system (Equation 7.16). Rearrangement of the continu- 7.20). The ERO, also referred to as ROA, and regurgitant vol-
ity equation allows calculation of stenotic and regurgitant ori- ume can be calculated with the continuity equation and the peak
fice areas by measuring 3 variables and solving for the fourth velocity and TVI of the continuous-wave mitral regurgitant sig-
(Equation 7.17). nal (Equations 7.21 and 7.22). Variations of the PISA technique
also allow calculation of flow rate and volume and orifice area of
Equation 7.16. Continuity Equation stenotic mitral valves, atrial and ventricular septal defects, and
aortic coarctation.
Flowproximal = Flowdistal
A1 × TVI1 = A2 × TVI2 Equation 7.20. Proximal Isovelocity Surface Area (PISA)
Flow Rate, mL/s
A1 × TVI1 = proximal flow
Flow = 2π2 × vr
A2 × TVI2 = flow across valve
Flow = instantaneous flow rate (mL/s)
A1 = reference area
r = radial distance of isovelocity shell from orifice (cm)
A2 = area of the stenotic valve (cm2)
vr = flow velocity radius r (cm/s)
TVI = time-velocity integral
A
Equation 7.17. Valve Area, cm2
(Rearrangement of continuity equation [Equation 7.16])
Ao

TVI1
A2 A1 ×
TVI 2 PISA
LV LA
LA
A1 = reference area
A2 = area of the stenotic valve (cm2)
TVI = time-velocity integral
Mitral valve area can be measured with the continuity equa-
tion (Equation 16) or the pressure half-time method (Equations
7.18 and 7.19). B
PISA flow = MR flow
Equation 7.18. Pressure Half-time (PHT), ms
2πr2 × PISA v = ERO × MR v
PHT = DT × 0.29 r PISA
2πr2 × Alias v = ERO × MR v
PHT = time required for the peak gradient to decrease by
one-half 2πr2 × Alias v
MR ERO =
DT = deceleration time (time [ms] from the maximal velocity to MR v
zero velocity)
6.28r2 × Alias v
0.29 = an algebraic constant that converts velocity to gradient =
MR v
Equation 7.19. Mitral Valve Area (MVA) by Pressure Half- Figure 7.16. Proximal Isovelocity Surface Area (PISA) Method.
time Measurement, cm2 A, Diagram of PISA (arrows) of mitral regurgitation. As blood flow
converges toward the mitral regurgitant orifice, blood-flow velocity
220 759 increases gradually and forms multiple isovelocity hemispheric shells.
MVA = or
PHT DT The flow rate calculated at the surface of the hemisphere is equal to
the flow rate going through the mitral regurgitant orifice. Ao indicates
220 and 759 = empirical time constants equating to an MVA of aorta; LA, left atrium; LV, left ventricle. B, Calculation and derivation
approximately 1 cm2 of effective regurgitant orifice (ERO) area of mitral regurgitation (MR)
with the PISA method. r indicates PISA radius; v, velocity. (Previously
(Definitions as in Equation 7.18) published. See “Credit Lines” section.)
90 II Noninvasive Imaging
Equation 7.21. Effective Regurgitant Orifice (ERO) (cm2) for
Quantifying Mitral Regurgitation
Flow (mL/s)
ERO =
vMR (cm /s)
s bicuspid, or tricuspid) and calcification.
v MR = peak velocity of continuous-wave mitral regurgitant
signal
Equation 7.22. Regurgitant Volume (mL) for Quantifying
Mitral Regurgitation
Regurgitant Volume = ERO (cm2) × TVIMR (cm)
ERO = effective regurgitant orifice
TVIMR = time-velocity integral of continuous-wave mitral regur-
gitant signal
For cardiology examinations, be able to identify the Doppler
signals and assess the hemodynamic significance of the follow-
ing conditions:
1. Aortic stenosis—transvalvular velocity, gradient, and aortic valve
area by the continuity equation (Figure 7.17)
2. Aortic regurgitation—pressure half-time and diastolic flow reversals
in the aorta (Figure 7.18)
3. Mitral stenosis—transvalvular gradient, pressure half-time, and
mitral valve area (Figure 7.19)
4. Mitral regurgitation—regurgitant volume, fraction, and systolic flow
reversals in pulmonary veins
5. Pulmonary artery pressure—tricuspid regurgitant velocity
6. Hypertrophic cardiomyopathy—LV outflow tract gradient
7. Tricuspid regurgitation—systolic flow reversals in hepatic veins and
markedly dilated inferior vena cava and hepatic veins
Figure 7.17. Aortic Stenosis. Doppler signal obtained from the api-
cal window in a patient with severe, symptomatic calcific aortic steno-
sis. Left ventricular outflow tract (LVOT) velocity (vel) = 1 m/s; LVOT
time-velocity integral (TVI) = 20 cm; and LVOT diameter = 2.0 cm.
By the continuity equation, the aortic valve (AV) area = 0.47 cm2. AV
vel = 5 m/s; AV TVI = 135 cm; and mean gradient across the aortic
valve = 54 mm Hg.
Evaluation of Specific Disorders
Aortic Stenosis
1. M-mode and 2D echocardiography—valve morphology (unicuspid,
2. Doppler echocardiography—peak aortic velocity, TVI, mean gradi-
ent (Figure 7.17), and aortic valve area by the continuity equation
(Equation 7.17).
Severe aortic stenosis is usually present if the peak aor-
tic velocity is 4.0 m/s or greater, the mean pressure gradient is
greater than 40 mm Hg, the valve area is less than 1.0 cm2, and
the indexed valve area is less than 0.6 cm2 /m2. A small calcu-
lated aortic valve area associated with a low gradient and a low
cardiac output state requires careful evaluation to differentiate
decreased LV systolic function due to truly severe aortic stenosis
from milder aortic stenosis and the presence of unrelated myo-
cardial dysfunction. Dobutamine echocardiography has been
used to increase contractility and to increase cardiac output to
differentiate anatomical aortic stenosis from “relative” aortic
stenosis.
The main pitfalls in assessing aortic stenosis are underesti-
mating the gradient and overestimating the valve area when the
highest velocity Doppler signal is not obtained because of tech-
nical or anatomical factors. When there is a discrepancy between
clinical assessment and calculated valve area by transthoracic
study, TEE may be required for more sensitive assessment of the
valve morphology and degree of stenosis, and planimetry of the
valve area can also be performed.
Mitral Stenosis
1. M-mode and 2D echocardiography—valve morphology, doming or
“hockey stick” (long-axis view) (Figure 7.20), “fish mouth” (short-
axis view), leaflet and subvalvular thickening, calcification and
mobility (Abascal echocardiographic score), commissural anatomy,
and LA size.
2. Doppler echocardiography—mean gradient; mitral valve area by
continuity equation (Equation 7.17), pressure half-time (Equation
7.19), and planimetry methods; pulmonary artery pressure; and
degree of mitral regurgitation. All 3 methods of echocardiographic
assessment of mitral valve area correlate well with invasive mea-
sures, but each has unique features that render it more or less accu-
rate in a given patient (Box 7.3). Therefore, all 3 methods should be
performed to achieve an integrated approach to the severity of mitral
stenosis.
A high transvalvular gradient with normal pressure half-
time may reflect severe mitral regurgitation rather than mitral
stenosis. Severe mitral stenosis is usually present if the mitral
valve area is 1.0 cm2 or less, the mean resting pressure gradient
is 10 mm Hg or greater, or the pressure half-time is 220 ms or
longer. Exercise Doppler echocardiography can be very useful to
assess stress-induced changes in gradient, mitral valve area, and
pulmonary artery pressures.
TEE is essential before percutaneous mitral balloon valvu-
loplasty and can help define further the presence or absence of
commissural fusion and calcification. The presence of heavy cal-
cification at both commissures, significant subvalvular disease,
and marked leaflet thickening and immobility predict subopti-
mal results for valvuloplasty. Significant mitral regurgitation,
which may worsen after valvuloplasty, also is a contraindication
to valvuloplasty. LA thrombus must be excluded to avoid embolic
complications.
 7 Principles of Echocardiography 91

A B

Systole Diastole

Arch Ao Ao
PA

C D

Figure 7.18. Aortic Regurgitation. A, Holodiastolic reversal flow (arrows) in the descending aorta indicates severe aortic regurgitation. Similar
diastolic reversal can be seen in a descending thoracic aneurysm or shunt into the aorta during diastole (as in Blalock-Taussig shunt). The sample vol-
ume usually is located just distal to the takeoff of the left subclavian artery. PA indicates pulmonary artery. B, Two-dimensional color flow imaging
of the descending thoracic aorta during diastole. The orange-red flow in the descending aorta during diastole indicates flow toward the transducer
(ie, reversal flow due to severe aortic regurgitation). Ao indicates aorta. C, Color M-mode echocardiogram from the descending thoracic aorta shows
holodiastolic reversal flow (arrows). D, Pulsed wave Doppler recording of abdominal aorta shows diastolic flow reversal (arrows) in severe aortic
regurgitation. (Previously published. See “Credit Lines” section.)

Aortic Regurgitation
1. M-mode and 2D echocardiography—valve morphology, LV size
and function, premature mitral valve closure, diastolic opening of
the aortic valve (severe aortic regurgitation), fluttering of the mitral
valve, and cause of aortic regurgitation (Marfan syndrome, bicuspid
aortic valve, endocarditis, and dissection).
2. Color flow imaging—ratio of jet width or area to LV outflow tract
width or area (mild, <30%; severe, >60%).
3. Pulsed wave Doppler echocardiography—holodiastolic flow rever-
sals in the descending or abdominal aorta are indicative of signifi-
cant regurgitation.
4. Continuous wave Doppler echocardiography—pressure half-time
(mild, ≥400 ms; severe, ≤250 ms). High LV end-diastolic pressure
can shorten pressure half-time, causing overestimation of the sever-
ity of regurgitation.
5. Quantitative methods—Regurgitant fraction (mild, <30%; severe,
>55%) or regurgitant volume (mild, <30 mL per beat; severe,
≥60 mL per beat); effective regurgitant orifice (mild, <0.10 cm2;
severe, ≥0.30 cm2); vena contracta (ie, the width of the regurgitant
flow at the orifice), a surrogate measurement for the size of the ori-
fice (severe, ≥0.5 cm); and LV diastolic dimension in chronic aortic
regurgitation (mild, <6.0 cm; severe, ≥7.5 cm).
An integrated approach that incorporates these quantita-
tive and semiquantitative methods of evaluation should be used
because all the above can be influenced by factors other than the
degree of aortic regurgitation (Figure 7.18).
A restrictive mitral inflow pattern may be seen in acute severe
aortic regurgitation.
Mitral Regurgitation
1. M-mode and 2D echocardiography—valve morphology, LV size and
function, and cause of mitral regurgitation (mitral valve prolapse,
flail leaflet, mitral annular calcification, papillary muscle dysfunc-
tion or rupture, and endocarditis).
2. Color flow imaging—jet size and ratio of jet size to LA area. Color
flow imaging of jet size is influenced by instrument settings (pulse
repetition frequency, depth, etc), loading conditions, and jet direc-
tion. An eccentric jet, or one that “hugs” the LA wall, carries more
regurgitant volume than a similarly sized “central” or “free” jet.
3. Pulsed wave Doppler echocardiography—systolic reversals in the
pulmonary vein indicate severe mitral regurgitation.
4. Quantitative methods—regurgitant volume and fraction. The PISA
method allows assessment of regurgitant volume and ERO area
using the concept of the continuity equation and flow convergence.
5. TEE—useful in assessing mitral valve morphology and the cause of
regurgitation, useful for visualizing the color flow jet and pulmonary
veins, and useful intraoperatively before and after mitral valve repair
or replacement.
92 II Noninvasive Imaging
Mean 16 mm Hg
t½ 210 ms
Figure 7.19. Mitral Stenosis. Continuous wave Doppler signal
from a patient with severe mitral stenosis. Mean gradient is 16 mm Hg.
Pressure half-time (t1/2) is 210 ms. Mitral valve area by the pressure half-
time method is 1.0 cm 2.
Tricuspid Regurgitation
M-mode and 2D echocardiography are used to evaluate valve
morphology and right ventricular size and function to determine
the cause of tricuspid regurgitation (rheumatic valve, prolapse,
Ebstein anomaly, carcinoid valve, right ventricular infarct, pul-
monary hypertension, or tricuspid valve injury). Severe tricuspid
regurgitation is suggested by a color flow regurgitant jet area of
30% or more of the right atrium, annular dilatation of 4 cm or
more, increased tricuspid inflow velocity greater than 1.0 m/s, or
systolic flow reversals in the hepatic veins.
Figure 7.20. Mitral Stenosis. Rheumatic mitral stenosis with left
atrial (LA) enlargement and obvious doming of the anterior mitral leaf-
let. LV indicates left ventricle; MV, mitral valve; RA, right atrium; RV,
right ventricle.
Box 7.3. Limitations and Pitfalls in Assessing Mitral
Valve Area
Two-dimensional (2D) planimetry
Dependent on 2D image quality, gain-setting, and
ability to visualize the minimal orifice area
Less accurate when extensive calcification is present
Difficult after commissurotomy because of irregular
orifice
Doppler pressure half-time
Tachycardia
Nonlinear pressure decay
Significant or acute aortic regurgitation increases
the rate of increase in left ventricular pressure and
shortens pressure half-time (mitral valve area is
overestimated)
Immediately after percutaneous mitral valvuloplasty
when hemodynamics are not stable (mitral valve
area is overestimated)
Continuity equation
Cumbersome to perform, and multiple
measurements are subject to error
Mitral valve area is underestimated when significant
mitral regurgitation is present
Prosthetic Valves
The range of “normal” hemodynamic variables (gradient, effec-
tive orifice area, etc) for a given prosthetic valve type and loca-
tion is broad. Published reference values for these variables serve
as guidelines. The best approach for assessing a patient is to per-
form a baseline transthoracic 2D and hemodynamic evaluation
early postoperatively to establish the patient’s own “normal val-
ues” for later comparison.
Regurgitation is normally present with virtually all prosthetic
valves and has been well characterized in vitro and in vivo. This
“physiologic” regurgitation is usually of low volume and veloc-
ity, appearing as a nonaliased jet, and should be differentiated
from pathologic regurgitation. Prosthetic valves are inherently
stenotic, with higher transvalvular gradients than native valves.
Prosthetic valve dysfunction includes valvular and perival-
vular regurgitation, pannus formation, obstruction, endocarditis,
abscess, dehiscence, and thromboembolism. A complete trans-
thoracic 2D evaluation may be limited by acoustical shadowing
from the prosthesis. Doppler echocardiography usually can assess
valve gradients and effective orifice areas accurately, detect and
quantify regurgitation, and provide ancillary information about
pulmonary pressures and LV systolic and diastolic function.
Unexpectedly high transvalvular gradients or small effective
orifice areas should be assessed further to exclude valve dysfunc-
tion. If available, comparison with a previous echocardiographic
study is invaluable. If no change has occurred and the patient is
clinically stable, it is likely that the hemodynamics are normal
for that patient and valve or that a prosthesis-patient mismatch is
present; that is, the valve is relatively undersized for the patient’s
body size and hemodynamics. High gradients may also occur in
the presence of increased transvalvular flow, such as anemia or
other high-output states, but effective orifice areas should remain
relatively normal in these conditions. High velocities present in
 7 Principles of Echocardiography 93

otherwise normal valves may be due to localized high-velocity Hypertrophic Cardiomyopathy

jets and distal pressure recovery, which may lead to Doppler gra-
M-mode and 2D echocardiography are useful in establishing the
dients that are higher than those measured by catheter. This has
diagnosis of hypertrophic cardiomyopathy, evaluating the sever-
been observed most commonly in smaller Starr-Edwards and St
ity of hypertrophy and its morphology (asymmetric, symmetric,
Jude prosthetic valves.
apical, etc), and assessing the presence and degree of LV outflow
TEE is invaluable and often complementary for visualizing
obstruction. Typical M-mode features of hypertrophic cardiomy-
valve motion, ring abscess, thrombus, pannus, endocarditis,
opathy include mid-systolic aortic valve notching and systolic
or the degree of regurgitation if a transthoracic study cannot
anterior motion of the mitral apparatus. Also, 2D echocardiog-
address the clinical question or concern adequately, or if there
raphy is useful for demonstrating presence or absence of systolic
are abnormalities detected from transthoracic examination but
anterior motion and for assessing mitral valve morphology.
the cause remains uncertain.
Pulsed wave Doppler and color flow imaging are useful in
localizing the presence and site of LV outflow tract or mid-ven-
Chest Pain and Acute Myocardial Infarction tricular obstruction. The degree of obstruction (pressure gra-
dient) is defined by a characteristic continuous wave Doppler
Echocardiography is useful for assessing the causes of chest pain
late-peaking, dagger-shaped signal. The peak gradient is calcu-
(pericarditis, large pulmonary embolus, aortic dissection, etc),
lated with the modified Bernoulli equation (4v2). Measurement
global and regional LV systolic function, and region and extent of
of the gradient during the Valsalva maneuver, administration of
myocardial infarction and for identifying patients who may ben-
amyl nitrite, or exercise can demonstrate the dynamic nature of
efit from revascularization. The absence of regional wall motion
the obstruction.
abnormalities during chest pain makes ischemia unlikely as a
Diastolic abnormalities in hypertrophic cardiomyopathy are
cause of the chest pain and so may be useful to aid in triaging
strongly associated with symptoms of dyspnea and exercise intol-
patients presenting emergently. A restrictive pattern of LV dia-
erance and should be carefully assessed. Isovolumic relaxation
stolic filling or a high wall motion score index (or both) predicts
period flow is present occasionally and is due to asynchronous
a poor prognosis. Infarct-related complications may be readily
ventricular relaxation. Normally, there is little or no flow during
assessed with echocardiography (Box 7.4).
the isovolumic relaxation period when both the mitral and the
aortic valves are closed. It is important not to confuse this flow
with the mitral E wave.
Box 7.4. Detection of Complications of Myocardial Infective Endocarditis
Infarction by Echocardiography
Echocardiography is the diagnostic procedure of choice for
Right ventricular infarction
detecting valvular vegetations (Figure 7.21). It has the additional
Dilated right atrium and right ventricle with regional benefit of being useful for detecting abscesses, valve perforation,
wall motion abnormalities rupture or aneurysm, fistula, dehiscence of a prosthetic valve,
Significant tricuspid regurgitation and hemodynamic consequences (shunt or regurgitation). The
combination of transthoracic echocardiography and TEE has a
Inferior vena cava dilatation or plethora
Pericardial effusion or tamponade
Mitral regurgitation
Ischemic—papillary muscle dysfunction or annular
dilatation
Ruptured papillary muscle
Ventricular septal defect
Color flow localization
Right ventricular dilatation
Elevated right ventricular pressure
Inferior vena cava plethora
Left ventricular free wall rupture
Pericardial effusion or tamponade
Pericardial thrombus
Extracardiac flow
Pseudoaneurysm (contained rupture)
Narrow neck, thin-walled
Aneurysm
Myocardial thinning, 90% located at apex
Figure 7.21. Valvular Vegetation. Vegetation (arrow) on the mitral
Left ventricular thrombus valve with flail in endocarditis. LA indicates left atrium; LV, left ventri-
cle; RA, right atrium; RV, right ventricle.
94 II Noninvasive Imaging

sensitivity for vegetations in the range of 90% to 95% for native
valves and 85% to 90% for prosthetic valves. Patients with sus-
pected infective endocarditis should have a baseline transtho-
racic study and, in most cases, a transesophageal study. TEE is
superior to transthoracic echocardiography for diagnosing valve
ring abscess. Serial echocardiographic examinations may be
helpful, especially in patients with congestive heart failure, fever,
or persistently positive blood cultures.
The false-negative rate for detection of vegetations is low
(<5%), but in patients with clinical features consistent with
infective endocarditis and negative initial TEE findings, it may
be reasonable to repeat the study in 1 to 2 weeks.
Pericardial Disease
Effusion
Echocardiography is the diagnostic procedure of choice for detect-
ing and evaluating pericardial effusion. Effusion is defined as an
echo-free space present throughout the cardiac cycle. Large effu-
sions may be associated with a “swinging heart” (Figure 7.22A).
Tamponade
M-mode and 2D features of tamponade are not sensitive but
can be quite specific. These include diastolic collapse of the
right atrium or right ventricle (Figure 7.22B) and inferior vena
cava plethora with blunted inspiratory collapse. Doppler find-
ings of cardiac tamponade are more sensitive and are based on
ventricular interdependence due to the relatively fixed cardiac
volume and reduced response of intrapericardial pressures to
changes in intrathoracic pressures. With inspiration, LV fil-
ling is impaired, whereas right ventricular filling is favored.
Doppler findings include an inspiratory increase in IVRT and
decreased mitral E-wave velocity, with reciprocal changes in
tricuspid valve inflow tracings. Pulmonary venous, hepatic
venous, and LV outflow tract tracings show similar respira-
tory flow changes. Echocardiographically guided pericardio-
centesis is the initial therapy of choice for most patients with
tamponade (except for patients who have aortic dissection with
tamponade). The use of an indwelling pigtail catheter for com-
plete drainage (until fluid return is <25 mL over 24 hours) is
associated with a much lower likelihood of recurrence of per-
icardial effusion. The need for surgical management of peri-
cardial effusion has become uncommon with the introduction
of echocardiographically guided pericardiocentesis techniques
and the adaptation of pigtail catheter drainage for decreasing
recurrence of effusion. Sclerotherapy is no longer used at Mayo
Clinic and is generally not recommended because of signifi-
cant pain associated with instilling a sclerosing agent into the
pericardial space.

Figure 7.22. Cardiac Tamponade. A, “Swinging heart” in cardiac tamponade. With a large amount of fluid, the position of the heart changes
dramatically during the cardiac cycle. Left, Diastole with right atrial inversion (arrow). Right, Systole. B, Diastolic collapse of right ventricle (RV) in
cardiac tamponade. Parasternal long-axis view of tamponade during systole (left) and diastole (right). The RV collapses during diastole. The single
arrows indicate anterior pericardial effusion; the pairs of arrows indicate posterior pericardial effusion compressing the RV free wall. LA indicates
left atrium; LV, left ventricle; PE, pericardial effusion; VS, ventricular septum.
 7 Principles of Echocardiography 95

Normal Constriction Restriction

ECG

Resp i e i e i e
E
A
MV
DT ≥160 DT <160

TV

SR DR
HV
D
S DR

Figure 7.23. Constrictive and Restrictive Pericarditis. Schema of Doppler velocities from mitral inflow (MV), tricuspid inflow (TV), and hepatic
vein (HV). Electrocardiographic (ECG) and respirometric (Resp) recordings with inspiration (i) and expiration (e) are also represented. The rela-
tive changes from normal caused by restrictive or constrictive pericarditis are represented. Both restriction and constriction are characterized by
short deceleration time (DT), but patients with constriction show reciprocal changes in filling of the left and right sides of the heart with respiration,
whereas patients with restriction do not. A indicates atrial contraction; D, diastolic; DR, diastolic reversal; E, early filling phase; S, systolic; SR, sys-
tolic reversal. (Previously published. See “Credit Lines” section.)

Constrictive Pericarditis similar to those of tamponade, with an inspiratory decrease in

M-mode and 2D features of constrictive pericarditis include left-sided flow (Figure 7.23). Expiratory hepatic vein diastolic
thickened or hyperechoic pericardium, abnormal “jerky” sep- flow reversals are often prominent. There is an absence of sig-
tal motion, respiratory variation in ventricular size, and a nificant inspiratory augmentation of systolic forward flow in the
dilated inferior vena cava. Doppler features of constriction are superior vena cava. Restrictive cardiomyopathy usually shows

A B

Figure 7.24. Aortic Dissection. A, Acute aortic dissection complicated by pleural effusion. Left pleural effusion within the posteromedial costo-
phrenic angle highlights the dissected (arrow) descending thoracic aorta (Ao); a portion of the left lung (L) is also noted within the effusion. PE
indicates pleural effusion. B, Transesophageal echocardiogram of the descending thoracic aorta with dissection present. FL indicates false lumen;
TL, true lumen. (Previously published. See “Credit Lines” section.)
96 II Noninvasive Imaging
Figure 7.25. Intracardiac Mass. Left atrial (LA) myxoma; transverse
4-chamber transesophageal echocardiographic plane. Cystic echolucen-
cies (arrows) are clearly seen within this myxoma (T); they were not evi-
dent on transthoracic examination. The myxoma appears to be attached
to the mitral valve but was found to be attached to the mitral annulus on
off-axis imaging. LV indicates left ventricle; RA, right atrium; RV, right
ventricle. (Previously published. See “Credit Lines” section.)
no significant respiratory changes in mitral inflow; therefore,
Doppler echocardiography is useful in differentiating constric-
tion from restriction. Tissue Doppler imaging is also helpful in
distinguishing constriction from restriction. In constriction, e′ is
well preserved (usually >0.08 m/s). In restriction, e′ is usually
of low velocity.
Thoracic Aorta
Although transthoracic echocardiography is usually useful
for visualizing the aortic root and arch, most of the aorta can-
not be evaluated satisfactorily. With TEE, the entire thoracic
Figure 7.26. Intracardiac Thrombus. Left atrial (LA) appendage
thrombus; transverse transesophageal echocardiographic plane, basal
short-axis view. A protruding thrombus (arrowheads) fills the LA
appendage. This thrombus was slightly mobile on real-time evaluation.
AV indicates aortic valve. (Previously published. See “Credit Lines”
section.)
B
Figure 7.27. Patent Foramen Ovale. Examples of transesophageal
echocardiographic findings in patients with paradoxical embolism. In
each case, a thrombus (arrows or arrowheads) is crossing through a pat-
ent foramen ovale. AV indicates aortic valve; IAS, interatrial septum;
LA, left atrium; LV, left ventricle; RA, right atrium; RV, right ventricle.
(Previously published. See “Credit Lines” section.)
aorta can be seen in most patients. Dissection can be evaluated
with TEE (Figure 7.24) as well as with computed tomography,
magnetic resonance imaging, or aortography. Because time is
often critical in suspected acute dissection, TEE has the added
advantages of 1) portability to bedside for any patient whose
condition is unstable, 2) simultaneous assessment of cardiac
function and associated conditions (aortic regurgitation and
pericardial effusion), and 3) no need for contrast agents. Other
thoracic aorta conditions that can be readily evaluated with
TEE include aortic aneurysm, rupture, ulcer, debris, abscess,
and coarctation.
Source of Embolus
Cardiovascular sources of emboli may account for 20% to 40%
of all strokes. Potential sources of emboli detectable with tran-
sthoracic echocardiography or TEE include intracardiac mass
(Figure 7.25) or thrombus (Figure 7.26), valvular vegetation,
thoracic aortic debris, atrial septal aneurysm, and patent fora-
men ovale. In the absence of overt cardiac disease on the basis
 7 Principles of Echocardiography 97

1 cm
VS
Ao

LV LA

PW

Figure 7.28. Hypertrophic Obstructive Cardiomyopathy. Ao indi-
cates aorta; LA, left atrium; LV, left ventricle; PW, posterior wall; VS,
ventricular septum.
of history, physical examination, or electrocardiographic find-
ings, the yield from a transthoracic echocardiogram for identi-
fication of a cardiac source of embolus is less than 1% and is
not routinely recommended. The transthoracic examination,
if performed, should focus on LV function and on excluding
abnormalities such as valvular heart disease and tumors. Several
Figure 7.30. Aortic Debris. Extensive atherosclerotic debris within
the descending thoracic aorta; transverse transesophageal echocardio-
graphic plane. A “shaggy” appearance of the lumen of the aorta (Ao) is
produced by several partially mobile lesions (arrowheads) projecting far
into the aortic lumen. This patient presented with diffuse atheroembolic
cutaneous infarcts of the feet and “blue toe” syndrome. (Previously pub-
lished. See “Credit Lines” section.)
studies have concluded that proceeding directly to TEE is a clin-
ically useful and cost-effective strategy for evaluation of stroke.
TEE is particularly well suited for excluding left atrial and left
atrial appendage thrombi, spontaneous echocardiographic con-
trast, patent foramen ovale, and mobile lesions in the thoracic
aorta. In recent years, echocardiographically guided percutane-
ous device closure of patent foramen ovale has been performed
for the prevention of recurrent cerebrovascular events in selected
patients (ie, patients whose patent foramen ovale is thought to
be a potential culprit for their cerebrovascular events), generally
after thorough investigations that have eliminated other potential
causes or sources of stroke.

Selected Abnormalities That

Are Diagnosed With TEE or
Transthoracic Echocardiography

1. Left atrial myxoma (Figure 7.25)

2. Left atrial thrombus (Figure 7.26)
3. Patent foramen ovale (Figure 7.27)
4. Hypertrophic obstructive cardiomyopathy (Figure 7.28)
5. Ruptured mitral valve chordae tendineae with flail leaflet
(Figure 7.29)
6. Aortic dissection (Figure 7.24)
7. Aortic debris (Figure 7.30)
8. Atrial septal defect: sinus venosus type (Figure 7.31) and secundum
Figure 7.29. Ruptured Chordae Tendineae. Mitral valve prolapse type (Figure 7.32)
with flail leaflet; left ventricular (LV) outflow view in the transverse 9. Rheumatic mitral stenosis (Figure 7.20)
plane. There is prolapse of the posterior mitral leaflet, with a flail leaflet 10. Pulmonary hypertension (Figure 7.33)
segment (arrow) producing a large deficiency in coaptation (arrowhead) 11. Cardiac effusion and tamponade: “swinging heart” (Figure 7.22A)
with the anterior leaflet. AV indicates aortic valve; LA, left atrium. and right ventricular diastolic collapse (Figure 7.22B)
(Previously published. See “Credit Lines” section.) 12. Endocarditis (Figure 7.21)
98 II Noninvasive Imaging

A B

Figure 7.31. Sinus Venosus Atrial Septal Defect. Short-axis scan (midesophageal transducer, horizontal plane, withdrawn to the high atrial level
just below the superior vena cava at the caval-atrial junction [SVC-RA]). A, The transducer is posterior to the left atrium (LA). Note the typical sinus
venosus atrial septal defect (arrows) between the LA and the SVC-RA. B, Color flow Doppler study. The color polarity has been reversed to better
illustrate the atrial septal defect left-to-right shunt (ie, LA to SVC-RA). Ao indicates ascending aorta. (Previously published. See “Credit Lines”
section.)

Figure 7.32. Secundum Atrial Septal Defect. Contrast and color flow echocardiographic studies show longitudinal scans of the atrial septum
from the midesophagus. Top, Long-axis view of the atrial septum with a 1-cm atrial septal defect (arrow). The transducer is within the esophagus
posterior and adjacent to the left atrium (LA). Anteriorly, the right atrium (RA) and superior vena cava (SVC) are visualized. Bottom left, After an
upper extremity venous injection, the RA is densely opacified. Left-to-right shunt (arrow) across the atrial septal defect appears as a negative contrast
effect in the RA. Bottom right, Color flow Doppler study shows left-to-right shunt (arrow) across the same septal defect. (Previously published. See
“Credit Lines” section.)
 7 Principles of Echocardiography 99

A B

Figure 7.33. Pulmonary Hypertension. A, parasternal short-axis view demonstrating the D-shaped left ventricular cavity and enlarged right ven-
tricular (RV) cavity in pulmonary hypertension. Similar appearances are present in RV volume overload; however, flattening of the ventricular sep-
tum (VS) persists during the entire cardiac cycle in RV and pulmonary artery pressure overload, whereas it disappears during systole in RV volume
overload. MV indicates mitral valve. B, Corresponding pathology specimen. (Previously published. See “Credit Lines” section.)

Abbreviations
LVEDD left ventricular end-diastolic diameter
A late diastolic mitral inflow velocity (atrial contraction) PISA proximal isovelocity surface area
2D 2-dimensional PWT posterior wall thickness
E early diastolic mitral inflow velocity (early filling) Qp /Qs ratio of pulmonary stroke volume to systemic stroke
e′ early diastolic velocity of the mitral annulus volume
ERO effective regurgitant orifice ROA regurgitant orifice area
IVRT isovolumic relaxation time SWT septal wall thickness
LA left atrial TEE transesophageal echocardiography
LV left ventricular TVI time-velocity integral
 8
Stress Echocardiography
PATRICIA A. PELLIKKA, MD
Stress echocardiography, introduced in 1979, is used for detec-
tion of coronary artery disease. The technique has evolved into a
widely used, versatile technique not only for diagnosis of ische-
mic heart disease but also for determination of prognosis. The
rationale for stress echocardiography is that stress results in wall
motion abnormalities in regions subtended by a stenosed cor-
onary artery; these wall motion abnormalities can be detected
with echocardiography.
Methods
Various methods of stress have been used in combination with
echocardiography. Treadmill exercise echocardiography is the
most widely used form of exercise echocardiography. Images are
obtained before and immediately after symptom-limited tread-
mill exercise. Attention is directed toward assessment of regional
wall motion and evaluation of changes in ejection fraction and
systolic volume. The standard views are parasternal long- and
short-axis and apical 4- and 2-chamber views. Additional views,
including apical long-axis and apical short-axis, also are obtained.
Rest and stress images are compared side by side to appreciate
subtle changes. Alternatively, the test may be performed during
either supine or upright bicycling. Bicycle imaging offers the
advantage that images can be obtained during exercise. This is
useful if Doppler data are obtained in addition to assessment of
regional wall motion.
For patients who are unable to perform physical exercise,
pharmacologic stress testing with dobutamine or the vasodilators
dipyridamole or adenosine can be used. Dobutamine is the most
common pharmacologic stress agent used in combination with
echocardiography. Dobutamine is predominantly a β1-adrenergic
stimulating agent. Its half-life in plasma is approximately 2 min-
utes, and therefore it must be administered with a continuous
100
intravenous infusion. A typical protocol involves administra-
tion at a starting dosage of 5 mcg/kg per minute, increasing at
3-minute intervals to 10, 20, 30, and 40 mcg/kg per minute.
End points are intolerable symptoms, uncontrolled hyperten-
sion, hypotension, or considerable arrhythmias. The infusion is
continued to achievement of 85% of the age-predicted maximal
heart rate (220 − age). If this is not achieved with dobutamine
infusion, atropine at a dose of 0.5 mg is administered intrave-
nously and repeated at 1-minute intervals to a maximal total dose
of 2 mg. Used in this way, atropine has been shown to increase
the sensitivity of dobutamine stress echocardiography, especially
in patients receiving β-blocker therapy. Dynamic intracavitary
or left ventricular outflow tract obstruction can be detected with
Doppler echocardiography in approximately 20% of patients
undergoing dobutamine stress echocardiography.
Dipyridamole is a coronary vasodilator, the effects of which
are mediated by increased interstitial levels of endogenous aden-
osine. Dipyridamole decreases coronary vascular resistance and
increases coronary blood flow but seems to have little effect on
vascular resistance in ischemic areas where small vessels are
already maximally dilated. Compared with dipyridamole, aden-
osine offers the advantages of greater coronary vasodilation
and a shorter half-life (less than 10 seconds). With vasodilator
stress echocardiography, the typical effects are a mild to moder-
ate increase in heart rate and a mild decrease in blood pressure.
Contraindications include bronchospastic lung disease, severe
obstructive lung disease, or current use of aminophylline. The
methylxanthine caffeine, another antagonist of adenosine, should
not be ingested 12 hours before testing. Use of oral dipyridamole
should be discontinued 24 hours before testing. Adenosine is
contraindicated in patients with heart block.
A newer method for patients who are unable to exercise is
transesophageal atrial pacing stress echocardiography. A 10F
 8 Stress Echocardiography 101

flexible catheter is inserted orally or nasally after a topical anes-
thetic is applied to the patient’s posterior pharynx. Initial pac-
ing is set at 10 beats per minute more than a patient’s baseline
heart rate at the lowest current that provides stable atrial capture
(usually 15 mA). At 2-minute intervals, the paced heart rate is
increased to 85% and 100% of the age-predicted maximal heart
rate. If Wenckebach second-degree heart block develops, admin-
istration of atropine is necessary. The advantage of this protocol
is very few adverse effects related to the method and rapid return
to baseline conditions on discontinuation of atrial pacing.
In patients with a temporary or permanent pacemaker, a pacing
stress echocardiogram can be obtained by temporarily reprogram-
ming the pacemaker to a higher heart rate. Ergonovine or hyperven-
tilation has been used in conjunction with stress echocardiography
to detect coronary vasospastic disease. Ergonovine testing has the
potential to produce severe or prolonged ischemia and should not
be performed in patients with previous infarction or documented
ischemia. This form of testing is most safely performed in the angi-
ography laboratory, where nitrates can be infused locally and the
coronary artery opened mechanically if complications occur.
Interpretation
During stress echocardiography, images obtained at rest are
compared with those obtained during stress. With supine bicycle
or dobutamine stress echocardiography, imaging is performed
during gradual increases in stress. This approach permits rec-
ognition of the heart rate or level of stress at which ischemia
first develops. The normal response to stress is the development
of hyperdynamic wall motion, a decrease in end-systolic vol-
ume, and an increase in ejection fraction (Figure 8.1). Regional
wall motion is assessed in each left ventricular segment at rest
and with stress. A 16-segment model of the left ventricle is most
frequently used; a 17-segment model, which includes an addi-
tional segment at the left ventricular apex, has been proposed for
echocardiographic perfusion imaging. The development of new
or worsening regional wall motion abnormalities is considered
a manifestation of ischemia (Figure 8.2). Resting wall motion
abnormalities that are unchanged with stress are considered to
represent infarction. A biphasic response, that is, with a low level
of exercise or pharmacologic stress, an improvement in contrac-
tility of regions that are hypokinetic or akinetic at rest followed
by worsening with continued stress, is considered a marker of
viability. This may be seen in segments subtended by severe
coronary stenosis. Regional wall motion abnormalities corre-
late with coronary artery anatomical distribution of blood flow.
A regional wall motion score is calculated at rest and at stress
as a sum of the scores of the individual segments divided by
the number of segments. A 5-point scoring system is most com-
monly used, in which 1 = normal, 2 = hypokinesis, 3 = akinesis,
4 = dyskinesis, and 5 = aneurysm. A decrease in the ejection
fraction or an increase in end-systolic volume with stress is a
marker of extensive ischemia.
Harmonic imaging is beneficial for improving image quality.
Intravenous administration of contrast (sonicated microbubbles)
is recommended if images are technically inadequate or if more
than one segment cannot be adequately visualized at rest. With
current state-of-the-art ultrasound equipment and use of contrast
agent as needed, technically adequate transthoracic images can
be expected in at least 97% of patients.
• A biphasic response, that is, with a low level of exercise or pharma-
cologic stress, an improvement in contractility of regions that are
hypokinetic or akinetic at rest followed by worsening with continued
stress, is considered a marker of viability.

Rest Post Exercise

Post

Figure 8.1. Parasternal long- (top) and short- (bottom) axis images at rest (left) and immediately after (Post) exercise (right) show normal systolic
contraction at rest and with stress. With exercise, ejection fraction markedly increases and all walls become hyperdynamic. Mild concentric left ven-
tricular hypertrophy also is noted in this patient, who had a history of hypertension.
102 II Noninvasive Imaging

Rest 54 bpm Exercise 91 bpm

Rest 54 bpm Exercise 103 bpm

Figure 8.2. Parasternal (A) and apical (B) images show the development of hypokinesis of the apex and anterior wall after the stress of exercise.
In contrast to Figure 8.1, there has been an increase in systolic cavity size with exercise, which was accompanied by a decrease in ejection fraction.
Aortic valve sclerosis also was noted. bpm indicates beats per minute.
 8 Stress Echocardiography 103

Usefulness for Detection of Coronary with diabetes mellitus. Not only the presence of ischemia but
Artery Disease also the extent and severity of ischemia as shown by the percent-
age of abnormal segments at peak stress, the stress wall motion
Stress echocardiography is widely used for the detection of
score index, a multivessel distribution of abnormalities, the
coronary artery disease and for assessment of its functional
change in wall motion score index with stress, the ejection frac-
significance. Coronary artery disease may be manifested by
tion response to stress, and end-systolic volume response to stress
either a resting wall motion abnormality or a stress-induced
are useful for identifying patients at highest risk. Risk indices
abnormality. The accuracy of stress echocardiography has been
combining clinical, exercise test, and stress echocardiographic
shown to be superior to that of exercise electrocardiography. The
variables have been developed and validated. The prognosis is
accuracy of stress echocardiography has been directly compared
excellent for patients with normal results on exercise echocardi-
with that of radionuclide perfusion imaging in laboratories of
ography; event rates, including cardiac death, myocardial infarc-
similar proficiency. In this situation, the techniques performed
tion, or coronary revascularization, are less than 1% per year.
similarly. Recent meta-analyses also have compared stress
Exercise echocardiography has been shown to be a cost-effective
echocardiographic with radionuclide techniques and have found
method for assessment of patients with known or suspected cor-
a higher specificity with stress echocardiography.
onary artery disease.
Advantages of stress echocardiography include its relatively
Stress echocardiography has been used to predict risk in
lower cost than other imaging tests and its versatility, that is, car-
patients who have had a myocardial infarction. The presence of
diac structure and function, including wall thicknesses, chamber
residual or remote ischemia, which manifests as a stress-induced
sizes, valves, the proximal aortic root, and presence of pericar-
wall motion abnormality or a worsening of ventricular function
dial effusion, can be evaluated simultaneously. For the detection
with stress, indicates a worse prognosis. Stress testing may be
of coronary artery disease, the sensitivity has been reported to
done early after myocardial infarction, and dobutamine stress
range from 72% to 97% depending on lesion severity and extent
testing allows recognition of myocardial viability.
of coronary artery disease. As with all forms of stress testing,
Assessment of cardiac risk before noncardiac operation is
the sensitivity of stress echocardiography for detecting single-
a frequent application of stress echocardiography and is espe-
vessel disease is lower than that for detecting multivessel disease.
cially beneficial in patients with risk factors, cardiac symptoms,
Sensitivity is highest if coronary artery disease is defined as ste-
or known coronary artery disease. Stress echocardiography has
nosis that narrows the diameter by 70% or more, and sensitivity
been shown to provide information beyond that which can be
is less when coronary artery disease is defined as stenosis that
obtained from clinical variables, resting left ventricular function,
narrows the diameter by 50% or more. However, specificity var-
or exercise electrocardiography. This advantage has been shown
ies conversely. The apparent accuracy of stress echocardiography
in patients undergoing either vascular and nonvascular operation.
can be affected by referral bias in that only patients with positive
Pharmacologic stress echocardiography with dobutamine is used
test results are likely to be referred for angiography, a practice
frequently because orthopedic, peripheral vascular, or other
leading to an artificially higher sensitivity and lower specificity.
comorbid conditions may limit a patient’s ability to exercise.
Although regional wall motion abnormalities are the hall-
With dobutamine stress echocardiography, the ischemic thres-
mark of ischemic heart disease, they also may occur in cardi-
hold, that is, the heart rate at which ischemia first develops, can be
omyopathy and microvascular disease and may be precipitated
used to monitor a patient perioperatively. If ischemia is extensive
by a hypertensive response to stress. False-negative studies can
or occurs at a low heart rate, preoperative coronary angiography
occur in patients with single-vessel disease or in those in whom
and revascularization are indicated. Alternatively, if the ische-
a small region of myocardium is subtended by a stenosed ves-
mia begins at a higher heart rate, perioperative β-blocker therapy
sel. False-negative results also can occur if there is a delay in
should suffice.
acquisition of images after peak exercise, if the patient performs
Stress echocardiography also has an important role for detec-
a low level of exercise, or if the heart rate response to exercise or
tion of myocardial viability. Systolic dysfunction may indicate
dobutamine stress is suboptimal.
hibernating or stunned myocardium. Augmentation of regional
• For the detection of coronary artery disease, the sensitivity of function in dysfunctional segments with a low dose of dobu-
stress echocardiography has been reported to range from 72% to tamine and reworsening of function with high doses (Figure 8.3)
97% depending on lesion severity and extent of coronary artery have been shown to be predictive of recovery of function after
disease. coronary revascularization. In stunned myocardium, that is, myo-
• As with all forms of stress testing, the sensitivity of stress echocar- cardium that is viable but dysfunctional after prolonged severe
diography for detecting single-vessel disease is lower than that for ischemia but for which blood flow has been restored, contractil-
detecting multivessel disease. ity also improves during dobutamine administration. Myocardial
thickness is also an indicator of viability. In segments with
a thickness less than 5 or 6 mm, the likelihood of recovery of
Prognostic Value
function after coronary revascularization is low. In patients with
Numerous large studies from various centers have shown the myocardial viability and ischemia who do not undergo revascu-
prognostic value of stress echocardiography. It provides informa- larization, the prognosis is unfavorable.
tion incremental to that which can be gleaned by assessing clin-
ical variables, exercise duration, electrocardiographic changes,
and resting echocardiographic function to identify patients at
Stress Echocardiography and
risk of all-cause mortality and cardiac events, including cardiac
Nonischemic Heart Disease
death and myocardial infarction. This prognostic value has been Stress echocardiography also may be used to assess valvular
found in various subgroups, including both men and women, the heart disease, hypertrophic cardiomyopathy, and pulmonary
elderly, patients who have had coronary artery bypass grafting, hypertension and to recognize abnormalities of diastolic func-
patients who have had percutaneous intervention, and patients tion that occur with stress. In most patients with valvular heart
104 II Noninvasive Imaging

A
Baseline 10 mcg/kg per minute

Pre-peak Peak

Baseline 10 mcg/kg per minute

Pre-peak Peak

Figure 8.3. Dobutamine stress echocardiographic parasternal images (A, long axis; B, short axis) in a patient with a recent anterior wall myo-
cardial infarction show improvement of contraction of the distal anteroseptum, apex, and mid anterior wall with low-dose dobutamine. At peak
dose, there is reworsening of systolic function, accompanied by an increase in end-systolic size. This biphasic response—with initial improvement
of hypokinetic segments followed by worsening at higher doses of dobutamine—is characteristic of hibernating myocardium. The patient, who had
received thrombolysis at the time of acute infarction, was found to have a residual high-grade stenosis of the left anterior descending coronary artery
and underwent percutaneous coronary intervention.
 8 Stress Echocardiography
disease, resting echocardiography and Doppler assessment
suffice. However, for patients in whom exertional symptoms do
not correlate with resting echocardiographic findings, a stress
test may be beneficial. In patients with mitral stenosis or regur-
gitation, changes in severity of regurgitation may occur with
exercise. Alternatively, marked increases in the gradient across
the mitral valve may occur with exercise in patients with mitral
stenosis, contributing to exertional symptoms. Lung disease and
mitral valve disease can be differentiated by assessing relative
changes in the gradient across the mitral valve with relative
changes in pulmonary artery pressure as assessed by tricuspid
regurgitation velocity.
In patients with aortic stenosis and reduced left ventricular
systolic function (low-output, low-gradient aortic stenosis), low
doses of dobutamine can be administered to augment the stroke
volume; the Doppler-derived aortic valve area can then be recal-
culated under these different hemodynamic conditions. If the
stenosis is considerable, the Doppler-derived aortic valve area
remains reduced despite augmentation of ventricular function.
Furthermore, dobutamine administration can be used to iden-
tify cases in which no contractile reserve is present. Changes in
pulmonary artery systolic pressure with exercise echocardiogra-
phy may be useful for identifying patients limited by pulmonary
hypertension. Changes in diastolic function, most commonly
measured from the mitral inflow and mitral annulus tissue
Doppler profiles and the ratio of the early diastolic mitral inflow
velocity (E) to mitral annulus velocity (e′), can identify patients
with increases in left ventricular filling pressure during exercise.
105
• In patients with low-output, low-gradient aortic stenosis, the
Doppler derived aortic valve area remains reduced despite aug-
mentation of ventricular function by dobutamine infusion.
New Developments
New technologies, including color kinesis, tissue Doppler, and
strain and strain rate imaging, can be applied to assess segmental
function and may provide quantitative assessment of left ventric-
ular response to stress. Myocardial perfusion may be assessed
with contrast echocardiography performed in conjunction with
regional wall motion assessment. These techniques have been
shown to be feasible with exercise, vasodilator, or dobutamine
stress. Real-time 3-dimensional imaging has become feasible
and permits acquisition of stress echocardiographic data very
quickly at or immediately after peak stress. These techniques
all offer the potential to further improve the accuracy of stress
echocardiography.
Summary
Stress echocardiography is a well-validated, relatively inexpen-
sive, and widely available means of detecting coronary artery
disease and assessing prognosis. Newer techniques will permit
increased accuracy and quantification of the extent and sever-
ity of ischemia. The test is highly versatile and additionally has
applications to evaluation of valvular heart disease, diastolic
dysfunction, and exertional cardiac symptoms.
 9
Transesophageal Echocardiography
SARINYA PUWANANT, MD, LAWRENCE J. SINAK, MD, and
KRISHNASWAMY CHANDRASEKARAN, MD
Multiplane TEE allows excellent visualization of all cardiac
structures and great vessels. The common clinical indications for
TEE are assessment of complications of myocardial infarction,
detection of aortic dissection, diagnosis of infective endocarditis
and its complications in native and prosthetic valves, determina-
tion of embolic source (left atrial appendage, ventricle, valves,
aorta), visualization of cardiac tumors, evaluation of congenital
heart diseases, and assessment of critically ill patients. TEE is
also a valuable adjunctive imaging method during cardiovascular
surgery and percutaneous cardiac interventions (Table 9.1).
Complications of Acute
Myocardial Infarction
TEE is an excellent diagnostic imaging method in patients with
acute myocardial infarction (Figure 9.1) with heart failure,
hemodynamic collapse, or hypoxia. TEE can identify the cause
of ventricular septal rupture, acute mitral regurgitation from pap-
illary muscle or chordal rupture, acute hypoxia from right-to-left
shunt across a patent foramen ovale, and cardiogenic shock due
to ventricular pump failure or right ventricular infarction.
Aortic Dissection
TEE is the diagnostic imaging method of choice when aortic
dissection is suspected (Figure 9.2). It has a sensitivity of 97%
to 99% and a specificity of 98%. Mortality from aortic dissec-
tion increases at the rate of 1% to 2% per hour; thus, the most
important role of TEE is the reliable, rapid diagnosis of aortic
dissection. It can be performed within 15 minutes at the bedside.
The diagnosis of aortic dissection is based on identification of an
intimal flap, which creates a double lumen in the aorta. TEE can
106
identify all major complications of aortic dissection, including
pericardial effusion, tamponade, aortic regurgitation, and coro-
nary artery dissection. Important information for surgical plan-
ning includes the finding of involvement of the ascending aorta
(type A dissection), the entry site of the intimal tear, the extent of
dissection, the morphologic features of the true and false lumen,
aortic root integrity, and the presence of pericardial hematoma.
TEE provides valuable information about the entire thoracic
aorta; however, visualization of the distal part of the ascending
aorta and aortic arch frequently is obscured by the air-filled left
main bronchus.
Infective Endocarditis and Its Complications
TEE aids in the diagnosis of infective endocarditis (Duke major
criteria) and is useful for detecting complications of native valve
and prosthetic valve endocarditis (Figure 9.3). TEE is more sen-
sitive than TTE for detecting vegetations, perivalvular abscess or
fistula (Figure 9.4), valve leaflet perforation (Figure 9.5), biopros-
thetic valve dehiscence, and paravalvular leakage, all suggestive
echocardiographic signs of infective endocarditis. With use of
high-resolution imaging, TEE is superior to TTE for detecting
small vegetations (1–2 mm), prosthetic valve endocarditis, and
pulmonary valve endocarditis, which can be missed by TTE. The
anterior portion of an aortic prosthesis or of the aortic root is
often better visualized with TTE than TEE. Recent data have
shown that TEE improves the sensitivity of the Duke criteria to
diagnose definite infective endocarditis and seems particularly
useful for evaluation of patients with suspected prosthetic valve
endocarditis. The specificity of TEE for the diagnosis of infec-
tive endocarditis is 85% to 98%. False-positive findings of vege-
tation may be caused by thrombus, suture materials, and pannus
 9 Transesophageal Echocardiography 107

Table 9.1. Summary of Clinical Applications of Transesophageal Echocardiography

Diseases of aorta
Aortic dissection
Anatomical diagnostic information Surgical planning
Intimal flap and mobility Involvement of ascending aorta
Site of intimal tear Entry site and extent of dissection
True and false lumen Morphologic features of the true and false lumen
Size of false lumen Aortic root integrity
Location of major blood flow Pericardial hematoma
Management information Expansion of false lumen
Pericardial effusion and tamponade
Coronary involvement: RWMA
Expansion of false lumen
Aortic regurgitation
Aortic valve and root integrity
Intramural hematoma
Aortic ulcer
Aortic rupture or transection
Aortic atheromatous disease: assessment of extent, complexity, and thickness of plaques
Aortoarteritis syndrome: assessment of aortic root in Marfan syndrome
Infective endocarditis and its complications
Diagnosis
Suspected endocarditis with poor TTE
Persistent bacteremia with negative TTE
Persistent bacteremia in a patient with prosthetic valve
Bacteremia in patients with devices (pacemaker lead, AICD)
Preoperatively for endocarditis
TEE after TTE should be performed in patients with the following:
Prosthetic valves
Virulent organisms
Clinical suspicion of development of perivalvular extension
Congenital defects
Persistent bacteremia and clinical deterioration
Initial TEE should be performed in patients with the following:
Staphylococcus aureus bacteremia with indwelling catheter, pacer lead, AICD lead
Intermediate probability (Duke criteria)
Serial TEE should be performed in patients with the following:
An initial negative TEE but with high clinical suspicion of infective endocarditis
Clinical deterioration or persistent bacteremia for assessment of disease progression
Assessment of prosthetic valves
Mechanism of stenosis (thrombus, leaflet degeneration, pannus)
Mechanism of regurgitation: prosthetic (malfunction of occluder or disk, leaflet degeneration) or periprosthetic (location and severity)
Unexplained anemia (hemolytic RBC morphologic features)
Assessment of native valves
Mitral valve morphologic features (balloon valvuloplasty, mechanism of regurgitation, mass lesions)
Aortic valve morphology (poor TTE with enlarged aortic root, unusually high gradients with normal morphologic features—subaortic membrane, mass
lesions)
Tricuspid valve morphologic features (Ebstein anomaly, mechanism of regurgitation in patients with a device)
Suspected endocarditis with negative TTE or infection with organisms of low virulency
Determination of source of embolism
LAA and aortic thrombus
Spontaneous echo contrast
Intracardiac mass (vegetation, tumor)
Interatrial shunt
Congenital heart disease
Atrial septal defects, Ebstein anomaly, coarctation of the aorta, etc.
Intraoperative TEE
Mitral valve repair
Preoperative data (mechanisms and pathology of valvular dysfunction)
Postoperative data (residual regurgitation, LVOT obstruction from SAM)
Valve replacement
Postoperative valve function and paravalvular regurgitation, aortic valve integrity
Myectomy in patients with hypertrophic cardiomyopathy
Preoperative data (myectomy site, mitral apparatus and location of papillary muscle, mitral valve disease, SAM, subaortic membrane)
Postoperative data (adequacy of myectomy, aortic valve integrity, aortic regurgitation, ventricular septal defect, SAM)
(continued)
108 II Noninvasive Imaging

Table 9.1. (Continued)

Coronary artery bypass grafting (RWMA, ventricular function, position of tips of intra-aortic balloon pump)
Detecting air in ventricular chambers and great vessels
Surgery of the aorta (evaluation of aortic graft anastomosis)
Heart or heart-lung transplantation (evaluation of anastomosis; LV and RV function)
Critically ill patients
Unexplained hypotension (volume status, LV contractility, RV contractility, RWMA, pericardial effusion, pulmonary thromboembolism)
Unexplained hypoxemia (right-to-left shunt across PFO, undetected ASD, shunt from arteriovenous malformation, pulmonary embolism)
Sepsis (endocarditis, line infection)
Evaluation of cardiac function in patients with brain death who are candidates for transplantation
Cardioversion
Identification of thrombus in LA or LAA before cardioversion
Assessment of LV function and LAA function in terms of predictor of successful restoration of sinus rhythm
Interventional cardiology
Transseptal puncture
Guiding the position of catheter during the following:
Radiofrequency ablation
Percutaneous valvular intervention
Transcatheter closure of interatrial communication and VSD
Novel transcatheter LAA occluder device
Percutaneous stent graft of aorta
Stent angioplasty of coarctation of aorta
Endomyocardial biopsy
Pacemaker lead extraction
Abbreviations: AICD, automatic implantable cardioverter-defibrillator; ASD, atrial septal defect; LA, left atrium; LAA, left atrial appendage; LV, left ventricular;
LVOT, left ventricular outflow tract; PFO, patent foramen ovale; RBC, red blood cell; RV, right ventricular; RWMA, regional wall motion abnormality; SAM, systolic
anterior motion; TEE, transesophageal echocardiography; TTE, transthoracic echocardiography; VSD, ventricular septal defect.

in patients with prosthetic valves. Although negative results of
TEE have a negative predictive value of 98% to 100% in native
valve endocarditis and of 90% in prosthetic valve endocarditis,
serial TEE should be considered if clinical suspicion is high. A
false-negative result of TEE can occur because of early infection
when vegetation and abscess are not well formed, unsatisfactory
image quality due to prosthetic valve artifact, or a previously
embolized vegetation.
Even when the results of TTE support the diagnosis of infec-
tive endocarditis, TEE is required for evaluation of the extent and
complications of infective endocarditis, particularly in patients
with virulent organisms, clinical suspicion of perivalvular exten-
sion, specific congenital heart defects, prosthetic valves, aortic
valve endocarditis, persistent bacteremia, and clinical hemo-
dynamic deterioration. Recent studies support initial TEE as a
useful and cost-effective diagnostic approach in patients with
S aureus bacteremia or an intermediate clinical probability of
infective endocarditis.
When TEE shows perivalvular abscess or fistula, mycotic
aortic aneurysm, new valvular dehiscence, or paravalvular leak-
age, cardiac surgery should be considered (Figure 9.6). Other
indications for cardiac surgery are early prosthetic valve endo-
carditis (<2 months after surgery), prosthetic valve endocarditis
associated with heart failure, infective endocarditis not respon-
sive to standard treatment, or infection with virulent organisms
(Staphylococcus aureus) or difficult-to-eradicate organisms
(fungi). Echocardiographic data that relate vegetation size and
the need for surgery with the risk of subsequent embolization are
controversial.
Critically Ill Patients
Many critically ill patients are managed with mechanical ventila-
tion with an endotracheal tube. This both hinders image quality
on TTE and facilitates examination with TEE. TEE is useful in
critically ill patients with unexplained hypotension, hypoxemia,
or systemic and pulmonary embolism (Figure 9.7). In patients
declared brain dead who are donor candidates for organ trans-
plantation, TEE can determine the suitability of the donor organ
for transplant.
Cardioversion
The ACUTE study found that cardioversion in patients with atrial
fibrillation can be safely performed when TEE shows an absence
of left atrial or left atrial appendage thrombus (Figure 9.8). This
is also a more clinically effective alternative strategy to conven-
tional anticoagulation therapy for several weeks before elective
cardioversion. In addition to the safety consideration, TEE is
helpful for identifying patients who are most likely to recover
sinus rhythm. Left ventricular function and left atrial appendage
anatomy and function assessed by 2-dimensional and Doppler
TEE have been reported to accurately predict rhythm restoration
and maintenance of sinus rhythm after cardioversion.
Intraoperative TEE
Intraoperative TEE is now widely used before, during, and after
cardiac surgery. It is particularly useful in mitral valve repair, for
providing information to aid the surgical plan, including preoper-
ative details of the mechanisms of regurgitation and valve abnor-
mality (Figure 9.9), and for providing postoperative details of
residual regurgitation (Figure 9.10) and systolic anterior motion.
Additionally, after aortic or mitral valve replacement, intraopera-
tive TEE can recognize prosthetic valve dysfunction and deter-
mine the significance of paravalvular regurgitation. It also has
been recommended to help determine the extent and site of ven-
tricular myectomy in patients with hypertrophic cardiomyopathy
 9 Transesophageal Echocardiography 109

Figure 9.1. Patient With Inferior Myocardial Infarction and Heart Failure. A, Multiplane transesophageal echocardiogram, showing complex
intramyocardial rupture (arrows) through posterior wall of left ventricle (LV) resulting in acquired ventricular septal defect. B, Transgastric short-
axis view, showing disruption of posterior wall of LV (open arrows) that extends through posterior wall of right ventricle (RV) and septum with
left-to-right shunt. LA indicates left atrium.
110 II Noninvasive Imaging

Figure 9.2. Transesophageal Echocardiogram Obtained at Bedside on an Emergency Basis for Patient With Shock and Chest Pain. Basal short-
axis view at the 158° transducer position, showing type A aortic dissection with a mobile intimal flap separating true lumen from false lumen,
and intimal flap (arrowheads) extending into ostium of right coronary artery (arrow). Small amount of pericardial effusion is also seen (asterisk).
Ao indicates ascending aorta; LA, left atrium; LV, left ventricle; RVOT, right ventricular outflow tract.

(Figure 9.11). Furthermore, it is helpful for identifying mitral
valve structural abnormalities associated with hypertrophic car-
diomyopathy, which may require valve repair, and complications
associated with postoperative myectomy, including aortic regur-
gitation and ventricular septal defect. In patients undergoing
coronary artery bypass grafting, intraoperative TEE is helpful
for defining chamber size and function (to help guide fluid and
drug administraton) and for evaluating segmental regional wall
motion abnormalities. In addition, intraoperative TEE is par-
ticularly useful for detecting air in the left ventricle and aorta
immediately after cardiopulmonary bypass surgery, which if
undetected can lead to myocardial ischemia due to coronary
artery air embolization. Moreover, in high-risk patients under-
going coronary artery bypass grafting in which an intra-aortic
balloon pump is used, intraoperative TEE is helpful for position-
ing the tip of the balloon pump in the aorta. It is also valuable in
the assessment of anastomoses of great vessels in surgery of the
aorta and heart-lung transplantation.
Cardiovascular Interventions
TEE-guided cardiovascular intervention results in a substan-
tial reduction of fluoroscopic X-ray exposure to patients and
operator. TEE is useful for guiding the position of the cath-
eter during radiofrequency ablation, transseptal puncture,
balloon valvuloplasty, novel percutaneous valvular interven-
tion, novel percutaneous left atrial appendage transcatheter
occlusion (Figure 9.12), and transcatheter device closure of
interatrial communication and ventricular septal defect.
Furthermore, TEE is helpful for guiding the placement of
percutaneous stent graft of the aorta and stent angioplasty of
coarctation of the aorta. Sporadically, TEE has been used as a
guide during high-risk endomyocardial biopsy and pacemaker
implantation.
Determination of Sources of Embolism
Ischemic stroke is the leading cause of morbidity and mortal-
ity. TEE permits excellent visualization of potential thromboem-
bolic sites: the left atrial appendage, the arch of the aorta, and
the descending thoracic aorta. In addition, it can image the atrial
septum for possible shunts, which have been implicated in para-
doxical embolism. However, the yield of TEE for detecting the
cause of embolism is high in younger individuals. Improved vis-
ualization by TEE allows easy recognition of potential sources
of embolism, including small thrombi, cardiac tumors such as
 9 Transesophageal Echocardiography 111

Figure 9.3. Transesophageal Echocardiograms, Long-Axis View, From Patient With Bicuspid Aortic Valve and Persistent Bacteremia.
End-diastolic (A) and end-systolic (B) frames, showing oscillating vegetation on right coronary cusp of aortic valve (arrow). Ao indicates aorta;
LA, left atrium; LV, left ventricle.
112 II Noninvasive Imaging

Figure 9.4. A, Transesophageal echocardiogram, short-axis view, from patient with aortic prosthetic valve endocarditis, showing abscess cavity
in base of anterior septum (asterisk). B, Color Doppler study, showing 4-mm communication (arrows) from anterior aortic root just below prosthetic
valve into abscess cavity. Ao indicates aorta; LA, left atrium; RA, right atrium; RV, right ventricle.
 9 Transesophageal Echocardiography 113

Figure 9.5. A, Intraoperative transesophageal echocardiogram, long-axis view systolic frame, from patient undergoing mitral valve operation for
regurgitation, showing prolapse of myxomatous mitral valve and perforated (arrowheads) mycotic aneurysm of anterior mitral leaflet from healed
endocarditis. B, Color Doppler study, showing severe mitral valve regurgitation through perforation (arrows). Ao indicates aorta; AML, anterior
mitral leaflet; LA, left atrium; LV, left ventricle; PML, posterior mitral leaflet.
114 II Noninvasive Imaging

Figure 9.6. Transesophageal echocardiogram, long-axis view, showing large mitral aortic intervalvular fibrosa pseudoaneurysm (arrowhead)
below aortic valve prosthesis (arrows) in a patient who had aortic valve replacement for aortic valve endocarditis. Asc.Ao indicates ascending aorta;
LA, left atrium; LV, left ventricle; MV, mitral valve; RVOT, right ventricular outflow tract.

Figure 9.7. Transesophageal echocardiogram, basal short-axis

view, for critically ill patient with hypotension, showing large throm- Figure 9.8. Transesophageal echocardiogram obtained before car-
bus at bifurcation of main pulmonary artery and right pulmonary artery dioversion in patient with atrial fibrillation, showing large thrombus
(arrowhead). Ao indicates aorta; RPA, right pulmonary artery; SVC, (arrowheads) in left atrial appendage (LAA) with spontaneous echo
superior vena cava. contrast (arrows). LA indicates left atrium; PA, pulmonary artery.
 A

Figure 9.9. A, Intraoperative transesophageal echocardiogram, showing mitral valve prolapse with flail (arrows) middle scallop of posterior leaf-
let. B, Color Doppler study, showing severe mitral valve regurgitation. Ao indicates aorta; LA, left atrium; LV, left ventricle; RV, right ventricle.

A B

Figure 9.10. A, Transesophageal echocardiogram, systolic frame, obtained after bypass for same patient described in Figure 9.5, showing repair
of anterior mitral valve perforation and ring anuloplasty (arrow). B, Color Doppler study, showing no residual mitral valve regurgitation. Ao indicates
aorta; LA, left atrium; LV, left ventricle.
115
116 II Noninvasive Imaging

A B

Figure 9.11. A, Preoperataive transesophageal echocardiogram, long-axis systolic frame, in patient with hypertrophic cardiomyopathy, showing
markedly increased thickness of basal septum (double-headed arrow) and systolic anterior motion of mitral valve (arrow). B, Color Doppler study,
showing systolic flow turbulence in left ventricular outflow tract (asterisk) and eccentric mitral regurgitation, a predominant posterior-directed jet
(arrowheads), and a less severe anterior jet (open arrows). Ao indicates aorta; LA, left atrium; LV, left ventricle; RVOT, right ventricular outflow tract.

myxoma (Figure 9.13), papillary fibroelastoma, vegetations, and commonly on the atrial side of the prosthesis; hence, TEE is
interatrial septum abnormality, especially patent foramen ovale essential to identify them. Prosthetic valve obstruction in the
and atrial septal defect. Previous studies have demonstrated that a aortic position often results from pannus, which may be difficult
protruding, noncalcified aortic plaque ≥4 mm detected by TEE is a to separate from the housing unit by TTE, and TEE can define
significant risk factor of ischemic stoke and peripheral embolism. them more easily in most cases. TEE is essential to define the
With TEE, those atheroma as well as aortic plaque compositions site and mechanism of regurgitation for both aortic and mitral
can be easily detected and visualized (Figure 9.14). prostheses (Figure 9.15). Furthermore, in cases of suspected
endocarditis, TEE is essential not only to confi rm the diagnosis
but also to identify high-risk patients, those with abscesses and
Prosthetic Valve Evaluation fistulae.
Clinically suspected prosthetic valve malfunction commonly
requires both TTE and TEE. TTE provides hemodynamic
information indicating the dysfunction; however, delineating A
the mechanism of dysfunction often requires TEE. Although
the ventricular aspect of the mitral and aortic prosthesis can
be well visualized by TTE, the atrial and aortic aspects of the
prosthesis are difficult to visualize by TTE because of ultrason-
ographic artifacts as well as masking. Prosthetic valve obstruc-
tion often results from thrombus in the mitral position and is

Figure 9.13. Transesophageal echocardiograms, showing left atrial

myxoma (asterisk) attached to interatrial septum in short-axis (A) and
Figure 9.12. Transesophageal Echocardiogram, Showing Percu- four-chamber (B) views. Ao indicates aorta; LA, left atrium; LV, left
taneous Left Atrial Appendage (LAA) Occluder Device (arrows). ventricle; PA, pulmonary artery; RA, right atrium; RV, right ventricle.
 9 Transesophageal Echocardiography 117

Figure 9.14. Transesophageal echocardiography in a patient with a transient ischemic attack demonstrates a 5-mm-thick complex atherosclerosis
(double-headed arrow) of the proximal descending thoracic aorta (Ao) (A) and a highly mobile thrombus (arrows) just distal to the left subclavian
artery origin (B).

Congenital Heart Diseases
TEE is helpful in evaluation of complex congenital as well as sim-
ple congenital lesions, such as atrial septal defect (Figure 9.16),
the cleft of atrioventricular valves, and anomalies of the aorta
associated with bicuspid aortic valve. In addition, it is essential
in previously operated patients to identify the residual defects,
right ventricular outflow tract or pulmonary obstruction, the
patency of the conduits and baffles, the patency of the superior
vena cava or inferior vena cava, especially in the post-Fontan
procedure, and the function of an atrioventricular valve pros-
thesis in individuals who require atrioventricular valve replace-
ment. Furthermore, in post-Fontan patients, right pulmonary
vein compression can occur from an enlarged right atrium or
atrial baffle bulging into the left atrium. In those patients, TEE
is superior to TTE in the assessment of the pulmonary venous
flow pattern.
118 II Noninvasive Imaging

Figure 9.15. A, Transesophageal echocardiographic diastolic frame in a patient with mitral valve replacement who had recurrent heart failure.
A mobile thrombus (arrow) is on the ventricular surface of the prosthesis. Arrowheads indicate normal open position of both leaflets of St. Jude
mechanical prosthesis. B, Color Doppler image demonstrates that the thrombus protrudes in and out, causing intermittent impaired closure of the
lateral orifice of the mitral prosthetic valve, resulting in moderate to severe mitral prosthetic regurgitation (open arrows). LA indicates left atrium;
LV, left ventricle; RA, right atrium; RV, right ventricle.
 9 Transesophageal Echocardiography 119

A B

Figure 9.16. A, Transesophageal echocardiography in a patient with a heart murmur demonstrates a sinus venosus atrial septal defect (ASD)
(arrow) in the superior portion of the interatrial septum (IAS) near the orifice of the superior vena cava (SVC) (asterisk). B, Color Doppler shows a
left-to-right shunt through the ASD and the flow from the pulmonary vein draining into the posterior aspect of the SVC (asterisk). C, Contrast echo-
cardiography demonstrates a negative contrast created by the flow from the left atrium (LA) to the right atrium (RA) across the ASD and by the right
upper pulmonary vein flow draining into the SVC (asterisk).

Abbreviations Clinical Trial

TEE transesophageal echocardiography ACUTE Assessment of Cardioversion Using Transesophageal
TTE transthoracic echocardiography Echocardiography

Nuclear Imaging
J. WELLS ASKEW III, MD, and TODD D. MILLER, MD
Nuclear imaging techniques are commonly used to evaluate
myocardial perfusion at rest and with stress, to measure left ven-
tricular function and volume, and to determine the presence or
absence of myocardial viability.
Radionuclide Imaging
Radiopharmaceuticals
Nuclear cardiology is based on the administration of a radiophar-
maceutical consisting of a radionuclide (isotope) with or without
a complexing agent (depending on the type of study) in order
to image the cardiovascular system. A radionuclide is an unsta-
ble element that decays spontaneously and, as a result, emits
energy in the form of radiation or charged particles. Radiation
that is emitted from the decay of the nucleus of an atom is called
gamma radiation. With some isotopes, the radiation is reabsorbed
by the orbiting electrons and then re-emitted as x-rays. Gamma
rays are high-energy electromagnetic radiation. By comparison,
light is considered low-energy electromagnetic radiation. The
energy of a gamma ray is described in units of electron volts,
typically between 50,000 and 200,000 electron volts (abbrevi-
ated as 50–200 keV). Radiopharmaceuticals also may consist of
a complexing agent (sestamibi or tetrofosmin) that helps to facil-
itate stabilization, biodistribution, and delivery of the radionu-
clide to the intended target. If the emitted gamma rays from the
administered radionuclide are not absorbed or scattered by soft
tissue or bone and pass out of the body, they can be detected and
used to create an image.
Abbreviations and acronyms are expanded at the end of this chapter.
120
10
Radioisotopes
Two important radioisotopes (Table 10.1) used in nuclear cardi-
ology are 201Tl and 99mTc.
201
Tl is a metallic element in group IIIA of the periodic table
and is a cyclotron-generated, monovalent cation with properties
similar to that of potassium. 201Tl is rapidly extracted from the
blood (extraction fraction close to 85%), enters the myocardium
by using the Na-K-ATPase pump, and has an initial distribution
proportional to that of myocardial blood flow and an equilibrium
distribution proportional to that of potassium. 201Tl displays the
property of redistribution; after its initial extraction, there is a
continuous exchange involving the myocyte and the extracellu-
lar space. The rate of “wash in” and “wash out” is largely due
to coronary blood flow. The maximal percentage of myocardial
uptake of the original dose is between 4% and 5%. 201Tl decays
to mercury 201 by electron capture. 201Tl emits some gamma rays
with energies of 135 and 167 keV; however, the principal (most
abundant) radiation emissions are x-rays emitted in the range of
69 to 83 keV. The relatively low-energy 80-keV protons are more
susceptible to scattering and attenuation. The long physical half-
life (73 hours) limits the ability to administer higher doses of
201
Tl. The low x-ray energy and low administered dose can result
in a reduction of image quality.
99m
Tc, a lipophilic monovalent cation, is formed through the
decay of 99Mo, which is available commercially from a 99Mo gen-
erator. The “m” in 99mTc is present because it is metastable, even-
tually decaying to 99Tc. Sodium 99mTcO4 is obtained by passing a
sodium chloride solution through the 99Mo generator to elute the
99m
Tc continuously being formed by the constant decay of 99Mo
(half-life, 66 hours). Because 99mTc is acquired by eluting a 99Mo
generator, an important quality control step that should be per-
formed the first time the generator is eluted is to ensure that very
 10 Nuclear Imaging 121

Table 10.1. Isotope Characteristics

Isotope

Characteristic Thallium 201 Technetium 99m sestamibi

Class Element Isonitrile
Charge Cation Cation
Production Cyclotron generated Molybdenum-99m generator
Half-life, h 73 6
Extraction fraction, % 85 65
Maximal myocardial uptake, % injected dose 3–5 1.2–1.5
Mechanism of myocardial uptake Na-K-ATPase pump Mitochondrial-derived gradient
Whole-body radiation dose, mrad/mCi 240 16
Gamma rays/x-rays (photopeak), keV 68–80, 135, 167 140
Myocardial redistribution Yes Minimal

little 99Mo is eluted from the generator. Because of the long half-
life of 99Mo, too much 99Mo results in an increased radiation dose
to the patient. The permissible amount must be less than 0.15 mCi
99
Mo/mCi 99mTc. 99mTc has a lower myocardial extraction fraction
(65%) than 201Tl and is sequestered in the cardiac myocyte by the
mitochondria. Uptake of both 99mTc sestamibi and tetrofosmin
plateaus at a lower myocardial blood flow rate than 201Tl. In con-
trast to 201Tl, 99mTc does not seem to be dependent on the Na-K-
ATPase pump and undergoes minimal redistribution. The decay
of 99mTc emits a higher-energy gamma ray with a photopeak of
140 keV. The shorter half-life (6 hours) and the higher-energy
photon (140 keV) improve image resolution through the ability to
administer higher doses and limit photon scatter and attenuation.
The ability to use a higher-energy gamma ray can limit the num-
ber of scattered photons, thereby accepting a greater percentage
of valid counts, and consequently image quality is improved and
gating is enhanced.
sensitivity, improved energy and spatial resolution, and faster
image acquisition times.
A separate detection system occasionally used in nuclear car-
diology is the multicrystal camera, which contains several sodium
iodide crystals with separate photomultiplier tubes. The multi-
crystal camera allows for high count sensitivity but low spatial
resolution. The improved temporal resolution of the multicrystal
camera enables its use for first-pass imaging (described below).
Numerous computer programs are available for display and
analysis of the planar/projection images and the reconstructed
tomographic slices. The standard planes (short axis, horizontal
long axis, and vertical long axis) used for analysis of the recon-
structed tomographic slices are shown in Figure 10.1. The recon-
structed short-axis slices typically are divided into 16 segments
and can be assigned to coronary artery perfusion territories
(Figure 10.2).

Instrumentation
In nuclear cardiology, the most commonly used device to detect RV LV
photons is the gamma camera. The gamma camera consists of
multiple components with the primary function of detecting
photons and converting that energy into an electrical current. Short axis
The primary components of the gamma camera are the colli-
Apex
mator, the sodium iodide crystal, and the photomultiplier tubes.
The collimator serves as a shield by allowing photons travel-
ing at an approximate 90° angle to pass through; image reso- RV
lution is thereby improved in the intended area of interest. Not Lateral
all photons passing through the collimator result in an accurate
tum

image because photons from other areas may be scattered and

Sep

ultimately pass through the collimator. In an effort to correct Base

these events, an energy window is used to identify scattered pho-
tons, because they typically will have a lower-energy level and Horizontal long axis
thus can be identified and excluded from analysis. Appropriate
incoming photons strike the sodium iodide crystal and are con- Anterior
verted to visible light. The light strikes the photomultiplier tube
and is amplified and converted to a measurable electrical signal.
With the aid of computer processing, the spatial image of interest Base Apex
is reconstructed.
Advancements in software using newer iterative reconstruc- Inferior
tion algorithms and resolution recovery can reduce imaging
Vertical long axis
time without degradation in image quality. New ultrafast, solid-
state cardiac cameras using cadmium-zinc-telluride detectors Figure 10.1. Standard imaging planes used in nuclear cardiology:
as opposed to sodium iodide crystals are also available. These short axis, horitzontal long axis, and vertical long axis. LV indicates left
advancements in camera technology provide increased photon ventricle; RV, right ventricle.
122 II Noninvasive Imaging
Anterior
Anteroseptal
Anterolateral
Base
Inferolateral
Inferoseptal
Inferior
Anterior
Anteroseptal
Anterolateral
Mid
Inferolateral
Inferoseptal
Inferior
LAD
RCA
Anterior LCX
Apex Septal Lateral
Inferior
Figure 10.2. Nomenclature and location of the 16 myocardial seg-
ments and assignment of coronary arterial perfusion territories on the
short-axis model. LAD indicates left anterior descending coronary
artery; LCX, left circumflex coronary artery; RCA, right coronary
artery.
Gated SPECT
Electrocardiogram-gated SPECT is an accurate and reproduci-
ble technique to measure LVEF and LV volume. In addition to
quantifying LV function and volume, gating permits the assess-
ment of regional wall motion. Performing a gated SPECT study
requires the typical equipment used in acquiring a standard
SPECT image set with the addition of a 3-lead electrocardio-
gram-gating device. Each cardiac cycle is identified using the
RR interval. A gating rate is set (typically 8 frames for each
RR interval). All of the counts acquired at each projection are
placed into a designated bin (frame) that corresponds to the time
the counts were acquired during the cardiac cycle. The pre-set
RR interval can remain the same for the entire study (fixed
acquisition mode) or can be monitored and recalculated during
each projection (variable acquisition mode). Because the heart
rate may be dynamic, acceptance windows can be specified to
reject cycle lengths that do not fall within a predetermined range
of the expected RR interval. Use of acceptance windows allows
for improved accuracy and reliability of the gated information
and avoids potential perfusion image artifacts (flashing, blur-
ring, and streaking) related to arrhythmias. In contrast to the
count-based technique used to calculate EF by blood-pool imag-
ing, gated SPECT is based on volume. Computer algorithms are
applied to identify the endocardial borders. EDV and ESV are
measured, and LVEF can then be calculated, as in the following
equation:
LVEF (%) = (EDV − ESV)/EDV × 100
Limitations to quantifying LVEF can occur in patients with a
small LV cavity, in that the ESV is underestimated because of
obliteration of the LV cavity, with subsequent overestimation of
the LVEF.
Myocardial Perfusion Imaging
Stress SPECT is the standard for assessing myocardial perfusion
in clinical practice. Stress SPECT has been extensively validated
for diagnostic and prognostic purposes. The hallmark of myocar-
dial ischemia is a reversible defect, which may be either partial
or complete (Figure 10.3). Severe fixed defects represent myo-
cardial infarction (Figure 10.4). Mild fixed defects can represent
nontransmural infarction or attenuation artifacts. Assessment of
regional wall motion on the gated images can help resolve this
issue. SPECT imaging is a relative perfusion technique; because
myocardial blood flow during stress is reduced in regions sub-
tended by hemodynamically significant stenoses, the radioisotope
tracer content in these regions also is reduced. Global reduction
in flow due to 3-vessel coronary artery disease (balanced ische-
mia) can result in normal perfusion images. Available data sug-
gest that this occurs uncommonly in clinical practice and can be
detected by other variables (see below).
Even though assessment of myocardial perfusion is the pri-
mary reason for performing SPECT studies, other important
findings can be seen on the planar/rotating images. Cardiac size
at rest and after stress can be qualitatively assessed. Transient
ischemic dilatation refers to the apparent increase in LV size on
the stress images compared with the rest images and can be meas-
ured qualitatively or quantitatively (Figure 10.5). Transient ische-
mic dilatation implies extensive myocardial ischemia or severe
(left main or multivessel) coronary artery disease. Severe stress-
induced ischemia may result in prolonged wall motion abnormal-
ities visible on the post-stress gated images. Isotope uptake in the
right ventricle can occasionally be seen and allow for evaluation
of right ventricular enlargement and hypertrophy (Figure 10.6).
Increased lung tracer uptake (lung-heart ratio >0.5) with 201Tl is
suggestive of increased LV filling pressures, LV dysfunction, or
extent of ischemia (Figure 10.7). Additional noncardiac findings
can relate to areas of increased isotope uptake (thyroid disease,
ectopic parathyroids, lymphomas, breast and lung malignancies)
and, conversely, reduced isotope uptake (ascites, cysts, pleural
and pericardial effusions).
Types
The most common forms of stress testing performed in con-
junction with SPECT imaging include treadmill exercise or
pharmacologic stress (adenosine, dipyridamole, dobutamine,
or regadenoson). Pharmacologic stress tests are performed in
patients unable to adequately exercise or in patients with selected
electrocardiographic findings (left bundle branch block or paced
ventricular rhythm).
Adenosine increases myocardial blood flow through its inter-
action with adenosine receptors (A1, A2A, A2B, A3) in which
activation of the A2 receptors results in coronary and systemic
vasodilatation. Regadenoson is a selective A2A receptor agonist
that increases myocardial blood flow through coronary vasodi-
latation and may limit the undesirable adverse effects mediated
by activation of the other adenosine receptors, such as chest pain,
flushing, atrioventricular block, and bronchoconstriction.
Dipyridamole exerts its vasodilatory actions by blocking the
intracellular transport of adenosine, thus raising extracellular
 10 Nuclear Imaging 123

Figure 10.3. Stress images (left) and rest images (right) show a large area of ischemia (involving the apex, septum, anterior, and anterolateral
segments) on the short-axis (A), horizontal long-axis (B), and vertical long-axis (C) views.

adenosine levels which in turn activate A2 receptors, resulting in
coronary vasodilatation.
Potential adverse effects of adenosine and dipyridamole
include light-headedness, headache, flushing, nausea, chest
discomfort, abdominal discomfort, and dyspnea. More seri-
ous adverse effects can include hypotension, profound brady-
cardia, second- or third-degree heart block, and bronchospasm
(these effects usually resolve with decreasing the dose, discon-
tinuation of the infusion, or administration of aminophylline).
Regadenoson typically produces adverse effects similar to those
of adenosine but of lesser severity because of its affinity for the
A2A adenosine receptors. Changes on the electrocardiogram can
occur with adenosine. ST-segment depression (≥1 mm) is sugges-
tive of ischemia and may indicate severe (left main or 3-vessel)
coronary artery disease.
In patients with contraindications to vasodilating agents,
dobutamine can be used. Dobutamine is an inotropic agent that
exerts its catecholamine effect of increasing heart rate and con-
tractility through activation of cardiac adrenergic receptors.
Potential adverse effects include chest pain, dyspnea, flushing,
palpitations, and nausea.
Indications and relative contraindications for adenosine,
dipyridamole, and dobutamine, and regadenoson are listed in
Table 10.2. Protocols for administering adenosine, dipyridamole,
dobutamine, and regadenoson vary. Standard protocols include
adenosine infusion of 140 mcg/kg per minute for 6 minutes with
radioisotope injection 3 minutes into the infusion, dipyridamole
infusion of 0.56 mcg/kg over 4 minutes with radioisotope injec-
tion 3 to 4 minutes after infusion is complete, stepwise titra-
tion of dobutamine up to 40 to 50 mcg/kg per minute with the
124 II Noninvasive Imaging

Figure 10.4. Stress images (left) and rest images (right) show a large, dense infarction (involving the lateral and inferior walls) on the short-axis
(A), horizontal long-axis (B), and vertical long-axis (C) views.

addition of atropine (to a total dose ≤2 mg) as needed to achieve
85% of the age-adjusted maximal heart rate, and regadenoson
intravenous bolus administration of 400 mcg (<10 seconds) fol-
lowed by saline flush and radioisotope injection 10 to 20 seconds
later.
Imaging Protocols
Various imaging protocols can be used, including same-day
rest-stress or stress-rest imaging with a single isotope, a 2-day
protocol with the same isotope, or a dual-isotope study in which
the rest imaging is performed with one isotope and the stress
imaging, with another (201Tl, 99mTc sestamibi). Acquisition pro-
tocols vary but typically consist of acquiring 60 to 64 projec-
tions at 15 to 20 seconds per projection over a 180° arc (45° right
anterior oblique to 45° left posterior oblique) with the patient in
the supine position. Prone imaging is occasionally performed in
addition to supine imaging because it can minimize diaphrag-
matic attenuation artifacts by increasing the separation of the
inferior wall and the diaphragm.
 10 Nuclear Imaging 125

Figure 10.5. Stress images (left) and rest images (right) show transient ischemic dilatation of the left ventricle on the planar projection (A) and
short-axis (B) view. Coronary angiography showed multivessel coronary artery disease.

Figure 10.6. Severe right ventricular enlargement and hypertrophy noted on the planar projection (A) and short-axis (B) image.
126 II Noninvasive Imaging

Figure 10.7. Thallium planar projection image with increased pulmonary uptake (lung-heart ratio, 0.66).

The total acquisition time depends on the number and con-
figuration of the detectors, the number of degrees between each
projection, and the length of acquisition time at each projection.
The total acquisition time to obtain either a rest or a stress image
with a conventional sodium iodide gamma camera (dual-head
detectors) with a 180° orbit and 60 projections at 20 seconds per
projection and standard software can be under 12 minutes. The
new ultrafast cadmium-zinc-telluride cameras allow completion
of image acquisition in 2 minutes. Time from injection of the
isotope to image acquisition varies according to the imaging pro-
tocol, the type of isotope (shorter times for 201Tl), and the method
of stress (shorter times with exercise). These times vary and are
in an effort to minimize artifacts by allowing for activity in the
liver to decrease yet avoiding the impact of subdiaphragmatic
activity from isotope activity in the stomach and intestines.
Several types of artifacts can occur and may be related to
the patient and the imaging equipment. Patient-related artifacts
include attenuation from soft tissue, breast, chest wall, dia-
phragm, and overlapping visceral activity. Patient movement can
occur during the acquisition of images and may result in ver-
tical or horizontal motion artifacts. A phenomenon referred to
as “upward creep” has been described with exercise 201Tl stress
studies when the stress images are acquired soon after exercise.
Immediately after exercise, diaphragmatic excursion is increased
because of the depth of respiration; as the depth of respiration
decreases, the position of the diaphragm rises in the chest with a
resulting upward movement of the heart during the acquisition of
images, giving rise to the diaphragmatic “creep” artifact. Septal
artifacts can be seen with left bundle branch block or paced
rhythms. The dyssynchronous septal motion due to left bundle

Table 10.2. Indications for Choosing Adenosine, Dipyridamole, Regadenoson, or Dobutamine

Indication Adenosine Dipyridamile Regadenoson Dobutamine
a a b
Asthma No No No Yes
COPD Noa Noa Perhapsa,b Yes
Patient taking dipyridamole No Yes No Yes
Patient taking theophylline Noc Noc Noc Yes
High-grade AV block No No No Yes
Caffeine use (<24 h) Nod Nod Nod Yes
Left bundle branch block Yes Yes Yes No
Abbreviations: AV, atrioventricular; COPD, chronic obstructive pulmonary disease.
a
May be tolerated if forced expiratory volume in 1 second is more than 40 L/s and bronchodilator change less than 31%.
b
Limited data available.
c
May use if theophylline use is discontinued 48 hours or more before the test.
d
May use; however, potential problem of false-negative results.
 10 Nuclear Imaging 127

branch block seen in many patients at rest is intensified with the

heart rate acceleration that occurs with exercise. Studies suggest Basal
that septal blood flow may be altered. Substituting a pharma- Anterolateral
cologic vasodilator stress agent (dipyridamole, adenosine, or
regadenoson) can reduce this problem by avoiding significant Mid
tachycardia and is generally recommended in patients with a left Basal Anterior
bundle branch block or paced rhythm.
Inferoseptal
Mid
Assessment of LV Size and Function
Equilibrium radionuclide angiography, commonly known as Apical
MUGA, is an imaging technique of the intravascular blood pool
using an isotope. The accuracy and reproducibility of the tech-
nique as a noninvasive assessment of ventricular function have
made it a valuable imaging tool for many years. 99mTc is the
most commonly used radioisotope for equilibrium radionuclide Septal Lateral
angiography. If images that reflect the intravascular blood volume
are to be obtained, the isotope (99mTc) must be linked to a carrier
that will remain within the circulation and be uniformly distrib- LAO
uted. Currently, red blood cells are used as the carrier. Labeling
the patient’s red blood cells with 99mTc can be accomplished with
Inferoseptal Inferolateral
either in vitro or in vivo techniques. Both techniques require the
use of a reducing agent (stannous ions) that helps facilitate red
blood cell binding to 99mTc by providing a lower oxidation state. Inferoapical
After labeling of the red blood cells, a gamma camera is used to
acquire the images. An electrocardiographic monitoring device is
used for R-wave triggering, which allows each cardiac cycle to be
timed. Each cardiac cycle is divided into “frames” (usually between Basal
16 and 32). The summation of data from the same frames of each Anterior
cardiac cycle can be processed and displayed for review, provid-
ing images of the cardiac chambers acquired over multiple cardiac Mid
cycles. Typically more than 2 million counts are needed from the
Left
field of view for acceptable image resolution, accurate assessment lateral
Basal
of wall motion, and reproducible quantification calculations, requir-
ing the sampling of many (>200) cardiac cycles. As with all gating
Inferior
techniques, a fairly regular heart rate is required. Orientation and Mid
separation of the cardiac chambers are important for assessing wall
Apex
motion and quantitating ventricular volumes and EFs because all Figure 10.8. Standard views (anterior, left anterior oblique [LAO],
chambers will contain radiolabeled red blood cells. left lateral) and nomenclature of the left ventricle for equilibrium radio-
Standard views of the left ventricle at rest include anterior, nuclide angiography.
left anterior oblique (45°), and left lateral (Figure 10.8). The left
anterior oblique view results in optimal separation of the ven-
tricular chambers. Stress equilibrium radionuclide angiography left atrium, left ventricle, and ascending aorta with a multicrystal
can be performed during cycle ergometry exercise with 3-min- camera to enhance temporal resolution of the bolus as it passes
ute stages. Usually, only one left anterior oblique view is used through each region of interest. The most common application of
because of acquisition times at each exercise level (approxi- the first-pass technique is quantification of LVEF. Other applica-
mately 2 minutes) and the importance of measuring LVEF dur- tions include quantification of right ventricular EF, assessment
ing exercise. Equilibrium radionuclide angiography can be used of LV wall motion, quantification of ventricular volumes, and
to assess ventricular volumes, LVEF, regional wall motion, right detection of right-to-left shunts.
ventricular function, and diastolic variables (early diastolic fill-
ing, rate of peak filling, and time to peak filling). Because the
measurements are count-based and do not require assumptions Clinical Applications
about cardiac (ventricular) geometry, the technique provides an Chronic Coronary Artery Disease
accurate, reproducible, and operator-independent assessment of (Known or Suspected)
ventricular function.
First-pass radionuclide angiography, although technically dif- Diagnostic Assessment
ferent from equilibrium radionuclide angiography, is an alterna- Stress SPECT is most beneficial for diagnostic purposes in
tive to quantifying ventricular function. Red blood cell labeling patients at intermediate risk for coronary artery disease (Bayes
is not required because the radioisotope (commonly 99mTc) is theorem) who are not candidates for standard exercise tread-
directly administered as a bolus injection into the venous circu- mill testing because of an inability to exercise or baseline
lation and recorded during its first pass through the circulation. electrocardiographic abnormalities, including preexcitation
The radioisotope bolus is tracked from the superior vena cava (Wolff-Parkinson-White syndrome), electronically paced ven-
through the right atrium, right ventricle, pulmonary circulation, tricular rhythm, left bundle branch block, or more than 1 mm
128 II Noninvasive Imaging
Table 10.3. Sensitivity and Specificitya of Myocardial Per-
fusion SPECT for Detecting Coronary Artery Disease
Test Sensitivity, % Specificity, %
Exercise SPECT 87 73
Vasodilator SPECT 89 75
Abbreviation: SPECT, single-photon emission computed tomography.
a
Because of the impact of referral bias on specificity, the normalcy rate is
occasionally used and refers to the frequency of normal test results in patients
with a low likelihood of coronary artery disease (91% for stress SPECT).
of resting ST-segment depression. Published values are 85% to
90% for sensitivity and 70% to 75% for specificity for the detec-
tion of angiographically significant coronary artery disease
(Table 10.3). These values have not been corrected for referral
bias, also known as verification bias. Referral bias describes the
effect of primarily referring patients with abnormal test results to
angiography to verify the results of the stress test. The number of
patients in the angiographic cohort of patients (in whom sensitiv-
ity and specificity are determined) with positive test results (both
true-positive and false-positive) greatly exceeds the number of
patients with negative test results (both true-negative and false-
negative). The net impact of referral bias is overestimation of test
sensitivity and underestimation of test specificity.
Similar sensitivity and specificity values (also not corrected
for referral bias) have been published for stress echocardiography.
A major advantage of stress imaging procedures over exercise
electrocardiography is higher sensitivity. Advantages of stress
SPECT over stress echocardiography include obtaining quality
images in a greater percentage of patients despite body habitus
or other confounding factors (eg, chronic obstructive pulmonary
disease) and greater accuracy for detecting ischemia in the pres-
ence of resting wall motion abnormalities or dyssynchronous
Figure 10.9. Thallium viability study (images from left to right represent time 0, 4 hours, and 24 hours), showing viability of the lateral wall on
the short-axis (A) and horizontal long-axis (B) views.
contraction patterns related to left bundle branch block or paced
ventricular rhythm.
Prognostic Assessment
Stress SPECT has the largest published prognostic database of
any noninvasive stress imaging technique. The prognostic value
of stress SPECT imaging has been well validated in numerous
studies. Many prognostic variables have been identified, most
importantly LVEF and the extent and severity of the perfusion
defect. Patients with normal perfusion images, in general, are
at very low risk and have an annual “hard” event rate (cardiac
death or nonfatal myocardial infarction) of less than 1%. This
statement may not apply to selected subsets of patients, including
those with diabetes and those undergoing pharmacologic stress,
especially if they have ischemic electrocardiographic changes
with adenosine or dipyridamole. In patients with abnormal
images, the risk of a cardiac event is proportional to the degree
of abnormality. Both the size and the severity of the perfusion
defect are prognostically important and have been incorporated
into the calculation of summed scores. The summed stress score
is a reflection of both infarcted and ischemic myocardium and
has modestly greater prognostic value than the summed differ-
ence score (the extent and severity of ischemic myocardium).
The prognostic information obtained with stress SPECT adds
incremental value over clinical and exercise variables. The prog-
nostic value of stress SPECT has been shown in numerous patient
subsets, including women, the elderly, patients with diabetes, and
patients with an intermediate-risk treadmill test result. Stress
SPECT has also been well validated for preoperative assessment
of patients at intermediate clinical risk undergoing an intermedi-
ate-risk or high-risk operation. The risk of a perioperative event
is 15% to 20% in patients with ischemic images and 2% to 3% in
patients with normal images.
 10 Nuclear Imaging
Assessment After Revascularization
Stress SPECT has been validated for risk stratification in patients
who previously have had coronary artery bypass grafting or per-
cutaneous intervention. Guidelines recommend stress imaging
over treadmill testing for assessment of patients who have had
revascularization and experience a change in symptom status
(class I indication). The frequency with which stress imaging
should be performed in asymptomatic patients after revascular-
ization is controversial. Guidelines currently suggest that testing
should not be performed more often than every 3 years.
Acute Coronary Syndromes
Myocardial perfusion imaging plays an important role in assess-
ing patients who present with symptoms ranging from atypical
chest pain to acute myocardial infarction. Rest myocardial perfu-
sion imaging is occasionally used to evaluate patients presenting
to the emergency department with chest pain and nondiagnostic
electrocardiography who are suspected of having an acute cor-
onary syndrome. The characteristic of minimal redistribution
with 99mTc-labeled agents allows them to be injected while the
patient is having chest pain and imaging is performed at a later
time, after resolution of the chest pain. The sensitivity and nega-
tive predictive value of acute rest imaging are high; however,
if perfusion defects are present, distinguishing between acute
ischemia, acute infarct, or chronic infarct is not possible on the
basis of rest images alone.
Stress SPECT imaging in patients with unstable angina or
non–ST-segment elevation myocardial infarction also can be used
for risk stratification in patients treated with an early conserva-
tive strategy or for assessment of the hemodynamic significance
of a coronary stenosis after coronary angiography. Patients with
an ST-segment elevation myocardial infarction are not candi-
dates for acute imaging. In clinically low- and intermediate-risk
patients in the chronic phase of ST-segment elevation myocar-
dial infarction (days 3–21) treated with an initial conservative
approach, stress SPECT can be used for risk stratification and
for identification of appropriate patients for coronary angiogra-
phy. Useful information includes perfusion, LVEF, regional wall
motion, and detection of viable myocardium. The presence of
ischemia or a perfusion defect outside the infarct zone identifies
patients at higher risk.
Myocardial Viability
The term hibernation refers to a chronic condition of contrac-
tile dysfunction in patients with coronary artery disease due to
long-standing underperfusion in whom restoration of myocardial
blood flow results in recovery of function. Hibernation is usually
distinguished from myocardial stunning, which is a reduction
in myocardial contractility resulting from transient reduction in
blood flow. Viable myocardium typically has characteristics of
preserved cell membrane integrity, preserved glucose metabo-
lism, and inotropic reserve. These characteristics can be evalu-
ated with nuclear (SPECT and positron emission tomography)
techniques.
Rest 201Tl viability protocols consist of immediate images
and delayed images at 4 and 24 hours. Initial 201Tl uptake on
the immediate images is proportional to myocardial blood flow.
Washout of 201Tl occurs more rapidly in normally perfused
129
myocardium and more slowly in areas that are poorly perfused.
Over time, the 201Tl concentration can redistribute and equalize
between these areas, thereby showing viability on the delayed
images (Figure 10.9). Myocardial segments that have reduced or
absent 201Tl uptake at rest and no further uptake over time reflect
areas of myocardial scar. Viability studies with 99mTc-sestamibi
also can be performed, in which uptake is dependent on an intact
cell membrane and functioning mitochondria. Studies with 99mTc-
sestamibi may underestimate viability in areas with a critically
severe stenosis because of its minimal redistribution. Efforts
to improve the ability of 99mTc-sestamibi to correctly identify
viable myocardium include performing a 4-hour redistribution
image, quantifying the degree of 99mTc activity in a perfusion
defect, using gated wall motion in combination with radioisotope
uptake, and obtaining a nitrate-enhanced image. Currently, the
optimal technique for assessing viability is the demonstration of
preserved glucose metabolism using fluorine-18-labeled fluoro-
deoxyglucose on positron emission tomography.
Additional Applications
Although metabolic imaging is ideally assessed with positron
emission tomography, imaging of iodine 123-labeled fatty acids
can be performed with SPECT to assess fatty acid metabolism
for detection of myocardial ischemia. Radionuclide imaging has
been used to visualize areas of myocardial infarct through the
use of 99mTc-pyrophosphate or indium 111-labeled antibodies
to cardiac myosin (hot-spot imaging). These agents localize to
areas of infarct. Unfortunately, the images are often of poor qual-
ity, and these approaches are not being currently used. In addi-
tion to use in myocardial infarct imaging, antimyosin imaging
has been used to assess myocarditis and rejection after cardiac
transplant. Imaging in congestive heart failure is expanding to
include the use of iodine 123-m-iodobenzylguanidine and car-
bon 11-hydroxyephedrine for evaluating cardiac innervation and
the potential use of 99mTc-labeled annexin V for the noninvasive
imaging of apoptosis. Further research is ongoing in radionuclide
imaging of atherosclerotic lesions and vulnerable plaques.
Limitations
Nuclear cardiology is a valuable technique in clinical practice,
but it does have limitations. Accurate nuclear cardiology imag-
ing necessitates well-qualified and experienced personnel at all
steps in the process, including image acquisition, processing, and
interpretation. Nuclear imaging requires radiation exposure and
is expensive. For these reasons, it should be applied in patients
who have appropriate indications.
Abbreviations
EDV End-diastolic volume
EF Ejection fraction
ESV End-systolic volume
LV Left ventricular
LVEF Left ventricular ejection fraction
99m
Mo Molybdenum 99
MUGA Multiple-gated blood pool analysis
SPECT Single-photon emission computed tomography
99m
Tc Technetium 99m
99m
TcO4 Pertechnetate
201
Tl Thallium 201
 11

Positron Emission Tomography

PANITHAYA CHAREONTHAITAWEE, MD

Principles of Positron Emission Tomography gadolinium oxyorthosilicate, and sodium iodide. The choice of
crystal depends in part on its ability to stop the 511-keV photons,
Since the late 1990s, PET has emerged as a valuable clinical
its light output and energy resolution, and cost. Until recently,
tool for the management of known or suspected CAD. Several
bismuth germinate was the most commonly used, owing to its
factors have contributed to its increased use: greater availability
superior ability to stop the photons, its acceptable random event
of PET scanners, FDA approval of PET radiotracers for clinical
rates, and its overall suitability for cardiac imaging. However,
cardiac application, reimbursement by the Centers for Medicare
the needs of whole-body oncology PET have led to the use of
and Medicaid Services for PET imaging to assess myocardial
other crystals and alternative methods to increase sensitivity and
perfusion and viability, and greater access to the tracers 82Rb and
reduce acquisition time, which may also benefit cardiac PET
FDG. Furthermore, the similar costs and lower patient radiation
imaging.
exposure of cardiac PET imaging compared with conventional
Attenuation correction is integral in PET imaging and has
nuclear imaging have added to the growth of cardiac PET imag-
traditionally been robust. For a photon of interest to be stopped
ing. With increased use of cardiac PET imaging, evidence for its
by the detectors, it must first leave the body. The likelihood of
clinical usefulness is rapidly expanding and demonstrating high
that depends in part on the distance the photons must travel
diagnostic accuracy and incremental prognostic value.
(ie, patient thickness) and the tissue density. If a person has a
Several inherent characteristics of PET contribute to its high
large body habitus, dense tissues (eg, breasts), or an overly-
spatial, temporal, and contrast resolution. These characteristics
ing diaphragm, the interactions of scattering and photoelectric
include coincidence detection, built-in attenuation correction, and
absorption in these tissues before photon detection can lead to
high-energy tracers. Coincidence detection is a unique scheme
attenuation artifacts and a heterogeneous, uncorrected emis-
for forming tomographic images in PET. PET tracers emit posi-
sion image that mimics either ischemia or infarction even in a
trons, which collide with electrons of nearby atoms in annihila-
healthy person (Figure 11.1, top row). An attenuation road map
tion reactions. Each reaction releases 2 high-energy (511 keV)
(Figure 11.1, middle row) obtained with an external radiation
photons that travel at 180° to each other. When a pair of radia-
source on the PET scanner, such as germanium-68 rod sources,
tion detectors is set up at opposite ends of the 180° line (ie, line
is generally acquired before tracer administration and is applied
of response), the time difference between 1 photon striking 1
to the subsequent emission image after tracer administration
detector and the other photon striking the other detector is used
to yield a more homogeneous attenuation-corrected image in
to localize precisely the source of the annihilation event. Many
the same healthy person (Figure 11.1, bottom row). Alignment
pairs of radiation detectors are arranged in a circular fashion
between the transmission and emission images is highly impor-
around the PET gantry and provide high sensitivity for detection
tant. Newer-generation PET scanners are now combined with a
of the photons. Several types of crystals are used in PET detec-
CT scanner, which can be used to perform attenuation correction
tors, including bismuth germinate, lutetium oxyorthosilicate,
instead of germanium-68 rod sources. The main advantage of
CT for attenuation correction of PET emission images is speed.
Abbreviations and acronyms are expanded at the end of this chapter. However, potential areas of concern include 1) misalignment
130
 11 Positron Emission Tomography 131

Figure 11.1. Top row, positron emission tomographic (PET) rubidium-82 emission image without attenuation correction demonstrates hetero-
geneity in myocardial uptake related to attenuation. Middle row, PET transmission image with germanium-68 line sources. Bottom row, Same PET
rubidium-82 image after attenuation correction demonstrates more homogenous tracer uptake with overall improved image quality.

between the CT transmission and PET emission data, which is
largely characterized by anterior or lateral defects and a high
false-positive rate and is more problematic than alignment with
traditional transmission data and 2) scaling the CT Hounsfield
units to 511-keV attenuation coefficients.
Cardiac PET Radiopharmaceuticals
The 3 PET tracers currently approved by the FDA for clinical
cardiac use are 13NH3, 82Rb, and FDG. Their major character-
istics are listed in Table 11.1. In clinical cardiac PET imaging,
these tracers can be divided into 2 main categories: perfusion
tracers and metabolism tracers.
PET Perfusion Tracers
The most commonly used PET perfusion tracers in clinical prac-
tice in the United States are 13NH3 and 82Rb-chloride. The tracer
oxygen-15 water deserves mention because it has properties of an
ideal perfusion tracer (ie, linearity with flow).
Nitrogen-13 Ammonia
The physical half-life of 13NH3 is only 9.9 minutes, so it must
be produced by an in-house or nearby cyclotron. The first-pass
myocardial extraction fraction of 13NH3 is high (nearly 100%),
and tissue retention of 13NH3 is prolonged owing to metabolic
trapping by the glutamine synthetase pathway. The net extraction
fraction is approximately 80% with flows in the resting range,
but decreases with hyperemic flow values. The presence of 13NH3
myocardial uptake is also a marker of myocardial viability.
However, absent uptake may reflect poor perfusion, so that addi-
tional viability (metabolic) imaging is required. Despite some
dependence on myocardial metabolism and possible flow under-
estimation at very high flow rates, 13NH3 provides the best-quality
perfusion images for visual analysis owing to the relatively long

Table 11.1. FDA-Approved PET Tracers and Their Characteristics

13 82
Characteristic NH3 Rb FDG
Production Cyclotron Generator (strontium 82) Cyclotron
Physical half-life 9.9 min 75 s 110 min
Positron range, mm 0.4 2.8 0.3
Physiology Extracted (first-pass 100%; Extracted (60% at rest) Glucose analogue
net 80% at rest)
Visual image quality Excellent Very good Excellent but requires metabolic
preparation
Measures Predominantly perfusion Predominantly perfusion Myocardial viability
Medicare reimbursement Yes Yes Yes
Total body effective radiation exposure 1.5 mSv (per 20-mCi dose) 2.5 mSv (per 45-mCi dose) 10 mSv (per 15-mCi dose)
Abbreviations: FDA, US Food and Drug Administration; FDG, fluorine-18 fluorodeoxyglucose; 13NH3, nitrogen-13 ammonia; PET, positron
emission tomography; 82Rb, rubidium 82.
132 II Noninvasive Imaging
Short-axis
Stress Rest
Apex
Mid
Base
Figure 11.2. Short-axis, horizontal-axis (HLA), and vertical long-axis (VLA) views of normal positron emission tomographic nitrogen-13 ammo-
nia rest and stress myocardial perfusion images. Overall image quality is very high.
physical half-life, the relatively high extraction fraction, and the
low background activity (Figure 11.2).
Absolute quantification of myocardial blood flow (in milli-
liters per minute per gram) can also be performed with 13NH3
with use of dynamic imaging and well-validated tracer kinetic
models. The duration of each 13NH3 image acquisition is approx-
imately 10 to 20 minutes, depending on the mode of acquisition
(static or dynamic; 2-dimensional or 3-dimensional) and the
PET scanner characteristics. Simultaneous gated acquisition
for assessment of LV volumes and systolic function is feasible
and adds incremental value to the perfusion data. Use of 13NH3
is approved by the FDA and is reimbursed by Medicare for the
assessment of myocardial perfusion. The total body effective
radiation dose with 13NH3 is very low (approximately 1.5 mSv
for a 20-mCi dose). The 2 major disadvantages of 13NH3 are its
cyclotron requirements, which can add considerable capital and
maintenance costs, and decreased uptake of the tracer in the
lateral wall, a pattern that should be recognized when visually
interpreting clinical images.
Rubidium 82
A commercially available generator for strontium 82 (physical
half-life, 23 days) can elute 82Rb (physical half-life, 75 seconds)
every 10 minutes. The uptake of 82Rb in tissue is similar to
that of potassium. Like thallium 201, 82Rb requires sodium-
potassium adenosine triphosphatase for intracellular transport.
The fi rst-pass extraction fraction is 50% to 60% for resting flow
but decreases to 25% to 30% with high flow and is reduced
in ischemic myocardium. In addition to its role as a perfusion
tracer, 82Rb uptake is a marker of viability since with cell mem-
brane disruption it may be lost rapidly from the myocardium.
The short physical half-life of 82Rb and its long positron track
Stress Rest
HLA
VLA
mildly impair the quality of the images (Figure 11.3). Absolute
quantification of myocardial blood flow with 82Rb is feasible
but is not as well validated as for 13NH3 or oxygen-15 water.
The duration of each 82Rb image acquisition is approximately
5 to 10 minutes. Simultaneous gated acquisition is also pos-
sible. Use of 82Rb is approved by the FDA and is reimbursed
by Medicare for the assessment of myocardial perfusion. The
total body effective radiation dose with 82Rb is approximately
2.5 mSv for a 45-mCi dose; however, larger doses of up to
60 mCi are often used for patients with large body habitus.
Although not requiring a cyclotron for production, 82Rb requires
a strontium-82 generator, which must be replaced every month.
This substantial expense should be considered in the context of
patient volumes for economic sustainability.
PET Metabolic Tracers
The only FDA-approved PET myocardial metabolic tracer is
FDG, and its use is reimbursed by Medicare for assessment of
myocardial viability. FDG is produced by a cyclotron, but its
longer physical half-life of 110 minutes allows transport, obviat-
ing the need for an on-site cyclotron. FDG is a glucose analogue
and is transported intracellularly by GLUT-1 and GLUT-4. FDG
traces the initial transmembranous exchange of glucose from
blood into tissue and its subsequent hexokinase-mediated phos-
phorylation into glucose-6-phosphate. Unlike phosphorylated
glucose, FDG-6-phosphate is not a substrate for the glycolytic
pathway, the pentose shunt, glycogenesis, or dephosphorylation
and remains trapped within the myocytes. Under normal fasting
conditions, the human heart primarily uses free fatty acids for
its energy production. With resting or stress-induced ischemia, a
switch from aerobic to anaerobic metabolism is in part responsi-
ble for increased myocardial glucose utilization. Uptake of FDG
 11 Positron Emission Tomography
Short-axis
Stress Rest
Apex
Mid
Base
Figure 11.3. Short-axis, horizontal-axis (HLA), and vertical long-axis (VLA) views of normal stress and rest positron emission tomographic
rubidium-82 myocardial perfusion images. Overall image quality is high but slightly inferior to nitrogen-13 ammonia images.
depends on myocardial substrate use: uptake is low when plasma
free fatty acid levels are high, and uptake is high when plasma
glucose and insulin levels are high. Adequate patient preparation
is therefore crucial to obtain images of high diagnostic quality,
with fasting conditions yielding more inadequate images than
glucose-loading protocols. The hyperinsulinemic euglycemic
clamp technique provides controlled metabolic conditions with
steady-state plasma insulin and glucose levels and yields high-
quality FDG images but is time-consuming and cumbersome.
For the assessment of myocardial viability, visual image analysis
is generally preferred, but quantification of absolute myocardial
glucose utilization is also feasible if the hyperinsulinemic eug-
lycemic clamp technique is used. The duration of FDG acquisi-
tion varies from 20 to 60 minutes, depending on the mode of
acquisition and scanner type. Electrocardiographic gating pro-
vides incremental value to the metabolic data. The total body
effective radiation dose with FDG is approximately 10 mSv for
a 15-mCi dose.
Clinical Applications
Diagnosis and Prognosis of CAD
Similar to SPECT, PET MPI is performed for diagnosis and
prognosis when patients have known or suspected CAD. Because
both FDA-approved PET perfusion tracers have short physical
half-lives, image acquisition must occur nearly simultaneously
with tracer administration to avoid significant radioactivity
decay. Therefore, pharmacologic stress is highly preferable;
generally, vasodilators such as adenosine or dipyridamole are
used, but dobutamine may be used in patients with contrain-
dications to vasodilators. Treadmill exercise PET MPI is fea-
sible, particularly with 13NH3, which is injected at peak stress
133
Stress Rest
HLA
VLA
on the treadmill; the patient is then immediately positioned in
the scanner for image acquisition, so this procedure requires
considerable technical expertise. Supine bicycle stress testing
may be performed simultaneously with image acquisition but
can introduce significant upper body motion. Exercise 82Rb PET
MPI is even more challenging owing to its very short physical
half-life.
Interpretation by visual analysis of PET myocardial perfusion
images is similar to that of SPECT images. Tracer uptake can
be graded on a semiquantitative scale (usually a 5-point scale
from 0 [normal] to 4 [no uptake]), with or without guidance from
quantitative software. A reversible defect is defined by reduced
tracer uptake on the stress image with relatively preserved or
improved uptake on the rest image and is consistent with isch-
emia (Figure 11.4). A fixed defect is defined by reduced tracer
uptake in a myocardial region on both rest images and stress
images and is consistent with infarction (Figure 11.5). Cardiac
size and global left ventricular systolic function are generally
visually assessed but can be aided by software-generated electro-
cardiographically gated volumes and ejection fraction. In addi-
tion, PET can quantify myocardial blood flow and flow reserve in
absolute terms, which is increasingly used clinically.
Compared with SPECT MPI, PET MPI has a significantly
higher interpretive certainty (Figure 11.6). The technical advan-
tages of PET, along with its superior image quality and high
interpretive certainty, translate into high diagnostic accuracy. In
9 published studies (877 patients) comparing PET MPI (13NH3
or 82Rb) to invasive coronary angiography, the weighted mean
sensitivity and specificity of PET MPI were 90% and 89%,
respectively, for detecting significant angiographic stenosis with-
out correction for referral bias. Compared with SPECT MPI,
the diagnostic accuracy of PET MPI is particularly superior in
women and in obese patients. However, these observations are
134 II Noninvasive Imaging

Short-axis

Stress Rest Stress Rest

Apex

HLA

Mid

VLA

Base

Figure 11.4. Short-axis, horizontal-axis (HLA), and vertical long-axis (VLA) views of abnormal stress and normal rest positron emission tomo-
graphic rubidium-82 myocardial perfusion images. A large, reversible apical, anterior, septal, and lateral defect (arrowheads) consistent with ischemia
and poststress left ventricular dilatation are evident in this patient with severe multivessel native and bypass graft coronary artery disease.

Short-axis

Stress Rest Stress Rest

Apex
HLA

Mid

VLA

Base

Figure 11.5. Short-axis, horizontal-axis (HLA), and vertical long-axis (VLA) views of abnormal stress and rest positron emission tomographic
nitrogen-13 ammonia myocardial perfusion images. A large, fixed apical, inferior, inferolateral, and inferoseptal defect (arrowheads) consistent with
infarction and left ventricular enlargement is evident in this patient with severe left ventricular systolic dysfunction.
 11 Positron Emission Tomography 135

P<.001

100 96
SPECT (n=112)
PET (n=112) 81
80

60
Percent

40

P=.004 P=.007
20

7 6 5 4
0 0
0
Probably normal Equivocal Probably abnormal Definitely normal
or abnormal
Figure 11.6. Results are shown from a study of 112 patients with known or suspected coronary artery disease (CAD) who underwent positron
emission tomographic (PET) myocardial perfusion imaging and were matched by age, sex, body mass index, and extent of CAD to 112 patients who
underwent single-photon emission computed tomography (SPECT) myocardial perfusion imaging. The proportion of studies interpreted as being
definitely normal or definitely abnormal was significantly higher with PET (96%) than with SPECT (81%). No PET studies, but 13% of SPECT
studies, were interpreted as probably normal or equivocal. (Previously published. See “Credit Lines” section.)

challenged by several confounders, such as referral bias and lack
of a robust database of head-to-head comparisons between tech-
niques in the same patients.
The prognostic accuracy of PET MPI has been examined in
5 published studies encompassing more than 4,000 patients. In
general, among these studies the prognostic value of PET MPI
is similar, but there are wide differences in mortality and car-
diac event rates (Table 11.2). These differences probably resulted
from population heterogeneity (selection bias), misclassification
bias, censoring variability related to revascularization, and less

Table 11.2. Approximate Annual Event Rates in Relation to PET-MPI Results

All-Cause Death, % Cardiac Events, %

Mildly Moderately Severely Mildly Moderately Severely

Reference Patients, No. Tracer Known CAD, % Normal Abnormal Abnormal Abnormal Normal Abnormal Abnormal Abnormal
Marwick et ala 657 82
Rb >50b 0.8 2.5 5.8 5.7 3.2 4.3 7.0 5.6
Yoshinaga et alc 367 82
Rb 40 NA NA NA NA 0.4 2.3 7.0d
Lertsburapa et ale 1,441 82
Rb 54 2.4f 4.1 6.9 NA NA NA NA
Dorbala et alg 1,432 82
Rb 31 2.5 5.0 8.0 10.0 0.5 2.5 5.5 10.0
Herzog et al h 229 13
NH3 66 NA NA NA NA 1.7 4.7i
Abbreviations: CAD, coronary artery disease; MPI, myocardial perfusion imaging; NA, not available; 13NH3, nitrogen-13 ammonia; PET, positron emission
tomography, 82Rb, rubidium 82.
a
Am J Cardiol. 1997 Oct 1;80(7):865–70.
b
Prior myocardial infarction, 48%; prior revascularization, 37%; receiving medical therapy, >50%.
c
J Am Coll Cardiol. 2006 Sep 5;48(5):1029–39. Epub 2006 Aug 17.
d
Value is for moderately abnormal and severely abnormal combined.
e
J Nucl Cardiol. 2008 Nov-Dec;15(6):745–53. Epub 2008 Sep 12.
f
Value is for normal and mildly abnormal combined.
g
JACC Cardiovasc Imaging. 2009 Jul;2(7):846–54.
h
J Am Coll Cardiol. 2009 Jul 7;54(2):150–6.
i
Value is for mildly, moderately, and severely abnormal combined.
(Previously published. See “Credit Lines” section.)
136 II Noninvasive Imaging

standardization in the reporting of imaging variables as com-
pared with SPECT MPI. The assessment of global and regional
LV systolic function also adds incremental value to the assess-
ment of perfusion alone. Wider use of PET MPI is supported by
its incremental value, greater diagnostic accuracy, and interpreta-
tive certainty. In addition, radiation exposure is less for the patient
compared with conventional SPECT imaging, and Medicare
reimbursement is similar between PET and SPECT MPI in many
geographic regions.
The quantification of PET absolute myocardial blood flow
and flow reserve has many potential clinical roles, including the
functional assessment of known epicardial coronary stenosis and
uncovering “balanced ischemia,” but more studies are needed to
determine normal and abnormal values, their clinical value, and
their prognostic significance.
Identification of Viable Myocardium
Revascularization may improve contractile function, survival,
and symptoms for patients who have moderate to severe ischemic
LV systolic dysfunction and severe CAD, but periprocedural
morbidity and mortality are high. Noninvasive viability imaging,
specifically PET, has traditionally been used to identify patients
most likely to benefit from revascularization.
Viable myocardium includes normal, stunned, and hibernat-
ing myocardium. Myocardial stunning refers to the temporarily
decreased contractile function that occurs after myocardial blood
flow has been temporarily decreased or stopped. Even after nor-
mal blood flow is restored, myocardial stunning may continue
for a while. Myocardial hibernation refers to a longer-term pro-
cess in which a chronic low-flow state or repetitive myocardial
stunning contributes to adaptive processes that lead to contrac-
tile dysfunction. Contractile function may improve if myocardial
blood flow is restored or if revascularization decreases myocar-
dial stunning. Several cellular mechanisms have been proposed
for both myocardial stunning and hibernation, which are 2 dis-
tinct pathophysiologic states. Practically though, they may be
various degrees along a continuum of the same process and may
be present in the same patient or same myocardial region.
The most conventional PET technique for assessing viability
is the combined examination of resting myocardial perfusion
(with either 13NH3 or 82Rb) and glucose metabolism (with FDG).
Visual assessment of regional perfusion and regional glucose
uptake can be performed with a semiquantitative approach sim-
ilar to MPI, which may be aided by quantification software. A
PET perfusion-metabolism mismatch (decreased perfusion and
preserved metabolism with contractile dysfunction) is the most
specific indicator of contractile recovery after revascularization

Short-axis

13 13
NH3 FDG NH3 FDG

Apex
HLA

Mid
VLA

Base

Figure 11.7. Short-axis, horizontal-axis (HLA), and vertical long-axis (VLA) views of viability positron emission tomographic (PET) study with
resting nitrogen-13 ammonia (13NH3) and resting fluorine-18 fluorodeoxyglucose (FDG) images from a 65-year-old woman with heart failure, severe
left ventricular (LV) systolic dysfunction, and severe coronary artery disease. LV enlargement is present. The calculated LV ejection fraction by
gated PET images (not shown) was 26%. A very large apical, septal, anterior, and lateral perfusion-metabolism mismatch (arrowheads) consistent
with hibernating myocardium is present. The patient underwent coronary artery bypass grafting surgery with improvement in LV systolic function
(ejection fraction increased to 45% postoperatively). All regions identified as viable recovered contractile function after revascularization.
 11 Positron Emission Tomography 137

and is the hallmark of hibernating myocardium (Figure 11.7).
Hypocontractile regions with FDG uptake of more than 50%
compared with normal or remote regions also might recover
contractile function after revascularization (but at a lower rate),
probably because of subendocardial scarring. Similarly, there
is uncertainty about contractile recovery of regions with mild
to moderate matched reduction in PET perfusion and metabo-
lism. A region with severely reduced perfusion and metabolism
indicates scar and has a low likelihood of contractile recovery
after revascularization (Figure 11.8). The reverse mismatch
pattern (normal perfusion and reduced FDG uptake) has been
described with myocardial stunning, left bundle branch block,
and paced rhythm. Evaluation of regional contractile function
by electrocardiographic gating adds incremental value to the
perfusion and metabolic data. Other important information in
the interpretation of PET viability studies includes the clinical
history, the patient’s metabolic status and preparation before
FDG administration, and the coronary anatomy, if available.
A 2007 analysis of established noninvasive viability imaging
techniques showed that recovery of regional contractile function
was best predicted with PET (sensitivity, 93%; negative predic-
tive value, 87%). PET also predicts global recovery of function
after revascularization. Comparative studies have shown that
PET may provide evidence of viability in about 15% of patients
with large, severe fixed SPECT defects. The diagnostic accuracy
appears to be less affected by the severity of regional and global
LV systolic dysfunction than other techniques that rely on assess-
ing contractile reserve.
From observational studies, the magnitude of improvement in
global LV systolic function after revascularization appears to be
directly related to the amount of viable myocardium identified by
PET before revascularization. This amount has ranged from 17%
to 67% of the LV, but 25% has been suggested as the minimum
required before revascularization to observe an improvement in
global LV systolic function after revascularization.
Observational studies also suggest that PET viability imag-
ing can predict prognosis and functional outcome among patients
with moderate to severe LV systolic dysfunction. When viabil-
ity has been identified by FDG PET, medically treated patients
seem to have the lowest survival rate. Compared with medical
therapy, revascularization significantly improves survival only
if viability is identified. Similarly, there is no significant differ-
ence in survival between patients who received medical therapy
and patients who underwent revascularization if viability is not
identified. In several studies involving PET, there was a mod-
est relationship between the extent of dysfunctional but viable
myocardium preoperatively and improvement in symptoms and
exercise capacity after revascularization.

Short-axis

13 13
NH3 FDG NH3 FDG

Apex
HLA

Mid

VLA

Base

Figure 11.8. Short-axis, horizontal-axis (HLA), and vertical long-axis (VLA) views of viability positron emission tomographic (PET) study with
resting nitrogen-13 ammonia (13NH3) and resting fluorine-18 fluorodeoxyglucose (FDG) images from a 72-year-old man with heart failure, severe left
ventricular (LV) systolic dysfunction, and severe coronary artery disease. LV enlargement is present. The calculated LV ejection fraction by gated
PET images (not shown) was 28%. A very large apical, septal, anterior, inferior, and inferolateral defect with moderate to severe reduction in both
perfusion and metabolism (perfusion-metabolism match) (arrowheads) is evident. This defect is consistent with scar and has a very low likelihood of
recovery of function after revascularization.
138 II Noninvasive Imaging

Limitations of published studies on myocardial viability Future Directions

include the lack of a robust database that compares different
Newer hybrid PET–multidetector CT technology provides the
techniques in the same patients, the observational and nonran-
potential to combine assessments of coronary and cardiac anat-
domized nature of the majority of studies, the limited timing of
omy, atherosclerotic burden, plaque characterization, physiol-
contractile assessment and follow-up after revascularization, the
ogy, and function but may involve greater radiation exposure and
small sample size of studies, referral bias, use of various pro-
costs. Since many PET-CT systems are routinely used for myo-
tocols, limited follow-up, and lack of information on graft or
cardial perfusion imaging, problems of coregistration between
vessel patency at follow-up. There have been only 2 randomized
the CT transmission and PET emission data should be recog-
studies of noninvasive viability imaging and patient outcomes.
nized when interpreting these studies.
In an earlier study that included patients who had relatively pre-
served LV systolic function, there was no significant difference
in accuracy or survival between patients who underwent SPECT Abbreviations
and patients who underwent PET. The PARR-2 study was the CAD coronary artery disease
first large (N = 430), randomized trial that used FDG-PET for CT computed tomography
patients who had CAD and moderate to severe LV systolic dys- FDA US Food and Drug Administration
function. There were no significant differences between patients FDG fluorine-18 fluorodeoxyglucose
GLUT glucose transporter
who underwent FDG-PET and the control group for the primary
LV left ventricular
end point of cardiac death, myocardial infarction, or hospital MPI myocardial perfusion imaging
stay for cardiac cause within 1 year of revascularization. The 13
NH3 nitrogen-13 ammonia
statistical power was decreased, however, because the observed PET positron emission tomography
event rates were lower than initially projected. In addition, some 82
Rb rubidium 82
patients in the control group underwent alternative viability test- SPECT single-photon emission computed tomography
ing and some patients in the FDG-PET group did not adhere to
recommendations for treatment by the study. With the exclusion
of patients who did not adhere to the PET recommendations for Name of Clinical Trial
therapy, survival benefits were significant. PARR-2 PET and Recovery Following Revascularization-2

Cardiovascular Computed Tomography
and Magnetic Resonance Imaging
THOMAS C. GERBER, MD, PHD, and ERIC M. WALSER, MD
Image formation in clinical CT relies on the mathematical con-
version, by filtered back-projection, of projection data that have
been obtained by measuring, with detectors, the attenuation of
a fan-shaped x-ray beam from many angles around the patient.
Image formation in clinical MRI relies on the alignment of
hydrogen nuclei or protons along an external magnetic field. The
alignment and angular momentum (spin) of these particles can
be excited if radiofrequency pulses are applied at the so-called
Lamor frequency. The relaxation of the particles toward their
original alignment in the magnetic field produces an MRI signal
that can be measured by external receiver coils. Each excitation-
relaxation sequence results in one line of data to be used for
image reconstruction, typically by Fourier transformation.
With current technology, CT examinations are faster and eas-
ier to perform than MRI examinations and, therefore, are more
widely offered. CT scans result in a 3-dimensional dataset that
can be reformatted in any arbitrary plane after the examination
is complete. Most MR pulse sequences acquire multiple parallel
slices or slab-shaped volumes that do, for example, not encom-
pass the entire heart. Most cardiovascular MRI examinations
consist of many image acquisitions, and experience is needed to
obtain the desired views. As a means to avoid misregistration and
motion artifacts, data acquisition or image reconstruction with
both imaging methods is performed relative to the electrocardio-
gram and, in MRI, often also to the respiratory cycle. However,
such “gating” can increase the time required to complete a scan.
Risks of and Contraindications to
Cardiac CT and MRI
As a result of the rapidly increasing use of cardiovascular CT
imaging, the dose and risks of ionizing radiation received from
Abbreviations and acronyms are expanded at the end of this chapter.
12
such examinations are discussed frequently. The effective radia-
tion dose of cardiovascular CT depends on the type of study. For
example, coronary artery calcium scanning imparts a lower radi-
ation dose than coronary CT angiography. The typical radiation
doses received from selected cardiovascular CT studies are listed
in Table 12.1. New CT imaging protocols, in particular so-called
prospective triggering, have substantially decreased the radiation
dose of cardiovascular CT examinations.
The risk of cancer with the radiation doses used in medical
imaging (<100 mSv) is controversial, but it is almost certainly
very low compared with the intrinsic lifetime risk of cancer.
Intravenous injection of iodinated contrast medium is needed for
some but not all types of cardiovascular CT studies. For exam-
ple, coronary artery calcium scanning does not require contrast
medium, whereas coronary CT angiography does.
Compared with CT, MRI can have higher temporal resolution
but has lower spatial resolution. Gadolinium, the contrast agent
used in MRI, has been linked to nephrogenic systemic fibrosis.
Nephrogenic systemic fibrosis occurs exclusively in patients with
renal failure. As of late 2008, 200 cases had been reported, none
of them before 1997. The risk of nephrogenic systemic fibrosis
in patients with end-stage renal disease is 2.5% to 5%. The risk
is probably dose-dependent and may also differ between vari-
ous gadolinium-containing contrast agents. To minimize the risk
of nephrogenic systemic fibrosis, use of gadolinium-containing
contrast agents should be avoided in patients with a glomerular
filtration rate of less than 30 mL/min.
MRI cannot safely be performed in the presence of many
metallic medical implants, including the growing number of
implantable cardioverter-defibrillators used in patients with cor-
onary artery disease. It is important to carefully screen patients
for MRI-incompatible implants, to accurately determine the type
of device or implant, to analyze the risk and benefit of MRI for
each specific clinical situation, and to obtain informed consent if
139
140 II Noninvasive Imaging

Table 12.1. Representative Values and Two types of CT scanners can be used for cardiac CT: multi-
Reported Ranges of Effective Radiation Dose row detector CT and electron beam CT. Multirow detector CT is
for Selected Cardiac Imaging Studies much more widely available than electron beam CT, but much of
the evidence base, particularly for coronary artery calcium scan-
Radiation Dose, mSv ning, has been developed with electron beam CT. Contemporary
Study Effective Range
multirow detector CT scanners acquire at least 64 (and as many
as 320) image slices simultaneously, and each gantry rotation
CAC scanning 3 1–8 lasts 330 milliseconds or less.
Coronary CTA Most cardiac MRI examinations are performed on scan-
Standard 15 12–18 ners with a field strength of 1.5 T, but initial experience with
Newest 3 2–4
3-T scanners has been reported. Spin echo sequences (black-
One-day Tc-99m sestamibi 12 —
—
blood technique) are used for anatomical imaging, and gradient
Rb-82 perfusion PET 10
FDG-PET viability 14 — echo sequences (bright-blood technique) are used for functional
Diagnostic catheterization 7 2–16 imaging.
With current technology, cardiac CT and cardiac MRI have
Abbreviations: CAC, coronary artery calcium; CTA, distinct areas of strength, and which method is to be used often
computed tomography angiography; FDG-PET,
[18]-fluorodeoxyglucose positron emission tomography;
depends on the clinical objective. Several scientific statements
PET, positron emission tomography; Rb-82, rubidium 82; and science advisories (but no specific guidelines) on the use
Tc-99m, technetium 99m. of cardiac CT and cardiac MRI have been published by the
American Heart Association and the American College of
Cardiology.
MRI is deemed necessary despite the presence of incompatible The field of view in cardiac CT and cardiac MRI is not limited
implants. The presence of coronary artery stents is not a contra- to the heart. Frequent incidental findings on these examinations
indication to performing MRI, even if placed more recently than include patent foramen ovale (Figure 12.1), atrial or ventricular
6 to 8 weeks previously. Up-to-date information on device com- thrombi (Figure 12.2), and calcified lymph nodes and pulmonary
patibility and MRI safety is available on the Internet at http:// nodules.
www.mrisafety.com/ and from other sources, such as the US
Food and Drug Administration.
Cardiac CT
Clinical Use of Cardiac CT and MRI A frequent, appropriate noncoronary indication for cardiac CT is
pericardial disease, for example, to visualize pericardial thick-
High temporal and spatial resolution are needed to obtain images ness and calcifications (Figure 12.3).
of the beating heart and of the small-caliber, often tortuous
coronary arteries that are detailed and free of motion artifact.
Attaining high temporal and spatial resolution is difficult and Coronary Artery Calcium Scanning
can be mutually exclusive. Recent developments in CT scan- Coronary artery calcification is an active process that begins in
ner technology and the programming of MRI pulse sequences the early stages of atherosclerosis and is regulated similar to the
have made cardiac CT and cardiac MRI widely available and calcification of bone. Because calcium has high x-ray attenuation,
frequently used. it can be detected sensitively by CT. The quantity of coronary

A B

RV

LA

RA
LV RA

LA

Figure 12.1. Computed Tomograms of Patent Foramen Ovale (PFO). A, Horizontal long-axis view. The fossa ovalis membrane (arrow) is visible.
Line indicates plane in which Figure 1B is reformatted. B, A small amount of contrast (arrow) traverses through the PFO from the right atrium (RA)
into the left atrium (LA). LV indicates left ventricle; RV, right ventricle.
 12 Cardiovascular Computed Tomography and Magnetic Resonance Imaging
LA
LV
Figure 12.2. Computed Tomogram of Thrombus. Vertical long-axis
view shows thrombus (arrow) in left ventricular apex. LA indicates left
atrium; LV, left ventricle. (See “Credit Lines” section.)
calcium, usually determined as Agatston score, volume score,
or calcium mass, is roughly proportional to, but represents only
approximately one-fifth of, the coronary plaque burden. Race
and socioeconomic factors affect the prevalence of coronary
artery calcification in ways not explained solely by differences
in risk factor profiles.
Because of the variable biologic process of vascular remodel-
ing, not all plaque results in luminal narrowing. A high quan-
tity of calcium may be present on cardiac CT in the absence of
coronary stenoses. In asymptomatic patients, a high coronary
calcium score alone should not trigger a decision to perform cor-
onary angiography. However, in patients symptomatic with chest
pain, the presence of coronary calcification predicts the presence
Figure 12.3. Computed Tomogram of Thickened and Calcified
Anterior Pericardium (arrows). Patient had catheter-proven constrictive
pericarditis.
141
of coronary luminal diameter stenoses exceeding 50% of a ref-
erence segment with high sensitivity (∼90%) but only moderate
specificity (∼50%). Specificity can be improved at the expense of
sensitivity if the quantity, not just the presence, of coronary cal-
cium is considered and if age- and sex-specific threshold values
for the calcium score are used to predict the presence of coronary
stenoses. The absence of coronary artery calcium predicts the
absence of coronary artery stenoses with very high accuracy.
The exact relationship between coronary artery calcification
and the presence of plaque likely to cause acute coronary syn-
dromes is poorly understood. Various theories hold that calcifi-
cation stabilizes plaque and makes it less likely to rupture, that
spotty calcification of plaques creates interfaces between tissue
with differing mechanical properties at which shear stress can
mechanically induce plaque rupture, or that calcified plaques and
plaques likely to rupture colocalize. Despite these uncertainties,
coronary artery calcium scanning with CT is widely used as a
screening technique to predict future cardiac events in asymp-
tomatic patients.
Most of the studies examining the prognostic value of cor-
onary calcium reported to date have important methodologic
shortcomings, and well-designed population-based studies are
currently ongoing. The current data are strongest for non-Hispanic
white men and suggest that the rates of adverse cardiac events are
higher in patients with coronary calcification (Figure 12.4A) than
in patients without (Figure 12.4B) and that the odds ratio and rel-
ative risk increase with increasing calcium scores. Current data
indicate that, in the absence of coronary calcium, the 2- to 5-year
risk of cardiac events is less than 0.1%. However, absence of cor-
onary calcium should not be a reason to lower the intensity of
risk factor modification in individual patients. The presence of
any coronary calcium is associated with a relative risk ratio of 4.6
for cardiac events over 3 to 5 years. The relative risk is 1.9 in the
lowest tertile of calcium scores (0–112), and 10.8 in the highest
tertile (>1,000).
Coronary calcification may allow refining risk prediction
in patients with an intermediate risk of cardiovascular events
based on the Framingham risk score, particularly if the calcium
score is high (eg, Agatston score >300). Age- and sex-adjusted
percentiles do not predict cardiovascular risk better than abso-
lute calcium scores. There is no evidence that the rate of pro-
gression of coronary calcification measured by serial CT can
be attenuated by pharmacologic preventive therapy. Similarly,
no outcomes studies to date suggest that making coronary cal-
cium scores the basis of risk factor management decisions will
improve the prognosis of patients with high coronary artery
calcium scores.
Coronary CTA
Under many circumstances, CTA can create detailed and
appealing images of epicardial coronary artery branches with a
diameter of approximately 1.5 mm or more (Figure 12.5). The
diagnostic accuracy of coronary CTA has improved with every
new generation of CT scanners. Data from the few available
multicenter studies at experienced institutions, each includ-
ing limited numbers (<360) of patients, suggest that coro-
nary CTA with contemporary multidetector scanners can, in
patients referred for catheter-based, selective coronary angiog-
raphy, identify or exclude coronary luminal diameter stenoses
exceeding 50% of a reference segment with high negative
predictive value (89%-98%) but low positive predictive value
(59%-82%) (Figure 12.6). A good-quality, negative coronary
142 II Noninvasive Imaging

A B

RVOT RVOT

LAD
Ao Ao

LA
LA

Figure 12.4. Transaxial Computed Tomograms Without Contrast Enhancement. A, Patient with calcification of left anterior descending artery
(arrow) and intermediate branch (arrowhead). B, Patient without coronary artery calcifications. Ao indicates aorta; LA, left atrium; LAD, left anterior
descending artery; RVOT, right ventricular outflow tract.

CTA reliably excludes high-grade coronary artery stenoses.
An abnormal coronary CTA may require further noninvasive
work-up. Predictors of poor image quality include known coro-
nary artery disease, fast (>60–70 beats per minute) heart rate,
and lack of patient cooperation with lying still and with breath-
ing instructions.
To date, no studies support the use of coronary CTA as a first-
line approach to the diagnosis of coronary artery disease in place
of stress testing. Similarly, although several studies in small num-
bers of patients and with limited follow-up have examined the
prognostic value of screening coronary CTA in asymptomatic
patients at intermediate or high risk of cardiac events, the prog-
nostic implications of detecting noncalcified plaque (Figure 12.7)
not large enough to cause high-grade luminal stenosis are not
established. Some comparative studies of coronary CTA and his-
topathologic results or intravascular ultrasonographic findings
suggest that such plaques can have features associated with an
increased likelihood of vulnerability or rupture.
Coronary CTA can, with sensitivity and specificity of 95%
or more, detect luminal diameter stenoses exceeding 50% of a
reference segment in, or occlusion of, coronary artery bypass
grafts (Figure 12.8). Coronary CTA can also, with greater ease
than invasive, catheter-based coronary angiography, determine
whether the proximal course of coronary arteries with congen-
itally abnormal origin is between the pulmonary artery and the
aorta (Figure 12.9). It is important to unequivocally make this
determination because such congenital coronary anomalies can
be associated with sudden cardiac death. However, in younger

A B

Ao
RVOT
PA
Diag
D ag
Di
LAD
Ao

LAD
LAD
LA
LV
RV
RV
LA

Figure 12.5. Coronary Computed Tomographic Angiograms Obtained From a Patient Without Coronary Calcifications or Coronary Artery
Stenoses. A, Contrast-enhanced transaxial view. B, Volume rendering. Ao indicates aorta; Diag, diagonal branch; LA, left atrium; LAD, left anterior
descending artery; LV, left ventricle; PA, pulmonary artery; RV, right ventricle; RVOT, right ventricular outflow tract.
 12 Cardiovascular Computed Tomography and Magnetic Resonance Imaging
A
LA
LV
B
Figure 12.6. Tandem High-Grade Coronary Artery Stenoses
(arrows) in Left Anterior Descending Artery, Proximal and Distal to
a Diagonal Branch. A, Coronary computed tomographic angiogram,
reformatted in vertical long axis. B, Invasive selective coronary angio-
gram. LA indicates left atrium; LV, left ventricle.
patients, cardiac MRI may be preferred to coronary CTA for this
purpose to avoid radiation exposure.
The following might be considered appropriate indications
for coronary CTA: clarifying the anatomy of suspected or known
congenital anomalies of coronary artery origin (this is the only
indication endorsed by current appropriateness criteria and sci-
entific statements), examining clinically important coronary
artery bypass grafts that could not be engaged selectively during
cardiac catheterization, patients with typical symptoms or abnor-
mal results of stress testing who refuse cardiac catheterization
or are at high risk for atheroembolic complications, and patients
with atypical symptoms or nondiagnostic results of stress tests in
whom coronary disease could not convincingly be established.
Coronary CTA with current technology should not be con-
sidered an alternative if invasive, selective coronary angiog-
raphy seems indicated. In particular, the following scenarios
are not acceptable indications for coronary CTA: screening
143
LAD
LA
LV
Figure 12.7. Computed Tomogram, Vertical Long-Axis View.
Partially obstructive noncalcified plaque (arrow) is seen in proximal seg-
ment of left anterior descending artery (LAD). Diffuse calcification is
distal to plaque. LA indicates left atrium; LV, left ventricle. (Previously
published. See “Credit Lines” section.)
of asymptomatic patients with risk factors for coronary artery
disease, establishing or ruling out progession of disease in
patients with known coronary artery stenoses, possible in-stent
restenosis (except stents in the left main coronary artery), and
unequivocally abnormal results of stress testing (if patient is
willing and able to undergo cardiac catheterization).
Cardiac MRI
Cardiac MRI determines left and right ventricular volumes with-
out geometric assumptions and provides accurate measurements
of stroke volume, ejection fraction, and myocardial mass.
In patients with hypertrophic cardiomyopathy, the precise
anatomical distribution of myocardial thickening can be well
characterized with cardiac MRI (Figure 12.10), and the systolic
anterior motion of the mitral valve and the dynamic outflow tract
obstruction can be visualized. In patients with valvular heart
disease, gradient echo cardiac MRI can show the turbulent flow
created by valvular stenosis and regurgitation. Aortic and mitral
valve areas determined planimetrically from cardiac MRI corre-
late well with data derived from echocardiographic assessment.
Velocity-encoded cardiac MRI, a variant of gradient echo, allows
quantitation of regurgitant volumes and calculation of regurgita-
tion fractions or pressure gradients.
In patients with simple (Figure 12.11) or complex (Figure 12.12)
congenital heart disease, cardiac MRI can determine with high
resolution the connections between the heart chambers and
the great vessels, the function of the heart chambers, and the
hemodynamic consequences (eg, shunt volume) of the anomaly.
Although cardiac CT may image anatomy with greater spatial
resolution, many imagers prefer cardiac MRI for the assessment
of congenital cardiovascular disease to avoid repeated radiation
exposure in the often young or very young patients.
Stress Imaging
Although coronary MRA is currently inferior to coronary CTA,
cardiac MRI is useful for the functional assessment of coronary
144 II Noninvasive Imaging

A B

LIMA
LIMA
LI
OM-SVG

RCA-SVG
RCA-S
SV
SVG
VG

Figure 12.8. High-Grade Stenosis (arrowheads) in Proximal Portion of Saphenous Vein Graft to Diagonal Branch. A, Volume-rendered coronary
computed tomographic angiogram. B, Invasive, selective coronary angiogram. LIMA indicates left internal mammary artery; OM-SVG, saphenous
vein graft to obtuse marginal branch; RCA-SVG, saphenous vein graft to right coronary artery.

artery disease. Exercise cardiac MRI is logistically difficult
because of MRI compatibility issues with the required exercise
equipment. Among the pharmacologic stress approaches, aden-
osine stress cardiac MRI is more widely performed than dobu-
tamine stress cardiac MRI, mostly because of higher patient
acceptance. Functional assessment can take the form of assessing
inducible wall motion abnormalities or assessing first-pass myo-
cardial perfusion defects. Dobutamine stress-induced regional
wall motion abnormalities are the overall most accurate pre-
dictor (sensitivity, 90%; specificity, 80%) of hemodynamically
significant coronary artery stenoses. Adenosine stress-induced
regional wall motion abnormalities are highly specific (95%)
but not sufficiently sensitive (40%). Adenosine stress-induced
myocardial perfusion abnormalities, currently the most widely
used stress cardiac MRI approach, are very sensitive (90%) for
the detection of hemodynamically significant coronary artery
stenoses. Quantitative measurement of myocardial perfusion by
MRI is possible but requires complex mathematical modeling.
Therefore, assessment of myocardial perfusion is typically per-
formed by visual assessment of myocardial first-pass enhance-
ment by a bolus of gadolinium.
Delayed Myocardial Enhancement
A bolus of gadolinium washes in and out of normal myocar-
dium quickly. Expanded extracellular space in the presence of

A B
Ao PA
PA

Ao
RCA

LAD
LA

LCX
LV RA
RV

Figure 12.9. Volume-Rendered Coronary Computed Tomographic Angiograms of a Single Coronary Artery Arising From the Right Sinus of
Valsalva. A, Anterior view. Left anterior descending artery (LAD) courses in front of the pulmonary artery (PA). B, Posterior view. Circumflex cor-
onary artery (LCX) courses behind aorta (Ao). Parts of left atrium (LA) and right atrium (RA) have been removed by image processing. LV indicates
left ventricle; RCA, right coronary artery; RV, right ventricle. (Previously published. See “Credit Lines” section.)
 12 Cardiovascular Computed Tomography and Magnetic Resonance Imaging 145

A B

RV
LA
LLV
V

LLV
V

Figure 12.10. Gradient Echo Cardiac Magnetic Resonance Images of Hypertrophic Cardiomyopathy Affecting Anterior Wall and Anterior
Septum (arrows). A, Short-axis view at mid ventricular level. B, Vertical long-axis view. Maximal wall thickness was 26 mm; left ventricular outflow
tract obstruction was not present. LA indicates left atrium; LV, left ventricle; RV, right ventricle.

myocyte necrosis can cause retention of gadolinium molecules, In contrast, myocardial damage due to idiopathic dilated
and DME (15–20 minutes after gadolinium administration) on cardiomyopathy, hypertrophic cardiomyopathy, myocarditis,
inversion-recovery prepared gradient echo images signifies myo- sarcoidosis, Anderson-Fabry disease, or Chagas disease typi-
cardial damage. cally shows a midmyocardial or subepicardial pattern of DME.
Different patterns of DME can help distinguish between Amyloidosis can have a subendocardial pattern of DME similar
different causes of myocardial damage. DME as the result of to that of myocardial infarction, but the DME is patchy, does not
myocardial infarction typically occurs in coronary perfusion ter- follow coronary perfusion territories, and may involve the right
ritories and involves at least the subendocardium. The volume ventricular myocardium (Figure 12.14).
of DME correlates well with myocyte necrosis on triphenyltet-
razolium chloride staining and can therefore be used to quantify
infarct size. The transmural extent of DME is a measure of myo-
cardial viability and predicts the likelihood of functional recov-
ery after revascularization of hypokinetic or akinetic myocardial
segments. For example, if more than 75% of myocardial thick-
ness shows delayed enhancement, the likelihood of functional
recovery is less than 2% (Figure 12.13).

Ao

PA
PA
RV

RA
LV SV

LA

Figure 12.11. Gradient Echo Cardiac Magnetic Resonance Image.

An atrial septal aneurysm (arrow) and an atrial septal defect of 5 mm Figure 12.12. Spin Echo Cardiac Magnetic Resonance Image of
are seen with enlargement of right atrium (RA) and right ventricle (RV). Repaired Complex Congenital Heart Disease With d-Transposition of
Pulmonary:systemic blood flow ratio was 2.0. LA indicates left atrium; Great Arteries. Aorta (Ao) and pulmonary artery (PA) arise from a sin-
LV, left ventricle. gle ventricle (SV) with left ventricular morphologic features.
146 II Noninvasive Imaging

A B

LLV
V
LLV
V
RV

Figure 12.13. Magnetic Resonance Images of Anterior Myocardial Infarction After Failed Placement of Stent, Complicated by Vessel Occlusion,
in Left Anterior Descending Artery. More than 75% of thickness of anterior wall and anterior septum from base to apex show delayed myocardial
enhancement (arrows). A, Short-axis view at mid ventricular level. B, Vertical long-axis view. Arrowhead is at apical thrombus. LV indicates left
ventricle; RV, right ventricle.

LV
RV

Figure 12.14. Magnetic Resonance Images of Biopsy-Proven Cardiac Amyloid. A, Gradient echo cardiac magnetic resonance short-axis view
at mid ventricular level, showing symmetric left ventricular hypertrophy. LV indicates left ventricle; RV, right ventricle. B, Corresponding inversion
recovery-prepared image, showing patchy, delayed myocardial enhancement that does not follow any coronary artery perfusion territory and includes
right ventricle.
 A B

Figure 12.15. Magnetic Resonance Angiograms of Abdomen, Pelvis, and Lower Extremities. Findings include abdominal aortic aneurysm (A),
aneurysmal degeneration of right femoral popliteal bypass graft (arrow in B), and anatomical location of degenerated bypass and thrombus within
aneurysm in coronal sources image (arrow in C).
147
148 II Noninvasive Imaging

Figure 12.16. Images From a Renal Transplant Recipient. A, Magnetic resonance angiogram of pelvis. Donor renal artery is normal, but there is
unusual scalloping along the common and external iliac arteries. B, Plain radiograph, showing multiple metal clips along course of iliac vessels. C,
Angiogram, showing no iliac arterial stenoses. The pseudostenoses were artifacts from metal clips during magnetic resonance angiography of pelvis.

Vascular CT and MRI
Technical Issues
Recent technical developments in CT and MRI have facilitated
vascular imaging because an entire scan can now be completed
during the transit time of one intravenous bolus administration of
contrast material. The newer, 64-slice CT scanners and the newer
pulse sequences for MRI have essentially replaced diagnostic angi-
ography for large and medium-sized vessels. Additionally, a sin-
gle intravenous injection of contrast agent permits 3-dimensional
and multiplanar representation not only of the vessel lumen and
wall in multiple projections but also of surrounding structures
(Figure 12.15), whereas during standard catheter-based angiog-
raphy only intraluminal anatomy can be visualized.
CTA Versus MRA
Because MRA requires no iodinated contrast material and
allows imaging in multiple different planes of acquisition, it is
the safest and most flexible imaging technique for evaluating
 12 Cardiovascular Computed Tomography and Magnetic Resonance Imaging 149

Figure 12.17. Imaging Findings Used to Diagnose Acute Embolic Disease to the Lower Extremities. A, Posterior volume-rendered computed
tomographic angiogram of knees, showing left popliteal occlusion (arrow). B, Thin-section transverse image, showing tiny filling defect in left pro-
funda femoral artery (arrow).

Figure 12.18. Magnetic Resonance Angiogram. A patent aortic stent

graft has small endoleak (small arrow). Metal-related artifact at proxi- Figure 12.19. Computed Tomographic Angiogram of Aorta and Iliac
mal and iliac limbs of stent graft is due to metal stents in these locations Arteries. Image was obtained after stent grafting. This method has bet-
(large arrows). Arteriography showed no stenoses at iliac segments. ter precision and fewer artifacts than magnetic resonance angiography.
150 II Noninvasive Imaging

Figure 12.20. Pulmonary Embolus in Third-Generation Branch of Pulmonary Artery. A, Computed tomographic angiogram, showing small
filling defect in a basilar segmental right pulmonary artery (arrow). B, Pulmonary angiogram confirmed defect (arrow). Embolus is near the limit of
resolution of computed tomographic angiography, although sensitivity continues to improve with newer-generation scanners.

large and medium-sized blood vessels. MRA is not as affected

Figure 12.21. Computed T omographic Angiogram. Large central
pulmonary emboli are seen on the left and right (arrows) in acutely
dyspneic patient. Although the patient was severely ill and marginally
stable, diagnostic images were obtained with minimal breath holding.
by severe vascular calcification as is CT. Heavily calcified arter-
ies may still induce artifact by MRA, but luminal narrowings
and irregularities are better seen with MRA in this situation.
However, MRI is very sensitive to the presence of metal, and
clips or stents can severely distort MRA and lead to false diag-
noses (Figure 12.16).
CTA, especially when done with the newer and faster scan-
ners, allows very high-resolution axial imaging. Complex vessel
morphologic features, as seen in dissections, irregular aneu-
rysms, and vascular tortuosity, are better delineated by CT
because of artifacts that turbulent or multidirectional blood flow
can cause on MRI. Indications for CTA overlap those for MRA,
but there is bias for CTA in cases requiring a high degree of
image detail (Figure 12.17). Additionally, if the surrounding
anatomy is important to evaluate, CTA provides better resolu-
tion of solid organs, particularly the lungs. Because of its high
spatial resolution, CTA is also very well suited for the evaluation
of aneurysms and dissecting hematomas, especially abdominal
aortic aneurysms and intracerebral aneurysms. CTA provides the
best imaging results for planning vascular surgical or endovascu-
lar procedures. CTA is particularly important for the increasing
numbers of endovascular stent repairs of thoracic and abdomi-
nal aortic aneurysms and dissections. Planning these procedures
requires very exact measurements of the vessel diameters, angu-
lations, and lengths in order to choose the appropriate device. In
patients who already have vascular stents in place, better images
are obtained with CTA than MRA because of the metallic
 12 Cardiovascular Computed Tomography and Magnetic Resonance Imaging 151

Figure 12.22. Images Used to Diagnose Vascular Stenosis. A, Magnetic resonance angiogram, showing irregular severe stenosis in proximal right
subclavian artery (arrow). B and C, Catheter angiograms confirmed the diagnosis and allowed successful angioplasty and stenting of this lesion.
152 II Noninvasive Imaging
Figure 12.23. Computed Tomographic Angiogram Obtained From
a Patient Who Had Carotid Subclavian Bypass for Right Subclavian
Artery Occlusion. Patient had recurrent ischemic right arm pain, and
computed tomographic angiogram showed intimal hyperplasia narrow-
ing the proximal anastomosis (arrow). Notice inclusion of carotid bifur-
cation (double arrows). Significant internal carotid stenosis would alter
surgical management of the dysfunctional bypass graft because carotid
endarterectomy also might be required.
artifact created in MRA (Figures 12.18 and 12.19). Other metal-
lic structures such as clips or prostheses also tip the balance in
favor of CTA rather than MRA, although CTA images also may
be degraded by streak artifact from large metallic objects such
as hip prostheses.
Technique of CTA and MRA
MRA and CTA are tailored to the specific purpose of the exam-
ination. Standard principles of cross-sectional imaging to facil-
itate reformatting images in 3 dimensions include thin sections
(2–5 mm) and rapid bolus administration of contrast material
at 4 to 5 mL/s. Large-bore venous access is required to perform
CTA or MRA (18 gauge at the least). Secondary multiplanar or
3-dimensional reconstructions of vascular anatomy are now easily
done with standard software; however, such image processing can
introduce artifacts and errors of its own, and reformatted images
should always be reviewed in conjunction with source images.
Because of the current limitations in resolution, MRA and
CTA are not appropriate imaging methods for evaluating small
vessels and diagnosing problems such as vasculitis or embolic
disease to the hand or foot. CTA and particularly MRA tend
to overestimate stenoses, a feature that makes them sensitive
screening tools. False-positive findings occur most often in obese
patients, uncooperative patients, or patients with tortuous vessels,
in whom reduced signal-to-noise ratio, motion, or turbulent flow,
respectively, create image artifacts.
Specific Indications for MRA and CTA
Pulmonary Arteries
Pulmonary CTA has evolved into the primary imaging method
of screening for pulmonary embolism, effectively replacing
ventilation-perfusion imaging and catheter-based pulmonary
angiography. Ventilation-perfusion scanning and pulmonary
angiography cannot detect other abnormalities in the chest that
can cause acute-onset dyspnea and chest pain. In patients allergic
to iodinated contrast material or with decreased renal function,
diagnostic multidetector pulmonary CTA can be performed with
intravenous gadolinium, although gadolinium is less radiopaque
than iodine-based contrast material. Pulmonary MRA is cur-
rently limited by lack of resolution.
Studies comparing older single-detector CTA with pulmo-
nary angiography show a sensitivity of 60% to 100% and speci-
ficity of 80% to 100% for the diagnosis of pulmonary embolus.
Pulmonary artery CTA is very sensitive for central pulmonary
emboli, and the lower sensitivities and specificities were prima-
rily due to very peripheral emboli in third- or higher-generation
pulmonary artery branches (Figure 12.20).
Multidetector CTA scanners are now fast enough to image the
entire chest and pulmonary circulation with 2-mm collimation in
one breath hold, even in acutely dyspneic patients (Figure 12.21).
Up to 30% of patients undergoing CTA for suspected pulmo-
nary embolus have some other cause for their symptoms which
is readily apparent on the high-resolution CTA scans. For exam-
ple, spiral CTA of the pulmonary arteries is able to detect even
a small patent ductus arteriosus because of the unopacified jet
of blood entering the pulmonary artery at the level of the aorti-
copulmonary window.
Aortic Arch Branches and Central Venous Structures
Vascular stenoses of the great vessels are also easily seen with
CTA or MRA, similar to stenoses in other, medium-sized to large
vessels (Figure 12.22). MRA detects great artery stenoses with a
sensitivity of 90% and a specificity of 96%. In brachiocephalic
or subclavian arterial stenoses, delayed imaging helps to visual-
ize the retrograde vertebral artery and distal subclavian artery in
cases of subclavian steal. In the presence of great vessel stenoses, it
is important to include the carotid bifurcations in the field of view,
particularly if surgical subclavian carotid bypass grafting is an
option (Figure 12.23). Anomalies of the aortic arch, such as aber-
rant subclavian arteries and coarctation and bicuspid aortic valves,
can be seen with both CTA and MRA (Figures 12.24 and 12.25).
Aortic Arch and Thoracic Aorta
MRA and CTA are extremely accurate for the evaluation of
aortic aneurysms and dissections. In the case of dissections,
multidetector CTA is preferred because of its improved resolu-
tion and ability to visualize small intimal flaps or ulcerations.
Recent studies report 99% sensitivity and 100% specificity for
multirow detector CTA of acute dissections (Figure 12.26). CTA
performed in the setting of back pain in hypertensive patients
is positive for dissection in about 18% of patients. The rare
false-negative examination usually occurs in patients with small
penetrating atherosclerotic ulcers or minimal intramural hema-
tomas that may escape detection by CTA.
 12 Cardiovascular Computed Tomography and Magnetic Resonance Imaging 153

Figure 12.24. Images Obtained From Patient With Coarctation of the Aorta. A and B, Computed tomographic angiograms, showing calcific arch
stenosis (arrow in A) and enlarged internal mammary artery and intercostal collaterals (double arrows in A). A bicuspid aortic valve is also apparent
(arrow in B). C, Magnetic resonance angiogram, showing similar findings with coarctation visible in arch (arrow).
154 II Noninvasive Imaging

A B

Figure 12.25. Computed Tomographic Angiograms of Right Aortic Arch. A, Vascular ring is formed by left subclavian artery and aortic diver-
ticulum (arrow) surrounding trachea (double arrow). B, Left oblique view, showing vascular structures forming the ring.

Figure 12.26. Computed Tomographic Angiograms of Type III Aortic Dissection. A, Volume-rendered image, showing thin dissection flap extend-
ing from left subclavian artery to distal aortic bifurcation (arrow). B, Sagittal reformatted image, showing intimal flap and mesenteric arteries.
 12 Cardiovascular Computed Tomography and Magnetic Resonance Imaging 155

Figure 12.27. A and B, Magnetic Resonance Angiograms of Thoracoabdominal Aneurysm. Transverse image (B) allows evaluation of luminal
thrombus and aneurysm diameter.

Figure 12.28. Computed Tomographic Angiogram. Image was Figure 12.29. Computed Tomographic Angiogram. Image was
obtained after aorto–uni-iliac stent graft placement and left iliac obtained after aortic stent graft repair of infrarenal aortic aneurysm.
artery occlusion with crossed femoral bypass for aortoiliac aneurysmal There is a large endoleak (arrow) with persistent enlargement and
disease. enhancement of aneurysm sac.
156 II Noninvasive Imaging

Figure 12.30. Renal Angiograms. A, Magnetic resonance angiogram of right renal artery, showing corrugation and possible stenosis of proximal
renal artery (arrow). B, Angiogram showing fibromuscular dysplasia and stenosis, successfully treated with angioplasty.

Thoracic aneurysms can be similarly imaged with MRA or
CTA, but some advantage is given to MRA because it does not
require iodinated contrast and radiation exposure. When a tho-
racic aneurysm is being followed, resolution requirements are
reduced because only aortic diameter and branch vessel patency
are evaluated (Figure 12.27).
Abdominal Aorta
MRA and CTA of the abdominal aorta are used to evaluate
abdominal aneurysms or dissections and to image the inflow
arteries in patients with known peripheral vascular disease. In
patients with aneurysms considered for stent-graft repair, CTA is
preferred because of its improved resolution (Figure 12.28). CTA
shows the contraction of the aneurysm sac, indicating successful
treatment, and monitors for endoleaks, which appear as contrast
material filling the persistent or enlarging aneurysm sac. After a
patient has undergone endovascular stenting, it is important for
early and delayed scanning to be performed, because endoleaks
around the stent graft may become visible only late after contrast
injection (Figure 12.29). Branch vessel or iliac occlusions can
complicate endovascular stent grafts and can be easily seen and
characterized with CTA.
Renal Artery Stenosis
It is generally agreed that screening patients for renal artery ste-
nosis is best done with CTA or MRA. CTA and MRA tend to
overestimate stenoses, and a small percentage of false-positive
screening results, which are insignificant lesions by angiogra-
phy, can be expected. MR and CT are equivalent for evaluating
suspected renal artery stenosis, but MR generally is preferred
because of the diminished renal function in many patients
(Figures 12.30 through 12.32). For renal artery stenosis, MRA
for the detection of hemodynamically significant arterial stenosis
has a sensitivity of 93% and a specificity of 99%. Similarly, CTA
has a sensitivity of 92% and a specificity of 99%.
A second common application of MRA and CTA is evalua-
tion of vascular anatomy in living renal donors or, more recently,
liver donors before transplant. Advantages of MRA and CTA
over standard angiography in this setting include the ability to
see renal parenchyma and the rest of the abdomen to screen for
potential tumors, cysts, or stone disease before transplant.
Mesenteric Ischemia
Similar to the situation for renal arteries, MRA and CTA have
essentially replaced angiography for the evaluation of patients

Figure 12.31. Computed Tomographic Angiogram of Left Renal Artery Ostial Stenosis (arrow).
 12 Cardiovascular Computed Tomography and Magnetic Resonance Imaging 157

with postprandial abdominal pain and weight loss. MRA eval-

uation of the abdominal aorta in suspected mesenteric ischemia
generally is performed in the sagittal plane rather than the axial
plane because of the orientation of the mesenteric vessels along
the long axis of the aorta compared with the lateral exit of the
renal arteries off the aorta (Figure 12.33). For detection of mes-
enteric artery stenosis, CT has a reported sensitivity of 100% and
a specificity of 89%. Similarly, MRA has a reported sensitivity
of 100% and a specificity of 95% for mesenteric arterial stenoses
greater than 75%.

Bluemke DA, Achenbach S, Budoff M, Gerber TC, Gersh B, Hillis
Figure 12.32. Magnetic Resonance Angiogram of Kidneys.
Budoff MJ, Achenbach S, Blumenthal RS, Carr JJ, Goldin JG,
Findings were compensatory hypertrophy of right kidney and severe,
long-segment stenosis and distal occlusion of left renal artery (arrow).
There was little function in the left kidney, and it was surgically
resected.
Gerber TC, Carr JJ, Arai AE, Dixon RL, Ferrari VA, Gomes AS, et al.
Greenland P, Bonow RO, Brundage BH, Budoff MJ, Eisenberg MJ,
Hendel RC, Patel MR, Kramer CM, Poon M, Hendel RC, Carr JC,
Figure 12.33. Magnetic Resonance Angiogram From a Patient
With Weight Loss and Abdominal Pain. Superior mesenteric artery has
long-segment occlusion (arrow). Although the inferior mesenteric artery
(double arrows) was patent, it was severely stenotic at its origin. Chronic
mesenteric ischemia symptoms are generally not evident until 2 of the 3
mesenteric arteries are significantly stenosed or occluded.
Suggested Reading
LD, et al. Noninvasive coronary artery imaging: magnetic resonance
angiography and multidetector computed tomography angiography:
a scientific statement from the american heart association commit-
tee on cardiovascular imaging and intervention of the council on car-
diovascular radiology and intervention, and the councils on clinical
cardiology and cardiovascular disease in the young. Circulation. 2008
Jul 29;118(5):586–606. Epub 2008 Jun 27.
Greenland P, et al; American Heart Association Committee
on Cardiovascular Imaging and Intervention; American Heart
Association Council on Cardiovascular Radiology and Intervention;
American Heart Association Committee on Cardiac Imaging, Council
on Clinical Cardiology. Assessment of coronary artery disease by car-
diac computed tomography: a scientific statement from the American
Heart Association Committee on Cardiovascular Imaging and
Intervention, Council on Cardiovascular Radiology and Intervention,
and Committee on Cardiac Imaging, Council on Clinical Cardiology.
Circulation. 2006 Oct 17;114(16):1761–91. Epub 2006 Oct 2.
Ionizing radiation in cardiac imaging: a science advisory from the
American Heart Association Committee on Cardiac Imaging of the
Council on Clinical Cardiology and Committee on Cardiovascular
Imaging and Intervention of the Council on Cardiovascular Radiology
and Intervention. Circulation. 2009 Feb 24;119(7):1056–65. Epub
2009 Feb 2.
Grundy SM, et al; American College of Cardiology Foundation
Clinical Expert Consensus Task Force (ACCF/AHA Writing
Committee to Update the 2000 Expert Consensus Document on
Electron Beam Computed Tomography); Society of Atherosclerosis
Imaging and Prevention; Society of Cardiovascular Computed
Tomography. ACCF/AHA 2007 clinical expert consensus document
on coronary artery calcium scoring by computed tomography in
global cardiovascular risk assessment and in evaluation of patients
with chest pain: a report of the American College of Cardiology
Foundation Clinical Expert Consensus Task Force (ACCF/AHA
Writing Committee to Update the 2000 Expert Consensus Document
on Electron Beam Computed Tomography) developed in collabora-
tion with the Society of Atherosclerosis Imaging and Prevention and
the Society of Cardiovascular Computed Tomography. J Am Coll
Cardiol. 2007 Jan 23;49(3):378–402.
et al; American College of Cardiology Foundation Quality Strategic
Directions Committee Appropriateness Criteria Working Group;
American College of Radiology; Society of Cardiovascular Computed
Tomography; Society for Cardiovascular Magnetic Resonance;
American Society of Nuclear Cardiology; North American Society
for Cardiac Imaging; Society for Cardiovascular Angiography and
Interventions; Society of Interventional Radiology. ACCF/ACR/
SCCT/SCMR/ASNC/NASCI/SCAI/SIR 2006 appropriateness cri-
teria for cardiac computed tomography and cardiac magnetic res-
onance imaging: a report of the American College of Cardiology
Foundation Quality Strategic Directions Committee Appropriateness
Criteria Working Group, American College of Radiology, Society of
158 II Noninvasive Imaging

Cardiovascular Computed Tomography, Society for Cardiovascular Clinical Cardiology, and the Council on Cardiovascular Radiology
Magnetic Resonance, American Society of Nuclear Cardiology, North and Intervention: endorsed by the American College of Cardiology
American Society for Cardiac Imaging, Society for Cardiovascular Foundation, the North American Society for Cardiac Imaging, and
Angiography and Interventions, and Society of Interventional the Society for Cardiovascular Magnetic Resonance. Circulation.
Radiology. J Am Coll Cardiol. 2006 Oct 3;48(7):1475–97. 2007 Dec 11;116(24):2878–91. Epub 2007 Nov 19.
Levine GN, Gomes AS, Arai AE, Bluemke DA, Flamm SD, Kanal E,
et al; American Heart Association Committee on Diagnostic and
Abbreviations
Interventional Cardiac Catheterization; American Heart Association
Council on Clinical Cardiology; American Heart Association Council CT computed tomography
on Cardiovascular Radiology and Intervention. Safety of mag- CTA computed tomography angiography
netic resonance imaging in patients with cardiovascular devices: an DME delayed myocardial enhancement
American Heart Association scientific statement from the Committee MRA magnetic resonance angiography
on Diagnostic and Interventional Cardiac Catheterization, Council on MRI magnetic resonance imaging
 13
Cardiac Radiography
JEROME F. BREEN, MD, MARK J. CALLAHAN, MD,†
and MARGARET A. LLOYD, MD
The conventional upright posteroanterior and lateral x-ray pro-
jections of the chest are obtained with high kilovoltage technique
at maximal inspiration to permit short exposure times, which
freeze cardiac motion. Interstitial markings are accentuated on a
poor inspiratory effort radiograph. A tube-to-film distance of at
least 6 feet minimizes distortion and magnification.
• Chest radiographs that show cardiac abnormality are a very impor-
tant part of cardiology examinations.
• Always take a systematic approach to interpreting chest radio-
graphs. Always identify the border-forming structures of the heart
on both the frontal and the lateral views. Use the pulmonary blood
vessels to help explain all abnormal contours.
• Always try to compare a chest radiograph with any available pre-
vious study.
• Postoperatively, suspect new abnormalities on chest radiographs to
be related to the surgical procedure.
• If a computed tomogram or magnetic resonance image is shown
on the cardiology boards, look carefully for pericardial or aortic
disease.
Cardiac Silhouette and Chamber Size
The image of the heart and great vessels on the chest radiograph
is a 2-dimensional display of dynamic 3-dimensional structures
(Figures 13.1 through 13.10). The cardiovascular silhouette var-
ies not only with the abnormality but also with body habitus, age,
respiratory depth, cardiac cycle, and position of the patient.
Posteroanterior Projection
The right mediastinal contour consists of a straight upper vertical
border formed by the superior vena cava and a smooth convex
† Died December 28, 2011.
lower cardiac contour formed by the right atrium. Occasionally,
a short segment of inferior vena cava is seen where the right
atrium meets the diaphragm.
The normal left mediastinal contour is formed by a series of
convexities: from superior to inferior, the aortic knob, the pul-
monary trunk, and the left ventricle abutting the diaphragm. The
left atrial appendage may be seen projecting between the pulmo-
nary trunk and the left ventricle in the normal heart, primarily
in young females. The shape of the pulmonary trunk segment
varies with age and body habitus. Most frequently, this segment
is only slightly convex; however, it can be prominent in women
20 to 40 years old and straight or even concave in older patients
and still be within normal limits. Occasionally, the cardiophrenic
junction of the cardiac silhouette is not formed by the left ven-
tricle but by a fat pad. Less common is a border-forming fat pad
in the right cardiophrenic angle, which should not be confused
with a cardiac mass.
• The left atrial appendage may be seen projecting between the pul-
monary trunk and left ventricle, especially in young females.
• The left cardiophrenic junction may be formed by a fat pad and
give a false impression of cardiomegaly.
Lateral Projection
Routinely, the patient’s left side is positioned against the film
cassette to minimize distortion of the heart due to geometric
magnification. Superiorly, the anterior border is formed by the
ascending aorta posterior to the retrosternal air space; inferi-
orly, the right ventricle and right ventricular outflow tract abut
the sternum and blend into the main pulmonary artery, which
then courses posteriorly to its bifurcation. The posterior cardiac
contour is formed by the left atrium superiorly beneath the carina
and the left ventricle curving inferiorly to the diaphragm, where
159
160 II Noninvasive Imaging

Left
Subclavian
Artery
SVC
Aortic
Knob
PA

Main
Pulmonary
Artery Ao
Aorta Left Atrial
Appendage
RA
LV
Right Left
Atrium Ventricle

Figure 13.1. Posteroanterior Projection of the Heart. SVC indicates

superior vena cava.
the straight vertical edge of the inferior vena cava is often appar-
ent within the thorax as it enters the right atrium.
Heart Size on Chest Radiographs
The cardiothoracic ratio—the ratio of the transverse cardiac
diameter to the maximal internal diameter of the thorax at
the level of the diaphragm on an upright posteroanterior chest
radiograph—corrects for body size and magnification produced
by slight differences in radiographic techniques. In adults, a ratio
more than 0.5 is considered to indicate cardiomegaly. In aor-
tic regurgitation, the left ventricle is often enlarged downward
rather than horizontally. A high diaphragm position, as seen with
obesity or shallow inspiration, produces an erroneous ratio more
than 0.5. Pectus excavatum and the absence of pericardium dis-
place the heart posteriorly and rotate the apex laterally, resulting
Figure 13.3. Magnetic Resonance Image of the Heart in Frontal
Plane. Ao indicates aorta; LV, left ventricle; PA, pulmonary artery; RA,
right atrium.
in a ratio more than 0.5 in the presence of a normal-sized heart.
Large pericardial fat pads may give a falsely increased cardio-
thoracic ratio. Because of these factors, relying on the cardiotho-
racic ratio alone to diagnose cardiomegaly can be misleading;
however, it does serve as a baseline for future comparisons.
• A cardiothoracic ratio >0.5 with a normal heart size occurs
with absent pericardium, pectus excavatum, obesity, and poor
inspiration.
Generalized Cardiac Enlargement
Global heart enlargement, with maintenance of an otherwise
normal cardiac contour, usually is due to diffuse myocardial

Aorta

Pulmonary
Artery

Right Left
Ventricle Atrium

Left
Ventricle

Figure 13.4. Tetralogy of Fallot. Indentation in region of left pul-

monary artery and elevation of apex due to right ventricular hypertrophy
give rise to boot-shaped contour, typical for this condition. (Previously
Figure 13.2. Lateral Projection of the Heart. published. See “Credit Lines” section.)
 13 Cardiac Radiography
LPA
RPA
RA
RV
Figure 13.5. Relative Positions of Right Heart Chambers on
Posteroanterior Angiogram. LPA indicates left pulmonary artery; RA,
right atrium; RPA, right pulmonary artery; RV, right ventricle.
disease, abnormal volume or pressure overload as a consequence
of valvular heart disease, hyperthyroidism, hypothyroidism, or
anemia. Pericardial effusions also produce generalized enlarge-
ment of the cardiac silhouette (Figure 13.11). Asymmetric
enlargement with left ventricular prominence can be seen in the
late stages of essential hypertension and other left-sided obstruc-
tive lesions with secondary left ventricular failure or in left-sided
regurgitant valvular lesions (Figure 13.12).
Left Atrial Enlargement
The left atrium sits just below the angle of the carina, in proximity
with the left bronchus and esophagus; thus, enlargement is read-
ily reflected by the displacement of these neighboring structures.
Enlargement usually produces a double density behind the right
atrial margin on a frontal projection as the left atrium bulges out
from the mediastinum into the right lung. Occasionally, a double
PV
PV
LA
PV PV
Figure 13.6. Drainage of Pulmonary Veins (PV) Into Left Atrium
(LA) on Angiogram.
161
LPA
MPA
RPA
RA
RV
Figure 13.7. Relative Positions of Right Heart Chambers on Lateral
Angiogram. LPA indicates left pulmonary artery; MPA, main pulmo-
nary artery; RA, right atrium; RPA, right pulmonary artery; RV, right
ventricle.
density is seen in the presence of a normal-sized left atrium in
patients with a prominent right pulmonary venous confluence.
Additional signs of left atrial enlargement on the posteroan-
terior projection include upward and posterior displacement of
the left main bronchus, resulting in a less acute carinal angle.
Enlargement of the left atrial appendage initially causes straight-
ening and, subsequently, a convexity in the upper left cardiac con-
tour. In the presence of a giant left atrium, the left atrium itself
may project beyond the right atrium and form a portion of the right
cardiac contour. On the lateral projection, left atrial enlargement
can be recognized by posterior and upward displacement of the
Ao
PA
LA
RA
IVC
Figure 13.8. Close Anatomical Relationship of Right and Left Atria
(RA and LA), Aorta (Ao), and Pulmonary Artery (PA) on Magnetic
Resonance Image. IVC indicates inferior vena cava.
162 II Noninvasive Imaging

A B

Ao

Ao
LA

LA

LV

LV

Figure 13.9. Angiograms Showing Relative Position of Left Heart Chambers on Frontal (A) and Lateral (B) Angiograms. Ao indicates aorta;
LA, left atrium; LV, left ventricle.

left main stem bronchus. The left atrium itself enlarges upward
and posteriorly to form an increasing convex density.
• Signs of left atrial enlargement are double density of right heart
border and upward and posterior displacement of left main bron-
chus; these result in widening of the carinal angle.
Isolated left atrial enlargement most commonly is due to
mitral valve stenosis caused by rheumatic heart disease (Figures
13.13 and 13.14). Left atrial myxoma and cor triatriatum can also
cause isolated left atrial enlargement. Isolated enlargement of the
left atrial appendage or apparent enlargement due to a pericardial
defect and focal herniation of the appendage may cause a local-
ized bulge in the upper left cardiac contour without other signs of
left atrial dilatation. Left atrial enlargement in combination with
additional chamber involvement may be produced by various
conditions, such as left ventricular failure, left-sided obstructive
lesions, and certain shunts (eg, ventricular septal defect, pat-
ent ductus arteriosus, and aortopulmonary window). However,
left atrial enlargement does not occur with simple atrial septal
defects. When left atrial enlargement is marked, it most often is
due to rheumatic valvular disease.
• Isolated left atrial enlargement most commonly is due to mitral
valve disease.
• Rarely, cor triatriatum or left atrial myxoma causes isolated left
atrial enlargement.
• Left atrial enlargement does not occur with simple atrial septal
defects.

A B

AVR

AVR

MVR

MVR

TVR
TVR

Figure 13.10. Prosthetic Starr-Edwards Valves on Frontal (A) and Lateral (B) Projections. AVR indicates aortic valve prosthesis; MVR, mitral
valve prosthesis; TVR, tricuspid valve prosthesis.
 13 Cardiac Radiography 163

Figure 13.11. Markedly Enlarged Cardiac Silhouette. This abnor-

mality was primarily due to a large malignant pericardial effusion
resulting from a sarcoma invading the heart chambers on the right. B

Left Ventricular Enlargement

Left ventricular enlargement can be due to dilatation or hypertro-
phy or both. Considerable hypertrophy must be present to cause
the cardiac shadow to enlarge appreciably. The classic appearance
of left ventricular hypertrophy on the posteroanterior projection
is rounding of the cardiac apex, with downward and lateral dis-
placement without cardiac enlargement. Left ventricular dilata-
tion causes an increase in the transverse diameter of the heart and
cardiothoracic ratio, together with an apparent increase in the
length of the left heart border. The cardiac apex may be displaced
to the extent that it projects below the diaphragm. On the lateral
projection, dilatation increases the posterior convexity of the left
ventricular contour, which will project behind the edge of the
vertical inferior vena cava. Obstruction to left ventricular empty-
ing or increased afterload, as caused by systemic hypertension,
aortic coarctation, or aortic valve stenosis, leads to hypertrophy
Figure 13.13. Mitral Stenosis Resulting in Left Atrial Enlargement,
Pulmonary Venous Hypertension, and Right Ventricular Dilatation.
Note double density projection over the right atrium (arrows) because
of dilatation of left atrium.

initially, with rounding of the cardiac apex (Figure 13.15). Left

Figure 13.12. Multichamber Cardiac Enlargement Resulting From
Rheumatic Heart Disease. The left atrium is the most dilated chamber.
Note prominence of the left atrial appendage (white arrows) and marked
splaying of the carina (arrowheads). A Hancock valve is in mitral
position (black arrow).
ventricular dilatation with cardiac failure may follow. Dilated
cardiomyopathy, especially ischemic cardiomyopathy, primarily
enlarges the left ventricle. Aortic valve regurgitation and mitral
valve regurgitation enlarge the left ventricle and are associated
with dilatation of the aorta and left atrium, respectively.
Left ventricular aneurysms, usually the result of a previous
myocardial infarction, occasionally result in a localized bulge
that projects beyond the normal ventricular contour or an angula-
tion of the left ventricular contour (Figure 13.16). A large apical
aneurysm can appear similar to simple left ventricular chamber
dilatation. Sometimes with true aneurysms of the left ventricle,
the heart appears normal in size and contour. False aneurysms
often are paracardiac in location, posterior and inferior to the
left ventricle. All cardiac chambers have been reported to be
involved with aneurysm formation, although atrial aneurysms
are extremely rare.
• In the absence of heart failure, left ventricular hypertrophy must be
massive before the heart shadow enlarges.
164 II Noninvasive Imaging

A B

Figure 13.14. Mitral Stenosis With Left Atrial Enlargement (arrows) and Calcification of Atrial Wall. A, Posteroanterior projection. B, Lateral
projection. A ball-cage valve prosthesis has been placed in the mitral position.

Right Atrial Enlargement
Isolated right atrial enlargement is detected best on a frontal radi-
ograph. Enlargement is to the right and causes increased fullness
and convexity of the right cardiac contour and angulation of the
junction of the superior vena cava and right atrium. There may
be associated dilatation of the superior and inferior venae cavae
that causes widening of the right superior mediastinum and an
additional border in the right cardiophrenic angle. On the lateral
projection, right atrial dilatation is often difficult to appreciate.
It causes a filling-in of the retrosternal clear space anteriorly and
superiorly, with the cardiac silhouette extending behind the ster-
num more than one-third the way above the cardiophrenic angle,
similar to that seen with right ventricular enlargement. There
may be a double density that merges with the inferior vena caval
shadow, which may be a slightly convex structure. Left atrial
enlargement can be simulated by marked right atrial dilatation.
• Right atrial enlargement fills in the retrosternal clear space on the
lateral projection.
Isolated right atrial enlargement is uncommon and usually is
due to tricuspid stenosis or right atrial tumor. Right atrial dil-
atation associated with other chamber enlargement, primarily
right ventricular enlargement, occurs in several conditions, such
as tricuspid regurgitation, pulmonary arterial hypertension,
shunts to the right atrium, and cardiomyopathies (Figures 13.17

A B

Figure 13.15. Enlarged Left Ventricle With Dilatation of Ascending Aorta Due to Combined Aortic Insufficiency and Aortic Stenosis. The aortic
valve is calcified (arrows) and pulmonary arteries are enlarged in this patient, who also has chronic obstructive pulmonary disease. A, Posteroanterior
projection. B, Lateral projection.
 13 Cardiac Radiography
A B
Figure 13.16. Marked Enlargement of Left Ventricle Due to Left Ventricular Aneurysm. A, Posteroanterior projection. B, Curvilinear calcifica-
tion outlines aneurysm (arrows).
and 13.18). Marked isolated right atrial enlargement resulting in
a box-shaped heart occurs in Ebstein malformation of the tricus-
pid valve (Figure 13.19). This configuration of the heart is the
result of marked angulation at the superior vena caval-right atrial
junction as the right atrium enlarges.
• Ebstein anomaly causes a box-shaped heart.
Right Ventricular Enlargement
The right ventricle enlarges by broadening its triangular shape
in the superior and leftward direction. With increasing right ven-
tricular enlargement, the entire heart rotates to the left around
Figure 13.17. Marked Right Atrial Dilatation and Right Ventricular
Dilatation Due to Severe Tricuspid Regurgitation. Regurgitation was
related to traumatic injury of tricuspid valve.
165
its long axis and displaces the left ventricle posteriorly. This dis-
placement causes increased convexity of the left upper heart bor-
der and elevation of the cardiac apex. The rotation also makes
the pulmonary trunk appear relatively small. With marked dil-
atation, the right ventricle may form the left heart border on the
posteroanterior projection.
On the lateral projection, the right ventricle extends cranially
behind the sternum, with increased bulk anteriorly. Normally,
the heart does not extend more than one-third of the distance
from the cardiophrenic angle to the sternal angle or the level of
the carina; however, normal extension can vary with body habi-
tus. Isolated right ventricular enlargement is very unusual. More
typically, there is associated prominence of the right atrium and
pulmonary trunk.
Pulmonary Vasculature
Because the pulmonary vasculature reflects the physiologic
effects of a cardiac lesion, it provides important clues to the diag-
nosis. Radiographic abnormalities are primarily the result of an
increase in pulmonary blood flow or an obstruction to flow some-
where in the pulmonary circuit.
Normal Pulmonary Blood Flow
The pulmonary arteries and veins extend outward from each hilum
in an orderly branching fashion, gradually tapering peripherally.
The hilar density is composed of the proximal pulmonary arter-
ies, and the left hilum normally projects more cranially than the
right one because of the course of the left pulmonary artery over
the left main bronchus. In the upper lobes, the veins and arter-
ies are essentially parallel, the veins lying lateral to their corre-
sponding arteries. The major arteries and veins in the lower lung
fields cross each other, the veins taking a more horizontal course
toward the left atrium. In the upright position, there is increased
flow to the base of the lungs (largely due to the effects of gravity),
which causes the lower-lobe vessels to increase in size. It may be
difficult to identify the apical vessels clearly because pulmonary
166 II Noninvasive Imaging
A B
Figure 13.18. Combined Mitral Stenosis and Mitral Insufficiency Resulting in Left Atrial Enlargement, Marked Right Ventricular Enlargement,
and Slight Left Ventricular Enlargement. A, Posteroanterior projection. B, Lateral projection. Dilated ventricles (arrows) are appreciated best on
lateral view.
flow to the apices is negligible in the upright position. Therefore,
position has a marked effect on flow distribution.
Increased Pulmonary Blood Flow
As pulmonary flow increases, the pulmonary vessels, both arter-
ies and veins, become enlarged. These enlarged vessels become
apparent when pulmonary flow is approximately twice normal.
The overcirculation pattern may be symmetric or asymmetric.
High-output states with increased circulating blood volume,
such as anemia, pregnancy, thyrotoxicosis, overhydration, and
fever, result in a symmetric increase in vascularity, as do var-
ious congenital defects characterized by left-to-right shunts
(Figures 13.20 and 13.21). An asymmetric increase in pulmonary
flow may be congenital in origin (eg, pulmonary arteriovenous
malformation, anomalous origin of a pulmonary artery) but is
more commonly the result of surgical intervention to create a
Figure 13.19. Enlarged Box-shaped Heart With Decreased
Pulmonary Vascularity Typical of Ebstein Anomaly.
systemic-to-pulmonary shunt to improve pulmonary blood flow
in the presence of severe pulmonary stenosis or atresia (eg, a
Blalock-Taussig shunt).
Decreased Pulmonary Blood Flow
Essentially all the linear shadows in the normal lung fields are
due to pulmonary vasculature. When flow and, therefore, ves-
sel size are diminished, the lung fields appear abnormally radi-
olucent. Both symmetric and asymmetric patterns of abnormal
vascularity can be observed. Generalized undercirculation can
be due to an obstructive lesion in the right heart, as in tetralogy
of Fallot, pulmonary atresia, right ventricular tumor, or tricus-
pid valve atresia. Small-caliber pulmonary vessels with relatively
hyperlucent lungs and a small heart are evidence of a marked
decrease in the circulating blood volume (eg, in Addison disease,
hemorrhage). Chronic obstructive pulmonary disease may result
in generalized lung destruction or, more commonly, a patchy
distribution of decreased vascularity. Segmental and asymmet-
ric decreases in pulmonary vascularity are seen with pulmonary
embolic disease (Westermark sign), segmental chronic obstruc-
tive pulmonary disease, partial pneumonectomy, and branch pul-
monary artery stenoses (Figure 13.22). Rarely, postinflammatory
changes (eg, granulomatous mediastinitis), extrinsic compression
(eg, aortic aneurysm), and congenital hypoplasia as seen in the
scimitar syndrome result in areas of decreased pulmonary flow.
Bronchial collateral circulation may become prominent, with a
somewhat disordered pattern, when there is a decrease in pul-
monary artery blood flow, and occasionally it gives the illusion
that the overall vascularity is actually normal or even increased.
Small hila in tetralogy of Fallot or pulmonary atresia and loss of
the normal branching pattern of pulmonary vasculature should
be evident on the chest radiograph.
• If the lung hila are small, consider tetralogy of Fallot or pulmo-
nary atresia.
Increased Resistance to Pulmonary Blood Flow
Pulmonary hypertension with redistribution of flow is the result
of increased resistance in the pulmonary circuit. Recognition
 13 Cardiac Radiography
A B
Figure 13.20. Two Examples of Atrial Septal Defects. A, Mild right ventricular dilatation in an asymptomatic patient. B, More prominent shunt
vascularity and cardiac enlargement in a patient with very mild dyspnea on exertion.
of the various redistribution patterns seen on chest radiographs
often allows the level of the increased resistance and the possible
underlying abnormality to be identified.
Pulmonary Venous Hypertension
Lesions acting beyond the pulmonary capillary level result in
increase of the pulmonary venous pressure. Left ventricular dys-
function and mitral valve disease are the most common causes of
pulmonary venous hypertension; other obstructive lesions at the
left atrial level (eg, atrial myxoma, cor triatriatum, thrombus)
Figure 13.21. Patent Ductus Arteriosus Resulting in Prominence
of the Aortic Arch and Shunt Vascularity. There is mild left ventricular
enlargement.
167
or pulmonary vein level (eg, stenosis, veno-occlusive disease, or
thrombosis) are relatively rare.
Initially, because of the increase in venous pressure, venous
dilatation occurs throughout the lungs. However, the typical radi-
ographic pattern is prominent upper lung vessels, both arteries
and veins. This phenomenon is thought to be due to a localized
segmental reflex initiated by the increase in pulmonary venous
pressure above a critical level of about 10 to 15 mm Hg. An
additional factor is the accumulation of fluid around compress-
ible small vessels when plasma oncotic pressure is exceeded by
pulmonary venous pressure. When a person is in the upright
position, the pressure in the lower lung is greater because of
hydrostatic forces; therefore, vasoconstriction of both arter-
ies and veins occurs here first and increases resistance to flow,
thereby reducing the circulatory volume through these vessels.
To overcome the increased resistance and to maintain a gradi-
ent in the presence of increased pulmonary venous pressure, the
pulmonary artery pressure must increase, resulting in increased
flow to the apices. The diverted pulmonary flow increases the
size and visibility of the upper-lobe vessels (Figure 13.23). As
pulmonary venous hypertension increases to the order of 25 mm
Hg, there is increased transudation of plasma from the lower lung
capillaries, and this results in interstitial edema. In addition to
obscuring further the now smaller and crowded lower-lobe ves-
sels, this transudation results in the radiographic appearance
of septal lines (Kerley lines), which are due to fluid within the
interlobular septa (Figure 13.24). Still further increase in pul-
monary venous pressure results in transudation of plasma into
the alveoli, producing classic alveolar edema when the pressure
exceeds 30 mm Hg.
Pulmonary Arterial Hypertension
Increased resistance at the pulmonary capillary or arteriolar
level increases pulmonary artery pressure. The causes of pul-
monary arterial hypertension include 1) obstructive processes
(eg, chronic pulmonary emboli, idiopathic or primary pulmo-
nary arterial hypertension, pulmonary schistosomiasis), 2) oblit-
erative processes (eg, pulmonary fibrosis, chronic obstructive
168 II Noninvasive Imaging

A B

Figure 13.22. A, Posteroanterior projection showing decreased vascular markings in both lungs, most marked in right upper lobe. B, Angiogram
showing large bilateral emboli, resulting in little flow to right upper lobe and left lower lobe.

pulmonary disease), 3) constrictive processes (eg, chronic
hypoxia), and 4) increased flow as seen in large left-to-right
shunts with development of Eisenmenger syndrome (Figures
13.25 and 13.26). Radiographically, pulmonary arteries are
dilated centrally, with an abrupt disparity in the caliber of the
central and intrapulmonary arteries or “pruning” of the intra-
pulmonary branches. This uneven response is thought to be
due to constriction of the muscular intrapulmonary branches in
response to the increased intraluminal pressures, with dilatation
of the more elastic central arteries.
• The classic chest radiographic findings in pulmonary arterial
hypertension are dilated distal proximal pulmonary arteries and
“pruning” of intrapulmonary branches.

Pericardial Disease
Normal pericardium is seldom identified on plain chest radio-
graphs. It may be visible as a sharp line at the cardiac apex,
outlined by epicardial and mediastinal fat.

A B

C D

Figure 13.23. A-D, Serial Radiographs Showing Development of

Pulmonary Venous Hypertension Continuing on to Florid Pulmonary
Edema in a Patient with a Large Myocardial Infarction. Note progressive
redistribution of the prominence of the pulmonary vessels to the right upper Figure 13.24. Interstitial Edema With Appearance of Kerley Lines
lobe. The hilar vessels become much less distinct as the edema develops. (arrows). These are due to fluid within the interlobular septa.
 13 Cardiac Radiography 169

A B

Figure 13.25. Pulmonary Hypertension Due to Chronic Pulmonary Emboli. Note the central pulmonary artery enlargement (arrows) (A) and
right ventricular dilatation seen best in the lateral projection (B). The angiogram shows classic arterial occlusions and stenoses (arrowheads) (C).

Pericardial Effusion
A pericardial stripe wider than 2 mm that parallels the lower
heart border, usually in the lateral projection and best identified
in the sternophrenic angle, is diagnostic of a pericardial effusion.
The only clue to a relatively small effusion may be a noticea-
ble change in heart size compared with that on previous studies.
The classic water-flask configuration of a large effusion may not
be present, and the appearance of the cardiac silhouette may be
identical to that in dilated cardiomyopathy with no significant
distortion other than enlargement. A large heart with a prom-
inent superior vena cava and azygos vein in combination with
decreased pulmonary vasculature should raise the question of
cardiac tamponade. Acutely, a relatively small effusion can cause
tamponade with minimal enlargement of the cardiac shadow.
Pericardial Calcification
Constrictive pericarditis may occur as the result of pericarditis
and pericardial effusion of any cause. Calcification of the pericar-
dium is highly suggestive but not pathognomonic of constrictive
pericarditis. In more than 50% of patients with constrictive peri-
carditis, calcifications do not show on the plain chest radiograph.
Calcifications are found frequently on the anterior and diaphrag-
matic surfaces, but they may be over any part of the heart. Linear
or plaquelike calcifications, often best seen on the lateral view,
are typically projected over the right ventricle or the atrioventric-
ular groove (Figure 13.27). The entire heart may appear encased
in a shell. The calcification may be dense and thick.
• In more than 50% of patients with constrictive pericarditis, peri-
cardial calcification does not show on chest radiography.
Pericardial Defects
Congenital or surgical absence of the pericardium may result
in changes in the cardiac contours. Congenital absence is more
commonly left-sided and rarely right-sided. Partial defects may
allow a portion of the heart (usually the left atrial appendage
in congenital defects) to herniate outside the pericardial sac,
with the herniated portion producing a bulge in the contour of
the heart. “Complete” absence of the pericardium is actually a
170 II Noninvasive Imaging

Figure 13.26. Marked Enlargement of the Pulmonary Arteries

Centrally Due to Eisenmenger Syndrome Caused by Long-Standing
Atrial Septal Defect.

unilateral defect and nearly always left-sided. The heart appears

shifted to the left without a shift in the trachea (Figure 13.28).
The left cardiac contour has an elongated appearance. The pul-
monary artery often appears prominent and sharply defined. A
somewhat similar appearance is seen on the frontal projection
when the heart is rotated because of compression of the chest
wall in patients with pectus excavatum deformity.
• Partial or “complete” absence of the pericardium is usually left-
sided.

Cardiac Masses
C
The role of plain chest radiographs in the identification of cardiac
masses is often limited. Radiographic manifestations are depend-
ent on tumor size, location, and type. With many intracavitary
and intramural tumors of even moderate size, no changes are
seen on plain radiographs unless hemodynamic alterations are
produced, such as the mimicking of mitral stenosis by a left atrial
myxoma. Left ventricular aneurysms, pericardial cysts, extracar-
diac mediastinal masses, loculated pericardial cysts, and locu-
lated pericardial effusions are all causes of abnormal contours
that can be indistinguishable from neoplasms (Figures 13.29 and
13.30). The presence of calcification may help in the detection of
a mass, but calcification patterns are not specific, and differenti-
ation from calcification of thrombus or normal structures usually
requires additional imaging methods.

Aortic Disease
The aortic knob, representing the foreshortened transverse aor-
tic arch, is the only border-forming portion of the normal tho-
racic aorta that is otherwise hidden within the mediastinum. The
descending thoracic aorta parallels the thoracic spine on the left.
With the development of atherosclerotic aortic disease, unfolding
and ectasia (dilatation and elongation) of the aorta occur. As the
Figure 13.27. A and B, Plain radiographs showing constrictive
pericarditis with circumferential pericardial calcifications (arrows).
Note the pulmonary venous hypertension and right pleural effusion. C,
Computed tomographic image through the mid ventricles better shows
the circumferential nature of the relatively coarse calcifications.
 13 Cardiac Radiography 171

Figure 13.28. Absence of Pericardium Resulting in Displacement

of Cardiac Apex to the Left, Mimicking an Oblique Projection.

Figure 13.30. Well-Defined Rounded Mass Projected Adjacent to

Right Atrium. Heart and pulmonary vasculature are otherwise normal.
Appearance is typical of a pericardial cyst. A, Posteroanterior projec-
tion. B, Lateral projection.

Figure 13.29. A, Mass (arrows) in right cardiophrenic angle con-

sistent with a prominent cardiac fat pad. B, High signal of adipose tissue
(arrows) is shown with magnetic resonance imaging of this region.
172 II Noninvasive Imaging

A B

Figure 13.31. Marked Enlargement of Aortic Knob and Widening of Mediastinum Due to a Large Ascending Aortic Aneurysm. A, Posteroanterior
projection. B, Lateral projection. The size of the ascending aorta is appreciated better on the lateral view. The patient has significant aortic regurgi-
tation with bilateral pleural effusions.

descending aorta swings into the left chest, more and more of the
contour becomes silhouetted by lung; on the lateral projection,
a portion of the descending aorta may be shown, and only then
is a clue to the presence of an aneurysm obtained. Unfolding or
ectasia of the ascending aorta produces a convexity of the right
superior mediastinum. These findings may be indistinguishable
from those present with an aortic aneurysm. The most common
A
finding of an aortic aneurysm on a frontal chest radiograph is
widening of the superior mediastinum (Figure 13.31). Other
chest radiographic findings suggestive of a thoracic aortic aneu-
rysm, whether atherosclerotic, luetic, dissecting, or traumatic,
include displacement or compression (or both) of the trachea and
esophagus either to the left and posteriorly by an aneurysm of
the ascending aorta or to the right and anteriorly by an aneurysm
of the descending aorta. Calcification in the aorta is a common
finding in atherosclerotic aortic disease. Because the aorta is
largely hidden by the mediastinal silhouette, the cross-sectional
methods, such as computed tomography and magnetic resonance
imaging, provide greater detail in the evaluation and follow-up of
aortic disease (Figures 13.32 through 13.34).

Figure 13.32. A, Posteroanterior projection showing marked enlargement of ascending aorta caused by a dissecting aneurysm. B, This is appre-
ciated better on a computed tomographic scan. Note discrepancy in size between the ascending aorta (arrows) and the main pulmonary artery
(arrowheads).
 13 Cardiac Radiography 173

Figure 13.33. A, Enlargement of aortic arch containing curvilinear calcification (arrow). B, Computed tomographic scan shows better the small
pseudoaneurysm containing peripheral calcification (arrows), the result of remote trauma.

A B

Figure 13.34. A, Double contour to aortic arch typical of coarctation. B, Magnetic resonance image nicely shows the focal narrowing and result-
ing kinking (arrows) of proximal descending aorta.
174 II Noninvasive Imaging

A B

Figure 13.35. St. Jude Bileaflet Prosthetic Valve in the Mitral Position. Only the valve ring is visible (arrow). A, Posteroanterior projection. B,
Lateral projection.

A B

Figure 13.36. Mitroflow Pericardial Valve in the Aortic Position. Only the valve ring is visible (arrow). A, Posteroanterior projection. B, Lateral
projection.
 13 Cardiac Radiography
Figure 13.37. Carpentier-Edwards Valve in the Aortic Position. The
valve ring and struts are visible in this lateral projection (arrow).
Implanted Cardiovascular Devices
An increasing number of patients have prosthetic cardiac valves,
septal occluders, and rhythm management devices. It is impor-
tant to be able to recognize cardiac hardware on the chest radio-
graph and to be able to identify normal and abnormal appearance
and position.
Figure 13.38. Edwards-Sapien Transcutaneous Aortic Valve on
Lateral Projection (arrow).
175
Valves
More than 80 models of heart valves have been in use since their
introduction. Although it is impossible to be able recognize all
models on the chest radiograph, it is helpful to be familiar with
the appearance of commonly used prosthetic valves and to be able
to ascertain the type and position of a prosthetic valve. The mitral
and aortic valves are the most commonly replaced. Figure 13.10
conduit
conduit
B
conduit
conduit
Figure 13.39. Ventricular Assist Devices. A and B, Implanted left
ventricular assist device. The inflow conduit exits from the left ventric-
ular apex. The outflow conduit connects the device to the aorta. The
cable connects the device to an external power source and system con-
trol module. A dual-chamber, biventricular implantable cardioverter-
defibrillator with leads in the right atrium, right ventricle, and coronary
sinus is also seen. C and D, Total implantable heart. Both ventricles and
all 4 native cardiac valves are removed; pneumatic chambers (arrows)
and an external power source are noted. The valves seen are part of the
implanted device. (See “Credit Lines” section.)
176 II Noninvasive Imaging
D visible in lateral projection (arrow).
Figure 13.39. (Continued)
illustrates ball-cage Starr-Edwards mechanical valves in the aor-
tic, mitral, and tricuspid positions. Figure 13.12 is an example of a
Hancock bioprosthetic valve in the mitral position. Figures 13.35
through 13.37 illustrate additional prosthetic valves.
Evidence of valve repair is not typically visible on the chest
radiograph. Annuloplasty rings may be seen in the tricuspid and
mitral positions, and they may be closed or open.
Figure 13.40. Atrial Septal Occluder Device. Radiographic mark is
Transcatheter Aortic Valve Implant Devices
Prosthetic aortic valves implanted transcutaneously are best
viewed on lateral chest radiographs. Although the bioprosthetic
valve leaflets are not visible, the mesh framework should be fully
expanded (Figure 13.38).
Ventricular Assist Devices
Left ventricular assist devices are implanted to augment cardiac
output in patients with severe left ventricular dysfunction and
failure. As illustrated in Figure 13.37, patients frequently have
evidence of prior cardiac interventions, such as placement of
rhythm management devices, coronary artery bypass grafting,
or valvular surgery, that can be detected on the chest radiograph.
Patients may have a right ventricular assist device, a left ventric-
ular assist device, a biventricular device, or a total artificial heart
in place (Figure 13.39).
Septal Occluder Devices
Septal occluder devices allow percutaneous treatment of patent
foramen ovale and atrial septal defect meeting anatomic criteria.
Device markers are visualized most easily on the lateral chest
radiograph (Figure 13.40).
 14

Atlas of Congenital Heart Defects

SABRINA D. PHILLIPS, MD, and FRANK CETTA JR, MD

Figure 14.1. Scimitar Syndrome. Scimitar syndrome consists of Figure 14.2. Atrial Septal Defect. The central pulmonary arteries
partial anomalous pulmonary venous return of right pulmonary vein(s) are enlarged, and vascularity is prominent throughout the lung fields.
to the inferior vena cava, below the diaphragm. This results in dextropo- There is mild cardiac enlargement. This is consistent with pulmonary
sition of the heart, right lung hypoplasia with a small right hemithorax, overcirculation from a substantial left-to-right intracardiac shunt.
and a curvilinear structure (arrow) in the right lung field (the scimitar),
which represents the anomalous pulmonary vein(s). All of these fea-
tures are shown here.

177
178 II Noninvasive Imaging

Figure 14.3. Ebstein Anomaly. A, Cardiac enlargement with a rounded contour of the cardiac silhouette is consistent with right chamber enlarge-
ment. Aortic arch (arrow) is small. B, Obliteration of retrosternal space (arrow) in lateral projection confirms right-sided chamber enlargement.

Figure 14.4. Eisenmenger Syndrome. Note marked cardiac enlarge- Figure 14.5. Coarctation of Aorta. Note rib notching due to inter-
ment with large central pulmonary arteries and “pruned” distal pulmo- costal collaterals (orange arrow). Contour of the descending aorta is
nary arteries. abnormal, consistent with coarctation (white arrow).
 14 Atlas of Congenital Heart Defects 179

Figure 14.6. Coarctation of Aorta. Magnetic resonance angiogram from a patient with prior repair shows aneurysm formation at site of previous
patch enlargement of aorta (arrow). Complications of coarctation repair include both recurrent obstruction and aneurysm formation.

A B

Figure 14.7. Coarctation of Aorta, Angiographic Findings. A, Arrow points to coarctation. B, Stent implantation in the coarctation shown in A.
C, After stent placement.
180 II Noninvasive Imaging

A
B

Figure 14.8. Dextrotransposition of the Great Arteries After Mustard Operation (Atrial Switch). A, Rounded cardiac border is indicative of mor-
phologic right ventricular enlargement. The right ventricle serves as the systemic ventricle in this situation. B, Pacemaker is in the posterior chamber
of the heart, which is the morphologic left ventricle serving as the subpulmonic ventricle after a Mustard procedure.

A B
Pulmonary
artery
Aorta

Figure 14.9. Complete Transposition of the Great Arteries (D-TGA) After Atrial Switch Procedure. A, Systemic ventriculogram. Ventricle is
heavily trabeculated and anterior (morphologic right ventricle). Note pacing leads in posterior ventricular apex. B, Subpulmonary ventriculogram.
Ventricle is smooth-walled and posterior (morphologic left ventricle).
 14 Atlas of Congenital Heart Defects 181

A B

Figure 14.10. Congenitally Corrected Transposition of the Great Arteries (L-TGA). A, Subpulmonary ventriculogram. Ventricle is morphologi-
cally a left ventricle and is posterior. B, Systemic ventriculogram. Systemic ventricle is a morphologic right ventricle and is anterior and leftward.

Figure 14.11. Unrepaired Tetralogy of Fallot in an Infant. A, Posteroanterior radiograph. B, Lateral radiograph. Right atrium is enlarged,
and cardiac apex contour is consistent with right ventricular enlargement. Paucity of flow into pulmonary bed is due to right ventricular outflow
obstruction.
182 II Noninvasive Imaging

Figure 14.12. Unrepaired Tetralogy of Fallot in an Infant. Angiogram shows increased muscular trabeculation in right ventricle, right ventricular
outflow tract narrowing (arrow), and filling of aorta with injection due to flow through ventricular septal defect.

B
A

Figure 14.13. Adult With Previous Repair of Tetralogy of Fallot, Now With Severe Pulmonary Valve Regurgitation. A, Posteroanterior radio-
graph. B, Lateral radiograph. Note cardiac enlargement, specifically right atrial enlargement (arrow in Figure 14.13A), central pulmonary enlarge-
ment, and rounded contour of ventricle, which is consistent with right ventricular dilatation. An intracardiac defibrillator has been implanted for
ventricular tachycardia, which is a consequence of severe right ventricular enlargement and dysfunction related to pulmonary valve regurgitation.
 14 Atlas of Congenital Heart Defects 183

B
A

PA

RV

Figure 14.14. Severe Pulmonary Valve Regurgitation After Previous Pulmonary Conduit Implantation. A, Lateral angiogram shows catheter
traversing right ventricle (RV) and conduit. Distal tip of catheter is in pulmonary artery (PA). B, Angiogram obtained after deployment of Melody
valve (percutaneous tissue valve) in conduit shown in Figure 14A shows that contrast does not fill right ventricle because a competent valve is now
in place.

A B

Figure 14.15. Large Patent Ductus Arteriosus. A, Lateral angiogram shows large ductus (arrows) with injection of contrast into aorta. B, Balloon
sizing of patent ductus arteriosus. Note that balloon catheter course is from inferior vena cava to aorta. Pigtail catheter (single arrow) is in aorta,
and there is an end-hole catheter (double arrow) in pulmonary artery for reference. C, Release of vascular plug (arrow) to occlude patent ductus
arteriosus.
184 II Noninvasive Imaging

Figure 14.15. (Continued)

A B

SVC

Figure 14.16. Right Coronary Artery Fistula to Superior Vena Cava (SVC) on Angiography. A, Segment of coronary artery proximal to fistula is
severely dilated (arrow). B, Preparation for occlusion of coronary fistula. Guide catheter is in SVC. Guidewire traverses aorta, coronary artery, and
fistula, terminating in SVC.

Cardiopulmonary Exercise Testing
in Clinical Medicine
THOMAS G. ALLISON, PHD, MPH
History of CPX
. diopulmonary measurements to the standard exercise test. The
CPX—also called metabolic stress testing and Vo2 max testing—
is an important tool in a state-of-the-art, comprehensive cardi-
ovascular center (Table 15.1). Originally, CPX was developed
for testing the fitness of athletes,. such as long-distance runners
and cross-country skiers since Vo2 max is the most important
predictor
. of performance in endurance events. Measurement of
Vo2 during various forms of physical labor was also an important
early use of CPX. . bon dioxide concentrations are measured, and a forehead sen-
Initially, measurement of Vo2 was cumbersome and labor
intensive because expired air was first collected in meteorologic
balloons or large canvas bags (Douglas bags). They were evacu-
ated into large, calibrated tanks (Tissot tanks) to measure the
volume of expired gas. Samples of mixed gas were analyzed
manually for oxygen and carbon dioxide content with instru-
ments that measured the gas volume before and after oxygen
or carbon dioxide was selectively removed
. by chemicals (with
a Schollander analyzer). Finally, Vo2 was calculated from the
measurements with specified equations that used the Haldane
conversion, which is based on the nonmetabolic properties of
nitrogen with corrections for temperature, barometric pressure,
and relative humidity of the inspired air.
Fortunately, fully automated, accurate systems now enable
breath-to-breath measurement of oxygen consumption without
the need to collect large volumes of gas to perform gas analysis
or calculations manually. The automated systems are compact
(especially when a laptop computer is used), they autocalibrate
quickly, they integrate with a treadmill or cycle ergometer and
electrocardiographic stress monitors, and relatively little extra
Abbreviations and acronyms are expanded at the end of this chapter.
15
technician time or patient inconvenience is involved to add car-
additional time and cost are relatively low for CPX compared
with conventional treadmill testing with electrocardiographic
monitoring only, and CPX procedures are reimbursable through
established Current Procedural Terminology codes.
Additional CPX instrumentation includes a small, light, dis-
posable mouthpiece with nose clip (or a face mask), through
which the flow of expired gas and continuous oxygen and car-
sor to measure oxygen saturation transcutaneously (Figure 15.1).
Fingertip oximetry is less reliable; arterial blood gases can be
tested if high accuracy
. is critical. Numeric and graphic real-
time displays of Vo2 and other CPX variables are provided, and
printed tables and graphs are available within 5 minutes after the
test is completed.
.
Relationship of VO2 to Cardiac Output
.
It is important to understand that Vo2 as. measured during CPX
refers to oxygen
. uptake in the lungs. Vo2 at the tissue level is
equal to Vo2 over time,. but there may be instantaneous differ-
ences. For example, Vo2 is higher when a person is standing
compared with lying supine, but when the position is changed
from supine to standing, oxygen uptake decreases momentarily
because of reduced venous return and subsequently decreased
cardiac output and flow of blood through the lungs—equilibrated
quickly by increased.heart rate.
Mathematically, Vo2 is equal to the product of cardiac output
and arteriovenous oxygen difference. In turn, cardiac output is
a product of heart rate and stroke volume (Figure 15.2). Stroke
185
186 II Noninvasive Imaging

Table 15.1. Principles of Maximal Oxygen Consumption

.
Cardiac output
Vo2 max is an individual’s maximum capacity to transport and use
oxygen during incremental exercise and reflects overall cardiovascular
. tness
fi
Vo2 max is expressed in liters of oxygen per minute (L/min) or, more Heart rate Stroke volume
commonly, on a weight-adjusted basis as mL of oxygen per Kg per
. minute (mL/kg/min)
Vo2 max is.considered the single-best measure of cardiovascular fitness
A normal Vo2 max level will usually exclude the presence of significant End-diastolic volume End-systolic volume
. cardiovascular, pulmonary, or neuromuscular. dysfunction
Vo2 increases linearly with workload,
. and Vo2 max is identified when there
is a plateau in. the curve of Vo2 versus workload.
Filling pressure Compliance Contractility Afterload
Age decreases Vo2 max level; training increases Vo2 max level
Figure 15.2. Cardiac Output. Cardiac output is a product of heart
rate and stroke volume.

volume is the difference between end-diastolic and end-systolic

volumes. End-diastolic volume is affected by compliance of
the ventricle and by filling pressure (blood volume); end-systolic
volume is related to inherent contractility, preload, and afterload.
It is well known from the Starling principle that modest increases
in preload and end-diastolic volume increase stroke volume, even
in a heart weakened by severe ischemia or various myopathic
processes, whereas larger increases may distend the weak ventri-
cle to a point at which contractility decreases and stroke volume
is reduced.
Arterial oxygen content is affected by hemoglobin concentra-
tion, alveolar ventilation and oxygen tension, and pulmonary dif-
fusing capacity (Fig 15.3). The oxygen content of mixed venous
blood is determined by the mass of the working muscles and
the amount of oxygen extracted, which is dependent on regional
blood flow, capillary density, and oxidative enzyme levels in the
active muscle mass.
In healthy
. persons, several of these factors affect peak
exercise Vo2:
1. Genetics particularly affects heart and lung size, distribution of mus-
cle fiber types (the relative mix of fast twitch glycolytic fibers and
slow twitch oxidative fibers), and maximal heart rate
.
2. Sex influences peak exercise Vo2 through effects on heart and lung
size, skeletal muscle mass, and hemoglobin concentration
3. Aging reduces muscle mass, peak heart rate, and vital capacity of
the lung
4. Physical training, particularly aerobic
. exercise training, is an impor-
tant influence on peak exercise Vo2 through its effect on skeletal
muscle mass and oxidative enzyme capacity, peripheral muscle cap-
illary density, blood volume, and contractile properties of the heart
Before
. discussing the influence of body weight .on peak exer-
cise Vo2, it is essential to distinguish peak exercise Vo2 (expressed
.
as milliliters per minute or liters per minute) from Vo2 max
(expressed as milliliters of oxygen per kilogram. of body weight
per minute). (In this chapter, by convention, Vo2 refers to the .
unadjusted measurement in milliliters per minute, whereas Vo2
max
. has been adjusted for body weight.) The advantage of using
Vo2 max is that it helps standardize patients from a performance
standpoint: A small, lean woman and a large, lean man who can
perform at approximately the same . level in endurance events (eg,
running) will have a similar Vo2 max, . even though the man will
have a much higher peak exercise Vo2. In
. contrast, an overweight
man, who. has a higher peak exercise Vo2 than the woman, has
a .lower Vo2 max that is consistent with his lower performance.
(Vo2 max would correlate well with functional aerobic capacity,
which indexes exercise capacity to age and sex in nonmetabolic
exercise testing. Weight is not considered in calculations of func-
tional aerobic capacity.) By analogy, cardiac output is measured
in liters per minute but is often indexed according to body sur-
face area to determine adequacy across a broad range of patients.

Hemoglobin 2 diff

Arterial O2 content Mixed venous O2 content

PAO2 Pulmonary O2 extraction Muscle

diffusing mass
capacity
Alveolar Capillary Regional
ventilation density
Oxidative
enzyme
Figure 15.1. Cardiopulmonary Exercise Test Instrumentation. Use levels
of mouthpiece and nose clip. A 1-piece breathing mask may be used . .
instead of a mouthpiece and nose clip. (Previously published. See Figure 15.3. Arteriovenous Oxygen Difference (AVo2diff). AVo2
“Credit Lines” section.) diff is affected by several factors. O2 indicates oxygen.
 15 Cardiopulmonary Exercise Testing in Clinical Medicine
. .
A problem arises, however, when Vo2 is used not as an indication
of cardiovascular fitness but as a surrogate for cardiac output.
Excess adipose
. tissue, which is relatively metabolically inac-
tive, reduces Vo2 max, but it adds to the energy cost of perform-
ing weight-bearing work (eg, treadmill running) and increases
demand for. cardiac output. The solution is to develop predicted
values for Vo2 that account for not only age and sex but also body
weight (and sometimes height). .
These various influences on Vo2 max combine to produce a
wide
. range of “normal” values. Women have 80% to 90% of the
Vo2 max of men at similar levels of training because women have
smaller hearts. in relation to body weight and a higher percentage
of body fat. Vo2 max declines by 5% to 10% per decade of age,
depending on level of physical. activity (more active persons have
a slower decline with age). A Vo2 max of 20 mL/kg/min might be
considered “normal” for a sedentary, overweight woman of 70,
but a healthy
. and lean elite male endurance athlete in his 20s may
have a Vo2 max of 70 mL/kg/min . or more. Many equations have
been published for predicting Vo2 max during treadmill exercise
in healthy adults. The following are . examples of simple equa-
tions describing average values for Vo2 max by age and sex:
.
• Men: Predicted Vo2 max = 60 − (0.5 × Age) mL/kg/min
•
.
Women: Predicted Vo2 max = 48 − (0.4 × Age) mL/kg/min
As noted above, these equations account for the patient’s
weight, which affects the oxygen demand of weight-bearing
work (eg, use of a treadmill). Automated analyzers come with
a choice of several more sophisticated .prediction equations that
estimate the percentage of predicted Vo2 max with the use of
weight and sometimes height in addition to age and sex. Whereas
a simple prediction equation will accurately assess the aerobic
fitness level and exercise performance potential of the patient, an
equation that adjusts for the increased oxygen demand of excess
weight may more accurately reflect cardiac output.
Different terminology may be used to describe . oxygen uptake
during a maximal exercise test. A theoretical Vo2 max is achieved
during high-intensity exercise . when a further increase in . work-
load fails to elicit higher Vo2. Achievement of a true Vo2 max
generally requires not only a high level of effort but also practice
on the specific ergometer being used, and not all patients can
perform such hard exercise,
. particularly on their first attempt.
Therefore,
. the term V O 2 peak is often used to describe the peak
Vo2 that is measured during a symptom-limited exercise test.
Advantages of CPX Over Conventional
Exercise Electrocardiography
The advantages of CPX over conventional exercise testing
include the following:
1. An objective measure
. .of fatigue is provided. This is the RER, which
is. calculated as Vco2 /Vo2.
2. Vo2 max provides a better measure of functional capacity than tread-
mill time alone. Patients with heart failure
. and poor left ventricu-
lar contractility often have reduced Vo2 compared . with workload,
and various gait abnormalities may increase Vo2 compared with
workload.
. workload. Excessive use of handrails
3. The reasons for a low Vo2 max can be generally identified: poor car-
diac reserve, ventilation limitation or other pulmonary abnormali-
ties, claudication, musculoskeletal problems, poor effort, or severe
deconditioning.
4. Pulmonary abnormalities during exercise can be detected. This is
particularly important in assessing patients with dyspnea on exertion
or in evaluating patients with both cardiovascular and pulmonary
disease.
187
5. In addition .to Vo2 max, CPX testing provides a second prognostic
index, Ve/Vco2, which is not available—or predictable—during
nonmetabolic exercise testing.
During conventional exercise testing, the degree of effort or
fatigue is estimated from the following:
1. Subjective reports by the patient, with use of ratings of perceived
exertion (the standard Borg scale of 6–20 is most commonly used,
but other versions of the Borg scale and other perceived exertion
scales can be used)
2. The percentage of predicted maximal heart rate to which the patient
exercises
3. The opinion of the personnel monitoring the test
Ratings of perceived exertion do not always accurately reflect
the degree of cardiopulmonary effort or fatigue because they
may be influenced by factors such as claudication or musculo-
skeletal pain from another cause, pulmonary abnormalities, or
fearfulness or anxiety during the test. Maximal heart rate has
a standard deviation of approximately 12 beats per minute and
is affected by drugs commonly used to treat cardiovascular
disease (most notably β-blockers) and by physiologic effects of
various disease states (eg, autonomic impairment in diabetes).
During CPX,. the
. monitored objective measure of fatigue is RER
(RER = Vco2 /Vo2). An RER greater than 1.0 indicates the pres-
ence of some degree of anaerobic metabolism. Empirically, it
has been determined that
. exercising to an RER greater than 1.15
is necessary
. to attain V o2 max, at least in healthy persons. The
excess Vco2 to make RER greater than 1.0 during anaerobic exer-
cise is given by the following equation:
HLac + NaHCO3 → NaLac + H2CO3,
where HLac indicates lactic acid; NaHCO3, sodium bicarbonate;
NaLac, sodium lactate; and H2CO3, carbonic acid.
Lactic acid released by muscle cells is buffered by sodium
bicarbonate to form sodium lactate and carbonic acid. In turn,
carbonic acid dissociates to water and carbon dioxide. The reac-
tions proceed to the right as carbon dioxide is blown off in the
lungs, resulting in a neutral pH despite addition of the hydrogen
ions from lactic acid, as follows:
H2CO3 → H2O + CO2,
where H2CO3 indicates carbonic acid; H2O, water; and CO2, car-
bon dioxide.
Without anaerobic metabolism, the highest value for RER
would be 1.0, which represents the metabolism of glucose during
which 6 moles of carbon dioxide are produced when 1 mole of
glucose combines with 6 moles of oxygen.
In the United States, most CPX is done on a motor-driven
treadmill. In young, healthy patients, there. is a high correlation
between treadmill performance time and Vo2 max, but the cor-
relation is much weaker for older patients, especially those with
cardiac disease. Patients with left ventricular dysfunction have
impaired cardiac output and thus rely more on anaerobic energy
.
sources for treadmill work, so their Vo2 is lower at any given
.
. for support also lowers Vo2
at a given workload. Thus, Vo2 max—and cardiac reserve—will
be overestimated from the performance time on the test. In con-
trast, gait abnormalities due to degenerative joint disease of the
lower extremities, obesity, or lack of experience
. with walking on
a motor-driven treadmill will increase Vo2 at a given workload
and cause an underestimation of cardiac reserve for a given per-
formance time. So, although the estimated metabolic equivalent
188 II Noninvasive Imaging
.
task level from a treadmill is a reasonable surrogate for Vo2 max,
it is often an imperfect surrogate, especially in various cardiac
disease states. Cycle ergometer
. exercise tests yield a more consis-
tent relationship between Vo2 and performance time, but patients
generally do not achieve equally high cardiac output levels as
on treadmill tests (generally 80%–90%), and most of the norms
for functional capacity with which US cardiologists are familiar
are based on treadmill time. Sometimes nonstandard ergometers
(eg, supine
. bicycle or arm cycle) are used for exercise testing, and
peak Vo2 from these tests is generally even lower (about 65% of
what can be achieved on a treadmill test).
Uses of CPX
1. Accurate assessment of aerobic capacity
2. Establishment of prognosis in various forms of cardiac disease
3. Verification of symptomatic status (functional class)
4. Determination of appropriateness and timing of surgery for patients
who have cardiac disease
5. Exercise prescription
6. Evaluation of benefits of various forms of intervention, includ-
ing surgery, device implantation, medical therapy, and exercise
rehabilitation
7. Establishment of differential diagnosis of dyspnea on exertion
Interpretation of CPX
. 3. Diseases that reintroduce arterialized blood with low levels of car-
A low peak exercise Vo2 may be a sign of poor cardiac reserve
if the RER reaches an adequate level, but obese and .severely
deconditioned patients may have low peak exercise Vo2 with
normal cardiac function. Therefore, several other measures of
cardiac performance are examined to determine whether the
limitation is truly from. poor cardiac reserve. The first of these
is an early plateau in Vo. 2 as plotted against time during graded
exercise. (A plateau in Vo2 might be the result of an athletically
. failure and a potential means of identifying the presence and esti-
trained patient exercising
. hard enough to achieve a true Vo2 max,
but in that case the Vo2 is not likely low and there will be confir-
matory signs such as a plateau in heart rate . and a low breathing
reserve.) Extending the measurement . of Vo2 during active recov-
ery increases the sensitivity of the. Vo2 versus time curve because
a plateau or further increase in Vo2 during active recovery has
also been associated with poor cardiac reserve. Examples of nor-
mal and abnormal curves are given below.
In addition,
. the curve of oxygen pulse plotted. against exercise
time or Vo2 can be examined. Oxygen pulse (Vo2 /heart rate) is
generally expressed as milliliters per minute without adjustment
for weight and is mathematically equivalent to the product of
stroke. volume and arteriovenous oxygen difference. In the way
that Vo2 is the noninvasive analogue for cardiac output, oxygen
pulse is the noninvasive analogue for stroke volume. . It is often
even more sensitive to cardiac impairment than Vo2 itself. A low
oxygen pulse alone may be an indication of low stroke volume,
but it is clearer if there is also an early plateau or even a drop in
oxygen pulse with increasing workload. This would generally be
considered an indication of decreasing stroke volume. Whereas
stroke volume increases during early exercise and remains rela-
tively constant during higher levels of exercise, an actual decrease
in stroke volume with increasing exercise generally reflects some
sort of cardiac abnormality, such as valvular lesions (both ste-
nosis and insufficiency), poor contractility, left-to-right shunts,
inflow or outflow tract obstruction, restrictive and constrictive
cardiomyopathies, various forms of congenital heart disease,
and lack of atrial contractility and atrioventricular synchrony
in atrial fibrillation. Oxygen pulse should be decreased during
tachyarrhythmias and increased during bradyarrhythmias or
with β-blockade.
Other important factors that may indicate inadequate cardiac
output during exercise include the following:
1. Ventilatory equivalent for carbon dioxide—that is, the .ratio. of
expired volume to carbon dioxide production. expressed as Ve/Vo2
2. Breathing reserve—expressed as BR = 1 − Ve /MVV, where MVV
indicates maximal voluntary ventilation measured by spirometry or
estimated from age, sex, and height
. .
Ve/Vco2 can be defined in various ways. Initially, the value at
peak exercise was used, but that could be artificially elevated
from volitional hyperventilation. Now the simplest technique
used is to take the lowest value during exercise (ie, the nadir).. A
more involved method
. is to calculate the slope of the curve of Ve2
plotted against Vco2, although the nadir and the slope are gen-
erally identical or very close mathematically. For young adults
without heart
. . or lung disease, the reference range for the nadir or
slope of Ve/Vco2 is 25 to 29. The ratio normally increases to the
mid 30s with age because aging lung membranes slightly reduce
the
. diffusion
. of carbon dioxide from the blood to the lungs. The
Ve/Vco2 ratio also increases with restrictive lung disease and
with various forms of cardiac disease:
1. Diseases that interfere with movement of blood through the lungs
and cause increased intrapulmonary pressures
2. Diseases that result in a ventilation to perfusion mismatch
bon dioxide into the lung
. .
The Ve/Vco2 ratio often increases with severe heart failure, mitral
stenosis or severe mitral regurgitation, restrictive and constrictive
cardiomyopathies, pulmonary vein stenosis, pulmonary hyper-
tension, pulmonary
. . embolus, and various forms of left-to-right
shunting. Ve/Vco2 is an independent predictor of survival in heart
mating the severity of other lesions described above.
Both heart disease and lung disease affect BR. A normal BR
appears to be 15% to 40%, depending on physical activity level.
Preexercise spirometry should be performed before CPX. The
maximal voluntary ventilation can be measured (if time per-
mits) or estimated as 40 times FEV1. If cardiac output decreases,
the stimulus on ventilation is less and BR increases. However,
coexisting lung disease may cause pseudonormalization . of the
BR by decreasing both the numerator (peak exercise Ve) and the
denominator (maximal breathing capacity) of the ratio. Chronic
heart failure frequently produces pseudorestrictive lung disease;
obesity may likewise proportionally reduce both forced vital
capacity and FEV1. Therefore, use of a large BR alone to indi-
cate reduced cardiac output is not recommended. In contrast,
highly trained endurance athletes have high cardiac outputs
that reduce the BR and may result in mild desaturation during
heavy exercise. A patient can be described as “ventilation lim-
ited” when the BR is absent or low (<10%), RER is low, or other
signs of maximal cardiac performance are absent. Other pulmo-
nary variables may be useful in verifying a pulmonary limita-
tion, including desaturation, high. .respiratory rate with limited
increase in tidal volume, high Ve/Vco2, and pressure of low end
tidal carbon dioxide.
Another indication of poor cardiac output that is occasionally
seen with severe heart failure is a periodic or sinusoidal breathing
pattern. This is likely an extension of Cheyne-Stokes respiration
with large, cyclical oscillations in tidal volume and respiratory
rate. An example of periodic breathing during exercise is shown
below.
 15 Cardiopulmonary Exercise Testing in Clinical Medicine 189

Thus, with the additional information gained from CPX, the

severity of cardiac output limitation in various cardiac disorders Box 15.1. Case 1
can be defined more fully than by looking only at treadmill per- Man aged 50 years with treated hyperlipidemia and hypertension
formance along with heart rate and blood pressure responses. and no known heart disease
Termination of the test at a low RER may indicate poor effort,
Electrocardiogram: sinus rhythm, HR = 71 bpm
musculoskeletal limitations, or ventilation limitation. Comparison
of maximal voluntary ventilation BMI = 23.9
. (as measured during standard .
spirometry) with the maximal Ve during exercise should provide Peak Vo2 = 41.1 mL/kg/min (118% of predicted value)
an indication of whether exercise was ventilation limited. The Exercise time = 13.3 min (FAC = 117%)
distinction between poor effort and musculoskeletal limitation is Peak HR = 171 bpm
based on the specific complaints identified by the patient during Peak RER = 1.11
the test and on the findings from the physical examination and .
Peak Ve = 91.4 L/min
medical history. CPX interpretation is summarized in Table 15.2.
Breathing reserve = 39%
. .
Ve / Vco2 nadir = 25
Sample Cardiopulmonary Tests
Oxygen saturation at peak exercise = 96% by forehead oximetry
Examples of actual CPX from clinical practice are given below.
These illustrate important points in the use of CPX discussed in Abbreviations: BMI, body mass index (calculated as weight in
the previous section. kilograms divided by height in meters squared); bpm, beats per
minute; FAC, functional aerobic
. capacity; HR, heart rate; RER,
.
respiratory exchange ratio; Vco . 2
, carbon dioxide production; Ve,
Case 1 expired volume per unit time; Vo2, oxygen consumption.
Patient data .are presented
. in Box 15.1. In Figure 15.4, the patient’s
heart rate, Vo2, and Vco2 are plotted against exercise time. The
heart rate increases continuously during exercise and shows a The patient is slightly tachypneic during the first minute of exer-
sharp recovery during the first .minute of an active recovery at 1.7 cise, but then the respiratory rate returns to roughly 18 breaths per
mph and 0% grade. Note that Vo2 increases normally . throughout minute. It is typical.that increased tidal volume accounts for most
exercise and decreases .rapidly in active recovery. Vco2 increases of the increase in Ve during the early stages of exercise, with a
to a higher level than Vo2 because of the contribution of anaer- delayed increase in the respiratory rate until later stages. The BR
obic metabolism, indicating a near-maximal effort with a peak (39%) is near the upper limit of normal, probably because of the
RER of 1.11. . near-maximal nature of the test, with a peak RER of 1.11.
In Figure 15.5, oxygen pulse (Vo2 /heart rate) is plotted against
time. Oxygen pulse increases rapidly during early exercise .but
increases more slowly in the remainder of the test. As with Vo2,
the peak value for oxygen pulse slightly exceeds the value pre-
dicted for age and sex. Figures 15.4 and 15.5 together give a
strong indication of normal cardiac function with normal cardiac
output and stroke volume.
Figure
. 15.6 shows the basic mechanics of ventilation for this
case. Ve increases continuously in a slightly curvilinear fashion.

Table 15.2. Interpretation of Cardiopulmonary Exercise Test

Results According to Selected Exercise Parameters
Cardiac Ventilation Poor Poor
Parameter Normal Limited Limited Fitness Effort
.
Vo2 max Normal Low Low Low Low
vs time Rise Flat Rise Rise Rise
RER >1.15 >1.15 <1.15 >1.15 <1.15
BR 15%–40% >40% <10% 25%–60% a >40%
. .
Ve/Vco2 <35b Normal High Normal Normal
or highc
O2 pulse Normal Low Low Low Low
vs time Rise Flat Rise Rise Rise
O2 sats >90% Normal or Generally >90% >90%
reducedc <90% d
Figure
.
15.4. Case 1: Heart Rate (HR), .
Oxygen Consumption
Abbreviations: BR, breathing reserve; O2 pulse, ratio of oxygen consumption (Vo2), and Carbon Dioxide Production (Vco2) Plotted Against Test
to heart. rate;
. O2 sats, arterial oxygen saturation; RER, respiratory exchange
. Time. Exercise started at running time 1 minute after a preexercise gas
ratio; Ve/Vco2, ratio of expired volume to carbon dioxide production; Vo2 max,
maximum oxygen consumption.
exchange measurement. An additional minute of gas exchange meas-
a
More severe deconditioning is associated with higher BR. urement was provided during active recovery (Rec) at 1.7 mph and 0%
b
Values up to 40 may be normal for older patients. grade after achievement of peak exercise at running time 14.3 minutes.
c
Varies according to type of cardiac disease. AT indicates anaerobic threshold estimated by the V-slope method of
d
Varies according to type of pulmonary disease. Wasserman; bpm, beats per minute.
190 II Noninvasive Imaging
Figure 15.5. . Case 1: Oxygen Pulse (Ratio of Oxygen Consumption
to Heart Rate [Vo2 /HR]) Plotted Against Test Time. Exercise started at
running time 1 minute after a preexercise gas exchange measurement.
An additional minute of gas exchange measurement was provided dur-
ing active recovery (Rec) at 1.7 mph and 0% grade after achievement
of peak exercise at running time 14.3 minutes. AT indicates anaerobic
threshold estimated by the V-slope method of Wasserman.
. Dioxide (Petco2) Plotted Against Test Time. Exercise started at run-
Figure 15.6. Case 1: Expired Volume (Ve), Respiratory Rate (RR),
and Tidal Volume (Vt) Plotted Against Test Time. Exercise started at
running time 1 minute after a preexercise gas exchange measurement.
An additional minute of gas exchange measurement was provided dur-
ing active recovery (Rec) at 1.7 mph and 0% grade after achievement
of peak exercise at running time 14.3 minutes. AT indicates anaerobic
threshold estimated by the V-slope method of Wasserman; BTPS, body
temperature, pressure, saturated with water.
Figure 15.7 shows additional analyses of gas exchange plotted
against test time. Figure 15.8 shows determination of the.anaer-
obic threshold by. the V-slope method of Wasserman. Vco2 is
plotted against Vo2 on equal interval axes.
. The anaerobic thresh-
old is identified as the point at which. Vco2 begins to increase
at an accelerated rate compared with Vo2. Table 15.3 shows the
completed interpretation matrix indicating normal CPX results.
Except for the near-maximal RER of 1.11, all the exercise param-
eters are consistent with a normal study. None of the other inter-
pretation classifications are met as well by the test data.
Case 2
Patient data .are presented
. in Box 15.2. In Figure 15.9, the patient’s
heart rate, Vo2, and Vco2 are plotted against exercise time. The
heart rate increases continuously during exercise and shows a
normal recovery during the first minute
. of an active recovery at
1.7 mph and 0% grade. Note that Vo2 plateaus early during exer- .
cise and does not decrease normally in active recovery. Peak Vo2
(indicated by the red vertical marker)
. occurs more than 1 minute
before the end of the test. Peak Vo2 is well
. below the predicted
value of approximately
. 2,300 mL/min. V co2
increases to a higher
level than Vo2 because of the contribution of anaerobic metabo-
lism, indicating a maximal effort with a peak RER of 1.21.
Low oxygen pulse with a distinct plateau. is. seen in
Figure 15.10.
. .Figure 15.11 .illustrates
. the elevated Ve/Vco2 ratio.
Both the Ve/Vco2 and the Ve/Vo2 curves are suggestive of a mild
periodic breathing pattern. Overall analysis of the CPX test for
the case is presented in Table 15.4. All the CPX results fit the
“Cardiac Limit” classification.
Figure
. . 15.7. Case 1: Ventilatory Equivalent for Carbon Dioxide
(Ve/Vco2), Dead Space to. Tidal
. Volume Ratio (Vd/Vt), Ventilatory
Equivalent for Oxygen (Ve/Vo2), and Pressure of End-Tidal Carbon
ning time 1 minute after a preexercise gas exchange measurement. An
additional minute of gas exchange measurement was provided dur-
ing active recovery (Rec) at 1.7 mph and 0% grade after achievement
of peak exercise at running time 14.3 . minutes. Petco2 was measured
as millimeters of mercury . ; V co2, milliliters .per kilogram per
minute; Vd, milliliters ; Ve, liters per minute; Vo2, milliliters per
kilogram per minute; and Vt, liters. AT indicates anaerobic threshold
estimated by the V-slope method of Wasserman; est, estimated.
 15 Cardiopulmonary Exercise Testing in Clinical Medicine 191

Box 15.2. Case 2

Man aged 59 years with chronic heart failure due to nonischemic
cardiomyopathy
Echocardiogram: left ventricular ejection fraction = 15%
Electrocardiogram: sinus rhythm, rate = 67 bpm
BMI = 22.6
.
Peak VO2 = 17.1 mL/kg/min (57% of predicted value)
Exercise time = 7.0 min (FAC = 67%)
Peak HR = 115 bpm
Peak RER = 1.21
.
Peak VE = 50.9 L/min
Breathing reserve = 65%
. .
Ve / Vco2 nadir = 38
Oxygen saturation at peak exercise = 99% by forehead oximetry

Abbreviations: BMI, body mass index (calculated as weight in

kilograms divided by height in meters squared); bpm, beats per
minute; FAC, functional aerobic
. capacity; HR, heart rate; RER,
.
respiratory exchange ratio; Vco . 2, carbon dioxide production; Ve,
expired volume per unit time; Vo2, oxygen consumption.

Figure 15.8. Case 1: Anaerobic Threshold .(AT) Determination

Estimated by the V-Slope
. Method of Wasserman. Vco2 indicates carbon
dioxide production; Vo2 , oxygen consumption. . Figures 15.13 and 15.14 show further abnormalities, with low
Vo2 and oxygen pulse curves that plateau before peak exercise and
do not decrease normally in active recovery. Contrast the curves
Case 3 in those figures with the curves in Figures 15.4 and 15.5 from
the
. normal study. There are likely multiple reasons for the low
Patient data are presented in Box 15.3. The prominent features of Vo2 and oxygen pulse, including effects of recent cardiac surgery
pulmonary function during exercise seen in Figure 15.12 include
a very limited tidal volume (<0.85 L), corresponding
. to the
low FEV1 seen on preexercise spirometry. Ve is low and shows
a distinct plateau .during the final minute of exercise. Contrast
the shape of. the Ve curve in Figure 15.12 with the curvilinear
increase in Ve in Figure 15.6, the normal study.

Table 15.3. Case 1: Interpretation of the Cardiopulmonary

Exercise Test Results According to Selected Exercise
Parametersa
Cardiac Ventilation Poor Poor
Parameter Normal Limited Limited Fitness Effort
.
Vo2 max Normal Low Low Low Low
vs time Rise Flat Rise Rise Rise
RER >1.15 >1.15 <1.15 >1.15 <1.15
BR 15%–40% >40% <10% 25%–60%b >40%
. .
Ve/Vco2 <35c Normal High Normal Normal
or highd
O2 pulse Normal Low Low Low Low
vs time Rise Flat Rise Rise Rise
O2 sats >90% Normal or Generally >90% >90%
reducedd <90% e
Abbreviations: BR, breathing reserve; O2 pulse, ratio of oxygen consumption
to heart. rate;
. O2 sats, arterial oxygen saturation; RER, respiratory exchange
. Figure 15.9. Case 2: Heart Rate (HR), Oxygen Consumption
ratio; Ve/Vco2, ratio of expired volume to carbon dioxide production; Vo2max, . .
(Vo2), and Carbon Dioxide Production (Vco2) Plotted Against Test
maximum oxygen consumption.
a
Time. Exercise started at running time 1 minute after a preexercise gas
Results in boldface type indicate results that support the “Normal”
interpretation classification (highlighted in light gray).
exchange measurement. An additional minute of gas exchange meas-
b
More severe deconditioning is associated with higher BR. urement was provided during active recovery (Rec) at 1.7 mph and 0%
c
Values up to 40 may be normal for older patients. grade after achievement of peak exercise at running time 14.3 minutes.
d
Varies according to type of cardiac disease. AT indicates anaerobic threshold estimated by the V-slope method of
e
Varies according to type of pulmonary disease. Wasserman; bpm, beats per minute.
192 II Noninvasive Imaging

Table 15.4. Case 2: Interpretation of the Cardiopulmonary

Exercise Test Results According to Selected Exercise
Parametersa
Cardiac Ventilation Poor Poor
Parameter Normal Limited Limited Fitness Effort
.
Vo2 max Normal Low Low Low Low
vs time Rise Flat Rise Rise Rise
RER >1.15 >1.15 <1.15 >1.15 <1.15
BR 15%–40% >40% <10% 25%–60% b >40%
. .
Ve/Vco2 <35c Normal High Normal Normal
or highd
O2 pulse Normal Low Low Low Low
vs time Rise Flat Rise Rise Rise
O2 sats >90% Normal or Generally >90% >90%
reducedd <90% e
Abbreviations: BR, breathing reserve; O2 pulse, ratio of oxygen consumption
to heart. rate;
. O2 sats, arterial oxygen saturation; RER, respiratory exchange
.
ratio; Ve/Vco2, ratio of expired volume to carbon dioxide production; Vo2 max,
maximum oxygen consumption.
a
Results in boldface type indicate results that support the “Cardiac Limited”
interpretation classification (highlighted in light gray).
b
More severe deconditioning is associated with higher BR.
c
Values up to 40 may be normal for older patients.
Figure 15.10.. Case 2: Oxygen Pulse (Ratio of Oxygen Consumption d
Varies according to type of cardiac disease.
to Heart Rate [Vco2 /HR]) Plotted Against Test Time. Exercise started at e
Varies according to type of pulmonary disease.
running time 1 minute after a preexercise gas exchange measurement.
An additional minute of gas exchange measurement was provided dur-
ing active recovery (Rec) at 1.7 mph and 0% grade after achievement
of peak exercise at running time 14.3 minutes. AT indicates anaerobic both on cardiac function and on overall level of physical activ-
threshold estimated by the V-slope method of Wasserman. ity, residual impairment of cardiac function (mildly reduced left
and right ventricular function, mitral and aortic valve abnormali-
ties, possible diastolic dysfunction), and ventilatory constraints
on oxygen uptake. Table 15.5 indicates that the test is primar-
ily ventilation limited but with features consistent with impaired
cardiac output and deconditioning due to recent surgery.
.
Prognosis Based on VO2 max
Perhaps the most widely recognized use of CPX in clinical cardi-
ology is for the evaluation of patients .with chronic heart failure.
Several studies have established that Vo2 max is a strong predic-
tor of survival in. this patient group (Figure 15.15).
As a result, Vo2 max has been used in the risk stratification of
heart failure patients who need a cardiac transplant, and serial
CPX is routinely performed to adjust risk level and to time the
transplant as heart failure progresses (or stabilizes). Although
criteria
. for cardiac transplant have broadened in recent years,
Vo2 max remains 1 of several critical factors used by insurance
companies in determining the appropriateness (and hence reim-
bursement)
. of cardiac transplant.
Vo2 max has also been reported to predict long-term survival
for cardiac rehabilitation patients (Figure 15.16). As a group, they
are considerably healthier than heart failure patients and have a
much broader spectrum of cardiovascular disease.
Figure
. . 15.11. Case 2: Ventilatory Equivalent for Carbon Dioxide Other types of cardiac patients who might be referred for
(Ve / Vco2), Dead Space to. Tidal
. Volume Ratio (Vd/Vt), Ventilatory CPX include those who have hypertrophic cardiomyopathy,
Equivalent for Oxygen (Ve / Vo2), and Pressure of End-Tidal Carbon valvular heart disease, congenital heart disease, or pulmonary
Dioxide (Petco2) Plotted Against Test Time. Exercise started at run- hypertension. CPX not only helps to determine the cardiac
ning time 1 minute after a preexercise gas exchange measurement. An reserve and establish prognosis for these patients, but it also
additional minute of gas exchange measurement was provided dur-
helps in determining the effects of various therapies (eg, drugs,
ing active recovery (Rec) at 1.7 mph and 0% grade after achievement
of peak exercise at running time 14.3. minutes. Petco2 was measured pacing, and rehabilitation) and allows for periodic evaluation for
as millimeters of mercury ; Vo2 , milliliters per. kilogram per the timing of surgical interventions.
.
minute; Vd, milliliters ; Ve, liters per minute; Vco2 , milliliters Another important role of CPX is to verify symptoms and
per kilogram per minute; and Vt, liters. AT indicates anaerobic thresh- functional class as determined in the office. Some patients
old estimated by the V-slope method of Wasserman; est, estimated. may reduce their physical activity to avoid symptoms and thus
 15 Cardiopulmonary Exercise Testing in Clinical Medicine 193

Box 15.3. Case 3

Man aged 55 years with history of paroxysmal atrial fibrillation,
history of mitral regurgitation with recent mitral valve
replacement, and chronic obstructive pulmonary disease
Pulmonary function test: FVC = 2.13 L (68% of predicted value)
FEV1 = 0.64 L (25% of predicted value)
Dlco adjusted for hemoglobin = 13.1 mL ⋅ min −1 ⋅ mm Hg−1 (55%
of predicted value)
Electrocardiogram: sinus rhythm, rate = 74 bpm
Echocardiogram: left ventricular ejection fraction = 53%,
mild mitral regurgitation, thickened aortic valve with mild
regurgitation, mildly decreased right ventricular systolic
function, and moderate left atrial enlargement
BMI = 20.1
.
Peak Vo2 = 18.0 mL/kg/min (56% of predicted value)
Exercise time = 7.8 min (FAC = 72%)
Peak HR = 121 bpm
Peak RER = 1.04
.
Peak Ve = 25.9 L/min
Breathing reserve = −4%
. . Figure 15.13. Case 3: Heart Rate (HR), Oxygen Consumption
Ve / Vco2 nadir = 26 . .
(Vo2), and Carbon Dioxide Production (Vco2) Plotted Against Test
Oxygen saturation at peak exercise = 77% by forehead oximetry Time. Exercise started at running time 1 minute after a preexercise gas
exchange measurement. An additional minute of gas exchange meas-
Abbreviations: BMI, body mass index (calculated as weight in urement was provided during active recovery (Rec) at 1.7 mph and 0%
kilograms divided by height in meters squared); bpm, beats per grade after achievement of peak exercise at running time 14.3 minutes.
minute; Dlco, diffusing capacity of lung for carbon monoxide; FAC, AT indicates anaerobic threshold estimated by the V-slope method of
functional aerobic capacity; FEV1, forced expiratory volume in the Wasserman; bpm, beats per minute.
first second of expiration; FVC, forced vital capacity; HR, heart rate;
. .
RER, respiratory exchange ratio; V o , carbon dioxide production; Ve,
. 2
expired volume per unit time; Vo2, oxygen consumption.

.
Figure 15.12. Case 3. Expired Volume (VE), Respiratory Rate (RR),
and Tidal Volume (VT) Plotted Against Test Time. Exercise started at
running time 1 minute after a preexercise gas exchange measurement.
An additional minute of gas exchange measurement was provided dur-
ing active recovery (Rec) at 1.7 mph and 0% grade after achievement
of peak exercise at running time 14.3 minutes. AT indicates anaerobic
threshold estimated by the V-slope method of Wasserman; BTPS, body
temperature, pressure, saturated with water.
Figure 15.14.. Case 3: Oxygen Pulse (Ratio of Oxygen Consumption
to Heart Rate [Vo2 /HR]) Plotted Against Test Time. Exercise started at
running time 1 minute after a preexercise gas exchange measurement.
An additional minute of gas exchange measurement was provided dur-
ing active recovery (Rec) at 1.7 mph and 0% grade after achievement
of peak exercise at running time 14.3 minutes. AT indicates anaerobic
threshold estimated by the V-slope method of Wasserman.
194 II Noninvasive Imaging

Table 15.5. Case 3: Interpretation of the Cardiopulmonary 60 Low peak VO2

Exercise Test Results According to Selected Exercise Preserved peak VO2, HRREC <13
Parametersa 50 Preserved peak VO2, HRREC ≥13

Cardiac Ventilation Poor Poor 40

Mortality, %
Parameter Normal Limited Limited Fitness Effort
.
Vo2 max Normal Low Low Low Low 30
vs time Rise Flat Rise Rise Rise
RER >1.15 >1.15 <1.15 >1.15 <1.15 20
BR 15%–40% >40% <10% 25%–60% b >40%
. .
Ve/Vco2 <35c Normal High Normal Normal 10
or highd
O2 pulse Normal Low Low Low Low 0
vs time Rise Flat Rise Rise Rise
0 1,000 2,000 3,000 4,000 5,000
O2 sats >90% Normal or Generally >90% >90%
reducedd <90%e Days After Exercise Test
Abbreviations: BR, breathing reserve; O2 pulse, ratio of oxygen consumption Figure 15.16. Mortality After Cardiopulmonary Exercise Testing
to heart rate; O2 sats, arterial oxygen saturation; RER, respiratory exchange Among Cardiac Rehabilitation Patients With Stable Coronary Artery
. . .
ratio; Ve/Vco2 , ratio of expired volume to carbon dioxide production; Vo2 max, Disease. Patients were grouped according to whether maximum oxy-
.
. was low (<15 mL/kg/min for women or <19
maximum oxygen consumption. gen consumption (Vo2max)
a
Results in boldface type indicate results that support the interpretation of mL/kg/min for men), Vo2 max was preserved and heart rate recovery
primarily ventilation limited with features consistent with impaired cardiac
(HRREC) was normal (decrease
. of ≥13 beats per minute in first minute
output and deconditioning due to recent surgery (highlighted in light gray).
b
More severe deconditioning is associated with higher BR. of active recovery), or Vo2 max was preserved and HRREC was abnor-
c
Values up to 40 may be normal for older patients. mal (decrease of <13 beats per minute in first minute of active recovery).
d
Varies according to type of cardiac disease. (Previously published. See “Credit Lines” section.)
e
Varies according to type of pulmonary disease.

provides evidence of impaired cardiac output. (The small spike

in oxygen pulse in active recovery is an artifact of a drop of a run
describe a “false-negative” symptomatic status to the physi-
of frequent premature ventricular contractions, which slowed. . the
cian; with CPX, they may quickly become symptomatic. Other
heart rate momentarily.) Figure 15.18 shows the elevated Ve/Vco2
patients may exaggerate their symptoms and describe themselves
ratio and low pressure of end-tidal carbon dioxide, which suggest
as being more limited than objective testing shows.
impaired gas exchange in the lungs. Computed tomography of
CPX for Dyspnea on Exertion
Dyspnea on exertion is a common indication for CPX to sort out
Box 15.4. Case 4
possible causes of dyspnea, including cardiac output impairment,
deconditioning, pulmonary emboli, and various forms of pulmo- Woman aged 44 years presents with a 2-week history of dyspnea on
nary disease. Skeletal abnormalities such as severe scoliosis or exertion. She is active and had been exercising at a gym regularly
pectus excavatum may cause a ventilation limitation in exercise. for many years. Now she reports dyspnea with 1 flight of stairs.
Physical examination findings: unremarkable
Case 4 Hemoglobin: normal
Case 4 is an example of pulmonary emboli suggested by CPX Baseline spirometry: unremarkable with FVC at 98% of
(Box 15.4). In Figure 15.17, the low oxygen pulse with plateau predicted value and FEV1 at 107% of predicted value
Resting electrocardiogram: normal with slightly increased heart
rate of 93 bpm
>18 mL/kg/min BMI = 26.8
100 .
>14 through 18 mL/kg/min
Peak Vo2 = 18.6 mL/kg/min (62% of predicted value)
Exercise time = 7.5 min (FAC = 73%)
75 Peak HR = 176 bpm
Survival, %

>10 through 14 mL/kg/min

Peak RER = 1.15
50 .
Peak Ve = 56.8 L/min
Breathing reserve = 45%
25 ≤10 mL/kg/min . .
Ve / Vco2 nadir = 41
Oxygen saturation at peak exercise = 95% by forehead oximetry
0
0 3 6 9 12 15 Abbreviations: BMI, body mass index (calculated as weight in
Months kilograms divided by height in meters squared); bpm, beats per minute;
FAC, functional aerobic capacity; FEV1, forced expiratory volume in
Figure 15.15. Survival for Patients in Heart Failure With Reduced
.
the first second of expiration; FVC, forced
. vital capacity; HR, heart
Systolic rate; RER, respiratory exchange ratio; Vco2, carbon dioxide production;
. Function, According to Peak Oxygen . Consumption (Vo2). . .
(For Vo2 ≤10 mL/kg/min compared with Vo2 >14 mL/kg/min, P<.05.) Ve, expired volume per unit time; Vo2, oxygen consumption.
(Previously published. See “Credit Lines” section.)
 15 Cardiopulmonary Exercise Testing in Clinical Medicine 195

Table 15.6. Examples

. of Various Types of Patients and
Typical Values for Vo2 max
.
Patient Vo2 max, mL/kg/min

30-year-old lean, active man 52

30-year-old female competitive long distance 60
runner
40-year-old overweight, deconditioned man 35
40-year-old woman with NYHA class II heart 18
failure
60-year-old man who has asymptomatic coronary 25
artery disease and takes a β-blocker
60-year-old lean, active woman 28
.
Abbreviation: Vo2 max, maximum oxygen consumption.

.
Ranges for VO2 max
.
By the prediction equations given above, the average Vo2 max for
Figure 15.17. . Case 4: Oxygen Pulse (Ratio of Oxygen Consumption a 50-year-old healthy but untrained man would be 35 mL/kg/min
to Heart Rate [Vo2 /HR]) Plotted Against Test Time. Exercise started at and for a 50-year-old similarly healthy .but untrained woman,
running time 1 minute after a preexercise gas exchange measurement. 28 mL/kg/min. However, the range for Vo2 max is quite broad,
An additional minute of gas exchange measurement was provided dur- with values greater than 80 mL/kg/min for champion endurance
ing active recovery (Rec) at 1.7 mph and 0% grade after achievement athletes and values less than 10 mL/kg/min for patients with
of peak exercise at running time 14.3 minutes. AT indicates anaerobic severe heart failure. As previously noted, many factors influence
.
threshold estimated by the V-slope method of Wasserman. Vo2 max, including age, sex, body mass index, level of physical
activity, left ventricular function, and ability to
. achieve a normal
maximal heart rate. Examples of values for Vo2 max that might
the chest showed multiple pulmonary emboli in the lungs. On be expected for various types of patients are given in Table 15.6.
further questioning, the patient described an episode of calf ten-
derness about 4 weeks before the onset of dyspnea on exertion.
Ultrasonography of the left lower extremity showed deep vein Table 15.7. Uses of Cardiopulmonary Exercise Testing in
thrombosis beginning at the level of the lower popliteal vein and Various Situations
extending into the posterior tibial veins.
Use of Cardiopulmonary
Situation Exercise Testing
Heart failure Establish prognosis
Evaluate before transplant
Evaluate therapy (eg, biventricular
pacing)
Valvular heart disease Determine functional status
Determine severity of impairment
Establish prognosis
Hypertrophic cardiomyopathy Determine severity of impairment
Evaluate therapy
Collect limited data on prognosis
Coronary artery disease Establish prognosis
Prescribe exercise
Cardiopulmonary disease Determine primary cause of dyspnea
Disability evaluation Provide objective assessment of effort
Determine true functional capacity
Congenital heart disease Determine functional capacity
Identify cardiac and noncardiac
limitations
Pulmonary hypertension Determine severity of impairment
Figure
. . 15.18. Case 4: Ventilatory Equivalent for Carbon Dioxide Evaluate therapy
(Ve /Vco2), Dead Space to. Tidal
. Volume Ratio (Vd/Vt), Ventilatory Collect limited data on prognosis
Equivalent for Oxygen (Ve /Vo2), and Pressure of End-Tidal Carbon Chronic fatigue or exercise Provide objective assessment of effort
Dioxide (Petco2) Plotted Against Test Time. Exercise started at run- intolerance of unknown cause Determine true functional capacity
ning time 1 minute after a preexercise gas exchange measurement. An Rule out cardiac limitation
additional minute of gas exchange measurement was provided dur- Orthopedic limitations Accurately assess fitness despite
ing active recovery (Rec) at 1.7 mph and 0% grade after achievement gait abnormality or nonstandard
of peak exercise at running time 14.3. minutes. Petco2 was measured ergometer exercise test
as millimeters of mercury . ; Vo2, milliliters per
. kilogram per Endurance athlete Determine performance potential
minute; Vd, milliliters ; Ve, liters per minute; Vco2 , milliliters Evaluate training
per kilogram per minute; and Vt, liters. AT indicates anaerobic thresh- Prescribe exercise
old estimated by the V-slope method of Wasserman; est, estimated.
196 II Noninvasive Imaging

The American Heart. Association suggests that 18 mL/kg/min scientific statement from the American Heart Association Committee
is a critical value for Vo2 max, below which the degree of dis- on Exercise, Rehabilitation, and Prevention of the Council on Clinical
ability Cardiology and the Council on Cardiovascular Nursing. [cited 2012
. is high and the long-term prognosis is poor. The range Mar 13]. c2007. American Heart Association. Available from: http://
of Vo2 max values from 18 to 22.5 mL/kg/min is considered
circ.ahajournals.org/content/116/3/329.full.pdf.
intermediate risk.
American Thoracic Society/American College of Chest Physicians
statement on cardiopulmonary exercise testing. [cited 2012 Mar
13]. Available from: http://ajrccm.atsjournals.org/cgi/content/full/
Summary 167/2/211.
.
CPX provides more accurate assessments of both Vo2 max and Milani RV, Lavie CJ, Mehra MR. Cardiopulmonary exercise testing:
cardiac reserve than conventional treadmill testing in many how do we differentiate the cause of dyspnea? [cited 2012 Mar 13].
cardiac disease states. Thus, the effect of the disease on func- c2007. American Heart Association. Available from: http://circ.
ahajournals.org/cgi/content/full/110/4/e27.
tional capacity and long-term prognosis is more accurately
established.
Table 15.7 lists uses of CPX. Although conventional exercise Abbreviations
testing is considered useful in many of these situations, CPX has BR breathing reserve
the potential to provide additional information relevant to mak- CPX cardiopulmonary exercise testing
ing the correct diagnosis or establishing an accurate prognosis. FEV1 forced expiratory volume in the first second of expiration
RER
. respiratory exchange ratio
V. CO2 carbon dioxide production
Suggested Reading V. E expired volume
Arena R, Myers J, Williams MA, Gulati M, Kligfield P, Balady GJ, et al. V. O2 oxygen consumption
Assessment of functional capacity in clinical and research settings: a VO2 max maximum oxygen consumption

Stress Test Selectiona
THOMAS G. ALLISON, PHD, MPH
Cardiac stress testing is an integral part of clinical cardiology.
The safe performance of high-quality stress testing with imaging
techniques has led to an increase in the number of stress pro-
cedures in recent years. This has prompted the development of
national guidelines and appropriateness criteria for cardiac stress
testing and imaging techniques. Although the particular tech-
niques of exercise ECG stress testing, stress echocardiography,
and myocardial perfusion imaging are discussed in more detail
in other chapters of this text, this chapter focuses on the relative
advantages and limitations of each technique to ensure that the
appropriate test is performed on the appropriate patient.
This chapter is addressing stress testing from the standpoint
of making a diagnosis or establishing prognosis in patients with
suspected or known CAD. Assessment of patients with complex
cardiovascular disease including chronic heart failure, various
forms of cardiomyopathy, valvular heart disease, congenital
heart disease, pulmonary hypertension, and heart transplant is
best performed by CPX, which is covered in a separate chapter.
CPX is also indicated for the differential diagnosis of dyspnea on
exertion, especially when there is known or suspected pulmonary
disease, and for other purposes such as evaluation of competitive
athletes, sophisticated exercise prescription requiring anaerobic
threshold, disability determination, and assessment of patients
for whom nonstandard ergometry is necessary due to orthopedic
problems. Exercise testing may also be indicated in the evalu-
ation and management of cardiac arrhythmias; imaging is not
generally required in the evaluation of arrhythmias.
a
The author acknowledges the contributions of Stuart D. Christenson,
MD, who authored this chapter in the previous edition.
Abbreviations and acronyms are expanded at the end of this chapter.
16
Stress Testing for the Diagnosis of CAD
Three basic questions need to be addressed when considering
stress testing for the diagnosis of CAD:
1. What is the pretest probability of CAD?
2. Are there conditions precluding a diagnostic exercise ECG stress
test?
3. Can the patient exercise?
Pretest Probability
An estimate of the pretest probability of having CAD can be
determined for each patient on the basis of clinical history. Age,
sex, and the characterization of chest pain are powerful predic-
tors of CAD. Chest pain can be categorized into one of three
groups:
1. Typical or definite angina
a) Substernal chest discomfort with characteristic quality and
duration
b) Provoked by exertion or emotional stress
c) Relieved with rest or nitroglycerin
2. Atypical or probable angina: chest discomfort with two of the above
typical angina characteristics
3. Noncardiac chest pain: chest discomfort with one or none of the
above typical angina characteristics
Additional clinical characteristics that have strengthened
prediction models include diabetes mellitus, smoking, hyper-
cholesterolemia, and resting ECG changes. With these models,
patients can be classified as having a low (<10%), intermediate
(10%-90%), or high probability (>90%) of having CAD. Current
guidelines suggest that stress testing for the diagnosis of CAD
be performed only in patients with an intermediate likelihood
of having CAD. Patients with a low pretest probability of CAD
197
198 II Noninvasive Imaging
are unlikely to benefit from further testing owing to a high rate
of false-positive test results. Similarly, stress testing offers little
additional diagnostic information for patients with high pretest
probability because the posttest probability of CAD remains high
even after a negative test result. In clinical practice, stress testing
is still performed at times in these high-probability patients to
gather prognostic information that can be used to guide therapy
beyond the initial detection of disease.
Standard Stress Test Techniques
Exercise is the stress test modality of choice in most patients
capable of exercise. It is generally safe, widely available, and
the least costly of available diagnostic stress options. However,
exercise ECG testing alone is limited in certain patient groups.
Patients who should be considered for stress testing with myo-
cardial imaging techniques include those with any of the
following:
1. More than 1-mm ST-segment depression on baseline ECG
2. Complete LBBB
3. Ventricular paced rhythms
4. Pre-excitation syndromes
5. Left ventricular hypertrophy
6. Digoxin use
7. Inability to adequately exercise
8. Previous revascularization
Stress echocardiography can be performed with either exer-
cise or pharmacologic stress. Additional information obtained
from echocardiographic images includes left ventricular size
and function, global and regional wall motion and thickening,
and further assessment of ischemic threshold. Myocardial con-
trast perfusion echocardiography may have a valuable role in the
future but is not currently part of standard practice.
Myocardial perfusion imaging typically uses the radionu-
clides thallium Tl 201– or technetium Tc 99m–labeled ses-
tamibi or tetrofosmin, all of which appear to provide similar
diagnostic accuracy in CAD. Imaging with SPECT is now per-
formed more commonly than planar techniques. Testing can
be performed with either exercise or pharmacologic stress.
Additional information obtained from perfusion imaging
includes the severity and extent of perfusion deficits, left ven-
tricular ejection fraction, and limited information on global and
regional wall motion.
PET is also being used increasingly for myocardial perfusion
imaging with improved accuracy in patients with indeterminate
SPECT scans or soft tissue attenuation. PET has several advan-
tages over SPECT imaging including correction for attenua-
tion, better tissue penetration in obese subjects, assessment of
perfusion and function, and slightly better diagnostic accuracy.
Limitations include less widespread availability, especially out-
side of academic centers and large medical complexes, and dif-
ficulty using exercise as the stress modality. Most PET scans
are done with dipyridamole as the pharmacologic stress agent.
In most institutions, PET is not the initial choice of imaging
modality for detection of myocardial ischemia for stress except
in markedly obese patients.
Stress radionuclide angiography is now performed infre-
quently and will not be discussed further.
CT and MRI techniques are also used to assess cardiovascu-
lar risk and prognosis. Although emerging data are beginning to
validate the techniques, their role in the evaluation of patients
with chest pain is still being determined.
Advantages and Limitations of Stress
Test Modalities
Exercise ECG Testing
Standard treadmill assessment is widely available and can be
performed with limited expense. Its use has been well vali-
dated in several different populations. The sensitivity of exercise
ECG testing (weighted mean of 147 studies = 68%) is gener-
ally reported to be lower than that of stress imaging techniques
(77%-78% in the limited studies without ascertainment bias).
The sensitivity appears to be higher in elderly patients and in
those with multivessel disease. Exercise ECG testing maintains a
relatively high specificity (weighted mean of 147 studies = 77%),
but it is decreased in patients with resting ST-segment depres-
sion, left ventricular hypertrophy, LBBB, ventricular paced
rhythms, or valvular heart disease and in those taking digoxin.
(It is important to recognize, however, that very few published
studies have complete ascertainment of sensitivity and specificity
for stress testing in that many patients with negative stress tests
never undergo coronary angiography.) In addition, there appears
to be a higher false-positive rate in women, possibly reflecting
the decreased prevalence of disease in women than in men of
similar age. Finally, exercise ECG testing alone is not useful
for localizing the distribution or extent of myocardial ischemia,
which can influence patient management decisions.
Exercise Stress Echocardiography
Stress echocardiography is widely available and can be per-
formed at an intermediate cost (Table 16.1). Published figures
on sensitivity (range, 70%-97%; overall average, 85%) and
specificity (range, 72%-100%; overall average, 86%) are both
greater with stress echocardiography than with exercise ECG
testing alone, especially in patients with an abnormal baseline
electrocardiogram. However, ascertainment bias (failure to per-
form angiography on all patients who undergo stress echocar-
diography) likely inflates those figures. Echocardiography can
provide pertinent information on the distribution and extent of
CAD, chamber size, global and regional function, and valvular
function. Imaging allows accurate detection of CAD even in the
presence of resting ECG abnormalities and digoxin use. Image
interpretation can be more difficult when resting regional wall
motion abnormalities exist, and interobserver variability remains
a limitation. Finally, image quality can be reduced in certain
patients because of body habitus or pulmonary disease.
Exercise Myocardial Perfusion Imaging
Perfusion imaging has been well validated for both the detection
of CAD and the assessment of prognosis, with widely reproduci-
ble results (Table 16.1). In general, the sensitivity and specificity
for detecting CAD with myocardial perfusion imaging are sim-
ilar to those with stress echocardiography, though several meta-
analyses have suggested that sensitivity is higher and specificity
is lower with myocardial perfusion imaging than with stress ech-
ocardiography. Myocardial perfusion imaging can accurately
localize the distribution and extent of ischemia as well as provide
basic information on myocardial viability, global ventricular size
and function, and regional wall motion in patients with regular
rhythms.
Limitations of exercise myocardial perfusion imaging include
additional equipment and personnel requirements, radionu-
clide administration with both rest and stress imaging, and
an increased cost compared with other stress test modalities.
 16 Stress Test Selection
Table 16.1. Relative Benefit of Stress Test Modalities
Standard Exercise
Feature ECG Test
Sensitivity for CAD Adequate
Specificity for CAD Good
Accuracy with limited ability to exercise Poor
Accuracy with baseline ST-segment abnormalities Poor
Accuracy for ischemia with baseline wall motion Not applicable
abnormalities
Ability to localize ischemia Poor
Ability to assess viability Poor
Ability to assess prognosis Good
Cost Relatively inexpensive
Abbreviations: CAD, coronary artery disease; ECG, electrocardiographic.
a
Several studies suggest that sensitivity is higher and specificity lower with myocardial perfusion imaging than with stress echocardiography.
b
Data on prognosis are more limited with stress echocardiography than with the other modalities.
Soft-tissue attenuation and motion artifact may also limit image
interpretation. Finally, patients who have LBBB or ventricular
paced rhythms have an increased risk of false-positive results
with exercise myocardial perfusion imaging.
PET scanning, if available, is recommended for patients with
significant obesity or if diagnostic quality of a completed SPECT
or stress echocardiographic study is inadequate.
Pharmacologic Stress Testing
The vasodilators adenosine and dipyridamole are the pharma-
cologic agents of choice when performing myocardial perfusion
imaging in patients who cannot exercise. Despite frequent mild
side effects (50% with dipyridamole and 80% with adenosine),
both agents are safe and well tolerated. Side effects from dipyri-
damole can be attenuated with aminophylline, but this is ordi-
narily not needed with adenosine owing to its very short half-life.
Both agents may cause severe bronchospasm in patients who have
asthma or chronic obstructive pulmonary disease with reversible
airway disease, so they should be used with extreme caution or
avoided altogether in these patients. Their use is also limited in
patients with heart block, oral dipyridamole or theophylline use,
and recent caffeine ingestion. Vasodilator stress with perfusion
imaging has been well validated and can detect the presence of
CAD with a high sensitivity and an acceptable specificity that
are similar to those of exercise imaging. Pharmacologic stress
perfusion imaging appears to produce fewer false-positive test
results in patients with LBBB or ventricular paced rhythms than
exercise perfusion imaging and should be strongly considered as
the preferred stress modality in those patients.
Dobutamine is the stress agent of choice for pharmacologic
stress echocardiography. By increasing heart rate, systolic blood
pressure, and myocardial contractility, dobutamine increases myo-
cardial oxygen demand and may provoke ischemia. Dobutamine
is also relatively safe, and side effects may be terminated with
discontinuation of the infusion or administration of a β-blocker.
Patients with severe CAD, depressed left ventricular function,
or recent myocardial injury have an increased risk of ventricu-
lar arrhythmias during dobutamine infusion. Dobutamine stress
echocardiography has a much higher sensitivity for detection of
CAD than vasodilator stress echocardiography (which is used
infrequently in the United States), with an estimated sensitivity
of 82% and specificity of 85%. In addition, dobutamine echocar-
diography can provide additional information on myocardial
viability and ischemic threshold. Dobutamine stress can also be
199
Stress Myocardial
Stress Echocardiography Perfusion Imaging
Very good Very gooda
Very good Very gooda
Very good Very good
Very good Very good
Good Very good
Very good Very good
Good Good
Very goodb Very good
Intermediate expense Expensive
used during myocardial perfusion imaging with reasonable diag-
nostic accuracy, although its use should be restricted to patients
with contraindications to adenosine or dipyridamole since it does
not provoke as great an increase in coronary flow.
Choice of Appropriate Stress Test Modality
A recommended strategy for determining the appropriate stress
test modality (Figure 16.1) is based on the three questions noted
previously.
1. What is the pretest probability of CAD? If a patient has a very low
(<10%) or very high (>90%) pretest probability of CAD by clinical
history, stress testing may not be an appropriate diagnostic strategy
since the results would be unlikely to alter management decisions. If
a patient has contraindications to stress testing or high-risk clinical
findings, consider direct referral for coronary angiography.
2. Are there conditions precluding a diagnostic exercise ECG stress
test? These conditions include baseline ECG abnormalities, digoxin
use, and history of prior revascularization. If none of these barri-
ers are present, standard exercise ECG testing can be performed.
Otherwise, consider exercise testing with imaging. Patients with
LBBB or ventricular paced rhythms may require special consider-
ation because pharmacologic myocardial perfusion imaging may
offer more diagnostic utility than exercise stress imaging. (In some
cases, however, determination of functional capacity or heart rate
response [in the case of paced rhythm] may be a more important
consideration than diagnosis of ischemia, so an exercise test is per-
formed despite the reduced diagnostic accuracy.)
3. Can the patient exercise? If a patient has an intermediate likelihood
of CAD and no contraindications to stress testing, the patient’s abil-
ity to exercise should be assessed. If the patient cannot exercise, con-
sider a pharmacologic imaging study.
With this strategy in mind, review the following three cases.
Case 1
Report of Case
A 45-year-old woman presents with substernal chest pain that
can last for hours and is most prominent while lying supine at
night. She has no cardiovascular risk factors and has a normal
ECG.
Comment
This patient’s age and sex and the noncardiac character of
chest pain would place her at low pretest probability of CAD
200 II Noninvasive Imaging

Patient with stable chest pain

or low-risk or intermediate-risk unstable angina
or previous MI
or post-revascularization

CAD diagnosis Yes Need for risk/prognostic No Need to guide No

certain? assessment? medical management?
Yes
No
Yes

Contraindications Yes Consider coronary Continue/initiate/

to stress testing? angiogram modify medical rx

No

Symptoms warranting Yes

angiography?

No

Can patient No Pharmacologic

exercise? imaging study

Yes

Is resting ECG No
Exercise imaging study
interpretable?*

Yes

Exercise test

Consider coronary
Is test result Yes
angiography/
high risk?**
revascularization
No

Is diagnosis and No Consider imaging

prognosis certain? study/angiography
Yes

Continue/initiate/modify
rx as appropriate

Figure 16.1. Clinic context for stress testing in patients with suspected coronary artery disease. *Electrocardiogram interpretable unless preexci-
tation, electronically paced rhythm, left bundle branch block, or resting ST-segment depression greater than 1 mm. **For example, high-risk if Duke
Treadmill Score predicts average annual cardiovascular mortality greater than 3%. (Previously published. See “Credit Lines” section.)

(approximately 2%). Her lack of other risk factors further sup-
ports a low pretest probability of CAD. She does not require
any further cardiac stress testing. Even though stress testing is
often performed clinically, the likelihood of false-positive results
prompting further unnecessary testing may lead to unwarranted
risk and expense.
Case 2
Report of Case
A 45-year-old woman presents with substernal chest pain that
occurs at times of high anxiety and is relieved after several
minutes of relaxation. She has a history of asthma and diabetes
 16 Stress Test Selection
mellitus but denies other cardiovascular risk factors and has a
normal ECG.
Comment
This patient’s pretest probability is greater than that for the
patient in case 1, primarily because of the more typical char-
acter of her chest pain, placing her at intermediate probability
of CAD (approximately 40%-50%). Diabetes mellitus is another
strong predictor of her intermediate risk. If she can exercise, a
standard exercise ECG stress test should be performed since she
has a normal baseline ECG and has not had prior revascular-
ization. (Although this is still an area of controversy because of
the risk of false-positive results in females with exercise ECG
testing, some cardiologists would recommend that she have an
exercise ECG stress test. If she is still at intermediate risk after
a positive exercise ECG test, she could proceed to subsequent
stress imaging.) If she cannot exercise, a pharmacologic stress
test with dobutamine (typically stress echocardiography) should
be performed. Note that she has asthma, and stress with adenos-
ine or dipyridamole should be avoided owing to the risk of life-
threatening bronchospasm.
Case 3
Report of Case
A 55-year-old man presents with substernal chest pain that
occurs with activity and improves with ibuprofen but not neces-
sarily rest. He is a current smoker but denies other cardiovascular
risk factors, and his ECG demonstrates an LBBB.
Comment
Although sometimes difficult to assess, this patient’s chest pain
is most characteristic of atypical angina. Along with his age,
sex, and smoking history, this would place him at intermediate
probability of CAD (50%-70%). However, exercise ECG test-
ing would be nondiagnostic for the presence of CAD because
he has LBBB. Patients with these characteristics are typically
referred for adenosine or dipyridamole stress myocardial perfu-
sion imaging. Exercise perfusion imaging in patients with LBBB
may lead to a false-positive test, and the baseline regional wall
motion abnormalities from LBBB make stress echocardiography
interpretation more difficult.
Stress Testing for Risk Stratification
and Prognosis
Although stress testing for the diagnosis of CAD should be
restricted to patients at intermediate risk of CAD, prognostic
information can be obtained from patients at both intermedi-
ate and high risk of CAD. Indeed, stress testing is indicated and
often performed in patients with known coronary artery or other
cardiac disease. Patients with a low pretest probability of CAD
have a low event rate regardless of stress testing results and there-
fore do not require further testing. In general, the same princi-
ples that determined appropriate stress testing for the diagnosis
of CAD may also be applied to risk stratification. Exercise test-
ing should be performed with patients who can do the requisite
level of activity. The inability to perform an exercise test has by
itself been shown to be a marker of poor prognosis. Functional
capacity (sometimes described as Functional Aerobic Capacity—
FAC—which expresses exercise capacity as a percent of age-sex
201
predicted) is the strongest predictor of survival. Numerous publi-
cations show the relationship of survival to functional capacity as
f(x)=1/x, meaning that risk increases rapidly as functional capac-
ity decreases below 80% to 85% predicted, while there is little
reduction in risk as functional capacity beyond 100% predicted
for age and sex (Figure 16.2).
Numerous studies have shown that prognosis is good in
patients who can perform at least 10 METs of exercise no matter
what the electrocardiogram or images show. Several predictors
of poor outcome have been identified, including poor exercise
capacity (<5 METs), the degree (>2 mm) and duration (>5 min)
of ST-segment depression, ST-segment elevation, and a low
Duke Treadmill Score. The Duke Treadmill Score is computed
as follows:
Time on Bruce protocol (or equivalent METs)
−5 × maximum ST deviation
–4 × angina index [0 = none; 1 = non-limiting, 2 = limiting]
A Duke Treadmill Score ≥5 indicates low risk for cardiovascular
death. Intermediate risk is given at a score of −11 to +4, and high
risk is indicated by a Duke Treadmill Score of <−11. The Duke
Treadmill Score is widely used to establish prognosis, though
it has 2 major limitations. First of all, the angina index is very
subjective in that the designation of chest pain as angina or not
can be strongly influenced by the pretest probability of CAD and
also by ST changes during exercise. A second problem is that
other validated prognostic measures, particularly chronotropic
incompetence and impaired HR recovery are not included in the
Duke Treadmill Score.
Chronotropic incompetence is defined as failure to achieve
at least 85% of age-predicted heart rate—generally given as
220-age—on the exercise test. Chronotropic incompetence
indicates some failure of the sympathetic nervous system.
Unfortunately, patients taking β-blockers and even certain cal-
cium channel blockers must be excluded from determination of
chronotropic incompetence.
Impaired heart rate recovery is calculated as peak exercise
heart rate—heart rate at 1-minute of active recovery. Values < 13
beats per minute are considered abnormal. A standardized active
Least
5 Fit
Relative Risk of Death
4
3
2
Most
Fit
1
0
1 2 3 4 5
(1.0 - <6.0 (6.0 - <8.0 (8.0 - <10.0 (10.0 - <13.0 (≥13.0
METs) METs) METs) METs) METs)
Categories of Exercise Capacity
Figure 16.2. All-cause mortality versus exercise capacity in healthy
men showing an f(x) = 1/x relationship of risk to exercise capacity.
(Previously published. See “Credit Lines” section.)
202 II Noninvasive Imaging
recovery is necessary for computation of heart rate recovery, so
it is generally not available for an exercise echo (with no active
recovery) or for any of the pharmacologic stress tests. Heart rate
recovery is a function of parasympathetic activity, so it is not
affected by β-blockade.
Other prognostic signs with less extensive validation include
low peak systolic blood pressure (<130 mm Hg) and exercise-
induced LBBB. Asymptomatic ventricular ectopy, even short
runs of ventricular tachycardia, on the exercise test is generally
not considered to have prognostic significance, though one paper
by Lauer et al suggested that frequent ventricular ectopy during
early active recovery was predictive of increased mortality.
In addition to the Duke Treadmill Score, numerous other tread-
mill scoring systems or scores have been developed, some taking
into account only responses during the exercise test, others incor-
porating clinical variables or results of other tests such as a resting
echocardiogram or various laboratory parameters. A publication
from the exercise laboratory at Mayo Clinic suggests that prog-
nosis on a standard exercise ECG stress test can be simply deter-
mined by the sum of abnormalities (poor functional capacity, ST
depression > 1.0 mm, and abnormal HR response) from 0 to 3.
In patients with a normal resting ECG who are not taking dig-
oxin, the modest incremental benefit of additional stress imaging
for the prediction of subsequent events does not appear to justify
the added cost. However, a stress imaging technique should be
considered as the initial stress test in patients with ST-segment
depression of more than 1 mm, LBBB, ventricular paced rhythms,
or pre-excitation. The extent and severity of exercise-induced left
ventricular dysfunction that occurs with exercise echocardiog-
raphy may add incremental prognostic value beyond the Duke
Treadmill Score alone. Similarly, with dobutamine echocardi-
ography, the number of territories with both resting and exer-
cise-induced regional wall motion abnormalities is a predictor
of mortality.
The prognostic value of myocardial perfusion imaging with
either exercise or pharmacologic stress has also been well vali-
dated. The presence, extent, and severity of perfusion defects
can provide independent prognostic information beyond exercise
data, except in patients with a high exercise capacity. A normal
perfusion scan predicts a low event rate comparable with that
of normal stress echocardiography. A mildly abnormal scan
identifies a group with low mortality but increased risk of car-
diac events requiring more aggressive medical management.
Moderate or severely abnormal scans predict a poor prognosis
and suggest the need for aggressive management with angiogra-
phy and possible revascularization.
In summary, exercise ECG testing can ultimately provide
adequate detection of CAD and risk stratification in the major-
ity of patients. In those who require further assessment with
imaging, the choice of stress imaging technique requires careful
consideration of the individual patient characteristics. In many
cases, either stress echocardiography or myocardial perfusion
imaging may be appropriate; the preferred test often depends on
local expertise, available facilities, and considerations of cost-
effectiveness.
For patients with complex cardiovascular disease, patients
presenting with dyspnea on exertion in the setting of known
or suspected pulmonary disease, competitive athletes, patients
requiring a sophisticated exercise prescription, and patients
undergoing disability determination, cardiopulmonary exercise
testing is recommended.
Suggested Reading
Brindis RG, Douglas PS, Hendel RC, et al. ACC/ASNC appropriateness
criteria for single-photon emission computed tomography myocardial
perfusion imaging (SPECT MPI): a report of the American College
of Cardiology Foundation Quality Strategic Directions Committee
Appropriateness Criteria Working Group and the American Society
of Nuclear Cardiology. Available from: http://www.acc.org/clinical/
pdfs/SPECTMPIACPubFile.pdf.
Gibbons RJ, Antman EM, Alpert JS, et al. ACC/AHA 2002 guideline
update for the management of patients with chronic stable angina.
Available from: http://www.acc.org/clinical/guidelines/stable/
update_index.htm.
Gibbons RJ, Balady GJ, Bricker JT, et al. ACC/AHA 2002 guideline
update for exercise testing—full text. Available from: http://www.
acc.org/clinical/guidelines/exercise/dirindex.htm.
Abbreviations
CAD coronary artery disease
CPX cardiopulmonary exercise testing
CT computed tomography, computed tomographic
ECG electrocardiography, electrocardiographic
LBBB left bundle branch block
MET metabolic equivalent
MRI magnetic resonance imaging
PET positron emission tomography
SPECT single-photon emission computed tomography
 Section III

Electrophysiology
This page intentionally left blank
 17

Electrocardiographic Diagnoses: Criteria

and Definitions of Abnormalities
STEPHEN C. HAMMILL, MD

This chapter is a diagnostic guide for all physicians who evaluate of Wolff-Parkinson-White syndrome, QT interval prolongation,
ECGs, and the criteria are of specific value to persons taking pro- the prominent U waves of hypokalemia, the tall R wave in lead
fessional examinations in cardiology and internal medicine. The V1 associated with a posterior wall infarct, and PR-segment
criteria are those used in clinical practice at Mayo Clinic, and depression in acute pericarditis.
thus they may differ in minor ways from those used on examina-
tions. Check the specific diagnoses available on the answer sheet. Systematic Evaluation of an ECG Tracing
These criteria are not endorsed by any professional examination
body or organization. 1. Standardization and leads shown; review for reversed leads, right
chest leads, and one-half voltage standardization in left ventricular
hypertrophy
Examination Strategy 2. Heart rate: determine separately for QRS and P waves if the rhythm
is other than sinus
When reviewing ECGs, be sure to evaluate the tracing system- 3. Heart rhythm
atically for abnormalities of the rate, rhythm, and axis and to 4. Cardiac axis
examine the configuration, duration, and relationship of the P, 5. Configuration and duration of P, QRS, and T waves
QRS, and T waves. Most ECGs used in examinations have 1 to 3 6. Relationship of P wave to QRS complex
major findings. Examinations in cardiology emphasize the cor- 7. Intervals: PR, QRS, and QT
rect identification of the major diagnoses on the tracing; points 8. ECG diagnosis
are subtracted for important oversights and misdiagnoses. The 9. Suggested clinical diagnosis
ancillary minor diagnoses are given minimal or neutral credit.
This chapter includes a score sheet similar to that used on ECG Diagnoses
examinations in cardiology. Minor changes are made to the num-
bering of the examination answer codes from year to year. In this 1. General Features
chapter, letters are purposely used for codes to avoid confusion a. Normal ECG
between the numbering scheme of examinations and the criteria b. Borderline normal ECG or normal variant
in this chapter. Initially, this chapter lists all the ECG diagnoses; c. Incorrect electrode placement
in subsequent pages, criteria for each diagnosis are defined on d. Artifact due to tremor
the score sheet.
Do not guess the ECG diagnoses on the examination, 2. Atrial Rhythms
because wrong diagnoses receive negative credits. If the ECG
a. Sinus rhythm
looks normal, be aware that it may not be normal. Always
b. Sinus arrhythmia
double-check for subtle ECG changes, including the delta wave c. Sinus bradycardia (<60 beats per minute)
d. Sinus tachycardia (>100 beats per minute)
Abbreviations and acronyms are expanded at the end of this chapter. e. Sinus pause or arrest
205
206 III Electrophysiology

f. Sinoatrial exit block i. Aberrant intraventricular conduction with supraventricular arrhyth-

g. Ectopic atrial rhythm mia (specify rhythm)
h. Wandering atrial pacemaker
i. Atrial premature complexes, normally conducted
j. Atrial premature complexes, nonconducted 8. P-Wave Abnormalities
k. Atrial premature complexes with aberrant intraventricular conduction a. Right atrial abnormality
l. Atrial tachycardia (regular, sustained, 1:1 conduction) b. Left atrial abnormality
m. Atrial tachycardia, repetitive (short paroxysms) c. Nonspecific atrial abnormality
n. Atrial tachycardia, multifocal (chaotic atrial tachycardia)
o. Atrial tachycardia with AV block
p. Supraventricular tachycardia, unspecified 9. Abnormalities of QRS Voltage or Axis
q. Supraventricular tachycardia, paroxysmal a. Low voltage, limb leads only
r. Atrial flutter b. Low voltage, limb and precordial leads
s. Atrial fibrillation c. Left-axis deviation (>−30°)
t. Retrograde atrial activation d. Right-axis deviation (>+100°)
e. Electrical alternans
3. AV Junctional Rhythms
a. AV junctional premature complexes 10. Ventricular Hypertrophy
b. AV junctional escape complexes
a. Left ventricular hypertrophy by voltage only
c. AV junctional rhythm, accelerated
b. Left ventricular hypertrophy by both voltage and ST-T–segment
d. AV junctional rhythm
abnormalities
c. Right ventricular hypertrophy
4. Ventricular Rhythms d. Combined ventricular hypertrophy
a. Ventricular premature complex(es), uniform, fixed coupling
b. Ventricular premature complex(es), uniform, nonfixed coupling 11. Transmural Myocardial Infarction
c. Ventricular premature complex(es), multiform
d. Ventricular premature complexes, in pairs (2 consecutive)
Probably acute Probably old or
e. Ventricular parasystole
f. Ventricular tachycardia (≥3 consecutive beats)
or recent age indeterminate
g. Accelerated idioventricular rhythm Anterolateral a. g.
h. Ventricular escape complexes or rhythm
Anterior b. h.
i. Ventricular fibrillation
Anteroseptal c. i.
Lateral or high lateral d. j.
5. Atrioventricular Interactions Inferior (diaphragmatic) e. k.
in Arrhythmias Posterior f. l.
a. Fusion complexes
b. Reciprocal (echo) complexes m. Probable ventricular aneurysm
c. Ventricular capture complexes
d. AV dissociation
e. Ventriculophasic sinus arrhythmia
12. ST-, T-, and U-Wave Abnormalities
a. Normal variant, early repolarization
b. Normal variant, juvenile T waves
6. AV Conduction Abnormalities c. Nonspecific ST- or T-wave abnormalities
a. AV block, first degree d. ST-segment or T-wave abnormalities suggesting myocardial ischemia
b. AV block, second degree–Mobitz type I (Wenckebach) e. ST-segment or T-wave abnormalities suggesting myocardial injury
c. AV block, second degree–Mobitz type II f. ST-segment or T-wave abnormalities suggesting acute pericarditis
d. AV block, 2:1 g. ST-segment or T-wave abnormalities due to intraventricular conduc-
e. AV block, third degree tion disturbance or hypertrophy
f. AV block, variable h. Post-extrasystolic T-wave abnormality
g. Short PR interval (with sinus rhythm and normal QRS duration) i. Isolated J-point depression
h. Wolff-Parkinson-White pattern j. Peaked T waves
k. Prolonged QT interval
l. Prominent U waves
7. Intraventricular Conduction
Disturbances
13. Pacemaker Function and Rhythm
a. RBBB, incomplete a. Atrial or coronary sinus pacing
b. RBBB, complete b. Ventricular demand pacing
c. Left anterior fascicular block c. AV sequential pacing
d. Left posterior fascicular block d. Ventricular pacing, fixed rate (asynchronous)
e. LBBB, complete with ST-T waves suggestive of acute myocardial e. Dual-chamber, atrial-sensing pacemaker
injury or infarction f. Pacemaker malfunction, not constantly capturing (atrium or ventricle)
f. LBBB, complete g. Pacemaker malfunction, not constantly sensing (atrium or ventricle)
g. LBBB, intermittent h. Pacemaker malfunction, not firing
h. Intraventricular conduction disturbance, nonspecific type i. Pacemaker malfunction, slowing
 17 Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities 207

14. Suggested or Probable Clinical Disorders • rSR′ or rSr′ in lead V1 (2.4% of normals)
a. Digitalis effect ° QRS duration <0.10 second and <7 mm in height
b. Digitalis toxicity ° Amplitude of r′ smaller than amplitude of r or S
c. Antiarrhythmic drug effect c. Incorrect electrode placement—most commonly the following:
d. Antiarrhythmic drug toxicity • Reversal of right and left arm leads (Figure 17.2)
e. Hyperkalemia
f. Hypokalemia Resultant ECG mimics dextrocardia in limb leads with P,
g. Hypercalcemia QRS, and T inversion in limb leads I and aVL. However, the
h. Hypocalcemia precordial leads remain normal and thus rule out dextrocardia.
i. Atrial septal defect, secundum
j. Atrial septal defect, primum • Reversal of chest V leads (Figure 17.3). There is a sudden decrease in
k. Dextrocardia, mirror image the R-wave amplitude with return in the next V lead
l. Mitral valve disease
d. Artifact due to tremor
m. Chronic lung disease
• Parkinson tremor simulates atrial flutter with a rate of approxi-
n. Acute cor pulmonale, including pulmonary embolus
mately 300/min (4–6/s) (Figure 17.4)
o. Pericardial effusion
• Physiologic tremor rate is 500/min (7–9/s)
p. Acute pericarditis
• Most prominent in limb leads
q. Hypertrophic cardiomyopathy
r. Coronary artery disease
s. Central nervous system disorder 2. Atrial Rhythms
t. Myxedema
u. Hypothermia a. Sinus rhythm
v. Sick sinus syndrome • Rate 60–100 beats per minute
w. Ebstein anomaly • P axis normal (+15° to +75°)
b. Sinus arrhythmia—requires the following:
Criteria for Score Sheet Diagnosis • P-wave morphology and axis normal
• PP interval varies by >0.16 second or 10%
1. General Features
c. Sinus bradycardia (<60 beats per minute)—requires the following:
a. Normal ECG • Rate <60 beats per minute
• No abnormalities of the rhythm, rate, or axis • Normal P-wave axis
• The configurations of the P wave, QRS complex, and T wave are
within normal limits (Figure 17.1) Note: If rate <40 beats per minute, consider 2:1 sinoatrial exit
block.
b. Borderline normal ECG or normal variant
• Early repolarization (see item 12a) d. Sinus tachycardia (>100 beats per minute)—requires the following:
• Juvenile T waves (see item 12b) • Rate >100 beats per minute
• S1, S2, and S3 • Normal P-wave axis
A terminal negative deflection is present in the QRS com- Note: P amplitude often increases and PR interval shortens with
plexes in the standard limb leads in up to 20% of healthy adults— increasing rate.
it should be distinguished from abnormal left axis deviation.
e. Sinus pause or arrest—pause >2.0 seconds without a P wave

Note: The differential diagnosis includes the following:

• Sinus arrhythmia—phasic, gradual PP-interval change
• Sinoatrial exit block—pause is multiple (2 times, 3 times, etc) of
R usual PP interval
• Nonconducted atrial premature complexes—look for P wave deform-
ing the preceding T wave
PR ST • Sinus arrest
segment segment T f. Sinoatrial exit block
P • First- and third-degree not detectable on surface ECG
U • Second-degree
° Type I (Mobitz I)—Group beating with shortening of the PP
T interval, constant PR interval, and a PP pause that is less than
P
Q twice the normal PP interval
S ° Type II (Mobitz II)—Constant PP interval followed by a dropped
PR ST P wave, the pause being a near multiple of the normal PP inter-
interval interval val. Interval of pause may be slightly less than twice the normal
QRS PP interval but usually within 0.10 second
g. Ectopic atrial rhythm—requires all the following:
QT • P-wave axis or morphology different from sinus rhythm
interval • Rate <100 beats per minute
• PR interval >0.11 second
Figure 17.1. Components of the Scalar Electrocardiogram. Standard
wave labeling (P, Q, R, S, T, U) and clinically useful interval and seg- Note: Low atrial focus may activate atrium retrogradely
ment measurements are shown. (Previously published. See “Credit (P inverted in leads II, III, and aVF), but PR interval >0.11
Lines” section.) second, distinguishing it from AV junctional rhythm.
208 III Electrophysiology

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

Figure 17.2. Reversal of Right and Left Arm Leads.

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

Figure 17.3. Reversal of V2 and V3 Leads in a 61-Year-Old Woman.

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

V1

V5

Figure 17.4. Parkinson Tremor Causing Artifact.

 17 Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities
h. Wandering atrial pacemaker—requires the following:
• Rate <100 beats per minute
• Varying P waves with ≥3 morphologic patterns
i. Atrial premature complexes, normally conducted—suggested by the
following:
• Premature P wave in relation to normal sinus rhythm
• P wave usually abnormal in configuration
• PR interval may be normal, increased, or decreased
• Post-extrasystolic pause is noncompensatory unless sinoatrial
entrance block is present and sinoatrial node is not reset, result-
ing in either an interpolated beat or a full compensatory pause
• QRS complex similar in morphology to the QRS complex pres-
ent during sinus rhythm
j. Atrial premature complexes, nonconducted (Figure 17.5 and 17.6)—
suggested by the following:
• Premature P waves that are abnormal in morphology but not fol-
lowed by QRS
• P waves that are often hidden in T wave (look for deformed T
wave)
• The sinus node is usually reset, resulting in RR interval pause
k. Atrial premature complexes with aberrant intraventricular conduc-
tion (Figure 17.7)—suggested by the following:
• P wave that occurs very early
• RBBB pattern is most common, but LBBB or even variable QRS
morphology may occur
Note: Also see item 7i.
l. Atrial tachycardia (regular, sustained, 1:1 conduction)—suggested
by the following:
• Abnormal P waves that are different in morphology from sinus
P waves
• Three or more beats in succession
• The atrial rate is generally 100–180 beats per minute
• Regular rhythm (constant RR interval), except for a warm-up
period in the automatic type
• A QRS complex follows each P wave—the QRS complex usually
resembles the morphology present during sinus rhythm unless
aberrantly conducted
• PR interval may be within normal limits or prolonged
• Secondary ST- and T-wave changes may occur
m. Atrial tachycardia, repetitive (short paroxysms)
• Characterized by recurring short runs of atrial tachycardia inter-
rupted by normal sinus rhythm
Note: Refer to item 2l for definition of atrial tachycardia.
n. Atrial tachycardia, multifocal (chaotic atrial tachycardia)—requires
all of the following:
• P waves of at least 3 morphologic patterns
• Absence of 1 dominant atrial pacemaker (in contradistinc-
tion to normal sinus rhythm with multifocal atrial premature
complexes)
• Variable PR, RR, and RP intervals
Figure 17.5. Nonconducted Atrial Premature Complex (arrow) Causing a Pause.
209
• Atrial rate >100 beats per minute
• Isoelectric baseline between P waves
o. Atrial tachycardia with AV block—requires all of the following:
• Abnormal P waves that are different in morphology from P waves
of sinus rhythm
• Atrial rate usually 150–240 beats per minute
• Isoelectric intervals between P waves in all leads (unlike atrial
flutter)
• AV block to a degree beyond simple PR prolongation (second or
third degree)
• Rhythm is regular, but ventriculophasic sinus arrhythmia may
occur (refer to item 5e for definition)
Note: Most cases are due to digitalis toxicity (refer to item 14b).
p. Supraventricular tachycardia, unspecified
• Rhythm is regular
• P wave not easily identified
• QRS complex usually narrow (occasionally aberrant)
Note: If rate is 150 beats per minute, rule out atrial flutter with
2:1 block.
q. Supraventricular tachycardia, paroxysmal
• Onset and termination sudden
• May have retrograde P wave (see item 2t)
• Refer to item 2p for definition of supraventricular tachycardia
r. Atrial flutter
• Rapid regular undulations (F waves), saw-tooth pattern usually
seen best in leads II, III, aVF, and V1
• Atrial rate 240–340 beats per minute. May be faster in children.
May be slower in the presence of class IA, IC, and III antiar-
rhythmic drugs
• QRS complex may be normal or aberrantly conducted
• Rate and regularity of QRS complexes are variable and depend
on the AV conduction sequence
• AV conduction
° Complete block may occur with or without AV junctional
tachycardia (usually digoxin toxicity)
° May have varying degrees of block (2:1, 4:1, or more)
s. Atrial fibrillation
• P waves are absent. Atrial activity is represented by fibrillation
(f) waves of varying amplitudes, duration, and morphology caus-
ing random oscillation of the baseline
• The ventricular rhythm, in the absence of third-degree AV block,
is irregularly irregular
• Atrial activity is best seen in the right precordial and inferior
leads
• Rate is usually 100–180 beats per minute in the absence of
drugs. If rate is <100 beats per minute, conduction system
disease is likely to be present. If rate is >200 beats per minute
with a QRS complex >0.12 second in duration, consider Wolff-
Parkinson-White syndrome
210 III Electrophysiology
Figure 17.6. Nonconducted Atrial Premature Complexes (arrows) in Bigeminy Causing Bradycardia. Strip is continuous.
• Differential diagnosis:
° Multifocal atrial tachycardia
° Paroxysmal atrial tachycardia with block
° Atrial flutter
t. Retrograde atrial activation
• Inverted P waves in leads II, III, and aVF
• Look for retrograde P waves after ventricular premature com-
plexes and other ectopic junctional or ventricular beats
3. AV Junctional Rhythms
a. AV junctional premature complexes—require all of the following:
• Occur early in cycle, in contrast to escape beats
• P wave is inverted in leads II, III, and aVF and upright in leads
I and aVL
• P wave may precede the QRS by 0.11 second or may be superim-
posed on or follow the QRS
• Ventricular complex may show aberration
• Coupling interval is usually constant
• Noncompensatory pause is usually seen
Note: Consider also item 2t.
b. AV junctional escape complexes
• There is decreased sinus impulse formation or conduction from
the sinoatrial node, or high-degree AV block at or proximal to the
bundle of His. Atrial mechanism may be sinus rhythm, paroxys-
mal atrial tachycardia, atrial flutter, or atrial fibrillation
Figure 17.7. Atrial Premature Complexes With Aberrant Intraventricular Conduction (arrows).
• May be seen with post-extrasystolic pause after atrial tachycar-
dia, atrial flutter, or atrial fibrillation
c. AV junctional rhythm, accelerated—requires all of the following:
• Rate >60 beats per minute
• Variable relationship between atrial and ventricular rates. If ret-
rograde block is present, atria remain in sinus rhythm and AV
dissociation will be present. If retrograde activation occurs, con-
stant QRS-P interval will be present
• May be seen with atrial fibrillation or atrial flutter with complete
heart block (consider digoxin toxicity)
• Exit block also occurs with digoxin toxicity
Note: Consider items 2t and 14b.
d. AV junctional rhythm (rate ≤60 beats per minute)
• RR interval of escape rhythm usually constant (<0.04-second
variation)
• May have isorhythmic AV dissociation (consider item 5d)
• P wave inverted in leads II, III, and aVF and upright in leads I
and aVL (consider item 2t)
4. Ventricular Rhythms
a. Ventricular premature complex(es), uniform, fixed coupling
(Figure 17.8)—all of the following are required:
• Premature in relation to normal cycles, not preceded by P wave
(or shorter than expected PR interval or “collapsing PR”)
 17 Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities
• Coupling interval usually the same for each site or focus (varia-
tion usually <0.08 second)
• Abnormal QRS configuration that is almost always >0.12 second
in duration
• Retrograde capture of atria may occur (consider item 2t)
• Initial direction of QRS complex is often different from that
observed during sinus rhythm
• Usually full compensatory pause is noted
• Compensatory pause requires an undisturbed sinus depolariza-
tion due to one of the following:
° Ventriculoatrial block
° Sinoatrial entrance block if atrial capture occurs
° Sinoatrial node discharged before arrival of retrograde wave-
front, and thus refractory
b. Ventricular premature complex(es), uniform, nonfixed coupling
• Ventricular premature complexes with variable temporal rela-
tionship to regular sinus beat
c. Ventricular premature complex(es), multiform
• Two or more morphologic patterns of ventricular premature com-
plexes present
d. Ventricular premature complexes, in pairs (2 consecutive)
• Two consecutive ventricular premature complexes of not neces-
sarily the same morphology
Note: Refer to item 4a for criteria.
e. Ventricular parasystole—an automatic ventricular focus with
entrance block and all of the following:
• Rates usually 30–56 beats per minute
• Varying relationship with the preceding sinus beats
• All interectopic intervals are a multiple of a constant shortest
interval
• When fusion beats and lack of fixed coupling are noted, consider
parasystole
f. Ventricular tachycardia (≥3 consecutive beats)—rapid succession of
≥3 beats of ventricular origin (Figure 17.9 and 17.10)
• Abnormal and wide QRS complexes with secondary ST-T
changes
Note: Ventricular tachycardia originating in the septum near the
normal conduction system may have a narrow QRS complex.
Figure 17.8. Ventricular Premature Complex Resulting in Concealed Retrograde Conduction During Atrial Fibrillation.
211
• Rate >100 beats per minute
• Regular or slightly irregular
• Abrupt onset and termination
• AV dissociation is common. On occasion, retrograde conduction and
capture of the atria may occur
• Look for ventricular capture and fusion beats as a marker for ven-
tricular tachycardia
Ventricular origin is favored with the following:
• QRS complexes like those of ventricular premature complexes
• Tachyarrhythmia initiated by ventricular premature complexes
• AV dissociation
• Capture or fusion beats
• QRS ≥0.14 second if RBBB morphology and ≥0.16 second if LBBB
morphology when QRS during sinus rhythm <0.12 second
• Left or northwest axis deviation
• All positive or all negative complexes in precordial leads
• In lead V1, R > r′ (left rabbit ear taller than right)
Supraventricular origin is favored with the following:
• QRS complex like aberrantly conducted atrial premature complexes
or QRS in sinus rhythm
• Tachyarrhythmia initiated by atrial premature complexes
• RBBB configuration with rSR′ in lead V1
g. Accelerated idioventricular rhythm—requires all of the following:
• Regular rhythm, rate 60–110 beats per minute
• QRS complexes are abnormal and wide
• Usually AV dissociation
• Capture and fusion beats are common because of slower rate
Note: Also consider items 5a and 5c.
h. Ventricular escape complexes or rhythm—requires all of the following:
• Rate is usually 30–40 beats per minute (can be 20–50 beats per
minute)
• QRS complexes are abnormal and wide
• Occurs when the rate of supraventricular impulse arriving at the
ventricle is slower than the inherent rate of the ectopic ventricular
pacemaker
i. Ventricular fibrillation
• Chaotic and irregular deflections of varying amplitude and contour
• No P waves, QRS complexes, or T waves
212 III Electrophysiology
Figure 17.9. Torsades de Pointes in a Patient With Long QT Syndrome.
5. Atrioventricular Interactions
in Arrhythmias
a. Fusion complexes—caused by simultaneous activation of the ventri-
cle from 2 sources and may occur with the following:
• Ventricular premature complexes
• Wolff-Parkinson-White syndrome
• Ventricular tachycardia
• Ventricular parasystole
• Accelerated idioventricular rhythm
• Paced rhythm
b. Reciprocal (echo) complexes
• The impulse activates a chamber (atrium or ventricle), returns,
and reactivates the same chamber
c. Ventricular capture complexes
• Ventricular capture by conducted supraventricular impulses
resulting in a fusion beat or a QRS morphology similar to that
during sinus rhythm
• Strong but not infallible evidence for rhythm of ventricular
origin
d. AV dissociation—requires all of the following:
• Atrial and ventricular activities that are independent
• Ventricular rate that is faster than atrial rate
• Always the result of some other disturbance of cardiac rhythm
e. Ventriculophasic sinus arrhythmia
• A PP interval containing a QRS complex is shorter than a PP
interval without a QRS
• Common in presence of high-grade AV block
6. AV Conduction Abnormalities
a. AV block, first degree—requires all of the following:
• PR interval ≥0.20 second (usually 0.21–0.40 second, but may be
as long as 0.80 second)
• Each P wave is followed by a QRS complex
• Usually a constant PR interval (Figure 17.11)
b. AV block, second degree–Mobitz type I (Wenckebach) (Figure 17.12
and 17.13)—requires all of the following:
• Progressive prolongation of PR interval until P wave fails to con-
duct to the ventricle
• Progressive shortening of RR interval until P wave is not
conducted
Figure 17.10. Monomorphic Ventricular Tachycardia With Fusion Complexes.
• The RR interval containing the nonconducted P wave is shorter
than the sum of 2 PP intervals
• Results in “group” or pattern beating
c. AV block, second degree–Mobitz type II (Figure 17.14)—requires
all of the following:
• There are intermittent nonconducted P waves with no evidence
of atrial prematurity
• In conducted beats, PR intervals stay constant
• The RR interval containing the nonconducted P wave is equal to
2 PP intervals
• A 2:1 AV block can be Mobitz type I or II and cannot be distin-
guished without the following information:
° Results of maneuvers used to increase heart rate and improve
AV conduction (atropine and exercise typically decrease type
I block and increase type II block) (Figure 17.15), or
° Classic Mobitz type I seen on another part of the ECG (then
probably type I), or
° QRS conduction is abnormal with bundle branch block or
bifascicular block (then usually Mobitz type II)
d. AV block, 2:1
• There are 2 P waves for each QRS complex (every other P wave
is nonconducted)
Note: Refer to item 6c.
e. AV block, third degree (Figure 17.16)—requires all of the following:
• Independent atrial and ventricular activities
• Atrial rate faster than ventricular rate
• Ventricular rhythm maintained by a junctional or idioventricular
escape rhythm or ventricular pacemaker
• Ventriculophasic sinus arrhythmia in 30% of cases (consider
item 5e)
• When the ventricular rate is faster than the atrial rate, AV disso-
ciation (not AV block, third degree) is present
f. AV block, variable
• Varying degrees of AV block, including 3:2, 2:1, and 4:1 present
on the same ECG
• Consider this in atrial flutter with variable RR intervals (flutter
wave to R wave) after ruling out third-degree AV block
g. Short PR interval (with sinus rhythm and normal QRS duration)—
requires the following:
• Sinus P wave
• PR interval <0.12 second
 17 Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities
Figure 17.11. Changing RP Interval Affecting the Subsequent PR Interval During Sinus Arrhythmia.
h. Wolff-Parkinson-White pattern (Figure 17.17)—suggested by the
following:
• Normal P wave with PR interval <0.12 second (rarely >0.12
second)
• Abnormally wide QRS >0.10 second
• Initial slurring of QRS (delta wave)
• PJ interval is constant and ≤0.26 second
Note: Atrial fibrillation or flutter that has QRS with varying
width (generally wide) and rate >200 beats per minute suggests
Wolff-Parkinson-White syndrome.
7. Intraventricular Conduction Disturbances
a. RBBB, incomplete—RBBB morphology (rSR′), but QRS duration is
0.09–0.11 second
b. RBBB, complete—requires all of the following (Figure 17.18):
• Prolonged QRS ≥0.12 second
• Secondary R wave (R′) in right precordial leads, usually with
R′ > initial R
• Delayed intrinsicoid deflection in right precordial leads >0.05
second
• Wide S wave in I, V5, and V6
• Secondary ST-T segment changes in leads V1 through V3
• Axis as determined by initial 0.06 to 0.08 second of QRS should
be normal unless concomitant left anterior fascicular block is
present
Note: Consider item 12g.
c. Left anterior fascicular block (Figure 17.19)—requires all of the
following:
• Displacement of mean QRS axis to between −45° and −90°
• qR complex (or an R wave) in leads I and aVL; rS in lead III
• Normal or slightly prolonged QRS duration (0.08–0.10 second)
• No other factors responsible for left axis deviation, such as the
following:
° Left ventricular hypertrophy
° Inferior infarct
° Emphysema (chronic lung disease)
Figure 17.12. Atrioventricular Block, Second Degree–Mobitz Type I (Wenckebach).
213
d. Left posterior fascicular block (Figure 17.20)—requires all of the
following:
• Frontal plane QRS axis of +100° to +180°
• S1 Q3 pattern (deep S wave in lead I, with Q wave in lead III)
• Normal or slightly prolonged QRS duration (0.08–0.10 second)
• No other factors responsible for right axis deviation, such as the
following:
° Right ventricular hypertrophy
° Vertical heart
° Emphysema (chronic lung disease)
° Lateral wall myocardial infarction
e. LBBB, complete with ST-T waves suggestive of acute myocardial
injury or infarction (Figures 17.21, 17.22, and 17.23)—requires all of
the following:
• Fulfills criteria for LBBB (see item 7f)
• ≥1 mm ST elevation concordant with QRS
• ≥1 mm ST depression in leads V1 through V3
• ≥5 mm ST elevation discordant with QRS
• Criteria valid with right ventricular artificial pacemaker
f. LBBB, complete—requires all of the following:
• Prolonged QRS duration ≥0.12 second
• Delayed intrinsicoid deflection in left precordial leads and lead
I >0.05 second
• Broad monophasic R in leads I, V5, and V6, which is usually
notched or slurred
Note: Also consider items 9c and 12g.
g. LBBB, intermittent—more common at high rates but also may be
bradycardia-dependent
h. Intraventricular conduction disturbance, nonspecific type
• QRS >0.11 second, but QRS morphology does not satisfy criteria
for either LBBB or RBBB
• May also be used when there is abnormal notching of the QRS
complex without prolongation
• May occur with antiarrhythmic drug toxicity, hyperkalemia, and
hypothermia
Note: Also consider items 12g, 14d, and 14e.
214 III Electrophysiology

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

Figure 17.13. Acute Inferior Myocardial Infarction With Atrioventricular Block, Second Degree–Mobitz Type I (Wenckebach), in a 69-Year-Old
Man. Arrows indicate P waves preceding dropped beats.

i. Aberrant intraventricular conduction with supraventricular arrhyth- Biatrial enlargement is suggested by any of the following:
mia (specify rhythm)
• Large biphasic P wave in lead V1 with initial positive component
Note: See item 4f for criteria of supraventricular versus ventricu- of 1.5 mm and the P terminal force with a negative amplitude of
lar tachycardia. 1 mm and a duration of 0.04 second
• Tall, peaked P waves (>1.5 mm) in right precordial leads (V1
through V3) and wide, notched P waves in left precordial leads
8. P-Wave Abnormalities (V5 and V6)
a. Right atrial abnormality • P-wave amplitude ≥2.5 mm and duration ≥0.12 second in limb
• Amplitude >2.5 mm in leads II, III, or aVF with a normal P-wave leads
duration (P pulmonale), or c. Nonspecific atrial abnormality
• Positive amplitude >1.5 mm in leads V1 or V2, or • Abnormal P-wave morphology but not fulfi lling criteria for right
• P-wave frontal axis ≥70° (rightward axis) or left atrial enlargement
b. Left atrial abnormality
• Notched P wave with a duration ≥0.12 second in leads II, III, or
9. Abnormalities of QRS Voltage or Axis
aVF (P mitrale), or
• Downward terminal deflection of P wave in lead V1 with a a. Low voltage, limb leads only
negative amplitude of 1 mm and with duration of 0.04 second • Amplitude of the entire QRS complex (R + S) is <5 mm in all
(Figure 17.24) limb leads

I aVR V1 V4

II aVL V2
V5

III aVF V3 V6

II

Figure 17.14. Sinus Rhythm With Atrioventricular Block, Second Degree–Mobitz Type II, and Right Bundle Branch Block. Arrow indicates P
wave before dropped beat.
 17 Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities 215

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

B V4

V5

V6

C HRA

HBE V
A H A H

HBE

V1

aVF

Figure 17.15. A and B, Electrocardiograms (ECGs) From a 59-Year-Old Man. A, Resting ECG with right bundle branch block. B, Exercise ECG
with a 2:1 atrioventricular block. Arrows indicate P waves with conduction of every second beat. C, His Bundle Recording. Right bundle branch
block and atrioventricular block distal to His. The highlighted complex shows an atrial (A) followed by a His (H) depolarization but no ventricular
(V) depolarization. HBE indicates His bundle electrogram; HRA, high right atrium.
216 III Electrophysiology

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

P P P P P P P P P P P P
Y

Figure 17.16. Complete Heart Block (Third-Degree Atrioventricular Block).

A
I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

V1

V5

B
Location V1 aVF aVL

Left lateral + + -
Left posterior/septal + - +
Right posterior/septal - - +
Right lateral/anterior - + +

Figure 17.17. Wolff-Parkinson-White Pattern. A, Pattern as seen on electrocardiogram. B, Pathway location in Wolff-Parkinson-White pattern
based on the polarity of the delta wave.
 17 Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities 217

Sinus
node
Left
bundle
branch
AV
node
Right
bundle
branch
Block Block

V1 V6 V1 V6

Figure 17.18. Bundle Branch Block. Left, location of right bundle branch block with characteristic V1 and V6 scalar electrocardiographic (ECG)
changes with rSR′ and terminal sagging of S wave. Right, Location of left bundle branch block with characteristic scalar ECG changes in leads V1 and
V6, deep S waves, and broad, notched wide QRS. AV indicates atrioventricular. (Previously published. See “Credit Lines” section.)

Note: Consider items 14m, 14o, and 14t. Note: Consider items 14m, 14o, and 14t.
b. Low voltage, limb and precordial leads c. Left axis deviation (>−30°)
• Amplitude of the entire QRS complex (R + S) is <10 mm in each • Axis −30° to −90°
precordial lead • Be careful about diagnosing left axis deviation in the presence of
• Amplitude of R + S in limb leads is <5 mm an inferior infarct
• Low voltage can also occur with obesity, pleural effusion, restrictive
or infiltrative cardiomyopathies, and diffuse myocardial disease Note: Consider item 7c.

I aVR V1 V4

II aVL V2 V5

III aVF V3
V6

II

Figure 17.19. Complete Heart Block With Right Bundle Branch Block and Left Anterior Fascicular Block.
218 III Electrophysiology

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

Figure 17.20. Left Posterior Fascicular Block and Right Bundle Branch Block in a 68-Year-Old Man.

d. Right axis deviation (>+100°) • Cornell criteria

• Axis +101° to +270°
• Pure right axis deviation (left posterior fascicular block) should
have an S1 Q3 pattern
• Causes include the following:
° Right ventricular hypertrophy (especially if axis >+100°);
consider item 10c
° Vertical heart
° Chronic lung disease; consider item 14m
° Pulmonary embolus; consider item 14n
° Left posterior fascicular block; consider item 7d
e. Electrical alternans (Figure 17.25)
• Alternation of amplitude of the P, QRS, or T waves
• Causes include the following:
° Pericardial effusion causes only a third of cases. But if elec-
trical alternans involves the P wave, QRS complex, and T
wave, significant effusion is usually present; 12% of simple
pericardial effusions have electrical alternans
° Severe left ventricular failure
° Hypertension
° Coronary artery disease
° Rheumatic heart disease
° Supraventricular or ventricular tachycardia
10. Ventricular Hypertrophy
a. Left ventricular hypertrophy by voltage only
° R wave in lead aVL and S wave in lead V3
° 28 mm in males
° >20 mm in females
• Precordial leads
° The sum of the R wave in lead V5 or V6 and the S wave in lead
V1 is >35 mm in adults older than 30 years (40 mm in those
20–30 years old and 60 mm in those 16–20 years old), or
° The sum of the maximal R and the deepest S waves in the
precordial leads is >45 mm, or
° The amplitude of the R wave in lead V5 is >26 mm, or
° The amplitude of the R wave in lead V6 is >20 mm
• Limb leads
° The sum of the R wave in lead I and the S wave in lead II is
≥26 mm, or
° The amplitude of the R wave in lead I is ≥14 mm, or
° The amplitude of the S wave in lead aVR is ≥15 mm, or
° The amplitude of the R wave in lead aVF is ≥21 mm, or
° The amplitude of the R wave in lead aVL is ≥12 mm (a highly
specific, if insensitive, finding)
b. Left ventricular hypertrophy both by voltage criteria (refer to
item 10a) and by ST-T segment abnormalities, which include the
following:
• ST-segment depression in any or all of the following leads: I,
aVL, III, aVF, and V4 through V6
• Subtle ST elevation in leads V1 through V3
• Inverted T waves I, aVL, and V4 through V6

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

Figure 17.21. Left Bundle Branch Block With ST Changes of Acute Anterior Injury in a 76-Year-Old Woman. Patient presented with severe
dyspnea.
 17 Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities 219

A
I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

V1

V6

I aVR V1 V4

II aVL V2 V5

III V3 V6
aVF

II

V1

V6

Figure 17.22. Findings in a 78-Year-Old Man Presenting With Severe Chest Pain and Permanent Ventricular Pacemaker. A, Inferior ST elevation.
B, Acute inferior myocardial infarction with complete heart block; pacemaker is off.

• Prominent or inverted U waves
• Nonvoltage-related criteria for left ventricular hypertrophy (often
with or without prominent voltage and ST-T segment changes in
patients with left ventricular hypertrophy)
° Left atrial abnormality
° Left axis deviation
° Nonspecific intraventricular conduction delay
° Delayed intrinsicoid deflection
° Low or absent R waves in lead V1, V2, or V3
° Absent Q waves in left precordial leads
° Abnormal Q waves in inferior leads due to left axis deviation
° Prominent U waves
• Left ventricular hypertrophy should not be diagnosed in the
presence of left axis deviation due to left anterior fascicular
block when the only criterion for the hypertrophy is in lead aVL
(Figure 17.26)
c. Right ventricular hypertrophy—suggested by one or more of the
following:
• Right axis deviation ≥+100°
220 III Electrophysiology

SAN

LCx

His
bundle LAD

RCA

AVN

Post div
of LBB

PDA

RBB
Ant div
of LBB

Figure 17.23. The Conduction System and Its Blood Supply. Ant div indicates anterior division; AVN, atrioventricular node; LAD, left anterior
descending coronary artery; LBB, left bundle branch; LCx, left circumflex coronary artery; PDA, posterior descending branch of right coronary
artery; Post div, posterior division; RBB, right bundle branch; RCA, right coronary artery; SAN, sinoatrial node. (Previously published. See “Credit
Lines” section.)

R L

110 ms
Normal P wave

Left atrial enlargement

Right atrial enlargement

Figure 17.24. P-Wave Morphologic Patterns. Top, Normal surface P wave, illustrating right and left atrial activation sequence and normal dura-
tion. Middle, Left atrial enlargement (broken line), illustrating delayed peak, and left atrial activation time, producing a prolonged P-wave duration
and notched P-wave morphology. Bottom, Right atrial enlargement (broken line), illustrating effective combined right and left atrial voltage peaks
occurring at the same time with resulting tall peaked P waves. (Previously published. See “Credit Lines” section.)
 17 Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities 221

I aVR V1 V4
Low voltage Electrical alternans

II aVL V2
V5

III aVF V3 V6

Figure 17.25. Electrical Alternans in a 51-Year-Old Woman With Pericardial Tamponade. Note the low QRS voltage in the limb leads and the
varying height of the QRS complex on consecutive beats. The RR interval and complex morphologic patterns are constant throughout, differentiating
electrical alternans from a bigeminal rhythm.

• R/S ratio in lead V1 or V3R >1 • Right axis deviation of initial vector (unblocked forces)
• R wave in lead V1 + S wave in lead V5 or V6 >10.5 mm
d. Combined ventricular hypertrophy (Figure 17.28)—suggested by
• Right atrial enlargement
any of the following:
• R wave in lead V1 ≥7 mm
• ECG meets one or more diagnostic criteria for both isolated left
• qR in lead V1
and right ventricular hypertrophy
• R/S ratio in lead V5 or V6 ≤1
• Precordial leads show left ventricular hypertrophy, but QRS axis
• rSR′ in lead V1 with R′ >10 mm
in frontal plane >+90°
• Secondary ST-T segment changes in right precordial leads
• R>Q in lead aVR, S >R in lead V5, and T inversion in lead V1 in
To diagnose right ventricular hypertrophy, exclude the following: conjunction with signs of left ventricular hypertrophy
• Kutz-Wachtel phenomenon—high-voltage equiphasic (R = S)
• Posterior wall myocardial infarction complexes in mid-precordial leads
• Right bundle branch block • Right atrial enlargement with left ventricular hypertrophy pattern
• Wolff-Parkinson-White syndrome (type A) in precordial leads
• Dextroposition
• Left posterior fascicular block (lateral wall infarct)
• Normal variant (especially in children)
11. Transmural Myocardial Infarction
Criteria to diagnose right ventricular hypertrophy in the presence
General considerations include the following:
of RBBB are the following (Figure 17.27):
• Acute MI: Q waves and ST elevation with or without reciprocal
• r′ ≥15 mm in lead V1 ST depression

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

Figure 17.26. Left Axis Deviation With Increased Voltage in Lead aVL but Not Left Ventricular Hypertrophy. The left axis deviation is due to
left anterior fascicular block.
222 III Electrophysiology

aVR
I V4
V1

II aVL V5
V2

III V6
aVF V3

Figure 17.27. Right Bundle Branch Block With Right Ventricular Hypertrophy (Right Axis Deviation, r′≥15 mm in Lead V1).

I aVR V1 V4

II aVL V2

V5

III aVF V3

V6

Figure 17.28. Combined Ventricular Hypertrophy. Both right ventricular hypertrophy (right axis deviation and amplitude of R wave >S wave in
lead V1) and left ventricular hypertrophy (R wave amplitude >21 mm in lead aVF) are present.
 17 Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities 223

• Recent MI: Q waves, isoelectric ST segments, and ischemic T • Q waves in leads II, III, and aVF
waves • ST-segment elevation, often with reciprocal ST depression in
• Old MI: Q waves, isoelectric ST segments, nonspecific T-wave leads I, aVL, V1, and V2
abnormalities, or normal ST and T waves
f. Posterior infarction, probably acute or recent (Figure 17.34 and
• Significant Q waves
17.35)
° Duration of Q wave ≥0.04 second • Initial R wave in leads V1 or V2 ≥0.04 second with R ≥S and
° Amplitude varies according to region of infarct ST-segment depression and upright T wave in anterior precordial
• ST elevation can persist 48 hours to 4 weeks after MI; >1 month
leads
suggests aneurysm
• Usually associated with inferior MI
• T-wave inversions may persist indefinitely
• Watch for pseudoinfarctions, such as the following: g. Anterolateral infarction, probably old or age indeterminate
• See above, no ST elevation
Condition Pseudoinfarct h. Anterior infarction, probably old or age indeterminate
• See above, no ST elevation
Wolff-Parkinson-White pattern Inferior, anteroseptal, posterior
(Figure 17.29) i. Anteroseptal infarction, probably old or age indeterminate
Hypertrophic cardiomyopathy Inferior, posterior, lateral, anteroseptal • See above, no ST elevation
LBBB Anteroseptal, anterior, inferior j. Lateral or high-lateral infarction, probably old or age indeterminate
Left ventricular hypertrophy Anteroseptal, anterior, inferior, lateral • See above, no ST elevation
(Figure 17.30)
Left anterior fascicular block Inferior, anterior, lateral k. Inferior (diaphragmatic) infarction, probably old or age
Chronic lung disease or right Inferior, posterior, anteroseptal, indeterminate
ventricular hypertrophy anterior • See above, no ST elevation
l. Posterior infarction, probably old or age indeterminate
• A Q wave may be present intermittently in lead aVF in the • See above, no ST changes
absence of MI as a result of respiratory effects and low voltage
inferiorly (Figure 17.31) m. Probable ventricular aneurysm
• In RBBB, Q-wave criteria apply for all infarcts (Figure 17.32) • Persistent ST-segment elevation of ≥1 mm in one or more leads
• It is difficult to diagnose any infarct in the presence of LBBB with associated Q waves; timing must be ≥1 month after infarc-
(Figure 17.21 and 17.22) tion to make this diagnosis on the basis of the ECG

a. Anterolateral infarction, probably acute or recent

• Abnormal Q waves in leads V4, V5, and V6 12. ST-, T-, and U-Wave Abnormalities
• ST-segment elevation
a. Normal variant, early repolarization
b. Anterior infarction, probably acute or recent • Some degree of ST-segment elevation is present in most young,
• rS in lead V1 followed by QS or QR in leads V2 through V4 healthy individuals, especially in the precordial leads—suggested
• ST-segment elevation by the following:
c. Anteroseptal infarction, probably acute or recent (Figure 17.32) ° Elevated takeoff of ST segment at J junction with QRS
• Q or QS deflection in leads V1 through V3 and sometimes in ° Distinct notch or slur on downstroke of R wave
lead V4 ° Upward concavity of ST segment
• Q in lead V1 helps distinguish anteroseptal from anterior infarction ° Symmetrically limbed T waves, which are often of large
amplitude
• ST-segment elevation
° Most commonly involves leads V2 through V5, rarely lead V6
d. Lateral or high lateral infarction, probably acute or recent (Figure • Sometimes also seen in leads II, III, and aVF
17.33) • No reciprocal changes
• Q waves in leads I and aVL
b. Normal variant, juvenile T waves—suggested by the following:
• ST-segment elevation
• Persistence of negative T wave in leads V1 through V3
e. Inferior (diaphragmatic) infarction, probably acute or recent • Most frequent in young, healthy females

I aVR V1 V4

II
aVL V2 V5

III aVF
V3 V6

Figure 17.29. Pseudo–inferior-posterior Myocardial Infarction Due to Wolff-Parkinson-White Pattern.

224 III Electrophysiology

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

Figure 17.30. Left Ventricular Hypertrophy Simulating an Anteroseptal Myocardial Infarction.

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

Figure 17.31. A-C Electrocardiograms (ECGs) in a 73-Year-Old Man. A, asymptomatic, preoperative ECG on July 9. B, Asymptomatic, last
prior ECG on January 14, 6 months before noncardiac operation. C, Asymptomatic, postoperative ECG on July 10. D, Inferior myocardial infarction:
Q-wave inconsistency on ECG. (See “Credit Lines” section.)
 17 Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities 225

I V4
aVR V1

II aVL V2 V5

III aVF V3 V6

II

D
Q-Wave Inconsistency:
1. Disappearance or reduction of Q
waves in aVF on 2 consecutive daily
ECGs: present on day 1, absent on
day 2
2. Occurred in 33% of 167 patients
3. Due to respiratory effects on axis shift
and low-voltage QRS inferiorly

Figure 17.31. (Continued)

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

B
I
aVR V1 V4

II
aVL V2 V5
III

aVF V3 V6

II

Figure 17.32. Electrocardiograms in a 62-Year-Old Man. A, Anteroseptal myocardial infarction. B, Anteroseptal myocardial infarction in the
presence of right bundle branch block.
226 III Electrophysiology

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

Figure 17.33. Acute High Lateral Wall Myocardial Infarction in an 80-Year-Old Woman. Electrocardiogram was interpreted as normal by a com-
puter electrocardiographic reading program.

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

Figure 17.34. Electrocardiograms in a 71-Year-Old Man. A, With severe chest pain. B, Acute inferior-posterior myocardial infarction. ST changes
in leads V1 through V3 represent posterior injury. C, Evolving inferior-posterior myocardial infarction 1 day after presentation. D, Inferior-posterior
infarction 6 months after presentation.
 17 Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities 227

I
aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

Figure 17.34. (Continued)

• Usually not symmetrical or deep
• T waves still upright in left precordial leads I and II
c. Nonspecific ST- or T-wave abnormalities—suggested by any of the
following:
• Slight ST-segment depression or elevation
• T wave flat, low, or slightly inverted; T wave normally should be
≥0.5 mm in leads I and II
d. ST- or T-wave abnormalities suggesting myocardial ischemia—
ischemic T-wave changes:
• Abnormally tall, symmetrical, upright T waves; QT usually pro-
longed, and there may be reciprocal changes
Differential diagnosis:
° Hyperkalemia—tall, peaked (tented), symmetrical (QT nor-
mal or short)
° Intracranial bleeding—QT long, prominent U waves
° Normal variant
• Symmetrically or deeply inverted T waves
Differential diagnosis:
° Giant T inversion from Stokes-Adams attack
° Post-tachycardia T-wave inversion
° Apical hypertrophic cardiomyopathy
° Post-extrasystolic or pacemaker T-wave inversion
° Central nervous system disease (eg, intracranial hemorrhage)
• Pseudonormalization of T waves during exercise
Ischemic ST-segment changes are the following:
• Horizontal or downsloping ST-segment depression with or with-
out T-wave inversion
• Prinzmetal angina typically manifests as ST-segment elevation
without Q waves
e. ST- or T-wave abnormalities suggesting myocardial injury
• Acute ST-segment elevation with upward convexity in the leads
representing the area of infarction
• Reciprocal ST-segment depression in the opposite leads
• Acute posterior injury often has horizontal or downsloping
ST-segment depression with upright T waves in leads V1 through
V3, with or without a prominent R wave in these same leads
f. ST- or T-wave abnormalities suggesting acute pericarditis (Figure
17.36)—4 stages:
• Stage 1: ST-segment elevation (upwardly concave) in almost all
leads except aVR; no reciprocal changes
• Stage 2: ST junction (J point) returns to the baseline and the
T-wave amplitude begins to decrease
• Stage 3: T waves are inverted
• Stage 4: ECG resolution
Other clues include the following:
• PR-segment depression early
• Low-voltage QRS with pericardial effusion (consider item 14o)
• Electrical alternans (consider items 9e and 14o)

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

Figure 17.35. Acute Inferior-Posterior-Lateral Myocardial Infarction.

228 III Electrophysiology

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

Figure 17.36. Acute Pericarditis With PR-Segment Depression (arrow).

• Sinus tachycardia • RBBB

• Regional pericarditis after acute MI (Figure 17.37 and 17.38)
° T waves typically invert by 48 hours and slowly return to nor-
mal over several days to weeks
° Abnormal T-wave evolution due to regional pericarditis is
characterized by either persistently positive T waves or early
reversal of typical T-wave inversion
Note: Also consider item 14p.
g. ST-segment or T-wave abnormalities due to intraventricular conduc-
tion disturbance or hypertrophy
• Left ventricular hypertrophy
° ST-segment depression with upward convexity and T-wave
inversion in left precordial leads
° Reciprocal changes in right precordial leads
° In limb leads, ST-segment and T-wave vectors are usually
opposite the QRS vector
• Right ventricular hypertrophy: ST-segment depression and
T-wave inversion in right precordial leads (V1 through V3) and
sometimes in inferior leads (II, III, and aVF) LBBB: ST-segment
and T-wave displacement are opposite the main QRS deflection
° Uncomplicated RBBB has little ST displacement
° T-wave vector is opposite of the terminal of the slowed por-
tion of QRS: upright in leads I, V5, and V6 and inverted in
right precordial leads
h. Post-extrasystolic T-wave abnormality—any alteration in contour,
amplitude, or direction of T wave in sinus beat(s) after ectopic beat(s)
i. Isolated J-point depression
• Most frequent in exercise testing
• In normals, ST segment should be within 1 mm of the isoelectric
line by 0.08 second after the J point
j. Peaked T waves—require either of the following:
• T wave >6 mm in limb leads
• T wave >10 mm in precordial leads
Differential diagnosis:
° Acute MI. Also see item 12k; consider item 12e
° Normal variant; most commonly mid-precordial leads; may
be >10 mm
° Hyperkalemia; QT normal; consider item 14e

V1 V2 V3 V4 V5 V6

1/9

1/12

Figure 17.37. Regional Pericarditis After Acute Myocardial Infarction. b indicates acute infarction; f, persistent ST elevation in leads V2 to V4.
 17 Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities
V1 V2 V3
Day 1
Day 3
Day 4
Day 9
Figure 17.38. Persistently Positive T Waves (Leads V2 Through V6) Due to Regional Pericarditis After Acute Myocardial Infarction.
° Intracranial bleeding; QT prolonged; prominent U waves;
consider item 14r
k. Prolonged QT interval (Figure 17.39)
• QT interval varies inversely with heart rate
• Measure lead with a large T wave and distinct termination
• Corrected QT interval (QTc) = QT ÷ the square root of the RR
interval (normal, <0.44 second in males and <0.46 second in
females)
Easier methods include the following:
° Use 0.40 second as the normal QT interval for heart rate of 70
beats per minute. For every 10-beats per minute change in heart
rate from 70 beats per minute, adjust by 0.02 second appropri-
ately. Measured value should be within 0.07 second of the calcu-
lated normal. (Example: For a heart rate of 100 beats per minute,
the calculated “normal” QT interval would be 0.34; for a heart
rate of 50 beats per minute, the calculated “normal” QT interval
would be 0.44.)
° Should be less than half the RR interval
Note: Also consider items 14c, 14d, 14h, 14s, and 14u.
l. Prominent U waves
• Largest in leads V2 and V3
• Normally 5%-25% of T wave
• Considered large when amplitude is ≥1.5 mm
13. Pacemaker Function and Rhythm
a. Atrial or coronary sinus pacing
• Pacemaker stimulus followed by atrial depolarization
b. Ventricular demand pacing
229
V4 V5 V6
• Pacemaker stimulus followed by a QRS complex of different
morphology than intrinsic QRS
• Must demonstrate inhibition of pacemaker output in response to
intrinsic QRS
c. AV sequential pacing
• Atrial followed by ventricular pacing
• Could be DVI, DDD, DDI, or DOO pacing mode
d. Ventricular pacing, fixed rate (asynchronous)
• Ventricular pacing with no demonstrable output inhibition by
intrinsic QRS complexes
e. Dual-chamber, atrial-sensing pacemaker
• DDD and possibly VAT or VDD
• For atrial sensing, need to demonstrate inhibition of atrial output
or triggering of ventricular stimulus in response to intrinsic atrial
depolarization
f. Pacemaker malfunction, not constantly capturing (atrium or ven-
tricle) (Figure 17.40)
• Failure of pacemaker stimulus to be followed by depolarization
• Rule out pseudomalfunction (eg, pacer stimulus falling into
refractory period)
g. Pacemaker malfunction, not constantly sensing (atrium or ventricle)
(Figure 17.41)
• For pacemakers in inhibited mode, it is failure of pacemaker to
be inhibited by an appropriate intrinsic depolarization
• For pacemakers in triggered mode, it is failure of pacemaker to
be triggered by an appropriate intrinsic depolarization
• Watch for pseudomalfunction
230 III Electrophysiology

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

V1

V5

Figure 17.39. Complete Heart Block, Right Bundle Branch Block, QT Prolongation, and Polymorphic Ventricular Tachycardia in a 76-Year-Old
Woman With Acute Myocardial Infarction.

* * *

Figure 17.40. VVI Pacing. Failure to capture (first asterisk), normal capture (second asterisk), and functional noncapture (third asterisk) occur
because of pacing artifact during ventricular refractoriness. The pacemaker fails to sense the native QRS complexes. All asterisks identify pacemaker
spikes. (Previously published. See “Credit Lines” section.)

* * * * *

Figure 17.41. VVI Pacing With Undersensing. Result is potentially undesirable competitive ventricular stimulation (first and fourth asterisks). All
asterisks identify pacemaker spikes. (Previously published. See “Credit Lines” section.)
 17 Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities
Premature depolarizations may not be sensed if the following
are true:
° They fall within the programmed refractory period of the
pacemaker
° They have insufficient amplitude at the sensing electrode site
Note: Any stimulus falling within the QRS complex probably
does not represent sensing malfunction. Common with right
ventricular electrodes in RBBB.
h. Pacemaker malfunction, not firing (Figure 17.42)
• Failure of appropriate pacemaker output
i. Pacemaker malfunction, slowing
• An increase in stimulus intervals over the programmed intervals
• Usually an indicator of end of battery life
• Often noted first during magnet application
14. Suggested or Probable Clinical Disorders
a. Digitalis effect—suggested by the following:
• Sagging ST-segment depression with upward concavity
• Decreased T-wave amplitude; T wave may be biphasic
• QT shortening
• Increased U-wave amplitude
• Lengthened PR interval
In left or right ventricular hypertrophy or bundle branch block,
ST changes are difficult to interpret, but if typical sagging ST
segments are present and QT is shortened, consider digitalis
effect.
b. Digitalis toxicity (Figure 17.43)
• Almost any type of cardiac arrhythmia resulting from either a
disturbance in impulse formation or an impairment of conduc-
tion, except bundle branch block
• Typical abnormalities include the following:
° Paroxysmal atrial tachycardia with block
° Atrial fibrillation with complete heart block
° Bidirectional tachycardia
° Second- or third-degree AV block with digitalis effect
° Complete heart block with accelerated junctional or idioven-
tricular rhythm
c. Antiarrhythmic drug effect—suggested by the following:
• Decrease in the amplitude of the T wave or T-wave inversion
• ST-segment depression
• Prominent U waves—one of the earliest findings
• Prolongation of the QTc interval
• Notching and widening of the P waves
• Decrease in the atrial flutter rate
* *
Figure 17.42. Oversensing. Result is an inappropriate, irregular pacemaker bradycardia. It is impossible to tell what is being oversensed in this
electrocardiographic tracing. (Previously published. See “Credit Lines” section.)
231
d. Antiarrhythmic drug toxicity—suggested by the following:
• Widening of the QRS
• Various degrees of AV block
• Ventricular arrhythmias—torsades de pointes
• Marked sinus bradycardia, sinus arrest, or sinoatrial block
e. Hyperkalemia (Figure 17.44)
• Potassium value 5.5–7.5 mEq/L
° Reversible left anterior fascicular block or left posterior fas-
cicular block
° Tall, peaked, narrow-based T waves
• Potassium value >7.5–10.0 mEq/L
° First-degree AV block
° Flattening and widening of P waves, later disappearance of
P waves (sinoventricular conduction) or sinus arrest
° ST-segment depression
• Potassium value >10.0 mEq/L
° LBBB, RBBB, markedly widened, diffuse intraventricular
conduction delay
° Ventricular tachycardia or fibrillation, idioventricular rhythm
f. Hypokalemia—suggested by the following:
• Prominent U waves
• ST-segment depression, decreased T-wave amplitude
• Increase in amplitude and duration of the P wave
• Cardiac arrhythmias and AV block may be digitalis-related
g. Hypercalcemia
• Major ECG change is shortened QTc
• Little effect on QRS, P, and T waves; may see PR-interval
prolongation
h. Hypocalcemia
• Earliest and most common finding is prolongation of QTc; results
from ST-segment prolongation
• ST-segment prolongation occurs without changing the duration
of the T waves; only hypothermia and hypocalcemia do this
• There can be flattening, peaking, or inversion of T waves
i. Atrial septal defect, secundum (Figure 17.45)—suggested by the
following:
• Typical RSR′ or rSR′ in lead V1 with duration <0.11 second;
right ventricular conduction delay in 90% (most are incomplete
RBBB)
• Right axis deviation due to right ventricular hypertrophy
• Right atrial enlargement in 36%
• PR interval prolonged in <20%
• Presence of a sharp deflection (termed crochetage) within
the QRS complex in 1 or more of the inferior leads II, III,
and aVF
* * *
232 III Electrophysiology

V4
I aVR V1

II aVL V2 V5

III aVF V3
V6

Figure 17.43. Paroxysmal Atrial Tachycardia With Atrioventricular Block Due to Digitalis Toxicity in a 79-Year-Old Man.

j. Atrial septal defect, primum (Figure 17.46)—suggested by the
following:
• Most have left axis deviation (in contradistinction to right axis
deviation in secundum atrial septal defect)
• PR-interval prolongation in 15%-40%
• Far-advanced cases have biventricular hypertrophy
k. Dextrocardia, mirror image (Figure 17.47)—suggested by the following:
• Decreasing R-wave amplitude from leads V1 to V6
• The P, QRS, and T waves in leads I and aVL are inverted, or
“upside down”
• Be wary of lead malposition producing similar findings in leads
I and aVL but not in leads V1 through V6

A V4

I aVR V1

II aVL V2
V5

III aVF V3 V6

II

V1

V6

Figure 17.44. Effect of Hyperkalemia. A, Unresponsive 35-year-old man with insulin-dependent diabetes and hyperkalemia (potassium,
7.4 mEq/L). B, and C, Asymptomatic 66-year-old man with hyperkalemia (potassium: B, 6.0 mEq/L; C, 9.2 mEq/L).
 17 Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities 233

B
I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

V1

V6

C
I aVR V1 V4

V2
II aVL V5

V3
III aVF V6

II

V1

V6

Figure 17.44. (Continued)

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

Figure 17.45. Right Axis Deviation and RSR′ in Lead V1 With Duration Less Than 0.11 Second in a 6-Year-Old Girl With Secundum Atrial
Septal Defect.
234 III Electrophysiology

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

Figure 17.46. Left Axis Deviation and rSR′ in Lead V1 With Duration Less Than 0.11 Second in a 1-Year-Old Boy With Primum Atrial Septal
Defect.

l. Mitral valve disease n. Acute cor pulmonale, including pulmonary embolus (Figure 17.48)
• Mitral stenosis • ECG abnormalities are frequently transient
° No diagnostic findings • Sinus tachycardia most common
° Combination of right ventricular hypertrophy and left atrial • Findings consistent with right ventricular pressure overload
abnormality is suggestive include the following:
• Mitral valve prolapse—may see any of the following: ° Right atrial abnormality
° Flattened or inverted T waves in leads II, III, and aVF with ° Inverted T waves in leads V1 through V3
or without ST-segment depression (sometimes left precordial ° Right axis deviation
leads); T-wave changes in the right precordial leads can be ° S1 Q3 and S1 Q3 T3 patterns
associated with prolapse of leaflets ° Pseudoinfarct pattern in inferior leads
° Prominent U waves, QT prolongation ° Transient RBBB
m. Chronic lung disease—suggested by any of the following: ° Various supraventricular tachyarrhythmias
• Right ventricular hypertrophy o. Pericardial effusion (Figure 17.25)—suggested by either of the
• Right axis deviation following:
• Right atrial abnormality • Low-voltage QRS (consider items 9a and 9b)
• Shift of transitional zone counterclockwise • Electrical alternans (consider item 9e)
• Low voltage
• Pseudoanteroseptal infarct p. Acute pericarditis
Findings of right ventricular hypertrophy in the setting of chronic • Refer to item 12f for criteria
lung disease: q. Hypertrophic cardiomyopathy (Figure 17.49 and 17.50)—suggested
• Rightward shift of QRS by the following:
• T-wave abnormalities in right precordial leads • Left atrial abnormality common
• ST depression inferiorly • Majority of cases have abnormal QRS
• Transient RBBB ° Left axis deviation in 20%
• rSR′ or QR in lead V1 ° High-voltage QRS

I aVR V1 V4

V2 V5
II aVL

III V6
aVF V3

Figure 17.47. Dextrocardia.

 17 Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities 235

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

Figure 17.48. Electrocardiogram in a 76-Year-Old Man With Massive Pulmonary Embolus.

° Large abnormal Q waves can give pseudo-infarct patterns in ° Prolonged QT interval

inferior, lateral, and precordial leads ° Prominent U waves
° Tall R wave in lead V1 with inverted T waves simulating right • Other changes include the following:
ventricular hypertrophy ° T-wave notching, loss of amplitude
• ST-T wave abnormalities common ° Diffuse ST-segment elevation imitating pericarditis or focal
° ST-T wave changes due to ventricular hypertrophy or conduc- ST-segment elevation imitating acute myocardial injury
tion abnormalities pattern
° Apical variants of hypertrophic obstructive cardiomyopa- ° Abnormal Q waves imitating MI
thy have deep lateral, precordial T-wave inversions (Figure ° Almost any rhythm abnormality
17.50)
r. Central nervous system disorder (Figure 17.51) Differential diagnosis:
• “Classic changes,” usually in precordial leads, include the ° Acute MI
following: ° Acute pericarditis
° Large upright or deeply inverted T waves ° Drug effect

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

Figure 17.49. Electrocardiogram in a 21-Year-Old-Man With Hypertrophic Cardiomyopathy.

236 III Electrophysiology

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

Figure 17.50. Electrocardiogram in a 72-Year-Old Man With Apical Hypertrophic Cardiomyopathy. Note deep symmetrical T-wave inversion
in leads V3 through V6.

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

Figure 17.51. Electrocardiogram in a 65-Year-Old Woman With Acute Subarachnoid Hemorrhage.

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

Figure 17.52. Electrocardiogram in a Patient With Myxedema.

 17 Electrocardiographic Diagnoses: Criteria and Definitions of Abnormalities 237

J wave

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

Figure 17.53. Electrocardiogram in a 63-Year-Old Woman With Hypothermia.

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

B
I aVR V1 V4

II aVL V2
V5

III aVF V3 V6

Figure 17.54. Electrocardiograms From Patients With Ebstein Anomaly. A, A 25-year-old man. B, A 10-year-old boy.
238 III Electrophysiology

s. Myxedema (Figure 17.52) • Atrial fibrillation with slow ventricular response preceded or fol-
• Low voltage of all complexes lowed by sinus bradycardia, sinus arrest, or sinoatrial block
• Sinus bradycardia • Prolonged sinus node recovery time after atrial premature com-
• Flattening or inversion of the T waves plex or atrial tachyarrhythmias
• PR interval may be prolonged • AV junctional escape rhythm
• Frequently associated with pericardial effusion • Additional conduction system disease is often present
t. Hypothermia (Figure 17.53) v. Ebstein anomaly (Figure 17.54)
• Sinus bradycardia • Characteristic RBBB with abnormal terminal forces (R′ in lead
• PR, QRS, and QT prolongation V1, S in I and aVL)
• J waves that may be quite prominent (Osborne, or “camel-hump”
sign)
• 50%-60% have atrial fibrillation Abbreviations
• Other arrhythmias occur
AV atrioventricular
u. Sick sinus syndrome—frequently manifests as one or more of the ECG electrocardiogram
following: LBBB left bundle branch block
• Sinus bradycardia of marked degree MI myocardial infarction
• Sinus arrest or sinoatrial block RBBB right bundle branch block
• Bradycardia alternating with tachycardia S1,2,3 first, second, third heart sound
 18
Cardiac Cellular Electrophysiology
HON-CHI LEE, MD, PHD
Ion Channels: Structure and Function
Ion channels are integral membrane proteins that regulate the
traffic of ions in heart cells. They are determinants of the car-
diac action potential, which in turn underlies cardiac impulse
conduction, excitation-contraction coupling, automaticity, and
arrhythmogenesis.
In the heart, most of the important ion channels assume 1 of
the following structural motifs, presented in increasing order of
complexity:
1. The Kir channels are proteins with 2 transmembrane segments that
sandwich a channel pore loop (Figure 18.1A). Four of these subunits
assemble to form a functional channel. These channels are not volt-
age-sensitive and are frequently gated by ligands. Examples include
the K ATP channels and the strong inward rectifier K + channel that
gives rise to IK1.
2. The Kv channels are proteins that have 4 subunits consisting of 6
transmembrane segments (S1 through S6) (Figure 18.1B) with
cytoplasmic amino and carboxyl termini. The voltage sensor is
located in S4, which contains a high density of positively charged
amino acids (lysine and arginine) that move according to changes
in membrane potential. This movement causes the channel to open
and close through allosteric and conformational changes in the ion
channel structure. S5 and S6 sandwich a pore loop. Four of these
pore-forming channel protein subunits assemble to form a functional
channel. Examples include the transient outward K + channels and
the delayed rectifier K + channels.
3. Voltage-gated Na + and Ca2+ channel proteins are the most structur-
ally complex. The channel is a single peptide consisting of 4 homolo-
gous domains, each of which has 6 transmembrane segments that
include a voltage-sensing S4 and a pore loop between S5 and S6
(Figure 18.1C).
Abbreviations and acronyms are expanded at the end of this chapter.
• Ion channels in the heart have a 4-fold structural symmetry.
• The ion channel pore is lined by 4 pore loops.
• The voltage-gated channels contain the voltage-sensing S4.
The function of ion channels is defined by 2 basic properties:
conductance and gating.
1. Conductance describes which ions are allowed to permeate the channel
and at what rate. Ion channel conductance is determined by the selectiv-
ity filter in the pore loop that lines the narrowest portion of the channel
pore. Na + channels preferentially conduct Na + over K+ (12:1) and Ca2+
(10:1), whereas K+ channels are selective for K+ over Na + (1,000:1).
Most cardiac ion channels show rectification, which refers to
the preferential conduction of ions in the outward or inward
direction (Figure 18.2A). Ion channel rectification can be influ-
enced by unequal ion concentration across the membrane, the
range of voltages that open voltage-gated ion channels (eg, Ito
and If ), or blockade of the channel by intracellular magnesium or
polyamines at positive voltages (eg, Kir channels).
2. Gating describes what governs the opening and closing of the chan-
nel. Voltage-gated channels open and close according to changes in
membrane potential: INa, ICa,L, ICa,T, Ito, IKr, and IKs are all activated
by membrane depolarization, whereas If is activated by membrane
hyperpolarization. Voltage-gated ion channels all contain a voltage-
sensing S4 in which every third amino acid is a positively charged
lysine or arginine. Ligand-gated channels are activated or inacti-
vated by the binding of chemical ligands: IKACh is activated by ace-
tylcholine, whereas IKATP is inhibited by ATP.
Open channels can be inactivated by different mechanisms,
including the following:
1. Physical occlusion of the channel pore by cytoplasmic portions of
the channel (eg, a “ball-and-chain” mechanism in Ito or a “hinge lid”
mechanism in I Na)
239
240 III Electrophysiology

A B

Extracellular Kv
Extracellular Kir +
+
1 2 3 4 5 6
+
Cytoplasmic +
Cytoplasmic

NH2 NH2
COOH
COOH

C I II III IV Nav/Cav
Extracellular
+ + + +
+ + + + Cellular
1 2 345 6 1 2 345 6 1 2 345 6 1 2 345 6
+ + + + membrane
+ + + +
Cytoplasmic

NH2 COOH

Figure 18.1. Structure of Cardiac Ion Channels. A, Inwardly rectifying potassium (Kir) channels have 2 transmembrane segments and an inter-
vening pore loop. Four of these subunits assemble to form a functional channel. B, Voltage-gated potassium (Kv) channels have 6 transmembrane
segments, including the voltage sensor in segment 4 and the pore loop between segments 5 and 6. Four of these subunits assemble to form a functional
channel. C, Voltage-gated sodium (Nav) and calcium (Cav) channels are channel proteins that consist of single polypeptide chains that contain 4
repeats of 6 transmembrane segments. COOH indicates the carboxyl group of amino acids; NH2, the amino group of amino acids.

2. Conformational changes in channel structure (eg, C-type inactiva-
tion in Ito)
3. Increase in intracellular Ca2+ concentration, which promotes Ca2+ -
calmodulin binding to the C-terminus of the channel (eg, ICa,L)
4. Chemical ligand binding (eg, IKATP)
Some channels, such as the delayed rectifier K+ channels, are
noninactivating. In these channels, K+ conduction stops when
membrane potentials are hyperpolarized, resulting in channel
deactivation (Figure 18.2B).
Ion Channels and the Cardiac Action Potential
Conceptually, the configuration of the cardiac action poten-
tial is determined by the sum of ionic current activity at any
given time point during the cardiac cycle. The upstroke of the
action potential, phase 0, is associated with the opening of the
Na + channels. Inactivation of the Na + currents and activation
of the transient outward K + currents give rise to early rapid
partial repolarization, or phase 1 of the action potential. The
plateau of the action potential, phase 2, reflects the balance
between inactivation of transient outward K + currents, activa-
tion of L-type Ca2+ currents, delayed rectifier K + currents, and
the Na + -Ca 2+ exchange currents. Phase 3 represents the fi nal
rapid repolarization of the action potential and is the result of
further activation of the delayed rectifier K + currents and inac-
tivation of the Ca 2+ currents. Toward the terminal portion of
phase 3, the strong I K1 currents are activated, leading to rapid
repolarization.
During diastole, or phase 4, atrial and ventricular myocytes
are normally electrically quiescent. However, pacemaker tissues
such as the sinus node, AV node, and His-Purkinje fibers show
slow diastolic depolarization, indicating the presence of pace-
maker activity caused by the activation of If and Ca2+ currents
and the inactivation of K+ currents. The rate of phase 4 depo-
larization and spontaneous action potential generation is fastest
in sinus node cells, which form the dominant pacemaker of the
heart. Activities of the different ionic currents during the cardiac
action potential are shown in Figure 18.3.
 18 Cardiac Cellular Electrophysiology 241

1500 – Specific cardiac ion channels are discussed next along with
A
their clinical relevance according to their participation in the dif-

Current (pA)
–
ferent phases of the cardiac action potential.
Linear (ohmic) 1000 –
–
Phase 0
500 –
Sodium Channel
–
0–
Atrial and ventricular myocytes as well as Purkinje fibers are
l
-150
l l
-100
l l
-50
l l l l densely populated with voltage-gated Na + channels. These chan-
–0 50
nels open very briefly (≤1 ms) when the membrane is depolarized
Membrane
-500 – Potential (mV) to −50 mV, producing a rapid influx of Na + with a fast upstroke
– of 100 to 200 V/s in the action potential. The inactivated Na +
channels require repolarization to the normal resting potential of
-1000 –
−90 mV in atrial and ventricular myocytes for complete recov-
ery. The relationship between Na + channel availability and mem-
brane potential is an important determinant of conduction and
B refractoriness. Depolarized resting potentials in injured or ische-
I Cardiac Myocytes mic myocardium prevent complete recovery of Na + channels,
resulting in reduced INa and action potential upstroke velocity. INa
is inactivated by voltage with physical occlusion of the channel
-30 mV
pore by the cytoplasmic linker between domains III and IV in a
“hinge lid” mechanism. Na + channels open very briefly and are
-80 mV rapidly inactivated. However, some channels stay open or reo-
pen and carry a small INaL, which becomes important in disease
Na+ currents
states. Nodal cells are sparsely populated with Na + channels and
have normal resting potentials of −50 to −70 mV; they have few
INa currents and their action potential upstrokes are slow and
Inactivation
dependent on Ca2+ currents.
Activation
INa in Clinical Practice
II • Class I antiarrhythmic drugs are INa blockers.
+100 mV
• Mutations in the cardiac Na + channel gene SCN5A are associated
with the following:
-40 mV
1. LQT3 with defects in I Na inactivation resulting in enhanced late
I NaL opening. Class I antiarrhythmic drugs, such as mexiletine,
Deactivation may be helpful in the treatment of LQT3.
Activation
2. Brugada syndrome with loss-of-function mutations. Class I
K+ currents
antiarrhythmic drugs exacerbate the Brugada phenotype and
may be helpful in diagnosis of the condition.
3. Autosomal recessive congenital sick sinus syndrome with loss-
of-function mutations and reduced cardiac excitability.
Figure 18.2. Electrophysiologic Properties of Cardiac Ion Channels. • In congestive heart failure, INa is reduced but I NaL is increased sec-
A, Current rectification. The current-voltage (IV) relationship of a non- ondary to increased channel phosphorylation due to an increased
rectifying ion channel shows a linear (ie, ohmic) relationship (black sympathetic state.
curve). Most ion channels in the heart exhibit nonlinear IV relation- • In atrial fibrillation, I Na has been reported to be reduced.
ships. The inwardly rectifying potassium family of channels shows
inward rectification (red curve), in which large currents are conducted • In myocardial infarction, I Na is reduced in the peri-infarct zone
at potentials negative to the equilibrium potential for potassium, but with altered channel gating.
very small currents are conducted at positive potentials. Key exam- • Na + channels are associated with other membrane and cytoskeletal
ples are the inwardly rectifying potassium current and the adenosine proteins in macromolecular complexes that regulate the channel
triphosphate–sensitive potassium current. In contrast, outwardly rec- function. Mutations in several of these Na + channel–associated
tifying channels (green curve) conduct mainly repolarizing currents proteins produce important channelopathy syndromes, including
with very small inward currents at negative potentials. Key examples the following:
are the rapidly activating potassium current and the slowly activating 1. Caveolin-3 (LQT9)
potassium current. B, Two models of current activation, inactivation, 2. α1-Syntrophin (LQT12)
and deactivation. I, In response to a depolarizing voltage step, the chan-
nel is activated, producing a rapid inward current. The current subse- 3. β4 Subunit (LQT10)
quently undergoes time-dependent inactivation, returning to baseline 4. β1 And β3 subunits with loss of function (Brugada syndrome)
even when the voltage stimulus is sustained. A key example is a sodium 5. GDP-1L (Brugada syndrome with impaired channel trafficking)
(Na +) current. II, In response to a voltage step, the channel is activated,
producing an outward current. The current is noninactivating and is sus- Gap Junction Channels
tained as long as the voltage stimulus persists. The current is reduced or
deactivated only when the voltage returns to negative potentials. A key In addition to the voltage-gated Na + channels, gap junction chan-
example is a slowly activating potassium (K +) current. nels are major determinants of electrical impulse conduction and
242 III Electrophysiology
Depolarizing Currents
INa
ICa,L
INa-Ca Ex
Ito
ICa,L
IKr,IKs
INa INa
IK1
Ventricle
Repolarizing Currents
IK1
Ito
IKr
IKs
IKur
Figure 18.3. Ionic Currents That Contribute to the Cardiac Action Potentials. Depolarizing (red) and repolarizing (blue) currents contribute
to the action potentials in the ventricle (left), atrium (middle), and nodal tissue (right). ICa,L indicates L-type calcium current; ICa,T, transient-type
calcium current; If, pacemaker current; IK, potassium current; IK1, strong inwardly rectifying potassium current; IKr, rapidly activating potassium
current; IKs, slowly activating potassium current; IKur, ultrarapid delayed rectifier potassium current; I Na, sodium current; INa-Ca Ex, sodium-calcium
exchange current; Ito, transient outward potassium current.
propagation. Gap junction channels are hexomeric complexes of
connexin (connexin 40 and connexin 43) that form hemichan-
nels on the cell surface. Two hemichannels from 2 different
cells are united to form a functional gap junction. Opening and
closing of the gap junction channels that regulate intercellular
communication are modulated by several metabolic factors,
including pH, intracellular Ca2+ , and channel phosphorylation.
Because intercellular transfer of currents can occur only where
cells share gap junctions, the activity and distribution of the
gap junction channels can profoundly affect electrical impulse
conduction, especially under conditions of ischemia. In normal
hearts, most of the connexins are localized at the longitudinal
ends of the cells, promoting impulse propagation in the longitu-
dinal direction.
Gap Junction Channels in Clinical Practice
• Gap junction channels are the target of rotigaptide and AAPs, a new
class of agents that improve cardiac gap junction communication.
• Loss-of-function mutations in connexin 40, the major gap junction
channels in the atrium, are associated with atrial fibrillation.
• Gap junctions undergo cellular redistribution in myocardial infarc-
tion (including the peri-infarct zone), atrial fibrillation, and other
disease states. This leads to increased lateralization and improper
coupling, resulting in loss of conduction.
Phase 1
Transient Outward Potassium Channel
Produced by currents from several channels (including Kv4.3,
Kv4.2, and Kv1.4), Ito is present in atrial, ventricular, and
Zero
ICa,T
ICa,L
If
INa-Ca Ex
Ito
IKur
ICa,L ICa,L
IKr,IKs IK (IKr,IKs)
ICa,T
If
IK1
Atrium Nodal
Zero
IK
conduction system cells. The rapid activation and inactiva-
tion of Ito contributes to phase 1 of the cardiac action potential.
In the ventricular myocardium, Ito is differentially expressed
(robustly in the epicardium and modestly in the endocardial
layers), leading to a transmural gradient that gives rise to J
waves or Osborne waves seen on ECGs in hypothermia. Ito is
mainly inactivated by the physical occlusion of the channel pore
by the N-terminal portion of the channel in a “ball-and-chain”
mechanism.
Ito in Clinical Practice
• Ito is increased in regions of the heart that have relatively short
action potential durations, such as the ventricular epicardium, the
right ventricle, and the septum.
• Ito is increased in hyperthyroidism and reduced in hypothyroidism,
with corresponding changes in the cardiac action potential.
• Ito is dynamically regulated and is downregulated in several disease
states, including atrial fibrillation, myocardial infarction, conges-
tive heart failure, and diabetes mellitus. The reduced Ito leads to
prolonged action potentials and QT intervals. In atrial fibrillation,
however, the atrial action potential is shortened because the down-
regulation of Ito is overcome by changes in other ionic currents.
• Ito is regulated by the activity of its accessory proteins. The follow-
ing are examples:
1. The transmural gradient in KCHIP2 expression contributes to
the transmural heterogeneous Ito activity in the ventricular wall
2. A MiRP2 (KCNE3) gain-of-function mutation enhances Ito and
aggravates the Brugada phenotype
3. A different MiRP2 mutation causes action potential shortening
by increasing Ito and is associated with a familial form of atrial
fibrillation
 18 Cardiac Cellular Electrophysiology
Phase 2
L-Type Calcium Channel
The L-type Ca2+ channels are present in all cell types in the heart.
ICa,L is activated by membrane depolarization but inactivates
much more slowly than INa. In nodal cells, ICa,L is responsible
for impulse generation and conduction. In atrial and ventricular
myocytes, ICa,L is a critical determinant of the action potential
plateau and has a crucial role in cardiac excitation-contraction
coupling. ICa,L is enhanced several-fold by sympathetic stimula-
tion. It is inactivated by depolarized voltage and by increases
in intracellular Ca2+ , which promote Ca2+ -calmodulin binding to
the C-terminus of the channel, inducing channel inactivation by
conformational changes.
ICa,L in Clinical Practice
• Class IV antiarrhythmic drugs are ICa,L blockers.
• Gain-of-function mutations with reduced channel inactivation are
associated with Timothy syndrome, a severe multisystem disorder
that includes prolonged QT (LQT8).
• Loss-of-function mutations are associated with SQTS, Brugada
phenotype, and sudden death syndromes.
• Atrial ICa,L is profoundly downregulated in atrial fibrillation, lead-
ing to shortening of the atrial action potential and refractoriness.
• Ventricular ICa,L is downregulated in congestive heart failure.
• In myocardial infarction, ICa,L is reduced in the infarct border
zone.
Ryanodine Receptors
Ryanodine receptors are important in the development of cardiac
arrhythmias, and their function is tightly coupled to the activities
of ICa,L. Ryanodine receptors are intracellular ion channels located
in the sarcoplasmic reticulum. They are responsible for releasing
Ca2+ from the sarcoplasmic reticulum and are crucial elements in
the regulation of intracellular Ca2+ homeostasis. Opening of the
ryanodine receptors is triggered by Ca2+ entry through the sarco-
lemmal Ca2+ channels, in a process known as Ca2+ -induced Ca2+
release, and is critical in regulating excitation-contraction cou-
pling of the heart. In the normal heart, ryanodine receptors are
closed during diastole. In disease states, however, these channels
become “leaky,” resulting in intracellular Ca2+ overload and the
development of arrhythmias.
Ryanodine Receptors in Clinical Practice
• Mutations are associated with catecholaminergic polymorphic VT,
a condition characterized by exercise- or stress-induced polymor-
phic VT, syncope, and sudden death.
• Ryanodine receptor function is compromised in hyperadrenergic
states, such as congestive heart failure, resulting in increased Ca2+
leak from the sarcoplasmic reticulum and promoting the develop-
ment of arrhythmias.
Phase 3
Delayed Rectifier Potassium Channels
Three types of delayed rectifier K+ currents have been identified:
IKur, IKr, and IKs. The rapidly activating and slowly inactivating
IKur is prominent in atrial myocytes, resulting in their short action
potential duration. IKr and IKs are activated very slowly during
243
the cardiac action potential and are responsible for phase 3 repo-
larization. IKr activates upon depolarization and is rapidly inac-
tivated; however, the channel recovers from inactivation during
early repolarization, contributing to major repolarizing currents
during phases 2 and 3 of the action potential. IKr then deacti-
vates upon phase 3 repolarization. IKs activates very slowly upon
depolarization and is noninactivating; K + conduction stops when
membrane potentials are hyperpolarized, resulting in channel
deactivation. The slow deactivation of IKs contributes to the short
action potential durations at high heart rates and is the basis of
the reverse use dependence of class III antiarrhythmic drugs that
mainly block IKr.
IKur in Clinical Practice
• IKur is mainly expressed in atrial myocardium and contributes to
atrial repolarization.
• Loss-of-function mutations in Kv1.5, which encodes IKur, have been
shown to be associated with familial atrial fibrillation.
• IKur is upregulated in atrial fibrillation, causing shortening of the
atrial action potential.
• Kv1.5 mRNA is downregulated in ischemia and the infarct zone.
IKr in Clinical Practice
• IKr is the target of many class III antiarrhythmic drugs (eg, sotalol,
dofetilide) and is modulated by the off-target effects of many com-
monly used drugs (eg, erythromycin, methadone, chlorpromazine)
to produce torsades de pointes.
• Loss-of-function mutations in IKr lead to LQT2.
• Gain-of-function mutations in IKr are associated with SQTS and
familial atrial fibrillation.
• IKr is downregulated in some disease conditions associated with a
prolonged action potential and QT interval (eg, myocardial infarc-
tion, diabetes mellitus).
• IKr is upregulated in acute myocardial ischemia, leading to shorten-
ing of the action potential and myocardial refractoriness.
• IKr is not changed in atrial fibrillation.
• Mutations in MiRP, the accessory subunit in IKr, are associated
with LQT6.
IKs in Clinical Practice
• IKs is the target of some class III antiarrhythmic drugs, including
amiodarone and azimilide (which also block IKr).
• IKs blockers do not exhibit reverse use dependence, unlike pure IKr
blockers.
• Loss-of-function mutations in IKs are associated with LQT1.
• Gain-of-function mutations of IKs are associated with SQTS and
familial atrial fibrillation.
• IKs is not altered in atrial fibrillation.
• The downregulation of IKs is associated with QT prolongation in
congestive heart failure and the peri-infarct zone in myocardial
infarction.
• Mutations in MinK, the accessory subunit of IKs, are associated
with LQT5.
G Protein–Gated Kir Channel
G protein–gated Kir channels encode IKACh and IKAdo and are
highly expressed in the atria and pacemaker cells of the sinus
node, AV node, and Purkinje fibers but not in the ventricular
244 III Electrophysiology
myocardium. The channels are activated by parasympathetic
stimulation and adenosine. Binding of acetylcholine or adeno-
sine to their respective receptors activates Gi, releasing its βγ
subunit, which in turn activates the channels, shortening action
potentials and producing membrane hyperpolarization.
IKACh and IKAdo in Clinical Practice
• IKACh is constitutively activated in atrial fibrillation. This leads to
shortening of the atrial action potential and decreased refractori-
ness, promoting development of atrial fibrillation.
• Activation of IKACh and IKAdo produces negative chronotropic and
dromotropic effects in sinus and AV nodal cells. Thus, these chan-
nels are therapeutic targets in the treatment of supraventricular
tachycardia.
K ATP Channel
IKATP is a Kir channel that is inhibited by physiologic intracel-
lular concentrations of ATP but is activated during ischemia,
upon depletion of ATP, or with a decrease in the ATP:ADP ratio.
Ventricular myocytes are endowed with high densities of these
channels. IKATP is a crucial element in mediating cardiac ischemic
preconditioning.
IKATP in Clinical Practice
• Sulfonylureas block IKATP, which regulates insulin secretion in the
beta cells in the pancreas.
• Loss-of-function mutations are associated with the following:
1. Familial persistent hyperinsulinemic hypoglycemia of infancy
2. Adrenergic-induced atrial fibrillation in the vein of Marshall
• Gain-of-function mutations are associated with transient neonatal
diabetes and permanent diabetes.
• Activation of IKATP accounts for the ST-segment elevation on ECG
during myocardial infarction.
Late Phase 3
Strong Inward Rectifier K+ Channel
The strong inward rectifier K+ channel IK1 is robust in ventricular
myocytes, weak in atrial myocytes, and absent in nodal cells. I K1
conducts very little outward current at depolarized membrane
potentials. It is crucial for maintaining the resting potential near
the K+ reversal potential of about −90 mV and is responsible for
the rapid terminal repolarization in phase 3. Damaged myocar-
dium with weakened IK1 is susceptible to the development of
abnormal automaticity.
IK1 in Clinical Practice
• Loss-of-function mutations are associated with Andersen-Tawil
syndrome and LQT7.
• Gain-of-function mutations are associated with SQTS and familial
atrial fibrillation.
• IK1 is downregulated in several disease states, such as congestive
heart failure, myocardial ischemia, and myocardial infarction.
• This promotes the development of arrhythmia through enhanced
automaticity and triggered activity.
Phase 4
Phase 4 is electrically quiescent in normal atrial and ventricu-
lar myocytes but is characterized by the presence of pacemaker
currents in the sinus node, AV node, and Purkinje fibers. Phase 4
diastolic depolarization occurs owing to inactivation of K+ cur-
rents and activation of pacemaker currents. The most important
pacemaker currents are contributed by If and ICa,T.
Hyperpolarization-Activated Cyclic
Nucleotide Gated Channel
If is activated by membrane hyperpolarization. It is a nonselec-
tive cationic channel that is responsible for pacemaker activity
and diastolic depolarization during phase 4 in the sinus node,
the AV node, and the Purkinje fibers. Activity of If is tightly
regulated by sympathetic activation and parasympathetic
inhibition.
If in Clinical Practice
• If is the target of ivabradine, a new bradycardia drug for treatment
of inappropriate sinus tachycardia.
• Mutations in If have been found to be associated with idiopathic
sinus bradycardia.
• If is enhanced in the following:
1. Atrial myocytes in atrial fibrillation
2. Ventricular myocytes in congestive heart failure
T-Type Calcium Channel
The T-type Ca2+ channels are more highly expressed in the atrial
myocardium, the conduction system, and nodal cells than in ven-
tricular myocytes. ICa,T is activated at hyperpolarized potentials
(positive to −70 mV) and is rapidly inactivated (−80 to −50 mV).
ICa,T is small in magnitude compared with ICa,L and is negligible
in ventricular cells.
ICa,T in Clinical Practice
• ICa,T is thought to be responsible for impulse generation in nodal
cells (the last two-thirds of phase 4 depolarization) and automatic-
ity in atrial myocytes and pulmonary veins.
Electrogenic Membrane Pumps and Exchangers
The Na + -K + pump is ubiquitous and is inhibited by digitalis. The
pump extrudes 3 Na + ions in exchange for the entry of 2 K + ions,
with hydrolysis of ATP; hence, the pump is electrogenic and con-
tributes a hyperpolarizing outward current to the cardiac action
potential. The Na + -Ca2+ exchange facilitates the exchange of 1
Ca2+ ion for 3 Na + ions. The direction of exchange depends on the
membrane potential. At the resting potential, Ca2+ is extruded,
allowing the cell to maintain low intracellular Ca2+ concentration.
At the plateau, the exchanger facilitates Ca2+ entry and contrib-
utes to cardiac excitation-contraction coupling. The exchanger is
electrogenic and is thought to mediate the development of DADs
and triggered activity.
• The major currents activated during different phases of the cardiac
action potential include the following:
1. Phase 0: Na + currents (atria, ventricles, and Purkinje fibers) and
Ca2+ currents (sinus and AV nodes)
2. Phase 1: Transient outward K + currents, Ito
3. Phase 2: Ca2+ currents, Na + -Ca2+ exchange currents, IKur, and IKr
4. Phase 3: Delayed rectifier K + currents, IKr and IKs, IK1 (late
phase 3)
5. Phase 4: Pacemaker currents, If and ICa,T
 18 Cardiac Cellular Electrophysiology 245

Regional Variations in Cardiac Action Potentials
Various types of cardiac tissues have different ion channel com-
positions and thus, different configurations of action potentials
(Figure 18.4).
Sinus Node
The sinus node is characterized by its phase 4 depolarization,
which gives rise to pacemaker activities. Phase 4 depolarization
is due to the high density of If and the lack of IK1, which also
accounts for the relatively depolarized state of the tissue (rest-
ing membrane potential at −50 to −65 mV). Na + channels are
sparse and the action potential upstroke is slow since it is medi-
ated mainly by ICa,L. There is no discernible phase 1 owing to the
lack of Ito. Action potential durations are short and the frequency
of depolarization is determined by the sympathetic and parasym-
pathetic modulation of If and ICa,L.
Atria
The atrial action potential has rapid upstrokes, allowing rapid
electrical impulse conduction from the right atrium to the left
atrium and from the sinus node to the AV node. It has a discernible
phase 1, followed by a short plateau phase and rapid repolariza-
tion. The short plateau probably results from the presence of IKur
and IKACh, as well as an enhanced Ito, and may partially explain
why an antiarrhythmic drug such as lidocaine, which exerts its
effects during the plateau phase, is ineffective in the treatment of
atrial arrhythmias. Normal atrial tissue has no phase 4 activity
owing to the presence of IK1.
AV Node
The AV node is similar to the sinus node in its lack of I Na and IK1.
Conduction through the AV node is mediated by ICa,L and prop-
agation is slow, accounting for its decremental conduction prop-
erties. ICa,L is significantly activated by sympathetic stimulation
and inhibited by parasympathetic influences; these are important
determinants of impulse conduction through the AV node. Phase
4 depolarization in the AV node is usually not as prominent as
in the sinus node.
His-Purkinje Fibers
His-Purkinje fibers have a high density of INa, which facilitates
the rapid conduction of impulses so that ventricular myocytes

Sinus node

Atria

AV node

His-Purkinje

Endocardium

Midmyocardium Ventricles

Epicardium

ECG

Figure 18.4. Action Potential Waveforms in Different Tissues in the Heart. Action potential configurations from various tissues of the heart
vary according to their specific roles in impulse generation, conduction, and contraction. The sinus and atrioventricular (AV) nodes are important
in impulse generation and have pacemaker activity. Atrial and ventricular myocardium are important for contraction. The His-Purkinje tissue has
the fastest upstroke velocity for rapid conduction of electrical impulse to activate the ventricles synchronously. The ventricular epicardium, endocar-
dium, and midmyocardium have distinct action potential configurations. The contributions of the action potentials from various tissues to the surface
electrocardiographic (ECG) signals are displayed.
246 III Electrophysiology
can be activated synchronously. In addition, the His-Purkinje
fibers have strong IK1 and weak pacemaker currents. Hence,
His-Purkinje tissue is characterized by a resting potential of
−90 mV (close to the reversal potential of K+) and a slow dias-
tolic depolarization that can provide tertiary pacemaker activity
when sinus and AV nodes fail.
Ventricles
In ventricular myocytes, the densities of I Na and IK1 are higher
than in atrial myocytes; thus, these cells rest near −90 mV and
are electrically quiescent. Configuration of the action potential
varies according to location in the left ventricle. Myocytes in the
epicardial layer have very strong Ito. This leads to marked repo-
larization in phase 1, followed by depolarization as Ca2+ currents
are activated, generating the characteristic “spike-and-dome”
configuration. In contrast, the endocardial layer has a much lower
Ito density, with reduced phase 1 amplitude and no spike-and-
dome. The midmyocardial layer is endowed with the so-called
M cells, which have strong Ito but weak delayed rectifier K+ cur-
rents and an enhanced slow component of the voltage-gated Na +
currents. Thus, the M cell action potential is characterized by
a spike-and-dome configuration and lengthened action potential
duration, exceeding those of epicardial and endocardial myo-
cytes. This regional heterogeneity in the electrophysiology of the
heart is thought to be the basis for the development of U waves
and triggered activity.
• Be able to correlate action potential configurations with various
cardiac tissues.
Normal Cardiac Rhythm Generation
and Conduction
Proper cardiac function relies on the coordinated regulation of
cardiac rhythm generation and conduction. Cardiac cells can
spontaneously generate action potentials (automaticity). The
hierarchy of automaticity, in descending order, is sinus node, AV
node, and His-Purkinje fibers. The function of the sinus node
can be evaluated in electrophysiologic studies by measuring the
sinus node recovery time (normal <1,500 ms) and the corrected
sinus node recovery time by subtracting the baseline sinus inter-
val from the sinus node recovery time (normal <550 ms). Sinus
node automaticity can also be assessed by measuring the intrinsic
heart rate in the presence of simultaneous β-adrenergic blockade
(propranolol 0.2 mg/kg) and muscarinic cholinergic blockade
(atropine 0.04 mg/kg).
• Failure in rhythm generation results in bradyarrhythmias.
• Enhanced automaticity may be associated with an increase in heart
rate with conditions such as inappropriate sinus tachycardia, accel-
erated junctional rhythm, and accelerated idioventricular rhythm.
• Abnormal automaticity and triggered automaticity results in
tachyarrhythmias.
Normal cardiac rhythm conduction proceeds from the sinus
node in an organized sequence to the atrial myocardium, AV node,
His-Purkinje fibers, and ventricular myocardium. Conduction
time in various cardiac structures can be evaluated by measuring
the PA interval (normal, 20–50 ms; atrial conduction), AH inter-
val (normal, 60-120 ms; AV nodal conduction), and HV interval
(normal, 35–55 ms; His-Purkinje conduction). In addition, the
QRS duration is a measure of intraventricular conduction.
Normal rhythm conduction is determined by tissue excit-
ability and refractoriness. The response of cardiac tissue to an
applied current defines its excitability. Since excitatory and depo-
larizing currents are inactivated after activation, and since reac-
tivation is required before the tissue becomes excitable again,
recovery of excitability is a fundamental determinant of rhythm
conduction. Excitability in the sinus and AV nodes is dependent
on the recovery of ICa,L, whereas excitability in atrial, ventricu-
lar, and His-Purkinje tissues is dependent on the recovery of INa.
Approximately 60% of the Na + channels in atrial and ventricular
myocardium must recover before tissue excitability can occur.
In depolarized tissue, such as myocardium during ischemia,
recovery of INa is incomplete, thereby slowing conduction.
There is a brief period at the end of repolarization in phase 3
when I Na has recovered sufficiently and the membrane potential
is closer to the threshold for action potential generation than it
is at the resting membrane potential. In this period, excitability
is enhanced, resulting in supernormal excitability. Supernormal
excitability underlies the vulnerable period of the cardiac cycle
and may contribute to a susceptibility toward the development
of arrhythmias. Refractoriness is the resistance to reexcita-
tion after a previous electrical activation. In electrophysiologic
studies, tissue refractoriness can be measured as the effective
refractory period, which is defined as the longest paired S1-S2
coupling interval that fails to excite and conduct through the
tissue.
• Conduction time in various cardiac structures can be evaluated by
measuring the PA interval (normal, 20–50 ms; atrial conduction),
AH interval (normal, 60–120 ms; AV nodal conduction), and HV
interval (normal, 35–55 ms; His-Purkinje conduction) during the
electrophysiology study.
• Failure of rhythm conduction results in various types of heart
blocks and bradyarrhythmias, including sinoatrial exit blocks, AV
nodal conduction blocks, and bundle branch blocks.
• Development of a unidirectional block may lead to the develop-
ment of reentry and tachyarrhythmias.
Mechanisms of Arrhythmogenesis: Abnormal
Rhythm Generation
Abnormal Automaticity
Abnormal automaticity refers to the spontaneous develop-
ment of impulses that are independent of preceding impulses
or stimulation. The underlying mechanism results from abnor-
mal diastolic depolarization in atrial or ventricular muscle
with tissue damage or from abnormal enhancement of pace-
maker activities in subsidiary tissues such as nodal cells or
His-Purkinje fibers.
• Important automatic tachyarrhythmias include ectopic atrial tachy-
cardias, pulmonary vein potentials in atrial fibrillation, accelerated
junctional tachycardias, and certain ventricular tachycardias that
are catecholamine dependent.
• Automatic tachycardias are usually not inducible by programmed
electrical stimulation during electrophysiologic studies. They are
characterized by the following:
1. Rate acceleration at the onset (warming up) and deceleration
before termination (cooling down)
2. Responsiveness to sympathomimetics and autonomic modula-
tion (Figure 18.5A)
Triggered Activity
Triggered activity is the development of abnormal impulses as a
result of the preceding impulse or impulses. Although triggered
 18 Cardiac Cellular Electrophysiology 247

A
Automaticity
↑ Sympathetic
Normal stimulation
myocardial
action
potential

↑ Diastolic Diastolic
depolarization depolarization

B
Triggered Activity-EAD
Normal AP ↑ APD EAD

↑ QT

ECG

C
Triggered Activity-DAD
Rapid pacing 1s
1s

0 mV 0 mV

Sustained
DADs triggered
activity
−50 mV

−100 mV
Normal

D
Reentry Area of slow Unidirectional
conduction block Reentry

Figure 18.5. Mechanisms of Cardiac Arrhythmias. A, Automaticity. Left, Automaticity is normally absent in atrial and ventricular myocardium.
Right, In injured or depolarized tissue, abnormal automaticity can occur with diastolic depolarization (red). Automaticity usually is augmented by
sympathetic or adrenergic stimulation, which enhances the rate of diastolic depolarization (green). B, Triggered activity: early afterdepolarization
(EAD). Left, A normal action potential (AP) is associated with a normal QT interval. Middle, With an increase in the action potential duration (APD)
(for causes, see Table 18.1), QT is prolonged. Right, With further prolongation of the AP, a depolarization occurs in late phase 2 as an EAD. Note that
the depolarization occurs before complete repolarization of the AP. C, Triggered activity: delayed afterdepolarization (DAD). Left, Normal tissue
shows no afterdepolarizations. Middle, In cardiac tissue, under conditions of intracellular calcium overload, rapid pacing induces DADs (red). Note
that the depolarizations occur after complete repolarization of the AP. Right, Further calcium overload results in sustained triggered activity (red).
D, Reentry. Left, For a reentry arrhythmia to occur, the presence of a substrate that contains more than 1 functionally distinct conduction pathway
and an area of slow conduction is required. Middle, Reentry is initiated by a unidirectional block. Right, The electrical impulse conducts through
the area of slow conduction in an adjacent pathway and returns in the reverse direction to the first pathway when the area of block recovers, allowing
reentry to occur.

activity is involved in only a small proportion of arrhythmias, Early Afterdepolarizations

its frequent association with life-threatening conditions makes it
important. Unless a proper diagnosis is made and the underlying
conditions that generate the triggered activity are removed, the
ensuing arrhythmias are potentially lethal. There are 2 types of
triggered activities: EADs and DADs.
EADs are depolarizations that occur during phase 2 or phase 3
of the cardiac action potential before complete repolarization
occurs (Figure 18.5B). Therefore, conditions associated with
markedly prolonged action potential durations and QT intervals
promote the development of EADs. Conditions that prolong QT
248 III Electrophysiology

intervals usually have heterogeneous regional effects, creating
dispersion of refractoriness. This together with the development
of EADs is thought to be the mechanism that underlies torsades
de pointes (Table 18.1).
• Torsades de pointes is characterized by the following:
1. Prolonged QT intervals
2. Exacerbation by bradycardia (which prolongs QT intervals and
increases dispersion of refractoriness)
3. Short-long coupling intervals
4. “Salvos” of nonsustained polymorphic VTs before degeneration
into sustained polymorphic VT and ventricular fibrillation
5. Polymorphic VT with characteristic “twisting around the axis”
morphology
Delayed Afterdepolarizations
DADs are depolarizations that occur after the action potential
has completely repolarized (Figure 18.5C). The mechanism that
underlies the development of DADs is intracellular Ca2+ over-
load. DADs are thought to be associated with conditions such as
digitalis toxicity, ischemic reperfusion arrhythmias, and ryano-
dine receptor dysfunction.
• Arrhythmias involving DADs are characterized by the following:
1. Intracellular Ca2+ overload
2. Exacerbation by tachycardia (which increases intracellular Ca2+)
3. Enhancement by sympathomimetics
Triggered activity is usually not inducible by programmed
electrical stimulation. Burst pacing may be able to induce DADs
and triggered activity, but current recording facilities in electro-
physiologic laboratories cannot reliably record or differentiate
between EADs and DADs.
Table 18.1. Conditions Known to Promote EADs and
Torsades de Pointes
Mechanisms of Arrhythmogenesis: Abnormal
Rhythm Conduction
Reentry
Reentry is the most common mechanism involved in many clin-
ically important cardiac arrhythmias, including AV nodal reen-
try tachycardia, AV reentry tachycardia using an AV accessory
connection, atrial flutter, and VT in hearts with infarct scars.
Conduction abnormalities that lead to reentrant arrhythmias in
the heart can occur when the following conditions are satisfied:
1. At least 2 functionally distinct conduction pathways are present
2. There is unidirectional block in 1 pathway
3. There is slow conduction down the second pathway with retrograde
conduction through the first pathway that was initially blocked
(Figure 18.5D)
Reentry arises as a result of altered rhythm conduction in car-
diac tissue that has more than 1 conduction pathway, each with
distinct electrophysiologic properties, and that contains an area
of slow conduction in the reentry circuit. The premature beat that
sets up reentry is blocked unidirectionally in a pathway that has
a relatively long refractory period, allowing the impulse to con-
duct down a second pathway that has an area of slow conduc-
tion. The impulse returns in the reverse direction to excite the
first pathway, which has recovered from refractoriness, to cre-
ate a reentry circuit. Reentry may involve anatomically defined
circuits such as in typical atrial flutter, AV nodal reentry tachy-
cardia, and AV reciprocating tachycardia in Wolff-Parkinson-
White syndrome. However, reentry can also occur when a fixed
anatomical substrate is absent but functional conduction blocks
are present. Functional reentry is dynamic, as exemplified by
Table 18.2. Mechanisms of Common Supraventricular and
Ventricular Tachyarrhythmias
Arrhythmia Mechanism

Condition Mechanism Sinus tachycardia Automatic: normal

Inappropriate sinus tachycardia Automatic: enhanced or abnormal
Electrolytes Sinoatrial reentry tachycardia Reentry
Hypokalemia ↓ K + channel activity Atrial tachycardia Automatic: ectopic atrial tachycardia
Hypomagnesemia Possibly ↑ Ca2+ channel activity Triggered: DAD with digitalis toxicity
Metabolic Reentry: intra-atrial reentry
Hypoxia ↑ I NaL Atrial fibrillation Reentry: multiple wavelets
↑ ICa,L (↑ sensitivity to Triggered: pulmonary vein potentials
β-adrenergic stimulation) Atrial flutter Reentry
Acidosis ↓ K + channel activity SVT: AV nodal reentry Reentry
Autonomic SVT: orthodromic and Reentry
Increased sympathetic tone, ↑ ICa,L, ↑ I NaL, ↑ spatial dispersion antidromic tachycardia
cathecholamines of repolarization (WPW and concealed bypass
Drugs tract)
Antiarrhythmic drugs Accelerated junctional Automatic: abnormal
Class IA K + channel blockade tachycardia (nonparoxysmal)
Class III K + channel blockade Accelerated idioventricular Automatic: abnormal
Sotalol, dofetilide, ibutilide Na + channel activation, K + rhythm
channel blockade VT Reentry: peri-infarct
Antihistamines K + channel blockade Automatic: catecholamine dependent
Macrolide antibiotics K + channel blockade Triggered: EAD, torsades de pointes
Congenital long QT syndrome Triggered: DAD, digitalis toxicity
Channelopathies (I Na, ICa,L, IKr, IKs, IK1) Prolonged action potentials VT: bundle branch reentry Reentry
VT: RVOT Possibly triggered: DAD

Abbreviations: Ca 2+ , calcium; ICa,L, L-type calcium current; IK1, strong inwardly Abbreviations: AV, atrioventricular; DAD, delayed afterdepolarization;
rectifying potassium current; IKr, rapidly activating potassium current; EAD, early afterdepolarization; RVOT, right ventricular outflow tract; SVT,
IKs, slowly activating potassium current; I Na, sodium current; I NaL, late sodium supraventricular tachycardia; VT, ventricular tachycardia; WPW, Wolff-
current; K + , potassium. Parkinson-White syndrome.
 18 Cardiac Cellular Electrophysiology 249

multiple reentry wavelets in atrial fibrillation and ventricular Ca2+ calcium

reentry during ischemia. DAD delayed afterdepolarization
To better understand reentry, it is important to consider the EAD early afterdepolarization
wavelength of the tachycardia, which is the minimum circuit size ECG electrocardiogram
GDP guanosine diphosphate
required for perpetuation of a reentrant tachycardia. The wave-
Gi inhibitory G protein
length of the reentry impulse is defined by the product of the con- HV His-ventricular
duction velocity and the refractoriness. Conditions that shorten ICa,L L-type calcium current
the wavelength by decreasing conduction velocity or decreasing ICa,T T-type calcium current
refractoriness (or both) promote reentry. These properties are If pacemaker (“funny”) current
important in determining the effectiveness of antiarrhythmic IK1 strong inwardly rectifying potassium current
drugs and overdrive pacing in the termination of reentry arrhyth- IKACh acetylcholine-sensitive potassium current
mias. An excitable gap exists when the tachycardia circuit is IKAdo adenosine-induced inwardly rectifying potassium current
longer than the tachycardia wavelength. The presence of an excit- IKATP adenosine triphosphate–sensitive potassium current
able gap allows electrical stimulations to invade the reentrant cir- IKr rapidly activating potassium current
IKs slowly activating potassium current
cuit and thereby to reset or terminate the tachycardia.
IKur ultrarapid delayed rectifier potassium current
• During electrophysiologic studies, reentrant arrhythmias usually INa sodium current
exhibit the following properties: INaL late sodium current
Ito transient outward potassium current
1. They are inducible by programmed electrical stimulation K+ potassium
2. They have an abrupt onset and offset K ATP adenosine triphosphate–sensitive potassium
3. During reentrant tachycardias, the R-R intervals are regular KCHIP2 potassium channel interacting protein 2
4. They can be reset and entrained by pacing Kir inwardly rectifying potassium
Kv voltage-gated potassium
Common cardiac arrhythmias and their underlying mecha- LQT long QT syndrome (with specific conditions numbered:
nisms are listed in Table 18.2. LQT1, LQT2, etc)
MiRP MinK-related protein
Na + sodium
Abbreviations PA P wave–atrial
AAP antiarrhythmic peptide S transmembrane segment
ADP adenosine diphosphate S1 fixed pacing train
AH atrial-His S2 premature stimulus
ATP adenosine triphosphate SQTS short QT syndrome
AV atrioventricular VT ventricular tachycardia
 19
Indications for Invasive and Noninvasive
Electrophysiologic Testing
BARRY A. BOILSON, MD, and PETER A. BRADY, MB, CHB, MD
Cardiac EP testing encompasses invasive and noninvasive
strategies to assess cardiac electrical function for diagnostic or
therapeutic purposes.
Invasive EP Testing
Invasive EP testing is a definitive means of evaluating the elec-
trical properties of the heart and of establishing the mechanism
and locating the origin of arrhythmias. The indications for inva-
sive cardiac EP study have been outlined by the American Heart
Association/American College of Cardiology task force for prac-
tice guidelines and are summarized in Box 19.1.
Goals of EP Testing
The goals of EP testing are as follows:
• Accurate diagnosis of an arrhythmia and its mechanisms.
• Establishment of the etiologic factors of syncope.
• Determination of prognosis in the clinical setting of arrhythmia.
• Risk stratification for sudden cardiac death.
• Acquisition of data regarding indications for ICD or pacemaker
implantation.
• Guidance of antiarrhythmic therapy with drugs, ablation, or surgery.
Invasive EP testing is not recommended in certain circumstances
(Box 19.2).
Principles and Methodology
of Invasive EP Testing
Invasive EP study is generally reserved for the evaluation of
syncope where a cardiac cause is considered likely, where an
Abbreviations and acronyms are expanded at the end of this chapter.
250
arrhythmic cause has been uncovered and further evaluation of
mechanism and prognosis is required, or where both situations
exist. In the present era, discovery of the precise mechanism of
arrhythmia frequently offers the possibility of definitive cure
through RF ablation.
The maneuvers performed during an EP study vary depend-
ing on the indication. However, the usual core procedures per-
formed are summarized as follows:
1. Conscious sedation is administered by anesthesiology staff.
2. Peripheral vascular access is obtained.
3. ECG leads and remote defibrillation pads are placed on the patient.
4. Central venous access is obtained using the Seldinger technique, via
the femoral venous routes and also usually via the internal jugular,
subclavian, or branchial veins to facilitate placement of a multipolar
catheter in the coronary sinus.
5. Electrode catheters are placed at the high right atrial position, across
the tricuspid annulus to permit a potential to be recorded at the bun-
dle of His, at the right ventricular apex, and within the coronary
sinus (Figure 19.1).
Baseline intracardiac intervals are recorded with measure-
ments of conduction through the atrium (PA interval), AV node
(AH interval), and the HPS (HV interval) (Figure 19.2).
Next, PES is performed, with the specific protocol used,
depending on whether tachyarrhythmia or bradyarrhythmia is
suspected or previously known. An overview of the core proto-
cols used by most laboratories follows.
PES for Suspected Bradyarrhythmia
Typically, where symptomatic bradycardia or cardiac pauses
are apparent, further EP testing is not necessary and appropri-
ate treatment with permanent pacing is instituted as needed.
Occasionally, bradyarrhythmia may be suspected but as yet
 19 Indications for Invasive and Noninvasive Electrophysiologic Testing
Box 19.1. Indications for Invasive Electrophysiologic
Study
Diagnosis and management of bradyarrhythmias
To evaluate sinus node function where sinus node
dysfunction is suspected
To define of the level of AV conduction abnormality
where HPS disease is suspected but unproved
Diagnosis and management of tachyarrhythmias
To define the mechanism of narrow complex
tachycardia where drug therapy is not tolerated or
ineffective
To define the mechanism of broad complex
tachycardia
Syncope of unknown cause in structural
or organic heart disease
To evaluate sinus and AV node function
To assess inducibility of VT
Miscellaneous
To evaluate survivors of unexplained cardiac arrest
To select optimal ICD parameters for therapy in
patients with VT/VF
Rarely, to assess antiarrhythmic drug response
not documented on testing or the findings have been borderline
and difficult to attribute as the cause of the patient’s symptoms.
Because a preliminary assessment of intracardiac conduction has
already been performed at this stage by measurement of base-
line intervals, more specific testing of sinus node and AV nodal
function is required.
Testing Sinus Node Function
These tests comprise assessments of abnormalities of conduc-
tion and automaticity by measurement of SACT and SNRT. In
practice, SACT is rarely assessed because it represents an indi-
rect measurement. The SNRT is the most common method of
assessing sinus node function. It requires the induction of sup-
pression of sinus node activity by pacing at a site close to the
sinoatrial node and then measuring the time for return of sinus
node activity (ie, SNRT). This interval varies according to the
sinus node (SCL) and therefore is often expressed as the CSNRT
value: CSNRT = SNRT − SCL.
Box 19.2. Contraindications to Invasive Electrophy-
siologic Study
Acute myocardial infarction
Unstable angina
Acute decompensated heart failure
Bacteremia or septicemia
Major bleeding diathesis
Uncooperativeness of the patient
251
Testing AV Node Function
When patients have symptomatic spontaneous second- and third-
degree AV block, further evaluation is usually not necessary and
permanent pacing is appropriate. However, occasionally, high-
degree AV block may be discovered in the absence of symptoms
or a patient may have symptoms of syncope or presyncope where
bradyarrhythmia is suspected but AV block has not been found.
In these cases, invasive evaluation of AV node function may be
helpful, primarily to assess risk and the need for permanent pac-
ing. The indications for invasive EP testing to evaluate AV nodal
function are shown in Box 19.3.
The baseline intervals recorded at the outset of the EP study
yield preliminary information on whether AV block exists,
and analysis of the intracardiac activation sequence clarifies at
which level. The presence of first-degree heart block on the sur-
face ECG is typically associated with the finding of AH interval
prolongation.
Wenckebach (or Mobitz I) second-degree AV block usually
results from block within the AV node and thus is typically mani-
fested by progressive prolongation of the AH interval and constant
HV interval during a run of Wenckebach rhythm, until an atrial
electrogram is not followed by a His bundle or ventricular elec-
trogram on the His catheter recording channel (Figure 19.2).
This situation is in contrast to the more worrisome finding
where a His bundle electrogram follows the atrial electrogram
but no ventricular electrogram follows (Figure 19.3). This find-
ing is suggestive of block, and thus disease, within the HPS. The
latter is typically found in the case of Mobitz II–type, second-
degree AV block and may progress to third-degree AV block, in
which the pattern described is persistent.
Third-degree or complete AV block may occur at any level
in the AV conduction system. The level of block determines the
nature of the escape rhythm. Third-degree AV block developing
within the AV node results in a junctional escape rhythm, where
each ventricular electrogram is preceded by a His bundle electro-
gram. However, complete AV block developing in the HPS results
in a ventricular escape rhythm and an HV dissociation is noted,
with a His bundle electrogram following each atrial electrogram.
The importance of making these distinctions is that the level of
third-degree AV block has an important bearing on the poten-
tial for hemodynamic tolerance of the resulting escape rhythm.
Block within the AV node and the associated junctional escape
rhythm may lessen with the enhanced AV node conduction seen
on increased catecholamine challenge (eg, during exercise) and
therefore may result in increased heart rate and resolution of
block. Hence, more favorable hemodynamics occur with activity.
However, where AV block is infra-Hisian, the opposite is the case
because a block in the HPS worsens at increased heart rates.
An invasive EP study is helpful in determining the site of 2:1
block when the site is uncertain. However, helpful surface ECG
clues are that block within the AV node is suggested by PR inter-
val prolongation before block; in contrast, a normal PR interval
before block suggests block in the HPS. In addition, block in the
HPS tends to occur with acceleration in the sinus rate, unlike
block in the AV node, which is more apparent at slower sinus
rates. At the time of invasive EP study, intracardiac electrograms
on the His channel usually confirm the level of 2:1 block.
It is important to determine the location of block because phys-
iologic second- or third-degree block in the AV node due to high
vagal tone is not usually an indication for pacing. However, con-
firmed infra-Hisian block in the clinical setting of second- or third-
degree block or symptoms of syncope, or both, warrant pacing.
252 III Electrophysiology

RAO LAO

Figure 19.1. Catheter Placement for Invasive Electrophysiologic Study. In this case, ventricular tachycardia arising from the right ventricular
outflow tract (RVOT) was identified after programmed stimulation and activation mapping. An intracardiac echocardiography catheter has also been
placed in the right atrium, as well as an Abl in the RVOT. Abl indicates ablation catheter; CS, coronary sinus; ICE, intracardiac ultrasound catheter;
His, His bundle; HRA, high right atrium; LAO, left anterior oblique; RAO, right anterior oblique; RV, right ventricle.

Figure 19.2. Measuring Baseline Intracardiac Intervals. A long PR interval is explained in this case by prolongation in AP interval and also pro-
longation in the AH interval. The HV interval is normal, rendering His-Purkinje disease unlikely. The effect of antiarrhythmic drugs is likely with
prolongation of resting intra-atrial and AV nodal conduction times. The second atrial activation sequence seen is premature and blocks the AV node
(no His potential, blocked PAC).
 19 Indications for Invasive and Noninvasive Electrophysiologic Testing 253

PES for Suspected Tachyarrhythmia

Box 19.3. Indications for EP Testing to Evaluate AV
Nodal Function
High-grade AV block (second or third degree)
without clear symptom correlation
To determine the level of block; AV nodal or in the HPS
2:1 AV block where block in the AV node cannot
be confirmed on the surface ECG
To determine the level of block; AV nodal, within the
His bundle (intra-Hisian), or below the His bundle
(infra-Hisian)
Atypical Wenckebach periodicity
PR prolongation not clear on surface ECG
To confirm AV nodal block vs block in the HPS
Exclude pseudo-AV block in all of the above
Confirmed by evidence of His extrasystoles that
block to the atrium and ventricle, and the presence
of nonconducted atrial impulses
The presence of HPS disease by itself is associated with myo-
cardial dysfunction and increased risk of sudden cardiac death.
Therefore, the finding of HPS disease may warrant further eval-
uation with a ventricular stimulation study. The findings on
invasive EP study that suggest increased risk of progression to
high-grade AV block are summarized in Box 19.4.
Where tachyarrhythmia is suspected or confirmed, an invasive
EP study can assist in confirming the precise mechanism of
arrhythmia, and subsequent curative RF ablation frequently can
be performed during the same session.
Specific Testing in Narrow-QRS Tachycardia
Almost all narrow-QRS tachycardia is supraventricular in nature,
and EP study is generally not warranted unless RF ablation is
being contemplated. Because SVT is typically not life-threat-
ening (unless in the clinical setting of structural heart disease
or an accessory pathway), the usual indication for proceeding
to RF ablation is incapacitating symptoms. An association of
tachycardia with symptoms of syncope or presyncope, angina, or
congestive heart failure also is an indication for proceeding with
EP study, particularly if the QRS is wide during tachycardia and
there is a need to exclude VT (Box 19.5).
During EP study, baseline intervals are recorded as described
herein, and then PES is performed from the high right atrial cath-
eter and, occasionally, from the coronary sinus. During atrial
extrastimulus testing, important observations may be made that
may suggest the presence of dual AV nodal physiology (and poten-
tial for AV nodal reentrant tachycardia) or of an accessory path-
way. The SVT frequently is induced by atrial pacing. Additional
observations—including the intra-atrial activation sequence,
means of onset, and cessation—and the effect of bundle branch
block are frequently helpful in clarifying the diagnosis.
Further maneuvers may be undertaken to clarify the mechan-
ism of tachycardia. Usually, multiple maneuvers are necessary to

A
Lead l

V1

HRA

AH=140
HBE 1

HV=140

B
Lead ll

V1

HRA
A

HBE 1
A H A H A H

Figure 19.3. Significant Infranodal Conduction System Disease. A, After procainamide injection (50 mg/min; maximal dose, 15 mg/kg), the HV
interval increased further (140 ms). B, This response was followed by spontaneous infra-Hisian block with no ventricular escape rhythm, which was
associated with complete loss of consciousness. Temporary pacing was required for 5 to 10 min before atrioventricular conduction was reestablished.
254 III Electrophysiology
Box 19.4. Findings on Invasive Study of the AV Node
Suggestive of High Risk of Progression to Complete AV
Block
Prolonged HV interval
HV prolongation beyond 55 ms in the presence of
symptoms of syncope/presyncope
Asymptomatic patient with HV prolongation beyond
100 ms
Abnormal prolongation of the HV interval with
procainamide challenge (doubling or increase
beyond 100 ms)
Infra-Hisian block
Development of infra-Hisian block at atrial rates less
than 150 bpm
Prolonged HPS ERP
Block occurring distal to the His bundle at an atrial
extrastimulus cycle length >450 ms (ERP) in the
presence of symptoms of syncope/presyncope or
other risk factors for progression to high-grade AV
block
Failure of the HPS ERP to shorten with higher atrial
rates
provide an accurate diagnosis. These include the introduction of
atrial or ventricular extrastimuli during tachycardia and, occa-
sionally, para-Hisian pacing.
Specific Testing in Broad-QRS Tachycardia
Invasive EP study is indicated when VT is suspected in a person
with previous syncope or presyncope or severely symptomatic
palpitations and also where VT has been confirmed but further
clarification of the mechanism is required to facilitate pharmaco-
logic or RF ablation therapy (Box 19.6).
PES for induction of VT is performed via a quadripolar
electrode placed in the right ventricle, and a second multipo-
lar catheter is placed at the His bundle position to record atrial
and His bundle activity. Ventricular extrastimulus testing is per-
formed with a drive train at a fixed cycle length and the addition
of single, double, and triple extrastimuli. The entire process is
repeated at a shorter drive train and at least one other location
in the right ventricle (usually apex and outflow tract). Typically,
extrastimulus coupling intervals are kept above the 180-ms level,
Box 19.5. Indications for EP Study for Narrow-QRS
Tachycardia
Severe symptoms or syncope in association
with arrhythmia
Preexcitation or short PR interval on surface ECG
Heart failure or angina even with relatively slow
rates
VT needs to be excluded
Failed pharmacological therapy and anticipated
ablation
Box 19.6. Indications for EP Study for Broad-QRS
Tachycardia
Severe symptoms or syncope and to investigate
VT as a mechanism
Known VT and determination of the mechanism to
guide therapy
To assess potential for ablation
To guide ICD and drug therapy
Assessment of VT inducibility and risk assessment
in special circumstances
Survivors of cardiac arrest where cause is unknown
Late-onset VT/VF post-MI (after 48 h)
because below this level, VT and ventricular fibrillation can be
induced in the normal heart and thus are a nonspecific finding.
Administration of isoproterenol hydrochloride also increases
inducibility of ventricular arrhythmias but at the expense of the
specificity of the findings.
When VT is induced, a 12-lead ECG is recorded immediately
to assess morphologic characteristics, axis, and cycle length. A
His bundle recording is essential during VT in order to diagnose
bundle branch reentry where present. Entrainment techniques
are helpful to prove a reentrant mechanism if VT is sustained and
well tolerated hemodynamically. If VT is poorly tolerated hemo-
dynamically, synchronized direct current (DC) cardioversion is
performed; otherwise, pace termination should be attempted. VT
terminated easily with timed ventricular extrastimuli suggests
macroreentry with a large excitable gap. Burst pacing and ramp
pacing may also be attempted, the results of which are helpful
in determining optimal ICD therapies where implantation is
planned.
Risks of Invasive EP Testing
The decision to proceed with invasive EP testing should include
consideration of risks to the patient versus potential yield, and
informed consent is essential (Box 19.7). When RF ablation is
undertaken, specific risks must be considered. Slow pathway abla-
tion for AV nodal reentrant tachycardia carries an approximately
1% risk of AV nodal damage requiring permanent pacemaker
placement, but for RF ablation of septal accessory pathways, this
risk approaches 3%. The risk of life-threatening complications,
such as myocardial infarction or stroke, is up to 0.5% overall and
is highest in patients with structural heart disease and in elderly
patients. The risk is higher with left-sided procedures, but risk
is minimized through a meticulous approach to anticoagulation,
prevention of air embolism, and clarification of the coronary
anatomy by angiography where indicated.
Summary: Appropriate Application
of Invasive EP Testing
• Invasive EP study is appropriate to further evaluate the mechanism
of suspected or confirmed bradyarrhythmia or tachyarrhythmia in
the clinical setting of known or suspected organic heart disease.
• Where there are no indications of an arrhythmic cause for symp-
toms and a low suspicion for organic heart disease, the findings
may be nonspecific and misleading.
 19 Indications for Invasive and Noninvasive Electrophysiologic Testing 255

• Where bradyarrhythmia is known or suspected, EP study permits

Box 19.7. Risks of Invasive EP Testing confirmation of sinus node disease or AV node disease.
Vascular complications (1%-5%) • Where AV node disease is documented, the level of block and the
degree of risk to the patient without pacing may be determined.
Bruising/hematoma (up to 5%)
Venous thrombosis, arterial occlusion (<1%) • Where tachyarrhythmia is known or suspected, the nature and
mechanism may be determined and treatment options defined—in
AV fistula formation (<1%) particular, whether ablation may be considered.
Pneumothorax and cardiac tamponade (<1%) • In some circumstances, the inducibility of VT may be helpful to
guide ICD implantation and therapies.
AV nodal injury requiring pacemaker
implantation • Patients with severe organic heart disease, decompensated heart
failure, or acute myocardial infarction should generally not be con-
Slow pathway ablation (∼1%) sidered for invasive testing.
Septal right-sided AV accessory pathway (2%–3%)
Arrhythmia induction requiring cardioversion Noninvasive EP Testing
Atrial fibrillation (∼10%) with programmed atrial Recommendations on indications, protocols, and end points in
stimulation noninvasive EP testing have been reviewed by the European
VT/VF with programmed ventricular stimulation
Society of Cardiology and were published in 2004. The evaluation
of the patient with suspected syncope (ie, loss of consciousness
Systemic thromboembolism (up to 0.5%) due to global cerebral hypoperfusion) requires a standard and
Transseptal puncture thorough methodological approach (Figure 19.4) that begins with
Left atrial ablation
a focused clinical history and physical examination, including
evaluation for bruits, asymmetry of peripheral pulses, and lying
Coronary artery trauma (up to 5%) and standing blood pressures, and an ECG. The history occasion-
Left-sided ablation ally makes it clear that loss of consciousness is not truly present
Radiation-induced injury or, if so, has a cause other than global cerebral hypoperfusion (eg,
carotid or vertebrobasilar ischemia, seizures, metabolic causes).
Skin burns (∼1 in 10,000 cases) In these instances, further evaluation and treatment should be
Malignancy (∼1 in 1,000 per h of fluoroscopy) directed appropriately. However, where the history indicates that
syncope has occurred, the physical examination (eg, orthostatic

Transient loss of consciousness

Clinical history and physical examination, lying and standing blood pressure, ECG

Syncope Nonsyncope

Diagnosis Suspected Diagnosis

certain diagnosis unknown

• Vasovagal confirmed Suspected Suspected • Falls

neurally Frequent Rare/single
by history cardiac • Epilepsy
mediated episodes episode
cause
• Situational confirmed • TIAs
by history • Metabolic causes
• Orthostatic confirmed • Holter monitoring/ • Tilt test • Psychogenic
No further
by physical event monitor • Carotid sinus pseudosyncope
examination evaluation
• Loop recorder massage
• Ischemic (cardiac) • Exercise stress • Loop recorder
confirmed by ECG test
• Arrhythmic confirmed • Invasive EP
by ECG testing
Appropriate
referral and
Treatment Reevaluation treatment treatment

Figure 19.4. Algorithm for Use of Noninvasive EP Testing. ECG indicates electrocardiography; EP, electrophysiology; TIA, transient ischemic
attack.
256 III Electrophysiology

Box 19.8. Indications for HUT Box 19.9. Relative Contraindications to HUT
Recurrent episodes of syncope in the absence of Proximal coronary artery stenosis
organic heart disease Critical mitral stenosis

Recurrent episodes of syncope in the presence of Clinically severe left ventricular outflow
organic heart disease where cardiac causes of obstruction
syncope have been excluded Known severe cerebrovascular stenosis
Unexplained single syncopal episode in high-risk
settings (eg, occupational hazard)
Older patients with unexplained falls The response to HUT varies by age. Younger patients are more
likely to exhibit a bradycardic response, whereas older patients
are more likely to demonstrate hypotension.
The sensitivity of HUT is up to 80%, depending on the proto-
hypotension) or ECG (Mobitz II–type or third-degree AV block,
col population studied, but this high sensitivity is at the expense
overt ischemic changes, or SVT/VT) may yield an almost certain
of a lowered specificity (ie, more false-positive results). Use of
diagnosis. Again, in these instances, appropriate specific testing
isoproterenol with HUT may increase the rate of false-positive
and treatment should be instituted. In other cases where syncope
response to up to 45% or more in patients without a history of
is confirmed by history, additional clues may be present in the
syncope.
history, examination, or ECG that may allow further evaluation
The absence of a gold standard test for diagnosis of neuro-
to be more specific and expeditious.
cardiogenic syncope makes the false-negative rate difficult to
The principal indications for noninvasive EP testing (HUT)
determine, but it has been estimated as up to 30%. Thus, a neg-
are outlined in Box 19.8.
ative HUT response cannot be considered to definitively exclude
Testing protocols vary, but patients typically are kept in a
neurocardiogenic syncope.
supine position for 20 to 45 minutes, followed by progressive
upright tilting with continuous monitoring of heart rate, blood
pressure, ECG, and, occasionally, pulse oximetry. Some investi- Summary: Appropriate Application of HUT
gators additionally monitor cerebral perfusion using transcranial
Doppler ultrasonography in supine horizontal position, during • Noninvasive EP testing is most useful in patients for whom the sus-
picion of organic heart disease is low or has been excluded.
and after HUT, using an 18-MHz ultrasound transducer placed
on the temporal bone. • The diagnostic value of a positive or negative HUT response varies
according to the population studied.
If a patient has loss of consciousness or is unable to maintain
posture in association with a substantial decrease in blood pres- • The clinical history, pertinent physical examination findings, and
sure or heart rate, the patient is returned to supine position and ECG are most useful in determining the modality of testing with
the highest diagnostic yield.
the test is considered positive. If, after 20 to 45 minutes, no symp-
toms have developed, the patient is returned to supine position. • Where HUT is used, care should be taken in the testing of patients
If the HUT response is negative, HUT may be repeated with with severe organic heart disease or cerebrovascular disease.
administration of such drugs as isoproterenol and glyceryl trini-
trate, which increase the sensitivity of the test but also decrease Abbreviations
its specificity. Carotid sinus massage testing for carotid sinus AV atrioventricular
hypersensitivity may also be performed where indicated. CSNRT corrected sinus node recovery time
HUT is generally a safe procedure, but judgment needs to be DC direct current
used in its application to patients with severe organic heart disease ECG electrocardiography
or cerebrovascular disease, where hypotension or arrhythmia EP electrophysiologic
may be poorly tolerated (Box 19.9). HPS His-Purkinje system
HUT head-up tilt
ICD implantable cardioverter-defibrillator
Interpretation of a Positive HUT Response PES programmed electrical stimulation
RF radiofrequency
In the absence of structural heart disease, a positive HUT
SACT sinoatrial conduction time
response may be considered diagnostic, and if syncope is repro- SCL sinus cycle length
duced, no other tests are necessary. Where structural heart SNRT sinus node recovery time
disease is present, arrhythmia and other cardiac causes of syn- SVT supraventricular tachycardia
cope should be excluded before considering a positive HUT VF ventricular fibrillation
response as diagnostic. VT ventricular tachycardia
 20
Heritable Cardiomyopathies and
Channelopathies: Clinical Presentations,
Genetics, and Implications of Genetic Testinga
J. MARTIJN BOS, MD, PHD, and MICHAEL J. ACKERMAN, MD, PHD
Since the early 1990s, the field of genetic cardiac diseases has
evolved and discoveries along the continuum of research have
been translated into clinical practice. Genetic insights have
increased our understanding of the heritable cardiomyopathies
and channelopathies, enabled identification of preclinical and
potentially at-risk family members, and opened the door for con-
sidering new therapeutic options. The genomic revolution, how-
ever, has also brought with it the need to acquire a new vocabulary
to understand the role of genetics and genetic testing in the eval-
uation and care of patients with these heritable heart diseases.
This chapter discusses the clinical presentations, the most impor-
tant genotype-phenotype relationships, and the implications of
genetic testing for clinical practice for some of the most common
heritable cardiovascular diseases: HCM, ARVC, LQTS, BrS,
and CPVT.
a
Portions previously published in the following: Tester DJ, Ackerman
MJ. Cardiomyopathic and channelopathic causes of sudden unexplained
death in infants and children. Annu Rev Med. 2009;60:69–84; Bos JM,
Towbin JA, Ackerman MJ. Diagnostic, prognostic, and therapeutic
implications of genetic testing for hypertrophic cardiomyopathy. J Am
Coll Cardiol. 2009 Jul 14;54(3):201–11; Landstrom AP, Tester DJ,
Ackerman MJ. Role of genetic testing for sudden death predisposing
heart conditions in athletes. In: Lawless CE, editor. Sports cardiology
essentials: evaluation, management and case studies. New York:
Springer; c2011. p. 85–100; Bos JM, Ommen SR, Ackerman MJ.
Genetics of hypertrophic cardiomyopathy: one, two, or more diseases?
Curr Opin Cardiol. 2007 May;22(3):193–9; Menon SC, Bos JM, Ommen
SR, Ackerman MJ. Arrhythmogenic malignancies in hypertrophic
cardiomyopathy. In: Gussak I, Antzelevitch C, Wilde AAM, Friedman
PA, Ackerman MJ, Shen W-K, editors. Electrical diseases of the heart:
genetics, mechanisms, treatment, prevention. London: Springer; c2008.
p 610–26. Used with permission.
Abbreviations and acronyms are expanded at the end of this chapter.
Hypertrophic Cardiomyopathy
Clinical Presentation of HCM
HCM is defined as unexplained left ventricular hypertrophy in
the absence of precipitating factors such as hypertension or aor-
tic stenosis. HCM is a disease of vast phenotypic and genotypic
heterogeneity. Affecting an estimated 1 in 500 people, HCM is
arguably the most prevalent genetic cardiovascular disease and,
more importantly, one of the most common causes of SCD before
the age of 40 years, especially among young athletes. HCM can
manifest with negligible to extreme hypertrophy, absent to severe
LVOTO, and distinct patterns of hypertrophy. Microscopically,
HCM is recognized by a classic triad of cardiomyocyte hyper-
trophy, interstitial fibrosis, and myofibrillar disarray. The clini-
cal presentation of HCM is underscored by extreme variability
that ranges from asymptomatic disease to severe heart failure,
arrhythmias, and SCD. Many patients remain asymptomatic
or only mildly symptomatic throughout life. HCM commonly
manifests between the second and fourth decades of life but can
manifest at the extremes of age. The most common symptoms at
presentation with disease are exertional dyspnea, chest pain, and
syncope or presyncope. Infants and young children who present
with severe hypertrophy leading to heart failure have a poor prog-
nosis. In addition, SCD can be the tragic sentinel event for HCM
in children, adolescents, or young adults. Approximately 5% of
patients with HCM progress to “end-stage” disease characterized
by LV dilatation and heart failure. In such cases, cardiac trans-
plant may be considered. Other life-threatening complications
include embolic stroke and cardiac arrhythmias.
Conventional 2-dimensional echocardiography is the initial
diagnostic imaging modality of choice for the clinical diagno-
sis of HCM. Echocardiography shows unexplained and usually
asymmetric, diffuse or segmental hypertrophy associated with
257
258 III Electrophysiology
a nondilated LV with or without LVOTO. A LV wall thickness
of 12 mm or less is typically regarded as normal, with mea-
surements of 13 to 15 mm labeled as “borderline hypertrophy.”
A maximal LV end-diastolic wall thickness exceeding 15 mm is
the absolute dimension generally accepted for the clinical diag-
nosis of HCM in adults (in children, ≥2 standard deviations from
the mean in relation to body surface area). In general, although
many morphologic phenotypes are associated with HCM, the 4
most common morphologic subtypes of HCM are sigmoid sep-
tum, reverse septal curvature, apical, and neutral contour variant
(Figure 20.1).
Dynamic LVOTO is a common feature of HCM, but its pres-
ence is not required for the diagnosis of HCM. LVOTO is diag-
nosed by demonstrating a resting or provoked Doppler gradient
of more than 30 mm Hg. LVOTO is produced by the interac-
tion of the hypertrophied septum and systolic anterior motion of
the mitral valve’s anterior leaflet. Approximately two-thirds of
patients with HCM have obstructive HCM (also called hyper-
trophic obstructive cardiomyopathy) that is equally divided
between obstruction at rest and labile or provoked obstruction.
Most patients at presentation manifest some degree of impaired
diastolic function ranging from abnormal relaxation to severe
myocardial stiffness, elevated LV end-diastolic pressure, and
elevated atrial pressure, which can progress to cause pulmonary
congestion, exercise intolerance, and fatigue. Systolic cardiac
function, as measured by ejection fraction, is usually preserved.
End-stage disease is characterized by LV dilatation, poor sys-
tolic function, and heart failure. Current pharmacotherapies
for managing symptomatic obstructive HCM include the use of
β-blockers, calcium channel blockers, and disopyramide alone
or in combination. Antiarrhythmic therapy in HCM is usually
unnecessary unless atrial fibrillation develops. For patients with
symptoms refractory to pharmacotherapy, surgical (septal myec-
tomy) and percutaneous (alcohol septal ablation) septal reduction
therapies are available.
Risk Stratification of SCD and ICD
Therapy in HCM
HCM is one of the most common causes of SCD in young
persons. For the entire population of patients with HCM, the
Sigmoid septum Reverse septal
(43%) curvature (31%)
22% Gene + 65% Gene +
Figure 20.1. Morphologic Subtypes of Hypertrophic Cardiomyopathy. Parenthetical percentages indicate the percentage of patients who have that
subtype of hypertrophic cardiomyopathy. The other percentages indicate the yield from genetic testing for the presence of the myofilament mutation
(Gene +).
overall mortality rate is less than 1% per year. For the highest-
risk subpopulation, the annual mortality approaches 6% to 8%.
Fortunately, only a small proportion of HCM patients have such
high-risk HCM. SCD is more frequent in adolescents and young
adults and could be the first presenting symptom. The ICD
assumes an important role in primary and secondary prevention
of the arrhythmias. In many young patients, ICD use prolongs
life substantially and offers a nearly normal life expectancy. In
a multicenter study of ICD use in patients with HCM, the device
intervened appropriately at a rate of 5% per year for primary
prevention and 11% per year for secondary prevention during an
average follow-up of 3 years.
ICD indications include 1) an out-of-hospital cardiac arrest
or 2) documented VF or VT. For primary prevention, no sin-
gle clinical, morphologic, genetic, or electrophysiologic factor
has emerged as the most reliable predictor of risk in HCM. A
decision to intervene with an ICD should not be based on the
presence or absence of a particular HCM-causing mutation.
Instead, the presence of 2 major risk factors or 1 major risk factor
and 1 or more minor disease gene modifiers identify the HCM
patient who may benefit most from a prophylactic ICD. In the
latest HCM guidelines, the 3 major risk factors are 1) extreme
hypertrophy (≥30 mm), 2) a family history of SCD, and 3) unex-
plained (exercise-induced) syncope. An abnormal blood pres-
sure response during exercise and nonsustained VT on Holter
monitoring are still considered major risk factors but may not be
as important as the other 3. Disease modifier factors (ie, minor
risk factors) include degree of obstruction, delayed gadolinium
hyperenhancement during cardiac magnetic resonance imaging,
or a positive genetic test result. There is general consensus for
recommending an ICD if 2 major risk factors are present and for
withholding ICD therapy in the absence of any major risk factor.
The management of an HCM patient with a single major risk fac-
tor with or without disease modifiers must be individualized.
Genetic Basis and Genotype-Phenotype
Correlations of HCM
The first locus for familial HCM on chromosome 14 was dis-
covered in 1989, and the first mutations involving the MYH7-
encoded β-myosin heavy chain were identified in 1990 as the
Apical Neutral
(7%) (19%)
26% Gene + 32% Gene +
 20 Heritable Cardiomyopathies and Channelopathies 259

pathogenic basis for HCM. Since then, several hundred mutations The spectrum of HCM-associated genes has expanded beyond
have been identified in more than 20 genes encoding various sar- the various cardiac myofilaments to encompass additional sub-
comeric proteins (myofilaments and Z-disc–associated proteins) groups that could be classified as Z-disc HCM and calcium-
and calcium-handling proteins and in genes encoding phenocop- handling HCM (Table 20.1). Attention has been focused on genes
ies for the disease (Table 20.1). The most common genetically encoding the proteins that comprise the cardiac Z-disc as potential
mediated subtype of HCM is myofilament HCM, with hundreds HCM-susceptibility genes because the Z-disc has several proper-
of disease-associated mutations in 9 genes encoding proteins ties of interest: 1) its proximity to the contractile apparatus of the
critical to the sarcomere’s thick myofilament (β-myosin heavy myofilament, 2) its specific cytoarchitectural structure-function
chain [MYH7], regulatory myosin light chain [MYL2], and relationship, and 3) its role in the stretch-sensor mechanism of
essential myosin light chain [MYL3]); intermediate myofilament the sarcomere. The Z-disc is an intricate assembly of proteins
(myosin-binding protein C [MYBPC3]); and thin myofilament at the Z line of the cardiomyocyte sarcomere. Proteins of the
(cardiac troponin T [TNNT2], α-tropomyosin [TPM1], cardiac Z-disc are important in the structural and mechanical stability of
troponin I [TNNI3], and actin [ACTC]). More recently, mutations the sarcomere—they seem to serve as a docking station for tran-
have been described in the myofilament protein α-myosin heavy scription factors, calcium-signaling proteins, kinases, and phos-
chain encoded by MYH6 and in cardiac troponin C encoded by phatases. In addition, this assembly of proteins seems to serve as
TNNC1. The prevalence of mutations in the 9 most common a way station for proteins that regulate transcription by aiding in
myofilament-associated genes ranges from 30% to 60% in dif- their controlled translocation between the nucleus and the Z-disc.
ferent international cohorts; many patients still have genetically Calcium is the critical ion in the excitation-contraction coupling
unexplained disease. of the cardiomyocyte, so proteins involved in calcium-induced
calcium release are of high interest in the pathogenesis of HCM;
several genes involved in calcium-handling of the cell have been
Table 20.1. Molecular Basis of Cardiomyopathies implicated in HCM (Table 20.1). Disease associations for most of
these genes have been identified by pedigree analyses (cosegre-
Gene Locus Protein gation) and functional in vitro or in vivo models (eg, transgenic
mice). Mutations in these genes are uncommon (<3%), however,
Myofilament (“Sarcomeric”) HCM and mutations in MYBPC3 and MYH7 are by far the largest con-
ACTC 15q14 α-Cardiac actin tributors to the disease, with a yield of 15% to 35% for each gene,
MYBPC3a 11p11.2 Cardiac myosin–binding protein C depending on the cohort studied.
MYH6 14q11.2-q12 α-Myosin heavy chain
In multiple studies since the early 1990s, attempts have been
MYH7a 14q11.2-q12 β-Myosin heavy chain
made to identify phenotypic characteristics that are most indic-
MYL2 12q23-q24.3 Regulatory myosin light chain
MYL3 3p21.2-p21.3 Essential myosin light chain
ative of myofilament HCM to facilitate genetic counseling and
TNNC1 3p21.1 Troponin C strategically direct clinical genetic testing. Recent genotype-
TNNI3 19p13.4 Cardiac troponin I morphologic subtype analyses have demonstrated that echocar-
TNNT2 1q32 Cardiac troponin T diographically guided genetic testing may be possible. Between
TPM1 15q22.1 α-Tropomyosin the 2 most common morphologic subtypes of disease, the yield
Z-disc HCM from HCM genetic testing is much greater for patients with
ACTN2 1q42-q43 α-Actinin 2 reverse septal curvature HCM (which is the type in about one-
CSRP3 11p15.1 Muscle LIM protein third of all HCM patients seen at Mayo Clinic) than the yield
LBD3 10q22.2-q23.3 LIM-binding domain 3 from patients with sigmoid septum HCM (65% vs 22%), which
MYOZ2 4q26-q27 Myozenin 2 is the most common morphologic type of HCM at Mayo Clinic
TCAP 17q12-q21.1 Telethonin (Figure 20.1). Compared with other phenotypic determinants,
VCL 10q22.1-q23 Vinculin/metavinculin septal contour was the strongest predictor for the presence of a
Calcium-Handling HCM myofilament mutation regardless of age (odds ratio 4.8, P<.001)
JPH2 20q12 Junctophilin 2 (Figure 20.1).
PLN 6q22.1 Phospholamban Early studies of genotype and risk stratification suggested
Metabolic Cardiac Hypertrophy (HCM Mimickers) that specific mutations in some of the genes responsible for myo-
FXN 9q13 Frataxin filament HCM may be inherently “benign” (eg, MYH7-L908V)
GLA Xq22 α-Galactosidase A or “malignant” (eg, MYH7-R453C). However, these early stud-
LAMP2 Xq24 Lysosome-associated membrane protein 2 ies were based on highly penetrant, single families with HCM.
PRKAG2 7q35-q36.36 AMP-activated protein kinase Subsequent genotype-phenotype studies involving a large cohort
RAF1 3p25.2 RAF serine/threonine kinase
of unrelated patients have indicated that great caution must be
Arrhythmogenic Right Ventricular Cardiomyopathy exercised with assigning particular prognostic significance to any
DSC2a 18q21 Desmocollin 2
particular mutation. Each discrete, HCM-causative mutation is
DSG2a 18q12.1q12.2 Desmoglein 2
rare and the previously noted clinical outcome may not hold true
DSPa 6p24 Desmoplakin
for the next unrelated patient. In addition, contemporary studies
JUP 17q21 Plakoglobin
PKP2a 12p11 Plakophilin 2
have shown that rather than being predictably the genetic disease
RYR2 1q42.1-q43 Cardiac ryanodine receptor of the young with greater hypertrophy (MYH7 HCM) or the HCM
TGFB3 14q23-q24 Transforming growth factor β-3 of the elderly (MYBPC3 HCM), these 2 common genetic sub-
TMEM43 3p25 Transmembrane protein 43 types of HCM are phenotypically similar. In contrast, patients
with myofilament HCM (ie, a positive genetic test) are more likely
Abbreviations: AMP, adenosine monophosphate; HCM, hypertrophic
cardiomyopathy.
than HCM patients with a negative genetic test to 1) receive a
a
Gene is responsible for at least 5% of the disease. diagnosis at a younger age, 2) have more hypertrophy, 3) have a
Previously published. See “Credit Lines” section. positive family history of HCM and SCD, and 4) receive an ICD.
260 III Electrophysiology
In addition, in a longitudinal study of a large cohort of unre-
lated Italian patients with HCM, patients with a positive HCM
genetic test involving any of the myofilament genes (compared
with patients with a negative genetic test) had an increased risk
of cardiovascular death, nonfatal stroke or progression to New
York Heart Association class III or IV (25% vs 7%, P = .002).
Multivariate analysis showed genotype-positive HCM to be the
strongest predictor of an adverse outcome (hazard ratio, 4.27;
95% confidence interval, 1.43–12.48; P = .008). Furthermore,
patients with genotype-positive HCM had a greater probability
of having severe LV systolic dysfunction (P = .021) and restric-
tive LV filling (P = .018). Patients with multiple mutations (ie,
compound or double heterozygotes), detected in 3% to 5% of
genotype-positive patients, have a more severe phenotype and
increased incidence of sudden death, suggesting that a gene-
dosage effect might contribute to disease severity. Interestingly,
in the majority of cases of compound heterozygosity, 1 of the
mutations usually involves MYBPC3. In a longitudinal study, a
similar trend showing that patients with double mutations (of
which 1 was usually MYBPC3) had more severe disease than
myofilament-negative patients or patients with a single MYBPC3
mutation, a thick filament mutation, or a thin filament mutation
combined (P<.05).
Genotype-phenotype relationships for the minor genetic
subtypes of HCM, especially cardiac Z-disc HCM, have been
even less informative. A main implication for the Z-disc is its
involvement in the cardiomyocyte stretch sensing and response
system, which has demonstrated that several HCM-susceptibility
genes are also implicated in dilated cardiomyopathy, with ACTC,
MYH7, TNNT2, TPM1, MYBPC3, TTN, ACTN2, LDB3, MLP,
TCAP, and VCL established as both HCM- and dilated cardio-
myopathy–susceptibility genes.
In summary, although clinical prognostication must be
rendered with great caution for specific gene domains or spe-
cific genetic mutations, a positive HCM genetic test in general
portends a greater likelihood for disease progression, particu-
larly as it pertains to systolic and diastolic dysfunction and pro-
pensity for symptoms to develop. Clinical genetic testing may
thereby aid in the prognostication of a patient’s disease outcome.
Furthermore, when a definite HCM-causative mutation has been
established for a given family, genetic testing can become a
gold standard diagnostic marker to separate the genetically
at-risk family members from the genetically unaffected ones.
Family members with a negative genetic test for the family’s
specific HCM-causing mutation and a normal echocardiogram
can potentially be dismissed. At a minimum, the frequency
and interval of subsequent echocardiographic surveillance can
be decreased. At the moment, no genotype-specific therapy is
available for HCM except perhaps for patients who genetically
have a very rare form of HCM secondary to LAMP2 mutations,
so-called Danon disease, for which early consideration is given
for transplant.
• HCM is one of the most common causes of SCD before the age of
40 years, especially in young athletes.
• The 4 most common morphologic subtypes of HCM are sigmoid
septum, reverse septal curvature, apical, and neutral contour
variant.
• No single clinical, morphologic, genetic, or electrophysiologic factor
has emerged as the most reliable predictor of risk of SCD in HCM.
• When a definite HCM-causative mutation has been established for
a given family, genetic testing can become a gold standard diag-
nostic marker to separate the genetically at-risk family members
from the genetically unaffected ones.
Arrhythmogenic Right Ventricular
Cardiomyopathy
Clinical Presentation of ARVC
ARVC, also known as arrhythmogenic right ventricular dyspla-
sia, is a heritable cardiomyopathy characterized by right ven-
tricular dysfunction and ventricular arrhythmias. Patients with
ARVC pathologically demonstrate fibrofatty replacement of the
myocardium of the right ventricle; however, both ventricles, or
in some instances only the LV, can be involved. Because ARVC
can involve the LV as well, some have proposed categorizing
this cardiomyopathy as arrhythmogenic cardiomyopathy. The
clinical picture may include 1) a subclinical phase with con-
cealed structural abnormalities, during which the patient may
present with cardiac arrest or SCD; 2) an overt electrical dis-
order with palpitations and syncope due to tachyarrhythmias
stemming from the right ventricle, often triggered during effort;
and 3) right ventricular or biventricular pump failure, possi-
bly requiring cardiac transplant. Perhaps 20% of SCDs may be
attributed to ARVC, but the percentage is even higher among
athletes who die suddenly. Palpitations and syncope (27% and
26%, respectively) are common presenting symptoms, and fatal
ventricular arrhythmias are the first manifestation in over 20%
of ARVC patients. Most patients present clinically after the onset
of adolescence, although the diagnosis has been given to patients
as young as 5 years.
Standard diagnostic criteria for ARVC are based on the recom-
mendations of a task force of the European Society of Cardiology
and International Society and Federation of Cardiology and
include major and minor criteria that were first proposed in 1994.
Revised task force criteria were proposed in 2010 (Table 20.2).
These criteria are based on evaluation of the genetic, electric,
anatomical, and functional status of the right ventricle. A diag-
nosis of ARVC requires the presence of 2 major criteria; 1 major
criterion and 2 minor criteria; or 4 minor criteria.
It is unclear whether any meaningful reverse remodeling to
the ARVC heart would result from pharmacotherapy used in the
treatment of the failing heart of patients with dilated cardiomy-
opathy (eg, β-blockers, angiotensin-converting enzyme inhibi-
tors, angiotensin receptor blockers, or aldosterone antagonists).
ARVC patients with arrhythmias are often treated with either
sotalol or amiodarone.
SCD and Indications for ICD in ARVC
ARVC may cause about 20% of all SCDs among persons
younger than 35 years. Of all SCD cases, 40% occur with exer-
cise, but a large proportion of ARVC patients have SCD during
sedentary activity. The occurrence of arrhythmic cardiac arrest
due to ARVC is substantially increased in athletes. As in HCM,
ICD placement is clearly indicated for secondary prevention
after an out-of-hospital cardiac arrest or sustained ventricular
tachycardia. In contrast, the role of an ICD for primary pre-
vention in patients with asymptomatic ARVC or their relatives
remains controversial. Because of the paucity of scientific data,
the decision must be based on individual risk assessment, physi-
cian judgment, patient preference, and economic considerations.
Recommendations for primary prevention in patients with
ARVC are the following: 1) presence of inducible ventricular
fibrillation during electrophysiologic study, 2) sustained ventric-
ular arrhythmia resistant to drug therapy, 3) extensive myocar-
dial involvement (left ventricular involvement with poor cardiac
function), 4) syncope or presyncope, or 5) a family history of
sudden death.
 20 Heritable Cardiomyopathies and Channelopathies 261

Table 20.2. Revised Task Force Criteria for Clinical Diagnosis of ARVC/ACM
Feature Major Criteria Minor Criteria

I. Global or regional By 2D echocardiography By 2D echocardiography

dysfunction and
structural alterations
II. Tissue characterization
of walls
III. ECG repolarization
abnormalities
IV. ECG depolarization
abnormalities
V. Arrhythmias
VI. Family history
Regional RV akinesia, dyskinesia, or aneurysm
And 1 of the following (end diastole):
PLAX RVOT ≥32 mm (corrected for by body
size ≥19 mm/m 2)
PSAX RVOT ≥36 mm (corrected for by body
size ≥21 mm/m 2)
Fractional area change ≤33%
By MRI
Regional RV akinesia or dyskinesia or
dyssynchronous RV contraction
And 1 of the following:
Ratio of RV end-diastolic volume to BSA
≥110 mL/m 2 (male) or ≥100 mL/m 2
(female)
RV ejection fraction ≤40%
By RV angiography
Regional RV akinesia, dyskinesia, or aneurysm
Residual myocytes <60% by morphometric analysis
(or <50% if estimated), with fibrous replacement
of the RV free wall myocardium in ≥1 sample,
with or without fibrofatty replacement of tissue on
endomyocardial biopsy
Inverted T waves in right precordial leads (V1, V2,
and V3) or beyond in individuals >14 y old (in the
absence of complete right bundle branch block
QRS ≥120 ms)
Epsilon wave (reproducible low-amplitude signals
from end of QRS complex to onset of T wave) in
right precordial leads (V1 through V3)
Nonsustained or sustained ventricular tachycardia of
left bundle branch morphology with superior axis
(negative or indeterminate QRS in leads II, III, and
aVF and positive in lead aVL)
ARVC confirmed in first-degree relative meeting task
force criteria
ARVC confirmed pathologically at autopsy or surgery
in first-degree relative
Identification of a pathogenic mutation categorized as
associated or probably associated with ARVC in
patient under evaluation
Regional RV akinesia, dyskinesia, or aneurysm
And 1 of the following (end diastole):
PLAX RVOT ≥29 to <32 mm (corrected for by body size ≥16
to <19 mm/m 2)
PSAX RVOT ≥32 to <36 mm (corrected for by body size 18 to
<21 mm/m 2)
Fractional area change >33% to ≤40%
By MRI
Regional RV akinesia or dyskinesia or dyssynchronous RV
contraction
And 1 of the following:
Ratio of RV end-diastolic volume to BSA ≥100 to <110 mL/m 2
(male) or ≥90 to <100 mL/m 2 (female)
RV ejection fraction >40% to ≤45%
Residual myocytes 60% to 75% by morphometric analysis (or
50%–65% if estimated), with fibrous replacement of the RV
free wall myocardium in ≥1 sample, with or without fibrofatty
replacement of tissue on endomyocardial biopsy
Inverted T waves in leads V1 and V2 in individuals >14 y old (in the
absence of complete right bundle branch block) or in V4, V5, or V6
Inverted T waves in leads V1, V2, V3, and V4 in individuals >14 y old
in the presence of complete right bundle branch block
Late potentials by SAECG in ≥1 of 3 parameters in the absence of a
QRS duration of ≥110 ms on the standard ECG
Filtered QRS duration ≥114 ms
Duration of QRS <40 μV (low-amplitude signal duration ≥38 ms)
Root mean square voltage of terminal 40 ms ≤20 μV
Terminal activation duration of QRS ≥55 ms measured from the
nadir of the S wave to the end of the QRS, including R′, in V1, V2,
or V3, in the absence of complete right bundle branch block
Nonsustained or sustained ventricular tachycardia of RV outflow
configuration, left bundle branch block morphology with inferior
axis (positive QRS in leads II, III, and aVF and negative in lead
aVL) or of unknown axis
>500 ventricular extrasystoles per 24 h (Holter monitor)
History of ARVC in first-degree relative in whom it is not possible
to determine whether family member meets task force criteria
Premature sudden death (<35 y old) due to suspected ARVC in first-
degree relative
ARVC confirmed pathologically or by current task force criteria in
second-degree relative

Abbreviations: ACM, arrhythmogenic cardiomyopathy; ARVC, arrhythmogenic right ventricular cardiomyopathy; BSA, body surface area; 2D, 2-dimensional; ECG,
electrocardiographic; MRI, magnetic resonance imaging; PLAX, parasternal long-axis view; PSAX, parasternal short-axis view; RV, right ventricular; RVOT, right
ventricular outflow tract; SAECG, signal-averaged electrocardiography.
Previously published. See “Credit Lines” section.

Genetic Basis and Genotype-Phenotype

Correlations of ARVC
About half of all cases of ARVC are familial, with an auto-
somal dominant inheritance with variable penetrance and
expressivity. The majority of cases involve causative genes
encoding proteins of mechanical cell junctions or the desmo-
some, prompting the designation of ARVC as a disease of the
desmosome or a cell-cell junction disorder. Besides their role in
cell-to-cell junctions, intracellularly, these proteins connect to
keratin filaments, ion channels, and sarcomere proteins, which
may explain the variable expressivity, pathologic changes, and
propensity for arrhythmias and SCD with this disease. To date,
8 ARVC-susceptibility genes and 3 additional genetic loci have
been identified (Table 20.1). Two forms of autosomal recessive
262 III Electrophysiology
ARVC have been identified: Naxos disease (17q21) and ARVC
with anterior polar cataract (14q24-q terminal). Naxos disease
also is associated with palmoplantar keratosis and woolly
hair. Unlike the incomplete penetrance associated with auto-
somal dominant ARVC, the penetrance of Naxos disease is
essentially 100%.
The ARVC genetic test offers negligible prognostic and thera-
peutic value. In addition, there is a relatively high rate of possible
false-positive results with the ARVC genetic test. If the current
ARVC genetic test is performed on otherwise healthy volunteers,
there is a 15% chance that there will be at least 1 rare variant
identified in 1 of the known ARVC-susceptibility genes. This
“background noise rate” implies that for any given “positive”
ARVC genetic test result, there is a 30% chance that the iden-
tified variant is not the patient’s true ARVC-causing mutation.
Therefore, unless other evidence supports the conclusion that a
specific ARVC mutation is a definite or highly probable cause of
ARVC, first-degree relatives should not be dismissed from car-
diologic evaluation on the basis of a negative genetic test result.
• The occurrence of arrhythmic cardiac arrest due to ARVC is sub-
stantially increased in athletes.
• About half of all cases of ARVC are familial, with an autosomal
dominant inheritance with variable penetrance and expressivity.
Long QT Syndrome
Clinical Presentation of LQTS
Congenital LQTS comprises a distinct group of cardiac channel-
opathies characterized by delayed repolarization of the myocar-
dium, QT prolongation (eg, QTc = 480 ms is in the 50th percentile
among patients with genetically proven LQTS and in the 99th
percentile among otherwise healthy women), and increased risk
of syncope, seizures, and SCD. The incidence of LQTS may be
as high as 1 in 2,000 persons. Individuals with LQTS may or
may not manifest QT prolongation on a resting 12-lead surface
ECG. In fact, about 30% to 40% of patients with genetically
proven LQTS have a QTc less than 460 ms at rest. This repolar-
ization syndrome almost always is without consequence; how-
ever, rarely, triggers (eg, exertion, swimming, emotion, auditory
stimuli such as an alarm clock) can cause the heart to become
electrically unstable and produce a potentially life-threatening
ventricular arrhythmia known as TdP. Although TdP most often
resolves spontaneously, resulting in only an episode of syncope,
5% of untreated and unsuspecting persons with LQTS have a
fatal arrhythmia as their sentinel event. However, it is estimated
that nearly half of persons experiencing SCD stemming from
this treatable arrhythmogenic disorder may have had unrecog-
nized warning signs (eg, exertional syncope, family history of
premature sudden death).
When evaluating a patient for LQTS, one must carefully dis-
tinguish between congenital LQTS and QT prolongation due
to an underlying disease or drugs (ie, acquired LQTS). Mutant
cardiac channels or cardiac channel–interacting proteins are not
the only causes of these conditions. Abnormal cardiac repolar-
ization, QT prolongation, and TdP can result from numerous
medical conditions (eg, pheochromocytoma, anorexia, diabetes
mellitus, HCM), electrolyte derangements (particularly hypoka-
lemia), and more than 50 FDA-approved medications that can
affect the QT interval (http://www.qtdrugs.org/). In fact, QT lia-
bility and drug-induced TdP with sudden death have been among
the most common reasons for withdrawing drugs from the mar-
ket since the mid 1990s (eg, terfenadine, astemizole, cisapride,
Table 20.3. Conditions Associated With Acquired LQTS
Category Condition
Arrhythmia Complete atrioventricular block, severe
bradycardia, sick sinus syndrome
Cardiac Congestive heart failure, dilated
cardiomyopathy, hypertrophic
cardiomyopathy, ischemic heart disease,
left ventricular noncompaction, myocardial
sclerosis, myocarditis, tumors
Drug induced >50 FDA-approved medications,
including antiarrhythmics, antibiotics,
antihistamines, psychotropics, CYP3A4
inhibitors (complete list at http://www.
qtdrugs.com/)
Electrolyte abnormality Acute and chronic hypokalemia, chronic
hypocalcemia, chronic hypomagnesemia
Endocrine Diabetes mellitus, hyperparathyroidism,
hypothyroidism, pheochromocytoma
Neurologic Autonomic neuropathy, cerebrovascular
accident, encephalitis, head trauma,
hypoxic ischemic encephalopathy,
subarachnoid hemorrhage
Nutritional Acute weight loss, alcoholism, anorexia
nervosa, liquid protein diet, obesity,
starvation
Other Drowning or near-drowning, hepatic and renal
impairment
Abbreviations: CYP3A4, cytochrome P450 3A4 isozyme; FDA, US Food and
Drug Administration; LQTS, long QT syndrome.
grepafloxacin, propoxyphene). Many causes of acquired LQTS
are listed in Table 20.3.
Prudent preventative measures include the following: 1) avoid
use of drugs that affect the QT interval; 2) replenish electrolytes
and fluids, particularly with vomiting and diarrheal illnesses;
3) judiciously use antipyretics to reduce fever; 4) possibly sup-
plement with fish oil or omega-3 fatty acids; and 5) begin use of
a personal automatic external defibrillator. Treatment options for
LQTS range from no active therapy to pharmacotherapy with
primarily β-blockers (a first-line treatment, which should be the
most common) to unilateral left cardiac sympathetic denerva-
tion (involving a lower-half left stellectomy plus extended sym-
pathectomy from T2 through T4) to device therapy with either a
single-chamber ICD or a dual-chamber pacemaker-defibrillator.
The majority of LQTS patients require only pharmacotherapy.
Indications for left cardiac sympathetic denervation include
intolerance to pharmacotherapy, LQTS-triggered event during
pharmacotherapy, Jervell and Lange-Nielsen syndrome, and his-
tory of VF-terminating ICD therapies. Reasonable indications
for ICD therapy in LQTS include history of aborted or resus-
citated cardiac arrest, LQTS-triggered event during pharmaco-
therapy, extreme QT prolongation (QTc ≥550 ms), and Jervell
and Lange-Nielsen syndrome.
Genetic Basis and Genotype-Phenotype
Correlations of LQTS
Previously known as Romano-Ward syndrome, LQTS is a genet-
ically heterogeneous disorder largely inherited in an autosomal
dominant pattern. Rarely, LQTS is inherited as a recessive trait
first described by Jervell and Lange-Nielsen and is character-
ized by a severe cardiac phenotype and sensorineural hearing
loss. Hundreds of mutations have now been identified in 13
 20 Heritable Cardiomyopathies and Channelopathies 263

LQTS-susceptibility genes with 2 of the fi rst 3 canonical LQTS- in these genes; these patients present at a younger age and with
susceptibility genes discovered in 1995. Approximately 75% of greater expressivity.
patients with a clinically robust diagnosis of LQTS host either The 10 minor LQTS-susceptibility genes encode for 1) key
loss-of-function or gain-of-function mutations in 1 of these 3 cardiac channel interacting proteins (ie, ChIPs), which gener-
major LQTS genes (Table 20.4), which are responsible for the ally regulate the native ion channel current, or for 2) structural
orchestration of the cardiac action potential: KCNQ1-encoded membrane scaffolding proteins that properly localize channels
IKs (Kv7.1) potassium channel (LQT1; about 35% of patients; to the plasma membrane and collectively explain perhaps 5% of
loss-of-function mutation); KCNH2-encoded IKr (Kv11.1) potas- LQTS cases. Importantly, there is no quintessential mutational
sium channel (LQT2; about 30% of patients; loss-of-function “hot spot” within these genes, since nearly all unrelated families
mutation); and SCN5A-encoded INa (Nav1.5) sodium chan- have a unique “private” mutation. Approximately 20% to 25% of
nel (LQT3; about 10% of patients; gain-of-function mutation). clinical definite cases of LQTS remain genetically elusive.
Approximately 5% to 10% of patients have multiple mutations Specific genotype-phenotype associations in LQTS have
emerged, suggesting relatively gene-specific triggers, ECG pat-
Table 20.4. Summary of Cardiac Channelopathy- terns, and responses to pharmacotherapy. Swimming-induced
Susceptibility Genes and exertion-induced cardiac events are strongly associated with
mutations in KCNQ1 (LQT1), whereas auditory triggers and
Gene Locus Protein events occurring during the postpartum period most often occur
in patients with LQT2. Events occurring during sleep or rest are
Long QT Syndrome
more common in LQT3. Characteristic gene-suggestive ECG
Major LQTS genes patterns have been described as well. LQT1 is associated with
KCNQ1a (LQT1) 11p15.5 IKs potassium channel α subunit a broad-based T wave, LQT2 with a low amplitude notched or
(KvLQT1, Kv7.1)
biphasic T wave, and LQT3 with a long isoelectric segment fol-
KCNH2a (LQT2) 7q35–36 IKr potassium channel α subunit
(HERG, Kv11.1)
lowed by a narrow-based T wave.
SCN5Aa (LQT3) 3p21-p24 Cardiac sodium channel α subunit Exceptions to these relatively gene-specific T-wave patterns
(Nav1.5) do exist, and appropriate caution must be exercised with predic-
Minor LQTS genesb
tions of a particular LQTS subtype before genetic test results
AKAP9 7q21-q22 Yotiao are available. Importantly, the most common clinical mim-
ANKB 4q25-q27 Ankyrin B icker of the ECG that suggests LQT3 is LQT1. This is key since
CACNA1C 12p13.3 Voltage-gated L-type calcium the underlying genetic basis heavily influences the response to
channel (Cav1.2) standard LQTS pharmacotherapy (β-blockers). β-Blockers are
CAV3 3p25 Caveolin-3 extremely protective in LQT1 patients but comparatively less
KCNE1 21q22.1 Potassium channel β subunit so for patients with LQT2 or LQT3. For patients with LQT3,
(MinK) the preferred β-blocker is propranolol because of its concomi-
KCNE2 21q22.1 Potassium channel β subunit tant inhibition of the late sodium current. In addition, targeting
(MiRP1) the pathologic, LQT3-associated late sodium current with agents
KCNJ2 17q23 IK1 potassium channel (Kir2.1)
such as mexiletine, flecainide, or ranolazine may provide addi-
KCNJ5 11q24 Inwardly rectifying potassium
tional gene-specific therapeutic options for LQT3. This strategy
channel (Kir3.4)
SCN4B 11q23.3 Sodium channel β-4 subunit
has attenuated repolarization with clinically apparent shortening
SNTA1 20q11.2 Syntrophin-α-1 in the QTc, although there has not been an evidence-based dem-
onstration of survival benefit.
Brugada Syndrome
SCN5Aa (BrS1) 3p21-p24 Cardiac sodium channel α subunit
In addition, intragenotype risk stratification has been realized
(Nav1.5) for the 2 most common subtypes of LQTS on the basis of muta-
CACNA1C 2p13.3 Voltage-gated l-type calcium tion type, mutation location, and cellular function. Patients with
channel (Cav1.2) LQT1 with Kv7.1 missense mutations localizing to the transmem-
CACNB2 10p12 Voltage-gated l-type calcium brane-spanning domains clinically have a 2-fold greater risk of an
channel β-2 subunit LQT1-triggered cardiac event than LQT1 patients with mutations
GPD1L 3p22.3 Glycerol-3-phosphate localizing to the C-terminal region. Trumping mutation location,
dehydrogenase 1-like Kv7.1 loss-of-function mutations that cause more damage at the
KCND3 1p13.2 Transient outward current (Ito) cellular in vitro level (ie, dominant negative) result in a 2-fold
Kv4.3 α subunit greater clinical risk than mutations that damage the Kv7.1 channel
KCNE3 11q13.4 Potassium channel β subunit
less severely (ie, haploinsufficiency). In addition to the traditional
(MiRP2)
clinical risk factors, molecular location and cellular function are
SCN1B 19q13 Sodium channel β-1
SCN3B 11q23.3 Sodium channel subunit β-3
independent risk factors used in evaluating patients with LQTS.
Akin to patients who have LQT1 and undergo molecular risk
Catecholaminergic Polymorphic Ventricular Tachycardia
RYR2a (CPVT1) 1q42.1-q43 Ryanodine receptor 2
stratification, patients with LQT2 due to mutations in the Kv11.1
CASQ2 (CPVT2) 1p13.3 Calsequestrin 2 pore region have a longer QTc, a more severe clinical manifes-
KCNJ2 17q23 IK1 potassium channel (Kir2.1) tation of the disorder, and significantly more arrhythmia-related
cardiac events occurring at a younger age than LQT2 patients
Abbreviations: BrS1, Brugada syndrome; IK1, inwardly rectifying potassium who do not have mutations in the Kv11.1 pore region. Additional
current; IKr, rapidly activating potassium current; IKs, slowly activating information has suggested that LQT2 patients with mutations
potassium current; LQT1, long QT syndrome 1; LQT2, long QT syndrome 2;
LQT3, long QT syndrome 3; LQTS, long QT syndrome.
involving the transmembrane pore region had the greatest risk
a
Gene is responsible for at least 5% of the disease. of cardiac events, those with frameshift or nonsense mutations
b
Listed alphabetically. in any region had an intermediate risk, and those with missense
264 III Electrophysiology
mutations in the C-terminal region had the lowest risk of cardiac
events.
• About 30%–40% of patients with genetically proven LQTS have a
QTc <460 ms at rest.
• About 5% of untreated and unsuspecting persons with LQTS have
a fatal arrhythmia as their sentinel event.
• QT liability and drug-induced TdP with sudden death have been
among the most common reasons for withdrawing drugs from the
market since the mid 1990s.
• Approximately 75% of patients with a clinically robust diagnosis
of LQTS host either loss-of-function or gain-of-function mutations
in 1 of 3 major LQTS genes.
• Swimming-induced and exertion-induced cardiac events are
strongly associated with mutations in KCNQ1 (LQT1).
• Auditory triggers and events occurring during the postpartum per-
iod most often occur in patients with LQT2.
• Events occurring during sleep or rest are more common in LQT3.
Brugada Syndrome
Clinical Presentation of BrS
BrS is a heritable arrhythmia syndrome characterized by an
ECG pattern consisting of coved ST-segment elevation (≥2 mm)
followed by a negative T wave in the right precordial leads V1
through V3 (often referred to as a type 1 Brugada ECG pattern)
and an increased risk of SCD resulting from episodes of poly-
morphic ventricular tachyarrhythmias. The highly variable pen-
etrance and expressivity result in a disorder that ranges from a
lifelong condition in asymptomatic persons to SCD during the
first year of life. BrS is generally considered a disorder involv-
ing young men, especially Southeast Asian males, with the
arrhythmogenic manifestation first occurring at an average age
of 40 years and SCD typically occurring during sleep. In fact,
sudden unexplained nocturnal death in young males is endemic
in Southeast Asia and is considered phenotypically, genetically,
and functionally the same disorder as BrS. However, BrS is not
solely confined to expression in adulthood. Although the pheno-
typic expression is most common in adult males, BrS was first
described in a child. Fever and exposure to BrS-offending medi-
cations (http://www.brugadadrugs.org/) are recognized triggers
for arrhythmic cardiac events, including syncope and SCD.
Reducing fever with antipyretics and avoiding offending med-
ication exposures are the primary interventions for the asympto-
matic patient who has a spontaneous or drug-provoked (ajmaline,
procainamide, or flecainide) type 1 Brugada ECG pattern; an
ICD is indicated for any patient with arrhythmic syncope and a
diagnostic ECG. Quinidine is the drug of choice for BrS patients
who have recurrent VF-terminating ICD therapies. Although
the subject of much international debate, programmed electrical
stimulation during an electrophysiology study to test for induc-
ibility of VF may provide relatively little prognostic value.
Genetic Basis and Genotype-Phenotype
Correlations of BrS
BrS is inherited generally as an autosomal dominant trait.
All known BrS-associated genes are shown in Table 20.4.
Approximately 20% to 30% of BrS cases stem from loss-of-
function mutations in the SCN5A-encoded cardiac sodium channel
and are classified as BrS1. In 2009, an international compendium
of SCN5A mutations in patients referred for BrS genetic testing
reported nearly 300 distinct mutations in 438 of 2,111 unrelated
patients (21%), and the mutation detection yield ranged from 11%
to 28% across 9 centers. The yield of mutation detection may be
significantly higher among familial forms than in sporadic cases.
Most of the mutations are missense (66%), followed by frameshift
(13%), nonsense (11%), splice-site (7%), and in-frame deletion or
insertion (3%) mutations. Approximately 3% of genotype-positive
patients host multiple putative pathogenic SCN5A mutations, and
like the genotype-phenotype observations in LQTS, patients host-
ing multiple SCN5A mutations tend to be younger at diagnosis
than those having a single mutation (mean ± standard deviation,
29.7 ± 16 years vs 39.2 ± 14.4 years). Again, like in LQTS there is
no particular mutational “hot spot”: nearly 80% of the BrS-related
SCN5A mutations occur as private single-family mutations.
However, nearly 10% of the 438 unrelated, SCN5A mutation-
positive patients hosted 1 of 4 mutations: E1784K, F861WfsX90,
D356N, and G1408R. Interestingly, the most common occurring
BrS1 mutation, E1784K, has also been reported as the most com-
monly seen LQT3-associated SCN5A mutation, illustrating how
the same DNA alteration in a given gene can lead to 2 distinct
cardiac arrhythmia syndromes, most likely as a result of other
environmental or genetic modifying factors.
Besides pathogenic mutations in SCN5A, common polymor-
phisms may have a modifying effect on the disorder. The most
important common polymorphism, H558R-SCN5A, is a mod-
ulator of the BrS phenotype, where the presence of the minor
allele R558 may be associated with a less severe clinical course.
In addition to primary mutations of the sodium channel, muta-
tions in genes that modulate the sodium channel function rarely
cause BrS (Table 20.4). Other minor causes of BrS include loss-
of-function mutations in the L-type calcium channel or its aux-
iliary subunits and mutations that result in genetic enhancement
of Kv4.3 (Ito). Mechanistically, either decreases in the inward
sodium or calcium currents or increases in the outward Kv4.3
potassium current can produce the clinical phenotype of BrS.
However, the genetic cause remains elusive for more than two-
thirds of clinically diagnosed BrS cases, suggesting that this dis-
order has a high degree of genetic heterogeneity.
Because the majority of BrS cases are genetically undefined,
genotype-phenotype correlations in BrS have not been analyzed
to the same degree as in LQTS. Patients with BrS1 tend to have
greater H-V intervals than patients with genotype-negative BrS.
BrS1 patients with either nonsense or frameshift, premature trun-
cation-causing mutations have a more severe phenotype than BrS1
patients with a missense mutation in SCN5A. Unlike LQTS genetic
testing, in which the triad of diagnostic, prognostic, and therapeutic
contributions have been clarified, BrS genetic testing is currently
limited by its lower yield (25% for BrS vs 75% for LQTS) and lack
of therapeutic contribution from not knowing the genotype.
• There is no particular mutational “hot spot”: nearly 80% of the
BrS-related SCN5A mutations occur as “private” single-family
mutations.
• The genetic cause remains elusive for more than two-thirds of clin-
ically diagnosed BrS cases, suggesting that this disorder has a high
degree of genetic heterogeneity.
• BrS genetic testing is currently limited by its low yield and lack of
therapeutic contribution from not knowing the genotype.
Catecholaminergic Polymorphic VT
Clinical Presentation of CPVT
CPVT is a heritable arrhythmia syndrome that classically mani-
fests with exercise-induced syncope or SCD, is predominately
 20 Heritable Cardiomyopathies and Channelopathies 265

expressed in the young, and closely mimics the LQT1 phenotype.
However, compared with LQT1, CPVT’s lethality or event rate
is much higher. Like in LQT1, swimming is a potentially lethal
arrhythmia-precipitating trigger in CPVT. In fact, both LQT1 and
CPVT have been shown to underlie several cases of unexplained
drowning or near-drowning involving young healthy swimmers.
Importantly, in contrast to LQTS, CPVT is associated with a
normal resting ECG (although bradycardia and U waves can be
observed) and is electrocardiographically suspected after either
exercise or catecholamine stress testing shows significant ven-
tricular ectopy with the hallmark feature of bidirectional VT.
Clinically, a presentation of exercise-induced syncope and
a QTc less than 460 ms should first prompt consideration of
CPVT. Further, exercise-induced premature ventricular com-
plexes in bigeminy are far more likely to be found than the more
specific but less sensitive finding of bidirectional VT. CPVT is
generally associated with a structurally normal heart. CPVT
was thought to manifest only during childhood, but studies now
suggest that the age of first presentation ranges from infancy to
40 years. The lethality of CPVT is illustrated by mortality rates
of 30% to 50% by age 35 years and documentation of a positive
family history of premature SCD in more than a third of CPVT
individuals and in as many as 60% of families hosting CPVT1-
associated RyR2 mutations. Moreover, approximately 15% of
autopsy-negative sudden unexplained deaths in the young and
some cases of sudden infant death syndrome have been attrib-
uted to CPVT.
Although β-blocker therapy is the first line of therapy in
CPVT, verapamil, flecainide, and left cardiac sympathetic dener-
vation therapy are often considered to minimize the likelihood
of CPVT-triggered SCD. Because of reports of ICD storms
(VF-shock-VF-shock cycles) in CPVT, the ICD is increasingly
only used for secondary prevention after a presentation of CPVT-
triggered cardiac arrest. Given the superior efficacy of β-blocker
therapy for LQT1 compared with CPVT and the greater lethality
in CPVT compared with LQT1, it is essential to distinguish these
2 syndromes. In 1 study, strikingly, nearly a third of patients
who received a diagnosis of “possible/atypical” LQTS (QTc
<480 ms) with exertion-induced syncope instead tested positive
for RYR2 mutation. Similarly, some patients with a diagnosis
of CPVT, based on the presence of bidirectional VT on exer-
cise, have been identified subsequently as positive for KCNJ2-
mediated Andersen-Tawil syndrome, a heritable arrhythmia
syndrome with much less lethality than CPVT. The misdiagnosis
of Andersen-Tawil syndrome as the potentially lethal disorder
CPVT may lead to unnecessarily aggressive prophylactic therapy
(ie, ICD implantation).
Genetic Basis and Genotype-Phenotype
Correlations of CPVT
Perturbations in key components of intracellular calcium-
induced calcium release from the sarcoplasmic reticulum serve
as the pathogenic basis for CPVT. The 3 main genetic causes
of CPVT are shown in Table 20.4. Inherited in an autosomal
dominant fashion, mutations in the RYR2-encoded cardiac
ryanodine receptor–calcium release channel result in the most
common genetic subtype of CPVT (CPVT1), accounting for
60% of clinically “strong” cases of CPVT. Gain-of-function
mutations in RYR2 produce leaky calcium release channels that
cause increased intracellular calcium levels during diastole, par-
ticularly during sympathetic stimulation. This can precipitate
calcium overload, delayed afterdepolarizations, and ventricular
arrhythmias. Again, most unrelated CPVT families are identi-
fied with their own unique “private” RYR2 mutation, and about
5% of unrelated mutation-positive patients host multiple putative
pathogenic mutations.
While there does not appear to be a specific mutation hot
spot in the large gene RYR2, there are 3 regional hot spots or
“domains” where unique mutations reside. This observation has
lent itself toward targeted genetic testing of RYR2 (about 61 exons
or about two-thirds of the complete gene) rather than a complete
scan of RYR2. In fact, two-thirds of all CPVT1-associated muta-
tions in RYR2 are confined to less than 20 of its 105 translated
exons. More than 90% of the RYR2 mutations discovered to
date are missense mutations; however, perhaps as many as 5%
of unrelated CPVT patients host large gene rearrangements con-
sistent with large whole-exon deletions, akin to what has been
observed in LQTS.
• In contrast to LQTS, CPVT is associated with a normal resting
ECG.
• As it is in LQT1, swimming is a potentially lethal arrhythmia-
precipitating trigger in CPVT.
Abbreviations
ARVC arrhythmogenic right ventricular cardiomyopathy
BrS Brugada syndrome (with specific type numbered BrS1)
CPVT catecholaminergic polymorphic ventricular tachycardia
(with genetic subtype numbered CPVT1)
ECG electrocardiogram
HCM hypertrophic cardiomyopathy
ICD implantable cardioverter-defibrillator
IKr rapidly activating potassium current
IKs slowly activating potassium current
INa sodium current
Ito transient outward potassium current
Kv voltage-gated potassium
LQTS long QT syndrome (with specific types numbered LQT1,
LQT2, LQT3)
LV left ventricle
LVOTO left ventricular outflow tract obstruction
QTc corrected QT interval
SCD sudden cardiac death
TdP torsades de pointes
VF ventricular fibrillation
VT ventricular tachycardia
 21
Pediatric Arrhythmiasa
BRYAN C. CANNON, MD
Although the mechanisms of dysrhythmias in pediatric and
adult patients are usually the same, the clinical presentations
and management plans may vary considerably. Many aspects of
the pediatric ECG such as QRS axis, PR interval, and ventricu-
lar hypertrophy voltages are age dependent; therefore, a table of
normal values based on age is necessary when evaluating ECGs
from pediatric patients. In addition, heart rates are much faster
in children, with resting heart rates up to 160 beats per minute
being common in children younger than 1 year. The maximum
sinus heart rate is also faster in younger patients, with the max-
imum rate typically being 220 beats per minute minus the age.
This is an important fact to note when distinguishing sinus tachy-
cardia from pathologic arrhythmias in this age group. This chap-
ter reviews the salient features of most pediatric dysrhythmias,
including their management. This chapter does not provide an
extensive treatment of dysrhythmias in patients with complex
congenital heart defects.
Accessory Pathway–Mediated Tachycardia
in Structurally Normal Hearts
Children can have accessory pathways, either manifest (WPW)
or concealed (unidirectional retrograde accessory pathway),
causing reentrant tachycardias. Orthodromic reciprocating
tachycardia is the most common mechanism and occurs when the
antegrade conduction is through the AV node and the retrograde
conduction is through the accessory pathway. Antidromic recip-
rocating tachycardia is very rare and occurs when the antegrade
conduction is through the accessory pathway and the retrograde
a
The author thanks Co-Burn J. Porter, MD, who wrote the chapter in the
previous edition of this book.
Abbreviations and acronyms are expanded at the end of this chapter.
266
conduction is through the AV node, which makes this tachycar-
dia a wide-complex tachycardia.
Patients with manifest preexcitation are at risk for rapid AV
conduction and sudden cardiac death if atrial flutter or fibrillation
develops; both dysrhythmias are very rare in pediatric patients.
However, any patient with syncope and a manifest accessory
pathway (WPW) should have a detailed cardiac evaluation. In
addition, even asymptomatic patients with manifest accessory
pathways should have risk stratification. If the accessory path-
way loses conduction in a single beat on an exercise treadmill
test, the pathway is most likely low risk. In atrial fibrillation, if
the shortest preexcited RR interval (shortest duration between 2
successive preexcited beats) is greater than 250 milliseconds, the
pathway is also low risk. The atrial fibrillation may be spontane-
ous or induced intentionally during an electrophysiology study.
APMT can present at any time, from in utero to the teenage
years, but if it presents when the patient is younger than 1 year, the
child has an approximately 80% chance of outgrowing the tachy-
cardia. Because young children cannot report a rapid heart rate,
parents must be vigilant for subtle signs: pallor, rapid breathing,
lethargy, irritability, and poor feeding. It is common for infants
to have heart rates of 250 to more than 300 beats per minute
(“too fast to count”) during supraventricular tachycardia.
In patients with manifest preexcitation (WPW), echocar-
diography should be done to look for congenital heart defects;
the incidence of associated congenital heart disease can be as
high as 30%, most commonly Ebstein anomaly, congenitally
corrected transposition, and hypertrophic cardiomyopathy.
Echocardiography is useful to assess ventricular dysfunction;
cardiac function is occasionally further depressed after conver-
sion of supraventricular tachycardia. Decreased ventricular func-
tion usually is temporary, but in some cases it is permanent and
part of a cardiomyopathy.
 21 Pediatric Arrhythmias
Conversion of supraventricular tachycardia depends on the
clinical situation. For example, it is preferable for a premature
baby to remain in utero as long as possible to prevent complica-
tions associated with early delivery. Therefore, control of tachy-
cardia for a fetus in utero is attempted by giving the mother
medication orally to convert the tachycardia and prevent progres-
sion to hydrops fetalis. The most common medication given to
mothers to convert and prevent recurrence of fetal tachycardia
is digoxin. Sotalol, flecainide, and amiodarone are also used as
second-line medications or in the presence of hydrops, which has
a high fetal mortality rate.
For infants and children not in great distress, a bag of ice can
be placed over the child’s face to produce a strong vagal response
to terminate supraventricular tachycardia. It is best to keep the ice
bag over the eyes, nose, and mouth for 10 seconds to promote the
vagal response without causing thermal injury. This is a suffo-
cation-type reflex and can be alarming to parents. Alternatively,
an intravenous line can be placed and adenosine administered.
It is important to place a large-bore intravenous line in a large
vein. Adenosine must be administered rapidly with an immedi-
ate flush because the half-life in the circulation is only seconds.
Intravenous calcium channel blockers are generally avoided in
young children, especially those younger than 1 year. It is best to
avoid intravenous amiodarone, except in refractory tachycardias.
The first-line therapy after conversion to sinus rhythm is typi-
cally an oral β-blocker. Chronic therapy with oral digoxin is not
used for patients with manifest preexcitation.
Most recurrent or sustained episodes of supraventricular
tachycardia in a pediatric patient warrant chronic daily medica-
tion or catheter ablation. For patients in whom supraventricular
tachycardia develops after the age of 1 or 2 years, the condition
will most likely be a lifelong problem. Orthodromic reciprocat-
ing tachycardia in a structurally normal heart is not typically a
life-threatening problem. Because APMT can occur at any time
and place (eg, on a family vacation, on a school field trip, or in
isolated areas), chronic daily medication (365 days a year) or
catheter ablation is typically indicated. The risk of complications
from ablation is increased in younger patients; therefore, most
centers prefer not to perform ablation until the patient is at least
4 or 5 years old. In children older than this, it can be performed
with no increased risk of adverse events compared with the adult
population. Currently, the success rate for ablation of pediatric
APMT is greater than 95%, with a 99% success rate for left-
sided pathways. Patients with appropriately evaluated and treated
APMT should be able to participate in all activities.
Atrioventricular Node Reentrant Tachycardia
AVNRT is very rare in infants, but the relative incidence
increases in children 5 to 10 years old. By the teenage years,
AVNRT and APMT account for about 95% of cases of supraven-
tricular tachycardia in children with a structurally normal heart,
and they occur with equal frequency. More than 90% of clinical
AVNRT episodes are of the typical form (slow pathway conduc-
tion to the ventricle, fast pathway conduction to the atrium). Like
most APMTs, AVNRT is a nuisance tachycardia and not life-
threatening; tachycardia rates depend on the adrenergic state,
sometimes with rates as low as 160 to 180 beats per minute. The
key to diagnosis is a narrow QRS tachycardia and the absence
of P waves, which are typically buried in the QRS complexes in
pediatric patients.
Conversion of AVNRT can be attempted with an ice slurry
while intravenous materials are being obtained. Other Valsalva
267
maneuvers such as bearing down or standing on one’s head can
be attempted in older patients. As in APMT, intravenous adeno-
sine is very effective for conversion.
Sustained AVNRT generally indicates the need for chronic
daily oral medication or catheter ablation. Currently, first-time
ablation is successful in 95% to 98% of cases. Although com-
plete heart block occurs in less than 1% of cases treated with
radiofrequency ablation, this risk can be almost eliminated by
using cryoablation techniques. The risk of recurrence, however,
is higher for cryoablation.
Atrial Ectopic Tachycardia
AET is a rare dysrhythmia: only 5% to 10% of pediatric
supraventricular tachycardias are AET, but it is a very common
form of incessant supraventricular tachycardia in children. AET
is believed to be due to increased automaticity of a nonsinus atrial
focus, has a high association with tachycardia-induced cardiomy-
opathy, is often refractory to medical therapy, and is not usually
responsive to direct-current cardioversion. Ventricular function
usually improves with control of ventricular rate or elimina-
tion of the atrial focus, although this may take several months
to recover. AET occurs predominantly in infants and children
with structurally normal hearts. Many patients are asymptomatic
or only mildly symptomatic, especially with slower tachycardia
rates. A frequent clue to the presence of AET is the presence of
first-degree AV block. AET is not abruptly terminated with ice or
adenosine, although rarely atrial tachycardias may be adenosine
sensitive. Typically, there is slowing of the atrial rate or evidence
of AV block with use of adenosine, with the rapid atrial rate then
continuing.
The options for treatment of AET are medication for tachycar-
dia or rate control and catheter ablation of the focus. Rarely, sur-
gical ablation is necessary for refractory cases. Radiofrequency
catheter ablation is successful in 75% to 100% of patients. Some
patients have multiple sites or a so-called chaotic atrial tachy-
cardia, which is not amenable to ablation. Spontaneous remis-
sion of AET has been reported in pediatric patients, with most
patients younger than 3 years having resolution of their tachy-
cardia. AET in patients older than 3 years is much less likely to
resolve spontaneously.
Junctional Ectopic Tachycardia
JET is a very rare form of supraventricular tachycardia and
occurs in 2 forms: 1) idiopathic chronic (usually congenital)
JET, which occurs in the setting of a structurally normal heart,
and 2) postoperative JET, which occurs after repair of certain
types of congenital heart disease. The pathophysiology of JET
is unclear, but it may be due to automatic or triggered activity.
Congenital JET may go undetected until signs of congestive
heart failure develop, unless the rate is very fast and it comes to
medical attention shortly after birth. Postoperative JET is more
common in younger patients and usually occurs 6 to 72 hours
after cardiopulmonary bypass for defects such as ventricular sep-
tal defect, tetralogy of Fallot, AV canal defect, truncus arterio-
sus, and single ventricle defects necessitating Fontan procedures.
ECG can show a narrow-complex tachycardia identical or very
similar to sinus rhythm, with 1:1 retrograde conduction or retro-
grade ventriculoatrial dissociation with occasional sinus capture
beats causing variation in the RR interval. ECG recordings from
atrial pacing wires can be obtained if it is difficult to determine
the relationship between atrial and ventricular contractions.
268 III Electrophysiology
Treatment of the 2 forms of JET is different. Congenital JET
usually is treated initially with antiarrhythmic medication, such
as amiodarone. If medication is not successful in controlling
the ventricular rate, then catheter ablation may be considered.
For postoperative JET, a multiprong approach is taken: 1) mini-
mizing inotropic infusions, 2) correcting electrolyte and other
metabolic abnormalities, 3) using controlled hypothermia with
paralysis and sedation, 4) infusing intravenous amiodarone,
5) trying atrial overdrive pacing to restore AV sequence and
improve cardiac output, and 6) in extreme cases considering use
of extracorporeal membrane oxygenation. Postoperative JET is
typically a perioperative arrhythmia and does not recur after the
immediate postsurgical period.
Atrial Flutter
Atrial flutter can be seen in the newborn period because of the
rapid conduction seen in neonatal myocardium. Atrial flutter even
occurs in utero; however, many such atrial flutters are converted
by the birth process and do not recur after delivery. A few fetuses
with sustained atrial flutter may have associated morbidity or
mortality; therefore, it is important to administer antiarrhyth-
mic medication to the mother to obtain ventricular rate control
at the very least. Once infants with atrial flutter undergo electri-
cal conversion, long-term antiarrhythmic medication usually is
not required if no other arrhythmias are present. Most fetuses
and neonates with atrial flutter have structurally normal hearts,
but some newborns and young children may have dilated atria;
therefore, echocardiography is warranted. Atrial flutter is a rare
condition in older children with structurally normal hearts, but it
can occur with some regularity in children with congenital heart
defects before or, especially, after surgical repairs. Many patients
with atrial flutter have 2:1 or 3:1 AV block and may not have any
symptoms; however, they have a fixed ventricular rate without
any normal variability, and this finding should be a tip-off to an
abnormal heart rhythm.
Atrial flutter also can be seen in patients with a congenital
heart defect, either before or after operation. Atrial flutter that
occurs around scar tissue is known as intra-atrial reentry tachy-
cardia. This condition can be a challenge to manage and may
necessitate medication, ablation, operation, pacemaker place-
ment, or any combination of these options.
Ventricular Arrhythmias
A ventricular arrhythmia may be an isolated and completely
benign finding, a marker of serious systemic disease or myop-
athy, or a mechanism for syncope and sudden cardiac death in
pediatric patients. Isolated premature ventricular contractions
are relatively common. If they disappear with exercise testing
and the results of ECG and echocardiography are otherwise nor-
mal, they are usually of no consequence. They can occur with
low daily frequency in as many as 40% of patients with appar-
ently normal hearts. Sustained ventricular arrhythmias are much
less frequent. Although sustained ventricular arrhythmias can
occur in apparently normal hearts, approximately 50% of these
patients have congenital heart disease or cardiomyopathy. The
incidence of sudden cardiac death in pediatric patients is low, but
clinical decisions about management are difficult and generally
require the expertise of a pediatric electrophysiologist. In patients
with congenital heart disease, cardiomyopathy, or decreased ven-
tricular function, the risk of sudden death is higher, and more
aggressive evaluation is therefore warranted.
Because neurally mediated syncope is common and life-
threatening arrhythmias are rare in the pediatric population, ven-
tricular arrhythmias are rarely the cause of syncope. However,
palpitations preceding syncope, syncope during strenuous exer-
cise, or a family history of sudden death should prompt evalua-
tion for an arrhythmic cause of syncope.
Although ventricular fibrillation is rare in children, one study
demonstrated ventricular fibrillation as the initial rhythm in 19%
of out-of-hospital pediatric cardiac arrests. Ventricular fibrilla-
tion is usually a degeneration of another dysrhythmia or the end
result of a primary respiratory arrest. In out-of-hospital arrests,
including those occurring in infants, an AED should be used if
available because AEDs are both accurate and effective in the
pediatric population. The AED preferably should have pediatric
pads attached, but if these are not available adult pads may be
used, even in small children.
In patients at high risk for sudden death from ventricular
arrhythmias or in patients who have been resuscitated from a
nonreversible cause of ventricular fibrillation, an ICD is war-
ranted. ICDs can be safely placed in pediatric patients, but a
nontraditional approach such as placing an ICD coil in the peri-
cardial sac and using an epicardial ventricular pacing lead for the
ICD sensing lead may be necessary in very young patients.
Sinus Node Dysfunction
Sinus node dysfunction is primarily manifested by sinus bradycar-
dia, with heart rates less than expected for age (a reference table of
normal values is needed for evaluation; Table 21.1), or sinus pauses
of more than 3 seconds. The current approach to the diagnosis of
sinus node dysfunction relies on noninvasive methods of testing
(eg, ECG, 24-hour ambulatory monitoring, and exercise testing
[chronotropic incompetence]). Invasive electrophysiologic test-
ing can result in false-positive and false-negative results and have
limited utility in the pediatric population. Nonsurgical causes of
sinus node dysfunction include genetic inheritance, certain types
of congenital heart defects, inflammatory disease (eg, myocardi-
tis), central nervous system disease, antiarrhythmic medications,
hypothyroidism, and hypothermia. Unrecognized blocked prema-
ture atrial contractions concealed in the preceding T wave can
appear to be sinus node dysfunction, which is especially true in
newborns. Only symptomatic sinus node dysfunction should be
treated with a permanent pacemaker. Occasionally, sinus node
dysfunction can occur in concert with tachycardia, giving rise
to the so-called bradycardia-tachycardia syndrome, which may
require a pacemaker and antiarrhythmic medication.
Second-Degree Atrioventricular Block
The implications of second-degree AV block depend on the under-
lying cause, its potential for progression to complete AV block,
Table 21.1. Approximate Normal Resting
Heart Rates in the Pediatric Population by Age
Age Resting Heart Rate, bpm
Birth–1 wk 90–160
>1 wk–1 y 100–170
1–2 y 80–150
3–7 y 70–135
8–10 y 65–130
11–15 y 60–120
Abbreviation: bpm, beats per minute.
 21 Pediatric Arrhythmias
and the location and rate of the potential subsidiary pacemaker.
Causes of second-degree AV block include mechanical (during
catheterization), tumors (rhabdomyoma), myopathy (limb-girdle
muscular dystrophy), infection, immunologic (maternal systemic
lupus erythematosus), genetic (long QT syndrome), metabolic,
and drug-induced. A subgroup of newborns with congenital long
QT syndrome may have 2:1 AV block, related to the His-Purkinje
system or ventricular myocardial refractoriness. Wenckebach, or
Mobitz type I, second-degree AV block with progressive prolon-
gation of the PR interval before the nonconducted atrial impulse
is most commonly seen during times of increased vagal tone. It
is frequently seen on 24-hour Holter monitoring during sleep.
Wenckebach block seen at increased heart rates is concerning for
underlying AV nodal pathology. Mobitz type II second-degree
AV block with no prolongation of the PR interval before the
nonconducted atrial impulse is almost never seen in pediatric
patients. This finding should warrant further evaluation for an
underlying cause. Temporary or permanent pacing is typically
not necessary but is dependent on symptoms and the likelihood
of progression of the block.
Third-Degree Atrioventricular Block
Third-degree AV block occurs in 2 forms, acquired and congen-
ital. The acquired form has multiple causes: associated congen-
ital heart disease (corrected transposition of the great arteries),
direct injury to the conduction system, infections (Lyme disease),
genetic causes (Kearns-Sayre syndrome), inflammatory disease
(rheumatic fever or myocarditis), tumors (mesotheliomas, lym-
phomas), and medications. Postsurgical third-degree AV block,
resulting from trauma to the AV node at operation, is the most
common cause of acquired AV block in children. Generally, all
patients with acquired third-degree AV block require permanent
pacing because of the unreliability of their escape rhythm. It is
prudent to wait 7 to 10 days after a traumatic injury to the AV
node during surgery or cardiac catheterization to place a pace-
maker, because conduction may return when local tissue edema
subsides.
Congenital third-degree AV block occurs in 2 forms. In
one group, the fetus or newborn has bradycardia because of an
embryologic disorder resulting in abnormal formation of the
AV node–His-Purkinje axis in the setting of a complex struc-
tural congenital heart defect (eg, L-transposition of the great
arteries or ventricular inversion). In the other group, heart block
develops in the setting of a structurally normal heart. About half
of mothers of these affected infants have antibodies (anti-La)
269
that pass through the placenta and affect the fetal AV conduc-
tion tissue. Congenital third-degree AV block generally may be
identified during prenatal examinations and may be progressive.
Medications are typically not effective at increasing the fetal
heart rate. Occasionally bradycardia is found late in the third
trimester and mothers are rushed to the delivery room unnec-
essarily for emergency cesarean delivery. It is important that a
thorough evaluation of the fetus is performed to avoid this unfor-
tunate circumstance.
Indications for pacing in newborns with congenital heart
block include congestive heart failure, a wide-complex ventricu-
lar escape rhythm or complex ventricular ectopy, or a heart rate
less than 50 beats per minute without congenital heart disease
or 70 beats per minute with coexisting congenital heart disease.
Typically, an epicardial system is placed in very young patients
to avoid occluding the subclavian vein. Pacemakers can be
placed in newborns (including premature infants), but these sys-
tems have a high incidence of lead fractures as the patient grows.
The QT interval should be examined because congenital long QT
syndrome can occasionally present with complete AV block, and
this subset of patients has a very high mortality rate.
Occasionally, third-degree AV block is found in an asympto-
matic older child with a structurally normal heart, in which case
the concern is that it is acquired. In most cases, however, the con-
dition is congenital block that escaped detection in the newborn
period. After the newborn period, the indications for permanent
pacemaker include 1) pauses longer than 3 seconds on Holter
monitoring, 2) progressive cardiomegaly with decreasing ven-
tricular function, 3) junctional instability or wide QRS escape
rhythm, 4) symptoms such as syncope or progressive fatigue, and
5) complex ventricular ectopy. In patients who weigh more than
20 kg, a transvenous pacing system is a possibility to provide bet-
ter long-term pacing thresholds. However, in patients with intrac-
ardiac shunts or single-ventricle physiology, an epicardial pacing
system should be used to prevent thrombi forming on the leads,
which may lead to a stroke.
Abbreviations
AET atrial ectopic tachycardia
APMT accessory pathway–mediated tachycardia
AV atrioventricular
AVNRT atrioventricular node reentrant tachycardia
ECG electrocardiogram, electrocardiography
JET junctional ectopic tachycardia
WPW Wolff-Parkinson-White syndrome
 22

Atrial Fibrillation and Fluttera

PETER A. BRADY, MB, CHB, MD

Epidemiology of AF and AFL • AF and AFL are closely related arrhythmias. In clinical practice,
AF is the most common sustained arrhythmia and is estimated the approach to patients with AF and AFL is similar.
to affect approximately 6 million Americans. The prevalence is
projected to increase 3-fold by 2050. AF preferentially affects Pathophysiology of AF and AFL
men and the elderly. The lifetime risk of AF is 1 in 4 among men
AF, especially if prolonged, is associated with atrial fibrosis and
and women older than 40 years. In most cases, AF is associated
myocyte hypertrophy. This leads to atrial dilatation and dysfunc-
with cardiac disease such as hypertension, congestive heart fail-
tion called structural remodeling. In turn, repeated episodes of
ure, coronary artery disease, myocardial infarction, and valvular
AF cause shortening of the atrial refractory period and allow
heart disease.
reentry within the atria; this process is known as electrical
AFL is closely related to AF. Approximately 200,000 cases
remodeling. Most forms of AF are now recognized to be associ-
of AFL are diagnosed each year in the United States, and AFL is
ated with automatic foci arising from within thoracic veins (usu-
the most common indication for catheter ablation. Most patients
ally the pulmonary veins). The multiple factors that modulate AF
with AFL also have AF at some time. For the purpose of clin-
(Figure 22.1) should be evaluated and treated.
ical management, AFL and AF are considered the same. In the
remainder of this chapter, comments apply to both AF and AFL • Most cases of AF arise from triggers within the pulmonary veins.
unless otherwise stated.
AF and Other Cardiac Diseases
Clinical Importance of AF and AFL
Conditions commonly associated with AF are summarized in
Without treatment, AF is associated with an increased risk of Box 22.1.
death that is more than 2-fold and a nearly 5-fold increase in
the risk of thromboembolism and stroke. Together, these fac-
Hypertension
tors lead to significantly reduced quality of life and symptoms
related to AF and stroke. AF also exacts a considerable cost to In the hypertensive patient, independent risk factors for new AF
the health care system, with a 2- to 3-fold increase in the need include age, left ventricular mass, and left atrial diameter. After
for hospitalization, electrical cardioversion, prescription drugs, adjustment for age, left ventricular hypertrophy in combination
and follow-up. with left atrial dilatation is associated with the highest risk of
recurrence of AF.
a
Portions previously published in Brady PA, Gersh BJ. Atrial fibrillation
and flutter. In: Willerson JT, Cohn JN, Wellens HJJ, Holmes DR Jr, Congestive Heart Failure
editors. Cardiovascular medicine. 3rd ed. London: Springer; c2007.
p. 715–95. Used with permission. More than two-thirds of patients who have a diagnosis of
Abbreviations and acronyms are expanded at the end of this chapter. heart failure and are older than 65 years are also likely to have
270
 22 Atrial Fibrillation and Flutter 271

Myocardial Infarction and Ischemia

Modulating factors for atrial fibrillation
Approximately 10% to 15% of persons with acute coronary syn-
drome have AF, which is associated with an increase in mortal-
ity and risk of stroke, especially if left ventricular dysfunction is
• Genetics
Reversible also present.
• Aging

Hypertrophic Cardiomyopathy
Inflammation Atrial stretch
• Pericarditis • Hypertension AF is present in nearly 25% of patients with hypertrophic cardio-
• Cardiac surgery • LV dysfunction myopathy and is associated with both functional deterioration
• Interleukin 6? • Valvular disease and an increased risk of heart failure. Independent predictors
• C-reactive protein? of AF in patients with hypertrophic cardiomyopathy include
age older than 50 years, New York Heart Association class II or
more, and increased left atrial size that exceeds 4.5 cm. In hyper-
Endothelial Coagulation
dysfunction factors trophic cardiomyopathy, the risk of thromboembolic complica-
tions is markedly increased, especially with significant outflow
tract obstruction, age younger than 50 years, and persistent or
Autonomic Hormonal permanent AF.
• Vagal • Thyroid
• Adrenergic • Diabetes
Preexcitation Syndromes
AF is present in up to 30% of patients with WPW syndrome
• Obesity and may conduct directly to the ventricles and cause ventricu-
• Sleep apnea
lar fibrillation, a possible mechanism of sudden death among
patients with WPW syndrome.
Figure 22.1. Factors That Modulate Atrial Fibrillation. LV indicates
left ventricular.
Thyrotoxicosis

coexisting AF. Moreover, the prevalence of AF increases with
increasing severity of heart failure symptoms. AF is also a pre-
dictor of increased mortality among patients with heart failure
and is a frequent cause of worsening of symptoms (especially
when ventricular rate control is inadequate) and hospitalization
for heart failure. A prolonged increase in ventricular rate owing
to persistent AF predisposes persons to progressive left ventricu-
lar dysfunction or tachycardia-induced cardiomyopathy, which
is most common in patients with few or no AF symptoms. If
diagnosed and treated promptly, most cases are reversible.
• AF may be a cause or a consequence of heart failure.
• If a patient has AF and a persistently elevated heart rate, exclude
the possibility of tachycardia-induced cardiomyopathy.
Box 22.1. Conditions Commonly Associated With
Atrial Fibrillation
Cardiac
Coronary artery disease
Congestive heart failure
Hypertension
Rheumatic mitral valve disease
Diastolic dysfunction and diastolic heart failure
Noncardiac
Uncontrolled hyperthyroidism
Chronic lung disorders
Excessive alcohol intake
Diabetes mellitus
Age
Untreated thyrotoxicosis predisposes persons to AF and should
be excluded as a cause of AF, especially if a patient has recent-
onset thyrotoxicosis or long-standing AF with a ventricular rate
that is difficult to control. Although the precise mechanism of
how thyrotoxicosis causes AF is unclear, AF likely results from
1) the action of thyroid hormone on atrial myocytes, shortening
the duration of the cellular action potential (promoting multiple
wavelets), or 2) increased triggered activity within the pulmo-
nary veins.
• Exclude secondary causes of AF (structural cardiac disease, pul-
monary disease, and hyperthyroidism).
Recognition of the Problem
The clinical presentation of patients with AF is highly variable,
at least in part owing to differences in heart rate control, hemody-
namics, and comorbid conditions. The most common symptoms
are palpitations, dyspnea, fatigue, and light-headedness. Syncope
without significant sinus node dysfunction or WPW syndrome
(preexcitation) is rare. In up to a third of persons, AF episodes
are associated with minimal or no symptoms.
Signs of AF include irregular pulse with a pulse deficit (ie,
the auscultated or palpated apical rate is greater than the rate
palpated at the wrist), variable intensity of the first heart sound,
and absence of an a wave in the jugular venous pulse. Signs of
secondary causes of AF (eg, thyrotoxicosis, WPW syndrome)
may be present.
Regularization of the ventricular rate in patients with AF may
result from conversion to NSR, transition to AFL with a fixed
conduction ratio, onset of a junctional tachycardia or ventricular
tachycardia, or development of complete heart block. The possi-
bility of digoxin toxicity should also be excluded.
Electrocardiographically, AF is characterized by a lack of
organized atrial activity, manifested as a lack of clear P waves
before each QRS complex and an irregular rhythm (Figure 22.2).
272 III Electrophysiology
Figure 22.2. Atrial Fibrillation. Electrocardiogram is characterized by the absence of P waves before each QRS complex, an uneven baseline,
and irregular QRS complexes.
In contrast to AF, AFL is characterized by a more organized
rhythm (Figure 22.3).
The atrial rate in AFL is usually regular owing to the pres-
ence of a fixed circuit within the atria. Occurrence of AFL is
most common within the right atrium where natural barriers
to conduction such as the crista terminalis, tricuspid valve, and
eustachian ridge exit and form a conduction channel for the wave
front. In the absence of prior surgery (especially mitral valve
replacement), congenital heart disease, or prior ablation, AFL
confined to the left atrium is much less common because the left
atrial wall is relatively smooth without natural barriers to con-
duction. Typically the atrial rate in AFL is 220 to 250 beats per
minute; with 2:1 atrioventricular conduction, the ventricular rate
is 120 to 125 beats per minute.
• In a patient with AF, rate regularization may result from transition
to AFL or development of complete heart block. Less commonly,
it may result from onset of a junctional or ventricular tachycardia.
If patients are taking digoxin, exclude the possibility of digoxin
toxicity.
• In the absence of prior surgery, left atrial ablation, or congeni-
tal heart defects affecting the left atrium, AFL arises in the right
atrium more commonly than in the left atrium.
Classification of AF
Three clinical patterns of AF are recognized:
1. Paroxysmal AF—spontaneous termination; usually lasts for less
than 7 days and most commonly for less than 48 hours
2. Persistent AF—not self-terminating; lasts for more than 7 days and
requires chemical or electrical cardioversion
3. Permanent AF or long-standing persistent AF—attempts at cardio-
version fail to restore sinus rhythm, or AF recurs within 24 hours
after successful DCCV
The rate of progression from paroxysmal AF to permanent
AF is about 8% per year with nearly 20% of patients in perma-
nent AF at 4 years (Figure 22.4).
The first detected episode of AF is defined as the first clinical
manifestation of the arrhythmia when the patient is seen with AF
that has been present for less than 48 hours. Since many episodes
are asymptomatic, the first detected episode is not necessarily
the first occurrence of AF.
Lone AF
AF occurring in the absence of apparent cardiac disease is often
termed lone AF. In most cases it arises from foci within the tho-
racic veins, in particular the pulmonary veins. Depending on
Figure 22.3. Atrial Flutter. Electrocardiogram shows flutter waves. QRS complexes may be regular (fixed atrioventricular nodal conduction) or
irregular (variable atrioventricular nodal conduction).
how lone AF is defined, it accounts for up to 45% of cases among
younger patients and up to 30% of cases overall.
Classification of AFL
The classification of AFL is often confusing. One useful classifi-
cation recognizes typical (or isthmus-dependent) AFL and atyp-
ical (or non–isthmus-dependent) AFL. Nearly all cases of AFL
can be categorized into these 2 groups on the basis of a 12-lead
electrocardiogram. Typical AFL is characterized by inverted
flutter waves in leads II, III aVF, and V6 and by upright flutter
waves in lead V1 (biphasic or negative deflections in V1 are less
common). Typical (isthmus-dependent) AFL is dependent on
conduction through a critical isthmus between the tricuspid valve
and the inferior vena cava (Figures 22.5 and 22.6). Distinguishing
typical AFL from atypical AFL is important since typical AFL
is more easily amenable to catheter ablation (>90% success) than
atypical AFL.
In patients with a history of cardiac surgery or atrial abla-
tion or with congenital heart disease, consider atypical (non–
isthmus-dependent) AFL due to reentry around a circuit created
by a surgical or ablation scar. Catheter ablation therapy for atyp-
ical (non–isthmus-dependent) AFL is less successful than for
typical AFL.
Initial Evaluation and Management
of a First Episode of AF or AFL
Primary goals in the management of patients with AF are relief
of symptoms and prevention of consequences (thromboem-
bolism and tachycardia-induced cardiomyopathy). Restoring
and maintaining NSR is preferred for patients with symptoms,
whereas controlling the rate may be appropriate for patients
who are asymptomatic or patients who have multiple comorbid
conditions.
Initial management is determined by the urgency of the clinical
presentation. Patients who present with a rapid ventricular response
or hypotension due to AF require initial ventricular rate control and
anticoagulation (usually with unfractionated heparin for a rapid
ventricular response) while longer-term goals are considered.
Initial history, physical examination, and preliminary investi-
gations should be directed toward excluding potentially reversible
causes and assessing the early and long-term risk of thromboem-
bolic stroke. In most cases, a 12-lead electrocardiogram, chest
radiographs, and thyroid function tests should be evaluated along
with an assessment of left ventricular function. Transthoracic
echocardiographic evaluation of left ventricular function usually
 22 Atrial Fibrillation and Flutter 273

provides important information on valve status, left atrial size,

First detected ventricular systolic and diastolic function (except when in AF),
presence of ventricular hypertrophy, and pulmonary arterial
pressure. Transesophageal echocardiography is generally per-
formed as an initial test only if left atrial thrombus is suspected
or if cardioversion is required.
Persistent As part of the initial assessment, a 24-hour Holter monitor
Paroxysmal (requires chemical
may be helpful to correlate the recorded rhythm with the pres-
(self-terminating) or electrical
cardioversion) ence or absence of symptoms and to assess average heart rate
control. An electrophysiologic study is rarely useful as part of the
assessment unless another supraventricular tachycardia or WPW
syndrome is suspected.
Permanent or
chronic Rhythm Control Compared With Rate Control
A strategy of rate control with anticoagulant therapy has been
Figure 22.4. Progression of Atrial Fibrillation. (Previously pub- compared with rhythm control in 5 randomized controlled tri-
lished. See “Credit Lines” section.) als (Table 22.1). Taken together, these trials confirmed that

I aVR
VR V1 V4

II aVL
VL V2 V5

III aVF
VF V3 V6

Figure 22.5. Typical (Isthmus-Dependent) Counterclockwise Atrial Flutter. The atrial heart rate is typically 250 to 300 beats per minute and has
negative sawtooth deflections in leads II, III, and aVF and positive deflections in lead V1.

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

Figure 22.6. Reverse Typical (Isthmus-Dependent) Clockwise Atrial Flutter. The electrocardiographic appearance of this arrhythmia may be
more variable but typically has positive deflections in leads II, III, and aVF and negative or biphasic deflections in lead V1. This arrhythmia uses the
same circuit as typical counterclockwise atrial flutter but in the opposite direction.
Table 22.1. Trials Comparing Rate Control With Rhythm Control for Atrial Fibrillation
Patients Receiving Rate Control or
Baseline Characteristic Antiarrhythmic Drugs, % End Points

Patients in
Mean NSR at End All-Cause Ischemic
Follow- AF Mean Female, HTN, CHF, CAD, of Study, Mortality, Strokes,
Trial Patients up, y Population Age, y % % % % Anticoagulation Rate Control Antiarrhythmic % % %

AFFIRM 4,060 3.5 Age ≥65 y and 69.7 39.3 70.8 23.1 38.2 Rate: required Digoxin, 70.6 Amiodarone, 62.8 Rhythm: 3 y, 73.3; 91.7 62.5
other risk Rhythm: physician discretion β-Blocker, 68.1 Sotalol, 41.4 5 y, 62.6
factors for if sinus rhythm is Diltiazem, 46.1 Propafenone, 14.5 Rate: 5 y, 34.6
stroke or death maintained Verapamil, 16.8
RACE 522 2.3 Recurrent or 68 36.6 49 50 27 Rate: required except if NR Sotalol as initial agent, Rhythm: 39 5.5 NR
persistent AF <65 y and no cardiac followed by other Rate: 10
or AFL disease agents if necessary
Rhythm: physician discretion
if sinus rhythm is
maintained
STAF 200 1.6 Persistent AF 65.8 36.5 62.5 55.5 43.5 ACCP guidelines β-Blocker, 45 Amiodarone, 42 Rhythm: 36 1.7 23
with moderate Calcium channel Sotalol, 22 Rate: 9
to high risk of blocker, 22 Class I, 12
recurrence Digoxin, 75
PIAF 252 1 Persistent 60.5 27 50 16.5 23 All patients β-Blocker, 9 Amiodarone, 100 Rhythm: 56 0.7 NR
symptomatic AF Digoxin, 70 Rate: 10
Diltiazem, 100
HOT CAFE 205 1.7 Persistent AF 60.8 34.6 64.4 62 43.9 Rate: ACCP guidelines β-Blocker, 89.1 Amiodarone, 56.7 Rhythm: 63.5% 0.5 14.6
Rhythm: physician discretion Calcium channel Propafenone, 36.5 Rate: NR
if sinus rhythm is blocker, 7.9 Sotalol, 24.0
maintained Digoxin, 42.6
Abbreviations: ACCP, American College of Chest Physicians; AF, atrial fibrillation; AFL, atrial flutter; CAD, coronary artery disease; CHF, congestive heart failure; HTN, hypertension; NR, not reported; NSR, normal sinus rhythm.
 22 Atrial Fibrillation and Flutter
outcomes were similar—at least among patients deemed suita-
ble for either a rate control strategy or a rhythm control strat-
egy. However, most patients had few or no symptoms, and young
patients and patients with congestive heart failure were under-
represented or not included. Moreover, only 60% of the patients
in the rhythm control group were in NSR at follow-up and were
not systematically or rigorously monitored for AF recurrences
between visits. Thus, these results cannot be generalized to all
patients with AF.
Rate Control Agents
For controlling ventricular rate, efficacy is similar between non-
dihydropyridine calcium channel blockers (eg, verapamil, dilti-
azem) and β-blockers (eg, esmolol, metoprolol, and propranolol).
Therefore, choice of initial agent is based on factors such as pref-
erence, comorbid conditions (coronary artery disease, conges-
tive heart failure, history of reactive airway disease), and known
intolerance of an agent. In refractory cases, intravenous amio-
darone may be effective. Because of its delayed onset of action
and reduced efficacy, especially in an increased adrenergic state,
digoxin is no longer considered first-line therapy. Similarly,
because of its short half-life, adenosine has no role in ventricular
rate control of AF.
• Digoxin is no longer considered a first-line agent for AF management.
Assessing Rate Control
Periodic assessment of the adequacy of rate control is essential.
The latest guidelines suggest aiming for a ventricular rate dur-
ing AF of 60 to 80 beats per minute at rest and 90 to 115 beats
per minute during exercise. With this definition, adequate rate
control is possible in about 70% of patients given β-blockers,
54% of patients given calcium channel blockers (with or without
digoxin), and 58% of patients given digoxin alone. Recently, for
patients with persistent AF, a more lenient approach to rate con-
trol (aiming for a resting heart rate of <110 beats per minute) has
been shown to be as effective as a stricter approach and easier to
achieve.
Cardioversion of Atrial Arrhythmias
Spontaneous conversion to NSR most commonly occurs in patients
with an AF duration of less than 24 hours; among these patients,
the rate of spontaneous conversion is nearly 70%. However, if the
duration of AF exceeds 24 hours, the rate of spontaneous cardio-
version decreases to less than 20%. Spontaneous cardioversion
to NSR is also more common in patients with normal left ven-
tricular function, but it is less common in the elderly, in females,
Figure 22.7. Early Recurrence of Atrial Fibrillation After Direct Current Cardioversion. An atrial premature contraction (arrow) initiated atrial
fibrillation.
275
in the presence of structural heart disease or left atrial enlarge-
ment, and with digoxin. β-Adrenergic receptor blockers have
modest, if any, efficacy at preventing AF recurrence, although
they may have a role in situations with adrenergic-mediated AF
(eg, postoperative AF) and in patients with AF secondary to
thyrotoxicosis.
Anticoagulation and Cardioversion
Risk of thromboembolism in patients undergoing DCCV is
significantly lower with use of anticoagulant therapy. In clini-
cally stable patients, the 2 approaches are 1) initiating oral anti-
coagulation therapy with warfarin (goal INR, 2–3) for at least
3 weeks followed by DCCV or 2) initiating anticoagulation ther-
apy followed by transesophageal echocardiographically guided
cardioversion. Current guidelines recommend oral anticoagula-
tion with warfarin (INR, 2–3) for at least 3 weeks before (if no
transesophageal echocardiography) and 4 weeks after elective
cardioversion.
Recurrence of AF After Successful DCCV
When AF recurs or persists after attempts to restore NSR, sev-
eral patterns are recognized: 1) failure to restore NSR (even for
a single beat), 2) elevated atrial defibrillation threshold in about
25% of patients who have had AF for more than 48 hours, and
3) early recurrence of AF (within minutes of successful cardio-
version), which suggests that a “trigger” mechanism (most com-
monly arising from 1 or more thoracic veins) is initiating AF
(Figure 22.7).
Adjuvant AAD Therapy to Facilitate DCCV
Pretreatment with ibutilide improves the success rate, low-
ers the energy required for cardioversion in 30% of patients,
and increases the likelihood of successful restoration of NSR
in cases of increased atrial defibrillation threshold. The risk
of proarrhythmia is increased and requires observation for 24
hours in the hospital. The risk may be increased in patients with
left ventricular dysfunction. Treatment of ibutilide-induced pol-
ymorphic ventricular tachycardia with intravenous magnesium
may be beneficial.
Pharmacologic Therapy for Maintenance of NSR
For a rhythm control strategy, antiarrhythmic agents in Vaughan
Williams class I or III are most commonly used (Table 22.2). Two
approaches to pharmacologic therapy may be useful: 1) long-term
276 III Electrophysiology
Table 22.2. Drugs Commonly Used for Maintenance of Sinus Rhythm
Drug Lone Atrial Fibrillation
First-line agents Flecainide
Propafenone
Second-line agents Sotalol
Procainamide
Disopyramide
Amiodarone
Agents to avoid ...
Abbreviations: CAD, coronary artery disease; CHF, congestive heart failure; EF, ejection fraction.
suppressive AAD therapy (ie, drug therapy used regularly to pre-
vent AF recurrences) or 2) use of pulsed therapy (ie, the pill-
in-the-pocket) approach, aimed at use of higher doses of a class I
agent to facilitate chemical cardioversion of AF episodes when
they occur. With suppressive AAD, the likelihood of remaining
in NSR at 1 year increases from about 30% without AAD to about
70% with AAD.
Dronedarone
Dronedarone (Multaq) is the most recently approved AAD with
a chemical structure similar to that of amiodarone but without
the iodine. Like amiodarone, dronedarone is indicated for the
treatment of AF and AFL (but not ventricular arrhythmias).
Compared with placebo, dronedarone is more efficacious at
maintaining NSR in patients with paroxysmal AF. In addition,
dronedarone is the only AAD that has been associated with a
decrease in hospitalization due to cardiovascular events and a
decrease in stroke risk (likely a type 1 error or false-positive
finding). Dronedarone has also been associated with increased
mortality risk in patients with severe heart failure (class IV) or
NYHA class II or III heart failure with recent decompensation;
dronedarone is contraindicated for these patients.
Initiation of AAD Therapy In Inpatients
and Outpatients
One of the most important consequences of AAD therapy is
proarrhythmia (most commonly torsades de pointes but virtually
Box 22.2. Risk Factors for Torsades de Pointes
Patient factors
Female sex
Bradycardia, especially pauses associated with
termination of atrial fibrillation
Hypokalemia, hypomagnesemia
Baseline QT prolongation
Drug factors
Concomitant use of drugs that prolong the QT
interval or interfere with the metabolism of a
QT-prolonging agent
Renal impairment of renally excreted agents that
prolong the QT interval (eg, sotalol, procainamide)
CHF, CAD CAD (Normal EF) Renal Failure
Dofetilide Sotalol Amiodarone
Amiodarone
... ... Propafenone
Flecainide Flecainide Sotalol
Propafenone Propafenone Procainamide
Dofetilide
any arrhythmia) (Box 22.2); sometimes an inpatient loading
dose (with dofetilide) for AAD is recommended or mandated.
In general, loading doses should be avoided in outpatients with
evidence of sinus node dysfunction, AV conduction disturbance,
bundle branch block, or structural heart disease, particularly
myocardial ischemia or baseline QT interval prolongation with
drugs that prolong the QT interval (class III AADs).
• Common scenarios include 1) sinus node dysfunction unmasked
by use of an AAD, leading to a prolonged sinus node recovery
(most commonly with termination of AF); 2) worsening of AV
node or His-Purkinje conduction (with propafenone, flecainide, or
procainamide).
• Organization of AF into AFL (with flecainide), which may con-
duct more rapidly via the AV node. In these cases, concomitant use
of an AV nodal blocking agent is advised.
Preexcited AF
Preexcited AF (Figure 22.8) is a special circumstance in which
antegrade conduction of AF occurs via the AV node and the
accessory pathway, giving a variable degree of preexcitation.
One problem is that use of agents that block AV nodal conduction
may increase accessory pathway conduction, potentially leading
to ventricular fibrillation. Thus, agents that also slow acces-
sory pathway conduction, such as procainamide or amiodarone,
should be used with immediate electrical cardioversion if hemo-
dynamic compromise exists.
Postoperative AF
AF commonly occurs postoperatively (it occurs in more than
one-third of patients after coronary artery bypass surgery and
in more than half after valve surgery). In most cases, it is the
patient’s first presentation with the arrhythmia. The incidence of
postoperative AF is highest on postoperative days 2 and 3 and is
associated with prolonged hospitalization and increased morbid-
ity and mortality. Risk factors for postoperative AF include older
age (among patients older than 70 years, the incidence of post-
operative AF is >30%), prior AF, valvular disease, congestive
heart failure, left ventricular dysfunction, hypertension, chronic
pulmonary disease, pneumonia, and the need for prolonged
mechanical ventilation postoperatively.
Management
Spontaneous restoration of NSR occurs in about 30% of patients,
depending on duration of follow-up. As in nonpostoperative
AF, initial therapy is directed toward ventricular rate control.
Owing to increased sympathetic tone in postoperative patients,
 22 Atrial Fibrillation and Flutter
I aVR
V V1
II aVL
aV
VL V2
III aVF
V V3
Figure 22.8. Preexcited Atrial Fibrillation. Electrocardiogram shows the appearance of atrial fibrillation in the presence of an accessory pathway.
The telltale features are variably wide complexes at irregular intervals.
β-blocking agents are most useful as initial therapy in the absence
of contraindications. Second-line agents include calcium chan-
nel blockers such as verapamil and diltiazem, which may be used
alone or in combination with β-blocker therapy for rate control.
Management approaches are summarized in Box 22.3.
Pharmacologic Therapy
Amiodarone
Amiodarone is frequently used postoperatively when ventricu-
lar rate control is inadequate despite use of β-blockers or cal-
cium channel blockers (or both). Amiodarone has a rapid onset
of action when given intravenously owing to its β-blocking and
calcium channel blocking actions and is generally well tolerated
even in patients with marginal hemodynamic status or conges-
tive heart failure. In most cases, amiodarone use is discontinued
after 3 months. Long-term efficacy of amiodarone in restoring
NSR postoperatively is less clear.
Prognosis
In the absence of preoperative AF, most patients return to
NSR within 6 weeks after surgery irrespective of AAD use.
Anticoagulation for postoperative AF is not warranted in the
absence of recurrent AF and risk factors for stroke (according to
patient’s CHADS2 score).
Stroke Risk and Prevention of
Thromboembolic Complications
Stroke is the most common and devastating complication of AF,
with an incidence of about 5% for all-cause stroke in patients
with AF. Stroke may be the first presenting symptom in up to
25% of patients with AF. AF is an independent risk factor for
stroke; in the United States, about 15% of all strokes are caused
by AF. The risk of stroke in AF increases with age and comorbid-
ity. Ischemic stroke associated with AF is often more severe than
stroke from other causes.
277
V4
V5
V6
Box 22.3. Summary of Management Approaches for
Postoperative AF
Noncardiac surgery
If patient is not already receiving BB therapy and if
patient does not have risk factors for AF, no therapy
is recommended
If patient is receiving BB therapy, it should be
continued perioperatively and postoperatively
Cardiopulmonary surgery
Administer BB therapy perioperatively and
postoperatively
If BB therapy is contraindicated, use amiodarone
In patients at high risk of perioperative AF (eg,
patients with history of AF or valve surgery),
amiodarone may be useful for prophylaxis
Rate control
Initial agent is a BB unless it is contraindicated
A short-acting agent such as esmolol is used
if patient has a relative contraindication for BB
therapy
If patient has an absolute contraindication for BB
therapy, a CCB can be used
If BB therapy does not control the rate, a CCB may be
substituted or used in combination with BB therapy
Consider cardioversion
Rhythm control
If patient is hemodynamically unstable, use
electrical cardioversion
If patient does not have structural heart disease,
consider flecainide, propafenone, or sotalol
(continued)
278 III Electrophysiology
Box 22.3. (Continued)
If patient has structural disease (especially unresolved
ischemia), avoid class I antiarrhythmic agents
Avoid dofetilide and sotalol if patient has renal
impairment
If patient has ventricular dysfunction, administer
amiodarone or dofetilide
If patient has ventricular dysfunction or coronary artery
disease with renal impairment, administer amiodarone
Abbreviations: AF, atrial fibrillation; BB, β-blocker; CCB, calcium
channel blocker.
Mechanisms of stroke in AF are complex and incompletely
understood (Figure 22.9). Although left atrial appendage throm-
bus is an important risk factor, up to 25% of AF-associated stroke
is a consequence of intrinsic cardiac disease such as atheroma
(especially within the proximal aorta) and hypertension. Thus,
these factors should be considered when offering advice on anti-
coagulation to patients who have AF.
Adjusted-dose oral anticoagulation is highly effective for pre-
venting all stroke (both ischemic and hemorrhagic), with a risk
reduction of 61% compared with placebo. Patient age and the
intensity of anticoagulation therapy are the most powerful pre-
dictors of major bleeding complications (Figure 22.10).
Risk Factors for Stroke in AF
Multiple risk factors for stroke in patients with AF have been
identified and incorporated into several scoring systems for
selecting patients most likely to benefit from oral anticoagulation
therapy (Figure 22.11 and Tables 22.3–22.5).
Although efficacious, warfarin has several limitations: slow
onset and offset of action, a narrow therapeutic window, multiple
Hypertension
• Compliance • LA dilatation
• Aortic plaque • Stasis
• Endothelial
dysfunction
Stroke risk
Figure 22.9. Mechanisms of Stroke in Atrial Fibrillation. LA indicates left atrial; LVH, left ventricular hypertrophy.
drug interactions, and need for frequent dose adjustment. The
benefit from oral anticoagulation therapy depends closely on the
time spent in the therapeutic range, but owing to the limitations
and problems with warfarin, the therapeutic range is low even in
well-conducted clinical trials (Figure 22.12).
Aspirin alone offers only modest benefit for protection against
stroke risk. Use of aspirin is reserved for patients who have no
risk factors for AF or who have 1 moderate risk factor and do not
wish to take warfarin.
• Warfarin has a demonstrated efficacy for stroke prophylaxis in
patients who have a CHADS2 score ≥2.
• Successful maintenance of NSR does not eliminate the risk of
thromboembolism in at-risk patients.
Novel Anticoagulants for Stroke
Risk Reduction in AF
Owing to the limitations of warfarin, alternative therapeutic
strategies have been evaluated in multiple clinical trials. The use
of aspirin alone and the use of aspirin in combination with clopi-
dogrel have been evaluated as alternatives for decreasing stroke
risk among patients with AF.
Other agents for thromboprophylaxis in AF include oral fac-
tor Xa inhibitors (apixaban, edoxaban, and rivaroxaban), and
direct thrombin inhibitors (dabigatran, which does not require
INR monitoring and dosage adjustment). Of these, dabigatran is
currently the only agent for which phase 3 data from AF patients
have been published. In the RE-LY trial, 150 mg dabigitran was
superior to warfarin in decreasing stroke risk without increasing
major, minor, or intracranial bleeding. Although no excess liver
toxicity was documented (as reported for ximelagatran), there
was a small but statistically significant increased risk of gastroin-
testinal tract bleeding and myocardial infarction with dabigatran.
In October 2010, the US Food and Drug Administration approved
dabigatran for use in patients with AF.
• Diastolic
dysfunction
• LVH
Thrombogenic
factors
 22 Atrial Fibrillation and Flutter 279

20 Nonpharmacologic Management of AF and AFL

Nonpharmacologic approaches to managing AF and AFL have
15 evolved considerably since the late 1990s. They include abla-
Odds Ratio

tion of the AV node (with permanent pacemaker implantation),

10 Intracranial bleeding left atrial ablation aimed at isolation of the pulmonary and other
thoracic veins, and surgical approaches (maze procedure, mini-
5 maze procedure).
Ischemic stroke
0 Ablation of the AV Node With Permanent
1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Pacemaker Implantation
INR Ablation of the AV junction with implantation of a permanent
Figure 22.10. Adjusted Odds Ratios for Ischemic Stroke and pacemaker is most effective at controlling high ventricular rates
Intracranial Bleeding in Relation to Intensity of Anticoagulation. after failure or intolerance of pharmacologic AV nodal blocking
Arrows indicate the range of international normalized ratio (INR) val- agents or abandonment of a rhythm control strategy in highly
ues within which the risks of intracranial bleeding and ischemic stroke symptomatic (and usually elderly) patients. In most cases, per-
are minimized. (Previously published. See “Credit Lines” section.) manent AV block can be achieved without significant risks

Predicted 5-Year Risk

Step 1 Step 4 of Stroke
Age, y Points Diabetes Points Total Points 5-Year Risk, %
55-59 0 No 0 0-1 5
60-62 1 Yes 5 2-3 6
63-66 2 4 7
67-71 3 5 8
72-74 4 Step 5 6-7 9
75-77 5 8 11
Prior Stroke or TIA Points
78-81 6 9 12
82-85 7 No 0 10 13
86-90 8 Yes 6 11 14
91-93 9 12 16
>93 10 13 18
Step 6 14 19
Add up points from steps 15 21
Step 2 1-5 16 24
17 26
Sex Points Look up predicted 5-year 18 28
Male 0 risk of stroke in table 19 31
Female 6 20 34
21 37
22 41
Step 3 23 44
24 48
Systolic Blood Points
25 51
Pressure, mm Hg
26 55
<120 0 27 59
120-139 1 28 63
140-159 2 29 67
160-179 3 30 71
>179 4 31 75

Figure 22.11. Predicted 5-Year Risk of Stroke in Atrial Fibrillation. Risk score was derived from data in the Framingham Heart Study. TIA indi-
cates transient ischemic attack. (Previously published. See “Credit Lines” section.)
280 III Electrophysiology

Table 22.3. CHADS2 Stroke Risk Assessment 1.0 Warfarin group

Risk Factor Score 71%-100%

Cumulative Survival
C—Congestive heart failure 1 0.9
H—Hypertension 1
A—Age ≥75 y 1 61%-70%
D—Diabetes mellitus 1 0.8
S2—History of stroke or transient ischemic attack 2 51%-60%
41%-50%
31%-40%
0.7 ≤30%
Table 22.4. Risk of Stroke in Newly Detected Atrial
Fibrillationa No warfarin
0.6
CHADS2 Adjusted Stroke CHADS2 Recommended
Scoreb Ratec (95% CI) Risk Level Anticoagulant 0 500 1,000 1,500
0 1.9 (1.2–3.0) Low Aspirin Survival Time to Stroke, d
1 2.8 (2.0–3.8) Low Aspirin or Warfarin
2 4.0 (3.1–5.1) Moderate Warfarin
Figure 22.12. Association Between Time Spent With the
International Normalized Ratio (INR) in the Therapeutic Range and
3 5.9 (4.6–7.3) Moderate Warfarin
Clinical Outcome. Warfarin groups were stratified by the time patients
4 8.5 (6.3–11.1) High Warfarin
spent with the INR in the therapeutic range (2.0–3.0). All patients had a
5 12.5 (8.2–17.5) High Warfarin
CHADS2 score of 2 or more (see Table 22.3 for CHADS2 score calcula-
6 18.2 (10.5–27.4) High Warfarin
tion). (Previously published. See “Credit Lines” section.)
a
From the Joint Panel of the American Academy of Family Physicians and the
American College of Physicians.
b
See Table 22.3 for CHADS2 score calculation. or robotic approaches may alter the indications for the procedure
c
Expected rate of stroke per 100 patient-years. and the selection of appropriate patients.
Previously published. See “Credit Lines” section.

Catheter Ablation of AF
Table 22.5. Stroke Risk Stratification in Atrial Fibrillationa Percutaneous catheter-based ablation of AF has emerged as
Less Validated or Moderate-Risk High-Risk
an increasingly successful approach with an acceptable risk of
Weaker Risk Factors Factors Factors complications for patients who have symptomatic AF. Although
specific approaches to ablation vary, almost all of them target
Female sex Age ≥75 y Previous stroke, the triggering foci that arise from the thoracic veins, most com-
TIA, or embolism monly within the pulmonary veins. Depending on factors such
Age 65–74 y Hypertension Mitral stenosis as duration of AF, left atrial size, and the presence of comorbid
Coronary artery disease Heart failure Prosthetic heart
conditions, the success of the percutaneous approaches ranges
valve
Thyrotoxicosis LVEF ≤35%
from 60% to 80% freedom from recurrence of AF and the risk of
Diabetes mellitus complications ranges from 1% to 5%.

Abbreviations: LVEF, left ventricular ejection fraction; TIA, transient ischemic

attack. Abbreviations
a
From the 2006 guidelines of the American College of Cardiology, the AAD antiarrhythmic drug
American Heart Association, and the European Society of Cardiology. AF atrial fibrillation
Previously published. See “Credit Lines” section.
AFL atrial flutter
AV atrioventricular
of complications, with alleviation of AF-related symptoms, CHADS2 congestive heart failure, hypertension, age older than
75 years, diabetes mellitus, previous stroke or transient
improved quality of life, and a reduced need for hospitalization.
ischemic attack
Complications of AV junction ablation include those related to DCCV direct current cardioversion
vascular access and permanent pacemaker implantation and INR international normalized ratio
the long-term need for anticoagulation, loss of AV synchrony, NSR normal sinus rhythm
potential increase in rate of progression to permanent AF, and NYHA New York Heart Association
pacemaker dependency. A fairly recent concern is the possible WPW Wolff-Parkinson-White
adverse effect of right ventricular apical pacing on left ventricu-
lar function. Names of Clinical Trials
AFFIRM Atrial Fibrillation Follow-up Investigation of Rhythm
Surgical Treatment of AF Management
HOT CAFE How to Treat Chronic Atrial Fibrillation
In the initial surgical maze procedure, linear lesions were cre-
PIAF Pharmacological Intervention in Atrial Fibrillation
ated within the atria to reduce effective atrial mass; the reported RACE Rate Control Versus Electrical Cardioversion for
long-term success rate was about 90%. The surgical maze pro- Persistent Atrial Fibrillation
cedure has evolved with the use of limited thoracotomy access RE-LY Randomized Evaluation of Long-Term Anticoagulation
and the use of cryoablative techniques rather than cut-and-sew Therapy
approaches. In the future, these minimally invasive techniques STAF Strategies of Treatment of Atrial Fibrillation
 23
Supraventricular Tachycardia
JOSEPH J. GARD, MD, AMMAR HABIB, MD,
and SAMUEL J. ASIRVATHAM, MD
SVT is a generic term encompassing various arrhythmias, includ-
ing atrial fibrillation, atrial flutter, and paroxysmal SVT. This
chapter focuses on paroxysmal SVT. This arrhythmia is among
the most commonly encountered in general cardiology practice,
and the 3 most common forms are AVNRT, AVRT, and AT.
Clinical Features
Although electrocardiography remains the mainstay for diagnos-
ing the presence and type of SVT, important clues can be ascer-
tained from the history and physical examination. The typical
symptom is palpitations. Several associated symptoms, includ-
ing light-headedness, chest discomfort, an uneasy sensation in
the neck, and an urge to micturate, may accompany the primary
symptom of palpitations. Syncope is uncommon but may occur
in elderly patients with or without coexisting cardiovascular
disease.
The palpitations are of sudden onset and equally abrupt offset
or termination with AVNRT and AVRT. In general, there is no
clear inciting factor or position that patients relate to the onset of
these tachycardias.
In contrast, with automatic AT, the onset is typically gradual
and there is sometimes also a gradual resolution back to normal
sinus rhythm. Some patients may find that a particular position,
maneuver, or inciting factor (such as exercise) triggers AT.
On physical examination during tachycardia with AVNRT and
AVRT, the first heart sound is usually loud, and regular cannon a
waves may be seen in the neck. With AT associated with variable
AV block, irregular cannon a waves and first heart sound of var-
iable intensity may be observed. Multiple fourth heart sounds are
occasionally heard in AT with high-grade AV block.
Abbreviations and acronyms are expanded at the end of this chapter.
• The 3 main SVTs are AVNRT, AVRT, and AT.
• AVNRT and AVRT are reentrant tachycardias with palpitations of
sudden onset and sudden offset.
• AT is an automatic tachycardia with gradual onset and usually
gradual resolution.
Mechanisms
As the name implies, SVTs are arrhythmias with rates more
than 100 beats per minute and have mechanisms requiring non-
ventricular cardiac tissue. Atrial tissue, the AV node, accessory
pathways, or a combination of these may be involved. Each of the
SVTs discussed in this chapter has distinguishing mechanisms
(Figure 23.1).
Atrioventricular Node Reentrant Tachycardia
Although the name AVNRT implies a circuit within the AV node,
this is in fact a misnomer. Parts of the AV node are essential for
this circuit, but the primary components of the circuit involve
atrial myocardium, specifically the fast and slow atrial myocar-
dial inputs to the AV node.
AVNRT is a reentrant tachycardia whose circuit utilizes the
slow and fast pathways to the AV node. These pathways are ana-
tomically and functionally distinct fibers connecting the atrial
tissue to the AV node, which is located on the interatrial septum
close to the tricuspid annulus. Normal AV conduction occurs via
the fast pathway such that the sinus node impulse primarily trav-
els superior to the fossa ovalis and posterior to the eustachian
ridge to reach the AV node. In contrast, the slow pathway is ante-
rior (ventricular) to the eustachian ridge from the region of the
coronary sinus to the AV node.
A premature atrial complex can trigger reentry within the slow
and fast pathways to initiate AVNRT. A premature complex that
281
282 III Electrophysiology

AV node Accessory pathway SA node, atrial

60% 30% 10%

V1 V1
I

Figure 23.1. Regular Narrow QRS Tachycardias. Left, Atrioventricular (AV) node reentry. Because AV activation proceeds from a common
turnaround point in or near AV node, the R wave and P wave may be nearly simultaneous producing a very short R-P interval, and the P wave may
be difficult to discern. Middle, Tachycardia in which an extranodal accessory pathway is used for retrograde conduction (orthodromic AV reentrant
tachycardia). A short R-P interval is seen with an easily discernible retrograde P wave. Right, A long R-P tachycardia in which each P wave is closer
to the succeeding rather than the preceding QRS is characteristic of sinus tachycardia and atrial tachycardia. SA indicates sinoatrial.

occurs early enough to reach the fast pathway during its refrac-
tory period may still conduct via the slow pathway to the AV
node. Once it reaches the AV node, retrograde conduction may
occur back to the atrial tissue via the fast pathway if it has recov-
ered from its refractory period. The echo beat (atrial activation)
may then conduct antegrade via the slow pathway, generating a
reentrant circuit. This is the typical form of AVNRT (slow-fast
reentry), whereby the slow pathway conducts antegrade and then
the fast pathway conducts retrograde.
• The slow and fast pathways are the atrial inputs to the AV node.
• Typical AVNRT is a circuit with antegrade conduction down the
slow pathway and retrograde conduction up the fast pathway.
Atrioventricular Reentrant Tachycardia
Unlike AVNRT and AT, AVRT requires both atrial and ventricu-
lar myocardium for propagation. In other words, AT and AVNRT
may occur in the presence of complete AV block, whereas 1:1
ventriculoatrial association is required for AVRT. This reentrant
tachycardia involves as one limb the AV node and the second
an accessory pathway. Accessory pathways are usually muscle
bundles that traverse the AV annulus. The most common form
of AVRT is orthodromic AVRT. During orthodromic AVRT,
antegrade conduction occurs via the AV node and then the His-
Purkinje system with retrograde conduction via an accessory
pathway. A premature atrial or ventricular complex may trigger
reentry within this circuit. Orthodromic AVRT may terminate as
a result of fatigue in AV node conduction, increased vagal tone
due to vagal maneuvers, or a premature extra systolic beat.
Antidromic AVRT is an uncommon tachycardia in which
reentry occurs with antegrade conduction via an accessory path-
way and retrograde conduction via the AV node. Because the AV
node is a component of the reentrant circuit, AV nodal blocking
agents or vagal maneuvers may terminate the rhythm.
• Accessory pathways may be involved with tachycardia in 3 cir-
cumstances: orthodromic AVRT, antidromic AVRT, and preex-
cited tachycardias.
• The orthodromic AVRT circuit has antegrade AV node and ret-
rograde accessory pathway activation and is a narrow complex
tachycardia.
• The antidromic AVRT circuit has antegrade conduction down the
accessory pathway and up the AV node and is a wide complex
tachycardia.
Atrial Tachycardia
When used as a generic term, AT includes several entities,
including macroreentrant AT, atrial flutter, scar-related AT, and
automatic AT. Tachycardias arising from the pulmonary veins
that can provoke atrial fibrillation are also sometimes referred
to as ATs. Atrial flutter and macroreentrant tachycardias are
discussed elsewhere in this textbook. In this chapter, the focus
is automatic AT. Narrow complex tachycardia (QRS duration
 23 Supraventricular Tachycardia
<120 milliseconds) implies a supraventricular origin for the
arrhythmia with a normal infra-hisian conduction system. With
rare exceptions (fascicular ventricular tachycardia), all ventricu-
lar tachycardias present as a wide complex tachycardia. Once a
narrow complex tachycardia is diagnosed from a 12-lead elec-
trocardiogram, the cardiologist must then determine whether
a paroxysmal SVT, sinus tachycardia, or a macroreentrant AT
is present. A methodical sequence of identifying the P waves,
defining whether a long R-P or short R-P tachycardia is present,
and further analyzing initiation and termination, when present,
increases the likelihood of accurate diagnosis.
Unlike the reentrant mechanism underlying AVNRT and
AVRT described above, the mechanism for automatic AT is a
focus of atrial tissue with inherent automaticity at a shorter cycle
length than that of the sinus node.
Electrocardiographic Recognition
Narrow Complex Tachycardia
A narrow complex tachycardia may represent sinus tachycar-
dia, atrial fibrillation, atrial flutter, or the SVTs discussed in this
chapter, except for antidromic AVRT. Fascicular VT is a rare
exception of a VT that has a narrow complex. It is reviewed in
detail elsewhere in this book, but it is notable that vagal maneu-
vers have been reported to terminate this rhythm.
The important first step in approaching a regular narrow com-
plex tachycardia is to identify the P waves. This allows classifi-
cation of the rhythm as a short R-P or a long R-P tachycardia. A
short R-P tachycardia has an R-P interval that is shorter than the
P-R interval, and, correspondingly, a long R-P tachycardia has
an R-P interval that is longer than the P-R interval. This classi-
fication is commonly used and helpful, but there are numerous
exceptions. In general, AVNRT and AVRT are short R-P tachy-
cardias, whereas AT is a long R-P tachycardia (Figure 23.2).
The R-P interval reflects the retrograde conduction time for
an impulse to travel from ventricular to atrial tissue. A short R-P
interval may occur with rapid retrograde conduction (orthodromic
I
II
III
aVR
aVL
aVF
V1
V2
V3
V4
V5
V6
Figure 23.2. Electrocardiogram From a Patient With Initially Regular Supraventricular Tachycardia and Then a Change in Ventricular Response
Rate. The underlying supraventricular arrhythmia is an atrial tachycardia (automatic or macroreentrant flutter). The abrupt changes in ventricular
responses may exacerbate symptoms, especially in patients already compromised with critical illness.
283
AVRT) or with near simultaneous activation of the ventricular and
atrial tissues (AVNRT and junctional tachycardia) (Figure 23.3).
Furthermore, the R-P interval in orthodromic AVRT (Figure 23.4)
is almost never less than 100 milliseconds because of the neces-
sary time it takes for the impulse to conduct through the ventricu-
lar tissue to reach the accessory pathway. Yet, the R-P interval can
be very short or even negative in AVNRT because there is near
simultaneous activation from a common point near the AV node
to the atrial and ventricular tissue. This very short R-P interval
can lead to the appearance of a pseudo-R’ wave seen when the
P wave occurs in the terminal portion of the QRS complex.
• Both AVRT and AVNRT are short R-P tachycardias. The R-P
interval may be zero or negative in AVNRT, whereas the R-P inter-
val is at least 100 milliseconds in AVRT.
• AT (and sinus tachycardia) are usually long R-P tachycardias.
The R-P interval classification is a helpful approach to nar-
row complex tachycardias, but there are exceptions that should
also be considered. Sluggish retrograde activation of the atrial
tissue can cause AVNRT to appear to be a long R-P tachycardia.
Orthodromic AVRT can appear to have a long R-P interval if
the retrograde conduction via the accessory pathway is slowed
(such as Ebstein anomaly) or if the pathway has previously been
damaged, possibly by a prior procedure, including prior ablation
attempts. A short R-P interval may be seen in AT when conduc-
tion system disease (AV delay) causes considerable delay in the
P-R interval and it is then longer than the R-P interval.
An example of an orthodromic AVRT with a long R-P inter-
val is the permanent form of junctional reciprocating tachycar-
dia. Despite the name, the arrhythmia is not a type of junctional
tachycardia but rather an orthodromic reciprocating tachycardia,
wherein the accessory pathway conducts slowly in a retrograde
fashion. As a result of this slow conduction, the arrhythmia itself
is relatively slow and may be mistaken for sinus tachycardia.
The arrhythmia is often incessant, and patients may present with
tachycardia-related cardiomyopathy because palpitations may
not be as prominent with the relatively slower rates.
284 III Electrophysiology

Figure 23.3. A 12-Lead Electrocardiogram of Typical Atrioventricular Nodal Reentrant Tachycardia (AVNRT). The P waves are readily recog-
nized just following the QRS complex. The regular tachycardia with short R-P interval should raise suspicion for this arrhythmia. The P waves are
typically very narrow in AVNRT as a result of the early septal activation during this tachycardia. Atrioventricular nodal blockade will terminate the
arrhythmia and likely prevent recurrence. This arrhythmia may be hemodynamically poorly tolerated even when relatively slow because of the near
simultaneous atrioventricular activation. This results in atrial contraction against a closed atrioventricular valve, producing increased back pressure
in the venous beds (systemic and pulmonary).

Figure 23.4. Orthodromic Atrioventricular Reentrant Tachycardia. When an extranodal accessory pathway is present, the most common arrhyth-
mia is orthodromic reciprocating tachycardia. Conduction occurs down the normal atrioventricular conduction system and up the accessory pathway,
producing a short R-P tachycardia with an R-P interval typically more than 100 milliseconds (arrow indicates p wave).
 23 Supraventricular Tachycardia 285

Wide Complex Tachycardia
SVT may present with a wide QRS in 2 circumstances: 1) when
bundle branch block exists during tachycardia and 2) when ante-
grade conduction occurs by an accessory bypass tract. The criti-
cal first steps in the assessment of a wide complex tachycardia
are to assess the patient’s stability and to determine whether the
tachycardia is VT or an SVT with a wide QRS complex due to
aberrancy or conduction via an accessory pathway. A detailed
review of VT and electrocardiographic distinction from SVT is
presented elsewhere in this textbook. Briefly, morphologic pat-
terns of AV dissociation, the presence of fusion or capture beats,
a markedly widened QRS (more than 140–160 milliseconds),
precordial concordance, northwest axis, and atypical bundle
branch block all favor VT rather than SVT with aberrancy.
• A wide QRS tachycardia may result from VT, SVT with bundle
branch block, or SVT with preexcitation.
• Precordial concordance with all R waves in the chest leads is con-
sistent with VT or an antidromic tachycardia.
• Precordial concordance with all S waves in the chest leads is con-
sistent with VT (arising near the apex).
• Precordial concordance is not seen with bundle branch block.
Wide QRS Complex Due to Bundle
Branch Block
An SVT may present with a wide complex tachycardia if there is
either a baseline bundle branch block or if a bundle branch block
develops with tachycardia (rate-related aberrancy). The initial
deflection of the QRS complex is rapid and appears normal; how-
ever, the later part of QRS complex is abnormal, reflecting the
intraventricular conduction delay due to the bundle branch block.
The ventricular activation pattern reflects the exit of the conduct-
ing bundle being midway between the base and apex such that
concordance will not be seen in the precordial leads (V1-V6).
Typically, there is a positive R wave in lead V1 (right bundle
pattern), but there are prominent negative S waves in leads V4
through V6. Finally, AV dissociation almost never occurs with
AVRT and is very rare with AVNRT. The widened QRS complex
may obscure the retrograde P wave in very short R-P tachycar-
dias such as AVNRT. When AV dissociation occurs with AT,
there are usually more P waves than QRS complexes. Thus, AV
dissociation with more QRS complexes than P waves suggests
that VT is the cause of the wide complex tachycardia.
Wide QRS Complex Due to Antegrade
Preexcitation
SVT may present with a wide QRS complex due to antegrade
preexcitation in 2 circumstances. First is antidromic AVRT
(Figure 23.5) in which ventricular activation occurs by ante-
grade conduction via an accessory pathway and reentry occurs
with retrograde conduction via the AV conduction system. The
resulting QRS complexes are wide because ventricular activa-
tion is entirely mediated by conduction of the accessory path-
way bypassing the specialized conduction system. Because the

Figure 23.5. Antidromic Reciprocating Tachycardia. When an extranodal accessory pathway is present, a regular wide complex tachycardia may
also result (antidromic reciprocating tachycardia). Here, the tachycardia circuit proceeds antegrade down the accessory pathway and up through the
atrioventricular nodal conduction system, producing a regular wide QRS tachycardia in which the QRS morphologic pattern depends on the site of
the accessory pathway.
286 III Electrophysiology
I
II
III
aVR
aVL
aVF
V1
V2
V3
V4
V5
V6
Figure 23.6. Regular Wide Complex Tachycardia. When a regular wide complex tachycardia is seen in the critical care setting, ventricular tachy-
cardia should always be considered. However, if the baseline electrocardiogram shows preexcitation, an antidromic tachycardia can be diagnosed
and easily terminated with any atrioventricular nodal blocking agent. If the baseline electrocardiogram is not available, wide QRS tachycardia with
consistent 1:1 R-P association in the absence of structural heart disease should raise suspicion for an accessory pathway-mediated mechanism.
insertion of most accessory pathways is basal near the annulus,
the resulting QRS vector is from base to apex. The wide QRS
complexes are regular, a reflection of the reentrant mechanism
underlying antidromic AVRT (Figure 23.6).
The second circumstance in which an SVT may present with
a wide QRS complex due to antegrade preexcitation is when
primary atrial tachyarrhythmia is not dependent on the acces-
sory pathway but does conduct to the ventricle via the accessory
pathway. For example, atrial fibrillation, atrial flutter, or AT may
conduct to the ventricles by both the AV node and the acces-
sory pathway. The result is a variably wide QRS complex during
tachycardia. Slowing AV nodal conduction will enhance preexci-
tation and can be dangerous because accessory pathways gener-
ally do not demonstrate decremental conduction. Thus, AV nodal
blocking agents are contraindicated in preexcited tachycardias
other than antidromic AVRT.
Treatment Options
Various treatment options are available for both the acute and
the long-term management of SVT. These treatment approaches
may be conservative, medical, or invasive. Patient preferences
and frequency and severity of symptoms are important factors
when considering long-term treatment options.
Treatment Options for AVNRT
AVNRT typically affects younger patients (aged 25–35 years)
and is slightly more common in females. It is generally not
considered a life-threatening rhythm unless there is substan-
tial coexisting cardiovascular disease. AVNRT is not associated
with an increased risk of atrial fibrillation, tachycardia-mediated
cardiomyopathy, stroke, or mortality. Thus, long-term treatment
is focused on symptomatic management. Depending on the fre-
quency and severity of symptoms, patients may be better suited
to conservative, medical, or invasive therapies.
If managed conservatively, then episodes may be treated on an
as-needed basis. Because the AV node is critical to its underlying
reentrant mechanism, vagotonic maneuvers, such as a Valsalva
maneuver, may be effective for terminating the tachycardia.
Patients can be instructed on how to perform these independently,
as needed (Figure 23.7). If these are ineffective, AV nodal block-
ing agents may be considered. Adenosine can be considered in
acute settings with appropriate cardiac monitoring. The medical
options to reduce the frequency of tachycardia episodes include
β-adrenergic blockers, calcium channel blockers, and possibly
digoxin. Some patients may continue to have symptoms or be
intolerant of adverse effects with medical treatment and prefer
an invasive ablative approach. Younger patients may also pre-
fer an initial ablative approach over taking medication over the
long term. Radiofrequency ablation has an excellent success rate
(95%) for eliminating tachycardia episodes and a low risk (2%)
of serious complications, such as AV block. Current practice tar-
gets the slow pathway for ablation. Fast pathway ablation is no
longer performed because of a higher incidence of AV block.
• AV nodal blocking agents are effective in managing AVNRT.
• Radiofrequency ablation is highly successful, and the slow path-
way input to the AV node is targeted for ablation.
Treatment Options for Accessory
Pathway-Related Tachycardias
Tachycardias mediated by accessory pathways include ortho-
dromic AVRT, antidromic AVRT, and other arrhythmias whose
genesis is dependent on an accessory pathway. As noted previ-
ously, slowing conduction in the AV node with vagal maneuvers
or AV nodal blocking agents can be effective for terminating
orthodromic and antidromic AVRT. Other tachycardias with
accessory pathway conduction should never be treated with AV
nodal blocking agents. Variation of the QRS morphologic pattern
during tachycardia is a diagnostic red flag that AV nodal agents
be absolutely avoided. Electrical cardioversion should be consid-
ered if the patient is hemodynamically unstable. If the patient is
hemodynamically stable and has normal left ventricular systolic
 23 Supraventricular Tachycardia 287

I
II
III
aVR

aVL
aVF
V1
V2
V3
V4
V5

V6

Figure 23.7. Abrupt Termination of a Narrow Complex Tachycardia During a Valsalva Maneuver. Note that a short R-P tachycardia with ret-
rograde P wave is seen just after the QRS interval. Termination of a tachycardia after a P wave suggests that the atrioventricular node is essential
for tachycardia perpetuation. Thus, atrial tachycardia can be virtually excluded. The very short R-P interval is diagnostic of atrioventricular node
reentry.

function, then intravenous treatment with flecainide, ibutilide, or
procainamide can be considered.
Catheter ablation is a treatment option for tachycardias
involving accessory pathways. The technical complexity and
risk associated with catheter ablation are largely determined by
the location of the accessory pathway. Careful analysis of the
pattern of preexcitation on the electrocardiogram (Figure 23.8)
can provide clues to the location of the accessory pathway. Exact
localization is beyond the scope of the general cardiology board
examination; however, an important skill on the examination
is recognition of a right bundle branch block pattern (positive
R wave in lead V1) in left-sided pathways and of a left bundle
branch block pattern (QS complex in lead V1) in right-sided
pathways.
Treatment Options for AT
Although AT may occur in the absence of structural heart
disease, it more commonly occurs in the setting of comorbid
heart disease. Particularly in the setting of significant coexist-
ing cardiac disease, AT may present acutely with hemody-
namic instability, in which case urgent cardioversion should be

I
II
III
aVR

aVL
aVF
V1
V2
V3
V4
V5

V6

Figure 23.8. Characteristic Electrocardiogram for Patient With Antegrade Preexcitation. Note the short P-R interval and the delta wave clearly
seen in the lateral precordial leads and lead II. The R wave seen in lead V1 and negative delta wave in lead I are consistent with the left-sided acces-
sory pathway. If atrial fibrillation develops in a patient with this baseline electrocardiogram, it should be treated as a medical emergency because of
the risk of ventricular fibrillation from the atrial fibrillatory waves conducting to the ventricle via this pathway without the intervening protective
effects of the atrioventricular node.
288 III Electrophysiology

considered. Electrical cardioversion is less effective in automatic 12-lead electrocardiogram is useful for localizing the focus
AT than in AT due to microreentry or triggered activity. Vagal generating the AT. The most common locations are in the right
maneuvers are rarely effective for terminating AT. Some ATs atrium, specifically the tricuspid annulus and crista terminalis.
are responsive to adenosine, β-adrenergic blockers, or calcium
channel blockers, although, as previously noted, these are con- Acknowledgement
traindicated if an accessory pathway is present. More aggressive
intravenous antiarrhythmic treatment options may include class The authors wish to acknowledge the contribution of Traci L.
IA and IC antiarrhythmic medications given under the direction Buescher, RN in preparation of the figures.
of a specialist. The approach to long-term management of AT
is determined by the frequency and severity of symptoms dur- Abbreviations
ing tachycardia and by the potential impact on coexisting heart
AT atrial tachycardia
disease. Again, if there is not an accessory pathway, treatment AV atrioventricular
with AV nodal blocking agents, such as β-adrenergic blockers AVNRT atrioventricular nodal reentrant tachycardia
and calcium channel blockers, can be attempted. If these fail, AVRT atrial ventricular reentrant tachycardia
cannot be tolerated, or are contraindicated, then other antiar- SVT supraventricular tachycardia
rhythmic agents or catheter ablation might be considered. The VT ventricular tachycardia

Ventricular Tachycardia and Ectopy a
THOMAS M. MUNGER, MD
Introduction
A relatively common arrhythmia in the United States, VT is
present in 1% to 2% of patients during the first year after MI.
This is the largest subgroup of patients with VT who require
treatment. Since the early 1980s, surgical and catheter abla-
tive techniques—using direct current and, later, radiofrequency
energy—have been used in an attempt to eradicate reentrant foci
that are responsible for VT in patients who have CAD. In addi-
tion to these patients, there are subsets of patients who have VT
that occurs in the absence of CAD (ie, with dilated cardiomyopa-
thy, ARVD, HCM, infiltrative diseases of the heart, and congen-
ital heart disease). Additionally, VT can emerge in patients with
structurally normal hearts; these patients may or may not have
associated repolarization syndromes. VT is an arrhythmogenic
cause of sudden death (Box 24.1), although in patients with nor-
mal repolarization and structurally normal hearts, sudden death
is rarely a complication.
Wide Complex Tachycardia and VT
VT rhythms have rates between 100 and 280 beats per minute and
can be monomorphic or polymorphic. Monomorphic rhythms
keep QRS morphology consistent from beat to beat and generally
keep rates constant. Warm-up and cooldown changes in rate can
occur, however, particularly in rhythms not caused by myocar-
dial reentry (triggered or automatic mechanisms). Polymorphic
VT rhythms are most frequent in patients with repolarization
a
Portions previously published in Hammill SC, Hubmayr RD. The
rapidly changing management of cardiac arrhythmias. Am J Respir Crit
Care Med. 2000 Apr;161(4 Pt 1):1070–3. Used with permission.
Abbreviations and acronyms are expanded at the end of this chapter.
24
abnormalities (LQTS, Brugada syndrome, acute myocardial
ischemia, myocardial inflammation such as myocarditis, or drug
toxicity); when the polymorphic form of VT turns about the iso-
electric line in a regular pattern, the VT is termed torsades de
pointes. VT generally has a wide QRS complex (fascicular VT
rhythms are an exception, with QRS durations <120–140 ms) on
the surface ECG.
Wide complex tachycardias encompass several different
arrhythmias, including paroxysmal supraventricular tachycar-
dia conducted aberrantly or anterogradely across an anomalous
AV pathway (eg, in Wolff-Parkinson-White syndrome or when
Mahaim fibers are present), VT, drug or electrolyte toxicity–
related arrhythmias, and device-related arrhythmias and arti-
facts (Box 24.2). Several clinical and ECG features have been
identified that favor VT as the cause of wide complex tachycardia
(Box 24.3).
In studies of patients with undifferentiated, wide QRS tachy-
cardia, patients with a clinical history of structural heart disease
had more than a 95% chance of having VT. A variable first heart
sound, cannon a waves, and identification of AV dissociation on
the ECG identify the presence of a ventricular rhythm indepen-
dent of the atrial rhythm, which is indicative of VT. Fusion beats,
which identify simultaneous activation of the ventricles through
the normal conduction system and an ectopic ventricular origin,
are also indicative of VT; capture beats are normal sinus beats
with a narrow QRS complex during VT in AV dissociation. With
a criterion of QRS duration longer than 160 ms, VT discrimina-
tion achieves 97% specificity. ECG examples of wide complex
tachycardias from various mechanisms are shown in Figures 24.1
through 24.4. The current American Heart Association and
advanced cardiac life support guidelines for the immediate man-
agement of wide complex tachycardia, monomorphic VT, and
polymorphic VT are shown in Figure 24.5.
289
290 III Electrophysiology

Box 24.1. Cardiac and Vascular Causes of Sudden Box 24.2. Wide Complex Tachycardias
Death Ventricular tachycardia
Coronary artery conditions Monomorphic
Atherosclerosis Polymorphic
Embolism
Aberrancy with PSVT (LBBB/RBBB/IVCD)
Dissection
Sinus tachycardia
Spasm
Junctional tachycardia
Coronary artery anomalies
Typical AVNRT
Myocardial diseases Atypical AVNRT
Dilated cardiomyopathy Orthodromic reciprocating tachycardia
Hypertrophic cardiomyopathy
Antidromic reciprocating tachycardia
Infiltrative disease (eg, amyloid or sarcoid)
Multiple pathway tachycardia
Myocarditis
Atrial fibrillation
Arrhythmogenic right ventricular dysplasia
Atrial flutter
Tumors
Ectopic atrial tachycardia
Primary arrhythmias Reentrant atrial tachycardia
Wolff-Parkinson-White syndrome
Accessory pathway with PSVT
Long QT syndrome
Sinus tachycardia
Catecholinergic polymorphic ventricular tachycardia
Typical AVNRT
Brugada syndrome
Atypical AVNRT
Blunt chest trauma (commotio cordis)
Orthodromic reciprocating tachycardia
Primary atrial arrhythmias
Bradycardias Antidromic reciprocating tachycardia
Multiple pathway tachycardia
Valvular diseases
Atrial fibrillation
Aortic stenosis
Atrial flutter
Mitral valve prolapse
Ectopic atrial tachycardia
Congenital (eg, tetralogy of Fallot)
Reentrant atrial tachycardia
Vascular or pericardial causes
Toxicity-related tachycardia
Aortic aneurysm or dissection
Electrolyte induced
Marfan syndrome
Drug induced
Pulmonary aneurysm or dissection
Pulmonary embolism Other tachycardia
Tamponade Pacemaker-mediated tachycardia
Primary pulmonary hypertension Artifactual tachycardia
Subarachnoid hemorrhage
Abbreviations: AVNRT, atrioventricular node reentry tachycardia;
Massive bleeding in gastrointestinal tract
IVCD, interventricular conduction delay; LBBB, left bundle branch
Drugs in circulatory system: torsades de pointes block; PSVT, paroxysmal supraventricular tachycardia; RBBB, right
Toxins in circulatory system bundle branch block.

Distinct Clinical Syndromes of VT

Ischemia and Acute Post-MI VT
In the early 1900s, case reports began to identify VF as a cause of
sudden death after the onset of MI. In the 1940s, it was noted that
the VF threshold in dogs decreased after coronary occlusion. The
incidence of VT and VF in the first 6 to 12 hours after MI was
first described in the 1960s after the institution of coronary care
units. The frequency of ischemic episodes increases in the early
morning and decreases as evening approaches. A similar pattern
is associated with the onset of MI and ventricular arrhythmias.
During the first 5 minutes of coronary occlusion, depolari-
zation of the ventricular myocytes develops from approximately
−90 mV to −60 mV because of an increase in intracellular calcium
levels and the activation of the IK ATP (which causes an increase in
extracellular potassium levels and a subsequent membrane depo-
larization). The sodium-potassium pump also fails to function
within the first 10 to 15 minutes after coronary occlusion, with a
subsequent excess accumulation of extracellular potassium. The
net rate of change of extracellular potassium is 1 to 1.5 mEq/L per
minute in the first 10 minutes of ischemia. Coincident with extra-
cellular augmentation of potassium is conduction slowing. At 15
minutes, myocytes at the central injury zone are electrically silent
as necrolysis begins. Lysophosphoglycerides are released from
cellular membranes. These agents electrically uncouple ventricu-
lar myocytes on the border zones of myocardial infarcts, thereby
exacerbating further conduction slowing. Local acidosis exacer-
bates these effects. Catecholamines are released locally. Distal
sympathetic fibers within the infarct zone undergo necrosis.
 24 Ventricular Tachycardia and Ectopy
Box 24.3. Features Favoring VT Over PSVT
Clinical
History of coronary artery disease or heart failure
Cannon a waves in jugular venous pulse contour
Variable first heart sound on auscultation
Electrocardiographic
AV dissociation
Fusion beats
Capture beats
QRS duration >140–160 ms
Precordial concordance
Northwest axis
Lead V1 RBBB with larger left peak
Lead V6 QRS with rS or S morphology
Abbreviations: AV, atrioventricular; PSVT, paroxysmal supraven-
tricular tachycardia; RBBB, right bundle branch block; VT, ventricular
tachycardia.
During the first 72 hours after MI, the incidence of PVCs,
nonsustained VTs, and sustained VTs increases steadily. With
sympathetic denervation in animals, ventricular ectopy can be
decreased or eliminated in some cases. The excess ventricular
ectopy during this time is caused by increased automaticity.
Ventricular ectopy (most often originating in the subendocardial
Purkinje fibers of the left ventricle) can be altered by varying
afterload, sheer stress on the infarct, collateral blood supply, and
patency of the infarct artery. Border zone ischemic cells remain
injured for several weeks after the healing phase of MI. These
cells typically have depolarized resting membrane potentials,
slowed conduction, and longer action potential phases. Complex
ectopy and the presence of late potentials on the signal-averaged
I aVR
II aVL
III aVF
Figure 24.1. Wide Complex Tachycardia Due to Aberrancy. Electrocardiogram shows orthodromic reciprocating tachycardia with right bundle
branch block. Note the “typical” rSR′ right bundle branch block pattern and relatively narrow QRS complex in lead V1.
291
ECG become more frequent until 6 weeks after MI and then
decrease thereafter. This time course correlates with the inci-
dence of inducibility at EP testing and the incidence of sudden
death following the very acute early phases after MI.
PVCs are absent in at least half of the patients who have VF
in the acute phase after MI and are as frequent in patients who
do not have VF. Because PVCs are not accurate predictors of VF
occurrence, prophylactic antiarrhythmics have no role in these
patients. Meta-analysis has shown that intravenous lidocaine
increases mortality when given prophylactically to all post-MI
patients. Revascularization strategies and use of β-blockers make
more sense in the early acute phase after MI. Accelerated idio-
ventricular rhythms occur at rates less than 100 to 120 beats per
minute and are noted in more than 90% of patients who undergo
successful revascularization. These rhythms carry no adverse
prognostic implications and do not require treatment.
Monomorphic VT is extremely uncommon during the first 48
hours after MI and is generally regarded as a triggered or auto-
matic rhythm. However, after the first 48 hours, most monomor-
phic VTs are due to scar reentry and are associated with higher
mortality rates. The short-term treatment of VT depends on the
hemodynamic status of the patient. If a patient is hemodynam-
ically unstable—with angina, shortness of breath, presyncope,
syncope, and hypotension—immediate sedation and synchro-
nized direct current cardioversion should be given. If the patient
is hemodynamically stable, intravenous amiodarone or lidocaine
is appropriate. The patient should be evaluated aggressively for
residual CAD with coronary angiography and treated if feasible.
A patient with late-onset VT should receive an ICD. Catheter
ablation or aneurysmectomy should be considered for continued
VT despite medical and device therapies.
VT or Sudden Death in the Patient With
CAD or Heart Failure
Animal experimentation, surgical mapping data, and pacing
interventions (eg, resetting, manifest entrainment, and concealed
entrainment) performed during VT have definitively shown that
V1 V4
V2 V5
V3 V6
292 III Electrophysiology

I aVR V1 V4

II V2 V5
aVL

III aVF V3 V6

Figure 24.2. Wide Complex Tachycardia Due to Slow Ventricular Tachycardia. Electrocardiogram is from a 79-year-old man. Note the northwest
axis, wide QRS complex, single R wave in lead V1, and prominent S wave in lead V6.

myocardial reentry is the cause of VT in patients with CAD.
This reentry involves “slow zone” areas of conduction that are
either adjacent to old MI scars or within the scars themselves.
Most infarct scars are not homogeneous collections of connec-
tive tissue but rather areas of scar tissue intermixed with inter-
digitating surviving myocardial fibers, which are poorly coupled
together electrically; the scars serve as excellent slow conduc-
tion zones (Figure 24.6). During EP studies, electrical signals
from such areas display low amplitude, complex fractionation,
and increased duration of electrical activity. On occasion, such
signals can appear biphasic or triphasic and can appear in the
mid-diastolic interval. On the surface ECG, this is manifested as
late potentials when analyzed with signal-averaged ECG.
These sites of slow conduction provide targets for ablative
techniques. The surgical techniques of subendocardial resection
and aneurysmectomy were used, particularly in the 1980s, for
eradicating VT in patients with CAD. The overall mortality from
the procedures was 5% to 8%, but cure rates were in excess of
70% to 80%. As early as 1983, direct current catheter ablation
was also used for treatment of VTs in patients with CAD; unfor-
tunately, the success rate for abolishing VT in these patients
was no better than 50%. Complications included pericardial

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

Figure 24.3. Idiopathic Polymorphic Ventricular Tachycardia From Left Ventricular Purkinje Condition System. This patient did not have long
QT syndrome. Note the alternating left bundle branch block and right bundle branch block patterns to the premature ventricular contractions.
 24 Ventricular Tachycardia and Ectopy 293

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

Figure 24.4. Antidromic Reciprocating Tachycardia Simulating Ventricular Tachycardia. This patient had a left-sided accessory bypass tract
(Wolff-Parkinson-White syndrome). Note the northwest axis and the duration of the QRS complex.

WCT
identified

Unstable
Stable hemodynamics
hemodynamics
Prepare for DCCV

Esophageal lead
Clinical parameters Monomorphic Polymorphic
12-Lead ECG VT VT
To facilitate diagnosis

Long QT interval
Preserved LV
Correct electrolytes
Identified SVT function
Therapy: magnesium,
or VT Therapy: procainamide,
overdrive pacing,
Treat these stable sotalol,
isoproterenol,
rhythms amiodarone, or
phenytoin, or
lidocaine
lidocaine

Normal QT interval
Unknown WCT Correct electrolytes
Impaired LV
Therapy: DCCV or function Treat ischemia
amiodarone (or Therapy: IV amiodarone Therapy: β-blockers,
procainamide if (or IV lidocaine) and amiodarone,
preserved LV then DCCV procainamide,
function) lidocaine, or
sotalol

Figure 24.5. American Heart Association and Advanced Cardiac Life Support Guidelines for Wide Complex Tachycardia (WCT). DCCV indicates
direct current cardioversion; ECG, electrocardiogram; IV, intravenous; LV, left ventricular; SVT, supraventricular tachycardia; VT, ventricular tachycardia.
294 III Electrophysiology
Figure 24.6. Myocardial Infarction Scars. Infarcted heart specimen (left), adjacent to normal heart specimen (right), shows nontransmural
(arrow) and transmural (arrowhead) scarred areas, which are potential zones of slow conduction for reentry.
tamponade, electrical mechanical dissociation, permanent heart
block, and cardiac disruption. The procedure-related mortality
rate among patients who underwent the procedure with direct
current techniques in the 1980s was approximately 6%.
Because of the considerable morbidity and mortality asso-
ciated with direct current catheter ablation of VT in patients
with CAD, radiofrequency catheter techniques were adopted.
Generally, ablation of VT with radiofrequency techniques (large
tip or cool tip, at times supplemented with epicardial access) is
thought to have a success rate similar to that of direct current
catheter ablation (50%–70%) but with less risk to the patient.
Ablation procedures should continue to be considered as pal-
liative treatment and as an adjunct to the ICD in the symptom-
atic patient who has CAD with VT. Most of these patients who
undergo VT radiofrequency catheter ablation already have an
ICD and have had no response to multiple antiarrhythmic drugs
(often including amiodarone) for suppression of frequent anti-
tachycardia pacing or shock therapies. With the high recurrence
rate of ventricular arrhythmias that have morphologies different
from the original VT, most patients cannot be considered cured
if a successful radiofrequency ablation was used as the primary
therapy. Radiofrequency ablation remains the primary therapy
for patients who present with incessant VT despite continuous
antiarrhythmic therapy, recurrent antitachycardia pacing, or
shock therapies from an ICD. Activation mapping can be done
if the VT is reproducibly inducible, although noninducible and
hemodynamically unstable VTs can be ablated with the use of
scar mapping (Figure 24.7).
Several medical therapies that are appropriate after MI
for prophylaxis of sudden death include aspirin, β-blockers,
angiotensin-converting enzyme inhibitors, and statins. Multiple
studies have documented the beneficial prophylactic effects of
ICD use in CAD patients with impaired left ventricular func-
tion, positive EP studies, and clinical heart failure. These pri-
mary prevention trials are summarized in Table 24.1. Some trials
required previous EP testing (MADIT and MUSTT), whereas the
MADIT-II and SCD-HeFT required only impaired ejection frac-
tion and a clinical history of MI or heart failure. As lower-risk
groups of patients have been investigated, the number needed to
treat for a similar benefit has steadily increased. Newer nonin-
vasive risk stratification testing (eg, T-wave alternans) has been
advocated to further risk stratify prospective ICD candidates.
Bundle Branch Reentry VT
Bundle branch reentry VT is perhaps the easiest of all monomor-
phic VT rhythms to ablate with catheter techniques. The tachy-
cardia is generally of left bundle branch morphology with a far
left axis deviation (Figure 24.8). The rhythm can be replicated by
pacing the right ventricle near the terminus of the right bundle at
the apex. It is truly a macroreentrant tachycardia, with activation
proceeding anterogradely over the right bundle and transeptally
to the left ventricular apex, returning retrogradely over the left
bundle system, and then connecting back again at the junction
of the bundles to proceed again anterogradely over the right bun-
dle. The critical portions of the circuit include the left and right
bundles and the ventricular septum. The His bundle is not part of
the circuit, although it is very common to note 1:1 retrograde His
activation during this tachycardia. It is estimated that in perhaps
5% of patients with CAD and in up to 50% of patients with dilated
 24 Ventricular Tachycardia and Ectopy 295

Figure 24.7. Mapping of Scarred Myocardium. Voltage maps show infarcted myocardium (red) adjacent to normal myocardium (purple) in a
patient with an antero-infero-septo-apical myocardial infarction and reentrant ventricular tachycardia (VT). Left, Right anterior oblique (RAO) view.
Right, Left anterior oblique view.

cardiomyopathy, this is the mechanism for clinical monomorphic
VT. Typically, a patient presenting with bundle branch reentry
VT has dilated cardiomyopathy, preexisting conduction disease
(manifested on the ECG as LBBB or nonspecific intraventricular
conduction delay), a clinical history of either cardiac arrest or
(more commonly) syncope, and an inducible, very rapid, mon-
omorphic VT at EP testing. Usually there is an associated pro-
longed H-V interval at baseline since conduction delay is critical
to maintenance of the circuit. Ablation of the right bundle cures
this VT, although most patients also require ICD placement for
prophylactic indications.
VT in the Patient With Congenital Heart Disease
Late ventricular arrhythmias are a source of concern and late
morbidity and death among patients with congenital heart
disease. VT contributes to cardiac arrest in these patients. In par-
ticular, after repair of tetralogy of Fallot, patients have frequent
ventricular arrhythmias that arise from either the ventriculo-
tomy scar or the patch site in the RVOT. The incidence of late
sudden death among these patients ranges from 1% to 5% dur-
ing mean follow-up of 7 to 20 years postoperatively. Right ven-
tricular failure, pulmonary hypertension, and number of years
postoperatively are predictive factors for ventricular arrhyth-
mias in these patients. Very limited data exist concerning the
EP characteristics of this group or the comparative efficacy of
the treatments, including pharmacologic therapy, defibrillation,
and ablation.
VT in ARVD, HCM, and Infiltrative
Diseases of the Heart
ARVD, HCM, and infiltrative diseases of the heart have distinc-
tive underlying pathologic features, but they share a common
mechanism for ventricular arrhythmias—myocardial reentry
that is usually of multiple sources. Lethal ventricular arrhythmias

Table 24.1. Primary Prevention of Sudden Cardiac Death in Patients With Structural Heart Disease and
an ICD
3-Year
Inclusion Criteria Mortality, %

Patients Saved Per

Study, Year of No. of No 1,000 Treated Per
Publication Patients CAD DCM Other LVEF, % ICD ICD Year With ICD
MADIT, 1996 196 + − VTns + EP ≤35 41 16 83
MUSTT, 1999 704 + − VTns + EP ≤40 35 16 63
No response to drug
MADIT-II, 2002 1,232 + − Prior MI ≤30 31 22 30
SCD-HeFT, 2005 2,521 + − CHF ≤35 22 17 17
Abbreviations: −, absent; +, present; CAD, coronary artery disease; CHF, congestive heart failure; DCM, dilated cardiomyopathy; EP,
electrophysiologic study; ICD, implantable cardioverter-defibrillator; LVEF, left ventricular ejection fraction; MI, myocardial infarction;
VTns, nonsustained ventricular tachycardia.
296 III Electrophysiology
II
aVF
V1
H H
RFdist
V V
RVA
Power
Figure 24.8. Bundle Branch Reentry Ventricular Tachycardia (VT). Initiation occurred with triple extrastimuli (paper speed, 100 mm/s). Note
the 1:1 association of His bundle (H) activation to ventricular (V) activation (left bundle branch block with a left axis deviation VT). A indicates atrial
pacing spike; RFdist, ablation catheter near bundle of His; RVA, right ventricular apex.
are a complication of all these disorders, including infiltrative
diseases and inflammatory diseases of the heart such as myocar-
ditis, sarcoidosis, amyloidosis, and cardiac tumors.
ARVD, a special type of cardiomyopathy, chiefly affects the
right ventricle, is present more often in younger patients, and
has a familial pattern of inheritance. The right ventricle is gen-
erally large, baggy, and thin, with a depressed ejection fraction
that often is not apparent in the left heart. Pathologically, the
right ventricular walls are infiltrated with adipose tissue, provid-
ing a source for slow ventricular conduction and thus myocardial
reentry (Figure 24.9). The diagnosis can be made with echocar-
diography, computed tomography, magnetic resonance imaging,
or endocardial biopsy. The VT is generally monomorphic with
LBBB morphology, and it can be confused with the more benign
RVOT VT variant. In these patients, the outcomes of surgical
and ablative procedures, as adjuncts to ICD and medical thera-
pies, have been variable, with success rates no greater than 50%
to 60% for a given patient.
HCM is characterized by myocardial fiber disarray (Figure
24.10), which contributes to disorganized myocyte cellular elec-
trical coupling, providing the substrate for myocardial reentry
and lethal polymorphic ventricular arrhythmias. Nonsustained
VT is present on Holter monitoring examinations 25% of the
time. For sustained ventricular arrhythmias, amiodarone is the
drug of choice. In general, for patients who have spontaneous
ventricular arrhythmias that have caused cardiac arrest or syn-
cope or for those with a positive family history of sudden cardiac
death, ICD implantation is recommended and possibly adjunc-
tive antiarrhythmic therapy.
Familial Channelopathy Syndromes and VT
Patients with familial channelopathy syndromes typically have
hearts with morphologically normal structure, abnormalities
of ion channel configuration, and poor repolarization reserves,
H
H
V V
A
making them susceptible to polymorphic VTs. The EAD
(Figure 24.11), which occurs during phase II and phase III of an
action potential, is the cellular event common to these rhythms.
EADs are easily produced experimentally by the inhibitor of the
IK, among others.
On the surface ECG, EADs produce polymorphic PVCs and,
when regenerated, can produce polymorphic VT or torsades de
pointes. Drugs that experimentally produce EADs clinically
prolong the QT interval in animals and humans; these same
agents also produce polymorphic VTs. EADs can be suppressed
experimentally with increased extracellular potassium levels,
acetylcholine (which hyperpolarizes the transmembrane poten-
tial), antiarrhythmic drugs such as lidocaine (which suppresses
the sodium window current, a slow sodium current that is also
responsible for prolongation of phase II), magnesium, β-blockers,
pacing to increase heart rate, and potassium channel openers
(which enhance potassium currents and thus hasten phase III).
EADs are the cellular cause for most proarrhythmia phenomena
seen with antiarrhythmic drugs and other noncardiac drugs (eg,
terfenadine, pentamidine, and erythromycin). During EP testing,
triggered rhythms can often be initiated and terminated with pro-
grammed stimulation, much like reentrant rhythms.
The congenital LQTS is characterized by long QT inter-
vals and episodes of paroxysmal VT that can produce syncope
and even cardiac arrest, particularly in younger persons. LQTS
affects women more often than it affects men. Autosomal domi-
nant (Romano-Ward syndrome) and recessive forms of the dis-
ease are recognized. Half of all LQTS-associated deaths occur
among patients younger than 20 years, and a relationship with
sudden infant death syndrome has been proposed. A molecular
diagnosis of LQT1 has been made post mortem in at least one
infant with sudden infant death syndrome. Higher-risk persons
include those who have family members who have died of sud-
den cardiac death and those who have syncope or QT intervals
in excess of 600 ms. Certain genotypes have been linked to
 24 Ventricular Tachycardia and Ectopy
Figure 24.9. Adipose Replacement of Right Ventricular Myocardium
by Arrhythmogenic Right Ventricular Dysplasia. Specimen is from a
19-year-old patient who died of sudden cardiac death. Fatty replacement
(arrow) provides anatomical substrate for ventricular tachycardia.
Figure 24.10. Myocardial Fiber Disarray. This is characteristic of patients who have hypertrophic cardiomyopathy and ventricular arrhythmias.
297
particular clinical symptoms (eg, drowning is associated with
sudden cardiac death with LQT1). β-Blockers, mexiletine, auto-
nomic surgery, pacing, and ICD therapy have all been used in
various groups of patients with this syndrome. Avoidance of
medications that can aggravate this problem is imperative.
KVLQT1 mutations of the SCN5A sodium channel (resulting
in a diminished current, in contrast to LQT1) cause the Brugada
phenotype and are also associated with polymorphic VT. Like
LQTS patients, certain patients with asymptomatic Brugada syn-
drome have been identified with characteristics meriting prophy-
lactic ICD implantation. (For further discussion see Chapter 20,
“HeritableCardiomyopathiesandChannelopathies:Clinical
Presentations,Genetics,andImplicationsofGeneticTesting.”)
Idiopathic VT in Patients With a Structurally
Normal Heart and No Repolarization
Abnormalities
Idiopathic VT causes less than 5% of VT cases in the general
population; nonetheless, it is an important entity to identify since
it is readily treatable. Unlike ventricular arrhythmias associated
with myocardial scar (eg, patients with CAD, cardiomyopathy, or
myocarditis), VT in a patient with a structurally normal heart is
overwhelmingly associated with an excellent prognosis. Sudden
cardiac death of these patients is extremely rare, and thus the goal
of treatment is relief of symptoms. The majority of patients pre-
sent between the ages of 20 and 50 years, although children and
the elderly can present with idiopathic VT as well. At presentation,
298 III Electrophysiology

Figure 24.11. Early Afterdepolarizations. These were recorded from a canine Purkinje fiber under conditions of 10 −7 M quinidine. Such activity
on the surface electrocardiogram produces polymorphic ventricular ectopy in series.

symptoms include palpitations in 80%, dizziness in 50%, and syn-
cope in 10%. Idiopathic VT can be divided into 2 main groups: ade-
nosine-sensitive VT and verapamil-sensitive intrafascicular VT.
Adenosine-Sensitive VT
Adenosine-sensitive VT accounts for more than 80% of all VT
cases among patients with a normal heart. This type of VT gen-
erally originates in the RVOT and manifests as 2 distinct sub-
types: 1) repetitive monomorphic VT (or Parkinson-Papp VT),
which occurs most frequently at rest, and 2) paroxysmal exer-
cise-induced sustained VT. Both subtypes have LBBB morphol-
ogy with a rightward (inferior) or normal axis (Figure 24.12).
Repetitive monomorphic (RVOT VT), originally described
in 1922, is associated with isolated PVCs, ventricular pairs, and
salvos of nonsustained VT intermixed with episodes of sinus
rhythm. Some patients have predominantly the exercise-induced
variant, which generally becomes sustained with catecholamine
stimulation. Remission of symptoms over time can be expected
in 5% to 20% of cases. Other LBBB tachycardias that should be
considered include Mahaim (atriofascicular) tachycardia, anti-
dromic reciprocating tachycardia or bystander tachycardias from
right-sided accessory pathways, VT due to ARVD, VT associated
with tetralogy of Fallot repair, and bundle branch reentry VT.
On echocardiography, the right ventricle is normal in more
than 90% of patients; a minority have tachycardia-induced cardi-
omyopathy changes, which are reversible. Occasionally, cases of
ARVD can be confused with idiopathic RVOT VT on echocardi-
ography and should be evaluated further with additional imaging
studies (computed tomography or magnetic resonance imaging)
if indicated. Adenosine-sensitive RVOT tachycardias are most
likely caused by a triggered DAD mechanism. Experimentally,
DADs can be produced or enhanced under conditions that
augment intracellular calcium loading, such as rapid pacing,
increased extracellular calcium concentration, digitalis glyco-
sides, drugs that enhance the intracellular calcium concentra-
tion, low extracellular potassium concentration, or endogenous
or exogenous catecholamines. Thus, DAD-dependent arrhyth-
mias, such as RVOT VT, can be terminated at multiple levels of
the β-receptor–adenyl cyclase cascade with muscarinic (vagal)
activation, β-blockade, calcium blockade (with agents such as
verapamil), the potassium channel opener nicorandil, or adeno-
sine. Verapamil can terminate triggered ventricular arrhythmias,
but it is not specific since the drug can also inhibit VT due to
abnormal automaticity, intrafascicular reentry, and myocardial
ischemia. In contrast, adenosine is very specific in its ability to
terminate triggered ventricular arrhythmias. Adenosine binds to
the A1 adenosine receptor, which is coupled to adenyl cyclase
through the inhibitory pertussis toxin–sensitive G-protein, Gi.
Through this inhibitory effect of Gi, the concentration of cAMP
is reduced, and the overall effect becomes antiadrenergic, thus
reducing DAD activity. Adenosine does not terminate ventricular
DAD activity due to cAMP-independent mechanisms, namely,
with digitalis, dibutyryl cAMP, or α1 inosine triphosphate–
dependent pathways. Also, adenosine does not suppress VT due
to catecholamine-facilitated reentry in the presence of structural
heart disease or VT due to EAD activity induced by quinidine or
calcium agonists like Bay K8644.

I aVR

II aVL

III aVF

V1

Figure 24.12. Right Ventricular Outflow Tract Ventricular Tachycardia (Left Bundle Branch Block–Right Axis Deviation Variant). These tachy-
cardias occur in both the right ventricular outflow tract and the left ventricular outflow tract.
 24 Ventricular Tachycardia and Ectopy 299

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

Figure 24.13. Verapamil-Sensitive Ventricular Tachycardia (Right Bundle Branch Block With Left Axis Deviation). This is typical of ventricular
tachycardia originating from the left posterior fascicular area of the left ventricular septum. Note the relatively narrow QRS complex.

At exercise testing and at EP testing, only 30% to 50% of
patients have sustained VT. Signal-averaged ECG is generally
negative and unhelpful. In most patients, VT is noninducible at
EP testing in the sedated state. Triggered VT is generally induced
with ventricular burst pacing or programmed ventricular stimula-
tion and cannot be entrained (like reentry). To facilitate induction
pharmacologically, isoproterenol, atropine, or aminophylline can
be administered and used in conjunction with programmed stim-
ulation. Support for the diagnosis in a patient with LBBB VT is
suggested by termination of the VT with adenosine, verapamil,
edrophonium, vagal maneuvers, or β-blockers. Unusual origins
for adenosine-sensitive VT include the left ventricular outflow

Figure 24.14. Idiopathic Polymorphic Ventricular Tachycardia Originating From the Posterior Left Ventricular Hemifascicle. Electroanatomical
map (right anterior oblique view) of the left ventricle (LV) in the same patient as in Figure 24.3 was displayed during normal sinus rhythm. The red
area is the His bundle, and the more apical yellow areas are the anterior and posterior hemifascicles. The maroon dots are ablation sites corresponding
to the electrograms in Figure 24.15.
300 III Electrophysiology
tract (where damage to the left main coronary artery, conduction
system, or aortic valve must be monitored for), left fascicular con-
duction system, epicardial left ventricle, and RVOT diverticula.
Verapamil-Sensitive Intrafascicular VT
Verapamil-sensitive intrafascicular VT, described in 1979, origi-
nates from the posteroinferior mid left ventricle (in the region of
the left posterior hemifascicle) and produces right bundle branch
block with left axis deviation (Figure 24.13). The ECG of this
VT generally has a short RS interval (<60–80 ms) and a short
QRS duration (<140 ms). Four features of this arrhythmia are
the morphology, the induction with atrial pacing, the verapamil
sensitivity, and the absence of structural heart disease. Akin to
RVOT VT, this arrhythmia generally occurs in younger patients
who have normal ventricular function and no CAD; 80% are
men. This VT can also be associated with tachycardia-mediated
cardiomyopathy. Sudden death virtually never occurs in associa-
tion with this form of VT, although the polymorphic form can
cause syncope and VF. At presentation, symptoms include pal-
pitations, dyspnea, dizziness, and syncope. The signal-averaged
ECG is generally normal when analyzed in the time domain.
At EP testing, in the region of the left posterior hemifasci-
cle, Purkinje fiber potentials that precede the earliest ventricu-
lar endocardial activation mapping site can be identified. Unlike
RVOT VT, this rhythm is thought to be caused by reentry and
not by a triggered mechanism. The zone of slow conduction has
been shown to be between the late diastolic potential area (the
common left bundle near the base of the left ventricular septum
[the entrance]) and the more apically positioned posterior hemi-
fascicle area (the exit). This area appears to depend on the slow
Figure 24.15. Ventricular Activation in Ventricular Tachycardia. Purkinje potentials (arrows) precede the local ventricular activations on the
ABL d channel on the first 4 beats of ventricular tachycardia.
inward calcium current since verapamil causes significant delay
in this zone, whereas lidocaine has minimal effect. It is postu-
lated that Purkinje potentials in the area of slow conduction are in
a depolarized state, more dependent on the slow inward calcium
current for maintenance of conduction. Entrainment is more eas-
ily demonstrated from the RVOT than from the right ventricular
apex. As the name implies, this type of VT is sensitive to vera-
pamil but is generally unresponsive to vagal maneuvers or aden-
osine, although it can be adenosine sensitive if catecholamines
are required for its initiation.
RVOT VT, if nonsustained and asymptomatic, requires no
therapy. β-Blockers or calcium channel blockers inhibit the
arrhythmia in 25% to 50% of cases and are first-line medical
therapies. Class I and class III antiarrhythmic drugs are effective
in up to 50% of cases. Acutely, carotid sinus massage, adenosine,
verapamil, and lidocaine facilitate termination. For verapamil-
sensitive left ventricular VT, calcium channel blockers are the
medical treatment of choice. Radiofrequency ablative therapy
should be recommended for patients who experience side effects
or inefficacy from drug therapy or for individuals who have
arrhythmia-related symptoms (eg, syncope or near syncope).
In RVOT VT, QRS duration greater than 140 ms with a tri-
phasic QRS complex in the inferior leads suggests a free wall
lateral origin, whereas QRS duration less than 140 ms with a
monophasic QRS complex suggests a septal origin. During VT,
precordial transition is generally seen in leads V2 through V4; for
transition in V2, a location just adjacent to the pulmonary valve
or a left ventricular outflow origin is suggested. Successful abla-
tion sites have been associated with activation times 10 to 45 ms
ahead of the surface QRS as well as pace map matches in 11 of
12 or in 12 of 12 ECG leads. Body surface potential mapping,
 24 Ventricular Tachycardia and Ectopy 301

Table 24.2. PVCs and Cardiomyopathy

Feature
Patient characteristics
Comorbid conditions
Ejection fraction
Frequency of PVCs
Pattern of PVCs
QRS morphology
Response to radiofrequency
ablation
Abbreviations: ICD, implantable cardioverter-defibrillator; PVC, premature ventricular contraction.
PVCs Associated With Cardiomyopathy
Older patients with known cardiovascular
disease
Hypertension, ischemic heart disease,
myocarditis, family history of myocardial
disease
Depressed
<5,000 in 24 h
Multiple morphologies
Nonspecific
Required only if associated with ventricular
tachycardia that has been triggering
frequent ICD shocks
No effect on the ventricular function
PVCs Causing Cardiomyopathy
Otherwise healthy individuals
Often no prior cardiac or major medical or family
history
Depressed
>10,000 in 24 h, often >20,000 in 24 h
Monomorphic
Outflow tract morphology (right or left bundle branch
block pattern with inferior axis)
Fascicular morphology (typical right bundle branch
block pattern with superior axis)
Normalization of ventricular function frequently seen

electroanatomical 3-dimensional mapping, and noncontact map- significant reduction in ejection fraction has already occurred.
ping have all been used to facilitate ablation of VT of RVOT Recovery of ventricular function after a short period of suppres-
origin. Successful ablation of RVOT VT occurs in 90% of cases, sion strongly suggests TICM and supports more aggressive inter-
with 10% recurrence. ventions such as radiofrequency ablation.
For verapamil-sensitive left ventricular VT, the usual suc- Ablative therapies are most successful in patients who have a
cessful sites of ablation are sites of earliest ventricular activation. single predominat PVC morphology arising from an endocardial
Success rates are approximately 85% to 90%. Pace mapping site in 1 of the outflow tracts. Ablative approaches and risks are
is not as useful as in RVOT VT, and the presence of Purkinje the same as those described earlier in this chapter.
potentials is not a necessary requirement for successful ablation
(the potentials can sometimes be recorded over a 2- to 3-cm2 Abbreviations
area of tissue). A similar region of targeting for ablation has
ARVD arrhythmogenic right ventricular dysplasia
recently been identified for “curing” patients with idiopathic VF
AV atrioventricular
(Figures 24.14 and 24.15). CAD coronary artery disease
Isolated PVCs and short runs of ventricular tachycardia in a cAMP cyclic adenosine monophosphate
normal heart are considered benign, but there is growing rec- DAD delayed afterdepolarization
ognition that high numbers of PVCs can lead to TICM. There EAD early afterdepolarization
are no established criteria to determine who is at risk of TICM, ECG electrocardiogram
but current experience suggests that 10,000 to 15,000 PVCs EP electrophysiologic
per 24 hours (documented with Holter monitor) over a period HCM hypertrophic cardiomyopathy
of weeks or months may confer risk. In the patient presenting ICD implantable cardioverter-defibrillator
with decreased left ventricular ejection fraction and worrisome IK inward potassium current
IK ATP adenosine triphosphate–sensitive potassium channel
ventricular ectopy, it is difficult to determine cause and effect,
LBBB left bundle branch block
although clinical characteristics can be helpful (Table 24.2). LQT1 long QT syndrome type 1
Additionally, some patients are extremely uncomfortable with LQTS long QT syndrome
relatively few PVCs. For those patients, treatment is warranted MI myocardial infarction
for symptomatic relief. PVC premature ventricular contraction
Initial evaluation of the patient who is symptomatic with RVOT right ventricular outflow tract
PVCs or who has high numbers of PVCs should include a search TICM tachycardia-induced cardiomyopathy
for underlying cardiac causes (such as CAD, definable cardio- VF ventricular fibrillation
myopathies, or myocarditis) and noncardiac causes (such as elec- VT ventricular tachycardia
trolyte and endocrine abnormalities).
β-Blockers or calcium channel blockers (or both) may Names of Clinical Trials
decrease the number of PVCs, but those agents rarely completely MADIT Multicenter Automatic Defibrillator Implantation Trial
suppress PVCs. With the exception of amiodarone, antiarrhyth- MADIT-II Multicenter Automatic Defibrillator Implantation Trial
mic drugs are usually not particularly effective in suppressing II
PVCs. Because of the potential side effects of amiodarone, a MUSTT Multicenter Unsustained Tachycardia Trial
3- to 6-month trial may be considered, especially if a clinically SCD-HeFT Sudden Cardiac Death in Heart Failure Trial

Arrhythmias in Congenital Heart Disease
CHRISTOPHER J. MCLEOD, MB, CHB, PHD,
and PETER A. BRADY, MB, CHB, MD
The number of persons with CHD who survive into adulthood has
increased dramatically in the past few decades. In this population,
arrhythmias are a major cause of late morbidity and mortality
and may herald worsening underlying hemodynamic changes and
associated poorer outcomes. In addition, the risk of sudden car-
diac death is high, especially in certain repaired malformations.
Device (pacing and defibrillator) therapy is often complicated and
may require thoracotomy for epicardial lead placement.
• Risk markers of arrhythmias in patients with CHD include the
following:
1. Presence of residual hemodynamic defects
2. Surgical repair later in life
3. Older age
Pathophysiology
Bradycardia or tachycardia (atrial or ventricular) may develop
inherently (congenital) or may be acquired (especially after sur-
gical intervention). Common associations are summarized in
Table 25.1.
Bradycardia
Congenital Bradycardia
Similar to patients without CHD, bradycardia may be due to
1) sinus node dysfunction or 2) heart block (AV node or His-
Purkinje level).
Congenital sinus node dysfunction is common and may
present as impaired heart rate response to exercise (chronotro-
pic incompetence), symptomatic sinus bradycardia, or sinus
pauses leading to presyncope or syncope. The risk of sinus node
Abbreviations and acronyms are expanded at the end of this chapter.
302
25
dysfunction is higher when the sinus node is anatomically dis-
placed (situs inversus or heterotaxy syndromes).
Congenital AV block occurs most commonly in the absence
of other cardiac disease. In most cases, block is proximal to the
bifurcation of the His bundle or within the AV node. Therefore,
the QRS is narrow (<100 milliseconds) and the ventricular rate
is usually more than 40 beats per minute. Although some contro-
versy exists, most agree that a permanent pacemaker is indicated
regardless of symptoms. Heart block necessitating a permanent
pacemaker will develop in up to a quarter of patients with con-
genitally corrected transposition of the great arteries at some
stage in their life, presumably through embryologic displacement
of this region of the heart.
Acquired Bradycardia
Acquired sinus node dysfunction is most common after atrial
surgery that may injure the sinus node, most commonly Senning,
Mustard, Fontan, Glenn, or atrial reduction procedure for Ebstein
anomaly. “Late” sinus node dysfunction is also common, a sug-
gestion that other mechanisms, such as sinoatrial ischemia or
fibrosis or the effects of drug therapy, play a role.
Acquired AV node dysfunction is also common in the early
postoperative period, especially after surgery close to the con-
duction tissue in the periaortic and interventricular septum
region. In many cases, AV block is transient; however, if it is
prolonged beyond 7 days, recovery is unlikely and a permanent
pacemaker is indicated.
• Acquired AV block is most common after:
1. Repair of AV septal defect (canal or endocardial cushion)
2. Closure of ventricular septal defect involving the perimembra-
nous region
3. Repair of subaortic stenosis
 25 Arrhythmias in Congenital Heart Disease 303

Table 25.1. Common Associations Between Arrhythmia and also have an accessory AV or atriofascicular pathway, which is
Congenital Heart Disease almost exclusively right-sided and often multiple. These pathways
can provide one “limb” of the reentry circuit for AV reentry tachy-
Arrhythmia Associated Defect cardia. In most cases, presentation is before the age of 35 years.
Bradycardia Because of concomitant hemodynamic and anatomic abnor-
Congenital malities, supraventricular arrhythmias are usually not well toler-
Congenital sinus node Situs inversus ated in patients with Ebstein anomaly, and patients frequently
dysfunction present with symptomatic deterioration or even presyncope,
Congenital AV block AV canal defects; CCTGA syncope, or sudden cardiac death.
Acquired In most patients undergoing surgical repair for manage-
Acquired sinus node Occurs after Mustard, Senning, ment of accessory pathways, even in the absence of prior symp-
dysfunction Fontan, Glenn operation
toms of palpitations, preoperative electrophysiologic testing is
Acquired AV node dysfunction Occurs after VSD, LVOT, aortic
appropriate. If an accessory pathway is found that can support
operation
Tachycardia
reentrant tachycardia, then either catheter ablation or surgical
Congenital cryoablation can be performed.
Accessory pathways Ebstein anomaly, CCTGA A similar approach is reasonable before creation of barriers to
Acquired future catheter access (ie, extracardiac Fontan tunnels).
Atrial tachycardia or flutter Atrial baffles; after maze procedure
Atrial fibrillation ASDs, mitral valve disease
Acquired Tachycardia
Ventricular tachycardia Tetralogy of Fallot, d-transposition
IART is the most common symptomatic rhythm disturbance in
Abbreviations: ASD, atrial septal defect; AV, atrioventricular; CCTGA,
congenitally corrected transposition of the great arteries; LVOT, left ventricular
adults with CHD. It usually develops many years after surgical
outflow tract; VSD, ventricular septal defect. repair and most often is localized within the right atrium, being
confined by anatomic and surgical boundaries and maintained
by a reentry circuit or multiple circuits (Figure 25.1).
Management of Bradycardia IART is most common late after extensive atrial surgical
Management of bradycardia in patients with CHD is similar to reconstructive surgery such as the Mustard, Senning, or original
that in patients without CHD. Unique to patients with CHD is Fontan operations. Other factors such as chronic hemodynamic
careful consideration (class IIb indication) when the resting heart stress also act on the atrial substrate, promoting IART.
rate is less than 40 beats per minute or pauses are longer than 3 On the surface 12-lead electrocardiogram, IART may be diffi-
seconds in the absence of symptoms. cult to distinguish from atrial flutter, and in some cases it may be
thought to be sinus tachycardia if the P waves are buried within
the QRS with 2:1 AV block. Like atrial flutter, IART is character-
Tachycardia ized by uniform flutter wave morphologic features, constant cycle
Congenital Tachycardia length, and sudden onset and offset (due to the reentrant mechan-
ism). However, IART is typically somewhat slower than isthmus-
The mechanisms of most “supraventricular” tachycardias in
dependent (classic) atrial flutter (with atrial rates in the range of
patients with CHD are similar to those in patients without CHD
150–250 per minute); for this reason, it can often conduct more
(paroxysmal supraventricular tachycardia, atrial tachycardia, and
rapidly through the AV node and may precipitate (often pro-
flutter). However, because of anatomic, hemodynamic, and cel-
found) hemodynamic decompensation. In addition, IART may
lular changes (eg, fibrosis), the likelihood of atrial arrhythmias,
be difficult to terminate with antiarrhythmic drug therapy and
especially atrial flutter, is higher.
to slow with AV nodal blocking agents. Because of this presen-
Because CHD and accessory pathways (Wolff-Parkinson-
tation, IART should be suspected in a previously stable patient
White syndrome) are relatively common (CHD, 1% or 2% of live
who experiences sudden deterioration in clinical status even if
births; accessory pathways, 1% of live births), the 2 may coexist
the heart rate is not substantially increased from baseline.
by chance. However, certain types of CHD are more frequently
associated with accessory pathways. This association is believed • IART is most common with the following:
to develop because embryologic malformation itself leads to dis-
1. Repair of atrial septal defect
ruption of the central fibrous body and interruption of the AV
rings, mechanisms that help to create muscular bands that cross 2. Mustard or Senning procedure
the AV groove, bypassing the normal conduction system and 3. Fontan procedure
allowing for ventricular “preexcitation.” 4. Tetralogy of Fallot
5. Repair of anomalous pulmonary venous drainage
• Common heart defects associated with the highest likelihood of
accessory pathways are Typical, or isthmus-dependent or common, atrial flutter is
1. Ebstein anomaly of the tricuspid valve so called because the critical zone of the circuit is within the
2. Congenitally corrected transposition of the great vessels cavotricuspid region. The wavefront travels around the tricus-
3. AV septal defect pid valve annulus and is characterized by a sawtooth pattern in
leads II, III, and aVF without a defined isoelectric line between
Ebstein anomaly of the tricuspid valve is characterized by the flutter waves. In counterclockwise atrial flutter, the P-wave
downward displacement of the functional annulus, malforma- axis is negative in these inferior leads and in lead V1, where an
tion of the tricuspid valve and right-sided chambers, atrial septal isoelectric baseline can be appreciated. Recognition of typical
defect, and enlargement of the right atrium and ventricle in most isthmus-dependent atrial flutter is important because it is ame-
patients. Approximately 25% of patients with Ebstein anomaly nable to catheter ablation, which has a high likelihood of success
304 III Electrophysiology
Figure 25.1. Septal Patch and Atriotomy Scar. The presence of
these structures, which form a nonconductive boundary, allows reentry
circuits to develop within the atrium. These may lead to sustained atrial
flutters that are non–isthmus-dependent and, therefore, in most cases
have an atypical appearance. Ablation of these atrial flutters requires
identification of these circuits and the use of linear ablation between scar
and other nonconductive boundaries, such as the tricuspid valve annulus
or the inferior vena cava. CS indicates carotid sinus; IVC, inferior vena
cava; SVC, superior vena cava; TV, tricuspid valve.
(>95%) and a low likelihood of recurrence (<2%). Because of
enlargement of either the right or the left atrial chamber (or both)
or surgical scar (atriotomy), the typical appearance of typical
atrial flutter may be distorted on the surface electrocardiogram.
Any other sustained narrow complex tachycardia with a stable
rate that is confined to the atria is termed atypical atrial flutter.
In patients with CHD, this rhythm is most likely IART.
Atrial fibrillation is characterized by disorganized, rapid, and
irregular electric activity and may occur in up to 30% of patients
with CHD. Unlike IART, atrial fibrillation is more common in
patients with residual left-sided obstructive lesions or ventricu-
lar dysfunction causing left atrial pressure or volume overload.
Similar to other atrial arrhythmias, loss of AV synchrony is par-
ticularly problematic, and efforts to restore and maintain sinus
rhythm are usually necessary, but difficult. Patients are often
committed to protracted courses of antiarrhythmic therapy, but
overall success rates are disappointing.
• Atrial fibrillation is more common with residual left-sided
disease:
1. Aortic stenosis.
2. Mitral valve deformities.
3. Unrepaired single ventricle of left ventricular morphology.
Therapeutic strategies for management of atrial arrhyth-
mias in patients with CHD are not dissimilar from those in the
non-CHD population: antiarrhythmic drugs, catheter ablation,
and surgical approaches (maze procedure), especially if sur-
gery is indicated for other reasons. Of primary importance in
patients with atrial arrhythmias and CHD is the issue of systemic
long-term anticoagulation, especially for slow flow circulations
such as the Fontan circulation.
For acute management, initiation of rate control (AV nodal
blocking agent) is the first-line therapy in an otherwise stable
patient. However, drugs that block the AV node may not be well
tolerated, especially in patients with IART, because of the rela-
tively slow atrial rates and concomitant sinus node dysfunction.
Direct-current cardioversion should be considered early if hemo-
dynamic compromise is present.
Chronic management includes antiarrhythmic drug therapy,
catheter ablation, and surgical therapy.
In highly symptomatic patients in whom catheter ablation is
neither feasible nor desired, antiarrhythmic drug therapy with
sotalol or amiodarone may be used after consideration of risks
and benefits and should be initiated in the hospital. Although
both class I and III agents have been used extensively in the
management of atrial arrhythmias in patients with CHD, overall
efficacy is poor. Moreover, the safety of these agents in this pop-
ulation (especially with concomitant sinus node or conduction
system disease) has not been proved. Thus, these agents, if used
at all, should be used with caution (especially in cardiovascular
board scenarios).
With the advent of more sophisticated intracardiac map-
ping, the efficacy and safety of catheter ablation have increased.
Catheter ablation of typical (isthmus-dependent) atrial flutter is
associated with a long-term success (defined as no recurrence of
the atrial flutter) rate of more than 95% and should be considered
early. One drawback of this approach is that even after success-
ful ablation of this atrial flutter, other atrial arrhythmias (such
as IART) may supervene. However, these can now be mapped
and ablated with high success in this population and should be
strongly considered as a therapeutic option.
Prevention of established atrial arrhythmias by surgical revi-
sion of a conduit or valve is often of limited efficacy. In most
cases, adjunctive cryoablative therapy at the time of surgical
repair is performed between anatomic and surgical barriers to
conduction. Modifications of surgical techniques may help to
prevent or reduce the likelihood of future atrial arrhythmias, but
few data currently pertain to this approach.
In patients with CHD, ventricular tachycardia may be either
scar-mediated (as in IART) or develop in a myopathic ventri-
cle. Scar-mediated ventricular tachycardia is more common and
occurs most frequently around a ventriculotomy or ventricular
patch site, most commonly in patients with repaired tetralogy of
Fallot.
Myopathic ventricular tachycardia occurs in patients with
impaired ventricular function (systemic or pulmonic). Most
common is ventricular tachycardia arising from a systemic right
ventricle (not teleologically designed for the higher pressures)
and in patients with substantial aortic valve disease and single
ventricle physiology. Interestingly, repair of the pulmonary valve
in repaired tetralogy of Fallot with subsequent amelioration of
the hemodynamic strain does not seem to reduce the risk of
malignant ventricular arrhythmias.
Also in the differential diagnosis of a wide complex tachycar-
dia is rapid conduction of an atrial arrhythmia over an accessory
pathway. In such cases, Ebstein anomaly or IART with a rapid
ventricular response in a patient who has had a Fontan, Mustard,
or Senning procedure should be excluded.
 25 Arrhythmias in Congenital Heart Disease
Sudden Cardiac Death and Risk Assessment
Although annual mortality in CHD, even in high-risk patients,
is low (<2%), the duration of risk exposure is extremely long
given the younger age of this patient population, and thus the
potential for years of life saved is high. The risk of sudden car-
diac death increases with the duration of follow-up; the great-
est risk is in patients more than 25 years from surgical repair.
Unfortunately, risk stratification of patients with CHD is chal-
lenging with no large prospective trials. Often, clinical decision
making is extrapolated from adults with ischemic heart disease
and ischemic cardiomyopathy, and our current understanding of
right ventricular arrhythmogenesis is rudimentary.
Secondary Prevention
Patients who present with hemodynamically unstable ventricu-
lar tachycardia or out-of-hospital cardiac arrest should undergo
implantation of a cardioverter-defibrillator. However, such a
presentation is uncommon, and most sudden cardiac death events
occur in the absence of prior symptoms.
Asystole and bradycardia-associated risk should initially be
excluded. Unlike acquired heart disease, there is no single risk
factor that dictates implantation of a cardioverter-defibrillator.
An individualized and careful assessment is necessary to iden-
tify potential risk factors, and this should be based on a detailed
enquiry to identify unrecognized presyncopal symptoms and
palpitations—in addition to frequent Holter and electrocardio-
graphic recordings.
Primary Prevention
Primary prevention of sudden cardiac arrest is more problematic
and relies on recognition of risk markers.
Risk Markers for Sudden Cardiac Death
Most data regarding risk for sudden cardiac death pertain to
patients with repaired tetralogy of Fallot. Although prognosis
in this population is generally excellent (about 90% survival at
30 years), the prevalence of ventricular tachycardia is between
3% and 14%, and it is the single most common condition associ-
ated with premature sudden cardiac death in patients with CHD
(Box 25.1).
Synthesis of these factors to determine a patient’s risk may
still leave the clinician in a predicament with regard to prophy-
lactic implantation of a cardioverter-defibrillator, and electro-
physiologic testing with programmed ventricular stimulation is
frequently used as an adjunct in this setting. It has reasonable
predictive power if the study identifies symptomatic sustained
monomorphic ventricular tachycardia—and ablation of the focus
at the time is then also possible. The decision to implant a device
can then potentially be deferred if the arrhythmia is rendered
noninducible. Although induction of sustained ventricular tachy-
cardia at electrophysiologic study is considered an independent
risk factor for subsequent clinical ventricular tachycardia and
sudden cardiac death, it has a false-positive rate of up to 30%.
Common Anomalies and Complexes
Atrial Septal Defects
Atrial septal defects lead to dilatation of the right-sided cham-
bers if the left-to-right shunting is large enough, the result of
which is atrial myocardial fiber injury and delayed intra-atrial
305
Box 25.1. Risk Factors for Sudden Cardiac Death
Clinical
Residual hemodynamic defect
Late surgical repair
Older age
Electrocardiographic
QRS prolongation >180 milliseconds (suggesting
slowed ventricular conduction)
Nonsustained ventricular tachycardia
QT duration
Hemodynamic
Systemic ventricular dysfunction or dilatation
Presence of a significant shunt
Pulmonary hypertension
Valve or conduit regurgitation
conduction that predisposes to reentrant arrhythmias and sino-
atrial node dysfunction. This condition may present with palpita-
tions, dyspnea, presyncope, and syncope (due to bradycardia or
sinus pauses).
• In the absence of surgical repair, the most common presenting
arrhythmias are
1. Typical isthmus-dependent atrial flutter
2. Late sinus node dysfunction
After repair, isthmus-dependent flutter remains common, yet
the surgical incisions (atriotomy scar) increase the likelihood
of a non–isthmus-dependent IART; frequently, these 2 rhythms
coexist in the same patient. Complex double-loop circuits are
also not uncommon, increasing the complexity of the rhythm and
attempted catheter ablation.
• Risk factors for atrial arrhythmias in surgically repaired atrial
septal defect include older age at repair and long duration of
follow-up.
AV Septal Defects
AV septal defects are associated with displacement of the AV
node and with varying degrees of malformation of the left
bundle branch, usually with deviation of the axis in a left supe-
rior direction. In most cases of complete AV septal defect, the
electrocardiogram shows biventricular hypertrophy and biatrial
enlargement. Without surgical repair, AV septal defects may
lead to development of Eisenmenger physiology, right ventricu-
lar hypertrophy, and risk of ventricular arrhythmias. AV septal
defects are also associated with a higher likelihood of accessory
AV connections and paroxysmal supraventricular tachycardias.
Ventricular Septal Defects
Although defects occur in all portions of the interventricular
septum, the effect on the cardiac conduction system and risk
of arrhythmias in unoperated patients probably result mostly
from hemodynamic effect rather than the precise location of the
defect. Even as an isolated anomaly, associated arrhythmias are
common in children and adults.
306 III Electrophysiology
• Independent predictors of arrhythmias in patients with ventricular
septal defect include
1. Increased pulmonary artery pressure at the time of operation
2. Older age
Although the risk of ventricular arrhythmias and sudden car-
diac death after repair of ventricular septal defect is low, sudden
cardiac death may occur when ventricular hypertrophy and con-
duction system fibrosis are present. Late sudden death occurs in
less than 5% of patients with surgically repaired ventricular sep-
tal defect. Risk factors for sudden cardiac death include opera-
tion after 5 years of age, increased pulmonary vascular resistance
(especially Eisenmenger physiology), and untreated heart block.
Tetralogy of Fallot
Complete repair of tetralogy of Fallot may involve patch closure
of the ventricular septal defect and resection of a considerable
amount of right ventricular myocardium. In early surgical prac-
tice, this procedure was performed via a ventriculotomy rather
than atriotomy, an approach that thereby provided a potential
circuit for ventricular tachycardia. The most common hemody-
namic sequela occurring late after right ventricular outflow tract
reconstruction and repair is pulmonary regurgitation, which pre-
disposes to volume overload of the right ventricle that may, in
turn, predispose to ventricular arrhythmias.
Mustard, Senning, and Fontan Procedures
Sinus node dysfunction and atrial arrhythmias are com-
mon after Mustard and Senning procedures done for repair of
d-transposition of the great arteries. During operation, creation of
the intra-atrial baffle may damage the sinus and perinodal struc-
tures within the atrium, a result predisposing to IART and atrial
flutter. Risk factors for development of atrial arrhythmias after
Mustard and Senning procedures include pulmonary hypertension,
right ventricular dysfunction, and persistent junctional rhythm.
Although the Fontan operation has multiple variations, all are
functionally similar in their intention to direct systemic venous
return to the pulmonary arteries and bypass the right ventricle.
Older Fontan procedures incorporated the right atrium as part
of the connection; in patients who had these procedures, atrial
arrhythmias are common because of myocardial scarring and
chronic increases in right atrial pressure elevations. Newer
Fontan circulations aim to avoid the right atrium and instead
utilize “extracardiac” synthetic conduits that direct inferior
vena cava blood to the pulmonary system. This approach thus
avoids the potential for right atrial dilatation and its arrhythmic
consequences.
Ebstein Anomaly
After surgical repair of Ebstein anomaly, enlargement of the right
atrium and the presence of an atriotomy provide the substrate
for IART or atrial flutter. The likelihood of postoperative IART
is decreased with the addition of right-sided maze or cryoabla-
tion procedure independent of corrective surgery and does not
increase operative mortality.
Arrhythmias Occurring After
Congenital Heart Surgery
Junctional tachycardia (junctional ectopic or His-bundle tachy-
cardia) (JT) is most common in the first 24 hours after repair of
a congenital heart defect. Procedures associated with a higher
likelihood of JT include repair of tetralogy of Fallot, Mustard
and Senning procedures for d-transposition of the great arter-
ies, closure of ventricular septal defect, repair of total anomalous
venous return, and Fontan operation.
On electrocardiography, JT is typically regular with rates
more than 200 beats per minute and a QRS morphologic pat-
tern identical to that during sinus rhythm. In most cases, ven-
triculoatrial dissociation is present (ventricular rate faster than
atrial rate) with occasional sinus capture. In some cases, it may be
confused with atrial tachycardia with first-degree AV block. The
precise mechanism of JT is not well understood, but it may be due
to surgical trauma to the conduction system, coupled with pulmo-
nary arterial hypertension, inflammatory changes, and hypoxia.
Overdrive suppression is common, and typically the rhythm
shows a warm-up phenomenon after suppression, increasing
gradually up to the previous rate. Although uncommon in the
adult population, recognition of JT is important because treat-
ment is directed at the underlying medical condition (usually a
complicated postoperative course) rather than the rhythm itself.
Because the mechanism of JT is a triggered activity and not reen-
try, direct-current cardioversion is rarely useful because recur-
rence rates are very high.
Instead, management of this arrhythmia involves attention to
ameliorating the hemodynamic milieu and allowing for sponta-
neous resolution of the rhythm over time. The goal of therapy is
to decrease the ventricular rate to less than 150 beats per minute.
Some success has been reported with the use of both class I
(flecainide and propafenone) and class III agents (mostly amio-
darone). Restoration of AV synchrony with atrial pacing may be
beneficial in some patients.
• Occurrence of arrhythmias (mostly atrial) late after surgical repair
is common and associated with considerable morbidity.
• Patients who have stable systemic ventricular function and few
symptoms and who are within 25 years of surgical repair have
excellent survival and can be managed conservatively.
• Patients with high-risk clinical characteristics, especially those
with symptoms such as presyncope or syncope, should undergo
careful evaluation and risk stratification (including hemody-
namic and electrophysiologic assessment) to determine the precise
arrhythmia and risk of sudden cardiac death.
• Patients with spontaneous sustained ventricular arrhythmias or
a high-risk clinical profile and inducible ventricular tachycardia
should undergo implantation of a cardioverter-defibrillator.
• Cardiovascular board questions typically focus on common clini-
cal scenarios:
1. In a previously stable patient with CHD who presents with clini-
cal deterioration, suspect onset of an atrial arrhythmia (IART,
atrial flutter, or atrial fibrillation).
2. Know broad treatment options and pacing indications.
3. Avoid use of class I (and class III) antiarrhythmic drugs because
of the risk of proarrhythmia.
4. Because of limited data regarding primary prevention of sud-
den cardiac death in this population, questions on this topic are
unlikely.
Abbreviations
AV atrioventricular
CHD congenital heart disease
IART intra-atrial reentrant tachycardia
JT junctional tachycardia
 26
Evaluation and Management of
Pregnancy-Associated Arrhythmias
PETER A. BRADY, MB, ChB, MD
Arrhythmias that occur for the first time during pregnancy are
uncommon and are usually due to exacerbation of a known
rhythm disorder or an arrhythmic substrate (eg, congenital heart
disease).
Basic Principles of Rhythm Management
During Pregnancy and Lactation
The presentation and management of heart rhythm disorders
arising during pregnancy and lactation are similar to those in
nonpregnant patients. However, patients who present with a heart
rhythm disorder during pregnancy deserve special consideration
for the following reasons:
1. Any heart rhythm affects both the mother and the fetus and may
continue throughout pregnancy, labor, and breast feeding. Thus, the
potential effects of the arrhythmia and the therapy on both maternal
and fetal well-being need to be considered.
2. Pregnancy is accompanied by complex physiologic changes that
alter drug efficacy and the risk of toxicity.
3. Almost all pharmacologic agents cross the placenta and appear in
breast milk, affecting the fetus and the newborn.
4. Patients with structural heart disease (most commonly congenital
heart disease) represent a special population that requires coordina-
tion of care with a specialist in adult congenital heart disease.
5. In most instances, patients can be effectively evaluated noninva-
sively and treated conservatively.
6. Decisions regarding therapeutic approaches to the pregnant patient
presenting with a heart rhythm disorder should be based on care-
ful consideration of 1) severity of symptoms and 2) risks versus
benefits.
These considerations are important in any patient, but they are
especially important in the pregnant patient.
Physiologic Changes During Pregnancy
Physiologic changes associated with normal pregnancy include
the following:
1. An increase in total body water (approximately 5–8 L), leading to a
40% increase in cardiac output beginning during the first trimester
and continuing until mid pregnancy, after which the amount of body
water stabilizes and then begins to decline in the last week.
2. Decreased peripheral vascular resistance, which occurs throughout
pregnancy. (Mean blood pressure begins to decrease early in preg-
nancy, reaches a minimum in mid pregnancy, and returns to baseline
levels at term.)
For a comprehensive review of physiologic changes during
pregnancy, see the chapter “Pregnancy and the Heart” elsewhere
in this book.
Evaluation of Heart Rhythm Symptoms
Accurate rhythm diagnosis and assessment of symptom severity
(especially hemodynamic symptoms) are mandatory. Arrhythmias
that result in hemodynamic compromise (usually in patients with
structural heart disease) are of major concern—not only because
of potential harm to the mother but also because of potential harm
to the fetus through reduced placental blood flow. In general,
short-lived arrhythmias causing minor or minimal symptoms in
a patient with a normal heart can be treated conservatively. In all
cases, evaluation should aim to 1) correlate symptoms with the
arrhythmia and 2) exclude preexisting structural cardiac disease.
Electrocardiography
Electrocardiographic changes that occur during normal preg-
nancy include a leftward shift in the frontal QRS axis, small
307
308 III Electrophysiology
Q wave, and an inverted T wave in lead III due to a gradual shift
in the position of the heart within the thorax.
A baseline 12-lead electrocardiogram may be helpful to
exclude the following:
1. Abnormal ventricular preexcitation (short PR interval and delta wave)
2. Presence of conduction system disease (most common in patients
with congenital heart disease)
3. Evidence for arrhythmogenic right ventricular dysplasia or cardio-
myopathy (inverted T waves beyond V1 or an epsilon wave)
4. QT interval prolongation (congenital or acquired long QT syndrome)
Echocardiography
Transthoracic echocardiography is useful to exclude structural
heart disease such as undiagnosed congenital or acquired cardiac
defects, nonischemic (peripartum) cardiomyopathy, and arrhyth-
mogenic right ventricular dysplasia.
Tilt Table Testing
Tilt table testing is generally safe during pregnancy. However,
it is not frequently indicated in pregnant patients because of the
apparent decrease in the frequency of neurocardiogenic syncope
during pregnancy (perhaps due to increased volume load).
Exercise Treadmill Testing
Exercise treadmill testing is generally safe to perform during
early pregnancy and is most useful in patients with exertional
symptoms, such as a right ventricular outflow tract tachycardia.
Ambulatory Monitoring (Holter
and Event Recorder)
Ambulatory monitoring with either 24-hour or 48-hour Holter
monitoring is helpful in patients who have frequent symptoms.
If symptoms are less frequent, an event recorder may be more
appropriate.
Electrophysiologic Studies
Invasive electrophysiologic studies are rarely needed in the preg-
nant patient and, in most cases, can be avoided because of the risks
of radiation exposure to the fetus during the positioning of cathe-
ters and the use of contrast agents and heparin. If invasive electro-
physiologic testing is deemed necessary, appropriate precautions
include placing a lead apron under the abdominal area to reduce
radiation exposure to the fetus. Nonfluoroscopic imaging such as
intra-cardiac echocardiography or electroanatomic mapping sys-
tems should be considered to minimize radiation exposure.
• The risk of radiation exposure to the fetus is greatest during the
first and second trimesters of pregnancy.
• Increased exposure has been linked to congenital malformations,
mental retardation, and increased risk of childhood malignancies,
particularly leukemia.
Drug Therapy in Pregnancy
Risks of antiarrhythmic drug therapy during pregnancy include
risk to the mother and risk to the fetus.
Maternal factors include 1) changes in absorption, distribu-
tion, and excretion of drugs (varies according to the stage of
pregnancy) and 2) drug effects at the time of labor and delivery.
Fetal factors include placental transfer of drugs during preg-
nancy and breast feeding.
Pharmacologic Therapy of Arrhythmias During
Pregnancy
Pharmacokinetic Changes
Pharmacokinetic changes during pregnancy and the peripartum
period are complex, leading to variable and unpredictable effects
of drugs. Therefore, careful monitoring of drug dosage and effect
is mandatory.
Absorption. Oral absorption of drugs, and thus bioavailabil-
ity, is altered in unpredictable ways, decreasing or increasing,
because of changes in gastric motility and secretion. In addition,
unpredictable changes in gastric pH alter the rate and degree of
absorption of drugs.
Distribution. Blood volume increases and plasma protein
concentration decreases. Increased blood volume (ie, increased
volume of distribution) decreases the drug concentration in the
central compartment and increases the elimination half-life,
whereas decreased plasma protein concentration decreases the
protein binding of medications and increases the drug effect.
Excretion. Renal blood flow is increased by 60% to 80%,
which increases elimination of renally excreted drugs. Increased
progesterone levels increase the hepatic clearance of hepatically
metabolized drugs, decreasing the drug levels.
Effects on the Fetus. During the first 8 weeks after fertili-
zation, the teratogenic risk is greatest. After that time, the risk
decreases considerably.
General Principles of Drug Therapy
for Arrhythmias During Pregnancy
Most drugs used in the treatment of heart rhythm disorders are
classified as category C by the US Food and Drug Administration
(Tables 26.1 and 26.2), and thus risk cannot be ruled out.
• No drug is completely safe during pregnancy, although many are
well tolerated and associated with low risk.
• Drug therapy should be avoided (unless absolutely necessary) in
the first trimester.
• Patients should be managed conservatively if possible.
Specific Drugs
Atrioventricular Nodal Blocking Agents
Adenosine
Although classified as a category C drug, adenosine, because of
its rapid onset and short duration of action, appears to be safe for
use during pregnancy and, on the basis of limited data, appears
to have no direct effect on the fetus after bolus intravenous
administration.
Verapamil
Verapamil is rapidly absorbed, but first-pass metabolism is high,
with only a small proportion of the drug excreted unchanged in
the urine; 90% of the drug is bound to plasma proteins. Verapamil
crosses the placenta and affects the fetal cardiovascular system,
but there are no reports of congenital defects associated with
its use.
 26 Evaluation and Management of Pregnancy-Associated Arrhythmias 309

Table 26.1. US Food and Drug Administration Classification
of Drugs for Use During Pregnancy
Category Interpretation
A Adequate, well-controlled studies in pregnant women
have failed to demonstrate a risk to the mother or
fetus in any trimester
B Adequate, well-controlled studies in pregnant women
have not shown increased risk of fetal abnormalities
despite adverse findings in animals or, in the
absence of adequate human studies, animal studies
show no fetal risk; the chance of fetal harm is
remote but remains a possibility
C Risk cannot be ruled out: adequate, well-controlled
human studies are lacking, and animal studies
have shown a risk to the fetus or are lacking; fetal
harm is possible if the drug is administered during
pregnancy, but the potential benefits may outweigh
the risk
D Positive evidence of risk: studies in humans or
investigational or postmarketing data have shown
a fetal risk; however, the benefit may outweigh the
potential risk (eg, use of the drug may be acceptable
in a life-threatening situation or serious disease for
which safer drugs cannot be used or are ineffective)
X Contraindicated in pregnancy: studies in animals or
humans or investigational or postmarketing reports
have shown clear evidence of fetal abnormalities or
risk that outweighs possible benefit
Diltiazem
Although newer than verapamil, available data suggest that dilti-
azem is probably safe to use for rate control because it has been
used in the treatment of premature labor without a report of con-
genital anomalies.
Digoxin
Digoxin is classified as a category C drug, but it is probably safe
to use during pregnancy. It crosses the placenta readily; within
30 minutes, fetal plasma concentrations are similar to maternal
values. Elimination is predominantly renal. Cardiac glycosides
have a long history of use in pregnant patients and are frequently
used in the management of supraventricular arrhythmias.
• Adenosine, verapamil, diltiazem, and digoxin are probably safe to
use during pregnancy.
β-Blockers
Like digoxin, β-blockers (in particular, propranolol) have been
used extensively in pregnancy (Table 26.3). Adverse outcomes
with β-blockers include fetal bradycardia, hypotonia, apnea,
and hypoglycemia. No studies implicate β-blockers in fetal
malformation.
Acebutolol and pindolol were recently reclassified as category
B drugs and are therefore preferred as first-line agents. However,
atenolol was recently reclassified into category D, meaning that
there is positive evidence of risk.
• Avoid use of atenolol.
• The best choices for β-blocker therapy are acebutolol (which is in
category B and is cardioselective) and propranolol (which has a
long history of safe use in pregnancy).
Vaughan Williams Classification of Drugs
Class IA
Quinidine. Quinidine readily crosses the placenta and has
been used to terminate fetal arrhythmias. It is generally con-
sidered safe, but fetal thrombocytopenia and cranial nerve VIII
palsy have been reported.
• Given the long history and safe use of quinidine in pregnancy, it is
the drug of choice among class IA drugs.
Procainamide. Procainamide readily crosses the placenta
and is also used to treat fetal arrhythmias. No adverse fetal out-
comes have been reported.
Disopyramide. Disopyramide may cause uterine contrac-
tions; therefore, it is not desirable for use in the later stages of
pregnancy.

Table 26.2. Antiarrhythmic Drugs During Pregnancy Class IC

Both flecainide and propafenone cross the placenta, yielding
Antiarrhythmic Vaughan Williams Safety During high drug levels in the fetal circulation. Although classified in
Agent Classification FDA Category Lactation
Disopyramide IA C S
Table 26.3. β-Blocker Therapy During Pregnancy
Procainamide IA C S
Quinidine IA C S Safety During
Lidocaine IB B S β-Blocker FDA Category Lactation Cardioselectivity
Mexiletine IB C S
Flecainide IC C S Acebutolol B S +
Moricizine IC B ? Atenolol D S +
Propafenone IC C ? Bisoprolol C ? +
Amiodarone III D NS Esmolol C ? +
Azimilide III None ? Inderal C S −
Dofetilide III C ? Labetalol C S −
Ibutilide III C ? Lopressor C S +
Sotalol III B S Metoprolol C S +
Adenosine V C ? Nadolol C S −
Verapamil IV C S Pindolol B ? −
Diltiazem IV C S Propranolol C S −
Timolol C S −
Abbreviations: FDA, US Food and Drug Administration; NS, not safe; S, safe;
?, unknown. Abbreviations: FDA, US Food and Drug Administration; S, safe; ?, unknown.
310 III Electrophysiology

category C, they have no reported teratogenic effects. Available Supraventricular Tachycardia

data suggest that flecainide is safe during pregnancy; therefore, Paroxysmal supraventricular tachycardia and atrial tachycardia
it is a reasonable choice for treatment of atrial arrhythmias in may increase and become more symptomatic during pregnancy
patients without structural heart disease. or they may decrease. Although the precise mechanism for this
is unclear, it probably relates to increased ectopy (eg, due to vol-
Class III ume or hormonal changes) causing more frequent initiation of
sustained tachycardia.
Sotalol. There are reports of the use of sotalol without adverse In patients with structurally normal hearts, atrioventricular
effects in pregnant patients. Classified in category B, sotalol is reentrant (Wolff-Parkinson-White syndrome and variants) and
therefore a good alternative to class I agents. atrioventricular nodal reentrant tachycardia are the most common
sustained arrhythmias. Atrial tachycardia and atrial fibrillation
Amiodarone. Use of amiodarone should be avoided, even for
are less common. The risk of paroxysmal supraventricular tachy-
a short duration, during pregnancy. Amiodarone has been asso-
cardia is equally distributed throughout pregnancy regardless of
ciated with several problems involving the fetus, including the
the mechanism; management strategies are shown in Figure 26.1.
following:
Because of the increased frequency of arrhythmias during
1. Bradycardia pregnancy and the problems with management, female patients
2. Hypothyroidism with prior symptoms who are planning on pregnancy should be
3. Congenital malformation counseled about the benefit of definitive therapy (such as catheter
4. Premature labor ablation) before becoming pregnant.
5. Death

Arrhythmias in Patients With

Anticoagulation in Pregnancy
Structural Heart Disease
Warfarin is a category X drug, meaning that it is contraindicated
Atrial arrhythmias such as atrial tachycardia and flutter are most
in pregnancy because it crosses the placental barrier. Potential
common in patients with congenital heart disease. Atrial flutter
adverse effects include spontaneous abortion, fetal hemorrhage,
is especially common in patients with prior surgical repair in that
mental retardation, and birth malformations. Table 26.4 lists
scar formation provides the substrate for reentry. The approach
advantages and disadvantages of anticoagulation therapy in
to patients presenting with these arrhythmias is similar to that in
pregnancy. See the chapter “Pregnancy and the Heart” for a full
nonpregnant patients. Because these arrhythmias are not depend-
discussion of anticoagulation during pregnancy.
ent on the atrioventricular node, vagal maneuvers and adenosine
are rarely effective. In all cases, if the arrhythmia is refractory to
Management of Specific Arrhythmias drug therapy or if it is associated with hemodynamic instability,
prompt electrical cardioversion may be required (Figure 26.1).
Arrhythmias in the Normal Heart
Palpitations • In patients with repaired congenital heart disease, atrial arrhyth-
mias such as atrial tachycardia, atrial flutter, or (less likely) atrial
Pregnancy is associated with an increase in “background” ectopic fibrillation are most common.
atrial and ventricular activity (most likely due to volume and hor-
Long-term management of arrhythmias is similar to that for
monal changes) that may present as “palpitations.” In patients
the nonpregnant patient and should be strongly advised after
with a structurally normal heart, these rhythm disorders are usu-
delivery.
ally benign and well tolerated; appropriate management includes
reassurance and avoidance of any identified triggers. Drug or
invasive therapy should be avoided unless hemodynamic insta- Ventricular Tachycardia
bility or intolerable symptoms occur. In the absence of congenital heart disease and with normal right
and left ventricular function, the likelihood of ventricular tachy-
cardia during pregnancy is low.
Table 26.4. Anticoagulant Therapy During Pregnancy
The cause of ventricular tachycardia in the pregnant patient is
Anticoagulant Advantages Disadvantages similar to that in the nonpregnant patient. In patients with struc-
tural heart disease, common causes include arrhythmogenic
Heparin Does not cross placenta Must be administered right ventricular dysplasia, hypertrophic cardiomyopathy, nonis-
Easily and rapidly reversed parenterally
chemic (peripartum) cardiomyopathy, and right ventricular out-
Short half-life Bleeding risk to mother
Maternal osteopenia and
flow tract ventricular tachycardia arising from the region of the
thrombocytopenia patch repair in patients with tetralogy of Fallot. In patients with-
Risk of maternal valve out structural heart disease, idiopathic outflow tract ventricular
thrombosis tachycardia (left bundle branch block inferior axis) arising from
Systemic infection the subpulmonary region of the right ventricle is most common.
Warfarin Most effective for preventing Crosses the placenta These arrhythmias are frequently exercise-induced and respond
thromboembolic events Not easily reversed to β-blockade. Ablation is frequently successful. Arrhythmias
Teratogenic in first due to anomalous coronary artery and, rarely, coronary artery
trimester disease are most likely to be polymorphic ventricular tachycar-
Risk of hemorrhage
dia or ventricular fibrillation.
during labor and
delivery
Although many “idiopathic” ventricular tachycardias
are well tolerated, immediate direct current cardioversion is
 26 Evaluation and Management of Pregnancy-Associated Arrhythmias 311

Figure 26.1. Management of Supraventricular Tachycardia in Pregnancy. DCCV indicates direct current cardioversion.

appropriate for a hemodynamically unstable ventricular tachy- therefore, evaluation of syncope in the pregnant female should
cardia. Intravenous antiarrhythmic drugs may also be used. In focus on excluding other causes.
the pregnant patient, lidocaine is the drug of first choice. Most
drugs used for the treatment of ventricular tachycardia are in cat- Cardiac Arrest
egory C. Exceptions include atenolol and amiodarone, which are
in category D, and sotalol and lidocaine, which are in category B. Cardiac arrest is rare in women of childbearing age, occurring
The use of procainamide and flecainide in pregnant patients has in approximately 1 in 30,000 deliveries. Causes of cardiac arrest
had good effect and no adverse outcomes. For patients who have during pregnancy include the following:
normal renal function, sotalol may be a good choice because it is 1. Postpartum hemorrhage
now a category B drug. 2. Pulmonary thromboembolism or amniotic fluid embolism
3. Eclampsia
4. Anaphylaxis or drug toxicity
Bradycardia 5. Peripartum cardiomyopathy
In most patients, congenital heart block is diagnosed during child- 6. Aortic dissection
hood; however, many cases are discovered incidentally during In general terms, standard guidelines of advanced cardiac
pregnancy. For pregnant patients who are asymptomatic, acute life support apply for medications, intubation, and defibrillation,
intervention is not needed. For patients who are symptomatic in except that amiodarone should be avoided because of the risk
the first or second trimesters, no definite guidelines exist, but of adverse fetal effects. Lidocaine or procainamide should be
permanent pacemaker implantation is probably indicated, with substituted.
avoidance of fluoroscopy if possible. For symptomatic patients
who present at or near term, temporary pacing with induction of Special Issues in Resuscitation
labor may be the procedure of choice. of the Pregnant Female
Until the fetus becomes viable, at approximately 25 weeks,
Syncope resuscitation should be performed as in the nonpregnant patient.
Neurocardiogenic mechanisms are the most common cause of In late-term pregnancy, aortocaval obstruction due to the gravid
syncope in the younger nonpregnant female, accounting for more uterus may reduce venous return and forward flow during chest
than 20% of unexplained cases of syncope in women of child- compressions, thereby limiting efficacy. In such circumstances,
bearing age. Vasovagal syncope, however, is much less com- venous return may be improved by performing cardiopulmo-
mon during pregnancy (perhaps due to increased volume load); nary resuscitation with the patient tilted on her side or in the
312 III Electrophysiology
semirecumbent position. In addition, chest compressions should
usually be performed higher on the chest to accommodate the
shift of pelvic and abdominal contents toward the head.
After 25 weeks, if resuscitation is prolonged (>5 minutes),
emergency cesarean section should be considered in order to
save the fetus. In most cases, external defibrillation (up to 300 J)
can be performed without substantially affecting the fetus and
with a low risk of inducing fetal arrhythmias.
Implantable Cardioverter-Defibrillator Therapy
Implantable cardioverter-defibrillator therapy appears safe in
pregnancy. In addition, the risk of implantable cardioverter-defi-
brillator–related complications or appropriate or inappropriate
discharges is not increased during pregnancy.
Long QT Syndrome
The risk of serious cardiac events in patients with congenital
long QT syndrome is not increased during pregnancy. However,
the risk is increased in the postpartum period. Thus, syncope
in the postpartum period should be evaluated very carefully
to exclude a cardiogenic mechanism. In patients at higher
risk, prolonged hospitalization or (rarely) consideration of an
implantable cardioverter-defibrillator during pregnancy may be
necessary.
Hypertrophic Cardiomyopathy
Because of the hemodynamic changes that occur during preg-
nancy, patients with hypertrophic cardiomyopathy usually expe-
rience symptomatic improvement during pregnancy. However,
maternal and fetal deaths due to ventricular arrhythmias in
patients with hypertrophic cardiomyopathy have been reported.
One contributor may be aortocaval compression, particularly
toward the end of pregnancy. It reduces venous return (and is
exacerbated by blood loss during delivery), which can dramat-
ically decrease preload and lead to hemodynamic collapse and
difficulties in adequate resuscitation.
Labor and Delivery
Cardiac output increases immediately after delivery but subse-
quently declines, reaching pre-pregnancy levels around 2 weeks
post partum. Supraventricular arrhythmias occurring in the peri-
partum period can be managed as for the nonpregnant patient.
Adenosine is safe to use.
Lactation and Breast Feeding
Considerations for drug therapy, including antiarrhythmic drugs,
during breast feeding are similar to those during pregnancy.
Most antiarrhythmic drugs and atrioventricular nodal blocking
agents are excreted in breast milk; thus, continuing their use
during breast feeding must be considered case by case. Use of
warfarin and heparin is safe in nursing mothers.
Summary
• In the majority of cases, arrhythmias arising during pregnancy
can be safely managed conservatively or with minimal medical
therapy.
• In all therapeutic interventions (drug or ablation or device), the
impact on both the mother and the fetus should be considered, and
the least number of medications at the lowest effective dose should
be used.
• Cardioversion is safe during pregnancy and should be an early
option in the management of arrhythmia, especially in the pres-
ence of hemodynamic compromise.
 27

Syncope: Diagnosis and Treatment

WIN-KUANG SHEN, MD

Introduction electrophysiologic study. Instead, this chapter summarizes key

areas for patient evaluation and management, including defini-
Diagnosis and prognosis are 2 key objectives in syncope evalua-
tion and differential diagnosis of syncope, pertinent history and
tion. The diagnosis of the cause of syncope determines the appro-
physical examination findings, appropriate use of diagnostic test-
priate therapy to prevent recurrent syncope. The prognosis of a
ing, and treatment of various conditions causing syncope. For
patient with syncope has broader implications, including the risk
a comprehensive review, refer to “Guidelines for the Diagnosis
of recurrent syncope, sudden cardiac death, and overall survival.
and Management of Syncope (version 2009)” (see “Suggested
Prevention of recurrent syncope is determined by the effective-
Reading” at the end of this chapter).
ness of therapy for a given condition. For example, if sinus node
dysfunction associated with a 10-second pause causes a syncopal
event, successful implantation of a pacemaker will essentially Definition
eliminate any recurrence of syncope due to intermittent sinus
Syncope is a symptom. Syncope is defined as a transient,
pauses or bradycardia. If a syncopal event is caused by a vasova-
self-limited loss of consciousness associated with loss of postural
gal mechanism, the risk of recurrent syncope will remain owing
tone. The onset of syncope is relatively rapid and the subsequent
to the lack of an overall effective therapy for vasovagal syncope.
recovery is spontaneous, complete, and relatively prompt. The
If the patient with a 10-second pause also has cardiomyopathy
underlying mechanism is a transient global cerebral hypoperfu-
with an ejection fraction of 20%, the pacemaker may be effective
sion. Conceptually, one should distinguish syncope from other
in preventing recurrent bradycardia-mediated syncope; however,
forms of TLOC such as concussion due to trauma, epilepsy due to
the overall survival of the patient will be poor unless the under-
primary cerebral electrical abnormality, and “syncope-mimics”
lying comorbid conditions are evaluated and effectively treated.
such as psychiatric problems. The conditions resulting in non-
For the patient who has vasovagal syncope without any other
syncopal TLOC or syncope-mimics are not caused by cerebral
cardiovascular risk factors, recurrent syncope is always possi-
hypoperfusion. Syncope in the context of TLOC is delineated in
ble; however, the long-term prognosis is excellent owing to the
Figure 27.1.
absence of any significant comorbidities. A clear understanding
of the interrelationships between diagnosis and prognosis dur-
ing syncope evaluation allows physicians to deduce an effective Clinical Spectrum and Diagnostic Strategy
management approach.
This chapter is constructed differently from a standard text- The population of patients with syncope is large and heteroge-
book chapter on syncope. It does not review in detail the var- neous, ranging from infants to the very elderly. The spectrum
ious conditions causing syncope, the complex mechanisms of of physiologic and pathophysiologic conditions that may cause
vasovagal syncope, or how to perform tilt-table testing or an syncope ranges from common, benign faints to severe, life-
threatening cardiac structural or rhythm abnormalities. The
prognosis for a patient with syncope depends on the presence or
Abbreviations and acronyms are expanded at the end of this chapter. severity of any underlying organic disease and the severity of any
313
314 III Electrophysiology
Clinical presentation
Loss of consciousness?
Yes
Transient?
Rapid onset?
Short duration?
Spontaneous recovery?
Yes
TLOC
Nontraumatic
Epileptic
Syncope seizure Psychogenic
Figure 27.1. Algorithm for Evaluation of Syncope and Transient Loss of Consciousness (TLOC). SCD indicates sudden cardiac death. (Previously
published. See “Credit Lines” section.)
traumatic injuries that may be sustained after syncope. Because
syncope is often episodic and infrequent in occurrence, estab-
lishing a cause-and-effect relationship can be challenging.
Several key elements are critical to the formulation of a logi-
cal diagnostic strategy:
1. One must be familiar with the overall organization and broad catego-
ries of syncope and nonsyncopal conditions with or without TLOC.
In the evaluation of any patient presenting with syncope or syncope-
like symptoms, the following differential scheme can be useful: neu-
rally mediated conditions (most common; present in approximately
40%–60% of the entire syncope population), orthostatic hypotension,
cardiopulmonary conditions, cerebrovascular disease, and nonsynco-
pal conditions such as seizures and metabolic or psychiatric disor-
ders. A classification of syncope is shown in Box 27.1, and conditions
that are incorrectly diagnosed as syncope are listed in Box 27.2.
2. “Standard of care” initial evaluation includes a thorough history,
physical examination, orthostatic blood pressure checks, and a
12-lead electrocardiogram.
3. After initial evaluation, the presumed cause of syncope can be catego-
rized as confirmed or certain, suspected or probable, or unexplained.
4. When the cause of syncope is suspected or unexplained after ini-
tial investigation, further evaluation should be individualized and
based on the frequency and severity of the patient’s clinical presenta-
tion, the presence or absence of underlying heart disease (including
familial conditions), and the anticipated prognosis.
No
Altered
Fall consciousness
No
Coma Aborted Other
SCD
Traumatic
Rare causes
History and Physical Examination
The most important and fruitful elements of the evaluation are
a detailed clinical history and a careful physical examination.
A presumptive diagnosis can be established about 50% of the
time after a thorough history and physical examination by an
experienced physician. Impressions from the initial clinical visit
are critical in further triaging patients for appropriate subsequent
evaluation and management.
During acquisition of the clinical history and physical exami-
nation, the following components should be considered:
1. Age and sex—Among elderly patients, clinical presentation is often
less typical and potential causes of syncope can be multiple. Cardiac
causes are more common in the elderly because cardiovascular dis-
eases are more prevalent. Orthostatic intolerance is common among
young women with syncope.
2. Position—Cardiac causes of syncope can occur in any position;
neurally mediated syncope does not usually occur in the supine
position.
3. Surrounding circumstances—Are there any obvious precipitants
such as physical or emotional distress, pain, time relationship to
meals, drugs, micturition, defecation, cough, swallowing, exercise,
or neck turning? Did syncope occur during or after exercise?
4. Premonitory symptoms—Shortly before the syncopal event, were
there symptoms of light-headedness, tunnel vision, nausea, vomiting,
 27 Syncope: Diagnosis and Treatment 315

Box 27.1. (Continued)

Box 27.1. Classification of Syncope
Congenital anomalies of coronary arteries
Reflex (neurally mediated) syncope Prosthetic valve dysfunction
Vasovagal Other
Mediated by emotional distress: fear, pain, instrumentation, Pulmonary embolus
blood phobia
Acute aortic dissection
Mediated by orthostatic stress
Pulmonary hypertension
Situational
Cough, sneeze
abdominal discomfort, sweating, aura, chest pain, or palpitations?
Gastrointestinal tract stimulation (swallow, defecation, visceral
What was the duration of the symptoms?
pain)
5. The index event—Was the index event witnessed? If it was wit-
Micturition (postmicturition) nessed, was the fall or slumping abrupt? Was the patient breath-
ing? Did the patient have pallor, cyanosis, or shaking or tonic-clonic
Postexercise
movements? If the event was not witnessed, was there associated
Postprandial injury, urinary or fecal incontinence, or tongue biting? What was
the estimated duration of unresponsiveness?
Others (eg, laughing, brass instrument playing, weight lifting)
6. Recovery—Was the recovery immediate or prolonged? Did the
Carotid sinus syncope patient have confusion or any recall of the event?
7. Past medical history—Was the index event isolated? Was it the first
Atypical forms (without apparent triggers or occurrence or a recurrence? If recurrent, were previous episodes
atypical presentation) similar or different? A comprehensive review of coexisting medical
and cardiovascular conditions is required.
Syncope due to orthostatic hypotension
8. Medications—A complete list of prescribed and over-the-counter
Primary autonomic failure medications should be documented. Pay particular attention to
negative chronotropic or dromotropic drugs (β-blockers, calcium
Pure autonomic failure, multiple system atrophy, Parkinson channel blockers, digoxin, and vagotonic agents), antihypertensive
disease with autonomic failure, Lewy body dementia drugs or vasodilators (diuretics, α-antagonists, angiotensin receptor
Secondary autonomic failure blockers, and angiotensin-converting enzyme inhibitors), and drugs
prolonging cardiac repolarization or the QT interval (Box 27.3).
Diabetes mellitus, amyloidosis, uremia, spinal cord injuries 9. Family history—Family history should be highlighted when a
Drug-induced orthostatic hypotension young patient presents with unexplained syncope without preex-
isting structural heart disease. The ethnicity of the patient’s fam-
Alcohol, vasodilators, diuretics, phenothiazine, antidepressants ily should be reviewed. Brugada syndrome was first reported in
Europe (Flemish families and southern European countries) and
Volume depletion
southern Asia. Arrhythmogenic right ventricular cardiomyopathy
Hemorrhage, diarrhea, vomiting, etc was reported in northern Italy and France. With the advancement of
medical science and increased awareness of these conditions, inter-
Cardiac syncope: cardiovascular national registries have been documenting accumulated experi-
arrhythmia as primary cause ences from around the world. Be aware of first-degree relatives with
Bradycardia sudden, unexplained, premature death (younger than 50 years). In
conjunction with appropriate diagnostic laboratory evaluation, a
Sinus node dysfunction (including bradycardia or tachycardia positive family history of premature, unexplained sudden death is a
syndrome) critical factor in the risk stratification for sudden death of a patient
Atrioventricular conduction system disease who presents with unexplained syncope.
10. Physical examination—Blood pressure and heart rate should be
Implanted device malfunction measured in the supine, sitting, and standing positions. A postural
decrease in systolic blood pressure of more than 20 mm Hg is con-
Tachycardia
sidered significant. Measuring blood pressure in both arms and lis-
Supraventricular tening for bruits in the carotid, subclavian, and temporal areas may
identify patients with vascular disorders, such as cerebrovascular
Ventricular (idiopathic; secondary to structural heart disease or
disease, Takayasu disease, or subclavian steal syndrome. Cardiac
channelopathies)
examination should focus on identifying signs of overall cardiac
Drug-induced bradycardia and tachyarrhythmias

Structural disease
Box 27.2. Conditions Incorrectly Diagnosed as Syncope
Cardiac valvular disease
Cataplexy
Acute myocardial infarction or ischemia
Drop attacks
Hypertrophic cardiomyopathy Falls
Cardiac masses (atrial myxoma, tumors, etc) Functional conditions (psychogenic
pseudosyncope)
Pericardial disease or tamponade
Transient ischemic attack of carotid origin
(continued)
316 III Electrophysiology
Box 27.3. Factors Responsible for QT Prolongation
Antiarrhythmics
Class IA: quinidine, procainamide, disopyramide
Class III: sotalol, N-acetylprocainamide, ibutilide,
dofetilide, azimilide, amiodarone
Class IV: bepridil
Antimicrobials
Antibiotics: macrolides (erythromycin),
trimethoprim-sulfamethoxazole
Antifungals: itraconazole, ketoconazole
Antimalarials: chloroquine
Antiparasitic: pentamidine
Antivirals: amantadine
Antihistamine: terfenadine, astemizole
Antidepressants: tricyclics (amitriptyline),
tetracyclics
Psychotropics: haloperidol, droperidol,
phenothiazines
Miscellaneous: cisapride, probucol, ketanserin,
vasopressin
Organophosphate poisoning, chloral hydrate
overdose
Electrolyte abnormalities: hypokalemia,
hypomagnesemia, hypocalcemia
Severe bradyarrhythmias: sinus node dysfunction,
complete atrioventricular block
Intrinsic cardiac disease: myocarditis
Hypothyroidism
Intracranial pathology: subarachnoid
hemorrhage, head trauma, encephalitis
Right radical neck dissection
Nutritional
Starvation, anorexia nervosa, liquid protein diet
function, valvular heart disease, hypertrophic cardiomyopathy, and
pulmonary hypertension. Carotid sinus massage should be a man-
datory part of the examination in older patients with syncope. It is
imperative, however, to exclude carotid artery disease before per-
forming carotid sinus massage. The presence of carotid bruits or a
prior history of stroke or transient ischemic attacks is considered a
contraindication for carotid sinus massage.
Pertinent clinical features associated with specific causes of
syncope are summarized in Box 27.4. The specificity and sen-
sitivity of these clinical features vary significantly in different
patient populations and from time to time. Combinations of these
characteristics, rather than a single feature, provide the basis for
management strategies.
Laboratory Investigation
Diagnostic testing is summarized in Box 27.5. Although “rou-
tine” tests for syncope evaluation have not been established, the
general consensus is that results of a complete blood cell count,
electrolytes, and fasting glucose and a standard ECG should be
obtained during the initial evaluation. The decision to pursue
additional testing depends on the patient’s clinical presentation
Box 27.4. Clinical Features That Can Suggest the
Cause of Syncope on Initial Evaluation
Neurally mediated syncope
Absence of heart disease
Long history of recurrent syncope
After sudden unexpected unpleasant sight, sound,
smell, or pain
Prolonged standing or being in crowded, hot places
Nausea or vomiting associated with syncope
During or after a meal
With head rotation or pressure on carotid sinus (as
with tumors, shaving, tight collars)
After exertion
Syncope due to orthostatic hypotension
After standing up
Temporal relationship with new use or changed
dosage of vasodepressive drugs, leading to
hypotension
Prolonged standing, especially in crowded, hot
places
Presence of autonomic neuropathy or parkinsonism
Standing after exertion
Cardiovascular syncope
Presence of definite structural heart disease
Family history of unexplained sudden death or
channelopathy during exertion or while supine
Abnormal electrocardiogram
Sudden-onset palpitation immediately followed by
syncope
Previously published. See “Credit Lines” section.
and the ultimate diagnostic, therapeutic, and educational goals
of the procedure. The risk-benefit ratio should be clearly defined
before any diagnostic test is undertaken.
Risk Stratification
When a patient has syncope, the fundamental and comprehen-
sive diagnostic goal is to determine the cause of the symptoms;
however, mortality risk assessment in determining prognosis
and guiding effective therapy is an important component of the
overall objectives. Although syncope alone does not independ-
ently predict death, syncope predicts increased mortality among
patients who have cardiovascular disease (which increases the
chances that the syncope has a cardiogenic cause). Factors asso-
ciated with cardiac and noncardiac causes of syncope are sum-
marized in Box 27.6.
Selected Diagnostic Testing
Specific, additional diagnostic evaluation recommendations
based on the results of the initial evaluation have been provided
in a recent update of the ESC guidelines. A brief review of the
indications and contraindications for tilt-table testing, electro-
physiologic study, and implantable loop recorders follows.
 27 Syncope: Diagnosis and Treatment 317

Box 27.5. Syncope Evaluation Box 27.6. Factors Associated With Cardiac and
“Routine” testing Noncardiac Causes of Syncope
Complete blood cell count Factors associated with cardiogenic syncope
Electrolytes History and physical examination
Blood glucose Coronary artery disease or prior myocardial infarction
Electrocardiography Congestive heart failure
“Elective” testing Older age
Echocardiography Abrupt onset, during exertion, or when supine
Holter monitor Serious injuries

Ambulatory continuous blood pressure monitor

Abnormal cardiovascular examination findings

Portable event recorder Laboratory findings

Implantable loop recorder Abnormal electrocardiogram: Q wave, bundle branch block, or
sinus bradycardia
Tilt-table testing
Structural heart disease
Electrophysiologic testing
Left ventricular dysfunction
Neurologic testing
Factors associated with noncardiogenic syncope
Autonomic testing
History and physical examination
Electroencephalography
Isolated syncope without underlying cardiovascular disease
Ophthalmography
Young age
Carotid ultrasonography
Symptoms consistent with a vasovagal cause
Transcranial Doppler ultrasonography
Normal cardiovascular examination findings
Computed tomography
Magnetic resonance imaging Laboratory findings

Endocrinologic testing
Normal electrocardiogram

Serum catecholamines
Urine metanephrines
Other cardiac testing 1. Mixed response—manifested by coexisting bradycardia and hypo-
Treadmill exercise test tension (Figure 27.2)
2. Cardioinhibitory response—manifested by persistent bradycardia or
Coronary angiography prolonged pauses and an absence of significant hypotension when
bradycardia is prevented by pacing or a vagolytic agent such as atro-
pine (Figure 27.3)
3. Vasodepressor response—manifested by significant hypotension in
Tilt-Table Testing the absence of bradycardia (Figure 27.4)

The most common cause of syncope and neurally mediated syn-
cope is vasovagal (common faint). Although the pathophysiology
of vasovagal response is incompletely understood, it is generally
accepted that certain physical or emotional distresses trigger a
chain of events that culminate in vasodilation or bradycardia
(or both). This in turn leads to the hypotension and loss of con-
sciousness associated with vasovagal syncope.
It is generally thought that tilt-table testing provokes a vaso-
vagal response by venous pooling and orthostatic distress.
Protocols for tilt-table testing have not been standardized. Most
recent guidelines suggest a tilt-table test duration of 20 to 45
minutes at 60° to 70°. Pharmacologic agents are often used to
provoke a positive response if vasovagal syncope is not induced
by a passive tilt-table test alone. Intravenous infusion of isopro-
terenol and sublingual nitroglycerin are the 2 most frequently
used provocative agents in conjunction with tilt-table testing. Be
aware that increased sensitivity (a positive response) from use of
a pharmacologic agent during tilt-table testing is inevitably asso-
ciated with a decrease in the specificity of the test. Carefully seek
a correlation of the symptoms induced during tilt-table testing
and the symptoms during the spontaneous clinical event.
The vasovagal response induced during tilt-table testing can
be classified into 3 subtypes:
Definitions for this classification are provided in Box 27.7.
The consensus is that tilt-table testing is indicated if
patients have presumed vasovagal syncope or 1 or more of the
following:
1. Syncope without evidence of organic heart disease
2. One episode of syncope that occurred with an injury or a motor vehi-
cle accident or in a high-risk situation
3. Syncope with a known cause and a treatment that vasovagal syncope
could affect
Tilt-table testing is not indicated if patients have had only 1
episode of syncope that occurred without injury in a low-risk
situation and if there is no clinical reason to strongly suspect
vasovagal syncope. Contraindications for tilt-table testing are
critical obstructive cardiac disease (eg, critical stenosis of the
mitral valve or a proximal segment of a coronary artery, severe
obstruction of the left ventricular outflow tract) or critical cere-
brovascular stenosis. Indications for tilt-table testing from the
ESC guidelines are summarized in Table 27.1.
Electrophysiologic Testing
When syncope is unexplained after the initial evaluation, an
electrophysiologic study should be considered if the risk of an
318 III Electrophysiology

Beginning tilt 10 min after tilt AV pacing during tilt

Baseline BP = 136/67 mm Hg BP = 54/30 mm Hg BP = 47/29 mm Hg

HR = 115 bpm HR = 39 bpm PCL = 700 ms
I
I I

200 V6 V6 V6
BP, mm Hg

BP
100
BP BP

1s
0

Figure 27.2. Mixed Response During Tilt-Table Testing. Electrocardiographic and arterial blood pressure (BP) recordings are shown. Left,
Asymptomatic. Middle, During syncope. Heart rate and blood pressure decrease markedly. Patient experiences typical symptoms. Right, dual-
chamber pacing with atrioventricular (AV) pacing cycle length (PCL) of 700 ms results in an increased heart rate, but the patient is still hypotensive
and symptomatic. The results of this test suggest that the patient’s symptoms are primarily due to the vasodepressor component, and they are unlikely
to be relieved by permanent pacing. bpm indicates beats per minute; HR, heart rate.

arrhythmic cause of syncope is high. During an electrophysi-
ologic study for syncope evaluation, the following assessment
should be made: 1) sinus node function, 2) atrioventricular
node and His-Purkinje system conduction, and 3) inducibility
of supraventricular and ventricular arrhythmias. Details of the
programmed stimulation protocols during an electrophysiologic
study are provided in Chapter 19 (“Indications for Invasive and
Noninvasive Electrophysiologic Testing”).
Indications for electrophysiologic testing from the ESC guide-
lines are summarized in Table 27.2. Depending on the patient

Tilt
I 9.1 s

V6
200
BP, mm Hg

BP
100

0 1s

AV PCL = 800 ms
I

V1
HRA
200
HBE
BP, mm Hg

RV
BP 100

1s
0

Figure 27.3. Cardioinhibitory Response During Tilt-Table Testing. Top, Prolonged pauses due to sinus arrest. Lower, Absence of significant hypo-
tension when bradycardia is prevented by pacing. Electrocardiographic and arterial blood pressure (BP) recordings are shown. AV indicates atrioven-
tricular; HBE, His bundle electrogram; HRA, high right atrial electrogram; PCL, pacing cycle length; RV, right ventricular electrogram.
 27 Syncope: Diagnosis and Treatment 319

Baseline, supine 8 min after tilt Recovery, supine

BP = 150/83 mm Hg BP = 70/50 mm Hg BP = 138/78 mm Hg

HR = 85 bpm HR = 140 bpm HR = 145 bpm

V6

BP

Figure 27.4. Vasodepressor Response During Tilt-Table Testing. Significant hypotension develops despite tachycardic heart rate (middle panel).
Electrocardiographic and arterial blood pressure (BP) recordings are shown. bpm indicates beats per minute; HR, heart rate.

population studied and the diagnostic end points, the yield from
electrophysiologic testing varies considerably. It is uncommon
(<10%–20%) for an electrophysiologic study to identify an
arrhythmic cause of unexplained syncope in patients who do
not have underlying heart disease or abnormalities on ECG. The
yield from an electrophysiologic study is higher among patients
who have underlying heart disease in association with left ven-
tricular dysfunction and an abnormal ECG as a consequence of
Box 27.7. Classification of Positive Responses to Tilt-
Table Testing
Type 1—mixed
HR decreases at syncope, but the ventricular rate
does not decrease to <40 bpm or decreases to <40
bpm for <10 s with or without asystole of <3 s
BP decreases before HR decreases
Type 2A—cardioinhibition without asystole
HR decreases to a ventricular rate <40 bpm for >10 s
without asystole of >3 s
BP decreases before HR decreases
Type 2B—cardioinhibition with asystole
Asystole occurs for >3 s
Decrease in BP occurs with or before decrease in HR
Type 3—vasodepressor
HR does not decrease >10% from its peak at
syncope
Abbreviations: BP, blood pressure; bpm, beats per minute; HR,
heart rate.
coronary artery disease or prior myocardial infarction (or both).
The value of electrophysiologic testing is less well established
among patients who have nonischemic cardiomyopathy and are
undergoing syncope evaluation. Although an electrophysiologic
study is indicated for this patient population with syncope, a neg-
ative study does not predict low mortality or low risk of sudden
cardiac death for patients who have a low ejection fraction.
As the indications for ICDs rapidly expand, the indications
for electrophysiologic testing are rapidly evolving. The prophy-
lactic use of ICDs for prevention of primary sudden cardiac
death is now considered to be the standard of care for asympto-
matic patients with an ejection fraction less than 35% with either
an ischemic or a nonischemic cause and NYHA class II or III
functional capacity. After patients in this population experience
syncope, electrophysiologic testing is no longer deemed neces-
sary since ICD implantation would provide appropriate therapy
for syncope caused by intermittent bradycardia or ventricular
tachycardia.
Implantable Loop Recorders
The implantable loop recorder has the capability of long-term
(14-16 months) continuous rhythm monitoring in patients with
infrequent episodes of syncope that may have an arrhythmic
cause. The advantage of this approach is that the rhythm can be
documented when the patient experiences a recurrent syncopal
spell after the loop recorder is implanted. The potential disad-
vantage of this “wait-and-watch” approach is the uncertain risk
of increased morbidity or mortality for the patient from waiting
for another event to occur. Although several clinical studies have
been conducted, the precise role of the implantable loop recorder
in the evaluation of syncope has not been determined. Indications
for consideration of an implantable loop recorder from the recent
ESC guidelines are summarized in Table 27.3. In general, the use
of an implantable loop recorder is considered when infrequent
episodes remain unexplained and when an arrhythmic cause
cannot be excluded.
320 III Electrophysiology

Table 27.1. Recommendations for Tilt-Table Testing

Class of
Indication Recommendation Level of Evidence
Tilt-table testing is indicated in the case of an unexplained single I B
syncopal episode in high-risk settings (eg, occurrence of or
potential risk of physical injury, or occupational implications),
or recurrent episodes in the absence of organic heart disease, or
in the presence of organic heart disease after cardiac causes of
syncope have been excluded
Tilt-table testing is indicated when it is of clinical value to I C
demonstrate susceptibility to reflex syncope to the patient
Tilt-table testing should be considered to discriminate between IIa C
reflex and orthostatic hypotension syncope
Tilt-table testing may be considered for differentiating syncope IIb C
with jerking movements from epilepsy
Tilt-table testing may be indicated for evaluating patients with IIb C
recurrent unexplained falls
Tilt-table testing may be indicated for evaluating patients with IIb C
frequent syncope and psychiatric disease
Tilt-table testing is not recommended for assessment of treatment III B
Isoproterenol tilt-table testing is contraindicated in patients with III C
ischemic heart disease
Previously published. See “Credit Lines” section.

Therapeutic Management Vasovagal Syncope (Common Faint)

The appropriate treatment of syncope is usually quite obvious
after the cause of syncope has been determined. Indications
for pacemaker therapy in patients with documented brady-
cardia associated with syncope are discussed in Chapter 28
(“Pacemakers”), appropriate therapy for symptomatic supraven-
tricular arrhythmias (drugs or ablation) is discussed in Chapter 23
(“Supraventricular Tachycardia and Ectopy”), and therapy for
ventricular arrhythmias (drugs, ablation, or ICD) is discussed in
Chapter 24 (“Ventricular Tachycardia”). The following sections
provide a brief review of 2 areas related to syncope management
that are challenging in clinical practice: 1) therapy for vasovagal
syncope and 2) the role of ICDs after a syncopal event.
The goals of therapy for patients with vasovagal syncope are
3-fold: 1) educating patients on the causes, prognosis, and man-
agement; 2) improving quality of life; and 3) reducing recur-
rences and associated morbidity. Although most patients with
vasovagal syncope respond to conservative measures (patient
education, adequate hydration, liberalization of salt intake, com-
pression stockings, and orthostatic training), some patients con-
tinue to have recurrent syncope. Pharmacologic therapy could be
considered in this group of patients.
Although the autonomic reflex underlying the vasovagal
response is complex and is not fully understood, it is gener-
ally accepted that the Bezold-Jarisch reflex is important in the

Table 27.2. Recommendations for Electrophysiologic Study

Class of
Indication Recommendation Level of Evidence
For patients with ischemic heart disease, EPS is indicated when I B
initial evaluation suggests an arrhythmic cause of syncope
unless there is already an established indication for ICD
For patients with bundle branch block, EPS should be considered IIa B
when noninvasive tests have failed to make the diagnosis
For patients with syncope preceded by sudden and brief IIb B
palpitations, EPS may be performed when other noninvasive
tests have failed to make the diagnosis
For selected patients with Brugada syndrome, ARVC, and IIb C
hypertrophic cardiomyopathy, an EPS may be performed
For selected patients with high-risk occupations for whom IIb C
every effort to exclude a cardiovascular cause of syncope is
warranted, an EPS may be performed
EPS is not recommended for patients with a normal III B
electrocardiogram, no heart disease, and no palpitations
Abbreviations: ARVC, arrhythmogenic right ventricular cardiomyopathy; EPS, electrophysiologic study; ICD,
implantable cardioverter-defibrillator.
Previously published. See “Credit Lines” section.
 27 Syncope: Diagnosis and Treatment 321

Table 27.3. Recommendations for Implantable Loop Recorder (ILR)

Class of
Indication Recommendation Level of Evidence
An early phase of evaluation for patients with recurrent syncope I B
of uncertain origin, absence of high-risk criteria, and a high
likelihood of recurrence within the battery longevity of the device
High-risk patients for whom a comprehensive evaluation did not I B
demonstrate a cause of syncope or lead to a specific treatment
ILR should be considered to assess the contribution of bradycardia IIa B
before embarking on cardiac pacing in patients with suspected
or certain reflex syncope presenting with frequent or traumatic
syncopal episodes
External loop recorders should be considered for patients who have IIa B
an intersymptom interval ≤4 wk
Previously published. See “Credit Lines” section.

pathogenesis of vasovagal syncope. In this reflex, activation of
mechanoreceptors in the left ventricle (as a result of increased
cardiac contractility from sympathetic activation) stimulates C
fibers, which in turn leads to vagal activation and a withdrawal
of sympathetic outflow. Conceptually, the aim of most pharma-
cologic therapies is to interrupt 1 or more components of this
reflex arc (Figure 27.5).
One of the most challenging aspects of assessing the effi-
cacy of any therapeutic intervention for vasovagal syncope is the
sporadic, infrequent, and sometimes clustered nature of this con-
dition in a highly heterogeneous population. Most of the clinical
trials with drug therapy for vasovagal syncope have enrolled rel-
atively few patients and have had limited follow-up. β-Blockers
have been widely used for many years as therapy for recurrent
vasovagal syncope. However, evidence from randomized, pla-
cebo-controlled, double-blind clinical trials clearly does not sup-
port this widespread practice. According to data from limited,
randomized, placebo-controlled, double-blind clinical trials,

Fludrocortisone
Midodrine
Salt/fluids
Midodrine Stockings
Sympathetic efferent
S Theophylline

V
Vag
Vagal efferent

Selective serotonin
reuptake inhibitors

Anticholinergics

β-Blockers
Blockers
Disopyramide
Figure 27.5. Neurocardiogenic Reflexes and Therapeutic Targets.
322 III Electrophysiology

midodrine appears to be effective in selected patient groups.
When used under proper supervision, it is usually well tolerated
with minimal side effects.
An area of ongoing investigation is cardiac pacing for vaso-
vagal syncope. Although it was listed as a class IIA indication
for recurrent vasovagal syncope in the 2002 guidelines, recent
data suggest that its efficacy is equivocal and that it should be
reserved only for severe and refractory cases in patients with
documented bradycardia.
In summary, the treatment of vasovagal syncope should be
individualized. In some cases, reassurance may suffice. In oth-
ers, augmenting central blood volume by increasing fluid or
salt intake (or both) is effective. The role of nonpharmacologic
physical maneuvers is increasingly recognized given that recent
clinical trial data support their efficacy. Of the many pharma-
cologic agents, the most promising is midodrine. Its use should
be reserved for patients with recurrent and refractory syncope.
Recommendations for treatment of reflex syncope from the ESC
guidelines are summarized in Table 27.4.
ICD Implantation and Syncope Management
One should keep a clear perspective in assessing the mor-
tality risk of syncope patients. The vast majority of syncope
patients have disturbing or disabling symptoms that are not life-
threatening but may cause injuries and substantially decrease
their quality of life. However, syncope could be the fi rst (and
perhaps last) clinical event in patients with an increased risk
of cardiac and arrhythmic death. These patients must be rec-
ognized during the evaluation, and therapy (ie, ICD implan-
tation) must be implemented if appropriate. One must also be
cognizant that although ICDs may prevent arrhythmic death,
they may not alleviate syncope. This aspect of the treatment of
the high-risk fainter is a common dilemma in ischemic heart
disease and cardiomyopathy patients as well as in persons with
familial conditions such as long QT syndrome, Brugada syn-
drome, arrhythmogenic right ventricular cardiomyopathy, and
hypertrophic cardiomyopathy.
Indications from the most recent guidelines for ICD implan-
tation in patients with increased risk of sudden cardiac death
are reviewed elsewhere in this book (Chapter 31, “Implantable
Cardioverter-Defibrillator Trials and Prevention of Sudden
Cardiac Death”). In brief, some of the key elements for ICD con-
sideration during syncope evaluation are the following:
1. When unexplained syncope occurs in patients who meet the current
criteria for ICD implantation for primary prevention of sudden car-
diac death (ischemic or nonischemic cardiomyopathy with ejection
fraction ≤35% and NYHA functional class II or III, or with ejection
fraction ≤30% and NYHA functional class I), ICD implantation is
indicated without an electrophysiologic study.
2. For syncope patients who have ischemic or nonischemic cardiomy-
opathy and an ejection fraction greater than 35%, ICD implantation
is indicated when sustained ventricular arrhythmia is inducible by
electrophysiologic testing.
3. Among patients with primary arrhythmic congenital conditions
without structural abnormalities (long QT syndrome or Brugada
syndrome), unexplained syncope is recognized as a risk factor asso-
ciated with increased sudden death. Data have been accumulated
from international registries for these conditions. ICD implanta-
tion is considered a class II indication for these patients. There is
no proven role for electrophysiologic testing in long QT syndrome
patients with syncope. The value of electrophysiologic testing in
Brugada syndrome patients continues to evolve as a risk assessment
tool for asymptomatic patients.
4. Among patients with primary congenital structural conditions
(eg, hypertrophic cardiomyopathy, arrhythmogenic right ventric-
ular cardiomyopathy), unexplained syncope is also a risk factor
associated with increased risk of sudden death. ICD implantation
is considered a class II indication for these patients. The electro-
physiologic study is of limited use for risk stratification because of
the uncertain negative predictive value of the test in these patient
populations.

Table 27.4. Recommendations for Treatment of Reflex Syncope

Class of
Indication Recommendation Level of Evidence
Explanation of diagnosis, provision of reassurance, and explanation I C
of risk of recurrence are indicated for all patients
Isometric physical counter maneuvers are indicated for patients I B
with prodrome
Cardiac pacing should be considered for patients with dominant IIa B
cardioinhibitory carotid sinus syndrome
Cardiac pacing should be considered for patients who have IIa B
frequent, recurrent reflex syncope; are older than 40 y; and have
a documented spontaneous cardioinhibitory response during
monitoring
Midodrine may be indicated for patients with vasovagal syncope IIb B
refractory to lifestyle measures
Tilt training may be useful for education of patients, but long-term IIb B
benefit depends on compliance
Cardiac pacing may be indicated for patients who have a tilt- IIb C
induced cardioinhibitory response and recurrent, frequent,
unpredictable syncope and are older than 40 y after alternative
therapy has failed
Cardiac pacing is not indicated in the absence of a documented III C
cardioinhibitory reflex
β-Adrenergic blocking drugs are not indicated III A
Previously published. See “Credit Lines” section.
 27 Syncope: Diagnosis and Treatment
Suggested Reading
Benditt DG, Ferguson DW, Grubb BP, Kapoor WN, Kugler J, Lerman
BB, et al. Tilt table testing for assessing syncope. American College of
Cardiology. J Am Coll Cardiol. 1996 Jul;28(1):263–75.
Blomstrom-Lundqvist C, Scheinman MM, Aliot EM, Alpert JS,
Calkins H, Camm AJ, et al; European Society of Cardiology Com-
mittee, NASPE-Heart Rhythm Society. ACC/AHA/ESC guidelines
for the management of patients with supraventricular arrhythmias:
executive summary: a report of the American College of Cardiology/
American Heart Association Task Force on Practice Guidelines and
the European Society of Cardiology Committee for Practice Guide-
lines (writing committee to develop guidelines for the management
of patients with supraventricular arrhythmias) developed in collabora-
tion with NASPE-Heart Rhythm Society. J Am Coll Cardiol. 2003 Oct
15;42(8):1493–531.
Brignole M, Alboni P, Benditt D, Bergfeldt L, Blanc JJ, Bloch Thomsen
PE, et al; Task Force on Syncope, European Society of Cardiology.
Guidelines on management (diagnosis and treatment) of syncope. Eur
Heart J. 2001 Aug;22(15):1256–306.
Brignole M, Alboni P, Benditt DG, Bergfeldt L, Blanc JJ, Bloch Thom-
sen PE, et al; Task Force on Syncope, European Society of Cardiol-
ogy. Guidelines on management (diagnosis and treatment) of syncope:
update 2004. Europace. 2004 Nov;6(6):467–537.
Epstein AE, DiMarco JP, Ellenbogen KA, Estes NA 3rd, Freedman RA,
Gettes LS, et al; American College of Cardiology/American Heart
Association Task Force on Practice Guidelines (Writing Committee
to Revise the ACC/AHA/NASPE 2002 Guideline Update for Implan-
tation of Cardiac Pacemakers and Antiarrhythmia Devices); American
Association for Thoracic Surgery; Society of Thoracic Surgeons. ACC/
AHA/HRS 2008 Guidelines for Device-Based Therapy of Cardiac
Rhythm Abnormalities: a report of the American College of Cardiol-
ogy/American Heart Association Task Force on Practice Guidelines
(Writing Committee to Revise the ACC/AHA/NASPE 2002 Guideline
Update for Implantation of Cardiac Pacemakers and Antiarrhythmia
Devices): developed in collaboration with the American Association
for Thoracic Surgery and Society of Thoracic Surgeons. Circulation.
2008 May 27;117(21):e350–408. Epub 2008 May 15. Erratum in:
Circulation. 2009 Aug 4;120(5):e34–5.
Gregoratos G, Abrams J, Epstein AE, Freedman RA, Hayes DL, Hlatky
MA, et al; American College of Cardiology/American Heart Asso-
ciation Task Force on Practice Guidelines/North American Society
for Pacing and Electrophysiology Committee to Update the 1998
Pacemaker Guidelines. ACC/AHA/NASPE 2002 guideline update
for implantation of cardiac pacemakers and antiarrhythmia devices:
summary article: a report of the American College of Cardiology/
American Heart Association Task Force on Practice Guidelines (ACC/
323
AHA/NASPE Committee to Update the 1998 Pacemaker Guidelines).
Circulation. 2002 Oct 15;106(16):2145–61.
Hunt SA, Abraham WT, Chin MH, Feldman AM, Francis GS, Ganiats
TG, et al. 2009 focused update incorporated into the ACC/AHA 2005
Guidelines for the Diagnosis and Management of Heart Failure in
Adults: a report of the American College of Cardiology Foundation/
American Heart Association Task Force on Practice Guidelines: devel-
oped in collaboration with the International Society for Heart and Lung
Transplantation. Circulation. 2009 Apr 14;119(14):e391–479. Epub
2009 Mar 26. Erratum in: Circulation. 2010 Mar 30;121(12):e258.
Strickberger SA, Benson DW, Biaggioni I, Callans DJ, Cohen MI, Ellen-
bogen KA, et al; American Heart Association Councils on Clinical
Cardiology, Cardiovascular Nursing, Cardiovascular Disease in the
Young, and Stroke; Quality of Care and Outcomes Research Interdisci-
plinary Working Group; American College of Cardiology Foundation;
Heart Rhythm Society; American Autonomic Society. AHA/ACCF
Scientific Statement on the evaluation of syncope: from the Ameri-
can Heart Association Councils on Clinical Cardiology, Cardiovas-
cular Nursing, Cardiovascular Disease in the Young, and Stroke, and
the Quality of Care and Outcomes Research Interdisciplinary Working
Group; and the American College of Cardiology Foundation: in col-
laboration with the Heart Rhythm Society: endorsed by the American
Autonomic Society. Circulation. 2006 Jan 17;113(2):316–27. Erratum
in: Circulation. 2006 Apr 11;113(14):e697.
Task Force for the Diagnosis and Management of Syncope; European
Society of Cardiology (ESC); European Heart Rhythm Association
(EHRA); Heart Failure Association (HFA); Heart Rhythm Society
(HRS); Moya A, Sutton R, Ammirati F, Blanc JJ, Brignole M, Dahm JB,
et al. Guidelines for the diagnosis and management of syncope (version
2009). Eur Heart J. 2009 Nov;30(21):2631–71. Epub 2009 Aug 27.
Zipes DP, DiMarco JP, Gillette PC, Jackman WM, Myerburg RJ, Rahim-
toola SH, et al. Guidelines for clinical intracardiac electrophysiologi-
cal and catheter ablation procedures. A report of the American College
of Cardiology/American Heart Association Task Force on Practice
Guidelines (Committee on Clinical Intracardiac Electrophysiologic
and Catheter Ablation Procedures), developed in collaboration with
the North American Society of Pacing and Electrophysiology. J Am
Coll Cardiol. 1995 Aug;26(2):555–73.
Abbreviations
ECG electrocardiogram
ESC European Society of Cardiology
ICD implantable cardioverter-defibrillator
NYHA New York Heart Association
TLOC transient loss of consciousness
 28
Pacemakers
YONG-MEI CHA, MD, and DAVID L. HAYES, MD
Indications for Permanent Pacing
Guidelines for permanent pacemaker implantation were estab-
lished in 1984 by a joint task force of the American Heart
Association and American College of Cardiology. The guide-
lines have been updated multiple times (most recently in 2008).
Indications for permanent pacing are divided into 3 classes:
1. Class I indications include those for which pacing is considered nec-
essary, provided that the indication is chronic or recurrent and is not
due to transient underlying causes, such as drugs, electrolyte imbal-
ance, or acute myocardial infarction. A single symptomatic episode
is sufficient to establish the necessity for pacing. Symptoms must be
clearly related to the rhythm disturbance and include weakness, lim-
ited exercise tolerance, syncope or presyncope, confusion, seizures,
or congestive heart failure. Symptoms may be subtle, especially in
elderly patients.
2. Class II indications include those for which permanent pacing may
be necessary, provided that the potential benefit to the patient can be
documented.
3. Class III indications are those for which permanent pacing is unlikely
to be of benefit; therefore, pacing is generally inappropriate.
Sinus Node Dysfunction and Dysregulation
Class I
1. Symptomatic sinus bradycardia (heart rate <40 beats per minute),
sinus pauses (>3 seconds while awake), or symptomatic chronotro-
pic incompetence. Pacing in patients with a heart rate of more than
40 beats per minute may be considered; however, careful documen-
tation of symptoms correlated with bradycardia is required.
2. Symptomatic sinus bradycardia due to drug treatment for which
there is no acceptable alternative (eg, amiodarone or β-blockers).
Abbreviations and acronyms are expanded at the end of this chapter.
324
3. Sinus node dysfunction with life-threatening, bradycardia-
dependent arrhythmias (eg, pause-dependent ventricular tachycar-
dia). Bradycardia itself need not be symptomatic.
4. Recurrent syncope with minimal carotid sinus pressure that causes
ventricular asystole (>3 seconds) without medication suppressing
the sinus node or AV node.
Class II
1. Sinus node dysfunction with a heart rate less than 40 beats per
minute when an association between clinically significant symptoms
(eg, syncope) and bradycardia has not been documented.
2. Patients with syncope of unexplained origin when evidence of sinus
node dysfunction is discovered in electrophysiology studies.
3. Recurrent syncope with a hypersensitive cardioinhibitory response
in the absence of provocation.
4. Neurocardiogenic syncope that is recurrent and symptomatic and is
accompanied by documented bradycardia (spontaneous or occurring
at tilt-table testing).
Class III
1. Asymptomatic sinus bradycardia, sinus arrest, or sinoatrial block.
2. Symptomatic bradycardia caused by nonessential drug therapy.
AV Conduction Abnormality
Class I
1. Acquired complete (third degree) and advanced second degree AV
block at any anatomical level. Exercise testing may provide evi-
dence of exercise intolerance. In asymptomatic patients, documented
ventricular asystole of more than 3.0 seconds or a heart rate of less
than 40 beats per minute during waking hours is also included as a
class I indication.
 28 Pacemakers
2. Third degree and advanced second degree AV block after cathe-
ter ablation of AV junction, or postoperative AV block that is not
expected to resolve after cardiac surgery.
3. Asymptomatic, persistent third degree AV block at more than 40
beats per minute with wide QRS escape complexes, or left ventricu-
lar dysfunction.
4. Symptomatic bradycardia due to drug therapy that is required to
control arrhythmia and other medical conditions.
5. Second degree AV block at any level of the conduction system if the
block is associated with symptomatic bradycardia.
6. Chronic bifascicular or trifascicular block with intermittent com-
plete heart block: intermittent type II second degree AV block or
alternating bundle branch blocks.
7. Neuromuscular disease (eg, myotonic muscular dystrophy, Kearns-
Sayre syndrome, limb-girdle muscular dystrophy, peroneal muscu-
lar atrophy) with AV block; patient may be asymptomatic but have
AV conduction disease that progresses unpredictably.
Class II
1. Asymptomatic third degree AV block with a narrow QRS complex
and an average heart rate during waking hours of more than 40 beats
per minute, and asymptomatic type II second degree AV block with
a narrow QRS complex.
2. First or second degree AV block with effective loss of AV synchrony
and symptoms similar to those in pacemaker syndrome.
3. Neuromuscular disease with any degree of AV block (including first
degree AV block), with or without symptoms.
4. Incidental electrophysiologic study finding of a markedly prolonged
His-ventricular interval (≥100 ms) in asymptomatic patients or a
pacing-induced infra-His block that is not physiologic.
Class III
1. Asymptomatic first degree or Mobitz type I second degree AV block
(Wenckebach block).
2. Fascicular block with first degree AV block with symptoms.
Acute Myocardial Infarction
Although the need for temporary pacing in the peri-infarct period
has decreased as a result of acute coronary intervention and other
aggressive approaches, various conduction disturbances may
still occur, including bradycardia and AV block. Disturbances
are usually related to the site of infarction and may be transient
or permanent. Temporary pacing during the peri-infarct period
is not necessarily an indication for permanent pacing. The guide-
lines for pacing after myocardial infarction are the following:
Class I
1. In the His-Purkinje system: a second degree AV block that is persis-
tent and associated with a bilateral bundle branch block. Within or
below the His-Purkinje system: a third degree AV block.
2. Second or third degree infranodal AV block that is transient and
associated with a bundle branch block. An electrophysiologic study
may be required to identify the site of the block.
3. Second or third degree AV block that is persistent and symptomatic.
Class II
1. Second or third degree infranodal AV block that is persistent.
Inferior Myocardial Infarction
Conduction disturbances that commonly occur in patients with
inferior myocardial infarction include sinus bradycardia, sinus
325
arrhythmia, sinus arrest, atrial fibrillation or flutter, and AV
block. First degree AV block and Mobitz type I second degree
block (Wenckebach block) occur more commonly; a minority of
patients have Mobitz type II second degree block or complete
(third degree) AV block. Patients who are hemodynamically
unstable may require temporary pacing. Conduction defects are
usually transient and rarely require permanent pacing.
Anterior Myocardial Infarction
Acute anterior myocardial infarction is more likely to be accom-
panied by persistent conduction disturbances. Patients with ante-
rior myocardial infarction and AV block have an unfavorable
prognosis and an increased incidence of sudden cardiac death,
which are related to the large area of myocardium involved in
this type of infarction. Temporary pacing—and usually perma-
nent pacing—is required in patients who have intermittent or
persistent complete heart block, new-onset bifascicular block, or
bilateral bundle branch block.
Nonbradycardiac Indications for Pacing
Hypertrophic Cardiomyopathy
Pacing for medically refractory hypertrophic cardiomyopathy is
a class II indication. Although enthusiasm was initially signifi-
cant for pacing in hypertrophic cardiomyopathy, pacing is now
used very selectively. When device therapy is used in patients
with hypertrophic cardiomyopathy, it is most commonly ICD
therapy for the prevention of sudden cardiac death.
Cardiac Resynchronization Therapy
The goal of CRT is to reestablish synchronous contraction
between the left ventricular free wall and the ventricular sep-
tum to improve left ventricular efficiency and functional class
(seeChapter29,“CardiacResynchronizationTherapy”).CRT
is now an accepted component of therapy for congestive heart
failure. The term CRT has generally been used to describe biven-
tricular pacing, but cardiac resynchronization can be achieved in
some patients by left ventricular pacing only.
Class I indications for CRT are the following:
1. New York Heart Association functional class III or IV.
2. QRS complex more than 120 ms.
3. Left ventricular ejection fraction 35% or less.
4. Optimized medical therapy.
5. Normal sinus rhythm.
Class II indications for CRT are the following:
1. Patients who meet the class I criteria and have atrial fibrillation.
2. Patients who must frequently depend on ventricular pacing.
Pacing Modes and Nomenclature
The North American Society of Pacing and Electrophysiology
and British Pacing and Electrophysiology Group Generic Code
is used to describe pacemaker function. The code has 5 positions
(Table 28.1).
1. Position I indicates the chamber paced. Five letters are commonly
used: A for atrium, V for ventricle, D if both chambers are paced,
and O if no pacing is to occur. Some manufacturers use S to indicate
pacing capability in a single-chamber device.
2. Position II indicates the chamber sensed. Five letters are commonly
used: A for atrium, V for ventricle, D if both chambers are sensed,
326 III Electrophysiology

Table 28.1. The Revised NASPE/BPEG Generic Code for Antibradycardiac Pacing
Position I: Chamber(s) Position II: Chamber(s) Position III: Response Position IV: Rate Position V: Multisite
Paced Sensed to Sensing Modulation Pacing
O = None O = None O = None O = None O = None
A = Atrium A = Atrium T = Triggered R = Rate modulation A = Atrium
V = Ventricle V = Ventricle I = Inhibited V = Ventricle
D = Dual (A + V) D = Dual (A + V) D = Dual (T + I) D = Dual (A + V)
Sa = Single (A or V) Sa = Single (A or V)
Abbreviation: NASPE/BPEG, North American Society of Pacing and Electrophysiology and British Pacing and Electrophysiology Group.
a
Manufacturers’ designation only.
(Previously published. See “Credit Lines” section.)

and O if no sensing is present in any chamber and asynchronous
pacing is to occur. Some manufacturers use S to indicate sensing
capability in a single-chamber device.
3. Position III indicates the response to a sensed signal. I indicates that
output is inhibited by a sensed event, T indicates that a stimulus is
triggered by a sensed event, and D indicates that a stimulus may be
triggered or inhibited by a sensed event. For example, in dual-cham-
ber devices, the atrial output may be inhibited by a sensed atrial
event, and the ventricular stimulus triggered by a sensed atrial event
(in the absence of a sensed ventricular event). The letter O indicates
that there is no mode of response, mandating that there likewise be
an O in the second (sensing) position.
4. Position IV reflects both programmability and rate modulation. R
indicates that the rate is modulated independently of intrinsic car-
diac activity (eg, activity or respiration) through a sensor in the
pacemaker. From a practical standpoint, R is the only indicator com-
monly used in the fourth position.
5. Position V, although rarely used clinically, indicates whether multi-
site pacing is present in none of the cardiac chambers (O), 1 or both
atria (A), 1 or both ventricles (V), or any combination of atria and
ventricles (D). To describe a patient with a DDDR (dual-chamber
rate-adaptive) pacemaker with biventricular stimulation, the code
would be DDDRV.
When choosing the appropriate pacing mode for an individual
patient, one must consider the underlying rhythm abnormality,
chronotropic status (ie, whether the patient can mount an appro-
priate rate response for a given physiologic activity), and activity
level. Some pacemakers are designed to change the pacing mode
automatically to promote intrinsic AV conduction and preserve
battery energy. For example, in a mode such as AAIR/DDDR,
the device is continuously monitoring AV conduction and can
automatically switch from AAIR mode to DDDR mode when
intrinsic conduction is lost or from DDDR mode to AAIR mode
when AV conduction is present (Figure 28.1). Table 28.2 summa-
rizes available pacing modes and appropriate indications.
Permanent Pacing Leads
A few basic facts about permanent pacing leads warrant discus-
sion. Pacing leads are either unipolar or bipolar. In a unipolar
lead the distal electrode is the negative pole and the pulse gener-
ator “can” serves as the positive pole. In a bipolar lead the distal
electrode is the negative pole and a more proximal electrode on
the pacing lead is the positive pole. Bipolar leads are less suscep-
tible to electromagnetic and electromechanical interference than
unipolar leads. (Some pulse generators are polarity programma-
ble. If a problem occurs when a bipolar lead is in service, repro-
gramming the pacemaker to unipolar pacing configuration may
restore normal function. Likewise, if a bipolar pulse generator is
in use and EMI is problematic when programmed to a unipolar
sensing configuration, reprogramming to a bipolar sensing con-
figuration may alleviate the problem.)
The outer insulation of pacing leads is made of silicone rub-
ber or polyurethane. There are advantages and disadvantages of
each type of insulating material.
All pacing leads have some mechanism of fixation, which is
classified as either active or passive. Passive fi xation leads usu-
ally have small tines that extend from the lead tip. The tines are

1 3
2

ECG
A A A A A A A A
P P P P P P P P

Marker channel
V V V V V V
B B P P P P

V-V intervals

Figure 28.1. Switching Pacing Modes. The first 2 beats are in AAIR pacing mode. After recognizing a single beat of atrioventricular (AV) block
(the third beat), a backup ventricular pacing beat occurs at 80 ms after the scheduled atrial pace (the fourth beat) (2). The fifth beat does not conduct
to the ventricle again, and the pacing mode is then switched to DDDR mode in programmed AV delay as shown in the sixth, seventh, and eighth beats
(3). The criterion to switch is the loss of AV conduction for 2 of the past 4 pacing cycles (1). AP indicates atrial pacing impulse; ECG, electrocardio-
gram; VB, native ventricular beat; VP, ventricular pacing impulse.
 28 Pacemakers 327

Table 28.2. Indications for Various Pacing Modes

Mode Generally Agreed Upon Indications
VVI Atrial fibrillation with symptomatic
bradycardia in the CC patient
VVIR Fixed atrial arrhythmias (atrial
fibrillation or flutter) with
symptomatic bradycardia in the CI
patient
AAI Symptomatic bradycardia as a result
of sinus node dysfunction in the
otherwise CC patient and when AV
conduction can be proved normal
AAIR Symptomatic bradycardia as a result
of sinus node dysfunction in the CI
patient and when AV conduction can
be proved normal
VVDa Congenital AV block
AV block when sinus node function can
be proved normal
DDI Need for dual-chamber pacing in the
presence of significant PSVT in the
CC patient
DDIRb AV block and sinus node dysfunction
in the CI patient in the presence of
significant PSVT
DDD AV block and sinus node dysfunction in
the CC patient
Need for AV synchrony (ie, to maximize
cardiac output) in active CC patients
Previous pacemaker syndrome
DDDR AV block and sinus node dysfunction in
the CI patient
AAIR/DDDR Intermittent AV block and sinus node
dysfunction in the CI patient
Controversial Indications
Symptomatic bradycardia in the patient
with associated terminal illness or other
medical conditions from which recovery
is not anticipated and pacing is life
sustaining only
As for VVI
Sinus node dysfunction in the absence
of AV block and in the presence of
significant PSVT in the CC patient
Sinus node dysfunction without AV block
in the CI patient in the presence of
significant PSVT
For any rhythm disturbance when atrial
sensing and capture are possible for
the potential purposes of minimizing
future atrial fibrillation and improving
morbidity and survival
As for DDD
As for DDD
Contraindications
Patient with known pacemaker syndrome
or hemodynamic deterioration with
ventricular pacing at implantation
CI patient who will benefit from rate response
Patient with hemodynamic need for dual-
chamber pacing
As for VVI
Sinus node dysfunction with associated AV
block demonstrated spontaneously or
during testing before implantation
When adequate atrial sensing cannot be
attained
As for AAI
Sinus node dysfunction
AV block when accompanied by sinus node
dysfunction
When adequate atrial sensing cannot be
attained
AV block when accompanied by paroxysmal
supraventricular tachycardias
CI patient with a demonstrated need or
improvement with rate responsiveness
Presence of chronic atrial fibrillation, atrial
flutter, giant inexcitable atrium, or other
frequent PSVT
When adequate atrial sensing cannot be
attained
As for DDD
As for DDD

Abbreviations: AV, atrioventricular; CC, chronotropically competent (ie, able to achieve an appropriate heart rate for a given physiologic activity); CI,
chronotropically incompetent (ie, unable to achieve an appropriate heart rate for a given physiologic activity); PSVT, paroxysmal supraventricular
tachycardia.
a
VDD as a stand-alone pacing mode (ie, a pacemaker capable of VDD as the only dual-chamber mode of operation) is currently used primarily as a single-
lead VDD system. If a dual-lead system is implanted, the capability of DDD pacing is desirable.
b
DDIR is being supplanted by DDD or DDDR pacemakers with mode-switching capability. That is, the pacemaker automatically reprograms to a mode
incapable of tracking the atrial rhythm in the presence of an atrial rhythm that the pacemaker classifies as pathologic. When the pacemaker recognizes
the atrial rhythm as physiologic, the pacemaker reprograms to the previously programmed mode.

designed to become entrapped in the endocardial trabeculae and
stabilize the lead until scar tissue forms around the lead. Active
fi xation leads usually have a screw that is screwed into the endo-
cardium. The screw may be permanently extended or it may be
extendable and retractable. Active fixation leads are the most
commonly used variety.
Most leads are designed to maintain a low stimulation or cap-
ture threshold. Steroid-eluting leads characteristically have lower
acute and chronic thresholds than non–steroid-eluting leads. In
addition to steroid-eluting leads, there are other low-threshold
design leads (eg, carbon-tipped and platinized electrodes). Low
thresholds maximize the battery life of the device.
Conventionally, the right atrial lead is placed in the appendage
that has trabecular endomyocardium and affords good lead stability.
The right ventricular lead is placed in the apex or septum (Figure
28.2). Right ventricular apical pacing may result in left ventricu-
lar systolic dysfunction in some pacemaker recipients, especially
in patients with left ventricular dysfunction. In these patients, the
right ventricular septum may be a preferred pacing site.
Pacemaker Syndrome
Pacemaker syndrome is a hemodynamic abnormality that can
result when use of ventricular pacing is inappropriate or when
328 III Electrophysiology

Figure 28.2. Normal Appearance of Dual-Chamber Pacing System. A, Posteroanterior chest radiograph. Pacemaker generator is implanted in the
upper left infraclavicular region. One atrial lead is present in the right atrial appendage, and 1 ventricular lead is present in the right ventricular apex.
Gentle redundancy is present on both leads. Ventricular lead is clearly visualized as bipolar. B, Lateral chest radiograph. Ventricular lead is clearly
anterior; therefore, it is in the right ventricular apex and not in the coronary sinus (in which case it would be pointing toward the spine). Atrial lead
is clearly visualized as bipolar.

ventricular pacing is uncoupled from the atrial contraction.
This syndrome is most common when the VVI mode is used in
patients with sinus rhythm, but it can occur in any pacing mode
if AV synchrony is lost (Figure 28.3).
Although the most common clinical presentation is general
malaise, patients may have a sensation of fullness in the head
and neck, syncope or presyncope, hypotension, cough, dyspnea,
congestive heart failure, or weakness. Physical findings include
cannon a waves in the neck veins and a lower blood pressure
when pacing than when in normal sinus rhythm. If pacemaker
syndrome occurs in the patient with a VVI or VVIR pacemaker,
the only definitive treatment is conversion to a dual-chamber
system. If episodes of symptomatic bradycardia are rare, the
symptoms of pacemaker syndrome may be alleviated by pro-
gramming the pacemaker to a lower rate limit and program-
ming hysteresis to the “on” mode. This setting would minimize
pacing and allow the patient to stay in normal sinus rhythm for
longer periods. If pacemaker syndrome occurs in a patient with
an atrial or dual-chamber pacing system, the cause of AV uncou-
pling must be identified and corrected (eg, in a patient with an

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

Figure 28.3. Pacemaker Syndrome with VVI Pacing. Retrograde ventriculoatrial conduction, with a retrograde P wave (arrow) following each
QRS complex, results in nearly simultaneous atrial and ventricular contractions.
 28 Pacemakers 329

AAIR pacing system and a very long AR interval and effective exceptions. Steroid-eluting leads characteristically have lower
AV uncoupling, ventricular pacing may be required to allow a acute and chronic thresholds than non–steroid-eluting leads.
shorter AV interval). Lead conductor fracture (Figure 28.6) may manifest as high lead
If single-chamber ventricular pacing is considered, a trial impedance, failure to capture, or oversensing with inappropri-
of ventricular pacing should be performed at implantation and ate output inhibition or muscle stimulation (Table 28.3). Lead
the blood pressure compared with that during sinus rhythm. If
blood pressure decreases with ventricular pacing or if the patient
has symptoms, dual-chamber pacing should be used; however, Box 28.1. Causes of Pacemaker Malfunction
pacemaker syndrome may develop even without symptoms or a
Failure to capture
decrease in blood pressure.
High thresholds with an inadequately programmed
output
Troubleshooting
Partial conductor coil fracture
Pacemaker questions on cardiology examinations typically relate
to troubleshooting. Most pacemaker problems are the result of Insulation defect
inappropriate programming, inappropriate mode selection, or Lead dislodgment or perforation
lead malfunction. Initial troubleshooting should always include
Impending total battery depletion
interrogation of the device and careful evaluation of electrocardio-
graphic tracings. Pacing and sensing thresholds should be eval- Functional noncapture
uated, and lead impedance should be noted. A chest radiograph Poor or incompatible connection at connector
should be obtained if lead impedances or electrocardiographic block
tracings suggest a potential lead problem.
Failure to sense or capture in the immediate postimplantation Circuit failure
period is most likely caused by microdislodgment or macrodis- Air in pocket (unipolar pacemaker)
lodgment of the lead or by a poor connection between the lead
Failure to output
and the set screw within the connector block of the pulse gener-
ator (Figure 28.4). Circuit failure
Failure to output is not synonymous with failure to capture Complete or intermittent conductor coil fracture
and is usually the result of oversensing (Box 28.1). Failure to
capture is characterized by a pacemaker artifact without subse- Intermittent or permanently loose set screw
quent depolarization of the chamber paced, and failure to output Incompatible lead or header
is characterized by a lack of pacemaker artifact at the appropri-
Total battery depletion
ate point in the timing cycle (Figure 28.5).
Internal insulation failure (bipolar lead)
Lead Abnormalities Far-field sensing or T-wave oversensing

Every lead has a characteristic range of normal impedance. Lead Oversensing any noncardiac activity
impedance is most commonly 400 to 1,500 ohms, but there are Lack of anodal connector contacta
Crosstalk
Undersensing
Morphology of intrinsic event different from that
measured at implantation
Lead dislodgment or poor lead positioning
Lead insulation failure
Circuit failure
Magnet application
Malfunction of reed switch
Electromagnetic interference
Battery depletion
Poor or incompatible connection at connector
block
Circuit failure
Air in pocket (unipolar pacemaker)

Figure 28.4. Atrial Lead Dislodgment. Posteroanterior chest radio- a

Examples include unipolar lead in a bipolar generator, bipolar lead in
graph shows dislodged atrial lead (arrow) the day after implantation a pacemaker programmed as unipolar, air in the pocket of a unipolar
of a permanent dual-chamber pacemaker. Position of the lead does not device, and a unipolar pacemaker not in the pocket.
permit atrial sensing or pacing.
330 III Electrophysiology

Figure 28.5. A, Dual-chamber pacemaker with atrial sensed and ventricular paced rhythm in the first 5 beats. Ventricular failure to output occurs
in the 6th to 8th beats (arrows) without ventricular pacing spikes seen. The underlying rhythm is complete atrioventricular block. B, DDD pacing in
the first 3 beats. The fourth beat (arrow) has a ventricular pacing spike that fails to result in ventricular depolarization; the narrow complex ventric-
ular depolarization is not a paced beat.

insulation failure usually manifests as low lead impedance,
undersensing or oversensing, failure to capture, muscle stimula-
tion, or early battery depletion.
Crosstalk
Crosstalk develops when an electrical event in 1 chamber is
sensed in the other chamber, with inappropriate inhibition of the
pacing stimulus in the second chamber (Figure 28.7). An exam-
ple is an atrial stimulus that is sensed by the ventricular lead as a
ventricular event, with consequent inhibition of ventricular output
and ventricular asystole. Crosstalk most commonly occurs when
the stimulus field is large or the stimulus output is high. Although
these conditions are more likely to develop when unipolar sens-
ing or pacing is used, crosstalk can occur in bipolar systems.
Crosstalk may be eliminated by altering the ventricular blank-
ing period. The blanking period is a portion of the timing cycle
that begins with the atrial pacing stimulus. During the blanking
period the sensing circuit is “turned off” so that no electrical
activity is sensed on the ventricular sensing circuit. The ventricu-
lar blanking period is almost always programmable. The adverse
outcome of crosstalk (ie, ventricular asystole) may be prevented
by using safety pacing; however, it is important to understand
that safety pacing does not correct crosstalk. With safety pacing,
any event that is sensed immediately after the blanking period in
the crosstalk sensing window is followed by a committed early
ventricular stimulus, producing a shortened AV delay, usually
from 100 to 110 ms (Figure 28.8). This portion of the timing
cycle is programmable in some pacemakers.
There are other ways to treat crosstalk in addition to lengthen-
ing the ventricular blanking period. One or more of the following
approaches can be used:
1. Decreasing the atrial output or programming the ventricular sen-
sitivity to a less sensitive value (predicated on the continued

Table 28.3. Intraoperative Evaluation of Pacing System

Voltage Current Lead
Defect Threshold Threshold Impedance
Wire fracture High High, normal, or low High
Figure 28.6. Lead Conductor Fracture. Posteroanterior chest radio- Insulation break Low High Low
graph shows the fracture site of a pacing lead (arrow). Inset, The frame Lead dislodgment High High Normal
outlines the portion of the radiograph that is enlarged to show the Exit block High High Normal
fracture.
 28 Pacemakers 331

Figure 28.7. Pacemaker Crosstalk. In the eighth complex (arrow) of this electrocardiographic strip, the ventricular output is inhibited as a result
of sensing of the atrial pacemaker artifact in the ventricular channel.

ability to achieve atrial capture or appropriate ventricular sensing,
respectively).
2. Switching from unipolar pacing or sensing to bipolar pacing or sens-
ing may eliminate crosstalk if the pacing system allows this pro-
grammable change in lead polarity configuration.
Pacemaker-Mediated Tachycardia
PMT can occur only with dual-chamber pacing systems with
intact atrial sensing (ie, DDD, DDDR, and VDD systems)
(Figure 28.9). Types of PMT include rapid tracking of atrial
fibrillation or flutter or rapid ventricular triggering from elec-
tromagnetic interference. Endless-loop tachycardia refers to a
specific type of PMT; intact ventriculoatrial conduction results
in retrograde P waves, which trigger another ventricular stim-
ulation, creating a loop. The situation can be corrected by
lengthening the PVARP beyond the retrograde P wave, so that
the retrograde P wave is not sensed and therefore does not ini-
tiate an AV timing cycle. Most devices have a programmable
option of PVARP extension after a premature ventricular con-
traction or some other algorithm to recognize and terminate
PMT. Premature ventricular contractions are the most common
initiators of endless-loop tachycardias.
Electromagnetic Interference
EMI can affect pacemaker performance. Devices are designed
to filter out much EMI; however, some sources of EMI cannot
be avoided.
EMI in the Hospital Environment
The hospital environment is a frequent source of EMI.
Electrocautery, for example, inhibits pacemakers; therefore, the
pacemaker should be programmed to an asynchronous mode
before starting the surgical procedure, especially in pacemaker-
dependent patients. The pacemaker internal circuitry can be dam-
aged if electrocautery, cardioversion, or defibrillation equipment
is used near the pacemaker; this equipment should be kept at least
15 cm away from the pacemaker during surgical procedures.
MRI uses magnetic and radiofrequency fields that cause all
pacemakers to pace asynchronously by closing the reed switch.
Additionally, the radiofrequency field theoretically has the
potential to induce rapid, asynchronous pacing. Although it is
recommended that patients with pacemakers avoid MRI, patients
who are not pacemaker dependent have undergone MRI with the
pacemaker output lowered below the capture threshold or pro-
grammed off. The patient must be carefully monitored through-
out the procedure.
Radiofrequency ablation can result in pacemaker inhibition
or reprogramming. The pacemaker programmer should be avail-
able during the procedure and the pacemaker interrogated at the
end of the procedure to document programmed parameters.
Therapeutic radiation may damage the pacemaker compo-
nents, resulting in damage to circuits or complete pacemaker
failure. Radiation-induced failure may manifest as sudden “no
output” or “runaway” pacemaker. This potential is of special
concern in patients receiving therapeutic radiation for breast or
chest malignancies. Damage is not related to the cumulative dose
but rather may occur at any time during the course of therapy. If
irradiation is being used for carcinoma of the breast or lung and
the pacemaker is located on the same side as the malignancy,
the pacemaker should be moved before initiating irradiation. For
irradiation of any other portion of the body, the generator should
be shielded.
Electroshock therapy for depressive disorders will not result in
pacemaker malfunction but may cause significant electrocardio-
graphic artifact and could potentially reprogram the pacemaker.
Extracorporeal shock wave lithotripsy may interfere with
pacemaker function. The lithotripter is usually synchronized to
the ventricular output or the pacemaker ventricular stimulus, so
programming the pacemaker to fixed-rate ventricular pacing is
safe. If the lithotripter is synchronized to the atrial stimulus, loss
of ventricular output may result and should be avoided. Therefore,
before lithotripsy the pacemaker should be programmed to the
VOO, VVI, or DOO modes. The focal point of the lithotripter

Figure 28.8. Safety Pacing. The fifth and eighth complexes (arrows) of this electrocardiographic strip show atrial pacing followed by a ventricular
pacing artifact at a fixed atrioventricular (AV) interval of 110 ms. At this interval, there is complete depolarization of the ventricle by the pacemaker.
The remainder of the complexes on the strip show an AV delay of 200 ms and ventricular fusion complexes.
332 III Electrophysiology
Figure 28.9. Pacemaker-Mediated Tachycardia. Electrocardiographic strip shows pacemaker-mediated tachycardia initiated by a ventricular
triplet (arrow).
should be at least 15 cm from the pulse generator to avoid dam-
age to the device.
Before and after procedures known to produce EMI, all
pacemakers should be interrogated to ensure that inadvertent
reprogramming or damage to the pacemaker has not occurred.
Appropriate programmers should be available during the proce-
dures in case problems arise.
EMI in the Nonhospital Environment
Permanent damage to pacing systems by electrical equipment
outside the hospital environment is unlikely; however, tempo-
rary interference may occur with exposure to certain devices
and electromagnetic fields. Potential sources of exposure include
heavy motors and arc welding. Devices such as airport metal
detectors may cause single-beat inhibition but should not cause
clinically significant sequelae. Microwave ovens do not interfere
with pacemaker function. Patients who work in an environment
that may cause significant interference with pacemaker function
may need to change occupations or at least avoid specific devices
in the workplace. If there is a question of risk in the workplace,
an engineer from the manufacturer can be consulted, or work site
testing can be performed.
Commercially available cellular phones do not usually cause
clinically significant EMI. Pacemaker patients who use a cellu-
lar phone should avoid carrying an activated phone in a pocket
directly over the pacemaker.
Figure 28.10. Pneumothorax. Posteroanterior chest radiograph shows pneumothorax (arrows) the day after pacemaker implantation.
Antitheft devices and electronic article surveillance equip-
ment can cause pacemaker interference. However, as long as the
pacemaker patient does not linger near the antitheft device, there
is little chance of clinically significant interference.
Miscellaneous Considerations
Several physiologic conditions can affect pacemaker function
and are sometimes interpreted as intrinsic pacemaker malfunc-
tion. Electrolyte and severe metabolic disturbances may result
in failure to capture or sense. Severe hyperkalemia is the most
common electrolyte disorder to cause pacing-related problems
and is most commonly encountered in patients with severe renal
insufficiency. Toxic levels of several antiarrhythmics increase
pacing thresholds; however, no significant problems have been
demonstrated at therapeutic levels of these drugs. Class IC
antiarrhythmics (flecainide, encainide, and propafenone) have
consistently been shown to significantly increase pacing thresh-
olds. This increase is sometimes dramatic, and these agents
should be avoided or used with caution in pacemaker-dependent
patients. Amiodarone does not consistently increase pacing
thresholds but often increases defibrillation thresholds in patients
who have an ICD. If amiodarone administration leads to hypo-
thyroidism, pacing thresholds may be elevated as a result of the
hypothyroid state.
Systemic corticosteroids can lower pacing thresholds and have
been used clinically for this purpose. Pacing thresholds usually
 28 Pacemakers
increase to pretreatment levels when use of systemic corticoster-
oids is discontinued.
Risks of Pacemaker Implantation
Patients should be advised that pacemaker implantation is a sur-
gical procedure and is associated with procedural risks such
as bleeding, infection, lead dislodgment, and pneumothorax
(Figure 28.10). Implantation should be avoided in patients with
active infection. Coagulation status should be determined before
the procedure to avoid a supratherapeutic international normal-
ized ratio, although implantation can usually be accomplished
with therapeutic prothrombin times. Chest radiography should
be performed after implantation to check for pneumothorax and
to verify lead position. Patients should also be advised of the
additional risk of coronary sinus dissection and phrenic nerve
or diaphragmatic stimulation if a coronary sinus lead is being
placed.
Temporary Pacing
Temporary pacing can be used in patients with symptomatic
bradycardia, either transiently if the cause is reversible or as a
bridge to permanent pacing. Common transvenous approaches
include the internal jugular vein, the subclavian vein, the femoral
vein, and, less commonly, the external jugular vein.
The 2 main types of temporary ventricular pacing catheters
are rigid, firm catheters and more flexible, balloon-tipped cathe-
ters. Although a balloon-tipped catheter may be easier to advance
into the right ventricular apex, a firmer catheter usually provides
a more stable catheter position. Pulmonary artery catheters are
available with an extra port through which a temporary pacing
wire can be placed. These are advantageous in that they can
be placed quickly without fluoroscopic guidance; however, the
pacing wire is relatively unstable. Atrial J-shaped catheters are
available if atrial pacing is required (eg, atrial overdrive pacing
333
or dual-chamber temporary pacing for patients who do not hemo-
dynamically tolerate single-chamber ventricular pacing).
Potential complications of temporary pacing include com-
plications of permanent pacing: bleeding, infection, lead dis-
lodgment, and pneumothorax. There is a higher risk of cardiac
perforation with temporary pacing catheters owing to the stiff-
ness of the catheters.
External pacing systems are widely available and can be used
in patients who have intermittent conduction disturbances or who
need to maintain rhythm until a transvenous pacing catheter can
be placed. Disadvantages of external pacing include occasional
difficulty maintaining capture as well as potentially significant
discomfort from the pacing stimulus for some patients.
Suggested Reading
Bernstein AD, Daubert JC, Fletcher RD, Hayes DL, Luderitz B, Reynolds
DW, et al; North American Society of Pacing and Electrophysiology/
British Pacing and Electrophysiology Group. The revised NASPE/
BPEG generic code for antibradycardia, adaptive-rate, and multisite
pacing. Pacing Clin Electrophysiol. 2002 Feb;25(2):260–4.
Epstein AE, Dimarco JP, Ellenbogen KA, Estes NA 3rd, Freedman RA,
Gettes LS, et al; American College of Cardiology; American Heart
Association Task Force on Practice Guidelines; American Association
for Thoracic Surgery; Society of Thoracic Surgeons. ACC/AHA/
HRS 2008 Guidelines for device-based therapy of cardiac rhythm
abnormalities. Heart Rhythm. 2008 Jun;5(6):e1–62. Epub 2008
May 21. Erratum in: Heart Rhythm. 2009 Jan;6(1):e2.
Abbreviations
AV atrioventricular
CRT cardiac resynchronization therapy
EMI electromagnetic interference
ICD implantable cardioverter-defibrillator
MRI magnetic resonance imaging
PMT pacemaker-mediated tachycardia
PVARP postventricular atrial refractory period

Cardiac Resynchronization Therapy
GRACE LIN, MD and DAVID L. HAYES, MD
Cardiac resynchronization therapy is the term applied to
reestablishing synchrony between left ventricular free wall
and ventricular septal contraction. Approximately one-third of
patients with dilated cardiomyopathy also have a conduction
delay due to left bundle branch block, resulting in left ventricular
dyssynchrony and worsening heart failure. In these patients, CRT
may promote left ventricular reverse remodeling and improve
symptoms and mortality from heart failure. CRT is achieved
through biventricular pacing, with placement of a left ventricular
lead in addition to a right ventricular lead.
CRT Clinical Trials
To date, more than 6,000 patients have been included in com-
pleted randomized clinical trials of CRT. Patients included in the
majority of trials have had NYHA class III or IV heart failure,
wide QRS complexes, and left ventricular ejection fraction of
35% or less. Primary end points of the CRT trials have reflected
changes in functional status; specifically, they have incorporated
the 6-minute walk test,. NYHA functional class, quality of life
assessments, and peak Vo2. Some trials have also included com-
posite clinical end points of cardiac and all-cause mortality and
hospitalizations for congestive heart failure.
Relatively consistent findings across the CRT trials are
improvement in the 6-minute walk test, NYHA functional class,
.
peak Vo2, and quality of life as assessed with the Minnesota
Living With Heart Failure Questionnaire and decreased risk of
death and heart failure hospitalizations (Table 29.1). Secondary
end points have included echocardiographic variables such as
left ventricular end-diastolic dimension, left ventricular volumes,
left ventricular ejection fraction, and mitral regurgitation. Most
Abbreviations and acronyms are expanded at the end of this chapter.
334
29
studies have consistently demonstrated a decrease in left ventric-
ular size and an increase in ejection fraction, findings that sug-
gest left ventricular reverse remodeling. Mitral regurgitation has
been shown to decrease after CRT, perhaps also reflecting bene-
ficial left ventricular reverse remodeling.
Indications for CRT
Findings from the CRT clinical trials have led to the following
class I indications for CRT:
• NYHA class III or IV heart failure while the patient is receiving
optimal medical therapy.
• Left ventricular ejection fraction ≤35%.
• QRS duration ≥120 ms.
• Sinus rhythm.
Patients who meet the above criteria and have atrial fibril-
lation have a class IIa indication for CRT. The benefit of CRT
in patients with QRS duration less than 120 ms is unclear and
has not been demonstrated in clinical trials (PROSPECT and
RethinQ). Currently, patients with a QRS duration less than
120 ms who are pacemaker dependent or require frequent ven-
tricular pacing would meet the criteria for a class IIa indication
for CRT.
Two CRT trials (REVERSE and MADIT-CRT) (Table 29.1)
have investigated the benefit of CRT in patients with NYHA
class I or II symptoms of heart failure and demonstrated
decreased risk of heart failure hospitalization and increased
left ventricular reverse remodeling with CRT. However, patients
with symptoms of heart failure that are less than NYHA class III
currently have a class IIb indication for CRT only if they are
receiving a permanent pacemaker or implantable cardioverter-
defibrillator, and frequent ventricular pacing is expected.
Table 29.1. Randomized Clinical Trials of Cardiac Resynchronization Therapy
Date of No. of
Publication Trial Design Patients NYHA Class Primary End Points Secondary End Points Results

2001 MUSTIC-SR Crossover 58 III 6MWT NYHA class Improvement

. in 6MWT, NYHA class, QOL,
Active vs inactive CRT QOL
.
and pVo2
pVo2 Reduced hospitalizations
Worsening HF
Patient pacing preference
Mortality
2002 MUSTIC-AF Crossover 43 III 6MWT NYHA class Improvement
. in 6MWT, NYHA class, QOL,
RV pacing vs CRT QOL
.
and pVo2
pVo2 Reduced hospitalizations
Worsening HF
Patient pacing preference
Mortality
2002 PATH-CHF Crossover 41 III or IV 6MWT Hospitalizations Improvement in 6MWT, QOL, and NYHA
.
RV or LV pacing vs CRT pVo2 NYHA class class
QOL
.
2002 MIRACLE Parallel 453 III or IV 6MWT pVo2 Improvement
.
in 6MWT, NYHA class, QOL,
Active vs inactive CRT NYHA class Exercise time pVo2, exercise time, LVEF, and LVEDD
QOL LVEF Decreased risk of death or HF event
LVEDD
MR
QRS duration
Clinical composite response
Mortality
.
2003 MIRACLE-ICD Parallel 555 III or IV 6MWT pVo2 Improvement in QOL and NYHA class
CRT-D vs ICD NYHA class Exercise time No change in 6MWT, LVEF, or LV volumes
(inactive CRT) QOL LVEF
LVEDD
LV volumes
MR
QRS duration
Clinical composite response
.
2003 PATH-CHF II Crossover 86 II or III pVo2 QOL Improvement in exercise time, 6MWT, QOL,
. .
No pacing vs LV pacing Anaerobic Vo2 6MWT NYHA class and pVo2
2003 CONTAK-CD Crossover and parallel 490 II-IV Composite of mortality, 6MWT Improvement in exercise capacity, QOL, and
Active vs inactive CRT HF hospitalization, NYHA class NYHA class
VT, and VF QOL Trend toward decreased morbidity and
.
pVo2 mortality
. . .
2004 MIRACLE-ICD II Parallel 186 II pVo2 Ve/Vo2 Improvement in composite
. . clinical response,
Active vs inactive CRT NYHA class LV volumes, and Ve/Vco2
QOL
6MWT
LV volumes
LVEF
Composite clinical response
(continued)
Table 29.1. (Continued)
Date of No. of
Publication Trial Design Patients NYHA Class Primary End Points Secondary End Points Results
2004 COMPANION Parallel 1,520 III or IV Composite of all-cause All-cause mortality Both CRT and CRT-D reduced primary
CRT vs CRT-D vs medical mortality or hospitalization QOL composite end point
therapy 6MWT Improvement in QOL, 6MWT, and NYHA
NYHA class class
2005 CARE-HF Parallel 813 III or IV All-cause mortality All-cause mortality and Improved symptoms of HF and QOL
CRT-P vs medical unplanned HF Reduced risk of death
therapy hospitalizations
NYHA class
QOL
2009 REVERSE Parallel 262 I or II Clinical composite response LVESVi Reduced risk of hospitalization or death
Active vs inactive CRT 6MWT Decreased LVESVi
QOL
2009 MADIT-CRT Parallel 1,820 I or II All-cause mortality LVEF Reduced risk of death or HF event
CRT-D vs ICD or HF event LV volumes Improvement in LV volumes and LVEF
Abbreviations: CRT, cardiac resynchronization therapy; CRT-D, cardiac resynchronization therapy device with defibrillator; CRT-P, cardiac resynchronization therapy pacemaker without defibrillator; HF, heart failure;
ICD, implantable cardioverter-defibrillator; LV, left ventricular; LVEDD, left ventricular
. end-diastolic dimension; LVEF, left ventricular ejection fraction; LVESVi,
. . left ventricular end-systolic volume index;
. MR, mitral
regurgitation; 6MWT, 6-minute walk test; NYHA, New York Heart Association; pVo2, peak oxygen consumption; QOL, quality of life; RV, right ventricular; Ve/Vco2, ventilatory response to exercise; Vo2, oxygen
consumption; VF, ventricular fibrillation; VT, ventricular tachycardia.
Previously published. See “Credit Line” section.
 29 Cardiac Resynchronization Therapy 337

Implantation Technique
Implantation of a CRT device requires placing a lead in the cor-
onary venous system, positioned to stimulate the left ventricular
free wall. There is marked variation in coronary venous anat-
omy, and lead placement is sometimes technically challenging
(Figures 29.1 and 29.2). With experience, the placement success
rate in the coronary venous system is more than 95%. When an
adequate position cannot be obtained, CRT can be accomplished
by placing an epicardial lead on the desired location of the left
ventricle.
In addition to the risks normally discussed with any patient
undergoing implantation of a pacemaker or cardioverter-
defibrillator, the coronary sinus or coronary veins can be
damaged (eg, dissection or perforation) with implantation of
a CRT device. In experienced hands, such complications are
uncommon, but they should be discussed with the patient.
Leads positioned in the coronary venous system also have a
higher potential to cause diaphragmatic stimulation. This com-
plication must be assessed at implantation. Even if diaphrag-
matic stimulation is absent at testing, slight movement of the lead
may cause this complication, which could subsequently require
reprogramming or lead repositioning.
CRT Responders and Prediction
of Response to CRT
There are different definitions for CRT responder. Clinical mea-
sures of response can be defined as an improvement
. in NYHA
functional class, 6-minute walk test, peak Vo2, or score on the
Minnesota Living With Heart Failure Questionnaire compared
with baseline. Echocardiographic measurements of response
include a decrease in left ventricular volumes and dimensions

Anterior
interventricular vein

LA

R L
CS
Ostium PLV

Lateral veins Branch of middle

cardiac vein
Branch of anterior
Middle cardiac interventricular vein LV
vein

Figure 29.1. A, Left anterior oblique cross-sectional view of heart at level of coronary sinus, showing most common location of cardiac veins in
relation to left ventricle. L indicates left; R, right. B, Left lateral view of heart, showing cardiac veins and usual left ventricular epicardial location.
CS indicates coronary sinus; LA, left atrium; LV, left ventricle; PLV, posterolateral vein. C, Fluoroscopic view of coronary sinus system. Inflated
balloon-tipped catheter is occluding coronary sinus (arrow) while contrast material is injected (outlining the coronary venous system). (Previously
published. See “Credit Line” section.)
338 III Electrophysiology

A B

Figure 29.2. Posteroanterior (A) and lateral (B) chest radiographs for patient with biventricular dual-chamber implantable cardioverter-defibril-
lator in place.

or an increase in left ventricular ejection fraction. Both features Names of Clinical Trials
indicate left ventricular reverse remodeling.
In the CRT clinical trials, up to one-third of patients were CARE-HF Cardiac Resynchronization in Heart Failure
nonresponders. Despite strict application of current guidelines COMPANION Comparison of Medical Therapy, Pacing, and
for CRT, identification of patients most likely to have clinical Defibrillation in Heart Failure
CONTAK-CD Guidant CONTAK CD CRT-D System Trial
and echocardiographic improvement with CRT continues to be
MADIT-CRT Multicenter Automatic Defibrillator Implan-
problematic. One method proposed to improve patient selection tation Trial With Cardiac Resynchronization
is echocardiographic assessment of left ventricular dyssynchrony Therapy
before implantation of a CRT device. Although many echocar- MIRACLE Multicenter InSync Randomized Clinical
diographic variables are available to assess intraventricular dys- Evaluation
synchrony of the left ventricle, a superior technique has not been MIRACLE-ICD Multicenter InSync ICD Randomized Clinical
established with certainty. For example, in the PROSPECT trial, Evaluation
which investigated several tissue Doppler imaging, Doppler, MIRACLE-ICD II Multicenter InSync ICD Randomized Clinical
and M-mode techniques, there was wide variation in sensitivity Evaluation II
and specificity among the techniques for predicting the clinical MUSTIC-AF Multisite Stimulation in Cardiomyopathies—
Atrial Fibrillation
and echocardiographic response. Current clinical guidelines for
MUSTIC-SR Multisite Stimulation in Cardiomyopathies—
patient selection for CRT do not incorporate echocardiographic Sinus Rhythm
measurements of dyssynchrony. PATH-CHF Pacing Therapies in Congestive Heart Failure
PATH-CHF II Pacing Therapies in Congestive Heart Failure II
Abbreviations
PROSPECT Predictors of Response to CRT
CRT Cardiac resynchronization therapy RethinQ Resynchronization Therapy in Narrow QRS
NYHA
. New York Heart Association REVERSE Resynchronization Reverses Remodeling in
VO2 oxygen consumption Systolic Left Ventricular Dysfunction
 30
Implantable Cardioverter-Defibrillator
Troubleshooting
BARRY A. BOILSON, MD, and MARGARET A. LLOYD, MD
This chapter reviews common problems that present on ICD
interrogation which are likely to be encountered in clinical
practice. Since their adoption over 20 years ago, ICDs are now
commonplace, and the ability to understand basic ICD interro-
gation and troubleshooting is an essential requirement in most
general cardiology practices. The evolution of ICD technology
has resulted in increasing complexity of arrhythmia-recognition
algorithms used by modern ICD microprocessors, but the basic
principles have not changed in that an ICD can still be considered
a sophisticated pacemaker device which has a limited capacity
to analyze and treat ventricular tachyarrhythmia. As such, like a
pacemaker, it is dependent on the ability to appropriately sense
cardiac signals and deliver energy appropriately. The difference
is the level of complexity of analysis which takes place, and the
form of energy delivery which follows, either as pacing (brady-
cardia or tachycardia pacing) or defibrillation (shock).
The hardware configuration in its simplest form consists of
1. The can, or generator, which is usually implanted prepectorally, but
occasionally subpectorally or abdominally
2. The pace-sense lead or leads (usually right ventricular but addi-
tionally right atrial or left ventricular). These are usually bipolar in
configuration.
3. The defibrillation coils, which are usually incorporated into the right
ventricular pace-sense lead.
A device interrogation report will contain a number of trac-
ings. The first is the shock channel, which is the far-field sig-
nal “seen” between the defibrillator coils. This approximates
a surface ECG electrode and can help to give an overview of
the nature of the presenting arrhythmia. Next are the local
EGMs (near-field EGMs) from the pace-sense lead or leads
Abbreviations and acronyms are expanded at the end of this chapter.
(Figure 30.1). Finally, there is a marker channel, in which the
device gives its “interpretation” of the EGMs. Dual-chamber
devices will include an atrial near-field EGM as well. The order
of these tracings on the screen and paper strip varies according
to device manufacturer and model.
• Comparing the marker channel (“device interpretation”) with the
ventricular (and when present, atrial) EGMs is extremely helpful in
interpreting device behavior.
Broadly speaking, patients will present for ICD trouble-
shooting in four principal ways:
1. Following shock delivery, in which case the role of the physician is to
determine if the shock was appropriate or inappropriate and to plan
the next step in treatment.
2. With symptoms possibly related to arrhythmia (eg, light-headedness,
syncope, or palpitations)
3. Having received an “alert,” either as an audible alarm or as a
vibration.
4. Asymptomatic, but with abnormalities of function identified on
routine device interrogation.
Following shock delivery or a symptomatic episode sug-
gestive of ventricular arrhythmia, the device should be inter-
rogated either remotely by telemetry (newest models) or in
the device clinic. Programmed detection parameters should
be verified, such as rate detection windows for VT, fast VT,
and VF. Programmed therapies (sequence of antitachycardia
pacing and shock therapies) for each ventricular arrhythmia
should also be verified. The following questions should then
be asked:
1) Was there an arrhythmia?
2) If so, did the device “see” and correctly interpret the arrhythmia?
3) Did the patient have symptoms?
4) Were appropriate therapies delivered?
339
340 III Electrophysiology

Figure 30.1. Lead placement and source of near- and far-field electrograms.

5) Were therapies successful in terminating the arrhythmia? If multi-
ple, sequential therapies programmed, which were successful and
which were not?
6) Have any new medications that might affect pacing or defibrillation
thresholds been added?
7) Did the patient have any relevant electrolyte abnormalities?
The key question is whether the device has delivered therapy
appropriately or not. In most cases, ICD shocks are intended
for the termination of ventricular arrhythmias not likely to be
tolerated hemodynamically by the patient. Some patients have
“phantom shocks;” the sensation or belief that they have received
a shock when in fact there has been no detected arrhythmia or
therapy delivered. Occasionally when a shock has been deliv-
ered appropriately, there has been inappropriate detection of
SVT, or other electrical signals of cardiac and extracardiac
origin.
The algorithm in Figure 30.2 outlines a systematic approach
to interpretation of data from device interrogation after a shock.
Additionally, posteroanterior and lateral chest radiographs
should be obtained to be sure that lead position is stable, there is

ICD Discharge

No tachyarrhythmia
Tachyarrhythmia
(oversensing)

SVT VT/VF Intracardiac Extracardiac

(innapropriate (appropriate signals signals
detection) detection) P or T waves Fracture, MP, EMI

Figure 30.2. Approach to ICD interrogation after shock.

 30 Implantable Cardioverter-Defibrillator Troubleshooting
no gross evidence of lead fracture, and that the pins are secure
in the header, as all may create problems in arrhythmia detection
and treatment.
In the examples below, common findings are demonstrated
and described, and solutions summarized.
Appropriate Therapy Delivery
Example 1 (Figure 30.3)
Complete atrioventricular dissociation is evident between the
atrial and ventricular EGMs. The atrial cycle length is 670 to
690 ms, the ventricular 290 to 350 ms. The device appropri-
ately detects the arrhythmia as VT and delivers burst pacing,
terminating the arrhythmia.
Example 2 (Figure 30.4)
VT is appropriately detected, and antitachycardia pacing deliv-
ered. Ventricular capture is evident on the “shock” (far-field)
channel following each pacing spike. The therapy is successful
and tachycardia is terminated.
• Antitachycardia pacing sequences are typically utilized for rela-
tively hemodynamically stable arrhythmias, with follow-up shocks
if pacing is unsuccessful in restoring sinus rhythm.
Figure 30.3. Appropriate antitachycardia pacing delivery.
341
• For poorly tolerated VT and for VF, antitachycardia pacing is
usually bypassed and defibrillation delivered immediately after
arrhythmia detection and charging of the capacitors.
Inappropriate Therapy Delivery
and Discriminators
Most commonly, inappropriate therapy is delivered in the setting
of SVT or sinus rhythm which is not effectively discriminated by
the ICD. The primary criterion used by ICDs to identify ventric-
ular tachyarrhythmias is rate. Additional criteria can be activated
to increase the ability of the device to discriminate between ven-
tricular and non-ventricular arrhythmias. ICDs discriminate VT
from SVT based on a number of criteria:
1. Ratio of atrial and ventricular depolarizations. Where atrial
EGMs outnumber ventricular EGMs, particularly if they are present
in an exact ratio, eg 2:1 or 3:1, the arrhythmia will be labeled as an
SVT by the device.
2. Stability of the ventricular cycle length. Where this is constant,
the device is more likely to treat as VT.
3. Ventricular EGM morphology analysis. In some ICD models, a
profile of the far field (shock lead) ventricular EGM in the patient’s
usual presenting rhythm (sinus, atrial fibrillation/flutter with con-
trolled ventricular response or paced) is stored. When tachycardia
occurs, the morphology of the ventricular EGMs is performed and
compared to those previously stored.
342 III Electrophysiology

Figure 30.4. Appropriate antitachycardia pacing delivery.

Example 3 (Figure 30.5) Example 5 (Figure 30.7): Morphology discrimination

successful.
In this case, the lower tracing (far-field channel) demonstrates
SVT is precipitated by a PAC. The cycle length meets criteria
a narrow complex tachycardia with an irregular R-R interval in
for tachycardia detection, but the QRS morphology remains at
keeping with atrial fibrillation. The patient has a single-chamber
100% of the stored morphology saved previously during sinus
ICD, and therefore only the ventricular EGM channel is available
rhythm. Therefore, VT detection is terminated. Following reanal-
to the device to use for detection. A VT is detected based on the
ysis and redetection, the arrhythmia is rejected as VT again due
cycle length.
to the constant 100% match on morphology.
Solution: Placement of an atrial lead and/or using SVT
discriminators.
Example 6 (Figure 30.8)
• Increasing the specificity (ie, the number of discriminator algo-
rithms) may reduce sensitivity of the device. The patient is in sinus tachycardia. The cycle length meets cri-
teria for VT detection, but the QRS morphology remains at
Example 4 (Figure 30.6) 100% of the stored morphology saved previously during sinus
The patient is in atrial tachycardia with 2:1 AV block. However, rhythm. Therefore, VT detection is terminated and no therapies
due to the nonconducted atrial EGMs falling in the PVARP, are delivered.
they are not sensed and the device fails to count them on the
marker channel. This results in failure to detect an atrial rate sig- Example 7 (Figure 30.9)
nificantly greater than the ventricular rate and an inappropriate
shock (lower panel, arrow) is delivered. VT is initiated by a premature ventricular contraction. VT is
Solution: Shortening the PVARP duration. detected, and the QRS morphology is a 0% match with that saved

Figure 30.5. Failure of SVT discrimination—insufficient discriminators available.

 30 Implantable Cardioverter-Defibrillator Troubleshooting 343

Figure 30.6. Failure of SVT discrimination—atrial lead undersensing.

Figure 30.7. Morphology discrimination successful.

344 III Electrophysiology
Figure 30.8. Morphology discrimination successful.
during sinus rhythm. In this particular case, the VT terminates
spontaneously and no therapy is required.
Occasionally, an inappropriate shock is delivered due to ven-
tricular lead oversensing. This is usually due to:
1. Far-field (atrial) or near-field (T wave) EGM oversensing, or
2. Sensing of electrical signals of noncardiac origin, due to electromag-
netic interference, myopotentials, or lead fracture.
Figure 30.9. Morphology discrimination successful.
Example 8 (Figure 30.10)
The marker channel demonstrates ventricular double counting
due to oversensing of T waves, with the result that the ventricular
rate estimated by the ICD is double that of the true value.
Solution: In this case, the best option is ventricular lead revi-
sion, as the T wave is of greater amplitude than the R wave with
the lead in this location and therefore adjusting the sensitivity
 30 Implantable Cardioverter-Defibrillator Troubleshooting
Figure 30.10. Inappropriate shock—T wave oversensing.
to avoid sensing of the T wave will also result in undersensing
of the R wave.
Example 9 (Figure 30.11)
The patient is in an AV sequentially paced rhythm at a cycle
length of 978 to 985 ms. On the ventricular EGM channel, ini-
tially intermittent low amplitude, high frequency myopotentials
are noted which become sustained, are inappropriately inter-
preted as PVCs; ventricular pacing is inhibited, and a ventricu-
lar cycle length of 158 to 268 ms is inappropriately marked and
detected as a VF episode.
Solution: Reprogramming from unipolar to bipolar, or lead
revision. This is also associated with integrated bipolar ventric-
ular sensing leads, and changing to a true bipolar sensing lead
may also be helpful.
Example 10 (Figure 30.12)
The above demonstrates a classic example of electromagnetic inter-
ference, with high frequency electrical signals on both the atrial
Figure 30.11. Inappropriate shock—myopotential oversensing.
345
and ventricular channels, resulting in rapid counting on the marker
channel, and inappropriate detection of atrial fibrillation and VF.
Solution: Avoidance of source of EMI (eg, arc welding, auto-
mobile alternators).
Example 11 (Figure 30.13)
The ventricular EGM channel and marker channels are shown.
With provocative maneuvers (movement of the upper extremities,
manipulation of the generator in the pocket) noncardiac electri-
cal signals are produced, labeled ‘a,’ resulting in inappropriate
overcounting on the marker channel.
Solution: Ventricular lead revision.
Failure to Deliver Appropriate Therapy
for Ventricular Arrhythmias
When delivery of therapy for ventricular arrhythmia fails, patients
may present with symptoms related to ventricular arrhythmia—
ie, syncope, presyncope, effort intolerance, or palpitations. This
may occur due to failure of appropriate ventricular arrhythmia
detection, or failure of delivered therapy.
Example 12 (Figure 30.14)
In this case, delayed detection of VT results from undersensing
of alternate ventricular depolarizations (see circled, non sensed
impulse on EGM).
Solution: Adjustment of ventricular lead sensitivity or lead
revision if this is not successful.
Example 13 (Figure 30.15)
In this case, the device functioned appropriately in accordance
with how it was programmed. All ventricular EGMs were “seen”
but the rate fell below the rate set for detection intermittently,
resulting in delayed therapies.
Solution: Reprogramming of ventricular arrhythmia detec-
tion zones to allow delivery of therapy at lower rates.
Example 14 (Figure 30.16)
VT is appropriately detected and a 41-joule shock is delivered,
but without termination of the tachycardia.
Solution: Lead revision or placement of a subcutaneous array.
Example 15 (Figure 30.17)
VT is appropriately detected and a antitachycardia pacing is
delivered, but without termination of the tachycardia.
346 III Electrophysiology
Figure 30.12. Inappropriate shock—electromagnetic interference.
Solution: Programming of further burst ATP at shorter
cycle lengths, addition of “ramp” pacing, or diverting to shocks
earlier.
Summary
• Was there an arrhythmia?
• If so, did the device “see” and correctly interpret the arrhythmia?
• Did the patient have associated symptoms?
• Were appropriate therapies delivered?
• Were therapies successful in terminating the arrhythmia?
• If sequential therapies programmed, which were successful in ter-
minating the arrhythmia?
Figure 30.13. Inappropriate shock—lead fracture.
• Have any new medications that might affect pacing or defibrilla-
tion thresholds been added?
• Any electrolyte abnormalities that might have precipitated an
arrhythmia?
• Are there any device or lead abnormalities on the chest radiograph?
In general cardiology practice, it is a good idea to review ICD
tracings as often as possible to improve familiarity with the nor-
mal functioning of ICDs. This will greatly facilitate the ability to
troubleshoot when abnormalities are evident. In practical terms,
the most common problem which requires prompt troubleshoot-
ing is the delivery of recurrent or inappropriate shocks. Where
shocks are recurrent and appropriate, aggressive treatment of VT
is warranted. Where shocks are inappropriate, careful review
 30 Implantable Cardioverter-Defibrillator Troubleshooting 347

Figure 30.14. Failure of ventricular arrhythmia detection—ventricular undersensing.

Figure 30.15. Failure of ventricular arrhythmia detection—suboptimal device programming.

348 III Electrophysiology

Figure 30.16. Failure of therapy—insufficient shock energy delivery.

Figure 30.17. Failure of therapy—antitachycardia pacing.

will usually reveal the answer, and temporary and long-term EGM electrogram
solutions may be applied. ICD implantable cardioverter-defibrillator
PVARP postventricular atrial refractory period
SVT supraventricular tachycardia
Abbreviations
VF ventricular fibrillation
ECG electrocardiogram VT ventricular tachycardia

Implantable Cardioverter-Defibrillator Trials
and Prevention of Sudden Cardiac Death
MARGARET A. LLOYD, MD, and BERNARD J. GERSH, MB, CHB, DPHIL
Since the early 1980s, the development of ICDs has revolution-
ized the approach to prevention of SCD. Multiple clinical trials
have defined the indications for ICD therapy in the prevention of
primary and secondary SCD. The current understanding of risk
stratification is a victory for evidence-based medicine; perhaps
no other area of cardiology has been so rigorously evaluated.
As such, these fact-driven clinical trials are rich fodder for the
cardiology board examination.
As many as 450,000 people may experience SCD each year in
the United States. This number may be an overestimation because
it is understood that perhaps 25% to 50% of deaths attributed
to “cardiac arrest,” when no other obvious cause is identified,
are noncardiac. It has long been recognized that patients who
experienced SCD and survived were at risk for a second event.
The challenge has been to identify at-risk patients before the first
event, because some studies suggest that less than 5% of patients
survive neurologically intact.
Although clinical trials have provided much information
about risk stratification and treatment, conclusions must be
evaluated critically. In retrospect, some trials have had flaws in
design, some were terminated too early to clearly test the hypo-
thesis, some had unacceptable crossover rates, and some used
surrogate end points (syncope, death without autopsy, ventricular
arrhythmia) that may or may not have been equivalent to SCD.
Differences in trial design make direct comparison of results
difficult (Table 31.1). Nevertheless, the accumulated data have
been used to formulate guidelines that help identify with more
certainty which patients have a greater risk of SCD and which
patients should be considered for ICD therapy. The evidence
demonstrates that ICDs are superior to antiarrhythmic drug ther-
apy in most instances for the primary and secondary prevention
Abbreviations and acronyms are expanded at the end of this chapter.
31
of SCD in patients at risk; antiarrhythmic drug therapy is now
relegated mainly to adjuvant therapy such as reducing the num-
ber of shocks.
A summary of secondary prevention trials is presented
in Table 31.2; primary prevention trials are summarized in
Table 31.3. Recent important trials are summarized in the text
that follows.
IRIS (2009)
Similar to the DINAMIT trial, the IRIS trial evaluated the effect
of early ICD implantation in MI patients deemed to be at high
risk of SCD (EF ≤40%, post-MI heart rate ≥90 beats per minute,
and runs of NSVT). Almost 900 patients were enrolled within
30 days of MI; half received standard therapy, and half received
standard therapy plus ICD. At a mean follow-up of 37 months,
survival in the 2 groups was identical. The reduction of SCD
in the ICD group was offset by the increase in non-SCD. These
findings demonstrate that SCD risk stratification based on cur-
rent criteria and ICD implantation early after MI does not reduce
overall mortality.
SCD-HeFT (2005)
The SCD-HeFT was a prospective, randomized, double-blind
trial involving 2,521 patients at 148 sites from 1997 to 2001.
The average age of patients was 60.1 years, the average EF was
25%, and the numbers of patients with ischemic and nonisch-
emic cardiomyopathy were nearly equal. Patients were randomly
assigned to receive amiodarone, ICD, or placebo. Medical man-
agement was optimized; blinding was imperfect because reveal-
ing side effects from the amiodarone developed in some patients,
and no sham devices were implanted. After at least 30 months
349
350 III Electrophysiology
Table 31.1. Inclusion Criteria for Selected Implantable
Cardioverter-Defibrillator Trials
Criteria
Ejection
Trial Fraction, % Other
IRIS ≤40 Acute MI within 30 d
Heart rate ≥90 beats per minute
NSVT
DEFINITE <36 NSVT or PVCs
SCD-HeFT ≤35 NYHA class II or III CHF
MUSTT <40 NSVT or inducible sustained VA
MADIT ≤30 Prior MI
MADIT II ≤30 Prior MI
DINAMIT ≤35 ↓ HRV
Abbreviations: CHF, congestive heart failure; HRV, heart rate variability; MI,
myocardial infarction; NSVT, nonsustained ventricular tachycardia; NYHA,
New York Heart Association; PVC, premature ventricular contraction; VA,
ventricular arrhythmia.
of follow-up, ICD decreased mortality by 23% compared with
placebo (Figure 31.1). Amiodarone did not improve survival.
Surprisingly, in subgroup analysis, mortality improved exclu-
sively in the NYHA II group (46%) and not in the NYHA III
group (0%). Subgroup analysis also did not show a benefit for
women, blacks, or patients with diabetes mellitus. Nonetheless,
the data were extrapolated to include NYHA classes II and III
heart failure as primary inclusion criteria, and the trial results
were critical in formulating the most recent guidelines of the
Centers for Medicare and Medicaid Services for ICD therapy.
DINAMIT (2004)
DINAMIT was a randomized, open-label trial that compared
ICD therapy with optimal medical therapy in 674 high-risk
patients (defined by an EF <35% and evidence of impaired car-
diac autonomic function) who were enrolled 6 to 40 days after
MI. The primary end point was death from any cause; a sec-
ondary end point was death from arrhythmia. There was no dif-
ference in baseline characteristics between the groups. During a
mean follow-up of 30 months, there was no difference in overall
mortality between the 2 treatment groups. A reduction in deaths
due to arrhythmia was balanced by an increase in overall mortal-
ity (cardiac but nonarrhythmic) in the ICD group.
The reason for this surprising finding is unclear. Use of
amiodarone was more common among patients in the medi-
cal therapy group, but since prior trials have found no survival
advantage with amiodarone use, this is unlikely to be respon-
sible. The rate of revascularization was low, which also may have
been a factor. More likely, the finding is related to the inclusion
criteria of this trial, which allowed for inclusion of patients who
would not have been included or who would have been specifi-
cally excluded in other trials—namely, the patients were enrolled
early after MI and had evidence of impaired cardiac autonomic
function. The inclusion of patients who had autonomic dys-
function may have preselected patients with a poorer prognosis
regardless of treatment strategy. The degree of cardiac autonomic
dysfunction was not reevaluated during the study, and improve-
ment or deterioration might have influenced the results.
Remodeling and infarct maturation, which may take months
or years to become evident, are the critical determinants, and
not arrhythmias or left ventricular dysfunction in the peri-infarct
period. In another study, recurrent infarct was the cause of most
deaths that occurred early after MI. Accordingly, device dis-
charge may be related to MI-induced arrhythmia. The MADIT
database showed no ICD benefit until 18 months after infarction.
The benefits of ICD therapy for the prevention of SCD may not
become evident until years after MI and may not have been cap-
tured in the mean 30-month follow-up of the DINAMIT study.
Current guidelines of the Centers for Medicare and Medicaid
Services specifically exclude prophylactic use of an ICD immedi-
ately after acute MI, although in some individual circumstances
it may be considered (eg, in patients with recurrent, sustained
arrhythmias).
DEFINITE (2004)
The DEFINITE trial included 458 patients with nonischemic
dilated cardiomyopathy and EF less than 36%, NSVT or prema-
ture ventricular contractions, and NYHA class I, II, or III heart
failure who were randomly assigned to receive standard medical
therapy or ICD. After a mean follow-up of 29 months, ICD implan-
tation (compared with optimal medical therapy only) substantially
reduced the all-cause death rate (34% relative decrease and 5.7%
absolute decrease) and the rate of death from arrhythmia. The
greatest benefit occurred in the NYHA III subgroup of patients
(in contrast to the findings of SCD-HeFT). Initially, women did
not seem to benefit from ICD implantation in this trial, but further
analysis suggested that the excess mortality in this subgroup was
unrelated to the ICD or underlying cardiac disease; the women
seemed to have an excess of malignancy, stroke, and other non-
cardiac, nonarrhythmic conditions. Another limitation is the rela-
tively small size of this trial, making subgroup analysis difficult.
COMPANION (2004)
The COMPANION trial was the first to test the effect of biven-
tricular pacing (cardiac resynchronization) in combination with
ICD on death from any cause or hospitalization. The 1,520 study
patients had advanced heart failure (ischemic or nonischemic)
and a QRS interval of more than 120 milliseconds. They were
randomly assigned in a 1:2:2 ratio to receive optimal medical
therapy alone or in combination with either biventricular pacing
or an ICD. Approximately 90% of patients assigned to device
therapy had a left ventricular lead successfully placed, a remark-
able feat given lead technology available at the time.
A statistical difficulty in this trial was the crossover from medi-
cal therapy to device therapy. Patients with cardiomyopathies fre-
quently require device therapy (pacing or ICD) as their disease
progresses, and crossover into the device therapy arm has been an
issue with many trials. During the COMPANION trial, biventricu-
lar ICDs became commercially available, and patients in the medi-
cal therapy alone group thus were free to have ICD implantation
outside the trial. In the medical therapy group, 26% of the patients
withdrew, and most of these patients eventually had devices placed
outside the trial. In addition to the problem of crossover, the 2 parts
of the combined end point of death or hospitalization for conges-
tive heart failure are not equivalent clinical end points.
The risk of the combined end point of death from or hospi-
talization for congestive heart failure was reduced by 34% with
biventricular pacing compared with medical therapy alone and
by 40% in the biventricular pacing–ICD group after 1 year of
follow-up. Given the defined end point and the crossover, it is
difficult to extrapolate the relative contributions of biventricular
pacing and ICD therapy to the mortality reduction for compari-
son with medical therapy alone.
Table 31.2. Trials of ICD Therapy for Secondary Prevention of SCD
Trial Inclusion Criteria End Point Treatment Arms Key Results
CASCADE (1993) Recent history of resuscitation for out- Cardiac arrest from VF Amiodarone (n = 113) Amiodarone performed better than
of-hospital VF Cardiac mortality EPS- or Holter-guided conventional AA conventional AA drugs for all end points at
No occurrence of an acute Q-wave MI at Syncope followed by ICD shock drug therapy, mostly quinidine or the end of 6 y
time of VF procainamide (n = 115) Percentage of patients with survival free of
cardiac death, resuscitated VF, or syncope
followed by ICD shock: 53% (amiodarone
group) vs 40% (conventional AA drug
group)
ESVEM (1993) Documented episode of sustained All-cause mortality or SCD survival EPS-guided AA drugs (n = 242) Both treatment arms provided similar accuracy
(>15 s) VT Overall cardiac death Holter-guided AA drugs (n = 244) of drug efficacy prediction
Resuscitated SCD or syncope Arrhythmic death Drugs evaluated: 6 class I AA drugs and Only sotalol was associated with a significant
Inducible on EPS sotalol reduction in risk of arrhythmia recurrence
CIDS (1993) Documented VF or All-cause mortality ICD (n = 328) ICDs provided a 20% relative risk reduction
Out-of-hospital cardiac arrest requiring Arrhythmic death Amiodarone (n = 331) in all-cause mortality and a 33% reduction
defibrillator or Nonfatal VF or VT in arrhythmic mortality compared with
Sustained VT ≥150 bpm causing Cause-specific mortality amiodarone
presyncope or angina with LVEF ≤35%
or
Unmonitored syncope with spontaneous or
inducible VT
CASH (1993) SCD survivors All-cause mortality Propafenone (n = 58) (this arm was stopped ICD treatment arm had a 37% reduction in
because of excess mortality compared mortality at 2 y and a 23% reduction in
with ICD arm) mortality over long-term follow-up of
Amiodarone (n = 92) 9 y compared with both amiodarone and
Metoprolol (n = 97) metoprolol
ICD (n = 99) ICD arm had a 63% reduction in mortality
compared with propafenone at 11 mo
Wever et al (1995) Post-MI survivors of cardiac arrest caused All-cause mortality Early ICDs (n = 29) Early ICD therapy reduced overall mortality
by documented VT or VF Costs and cost-effectiveness Tiered therapy, starting with AA drugs by 66% at 2 y: ICD group, 14% mortality;
Baseline inducibility of VT (mainly class III) (n = 31) tiered-therapy group, 35% mortality
Early ICD therapy resulted in cost savings of
$11,315 per life-year saved compared with
tiered therapy
AVID (1997) VF or All-cause mortality ICD therapy (n = 507) ICDs reduced total mortality 39% after 1 y,
Sustained VT with syncope or Quality of life EPS- or Holter-guided sotalol or empirical 27% after 2 y, and 31% after 3 y compared
Sustained VT without syncope and LVEF Costs and cost-effectiveness amiodarone (n = 509) with AA drugs
≤40% and SBP <80 mm Hg, chest pain,
or near syncope
Abbreviations: AA, antiarrhythmic; bpm, beats per minute; EPS, electrophysiologic study; ICD, implantable cardioverter-defibrillator; LVEF, left ventricular ejection fraction; MI, myocardial infarction; SBP, systolic blood
pressure; SCD, sudden cardiac death; VF, ventricular fibrillation; VT, ventricular tachycardia.
Previously published. See “Credit Lines” section.
Table 31.3. Trials of ICD Therapy for Primary Prevention of SCD
Trial Inclusion Criteria End Point Treatment Arms Key Results
BHAT (1982) Acute MI All-cause mortality β-Blocker: Propranolol reduced SCD mortality by 28%
ECG and enzyme changes Secondary: Propranolol (n = 1,916) and overall mortality by 26%
CAD death Placebo (n = 1,921) ≥90% of observed deaths were from
SCD cardiovascular causes
CAD death with definite nonfatal MI Both sudden and non-sudden arteriosclerotic
heart disease mortality rates were less in
propranolol group
CAST (1991) 6 d to 2 y after MI All-cause mortality or SCD survival Class I AA drugs: encainide (n = 432) or At 14-mo follow-up, total mortality for
>6 PVCs per hour Overall cardiac death flecainide (n = 323) patients in the class I AA drug group was
No VT ≥15 beats or ≥120 bpm Arrhythmic death Placebo (n = 743) 3.6 times higher than that for patients in
LVEF: placebo arm
≤55% if recruited within 90 d post-MI Study was stopped early
or
≤40% if recruited ≥90 d after MI
GESICA (1994) NYHA class II, III, or IV All-cause mortality Amiodarone (n = 260) Amiodarone reduced overall mortality at 2 y
2 or 3 indexes of systolic myocardial SCD Standard treatment (n = 256) by 28%: amiodarone, 33.5% mortality;
dysfunction: Death due to progressive heart failure placebo, 41% mortality
1. Chest radiograph: cardiothoracic ratio Hospital admissions
>0.55
2. LVEF ≤35%
3. LVEDD >3.2 cm/m 2 body surface
CHF-STAT (1995) NYHA class II, III, or IV All-cause mortality Amiodarone (n = 336) Amiodarone did not reduce total mortality
≥10 PVCs per hour SCD Placebo (n = 338) despite improved left ventricular function
LVEF ≤40% Suppression of ventricular arrhythmias ICD (n = 95) and suppressed ventricular arrhythmias
Effect on LVEF ICDs reduced overall mortality by 54%
Hospitalizations
MADIT (1996) Q-wave MI ≥3 wk All-cause mortality Conventional therapy (n = 101) ICDs cost $22,800 per life-year saved
Asymptomatic NSVT Costs and cost-effectiveness vs conventional therapy, assuming
LVEF ≤35% transvenous devices were used
Inducible, nonsuppressible VT on EPS
with procainamide
NYHA class I, II, or III
SWORD (1996) LVEF ≤40% All-cause mortality (n = 1,549)
D -Sotalol Prophylactic oral D-sotalol group had
Recent MI (6–42 d) or remote MI (>42 d) SCD Placebo (n = 1,572) excessive mortality (5%) vs placebo (3.1%)
with NYHA class II or III
DIAMOND (1997) CHF study: All-cause mortality CHF (n = 1,518) Treatment with dofetilide in CHF group with
Hospital admission for CHF Dofetilide (n = 762) left ventricular dysfunction and in post-MI
LVEF ≤35% Placebo (n = 756) patients did not result in any significant
MI study: MI (n = 1,510) reduction in mortality compared with
Acute MI in past 7 d Dofetilide (n = 749) placebo
LVEF ≤35% Placebo (n = 761)
CAMIAT (1997) Acute MI in past 6–45 d All-cause mortality Amiodarone (n = 606) No difference in overall mortality between
≤10 VPDs per hour or any VT on Holter Arrhythmic death Placebo (n = 596) amiodarone and placebo groups
monitoring Resuscitated VF Amiodarone reduced cumulative risk of
arrhythmic death or resuscitated VF by
48.5%
EMIAT (1997) 5–21 d after MI All-cause mortality Amiodarone (n = 385) No difference in overall mortality or cardiac
LVEF ≤40% Cardiac death Placebo (n = 444) mortality between amiodarone and placebo
Arrhythmic death patients
Amiodarone reduced arrhythmic death by
35%
CABG Patch (1997) Scheduled for elective CABG surgery All-cause mortality ICD (n = 446) Survival was not improved by prophylactic
LVEF ≤35% Standard treatment (n = 454) implantation of ICD at time of elective
Abnormalities on SAECG CABG surgery
MUSTT (1999) CAD Cardiac arrest or arrhythmic death EPS-guided therapy (n = 351) EPS-guided antiarrhymic therapy with ICDs,
LVEF ≤40% All-cause mortality No antiarrhythmic therapy (n = 353) but not with AA drug, reduced the risk
Asymptomatic NSVT (≥3 beats) Cardiac death of SCD in high-risk patients with CAD
Spontaneous, sustained VT (arrhythmic mortality, 9% for patients with
ICD vs 37% for patients without ICD at 5 y)
SCD-HeFT (2005) Ischemic or nonischemic NYHA class II or Primary: overall mortality Placebo (n = 847) Shock-only ICD decreased mortality by 23%
III CHF, LVEF ≤35% Secondary: subgroup analysis of QOL and ICD (n = 845) Amiodarone alone did not improve survival
cost-effectiveness Amiodarone (n = 829)
IRIS (2009) Acute MI within 30 d All-cause mortality Standard care (n = 453) Identical survival at 37 mo
LVEF ≤40% ICD (n = 445) Non-SCD offset reduction in SCD
HR ≥90 bpm
NSVT
Abbreviations: AA, antiarrhythmic; bpm, beats per minute; CABG, coronary artery bypass graft; CAD, coronary artery disease; CHF, congestive heart failure; ECG, electrocardiographic; EPS, electrophysiologic study; HR,
heart rate; ICD, implantable cardioverter-defibrillator; LVEDD, left ventricular end-diastolic diameter; LVEF, left ventricular ejection fraction; MI, myocardial infarction; NSVT, nonsustained ventricular tachycardia; NYHA,
New York Heart Association; PVC, premature ventricular contraction; QOL, quality of life; SAECG, signal-averaged electrocardiogram; SCD, sudden cardiac death; VF, ventricular fibrillation; VPDs, ventricular premature
depolarizations; VT, ventricular tachycardia.
Previously published. See “Credit Lines” section.
354 III Electrophysiology

Hazard Ratio (97.5% CI) P Value

Amiodarone vs placebo 1.06 (0.86-1.30) .53
ICD therapy vs placebo 0.77 (0.62-0.96) .007

40
Placebo
(244 deaths; 5-y event rate, 0.361)
30 Amiodarone
Mortality Rate, %

(240 deaths; 5-y event rate, 0.340)

ICD therapy
(182 deaths; 5-y event rate, 0.289)
20

10

0
0 12 24 36 48 60
Months of Follow-up
No. at Risk
Amiodarone 845 772 715 484 280 97
Placebo 847 797 724 505 304 89
ICD therapy 829 778 733 501 304 103
Figure 31.1. Kaplan-Meier Estimate of Death From Any Cause, by Study Group in the SCD-HeFT. ICD indicates implantable cardioverter-
defibrillator. (Previously published. See “Credit Lines” section.)

MADIT II (2002)
The multicenter MADIT II was the third major clinical trial to
evaluate the effectiveness of ICD therapy compared with con-
ventional therapy in high-risk patients who had had an MI. The
study enrolled 1,232 patients. Invasive testing for risk stratifica-
tion was not required; inclusion criteria included prior MI (>1
month) and an EF of 30% or less. The primary end point was
death from any cause. At an average follow-up of 20 months, the
mortality rate was 19.8% in the conventional therapy group and
14.2% in the ICD group.
This trial was novel in that there was no requirement for inva-
sive electrophysiologic testing or prior ventricular arrhythmia.
Survival benefit became apparent at 9 months after device
implantation.

30

Defibrillator group
Cumulative Mortality, %

20

10

Control group

0
0 12 24 36 48
Months of Follow-up
No. at Risk
Defibrillator group 446 384 313 213 61
Control group 454 399 308 199 57

Figure 31.2. Kaplan-Meier Estimate of Death, by Study Group in the CABG Patch Trial. (Previously published. See “Credit Lines” section.)
 31 Implantable Cardioverter-Defibrillator Trials and Prevention of Sudden Cardiac Death
MADIT II expanded on the findings of 2 trials: 1) MADIT,
which showed the superiority of ICD therapy in patients with
coronary artery disease, NSVT on Holter monitoring, and an EF
of 35% or less who had inducible sustained ventricular tachy-
cardia not suppressed by intravenous procainamide (subgroup
analysis), and 2) MUSTT, which showed ICD superiority among
patients who had an EF of 40% or less, coronary artery disease,
and inducible arrhythmias regardless of whether arrhythmias
were suppressible by drug therapy. Importantly, these trials
showed the lack of any role for Holter monitoring and invasive
electrophysiologic study and the effect of antiarrhythmic drugs
on inducibility status in predicting the risk of SCD for patients
with coronary artery disease and depressed EF.
CABG Patch Trial (1997)
Although the CABG Patch Trial was reported in 1997, it is impor-
tant because it is the only trial to address the effect of revascular-
ization on the risk of SCD. Inclusion criteria for the 900 patients
in the study were an EF less than 36%, abnormal signal-averaged
electrocardiograms, and scheduled coronary artery bypass graft-
ing. They were randomly assigned to receive an ICD and standard
medical care or standard medical care alone at the time of bypass
surgery. At an average of 32 months of follow-up, there was no
difference in the primary end point (overall mortality) between
the groups (Figure 31.2). Of note, 88 patients enrolled were not
assigned to a group because their conditions were too unstable
for ICD placement at the time of surgery. Additionally, EF and
signal-averaged electrocardiograms were not assessed postoper-
atively. The patients with depressed EF postoperatively may have
had a survival advantage with ICD therapy. Also, abnormalities
on signal-averaged electrocardiography may preselect patients
at higher risk, because both groups had relatively high mortal-
ity rates. Nevertheless, the results suggest that revascularization
should be performed when feasible and that SCD risk stratifica-
tion should be performed after revascularization.
Summary
On the basis of the results of the accumulated evidence from
clinical trials and professional advisory groups, the Centers for
Medicare and Medicaid Services issued new guidelines for ICD
implantation that went into effect in January 2005 (Box 31.1).
Exceptions occur, and clinical judgment may deem that a patient
who does not meet the guidelines may have risk sufficient to
Box 31.1. Summary of Coverage for Implantable
Cardioverter-Defibrillators: January 27, 2005
1. Documented episode of cardiac arrest due to VF and
not due to a transient or reversible cause (effective
July 1, 1991)
2. Documented sustained VT, either spontaneous or
induced by an EPS, not associated with an acute
MI and not due to a transient or reversible cause
(effective July 1, 1999)
3. Documented familial or inherited conditions with
a high risk of life-threatening VT, such as long QT
syndrome or hypertrophic cardiomyopathy (effective
July 1, 1999)
(continued)
355
Box 31.1. (Continued)
4. Coronary artery disease with a documented prior MI,
a measured LVEF ≤35%, and inducible sustained VT or
VF on EPS (the MI must have occurred >4 wk before
defibrillator insertion; the EPS must be performed
>4 wk after the qualifying MI)
5. Documented prior MI and a measured LVEF ≤30%.
Patients must not have any of the following:
a. NYHA class IV
b. Cardiogenic shock or symptomatic hypotension
while in a stable baseline rhythm
c. CABG or PTCA within past 3 mo
d. MI within past 40 d
e. Clinical symptoms or findings that would make
them a candidate for coronary revascularization
f. Any disease other than cardiac disease (eg,
cancer, uremia, liver failure) associated with a
likelihood of survival <1 y
6. Patients with ischemic dilated cardiomyopathy,
documented prior MI, NYHA class II or III heart failure,
and measured LVEF <35%
7. Patients with nonischemic dilated cardiomyopathy
>9 mo, NYHA class II or III heart failure, and measured
LVEF ≤35%
8. Patients who meet all current Centers for Medicare
and Medicaid Services coverage requirements for a
cardiac resynchronization therapy device and have
NYHA class IV heart failure
Indications 3 through 8 (primary prevention
of sudden cardiac death) must also meet the
following criteria:
a. Patients must be able to give informed consent
b. Patients must not have any of the following:
• Cardiogenic shock or symptomatic hypotension
while in a stable baseline rhythm
• CABG or PTCA within the past 3 mo
• Acute MI within the past 40 d
• Clinical symptoms or findings that would make
them a candidate for coronary revascularization
• Irreversible brain damage from preexisting
cerebral disease
• Any disease other than cardiac disease (eg,
cancer, uremia, liver failure) associated with a
likelihood of survival <1 y
c. Ejection fractions must be measured with
angiography, radionuclide scanning, or
echocardiography
d. MIs must be documented and defined according
to the consensus document of the Joint European
Society of Cardiology/American College of
Cardiology Committee for the Redefinition of
Myocardial Infarction
Abbreviations: CABG, coronary artery bypass graft; EPS,
electrophysiologic study; LVEF, left ventricular ejection fraction; MI,
myocardial infarction; NYHA, New York Heart Association; PTCA,
percutaneous transluminal coronary angioplasty; VF, ventricular
fibrillation; VT, ventricular tachyarrhythmia.
See “Credit Lines” section.
356 III Electrophysiology

justify ICD placement. Additionally, even if a patient qualifies NYHA New York Heart Association
for an ICD on the basis of the guidelines, device placement may SCD sudden cardiac death
be deemed inappropriate by family and caregivers (eg, a patient
with severe dementia living in a nursing home). Notably, in all Names of Clinical Trials
trials the mean ages of the patients were in the 60s or 70s; it is not
AVID Antiarrhythmics Versus Implantable Defibril-
at all clear that patients in their 80s or 90s will derive meaningful
lators
life extension from ICD placement. BHAT β-Blocker Heart Attack Trial
CABG Patch Coronary Artery Bypass Graft Patch
• ICDs are more effective than antiarrhythmic drugs for reducing
CAMIAT Canadian Amiodarone Myocardial Infarction
mortality.
Arrhythmia Trial
• β -Blockers reduce the risk of SCD and all-cause mortality. CASCADE Cardiac Arrest in Seattle, Conventional Versus
• Electrophysiologic-guided antiarrhythmic therapy is generally Amiodarone Drug Evaluation
ineffective for predicting the risk of SCD. CASH Cardiac Arrest Study Hamburg
• EF is the best predictor of the risk of SCD. Nevertheless, it is a CAST Cardiac Arrhythmia Suppression Trial
relatively “blunt instrument,” and future trials to further refine the CHF-STAT Survival Trial of Antiarrhythmic Therapy in
risk of SCD are needed. Congestive Heart Failure
CIDS Canadian Implantable Defibrillator Study
• ICD implantation in high-risk patients early after MI does not
COMPANION Comparison of Medical Therapy, Pacing, and
reduce overall mortality.
Defibrillation in Heart Failure
Guidelines for ICD use in patients with high-risk disorders DEFINITE Defibrillators in Non-Ischemic Cardiomyopathy
such as long QT syndromes, Brugada syndrome, arrhythmic Treatment Evaluation
right ventricular dysplasia, and hypertrophic cardiomyopathy are DIAMOND Danish Investigations of Arrhythmia and
discussed in other chapters in this book. Mortality on Dofetilide
DINAMIT Defibrillator in Acute Myocardial Infarction
Trial
Suggested Reading EMIAT European Myocardial Infarct Amiodarone
Centers for Medicare & Medicaid Services. [cited 2011 Aug 18]. Trial
Available from: http://www.cms.gov/MedicareApprovedFacilitie04_ ESVEM Electrophysiology Study Versus Electrocardio-
ICDregistry.asp. graphic Monitoring
Heart Rhythm Society. [cited 2011 Aug 18]. Available from: http:// GESICA Grupo de Estudio de la Sobrevida en la
www.hrsonline.org. Insuficiencia Cardiaca en Argentina
IRIS Immediate Risk Stratification Improves
Survival
Abbreviations
MADIT, MADIT II Multicenter Automatic Defibrillator Implan-
EF ejection fraction tation Trial, MADIT II
ICD implantable cardioverter-defibrillator MUSTT Multicenter Unsustained Tachycardia Trial
MI myocardial infarction SCD-HeFT Sudden Cardiac Death in Heart Failure Trial
NSVT nonsustained ventricular tachycardia SWORD Survival With Oral D-Sotalol
 32
Heart Disease in Athletes
STEPHEN C. HAMMILL, MD
The medical community, the public, and athletes all have con-
cerns about identifying a priori persons at risk for sudden cardiac
death. Physicians are frequently asked to assess the suitability
of an athlete to participate in a particular sport. Ideally, iden-
tification of persons at risk and resultant interventions should
be aimed at younger athletes at a stage in their lives when it is
easier for them to alter their lifestyles and redirect their interests
to other activities.
In high school, the athletic activity is extracurricular, and the
athlete is typically a minor who is considered an amateur. As
athletes age throughout college and gain independence as adults,
their sport becomes a vocation and there are considerable per-
sonal and institutional financial incentives to induce them to pur-
sue athletic competition. Professional athletes may have devoted
their entire lives to athletic training, and they have even stronger
financial and organizational interests in playing.
Typically, compared with nonathletes, athletes have a slower
minimal heart rate, a higher percentage of sinus pauses lasting
longer than 2 seconds, and more Mobitz I atrioventricular block
(ie, Wenckebach block) because of higher vagal tone that occurs
with conditioning. The slower heart rate causes a compensatory
increase in stroke volume to maintain cardiac output; therefore,
athletes often have an enlarged left ventricle without increased left
ventricular wall thickness. Sports associated with the largest left
ventricular size include rowing, cycling, and cross-country skiing.
The causes of sudden death in athletes depend on age. Among
athletes younger than age 35 years, approximately 25% who die
suddenly are found at autopsy to have hypertrophic cardiomy-
opathy, 20% commotio cordis, and 14% coronary abnormalities;
the remainder may have idiopathic left ventricular cardiomy-
opathy, ruptured aorta, long QT syndrome, Brugada syndrome,
polymorphic ventricular tachycardia, arrhythmogenic right ven-
tricular dysplasia, or idiopathic ventricular fibrillation. In con-
trast, among athletes 35 years or older who die suddenly, 80% are
found at autopsy to have severe coronary disease and 5% each
are found to have hypertrophic cardiomyopathy, mitral valve
prolapse, acquired valvular disease, or no identified cause.
Screening for heart disease that can lead to sudden death is
difficult because of the large number of persons who must be
screened to prevent only 1 such death. In asymptomatic persons
younger than age 35 years, approximately 200,000 athletes would
have to be screened to identify just l in whom sudden death could
be prevented (Figure 32.1). Screening is also difficult in com-
petitive athletes who are asymptomatic and older than age 35
years. In this group, 10,000 asymptomatic men would have to be
screened to identify 100 with risk factors and to prevent 1 sudden
death among those 100, but 4 sudden deaths would have occurred
in the group of 9,900 with no risk factors who had a normal stress
test (Figure 32.2). Obviously, intensive screening would not be
cost-effective in either group. Furthermore, published reports of
studies have focused almost exclusively on male athletes; their
applicability to female athletes is unknown.
Guidelines for determining athletic eligibility, which were
formulated in 2005 (and later updated in 2008) at the 36th
Bethesda Conference and in a European Consensus Document,
have been published to help guide the clinical approach used in
managing patients with cardiac abnormalities who are athletes.
These guidelines recommend that athletes be screened initially
by obtaining a history and conducting a physical examination.
If the history identifies previous near syncope, syncope, pal-
pitations, or a family history of sudden cardiac death, or if the
physical examination identifies a serious heart murmur or other
cardiac problem, the athlete should have an electrocardiogram,
an echocardiogram, and stress testing. Depending on the results
of these tests, electrophysiologic testing or long-term monitoring
should be considered. Hypertrophic cardiomyopathy is the most
common abnormality found during the screening of athletes for
athletic eligibility.
357
358 III Electrophysiology
Screening for sudden death risk
in asymptomatic competitive athletes
<35 years old
200,000 athletes screened
0.5% with congenital heart disease
1,000 with disease
1% with disease capable of sudden death
10 athletes at risk for sudden death
10% die suddenly
1 sudden death
Figure 32.1. Screening for risk of sudden death in asymptomatic
competitive athletes younger than age 35 years. (Previously published.
See “Credit Lines” section.)
The athlete who presents with syncope or palpitations should
provide a history and undergo a physical examination, along with
having an electrocardiogram, an echocardiogram, and stress
testing. If no heart disease is identified, further evaluation may
include tilt testing and event recording; if heart disease is identi-
fied, further evaluation may include electrophysiologic studies,
followed by tilt testing or event recording if no serious abnormal-
ity is identified. In this population, an implantable loop recorder
is extremely useful for capturing the electrocardiographic find-
ings that occur during infrequent symptoms.
In summary:
1. Athletes usually have an enlarged left ventricle without increased
left ventricular wall thickness.
10,000 asymptomatic men:
sudden death 0.5%/y
Exercise test +
risk factor screen
+100 –9,900
Sudden death Sudden death
0.8%/y 0.4%/y
1 death 4 deaths
Figure 32.2. Screening for risk of sudden death in asymptomatic
competitive athletes aged 35 years or older. (Previously published. See
“Credit Lines” section.)
2. Hypertrophic cardiomyopathy is the most common abnormality iden-
tified during the screening of athletes before athletic participation.
3. Among athletes younger than age 35 years, approximately 25%
who die suddenly are found at autopsy to have hypertrophic cardio-
myopathy.
4. Among athletes aged 35 years or older, 80% are found at autopsy to
have severe coronary artery disease.
5. Patients should always be excluded from competitive athletics if they
have any of the following conditions: hypertrophic cardiomyopathy,
Brugada syndrome, congenital long QT syndrome, arrhythmogenic
right ventricular dysplasia, or severe aortic stenosis.
Suggested Reading
Link MS, Estes NA. Athletes and arrhythmias. J Cardiovasc Electro-
physiol. 2010 Oct;21(10):1184–9.
Maron BJ, Zipes DP. Introduction: eligibility recommendations for
competitive athletes with cardiovascular abnormalities: general con-
siderations. J Am Coll Cardiol. 2005 Apr 19;45(8):1318–21.
Pelliccia A, Zipes DP, Maron BJ. Bethesda Conference #36 and the
European Society of Cardiology Consensus Recommendations revis-
ited a comparison of U.S. and European criteria for eligibility and
disqualification of competitive athletes with cardiovascular abnor-
malities. J Am Coll Cardiol. 2008 Dec 9;52(24):1990–6.
 33
Autonomic Function Testing
K. L. VENKATACHALAM, MD
Introduction
Syncope may be defined as a transient loss of consciousness due
to a temporary limitation in cerebral blood flow. Causes of syn-
cope include the following:
1. Low systemic blood pressure (arrhythmias, valvular disease, medi-
cations, and reflex causes)
2. Reduced cerebral blood flow (cerebral vasospasm and hyperven-
tilation)
3. Miscellaneous causes of cerebral dysfunction (metabolic, endocrine,
and electrolyte abnormalities)
The autonomic nervous system may have a role in syncopal and
near-syncopal episodes for the following reasons:
1. Inadequate venous return to the right heart
2. Low peripheral vascular resistance
3. Autonomic neuropathies (eg, pure autonomic failure, multiple sys-
tem atrophy)
Autonomic failure and autonomic function testing are vast and
complex areas of neurology, but the focus of this chapter is auto-
nomic function assessment related to cardiology and, specifi-
cally, to the evaluation of a patient referred for syncope.
Autonomic Anatomy and Physiology
Related to Cardiology
Tissue perfusion depends on sympathetic efferent pathways, sys-
temically acting hormones, and locally acting substances such as
nitric oxide. Cardiovascular afferents from the carotid sinus, aor-
tic arch, and cardiopulmonary region provide afferent inputs to
the cortical, limbic, hypothalamic, and medullary centers of the
Abbreviations and acronyms are expanded at the end of this chapter.
brain. Lesions in any of these areas can impair cardiovascular
reflex activity, resulting in abnormal control of the heart rate and
blood pressure, which can manifest as presyncope, orthostasis,
or frank syncope.
Autonomic Function Tests
Physiologic
Head-Up Tilt Test
To determine a patient’s predisposition to neurally mediated bra-
dycardia or hypotension (or both), a strong orthostatic stimulus is
very useful. The HUT is a safe and effective approach to produce
this stimulus.
The HUT may be performed with or without the use of pro-
vocative agents such as intravenous isoproterenol. The test is
usually performed in the morning after the patient has fasted
overnight. A tilt table with a weight-supporting footboard is
used to evaluate the patient in a supine and 70º inclined position.
Heart rate and rhythm and continuous blood pressure are mon-
itored during the test. The inclination is maintained for 30 to
45 minutes. If the test is negative at this point, carotid sinus mas-
sage and isoproterenol infusion are given to provoke an episode
of syncope or near-syncope. Increasing the tilt angle and admin-
istering isoproterenol increase the sensitivity of the test at the
expense of specificity. Several responses to this test have been
described:
1. Neurocardiogenic pattern: abrupt onset of hypotension or bradycar-
dia (or both)
2. Dysautonomic pattern: gradual decrease in systolic and diastolic
blood pressure with no significant change in heart rate and usually
accompanied by other signs of autonomic dysfunction (see discus-
sions of TST and QSART below)
359
360 III Electrophysiology

3. Postural orthostatic tachycardia syndrome pattern: a heart rate Sudomotor

increase of 30 beats per minute or more within 10 minutes of upright
tilt without significant hypotension Eccrine sweat glands receive a rich supply of sympathetic nerve
4. Cerebral syncope pattern: tilt-induced cerebral vasoconstriction fibers (cholinergic) and are primarily responsible for thermoregu-
(by transcranial Doppler ultrasound) without significant systemic lation. Long autonomic sudomotor fibers from the hypothalamus
hypotension are the efferent inputs to these glands. Interrupting the course of
5. Psychogenic pattern: syncope without a significant change in heart
rate, blood pressure, or cerebral blood flow

Carotid Sinus Massage

Sensory fibers arising from the carotid sinus (afferents) attach
to several cranial nerves (IX, X, and XII) and the cervical
sympathetic nerves. The efferent limb of this circuit involves
the sympathetic and parasympathetic innervation to the heart.
Changes in arterial blood pressure change the stretch on the
carotid sinus and modify the afferents and, subsequently, the
sympathetic input or parasympathetic input (or both) to the
heart. An exaggerated response to stimulation produces carotid
sinus hypersensitivity. Carotid sinus massage is used to deter-
mine heart rate and blood pressure response. It should not be
performed if patients have carotid bruits or known carotid
disease. Gentle pressure is applied at the angle of the jaw to
compress the carotid bulb against the transverse processes of
the adjacent cervical vertebrae. Three responses have been
described:
1. Cardioinhibitory: ventricular asystole for 3 seconds or more
2. Vasodepressor: a decrease in systolic blood pressure of 30 mm Hg
or more with symptoms, or a decrease in systolic blood pressure of
50 mm Hg or more without symptoms
3. Mixed: both asystole and hypotension are present

Pharmacologic
Norepinephrine (α-Adrenergic Receptor Function)
Norepinephrine predominantly affects α-adrenergic receptors
but has some effect on β-adrenergic receptors. Infusing low
doses of norepinephrine (starting at 0.01–0.05 mcg/kg/min)
while assessing the blood pressure response provides evidence
for vascular sensitivity and a possible response to sympathomi-
metic treatment. Patients with autonomic failure may have signif-
icant supine hypertension (which can be quickly corrected with
a head-up tilt). An impaired pressor response has been described
in systemic amyloidosis (presumably due to vascular infiltration
by amyloid protein).

Isoproterenol (β-Adrenergic Receptor Function)

Isoproterenol acts on β1- and β2-adrenergic receptors. Heart rate
and blood pressure are monitored during intravenous infusion
of the drug. An exaggerated decrease in systolic and diastolic
blood pressures may occur in quadriplegics and patients with
autonomic failure.

Atropine (Parasympathetic Function)
Atropine produces a strong muscarinic blockade of postsynap-
tic parasympathetic and sympathetic cholinergic receptors. It is
administered intravenously as 5-mcg/kg boluses every 2 minutes
to increase heart rates. Dose-response curves are constructed to
heart rates of about 110 beats per minute. A flat response is seen
when patients have pure autonomic failure or multiple system
atrophy.
Figure 33.1. Thermoregulatory Sweat Test Results Showing
Widespread Anhidrosis in Multiple System Atrophy. Sweating is shown
in purple. Oral temperature was 36.0°C before and 38.0°C after test-
ing. Body surface anhidrosis was 85%. Distribution was segmental.
Impression: Thermoregulatory sweating was decreased to absent over
the legs and lower trunk. With quantitative sudomotor axon reflex test-
ing, results show a severe segmental (preganglionic and postganglionic)
sudomotor deficit compatible with multiple system atrophy rather than
Parkinson disease.
 33 Autonomic Function Testing
Figure 33.2. Thermoregulatory Sweat Test Results Showing
Minimal Patchy Anhidrosis Consistent With Small Fiber Neuropathy.
Sweating is shown in purple. Oral temperature was 35.9°C before and
37.6°C after testing. Body surface anhidrosis was 2%. Distribution was
distal. Impression: Patchy anhidrosis affected the medial forehead and
the distal areas of the feet and toes. Results are compatible with a very
distal small fiber neuropathy.
the fibers impairs the sweat response. The TST allows evaluation
of the sweat response in preganglionic and postganglionic fibers.
Postganglionic fibers alone may be evaluated by the QSART (see
below).
Thermoregulatory Sweat Test
For the TST, the unclothed patient lies supine on a cart, and
exposed body surfaces are covered with an indicator powder
361
100
1.77 μL/cm2
Volume,
arbitrary 50
Sweat units
gland
0
Sweat 0 2 4 6 8 10 12
gland Minutes
Sudomotor axon
reflex
G E
C
F B
G D A
E
Figure 33.3. Quantitative Sudomotor Axon Reflex Test. Top, The
sudomotor axon reflex with a typical response. In the volume-time graph,
the arrow pointing down indicates initiation of stimulation; the arrow
pointing up indicates end of stimulation. Bottom, A multicompartment
sweat cell. By iontophoresis, acetylcholine passes through compartment
C with use of cannula E. Current is injected between anode F and a dis-
tant cathode. Humidified nitrogen through D is used to evaporate sweat
produced in compartment A. Compartment B is a dividing chamber.
Posts G are used to strap the cell in place. (Previously published. See
“Credit Lines” section.)
mixture (alizarin red, sodium carbonate, and cornstarch). The
patient lies in a sweat cabinet at 44°C to 50°C and 35% to 45%
relative humidity to maintain a skin temperature of 39°C to
40°C for 45 to 65 minutes. Under these circumstances, the mean
increase in oral temperature is 1.2°C. A color change between
areas of hidrosis and anhidrosis is recorded with a digital cam-
era, and the relative areas are quantified. Widespread (>80%)
anhidrosis is characteristically seen in multiple system atrophy
and pure autonomic failure (Figure 33.1).
Peripheral neuropathies produce distal anhidrosis due to post-
ganglionic denervation (Figure 33.2). In diabetic neuropathy,
preganglionic anhidrosis may also be present.
Quantitative Sudomotor Axon Reflex Test
The QSART allows assessment of the postganglionic sweat
response and is used in conjunction with the TST (see above) to
identify autonomic causes of orthostastic syncope. Figure 33.3
shows a multicompartment sweat cell and the physiologic basis
of this test. By means of iontophoresis, acetylcholine passes
through the outer compartment of the sweat cell. An antidromic
impulse due to the acetylcholine travels to the branch point and
is an afferent input to the spinal cord. Simultaneously, an ortho-
dromic signal travels to an adjoining sweat gland and causes the
release of sweat. The sweat is evaporated with a nitrogen stream
and quantified at 4 different sites on the body. An absent sweat
response indicates a lesion of the postganglionic axon. In pregan-
glionic or central disorders, the QSART is unimpaired.
362 III Electrophysiology

Biochemical range ≤100 mcg per 24 hours) may also be used for the diagnosis
of pheochromocytoma.
Plasma Norepinephrine Levels
(Supine Testing and HUT)
Since norepinephrine is the main neurotransmitter at sympa- Suggested Reading
thetic nerve endings, plasma norepinephrine is widely used to Mathias CJ, Bannister R. Autonomic failure: a textbook of clinical
assess increases in sympathetic activity. It is a reliable indicator disorders of the autonomic nervous system. 4th ed. Oxford; Oxford
of peripheral nerve traffic. The HUT increases plasma norepi- University Press; c1999.
nephrine levels, but no significant change occurs in patients with Robertson D, Biaggioni I, Burnstock G, Low PA, editors. Primer on the
autonomic nervous system. 2nd ed. Amsterdam; Elsevier Academic
sympathetic failure. Basal plasma norepinephrine levels may be
Press; c2004.
normal in multiple system atrophy but are usually low in pure
autonomic failure. In addition, plasma norepinephrine levels
are used in the diagnosis of pheochromocytoma (norepineph- Abbreviations
rine ≥1,000 pg/mL) because patients may present with postural HUT head-up tilt test
hypotension (due to reduced plasma volume and desensitization QSART quantitative sudomotor axon reflex test
of α-adrenergic receptors). Urine catecholamine levels (reference TST thermoregulatory sweat test
 34
Antiarrhythmic Drugsa
PETER A. BRADY, MB, CHB, MD
Effects of Antiarrhythmic Drugs
on Reentrant Circuits
Reentrant arrhythmias are sustained by a propagating wave front
that continually excites myocardial tissue. When the propagat-
ing wave front has passed, myocardial tissue is refractory to fur-
ther stimulation and needs to recover before another propagating
wave front can reexcite the tissue. The amount of tissue in the
reentrant circuit that fully recovers from depolarization and can
therefore be excited again is known as the excitable gap.
In general terms, antiarrhythmic drugs work in 1 of 2 ways:
to slow the velocity of the propagating wave front (the action of
drugs that predominantly block sodium channels) or to increase
the amount of cardiac tissue that is refractory and, therefore,
cannot be excited by the propagating wave front (predomi-
nantly drugs that block potassium channels). In this way, antiar-
rhythmic drugs act to alter the myocardial-electrical substrate
to decrease the stability of sustained reentrant circuits. In some
circumstances, drugs may either slow conduction too much or
alter the refractoriness of remote parts of the reentrant circuit
that increases, rather than decreases, the stability of the reen-
trant circuit and, in this way, may be proarrhythmic.
Classification of Antiarrhythmic Drugs
The most commonly used classification system for antiarrhyth-
mic drugs continues to be the Vaughan Williams classification.
a
Portions previously published in Brady PA, Gersh BJ. Atrial fibrillation
and flutter. In: Willerson JT, Cohn JN, Wellens HJJ, Holmes DR Jr,
editors. Cardiovascular medicine. 3rd edition. London: Springer Verlag;
c2007. p. 1955–78. Used with permission.
Abbreviations and acronyms are expanded at the end of this chapter.
This system, although not perfect, differentiates drugs on the
basis of the ion channel or receptor, or both, that is affected
(Figure 34.1). The classes are the following:
• Class I: predominantly sodium channel blockers.
• Class II: β-adrenergic receptor blockers.
• Class III: potassium channel blockers (all clinically used agents
block IKr and prolong the QT interval).
• Class IV: calcium channel blockers.
Despite the usefulness of the Vaughan Williams classifica-
tion, it is, of course, an oversimplification since many drugs exert
effects at multiple levels or on multiple ion channels (eg, amio-
darone or its metabolites block sodium, potassium, calcium, and
β-adrenergic receptors), or both, and drug effect may be altered
by such tissue factors as ischemia or injury.
• All antiarrhythmic drugs should be considered proarrhythmic.
• Likelihood of proarrhythmia is increased in patients with struc-
tural heart disease (especially myocardial ischemia and LV
systolic dysfunction).
• β -Adrenergic receptor blockers are the only class that has favor-
able effects on mortality rate.
• Amiodarone and dofetilide have a neutral impact on mortality rate
following MI and in the presence of structural heart disease.
Class I Antiarrhythmic Drugs
Class I drugs are further subdivided on the basis of potency of
sodium channel blockade (IC > IA > IB). In addition, class I drugs
bind to the sodium channel in such a way that the drug effect
is increased at faster heart rates. This phenomenon is known as
use-dependent action. The clinical importance of this effect is
2-fold: sodium channel blocking agents work best at faster heart
363
364 III Electrophysiology
Class IV
Ca++ blockers
Class I - 95
Class III
K+ blockers
Na + blockers
- -
- Class II
β-Blockers
Class II
Figure 34.1. Summary of the Effect and Site of Action of
Antiarrhythmic Drugs. Ca ++ indicates calcium; K + , potassium; Na + ,
sodium.
rates (so may be best for chemical cardioversion) and, through
a similar action, the proarrhythmic effect is increased at faster
heart rates.
• Use-dependent action is the reason why patients whose therapy is
started with a sodium channel blocking drug should undergo tread-
mill exercise testing to increase their heart rate and maximize the
likelihood of unmasking proarrhythmia.
Clinical Monitoring of Class I Agents
Because sodium channel blocking agents act to slow conduction,
the tissue effects of the drug on the surface ECG include PR inter-
val prolongation and an increase in QRS duration. Generally, at
a normal heart rate, a 10% or 15% increase in the QRS duration
suggests an adequate tissue level of a sodium channel blocking
drug. Other effects of sodium channel blocking drugs include an
increase in pacing threshold (block of sodium channels means
more energy is required to stimulate the heart to generate an
action potential).
Proarrhythmic Effects of Class I Agents
First evidence for the potential for proarrhythmia of class I agents
came from CAST. The hypothesis of this trial was that suppres-
sion of VPCs in patients shortly after MI would be beneficial in
terms of mortality rate since it was assumed that VPCs lead to
life-threatening ventricular arrhythmias. However, unexpectedly,
mortality rate was increased in the group randomly assigned to
flecainide or encainide (Figure 34.2).
CAST was conducted in a specific population of patients in
an era when thrombolysis or percutaneous intervention was not
routine. Nevertheless, its findings of increased mortality rate
with use of flecainide and encainide (not currently available in
the United States) indicate that caution is important in the use of
class I agents for patients with structural heart disease (especially
unrevascularized coronary disease).
Specific Class I Agents
Quinidine (Class IA)
Although quinidine is less commonly used now than in previous
times, questions about it continue to be testable for cardiovascu-
lar board examinations. It is used in the treatment of atrial and
ventricular arrhythmias.
100
Survival, %
90
Placebo (n=725)
85
Encainide or
flecainide (n=730) P=.0006
0
0 100 200 300 400 500
Days After Randomization
Figure 34.2. CAST Results. When used soon after myocardial
infarction, encainide and flecainide were associated with increased
mortality compared with placebo. (Previously published. See “Credit
Lines” section.)
Proarrhythmia
Proarrhythmia is dose independent and may occur with normal
or even subtherapeutic drug levels. Even though quinidine is a
class I agent, it does have potassium channel blocking effect, and
therefore the therapeutic effect of quinidine is determined with
reference to the corrected QT interval. In general, a QT prolon-
gation greater than 500 msec should prompt abandonment of this
drug class or a decrease in the drug dose. However, some degree
of QT prolongation from baseline suggests antiarrhythmic drug
effect due to prolonged repolarization.
Quinidine syncope is most frequently due to polymorphic
VT (incidence, 0.5%-4.4%) and is common to all class IA drugs
(concordance rate of approximately 30%) and is most common
at a median of 3 days after initiation of drug therapy. Thus, if
polymorphic VT (torsades de pointes) is observed with 1 class
IA agent, all other class IA agents should be avoided. Because of
the risk of proarrhythmia, all class IA drugs should be initiated
in the hospital.
• To reduce the risk of polymorphic VT, quinidine should be discon-
tinued at a corrected QT interval greater than 500 msec.
• If polymorphic VT occurs, then IV magnesium or isoproterenol or
temporary pacing at 100 to 110 pulses per minute, or both, can be
tried, which will suppress the underlying cellular mechanism of
torsades de pointes early after depolarization.
Hemodynamics
The α-adrenergic blocking property of quinidine, in addition to
peripheral vasodilation, may cause orthostatic hypotension and
reflex tachycardia. Quinidine does not cause clinically important
negative inotropic effect, even when LV dysfunction is severe.
The vagolytic effect of quinidine may enhance AV nodal conduc-
tion and increase ventricular response during atrial fibrillation or
flutter. Thus, an AV nodal blocking agent should always be used
before starting quinidine therapy.
Important Drug Interactions
Quinidine inhibits the cytochrome P450 2D6 system and thus
decreases the metabolism of propranolol, metoprolol, and
propafenone. Digoxin levels are increased (and may be doubled in
 34 Antiarrhythmic Drugs
many cases) because of decreased tissue binding (lower volume of
distribution) and decreased renal and biliary clearance. Therefore,
the digoxin dose should be halved when used with quinidine. The
anticoagulant effect of warfarin also may be increased. Heparin
displaces quinidine and increases the unbound fraction of the
plasma. Verapamil may increase quinidine levels by decreasing
its metabolism. Amiodarone increases quinidine levels and also
has an increased impact on QT interval prolongation.
• Quinidine has important interaction with many commonly used
cardiac drugs, including digoxin, warfarin, heparin, propranolol,
metoprolol, propafenone, calcium channel blockers, cimetidine,
and enzyme-inducing drugs (phenobarbital, phenytoin, and
rifampicin).
Adverse Effects
Adverse GI effects (abdominal cramping and diarrhea) often
limit the use of quinidine. In addition, adverse effects include
rash cinchonism (hearing decrease, tinnitus, and blurred vision,
and possible delirium when severe), thrombocytopenia, Coombs
test–positive hemolytic anemia, lupus syndrome with antihistone
antibodies, and granulomatous hepatitis (rare).
Procainamide (Class IA)
As with quinidine, procainamide is useful for both atrial and
ventricular arrhythmias. However, IV administration may lead to
hypotension because of a decrease in sympathetic efferent activity
(likely caused by ganglionic blockade). This effect is not seen with
oral loading of procainamide. In approximately 10% of patients,
procainamide is associated with a worsening of CHF symptoms.
Proarrhythmia
Proarrhythmic potential increases when procainamide is given
in combination with other drugs that prolong the QT interval
(especially class III agents).
Important Drug Interactions
Amiodarone increases procainamide levels.
Adverse Effects
GI adverse effects limit the use of procainamide in approxi-
mately 25% of patients. Other adverse effects that may be seen
are rash, fever, Raynaud phenomenon, agranulocytosis, Coombs
test–positive hemolytic anemia, depression, psychosis, and chole-
static jaundice.
Another adverse effect is drug-induced lupus. Antinuclear
antibodies form in about 80% of patients taking procainamide,
with most patients undergoing serologic conversion in the first
6 months of therapy. In approximately 30% of these patients,
clinical lupus develops with features that include fever, arthral-
gias, myalgias, serositis, pulmonary infiltrates, and hepatome-
galy. Antibodies to histones are reported in two-thirds of
affected patients, and antibodies to DNA may also be detected.
Antibodies to double-stranded DNA are less frequent than in the
idiopathic form of lupus.
• Drug-induced lupus differs from the idiopathic type in that renal,
bone marrow, and CNS involvement is rare.
• Considerable proportions of the US population are slow acetyla-
tors and have antibodies. In them, lupus syndrome develops sooner
and at a lower cumulative dose than in rapid acetylators.
365
• N-Acetylprocainamide is a major procainamide metabolite that is
not thought to produce lupus syndrome. Thus, patients with procain-
amide-induced lupus may be treated with N-acetylprocainamide.
Disopyramide (Class IA)
Disopyramide is useful in patients with both atrial and ventricu-
lar arrhythmias. An important problem with the clinical use of
disopyramide is the marked negative inotropic effect, which is
related directly to plasma levels and is especially problematic in
patients with underlying LV dysfunction. However, this property
makes disopyramide useful in patients with diastolic dysfunction
and particularly useful in patients with hypertrophic obstructive
cardiomyopathy.
Important Drug Interactions
Phenobarbital, phenytoin, and rifampicin increase hepatic
metabolism of disopyramide and decrease its plasma levels. The
negative inotropic effect of some drugs, such as β-adrenoreceptor
blockers, is additive when used in combination with disopyra-
mide. Erythromycin may increase disopyramide levels by inhib-
iting hepatic monodealkylation, and this effect has led to torsades
de pointes in some patients.
Adverse Effects
Anticholinergic effects such as dry mouth, blurred vision, con-
stipation, and urinary retention occur in about 30% of patients.
Bladder effects, typically seen in older men, may be counteracted
with bethanecol. Pyridostigmine prevents or diminishes the anti-
cholinergic effect of disopyramide and allows high tolerated
doses of the drug.
Disopyramide-induced hypoglycemia has been reported and
may be due to enhanced secretion of insulin. Predisposing factors
appear to include advanced age, malnutrition, and chronic renal
failure. Other important adverse events include nausea, vomit-
ing, rash, cholestatic jaundice, agranulocytosis, and increased
uterine contractions.
Lidocaine (Class IB)
Lidocaine is used exclusively for ventricular arrhythmias and
is most useful in partially depolarized (ischemic) myocardium.
One other advantage of lidocaine is that it has minimal effect on
hemodynamics.
Proarrhythmia
Clinically relevant proarrhythmias secondary to lidocaine are
infrequent.
Important Drug Interactions
Propranolol and metoprolol decrease hepatic blood flow and,
consequently, the metabolism of lidocaine, thereby increas-
ing the plasma level by up to 80%. Phenobarbital decreases the
plasma concentration of lidocaine.
Adverse Effects
CNS effects predominate, including perioral numbness,
paraesthesia, diplopia, hyperacusis, slurred speech, altered
consciousness, seizures, respiratory arrest, and coma. Infra-
His conduction block has been reported in some patients with
preexisting conduction system disease. Sinus node depression
366 III Electrophysiology
can be seen, especially in patients with underlying sinus node
disease.
• Lidocaine is hepatically metabolized and thus conditions that
decrease hepatic blood flow (especially CHF and use of β-blockers)
increase the risk of lidocaine toxicity.
Mexiletine (Class IB)
Mexiletine is best thought of as oral lidocaine and, similar
to lidocaine, is useful exclusively in patients with ventricular
arrhythmias.
Proarrhythmia
The incidence of serious proarrhythmia due to mexiletine is
low (<2%).
Hemodynamics
Mexiletine has minimal hemodynamic effect on heart rate, blood
pressure, cardiac output, or intracardiac pressures. In some
patients with severe LV dysfunction, mexiletine may cause an
increase in pulmonary capillary wedge pressure, but this effect
is variable. EF is generally unaltered by mexiletine therapy, even
in patients with structural heart disease.
Important Drug Interactions
Phenytoin, phenobarbital, and rifampicin increase the metabo-
lism of mexiletine. Digoxin and warfarin are not affected by
mexiletine.
Adverse Effects
GI and CNS effects are most prominent (dependent on dose and
concentration). Tremor is typically the first sign of CNS toxicity,
but blurred vision, dysarthria, ataxia, and confusion may occur.
Tremor may respond to β-adrenergic blocking agents. Abnormal
results on liver function tests, thrombocytopenia, and other blood
dyscrasias (rare) may be seen.
Phenytoin (Class IB)
Phenytoin is used in the treatment of ventricular arrhythmias.
Hemodynamics
Hypotension may occur during IV administration (probably
because of the dilutant in the IV preparation). No relevant ven-
tricular depressant effect develops, but worsening of sinus node
function may occur (especially in patients with underlying sinus
node disease).
Important Drug Interactions
Phenytoin induces hepatic enzymes and thereby increases metabo-
lism of numerous drugs, including quinidine, disopyramide, lido-
caine, mexiletine, and theophylline. Phenytoin levels are increased
with use of isoniazid, chloramphenicol, disulfiram, and some sul-
fonamides. Antacids decrease the absorption of phenytoin.
Adverse Effects
Gingival hyperplasia is common (incidence, approximately
50%). Elimination of preexisting periodontal disease and thor-
ough oral hygiene may prevent this hyperplasia. CNS effects,
such as nystagmus and ataxia, are concentration dependent.
Adverse effects that may be seen are rash, nausea, blood dys-
crasia, lupus syndrome, peripheral neuropathy, hyperglycemia,
lymphadenopathy, Stevens-Johnson syndrome, hirsutism, and
osteomalacia. Extravasation during IV administration may cause
serious tissue damage or limb loss.
• A rare, but important, adverse event is phenytoin hypersensitivity
syndrome, characterized by fever, skin lesions (ranging from acne
form to erythema multiforme), lymphadenopathy, hepatospleno-
megaly, and leukocytosis with eosinophilia.
Flecainide (Class IC)
This drug is useful in the treatment of atrial and ventricular
arrhythmias.
Proarrhythmia
The proarrhythmic effects of flecainide were demonstrated in
CAST.
• Exercise testing is recommended before hospital dismissal to
maximally increase heart rate—the use-dependent effect—so as
to increase the likelihood of detecting a proarrhythmic effect.
Hemodynamics
Heart rate is typically unaffected by flecainide except in the pres-
ence of underlying sinus node dysfunction. Negative inotropic
effect is similar to that of disopyramide, and initiation of flecain-
ide may occasionally exacerbate CHF in patients with LV dys-
function. No effect on EF occurs when EF is greater than 50%.
Important Drug Interactions
Flecainide increases digoxin levels by 25% (through decreased
clearance); amiodarone and cimetidine increase flecainide
levels. Quinidine inhibits hepatic metabolism of flecainide and
increases the elimination half-life by about 20%.
Adverse Effects
CNS effects, such as blurred vision, headache, and ataxia, are the
most common adverse effects of flecainide. CHF may develop
in patients with underlying LV systolic dysfunction. In patients
with atrial arrhythmias, flecainide is best used in combination
with a β-adrenergic blocker, to prevent atrial flutter with 1:1 AV
conduction.
Propafenone (Class IC)
Propafenone may be used to treat atrial and ventricular
arrhythmias.
Proarrhythmia
Polymorphic VT and VF have rarely been reported shortly after
initiation of propafenone therapy. Overall incidence of ven-
tricular proarrhythmia with propafenone is about 5%, and such
ventricular proarrhythmia is rare in patients with structurally
normal hearts. Molar sodium lactate reportedly reverses the
arrhythmogenic effect of propafenone.
Hemodynamics
Resting sinus rate is usually unchanged with propafenone ther-
apy, but heart rate response to exercise may be blunted (negative
chronotropic effect). Exacerbation of symptoms may occur in up
to 10% of patients with a history of CHF.
 34 Antiarrhythmic Drugs
Important Drug Interactions
Approximately 7% of the US population do not have cyto-
chrome P450 2D6 enzyme activity that metabolizes propafenone
to 5-hydroxy propafenone, an intermediate that has less
β-blockade than the parent compound. In patients without this
enzyme, propafenone levels are markedly higher while levels
of 5-hydroxy propafenone are lower. This result may cause
profound β-blockade effect. Digoxin levels are increased more
than 80%, possibly because of decreased nonrenal clearance
and decreased volume of distribution. Warfarin clearance is
decreased (prothrombin time is increased >40%). Propranolol
and metoprolol are both metabolized by the cytochrome
P450 2D6 system, so levels are increased in the presence of
propafenone. Theophylline, cyclosporine, and disopyramide
levels are increased also. Phenytoin, phenobarbital, and rifam-
picin increase the metabolism of propafenone. Quinidine blocks
the cytochrome P450 system and inhibits the conversion of
propafenone to 5-hydroxy propafenone in “extensive metabo-
lizers.” Cimetidine causes an increase in propafenone levels,
resulting in a small, but important, lengthening of the QRS
duration.
• Propafenone increases digoxin and warfarin effects.
Adverse Effects
Adverse effects are metallic taste (most common with dairy
products especially), blurred vision, paresthesias, constipation,
increased results on liver function test, exacerbation of asthma
(probably due to β-adrenergic blocking effects), and conduction
abnormalities.
CNS adverse effects are related to propafenone plasma con-
centration and are more frequent in “poor metabolizers.”
Class III Drugs
Since class III drugs prolong the QT interval, the most common
proarrhythmic response is polymorphic VT. However, other
proarrhythmic effects also occur.
Risk factors for proarrhythmia in patients taking QT-
prolonging drugs such as sotalol include bradycardia, base-
line QT interval prolongation, female sex, hypokalemia, and
hypomagnesemia.
• Proarrhythmia is typically dose dependent (especially in the case
of sotalol).
• Most class III agents (except amiodarone) reduce the defibril-
lation threshold in patients with an implantable cardioverter-
defibrillator.
• Reverse-use dependency (ie, greater efficacy at slower heart rate) is
a property especially associated with sotalol.
Specific Class III Agents
Amiodarone (Class III)
Amiodarone is a complex drug with multiple ion channel and
receptor actions. It is used in the treatment of atrial and ventricu-
lar arrhythmias.
Proarrhythmia
Amiodarone may cause life-threatening ventricular arrhythmias,
including VF and torsades de pointes, but these arrhythmias are
less common than with other antiarrhythmic agents.
367
Hemodynamics
Amiodarone is generally well tolerated by patients with impaired
LV function and CHF, such that the EF is unchanged even in
patients with severely reduced ventricular function. However,
administration of IV amiodarone may cause hypotension.
Important Drug Interactions
Amiodarone use increases digoxin, warfarin, and cyclosporine
concentrations, and therefore the physician should reduce or
halve the dose of these agents. Because of the β-blocking and
calcium channel blocking activities of amiodarone, doses of
other agents with this action should be reduced.
Adverse Events
Pulmonary, hepatic, and thyroid toxicities are most common.
Hyperthyroidism is less common than hypothyroidism but is
much more difficult to treat. (Usually, hyperthyroidism requires
antithyroid drugs or surgical removal of the gland; radioiodine
treatment is not effective.) Pulmonary toxicity may present as
dyspnea, nonproductive cough, fever, or pleuritic chest pain or
with asymptomatic changes on chest radiograph. Photosensitivity,
discoloration of the skin, and corneal microdeposits (rarely a
clinical problem) may occur. Monitoring for amiodarone toxicity
should be done (Table 34.1).
Important Clinical Trials
1. CAMIAT. This trial randomly assigned amiodarone therapy to
1,202 patients after MI who had 10 or more ventricular premature
depolarizations per hour. During a 1.79-year follow-up period, resus-
citative VF or arrhythmic death was less in the patients treated with
amiodarone. The investigators concluded that amiodarone reduces
VF and arrhythmic death after MI with frequent or repetitive ven-
tricular premature complexes. The absolute risk reduction was
greater in patients with previous MI or CHF.
2. EMIAT. This trial randomly assigned 1,486 patients with ischemic
ventricular dysfunction after MI who had an EF of less than 40%.
During a 21-month follow-up period, all-cause and cardiac mortal-
ity rates were similar in both groups. A 35% decrease in arrhyth-
mic death occurred in the patients who received amiodarone. Thus,
amiodarone therapy in this group of patients was not supported by
the study.
The conclusions of these trials differ, as did their entry trials.
CAMIAT focused on ventricular ectopy, whereas EMIAT focused
on ventricular dysfunction. Clearly, patients in EMIAT were
more sick and had a substantially greater overall mortality rate.
Interestingly, the details of CAMIAT indicate that, as in EMIAT,
the all-cause mortality rate was unaffected by amiodarone. The
ultimate message from these trials is ambiguous, but amiodarone
at the very least was shown to be no worse than placebo, supporting
Table 34.1. Recommendations for Routine Laboratory
Testing With Long-Term Amiodarone Therapy
Test Time Frame
Thyroid function (TSH and T4) Baseline at 3 mo and every 6 mo
thereafter
Liver function Baseline and every 6 mo
Chest radiograph Baseline and yearly
Pulmonary function Baseline and when symptoms
develop or change
Ophthalmic evaluation Baseline and when visual
symptoms develop or change
Abbreviations: T4, thyroxine; TSH, thyrotropin.
368 III Electrophysiology

at minimum its use in patients who require antiarrhythmic therapy
for other reasons.
3. GESICA. An important trial that demonstrated the relative safety
of amiodarone therapy in patients with impaired LV function and
CHF.
4. CHF STAT. Also an important trial that showed the relative safety of
amiodarone in patients with impaired LV function and CHF.
• Amiodarone is a useful drug in the suppression of symptom-
atic atrial and ventricular arrhythmias, but it is not a substitute
for implantable cardioverter-defibrillator therapy in patients
deemed at risk for sudden cardiac arrest.
Dronedarone (Class III)
Dronedarone is a recently approved drug for patients with atrial
fibrillation and flutter in whom amiodarone therapy has either
failed to be effective or been intolerable. Unlike amiodarone,
dronedarone is not currently approved for treatment of ven-
tricular arrhythmias. Based on current trial data, the efficacy of
dronedarone in patients with atrial fibrillation is superior to pla-
cebo but appears inferior to amiodarone and sotalol.
Proarrhythmia
The proarrhythmic potential of dronedarone is low and is similar
to amiodarone.
Hemodynamics
The ANDROMEDA study, which included patients with
impaired LV function and CHF, reported that dronedarone was
associated with an approximate doubling of mortality rate in
patients with moderate to severe CHF, although the precise mech-
anism is unknown (Figure 34.3). In view of these findings, drone-
darone is contraindicated (US Food and Drug Administration
Black Box Warning) in class IV CHF and in patients with recent
exacerbation of heart failure.
Adverse Events
Acute liver injury and failure requiring liver transplant have been
reported in patients receiving dronedarone therapy. Therefore,
in patients for whom dronedarone is prescribed, liver function
should be monitored, especially in the first 6 months of therapy.
Bradycardia and QT interval prolongation (class III effect), nau-
sea, rash, and increased creatinine concentration have also been
reported. Unlike amiodarone, dronedarone does not contain
iodine and therefore does not cause iodine-related adverse reac-
tion and adverse effects.
Important Clinical Trials
EURIDIS and ADONIS trials reported that when compared with
placebo, dronedarone was superior in maintaining sinus rhythm
after electrical, pharmacologic, or spontaneous conversion to
sinus rhythm from atrial fibrillation or flutter with no (short-
term) differences in lung and thyroid function (Figure 34.4).
The ATHENA trial reported that dronedarone was signifi-
cantly more effective than placebo in reducing the composite
end point of first hospitalization due to cardiovascular events or
death, with a significant reduction in the rate of cardiovascular
death but not in the rate of death from any cause. A post-hoc
analysis of the ATHENA trial also showed a significant reduc-
tion in the rate of stroke, although the clinical significance of this
is unclear.
• Dronedarone is contraindicated for patients with moderate to
severe heart failure (New York Heart Association class IV) or
class II or III heart failure with recent decompensation requiring
hospitalization.
• Dronedarone may be associated with liver injury and acute liver
failure.
Sotalol (Class III)
Sotalol is effective in treating atrial and ventricular arrhythmias.
Proarrhythmia
QT interval lengthening can result in VF or torsades de
pointes. Proarrhythmia rates of 3% to 5% have been reported.
Hypokalemia or hypomagnesia (ie, patients taking diuretics) and

All-Cause Mortality or
Hospitalization for
50 Worsening Heart Failure 50 All-Cause Mortality
Cumulative Incidence, %

Placebo
40 Dronedarone 40

30 30

20 20

10 10

0 0
0 30 60 90 120 150 180 210 0 30 60 90 120 150 180 210
Time, d Time, d
No. at Risk
Placebo 317 234 159 87 41 16 6 1 317 256 181 103 50 18 6 1
Dronedarone 310 232 151 87 49 19 4 1 310 257 174 104 59 22 5 1

Figure 34.3. ANDROMEDA Results. Dronedarone was associated with a significant increase in the rate of death among patients with moderate
to severe congestive heart failure and is contraindicated. (Previously published. See “Credit Lines” section.)
 34 Antiarrhythmic Drugs 369

100 nonvascular selective calcium channel blockers (ie, verapamil

Cumulative Incidence, %
Placebo
Dronedarone and diltiazem).
80
Adverse Effects
60
Most adverse effects are related to β-blocking action. Of these,
40 fatigue is most prominent. Clinically important bradycardia may
be seen.
20 HR, 0.75 (95% CI, 0.65-0.87)
P<.001 Dofetilide (Class III)
0
Dofetilide is a pure IKr blocker with no other ion channel effects.
0 60 120 180 270 360
It is approved for treatment of atrial fibrillation and flutter (it is
Time, d weakly active against ventricular arrhythmias) (Figure 34.6).
No. at Risk
Placebo 409 192 156 133 112 90
Dronedarone 828 450 389 347 307 262 Proarrhythmia

Figure 34.4. EURIDIS and ADONIS Results. Composite inci-
dence of atrial fibrillation in patients treated with dronedarone or pla-
cebo. When compared with placebo, dronedarone was associated with
a decrease in the incidence of atrial fibrillation. HR indicates hazard
ratio. (Previously published. See “Credit Lines” section.)
use of greater dose ranges (>160 mg twice daily) are associated
with increased risk of proarrhythmia.
The SWORD trial evaluated d-sotalol (a pure IKr potassium
channel blocker with no β-blocking action) in 3,121 patients
with LV dysfunction and recent MI or recent heart failure and a
remote MI. d-Sotalol was associated with an increased mortality
rate compared with placebo (Figure 34.5).
Hemodynamics
Sotalol has potent β-adrenergic blocking effect and so may cause
bradycardia and worsening of LV dysfunction.
Dofetilide is contraindicated in patients with a creatinine clear-
ance of less than 20 mL/min. Inpatient initiation of dofetilide is
mandatory. Risk of proarrhythmia is reduced when the dofetilide
dose is adjusted appropriately for patients with renal insuffi-
ciency. If creatinine clearance is 20 to 40 mL/min, the starting
dose of dofetilide is 125 mcg twice daily. If the creatinine clear-
ance is 40 to 60 mL/min, the starting dose is 250 mcg twice daily.
If creatinine clearance is greater than 60 mL/min, the starting
dose is 500 mcg twice daily. Dose adjustments are required when
the corrected QT interval increases by more than 15% after the
first dose.
Hemodynamics
Dofetilide has minimal or no effect on cardiac output, stroke vol-
ume, or systemic vascular resistance. Importantly, studies of this
drug in patients with CHF have shown no increase in signs or
symptoms of heart failure.

Important Drug Interactions Important Drug Interactions

The β-blocking action of sotalol may exacerbate the effects Several important drug interactions occur with dofetilide.
of other drugs that have AV nodal inhibitory effects, such as Concomitant use of cimetadine, verapamil, fluconazole, and
trimethoprim (alone or with sulfamethoxazole) is contraindi-
1.00 cated since each drug increases the dofetilide level. Amiodarone
Proportion Event Free

0.98

0.96 1.0
Possibility of Remaining

0.94 0.8
in Sinus Rhythm

0.92 Placebo
d-Sotalol 0.6
0.90
Z=-2.75, P=.006 0.4
0.88
0 60 120 180 240 300 0.2
Placebo (n=69)
Time From Randomization, d
Dofetilide (n=119)
No. at Risk 0.0
Placebo 1,572 1,170 874 551 330
0 12 24 36
d-Sotalol 1,549 1,150 844 544 323
Time, mo
Figure 34.5. SWORD Trial Results. In patients after myocar-
dial infarction, d-sotalol (no β-blocking effect) was associated with Figure 34.6. DIAMOND Trial Results. The trial showed that
an increased mortality rate compared with placebo. It is not clear dofetilide had efficacy in treating patients with atrial fibrillation when
whether the worse outcome associated with d-sotalol was due to lack of compared with placebo. (Previously published. See “Credit Lines”
β-blocking effect. (Previously published. See “Credit Lines” section.) section.)
370 III Electrophysiology

1.0 rhythm. Typically, ibutilide is administered as a 1-mg bolus and

Probability of Survival repeated at 10 minutes if cardioversion has not occurred before
0.8 electrical cardioversion, if necessary.

0.6 Proarrhythmia

0.4 Torsades de pointes is most common with impaired ventricular

function and prolonged baseline QT interval.
0.2 Placebo • Because of the risk of delayed torsades de pointes, patients given
Dofetilide ibutilide should be monitored for a minimum of 24 hours.
0.0
0 12 24 36
Adenosine (Unclassified)
Time, mo
No. at Risk Adenosine is an endogenous compound that transiently blocks
Dofetilide 762 564 214 6 the AV node. It is used for therapeutic or diagnostic purposes,
Placebo 759 536 199 1 or both.
Figure 34.7. DIAMOND Trial Results. Dofetilide had a neutral
effect on mortality rate among patients with prior myocardial infarc- Proarrhythmia
tion or congestive heart failure. No other antiarrhythmic drug has been
Atrial fibrillation (that can be sustained) may occur after admin-
shown to have a neutral effect on mortality rate. Error bars indicate 95%
confidence interval. (Previously published. See “Credit Lines” section.)
istration of adenosine because of its effect on decreasing atrial
refractoriness, providing a potential for microreentry in the
atria. This result may be a problem in patients with bypass
tract–mediated, narrow complex supraventricular tachycardia
therapy must be stopped for 3 months before initiation of converted to preexcited atrial fibrillation. Sinus bradycardia and
dofetilide (or the plasma amiodarone level must be less than 0.03 sinus arrest may also occur. Ventricular proarrhythmia usually
mcg/mL). consists of VPCs and nonsustained VT, but VF may also occur.

Important Clinical Trials Hemodynamics

On the basis of the results of the DIAMOND study, dofetilide
does not affect the mortality rate of patients with prior MI and
LV dysfunction. As a result of these findings, dofetilide is often
used in the management of atrial fibrillation in patients with LV
dysfunction (Figure 34.7).
• Concomitant use of dofetilide and cimetidine, verapamil, flucon-
azole, or trimethoprim (alone or with sulfamethoxazole) is con-
traindicated since each of these agents increases the tissue levels
of dofetilide.
• Uniquely, dofetilide is associated with a neutral effect on mortality
rate in patients with prior MI or CHF.
Ibutilide (Class III)
Ibutilide is available only as an IV preparation and is used mostly
to achieve chemical, in addition to electrical, cardioversion in
patients with “resistant atrial fibrillation and flutter” to sinus
The major advantage of adenosine is its short duration of action
(1–5 seconds) and minimal hemodynamic effects (ie, sinus
tachycardia, increased systolic arterial pressure, and decreased
diastolic arterial pressure). Direct vasodilator effect is offset by
sympathetic excitation through its actions on chemoreceptors.
Hyperventilation occurs because of chemoreceptor stimulation.
• Adenosine effect is augmented in patients taking dipyridamole
and is diminished in those taking theophylline compounds. This
response reflects both inhibition of adenosine uptake by dipyrida-
mole and block of adenosine receptors by theophylline.
• Dennervated hearts are sensitive to the electrophysiologic effects
of adenosine.
Adverse Effects
Chest pain and dyspnea are common during the administration
of adenosine.

Table 34.2. Drug Choice for Maintenance of Sinus Rhythm in Patients With Atrial
Fibrillation
Drug Therapy Lone Atrial Fibrillation CHF, CAD CAD (Normal EF) Renal Failure
First line Flecainide Dofetilide Sotalol Amiodarone
Propafenone Amiodarone Dronedarone
Second line Sotalol Propafenone
Procainamide
Disopyramide
Amiodarone
Avoid Flecainide Flecainide Sotalol
Propafenone Propafenone Procainamide
Dronedarone Dofetilide
Abbreviations: CAD, coronary artery disease; CHF, congestive heart failure; EF, ejection fraction.
 34 Antiarrhythmic Drugs 371

In-Hospital Versus Outpatient Initiation In most cases, antiarrhythmic drug therapy should be started
of Antiarrhythmic Drugs at the lowest possible dose and titrated upward and its therapeu-
tic efficacy should be monitored with reference to PR interval
The decision to initiate antiarrhythmic drug therapy in hospital
(flecainide, propafenone, sotalol, and amiodarone) and to QRS
versus in the outpatient setting is determined largely by the like-
(flecainide and propafenone) and QT intervals (sotalol, amio-
lihood of risk versus cost and the inconvenience of inpatient drug
darone, and disopyramide) at rest (sotalol) or with exercise
loading. In the case of dofetilide, in-hospital initiation is man-
(class IC agents). Measurement of serum drug levels is rarely
dated. Unfortunately, few prospective data are available regard-
helpful.
ing the safety of outpatient initiation of antiarrhythmic drug
therapy. In the case of atrial fibrillation, outpatient loading should
be avoided in patients with symptomatic sinus node dysfunction,
AV conduction disturbance, bundle branch block, structural Abbreviations
heart disease, and QT interval prolongation (Table 34.2). The AV arterioventricular
most important risk of antiarrhythmic therapy is that of proar- CHF congestive heart failure
rhythmia, most commonly due to torsades de pointes, but also CNS central nervous system
of virtually any other arrhythmia, such as VT and atrial arrhyth- ECG electrocardiography
mias (Box 34.1). In general, the following statements are true: EF ejection fraction
GI gastrointestinal
• In the absence of notable bradycardia and with normal ventricular IKr rapidly activating potassium current
function and QRS and QT intervals, proarrhythmia risk is low. IV intravenous
LV left ventricular
• Unsuspected sinus node dysfunction that is suppressed by the anti-
MI myocardial infarction
arrhythmic agent may lead to prolonged sinus node recovery with
VF ventricular fibrillation
termination of AF, leading to possible syncope.
VPC ventricular premature contraction
• Propafenone and flecainide also worsen AV node and His-Purkinje VT ventricular tachycardia
conduction; thus, if outpatient use is contemplated, initial cardio-
version in hospital is prudent.
• For sotalol, outpatient initiation is probably safe when the baseline
QT interval is less than 450 msec in the absence of renal dysfunc- Names of Clinical Trials
tion and risk factors for torsades de pointes. ADONIS American-Australian-African Trial With
• Amiodarone is generally considered safe to administer as an out- Dronedarone in Atrial Fibrillation or Flutter
patient therapy. Risk of proarrhythmia is low. Patients for the Maintenance of Sinus Rhythm
• In-hospital administration of dofetilide is mandatory. ANDROMEDA Antiarrhythmic Trial With Dronedarone in
Moderate to Severe Congestive Heart Failure
Evaluating Morbidity Decrease
ATHENA A Placebo-Controlled, Double-Blind, Parallel
Box 34.1. Risk Factors for Torsades de Pointes Arm Trial to Assess the Efficacy of Dronedarone
400 mg bid for the Prevention of Cardiovascular
Patient factors
Hospitalization or Death From any Cause in Patients
Female sex With Atrial Fibrillation/Atrial Flutter
CAMIAT Canadian Amiodarone Myocardial Infarction
Bradycardia, especially pauses associated with Arrhythmia Trial
termination of atrial fibrillation CAST Cardiac Arrhythmia Suppression Trial
CHF STAT Congestive Heart Failure: Survival Trial of
Hypokalemia, hypomagnesemia
Antiarrhythmic Therapy
Baseline QT interval prolongation DIAMOND Danish Investigations of Arrhythmia and Mortality
on Dofetilide
Drug factors
EMIAT European Myocardial Infarct Amiodarone Trial
Concomitant use of drugs that prolong QT interval EURIDIS European Trial in Atrial Fibrillation or Flutter
or interfere with metabolism of QT prolonging agent Patients Receiving Dronedarone for the Maintenance
of Sinus Rhythm
Renal impairment of renally excreted agents that GESICA Grupo de Estudio de la Sobrevida en la Insuficiencia
prolong QT interval (eg, sotalol, procainamide) Cardiaca en Argentina
SWORD Survival With Oral d-Sotalol

Intracardiac Electrophysiology Tracings
MALINI MADHAVAN, MBBS, CHRISTOPHER J. MCLEOD, MB, CHB, PHD,
DOUGLAS L. PACKER, MD, and SAMUEL J. ASIRVATHAM, MD
The field of analysis and interpretation of intracardiac EGMs is
fascinating but complex and often intimidates the general cardio-
vascular trainee. For the cardiovascular board examination, how-
ever, a very limited and fundamental knowledge of intracardiac
EGMs is expected. In this chapter, the normal intracardiac EGM
sequence is introduced and briefly explained; tracings illustrate
the main concepts that may be tested on the cardiovascular board
examination.
Intracardiac EGMs are an extension of the surface 12-lead
ECG. The more familiar a student is with analyzing 12-lead ECGs
obtained during slow and fast rhythms, the easier it will be to
understand invasive tracings. In some ways, intracardiac EGMs
are easier to interpret than ECGs. For example, for tachyarrhyth-
mia analysis on ECG, it can be difficult to know where the P
wave is located (sometimes buried within the QRS) and, without
this interpretation, the mechanism of arrhythmia cannot be diag-
nosed. However, with invasive tracings, as will be explained, the
cardiologist knows whether and when atrial activation is occur-
ring by simply looking at whether an EGM is being recorded by
the atrial catheter typically placed in the right atrium.
In preparing for the boards, the trainee should be familiar
with recognizing atrial and ventricular EGMs and especially
recognizing the His bundle EGMs recorded on a catheter placed
in the anatomic region of the His bundle. Additional concepts
include the following:
• Diagnosing whether AV block is either suprahisian or infrahisian
• Recognizing and determining the significance of dual AV node
physiology
• Recognizing whether there are more ventricular EGMs than atrial
EGMs or vice versa and interpreting the significance
• Distinguishing AV node reentry from accessory pathway–mediated
orthodromic reciprocating tachycardia
Abbreviations and acronyms are expanded at the end of this chapter.
372
35
• Understanding the difference between antidromic and orthodro-
mic reciprocating tachycardia
In essentially all electrophysiology tracings, at least 1 surface
ECG will be displayed. Typically, there are 3, and Figure 35.1
shows leads I, II, and V1. The student should start by looking at
the ECG. In Figure 35.1, sinus rhythm with right bundle branch
block will be noted.
Anywhere from 3 to more than 12 intracardiac catheters with
displayed EGMs can be obtained in an actual clinical study,
depending on the complexity of the arrhythmia. In all cases,
however, 3 basic catheters will be displayed, and the student
should be familiar with the typical tracing obtained:
High right atrial catheter: This is a recording catheter in
the atrium, typically high in the right atrium, often near the
superior vena cava-right atrium junction. Anatomically,
the sinus node is located at the cranial end of the sulcus
terminalis, epicardially near the superior vena cava-right
atrium junction. Thus, this electrode will show the earliest
atrial activation in sinus rhythm.
Right ventricular apical catheter: This electrode is
placed in the right ventricle near the region of the right
ventricular apex. A ventricular EGM will be recorded on
this electrode.
His bundle electrogram catheter: His bundle electrodes
are placed in the anatomic region of the His bundle on the
superior septal tricuspid annulus.
For analyzing the basic tracings, the following approach should
be used:
1. Identify the ventricular EGMs. The QRS should be identified on the
surface ECG. Mentally, a line should be dropped down from the sur-
face ECG QRSs to the EGMs being recorded. EGMs that generally
line up with the QRS will be ventricular EGMs. On the right ven-
tricular apical catheter, as expected, only a ventricular EGM is seen.
 35 Intracardiac Electrophysiology Tracings
II
V1
RVA
HRA
HBE2
AH=75 ms HV=42 ms
Figure 35.1. Normal Intracardiac Electrogram Sequence. (See end of chapter for abbreviations used in this figure.)
On the His bundle recording, there are 3 EGMs, but only 1 lines up
with the QRS and the right ventricular EGM. This is the recording
of the ventricular EGM near the region of the His bundle.
2. The next step is to identify the atrial EGMs. The atrial EGMs will
not line up with the surface QRS; in this case, the atrial EGM clearly
precedes the ventricular EGMs and is being recorded on the right
atrial catheter.
3. Once the ventricular and atrial EGMs have been recognized, then the
His bundle catheter tracings should be analyzed. Both an atrial and a
ventricular EGM are noted (atrial EGM lines up with the high right
atrial signal and the P wave; ventricular EGM lines up with the ventric-
ular EGM on the right ventricular apical catheter and QRS). Between
these 2 EGMs, a sharp deflection, the His bundle recording, is noted.
The student should be thoroughly familiar with this normal
sequence.
Two important intervals are measured. The first is the AH
interval, which is measured between the atrial EGM and the His
bundle recording. This represents conduction time primarily
through the AV node (normal, 45–125 milliseconds). The second
is the HV interval, which is measured from the His bundle EGM
to the beginning of ventricular activation recorded anywhere
(often the start of the QRS on the surface ECG). The HV inter-
val is a measure of infrahisian conduction (His-Purkinje system).
The normal value is 35 to 55 milliseconds. The AH interval and
the HV interval are both normal in Figure 35.1.
• Be able to recognize atrial, ventricular, and His bundle EGMs.
• Be able to measure and understand the significance of the AH and
HV intervals.
Localizing the Site of AV Block
For the cardiovascular boards, it is necessary to distinguish
between suprahisian (AV node level) block and infrahisian block
on the surface ECG. Typically, Mobitz type I AV block is supra-
hisian, and Mobitz type II AV block is infrahisian. In some cases,
373
V
A
1,000 ms
an electrophysiologic study is done to be sure of the level of block
because pacemaker implantation is usually indicated for infra-
hisian block.
In Figure 35.2, the surface ECGs, high right atrial, right ven-
tricular, and His bundle tracings are seen. In addition, the arterial
blood pressure recording is shown. Once the atrial, ventricular,
and His bundle tracings are recognized, as explained above,
the AH interval prolongs until there is AV block. Note that an
atrial EGM is present but there is no ventricular EGM. This is
the equivalent of an ECG with no QRS complex. Importantly,
when the QRS complex is not seen, there is absence of not only
the ventricular EGM but also the His bundle EGM. This find-
ing indicates that the electrical wavefront did not reach the His
bundle. Thus, the level of block is within the AV node.
Contrast Figure 35.2 with Figures 35.3 and 35.4. Again, the
student should recognize the His bundle tracing. In Figure 35.3,
the AH interval is normal but the HV interval is prolonged. In
Figure 35.4, there is high-grade AV block during atrial pacing
(see the pacing spikes before the atrial EGMs and P wave). A His
bundle tracing is seen clearly, but no QRS complex or ventricular
EGMs are seen for several of the beats. Thus, the electrical wave-
front did reach the His bundle but then is blocked downstream to
the ventricle (infrahisian block).
Distinguish between suprahisian and infrahisian block:
• If with blocked beat a His bundle EGM is seen (but no ventricular
EGM), the block is infrahisian.
• If with blocked beat only an atrial EGM is seen (no His bundle or
ventricular EGM), the block is suprahisian.
Sinus Node Dysfunction and
Evaluation of Syncope
The most common way of diagnosing whether a person with
symptoms has considerable sinus node dysfunction is noninvasive
374 III Electrophysiology

II

aVF

V1

HRA

AH AH AV node block

HBE2

RVA

BP

1,000 ms

Figure 35.2. Suprahisian Block. The atria are paced (dashed arrows). (See end of chapter for abbreviations used in this figure.)

extended rhythm monitoring (eg, loop recorder, implanted loop
recorder). Occasionally, however, electrophysiologic study is
done to assess sinus node dysfunction.
In Figure 35.5, pacing is being done from the high right atrium
(arrows). Note that there is Wenckebach block (prolongation of
the AH interval). In addition, however, the student should note
that when pacing stops, there is a pause before the sinus node
recovers (hatched arrow). This pause is called the sinus node
recovery time. Thus, overdrive suppression of the sinus node is
being done with the atrial pacing, and the time taken for the
sinus node to recover is being measured. The sinus node func-
tion, however, depends on the resting sinus node rate. Naturally,

II

aVF

V1

HRA

HBE1

AH=75 ms HV=85 ms

RVA

1,000 ms

Figure 35.3. Infrahisian Disease. HV is 85 milliseconds; normal is 35 to 55. (See end of chapter for abbreviations used in this figure.)
 35 Intracardiac Electrophysiology Tracings
II
aVF
V1
HRA
HBE1
RVOT
Figure 35.4. Infrahisian Block. The atria are paced. (See end of chapter for abbreviations used in this figure.)
when there is sinus bradycardia to begin with, the recovery time
will also be longer. Thus, the more clinically relevant interval
is the corrected sinus node recovery time. The corrected time
is obtained by subtracting the sinus cycle length from the sinus
node recovery time before starting atrial pacing. For exam-
ple, in Figure 35.5, the sinus node recovery time was 1,700
milliseconds. However, the sinus cycle length before atrial pac-
ing commenced was 850 milliseconds. Thus, the corrected sinus
node recovery time is 850 milliseconds. If the corrected time is
more than 500 milliseconds, sinus node dysfunction is likely.
Figure 35.6 was obtained during carotid sinus massage. Note
the prolonged pause. No atrial or ventricular EGMs are being
recorded. Therefore, this is a sinus pause (cardioinhibitory
response). Note that when the heart rate picks up (spontaneous
or paced), the blood pressure still remains low (vasodepressor
response). Often patients do have a mixed response (vasodepres-
sor and cardioinhibitory).
Analyzing Tachycardia: Are There
More Ventricular or Atrial EGMs,
or Is the Number Equal?
The 12-lead ECG in Figure 35.7 shows a wide complex tachy-
cardia. The differential diagnosis is ventricular tachycardia,
a supraventricular tachycardia with bundle branch block, or a
supraventricular tachycardia with antegrade accessory pathway
conduction. The diagnosis, however, is easily made when ana-
lyzing the intracardiac tracing. Note that there is a ventricular
EGM for each QRS complex; however, an atrial EGM is seen
for only some of the QRS complexes. Thus, there are more ven-
tricular than atrial EGMs, a finding diagnostic of ventricular
tachycardia.
• During tachycardia:
1. If there are more ventricular than atrial EGMs: ventricular
tachycardia.
375
1,000 ms
2. If there are more atrial than ventricular EGMs: likely atrial
tachycardia.
3. If there are equal numbers of atrial and ventricular EGMs: may be
any supraventricular tachycardia with 1:1 AV conduction or ventric-
ular tachycardia with retrograde 1:1 ventriculoatrial conduction.
Another example of wide complex tachycardia is shown in
Figure 35.8. Note that there are ventricular and atrial EGMs in
all the beats. Thus, there is 1:1 AV relationship. In this instance,
neither supraventricular tachycardia with 1:1 AV conduction nor
ventricular tachycardia with 1:1 retrograde ventriculoatrial con-
duction can be completely excluded. However, this tracing likely
represents AV node reentry because this is the only arrhythmia
that frequently produces near-simultaneous AV activation (SEE
THE CHAPTER “SUPRAVENTRICULAR TACHYCARDIA”).
• In AV node reentrant tachycardia, simultaneous or near-
simultaneous ventricular and atrial EGMs are being recorded.
• The ECG equivalent is the P wave being buried in or very closely
associated with the QRS complex.
Dual AV Node Physiology and Induction
of AV Node Reentrant Tachycardia
AV node reentry is a common supraventricular arrhythmia.
In this condition, there are 2 atrial myocardial connections to
the AV node: 1 called the fast pathway located superiorly just
behind the tendon of Todaro, and 1 called the slow pathway
located inferiorly near the coronary sinus ostium (SEE THE CHAP-
TER “SUPRAVENTRICULAR TACHYCARDIA”). During atrial pacing, as
one paces faster, there is always decrement in the AV node, as
evidenced on the surface ECG by prolongation of the PR inter-
val and on the intracardiac tracings by prolongation of the AH
interval. However, this AH prolongation with pacing faster is a
gradual, continuous process. In patients with dual AV node phys-
iology, there is an abrupt increase in the AH interval, despite
376 III Electrophysiology

II

aVF

V1

HRA

HBE2

SNRT=1,700 ms
CSRT=850 ms
RVA

BP

1,000 ms

Figure 35.5. Testing of Sinus Node Function. Atrial pacing (black arrows); spontaneous return of atrial activity (red dashed arrow). (See end of
chapter for abbreviations used in this figure.)

only a minimal decrease (going faster) in the atrial cycle length.
This phenomenon is illustrated in Figures 35.9.
In Figure 35.9A, atrial pacing (S1) is being performed at a
cycle length of 600 milliseconds, and then an atrial extra stim-
ulus (S2) is being placed at a coupling interval of 350 milli-
seconds. As expected, the AH interval, as a result of the extra
stimulus, prolongs (A2H2 130 milliseconds). In Figure 35.9B,
the extra stimulus coupling interval has been shortened by just
10 milliseconds (now 320 milliseconds). However, there is an
abrupt jump in the A2H2 interval to 230 milliseconds. This find-
ing is diagnostic of dual AV node physiology.
Dual AV node physiology is defined as an increase in the
A2H2 interval (or the H1H2 interval between the His bundle
EGM during pacing and the extra stimulus) by 50 milliseconds
or more when the atrial coupling interval has decreased by 10
milliseconds or less.

Figure 35.6. Carotid Sinus Massage. (See end of chapter for abbreviations used in this figure.)
 35 Intracardiac Electrophysiology Tracings 377

Figure 35.7. Wide Complex Tachycardia. (See end of chapter for abbreviations used in this figure.)

• Patients with AV node reentrant tachycardia have both a fast and a
slow pathway input to the AV node.
• Dual AV node physiology is diagnosed when there is an abrupt
increase in the AH interval by 50 milliseconds or more when the
atrial coupling interval has decreased by 10 milliseconds or less.
In Figure 35.9B, after the jump to the slow pathway has
occurred, a few beats of AV node reentrant tachycardia were
initiated.
Coronary Sinus EGMs and AV
Reentrant Tachycardia
In Figure 35.10, the surface ECG, the right ventricular, right
atrial, His bundle, and the proximal and distal coronary sinus
EGMs are shown. The coronary sinus catheter serves as a sur-
rogate for the left atrial and left ventricular electrical activation
in the region of the mitral annulus. The figure shows a narrow
complex tachycardia with cycle length of 290 milliseconds.

II

V1

RVA

HRA

H H H H
HBE

PCS

MCS

DCS

1,000 ms

Figure 35.8. Wide Complex Tachycardia. (See end of chapter for abbreviations used in this figure.)
378 III Electrophysiology

V1

HRA S1 600 S1 330 S2

A1 A1 A2 H2 A2H2=130
H1 H1
HBE2

H1H2=380

RVA

1,000 ms

V1

S1 600 S1 320 S2
HRA

A1 A1 A2 H2 A2H2=230
H1 H1

HBE2

H1H2=475

RVA

1,000 ms

Figure 35.9A and B. Dual Atrioventricular Node Physiology. (See end of chapter for abbreviations used in this figure.)

The differential diagnosis based on the ECG is an atrial
tachycardia, AV node reentrant tachycardia, or AV reentrant
tachycardia using a retrograde conducting accessory bypass
tract (SEE THE CHAPTER “SUPRAVENTRICULAR TACHYCARDIA”). On
careful scrutiny of the ECG, what looks like a P wave is seen
possibly just following the QRS complex. This short RP tachy-
cardia favors AV reentrant tachycardia or AV node reentrant
tachycardia.
When the intracardiac EGMs are evaluated, however, the
diagnosis becomes much more straightforward. The high right
atrial catheter clearly shows atrial signals. The atrial signals (A)
are also seen on the His bundle recording catheter and on the
coronary sinus electrodes.
Because the AV node is located near the His bundle, when
retrograde atrial activation is via the AV node (as in AV node
reentrant tachycardia), the atrial EGM (A) should be earliest on
 35 Intracardiac Electrophysiology Tracings
II
V1
VV=290
RVA
A A
HRA
A H A H
HBE2
A A
PCS
A A
DCS
VA=75
Figure 35.10. Orthodromic Reciprocating Tachycardia Using a Left-Sided Accessory Pathway (See end of chapter for abbreviations used in
this figure.)
the His bundle catheter. In Figure 35.10, however, the earliest
atrial EGM is located in the coronary sinus catheter. Thus, the
left atrium is activated before the AV node region. This patient
has a left-sided accessory pathway that conducts in the retrograde
direction (from V to A). To follow the conduction pattern dur-
ing this tachycardia, therefore, antegrade conduction is through
the AV node (AHV), and retrograde conduction is through the
left-sided accessory pathway (earliest A on the coronary sinus)—
orthodromic reciprocating tachycardia using a left-sided acces-
sory pathway.
• Coronary sinus electrodes are used to map the left atrium and left
ventricle because the coronary sinus is along the mitral annulus.
• When the coronary sinus EGMs are earliest during narrow com-
plex tachycardia or ventricular pacing, suspect a left-sided acces-
sory pathway.
Recognizing Antegrade Conduction
Through an Accessory Pathway
Antegrade conduction through an accessory pathway is shown
in Figure 35.11. Note that during sinus rhythm the ventricle is
activated in 2 ways: 1 through the normal AV node and 1 through
an accessory pathway (in this case in the left free wall). Because
of the accessory pathway conduction, the typical pattern of
Wolff-Parkinson-White syndrome is seen. There is a short PR
interval because conduction proceeds across the accessory path-
way, which does not show slow or decremental conduction, and
there is a delta wave because the early ventricular activation is
through the accessory pathway directly to the ventricle, rather
than through the conduction system.
In the inset of Figure 35.11, the accompanying intracardiac
EGMs are noted. The student should see that the ventricular
379
A A A A
A H A H A H
A A A
A A A
1,000 ms
EGM is earlier in the coronary sinus (mapping the left ventricle)
than the EGM of the His bundle catheter (located on the septum).
It is earlier because of conduction through the left-sided acces-
sory pathway. An ablation catheter has been placed on the mitral
annulus (often through transseptal access). On the ablation cath-
eter, both atrial and ventricular EGMs are recorded very close to
each other because there is direct conduction from the atrium to
the ventricle through the accessory pathway.
Figure 35.12 shows a regular wide complex tachycardia.
As noted above, this could represent ventricular tachycardia,
supraventricular tachycardia with right bundle branch block, or
tachycardia utilizing a left-sided accessory pathway.
Figure 35.13 shows the intracardiac tracings, which make it
considerably easier to diagnose the mechanism of arrhythmia.
Note that there is 1:1 AV association (for every QRS complex,
an atrial EGM is seen on the right atrial recording catheter and
coronary sinus recording catheter). The His bundle EGM, how-
ever, is seen between the ventricular and atrial EGMs (hatched
arrow). Remember that when there is antegrade conduction down
the AV node, the His bundle EGM will occur between the atrial
and ventricular EGMs. The earliest ventricular EGMs (arrows)
are noted in the coronary sinus catheter. Thus, the left ventricle
is activated first, consistent with activation, antegrade through a
left-sided accessory pathway. The student should trace the circuit
in his or her mind. Antegrade activation is from the atrium to
ventricle through a left-sided pathway (thus, the early ventricular
EGM in the coronary sinus and the wide QRS complexes). Then,
retrograde activation is from the ventricle to the His bundle
through the AV node and then finally to the atrium (His bundle
EGM between ventricular and atrial EGMs and the earliest atrial
EGMs seen on the His bundle catheter). This is the circuit of
antidromic tachycardia.
380 III Electrophysiology

HBE

CS

ABL

Figure 35.11. Antegrade Conduction Through a Left Free-Wall Pathway (Wolff-Parkinson-White Syndrome). (See end of chapter for abbrevia-
tions used in this figure.)

Contrast this pattern to intracardiac EGMs obtained during
orthodromic tachycardia using a left-sided accessory pathway
that conducted in the retrograde direction (Figure 35.14). First,
this is a narrow complex tachycardia since antegrade activa-
tion is through the AV node. Second, the His bundle EGM
(hatched arrow) is seen between the atrium and ventricular
EGMs, and, finally, the earliest atrial EGMs (arrow) are noted
in the mid coronary sinus (left-sided accessory pathway con-
ducting retrograde). Thus, the circuit is antegrade conduction
down the AV node, retrograde conduction through the acces-
sory pathway. This is the circuit of orthodromic reciprocating
tachycardia.

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

Figure 35.12. Twelve-Lead Electrocardiogram of Wide Complex Tachycardia.

 35 Intracardiac Electrophysiology Tracings 381

P1 ART
II
V1

RVp S
hRA T
HBE 2 C

HBE 1 C
CS 19, 20 C

CS 17, 18 C
CS 15, 16 C

CS 13, 14 C
CS 11, 12 C
CS 9, 10 C
CS 7, 8 C
CS 5, 6 C
CS 3, 4 C

CS 1, 2 C

Figure 35.13. Intracardiac Electrograms of Wide Complex Tachycardia Illustrated in Figure 35.12. (See end of chapter for abbreviations used in
this figure.)

P1 ART
II S
V1
RVa
HRA d
HIS 4
HIS 3
HIS 2
HIS 1
CS 19, 20
CS 17, 18
CS 15, 16
CS 13, 14
CS 11, 12
CS 9, 10
CS 7, 8
CS 5, 6
CS 3, 4
CS 1, 2 C

Figure 35.14. Left-sided Accessory Pathway With Retrograde Conduction. (See end of chapter for abbreviations used in this figure.)
382 III Electrophysiology

• Antidromic tachycardia involves antegrade conduction through an ECG electrocardiogram

accessory pathway. EGM electrogram
HV interval interval between His bundle and ventricular EGM
• If the pathway is left-sided, intracardiac EGMs will show the ear-
liest ventricular EGM during tachycardia on the coronary sinus
electrodes.
Abbreviations Used in Figures
• Orthodromic reciprocating tachycardia is a circuit in which the
accessory pathway conducts retrograde from ventricle to atrium. A atrial electrogram
ABL ablation
• With a left-sided accessory pathway, the earliest atrial EGM will
AH interval between atrial electrogram and His bundle
be on the coronary sinus.
electrogram
AV atrioventricular
Summary BP blood pressure
This chapter reviewed the basic invasive EGMs with which the CS coronary sinus
CSRT corrected sinus node recovery time
student preparing for the cardiovascular boards should be famil-
DCS distal coronary sinus
iar. Details of differential diagnosis of the various arrhythmias H His bundle electrogram
and maneuvers performed in the electrophysiology laboratory to HBE His bundle electrogram
make an exact diagnosis before ablation are beyond the scope of HRA high right atrium
this chapter and are not material tested on the boards. This mate- HV interval between His bundle electrogram and ventricu-
rial, however, could serve as a basis for the student interested in the lar electrogram
bewildering but fascinating world of invasive electrophysiology. MCS midcoronary sinus
PCS proximal coronary sinus
RVA right ventricular apical
Abbreviations
RVOT right ventricular outflow tract
AH interval interval between atrial and His bundle EGM SNRT sinus node recovery time
AV atrioventricular V ventricular electrogram
 Section IV

Valvular Heart Disease

This page intentionally left blank
 36
Valvular Stenosisa
RICK A. NISHIMURA, MD
Aortic Stenosis
Definition and Causes
Aortic stenosis is a disease in which progressive obstruction to
left ventricular outflow results in the following: 1) pressure hyper-
trophy of the left ventricle; 2) symptoms of angina, dyspnea, and
syncope; and 3) if untreated, death. The presentation, diagnosis,
and eventual treatment of aortic stenosis depend on the cause and
severity of the outflow obstruction. The causes can be divided
into supravalvular aortic stenosis, fixed subvalvular aortic steno-
sis, and valvular aortic stenosis (Box 36.1). Valvular aortic steno-
sis has many causes, including congenital unicuspid or bicuspid
aortic valve, rheumatic heart disease, and senile degenerative
disease. Two-dimensional and Doppler echocardiography can be
used to determine reliably the level of obstruction and to assess
the severity of obstruction.
Supravalvular Aortic Stenosis
Supravalvular aortic stenosis is a congenital abnormality
in which the ascending aorta superior to the aortic valve is
a
Portions previously published in Bonow RO, Carabello BA, Chatterjee
K, de Leon AC Jr, Faxon DP, Freed MD, 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.
J Am Coll Cardiol. 2006 Aug 1;48(3):e1–148. Erratum in: J Am Coll
Cardiol. 2007 Mar 6;49(9):1014. Used with permission.
Abbreviations and acronyms are expanded at the end of this chapter.
narrowed. This is the rarest site of aortic stenosis. The stenosis is
seen as either a single discrete constriction or a long tubular nar-
rowing. Important associations with supravalvular aortic steno-
sis include elfin facies, hypercalcemia, and peripheral pulmonic
stenosis. The diagnosis of supravalvular aortic stenosis should be
suspected in a young patient who has a left ventricular outflow
murmur. Typically, a thrill is felt on palpation of the right carotid
artery but not the left carotid artery, because the obstructive jet
is directed toward the innominate artery. The diagnosis can be
made initially on the basis of 2-dimensional echocardiography
(which allows visualization of the narrowed ascending aorta) and
Doppler echocardiography (which provides information about
the magnitude of the obstruction). Magnetic resonance angiog-
raphy or computed tomography is required to show the extent
of narrowing of the ascending aorta if surgical intervention is
contemplated.
Subvalvular Aortic Stenosis
Discrete subvalvular stenosis is seen in approximately 10% of
all patients with aortic stenosis. It can be secondary to a subval-
vular ridge that extends into the left ventricular outflow tract or
to a tunnel-like narrowing of the outflow tract. The obstruction
is frequently accompanied by aortic regurgitation due to malfor-
mation of the aortic valve from the high-velocity jet emanating
from the subvalvular obstruction. The diagnosis can be made at
echocardiography by visualization of a narrowing or discrete
subvalvular ridge extending into the left ventricular outflow
tract and a high-velocity turbulence on continuous-wave Doppler
echocardiography. A discrete ridge may be difficult to visualize
directly in older patients, but it should be suspected when there
is high-velocity flow across the outflow tract in the presence of
a structurally normal aortic valve with early systolic closure. If
385
386 IV Valvular Heart Disease
Box 36.1. Causes of Aortic Stenosis
Supravalvular
Subvalvular
Discrete
Tunnel
Valvular
Congenital unicuspid or bicuspid aortic valve
(1–30 y old)
Bicuspid aortic valve (40–60 y old)
Rheumatic heart disease (40–60 y old)
Senile degenerative disease (>70 y old)
the site of obstruction is not visualized on the initial echocardi-
ogram, transesophageal echocardiography should be performed
to confirm the diagnosis.
The diagnosis of subvalvular aortic stenosis needs to be dif-
ferentiated from the dynamic outflow obstruction of hypertrophic
obstructive cardiomyopathy because treatment differs. Some sur-
geons recommend resection for treatment of discrete subvalvular
stenosis in all patients with at least moderate obstruction who are
candidates for operation, both to relieve the degree of left ventric-
ular outflow obstruction and to prevent progressive aortic regur-
gitation. The treatment of patients with hypertrophic obstructive
cardiomyopathy is discussed in CHAPTER 100 (“HYPERTROPHIC
CARDIOMYOPATHY”).
• Supravalvular aortic stenosis is associated with elfin facies, hyper-
calcemia, and peripheral pulmonic stenosis.
• Supravalvular aortic stenosis should be suspected when there is a
palpable thrill in the right carotid artery.
• Patients with discrete subvalvular aortic stenosis have a high
Doppler velocity across the aortic outflow tract and a structurally
normal aortic valve on 2-dimensional echocardiography.
• Consider supravalvular or subvalvular aortic stenosis in a young
person presenting with signs and symptoms of aortic stenosis but
with a normal aortic valve on echocardiography.
Valvular Aortic Stenosis
Obstruction at the valvular level accounts for most cases of aortic
stenosis. The cause of the valve abnormality depends on the age
at presentation (Box 36.1). The most common presentation today
is an elderly patient who has senile degeneration of the valve,
with calcific deposits at the base of the cusps in the absence of
commissural fusion.
• The most common cause of aortic stenosis is senile degenerative
changes.
• In patients with aortic stenosis due to rheumatic disease, “silent”
mitral stenosis should be ruled out.
• A bicuspid valve or rheumatic cause should be suspected in a patient
with aortic stenosis who presents in the sixth decade of life.
Pathophysiology
Progressive left ventricular outflow obstruction results in con-
centric pressure hypertrophy of the left ventricle caused by an
increase in left ventricular wall thickness. The increase in wall
thickness is a compensatory mechanism to “normalize” wall
stress. In most patients, the size of the left ventricular cavity
remains normal and systolic function is usually well preserved.
Failure of the left ventricle to compensate for the long-standing
pressure overload results in ventricular dilatation and a progres-
sive decrease in systolic function.
The pathophysiologic processes of aortic stenosis result from
the following: 1) an increase in afterload, 2) a decrease in sys-
temic and coronary blood flow from obstruction, and 3) progres-
sive hypertrophy. These mechanisms cause the classic symptom
triad of dyspnea, angina, and syncope. Exertional dyspnea is
common, even with normal systolic function. Abnormalities of
diastolic function are common in patients with aortic stenosis
and cause increased left ventricular filling pressures that are
reflected onto the pulmonary circulation. Diastolic dysfunction
occurs from prolonged ventricular relaxation and decreased
compliance and is caused by myocardial ischemia, a thick non-
compliant left ventricle, and increased afterload. Symptoms of
exertional angina may be present without epicardial coronary
artery obstruction. Myocardial ischemia results from a mismatch
in myocardial oxygen supply and demand due to high diastolic
pressure, a decreased myocardial perfusion gradient, and an
increased myocardial mass. The cause of exertional syncope is
multifactorial and may include ventricular arrhythmias, a sudden
decrease in systemic flow caused by the obstruction, or abnor-
mal vasodepressor reflexes caused by the high left ventricular
intracavitary pressure. As a progressive, long-standing pressure
overload is placed on the left ventricle, systolic decompensation
may occur from the afterload mismatch and lead to symptoms of
both left-sided and right-sided heart failure.
A “death spiral” may occur in patients with critical aortic
stenosis. With the onset of systemic hypotension (due to either
drugs or a vasovagal reaction), perfusion of the coronary arter-
ies may decrease. This increases the mismatch in myocardial
oxygen supply and demand, resulting in myocardial ischemia.
The myocardial ischemia, in turn, reduces forward cardiac out-
put. Aortic diastolic pressure decreases, further decreasing cor-
onary perfusion pressure. Unless immediate steps are taken to
increase perfusion pressure with pressors and fluids, progressive
hemodynamic deterioration and death may occur.
• Diastolic dysfunction is a primary pathophysiologic process in
patients with aortic stenosis.
• Myocardial ischemia occurs in patients with aortic stenosis even
if they have normal epicardial coronary arteries, because of a mis-
match in myocardial oxygen supply and demand.
• The most common type of syncope in patients with aortic stenosis
is vasodepressor syncope.
• Suspect critical aortic stenosis in a patient with syncope and an
aortic murmur.
Clinical Presentation
The clinical presentation of patients with aortic stenosis varies.
Some patients are asymptomatic, but a heart murmur is detected
on physical examination. Others have 1 or more symptoms from
the classic triad of exertional dyspnea, angina, and syncope.
Uncommonly, patients with end-stage aortic stenosis and con-
comitant left ventricular dysfunction present with anasarca and
cardiac cachexia. Albeit rare, sudden death may be the initial
manifestation of aortic stenosis.
On physical examination, a patient with aortic stenosis has
classic, characteristic findings. Severe aortic stenosis is diag-
nosed on the basis of a dampened upstroke of the carotid artery,
a sustained bifid left ventricular impulse, an absent A2, and
 36 Valvular Stenosis
a late-peaking systolic ejection murmur. A concomitant systolic
thrill indicates the presence of aortic stenosis (mean gradient
>50 mm Hg). In some patients, the systolic ejection murmur may
be heard with equal intensity at the apex and at the base. It is not
necessarily the intensity of the murmur that corresponds to the
severity of aortic stenosis but rather the timing of the peak and
the duration of the murmur. The murmur of aortic stenosis must
be differentiated from that of hypertrophic obstructive cardio-
myopathy or mitral regurgitation due to a flail posterior leaflet.
The distinction is based on the carotid pulse contour and on the
response of the murmur to maneuvers.
• Hypertrophic cardiomyopathy can be distinguished from aortic
stenosis on the basis of physical examination findings.
Laboratory Tests
Electrocardiography and Radiography
Electrocardiography usually shows normal sinus rhythm with
left ventricular hypertrophy. If atrial fibrillation is present, con-
comitant mitral valve disease or thyroid disease must be sus-
pected. Chest radiography shows left ventricular predominance,
with dilatation of the ascending aorta. Aortic calcification is
frequently seen on lateral chest radiographs.
Echocardiography
Two-dimensional and Doppler echocardiography are the imaging
modalities of choice for diagnosing aortic stenosis and estimat-
ing its severity. The location of the obstruction (supravalvular,
valvular, or subvalvular) can be identified with 2-dimensional
echocardiography. In patients with valvular aortic stenosis, the
cause (bicuspid aortic valve, rheumatic heart disease, or senile
degenerative disease) may be assessed from the parasternal
short-axis view. In patients with a bicuspid aortic valve, echocar-
diography should be used to assess for dilatation of the aorta and
the presence of a coarctation. Although the presence or absence
of aortic stenosis is readily diagnosed on 2-dimensional echocar-
diography, the severity of the stenosis cannot be judged on the
basis of the 2-dimensional echocardiographic image alone.
Doppler echocardiography is excellent for assessing the sever-
ity of aortic stenosis. With use of the modified Bernoulli equa-
tion (ΔP = 4v2, where ΔP is the pressure difference across the
valve and v is the velocity of the blood flow at the valve), a maxi-
mal instantaneous and mean aortic valve gradient can usually be
derived from the continuous wave Doppler velocity across the
aortic valve. However, accurate measurement of the aortic valve
gradient requires a detailed, meticulous study with multiple sites
of interrogation to ensure that the Doppler beam is parallel to
the stenotic jet. In laboratories with experienced echocardiog-
raphers, the Doppler-derived aortic valve gradients are accurate
and reproducible and correlate well with those obtained with car-
diac catheterization. The mean gradient from the integral of the
aortic valve velocity curve should be used to determine the sever-
ity of aortic stenosis. If the mean gradient is greater than 40 mm
Hg, severe aortic stenosis can be diagnosed with certainty in the
absence of severe anemia or associated subvalvular stenosis. In
a patient with clinical findings of severe aortic stenosis and a
Doppler-derived mean gradient greater than 40 mm Hg, no other
hemodynamic information is needed to assess the severity of
stenosis.
Aortic valve gradients depend not only on the severity of
obstruction but also on flow. In patients with low cardiac out-
put, the stenosis may still be severe, with mean gradients less
387
than 40 mm Hg. To overcome these problems, an AVA has been
derived with use of the Gorlin formula. In the cardiac catheter-
ization laboratory, the AVA is calculated from the pressure gradi-
ent and an independent measure of cardiac output:
1, 000 × CO
AVA =
44 × SEP HR ΔP
where CO = cardiac output, HR = heart rate, ΔP = pressure dif-
ference across the valve, and SEP = systolic ejection period.
Two-dimensional and Doppler echocardiography can also
provide reliable estimations of AVA by the continuity equation:
Tarea × LVOT
LVOT TTVI
AVA =
AVVTVI
where AV = aortic valve flow velocity, LVOT = left ventricular
outflow tract, and TVI = time-velocity integral.
Severe aortic stenosis can be diagnosed if a patient has clinical
findings consistent with severe aortic stenosis, a mean gradient
greater than 40 mm Hg, and AVA less than 1.0 cm2 (Table 36.1).
There are limitations in using Doppler echocardiography
to estimate the severity of aortic stenosis. The biggest prob-
lem occurs when the Doppler beam is not parallel to the aortic
stenosis velocity jet, because the mean gradient will be under-
estimated. Thus, in a patient with the clinical features of severe
aortic stenosis but echocardiographic and Doppler findings of
mild or moderate stenosis, further evaluation with either another
Doppler echocardiographic study or cardiac catheterization is
required. Doppler echocardiography will not overestimate the
mean gradient except in rare instances of severe anemia (hemo-
globin <8.0 g/dL), a small aortic root, or sequential stenoses in
parallel (coexistent left ventricular outflow tract and valvular
obstruction). The calculation of AVA with echocardiography is
highly dependent on accurate measurement of the diameter of the
left ventricular outflow tract. Thus, special attention is required
when diagnosing severe aortic stenosis in patients with small
valve areas but relatively low mean gradients. In these instances,
correlation with clinical findings is essential.
If the clinical findings are not consistent with the Doppler
echocardiographic results, cardiac catheterization is recom-
mended for further hemodynamic assessment. Cardiac cath-
eterization should consist of the simultaneous measurement
of 2 pressures, 1 in the left ventricle and 1 in the aorta, from
which a mean gradient can be calculated. A “pull-back” trac-
ing from the left ventricle to the aorta may be used in patients
with normal sinus rhythm but is not accurate in patients with
irregular rhythms or low-output states. The use of simultaneous
left ventricular and femoral artery pressures is not accurate for
assessing the aortic valve gradient because there may be a sig-
nificant difference between central aortic pressure and femoral
artery pressure. At cardiac catheterization, cardiac output should
Table 36.1. Criteria for Determining Severity of
Aortic Stenosis
Severity Mean Gradient, mm Hg Aortic Valve Area, cm2
Mild <25 >1.5
Moderate 25–40 1.0–1.5
Severe >40 to ≤80 ≥0.7 to <1.0
Critical >80 <0.7
388 IV Valvular Heart Disease
be assessed for calculation of valve area, preferably by the Fick
method. Thermodilution dye curves are used, but these tests
have inherent limitations in patients with irregular heart rhythms
or low-output states. Coexistent aortic regurgitation may cause
errors in the calculation of valve area by cardiac catheterization.
• If atrial fibrillation is present on electrocardiography, suspect
mitral valve disease.
• A mean aortic valve gradient >40 mm Hg on Doppler echocardiog-
raphy indicates severe aortic stenosis.
• The biggest pitfall of Doppler echocardiography is underestima-
tion of the aortic valve gradient.
• Because the calculation is related to cardiac output, the AVA may
be underestimated by the Gorlin formula in low-flow states.
• Accurate determination of AVA by the continuity equation depends
on precise measurements of the LVOTarea.
Natural History and Treatment
The natural history of aortic stenosis is well known (Figure 36.1).
After symptoms occur in a patient with severe aortic steno-
sis, there is a rapidly progressive downhill course, with a 2- to
3-year mortality of 50%. Therefore, it has been recommended
that aortic valve replacement be performed in all patients with
severe aortic stenosis who have symptoms (Table 36.2). Aortic
valve replacement has a low perioperative mortality (<1%–2%
in young, healthy patients) and results in significantly improved
longevity.
Before aortic valve surgery, a complete hemodynamic assess-
ment of the aortic valve with either Doppler echocardiography or
cardiac catheterization is required. Left ventricular function and
concomitant mitral valve disease should be assessed. Coronary
angiography should be performed in older patients who have risk
factors for coronary artery disease, but it usually is not required
in men younger than 35 years or in premenopausal women who
do not have risk factors. Aortic valve replacement should be per-
formed in all patients with symptomatic severe aortic stenosis,
regardless of concomitant left ventricular function. If significant
mitral regurgitation is present, the degree of regurgitation should
be evaluated intraoperatively after replacement of the aortic valve
to determine the need for mitral valve repair or replacement,
unless there is intrinsic disease of the mitral valve apparatus.
100
90
Survival, %
80
70
60
50
40
30 40 50 60 70
Age, y
Figure 36.1. Natural History of Aortic Stenosis. At the onset of
symptoms (arrow), there is a rapid progression and survival is severely
limited. The 3 data points after the onset of symptoms represent death
related to the onset of (from left to right) heart failure, angina, and
syncope.
Table 36.2. Recommendations for Aortic Valve Replacement
in Aortic Stenosis
Indication Class
1. Symptomatic patients with severe AS I
2. Patients with severe AS undergoing coronary artery bypass I
surgery
3. Patients with severe AS undergoing surgery on the aorta or I
other heart valves
4. Patients with moderate AS who are undergoing coronary IIa
artery bypass graft surgery or surgery on the aorta or other
heart valves
5. Asymptomatic patients with severe AS and the following:
LV systolic dysfunction IIa
Abnormal response to exercise (eg, hypotension) IIa
Ventricular tachycardia IIb
Marked or excessive LV hypertrophy (≥15 mm) IIb
Valve area <0.6 cm2 IIb
6. Prevention of sudden death of asymptomatic patients with III
none of the findings listed under indication 5
Abbreviations: AS, aortic stenosis; LV, left ventricular.
An increasing number of elderly patients are presenting with
severe aortic stenosis owing to the aging population. The risk
of aortic valve replacement increases with age and concomitant
medical problems. In patients older than 80 years, operative mor-
tality can be as high as 30%. Percutaneous aortic balloon valvu-
loplasty was introduced in the 1980s as an alternative to valve
replacement and as a way to decrease the high operative mor-
tality among elderly patients. By inflating 1 or more large bal-
loons across the aortic valve from a percutaneous route, a modest
decrease in gradient and a significant improvement in symptoms
could be achieved in elderly critically ill patients. However,
follow-up has demonstrated a high rate of restenosis (>60% at
6 months and nearly 100% at 2 years), with no decrease in the
mortality rate after the procedure. Therefore, percutaneous aor-
tic balloon valvuloplasty is used only 1) for critically ill elderly
patients who are not candidates for surgical intervention or 2) as
a “bridge” for critically ill patients before aortic valve replace-
ment. Percutaneous aortic valve replacement through either a
transfemoral route or a transapical approach is a less invasive
alternative to open heart surgery, especially for patients deemed
to be high risk for the operative approach.
• Aortic valve replacement is indicated for patients with symptoms
and severe aortic stenosis, regardless of the left ventricular ejec-
tion fraction.
• Coronary angiography may not be required preoperatively in
younger patients without risk factors for coronary artery disease.
• Percutaneous aortic balloon valvuloplasty is used only for criti-
cally ill patients or as a “bridge” to surgery.
Controversial Issues in the Management
of Patients With Aortic Stenosis
Asymptomatic Patients With Severe Aortic Stenosis
Treatment of asymptomatic patients with severe aortic stenosis is
controversial. Advocates of “prophylactic” aortic valve replace-
ment in asymptomatic patients recommend that the procedure be
done to prevent sudden death. However, longitudinal studies have
shown that the incidence of sudden death among patients who
are truly asymptomatic is low (probably <1% per year). It is thus
reasonable to follow the asymptomatic patient closely as long as
 36 Valvular Stenosis
exercise tolerance is good and left ventricular systolic function is
preserved. Exercise testing may be performed carefully to docu-
ment exercise tolerance and the hemodynamic response to exer-
cise; operation is considered for those with a reduced exercise
tolerance or an abnormal hemodynamic response to exercise. If
the patient with asymptomatic severe aortic stenosis undergoes
close medical observation, surgery should be performed at the
onset of symptoms or left ventricular systolic dysfunction.
Asymptomatic patients with very high gradients and critical
aortic stenosis (gradients >80 mm Hg and valve areas <0.7 cm2)
may be at higher risk for cardiac events. In addition, 60% to 70%
of asymptomatic patients with severe stenosis and calcified valves
or rapid progression of the stenosis become symptomatic or die
in 3 to 5 years. Although still controversial, it may be reasonable
to offer surgical treatment in select subgroups of these patients
with severe asymptomatic aortic stenosis if the operative risk is
low (<1%) and the patient desires early operation.
Definition of Severe Aortic Stenosis
The definition of severe aortic stenosis varies. Criteria have
included AVA (<0.5 cm2, <0.7 cm2, or <1.0 cm2) and AVA
indexed to body surface area (<0.5 cm2 /m2). AVA, especially
from a single measurement, should not be used as the sole deter-
minant for the severity of stenosis. The reproducibility of AVA
varies by as much as 0.4 to 0.6 cm2. The AVA for a large man
may have a different hemodynamic consequence than the same
AVA for a small woman, supporting the concept that AVA should
be corrected for body surface area. Studies have shown that the
natural history of symptomatic patients with moderate aortic ste-
nosis is comparable to the classic natural history of symptomatic
patients with severe aortic stenosis.
The message is that patients should not be treated on the basis
of a single determination of AVA in isolation from clinical signs
and symptoms. Numerous factors, such as clinical presentation,
exercise tolerance, mean gradient, and left ventricular function,
should be considered when determining the need for aortic valve
surgery.
Low-Output/Low-Gradient Aortic Stenosis
Patients may present with low-output/low-gradient aortic steno-
sis and a low ejection fraction. In these patients, the mean aortic
valve gradient is less than 30 mm Hg and the calculated AVA is
in the range of severe aortic stenosis (<1.0 cm2). These patients
may have critical end-stage aortic stenosis in which ventricu-
lar function has deteriorated because of progressive afterload
on the left ventricle. Aortic valve replacement results in symp-
tomatic improvement, increased longevity, and improvement of
left ventricular systolic function. However, other patients may
have a combination of mild calcific valvular aortic stenosis and
concomitant left ventricular dysfunction due to another cause.
In these patients, the calculated AVA is low because the stroke
volume is not sufficient to completely open the mildly stenotic
aortic valve. It has been difficult to differentiate these 2 subsets
of patients. The use of dobutamine stress testing has been recom-
mended to select patients with true severe stenosis (the gradient
increases to >40 mm Hg with dobutamine).
There may also be a subset of patients with low-output/
low-gradient aortic stenosis and a normal ejection fraction. These
patients have a poor prognosis, and whether they would benefit
from aortic valve replacement is unclear. This subset of patients
is being investigated further.
389
Mitral Stenosis
Cause
Most cases of mitral stenosis have a rheumatic cause. The rheu-
matic process results in “immobility” and thickening of the
mitral valve leaflets, with fusion of the commissures. Leaflet
calcification and subvalvular fusion occur in the late stages of
the disease. In rare instances, the cause may be a congenital
abnormality of the mitral valve or a parachute mitral valve. Cor
triatriatum is an abnormality that simulates mitral stenosis. In
this condition, a thin membrane across the left atrium obstructs
pulmonary venous inflow. Patients with left atrial myxoma and
pulmonary vein obstruction may also present with signs and
symptoms similar to those of mitral stenosis.
Pathophysiology
The pathophysiologic features of mitral stenosis are related to the
increase in left atrial pressure from obstruction across the mitral
valve. This increased pressure is reflected onto the pulmonary
circulation, causing symptoms of dyspnea, orthopnea, and par-
oxysmal nocturnal dyspnea.
Unless mitral stenosis is severe, patients do not have symptoms
at rest; however, with exercise or the onset of atrial fibrillation,
left atrial pressure increases. This results from an increase in the
gradient and an increase in left atrial pressure, which occurs with
a shortened diastolic filling period.
In long-standing severe mitral stenosis, secondary pulmonary
hypertension may occur and lead to right ventricular failure and
tricuspid regurgitation. Symptoms of angina pectoris are rare but
may be due to right ventricular hypertrophy and ischemia of the
right ventricle. The left ventricle is not affected in pure mitral
stenosis.
Pathologically, rheumatic mitral stenosis results in commis-
sural fusion. Secondary effects of the long-standing rheumatic
process may involve progressive calcification and fibrosis of
the mitral valve leaflets. The rheumatic process can also affect
the subvalvular apparatus, leading to shortened and fibrotic
chordae.
• The differential diagnosis of mitral stenosis includes cor triatria-
tum, atrial myxoma, and pulmonary vein obstruction.
• The hallmark of mitral stenosis is commissural fusion.
• The left ventricle is not affected in pure mitral stenosis.
Clinical Presentation
The presentation of patients with mitral stenosis is related to
the increase in left atrial pressure, which produces symptoms
of dyspnea. The early course of the disease—before symptoms
develop—may last for decades. Symptoms begin insidiously,
with mild dyspnea only on exertion. Frequently, patients are
not aware of progressive limitations in exercise because their
activity level has decreased imperceptibly over the years. With
a severe increase in left atrial pressure, paroxysmal nocturnal
dyspnea and orthopnea occur. High pulmonary venous pressures
may cause distention of the bronchial veins, and rupture of these
veins may result in hemoptysis. Stasis occurs in the enlarged left
atrium, especially in the presence of atrial fibrillation, and pro-
duces a nidus for thrombus formation. Systemic embolic events
are seen in approximately one-third of patients with atrial fibril-
lation and mitral stenosis and may be the presenting event before
the diagnosis of mitral stenosis is made.
390 IV Valvular Heart Disease
The classic finding on physical examination of a patient with
mitral stenosis is a loud first heart sound, an OS, and a diastolic
rumble. The interval between aortic valve closure and the OS
(ie, the A2-OS interval) is related to left atrial pressure and, thus,
can be used to determine the severity of mitral stenosis. Patients
with severe mitral stenosis have A2-OS intervals shorter than 60
to 70 ms, and those with mild mitral stenosis have A2-OS inter-
vals longer than 100 to 110 ms. The intensity and duration of the
diastolic rumble increase as the gradient across the mitral valve
increases. However, because of differences in body habitus and
chest cavities, severe mitral stenosis may be present with a barely
audible diastolic rumble. If the mitral valve is pliable and non-
calcified, the first heart sound is loud and snappy and the OS is
very prominent. With progressive calcification and fibrosis of the
leaflets, the first heart sound may diminish in intensity and the OS
may disappear. The intensity of the pulmonic valve component
of the second heart sound is important to note in determining the
severity of coexistent pulmonary hypertension. In patients who
do not have a diastolic rumble on initial auscultation, increasing
heart rate by mild exercise (sit-ups or step-ups) may bring out a
diastolic rumble.
Echocardiography
The standard for diagnosis and determination of the severity of
mitral stenosis is 2-dimensional and Doppler echocardiography.
On 2-dimensional echocardiography, the typical hockey-stick
deformity of the mitral valve leaflets is easily visualized in the
parasternal long-axis view. Commissural fusion and narrowing
of the mitral valve opening area are seen on the short-axis view.
In patients with adequate echocardiographic images, the area
of the mitral valve can be determined planimetrically from the
short-axis view if the plane of the 2-dimensional view is at the
tip of the mitral valve leaflets. Two-dimensional echocardiogra-
phy is also important in identifying the morphology of the mitral
valve leaflets and the subvalvular apparatus.
A grading system has been assigned to determine suitability
for mitral valve valvotomy (surgical or percutaneous) based on
2-dimensional features of the following: 1) leaflet thickening,
2) leaflet calcification, 3) leaflet mobility, and 4) subvalvular
fusion. Each of the 4 morphologic features is assigned a score
from 1 (least involvement) to 4 (most severe involvement), and
the 4 scores are totaled. A total mitral valve score of 8 or less
indicates a pliable noncalcified valve that should be suitable
for balloon valvuloplasty or commissurotomy. Alternatively,
a score of 12 or more indicates a calcified fibrotic valve with
subvalvular fusion that may not be appropriate for valve repair.
Other factors must be considered when determining suitabil-
ity for mitral valvotomy. Calcification of the commissures may
preclude successful valvotomy. Isolated severe subvalvular
fusion, even with a pliable valve leaflet, may preclude a suc-
cessful valvotomy.
Determining the severity of obstruction across the mitral
valve requires measuring the mean mitral valve gradient and cal-
culating MVA. By conventional criteria, in mild mitral stenosis
the mean gradient is less than 5 mm Hg; in moderate stenosis,
5 to 10 mm Hg; and in severe stenosis, greater than 10 mm Hg
(Table 36.3). These values apply to patients with normal car-
diac output and heart rates within the physiologic range of 60
to 90 beats per minute. Previously, cardiac catheterization was
needed to determine the mitral valve gradient. However, Doppler
echocardiography can be used to measure the mean mitral valve
gradient accurately and reproducibly—it is more accurate than
Table 36.3. Criteria for Determining Severity
of Mitral Stenosis
Severity Gradient, mm Hg Mitral Valve Area, cm2
Mild <5 >1.5
Moderate 5–10 1.0–1.5
Severe >10 <1.0
conventional cardiac catheterization when using pulmonary
artery wedge pressure and left ventricular pressures.
The mean mitral valve gradient depends not only on the degree
of obstruction but also on the flow and the diastolic filling period.
A calculated MVA incorporates all these factors. By convention,
an MVA less than 1.0 cm2 indicates severe mitral stenosis; 1.0 to
1.5 cm2, moderate stenosis; and greater than 1.5 cm2, mild steno-
sis (Table 36.3). The Gorlin formula has been the standard for
calculating MVA in cardiac catheterization laboratories:
1, 000 × CO
MVA =
38 × HR × DFP ΔP
where CO = cardiac output, DFP = diastolic filling period, HR =
heart rate, and ΔP = pressure difference across the valve.
The Gorlin formula has limitations: It is not applicable at low
or high cardiac outputs, and results are erroneous with concomi-
tant mitral regurgitation. In addition, determining cardiac output
by cardiac catheterization can be misleading, especially with
atrial fibrillation and concomitant tricuspid regurgitation.
Doppler echocardiography uses the concept of a diastolic half-
time to estimate MVA. The diastolic half-time, initially described
at cardiac catheterization, is the time it takes for the maximal
mitral gradient to decrease by 50%. It is inversely related to valve
area. In most patients, an accurate measurement of MVA can be
obtained from the rate of velocity decrease during early and mid-
diastole, as assessed on the transmitral velocity curve:
t1 / 2 DT × 0 29
220
MVA =
t1 / 2
where t1/2 = half-time and DT = deceleration time.
There are limitations to using the diastolic half-time, espe-
cially when abnormalities of left atrial and left ventricular com-
pliance exist. In these instances, the MVA should be calculated
with the continuity equation:
TTVI × LVOTarea
LVOT
MVA =
MV VTVI
where LVOT = left ventricular outflow tract, MV = mitral valve,
and TVI = time-velocity integral.
• Doppler echocardiography is more accurate than cardiac catheter-
ization for determining the mean mitral valve gradient.
• Suitability for mitral valve valvuloplasty should be assessed with
2-dimensional echocardiography and be based on the mitral valve
score and appearance of the commissures.
• The continuity equation for MVA should be used when the area
derived from the diastolic half-time does not correlate with the
mean transmitral gradient.
 36 Valvular Stenosis
Signs
RF Class II
10 y AF
Class III or IV
10 y
10 y
10 y
Figure 36.2. Natural History of Mitral Stenosis. The course is one
of plateaus. After the onset of rheumatic fever (RF), the patient remains
asymptomatic for 10 to 20 years before auscultatory signs of mitral ste-
nosis develop. With the onset of mild symptoms, there is a long per-
iod before the development of atrial fibrillation (AF). After the onset
of New York Heart Association class III or IV symptoms, a patient’s
condition rapidly deteriorates.
Natural History and Treatment
The course of mitral stenosis is one of plateaus (Figure 36.2).
After the onset of rheumatic fever, a period of 1 to 2 decades pre-
cedes the appearance of signs of mitral stenosis. Another period
of 1 to 2 decades passes before mild symptoms occur. Mild
symptoms of dyspnea on exertion may be present for another 1
to 2 decades. During this time, the onset of atrial fibrillation may
cause further decompensation, but this can be treated by rate
control. Finally, severe New York Heart Association class III or
IV symptoms develop. Indications for mitral valve replacement
include a combination of severe mitral stenosis and New York
Heart Association functional class III or IV symptoms, with sig-
nificant limitation of lifestyle. With the advent of percutaneous
mitral balloon valvotomy, balloon valvotomy may be a reason-
able procedure to perform relatively early if the valve is a good
candidate morphologically.
In determining the need for intervention, the symptoms must
be correlated with the gradient and, subsequently, the MVA.
Some patients may have significant symptoms but have a mitral
valve gradient and an MVA that are consistent with less severe
mitral stenosis (gradient ≤10 mm Hg or MVA >1.5 cm2). A mild
degree of obstruction in 1 patient may be a significant degree of
obstruction in another.
In patients whose symptoms are out of proportion to the
calculated mitral valve indexes, the hemodynamic response
to exercise should be measured. Previously, this was assessed
during right and left heart catheterization during exercise.
However, exercise Doppler echocardiography can provide
all the information required. For this, the patient undergoes
a treadmill or supine bicycle exercise test until symptoms
occur. Mean mitral valve gradient and pulmonary artery sys-
tolic pressure should be measured at peak exercise. If the mean
gradient increases more than 15 mm Hg with symptoms, the
patient should be considered symptomatic on the basis of the
mitral valve disease. Alternatively, if symptoms occur and
the mitral valve gradient does not increase to those levels,
another cause for the symptoms must be pursued. In patients
with a mean mitral valve gradient that does not correlate with
MVA, other Doppler estimates of MVA must be considered,
including the continuity equation or proximal isovelocity
surface area calculations.
Operations for mitral stenosis have consisted of closed com-
missurotomy, open commissurotomy, and mitral valve replace-
ment. Closed commissurotomy was an effective procedure that
391
Table 36.4. Recommendations for Percutaneous Mitral
Balloon Valvotomy
Indication Class
1. Symptomatic patients (NYHA functional class II, III, or IV) I
with moderate or severe MS (MVA ≤1.5 cm 2) a and valve
morphology favorable for percutaneous balloon valvotomy in
the absence of left atrial thrombus or moderate or severe MR
2. Asymptomatic patients with moderate or severe MS (MVA IIa
≤1.5 cm 2) a and valve morphology favorable for percutaneous
balloon valvotomy who have pulmonary hypertension
(pulmonary artery systolic pressure >50 mm Hg at rest
or >60 mm Hg with exercise) in the absence of left atrial
thrombus or moderate or severe MR
3. Patients with NYHA functional class III or IV symptoms, IIa
moderate or severe MS (MVA ≤1.5 cm 2),a and a nonpliable
calcified valve who are high-risk candidates for surgery in the
absence of left atrial thrombus or moderate or severe MR
4. Asymptomatic patients with moderate or severe MS (MVA IIb
≤1.5 cm 2) a and valve morphology favorable for percutaneous
balloon valvotomy who have new-onset atrial fibrillation in the
absence of left atrial thrombus or moderate or severe MR
5. Patients with NYHA function class III or IV symptoms, IIb
moderate or severe MS (MVA ≤1.5 cm 2), and a nonpliable
calcified valve who are low-risk candidates for surgery
6. Patients with mild MS III
Abbreviations: MR, mitral regurgitation; MS, mitral stenosis; MVA, mitral
valve area; NYHA, New York Heart Association.
a
The American College of Cardiology/American Heart Association guidelines
committee recognizes that there may be variability in the measurement of MVA
and that the mean transmitral gradient, pulmonary artery wedge pressure, and
pulmonary artery pressure at rest or during exercise should also be considered.
was used before the institution of cardiopulmonary bypass.
Through a lateral thoracotomy, the surgeon would attempt to split
the commissural fusion with a finger or dilator. This was success-
ful in most patients with noncalcified valves but was inadequate
for those with calcified fibrotic valves and subvalvular fusion.
With the advent of cardiopulmonary bypass, open commissuro-
tomy became the procedure of choice. This requires a median
sternotomy and cardiopulmonary bypass. The surgeon directly
inspects the mitral valve apparatus and incises the commissures
under direct vision, with chordal reconstruction if necessary. If
the mitral valve is not suitable for commissurotomy, the mitral
valve is replaced.
Recently, percutaneous mitral balloon valvotomy has become
an acceptable alternative to mitral valve surgery in selected
patients (Table 36.4). Fused commissures can be split with 1
or more large balloons inflated across the mitral valve. This
can produce hemodynamic improvement comparable to that
of closed commissurotomy in selected patients with pliable
valves. Percutaneous mitral balloon valvotomy requires exper-
tise, including the capability to perform a transseptal puncture.
However, in experienced hands, the results are excellent and
comparable to those of surgery, with the valve area typically
increasing from 1.0 to 2.0 cm2. In several randomized trials
comparing mitral balloon valvotomy with closed and open surgi-
cal commissurotomy, the acute results and long-term outcomes
have been comparable among young patients with pliable valves.
Potential complications, such as systemic embolus, severe mitral
regurgitation, and left ventricular perforation, can be avoided
by preoperative assessment of mitral valve morphology and
documentation of the lack of atrial thrombus by transesophageal
392 IV Valvular Heart Disease

echocardiography. In selected patients with pliable valve leaflets, • Percutaneous mitral balloon valvotomy may be the procedure of
percutaneous mitral balloon valvotomy should be considered not choice for patients with mitral stenosis and a noncalcified, pliable
only in those with New York Heart Association class III or IV mitral valve.
symptoms but also in those with class I or II symptoms who have
high resting gradients and pulmonary hypertension.
Abbreviations
• The course of mitral stenosis is one of plateaus.
A2 aortic valve component of second heart sound
• Exercise hemodynamic studies should be performed if patients AVA aortic valve area
have symptoms out of proportion to the calculated mitral valve MVA mitral valve area
gradient and MVA. OS opening snap
 37
Valvular Regurgitation
RICK A. NISHIMURA, MD
Mitral Regurgitation
The mitral valve is a complicated structure that requires the cor-
rect functioning of the valve leaflets, valve commissures, mitral
annulus, papillary muscles, chordae tendineae, and left ventri-
cle for competence. Mitral regurgitation results from failure of 1
or more of the components of normal competence of the mitral
valve. The presentation and management of mitral regurgitation
depend on the underlying cause, duration, severity, symptoms
(including objective exercise tolerance), and left ventricular size
and systolic function.
Causes of Mitral Regurgitation
There are 3 basic mechanisms of mitral regurgitation: 1) altera-
tion of mitral leaflets, commissures, or annulus; 2) defective ten-
sor apparatus; and 3) alterations of left ventricular and left atrial
size and function.
Alteration of Mitral Leaflets, Commissures, or Annulus
Rheumatic Fever. Rheumatic mitral valve disease may
deform valves, shorten chordae, or fuse commissures and lead to
pure mitral regurgitation or predominant regurgitation in combi-
nation with stenosis.
Mitral Valve Prolapse. Mitral valve prolapse is the most
common cause of isolated severe mitral regurgitation. The pos-
terior leaflet is affected more frequently and severely than the
anterior leaflet. Mitral annular dilatation and calcification and
myxomatous changes of chordae tendineae may be associated.
The likelihood of mitral valve prolapse resulting in considerable
mitral regurgitation increases with age and is greater in men
than in women. Prolapse may occur alone or in association with
other conditions, including Marfan syndrome, Ehlers-Danlos
syndrome, and thoracic skeletal abnormalities.
Mitral Annulus Calcification. Isolated mitral annulus calci-
fication is usually age-related and occurs in older patients with
hypertension, hypertrophic obstructive cardiomyopathy, chronic
renal failure, or aortic stenosis.
Infective Endocarditis. Infective endocarditis may damage
valve leaflets by perforation or by the development of a vege-
tation that interferes with coaptation. Anatomical abnormalities
may persist even after the active infection has been eradicated.
Congenital. A congenital cleft of the anterior mitral leaflet
often is associated with primum atrial septal defect, but it can
exist in isolation without other features of a persistent atrioven-
tricular canal.
Rare Causes of Mitral Incompetence. Other, uncommon
causes of mitral leaflet abnormalities include endomyocardial
fibrosis, carcinoid disease with right-to-left shunting or bronchial
carcinoid-secreting adenomas, ergotamine toxicity, radiation
therapy, trauma, rheumatoid arthritis, systemic lupus erythema-
tosus (Libman-Sacks lesions), and diet-drug toxicity.
Defective Tensor Apparatus
Abnormal Chordae Tendineae. Ruptured chordae are
responsible for a considerable percentage of cases of mitral regur-
gitation. Rupture may be idiopathic, a complication of endocarditis,
a result of myxomatous degeneration in the setting of mitral valve
prolapse, or a result of blunt or direct penetrating thoracic trauma.
Papillary Muscle Dysfunction. Myocardial ischemia or
infarction can cause papillary muscle dysfunction (without
393
394 IV Valvular Heart Disease
rupture). The posteromedial papillary muscle is more vulnerable
to ischemia or infarction than the anterolateral papillary muscle
because of its single end-artery vascular supply. Nonischemic
causes of papillary muscle dysfunction include dilated cardio-
myopathy, myocarditis, and hypertension.
Myocardial infarction (either transmural or subendocardial)
can cause rupture of a papillary muscle. It usually occurs in the
first week after infarction at a time when the inflammatory cell
response is maximal and most frequently involves the posterome-
dial papillary muscle. Rarely, chest trauma can cause papillary
muscle rupture, usually with coexistent myocardial or ventricu-
lar septal rupture.
Alterations of Left Ventricular and Left
Atrial Size and Function
Global or regional left ventricular enlargement may alter the pos-
ition and axis of contraction of the papillary muscles in addi-
tion to causing dilatation of the mitral ring. Progressive left atrial
and ventricular enlargement associated with any type of chronic
mitral regurgitation, in turn, leads to more mitral regurgitation
by further altering the geometry of the chamber.
Mitral regurgitation in hypertrophic obstructive cardiomyop-
athy results from distortion of the mitral valve apparatus from
systolic anterior motion of the mitral valve, with the jet directed
posteriorly and laterally.
• Understanding the mechanism of mitral regurgitation helps define
the natural history and optimal treatment.
Acute volume overload
Left ventricle
Sarcomere stretch Left ventricular
diastolic pressure
Frank-Starling
mechanism
Left ventricular stroke volume
Left ventricular ejection fraction
Figure 37.1. Pathophysiology of Acute Severe Mitral Regurgitation.
• Mitral prolapse is the most common cause of isolated severe mitral
regurgitation.
• Ischemia can cause either dysfunction or rupture of papillary
muscle. The posteromedial papillary muscle is affected most often
because of its single end-artery blood supply.
• Left ventricular enlargement and abnormal contractile function
are common causes of significant mitral regurgitation.
Pathophysiology of Mitral Regurgitation
Mitral regurgitation has 3 pathophysiologic stages: 1) acute, 2)
chronic compensated, and 3) chronic decompensated.
Acute Stage
In acute severe mitral regurgitation, there is a sudden vol-
ume overload on an unprepared left ventricle and left atrium
(Figure 37.1). The left ventricle responds with increased sarco-
mere stretch and augmented left ventricular stroke volume via
the Frank-Starling stretch mechanism. However, the larger vol-
ume increases left ventricular diastolic pressure, which in turn
increases left atrial pressure. Because left atrial compliance is
normal in the acute stage, the large regurgitant volume markedly
increases left atrial pressure, which causes pulmonary conges-
tion, edema, and dyspnea.
The sudden opening of a new low-pressure pathway for sys-
tolic ejection decreases left ventricular afterload, permitting
more complete volume ejection from the ventricle. Although total
left ventricular stroke volume increases, forward stroke volume
Left atrium
(Direct)
(Indirect) Left atrial
pressure
Pulmonary
Dyspnea
congestion
Flow into left atrium
Left ventricular afterload
 37 Valvular Regurgitation 395

decreases. The combination of increased preload, decreased after-
load, and increased left ventricular contractile function increases
the measured ejection fraction to between 60% and 75%.
Chronic Compensated Stage
In the chronic compensated stage (Figure 37.2), left ventricular
volume overload elongates individual myocytes, causing com-
pensatory eccentric left ventricular hypertrophy and increasing
left ventricular end-diastolic volume. The Frank-Starling mech-
anism continues to augment total stroke volume. The left atrium
dilates, thus increasing its compliance. There is an increase in
preload from the volume overload and a decrease in afterload
from ejection into the low-impedance left atrium. The combina-
tion of increased end-diastolic volume, augmented contraction,
and decreased afterload continues to result in a high ejection
fraction. The dilated left ventricle and atrium allow the regur-
gitant volume to be accommodated at lower filling pressures,
thereby minimizing symptoms of pulmonary congestion.
Chronic Decompensated Stage
Eventually, left ventricular contractile function decreases as
a result of the development of fibrosis from longstanding vol-
ume overload, and the chronic decompensated stage of mitral
regurgitation begins (Figure 37.3). This downward spiral includes
an increase in end-systolic volume as left ventricular function
decreases. Left ventricular filling pressures increase and cause
pulmonary congestion. Increased ventricular pressure further
dilates the left ventricle, increasing systolic wall stress and after-
load. The increased afterload further reduces ventricular systolic
function, thus completing the downward cycle. As end-diastolic
and end-systolic volumes increase, the ejection fraction may
be in the normal range of 50% to 60%. In patients with severe
mitral regurgitation, an ejection fraction less than 60% is prob-
ably abnormal and indicative of ventricular dysfunction.
The time during which patients progress from compensated
to decompensated mitral regurgitation depends on the severity
of the regurgitation (which itself can change over time), variables
that affect afterload and ventricular contractility, and individual,
poorly understood, patient characteristics.
• Acute severe mitral regurgitation is characterized by normal-sized
chambers, high ejection fraction, and pulmonary congestion.
• Chronic compensated severe mitral regurgitation is typified by few
symptoms, enlargement of the left ventricle and atrium, and high
ejection fraction.
• A normal ejection fraction (50%–60%) in the setting of severe
mitral regurgitation usually implies left ventricular systolic
dysfunction.

Chronic volume overload

Left ventricular Increased compliance Left atrial

dilatation dilatation

Pulmonary congestion
Dyspnea
Left
ventricular
mass
Left ventricular Left ventricular
radius afterload

Frank-Starling Left ventricular

mechanism afterload

Left ventricular Left ventricular end- No net change in left

stroke volume diastolic volume ventricular afterload

Ejection fraction

Figure 37.2. Pathophysiology of Chronic Compensated Mitral Regurgitation.

396 IV Valvular Heart Disease
Chronic volume overload
Left ventricular
dilatation
Left ventricular Frank-Starling
radius mechanism
Left ventricular Left ventricular
afterload stroke volume
Ejection fraction
Figure 37.3. Pathophysiology of Chronic Decompensated Mitral Regurgitation.
Clinical Syndrome of Mitral Regurgitation
Acute Mitral Regurgitation
Acute mitral regurgitation usually results from infective endo-
carditis, infarction, ischemic heart disease, trauma, or chordal
rupture. If acute mitral regurgitation is severe, severe pulmo-
nary congestion is expected. High left atrial and left ventricu-
lar end-diastolic pressures account for the third and fourth heart
sounds. The systolic murmur of mitral regurgitation in this acute
condition may be short, soft, or completely absent. Rarely, there
may be only a small left ventricular-to-atrial pressure gradient
(because left atrial pressure has increased close to that of the
left ventricle). This may result in the absence of both an audible
murmur and Doppler color flow evidence of mitral regurgitation
(indicating almost no turbulence in the regurgitant flow).
Chronic Mitral Regurgitation
Patients with chronic mitral regurgitation may have a prolonged
asymptomatic interval. However, adverse ventricular changes
may develop during this period. Once symptoms arise, the low
cardiac output symptoms of fatigue and generalized weakness
predominate early. As left ventricular function deteriorates, exer-
tional dyspnea, orthopnea, and paroxysmal nocturnal dyspnea
become more prominent.
Examination of the precordium in a patient with severe
chronic mitral regurgitation reveals a brief, laterally displaced,
and enlarged apical impulse. A ventricular filling impulse may be
palpable. The presence of an apical thrill indicates severe mitral
regurgitation. As ventricular systolic function deteriorates, the
duration of the apical impulse increases.
Left ventricular end-
systolic volume
Left ventricular and
atrial pressure
Pulmonary
Dyspnea
congestion
Auscultation may reveal single or multiple nonejection clicks
if mitral prolapse is present. An accentuated pulmonic compo-
nent of the second heart sound indicates pulmonary hyperten-
sion. Aortic closure may be early because of decreased duration
of forward flow, causing persistent wide splitting of the second
heart sound (however, the duration of the murmur often obscures
the second heart sound, making the splitting hard to detect). An
apical third heart sound is common, but fourth heart sounds are
unusual.
A holosystolic murmur is the hallmark of mitral regurgitation.
The intensity of the murmur does not necessarily correlate with
the severity of the regurgitant flow. Although the murmur fre-
quently radiates to the axilla, a primary posterior leaflet abnor-
mality will direct the regurgitant flow anteriorly and the radiation
may be toward the aortic area. Characteristically, the intensity of
the mitral regurgitant murmur is constant despite different cycle
lengths; this feature helps, at the bedside, to distinguish it from
aortic outflow murmurs. The presence of a short diastolic apical
rumble in the absence of mitral stenosis implies high diastolic
transmitral flow and severe mitral regurgitation.
• Acute severe mitral regurgitation may have a short or soft murmur
because of the low left ventricular-to-atrial pressure gradient.
• Nonspecific fatigue and weakness may represent the early symp-
toms of chronic severe mitral regurgitation.
• A third heart sound in chronic mitral regurgitation is usually indic-
ative of severe regurgitation.
• Posterior leaflet prolapse often produces a murmur that radiates to
the aortic area.
• Lack of change in the intensity of a mitral regurgitant murmur with
variable cycle lengths distinguishes it from an outflow tract murmur.
 37 Valvular Regurgitation
Evaluation of Mitral Regurgitation
Electrocardiography
No electrocardiographic findings are diagnostic of mitral regur-
gitation. The electrocardiogram may show atrial fibrillation; left
atrial enlargement is expected if sinus rhythm persists. Left ven-
tricular hypertrophy and nonspecific ST-segment and T-wave
changes are also common.
Chest Radiography
Significant chronic mitral regurgitation causes the left ventric-
ular and left atrial enlargement seen on chest radiographs. Left
atrial enlargement might be recognized as straightening of the
left border of the heart, an atrial double density, or elevation of
the left mainstem bronchus. Pulmonary venous congestion may
be present in any stage of mitral regurgitation.
Echocardiography
Echocardiography is valuable for determining the effects of
mitral regurgitation both on left ventricular and atrial size and
function and on right ventricular function and hemodynamics.
Echocardiography is essential for morphologic assessment of
the mitral valve, annulus, commissures, and papillary muscles.
The severity of mitral regurgitation is assessed with an inte-
grated Doppler and 2-dimensional echocardiographic examina-
tion (Box 37.1). A flail mitral valve leaflet usually is associated
with severe mitral regurgitation. In some cases, transesophageal
echocardiography is required to better assess the anatomy of the
mitral valve and its supporting structures, to inspect the atria for
thrombus, and to gather supplemental data used in qualitative
and quantitative measures of regurgitation severity.
Cardiac Catheterization
Left ventriculography is used primarily when noninvasive data
are discordant or technically limited or differ from the clini-
cal perception of the severity of mitral regurgitation or ven-
tricular function. Angiographic grading of mitral regurgitation
Box 37.1. Doppler Indicators of Severe Mitral
Regurgitationa
Color jet area (>8.0 cm2; >1/3 LA area)
Wide vena contracta (color flow)
Regurgitant volume (>60 mL)b
Regurgitant fraction (>55%)b
ERO (>0.35 cm2)b
Pulmonary vein PW Doppler flow profile (systolic
flow reversal)
CW Doppler signal intensity (dense)
Transmitral PW flow velocity (E >1.5 m/s)
Abbreviations: CW, continuous-wave; E, early transmitral flow; ERO,
effective regurgitant orifice area; LA, left atrial; PW, pulsed-wave.
a
Measurements indicative of severe regurgitation are in parentheses.
b
Can be measured by volumetric Doppler, proximal isovelocity surface
area method, or a combination of these, including 2-dimensional
measures.
397
Box 37.2. Sellers Criteria for Invasive Estimation of
Degree of Mitral Regurgitation
Grade Criteria
1+ Contrast medium does not completely fill left
atrium
2+ Contrast medium completely opacifies left
atrium but does not reach intensity of that in left
ventricle
3+ Contrast medium completely opacifies left
atrium and reaches intensity of that in left
ventricle after 4 or 5 beats
4+ Contrast medium completely opacifies left
atrium and reaches intensity of that in left
ventricle within first 2 or 3 beats
Previously published. See “Credit Lines” section.
provides a semiquantitative measure of the severity of regurgi-
tation from the density of contrast agent going back into the left
atrium (Sellers criteria, Box 37.2). However, grading accuracy
is dependent on the volume and injection rate of the contrast
agent, catheter position, hemodynamics at the time of injection,
and volume of the left atrium, in addition to the severity of valve
regurgitation. Calculations of the invasively derived regurgitant
fraction are rarely used and are dependent on accurate measure-
ment of left ventricular volume and cardiac output. The pres-
ence or absence of prominent v waves on a pulmonary wedge
hemodynamic tracing depends not only on mitral regurgitant
flow at the time of the study but also on the compliance of the
left atrium.
• Echocardiography is invaluable for assessing the cause and sever-
ity of mitral regurgitation and its effects on the size and function of
the left ventricle, left atrium, and right ventricle.
• Left ventriculography is useful when noninvasive data are discor-
dant or technically limited or differ from the clinical impression of
the severity of mitral regurgitation or ventricular function.
Mitral Valve Prolapse
In patients with mitral valve prolapse, the systolic billowing of a
portion of the mitral leaflet into the left atrium may cause an audi-
ble click. If the prolapse is significant, mitral regurgitation results.
This condition is dynamic; the timing of the click and murmur
varies with preload and afterload. With maneuvers that decrease
the ventricular preload (Valsalva maneuver, standing) result in
the click and murmur will occur earlier in the cardiac cycle. The
converse is true with maneuvers that increase preload.
The natural history of most cases of mitral valve prolapse
is benign, although progression to severe mitral regurgitation
is more prevalent in men than women and with advancing age.
Rare complications of mitral valve prolapse such as endocarditis,
severe rhythm disturbances, and strokes occur predominantly in
patients with thickened valve leaflets.
Concomitant mitral valve prolapse and severe mitral regur-
gitation are managed similarly to severe mitral regurgitation
alone. If the regurgitation is mild, patients should be reassured
and followed up clinically for any changes in either symptoms or
physical findings.
Patients with mitral valve prolapse and palpitations or
increased adrenergic tone, atypical chest pain, anxiety, or
398 IV Valvular Heart Disease
fatigue should be counseled to minimize their use of exogenous
stimulants and may benefit from β-adrenergic blocker therapy.
Antiplatelet agents are advised for transient ischemic events in
the setting of mitral valve prolapse if no other cause is found.
Natural History of Mitral Regurgitation
The natural history of acute mitral regurgitation is dependent
on its cause and severity. Patients with papillary muscle rupture
or severe regurgitation from an unstable mitral prosthesis have
a poor short-term outlook without operation. Those with acute
regurgitation from endocarditis have a variable course depend-
ing on the response to antibiotic treatment. Patients with abrupt
chordal rupture have a natural history dependent primarily on
the severity of regurgitation.
Patients with chronic mitral regurgitation have a clinical
course characterized by an initial compensated phase followed
by progressive left ventricular enlargement and eventually sys-
tolic dysfunction. Patients with severe mitral regurgitation may
not have symptoms for years. However, the long-standing volume
overload causes progressive fibrosis and myocyte degeneration
with eventual left ventricular dysfunction. This left ventricular
dysfunction may occur before the onset of symptoms and por-
tends a poor prognosis even if the regurgitation is corrected sur-
gically. The natural history of severe mitral regurgitation is one
of considerable early mortality and morbidity if left untreated.
In patients with severe mitral regurgitation due to a flail leaflet,
more than 60% will have development of heart failure within
10 years of follow-up.
Treatment of Acute Severe Mitral Regurgitation
Patients with acute severe mitral regurgitation may be hemody-
namically stable or unstable at presentation. Any patient with
hemodynamic instability requires rapid evaluation and therapy
with intravenous vasodilators (usually sodium nitroprusside),
intravenous inotropes, and, perhaps, intra-aortic balloon counter-
pulsation. Patients with acute regurgitation often have a ruptured
papillary muscle or an unstable mitral prosthesis, and repair or
replacement of the mitral valve generally is indicated because
long-term medical therapy is ineffective.
Other therapies for acute mitral regurgitation include antibiot-
ics if endocarditis is present and antianginal drugs, angioplasty,
or stenting in some cases of ischemic papillary muscle dysfunc-
tion. In endocarditis, operation is often delayed in the hope of
sterilizing the mitral valve bed because ongoing active infection
increases the risk of prosthetic endocarditis. Patients with endo-
carditis should be considered for urgent surgical intervention if
progressive heart failure, infection unresponsive to antibiotics,
intracardiac abscess, or recurrent systemic embolization devel-
ops despite therapy.
Treatment of Chronic Mitral Regurgitation
All patients with mitral regurgitation should be instructed in
dental hygiene to prevent infective endocarditis. Treatment of
hypertension and myocardial ischemia (in patients with coro-
nary artery disease) and prevention of ventricular dilatation (in
patients with cardiomyopathy) may be helpful for preventing
the progression of severity of mitral regurgitation. There are no
data to support the use of vasodilators or angiotensin-converting
enzyme inhibitors in patients with mitral regurgitation unless
there is left ventricular dysfunction or hypertension.
Timing of operation is difficult. Early operation to implant a
prosthetic mitral valve has its own short- and long-term compli-
cations and thus should not be performed prematurely. However,
because of the insidious progressive left ventricular dysfunction
that occurs from the long-standing volume overload, by the time
symptoms occur the opportunity for operation may have been
missed.
Overall, operation is indicated in chronic severe mitral regur-
gitation with the onset of any symptoms. Although left ventricular
dysfunction may have already occurred, survival and outcomes
are still better with operation than with medical management.
In asymptomatic patients, measurements of left ventricular
size and function determine the need for operation. An ejection
fraction less than 60% or end-systolic dimension more than 40
mm indicates systolic dysfunction. Patients in whom these val-
ues are exceeded should be considered for operation. In the sub-
group of patients with heart failure and ejection fraction less than
35%, the risk associated with operation is high because systolic
function is severely reduced, and the decision to operate must
balance the operative risk versus the probability of symptomatic
improvement.
In patients with symptoms of left ventricular dysfunction
and ejection fraction greater than 35%, operation should be
considered regardless of the type of operation (replacement or
repair). However, in any patient, repair is preferred over replace-
ment because of better long-term outcome. The better outcome
with mitral valve repair is a result of better preservation of left
ventricular function, durability of the repair, and avoidance of
the potential complications of a valve prosthesis (and long-term
anticoagulation).
Some centers now advocate earlier operation in asymptomatic
patients with severe mitral regurgitation despite preserved ventric-
ular function. This early operation is to avoid the onset of ventric-
ular dysfunction, which, if it occurs, portends a poorer prognosis.
However, because of the attendant risks of a prosthetic valve, this
early operation should be done only if there is a high chance of
successful mitral valve repair (>90%) and a low operative risk.
• Patients with acute severe mitral regurgitation and hemodynamic
instability require rapid evaluation, aggressive stabilization, and
early valve operation.
• Indications for valve operation in endocarditis include progres-
sive heart failure, resistance to antibiotics, intracardiac abscess, or
recurrent systemic embolization despite therapy.
• Patients with severe chronic mitral regurgitation who are in
New York Heart Association class II, III, or IV, have an ejec-
tion fraction less than 60%, or have an end-diastolic diameter
more than 40 mm, should undergo valve surgery, if not otherwise
contraindicated.
• All patients are better served by valve repair than replacement.
• Successful valve repair is less likely in cases that are rheumatic,
ischemic, or due to infection, or when prolapse is anterior or bileaf-
let, or significant calcification of the leaflets or annulus is present.
Aortic Regurgitation
Aortic regurgitation may result from intrinsic structural abnor-
malities of the aortic valve or the ascending aorta or both.
Causes of Aortic Regurgitation
Intrinsic Valvular Disease
Rheumatic fever causes mild aortic regurgitation during the first
episodes, but with time the leaflets fibrose, shorten, and contract,
 37 Valvular Regurgitation
resulting in malalignment and loss of coaptation. Associated aor-
tic stenosis is usually present due to commissural fusion. It is
uncommon to have rheumatic aortic disease without coexistent
mitral disease.
The most common congenital cause of aortic regurgitation in
adults is a bicuspid valve with malcoaptation, diastolic prolapse
of a cusp, or both. In addition, chronic progressive aortic regur-
gitation also may be associated with ventricular septal defects
(due to weakening of the neighboring supporting aortic annulus)
or subaortic stenosis (causing turbulent high-velocity jets that hit
and damage the aortic leaflets).
Infective endocarditis usually involves previously abnormal
valves and leads to tissue destruction, vegetation interference
with proper alignment of the commissures during closure, or
invasion and structural distortion of the aortic valve annulus.
Even after medical eradication of infection, regurgitation may
progress because of contracture of healing cusps.
Collagen vascular diseases usually affect the aortic root, but
they also can affect the cusps themselves. The associated valvu-
litis leads to contracture of leaflets, with central regurgitation.
Perforation of leaflets is less common.
Senile degenerative aortic valve disease is usually important
clinically because of aortic stenosis, but some degree of aortic
regurgitation, especially in early stages, often occurs. Significant
regurgitation often occurs after either operative decalcification
or percutaneous balloon valvuloplasty for aortic stenosis.
Diseases of the Ascending Aorta
Acute destruction of the aortic root disrupts the supporting struc-
tures of the valve and results in regurgitation. Aortic dissection
longitudinally cleaves the aortic intima or media with a dissect-
ing column of blood and occurs most often in patients with idio-
pathic dilatation of the ascending aorta, hypertension, or Marfan
syndrome. It also can be associated with pregnancy, result from
blunt chest trauma, or follow acute aortitis complicating aortic
valve infective endocarditis.
Many diseases are associated with chronic dilatation of the
aortic root, and they cause regurgitation by stretching the valve
cusps. These diseases include:
1. Marfan syndrome, usually associated with progressive aortic dilata-
tion as a result of cystic medial necrosis
2. Progressive idiopathic aortic dilatation with cystic medial necrosis
3. Senile dilatation and annuloaortic ectasia of unknown cause
4. Syphilitic aortitis developing 15 to 25 years after the initial infection
and sparing the sinuses of Valsalva
5. Connective tissue disorders (rheumatoid arthritis, ankylosing spon-
dylitis, Reiter syndrome, relapsing polychondritis, giant cell arteri-
tis, and Whipple disease).
Marfan syndrome, progressive idiopathic aortic dilatation, and
some of the connective tissue disorders also can affect the mitral
leaflets and the proximal conducting system.
• It is essential to know whether a patient’s aortic regurgitation is due
to valvular disease or proximal aortic disease or both.
• Many causes of aortic regurgitation often have associated mitral
valve abnormalities: endocarditis, rheumatic fever, collagen vas-
cular disease, or Marfan syndrome.
Pathophysiology of Aortic Regurgitation
Acute or subacute significant aortic regurgitation causes the
abrupt introduction of a large volume of blood into a noncom-
pliant left ventricle, thus increasing left ventricular end-diastolic
399
and pulmonary venous pressures and leading to dyspnea or pul-
monary edema.
In chronic aortic regurgitation, compensatory left ventricular
changes occur over time. The excess volume load causes stretch-
ing and elongation of myocardial fibers, which in turn increase
wall stress. Wall stress is normalized by sarcomere replication
and hypertrophic thickening of the ventricular walls. Thus,
although the ratio of wall thickness to cavity radius remains
essentially normal, left ventricular mass increases (eccentric
hypertrophy).
Initially, ventricular enlargement increases the ejection frac-
tion through the Frank-Starling mechanism. However, further
enlargement exhausts preload reserve, and the ejection fraction
decreases to the “normal” range. Eventually, ejection fraction
decreases further, whereas end-systolic volume increases. This
end-systolic volume increase is a sensitive index of myocar-
dial dysfunction. When the ventricle can dilate no further, dia-
stolic pressure increases and results in dyspnea, another sign of
decompensation.
During exercise, the volume of aortic regurgitation tends to
decrease because of decreased systemic vascular resistance and
shortened diastolic period. However, there also are increases in
venous return that the enlarged left ventricle may not be able
to handle, thus causing a relative decrease in output (exertional
fatigue), increase in end-diastolic pressure (dyspnea), or both.
Patients with chronic aortic regurgitation may experience
anginal symptoms despite normal coronary arteries. Mechanisms
include increase in total myocardial oxygen consumption
(increased left ventricular myocardial mass and wall tension),
decreased subendocardial perfusion gradient due to compressed
intramyocardial coronary arterioles, decreased central aortic
diastolic driving pressure, and diminished coronary arteriolar
vasodilatory reserve.
• Acute severe aortic regurgitation is characterized by normal left
ventricular size, high ejection fraction, and dyspnea or pulmonary
edema.
• Chronic compensated aortic regurgitation is typified by minimal
symptoms and left ventricular enlargement.
• Decompensation is characterized by symptoms (initially with
exertion) and decreasing ejection fraction.
Clinical Syndrome of Aortic Regurgitation
Acute aortic regurgitation usually is due to aortic dissection or
infective endocarditis. In these circumstances, the manifesta-
tions of the underlying process usually predominate. Because
compensatory cardiac mechanisms cannot develop, significant
dyspnea occurs as a consequence of high left ventricular end-
diastolic and pulmonary venous pressures. A murmur may be
minimal because of the abrupt increase in left ventricular end-
diastolic pressure and rapidly decreasing aortic-left ventricular
diastolic pressure gradient. Peripheral manifestations (which
are caused by rapid volume runoff into the left ventricle) may be
absent because of acutely high ventricular diastolic pressures.
In chronic aortic regurgitation, the patient often remains
asymptomatic for decades. Exertional dyspnea occurs late.
Examination of the patient may reveal nodding of the head (de
Musset sign), visible capillary pulsation in the nail beds during
gentle pressure on the edge of the nail (Quincke sign), features of
Marfan syndrome, or stigmata of infective endocarditis.
Hemodynamically severe aortic regurgitation generally causes
a widened pulse pressure more than 100 mm Hg with a diastolic
400 IV Valvular Heart Disease
pressure less than 60 mm Hg. Pulse pressure may not accurately
reflect the severity of aortic regurgitation in young patients with
compliant vessels or in patients with accompanying left ventricu-
lar failure and increased left ventricular end-diastolic pressure.
Other signs of high-volume systolic ejection of blood with rapid
diastolic runoff include a sharp, rapid carotid upstroke, followed
by abnormal collapse (Corrigan pulse), a “pistol-shot” sound
heard over the femoral artery (Duroziez murmur), or a bipha-
sic bruit heard during mild compression of the femoral artery
with the stethoscope. The jugular venous pulse and pressure are
generally normal in isolated aortic regurgitation unless a dilated
ascending aorta compresses the superior vena cava.
The apical impulse in chronic aortic regurgitation is diffuse,
hyperdynamic, and displaced inferiorly and leftward. A third
heart sound may be palpated. A diastolic thrill at the base of the
heart signifies severe aortic regurgitation.
Severe aortic regurgitation may cause partial diastolic clo-
sure of the mitral valve, decreasing the intensity of the first heart
sound. An early systolic ejection click can signify either a bicus-
pid aortic valve or a large stroke volume entering a dilated aortic
root. The second heart sound may be normal or abnormal (if the
aortic valve does not close properly). A third heart sound may be
present even without significant ventricular dysfunction, because
of the rapid early diastolic filling of the ventricle by the sum of
the transmitral and aortic regurgitant blood flow.
The characteristic auscultatory finding is a high-pitched
diastolic decrescendo murmur best heard along the left sternal
border. If it is most audible at the right sternal border, significant
aortic root dilatation is suggested. The murmur is heard best
with the diaphragm of the stethoscope, with the patient lean-
ing forward with breath held in full expiration. The duration,
rather than its loudness, of the murmur and the amplitude of
the pulse pressures correlate best with the severity of regurgita-
tion. If the murmur is musical or cooing, a cusp fenestration
or perforation is suspected. A coexistent aortic systolic mur-
mur does not necessarily imply the presence of aortic stenosis
and may be a functional flow murmur due to the ejection of an
abnormally high volume of blood during systole. The carotid
upstroke helps define coexistent aortic stenosis. In significant
aortic regurgitation, an additional diastolic apical rumbling
(Austin Flint) murmur may be detected. Amyl nitrite inhala-
tion softens an Austin Flint murmur but makes the rumbling
murmur of mitral stenosis louder. Late in the course of disease,
ventricular dilatation causes mitral annular dilatation and resul-
tant mitral regurgitation.
• There may be few typical physical findings in acute aortic regurgi-
tation.
• In patients with aortic regurgitation, wide pulse pressure implies
low diastolic pressure in addition to the large difference between
systolic and diastolic pressures.
• Physical findings in severe aortic regurgitation include a long dia-
stolic murmur, apical diastolic rumble, enlarged and displaced
apex, and peripheral findings of high output and rapid runoff.
Evaluation of Aortic Regurgitation
Acute, subacute, and mild chronic aortic regurgitation are not
necessarily associated with electrocardiographic abnormalities.
Chronic moderate or severe aortic regurgitation usually causes
features of left ventricular hypertrophy, but a substantial minor-
ity of such patients may not have ventricular hypertrophy by volt-
age criteria.
Chronic significant aortic regurgitation, with its associated
enlargement of the left ventricle, increases the radiographic car-
diothoracic ratio. The ascending aorta may be dilated, but it can
appear normal because the most proximal portion of the ascend-
ing aorta is hidden within the cardiac silhouette.
Echocardiography visualizes the ascending aorta and the
aortic valve and measures left ventricular size and function.
Findings may include aortic leaflet prolapse, diastolic vibration
of the anterior mitral leaflet, or premature closure of the mitral
valve. Assessment of the severity of aortic regurgitation involves
an integrated assessment of left ventricular size in conjunction
with a comprehensive Doppler investigation. Transesophageal
echocardiography images the thoracic aorta more completely
and may better assess aortic valvular vegetations or infectious
complications involving the aortic leaflets or annulus.
Other noninvasive methods can be used to examine ven-
tricular dimensions and myocardial function. Radionuclide
angiography has been studied the most and provides accurate
measurements of left ventricular volume and ejection fraction.
Computed tomography and magnetic resonance imaging also
may provide accurate measurements of volumes and ventricu-
lar function. Magnetic resonance imaging and cine-computed
tomography have the advantage of obtaining additional informa-
tion about the thoracic aorta. Especially in patients with bicus-
pid aortic valve, it is necessary to examine the thoracic aorta for
aortic aneurysm, dissection, and coarctation.
Noninvasive assessment of the cause and severity of aortic
regurgitation is usually sufficient, as long as the findings cor-
relate with those from the physical examination. Aortic root
angiography is still appropriate, however, in patients in whom
noninvasive testing was technically inadequate or gave results
discordant with clinical findings. Exercise testing also may be
valuable for determining functional capacity.
• Echocardiography assesses the cause and severity of aortic regur-
gitation in addition to left ventricular size and function.
• Transesophageal echocardiography is a useful adjunct for anatomi-
cal assessment but does not add to the quantitative assessment of
aortic regurgitation.
• Echocardiographic assessment of the severity of regurgitation
should not rely exclusively on color flow data.
• Aortography is indicated when noninvasive data are discordant,
technically limited, or differ from the clinical impression of regur-
gitant severity.
Natural History and Treatment
of Acute Aortic Regurgitation
In patients with severe acute aortic regurgitation, dyspnea and
heart failure develop rapidly, because of the large volume of
blood flowing backward into an “unprepared” ventricle. As a
result, left ventricular diastolic pressure increases and aortic dia-
stolic pressure decreases rapidly. In these patients, the diastolic
murmur may be short or even inaudible. The half-time on the
aortic regurgitation Doppler velocity signal is shortened. One of
the older indicators of severe aortic regurgitation is premature
closure of the mitral valve on M-mode echocardiography. On the
Doppler mitral inflow, the deceleration time is very short.
In patients with severe acute aortic regurgitation due to an
aortic dissection, emergency operation is indicated. This proce-
dure replaces both the aortic valve and the dissected ascending
aorta. This is a surgical emergency and must be performed as
soon as possible.
 37 Valvular Regurgitation
In patients with acute severe aortic regurgitation due to infec-
tive endocarditis, early operation is indicated, especially if the
organism is Staphylococcus aureus. These patients will rapidly
deteriorate from heart failure and cardiogenic shock. Even if
the patient has not had a prolonged course of antibiotics, early
operation should be performed after bolus infusion of antibiot-
ics. If possible, aortic valve homografts should be considered,
because they seem to be most resistant to superimposed intra-
operative and perioperative infection. In the patient with hemo-
dynamically significant mild to moderate aortic regurgitation
and active infective endocarditis, continued antibiotic therapy
with close observation is indicated. Transesophageal echocar-
diography should be performed to rule out an aortic abscess.
However, surgical standby should be maintained because acute
severe aortic regurgitation may rapidly develop and require
urgent operation.
Natural History of Chronic Aortic Regurgitation
Patients with severe chronic aortic regurgitation may go for
decades without symptoms. However, as with mitral regurgita-
tion, the long-standing volume overload does cause progressive
fibrosis and myocyte degeneration with subsequent left ventricu-
lar dysfunction (Figure 37.4). Because exercise results in shorten-
ing of the diastolic filling, and thus a decrease in the regurgitant
volume, patients may not have development of symptoms with
exertion for decades.
In patients with aortic regurgitation, there is an increase in
ventricular volume and an increase in ventricular mass. These
effects represent an attempt to normalize left ventricular wall
stress imposed on the ventricle by the large regurgitant vol-
ume. Both the afterload and the preload on the left ventricle are
increased in patients with aortic regurgitation. The ejection frac-
tion, therefore, is an accurate assessment of ventricular function,
unlike with mitral regurgitation.
Because the ventricle is able to compensate for aortic regurgi-
tation with both an increase in ventricular volume and an increase
Onset
severe
Survival
regurgitation LV
dysfunction
Progressive
LV dilatation
Onset
symptoms
Years
Figure 37.4. Natural History of Regurgitant Lesions. After the onset
of severe regurgitation, patients remain asymptomatic for years, dur-
ing which progressive left ventricular (LV) dilatation occurs. Incipient
left ventricular dysfunction may occur before the onset of symptoms.
If operation is performed at the onset of symptoms (dotted line), sur-
vival will be better than if no operation is performed. However, because
incipient left ventricular dysfunction may have already occurred, there
is a decreased survival as opposed to what would have occurred with
earlier operation. Similarly, even if operation is performed at the onset
of left ventricular dysfunction (dotted line), survival is still poorer than
what may have occurred if operation had been performed earlier.
401
in wall thickness, patients may have severe asymptomatic aortic
regurgitation for decades and maintain normal ventricular func-
tion. In most patients, ventricular function begins to deteriorate
before the onset of symptoms. Therefore, it is very important to
continue to follow patients up every 6 to 12 months with serial
measurements of ventricular volume and ejection fraction.
Overall, the outlook for patients with severe aortic regurgitation
is much better than that for patients with severe mitral regur-
gitation because ventricular dysfunction occurs late and can be
detected by noninvasive imaging.
Treatment of Severe Chronic Aortic
Regurgitation
All patients with aortic regurgitation should be instructed in
dental hygiene to prevent infective endocarditis. Although
initial studies suggested that vasodilators such as calcium
channel blockers or angiotensin-converting enzyme inhibi-
tors may be effective for halting the progression of left ven-
tricular dilatation, subsequent studies have shown that these
medications are not effective in patients with normal blood
pressure. Therefore, afterload reduction is indicated only in
patients who have severe aortic regurgitation and concomitant
hypertension.
The timing of operation for aortic regurgitation is better
understood than the timing for mitral regurgitation. Early opera-
tion with a prosthetic aortic valve has its own long- and short-
term complications and, therefore, should not be performed
prematurely. Because ejection fraction is an accurate measure-
ment of ventricular function, it is safe to continue to observe
patients with periodic measurements of ventricular volume and
function.
Overall, operation is indicated for patients with chronic
severe aortic regurgitation with the onset of any symptoms. As
with mitral regurgitation, although left ventricular dysfunction
may have already occurred, survival and outcomes are better
with operation than with medical management.
In asymptomatic patients with severe aortic regurgitation,
measurements of left ventricular size and function determine the
need for operation. An ejection fraction of less than 50% or an
end-systolic dimension of more than 55 mm indicates the pres-
ence of left ventricular systolic dysfunction. Patients in whom
these values are exceeded should be considered for operation
with aortic valve replacement.
In patients who have end-systolic dimension between 50 and
55 mm, it is also reasonable to consider earlier operation, espe-
cially if there has been a rapid progression of ventricular size. In
patients with an end-systolic dimension of less than 50 mm in
whom ventricular function with ejection fraction is maintained,
serial follow-up should be performed. The follow-up should
be performed initially at 3 months to ensure rapid progression
has not occurred. Subsequently, the evaluation should be per-
formed every 6 months if there is moderate to severe dilatation
(end-systolic dimension of 45–50 mm) or every year if there is
mild to moderate dilatation (end-systolic dimension of less than
45 mm). Exercise testing is indicated in the initial screening to
determine whether the patient truly is asymptomatic; this evalu-
ation provides an objective measurement of exercise tolerance.
The ventricular response to exercise is not a clear-cut indication
for operation.
When operation is considered, it is necessary to also examine
the size of the ascending aorta. In patients with a dilated aorta
more than 5.0 cm in diameter, concomitant aortic replacement
402 IV Valvular Heart Disease

and aortic valve replacement is indicated. In patients who have
rapid dilatation of the ascending aorta to more than 5.5 cm in
diameter, even if the aortic regurgitation is not severe, opera-
tion is indicated. In patients with an aortopathy such as is pres-
ent with a bicuspid valve or Marfan syndrome, the threshold to
replace the aorta is lower.
Suggested Reading
ACC/AHA 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 (Committee
on Management of Patients with Valvular Heart Disease). J Am Coll
Cardiol. 1998 Nov;32(5):1486–588.

Rheumatic Heart Disease
ANDREW G. MOORE, MD
Introduction
Rheumatic heart disease is the most serious manifestation of
acute rheumatic fever and is the most common cause of death in
acute rheumatic fever. Rheumatic heart disease is the end result
of carditis, which affects 30% to 80% of patients who have acute
rheumatic fever. Damage to the cardiac valves, which is the hall-
mark of rheumatic heart disease, may be chronic and progres-
sive; in conjunction with left ventricular dysfunction, it can lead
to congestive heart failure and death. Although both acute rheu-
matic fever and rheumatic heart disease are rare in more afflu-
ent populations in North America and Europe, where valvular
disease is now largely degenerative, pockets of resurgence have
appeared in these regions in recent years. Acute rheumatic fever
continues essentially unchecked in many developing countries,
where it is an important public health problem.
Acute Rheumatic Fever
Epidemiology
Although there has been no documented decrease in the incidence
of group A streptococcal pharyngitis in industrialized countries
over the past 100 years, the incidence of acute rheumatic fever has
decreased during this period in those countries. This decrease,
which began in the late 1800s, was accelerated by the develop-
ment of antibiotics in the 1950s. In Denmark, for example, the
incidence was approximately 250 per 100,000 in the 1860s, but
by 1980, it was between 0.23 per 100,000 and 1.88 per 100,000.
This contrasts with much greater incidence rates reported from
some developing countries and specific populations. In Sudan,
Abbreviations and acronyms are expanded at the end of this chapter.
38
for example, the incidence exceeds 100 per 100,000. The highest
reported rates are from indigenous populations in certain Pacific
islands, Australia, and New Zealand. Rates as high as 374 per
100,000 have been documented among Aboriginal school-aged
children in parts of Australia. Reported incidence rates vary
depending on whether detection is based on clinical findings or
on echocardiographic evaluation.
Several relatively recent publications have reported a resur-
gence of acute rheumatic fever in isolated, intermountain
populations in the United States, such as in West Virginia and
Tennessee. Overcrowding, close person-to-person contact, and
poor health care facilities seem to be consistent predisposing
conditions for the high incidence of acute rheumatic fever in
all these populations. An association with ethnic origin has not
been identified.
Acute rheumatic fever is a disease of the young, occurring
most commonly in preadolescent children. It is much rarer in
children younger than 5 years and adults older than 35 years.
Quite frequently, recurrent episodes occur through adolescence
and into early adulthood, and it is thought that the cumulative
effect of recurrent episodes of acute rheumatic fever leads to the
development of rheumatic heart disease. Both acute rheumatic
fever and rheumatic heart disease are more common in females
in many populations for various reasons, including increased
exposure to group A streptococci through child rearing and less
access to preventive medical care in some cultures.
Pathogenesis
The pathogenesis of acute rheumatic fever is incompletely under-
stood. Although streptococci have not been found in the heart
tissues of patients with acute rheumatic fever, strong circumstan-
tial evidence indicates that acute rheumatic fever results from
403
404 IV Valvular Heart Disease
an exaggerated immune response to pharyngeal infection with
group A streptococci:
1. Epidemics of streptococcal pharyngitis or scarlet fever precede out-
breaks of acute rheumatic fever
2. The incidence of rheumatic fever is less when streptococcal infec-
tions are treated
3. Recurrent episodes of acute rheumatic fever are prevented with pro-
phylactic antibiotic therapy
4. Elevated titers to any of the antistreptococcal antibodies (strep-
tolysin O, hyaluronidase, and streptokinase) are usually associated
with acute rheumatic fever
5. Group A streptococcal infection of the pharynx is necessary for
acute rheumatic fever to develop
6. Documented outbreaks of impetigo have caused glomerulonephritis,
but they have not been shown to cause acute rheumatic fever except
possibly in Australian Aborigines.
• Circumstantial evidence suggests that acute rheumatic fever results
from an exaggerated immune response to group A streptococci.
• Pharyngeal streptococcal infection is necessary for acute rheu-
matic fever to develop.
Clinical Features
Onset of acute rheumatic fever is typically characterized by an
acute febrile illness 2 to 4 weeks after an episode of pharyn-
gitis. Diagnosis is primarily clinical and is based on a constel-
lation of signs and symptoms, which were initially established
as the Jones criteria in 1944 (Box 38.1). Although the clinical
Box 38.1. Jones Criteria (1992 Revision) for Diagnosis
of Acute Rheumatic Fevera
Major manifestations
1. Carditis
2. Polyarthritis
3. Chorea
4. Erythema marginatum
5. Subcutaneous nodules
Minor manifestations
1. Fever
2. Arthralgias
3. Previous rheumatic fever or rheumatic heart
disease
4. Elevated C-reactive protein concentration or
erythrocyte sedimentation rate
5. Prolonged PR interval on electrocardiogram
Evidence of antecedent group
A streptococcal infection
1. Positive throat culture or rapid antigen test
positive for group A streptococci
2. Increased or increasing streptococcal antibody
titer
a matic heart disease. Acute rheumatic fever develops in 500,000
A firm diagnosis requires 1) 2 major manifestations or 1 major and
2 minor manifestations and 2) evidence of a recent streptococcal
infection. However, when chorea or carditis is clearly present, evidence
of an antecedent group A streptococcal infection is not necessary.
Box 38.2. World Health Organization Criteria for
Diagnosis of Acute Rheumatic Fever
1. First episode—same as the Jones criteriaa
2. Recurrent episode in a patient without
established RHD—same as for first episode
3. Recurrent episode in a patient with established
RHD—requires 2 minor manifestations from the
Jones criteriaa and evidence of an antecedent
group A streptococcal infection
Abbreviation: RHD, rheumatic heart disease.
a
See Box 38.1.
usefulness of the Jones criteria has been recently reaffirmed,
the main features have been modified or updated several times
to increase the specificity of the criteria. The WHO has more
recently developed criteria that favor sensitivity over specificity
(Box 38.2); they may be the preferred guidelines in countries
with populations at high risk of acute rheumatic fever.
Beyond fever, arthritis is typically the earliest manifestation
of acute rheumatic fever. In untreated patients, the arthritis is
classically described as “migrating” from joint to joint in quick
succession. The knees and ankles are often the first affected.
The duration of joint inflammation is short (≤1 week), and the
synovial fluid is generally sterile when examined.
Chorea, also known as Sydenham chorea or St Vitus dance,
is a neurologic movement disorder characterized by abrupt, pur-
poseless involuntary movements of the muscles of the face, neck,
trunk, and limbs. Muscular weakness (hypotonia) and behavioral
disturbances such as obsessive-compulsive behaviors are consid-
ered to be additional findings. The course of the syndrome is
variable. Symptoms tend to develop subtly, progressively worsen
over 1 to 2 months, and spontaneously resolve gradually after
3 to 6 months. Residual waxing and waning of symptoms may
occur for a year or more, and 20% of patients have recurrences
within 2 years.
Two classic skin lesions with well-described identifying char-
acteristics occur in acute rheumatic fever. Subcutaneous nodules
with diameters up to 2 cm occur for approximately 2 weeks over
bony surfaces or near tendons. The nodules are described as firm
and painless, and the overlying skin is not inflamed. They are
typically smaller and shorter lived than nodules of rheumatoid
arthritis. The second condition, erythema marginatum, is a clas-
sic skin rash that occurs early in the course of acute rheumatic
fever. The rash is evanescent, pink to red, and nonpruritic. It typi-
cally occurs on the trunk or the proximal limbs. The rash appears
as a ring that extends centrifugally, while the appearance of the
skin in the center of the ring returns to normal. The rash can per-
sist or recur after other symptoms of acute rheumatic fever have
passed. Interestingly, erythema marginatum and the subcutane-
ous nodules usually occur only in patients with carditis.
Carditis and Rheumatic Heart Disease
According to the WHO, at least 15.6 million people have rheu-
people every year, and carditis develops in 30% to 80% of them.
Carditis, which leads to valvular lesions and eventually chronic
rheumatic heart disease, is directly responsible for 233,000
deaths annually.
 38 Rheumatic Heart Disease
The term carditis refers to diffuse inflammation of the peri-
cardium, epicardium, myocardium, and endocardium. Valvular
involvement, with leaflet thickening, occurs as a rule. In addition
to thickening, the valve leaflets frequently display small rows
of vegetations called verrucae along their apposing surfaces.
Symptoms include tachycardia and mild or moderate chest dis-
comfort that is commonly pleuritic. The cardiac physical exam-
ination is often noteworthy for the presence of a pericardial
friction rub and, typically, new or changing murmurs. In young
patients, mitral valve regurgitation is the predominant cardiac
lesion; a new apical systolic murmur is characteristic. Aortic
regurgitation develops less commonly; a new basal diastolic
murmur is characteristic. The pulmonary and tricuspid valves
are rarely involved. Mitral stenosis becomes progressively more
common in early to mid adulthood. Heart block of all degrees is
seen on the electrocardiogram, and the most common radiologic
finding is cardiomegaly.
In myocarditis associated with rheumatic heart disease, the
myocardium is infiltrated by mononuclear cells, vasculitis is
apparent, and the interstitial connective tissue shows degenera-
tive changes. In the proliferative stage of acute rheumatic fever,
30% to 40% of biopsy samples have Aschoff bodies, which are
pathognomonic.
• Arthritis is usually the first symptom of acute rheumatic fever.
• Erythema marginatum and subcutaneous nodules are characteris-
tic skin lesions associated with acute rheumatic fever.
• In the proliferative stage of acute rheumatic fever, Aschoff bodies
are pathognomonic.
The use of echocardiography has contributed much toward
understanding the pathogenesis of valvular regurgitation in rheu-
matic carditis. At least initially, regurgitation appears to result
from geometric changes and stresses affecting the left ventricle
rather than from the rheumatic process directly involving the
valve leaflets. Prolapse of the anterior leaflet of the mitral valve
is the most common feature and is invariably associated with
a posteriorly directed jet from mitral regurgitation as seen on
Figure 38.1. Thickened mitral valve leaflets and subvalvular apparatus in rheumatic mitral stenosis.
405
color Doppler imaging. An increase in the mitral annular diam-
eter, especially around the posterior annulus, appears to be the
primary cause of leaflet prolapse. The typical mean annular
diameter in patients with carditis and mitral regurgitation is 37
mm, compared with 23 mm in healthy persons. Elongation of
the chordae to the anterior mitral valve leaflet is another feature
of severe mitral regurgitation associated with rheumatic carditis.
The elongation is thought to be from involvement by the primary
rheumatic process and from secondary exposure to increased
tensile stresses occurring during ventricular systole. Early in the
course of the disease, 2-dimensional echocardiography shows
some important differences between the rheumatic mitral valve
and the prolapsed mitral valve associated with degenerative or
myxomatous mitral valve disease. In rheumatic carditis, the
mitral leaflets are not thickened, redundant, or billowing. By
contrast, in myxomatous mitral valve disease, which tends to
involve the posterior leaflet, the leaflets are thickened, large, and
redundant, and they billow toward the left atrium.
• Mitral regurgitation in rheumatic heart disease results from pro-
lapse of the anterior leaflet of the mitral valve; mitral regurgitation
in myxomatous prolapse tends to involve the posterior leaflet.
Acute rheumatic fever is also the predominant cause of mitral
stenosis. Some degree of rheumatic involvement is present in
nearly all stenotic mitral valves excised at valve replacement.
Approximately 25% of patients with rheumatic heart disease
have pure mitral stenosis. Mitral stenosis is characterized by pro-
gressive thickening, fibrosis, and calcification of the leaflets and
chordae tendineae (Figure 38.1): 30% of patients have thicken-
ing of the commissures alone, 15% have thickening of the cusps,
and 10% have isolated thickening of the chordae tendineae. The
leaflets show fibrous obliteration, and the mitral valve orifice
becomes funnel-shaped, like a fish mouth (Figure 38.2). This
feature and the classic hockey-stick appearance of the anterior
mitral valve leaflet in diastole are apparent on 2-dimensional
echocardiography (Figures 38.3 and 38.4). It is not certain
whether the progressive fibrosis of rheumatic mitral stenosis is
the end result of a smoldering rheumatic process or the result of
406 IV Valvular Heart Disease

Figure 38.2. Fish-mouth appearance of a mitral valve affected by chronic rheumatic heart disease (arrow).

constant trauma from turbulent blood flow after initial deforma-
tion of the valve.
Subclinical Rheumatic Heart Disease
and Echocardiography
The most devastating effects of rheumatic fever and rheu-
matic heart disease occur in children and in young adults
during their most productive years. Therefore, early diagno-
sis to prevent recurrence of rheumatic fever and subsequent
valvular dysfunction is crucial in the global effort to reduce
morbidity and mortality. In children, the primary valvular
lesion is mitral regurgitation, which is identified in nearly all
patients. Aortic regurgitation occurs in approximately one-
fifth of patients.
Cardiac auscultation on physical examination is the primary
means of initially detecting a regurgitant murmur. However, the
clinical sensitivity and specificity of auscultatory skills vary
widely among practitioners, and in some patients with a diag-
nosis of rheumatic fever, pathologic valvular regurgitation will
not be clinically evident. This very early stage of the disease,
called subclinical carditis or subclinical rheumatic heart disease,

Figure 38.3. Two-dimensional echocardiographic parasternal short-axis view of the stenotic mitral valve in rheumatic heart disease, showing the
fish-mouth appearance of the mitral valve orifice (arrow).
 38 Rheumatic Heart Disease
Figure 38.4. Two-dimensional echocardiographic parasternal long-axis view of the mitral valve during diastole, showing the classic hockey-stick
appearance of the anterior mitral valve leaflet (arrow).
is thought to occur in approximately 12% to 21% of persons with
acute rheumatic fever.
Echocardiography is uniquely suited as a diagnostic tool to
detect subclinical rheumatic heart disease. In most countries,
echocardiography is available at least in hospitals, and the por-
tability of echocardiographic equipment makes it ideal for use in
the community and in more remote settings. Where it is avail-
able, echocardiography can identify a substantial proportion of
patients with subclinical rheumatic heart disease. The implica-
tions for early detection and long-term care are so important that
an echocardiographic examination is a recommended part of the
routine evaluation of all patients with confirmed or suspected
rheumatic fever.
Left Ventricular Dysfunction
The development of left ventricular dysfunction is common in
long-standing, untreated rheumatic heart disease. The course of
left ventricular dysfunction can be slow and insidious or, some-
times, very rapid and fulminate, mimicking viral myocarditis.
The associated development of congestive heart failure results
in significant morbidity and is potentially life threatening, often
requiring aggressive management.
Rheumatic myocarditis is rare without rheumatic valvular
disease. Myocarditis is frequently documented histologically at
autopsy, but without hemodynamically significant mitral regurgi-
tation, left ventricular dilatation and heart failure are rare. Recent
studies of the development of left ventricular systolic dysfunction
have attempted to clarify the relative contributions of 1) volume
overload from valvular regurgitation and 2) the direct effect on
the cardiac myocytes from the rheumatic process. After replace-
ment of the mitral valve or the aortic valve (or both), left ventric-
ular dimensions decrease and fractional shortening is preserved.
407
These findings suggest that myocardial contractile dysfunction is
not involved in rheumatic carditis.
Treatment of Acute Rheumatic Fever
Treatment of acute rheumatic fever consists of relieving symp-
toms, eradicating group A streptococci, and providing prophy-
laxis against further infections to prevent long-term cardiac
disease.
Relief of Symptoms
Fever and arthritis resolve quickly when salicylates are given
for initial treatment of acute rheumatic fever. Carditis, however,
should not be treated with salicylates. The incidence of residual
rheumatic heart disease did not decrease when salicylate therapy
was compared with no treatment or bed rest alone.
Congestive heart failure associated with severe carditis should
be treated with conventional therapy for heart failure. Many
physicians also treat carditis with corticosteroids, believing that
myocardial dysfunction will resolve faster with corticosteroids.
Results of randomized trials performed before the development
of echocardiography did not show a benefit with corticosteroids
compared with placebo or salicylates, but with newer imaging
modalities and corticosteroid preparations, the outcome of these
trials might have been different. The typical dosage of oral pred-
nisone is 2 mg/kg per day for 1 to 2 weeks, with a tapered dosage
thereafter.
Antibiotic Therapy
Treatment with penicillin is considered mandatory to eradicate
group A streptococcal infection from the upper respiratory tract.
Antibiotic therapy should be continued for at least 10 days, even
408 IV Valvular Heart Disease
if pharyngitis is not present at the time of diagnosis. Penicillin
V is recommended in a dosage of 250 mg 2 or 3 times daily for
children and 500 mg 2 or 3 times daily for adults. If adherence
to therapy is a concern, a depot penicillin such as penicillin G
benzathine, can be given as 1 IM dose. Children should receive
600,000 units IM if they weigh less than 27 kg. The common
dose for children who weigh more than 27 kg and for adults is
1.2 million units IM.
Antibiotic Prophylaxis
For patients who have a history of acute rheumatic fever or rheu-
matic heart disease, long-term antibiotic therapy is used as sec-
ondary prophylaxis. According to WHO documents, empirical
evidence underscores the importance of national programs against
acute rheumatic fever that are based on secondary prophylaxis,
which is a more cost-effective and practical intervention than pri-
mary prophylaxis (eg, sore throat screening programs in school-
aged children). Recurrence is most common within 2 years of the
original infection but can happen at any time. Prophylaxis should
be initiated when the acute episode resolves.
The best drug for prophylaxis is penicillin G benzathine
administered IM every 3 to 4 weeks. In direct comparisons with
oral treatments, this regimen has been more efficacious. The
recommended dose for adults is 1.2 million units IM. For chil-
dren who weigh less than 27 kg, the WHO recommends 600,000
units IM. Oral regimens include 1) penicillin V 250 mg twice
daily or 2) sulfadiazine 500 mg daily for children who weigh
less than 27 kg and 1,000 mg daily for children who weigh more
than 27 kg and for adults. Patients allergic to either penicillin or
sulfadiazine can take oral erythromycin 250 mg twice daily.
The duration of prophylaxis depends on several variables,
including the age of the patient, the time since the last attack,
the number of attacks, the severity of existing heart disease, and
the risk of exposure to streptococcal infections. General WHO
guidelines recommend that if no carditis has developed, prophy-
laxis should be continued for 5 years after an acute attack or
until the age of 18 years (whichever is longer). In cases of mild
or healed carditis, treatment should continue for 10 years past
the latest attack or until age 25 years (whichever is longer). In
cases of more severe carditis or valve surgery, treatment should
be lifelong. For additional discussion of antibiotic treatment and
prophylaxis, SEE CHAPTER 83 (“INFECTIVE ENDOCARDITIS”).
• Penicillin is the mainstay for primary and secondary prophylaxis
of rheumatic heart disease.
Management of Rheumatic Valvular Disease
The management of clinically significant mitral regurgitation
associated with rheumatic heart disease differs somewhat from
the management of degenerative mitral regurgitation, which
is much more common in the developed world. Since the mid
1990s, there has been a strong push to intervene surgically with
mitral valve repair earlier in the course of degenerative mitral
regurgitation to preserve left ventricular dimensions and sys-
tolic function. The benefits of mitral valve repair are not nearly
as clear for patients with mitral regurgitation from a rheumatic
cause. In active carditis, progressive fibrosis and leaflet defor-
mity may preclude a durable, long-term result. Mitral valve
repair has been disappointing in patients with chronic rheu-
matic mitral regurgitation and no evidence of active infection.
Reoperation is frequently required, and in 1 study, overall free-
dom from valve failure was only 66% after a mean follow-up
of 5 years. In developing countries, the situation is complicated
by other associated problems such as a lack of adequate health
care facilities and cardiac surgeons skilled in the repair of the
mitral valve. It is well documented that mitral valve replacement
results in a consistent reduction in left ventricular systolic func-
tion. However, if replacement is timed to occur when the end-
systolic diameter of the left ventricle is between 40 and 50 mm,
the long-term durability and success of the procedure may be a
reasonable compromise.
Pregnancy and Rheumatic Heart Disease
Although the incidence of rheumatic heart disease associated
with pregnancy declined in the 20th century, its occurrence is
still sufficiently common that physicians (particularly obstetri-
cians) should be familiar with managing this potentially life-
threatening disease.
As described above, clinically important rheumatic cardiac
lesions are predominately valvular. The most frequently affected
valve is the mitral valve, and the rheumatic valvular lesion most
likely to lead to a potentially serious outcome is mitral stenosis.
Pregnancy is associated with a 20% to 100% increase in blood
volume and a corresponding increase in cardiac output. Although
the volume overload associated with mitral or aortic regurgita-
tion is reasonably well tolerated, significant mitral stenosis is not.
The hypervolemia and tachycardia of pregnancy exacerbate the
transmitral gradient, and in patients who have a mitral valve area
less than 1.5 cm2, symptoms of congestive heart failure are prone
to develop.
Associated pulmonary hypertension or atrial fibrillation fur-
ther complicates the clinical picture. The sudden development of
atrial fibrillation with an uncontrolled ventricular heart rate can
precipitate heart failure in a patient who was previously hemody-
namically stable. Diuretics, digoxin, and β-blockade are main-
stays of treatment in this situation. Anticoagulation regimens
during pregnancy are described in CHAPTER 80 (“PREGNANCY AND
THE H EART”).
If a pregnant woman who received optimal medical manage-
ment remains symptomatic with New York Heart Association
class III or IV symptoms of congestive heart failure, clinically
significant risk is posed to both the mother and the fetus, and the
stenosis should be relieved. In the past, surgical commissurotomy
was used as palliative therapy. Now the treatment of choice is
percutaneous balloon mitral valvuloplasty performed at the end
of the second trimester or at the beginning of the third. Cardiac
surgery for valve replacement with a biologic tissue valve is an
option of last resort owing to the risk of fetal loss.
• Clinically significant mitral stenosis is poorly tolerated during
pregnancy.
Conclusion
Before the development of antibiotics, acute rheumatic fever
was the most frequent cause of valvular heart disease. Although
incidence rates have decreased dramatically in most developed
countries, acute rheumatic fever and, consequently, rheumatic
heart disease continue to flourish in developing countries. Since
this disease can cause severe morbidity and death, all physicians
should be familiar with the diagnosis.
Abbreviations
IM Intramuscularly
WHO World Health Organization
 39
Carcinoid and Drug-Related Heart Diseasea
HEIDI M. CONNOLLY, MD, and PATRICIA A. PELLIKKA, MD
Carcinoid Heart Disease
Progress in the medical and surgical management of patients
with metastatic carcinoid tumor has resulted in improved symp-
toms and survival. However, carcinoid heart disease remains a
major cause of morbidity and death. Limited medical treatment
options are available for patients with carcinoid valvular disease;
for those with severe valvular heart disease, valve replacement
has been increasingly used.
Diagnosis
Carcinoid tumors are rare, occurring in 1.2 per 100,000 to 2.1 per
100,000 people in the general population each year. The tumors
arise from enterochromaffin cells and are usually located in the
gastrointestinal tract, but they may originate in the lungs or ovary.
In 20% to 30% of patients, the initial manifestation is carcinoid
syndrome. Malignant carcinoid syndrome consists of flushing,
gastrointestinal tract hypermotility (secretory diarrhea), bron-
chospasm, and carcinoid heart disease caused by the release of
several vasoactive substances: 5-hydroxytryptamine (serotonin),
5-hydroxytryptophan, histamine, bradykinins, tachykinins, and
prostaglandins. The test for urinary 5-HIAA (24-hour collec-
tion) is specific and reproducible, providing a reliable biologic
marker for assessing tumor activity and the response to treatment.
The level of circulating plasma chromogranin A, a protein pro-
duced by neuroendocrine cells, may also be used as a marker for
diagnosis and follow-up in selected cases.
a
Portions previously published in Connolly HM, Pellikka PA. Carcinoid
heart disease. Curr Cardiol Rep. 2006 Mar;8(2):96–101 and Connolly
HM, McGoon MD. Obesity drugs and the heart. Curr Probl Cardiol.
1999 Dec;24(12):745–92. Used with permission.
Abbreviations and acronyms are expanded at the end of this chapter.
Systemic and Regional Therapy for Metastatic
Carcinoid Tumor
Octreotide acetate (Sandostatin), a somatostatin analogue, is
a synthetic octapeptide that binds to subtypes of somatosta-
tin receptors and inhibits the secretion of bioactive substances
responsible for the carcinoid syndrome. Treatment relieves symp-
toms in more than 70% of patients and is now available as a long-
acting, once-a-month intramuscular injection (Sandostatin LAR
Depot). This microencapsulated depot formulation provides lon-
ger steady-state levels and similar therapeutic benefit with less
discomfort and inconvenience.
Patients presenting with bulky metastatic liver disease and
no other metastatic disease are candidates for surgical debulk-
ing (partial hepatectomy). Alternatively, tumors that cannot be
debulked surgically may be debulked by catheter-based hepatic
artery embolization. The use of the somatostatin analogue is rec-
ommended in conjunction with debulking because of its static
effect on the tumor.
Cardiac Involvement
Carcinoid heart disease eventually occurs in more than 50% of
patients with carcinoid syndrome, and it may be the initial man-
ifestation of metastatic carcinoid tumor in as many as 20% of
patients (Figure 39.1). Deposition of a matrix-like material on
the valves and endocardium of the right side of the heart results
in retraction and fixation of the tricuspid leaflets, reduced leaflet
motion, and lack of central coaptation. Clinically, these changes
produce severe tricuspid valve regurgitation and, less commonly,
tricuspid valve stenosis. The pulmonary valve is likewise affected,
with thickening and retraction of the leaflets, resulting in a com-
bination of valvular regurgitation and stenosis. In patients with
409
410 IV Valvular Heart Disease
Figure 39.1. Carcinoid Heart Disease. Schematic shows tricuspid and pulmonary valve disease and endocardial plaque deposition. Metastatic
lesions are also present.
advanced carcinoid heart disease, right-sided endocardial lesions
may develop.
The mechanism of valve injury in carcinoid heart disease is
incompletely understood but is likely due to activation of sero-
tonin 2B receptors on the surface of the valves by serotonin.
Circulating serotonin levels are higher in patients with carci-
noid heart disease than in carcinoid patients without cardiac
involvement, implying that serotonin directly contributes to the
development and progression of valvular disease. Bulky hepatic
metastases allow large quantities of vasoactive tumor products to
reach the right heart without being inactivated.
Rarely, carcinoid tumor originates in the ovary. Because the
ovarian veins bypass the portal circulation and enter the systemic
venous circulation directly, cardiac involvement can occur with-
out liver metastases.
The preferential involvement of right-sided cardiac structures
in carcinoid heart disease is most likely related to inactivation of
the humoral substances by the pulmonary circulation. Left-sided
valvular pathology is present in 10% of patients who have carci-
noid heart disease and is typically found in patients who have an
intracardiac shunt or primary bronchial carcinoid. Presumably,
serotonin-rich blood enters the left heart chambers without pul-
monary deactivation. Rarely, left-sided valvular disease occurs
in patients with severe, poorly controlled carcinoid syndrome.
Left-sided carcinoid valvular disease typically consists of valve
thickening and regurgitation.
Unfortunately, treatment of the carcinoid tumor does not
appear to result in regression of valvular disease. However, the
posttreatment levels of 5-HIAA seem to predict the development
or progression of valvular abnormalities, with a threshold of about
100 mg/24 hours. In a report of 71 Mayo Clinic patients who had
serial echocardiograms, the patients did not have a reduced risk of
progressive valvular disease with somatostatin analogue therapy,
hepatic dearterialization, or chemotherapy. In fact, chemother-
apy was statistically correlated with a higher rate of progressive
valvular disease, perhaps because this form of therapy is usually
reserved for patients with more aggressive carcinoid tumors.
Clinical Features of Carcinoid Heart Disease
Early in the course of carcinoid heart disease, symptoms are
insidious and include fatigue and dyspnea on exertion. Patients
with severe tricuspid and pulmonary valve disease may be
asymptomatic for months. Right-sided heart failure eventually
develops with progressive cardiac disease. Without treatment,
the median survival with malignant carcinoid syndrome is 38
months from the onset of systemic symptoms. Clinical evidence
of carcinoid heart disease with New York Heart Association
class III or IV symptoms is associated with a median survival of
only 11 months.
The physical findings among patients with carcinoid heart dis-
ease likewise may be subtle early in the course of the disease. The
murmurs of tricuspid and pulmonary valve disease are typically
difficult to detect because of the low pressure in the pulmonary
circulation. Elevation of the jugular venous pressure with a prom-
inent v wave is often the earliest finding on physical examina-
tion. As the valvular disease worsens, peripheral edema, ascites,
and pulsatile hepatomegaly occur. In addition to the murmurs
of tricuspid and pulmonary valve regurgitation, cardiac findings
include a palpable right ventricular impulse. Less frequently, mur-
murs of pulmonary stenosis and tricuspid stenosis may be noted.
For unknown reasons, the electrocardiogram in advanced car-
cinoid heart disease demonstrates low-voltage QRS complexes.
Chest radiography shows cardiomegaly with prominence of the
right cardiac chambers.
Echocardiographic Features
of Carcinoid Heart Disease
Thickening and retraction of immobile tricuspid valve leaflets
with associated severe tricuspid regurgitation are characteristic
echocardiographic features (Figure 39.2). Early in the course, the
tricuspid valve leaflets may maintain some mobility; less com-
monly, tricuspid valve stenosis is noted. The pulmonary valve
is usually also involved; the characteristic pulmonary valve fea-
tures include immobility of the valve cusps that may be difficult
 39 Carcinoid and Drug-Related Heart Disease 411

Figure 39.2. Advanced Carcinoid Tricuspid Valve Disease. A, Transthoracic echocardiographic right ventricular inflow view shows that the
septal and anterior tricuspid valve leaflets (arrows) are severely thickened and retracted. They are fixed in a semiopen position, resulting in marked
deficiency of central systolic coaptation. B, Color flow Doppler imaging of the same valve shows severe central tricuspid regurgitation through the
fixed open tricuspid orifice. RA indicates right atrium; RV, right ventricle; TR, tricuspid regurgitation.

to visualize owing to cusp retraction (Figure 39.3). Tricuspid
regurgitation and pulmonary valve regurgitation eventually
result in progressive right ventricular volume overload and right
ventricular diastolic pressure elevation. Other echocardiographic
findings among patients with carcinoid heart disease include
left-sided valvular pathology in approximately 10% and myocar-
dial metastases in less than 5%. Pericardial effusions are com-
monly noted, but these are rarely hemodynamically significant.
Pathology of Carcinoid Valvular Disease
The affected tricuspid and pulmonary valves in carcinoid heart
disease are white with thickened leaflets and chordal struc-
tures (Figures 39.4 and 39.5). The carcinoid plaque consists of
smooth muscle cells and myofibroblasts surrounded by an extra-
cellular matrix and an overlying endothelial cell layer. The mor-
phology of the valve leaflet is not disrupted, and the carcinoid
plaque generally affects the ventricular aspect of the tricuspid
valve leaflets and the arterial aspect of the pulmonary valve
cusps. Plaques may exhibit neovascularization and chronic
inflammation.
Management of Carcinoid Heart Disease
Limited medical therapeutic options are available for patients
with symptomatic right heart failure related to carcinoid heart
disease. Cardiac surgery is the only effective treatment of car-
cinoid heart disease and should be considered for symptomatic
patients whose metastatic carcinoid tumor and carcinoid syn-
drome are well controlled. Patients with severe carcinoid heart
disease should be considered for cardiac surgery when any of the
following develop: 1) symptoms of right heart failure, 2) impaired
exercise performance, 3) progressive right heart enlargement, or
4) right ventricular systolic dysfunction. Rarely, surgery is also
412 IV Valvular Heart Disease

A B

Figure 39.3. Carcinoid Involvement of the Pulmonary Valve. A, Transthoracic echocardiogram shows the difficulty in visualizing the pulmo-
nary valve (PV) (arrow), which is characteristic of carcinoid PV involvement. The pulmonary annulus is narrowed. B, Continuous wave Doppler
examination shows pulmonary stenosis and regurgitation.

performed in minimally symptomatic patients with severe carci- embolization or ligation may protect prosthetic valve tissue from
noid heart disease in anticipation of hepatic surgery. the adverse effects of serotonin and other vasoactive peptides,
Patients with severe valvular disease are not candidates for but premature bioprosthesis degeneration may occur.
partial hepatic resection or liver transplant owing to the intraop- Mechanical tricuspid prostheses have a more favorable hemo-
erative risk of hepatic hemorrhage induced by the elevated right- dynamic profile than most bioprostheses; however, mechanical
sided pressures. prostheses are not ideal for patients with carcinoid heart disease
Cardiac surgery has been successful in reducing or reliev- because subsequent surgical procedures for tumor control are
ing the cardiac symptoms of many patients with carcinoid heart often required and are complicated by anticoagulation manage-
disease. Data from Mayo Clinic patients with metastatic carci- ment. In addition, the risk of mechanical tricuspid prosthesis
noid syndrome suggest that early, regular cardiac evaluation and thrombosis is about 4% per year.
cardiac surgical intervention before advanced right heart failure
develops may decrease surgical mortality (currently <10%) and
Anesthetic Management
improve the prognosis.
Anesthesia can precipitate carcinoid crisis, which is character-
ized by profound flushing, extreme changes in blood pressure,
Choice of Valve Prosthesis
In the past, tricuspid valve replacement with a mechanical
prosthesis was recommended for patients who had carcinoid
heart disease because it was assumed that a bioprosthetic valve
could be damaged by vasoactive tumor substances. Treatment
with synthetic somatostatin and hepatic artery interruption by

Figure 39.4. Tricuspid Valve in Carcinoid Heart Disease. Gross Figure 39.5. Pulmonary Valve in Carcinoid Heart Disease. Gross
pathologic specimen. The tricuspid valve leaflets are thickened, pathologic specimen. The valve cusps are thickened, retracted, and
retracted, and shortened. (See “Credit Line” section.) shortened. (See “Credit Line” section.)
 39 Carcinoid and Drug-Related Heart Disease
bronchoconstriction, and arrhythmias. Carcinoid crisis can be
fatal. Thus, carcinoid symptoms should be controlled with an
octreotide analogue before anesthesia; large doses of octreotide
analogue are often required perioperatively and postoperatively.
Drug-Related Heart Disease
Anorexigens and Valvular Heart Disease
In the United States, the increasing prevalence of obesity, the
altered perception of desirable body habitus, and an effective
advertising campaign contributed to a striking increase in the
sales of the diet drugs phentermine, fenfluramine, and dexfen-
fluramine between 1994 and 1996. Fenfluramine and dexfenflu-
ramine increase the release of the neurotransmitter serotonin and
decrease its reuptake.
A report published in August 1997 described 24 women
who had valvular heart disease that was atypical for degen-
erative or rheumatic disease; 5 required valve surgery. All
these women had been treated with fenfluramine and phen-
termine for an average of 11 months. The FDA requested
reporting of similar cases, and by September 1997, more than
100 spontaneous occurrences that met the case definition
had been reported to the FDA. A case of diet-drug–related
valvular disease was defined as valvular disease in a person
who previously did not have known valvular disease but who
used appetite suppressants and subsequently had echocardio-
graphic evidence of 1 or both of the following: 1) aortic valve
regurgitation that was at least mild in severity and 2) mitral
valve regurgitation that was at least moderate in severity. This
degree of valvular heart disease is expected in less than 1% of
young healthy persons.
The prevalence of valvular disease meeting the case defini-
tion ranged from 30% to 38%. An important additional finding
was that the prevalence of disease was time dependent; 35% of
patients treated for longer than 6 months had disease, whereas
only 22% of those treated for less than 3 months exhibited val-
vular damage on echocardiography. In addition, 30% of patients
taking dexfenfluramine (with or without phentermine) had val-
vular regurgitation. Of the patients who had valvulopathy, 83%
had at least mild aortic regurgitation and 21% had at least moder-
ate mitral regurgitation either alone or in combination with other
valvular disease.
The 32% overall prevalence of valvular lesions meeting the
FDA case definition in exposed persons was markedly higher
than what would be expected in the general population (2%–5%).
Compared with a healthy population, diet-drug–treated individu-
als had a 15-fold increased risk of valvulopathy. On the basis
of the FDA prevalence data, fenfluramine and dexfenfluramine
were withdrawn from the US market in 1997. Subsequent reports
supported the association between the use of fenfluramine or
dexfenfluramine and valvular regurgitation but differed on the
strength and clinical significance of this association. The dura-
tion and dose of appetite suppressant use appeared to be related
to the incidence of valvular disease; clinically important disease
was not likely to develop after short-term exposure.
The American College of Cardiology and the American
Heart Association included a section on management of patients
exposed to anorexigens in their “Guidelines for the Management
of Patients with Valvular Heart Disease.” The recommendations
include the following:
1. All exposed individuals should be evaluated with a history and phys-
ical examination.
413
2. An echocardiogram is suggested for patients with symptoms or signs
of cardiovascular disease or an unreliable physical examination.
3. Treatment and follow-up depend on the type of valvular disease.
The optimal timing of cardiac surgery for patients with diet-
drug–related valvular disease is unknown.
The echocardiographic appearance of valvular disease related
to anorexigen use is characterized by thickening of 1 or more
valves with associated valvular regurgitation. The features are
similar to rheumatic valvular disease, but the predominant val-
vular lesion is regurgitation rather than stenosis.
There is unequivocal evidence that the diet drugs fenfluramine
and dexfenfluramine cause valvular heart disease and pulmonary
hypertension. Because valvular disease in patients treated with
diet drugs or ergot alkaloid derivatives is pathologically similar
to valvular disease in patients with carcinoid heart disease, sero-
tonin may have a role in the valvular injury. The development of
drug-induced valvular heart disease is probably a manifestation
of the complex interaction between various factors, including
serotonin, serotonin 2B receptors on valves, and the serotonin
transporter. Individual susceptibility to drug-induced valvular
disease could be partially attributable to the interplay of these
factors or to polymorphisms in gene expression.
Explanted valve pathologic specimens from patients treated
with diet drugs showed intact leaflet and cusp architecture with
a layer that appeared to be adhered to the valve (Figure 39.6).
The adherent layer consisted of abundant myofibroblasts in an
extracellular matrix. Although regression of valvular disease
after cessation of diet-drug therapy has been reported, the natu-
ral history is uncertain.
Ergot Alkaloids and Valvular Heart Disease
Methysergide and ergotamine are used in various preparations
for the prophylaxis and treatment of migraine headaches. The
chemical structures of ergotamine, methysergide, and serotonin
are similar. Ergotamine is a naturally occurring ergot alkaloid,
and methysergide is a semisynthetic derivative of the ergot alka-
loids. Ergotamine is a partial serotonin receptor antagonist in
various smooth muscles and a partial agonist in certain blood
vessels. Long-term ingestion of methysergide or ergotamine can
produce endocardial thickening that can involve the valves and
cause regurgitant or stenotic lesions.
Ergot alkaloid–associated heart disease is caused by endomyo-
cardial fibrosis that extends onto the valve structures and distorts
the anatomy of the valves, producing lesions that cause stenosis
or regurgitation (or both). All patients in reports of ergot alkaloid
valvular disease used preparations of ergotamine suppositories
or methysergide for more than 6 years (usually up to 20 years).
Long duration of exposure seems to be the most important fac-
tor in producing valvular disease. Methysergide and ergotamine
have been associated with lesions of all 4 valves; however, mitral
and aortic regurgitation are the most common valvular lesions.
Echocardiography in ergot alkaloid valvular disease typically
shows thickening of 1 or more valves, with associated regurgita-
tion and stenosis. The features are similar echocardiographically
to those in rheumatic valvular disease.
The structural similarity of ergotamine, methysergide, and
serotonin and the similarity of the valvular lesions in ergot alka-
loid valvular disease and carcinoid syndrome suggest a common
pathophysiologic mechanism. However, ergot alkaloid valvular
disease commonly produces left-sided valvular lesions, whereas
carcinoid-associated valvular disease is usually restricted to the
right side. The macroscopic and histopathologic features of ergot
414 IV Valvular Heart Disease

Figure 39.6. Regurgitant Mitral Valve From Patient Who Used Fenfluramine and Dexfenfluramine. A, Gross pathologic specimen. B, Microscopic
view (hematoxylin-eosin). (See “Credit Line” section.)

alkaloid valvular disease are identical to those of carcinoid and
diet-drug–related valvular disease. These pathologic similarities
support the concept that a complex interaction between various
factors is involved in the development of drug-related valvular
disease.
Regression of valvular disease has never been documented;
however, diminution of murmurs has been reported. Surgical
intervention is required in patients with advanced, symptomatic
valvular disease.
Pergolide and Valvular Heart Disease
Pergolide is an ergot-derived dopamine receptor agonist used to
treat Parkinson disease and restless legs syndrome. Long-term use
 39 Carcinoid and Drug-Related Heart Disease 415

has been associated with retroperitoneal, pleural, and pericardial Suggested Reading
fibrosis. Drug-induced restrictive valvular heart disease has also Bhattacharyya S, Schapira AH, Mikhailidis DP, Davar J. Drug-induced
been described. In echocardiographic and histologic evaluations fibrotic valvular heart disease. Lancet. 2009 Aug 15;374(9689):
of surgically explanted valves, the abnormalities suggested car- 577–85.
cinoid involvement, ergot alkaloid use, or anorexigen treatment. Moller JE, Pellikka PA, Bernheim AM, Schaff HV, Rubin J, Connolly
The echocardiographic appearance of pergolide-related valvular HM. Prognosis of carcinoid heart disease: analysis of 200 cases over
disease is notable for the thickening of 1 or more valves, with two decades. Circulation. 2005 Nov 22;112(21):3320–7. Epub 2005
associated regurgitation. The initial estimate of the frequency of Nov 14.
valvular heart disease in patients taking pergolide was very low
(1 in 20,000), although researchers later suggested a higher inci- Abbreviations
dence. Pergolide has been withdrawn from the US market, but FDA US Food and Drug Administration
pergolide and cabergoline are in use in the rest of the world. 5-HIAA 5-hydroxyindoleacetic acid

Prosthetic Heart Valves
MARTHA A. GROGAN, MD, and FLETCHER A. MILLER JR, MD
Prosthetic valves can only approximate normal human valve
hemodynamics, and they carry the risk of unique complications,
such as structural failure, thrombosis, hemolysis, and infections.
Valve Types
Valve prostheses are classified as mechanical and bioprosthetic
valves. Mechanical valves are subdivided into caged-ball, tilt-
ing-disk, and bileaflet (Figure 40.1). Bioprosthetic valves are
subdivided into stented heterografts, homografts, and stentless
heterografts. The homograft and stentless heterograft valves are
designed for implantation in the aortic or pulmonic position. All
the other valve types can be implanted in any valve position.
Specific types of prosthetic valves are listed in Box 40.1.
Each type of prosthesis is manufactured in several different
sizes, ranging from 19 mm to 33 mm in diameter. Aortic pros-
theses are generally available in odd-numbered sizes from 19
through 31, whereas mitral and tricuspid prostheses are available
in odd-numbered sizes from 23 through 33. The size refers to the
external diameter of the sewing ring in millimeters. The size of
the prosthetic valve greatly influences its hemodynamics, partic-
ularly in the aortic position. A large number of prosthetic valves
have been manufactured since the 1970s. Some of the most com-
monly used valves are reviewed below.
Starr-Edwards Valve
The currently available Starr-Edwards valves have a cage that is
constructed from the alloy Stellite 21 and a Silastic poppet (ball)
that is specially cured to prevent lipid accumulation (which can
Abbreviations and acronyms are expanded at the end of this chapter.
416
40
result in ball variance). The struts of the modern Starr-Edwards
prosthesis are not covered with cloth.
Medtronic-Hall Valve
The Medtronic-Hall valve has a tilting disk made of pyrolytic car-
bon. The valve housing, which is machined from a single block of
titanium, is composed of the valve ring with a sigmoid strut that
projects into the center of the ring and passes through a hole in the
center of the disk. The tilting disk is supported by a smaller strut
and 2 lugs that project from the ring. The disk tilts to an opening
angle of 75° for aortic prostheses and 70° for mitral prostheses.
Björk-Shiley Valve
The Björk-Shiley valve also has a disk made of pyrolytic carbon.
The standard disk is planar on one side and convex on the other.
This disk is restrained by an inlet strut and an outlet strut. In
1978, the 2 struts were modified and the disk was changed to a
convexoconcave profile (the C-C model). This new model was
available in versions that tilted either to 60° or to 70° of the
opening angle. Only the 60° valves were implanted in the United
States. Subsequently, the Björk-Shiley C-C model was found to
be subject to fractures of the outlet strut, with disk escape (see
below). A later model, referred to as the monostrut valve, used a
modified outlet strut. In this model, the entire ring and the struts
are machined from one piece with no welds.
St Jude Medical Valve
The most widely used mechanical prosthesis is a bileaflet valve
from St Jude Medical. Its housing and leaflets are manufactured
entirely from pyrolytic carbon. The leaflets move in a slot with
 40 Prosthetic Heart Valves 417

complex opening and closing motions that are a combination of
sliding and tilting. The leaflets open to a nearly parallel posi-
tion (85°). The closing angle varies from 120° to 130°, depend-
ing on valve size, with valves 25 mm or less having the smaller
closing angle. Other examples of bileaflet prostheses include
CarboMedics and Duromedics valves.
Heterograft Valves
For all porcine heterograft prostheses, a pig aortic valve is
used regardless of whether the valve is implanted in the aortic,
pulmonic, mitral, or tricuspid position. The pig aortic valve is
mounted on flexible stents covered with Dacron.
Heterograft valves also have been constructed from pericar-
dium sutured to flexible, cloth-covered stents. The Ionescu-Shiley
valve was one of the original pericardial valves. A design flaw
predisposed this valve to sudden rupture of the cusps. Currently,
the Carpentier-Edwards pericardial valve is commonly used, and
long-term studies have demonstrated a lower rate of structural
valve deterioration compared with that of porcine heterografts.
Although stentless porcine aortic prostheses were initially
reported to have superior hemodynamics, long-term studies have
not shown improved clinical outcomes compared with stented
bioprostheses. With 12-year follow-up, patients who received
the Toronto Stentless Porcine Valve had an increased rate of
reoperation for structural valve deterioration when compared
with the Carpentier-Edwards pericardial prosthesis, especially
in younger patients. Stentless prostheses, such as the Medtronic
Freestyle porcine prosthesis, include not only the valve but also
the pig ascending aorta and may be considered as an alterna-
tive to a homograft aortic valve and root replacement for patients
with active infective endocarditis. For most patients who do not
require aortic root replacement, the choice of prosthesis is a
stented heterograft or mechanical prosthesis.
Homograft Valves
Homografts are valves harvested from cadavers and cryopre-
served. They frequently are used in patients who have infective
endocarditis with perivalvular extension of infection. Homografts

Figure 40.1. Mechanical Prostheses. A, Starr-Edwards prosthesis in closed position. B, Medtronic-Hall prosthesis in fully open position (central
strut fits through hole in disk; open disk creates major and minor orifices). C and D, St Jude Medical and CarboMedics bileaflet prostheses respec-
tively in fully open position (there are 2 large orifices and a smaller central orifice).
418 IV Valvular Heart Disease
Figure 40.1. (Continued)
have a low rate of thromboembolism, but long-term studies have
not demonstrated improved durability when compared with het-
erograft prostheses. As with all human donor tissue, availability
of homograft valves is a limiting factor.
Principles of Prosthetic Valve Selection
Similar early and late mortality rates have been reported for
patients with either mechanical or tissue valves. Because biopros-
thetic valves are less durable, the need for reoperation is higher
than for those with mechanical valves. Mechanical valves have
higher thrombogenicity and a higher anticoagulation-related
bleeding rate than tissue valves. The choice of prosthesis for an
individual is complex and is best made with careful consider-
ation of clinical factors and a shared decision-making approach
with the patient. General guidelines for prosthetic valve selection
are outlined below.
• Mechanical valves are generally preferred in patients who require
long-term anticoagulation for other reasons, such as an existing
mechanical prosthesis in another position or atrial fibrillation.
• Patients unwilling or unable to take warfarin or to comply with
long-term monitoring should receive tissue valves.
Mechanical valves are generally preferred for patients younger
than 60 to 65 years who are undergoing aortic valve replacement
and for patients younger than 65 to 70 years who are undergo-
ing mitral valve replacement. An informed discussion with the
patient is crucial to discuss the risks of lifelong anticoagulation
compared with the need for reoperation.
Patients with a history of bleeding disorders or high risk
of bleeding and those with a life expectancy of less than 10 to
12 years should usually receive tissue valves. Previously, because
of concerns about accelerated calcification of the bioprosthesis,
mechanical prostheses were recommended for patients receiving
dialysis. However, studies demonstrating poor overall survival
after valve replacement surgery in dialysis patients suggest that a
bioprosthesis is preferable for most of these patients.
Women with critical valve disease who desire pregnancy
pose a therapeutic challenge. Warfarin increases the risk of fatal
fetal bleeding and is teratogenic, but heparin anticoagulation is
problematic because of the hypercoagulable state of pregnancy.
Bioprostheses are less durable in young women, and a second valve
operation will almost always be required later if a tissue valve is
implanted first. The choice of prosthesis in women of childbearing
age should be individualized on the basis of the overall clinical sit-
uation after an informed discussion of the therapeutic challenges.
 40 Prosthetic Heart Valves 419

more than 40,000 patients in Europe. The Edwards Sapien valve

Box 40.1. Types of Prosthetic Heart Valvesa has been approved by the FDA for use in patients with severe aor-
Bioprostheses tic stenosis who are at high risk for surgical aortic valve replace-
ment. The Medtronic Melody valve is approved for transcatheter
Porcine (stented) pulmonary valve replacement in patients with dysfunctional right
Hancock I ventricular outflow tract conduits.
Hancock II
Hancock MO (modified orifice) Complications of Prosthetic Valves
Carpentier-Edwards
Medtronic Intact Specific complications of prosthetic valves are outlined in
Box 40.2.
Pericardial
Ionescu-Shiley Perivalvular Leak
Carpentier-Edwards pericardial
Mitroflow Perivalvular regurgitation is always abnormal (Figure 40.2). The
clinical significance of perivalvular regurgitation is determined
Homograft
by the volume of regurgitation, and the presence or absence of
Porcine (stentless) mechanical hemolysis. Pathologic transvalvular regurgitation
Toronto Stentless Porcine Valve must be distinguished from the normal regurgitation that is
Medtronic Freestyle “built into” various prosthetic valves. All prosthetic valves have
associated closing volume regurgitation, which is the volume of
Mechanical prostheses blood displaced by the occluder when it closes (Figure 40.3).
Caged-ball Tilting-disk and bileaflet prostheses also have a built-in leakage
volume, true transvalvular regurgitation that travels between the
Starr-Edwards
Braunwald-Cutter disk or leaflets and the housing and also between closed bileaflets
Smeloff-Cutter (Figure 40.4). This leakage volume serves to wash the surface of
Magovern-Cromie the disk or leaflets. Earlier models that were designed without
any leakage volume have an unacceptably high incidence of valve
Tilting-disk thrombosis. Whereas closing volume depends on occluder size,
Björk-Shiley length of travel, and speed of closure, leakage volume depends
Björk-Shiley convexoconcave on the size of the gap between the occluder and the rim of the
Medtronic-Hall housing. It increases with valve size and with decreasing heart
Lillihei-Kaster rate. In the extreme, the sum of the closing and leakage volumes
Omniscience may be as great as 10 mL per beat.
Sorin
Bileaflet • Perivalvular leak is always abnormal.

St Jude Medical
CarboMedics
On-X
a
Valves in boldface are most likely to be encountered in modern
practice.
Ross Procedure
In the Ross procedure, a pulmonary autograft (the patient’s
own pulmonary valve and main pulmonary artery) is placed in
the aortic position, the coronary arteries are reimplanted, and a
homograft is placed in the pulmonary position. This procedure
usually is performed in children and adolescents. Advantages of
the pulmonary autograft include the lack of a need for anticoagula-
tion and growth of the valve and root in children and adolescents.
However, the procedure is technically demanding, and long-term
studies have demonstrated an increased need for reoperation,
which is often complex and carries an increased risk of mortality
compared with traditional heterograft valve replacement.
Transcatheter Valve Replacement
Box 40.2. Complications of Prosthetic Heart Valves
1. Structural deterioration of the valve leading to
stenosis or regurgitation
2. Nonstructural dysfunction—an abnormality,
not intrinsic to the valve, resulting in stenosis
or regurgitation (exclusive of infection and
thrombosis)
Pannus
Suture entrapment
Perivalvular leak
Inappropriate sizing (patient-prosthesis mismatch)
Clinically important hemolytic anemia
3. Thromboembolism
Neurologic deficit
Peripheral emboli
Acute myocardial infarction after operation, if
coronary arteries are known to be normal

Transcatheter valve replacement is an important advance in 4. Valve thrombosis

the treatment of patients with valvular heart disease. Its use 5. Anticoagulation-related hemorrhage
is expected to expand to wider application in coming years. 6. Prosthetic valve endocarditis
Transcatheter aortic valve replacement has been performed in
420 IV Valvular Heart Disease
LA
LV
RA
SR
Figure 40.2. Transesophageal echocardiogram, horizontal plane of
a Starr-Edwards mitral prosthesis, showing a relatively narrow peripros-
thetic jet (arrows) originating around the medial portion of the sewing
ring (SR). Note mosaic appearance. LA indicates left atrium; LV, left
ventricle; RA, right atrium. (Previously published. See “Credit Lines”
section.)
Structural Failure
Structural valve deterioration is far more common for bioprosthe-
ses than for mechanical prosetheses. Degenerative calcification
most often results in leaflet tears during transvalvular regurgi-
tation, but it may also result in stenosis. Nonstructural lesions
such as pannus and suture entrapment are more common with
mechanical prostheses, whereas perivalvular leaks occur with
both types of valves. Clinically important hemolytic anemia is
usually the result of perivalvular regurgitation, particularly if the
regurgitant jet is directed against prosthetic material.
Historically, the most common structural dysfunction of a
mechanical prosthesis has been fracture of the outlet strut, which
LA
RA
SR
Figure 40.3. Transesophageal echocardiogram, horizontal plane,
4-chamber view. Normal Starr-Edwards tricuspid prosthesis. In this
systolic frame, the poppet and cage are not visible in the right ventricle.
The ball-shaped low-velocity color map (arrow) in the right atrium (RA)
represents the volume of blood that is displaced as the poppet moves
to its closed position against the sewing ring (SR). This color array is
therefore referred to as the prosthetic closing volume. LA indicates left
atrium. (Previously published. See “Credit Lines” section.)
MP
LA
Figure 40.4. Omniplane transesophageal echocardiogram, with
color flow imaging, of a normal Lillihei-Kaster mitral prosthesis (MP).
This tilting-disk valve has a normal, small amount of transvalvular
regurgitation. Note the 2 separate jets (arrows) in the left atrium (LA).
Besides their small size, these jets clearly represent very mild regurgi-
tation, because they are uniformly red (nonturbulent). Very few blood
cells are traveling at higher velocities (ie, above the Nyquist limit);
therefore, minimal color aliasing is seen within these jets. (Previously
published. See “Credit Lines” section.)
occurs in the Björk-Shiley C-C prosthesis. The risk of outlet strut
fracture is significantly higher for 70° C-C valves than for 60°
C-C valves. Only 60° C-C valves were implanted in the United
States. The risk of strut fracture is highest for large valve sizes
(29, 31, and 33 mm). The largest valves are estimated to have
a potential strut fracture rate as high as 2.8% (280 per 10,000
valves implanted). Although the Björk-Shiley prosthesis is no
longer used, patients with these valves are still encountered
in clinical practice. For most patients, the risk of reoperation
exceeds the risk of strut fracture; thus, routine explantation is
not recommended.
Thromboembolism
Thromboembolism is a common problem of all prostheses,
although it is more common with mechanical valves than with
bioprostheses and with mitral than with aortic prostheses.
Thromboembolism should be clearly distinguished from valve
thrombosis. The latter can result in thromboembolism, but it also
has the potential for acute and severe hemodynamic disturbance
as a result of entrapment of the moving parts by severe stenosis
or severe regurgitation.
• The risk of thromboembolism is higher with mitral than with aor-
tic prostheses.
Prosthetic Valve Endocarditis
Prosthetic valve endocarditis can occur with any of the various
prostheses, and all patients with prosthetic heart valves should
receive endocarditis prophylaxis. With mechanical prostheses,
vegetations form on the sewing ring. With bioprostheses, veg-
etations can form on the ring and on the cusps. In either case,
the infection is difficult to eradicate without replacing the pros-
thesis. Perivalvular extension of infection, such as valve-ring
abscess formation, is a dreaded and all-too-common complica-
tion of prosthetic valve endocarditis. Staphylococci are the most
 40 Prosthetic Heart Valves
common isolate from patients who have early-onset prosthetic
valve infection, with Staphylococcus epidermidis causing a sub-
stantial percentage of cases. Streptococci are the predominant
microorganism causing late-onset prosthetic valve infection.
• Valve-ring abscess is a common complication of prosthetic valve
endocarditis.
Prosthesis-Patient Mismatch
Prosthesis-patient mismatch is defined as an EOA that is too
small for the patient’s body surface area, which results in abnor-
mally high transvalvular gradients. An aortic prosthesis EOA
less than 0.85 cm2 /m2 generally is considered indicative of pros-
thesis-patient mismatch, and an EOA less than 0.60 cm2 /m2 is
considered severe. Severe prosthesis-patient mismatch has been
shown to be associated with reduced survival. For patients under-
going aortic valve replacement, the EOA should be estimated. If
the patient is found to be at risk for prosthesis-patient mismatch,
strategies to reduce mismatch should be considered, including
aortic root enlargement and alternative choices of prosthesis.
Diagnosis of Prosthetic Valve Dysfunction
Patient History
The evaluation of patients who already have a prosthetic valve
should document the exact valve type, size, and model. Patients
may be unaware of this information, which is listed on the valve
identification card. In the serial number, the valve size (in mil-
limeters) usually precedes an “A” or an “M” indicating aortic or
mitral models, respectively.
Patients with prosthetic valves should be asked whether they
are taking antiplatelet agents or receiving anticoagulation. The
adequacy of anticoagulation should be assessed by checking
the international normalized ratio over several months. Patients
should be questioned about bleeding, symptoms consistent
with embolic events, and symptoms suggesting endocarditis. If
a patient has been aware of valve clicks, ask about any sudden
changes, because reduced or absent valve clicks may indicate
valve thrombosis.
Whether the patient is in sinus rhythm or atrial fibrillation
should be established, because atrial fibrillation increases the
incidence of thromboembolic events. The patient’s functional
status should be determined, including any symptoms of left-
sided or right-sided heart failure. As with patients who have
native aortic stenosis, patients who have aortic prostheses also
should be questioned about angina and exertional syncope or
near syncope.
Physical Examination
Typically, there is a brief systolic ejection murmur across nor-
mal aortic prostheses. Normal mitral prostheses, particularly
mitral bioprostheses, may create a brief, low-grade apical rumble
(Table 40.1). The bioprostheses create normal closing sounds (ie,
normal-sounding heart sounds) and do not create opening clicks.
Caged-ball mechanical prostheses produce prominent opening
and closing clicks, whereas tilting-disk and bileaflet prostheses
make prominent closing clicks but muffled, hard-to-hear open-
ing sounds.
• Absence or reduction of expected valve sounds is an important
sign of mechanical valve thrombosis.
421
Table 40.1. Auscultatory Findings in Patients With Normally
Functioning Prosthetic Heart Valves
Auscultatory Finding
Prosthesis Aortic Mitral
Starr-Edwards Sharp opening sound at S a
Sharp opening sound
1
ball valve Sharp closing sound at S2 70–150 ms after S2
Ball “rattles” during systole Sharp closing sound at S1
SEM Ball “rattles” during
diastole
SEM
St Jude Sharp closing sound at S2 Sharp closing sound at S1
SEM
Heterograft SEM Diastolic rumbleb
Abbreviations: S1, first heart sound; S2, second heart sound; SEM, systolic
ejection murmur.
a
Absence of opening or closing sounds with mechanical prosthesis usually
signifies severe prosthesis dysfunction.
b
Should be brief; if prolonged, indicates bioprosthetic obstruction or prosthesis-
patient mismatch.
(Previously published. See “Credit Lines” section.)
The examination of patients with prosthetic valve stenosis
or regurgitation is similar to that of patients with corresponding
native valve lesions. It includes checking for a decreased aortic-
closure-to-mitral-opening interval in patients with prosthetic
mitral valve stenosis. This finding is notable because it is par-
ticularly easy to elicit in patients with certain types of prostheses,
particularly the caged-ball mitral prosthesis.
Radiography
The sewing ring of most prostheses can be visualized on stan-
dard posterior-anterior and lateral chest radiographs. Heterograft
stents are radiopaque, as are many of the mechanical valve
occluders. For patients with valvular prostheses, however, the
standard chest radiograph is more useful for demonstrating signs
of heart failure, such as pulmonary venous hypertension.
For valves with radiopaque occluders, fluoroscopy can be
used to measure the opening and closing angles. Valve thrombo-
sis results in a substantially reduced opening or closing motion
(or both). The fluoroscopic appearance is useful not only for
diagnosing prosthetic valve thrombosis but also for assessing the
efficacy of thrombolytic therapy.
• Fluoroscopy is an excellent tool for visualizing prosthetic leaflet
motion.
Echocardiography
In most clinical situations, echocardiography provides a com-
plete hemodynamic assessment of valvular prostheses. Its use
has revolutionized the diagnostic approach to patients with sus-
pected prosthetic valve dysfunction.
Aortic Prostheses
The complete echocardiographic assessment of prosthetic aortic
valves includes measurement of peak systolic velocity, mean gra-
dient, and EOA. In addition, if the presence or absence of regur-
gitation is noted, the regurgitation is characterized as normal (ie,
closing volume or leakage volume) or pathologic. An attempt is
made to separate pathologic regurgitation into perivalvular or
422 IV Valvular Heart Disease
transvalvular, according to the origin of the jet. Semiquantitative
and quantitative measures are used to characterize the amount of
regurgitation.
Gradients across prosthetic valves are determined by the mod-
ified Bernoulli equation (SEE CHAPTER 8 “PRINCIPLES OF ECHOCAR-
DIOGRAPHY”). It is crucial to ensure that the highest velocity of the
aortic prosthesis has been obtained, which requires interrogation
from multiple windows as in the Doppler assessment of native
aortic stenosis. The EOA for prosthetic valves is determined with
the continuity equation and should be commonly performed for
aortic and mitral prostheses. The prosthesis sewing ring must be
visualized clearly to allow measurement of the diameter of the
left ventricular outflow tract. In rare cases in which this measure-
ment cannot be made confidently, it is acceptable to approximate
the measurement with the external diameter of the sewing ring
(ie, the valve size). This approximation slightly overestimates the
actual EOA.
In our experience, the average mean gradient is 13 to 15 mm
Hg for heterograft, Björk-Shiley, St Jude Medical, and Medtronic-
Hall prostheses. The average mean gradient is considerably
greater for Starr-Edwards aortic prostheses (23 mm Hg) and
considerably less for homograft prostheses (8 mm Hg). However,
it is important to be aware of patient-to-patient variability. With
all valve types, except homografts, some patients will have nor-
mally functioning prostheses with mean gradients as high as 35
to 45 mm Hg. In general, these patients have small prostheses,
and the high gradients are due to prosthesis-patient mismatch.
Because of this variability in mean gradient and EOA among
patients with normal prosthetic valve function, it is mandatory
to perform a baseline transthoracic Doppler echocardiographic
examination on each patient soon after implantation. Doing so
effectively “fingerprints” the individual prosthesis and serves
as a baseline for comparison, should symptoms develop that are
consistent with prosthetic valve dysfunction.
Assessment of Prosthetic Aortic Valve
Incompetence
Semiquantitation of aortic regurgitation is performed with
2-dimensional imaging, spectral Doppler, and color flow imag-
ing. A determination is made of the degree to which color flow
signals of regurgitation fill the left ventricular outflow tract in
diastole. In addition, assessments are made of the intensity of
high-velocity signals in the continuous-wave spectrum of aor-
tic regurgitation, the pressure half-time of the continuous-wave
signal, the amount of diastolic flow reversal in the descending
thoracic aorta (obtained by pulsed-wave Doppler), and the size
of the left ventricle (from 2-dimensional imaging). If a patient
has a native mitral valve that is competent, the aortic regurgitant
volume and regurgitant fraction can be calculated by comparing
the forward flow through the left ventricular outflow tract with
the forward flow across the mitral annulus.
Assessment of Prosthetic Mitral
and Tricuspid Valves
Mitral and tricuspid prostheses are assessed more easily from
Doppler hemodynamic data because the optimal window for
interrogation is always either apical or periapical. Complete
assessment requires measurement of the peak early velocity
(E velocity), the velocity with atrial contraction (A velocity) for
patients in sinus rhythm, the pressure half-time, the EOA, and the
presence and degree of regurgitation. The velocities and mean
gradient are measured from the continuous-wave Doppler sig-
nal. Although the EOA can be calculated from the pressure half-
time for obstructed prostheses, this method tends to overestimate
the EOA for normally functioning prostheses. Therefore, it is
preferable to report the pressure half-time independently and to
calculate the EOA by the continuity method. As with aortic pros-
theses, the EOA for mitral and tricuspid prostheses is obtained
by dividing the stroke volume of the left ventricular outflow tract
by the prosthesis time velocity integral. However, this method
cannot be used with mitral or tricuspid prostheses when there is
significant aortic regurgitation or significant prosthesis regurgi-
tation; under these circumstances, continuity of flow no longer
exists. In such cases, the pressure half-time should be reported
and the EOA should not be calculated.
Because mitral and tricuspid prostheses create reverberations
and acoustic shadowing within the atria, visualization of regur-
gitant jets by surface echocardiography is always suboptimal.
However, important clues to significant regurgitation may be
found on the surface examination. These include an increased E
velocity with normal pressure half-time, a dense continuous-wave
regurgitant signal, and color Doppler signals of flow convergence
on the ventricular side of the regurgitant orifice. Transesophageal
echocardiography provides complete visualization of color jets
due to prosthetic mitral or tricuspid regurgitation. It also allows
sampling of the left and right upper pulmonary veins for systolic
flow reversal.
The average mean gradient of a mitral prosthesis is 4 to 5 mm
Hg for heterograft, tilting-disk, bileaflet, and caged-ball prosthe-
ses. Occasionally, normal valves will have a mean gradient as
high as 10 mm Hg. For normal tricuspid prostheses, the mean
gradient is 2 to 3 mm Hg, with outliers as high as 6 mm Hg. For
all Doppler hemodynamic calculations, at least 3 cardiac cycles
should be averaged for patients in sinus rhythm and at least 5
cycles should be averaged for those in atrial fibrillation. For tri-
cuspid prostheses, 10 cycles must be averaged, even for patients
in sinus rhythm, because of significant variation in mean gradi-
ent with the respiratory cycle.
Transesophageal Echocardiography
Most prosthetic valve hemodynamic information is available
from surface echocardiography. Similarly, the amount of pros-
thetic aortic regurgitation usually can be assessed by the surface
examination. Complete visualization of mitral and tricuspid pros-
thesis regurgitant jets requires transesophageal echocardiography,
which is also necessary to determine the mechanism of regurgi-
tation or stenosis. Because transesophageal echocardiography is
sensitive enough to detect normal closing volume and leakage
volume, the echocardiographer must be aware of the normal pat-
terns for each type of prosthesis. Three-dimensional imaging is
particularly useful for localizing periprosthetic mitral regurgitant
jets and is used to guide percutaneous or surgical closure.
For patients with clinically significant prosthetic valve dys-
function, an echocardiographic examination usually can obvi-
ate the need for invasive hemodynamic tests before surgical
replacement.
Laboratory Tests and Hemolysis
For patients with prosthetic valves who are receiving long-term
anticoagulation therapy, the international normalized ratio should
be checked at least monthly. The hemoglobin value should be
checked periodically also, because a decrease could be due to
 40 Prosthetic Heart Valves
bleeding or significant hemolysis. On a peripheral blood smear,
sheared red blood cells will appear as schistocytes. The level of
serum haptoglobin will approach zero, and the level of lactate
dehydrogenase will increase. There is usually a compensatory
increase in reticulocytes. Urinary loss of iron, in the form of
hemosiderin, produces iron deficiency.
• Hemolysis is a potential complication of prosthetic valves that
leads to a decreased serum haptoglobin level and an increased lac-
tate dehydrogenase level.
Invasive Hemodynamics
The diagnosis of prosthetic valve dysfunction seldom requires
an invasive procedure. Catheters should not be passed across
mechanical aortic prostheses; thus, measurement of aortic pros-
thesis gradients requires a transseptal approach. In patients with
both mitral and aortic prostheses, measurement of the gradients
requires not only transseptal puncture but also left ventricular
puncture. For mitral prostheses, it is always necessary to measure
the gradient with a transseptal approach, with direct measure-
ment of left atrial pressure, rather than depending on pulmonary
artery wedge pressure. The latter nearly always results in signifi-
cant overestimation of the gradient, even when the phase delay is
taken into account.
Primary Prevention of Valve Dysfunction
Antithrombotic Therapy
Unless contraindicated, aspirin therapy (81 mg/day) should be
used in all patients with prosthetic heart valves, both mechanical
prostheses and bioprostheses. All patients with mechanical pros-
theses require anticoagulation with warfarin. The recommen-
dations for target international normalized ratio for long-term
anticoagulation (more than 3 months after valve implantation)
are outlined in Box 40.3. The intensity of anticoagulation must
Box 40.3. Recommendations for Long-term Anticoa-
gulation With Mechanical Prosthetic Heart Valvesa
Target INRb 2–3
AVR and no risk factor
Bileaflet
Medtronic-Hall
Target INRb 2.5–3.5
AVR
Other disk valves
AVR with risk factorc
All MVR
Abbreviations: AVR, aortic valve replacement (prosthesis); INR,
international normalized ratio; MVR, mitral valve replacement
(prosthesis).
a
Aspirin is recommended for all patients with prosthetic heart
valves.
b
More than 3 months after valve replacement.
c
Risk factors: atrial fibrillation, left ventricular systolic dysfunction,
previous thromboembolism, and hypercoagulable conditions.
Previously published. See “Credit Lines” section.
423
be tailored for each patient, taking into account such variables as
age, bleeding risk, gait stability, and any risk factors for throm-
bosis and thromboembolism.
• All patients with prosthetic valves (biologic or mechanical) should
receive aspirin, unless its use is contraindicated.
Patients with bioprostheses generally receive anticoagulation
treatment with warfarin for the first 3 months, although aspirin
alone is used in some centers. The recommended target interna-
tional normalized ratio for both aortic and mitral bioprostheses
is 2.0 to 3.0 for the first 3 months after implantation. In patients
with chronic atrial fibrillation, warfarin treatment should be con-
tinued indefinitely, which removes the major advantage of bio-
prostheses over mechanical prostheses.
Most patients with prosthetic valves will eventually require
short-term discontinuation of anticoagulation in order to undergo
an invasive procedure. Management options for “bridging” anti-
coagulation are based on individual risk (including the prosthesis
type and its position), the nature of the procedure, the risk of
bleeding, and the expected duration of the interruption of warfa-
rin therapy. Short-term discontinuation of warfarin without hep-
arin coverage is appropriate for low-risk patients, such as those
with a bileaflet aortic prosthesis, normal left ventricular func-
tion, and no additional risk factors. Risk factors for thromboem-
bolism during discontinuation of anticoagulation include atrial
fibrillation, previous thromboembolism, left ventricular dysfunc-
tion, hypercoagulable conditions, older-generation thrombogenic
valves, mechanical mitral valves, or more than one mechanical
valve (Box 40.4).
Warfarin therapy should be resumed in all patients with or
without heparin therapy as soon as possible after the proce-
dure. Recommendations for management of periprocedural
anticoagulation are outlined in Box 40.5. The use of low-
molecular-weight heparin for periprocedural anticoagulation of
prosthetic valves has not been approved by the FDA and is con-
troversial. Guidelines favor the use of unfractionated heparin;
however, studies are ongoing to address the safety and efficacy
of low-molecular-weight heparin for “bridging” therapy of pros-
thetic valves.
• Patients with a bileaflet aortic prosthesis and no risk factors do
not require heparin coverage during short-term discontinuation of
warfarin.
Box 40.4. Risk of Thrombosis or Thromboembolism
With Interruption of Anticoagulation in Prosthetic
Heart Valves
High-risk factors
Recent thrombus or embolus (1 year)
Previous event when not taking warfarin
Björk-Shiley valve
Other risk factors
Atrial fibrillation
History of thromboembolism
Hypercoagulable state
Decreased ejection fraction
See “Credit Lines” section.
424 IV Valvular Heart Disease
Box 40.5. Recommendations for Periprocedural
Anticoagulation in Patients With Mechanical Prosthetic
Heart Valves
No heparin
St Jude or Medtronic-Hall AVR
No risk factors
Heparin
Björk-Shiley valve
AVR: 1 risk factora
MVR
Abbreviations: AVR, aortic valve replacement (prosthesis); MVR,
mitral valve replacement (prosthesis).
a
Risk factors: atrial fibrillation, previous thromboembolism, left
ventricular systolic dysfunction, and hypercoagulable condition.
In patients with mechanical valves who require interruption
of warfarin therapy for noncardiac surgery, invasive proce-
dures, or dental care, high-dose vitamin K1 should not be given
routinely because this may create a hypercoagulable condition.
Endocarditis Prophylaxis
All patients with prosthetic valves should receive antibiotics
for endocarditis prophylaxis according to the guidelines of the
American College of Cardiology/American Heart Association
(SEE CHAPTER 83 “INFECTIVE ENDOCARDITIS”).
Management of Prosthetic Valve Complications
Replacement of a dysfunctional prosthetic valve carries a con-
siderably higher surgical risk than initial valve replacement.
Therefore, with most chronic or subacute problems, one should
be conservative when deciding whether to replace a dysfunc-
tional prosthesis. Indications for consideration of surgery include
1) severe prosthetic stenosis or regurgitation with resultant
symptoms or ventricular dysfunction, 2) transfusion-dependent
hemolysis, 3) recurrent emboli despite adequate anticoagulation
or antiplatelet therapy, and 4) prosthetic valve endocarditis with
hemodynamic compromise, significant perivalvular extension of
infection, very large vegetations, or recurrent embolic events.
Valve thrombosis is a life-threatening and dreaded compli-
cation of mechanical valves. Treatment options include throm-
bolysis and surgery. In patients with small clots and stable
hemodynamic values, a trial of heparin therapy may be con-
sidered. There is debate concerning the choice of thrombolytic
therapy versus cardiac surgery. In most cases of thrombosis in
right-sided prostheses, thrombolysis is considered the first-line
treatment. For left-sided prostheses, thrombolytic therapy car-
ries a substantial risk of embolism and stroke. Nevertheless,
many patients with thrombosed left-sided valves will also have
a very high operative mortality rate. In these cases, it is best
for the cardiologist and the cardiac surgeon to arrive jointly at
a decision, which will depend heavily on the estimated risk of
cardiac surgery.
Abbreviation
EOA effective orifice area
FDA US Food and Drug Administration
 41
Surgery for Cardiac Valve Diseasea
RAKESH M. SURI, MD, DPHIL, and THOMAS A. ORSZULAK, MD
The decision to perform an operation for valvular disease is
complex and requires a meld of multidisciplinary expertise and
considerable diagnostic effort. However, the history and physi-
cal examination are as important as everything else for making
diagnostic and therapeutic decisions. The indications for oper-
ation are based on historical events in cardiac surgery, natural
history studies, and current published studies.
This chapter discusses the approach to correcting valvular
heart disease and illustrates a major concept: Repair of any valve
is always preferable to replacement if it can be done safely and
durably.
Aortic Valve
The indications for operation in aortic valve disease, predomi-
nantly aortic stenosis (Figure 41.1) are dictated by the severity of
disease and the symptoms. The most common symptoms include
dyspnea on exertion, chest pain, and light-headedness or dizzi-
ness, usually with a sudden change of position, such as stand-
ing quickly. Although correction is certainly required in patients
with symptoms, some asymptomatic patients with aortic stenosis
need valve replacement or repair. Their situations require close
scrutiny. Most frequently these asymptomatic patients have an
excessively high gradient or a valve area less than 0.75 cm2. A
treadmill or exercise test can be done carefully to evaluate their
myocardial response to light exercise. If the ejection fraction
decreases at a very low work level, especially without symp-
toms, the patient’s condition is on the edge of deterioration and,
even though the patient is asymptomatic, an operation can be
a
Portions previously published in Minimally invasive robotic and
thoracoscopic cardiac surgery. Mayo Clin Cardiovasc Update.
2008;6(2):6–7.
performed. An exercise test will also determine the level of exer-
cise a patient actually can do without having symptoms.
Repair is not performed nearly as frequently for aortic valve
disease as for mitral valve disease. Aortic valve repair is pre-
dominantly for regurgitant lesions from healed endocarditis
with perforated leaflets that can be patched or for some forms of
commissural dilatation. With an aneurysm that is limited to the
ascending aorta distal to the sinotubular junction, splaying of the
commissures at the sinotubular junction can cause aortic regur-
gitation; repairing the aneurysm and returning the sinotubular
junction and commissures to their original or natural relationship
can preserve the valve. More direct approaches to valve lesions
can be performed with a technique forwarded by Yacoub and
David in which the valve leaflets are preserved, the aneurysm or
ascending aorta is removed, and the valve is resuspended within a
graft (Figure 41.2). These procedures are technically demanding
and are limited to patients who have Marfan disease or ascending
aortic aneurysm and minimal or mild aortic regurgitation.
In the acute stage of endocarditis, repair of the aortic valve is
impossible (Figure 41.3). Surgical principles for infection dictate
removing or sterilizing all infected tissue before reconstruction.
When the infection is limited to the leaflets alone, débridement
mimics a normal valve replacement. The difficulty arises when
an aggressive organism is present or when a delay in diagnosis or
antibiotic treatment allows the infection to breach the boundar-
ies of the leaflets and destroy annular tissue with abscess forma-
tion. This requires broader débridement and reconstruction with
autologous pericardium or homograft implantation (or both).
Autologous tissue and homografts have the lowest reinfection
rate in the aortic root with extensive acute infection.
For aortic valve regurgitation, echocardiographic findings
and symptoms help determine the timing of the operation. If
a patient’s left ventricular function starts to fail or if the left
425
426 IV Valvular Heart Disease

Figure 41.1. Rheumatic Aortic Valve Disease. Valve cusps are thickened and fused.

Figure 41.2. Surgical Repair Described by Yacoub and David. See text for discussion.
 41 Surgery for Cardiac Valve Disease 427

Figure 41.3. Endocarditis. A, Endocarditis of native aortic valve. B, Endocarditis of aortic tissue prosthesis.

ventricle becomes dilated, it is appropriate to recommend an
operation. For aortic stenosis, similar events are tantamount for
recommending valve replacement.
Repair of aortic stenosis with various forms of decalcification
are of historical interest only. In a Mayo Clinic series, patients
who had leaflet decalcification with ultrasonic débridement had
excellent early results (Figure 41.4). However, fulminant scar-
ring caused the leaflets to retract, and the patients returned
with severe aortic regurgitation requiring reoperation and valve
replacement.
Mitral Valve
Mitral regurgitation, especially from myxomatous disease or
ruptured chordae, is ideal for valve repair. Figure 41.5 shows
generally the steps that are involved in mitral valve repair.
Current guidelines suggest that patients who have severe
mitral regurgitation should undergo mitral valve repair before
the onset of symptoms or left ventricular dysfunction and
that they should be referred to centers where the success rate
of valve repair is greater than 90%. Repairs are quite dura-
ble. The chance of recurrent regurgitation requiring opera-
tion is approximately 0.5% per year for the posterior leaflet
(Figure 41.6) and 1% per year for the anterior leaflet. The true
advantage of valve repair for patients is the preservation of
left ventricular morphology and function. With repair, long-
term anticoagulation is avoided (except for 6–8 weeks to
allow endothelialization of the annuloplasty ring). The long-
term results with repair match those of a healthy age-matched
population.
428 IV Valvular Heart Disease

Figure 41.4. Calcified Aortic Valve. A, Repair at initial operation. B, Same patient at reoperation 8 months later. Valve leaflets are scarred and
retracted, resulting in severe aortic regurgitation.

Repair is also possible in some cases of endocarditis if the
operation is done in the early phases of the disease or in the
healed state (Figure 41.7). The principles of repair are basically
the same, but additional features are present with acute endo-
carditis. Any acute infection requires the removal of all infected
tissue, which limits repair when a large area of the valve has veg-
etations. When the infection is associated with abscess forma-
tion, débridement almost always eliminates the chance of repair.
There is frequently a balance between what can and should be
done with antibiotic treatment and operative intervention to
preserve valve tissue.
Mitral stenosis more frequently requires valve replacement.
Today any valve that would have been treated with an open or
closed commissurotomy is now treated with percutaneous bal-
loon dilatation. If a patient has a mitral stenosis that cannot be
dilated, generally the valve must be replaced. Unlike patients
in developing countries, most patients in the United States who
have mitral stenosis had rheumatic fever several years earlier,
and the scarring is established and replete with calcium that pro-
hibits or limits any form of repair (Figure 41.8). In developing
countries, however, young patients with fairly acute or recent
rheumatic fever may have pliable valves that can be repaired
effectively. Also, patients in the United States who need mitral
valve replacement for rheumatic disease are generally older, so
that a valve replacement is not as limiting as in an adolescent or
a young adult.
 41 Surgery for Cardiac Valve Disease 429

Figure 41.5. Mitral Valve Repair.

Tricuspid Valve however, the leaflets may then become fused to the lead so that
the valve must be replaced. In this situation, the leads are exte-
The tricuspid valve rarely requires replacement today. Carcinoid
riorized outside the prosthesis at valve replacement. More fre-
involvement and endocarditis are probably the 2 most common
quently, tissue valves are used at this location since they have a
situations in which replacement is required. Rheumatic fever
much lower incidence of thrombosis and they also allow a pace-
involving the tricuspid valve in the United States is quite rare.
maker lead to be passed through them should the need arise.
Most tricuspid lesions are regurgitant, functional, and secondary
In today’s cardiology practice, echocardiography provides
to left-sided lesions. They respond well to annuloplasty, which
a window through which valvular disease can be accurately
decreases the diameter of the dilated annulus to normal and
evaluated to provide patients with earlier surgical treatment,
recreates leaflet coaptation. There are several techniques, and
especially when the valve is repairable. When a valve is less
leaflet overlap must be recreated to achieve competence. Chordae
likely to be repairable, continuing medical treatment is a better
tendineae rupture in the tricuspid valve infrequently; the repair is
option than implanting a prosthesis unless the symptoms warrant
identical to the repair of the mitral valve.
intervention.
A situation that is becoming more frequent for tricuspid valve
replacement is significant dilatation from long-standing tricus-
pid regurgitation or, even more frequently, from trapping of a
device in the chordae or tethering of the leaflets. Pacemakers or
Valve Choice
implantable cardioverter-defibrillators that have leads crossing Discussions about valve choice can be distilled into 2 basic
the tricuspid valve limit the motion of the leaflets and can cause categories: mechanical valve and bioprosthesis. Both types are
severe tricuspid regurgitation. Occasionally, these leads can be extremely functional; neither is perfect. The choice is between 1)
positioned at the commissures without causing regurgitation; lifelong anticoagulation with warfarin with a mechanical valve
430 IV Valvular Heart Disease

Figure 41.6. Posterior Leaflet Repair. A, Before repair of the posterior leaflet (arrow). B, After repair.

and 2) the greater need for subsequent operation with a biopros-
thesis. The latest generation of bioprostheses can effectively
last 12 to 15 years in adults and longer in patients older than
65 years. The rate of deterioration is more rapid in adolescents
and young adults. Although there are many types of mechanical
valves, the bileaflet valve has the best mechanism for replace-
ment. Bioprosthetic data have shown that in the aortic position,
the bovine pericardial valve is more durable than the porcine
valve, although the latest generation of valves is being evaluated
in a randomized study to assess any real difference.
Today the patient must be involved in the choices for valve
prostheses. Historically, it was said that younger patients should
receive a mechanical prosthesis because they could look forward
to many years of life and that elderly patients should receive tissue
valves or a bioprosthesis. Now patients have much more informa-
tion before valve surgery, predominantly from the Internet and
their personal interest, and many patients are active and unwill-
ing to give up activities with increased risks of injury. This sit-
uation provides patients with the choice of a bioprosthesis that
allows them to have the most normal life possible. Knowing that
they will need a subsequent operation is critical in their decision.
For patients who do undergo elective reoperation for deteriora-
tion of a bioprosthesis without coronary bypass grafts, the oper-
ative risk is similar to that in the initial operation.
In general, patients who have aortic valve bioprostheses do
not require any form of anticoagulation other than one 325-mg
aspirin tablet daily. Mitral valve prostheses require anticoagula-
tion for at least 6 to 8 weeks because of the lower rate of blood
 41 Surgery for Cardiac Valve Disease 431

Figure 41.7. A, Repaired mitral valve after acute endocarditis. B, Chronic endocarditis of the mitral valve. Scarring and calcification of the valve
make repair impossible and valve replacement necessary.

flow across the prosthesis and the possibility of thrombus form-
ing on the sewing ring. After 6 to 8 weeks, anticoagulation can be
stopped. Frequently, atrial fibrillation occurs postoperatively, and
anticoagulation needs to be continued until sinus rhythm returns.
When patients are in chronic atrial fibrillation, a mechanical
valve is usually recommended, although this requires thoughtful
consideration in the elderly. Aging increases the chance of other
illness and possible operative procedures; gait instability can
lead to life-threatening falls when elderly patients are in a fully
anticoagulated state. It is also important for the surgeon to ligate
the atrial appendage in these patients to minimize the chance
of thrombus formation. Obviously, mechanical valves in either
position require anticoagulation early and indefinitely. Patients
who require a bioprosthesis instead of a mechanical prosthesis
include those who have easy bruising, ulcer disease, or a blood
dyscrasia for which warfarin is contraindicated. An additional cat-
egory is women of childbearing age. A full discussion of antico-
agulation in pregnancy is presented in CHAPTER 80 (“PREGNANCY
AND THE HEART”), but it is easier to carry a pregnancy to delivery
with a tissue prosthesis, which does not require warfarin.
Minimally Invasive Robotic and
Thoracoscopic Valve Surgery
Medical interventions are evolving toward less invasive proce-
dures. The use of high-definition imaging systems and robotic
432 IV Valvular Heart Disease
Figure 41.8. Rheumatic Mitral Valve Disease. Extensive scarring and deformity preclude surgical repair.
technology allows complex cardiac surgery to be performed
through small incisions without a sternotomy. The putative ben-
efits of such an approach include a shorter duration of ventilator
support, less need for blood transfusion, shorter hospital stay, less
postoperative functional limitation, and quicker return to normal
activity. The current role of minimally invasive and robotic sur-
gery at Mayo Clinic includes mitral valve repair or replacement,
tricuspid valve repair or replacement, atrial fibrillation surgery,
coronary artery bypass, and congenital surgery.
Although cardiac surgery performed through a median ster-
notomy provides excellent exposure to the heart and great ves-
sels, the necessary limitation in upper extremity mobility while
sternal union occurs often precludes early return to functional
activity. Alternate incisions evolved to avoid complete sternal
division, thereby also minimizing the length of the skin incision.
Early approaches included partial sternotomy and parasternal
incisions. From these efforts, the field of minimally invasive
valve surgery was born. Despite their appeal, all these strate-
gies suffer from some degree of impaired surgical visualization
of intracardiac structures and may be associated with the same
postoperative functional limitations as full sternotomy.
The early pioneers of cardiac surgery quickly established
that an incision in the right chest provides surgical access to the
mitral and tricuspid valves along with the atria and the interatrial
septum. A small right anterolateral thoracotomy permits direct
visualization of the atrial aspects of the atrioventricular valves.
This convenient alignment allows surgeons to perform intracar-
diac procedures through a small, well-tolerated incision in the
right chest wall.
Facilitating Technology
Whenever the heart is opened, even through incisions that are
“minimally invasive,” cardiopulmonary bypass is necessary to
ensure a still, bloodless field during surgical manipulation of
intracardiac structures. Near-percutaneous femoral artery and
vein cannulation permits full cardiopulmonary bypass support.
Cardiac arrest is achieved with the use of a transthoracic
cross-clamp and direct instillation of a cardioplegic agent into
the aortic root.
The distance to the atrioventricular valves makes the right
minithoracotomy approach with the naked eye more challenging
than conventional open surgery. The technically simplest option
to facilitate surgical visualization is a thoracoscopic approach
that uses a high-definition video thoracoscope and long-shafted
instruments. This option is appealing because it is simple,
requires minimal equipment, and may be quickly set up and
disassembled. The limitations of thoracoscopic surgery include
suboptimal exposure in certain situations and the relative diffi-
culty of performing fine technical manipulations at the valvar or
subvalvar level. Nevertheless, a purely thoracoscopic approach
can be used for mitral and tricuspid valve procedures and for
congenital, atrial septal, and arrhythmia surgery.
The next tier of technical sophistication in minimally inva-
sive valve surgery involves robotic technology. The newest
high-definition robotic system uses a central computer proces-
sor that allows the surgeon to manipulate multiple instruments
simultaneously (Figure 41.9). Concurrent use of 3 instruments
and a camera lets the surgeon shift quickly between control of
the dynamic atrial retractor and the robotic arms to expose, vis-
ualize, and manipulate cardiac tissue. Robotic arms mimic the
movements of a surgeon’s hands. The surgeon operates from a
console with a 3-dimensional view of the operative field. The
computer processor reproduces hand movements of the surgeon
in a scaled manner and translates them into real-time movement
of instruments mounted on robotic arms through chest wall ports.
The robotic instruments are not merely shafted but also wristed,
thereby permitting more natural manipulation of suture and tis-
sue. Robotic technology allows the surgeon to approximate or
exceed the level of technical precision possible with a traditional
open-chest approach.
Robotic Mitral Valve Repair
Mitral valve repair improves survival among asymptomatic
patients who have severe mitral valve regurgitation due to leaflet
 41 Surgery for Cardiac Valve Disease 433

Figure 41.10. Triangular resection (arrow) of a prolapsing segment

of a myxomatous mitral valve leaflet, as viewed with the robotic camera.
(Previously published. See “Credit Lines” section.)

valve repair to those who otherwise might delay or avoid having

Figure 41.9. The newest high-definition robotic system allows 3 an operation. Moreover, Mayo Clinic mitral valve repair tech-
instruments to be used concurrently with a high-definition camera. The niques are well suited to robotic technology. Robotically assisted
instrument arms, which are inserted gently between the patient’s ribs, repair for prolapse of the middle scallop of the posterior leaflet
were designed to mimic the movements of a surgeon’s hands. (Previously
of the mitral valve at Mayo Clinic currently involves a traditional
published. See “Credit Lines” section.)
posterior resection. The leaflet is then reconstructed robotically,
and a standard-length 63-mm posterior annuloplasty band is
prolapse—even when ventricular function is normal. The avail- anchored into the posterior mitral annulus between the right and
ability of reproducible valve repair techniques as a safe and left fibrous trigones. This technique reduces the posterior mitral
durable alternative to prosthetic replacement has dramatically annular circumference and protects the leaflet closure.
influenced the indications for surgical intervention in patients Visualization of the subvalvar apparatus (consisting of the
with mitral regurgitation due to leaflet prolapse. With success- papillary muscles and chordae tendineae) is sometimes difficult
ful valve repair, patients who maintain sinus rhythm can resume through a median sternotomy. This problem has led to anterior
full activities without the need for long-term anticoagulation. An or bileaflet prolapse being labeled a “complex” valve disease.
evolution in operative techniques has led to predictable and dura- In contrast, with robotic assistance, the surgeon’s view of the
ble results for mitral valve repair. In the current practice at Mayo mitral valve is often direct and uncompromised. The surgeon
Clinic, 95% to 99% of patients with pure mitral valve regurgita- can precisely determine whether 1 or more leaflet segments are
tion due to degenerative disease undergo valve repair rather than unsupported and resuspend them to the appropriate lengths with
prosthetic replacement. artificial neochordae. Robotic instruments are suited to tailoring
Posterior leaflet mitral valve repair strategy at Mayo Clinic
is based on the premise that eliminating only the redundant pro-
lapsing leaflet edge from the coaptation margin with the use of
a simple triangular resection (Figure 41.10) followed by suture
reconstruction (Figure 41.11) is sufficient, expedient, and dura-
ble. All repairs are protected with a 63-mm posterior annulo-
plasty band anchored between right and left fibrous trigones
(Figure 41.12). Anterior leaflet pathology is routinely corrected
by resuspending the leading edge with artificial neochordae.
Bileaflet disease is addressed with a combination of these tech-
niques. The annual risk of reoperation after use of these repair
methods at Mayo Clinic is about 0.5% for posterior leaflet pro-
lapse, 1.6% for anterior leaflet prolapse, and 0.9% for bileaflet
prolapse. Repair of any mitral leaflet prolapse is as durable as
replacement of the mitral valve, which has a risk of reoperation
of 0.74% per year overall.
Despite the proven advantages, the psychological barrier asso-
ciated with the referral of an otherwise healthy, asymptomatic
patient for cardiac surgery has prevented the widespread accept-
ance of early mitral valve repair for leaflet prolapse—even in Figure 41.11. Suture reconstruction (arrow) after triangular resection
centers where the incidence of repair is greater than 95%. Robotic of prolapsing leaflet tissue, as viewed with robotic camera. (Previously
techniques offer the opportunity to extend the benefits of mitral published. See “Credit Lines” section.)
434 IV Valvular Heart Disease
Figure 41.12. Completed repair (arrow) after triangular resection,
suture reconstruction, and posterior annuloplasty band insertion, as
viewed with the robotic camera. (Previously published. See “Credit
Lines” section.)
neochordae because they allow the surgeon to refer to the lengths
of more normal adjacent cords.
Future Developments
As nanotechnology evolves, robotic instruments are becoming
more intricate and control systems more intuitive. Robotic port
widths have decreased and are now smaller than the diameter
of a standard pencil. Continued improvements are expected
for 3-dimensional visualization and the manipulation of vid-
eoscopes, which already provide high definition and nearly
cinematic quality. Enhanced visualization of tissue strain with
high-definition robotic cameras provides the surgeon with
real-time data, and systems designed to convey tactile feed-
back are currently under development. Surrogates have been
developed for traditional stitches and intracorporeal knot tying,
including malleable clips and knots and vascular anastomotic
devices. These innovations have shortened operative times and
increased the range of procedures performed with robotic assis-
tance (eg, coronary revascularization). Finally, data from our
institution has shown that robotic mitral valve repair is safe,
effective, and may be associated with improved early quality of
life/return to work.
Reoperation
In general, the indications for reoperation are similar to those for
a primary operation; however, there are special circumstances.
Bioprostheses fail either by calcification of the leaflets and steno-
sis or by leaflet fatigue and regurgitation. These situations usually
are diagnosed with yearly surveillance echocardiographic stud-
ies. Occasionally, a bioprosthetic leaflet tears and causes sudden,
severe regurgitation, with an urgent need for reoperation.
Mechanical prostheses can function indefinitely, but several
external factors may cause failure. Thrombosis, endocarditis,
and perivalvular leak are the most common causes for failure
of mechanical prostheses. Almost invariably, patients with valve
thrombosis have not received adequate anticoagulation before
thrombosis. Some causes are iatrogenic, such as stopping use of
warfarin for noncardiac procedures or operations and delaying
the reinstitution of anticoagulation. These patients usually pre-
sent with a valve obstruction that is a surgical emergency. Rarely,
thrombolysis is used, but it is complicated by strokes and recur-
rences. Patients can also present with thromboemboli, which
may occur without valve obstruction.
Endocarditis involves mechanical valves and bioprosthetic
valves equally. Indications for operation include valve dehis-
cence, valve obstruction due to vegetations, multiple emboli,
persistent sepsis despite antibiotic treatment, and infection with
staphylococci or fungal organisms. The recurrence rate is higher
with early postoperative endocarditis (≤6 weeks).
Suggested Reading
Rehfeldt KH, Mauermann WJ, Burkhart HM, Suri RM. Robot-assisted
mitral valve repair. J Cardiothorac Vasc Anesth. 2011 Aug;25(4):
721–30. Epub 2011 May 26.
Suri RM, Antiel RM, Burkhart HM, Huebner M, Li Z, Eton DT,
Topilsky T, Sarano ME, Schaff HV. Quality of life after early mitral
valve repair using conventional and robotic approaches. Ann Thorac
Surg. 2012 Mar;93(3):761–9.
Suri RM, Burkhart HM, Daly RC, Dearani JA, Park SJ, Sundt TM 3rd,
Li Z, Enriquez-Sarano M, Schaff HV. Robotic mitral valve repair for
all prolapse subsets using techniques identical to open valvuloplasty:
establishing the benchmark against which percutaneous interventions
should be judged. J Thorac Cardiovasc Surg. 2011 Nov;142(5):970–9.
Epub 2011 Sep 10.
Suri RM, Burkhart HM, Rehfeldt KH, Enriquez-Sarano M, Daly RC,
Williamson EE, Li Z, Schaff HV. Robotic mitral valve repair for all
categories of leaflet prolapse: improving patient appeal and advanc-
ing standard of care. Mayo Clin Proc. 2011 Sep;86(9):838–44. Epub
2011 Jul 14.
 42
Pathogenesis of Valvular Heart Disease
NALINI M. RAJAMANNAN, MD, and JOSEPH G. MURPHY, MD
Aortic Valve
Aortic Valve Sclerosis
Aortic valve sclerosis is thickening of the aortic valve leaflets
without hemodynamic valve stenosis. It occurs in 25% of per-
sons older than 65 years and 50% of those older than 85 years.
Most patients have a mid-peaking systolic murmur best heard
in the aortic area with preservation of the aortic second sound;
this finding is in contrast to that in significant aortic stenosis, in
which the aortic second sound is diminished or absent.
Aortic Valve Stenosis
Calcific aortic valve disease is the commonest indication for
heart valve surgery worldwide. It was previously thought to be
due to a passive process, in which calcium nodules attach to the
valve leaflets and cause degeneration. Current research indicates
that an active biologic process similar to vascular atherosclerosis
is the cause. Box 42.1 lists the cardiovascular risk factors associ-
ated with calcific aortic valve disease.
• Traditional cardiovascular risk factors are similar for atheroscle-
rotic vascular disease and calcific aortic valve disease.
Animal models of atherosclerosis have tested the lipid
hypothesis of aortic valve disease (Figure 42.1). In the presence
of increased lipid levels, the valve myofibroblast cell has the plas-
ticity to differentiate into bone, cartilage, and fat. With the use of
statins in experimental models, these cells can reverse this effect.
The cellular mechanisms important in the development of cal-
cific aortic valve disease have been defined, as listed in Box 42.1.
Box 42.2 lists the numerous signaling pathways present in the
development of calcific aortic valve disease. The cardiovascular
Abbreviations and acronyms are expanded at the end of this chapter.
risk factors provide a necessary inflammatory environment, and
activation of the numerous signaling pathways has the potential
for cross talk within the aortic valve that, in turn, provides the
cellular information needed for development of an osteogenic
phenotype. The cellular cross talk present in calcifying aortic
valves is shown in Figure 42.2.
The final mechanism important in the development of cal-
cific aortic valve disease is genetic inheritance of specific single
nucleotide polymorphisms. Box 42.3 lists the different genetic
signaling pathways defined in the development of this disease.
• Atherosclerotic signaling mechanisms play a role in the develop-
ment of calcific aortic valve disease.
• Genetics inheritance plays a role in the development of calcific aor-
tic valve disease.
Statins in Aortic Valve Disease
Currently, there is no US Food and Drug Administration indica-
tion for the use of statins in the treatment of calcific aortic valve
disease. In addition, the American College of Cardiology and the
American Heart Association have not recommended the use of
statins to slow the progression of calcific aortic valve disease. On
the basis of echocardiographic criteria and measurement of valve
calcium with computed tomography, many retrospective studies
have shown that statins slow the progression of aortic valve dis-
ease (Table 42.1). In these studies, the clinical indication for the
use of statins was an increased low-density lipoprotein level and
not aortic valve disease. The positive effects of statin in these
database studies may be due to the long duration of treatment in
the patient populations. A recent retrospective study that sepa-
rated patients with aortic valve sclerosis from those with mild
to moderate aortic valve stenosis found that medical therapy in
earlier valve lesions was more beneficial than in more advanced
435
436 IV Valvular Heart Disease

and the fourth, an open-label study, had positive results. The first
Box 42.1. Cardiovascular Risk Factors Associated study tested the effects of atorvastatin in aortic valve disease
With Calcific Aortic Valve Disease (SALTIRE). In this double-blind, placebo-controlled trial,
Increased lipid levels patients with moderate to severe calcific aortic stenosis were
randomly assigned to receive either 80 mg of atorvastatin daily
Hypertension
or a matched placebo. Aortic valve stenosis was monitored with
Male sex Doppler echocardiography, and calcification was monitored with
Metabolic syndrome the use of helical computed tomography. The primary end points
Smoking were change in aortic jet velocity and aortic valve calcium score.
Diabetes Secondary end points were a composite of clinical end points:
Renal failure aortic valve replacement, death from any cause, hospitalization
for any cause, and hospitalization for cardiovascular causes. The
Increased C-reactive protein level
results of the SALTIRE trial were negative (Figure 42.3). The
Hyperhomocysteinemia investigators concluded that intensive lipid-lowering therapy
does not halt the progression of calcific aortic stenosis or induce
Previously published. See “Credit Lines” section. its regression.
The RAAVE trial assigned rosuvastatin to those patients who
had aortic stenosis and increased serum low-density lipoprotein
valve stenosis. Thus the initiation of therapy in early aortic valve levels and compared them with similar patients who had normal
disease may be of critical importance (Table 42.1). levels. The aim of the open-label study was to assess the effect
• Retrospective studies suggest a role for statins in aortic valve
of rosuvastatin on the hemodynamic progression and inflamma-
disease. tory markers of aortic stenosis by treating increased levels of
low-density lipoprotein according to the NCEP-ATP III guide-
lines for 1 year. The statin group had significant improvement
Prospective Trials in Calcific Aortic Valve in serum lipid and echocardiographic measures of aortic steno-
Disease sis (Figure 42.4). Treatment with rosuvastatin slowed progres-
Four prospective studies have evaluated the effects of statins in sion of aortic valve peak velocity and improved inflammatory
aortic valve disease. Three of the studies had negative results, biomarkers.

Lipids

Circulating monocytes

Endothelial cells

Macrophages

Aortic valve fibroblast cells

Proliferation
Statins Pre-osteoblast

Matrix synthesis
Osteopontin
Alk phos
Sox 9
Osteoblast

Bone formation

Figure 42.1. Mechanism of Development of Calcific Aortic Valve Disease in the Presence of Hypercholesterolemia. alk phos indicates alkaline
phosphatase; LDL, low-density lipoprotein. (Previously published. See “Credit Lines” section.)
 42 Pathogenesis of Valvular Heart Disease
Box 42.2. Osteoblast Regulatory Pathways Asso-
ciated With Calcific Aortic Disease
Angiogenesis and apoptosis
Bone matrix protein expression: OPN, OCN, ON,
alk phos, GAG, collagen, FN
Transcription factors: Cbfa1, Msx2, Sox9, Egr1
Cartilage: hypertrophic chondrocytes
Bone: endochondral ossification
Signaling markers: Lrp5/Wnt, TGF-β, TNF-α, FGF,
PDGF, IL6, RANKL, VEGF
Osteoclast bone resorption
MMP/tenascin/purine nucleotides
Inflammatory cells and lipids
Abbreviations: alk phos; alkaline phosphatase; Cbfa1; core-
binding factor α-1; Egr1, early growth response-1; FGF, fibroblast
growth factor; FN, fibronectin; GAG, glycosaminoglycan; IL-6,
interleukin-6; Lrp5/Wnt, low-density lipoprotein-receptor–related
protein 5/Wnt; MMP; matrix metalloproteinases; ON, osteonectin;
OPN, osteopontin; PDGF, platelet-derived growth factor receptor
β; RANKL, receptor activator of nuclear factor-κB ligand; TGF-β,
transforming growth factor-β; TNF-α, tumor necrosis factor-α;
VEGF, vascular endothelial growth factor.
Previously published. See “Credit Lines” section.
The SEAS study was the largest trial to date: 1,873 patients
at 173 study sites in 7 European countries. The mean serum level
of low-density lipoprotein remained unchanged in the placebo
group and decreased by 61.3% at 8 weeks in the simvastatin–
ezetimibe group. During the entire follow-up period, the mean
percentage reduction in low-density lipoprotein was 53.8% in
the simvastatin–ezetimibe group and 3.8% in the placebo group.
There was no difference between the 2 groups at any time in
terms of aortic stenosis progression, nor was there any significant
difference in the aortic valve area from baseline.
The most recent trial of statins in aortic valve disease, the
ASTRONOMER trial, was also designed to assess the effect of
cholesterol lowering with rosuvastatin on the progression of aor-
tic stenosis. This randomized, double-blind, placebo-controlled
trial included asymptomatic patients with mild to moderate
aortic stenosis and no other clinical indications for cholesterol
lowering. The 269 patients were randomized to receive either
placebo (n = 135) or rosuvastatin 40 mg daily (n = 134). Annual
echocardiograms were obtained to assess the progression of aor-
tic stenosis, which was the primary outcome. The median dura-
tion of follow-up was 3.5 years. Treatment with rosuvastatin was
not associated with reduction in the progression of aortic stenosis
in any of the predefined subgroups (Figure 42.5).
• Prospective studies demonstrate conflicting results regarding the
use of statins for treatment of aortic valve disease; however, differ-
ent trial designs may explain the differences in trial results.
Data on the effect of statins in aortic valve disease show con-
flicting results between positive retrospective studies and gener-
ally negative prospective studies. In patients with an increased
level of low-density lipoprotein and aortic valve disease, statins
are recommended on the basis of vascular disease prevention and
may have an added potential benefit of slowing aortic disease
progression. Statins are not indicated to reduce progression of
aortic stenosis per se.
437
Acute Aortic Valve Regurgitation
Acute aortic valve regurgitation most commonly is due to
either infective endocarditis or aortic dissection. Rarer causes
include blunt force chest trauma, deceleration injuries from
auto accidents, or iatrogenic injury during aortic valvuloplasty.
Infective endocarditis causes destruction and perforation of the
valve leaflets and perivalvular abscess formation; aortic dis-
section results in dilatation of the ascending aorta with poor
coaptation of the valve leaflets and in genesis of a flail valve
leaflet due to destruction of the valve leaflet aortic wall support
mechanism.
Chronic Aortic Valve Regurgitation
Chronic aortic valve regurgitation results from a valve leaflet
mechanism in 60% of cases (usually associated with a bicuspid
valve or infective endocarditis) and from aortic root dilatation,
injury, or dissection in the remaining 40% of cases. Systemic
hypertension, Marfan syndrome, and the aortopathy associated
with a bicuspid valve are important risk factors for aortic root
abnormality. Other associations with chronic aortic regurgita-
tion are ankylosing spondylitis, reactive arthritis, and syphilitic
aortitis.
Mitral Valve
Mitral Annular Calcification
Mitral annular calcification is probably caused by an atheroscle-
rotic process similar to aortic valve sclerosis, with which it is fre-
quently associated. Chronic renal failure, dialysis, hypertrophic
cardiomyopathy, and hyperparathyroidism are associated with
mitral annular calcification.
Mitral Valve Stenosis
Mitral stenosis is caused by postinflammatory thickening of the
mitral valve leaflets and chordal apparatus due almost solely to
rheumatic fever, although 40% of patients will have no clinical
history of prior rheumatic fever. Rare causes of mitral stenosis
include congenital malformation of the mitral valve, systemic
rheumatologic inflammatory disorders such as systemic lupus
erythematosus or rheumatoid arthritis, left-sided carcinoid heart
disease, or endomyocardial fibrosis.
Acute Mitral Valve Regurgitation
Acute mitral valve regurgitation occurs due to spontaneous
development of a flail mitral valve leaflet in mitral valve pro-
lapse syndrome. Valve leaflet or chordal apparatus destruction
is common in infective endocarditis or, in the developing world,
in acute rheumatic fever. Papillary muscle rupture may occur
in acute myocardial infarction. Severe ischemia of a papillary
muscle can also cause valve dysfunction without muscle rupture.
Traumatic rupture of the mitral valve occurs with blunt chest
trauma or deceleration injuries.
Chronic Mitral Valve Regurgitation
The mitral valve is a complex valve that requires close approx-
imation of valve leaflets, integrity of the chordae tendineae,
438 IV Valvular Heart Disease

Cholesterol
Cardiovasular Receptor Statin AT Blockade AT2
Risk Factors NOTCH1
APOE
Biology
Endothelial
Cells

Statin
PDGF
TNF-α
Cell IL-6
Cell Cell Cell
Proliferation BMP Proliferation Proliferation Proliferation
Cleaved NOTCH1 MMP NOTCH Bradykinin Bradykinin
Cathepsin Uncleaved
Wnt
Vitamin D Lrp5 Estrogen
Receptor Frizzled Receptor

Valvular
Myofibroblast
Cells

OPN Hrt
BSP Bone Matrix Bone Matrix Bone Matrix
Msx2
ON Synthesis Cbfa1 Wnt
Synthesis Synthesis
Alk Phos
Lrp5
Frizzled
Bone
Formation
Osteoblast
Phenotype

Hydroxy Apatite/Calcium

Figure 42.2. Signaling Pathways Identified in Calcific Aortic Valve Disease. ACE indicates angiotensin-converting enzyme; Alk Phos, alkaline
phosphatase; APOE, apolipoprotein E; AT, angiotensin; BMP, bone morphogenetic protein; BSP, bone sialoprotein; Cbfa1, osteoblast-specific tran-
scription factor for bone formation; eNOS, endothelial nitric oxide synthase; Hrt, hairy-related transcription factor; IL-6, interleukin-6; LDL, low-
density lipoprotein; Lrp5, LDL receptor–related protein 5; MMP, matrix metalloproteinase; NO, nitric oxide; ON, osteonectin; OPN, osteopontin;
PDGF, platelet-derived growth factor receptor β; TNF-α, tumor necrosis factor-α. (Previously published. See “Credit Lines” section.)

synchronized contraction of papillary muscles, and structural chordae tendineae. Rheumatologic disorders associated with
reliability of the left ventricular annulus to function normally. systemic inflammation (systemic lupus erythematosus, rheuma-
Myxomatous degeneration (mitral valve prolapse) is the com- toid arthritis, and scleroderma) and defective connective tissue
monest cause of chronic mitral regurgitation and causes degen- formation (Marfan syndrome, Ehlers-Danlos syndrome, pseu-
eration (cartilaginous changes) in both valve leaflets and the doxanthoma elasticum) are associated with mitral valve pro-
lapse. A congenitally abnormal mitral valve in conditions such
as endocardial cushion defect, parachute mitral valve, or cleft
Box 42.3. Genetic Factors Identified in Calcific mitral valve may lead to mitral regurgitation. Rheumatic valve is
Aortic Valve Disease a leading cause of mixed mitral stenosis and regurgitation in the
developing world.
Vitamin D receptor Mitral regurgitation due to dilatation of the left ventricular
Apolipoprotein AI, B, and E annulus can occur in ischemic and nonischemic dilated cardio-
Estrogen receptor-α gene myopathy, whereas hypertrophic cardiomyopathy can cause
NOTCH1 gene trauma to the mitral valve apparatus due to systolic anterior
Angiotensin-converting enzyme motion of the mitral valve. Ischemia or infarction of the papil-
lary muscles will also cause mitral regurgitation.
Interleukin-10
Advanced carcinoid syndrome or carcinoid heart disease in
Connective tissue growth factor the setting of an atrial left-to-right shunt (atrial septal defect,
Chemokine receptor 5 patent foramen ovale) or a bronchial carcinoid, all of which
PON1 gene allow high levels of serotonin into the left heart circulation, can
cause thickening and poor function of the mitral valve leaflets
Previously published. See “Credit Lines” section. and chordal apparatus, results leading to mitral regurgitation.
Similar effects occur with drugs that cause a chronic increase
Table 42.1. Retrospective Clinical Studies Evaluating Role of Statins in Calcific Aortic Valve Disease
Change in Mean Echocardiographic Outcome Variable

Mean Duration EBT of Aortic Valve—

of Follow-up Annualized Peak Gradient, Mean Gradient, Jet Velocity, Annualized Calcium
Retrospective Study (range), mo Study Group AVA, cm 2 mm Hg mm Hg m/s per y Score, %
Novaro et al, 2001 21 (12–40) No statin (n = 117) −0.11 9.5 5.8
Statin users (n = 57) −0.06 5.3 4.2
(P = .03)
Pohle et al, 2001 15 (10–36) LDL-C >130 mg/dL (n = 47) 43.2
LDL-C ≤130 mg/dL (n = 57) 9.1
(P<.001)
Aronow et al, 2001 33 (24–120) LDL-C ≥125 mg/dL, no statin (n = 69) 6.3
LDL-C ≥125 mg/dL, with statin (n = 62) 3.4 (P<.001)
LDL <125 mg/dL, no statin (n = 49) 3.1 (P<.001)
Shavelle et al, 2002 30 No statin (n = 37) 32.0
Statin users (n = 28) 12.1
Rosenhek et al, 2004 24 (6–42) No statin (n = 161) 0.39 ± 0.42
Statin users (n = 50) 0.10 ± 0.41
(P = .0001)
Bellamy et al, 2002 44 (28) No statin (n = 118) −0.09 17
Statin users (n = 38) −0.04 9
(P = .04)
Change in Aortic Valve Disease Severity

Mild Aortic Moderate Aortic

Aortic Sclerosis Stenosis Stenosis
Antonini-Canterin, 2008 67 (24–228) No statin (n = 737) 48% 55% 71%
Statin users (n = 309) 33% 37% 61%
P≤.01 P<.0001 P≤.10
Abbreviations: EBT, electron beam tomography; LDL-C, low-density lipoprotein cholesterol.
Previously published. See “Credit Lines” section.
440 IV Valvular Heart Disease

A A
5.0 5.0
Placebo Untreated patients
P=.007
Atorvastatin Treated patients
Aortic Jet Velocity, m/s
4.5 4.5

Aortic Jet Veloity, ms-1

4.0 4.0

3.5
3.5
3.0
3.0
2.5
2.5
0
0 12 24 36
Months 0 26 52 78
No. of Patients Weeks of treatment
Placebo 77 69 55 30 No. of Patients
Atorvastatin 77 65 60 34 Untreated 60 50 45 35
Treated 61 51 37 27

B
5.0
B
Placebo 70
Atorvastatin Untreated patients
Treated patients P=.049
4.5
Log CT Aortic Valve

60
Calcium Score, AU

Mean Gradient, mm Hg
4.0 50

3.5 40

3.0 30

2.5 20

0 10
0 12 24 36
Months
0 26 52 78
No. of Patients
Placebo 76 69 56 29 Weeks of treatment
Atorvastatin 77 64 60 34 No. of Patients
Untreated 60 50 45 35
Treated 61 51 37 27
C
200 P<.001 C
2.0 Untreated patients
Serum LDL Cholesterol
Concentration, mg/dL

Treated patients P=.041

150
Aortic Valve Area, cm2

1.5
100

50 1.0
Placebo
Atorvastatin
0 0.6
0 12 24 36
Months
No. of Patients 0 26 52 78
Placebo 78 70 54 31 Weeks of treatment
Atorvastatin 77 65 58 35 No. of Patients
Untreated 60 50 45 35
Treated 61 51 37 27
Figure 42.3. Results of SALTIRE Trial. A, Increases in aortic jet
velocity were 0.199 ± 0.210 m/s per year in the atorvastatin group and
0.203 ± 0.208 m/s per year in the placebo group (P = .95). B, Progression Figure 42.4. Results of RAAVE Trial During Mean Follow-up of
in valvular calcification was 22.3% ± 21.0% per year in the atorvasta- 73 ± 24 Weeks. A, The increase in aortic valve velocity was 0.24 ±
tin group and 21.7% ± 19.8% per year in the placebo group (P = .93). 0.30 m/s per year in the control group and 0.04 ± 0.38 m/s per year in
C, Serum concentrations of low-density lipoprotein remained at 130 ± the rosuvastatin group (P = .007). B, The change in mean gradient was
30 mg/dL in the placebo group and decreased to 63 ± 23 mg/dL in the +5.06 ± 7.17 mm Hg per year in the control group and +2.08 ± 8.15
atorvastatin group (P<.001). AU indicates arbitrary units; CT, computed mm Hg per year in the rosuvastatin group (P = .049). C, The change
tomography; LDL, low-density lipoprotein. (Previously published. See in aortic valve area was −0.10 ± 0.09 cm2 /year in the control group and
“Credit Lines” section.) −0.05 ± 0.12 cm2 /year in the rosuvastatin group (P = .041). (Previously
published. See “Credit Lines” section.)
 42 Pathogenesis of Valvular Heart Disease 441

A Tricuspid Valve
25
Rosuvastatin Tricuspid Valve Stenosis
Placebo
From Baseline, mm Hg
Peak Gradient Change

20 Tricuspid stenosis is a rare valve lesion that is usually due to

rheumatic heart disease occurring in association with rheumatic
15 mitral stenosis. Other causes include congenital stenosis of the
tricuspid valve and carcinoid syndrome.
10

5
Tricuspid Valve Regurgitation
The commonest cause of tricuspid regurgitation is functional
0 regurgitation (anatomically normal valve leaflet and chordal
0 12 24 36 48 60 apparatus) due to tricuspid annular dilatation in association
Months with left heart disease or pulmonary hypertension due to any
cause. Valvular tricuspid regurgitation is associated with Ebstein
No. of Patients
Rosuvastatin 134 128 106 99 69 34 anomaly of the tricuspid valve, trauma (blunt chest or abdominal
Placebo 135 126 104 88 56 31 injury), iatrogenic injury at the time of endomyocardial biopsy,
or valve leaflet tethering or injury in association with pacemaker
B or automatic implantable cardioverter-defibrillator lead place-
ment. Myxomatous degeneration and prolapse of the tricuspid
0.0
Aortic Valve Area Change

valve occurs frequently in association with mitral valve prolapse.

From Baseline, cm2

Right-sided infective endocarditis usually in association with

-0.1
intravenous drug abuse or noninfective marantic endocarditis
-0.2
can destroy the valve leaflet and chordal apparatus. Carcinoid
syndrome and the now-discontinued anorexia drugs fenflura-
-0.3 mine and phentermine are associated with fibrosis and regurgita-
tion of the tricuspid valve, as is the dopamine agonist pergolide
-0.4 Rosuvastatin used to treat Parkinson disease.
Placebo

0 12 24 36 48 60 Pulmonary Valve
Months
Pulmonary Valve Stenosis
No. of Patients
Rosuvastatin 130 118 98 92 62 34 Pulmonary valve stenosis is usually congenital in origin, most
Placebo 133 122 102 85 56 31 commonly as an isolated condition but also associated with con-
genital rubella syndrome, tetralogy of Fallot, and Noonan syn-
drome. Carcinoid heart disease is a rare noncongenital cause of
Figure 42.5. Results of ASTRONOMER Trial. A, Peak aortic ste-
nosis gradient increased from a baseline of 40.8 ± 11.1 mm Hg to 57.8 pulmonary valve stenosis.
± 22.7 mm Hg at the end of follow-up in patients receiving rosuvastatin
and from 41.6 ± 10.9 mm Hg at baseline to 54.8 ± 19.8 mm Hg at the Pulmonary Valve Regurgitation
end of follow-up in patients receiving placebo. The annualized increase
in the peak aortic stenosis gradient was 6.3 ± 6.9 mm Hg in the rosu- Pulmonary regurgitation is common after balloon valvuloplasty
vastatin group and 6.1 ± 8.2 mm Hg in the placebo group (P = .83). B, for pulmonary stenosis and after surgical repair of tetralogy of
The annual decrease in aortic valve area was 0.08 ± 0.21 cm2 in patients Fallot. Pulmonary hypertension due to left heart disease or pul-
receiving placebo and 0.07 ± 0.15 cm 2 in patients receiving rosuvastatin monary disease can cause secondary pulmonary regurgitation.
(P = .87). (Previously published. See “Credit Lines” section.)
Names of Clinical Trials
in serum serotonin levels or directly stimulate the serotonin ASTRONOMER Aortic Stenosis Progression Observation: Measu-
receptor. The now-discontinued anorexia drugs fenfluramine ring Effects of Rosuvastatin
NCEP-ATP III National Cholesterol Education Program Adult
and dexfenfluramine augment serotonergic activity, whereas
Treatment Panel III
ergotamine and methysergide used to treat migraine and the RAAVE Rosuvastatin Affecting Aortic Valve Endothelium
dopamine agonists pergolide and cabergoline used to treat par- SALTIRE Scottish Aortic Stenosis and Lipid-Lowering
kinsonism are structurally similar to serotonin and cause valve Trial, Impact on Regression
disease by stimulation of serotonin (5-hydroxytryptamine 2B) SEAS Intensive Lipid Lowering With Simvastatin and
receptors. Ezetimibe in Aortic Stenosis

Epidemiology of Valvular Heart Diseases
VUYISILE T. NKOMO, MD, MPH
Typically, valvular heart diseases are left sided and affect the
mitral and aortic valves, which can become regurgitant or
stenotic. Physiologic valve regurgitation is common in the gen-
eral population and is not associated with adverse outcomes, but
regurgitation that is not physiologic, particularly if it is mod-
erate or severe, negatively affects overall survival. Any degree
of valve stenosis is pathologic, and stenosis that is moderate
or severe is associated with decreased survival. Primary right-
sided (tricuspid and pulmonic) valve diseases are less common.
Tricuspid valve regurgitation is usually secondary to left-sided
heart disease or lung disease.
• Any degree of valve stenosis is pathologic, and stenosis that is
moderate or severe is associated with decreased survival.
• Tricuspid valve regurgitation is usually secondary to left-sided
heart disease or lung disease.
Valvular heart diseases are an important public health prob-
lem worldwide. They are classically related to RF and RHD. RF
and RHD are thought to result from an autoimmune reaction to
GAS pharyngeal infection, which leads initially to valve regurgi-
tation (mitral regurgitation more often than aortic regurgitation)
and later to valve stenosis (mitral stenosis more often than aortic
stenosis).
Although the incidence of RHD has decreased markedly
in economically developed countries owing to the dramatic
decrease in the incidence of RF, RHD persists in economically
developing countries owing to the persistently high incidence of
RF. The distribution of RF and RHD is linked to poor socio-
economic status: In developing countries, RHD may account
for up to 60% of cardiovascular diseases in children and young
adults. The course of the disease can be rapid, leading to death
Abbreviations and acronyms are expanded at the end of this chapter.
442
43
from heart failure caused by valve regurgitation or valve steno-
sis (or both) or by endocarditis in teenagers and young adults.
Prevention measures are difficult to implement owing to lack of
infrastructure and resources.
The paradigm that RF and RHD are consequences of only
pharyngeal GAS is not consistent with observations among
Australian Aborigines, who have a high incidence and high
prevalence of RF and RHD but a low incidence of GAS pha-
ryngeal infections. These people have non-GAS pharyngeal and
skin infections, suggesting that RF and RHD may be linked
to non-GAS pharyngeal infections. This situation would have
enormous implications for strategies to prevent RF and RHD,
but similar links to non-GAS pharyngeal infections have not
been found in other communities that have a high prevalence of
RF and RHD.
The prevalence of RHD is highest in sub-Saharan Africa
and among Aborigines of Australia and New Zealand, followed
by south-central Asia, North Africa and the Middle East, Latin
America, and Eastern Europe. However, precise epidemio-
logic estimates of the burden of RHD are difficult to ascertain
because most of the estimates are based on clinical detection
of valve disease, which is not sensitive to detecting subclin-
ical valve disease. For instance, clinical detection of valve
disease by physical examination would estimate the prevalence
of RHD in school-aged children in sub-Saharan Africa to be
5.7 per 1,000 and in south-central Asia to be 2.2 per 1,000.
Echocardiography, however, provides an incremental benefit
in detecting subclinical RHD, and the echocardiographically
determined prevalence of RHD is as high as 30.4 per 1,000
in parts of sub-Saharan Africa and 21.5 per 1,000 in parts of
south-central Asia.
• Epidemiologic estimates of the burden of RHD depend on the
manner of detection.
 43 Epidemiology of Valvular Heart Diseases
The natural history of echocardiographically determined sub-
clinical RHD is similar to that of clinically detected RHD in
patients presenting with an episode of RF. However, the Jones
criteria for RHD are based on clinical findings of valve disease
and not on echocardiographically determined subclinical valve
disease. Nevertheless, in geographic regions with a high preva-
lence of RHD, such as New Zealand and Australia, echocardio-
graphically determined valve findings have been incorporated
into the Jones criteria as part of the guidelines for diagnosing
RHD. This step reflects the higher sensitivity of echocardiog-
raphy compared with physical examination in detecting valve
disease and the high socioeconomic burden of RHD if it is
not detected and treated early. Some of the 2-dimensional and
Doppler criteria for subclinical RHD include valve thickening,
dog-leg or hockey-stick deformity of the anterior mitral leaflet,
leaflet restriction, subvalvular thickening, and valve regurgita-
tion that is not physiologic and is visualized in 2 planes. Because
echocardiography has shown that the prevalence of RHD is much
higher than suspected on the basis of clinical findings in develop-
ing countries and because the social and economic costs of sub-
sequent clinical disease are high, interest has been rejuvenated in
detecting and combating RF and RHD in these countries.
In developed countries, the decreased incidence of RF and
RHD has been associated with an increase in life expectancy and
an increase in age-related valve diseases, although mitral valve
stenosis (the classic lesion of RHD) is typically encountered in
elderly patients as a result of an episode of RF decades earlier.
Migrant populations from developing countries also contribute
to the prevalence of RHD in developing countries.
Age-related calcific aortic valve stenosis, once thought to
be a passive degenerative wear-and-tear phenomenon, has been
shown to be an active atherosclerotic, inflammatory, and calcifi-
cation process linked to similar risk factors for coronary artery
443
disease, such as age, hypertension, smoking, and hyperlipidemia.
A congenitally abnormal aortic valve (usually with bicuspid aor-
tic valve disease) is a common cause of aortic valve stenosis or
regurgitation in adults younger than 65.
• Age-related calcific aortic valve stenosis, once thought to be a pas-
sive degenerative wear-and-tear phenomenon, is an active athero-
sclerotic, inflammatory, and calcification process.
The leading indication for surgery for mitral valve regurgita-
tion in developed countries is mitral valve prolapse or flail mitral
valve from myxomatous mitral valve disease.
In developed countries, the prevalence of valve disease
increases substantially with aging: In the general population,
more than 12% of persons 75 years or older have some form of
left-sided valve disease of at least moderate severity. This degree
of valve disease has been associated with poor survival in popu-
lation-based studies and in community-based studies. Poor sur-
vivorship is associated with valve disease in part because of the
incipient nature of these conditions, the patient’s advanced age
and heart failure at presentation, or the presence of comorbidi-
ties that increase the risk of surgery. Nonetheless, the prevalence
of valve diseases is projected to increase substantially with the
aging population, and this has been an impetus for the develop-
ment of alternative therapies to open heart surgery to treat valve
diseases. Overall, survival increases with valve repair or replace-
ment (or valvuloplasty for mitral stenosis with suitable anatomy)
and can potentially be normalized among patients receiving
intervention at an optimal time.
Abbreviations
GAS group A streptococcal
RF rheumatic fever
RHD rheumatic heart disease
This page intentionally left blank
 Section V

Aorta and Peripheral Vascular Disease

This page intentionally left blank
 44
Peripheral Vascular Disease
PETER C. SPITTELL, MD
Claudication
Lower extremity arterial occlusive disease affects 8 to 12 million
Americans, but 75% have no substantial symptoms. Despite an
overall incidence of 30% in high-risk groups (age >70 years, dia-
betes, tobacco use), the condition is frequently not diagnosed
(>30% of cases). Furthermore, peripheral arterial disease can
provide an important clue to systemic atherosclerosis and iden-
tify persons at increased risk for left ventricular systolic dysfunc-
tion, myocardial infarction, and cerebrovascular accident.
The clinical presentation of lower extremity arterial occlusive
disease is variable and can range from no symptoms (or atypical
symptoms) to intermittent claudication to critical limb ischemia
(rest pain, ulceration, or gangrene). The degree of functional lim-
itation varies depending on the degree of arterial stenosis, collat-
eral circulation, exercise capacity, and comorbid conditions. The
discomfort of intermittent claudication (aching, cramping, or
tightness) is always exercise-induced, may involve 1 leg or both
legs, and occurs at a fairly constant walking distance. Relief is
obtained by standing still. Supine ABIs before and after tread-
mill exercise testing can confirm the diagnosis (Table 44.1).
Pseudoclaudication
Pseudoclaudication is caused by lumbar spinal stenosis and is
the condition most commonly confused with intermittent claudi-
cation. Pseudoclaudication is usually described as a paresthetic
discomfort that occurs with standing and walking (variable dis-
tances). Symptoms almost always are bilateral and relieved by sit-
ting or leaning forward. The patient often has a history of chronic
back pain or previous lumbosacral spinal surgery (Table 44.2).
Abbreviations and acronyms are expanded at the end of this chapter.
The diagnosis of lumbar spinal stenosis can be confirmed from
characteristic findings on CT or MRI of the lumbar spine, elec-
tromyography, and normal or minimally abnormal ABIs before
and after exercise.
• The discomfort of intermittent claudication is always exercise-
induced.
• Pseudoclaudication is usually described as a paresthetic discom-
fort that occurs with standing and walking.
Natural History of Peripheral Vascular Disease
Peripheral arterial occlusive disease is associated with consid-
erable mortality because of its association with coronary and
carotid atherosclerosis. Cardiovascular outcomes at 5 years in
patients with noncritical claudication include nonfatal myocar-
dial infarction or stroke in 20% and death in 15% to 30% (due
to cardiovascular causes in 75%). The prevalence of coronary
artery disease is more than 50% in patients with peripheral arte-
rial disease, and the presence of peripheral arterial disease is an
independent predictor for all-cause mortality in the 2 years after
coronary artery bypass grafting. In patients with critical limb
ischemia (ischemic rest pain, ulceration, and gangrene), 1-year
mortality is 25%. In patients with noncritical limb ischemia,
claudication remains stable in 70% to 80%, worsens in 10% to
20%, and progresses to critical limb ischemia in 1% to 2% over
5 years. When critical limb ischemia is present, the amputation
rate over 1 year is 25%.
Continued use of tobacco results in a 10-fold increase in the
risk for major amputation and a more than twofold increase in
mortality. The effect of diabetes mellitus on patients with inter-
mittent claudication deserves special mention because it is the
reason for the majority of amputations in a community (12-fold
increased risk of below-knee amputation and a cumulative risk of
447
448 V Aorta and Peripheral Vascular Disease

Table 44.1. Grading System for Lower Extremity Occlusive

Arterial Diseasea
ABI

Grade Supine Resting After Exercise

Normal >1.0 No change or increase
Mild disease 0.8–0.9 >0.5
Moderate disease 0.5–0.8 >0.2
Severe disease <0.5 <0.2
Abbreviation: ABI, ankle:brachial systolic pressure index.
a
After treadmill exercise (1–2 mph, 10% grade, 5 minutes or symptom-limited)
or active pedal plantar flexion (50 repetitions or symptom-limited).

major amputation exceeding 11% over 25 years). Diabetes mel-

litus also increases all-cause mortality in persons with periph-
eral arterial disease. The rate of limb loss is also increased in
patients with a progressively declining ankle to brachial systolic
pressure index (ABI) during follow-up. Independent predictors
of a decline in the ABI include advanced age, current tobacco
use, hypertension, diabetes mellitus, and increased level of low-
density lipoprotein cholesterol.
The location of arterial occlusive disease also affects the
overall prognosis. Patients with aortoiliac disease have a lower
5-year survival rate (73%) than those with predominantly fem-
oral artery disease (80%). The increased mortality is attrib-
utable primarily to complications of coronary artery disease.
Diabetes mellitus in combination with femoral artery disease
results in a further decrease in overall survival and an increased
incidence of major amputation. Patients with diabetes melli-
tus, particularly type 2 diabetes mellitus, have a distinct pat-
tern and distribution of atherosclerosis in the lower extremity
arteries. Compared with patients without diabetes, they have
less involvement of the aortoiliac segment, equal occurrence
in the femoropopliteal segments, and more extensive disease
in the infrapopliteal segments (tibial and peroneal arteries)
(Figure 44.1).
Arterial occlusive disease in other locations (eg, the subcla-
vian artery) is also important, especially in patients being con-
sidered for coronary artery bypass grafting and in those who have
recurrent angina after having a left internal mammary artery-left
anterior descending coronary artery bypass. Stenting of a subcla-
vian stenosis may improve myocardial perfusion in patients with
internal mammary artery grafts and an ipsilateral flow-limiting
proximal subclavian stenosis (ie, coronary steal).
• The 5-year mortality rate in patients with intermittent claudication
is 15%-30%, predominantly due to cardiovascular causes.
• Tobacco use, diabetes mellitus, and critical limb ischemia (ischemic
rest pain, ulceration, and gangrene) are associated with an increased
risk of limb loss in patients with intermittent claudication.
Table 44.2. Differential Diagnosis of True Claudication and
Pseudoclaudication
Feature Claudication Pseudoclaudication
Onset Walking Standing and walking
Character Cramp, ache Paresthetic
Bilateral ± +
Walking distance Fairly constant More variable
Cause Atherosclerosis Spinal stenosis
Relief Standing still Sitting down, leaning forward
Figure 44.1. Angiogram Showing Characteristic Infrapopliteal
Location of Arterial Disease in a Patient With Diabetes.
Diagnosis of Peripheral Vascular Disease
Peripheral angiography is rarely needed to diagnose intermittent
claudication. The diagnosis is made clinically and confirmed by
noninvasive testing (ABI before and after exercise). The ABI at
rest provides a measurement of disease severity in patients with
lower extremity arterial occlusive disease: normal, 1.0 to 1.4; mild
disease, 0.8 to 0.9; moderate disease, 0.5 to 0.8; severe disease,
less than 0.5. In addition, an abnormal resting ABI (<0.9) is asso-
ciated with an increased risk for functional impairment, stroke,
cardiovascular death, and all-cause mortality. Furthermore, an
abnormal resting ABI may improve the accuracy of cardiovas-
cular risk prediction beyond the Framingham risk score. An
abnormally high ABI (>1.4) is associated with an increased inci-
dence of foot ulcers, neuropathy, and higher rates of cardiovas-
cular risk. A progressive decline in ABI (>0.15) on follow-up is
independently associated with increased cardiovascular risk and
identifies persons who are candidates for more intensive cardio-
vascular risk-factor management.
Angiography is indicated for 1) defining vessel anatomy pre-
operatively, 2) evaluating therapy, and 3) documenting disease
(medicolegal issue). It is also indicated when an uncommon type
of arterial disease is suspected. MR angiography and CT angiog-
raphy have excellent diagnostic accuracy and are alternatives to
standard angiography. MR angiography is also useful for pre-
operative planning in patients with contraindications to invasive
angiography (ie, renal insufficiency or allergy to contrast media)
(Figure 44.2). Color Doppler imaging is as good as digital sub-
traction angiography for identifying hemodynamically signifi-
cant lesions in the aortoiliac and femoropopliteal regions but is
less accurate in the infrapopliteal segments.
• The diagnosis of intermittent claudication is made clinically
and confirmed with noninvasive testing (ABI before and after
exercise).
Treatment of Peripheral Vascular Disease
Initial medical management of intermittent claudication
involves 3 areas: risk-factor reduction, exercise training, and
 44 Peripheral Vascular Disease
Figure 44.2. Magnetic Resonance Angiogram Showing High-Grade
Stenosis of Left Common Iliac Artery and Occlusion of Right External
Iliac Artery With Collateral Formation.
pharmacologic therapy. In addition, weight reduction (if obese),
foot care and protection, and avoidance of vasoconstrictive drugs
are of benefit. Foot care and protection are of paramount impor-
tance in diabetic patients with peripheral arterial disease. The
combination of peripheral neuropathy, small vessel disease, or
peripheral arterial disease in diabetic patients makes foot trauma
more likely to be associated with a nonhealing wound or ulcer.
Modifiable risk factors for the development of peripheral arte-
rial disease (cigarette smoking, diabetes mellitus, hypertension,
and hyperlipidemia) should be aggressively treated. Tobacco ces-
sation has been shown to reduce the rates of disease progression
and amputation. Diabetes mellitus should be aggressively treated
(goal hemoglobin A1c of <7.0% and as close to 6.0% as possi-
ble). Although there are no data to confirm that modification of
hypertension alters the progression of claudication, hypertension
should be controlled to reduce morbidity and mortality from car-
diovascular and cerebrovascular disease.
All patients with peripheral arterial disease should be pre-
scribed an antiplatelet agent. Aspirin (81–325 mg/day) is effec-
tive in peripheral arterial disease, resulting in a decreased risk
of limb loss, reduced need for vascular surgery, and a decreased
incidence of major coronary and cerebrovascular events.
Clopidogrel (75 mg/day) has been shown to be more effective
than aspirin for preventing major atherosclerotic vascular events.
The CHARISMA trial found that dual-antiplatelet use (aspirin
and clopidogrel) is more effective than either agent alone for
preventing ischemic events in patients with peripheral vascular
disease. Oral anticoagulant therapy in addition to antiplatelet
449
therapy is not more effective than antiplatelet therapy alone in
patients with peripheral arterial disease and is associated with an
increased risk for bleeding.
Exercise training is effective in intermittent claudication. A
regular walking program (level ground, walking the distance
to claudication, stopping to rest for relief, repeatedly for 45–60
minutes per session, 4 or more days a week, continued for 6
months) results in a considerable increase in maximum walking
time (mean difference, 6.5 minutes). For patients who do not ade-
quately respond to a walking program, cilostazol may be useful.
Cilostazol, a phosphodiesterase III inhibitor, results in a more
considerable improvement in walking ability (an approximate
doubling of initial and absolute claudication distance) compared
with placebo and pentoxifylline. Cilostazol is contraindicated in
patients with congestive heart failure of any severity. The dose
is 100 mg orally daily (50 mg orally twice daily in patients tak-
ing diltiazem, ketoconazole, or other inhibitors of cytochrome
P450 3A4). Cilostazol is not recommended for use in all patients
with claudication because of its cost and modest clinical benefit.
Propionyl-L-carnitine seems to improve alterations in carnitine
metabolism in patients with a severe functional impairment from
intermittent claudication, resulting in improvement in maximal
walking distance and quality of life, but overall benefit is not
well established and clinical use is not yet recommended. Statin
pharmacotherapy is indicated in patients with peripheral arte-
rial disease. Statins may improve the symptoms of intermittent
claudication, they reduce the incidence of adverse cardiovas-
cular and cerebrovascular events, and they improve short- and
long-term survival in patients undergoing peripheral vascular
surgery. Angiotensin-converting enzyme inhibitor therapy also
reduces the risk of ischemic cardiovascular events in patients
with peripheral arterial disease; in addition, they have nephro-
protective effects in diabetic patients.
Indications for revascularization (endovascular or surgical) in
a patient with peripheral arterial occlusive disease are disabling
(lifestyle-limiting) symptoms despite optimal medical therapy,
diabetes mellitus with progressive symptoms, or critical limb
ischemia (rest pain, ischemic ulceration, or gangrene).
Revascularization is elective in nondiabetic patients with
intermittent claudication because 1) it does not improve coronary
artery or cerebrovascular disease, the major causes of mortal-
ity, and consequently does not affect overall long-term survival
and 2) the incidence of severe limb-threatening ischemia is rela-
tively low because distal runoff is usually adequate.
Revascularization in patients with rest pain or ischemic
ulceration or in those with diabetes mellitus with progressive
symptoms is indicated because 1) the incidence of limb loss is
increased without revascularization, 2) surgery may permit a
lower anatomical level of amputation, and 3) the risks of the pro-
cedure are generally less than the risk of amputation.
PTA is an effective alternative to surgical therapy in patients
with proximal disease, short, partial occlusions, and good distal
runoff. The ideal lesion for PTA is an iliac stenosis less than 5 cm
or a femoropopliteal occlusion or stenosis less than 10 cm in total
length (excluding lesions involving the origin of the superficial
femoral artery and those affecting the distal 2 cm of the poplit-
eal artery). Advantages of PTA over surgery include less mor-
bidity, shorter convalescence, lower cost, and preservation of the
saphenous vein for future use. The use of PTA in aortic or iliac
disease may also allow for an infrainguinal surgical procedure to
be performed at reduced perioperative risk (compared with intra-
abdominal aortic surgery).
Placement of an iliac stent is indicated when the hemody-
namic results of PTA are inadequate; however, more commonly,
450 V Aorta and Peripheral Vascular Disease
the majority of patients with aortoiliac disease undergo primary
stent placement due to the high rate of arterial dissection and
elastic recoil with PTA alone (Figure 44.3). A randomized trial
of 279 patients with intermittent claudication with an iliac artery
stenosis greater than 50% (proven by angiography) compared
direct stent placement with primary angioplasty, with subsequent
stent placement only in cases of residual gradient. The difference
in the clinical outcomes between the 2 treatment strategies was
not significant at either short-term or long-term follow-up.
• For intermittent claudication, medical management is impor-
tant as initial therapy; this includes antiplatelet agents,
cessation of tobacco use, lipid reduction, foot care and pro-
tection, a walking program, and, in selected cases, cilostazol.
• Diabetes mellitus with progressive symptoms, rest pain, ischemic
ulceration, and gangrene are all associated with an increased risk
of limb loss in patients with lower extremity arterial occlusive
disease. The presence of these features warrants angiography fol-
lowed by revascularization (endovascular or surgical).
Acute Arterial Occlusion
The symptoms of acute arterial occlusion are sudden in onset
(<5 hours) and include the “5 Ps”: pain, pallor, paresthesia
(numbness), poikilothermy (coldness), and absent pulses.
Features that suggest a thrombotic cause of acute arterial
occlusion include previous occlusive disease in the involved
A B
Figure 44.3. Angiograms Showing Stenosis of Right Common Iliac Artery Before (A) and After (B) Percutaneous Transluminal Angioplasty
and Stent Placement.
limb, occlusive disease involving other extremities, acute aortic
dissection, hematologic disease, arteritis, inflammatory bowel
disease, neoplasm, and ergotism.
An embolic cause of acute arterial occlusion is suggested by
the presence of ischemic or valvular heart disease, atrial fibrilla-
tion, proximal aneurysm, or atherosclerotic disease.
After institution of intravenous heparin and foot protection,
angiography is performed. After confirmation of the diagnosis
with angiography, the initial therapeutic options for acute arterial
occlusion include intra-arterial thrombolysis and surgical ther-
apy (thromboembolectomy). If thrombolysis is the initial treat-
ment, PTA or surgical therapy is often required subsequently to
treat the underlying stenosis (if present) to improve long-term
patency rates.
• The “5 Ps” suggestive of acute arterial occlusion include pain, pal-
lor, paresthesia, poikilothermy, and absent pulses.
Aneurysms
Because aneurysms are caused most commonly by atherosclero-
sis, they are more common in men 60 years or older. Coronary
artery and carotid artery occlusive disease are frequent comorbid
conditions. Other predisposing factors for aneurysmal disease
include hypertension, familial tendency, connective tissue
disease, trauma, infection, and inflammatory disease.
 44 Peripheral Vascular Disease 451

Most aneurysms are asymptomatic. Complications of aneu- occlusion with normal proximal vessels (Figure 44.4). Treatment
rysms include embolization, pressure on surrounding structures, of thromboangiitis obliterans is the same as that for other types
infection, and rupture. Aneurysms of certain arteries develop of occlusive peripheral arterial disease, but particular empha-
specific complications more often than other complications. For sis is placed on the need for permanent abstinence from all
example, the most common complication of aortic aneurysms is forms of tobacco. Smoking cessation ameliorates the course of
rupture, and a common complication of femoral and popliteal the disease but does not invariably stop further exacerbations.
artery aneurysms is embolism. Aortic aneurysms are discussed Abstinence from tobacco use also substantially reduces the risk
in the chapter on the aorta. of ulcer formation and amputation. Because the arteries involved
An iliac artery aneurysm usually occurs in association with an are small, arterial reconstruction for ischemia in patients with
abdominal aortic aneurysm, but it may occur as an isolated find- Buerger disease is technically challenging. Distal arterial recon-
ing. Iliac artery aneurysms may cause atheroembolism, obstruc- struction, if necessary, is indicated to prevent ischemic limb loss.
tive urologic symptoms, unexplained groin or perineal pain, or Collateral artery bypass is an option when the main arteries are
iliac vein obstruction. The preferred diagnostic procedure is CT affected by the disease. A patent but diseased artery should be
with intravenous contrast agent or MRI. Surgical resection is avoided as a target for reconstruction. In severe digital ischemia
indicated when an iliac artery aneurysm is symptomatic or larger with ulceration, sympathectomy may be useful to control pain
than 3 cm in diameter. and to improve cutaneous blood flow. Therapeutic angiogenesis
Popliteal artery aneurysms can be complicated by throm-
bosis, venous obstruction, embolization, popliteal neuropathy,
popliteal thrombophlebitis, rupture, and infection. They are
bilateral in 50% of patients, and 40% of patients have 1 or more
aneurysms at other sites, usually of the abdominal aorta. The
diagnosis is readily made with ultrasonography, but angiogra-
phy is necessary before surgical treatment to evaluate the proxi-
mal and distal arterial circulation. When a popliteal aneurysm is
diagnosed, surgery is the treatment of choice to prevent serious
thromboembolic complications.
• An iliac artery aneurysm usually occurs in association with an
abdominal aortic aneurysm, but it may occur as an isolated finding.
• Popliteal artery aneurysms are bilateral in 50% of patients, and
40% of patients have 1 or more aneurysms at other sites.

Uncommon Types of Arterial Occlusive Disease

The clinical features that suggest an uncommon type of periph-
eral arterial occlusive disease include young age, acute ischemia
without a history of arterial occlusive disease, and involvement
of only the upper extremity or digits. Uncommon types of arte-
rial occlusive disease include thromboangiitis obliterans, arteri-
tis associated with connective tissue disease, giant cell arteritis
(cranial and Takayasu disease), and arterial occlusive disease
due to blunt trauma or arterial entrapment.

Thromboangiitis Obliterans (Buerger Disease)

The diagnostic clinical criteria for thromboangiitis obliterans
(Buerger disease) are listed in Table 44.3. More definitive diag-
nosis of thromboangiitis obliterans requires angiography, which
usually shows multiple, bilateral focal segments of stenosis or

Table 44.3. Diagnostic Criteria for Thromboangiitis Obliterans

Age <40 years (often <30 years)
Sex Males most often
Habits Tobacco use
History Superficial phlebitis
Claudication, arch or calf
Raynaud phenomenon
Examination Small arteries involved
Upper extremity involved (positive Allen test)
Laboratory results Normal glucose, blood cell count, erythrocyte
sedimentation rate, lipids, and screening tests
for connective tissue disease Figure 44.4. Angiogram Showing Characteristic Abrupt Occlusions,
Radiography No arterial calcification Segmental Stenoses, and Corkscrew Collaterals of Infrapopliteal
Arteries in a Patient With Thromboangiitis Obliterans.
452 V Aorta and Peripheral Vascular Disease
with vascular endothelial growth factor gene transfer may be
beneficial in patients with advanced Buerger disease that is unre-
sponsive to standard medical or surgical treatment methods.
Popliteal Artery Entrapment
PAE is an uncommon congenital abnormality that is often over-
looked clinically. PAE is important because repeated compres-
sion of the popliteal artery can lead to localized atherosclerosis,
poststenotic dilatation, or thrombosis resulting in serious ische-
mia in the distal leg or foot. PAE occurs most often in young men,
who may present with a complaint of intermittent claudication in
the arch of the foot or calf. If the popliteal artery is not already
occluded, the finding of reduced pedal pulses with sustained
active plantarflexion should increase suspicion of the disorder.
PAE can occur by several mechanisms. The artery can be
compressed because of its anomalous relationship to the medial
head of the gastrocnemius muscle (looping around and under or
through the muscle), because of its displacement by an anom-
alous insertion of the plantaris muscle, or by passing beneath
rather than behind the popliteal muscle.
PAE can be diagnosed noninvasively with duplex ultrasonog-
raphy, CT, and MRI. MRI is superior to ultrasonography and CT
for defining the exact abnormality in PAE, with results similar to
those with digital subtraction angiography. The combined mor-
phologic and functional evaluation of the popliteal fossa makes
MRI the investigation of choice in young adults with intermittent
claudication. MRI is particularly useful when the popliteal artery
is occluded, in which situation ultrasonography and angiography
are of limited value.
Angiographic findings in PAE include irregularity of the
wall of the popliteal artery in an otherwise normal arterial tree,
often associated with prestenotic or poststenotic dilatation. If the
artery is still patent, medial displacement of the popliteal artery
from its normal position in the popliteal space and popliteal
artery compression with extension of the knee and dorsiflexion
of the foot are diagnostic angiographic findings. If the mecha-
nism of compression is by the plantaris or popliteal muscle, the
position of the artery may appear normal on angiography. If PAE
has been diagnosed in 1 limb, the contralateral limb should be
screened because bilateral disease occurs in more than 25% of
patients.
The management of PAE depends on the clinical presentation
and anatomical findings. Although the natural history of PAE
is not well defined, surgery has been advocated to prevent pro-
gression of the disease from repetitive arterial trauma. Detection
and treatment of PAE at an early stage appear to permit better
long-term results.
Thoracic Outlet Compression Syndrome
Compression of the subclavian artery in the thoracic outlet (tho-
racic outlet compression syndrome) can occur at several points,
but the most common site of compression is in the costoclavicu-
lar space between the uppermost rib (cervical rib or first rib) and
the clavicle. If the patient is symptomatic, the presentation may
be any of the following: Raynaud phenomenon in 1 or more fin-
gers of the ipsilateral hand, digital cyanosis or ulceration, and
“claudication” of the arm or forearm. Arterial occlusive disease
in the affected arm or hand is readily detected on examination
of the arterial pulses and with the Allen test. Compression of
the subclavian artery in the thoracic outlet can be demonstrated
by noting a decreased or absent pulse in the ipsilateral radial
artery during performance of thoracic outlet maneuvers. Venous
compression resulting in edema and deep vein thrombosis and
neurologic compression resulting in paresthesias or pain also can
occur in the involved extremity. The diagnosis is confirmed with
duplex ultrasonography, MR angiography, or angiography, with
the involved arm in the neutral and hyperabducted position.
The optimal therapy for thoracic outlet compression is con-
troversial. In general, treatment depends on the severity of symp-
toms. In minimally symptomatic patients, physical therapy and
education regarding the relationship of arm and body position
to arterial compression may be sufficient treatment. In patients
who have more severe symptoms, aneurysm formation, distal
embolization, or digital ischemia, surgical treatment is indicated.
Surgical resection of the first thoracic or cervical rib is the most
effective way to relieve the arterial compression. In some cases,
thrombectomy and reconstruction of the subclavian artery in
patients with subclavian artery stenosis, thrombosis, or occlusion
are also indicated. Sympathectomy may be used as an adjunctive
surgical procedure when there is extensive digital or hand ische-
mia. Stent placement in residual subclavian artery stenoses has
been successful but needs to be performed after surgical decom-
pression of the costoclavicular space to decrease the likelihood
of recurrent symptoms or damage to the stent.
• Compression of the subclavian artery in the thoracic outlet can be
demonstrated by noting a decreased or absent pulse in the ipsilateral
radial artery during performance of thoracic outlet maneuvers.
All the connective tissue disorders and giant cell arteritides
can involve peripheral arteries, and symptoms of peripheral arte-
rial involvement may dominate the clinical picture. Other than
the conservative measures already discussed for ischemic limbs,
therapy is directed mainly at the underlying disease. Only after
the inflammatory process is controlled should surgical revascu-
larization of chronically ischemic extremities be performed.
Heparin-Induced Thrombocytopenia
Heparin-induced thrombocytopenia affects between 5% and
10% of patients who receive heparin therapy, but the inci-
dence of arterial or venous thrombosis is less than 1% or 2%.
Heparin-induced thrombocytopenia (type 2) typically occurs 5
to 14 days after heparin exposure and is associated with arte-
rial thrombosis (arterial occlusion, ischemic strokes, myocar-
dial infarction) and venous thrombosis (pulmonary embolism,
phlegmasia cerulea dolens [venous gangrene], and sagittal sinus
thrombosis). The diagnosis of heparin-induced thrombocyto-
penia is primarily clinical—occurrence of thrombocytopenia
during heparin therapy, resolution of thrombocytopenia when
heparin therapy is discontinued, and exclusion of other causes
of thrombocytopenia—and can be confirmed by demonstration
in vitro of a heparin-dependent platelet antibody. Treatment of
type 2 heparin-induced thrombocytopenia includes discontinua-
tion of all forms of heparin exposure (subcutaneous, intravenous,
or heparin flushes and heparin-coated catheters), including low-
molecular-weight heparins. The current anticoagulant of choice
is a thrombin-specific inhibitor, that is, lepirudin, bivalirudin,
or argatroban. Warfarin therapy can be started once the patient
is clinically stabilized with a nonheparin anticoagulant and the
platelet count has recovered to more than 150,000/mcL.
Vasospastic Disorders
Vasospastic disorders are characterized by episodic color changes
of the skin resulting from intermittent spasm of the small arteries
 44 Peripheral Vascular Disease 453

Table 44.4. Comparison of Primary and Secondary Secondary Raynaud phenomenon affects men more often
Raynaud Phenomenon than women, and in most patients the onset is after age 40 years.
It is usually unilateral or asymmetric at onset. Associated pulse
Raynaud Phenomenon deficits, ischemic changes, and systemic signs and symptoms are
Feature Primary Secondary
often present. Identification of the underlying cause is basic to
appropriate treatment for secondary Raynaud phenomenon.
Age at onset <40 yr ≥40 yr The initial laboratory evaluation in a patient with Raynaud
Sex Women Men phenomenon includes complete blood count, erythrocyte sedi-
Bilateral + ± mentation rate, urinalysis, serum protein electrophoresis, and
Symmetric + ± antinuclear antibody test and tests for cryoglobulin, cryofi-
Toes involved + −
brinogen, and cold agglutinins and chest radiography to detect
Ischemic changes − +
disorders not identified by the medical history and physical
Systemic manifestations − +
examination.

• Primary Raynaud disease is more common in women than men,

and arterioles of the skin and digits. Vasospastic disorders are
and its onset usually is before age 40 years.
important because they frequently are a clue to another under-
• Secondary Raynaud phenomenon affects men more often than
lying disorder, such as occlusive arterial disease, connective
women, and in most patients the onset is after age 40 years.
tissue disorders, neurologic disorders, or endocrine disease.
Vasospastic disorders also can develop as adverse effects of some
drugs, specifically of ergot preparations, estrogen replacement Livedo Reticularis
therapy, certain β-blockers, chemotherapeutic agents, interferon,
cyclosporine, clonidine, narcotics, nicotine, and cocaine. Livedo reticularis, the bluish mottling of the skin in a lacy reticu-
lar pattern, is caused by spasm or occlusion of dermal arterioles.
Raynaud Phenomenon Primary livedo reticularis is idiopathic and not associated with
an identifiable underlying disorder. Secondary livedo reticularis
When Raynaud phenomenon is present, several clinical features
is suggested by an abrupt severe onset of symptoms, ischemic
can help differentiate primary Raynaud disease from secondary
changes, and systemic symptoms. Most commonly, it is the result
Raynaud phenomenon (Table 44.4).
of embolism of atheromatous debris from a proximal aneurysm
Primary Raynaud disease is more common in women than men,
or from proximal atheromatous plaques. The appearance of
and its onset usually is before age 40 years. Episodes are charac-
livedo reticularis in a patient older than 50 years should suggest
terized by triphasic color changes (white, blue, red). Symptoms
the possibility of atheroembolism. Other causes of secondary
are usually bilateral and often symmetric and precipitated by emo-
livedo reticularis include connective tissue disease, antiphospho-
tion or exposure to cold. Ischemic or gangrenous changes are not
lipid antibody syndrome, vasculitis, myeloproliferative disorders,
present. The absence of any causal condition and the presence of
dysproteinemias, reflex sympathetic dystrophy, cold injury, and
symptoms for at least 2 years are also required for the diagnosis.
as an adverse effect of amantadine hydrochloride (Symmetrel)
Raynaud disease is a benign condition, with treatment empha-
therapy.
sizing protection from cold exposure and other vasoconstrictive
influences. Occasionally, a patient with severe symptoms not con-
trolled by local measures may benefit from a trial of a calcium- • The appearance of livedo reticularis in a patient older than 50 years
channel blocker or an α1-adrenergic receptor antagonist. should suggest the possibility of atheroembolism.

Figure 44.5. Characteristic Lesions of Chronic Pernio.

454 V Aorta and Peripheral Vascular Disease
Chronic Pernio
Chronic pernio is a vasospastic disorder characterized by sensi-
tivity to cold in patients (usually women) with a past history of
cold injury. Chronic pernio presents with symmetric blueness of
the toes in the autumn and clearing in the spring (Figure 44.5).
Without treatment, the cyanosis may be accompanied by blis-
tering of the skin of the affected toes. The cyanosis is relieved
within a few days after initiating treatment with an α1-adrenergic
receptor antagonist, which can then be used to prevent recur-
rences during the colder months of the year.
Erythromelalgia
Erythromelalgia is the occurrence of red, hot, painful burning
fingers or toes (or both) on exposure to warm temperatures or
after exercise. It is not a vasospastic disorder but is associated
with color change of the skin. It may be primary (idiopathic) or
be secondary to an underlying disorder, most commonly myelo-
proliferative disease, diabetes mellitus, or small fiber neuropathy.
Treatment of the primary form includes avoidance of exposure
to warm temperatures, aspirin, and a β-blocker (nonselective),
which is helpful in some patients. In persons with secondary
erythromelalgia, treatment of the underlying disorder usually
relieves the symptoms.
Suggested Reading
Hirsch AT, Haskal ZJ, Hertzer NR, Bakal CW, Creager MA, Halperin JL,
et al. ACC/AHA 2005 guidelines for the management of patients with
peripheral arterial disease (lower extremity, renal, mesenteric, and
abdominal aortic): executive summary: a collaborative report from
the American Association for Vascular Surgery/Society for Vascular
Surgery, Society for Cardiovascular Angiography and Interventions,
Society for Vascular Medicine and Biology, Society of Interventional
Radiology, and the ACC/AHA Task Force on Practice Guidelines
(Writing Committee to Develop Guidelines for the Management of
Patients With Peripheral Arterial Disease). J Am Coll Cardiol. 2006
Mar 21;47(6):1239–312.
Abbreviations
ABI ankle:brachial systolic pressure index
CT computed tomography
MRI magnetic resonance imaging
PAE popliteal artery entrapment
PTA percutaneous transluminal angioplasty
Name of Clinical Trial
CHARISMA Clopidogrel for High Atherothrombotic Risk and
Ischemic Stabilization, Management, and Avoidance
 45
Cerebrovascular Disease and
Carotid Artery Stenting
PETER C. SPITTELL, MD, and DAVID R. HOLMES JR, MD
Carotid Artery Disease
Carotid artery disease is common in patients with atherosclero-
sis in other vascular beds (ie, coronary and peripheral arteries)
and has a wide spectrum of clinical presentations (asymptomatic
carotid bruit, TIA, or stroke). Stroke is the third leading cause
of death in the United States. Approximately 795,000 persons
experience either a new event (approximately 610,000) or recur-
rent event (185,000) per year. The risk of stroke is related to age.
Persons 85 years or older are at highest risk; among these, stroke
develops in more women than men because of their longer life
expectancy. The stroke rate is also considerably increased in
African Americans and Mexican Americans. Although stroke
remains the third leading cause of mortality, up to 150,000
deaths per year, national statistics indicate that stroke death rate
declined between 1995 and 2005. This observation may relate to
more optimal control of blood pressure. Stroke also remains the
leading cause of disability. Of stroke survivors at 3 months after
the onset of symptoms, 50% to 70% regain functional indepen-
dence, 15% to 30% are permanently disabled, and 20% require
institutional care. The total cost of stroke is approximately $70
billion per year.
An asymptomatic carotid bruit is present in approximately
13% of the population, and this percentage increases with age.
Patients with asymptomatic carotid bruits are at greater risk for
cerebral ischemic events than the general population and have
a higher overall mortality rate, primarily due to complications
of coronary artery disease. In asymptomatic patients, the annual
risk of stroke is strongly related to the severity of stenosis, being
less than 1% for stenoses of less than 60% and increasing to 3% to
5% for stenoses of more than 80%. For symptomatic patients, the
Abbreviations and acronyms are expanded at the end of this chapter.
annual stroke risk is higher than that for asymptomatic patients,
but it is also dependent on the severity of stenosis. In addition,
stroke risk is highest immediately after an initial ischemic event.
TIA is defined as a focal loss of brain function attributed to
cerebral ischemia lasting less than 24 hours and localized to a
limited region of the brain (Boxes 45.1 through 45.3). In contrast,
stroke is a permanent neurologic deficit. Generally, only 40% of
thrombotic strokes are preceded by a TIA. Among patients with
a TIA who do not die of another cause within 5 years, one-third
will have a stroke, 20% of which occur within the first month
after the TIA and 50% within the first year.
• An asymptomatic carotid bruit is present in approximately 13% of
the population, and this percentage increases with age.
Carotid stenosis severity 1-year risk of TIA or stroke
<60% <1%
>80% 3%–5%
Causes of Stroke
The cause of stroke varies from atherothromboembolic to cardio-
embolic to hemorrhagic. Statistics indicate that, of all strokes,
87% are ischemic, 10% are related to intracerebral hemorrhage,
and 3% are the result of subarachnoid hemorrhage. Of these
different categories, mortality is highest with intracerebral
hemorrhage.
Cardioembolic causes of stroke include intracardiac throm-
bus, intracardiac mass lesions, valvular heart disease, infectious
endocarditis, and paradoxic emboli. Other common causes of
stroke are large-vessel occlusive disease (ascending aorta, aor-
tic arch, major branches of the cerebrovascular circulation) and
small-vessel disease (diabetes mellitus, hypertension, arteritis).
Carotid atherosclerosis can produce symptoms either because
455
456 V Aorta and Peripheral Vascular Disease
Box 45.1. Clinical Features of Transient Ischemic
Attack (Transient Cerebral Ischemia)
Short-term symptoms: minutes to <24 hours
Rapid onset
Spontaneous occurrence
Focal neurologic signs
Patient is conscious
Patient has normal examination results between
attacks
of embolization of thrombus or plaque debris or because of
hypoperfusion. Less common causes include hematologic dis-
ease (polycythemia vera, thrombocytosis, leukemia, thrombo-
philia, antiphospholipid antibody syndromes, cryoglobulinemia,
paraproteinemia). Rare causes are air and fat emboli, cortical
vein thrombosis, and global hypoperfusion. Major risk factors
for stroke include advanced age, hypertension, diabetes mellitus,
hyperlipidemia, cigarette smoking, and excessive use of alcohol.
• Of all strokes, 87% are ischemic, 10% are related to intracerebral
hemorrhage, and 3% are the result of subarachnoid hemorrhage.
• Major risk factors for stroke include advanced age, hypertension,
diabetes mellitus, hyperlipidemia, cigarette smoking, and exces-
sive use of alcohol.
Noninvasive Imaging
Given the dire consequences of stroke and the fact that stroke
may not be preceded by warning symptoms or signs such as
TIA, there has been great interest in predicting and identifying
patients at risk. Identification of carotid artery stenosis has been
an important component of this process.
Noninvasive evaluation of a carotid bruit includes oculoplethys-
mography or duplex ultrasonography. Oculoplethysmography
measures ocular arterial pressure and is an indirect method of
determining whether a hemodynamically significant stenosis is
present in the ipsilateral carotid artery. Persons with an asymp-
tomatic carotid bruit and abnormal oculoplethysmographic result
have a 2-fold greater risk of stroke over 3 years compared with
an age-matched normal population. Limitations of oculoplethys-
mography include its inability to further localize disease and
contraindications in patients with glaucoma or previous ophthal-
mologic surgery.
Duplex ultrasonography (2-dimensional, pulsed-wave
Doppler, and color flow Doppler imaging) is able to provide
both anatomical and hemodynamic information about the
extracranial carotid artery and is the noninvasive test of choice
Box 45.2. Features of a Transient Ischemic Attack in
the Territory of the Carotid Artery
Monoparesis or hemiparesis
Numbness (unilateral)
Impaired vision (unilateral)
Aphasia, dominant hemisphere
Carotid bruit
Retinal findings (cholesterol emboli)
Box 45.3. Features of a Transient Ischemic Attack in
the Territory of the Vertebrobasilar Arterial System
Limb paresis
Drop attacks
Numbness (limbs and face)
Impaired vision (diplopia or bilateral visual field
defects)
Vertigo, nausea
Dysarthria
Ataxia
in the initial evaluation of a patient with a carotid bruit or cere-
bral ischemic symptoms. Duplex ultrasonography can be used to
quantify the severity of a carotid artery stenosis into categories
of diameter reduction and also permits qualitative estimates of
the degree of plaque and its location. Interpretation of the results
of Doppler imaging after stent placement is complex and may
be affected by the specific stent configuration. Evaluation of the
subclavian arteries for evidence of occlusive disease or aneurys-
mal disease and of the vertebral arteries for vessel patency and
direction of flow is also possible with ultrasonography.
Angiography
Magnetic resonance angiography can provide accurate diagnosis
of extracranial carotid artery disease and provide details about
intracerebral arterial anatomy. Magnetic resonance angiogra-
phy is primarily indicated for patients with symptomatic carotid
artery disease to further characterize the internal carotid disease
and to exclude intracerebral and arterial occlusive disease before
carotid endarterectomy. Multidetector computed tomography
angiography also demonstrates excellent agreement with selec-
tive carotid arteriography. Standard angiography is generally
indicated before carotid revascularization or endarterectomy
to define precisely the extent of the extracranial carotid artery
disease and to determine whether there is associated intracere-
bral arterial occlusive disease when the results of magnetic reso-
nance angiography are inconclusive (Figure 45.1). Angiography
is associated with a stroke risk of about 1%. This may be the
result of an embolic complication with selective catheter manip-
ulation in the aortic arch.
Carotid Endarterectomy
Carotid endarterectomy is superior to medical therapy for
patients with symptomatic carotid artery disease and a high-
grade stenosis, including carotid territory or retinal TIAs or
nondisabling stroke with an ipsilateral high-grade carotid artery
stenosis (70%–99%). Medical therapy is better for persons with
mild carotid artery stenosis (0%–29%), even when symptomatic.
The treatment of moderate carotid artery stenosis (30%–59%) is
controversial and must be individualized.
Veterans Administration Asymptomatic
Carotid Stenosis Study
The Veterans Administration study of 444 male patients with
asymptomatic carotid stenosis of more than 50% on angiography
found that carotid endarterectomy in combination with medical
therapy reduced the incidence of ipsilateral neurologic events to
 45 Cerebrovascular Disease and Carotid Artery Stenting
Figure 45.1. Cerebral Angiogram Showing Severe Stenosis of
Proximal Right Internal Carotid Artery.
8% (over a mean follow-up of 4 years) in comparison with a 21%
event rate with medical therapy alone. However, the combined
end-point risk of stroke and death was not statistically reduced
in this study, a finding emphasizing the need for individualized
patient management, because most persons die of their associ-
ated coronary artery disease. Lowering low-density lipoprotein
cholesterol levels with statin pharmacotherapy reduces the stroke
risk in patients with coronary artery disease.
• Carotid endarterectomy is superior to medical therapy for
patients with symptomatic carotid artery disease and a high-grade
stenosis.
• Medical therapy is better for persons with mild carotid artery
stenosis, even when symptomatic.
• Lowering low-density lipoprotein cholesterol levels with statin
medication reduces the stroke risk in patients with coronary artery
disease.
Asymptomatic Carotid Atherosclerosis Study
The ACAS showed that carotid endarterectomy is superior to
medical therapy (aspirin and reduction of risk factors) in persons
with asymptomatic carotid stenosis of more than 60% if their
general health is good and the medical center has a documented
combined perioperative morbidity and mortality rate of less than
3%. When these criteria were met, the aggregate risk for stroke
or death over 5 years was 5% in the surgical group and 11% in
the medical group. There was a marked sex effect in this study:
the resultant 66% relative risk reduction in men was statistically
457
significant, but the comparable risk reduction of 17% in women
was not statistically significant.
Currently, the approach to a patient with asymptomatic caro-
tid artery stenosis of less than 60% is aggressive risk factor
reduction with emphasis placed on meticulous blood pressure
control, discontinuation of tobacco use, control of diabetes mel-
litus, and treatment of hyperlipidemia (preferably with a statin).
Antiplatelet therapy is indicated in all patients unless absolute
contraindications exist. Follow-up includes both a clinical exam-
ination and noninvasive imaging at periodic intervals.
In a patient requiring coronary artery bypass grafting who
has considerable carotid artery disease, management needs to
be individualized because the available evidence is somewhat
controversial. Existing data for this group of patients suggest
that carotid endarterectomy (when indicated according to the
aforementioned criteria) performed simultaneously with coro-
nary artery bypass grafting results in a lower stroke rate than
delayed carotid endarterectomy performed within 2 weeks after
the coronary artery surgery.
In summary, carotid endarterectomy is beneficial for symp-
tomatic patients with recent nondisabling carotid artery ische-
mic events and ipsilateral carotid artery stenosis of 70% to
99%. Carotid endarterectomy is not beneficial for symptomatic
patients with stenosis of 0% to 29%. The potential benefit of
carotid endarterectomy for symptomatic patients with stenosis of
30% to 69% is uncertain. For asymptomatic patients, guidelines
for carotid endarterectomy include surgical risk less than 3% and
a life expectancy of at least 5 years in the presence of a carotid
artery stenosis of more than 60% or a carotid artery stenosis of
more than 60% in patients scheduled to undergo coronary artery
bypass grafting. There are no proven absolute indications for
carotid endarterectomy for patients with a surgical risk of 3%
or more. However, for patients with a surgical risk of 3% to 5%,
an acceptable indication for ipsilateral carotid endarterectomy is
stenosis of 75% or more in the presence of contralateral internal
carotid artery stenosis of 75% or more.
Carotid endarterectomy is superior to medical therapy for per-
sons with symptomatic carotid territory or retinal TIAs or non-
disabling stroke and ipsilateral high-grade carotid artery stenosis
(70%–99%) and for persons with asymptomatic carotid stenosis
of more than 60% if their general health is good and the medical
center has a documented combined perioperative morbidity and
mortality rate of less than 3%.
Carotid Artery Stenting
Although the average risk for stroke or death with carotid end-
arterectomy is 2% to 5%, in high-risk patients it may be as high
as 18%. The increased morbidity and mortality associated with
carotid endarterectomy in high-risk patients formed the ration-
ale for the application of carotid artery stenting. If carotid artery
stenting is considered, diagnostic angiography of both carotid
arteries is performed with attention to details of the common
carotid, external carotid, and internal carotid arteries; this is fol-
lowed by evaluation of the intracranial vessels. In some centers,
selective vertebral angiography is performed at the same time.
During angiography, careful attention must be paid to the ease
of access, the tortuosity both in the common carotid artery and
in the internal carotid artery, the presence or absence of calcifi-
cation in the lesion, and the presence or absence of a satisfactory
landing zone for embolic protection devices.
Procedural goals include delivery and subsequent retrieval
of embolization protection devices, successful prestenting and
458 V Aorta and Peripheral Vascular Disease
poststenting balloon dilation, and successful stenting, all with-
out the development of neurologic complications. Catheter
techniques and the equipment widely used for coronary and
peripheral interventions have been applied for the treatment of
carotid artery disease.
Initially developed for patients undergoing carotid endarter-
ectomy, who are at high risk of mortality, infarction, or stroke,
carotid stenting has now been tested in both asymptomatic and
symptomatic patients who not only are at high risk for these
events but also would be at low risk with carotid endarterectomy.
Carotid stenting has now been studied in single and multicenter
registries and in randomized trials of stenting compared with
carotid endarterectomy. These studies culminated in approval
of the technique by the US Food and Drug Administration in
August 2004 and then approval by the Centers for Medicare and
Medicaid Services in March 2005. Patient selection criteria for
carotid artery stenting continue to evolve. Currently, Medicare
reimburses qualified institutions and physicians for stenting in
symptomatic high-risk surgical patients with stenosis of 70% or
more or in patients enrolled in clinical trials sponsored by the US
Food and Drug Administration.
Access to the common carotid artery is obtained with various
long sheaths. Various embolic protection devices are available:
both proximal and distal protection devices have been tested.
The distal protection devices are easier to use and, accordingly,
are used more widely. Data suggest that if an embolic protection
device cannot be used for any reason, such as lack of a land-
ing zone, severe vessel tortuosity, or distal disease, then carotid
artery stenting should not be performed. The stent ideally cov-
ers the origin of the external carotid artery. Compromise of this
artery is almost always of no clinical consequence. Intravascular
ultrasonography is performed with increasing frequency and is
used to document lesion characteristics and to optimize stent
deployment. At the end of the procedure, the embolic protec-
tion device is retrieved, with care taken not to disrupt the stent.
Minimizing trauma to the vessel wall and minimizing the dura-
tion of the procedure are important to minimize complications.
Repeat angiography of the stent and the intracranial vessels is
performed before termination of the procedure to document that
there has been no arterial complication.
Adjunctive therapy is similar to that used for coronary inter-
ventions. In the past, anticoagulation was administered in the
form of heparin. An activated clotting time of approximately
300 seconds was the target. More recently, bivalirudin has been
widely used. Pretreatment with aspirin and clopidogrel is rou-
tine. If clopidogrel is not given before the procedure, 600 mg
should be administered. A IIb/IIIa agent is used very infre-
quently. After the procedure, aspirin and clopidogrel are given
orally; clopidogrel is given for 1 to 2 months in a dosage of 75 mg
a day, and aspirin therapy is continued indefinitely. It is impor-
tant to avoid wide swings in blood pressure during the procedure.
Typically, antihypertensive medications are withheld the day of
the procedure to avoid significant hypotension. Great care must
be taken to avoid hypertension, which can cause hyperperfusion
syndrome.
Clinical Trials
The choice between surgery and stenting has been controversial,
even with multiple meta-analyses and longer-term follow-up. The
issues have related to operator experience (with variable numbers
of procedures required for operator qualification), specific lesion
and patient selection criteria, use of specific distal protection
devices, and the techniques used to evaluate outcome. Using the
hard and major end points of death and myocardial infarction,
the trials have typically not shown a difference between coronary
artery stenting and carotid endarterectomy. Some of the trials,
but not all, have identified an early imbalance in periprocedural
stroke within 30 days, with this event occurring more frequently
with carotid artery stenting. This complication has led to an
emphasis on optimizing the initial safety of the procedure with
optimal equipment, experienced operators, and careful patient
selection. Patient accrual continues in both randomized clinical
trials and registries.
Multiple registry data sets of carotid stenting indicate a
success rate of 96% to 99%. The SAPPHIRE trial is the only
published randomized trial that compares modern carotid end-
arterectomy with carotid artery stenting. This trial included
symptomatic carotid stenosis of more than 50% or asympto-
matic carotid stenosis of more than 80%. Patients had to have at
least one high-risk carotid endarterectomy criterion. The 30-day
end point of stroke, myocardial infarction, and death was 4.8%
with carotid artery stenting and 9.8% with carotid endarterec-
tomy (P = .09). The primary end point of this trial was complex
and included a composite of stroke, myocardial infarction, and
death within 30 days plus death from neurologic causes during
the first year. This occurred in 12.2% of the carotid artery stent
group and 20.1% of the carotid endarterectomy group. A meta-
analysis of clinical trials of 1,269 patients managed with either
endovascular therapy or carotid endartectomy documented no
difference in the rate of stroke or mortality at 30 days and no dif-
ference between the 2 treatment strategies in preventing stroke
or death at 1 year.
The most important trial is CREST, one of the most eagerly
awaited trials addressing cardiovascular disease. This multi-
center trial enrolled 2,502 patients at 108 centers in the United
States and 9 in Canada. Centers were required to have a multi-
disciplinary team consisting of a neurologist, an interventional
physician, a surgeon, and a research coordinator. Specific ana-
tomical sites and selection criteria had been documented, and
all centers adhered to them. The interventionalists were certified
on the basis of their carotid stent results, participation in hands-
on training, and participation in a lead-in phase of training; the
surgeons also had to document their experience. These 2 crite-
ria satisfied some of the concerns and issues relating to operator
experience in prior trials.
This largest trial of 2,502 patients included both symptomatic
(53%) and asymptomatic (47%) patients with carotid artery ste-
nosis. The eligibility criteria used were stenosis of 50% or more
at the time of angiography, 70% or more at the time of duplex
ultrasonography, or 70% or more on computed tomographic or
magnetic resonance angiography. Criteria were expanded during
the course of the study as asymptomatic patients were included.
Patient exclusion criteria included prior stroke that was severe
enough to confound the assessment of study end points, another
potential cause for stroke such as atrial fibrillation, or unstable
angina. Of note, the patients could be randomized on the basis of
ultrasonographic results.
This randomization scheme has important implications; some
patients who were randomly assigned to carotid stenting at the
time of intervention were found to have anatomical findings that
made them unsuitable for carotid stenting, such as lack of a distal
landing zone for the embolic protection device. However, because
intention-to-treat survival analysis was used, these patients con-
tinued to be enrolled in the carotid stent limb of the trial. Patients
were followed for 2.5 years. Patients were randomized to either
 45 Cerebrovascular Disease and Carotid Artery Stenting 459

conventional endarterectomy or carotid stenting with a relatively

early-generation stent and a distal embolic protection filter.
The primary end point was a composite of any stroke, myo-
cardial infarction, or death during the periprocedural period or
an ipsilateral stroke within 4 years after randomization. There
was no significant difference in the primary end point between
the carotid artery stenting group and the carotid endarterectomy
group (7.2% vs 6.8%, respectively; hazard ratio, 1.11; 95% con-
fidence interval, 0.81–1.51; P = .51). During the periprocedural
period, the incidence of the primary end point was also simi-
lar. There were, however, differences in individual end points.
Although there was no difference in death (0.7% in the carotid
artery stenting group and 0.3% in the carotid endartectomy
group), there was a significant imbalance in the other end points.
Stroke, which was typically minor, occurred more frequently in
the carotid artery stenting group (4.1%) than in the carotid end-
arterectomy group (2.3%) (P = .01), whereas myocardial infarc-
tion occurred more frequently with carotid endartectomy (2.3%)
than carotid artery stenting (1.1%) (P = .03). There was a marked
imbalance in the frequency of cranial nerve palsy: 4.7% in the
surgical group and 0.3% in the carotid artery stenting group
(hazard ratio, 0.07; 95% confidence interval, 0.02–0.18). After
the periprocedural period, the rate of ipsilateral stroke was low in
each group: 2.1% with stenting and 2.4% with surgery.
The 4-year rate of stroke or death in the stenting group was
6.4% with stenting and 4.7% with surgery. In patients who were
symptomatic, the respective rates were 8.0% and 6.4% (haz-
ard ratio, 1.37; 95% confidence interval, 0.90–2.09; P = .14);
in asymptomatic patients, those respective rates were 4.5% and
2.7% (hazard ratio, 1.86; 95% confidence interval, 0.95–3.66;
P = .07).
The CREST results document that, in experienced hands,
carotid artery stenting and carotid endarterectomy are associated
with similar rates of primary composite outcome. Procedural
stroke, myocardial infarction, or death and subsequent ipsilateral
stroke among patients with either symptomatic or asymptomatic
carotid stenosis were not significantly different between patients
who had stenting and those who had endarterectomy. There was
a difference in the individual end points in that more patients in
the stenting group had a stroke (typically minor), whereas more
patients in the carotid endarterectomy group had myocardial
infarction and cranial nerve palsy.
Accordingly, the authors concluded that carotid revascular-
ization performed by highly qualified surgeons and interven-
tionalists is effective and safe and that “the low absolute risk of
recurrent stroke suggests that both carotid artery stenting and
carotid endarterectomy are clinically durable . . . .”
Cerebral Embolism
In patients with cerebral (or systemic) embolic events, the
identification of atherosclerotic plaque in the thoracic aorta by
transesophageal echocardiography has important clinical impli-
cations (Figure 45.2). Atherosclerotic plaque thickness of more
than 4 mm or mobile thrombus (any size) or both predict a high
incidence of recurrent embolic events. Because many patients
Figure 45.2. Transesophageal Echocardiogram for a Patient With
Recurrent Left Hemispheric Transient Ischemic Attacks. Transverse
view of the distal transverse aortic arch shows large mobile thrombus
(short arrow) adjacent to origin of the left common carotid artery (LCA,
long arrow).
with cerebral ischemic events and protruding and mobile athero-
mas of the thoracic aorta have coexistent carotid artery disease
(>70% stenosis), precise identification of the source of embo-
lism is difficult. Treatment in symptomatic patients with no
other identifiable source of embolism is surgical resection of the
involved aorta if the patient’s general medical condition permits.
Oral anticoagulation treatment for 3 months on the presumption
that the friable components will have organized, followed by
antiplatelet therapy, is an alternative. Statin pharmacotherapy is
also warranted.
• Many patients with cerebral ischemic events and protruding and
mobile atheromas of the thoracic aorta have coexistent carotid
artery disease.
Spontaneous Dissection of Cephalic Arteries
Spontaneous dissection of the cervical cephalic arteries is
uncommon but important for 2 reasons: 1) the clinical presenta-
tion is characteristic—either hemicrania with oculosympathetic
paresis (Horner syndrome) or hemicrania with delayed focal
cerebral ischemic symptoms and 2) the prognosis for recovery is
good and recurrences are rare.
Abbreviation
TIA transient ischemic attack
Names of Clinical Trials
ACAS Asymptomatic Carotid Atherosclerosis Study
CREST Carotid Revascularization Endarterectomy Versus
Stenting Trial
SAPPHIRE Stenting and Angioplasty With Protection in Patients at
High Risk for Endarterectomy

The Aorta
PETER C. SPITTELL, MD
Aortic Atheroembolism
Microemboli or macroemboli from atherosclerotic plaques and
thrombi in the aorta are important causes of cerebral and sys-
temic embolization. Cerebral atheroembolism suggests that the
source of embolic material is in the heart, the ascending aorta
or transverse aortic arch (or both), or the cerebrovascular cir-
culation. Lower extremity atheroembolism is caused most com-
monly by AAA or diffuse atherosclerotic disease. Unilateral
blue toes suggest that the embolic source is distal to the aortic
bifurcation.
Atheroembolism is characterized by livedo reticularis,
blue toes, palpable pulses, hypertension, renal insufficiency,
increased erythrocyte sedimentation rate, and eosinophilia (tran-
sient) (Figure 46.1). Atheroembolism can occur spontaneously
or result from medication (warfarin or thrombolytic therapy) or
angiographic or surgical procedures.
Thoracic aortic atherosclerotic plaque is most accurately
assessed with TEE (Figure 46.2). Plaque with a thickness of more
than 4 mm or a mobile thrombus (of any size) is associated with
an increased risk of embolism. Severe aortic atherosclerosis is
present in approximately 27% of patients with previous embolic
events and is also a strong predictor of coronary artery disease.
Aggressive modification of cardiovascular risk factors is war-
ranted for patients with aortic atherosclerosis. Antiplatelet agents
and a statin should be used in all patients with aortic embolic
events unless there are absolute contraindications. Angiotensin-
converting enzyme inhibitors also reduce the risk of ischemic
cardiovascular events in patients with aortic atherosclerosis.
Warfarin therapy may be beneficial for reducing subsequent
embolic events, but it may also exacerbate embolism in some
Abbreviations and acronyms are expanded at the end of this chapter.
460
46
patients; further randomized trials are required. The treatment
of choice is to identify the source of embolism and, if possible,
to surgically resect it.
• Atheroembolism is characterized by livedo reticularis, blue toes,
palpable pulses, hypertension, renal insufficiency, increased eryth-
rocyte sedimentation rate, and eosinophilia (transient).
• Emboli from atherosclerotic plaques and thrombi in the thoracic
aorta are important causes of stroke and peripheral embolization.
Thoracic Aortic Aneurysm
Thoracic aortic aneurysms are caused most commonly by ath-
erosclerosis, but they also occur as a result of systemic hyperten-
sion, Marfan syndrome, bicuspid aortic valve, giant cell arteritis
(cranial arteritis) (Figures 46.3 and 46.4), Takayasu arteritis
(Figure 46.5), infections (syphilis), familial aortopathy, and
trauma. Most thoracic aortic aneurysms are asymptomatic and
are discovered incidentally on chest radiography. CT, MRI, and
combined TTE and TEE are all accurate noninvasive techniques
for imaging thoracic aortic aneurysms (Figure 46.6). Indications
for surgical resection include the presence of symptoms attrib-
utable to the aneurysm, an aneurysm rapidly enlarging under
observation (particularly if the patient has hypertension), post-
traumatic aneurysm, pseudoaneurysm, and an aneurysm with a
diameter of 6.0 cm or more (5.5–6.0 cm in patients with a low
operative risk). In patients with Marfan syndrome and in patients
with bicuspid aortic valve, surgery is usually indicated when the
ascending aortic diameter is 5.0 cm or more.
Abdominal Aortic Aneurysm
In most persons, an AAA (Figure 46.7) does not cause any
symptoms. A pulsatile abdominal mass is the most common
 46 The Aorta
Figure 46.1. Livedo reticularis over both patellae and multiple
blue toes in a patient with atheroembolism from an abdominal aortic
aneurysm.
physical examination finding. For AAA detection, abdominal
palpation has a sensitivity of 43% overall: for larger aneurysms
(diameter ≥4.0 cm) the sensitivity is 57%; for smaller aneurysms
(diameter <4.0 cm) the sensitivity is 29%. AAA can be diag-
nosed reliably with ultrasonography, MRI, or CT (Figure 46.8).
Angiography is not required unless the renal or peripheral arte-
rial circulation needs to be visualized to plan treatment.
Because physical examination lacks sensitivity for AAA
detection, screening tests are indicated for patients at high risk
of AAA. Early detection of AAA can reduce mortality; further-
more, among men older than 65 years a single screening ultra-
sonographic examination can identify the majority of AAAs. The
US Preventive Services Task Force recommends a 1-time screen-
ing ultrasonographic examination for men aged 65 to 75 who
have ever smoked. For men who are either siblings or offspring
of patients with AAA, screening generally begins at age 60.
The most frequent complication of AAA is rupture, which
is related to aneurysm size: for an AAA diameter of 5.5 to 5.9
cm, the 1-year incidence of rupture is 9.4%; for a diameter of
6.0 to 6.9 cm, 10.2%; and for a diameter of 7.0 cm or more,
32.5%. Rupture of an AAA is characterized by severe abdomi-
nal pain and hypotension in the presence of a tender abdominal
mass. Other findings that may be present in patients who have
AAA are obstructive uropathy (from ureteral compression),
461
Figure 46.2. Transesophageal echocardiography shows advanced
immobile atherosclerosis (arrow) in the descending thoracic aorta.
gastrointestinal hemorrhage (if the aneurysm ruptured into the
intestinal tract), and high-output congestive heart failure (from
an aortocaval fistula).
Medical management of asymptomatic AAA includes aggres-
sive modification of risk factors with an emphasis on control-
ling systemic hypertension and discontinuing use of tobacco.
Treatment of associated coronary artery disease and carotid
artery disease is also warranted. Serial noninvasive imaging is
performed to assess both absolute aneurysm size and growth
rate. The frequency of surveillance is based on aneurysm size at
initial detection (surveillance intervals are 3 years for an AAA
diameter of 3.0–3.4 cm; 1 year for a diameter of 3.5–3.9 cm; and
6 months for a diameter of 4.0–4.9 cm).
Elective surgical or endovascular treatment of an AAA should
be considered for aneurysms with a diameter of 5.5 cm or more.
Open surgical repair is indicated for patients who have a low
perioperative risk or for good-risk patients who have an aneu-
rysm diameter of 5.5 to 6.0 cm. For patients who have a large or
symptomatic AAA and a comorbid condition that makes them
poor surgical candidates, a reasonable alternative is exclusion of
the aneurysm by an endovascular approach if the anatomy is suit-
able. Endovascular AAA repair is a safe and effective treatment
compared with open surgical repair for an infrarenal AAA with
appropriate anatomy. Mortality rates for percutaneous repair are
comparable to those for surgical repair, and percutaneous repair
Figure 46.3. Gross specimen shows isolated intimal tear (arrow) of
the ascending aorta in giant cell aortitis.
462 V Aorta and Peripheral Vascular Disease

Figure 46.4. Histologic specimen shows isolated intimal tear of the ascending aorta in giant cell aortitis.

may have improved short-term and long-term morbidity rates.
All patients treated with endovascular repair need continued life-
long follow-up with tomographic imaging.
• Elective surgical repair is definitely indicated when the aneurysm
diameter is ≥5.5 cm in good-risk patients.
An inflammatory AAA is suggested by the triad of back
pain, weight loss, and increased erythrocyte sedimentation rate.
Obstructive uropathy may occur with ureteral involvement. The
findings on CT are diagnostic (Figure 46.9). The treatment is
surgical resection. Trials of immunosuppressant therapy with
corticosteroids are ongoing.
Surgery is also indicated for AAAs that are symptomatic,
traumatic, or infectious in origin or are rapidly expanding
(>0.5 cm yearly).
• An inflammatory AAA is suggested by the triad of back pain,
weight loss, and increased erythrocyte sedimentation rate.

Figure 46.5. Takayasu arteritis in the descending thoracic aorta.

 46 The Aorta
A
Figure 46.6. Magnetic resonance angiography in a patient with an asymptomatic thoracic aortic aneurysm. Images in the transverse (A) and
longitudinal (B) planes show a large aneurysm (arrows) of the ascending aorta (7.8 cm) and moderate dilatation (4.5 cm) of the descending thoracic
aorta. Moderate aortic regurgitation is also apparent (C).
• Surgery is indicated for AAAs that are ≥5.5 cm in diameter, symp-
tomatic, traumatic, or infectious in origin or are rapidly expanding
(>0.5 cm yearly).
Aortic Dissection
Etiology
The most common predisposing factors for aortic dissection are
advanced age, male gender, hypertension, Marfan syndrome, and
congenital abnormalities of the aortic valve (bicuspid or unicus-
pid valve). When aortic dissection complicates pregnancy, it usu-
ally occurs in the third trimester. Iatrogenic aortic dissection can
also occur as a result of cardiac surgery or invasive angiographic
procedures.
Classification
Aortic dissection involving the ascending aorta is designated as
type I or type II (or proximal or type A), and dissection con-
fined to the descending thoracic aorta is designated as type III
(or distal or type B) (Figures 46.10 and 46.11).
Clinical Features
Acute onset of severe pain (often migratory) in the anterior chest,
back, or abdomen is the most suggestive clinical finding in aortic
463
C
dissection (sensitivity, 90%; specificity, 84%). Additional find-
ings include hypertension (49% of patients), an aortic diastolic
murmur (28% of patients), pulse deficits or blood pressure differ-
ential (31% of patients), and neurologic changes (17% of patients).
Syncope occurs when an aortic dissection ruptures into the per-
icardial space, producing cardiac tamponade. Congestive heart
failure most commonly results from severe aortic regurgitation.
Other cardiac manifestations are acute myocardial infarction
(most commonly inferior infarction due to right coronary artery
ostial dissection) and pericarditis.
Clues to type I aortic dissection include substernal pain, aortic
valve incompetence, decreased pulse or blood pressure in the right
arm, decreased right carotid pulse, pericardial friction rub, syncope,
ischemic electrocardiographic changes, and Marfan syndrome.
Clues to type III aortic dissection include interscapular pain,
hypertension, and left pleural effusion.
• Aortic dissection should always be included in the differential
diagnosis for a patient with a catastrophic presentation, systemic
hypertension, and unexplained physical findings of vascular origin
(especially with chest or back pain and an aortic murmur), and an
appropriate screening test should be performed emergently.
Laboratory Tests
Chest radiographic findings are abnormal in 60% to 90%
of patients. Suggestive findings include widening of the
464 V Aorta and Peripheral Vascular Disease

Figure 46.9. Computed tomographic scan of the abdomen of a

patient with an inflammatory abdominal aortic aneurysm. Note the high
attenuation change surrounding the aorta, representing inflammatory
change in the periaortic retroperitoneal tissue.

findings, most commonly left ventricular hypertrophy (with or

without strain). Additional electrocardiographic abnormali-
ties in acute aortic dissection include ST-segment depression,
ST-segment elevation, T-wave changes, and the changes of acute
pericarditis and complete heart block.

Diagnostic Imaging
Figure 46.7. Aneurysms of the abdominal aorta (long arrow) and Definitive diagnosis of aortic dissection can be made with
iliac artery (short arrow). echocardiography, CT, MRI, or aortography.

mediastinum and supracardiac aortic shadow, deviation of the

trachea to the right, a discrepancy in diameter between the
ascending aorta and descending aorta, and pleural effusion
(Figure 46.12). Normal findings on chest radiography do not
exclude aortic dissection because 12% or more of patients who
have acute aortic dissection have normal radiographic find-
ings. Up to 55% of patients have abnormal electrocardiographic

Type I Type II Type III

Figure 46.8. Computed tomography with intravenous contrast Type A (proximal) Type B (distal)
medium shows a large aneurysm (arrow) of the infrarenal abdominal
aorta. There is a small amount of laminated thrombus within the aneu- Figure 46.10. Classification of aortic dissection. Arrows indicate
rysm and dense peripheral calcification. locations of dissection.
Figure 46.11. Type III aortic dissection in the abdominal aorta.

Figure 46.12. Chest radiographs before (A) and after (B) aortic dissection. Note widening of the superior mediastinum after aortic dissection
(arrow).
466 V Aorta and Peripheral Vascular Disease
Echocardiography
The combination of TTE and TEE can be used to identify an
intimal flap, communication between the true and false lumina,
a dilated aortic root, thrombus formation, widening of the aor-
tic walls, aortic regurgitation, and pericardial effusion or tam-
ponade. TTE detects aortic dissection in the ascending aorta
with a sensitivity of 77% to 80% and a specificity of 93% to
96%; therefore, in patients with suspected acute aortic dissec-
tion, TTE negative for dissection does not exclude the diagnosis.
Multiplane TEE techniques have markedly improved the accu-
racy of TEE, resulting in a sensitivity of 99% and a specificity
of 98% (Figure 46.13). Advantages of TEE include portability,
safety, accuracy, rapid diagnosis, and the possibility of using
it intraoperatively or with patients who are hemodynamically
unstable.
Computed Tomography
Advantages of CT are that it is useful for accurately detect-
ing the intimal flap; identifying 2 lumina; demonstrating dis-
placed calcification, pericardial effusion, pleural effusion, and
abdominal aorta involvement; and providing accurate aortic
diameters (Figure 46.14). For suspected acute aortic dissec-
tion, CT has an overall accuracy of 96%. Cardiac gating is
essential to avoid cardiac motion artifacts. The disadvantages
include nonportability (limiting its use for patients who are
hemodynamically unstable) and the need for intravenous con-
trast agents.
Magnetic Resonance Imaging
For the diagnosis of aortic dissection, MRI is as accurate as
CT. In addition, with its inherent multiplanar imaging capabil-
ity, MRI can be used with or without contrast enhancement. All
the following can be demonstrated: the intimal flap, entry and
exit sites, thrombus formation, aortic regurgitation, pericardial
Figure 46.13. Transesophageal echocardiographic view of the
ascending aorta in the longitudinal plane shows an intimal flap (arrow)
originating in the right coronary sinus.
Figure 46.14. Computed tomography with intravenous contrast
agent shows dilatation of the descending thoracic aorta and an intimal
flap (arrow). Note the relatively equal opacification of the true and false
lumina.
effusion, pleural effusion, and abdominal aorta and branch ves-
sel involvement (Figure 46.15). Disadvantages of MRI include its
cost and nonportability.
Aortography
Aortography can be used to accurately diagnose aortic dissection
by showing the intimal flap, opacification of the false lumen, and
deformity of the true lumen. It can also show associated aortic
regurgitation and coronary artery anatomy. The disadvantages
include invasive risks, exposure to intravenous contrast agents,
delay in diagnosis, and nonportability.
When a patient has a suspected acute aortic dissection, the
choice of test (TTE, TEE, CT, MRI, or aortography) depends on
which is most readily available and on the hemodynamic stabil-
ity of the patient. Currently at Mayo Clinic, the test of choice
for suspected acute aortic dissection in hemodynamically stable
patients is CT or combined TTE and TEE. In hemodynamically
unstable patients, combined TTE and TEE is the initial test of
choice. The initial management of suspected acute aortic dissec-
tion is shown in Figure 46.16.
The most common cause of death among persons with aor-
tic dissection is rupture into the pericardial space, with cardiac
tamponade. Echocardiographically guided pericardiocentesis
in acute aortic dissection is associated with an increased risk
of aortic rupture and death. Cardiac tamponade due to aortic
dissection is a surgical emergency, and pericardial fluid should
be removed only in the operating room after cardiopulmonary
bypass has been instituted. Other causes of death include acute
congestive heart failure due to severe aortic regurgitation, rup-
ture through the aortic adventitia, rupture into the left pleural
space, and occlusion of vital arteries. In acute type III aortic dis-
section, a large false lumen and branch vessel involvement are
both predictors of inhospital complications. Factors that prop-
agate dissection include “impulse” pulsatile flow and increased
mean arterial pressure.
• Cardiac tamponade due to aortic dissection is a surgical emer-
gency, and pericardial fluid should be removed only in the operat-
ing room after cardiopulmonary bypass has been instituted.
 46 The Aorta 467

Chest pain

History, physical examination,

CXR, ECG, cardiac enzymes

Aortic dissection suspected

Initiate ? -blocker therapy

Hemodynamically stable

Yes No

TEE, CT, MRI, TTE and TEE (MRI, CT, or angio

or angio if TEE is not available, or rapid
transfer to tertiary facility)

Aortic dissection
Type I or II Type III

Surgery Coronary care unit

Figure 46.16. Initial management of suspected acute aortic dissec-
tion. Angio indicates angiography; CT, computed tomography; CXR,
chest radiography; ECG, electrocardiography; MRI, magnetic resonance
imaging; TEE, transesophageal echocardiography; TTE, transthoracic
echocardiography. (Previously published. See “Credit Lines” section.)

The initial pharmacologic therapy for aortic dissection is

Figure 46.15. Magnetic resonance angiography shows dissection
(arrow) of the mid and distal abdominal aorta in a patient who had a
remote history of sudden deceleration injury.
outlined in Box 46.1. When long-term pharmacologic therapy is
used for type III aortic dissection, indications for surgery include
development of saccular aneurysm, increasing aortic diameter,
or symptoms related to chronic dissection.

Penetrating Aortic Ulcer

Treatment
Pharmacologic therapy for aortic dissection should be instituted
as soon as the diagnosis is suspected (see below). Emergent sur-
gery is indicated for types I and II aortic dissection. Preoperative
coronary angiography is not indicated because it delays surgi-
cal treatment and does not improve survival. Pharmacologic
therapy in the coronary care unit is the preferred initial treat-
ment for type III aortic dissection, with delayed surgical therapy
(2–3 weeks) for selected patients whose general medical condi-
tion permits an operation. Endovascular thoracic aortic repair to
treat type III aortic dissection can be lifesaving in acute emer-
gencies, but in stable type III dissection endovascular repair has
not improved 2-year survival or adverse event rates.
Penetrating aortic ulcer occurs when an atherosclerotic plaque
undergoes ulceration and penetrates the internal elastic lamina.
The result is 1 of 4 possible consequences (Figure 46.17):
1. Formation of an intramural hematoma
2. Formation of a saccular aneurysm
3. Formation of a pseudoaneurysm
4. A transmural rupture
Penetrating aortic ulcer most commonly involves the mid or
distal descending thoracic aorta and less often the ascending or
abdominal aorta (Figure 46.18). The clinical features of pene-
trating aortic ulcer are similar to those of aortic dissection and
include acute onset of pain in the anterior or posterior chest (or
both) and hypertension. However, other findings in classic aortic
468 V Aorta and Peripheral Vascular Disease

dissection, such as pulse deficits, neurologic signs, and acute

Box 46.1. Initial Pharmacologic Therapy for Aortic cardiac disease (aortic regurgitation, myocardial infarction, or
Dissection pericardial effusion), are absent in penetrating aortic ulcer. The
Hypertensive patients diagnosis of penetrating aortic ulcer can be established with CT,
MRI, TEE, or aortography (Figure 46.19).
Sodium nitroprusside IV, 2.5–5.0 mcg/kg per minute
The treatment of penetrating aortic ulcer is usually nonopera-
with tive if only an intramural hematoma is present. With control of
Propranolol IV, 1 mg every 4–6 h hypertension, the intramural hematoma tends to resolve sponta-
(The goal is to have systolic blood pressure <110 mm
neously over time. Surgical therapy is indicated for patients who
Hg—a lower pressure is acceptable if urine output
have ascending aortic involvement, a saccular aneurysm, or a
is maintained to at least 25–30 mL/h—until oral
pseudoaneurysm or for patients who have an intramural hema-
therapy is started)
toma and persistent symptoms, increasing aortic diameter, or
poorly controlled hypertension.
or
IV esmolol, metoprolol, or atenolol (in place of • Penetrating aortic ulcer occurs when an atherosclerotic plaque
propranolol) undergoes ulceration and penetrates the internal elastic lamina.
• The treatment of penetrating aortic ulcer is similar to that of aortic
or
dissection (ascending aortic involvement is treated surgically, and
IV labetalol (in place of sodium nitroprusside and initial medical management is preferred for descending thoracic
a β-blocker) aortic involvement).
Normotensive patients
Propranolol, 1 mg IV every 4–6 h Aortic Intramural Hematoma
or
Characterized by the absence of an intimal tear, aortic intramu-
Propranolol, 20–40 mg orally every 6 h ral hematoma is associated with cystic medial necrosis. Aortic
(Metoprolol, atenolol, or labetalol may be used in intramural hematoma is increasingly recognized, largely owing
place of propranolol) to advances in noninvasive imaging techniques. The exact cause
is not well defined. Aortic intramural hematoma is diagnosed in
Abbreviation: IV, intravenously. the same manner as aortic dissection, the classification schemes
are identical, and traditionally the management has been

Figure 46.17. Possible consequences of a penetrating aortic ulcer.

 46 The Aorta 469

Figure 46.18. Grade 4 ulcerocalcific disease of the abdominal aorta.

C

similar: operation for type A intramural hematoma and medical

treatment for type B.
IM
Incomplete Aortic Rupture
Incomplete rupture of the thoracic aorta (in the region of the aor- Ao
tic isthmus) results from a sudden deceleration injury. It is seen
most often in motor vehicle accidents and should be suspected
when there is evidence of trauma to the chest wall, decreased
or absent leg pulses, left-sided hemothorax, or widening of the
superior mediastinum on chest radiography. At initial presenta-
tion, the patients usually are hypertensive and 40% to 50% are Figure 46.19. A, Computed tomography with intravenous contrast
hemodynamically unstable. The diagnosis can be confirmed agent shows intramural hematoma (arrow) of the transverse aortic arch
with TEE, CT, MRI, or angiography. Treatment is emergent sur- caused by a penetrating aortic ulcer. B, Magnetic resonance imaging
gical repair in patients who are suitable surgical candidates. No shows a penetrating aortic ulcer (arrow) of the transverse aortic arch. C,
clinical or imaging criteria accurately predict future complete Transesophageal echocardiography of the transverse aortic arch (trans-
rupture, so even if a patient presents with a chronic incomplete verse plane) shows an intramural hematoma (IM). Ao indicates aorta.
rupture, surgery is indicated. Furthermore, most of the patients
are young, and the risk of elective surgical repair is low, with an
otherwise good prognosis for long-term survival if aortic repair Abbreviations
is successful. AAA abdominal aortic aneurysm
CT computed tomography
• Incomplete rupture of the thoracic aorta (in the region of the aor- MRI magnetic resonance imaging
tic isthmus) results from a sudden deceleration injury, frequently a TEE transesophageal echocardiography
motor vehicle accident. TTE transthoracic echocardiography

Renal Artery Disease
VERGHESE MATHEW, MD
Obstructive Disease of the Renal Arteries
Obstructive disease of the renal arteries can be categorized
as either atherosclerosis (the majority of RAS cases) or FMD.
Atherosclerotic RAS may result in hypertension or ischemic
nephropathy (or both), and RAS is the most commonly identi-
fied cause of secondary hypertension. The true prevalence of ath-
erosclerotic renal disease in the general population is unknown,
although the prevalence in selected populations of patients
with identified atherosclerosis in other vascular beds has been
reported to be 10% to 30%. Among patients with identified RAS,
there are patients who have physiologically significant stenosis
and patients who do not. Despite the existence of noninvasive
and invasive diagnostic modalities to identify RAS, clarification
is still required to understand the true natural history of RAS and
outcomes after renal revascularization and to identify patients
who have physiologically significant RAS most likely to benefit
from revascularization. The clinical outcome data after renal
revascularization may be heterogeneous because physiologically
significant RAS has not been accurately defined (Figure 47.1).
Renal FMD occurs in younger patients, in females more than
males, and often in secondary or distal branches. Renal artery
revascularization for patients with renal FMD is often curative
for hypertension control. Historically, surgical revascularization,
and more recently percutaneous transluminal angioplasty, has
been used; stenting is not thought to be incrementally benefi-
cial in percutaneous treatment of FMD and may be detrimen-
tal (Figure 47.2). FMD may also coexist in other vascular beds,
including the coronary circulation. FMD occurring in isolation
usually is not associated with renal failure.
Abbreviations and acronyms are expanded at the end of this chapter.
470
47
• RAS is the most commonly identified cause of secondary
hypertension.
• FMD occurring in isolation usually is not associated with renal
failure.
Clinical Clues to Detecting RAS
Given the possible limitations for accurate diagnosis of physi-
ologically significant RAS, several clinical situations have been
proposed in which renovascular disease should be considered in
the clinical differential diagnosis:
1. Known atherosclerosis in other vascular beds (associated with the
presence of atherosclerotic RAS)
2. Onset of hypertension before age 30 (consider FMD) or after age 55
3. Worsening of previously controlled hypertension
4. Malignant or resistant hypertension
5. Abdominal or flank bruit (not highly predictive but is supportive)
6. Discrepancy in renal size (small kidney)—more likely a later stage
of renovascular disease
7. Azotemia otherwise unexplained or worsened by angiotensin-
converting enzyme inhibitors or angiotensin receptor blockers
8. Recurrent congestive heart failure or flash pulmonary edema in a
hypertensive patient, particularly one with preserved systolic left
ventricular function
9. Angina without significant coronary artery disease in a hypertensive
patient
Diagnostic Testing for RAS
Several diagnostic tests are available for evaluating RAS and
for assessing anatomical factors (parenchyma, arterial imag-
ing, and flow information) or physiologic effects of presumed
RAS. Intravenous pyelography, one of the earliest tests used for
the diagnosis of RAS, was associated with poor sensitivity and
a high false-positive rate; it is no longer considered a useful or
 47 Renal Artery Disease
Figure 47.1. Renal Artery Stenosis. A 65-year-old man with poorly
controlled hypertension (while taking 4 medications) had physiologi-
cally significant left renal artery stenosis. The stenosis was identified at
coronary angiography when he presented with unstable angina; screen-
ing aortography was requested by the referring cardiologist for refrac-
tory hypertension. Note the atheromatous disease of the aorta adjacent
to the left renal artery origin.
appropriate test for evaluation of RAS. Similarly, the use of plasma
renin activity has low predictive value; with administration of an
angiotensin-converting enzyme inhibitor (conventionally capto-
pril is used), specificity increases, although the value of the test
diminishes in bilateral RAS and the need to discontinue the use
Figure 47.2. Renal Fibromuscular Dysplasia. Aortography (left) and selective right renal arteriography (right) show the characteristic beaded
appearance (like a string of pearls) of renal fibromuscular dysplasia. The lesion is generally not located in the proximal portion of the renal artery.
471
of most antihypertensive agents is a practical limitation of the test.
Captopril scintigraphy is useful for diagnosis of unilateral RAS
in patients with preserved renal function, although it has limited
utility in patients with bilateral RAS and renal dysfunction; anti-
hypertensive agents also need to be withheld for this test.
In clinical practice, the more frequently used noninvasive
diagnostic modalities include ultrasonography, CTA, and MRA;
arteriography is considered the criterion standard for the diag-
nostic evaluation of RAS, although its invasive nature practically
precludes its widespread application for screening.
Duplex Ultrasonography
Duplex ultrasonography is relatively inexpensive and has a high
sensitivity and specificity. Limitations may include operator-
dependent variability and technical difficulties with large patients
or with overlying bowel gas. In addition, accessory renal arter-
ies, in which stenosis may still lead to renovascular hypertension,
may be difficult to identify. For patients in whom FMD is sus-
pected clinically, a negative ultrasonographic examination should
be followed with another imaging modality. In general, for the
evaluation of suspected atherosclerotic RAS, ultrasonography is a
reasonable screening test. A ratio of renal artery velocity to aortic
velocity of more than 3.5 supports the diagnosis of significant
RAS. The calculated resistive index has often been used to deter-
mine the likelihood of benefit with renal revascularization: a resis-
tive index of more than 0.8 suggests renal parenchymal disease,
which would not be expected to improve with revascularization.
Additionally, the medical literature suggests that kidneys 8 cm
or less in length are unlikely to benefit from revascularization,
although these guidelines should not be taken as absolute to deter-
mine whether revascularization should be offered in practice.
Magnetic Resonance Angiography
MRA has evolved into a useful imaging modality for the evalu-
ation of RAS. Advantages include images that have a visual
472 V Aorta and Peripheral Vascular Disease
appearance similar to those of aortography with 3-dimensional
capabilities, although image processing may introduce image
inaccuracies in inexperienced hands. Additionally, an iodinated
contrast agent is not used, which is a clear advantage for patients
in whom renal function is already compromised. However, in
patients with significant impairment of renal function, gado-
linium has been associated with nephrogenic systemic fibrosis,
a rare but untreatable fibrosing condition that primarily affects
the skin but may affect other organs. MRA interpretation may
be hampered in stented arterial segments because of signal
dropout. The usual limitations and contraindications for mag-
netic resonance examinations apply (eg, claustrophobic patients,
indwelling metal hardware such as cardiac pacemakers). MRA
can also be used to ascertain regional renal perfusion and assess
blood oxygenation, although these variables are generally not
part of routine clinical examinations at most centers but have
the potential to add to the understanding of renal physiology
(Figure 47.3).
Computed Tomographic Angiography
CTA is an excellent imaging modality for RAS. Although
CTA is not limited by some of the patient factors that preclude
Figure 47.3. Magnetic resonance angiogram (MRA) from a 75-year-
old woman with a prior history of diabetes mellitus and coronary revas-
cularization of the left anterior descending artery 2 years previously
in the Bypass Angioplasty Revascularization Investigation 2 Diabetes
(BARI 2D) trial. She presented with poorly controlled hypertension
(while using 3 agents) and insidious worsening of her serum creatinine
level, from 1.4 to 1.7 mg/dL, over 2 years. The MRA shows a single left
renal artery without significant stenosis. The main right renal artery has
a high-grade proximal stenosis (arrow); an accessory right renal artery
to the superior pole may have a proximal stenosis (arrowhead), although
this was difficult to characterize with certainty. On angiography, the
accessory renal artery was found to be free of obstructive disease.
MRA examination (ie, cardiac pacemakers), CTA does require
radiation. In addition, CTA requires iodinated contrast material,
which makes it a less attractive option for patients with renal dys-
function. Significant calcification may hamper the assessment of
stenosis severity. Even though the presence of stents may also
hamper the assessment of severity, CTA is preferred to MRA
when stents are present. Similar to MRA, CTA can be used to
measure renal perfusion and flows, although this is not performed
in standard clinical practice. Ultimately, however, modalities
such as MRA and CTA may be useful to simultaneously assess
for the anatomical presence of RAS and for the physiologic effect
of RAS on the kidney (Figure 47.4).
Invasive Arteriography
Invasive arteriography has long been considered the standard for
renal artery imaging. Potential limitations include its invasive
nature, making it less attractive as a screening test, and the use of
an iodinated contrast agent (although carbon dioxide angiography
and gadolinium angiography can also be performed with diagnos-
tic-image quality). The limitations of coronary angiography in
assessing coronary artery lesion severity are well known; angiog-
raphy can underestimate or overestimate lesion severity. With the
use of 0.35-mm pressure/flow wires, it is clear that angiographi-
cally mild or moderate coronary artery lesions may be physi-
ologically significant and lesions that appear angiographically
significant may not be flow limiting. This issue of indeterminate
angiography, or misdiagnosing the severity of stenosis from the
angiographic assessment, has not been appreciated or evaluated
to the same degree in the renal arterial bed; angiographically
Figure 47.4. Computed tomographic angiogram (CTA) from a
66-year-old man with diabetes mellitus, multiple prior percutaneous
coronary interventions, and carotid endarterectomy, with a recent right
renal artery stent. CTA shows a dominant superior left renal artery
stenosis (arrowhead); calcification confounds the assessment of steno-
sis severity, but pressure gradient measurements in this vessel with a
0.35-mm pressure wire at the time of right renal stenting demonstrated
no significant pressure gradient. A left lower pole accessory renal
artery is also identified. Note the stent in the proximal right renal artery
(arrow), which could confound assessment of in-stent stenosis, although
the cross-sectional images suggested absence of significant restenosis.
 47 Renal Artery Disease 473

significant RAS has been presumed to be physiologically sig- Revascularization for RAS
nificant, and angiographically insignificant findings have been
When FMD is the cause of renovascular hypertension, renal
thought to exclude important disease. Translesional pressure gra-
percutaneous transluminal angioplasty is an accepted, durable
dient determinations using 4F catheters have been performed in
treatment strategy. The role of renal artery intervention for ath-
the renal arterial vasculature for many years, but the effect of
erosclerotic renal artery disease, however, has been controver-
the catheter diameter itself may overestimate the translesional
sial since results of trials have been mixed but largely negative
gradient. Conventionally, a 10- to 20-mm Hg peak-to-peak gra-
for clinical benefit. The ASTRAL trial, published in 2009, is
dient across an RAS has been thought to be important, although
the largest study of its kind to date. In the ASTRAL trial, 806
these values are relatively arbitrary and have not correlated with
patients with atherosclerotic RAS were randomly assigned to
a measure of renal function or physiology. Recently, 0.35-mm
a strategy of either revascularization with medical therapy or
pressure/flow wires from coronary artery practice have been
medical therapy alone. The hypothesis was that renal interven-
used to assess RAS. A role for renal flow and pressure reserve
tion may attenuate the decline in renal function over time, but
assessments will probably emerge, although the complexities of
the trial results were negative—no benefit was demonstrated
renal autoregulation make it unlikely that simply measuring rest-
with revascularization. Patients with flash pulmonary edema,
ing or even inducible gradients across renal lesions will reveal
“critical” RAS, and “very severe” or uncontrolled hypertension
completely the physiologic significance of RAS.
were excluded; it should be noted that intervention in these sub-
sets is not supported by a high level of evidence but is often
• Noninvasive diagnostic modalities include ultrasonography, CTA, performed in clinical practice. Exclusion of such patients is a
and MRA. significant limitation to widespread generalizability of the trial
• Invasive arteriography is considered the standard diagnostic test results. The primary outcome measure was the ability of renal
for RAS. intervention to attenuate a decrease in renal function (the mean

A B

Figure 47.5. An 82-year-old woman with extensive vascular disease, severe chronic obstructive pulmonary disease, and unevaluated anemia
presented with a subacute progression of chronic renal failure. Serum creatinine was 1.7 to 2.0 mg/dL 1 year earlier, 4.8 mg/dL at presentation, and
5.6 mg/dL 48 hours after presentation. Ultrasonography of the kidneys showed mildly atrophic parenchyma; the right kidney length measured 8 cm
and the left 9.2 cm, with single renal arteries bilaterally and significantly elevated velocities in both. The resistive indexes were 0.7 to 0.8. The patient
began hemodialysis and had a cardiovascular consultation to evaluate her for renal revascularization. Although the kidneys appeared small with
borderline resistive indexes of both renal arteries, stenting was offered because the history suggested a recent decline in renal function and because
the patient would be committed to dialysis in any case. Magnetic resonance angiography was performed to aid in procedural planning; the patient
was unable to tolerate a magnetic resonance scan because of restlessness and claustrophobia. Thus, the patient was brought to the angiographic
suite; access was difficult through the left common femoral artery, and the right femoral artery was occluded. The aorta was extremely tortuous and
calcified; a 6F 45-cm sheath was used, and an internal mammary catheter could not be manipulated sufficiently close to the right renal artery safely
(without a risk of scraping the severely atherosclerotic aorta). Therefore, a Sos Omni catheter was used with a 0.875-mm Bentson wire to cannulate
the right renal artery. A, Gadolinium injection showed a critical ostial stenosis (arrow). B, The stenosis was crossed with a 0.35-mm Cordis Supersoft
Reflex wire, and the Sos Omni catheter was exchanged over this wire for a 6F internal mammary guide. Predilation was performed with a 3.0-mm
Powersail (coronary) balloon, and a 3.5×18-mm Guidant Vision (coronary) stent was placed with an excellent result (arrow). Similarly, the left renal
artery was identified as having a high-grade ostial stenosis and was stented with a 3.0×13-mm Guidant Vision stent. Gadolinium (60 mL) was used
with no iodinated contrast medium. The serum creatinine concentration decreased to 2.5 mg/dL within 72 hours, and hemodialysis was ceased. Four
months later, the creatinine was 1.4 mg/dL. Although visualization was somewhat compromised because of the use of gadolinium, the images were
diagnostic. In light of the issue of nephrogenic systemic fibrosis related to gadolinium use in patients with significant renal dysfunction, the use of
gadolinium is now discouraged in these patients.
474 V Aorta and Peripheral Vascular Disease

glomerular filtration rate was approximately 40 mL/min per congestive heart failure, or worsening renal function (with bilat-
1.73 m 2, which indicates chronic kidney disease stage 3). eral or perhaps unilateral disease), renal artery revascularization
The primary goal of this trial was to investigate renal func- may be considered with the caveats and uncertainties outlined
tion changes rather than hypertension control in response to above (Figure 47.5).
intervention.
The STAR trial, with a much smaller sample size (140 patients • Renal percutaneous transluminal angioplasty is an accepted,
durable treatment strategy for renovascular hypertension caused
with impaired renal function), arrived at a similar conclusion as
by FMD.
the ASTRAL trial.
In the National Institutes of Health CORAL trial (in prog- • The role of renal artery intervention for atherosclerotic renal artery
disease has been controversial because results of trials have been
ress), patients with RAS and significant hypertension who use
mixed.
2 or more antihypertensive agents are randomly assigned to
receive medical therapy alone or medical therapy with renal
stenting. This trial will shed light on the role of percutaneous Abbreviations
renal revascularization for managing hypertension and reducing
CTA computed tomographic angiography
adverse cardiovascular events.
FMD fibromuscular dysplasia
In clinical practice, heterogeneous treatment outcomes are MRA magnetic resonance angiography
observed with intervention: clinically, some patients improve, RAS renal artery stenosis
some are unchanged (either because of or in spite of the pro-
cedure), and some worsen (either because of or in spite of the
procedure). This heterogeneity may reflect a problem with the Names of Clinical Trials
procedure, but it also (at least in part) reflects an inability to ASTRAL Angioplasty and Stenting for Renal Artery Lesions
consistently and reliably establish causality related to atheroscle- CORAL Cardiovascular Outcomes in Renal Atherosclerotic
rotic renal artery disease when patients also have hypertension Lesions
or renal failure (or both). For patients who have atheroscle- STAR Stent Placement in Patients With Atherosclerotic Renal
rotic RAS and refractory hypertension, recurrent or refractory Artery Stenosis and Impaired Renal Function

Pathophysiology of Arterial Thrombosisa
ROBERT D. MCBANE, MD, and WALDEMAR E. WYSOKINSKI, MD, PHD
Atherosclerosis is the most common cause of major disability
and death in the United States. The most devastating complica-
tion of this disease occurs when a platelet-rich thrombus abruptly
occludes arterial blood flow, resulting in acute MI, stroke, or
sudden cardiac death. This chapter reviews the basic pathophysi-
ology of arterial thrombosis.
The pathophysiology of arterial thrombosis involves platelet-
rich thrombus formation over a ruptured atherosclerotic plaque.
This process can be partitioned into platelet adhesion, coagula-
tion factor activation, and thrombus propagation through platelet
accretion.
Endothelium
The endothelial lining presents an inert interface between the
vessel wall and circulating blood. More than only an inert surface,
endothelium synthesizes nitric oxide (also known as endothelial-
derived relaxant factor) and PGI2, which are potent vasodilators
and inhibitors of platelet adhesion and aggregation. The endothe-
lial glycocalyx is a rich source of proteoglycans such as heparan
sulfate, which in combination with antithrombin provides a local
source of anticoagulant. Endothelium also expresses thrombo-
modulin, a specific receptor that binds thrombin, changing its
activity from prothrombotic to anticoagulant through the protein
C activation pathway. The endothelium is therefore more than a
passive barrier—it actively participates in maintaining luminal
patency by inhibiting local thrombosis.
a
Portions previously published in McBane RD 2nd. Genetically
determined procoagulant states and heparin use. Semin Cardiothorac
Vasc Anesth. 2003 Dec;7(4):427–42.
Abbreviations and acronyms are expanded at the end of this chapter.
48
Platelet Adhesion
After endothelial injury, exposed collagen binds vWF, which
serves as an initial tether to which the platelet receptor GpIb-
IX-V binds (Figure 48.1). The platelet, tethered by vWF, trans-
locates along the injured surface in the direction of blood flow.
A second platelet receptor, GpIa-IIa (integrin α2β1), then engages
collagen, thus arresting further translocation. A third receptor,
GpVI, also engages collagen and serves as a signal-transducing
receptor that stimulates platelet activation, including GpIIb-
IIIa (integrin α2bβ3) receptor activation and dense granule ADP
secretion. ADP stimulates adjacent platelets and promotes fur-
ther platelet recruitment to the growing thrombus. This sequence
of platelet adhesion and receptor transduction–mediated platelet
activation is sufficient for the initial stages of platelet plug forma-
tion necessary for hemostasis.
Coagulation Factor Activation
Pathologic arterial thrombosis is a thrombin-mediated pro-
cess and thus requires activation of the coagulation cascade.
Coagulation factor activation occurs as a cascade of proenzymes
is cleaved from their inactive forms to their active forms. This
cascade of enzyme activation and amplification requires the
assembly of activation complexes on the phospholipid membrane
of activated platelets (Figure 48.2). These activation complexes
include activated platelet phospholipid membrane, activating
enzyme and proenzyme, an activated cofactor, and calcium. The
coagulation cascade may be initiated through 1 of 2 pathways:
the extrinsic (tissue factor) pathway or the intrinsic (contact acti-
vation) pathway (Figure 48.3). When an atherosclerotic plaque
ruptures, the tissue factor pathway of coagulation cascade ini-
tiation probably predominates. Exposed tissue factor within the
475
476 V Aorta and Peripheral Vascular Disease

RBCs

Platelets
Endothelial cells

GpIa-IIa

GpIb-IX-V
vWF

Figure 48.1. Platelet Adhesion in Response to Arterial Injury. A, Arterial injury. B, Platelet adhesion. Area of vessel wall marked with rectangle
in panel A is depicted as a close-up drawing in panel B. Gp indicates glycoprotein complex; RBCs, red blood cells; vWF, von Willebrand factor.

Factor Va
Activated
Thrombin
platelet Prothrombin
Factor Xa

Platelet
phospholipid
cell surface

Figure 48.2. Coagulation Factor Activation. Prothrombinase complex is depicted at the platelet phospholipid cell surface.
 48 Pathophysiology of Arterial Thrombosis 477

Figure 48.3. Coagulation Cascade. Both intrinsic and extrinsic pathways converge on prothrombinase complex, which activates prothrombin to
thrombin. TF indicates tissue factor.

vessel wall initiates the cascade by binding circulating factor
VIIa, which in turn cleaves factors IX and X to their active forms
(IXa and Xa, respectively). Factor Xa then combines with factor
Va on the platelet phospholipid membrane to form the prothrom-
binase complex, which cleaves prothrombin to thrombin.
by inducing further platelet granule secretion, cleaving fibrino-
gen to fibrin, and activating factor XIII, which cross-links fibrin
strands to form a stable thrombus. Thrombin generation is there-
fore necessary for thrombus propagation sufficient for arterial
occlusion.

Thrombus Propagation Anticoagulant System

Thrombin stimulation leads to further activation of GpIIb-IIIa, Almost as soon as the coagulation system is initiated, an endoge-
which binds fibrinogen, cross-linking adjacent platelets to form nous anticoagulant system begins to turn it off. Thrombin activates
aggregates (Figure 48.4). Thrombin then orchestrates thrombosis an endogenous anticoagulant system whose central components

Fibrinogen

Platelet
GpIIb-IIIa
activation

Platelet GpIIb-IIIa

Figure 48.4. Platelet Aggregation. Gp indicates glycoprotein complex.

478 V Aorta and Peripheral Vascular Disease

C4b-binding protein
Factors iVIIIa

VIIIa Protein S

Va

iVa

APC

Protein C
Thrombin

Thrombomodulin

Protein C receptor

Endothelial cell surface

Figure 48.5. Protein C Anticoagulant System. APC indicates activated protein C; i, inactivated form of factor.

include protein C, protein S, and antithrombin (Figure 48.5). Platelets

Thrombin binds to thrombomodulin on the luminal surface of
Platelets participate in every stage of arterial thrombosis, from
endothelial cells and activates protein C to activated protein C,
the formation of the initial hemostatic plug to the leukocyte
which (along with the cofactor protein S) downregulates the pro-
recruitment necessary for wound healing. Platelets are therefore
coagulant pathway by inactivating the procoagulant protein fac-
not merely passive bystanders in this process. The platelet has 4
tors VIIIa and Va. Ultimately, antithrombin binds irreversibly to
well-recognized storage granules: the α-granule, the dense body,
thrombin, and the complex is cleared by the liver (Figure 48.6).
the lysosomal granule, and the microperoxisome. These storage
granules contain many vasoactive amines, receptors, procoagu-
Fibrinolytic System lant and anticoagulant factors, and growth hormones (Box 48.1).
Platelet glycoprotein receptors mediate both platelet adhesion
Thrombin also stimulates endothelial cells to release tPA, which
to the arterial wall and aggregation (Table 48.1). Three receptors
activates plasminogen to plasmin, the central component of the
have an important role in initiating the steps of platelet adhesion:
fibrinolytic system. Plasminogen activation by tPA is markedly
GpIb-IX-V, GpIa-IIa, and GpVI. A fourth receptor, GpIIb-IIIa,
amplified in the presence of fibrin, specifically the C-terminal
is principally involved in platelet aggregate formation, which is
lysines of fibrin. The binding of both tPA and plasminogen to
essential for the later stages of thrombus growth. Platelet glyco-
the lysine-rich C-terminal of fibrin increases the catalytic effi-
protein receptors are highly polymorphic. These platelet receptor
ciency of plasminogen activation and the clot-specific nature of
polymorphisms alter platelet function and have each been impli-
fibrinolysis. The fibrinolytic pathway removes (ie, lyses) the arte-
cated in arterial thrombosis. Platelet receptor quality, quantity,
rial thrombus as part of the healing process of the arterial wall.
function, and polymorphism status have been studied exten-
Plasmin cleaves fibrin within the thrombus, thus dissolving the
sively, from basic science studies to large epidemiologic studies.
thrombus. Plasmin also inactivates several important components
Whereas the family history contributes importantly to the risk
of the procoagulant pathway, including factor V, factor VIII, and
of MI, interindividual differences in arterial thrombus formation
platelet glycoproteins GpIb-IX-V and GpIIb-IIIa. Consequently,
may in part be explained genetically. The following paragraphs
the overall action of plasmin is to promote thrombus dissolution
briefly review the platelet receptors that are relevant to arterial
and downregulate ongoing thrombus formation.
thrombosis.
Akin to the endogenous anticoagulant system, a system of
inhibitors downregulates fibrinolysis. PAI-1 is the principal
inhibitor of tPA. PAI-1 is primarily found in platelet α-granules.
Glycoprotein Ib/IX/V Complex
Upon platelet activation and secretion, rich concentrations of
PAI-1 are secreted into the thrombotic environment to protect GpIb-IX-V initiates platelet adhesion by binding tethered vWF
the thrombus from fibrinolysis. at the site of endothelial injury. There are approximately 25,000
TAFI is a carboxypeptidase produced by the liver and activa- copies of the receptor per platelet. The GpIb-IX-V comprises 4
ted by thrombin when thrombomodulin is present. This inhibitor polypeptide subunits: glycoproteins Ibα, Ibβ, IX, and V. Each
markedly attenuates fibrinolysis by cleaving the C-terminal polypeptide is present in duplicate in the complex except GpV.
lysines of fibrin, thus limiting plasminogen activation by tPA. The vWF binding domain is found on the glycoprotein Ibα
 48 Pathophysiology of Arterial Thrombosis 479

Thrombin-antithrombin
complex

Thrombin
Antithrombin

Thrombomodulin

Endothelial cell surface

Figure 48.6. Antithrombin Anticoagulant System. Antithrombin removes thrombin from thrombomodulin. The thrombin-antithrombin complex
is cleared by the liver.

subunit. There are 2 polymorphisms in the coding sequence After the initial tether of GpIb-IX-V to vWF, a second plate-
of the gene encoding the Ibα subunit of the receptor. These let receptor, GpIa-IIa (integrin α2β1), binds fibrillar collagen
polymorphisms lower the threshold for shear-induced platelet (types I, II, III, and V) and nonfibrillar collagen (types IV, VI,
adhesion and have a significant association with either stroke VII, and VIII), thereby arresting further translocation. This
or MI. receptor, comprising an α and a β subunit, exists on the plate-
let cell surface with fewer copies (3,000) than the more abun-
dant fibrinogen receptor, GpIIb-IIIa (50,000–80,000 copies).
The number of receptors, however, varies considerably among
Box 48.1. Contents of Platelet Storage Granules persons, spanning a 10-fold range. There is a direct relationship
between receptor density and platelet adhesion, whereby type I
α-Granule collagen adhesion varies over a 20-fold range and type III colla-
P-selectin gen adhesion varies over a 5-fold range, depending on the number
Growth factors (PDGF, TGF-β) of platelet receptors present. Receptor density is directly propor-
tional to platelet adhesion to type I collagen under high-shear
Fibrinogen
conditions typical of stenosed arteries. Increased receptor density
Factor V has been associated with both MI and stroke, especially in the
Receptors (GpIIb-IIIa, GpIb-IX, GpIV, GpV) young. In contrast, in patients with mild type I von Willebrand
Platelet factor 4
disease, characterized by a mild reduction of normally func-
tioning vWF, GpIa-IIa receptor density correlates with platelet
vWF adhesion under high-shear conditions. Patients deficient in this
Fibronectin receptor who have mild von Willebrand disease have more clini-
Vitronectin cal bleeding than patients who have von Willebrand disease and
normal receptor density.
Osteonectin (SPARC)
Dense body
Glycoprotein VI
Adenosine nucleotides (ADP, ATP)
Serotonin GpVI is a collagen receptor and a member of the immunoglobulin
superfamily of proteins. GpVI is important in signal transduction
Lysosomal granule
Hydrolases
Table 48.1. Platelet Receptors, Receptor Ligands,
LAMP-1 and LAMP-2
and Shear Stress
Microperoxisome
Catalase Receptor Ligand Optimal Shear Stress
GpIb-IX-V vWF High
Abbreviations: ADP, adenosine diphosphate; ATP, adenosine GpIa-IIa Collagen Low
triphosphate; Gp, glycoprotein complex; LAMP, lysosome-associated GpVI Collagen Low
membrane protein; PDGF, platelet-derived growth factor; SPARC, GpIIb-IIIa Fibrinogen, vWF Low
secreted protein acidic and rich in cysteine; TGF, transforming
P-selectin P-selectin binding ligand High
growth factor; vWF, von Willebrand factor.
Abbreviations: Gp, glycoprotein complex; vWF, von Willebrand factor.
480 V Aorta and Peripheral Vascular Disease
for collagen-induced platelet activation and therefore has a pivotal
role in the initial stages of platelet adhesion. Optimal activity of
GpVI requires a coreceptor, the Fc receptor γ chain. In genetically
altered mice deficient in the Fc receptor, collagen-induced plate-
let activation is impaired. A direct relationship exists between
GpVI receptor content and platelet-dependent prothrombinase
activity and thrombin generation. In platelets with a high GpVI
receptor density, prothrombin activation to thrombin is 3-fold
greater than in platelets with lower receptor content.
Glycoprotein IIb/IIIa Complex
The GpIIb-IIIa receptor (integrin α2bβ3) has a central role in
platelet thrombus propagation. This receptor is abundant, with
50,000 to 80,000 copies per cell, and requires activation before
binding its primary ligand, fibrinogen. Activation of the plate-
let results in a conformational change of the receptor from the
low-affinity state to the high-affinity state. Binding fibrinogen to
the activated receptor allows platelets to bind to each other dur-
ing platelet aggregation. The PLA2 polymorphism results from
a thymine-cytosine substitution at position 1565 and a leucine-
proline dimorphism at residue 33 of the β3 chain. The PLA2 poly-
morphism results in a lower threshold of platelet activation, with
increased P-selectin expression and fibrinogen binding. Platelets
with this polymorphism bind fibrinogen more tightly and exhibit
increased adhesion to immobilized fibrinogen with greater cell
spreading, actin reorganization, and clot retraction. Furthermore,
these platelets have a greater sensitivity to inhibition by thera-
peutic aspirin or abciximab (ReoPro, a chimeric monoclonal
antibody against GpIIb-IIIa). This polymorphism is common,
occurring on at least 1 allele in 25% of northern Europeans.
P-Selectin
P-selectin is expressed on the platelet cell membrane and medi-
ates the binding of activated platelets to leukocytes. P-selectin
is a component of platelet α-granules and is expressed on the
cell surface after platelet activation. Binding of platelets to leuko-
cytes by this receptor is mediated by P-selectin–binding ligand.
Tissue Factor (Extrinsic) Pathway
Tissue factor is an integral surface membrane glycoprotein
located in the wall of blood vessels (Figure 48.3). The blood pro-
coagulant system is initiated when vascular injury exposes tis-
sue factor to flowing blood at the site of injury. Vascular injury
results in rapid induction of tissue factor messenger RNA and
a 10-fold augmentation of procoagulant activity. In animal and
human studies, inhibitors of tissue factor effectively prevent arte-
rial thrombosis. After the site of vascular injury has been covered
by platelets and fibrin, exposure of flowing blood to the underly-
ing vascular tissue factor is restricted. Diffusion of procoagulant
factors from the blood to vascular tissue factor exposed by injury
is effectively blocked when the initially adherent platelet-fibrin
thrombus reaches a thickness of just a few microns. Thus, vas-
cular tissue factor cannot support continued thrombus growth
beyond a thickness of a few microns. It is now evident that tissue
factor also “circulates” within blood. Circulating tissue factor
appears to be “encrypted” within phospholipid microparticles
that “bud” off monocytes stimulated by cytokines (eg, tumor
necrosis factor α and IL-1). Circulating tissue factor is essential
for continued activation of the procoagulant system, accretion
of fibrin and activated platelets, and thrombus growth. Levels of
circulating tissue factor are elevated in patients with traditional
risk factors for atherosclerosis such as diabetes mellitus, hyper-
lipidemia, and tobacco exposure. Improved glycemic control in
diabetic patients is associated with a reduction of circulating tis-
sue factor and blood thrombogenicity assessed by ex vivo perfu-
sion chamber platelet deposition.
The primary inhibitor of tissue factor is TFPI, which binds
tissue factor in the presence of factor Xa. TFPI, a Kunitz-type
inhibitor, has several important functional domains. After TFPI
binds and inhibits factor Xa, the affinity for binding and inhib-
iting tissue factor and factor VIIa increases substantially. TFPI
limits thrombin generation (through the tissue factor pathway)
after minor thrombotic stimuli. TFPI is expressed on a number
of cell surfaces (particularly endothelial cells), it is secreted into
plasma, and it interacts with heparin and lipoproteins. The inhib-
itor is cleared by endothelial internalization through caveolar
sequestration and endocytosis.
Contact Activation (Intrinsic) Pathway
The extrinsic (tissue factor) and intrinsic (contact activation)
coagulation factor activation pathways converge on prothrom-
bin activation to thrombin (Figure 48.3). The intrinsic pathway
is initiated when factor XII comes into contact with negatively
charged surfaces. Neither the mechanism for factor XII activa-
tion nor its relevance to in vivo coagulation activation is known.
In contrast to factor VII deficiency, for example, patients with
factor XII deficiency do not experience excessive bleeding. From
this, many have assumed that thrombin generation in vivo is
largely the result of activation of the extrinsic (tissue factor) path-
way with little input from factor XII activation. This hypothesis
has been recently challenged by experiments with factor XII–
deficient mice. These mice did not bleed spontaneously and the
bleeding times were normal. After arterial injury, platelet adhe-
sion occurred normally; however, both the formation and the
stabilization of thrombi in these mice were severely impaired.
Thrombi did not occur at all in half of the injured mice. In the
other half, the thrombi that formed were unstable and detached
from the vessel wall within 1 minute of formation. Similar results
were noted in factor XI–deficient mice after arterial injury.
von Willebrand Factor
Functional domains of vWF subunits include binding of platelet
receptors GpIb and GpIIb-IIIa, coagulation factor VIII, heparin,
and collagen types I, III, and VI. vWF is synthesized in endothe-
lial cells (the principal source of circulating plasma protein) and
megakaryocytes. In endothelial cells, newly synthesized vWF is
either secreted constitutively or stored in Weibel-Palade bodies,
where polymerization of the protein occurs. This polymerization
process can be extensive. Circulating vWF exists in plasma as
multimers containing various numbers of subunits with molecu-
lar weights ranging from 500 kDa to more than 20,000 kDa. The
ultralarge vWF multimers, when released, form high-strength
bonds with platelet GpIb-IX-V spontaneously, resulting in
platelet adhesion, aggregation, and thrombus formation. Under
normal circumstances, thrombosis is prevented by rapid (but
partial) proteolysis by a metalloprotease called ADAMTS-13.
ADAMTS-13 cleaves the Y842/M843 peptide bond in the vWF
A2 domain, releasing 176-kDa and 140-kDa fragments that are
active only in the presence of modulators (ristocetin, botroce-
tin, and high-shear stress). Disturbances in proteolysis have a
devastating outcome: increased cleavage causes severe bleeding
 48 Pathophysiology of Arterial Thrombosis 481

(von Willebrand disease type IIA), while decreased proteolysis infection, and obesity. Moreover, an elevated fibrinogen level
induces multiorgan thrombosis (thrombotic thrombocytopenic seems to compound the attributable risk from the acquired
purpura). Patients with stable angina have not only an increased factors.
vWF antigen concentration but also a significant decrease in
ADAMTS-13 activity.
Old Paradigm of Arterial Thrombosis
In the old paradigm of arterial thrombosis, acute arterial luminal
Fibrinogen thrombosis resulting in MI is widely viewed as an unpredictable
Fibrinogen is a large protein with a molecular weight of 340,000 event governed predominantly by the rupture of an atheroscle-
Da. It is synthesized in the liver and is the most abundant coag- rotic plaque (Figure 48.7). Observations of autopsy specimens
ulation protein in the circulation. Fibrinogen is unique among have identified specific features associated with atherosclerotic
other coagulation proteins in that it is transcribed from 3 separate plaques more likely to rupture. Plaques exhibiting these features
genes. Fibrinogen is a dimer of 3 pairs of polypeptide chains: 2 are referred to as vulnerable plaques. Features associated with
Aα, 2 Bβ, and 2 γ chains. Each Aα, Bβ, and γ chain is covalently plaque vulnerability include a lipid-rich core with abundant lipid-
linked through disulfide bonds, and the 2 mirror-image 3-strand
chains are linked together through a disulfide bond. The struc-
ture of fibrinogen is therefore an elongated mirror image of two
3-chain strands (Aα, Bβ, and γ chains), which are further divided
into functional domains. The central E domain and the lateral D Arterial
lumen
domain are separated by a central coiled-coil domain. Fibrinogen
is cleaved to fibrin by the enzymatic removal of fibrinopeptides
A and B (central E domain) by thrombin. After it is cleaved, the Lipid-rich
soluble fibrin monomer is unstable and immediately undergoes core
polymerization with other fibrin monomers to form an extensive
fibrin multimer. This multimer is then further stabilized through Thin fibrous
the activity of factor XIII, which is also activated by thrombin. cap
Another important functional domain in fibrinogen is the
integrin α2bβ3 (GpIIb-IIIa) binding sequence in the D domain.
This arginine-glycine-aspartic acid sequence found on the Aα
chain within this domain is an important binding site for the
platelet-to-platelet aggregation that is necessary for platelet-rich
thrombus formation in arterial thrombi. Two important integ-
rin recognition sites are also present on the γ chain. All 3 sites
are probably necessary for proper platelet aggregation and clot
Inflammatory
retraction. infiltrate
Plasmatic factors have been studied extensively for their
contribution to arterial thrombosis. Several large epidemio-
logic studies have found an association between elevated levels
of fibrinogen and cardiovascular events. An elevated fibrinogen
level has been associated with an increased risk of MI, stroke,
venous thrombosis, and death among patients with peripheral
arterial disease. Fibrinogen is an acute-phase reactant that may
increase in response to inflammatory mediators such as IL-6. It is
therefore difficult to determine whether the disease pathogenesis
and thrombotic outcomes result from an elevated fibrinogen level Plaque rupture
or from the underlying inflammation of atherosclerosis. Both
genetic factors and environmental factors influence fibrinogen
levels. As much as 50% of fibrinogen variability is now thought
to be under genetic influence. For example, a polymorphism of
the Bβ gene (−455A allele) results in higher plasma fibrinogen
levels, particularly in the homozygous state (−455AA). In the
AIRGENE study, 895 survivors of MI were assessed for plasma
fibrinogen, IL-6, and C-reactive protein levels associated with
21 SNPs and the corresponding haplotypes in the 3 fibrinogen
genes. Fibrinogen was measured serially over time to minimize
variability. Eight SNPs were significantly associated with fibrin- Platelet-rich
ogen levels. These included alleles for genes encoding for both
thrombosis
the Aα chain and the Bβ chain. Fibrinogen levels correlated with
both IL-6 and C-reactive protein measures. In addition, IL-6
levels significantly modified the association between fibrinogen
levels and several of the haplotypes for both Aα and Bβ chains. Figure 48.7. Old Paradigm of Arterial Thrombosis. The 3 lower
Environmental factors known to increase fibrinogen lev- drawings depict close-up views of the steps occurring at the shoulder of
els include tobacco use, aging, hypertension, diabetes mellitus, the plaque (indicated by the square in the upper drawing).
482 V Aorta and Peripheral Vascular Disease
laden macrophage foam cells enclosed within a thin (<65 μm),
friable fibrous cap. The site of rupture is typically at the shoulder
of the plaque and is histologically characterized by an inflam-
matory infiltrate of activated macrophages and T lymphocytes.
These inflammatory cells secrete cytokines and matrix metallo-
proteinases that proteolytically degrade the extracellular matrix,
thus weakening the structure of the thin fibrous cap. The plaque
rupture may be spontaneous or be triggered by a surge of sym-
pathetic activity associated with emotional stress or physical
activity. Sudden increases in blood pressure and the force of car-
diac contractility may result in critical elevations of wall tension
along the coronary arterial wall. Increases in wall tension with
mechanical stress on the thin fibrous cap, which lacks sufficient
structural strength, may then lead to plaque rupture. After plaque
rupture, exposure of the lipid-rich core to flowing blood provides
a potent stimulus for platelet-rich thrombus formation.
Although atherosclerotic plaque rupture is held to be the ini-
tiating event in most cases of acute MI, not all ruptured plaques
result in thrombosis. Furthermore, many cases of coronary
thrombosis occur without demonstrable plaque disruption, and
between 4% and 30% of events occur in angiographically normal
coronary arteries. Plaque erosion, defined as endothelial cell loss
over a smooth muscle and proteoglycan-rich fibrous plaque, may
account for 15% to 44% of acute coronary thrombotic events.
These coronary lesions lack the features typical of the vulnerable
plaque, including the lipid-rich core. Endothelial denudation cen-
tral to the plaque erosion hypothesis may be a common phenom-
enon in the general population, infrequently resulting in acute
Normal artery
Thrombotic propensity
Yes Yes
1 2
4%-30% of MIs 15%-44% of MIs
Figure 48.8. New Paradigm of Arterial Thrombosis. See text for discussion of scenarios 1 through 4. MI indicates myocardial infarction.
luminal thrombosis. The poor correlation between atheroscle-
rosis and thrombosis suggests that additional factors other than
simply the extent of atherosclerosis are important in determining
in situ thrombosis. Acute arterial luminal thrombosis could be a
random process, or it could arise from histologic or biochemical
differences in atherosclerotic plaque composition, geometry, or
stability. Alternatively, interindividual differences in the propen-
sity for forming arterial thrombi could govern this process. These
differences may arise from blood-borne cellular or plasmatic fac-
tors and may be constitutive or constantly evolving in response to
neuroendocrine, inflammatory, or metabolic influences.
New Paradigm of Arterial Thrombosis
The new paradigm of arterial thrombosis suggests that the clin-
ical outcome is governed by the morphology, composition, and
stability of the atherosclerotic plaque and by the propensity of the
person to form platelet-rich arterial thrombi. In the new paradigm,
individuals may have a very high risk of thrombosis when both
atherosclerotic plaque rupture and an increased propensity for
arterial thrombosis are present, or they may have a very low risk
when both variables are absent. Four scenarios are recognized on
the basis of the presence or absence of variables that increase the
propensity for thrombosis and the presence or absence of athero-
sclerotic disease and plaque rupture (Figure 48.8):
1. Acute thrombosis and normal coronary artery—The relative fre-
quency of an acute MI developing is low in the absence of athero-
sclerosis; however, 4% to 30% of MIs occur with angiographically
Diseased artery
Yes No
3 4
Old paradigm
 48 Pathophysiology of Arterial Thrombosis 483

normal coronary arteries. Whether these individuals harbor cryptic GpIa-IIa glycoprotein Ia/IIa complex
atherosclerosis as a substrate for acute thrombosis is not known. GpIb glycoprotein Ib
2. Acute thrombosis and coronary plaque without rupture—In the GpIb-IX-V glycoprotein Ib/IX/V complex
reported 15% to 44% of MI patients who have plaque erosion rather GpIIb-IIIa glycoprotein IIb/IIIa complex
than rupture, the major stimulus may be the thrombotic propensity GpV glycoprotein V
rather than the arterial substrate. GpVI glycoprotein VI
3. Acute thrombosis and coronary plaque rupture—In the old para- IL interleukin
digm, the contribution of the thrombotic propensity to this clinical MI myocardial infarction
outcome is poorly understood. PAI plasminogen activator inhibitor
4. Plaque rupture without thrombosis—This situation may be relatively PGI2 prostacyclin
prevalent in the general population of asymptomatic persons and has SNP single-nucleotide polymorphism
been hypothesized to lead to plaque progression or growth. TAFI thrombin activatable fibrinolysis inhibitor
TFPI tissue factor pathway inhibitor
In summary, the thrombotic propensity may be governed by tPA tissue plasminogen activator
platelet variables, by plasmatic variables, or by synergistic inter- vWF von Willebrand factor
action between both types of variables.

Abbreviations Name of Clinical Trial

ADAMTS a disintegrin and metalloprotease with thrombospondin AIRGENE Air Pollution and Inflammatory Response in Myocardial
motif Infarction Survivors: Gene-Environment Interaction in
ADP adensosine diphosphate a High Risk Group

Treatment and Prevention of Arterial Thrombosis
ROBERT D. MCBANE, MD, and WALDEMAR E. WYSOKINSKI, MD, PHD
Each year, 1.45 million Americans have an MI and 800,000
Americans have a new or recurrent stroke. These events are
thrombotic in nature and are therefore potentially treatable and
preventable with the appropriate use of antithrombotic agents.
Current antithrombotic therapy targets 3 of the 4 stages of
thrombus formation: platelet activation, platelet aggregation, and
coagulation cascade activation. There are no effective means to
prevent the initial stage of platelet adhesion to injured arterial
walls. This chapter reviews the mechanism of action, pharmacol-
ogy, and clinical use of agents for the prevention and treatment
of arterial thrombosis.
• The 4 stages of thrombus formation are the following:
1. Platelet adhesion
2. Platelet activation
3. Platelet aggregation
4. Coagulation cascade activation
Platelet Inhibitors
Antiplatelet therapy is effective in preventing and treating arte-
rial thrombotic events. Large clinical trials have documented a
morbidity and mortality benefit with antiplatelet agents for dis-
eases of coronary, cerebral, and peripheral arterial circulations.
The 4 main classes of antiplatelet agents are COX inhibitors,
ADP-receptor antagonists, phosphodiesterase inhibitors, and
platelet GpIIb-IIIa receptor inhibitors. COX inhibitors, ADP-
receptor antagonists, and phosphodiesterase inhibitors block
platelet activation by agonist inhibition; agents in these 3 classes
are well suited for both primary and secondary prevention. The
Abbreviations and acronyms are expanded at the end of this chapter.
484
49
GpIIb-IIIa inhibitors interfere with the final common pathway of
platelet aggregation; these agents are effective in acute arterial
thrombosis.
• The 4 main classes of antiplatelet agents are the following:
1. COX inhibitors
2. ADP-receptor antagonists
3. Phosphodiesterase inhibitors
4. Platelet GpIIb-IIIa receptor inhibitors
COX Inhibitors
Aspirin
Aspirin permanently inhibits COX activity in the platelet
(Box 49.1). COX exists in 2 isoforms (COX-1 and COX-2), but
platelets contain only COX-1. In the platelet, COX-1 is respon-
sible for the first step in the conversion of arachidonic acid to its
ultimate product, TXA2. TXA2 induces platelet activation and
aggregation and promotes vasoconstriction (Figure 49.1). COX
is also responsible for the production of endothelial cell–derived
PGI2. In contrast to TXA2, PGI2 inhibits platelet aggregation and
induces vasodilation. At thromboprophylactic doses of aspirin,
platelet COX-1 is preferentially inhibited and vascular PGI2 pro-
duction continues relatively unabated.
Aspirin is rapidly absorbed from stomach mucosa, with peak
plasma levels achieved within 30 minutes of ingestion. Enteric
coating, however, may delay the absorption by 3 to 4 hours. In
an urgent setting, therefore, enteric-coated aspirin should be
chewed to remove the outer coating. The plasma half-life of aspi-
rin is short, ranging from 15 to 20 minutes. In healthy persons,
a single dose of 100 mg of aspirin results in 98% inhibition of
thromboxane production within 1 hour after ingestion. This same
 49 Treatment and Prevention of Arterial Thrombosis 485

cardiovascular protection. For patients at high risk of arterial

Box 49.1. Aspirin
Situations with the greatest benefit-to-risk ratio:
Lone atrial fibrillation (without additional risk factors)
Chronic stable angina
Prior myocardial infarction
Unstable angina
Acute myocardial infarction or stroke
Risk of major hemorrhage: approximately 0.7%
per year
No proven benefit for venous thrombosis
prophylaxis
inhibition is seen with 0.45 mg/kg daily (ie, 40 mg) given for
4 days. When aspirin use is discontinued, COX recovery occurs
as new platelets enter the circulation. Platelet lifespan averages 9
to 11 days; therefore, sensitive assays of platelet aggregation may
be abnormal for up to 10 days after a single dose of aspirin. Each
day, 10% of the circulating platelet population is renewed.
Aspirin therapy for patients with coronary artery disease is
associated with a 23% reduction in the occurrence of vascular
events. The search for the optimal dose of aspirin has led to mul-
tiple randomized controlled trials, with aspirin dosages ranging
from 30 to 1,500 mg daily. These trials have demonstrated the
efficacy of low-dose aspirin for both primary and secondary
thrombosis, the lowest effective dose is 75 mg. For patients with
either an acute MI or an acute stroke, the lowest effective dose
is 160 mg. Although aspirin is an effective antithrombotic agent
over a large dose range, the lower doses appear to be more effec-
tive than higher doses and result in fewer gastrointestinal side
effects and lower bleeding rates. The risk of major bleeding with
low-dose aspirin is less than 1%. Neither enteric-coated nor buff-
ered preparations have been shown to reduce this risk.
More than 70 completed trials, enrolling more than 100,000
patients, have assessed the efficacy of aspirin therapy. The big-
gest relative benefit is for patients with acute arterial thrombotic
events. In the ISIS-2 trial, for example, patients receiving aspirin
(162 mg daily) had a nearly 25% reduction in mortality and a
nearly 50% reduction in recurrent MI or stroke. In other trials,
among moderate- or high-risk patients (those with a remote his-
tory of MI, unstable angina, prior TIA, or stroke), aspirin use
was associated with a 25% risk reduction. In contrast, primary
prevention trials of aspirin have shown mixed results, largely
depending on the degree of cardiovascular risk of the population
studied. In very low-risk individuals, it has been difficult to show
an advantage of aspirin over placebo for primary prevention.
For atrial fibrillation, the consensus is that anticoagulant ther-
apy with coumarin derivatives reduces the risk of thromboembo-
lism more effectively than antiplatelet therapy for most patients.
Moreover, aspirin does not provide effective venous thrombosis
prophylaxis.
Some patients, however, do have a thrombotic event despite
daily aspirin therapy. One in 8 patients treated with aspirin has
an arterial thrombotic event in 2 to 3 years of follow-up. One

COOH

Arachidonate
ASA
COX-1

O
COOH

O
PGG2
OOH

O
COOH

O
PGH2
OH

COOH
O
O
OH
Thromboxane A2
OH
COOH COOH
O
O HO O
OH OH TXB2
Figure 49.1. Platelet Arachidonate Metabolism. Arachidonate is sequentially metabolized to thromboxane A2 within the platelet. Thromboxane
A2 is the metabolite that ultimately activates platelets before being converted to thromboxane B2 (TXB2) for renal excretion. ASA indicates aspirin;
COOH, carboxyl group; COX-1, cyclooxygenase 1; O, oxygen; OH, hydroxyl group; PGG2, prostaglandin G2; PGH2, prostaglandin H2.
486 V Aorta and Peripheral Vascular Disease
proposed explanation for these aspirin failures is that platelets in
these patients are not effectively inhibited and are therefore aspi-
rin resistant. Aspirin resistance is the failure of aspirin to produce
the expected inhibition of platelet function by in vitro testing or
by in vivo thrombotic outcomes. The prevalence of aspirin resist-
ance varies from 5.2% to 60% depending on the patient popula-
tion studied and the criteria used for defining the condition.
Of more than academic interest, aspirin resistance increases
the risk of thrombosis, as demonstrated in a nested case-con-
trol substudy of the HOPE trial. Aspirin-resistant patients had
a 2-fold increased risk of MI and a 3.5-fold increased risk of
cardiovascular death compared with aspirin-sensitive patients.
The prevalence of aspirin resistance in patients with PAD who
were undergoing elective angioplasty was nearly 60% in 1 trial;
subsequent arterial occlusions all occurred in aspirin-resistant
patients. In a study of stroke survivors, the main thrombotic end
points were nearly 10-fold higher for aspirin-resistant patients
than for aspirin-sensitive patients. The exact mechanism of aspi-
rin resistance is not known, and screening for aspirin resistance
has not been widely endorsed. The therapeutic adjustments that
should be instituted when it is present are unclear.
• At doses of aspirin used for thromboprophylaxis, platelet COX-1 is
preferentially inhibited.
• Platelet lifespan averages 9–11 days; therefore, sensitive assays
of platelet aggregation may be abnormal for up to 10 days after a
single dose of aspirin.
• For patients at high risk of arterial thrombosis, the lowest effective
dose of aspirin has been shown to be 75 mg.
• For patients with either an acute MI or an acute stroke, the lowest
effective daily dosage of aspirin is 160 mg.
• Aspirin resistance increases the risk of thrombosis.
ADP-Receptor Antagonists: Thienopyridines
ADP, a weak but important platelet agonist, stimulates platelet
function through the binding of the ADP receptor P2Y12. Upon
activation, ADP is released from platelet-dense granules and
from endothelial cells. The release of ADP amplifies recruitment
and aggregation of adjacent platelets, thereby promoting arterial
thrombus formation. Three structurally similar ADP receptor
antagonists (Box 49.2) are available: ticlopidine, clopidogrel, and
Box 49.2. Thienopyridines
Inhibit platelet ADP receptor P2Y12
Situations with the greatest benefit-to-risk ratio:
Peripheral arterial occlusive disease
Non–ST-segment elevation myocardial infarction
(combined with aspirin)
Coronary stent implantation (particularly with drug-
eluting stent)
Risk of major hemorrhage:
1.4% per year (clopidogrel alone)
2.0%–3.7% per year (combined with aspirin)
No proven benefit for venous thrombosis
prophylaxis
Abbreviation: ADP, adenosine diphosphate.
prasugrel. All 3 agents require in vivo hepatic transformation to
the short-lived active metabolite. Each agent inhibits platelet
function by the irreversible modification of the ADP receptor
P2Y12. Steady-state platelet inhibition occurs 4 to 7 days after
initiation of drug therapy. When clopidogrel is administered, the
inhibitory effect is achieved more rapidly with a 300-mg loading
dose. Higher loading doses of clopidogrel (450–900 mg) before
PCI may provide an additional and earlier inhibitory effect than
300 mg. By comparison, prasugrel therapy inhibits platelet func-
tion more rapidly and more extensively than clopidogrel therapy
regardless of the patient population studied.
Ticlopidine
Ticlopidine has been effective in trials for secondary preven-
tion of MI and may be superior to aspirin for stroke. Compared
with warfarin plus aspirin, combined antiplatelet therapy with
ticlopidine and aspirin after coronary artery stenting reduced the
incidence of MI, stent thrombosis, and the need for subsequent
intervention with fewer vascular access and hemorrhagic compli-
cations. In 1 study of high-risk patients undergoing PCI with bare
metal stents, this composite end point was reached by 1.5% of
those receiving ticlopidine plus aspirin and by 6.2% of those ran-
domly assigned to receive warfarin plus aspirin. The rate of stent
thrombosis decreased from 5.4% to 0.8% for patients receiving
antiplatelet therapy. In another study of moderate-risk patients,
the use of ticlopidine plus aspirin (compared with warfarin plus
aspirin) reduced the composite end point rate from 3.7% to 0.5%.
These antithrombotic benefits were durable throughout the first
year after the procedure. On the basis of this early experience,
the combination of these agents has become the standard of care
after PCI.
In the STIMS trial, which compared ticlopidine with placebo
in patients with intermittent claudication, relative risk decreased
30% for all-cause mortality and 50% for fatal vascular events.
Compared with placebo, ticlopidine significantly improves
long-term patency of saphenous vein bypass grafts to the lower
extremity. Diarrhea (in 2% of patients) and neutropenia (in 1%
of patients) are potential adverse effects that require monitoring
and discontinuing drug therapy.
Clopidogrel
The efficacy of clopidogrel has been studied extensively in ran-
domized controlled trials for several indications (Table 49.1).
For the best use for this drug, each indication should be assessed
individually (Box 49.2). CAPRIE and CHARISMA assessed
the role of clopidogrel in patients with high-risk atherosclerotic
disease. In the CAPRIE trial, patients were randomly assigned
to receive clopidogrel (75 mg daily) or aspirin (325 mg daily)
to determine which therapy resulted in the greatest reduction
in vascular events. Patients were distributed nearly equally into
3 subgroups: stroke, MI, or symptomatic atherosclerotic PAD.
Significantly fewer annual events (stroke, MI, or vascular death)
occurred in patients treated with clopidogrel than in patients
treated with aspirin (5.32% vs 5.83%). The biggest risk reduction
occurred among patients who had PAD. For these PAD patients,
the average event rate per year was 3.71% in the clopidogrel
group compared with 4.86% in the aspirin group.
The CHARISMA investigators extended these observations
by assessing the efficacy of dual antiplatelet therapy (clopidogrel
plus aspirin) compared with aspirin alone for high-risk athero-
sclerosis patients. The combination was not superior to aspirin
 49 Treatment and Prevention of Arterial Thrombosis 487

Table 49.1. Clinical Trials With Clopidogrel for Specific Indications

Clinical Trial Clinical Setting No. of Patients Comparison Outcome

Stable High-Risk Atherosclerosis

CAPRIE Stroke, MI, or PAD 19,185 Clopidogrel vs ASA Clopidogrel was favored over ASA to reduce
rates of CV events and mortality (PAD
patients)
CHARISMA Stroke, MI, or PAD 15,603 Clopidogrel + ASA vs ASA Clopidogrel + ASA was not superior to ASA
alone
ACS
CURE NSTEMI 12,562 Clopidogrel + ASA vs ASA In all 3 trials (CURE, COMMIT, and
CLARITY), clopidogrel + ASA was
favored over ASA alone to reduce rates of
CV events and mortality among patients
with ACS
COMMIT STEMI 45,852 Clopidogrel + ASA vs ASA
CLARITY STEMI 3,491 Clopidogrel + ASA vs ASA after
fibrinolysis
PCI
CREDO Elective PCI 2,116 Clopidogrel: loading dose vs no Loading dose did not improve risk of events
loading dose before PCI Pretreatment ≥6 h before PCI reduced CV
event rate
Treatment beyond 1 y reduced CV event rate
PCI-CLARITY STEMI 1,863 Clopidogrel + ASA vs ASA with In both trials (PCI-CLARITY and PCI-
fibrinolysis before PCI CURE), clopidogrel + ASA was favored
over ASA alone
PCI-CURE NSTEMI 2,658 Clopidogrel + ASA vs ASA
AF
ACTIVE A AF, “unsuitable” for warfarin 7,554 Clopidogrel + ASA vs ASA Clopidogrel + ASA reduced CV events,
including stroke
Benefit was offset by major bleeding
ACTIVE W AF “suitable” for warfarin 6,706 Clopidogrel + ASA vs warfarin Stopped early because of warfarin
superiority
Stroke
MATCH TIA or stroke 7,599 Clopidogrel + ASA vs clopidogrel Clopidogrel + ASA was not superior to
clopidogrel alone, and bleeding rates
were higher
PROFESS Stroke 20,332 Clopidogrel vs ASA + dipyridamole Both therapies had similar efficacy
Major bleeding rates were higher with
ASA + dipyridamole

Abbreviations: ACS, acute coronary syndrome; AF, atrial fibrillation; ASA, aspirin; CV, cardiovascular; MI, myocardial infarction; NSTEMI, non–
ST-segment elevation myocardial infarction; PAD, peripheral artery disease; PCI, percutaneous coronary intervention; STEMI, ST-segment elevation
myocardial infarction; TIA, transient ischemic attack.

alone in reducing the rate of cardiovascular events, and the rate of
serious or fatal hemorrhage did not differ between the 2 groups.
• Combined data suggest that clopidogrel (compared with aspirin)
may provide more effective prophylaxis in patients with PAD.
• For patients with high-risk but stable PAD, dual antiplatelet ther-
apy is no more effective than aspirin alone.
Clopidogrel has been extensively studied in patients with
acute coronary syndromes. The CURE investigators randomly
assigned patients with recent NSTEMI to receive clopidogrel
plus aspirin or aspirin alone. Patients receiving the combination
had a significantly lower rate of reaching the composite end point
(vascular death, nonfatal MI, or stroke) compared with patients
receiving aspirin alone (9.3% vs 11.4%). Clopidogrel therapy,
however, was associated with a significantly greater risk of major
bleeding. Patients who needed cardiac surgery were required to
wait for 5 days after stopping clopidogrel therapy because of the
increased risk of major hemorrhage when taking this drug.
The COMMIT investigators assessed dual antiplatelet ther-
apy with clopidogrel plus aspirin or aspirin alone in patients with
acute coronary syndrome (93% had STEMI or new left bundle
branch block). Clopidogrel plus aspirin significantly reduced
the rates of death, reinfarction, or stroke compared with aspirin
alone (9.2% vs 10.1%) without significantly increasing the rates
of major hemorrhage.
The CLARITY investigators sought to determine the added
efficacy of clopidogrel plus aspirin after fibrinolytic therapy in
patients with STEMI. The primary efficacy end point was a com-
posite of an infarct-related artery occlusion confirmed by angiog-
raphy, recurrent MI, or death. An absolute 6.7% decrease in the
occurrence of the primary efficacy end point was reported for
patients who were randomly assigned to dual antiplatelet therapy
(15.0% vs 21.7%).
• Dual antiplatelet therapy with clopidogrel plus aspirin reduces the
occurrence of ischemic events in patients with unstable angina,
NSTEMI, or STEMI and thus is considered standard therapy in
those instances.
The efficacy of clopidogrel plus aspirin has been extensively
studied in patients undergoing PCI. The PCI-CURE trial tested
488 V Aorta and Peripheral Vascular Disease
the hypothesis that clopidogrel plus aspirin was more effective
than aspirin alone, sustainable up to 1 year, in preventing major
ischemic events after PCI. The primary end point (composite of
cardiovascular death, MI, or urgent target vessel revasculariza-
tion within 30 days) was achieved in significantly fewer patients
in the clopidogrel group (4.5%) than in the placebo group (6.4%),
and the outcomes persisted to 1 year.
Similarly, PCI-CLARITY randomly assigned STEMI
patients to receive clopidogrel plus aspirin or aspirin alone when
fibrinolytic therapy was initiated. Pretreatment with clopidogrel
plus aspirin significantly reduced the occurrence of cardiovascu-
lar events before and after PCI through 30 days without affecting
bleeding rates.
The CREDO trial sought to determine both the benefit of
a preprocedural loading dose of clopidogrel and the efficacy
of long-term clopidogrel therapy after elective PCI. This study
showed that extending clopidogrel therapy for 1 year decreased
the occurrence of the combined end points (death, MI, or stroke)
by an absolute value of 3%. A 300-mg pretreatment loading
dose, however, did not further reduce the risk of these end points.
The time interval between the loading dose and PCI, however,
was important. Among patients receiving a loading dose at least
6 hours before PCI, the cardiovascular event rate was reduced. In
other studies, a larger, 600-mg loading dose of clopidogrel pro-
vided earlier and more complete platelet inhibition than a 300-
mg loading dose. Both the ISAR-REACT and the ISAR-CHOICE
trials showed that if a 600-mg loading dose is administered, there
is no advantage to increasing the pretreatment duration beyond
2 hours. Increasing the loading dose to 900 mg only marginally
increases the rapidity of platelet inhibition.
• The combination of clopidogrel and aspirin has become standard
therapy after coronary stent implantation, particularly with drug-
eluting stents.
• A thienopyridine combined with aspirin is superior to warfarin
plus aspirin for preventing thrombotic complications after stent
implantation.
Clopidogrel with or without aspirin has also been extensively
studied for secondary stroke prevention. The MATCH inves-
tigators randomly assigned patients with recent stroke or TIA
to receive clopidogrel or clopidogrel plus aspirin. The rates of
reaching the composite end points of stroke, MI, vascular death,
or rehospitalization for acute ischemia were comparable for the
clopidogrel-plus-aspirin (15.7%) and clopidogrel-alone (16.7%)
groups. The rate of life-threatening bleeding, however, was sig-
nificantly higher for patients receiving dual antiplatelet therapy
(2.6% vs 1.3%). The rate of intracranial hemorrhage was 1% for
both groups.
The PROFESS trial compared clopidogrel with aspirin plus
extended-release dipyridamole in stroke patients. After a mean
follow-up of 2.5 years, rates of recurrent stroke were simi-
lar for both groups (8.8% vs 9.0%). Major hemorrhagic events
were more common in the aspirin-dipyridamole group (4.1% vs
3.6%). These data suggest that clopidogrel provides similar effi-
cacy for secondary stroke prevention compared with aspirin plus
dipyridamole.
• Aspirin in combination with clopidogrel does not improve efficacy
in secondary stroke prevention; use of the combination is associ-
ated with excessive major bleeding and therefore is not advocated.
For preventing thromboembolism in atrial fibrillation, current
management strategies begin with a risk assessment of major vas-
cular events with use of the CHADS2 score system. According to
this strategy, antiplatelet therapy alone is an acceptable option for
low-risk patients (0 or 1 point). For atrial fibrillation patients who
are not low risk, warfarin therapy should be strongly considered
if the patients do not have contraindications. For these patients,
systems capable of providing high-quality warfarin management
should include specialized anticoagulation clinics and home
monitoring of INR. For patients who are not candidates for war-
farin therapy, clopidogrel in combination with aspirin would
probably improve the rate of major vascular events, including
ischemic stroke, but would also increase the risk of major hem-
orrhage (primarily in the gastrointestinal tract).
Multiple factors, both genetic and nongenetic, determine
clopidogrel metabolism, pharmacodynamic response, and effect
on cardiovascular events. Genetic variables govern absorption,
activation, and receptor physiology. Clopidogrel efficacy depends
on hepatic conversion to the active metabolite by the CYP sys-
tem, mainly CYP2C19. The overall effectiveness of the enzyme
in metabolizing clopidogrel depends on the genotype of that
enzyme. The wild-type for the metabolism alleles (CYP2C19*1)
results in normal activation of clopidogrel. The loss of the
function allele (CYP2C19*2), which may be either heterozy-
gous (CYP2C19*1/CYP2C19*2) or homozygous (CYP2C19*2/
CYP2C19*2), results in decreased activation of clopidogrel with
suboptimal platelet inhibition. The prevalence of these alleles
varies by race and ethnicity. Homozygous CYP2C19*2 alleles
occur in 2% of whites, 3% to 4% of African Americans, and
13% of East Asians. Heterozygous CYP2C19*2 alleles occur in
27% of whites.
Several studies have addressed the effect of these alleles on
clopidogrel responsiveness in healthy subjects and in patients
with vascular disease. In 1 study, heterozygous CYP2C19*1/
CYP2C19*2 carriers (compared with noncarriers) had a 32.4%
relative reduction of measurable active metabolite and lower max-
imal platelet inhibition in response to clopidogrel. Interventional
studies have suggested that carriers and persons with homozy-
gous alleles have more vascular events than noncarriers (with
wild-type alleles). Whether the allele carriers respond to higher
doses of clopidogrel is not completely clear. In a study of patients
receiving 75 mg daily, platelet function testing was assessed at
baseline and after receiving 150 mg daily for 7 days. Carriers had
much less baseline platelet inhibition compared with wild-type
noncarriers (18% vs 53%). Although platelet inhibition in car-
riers improved with higher clopidogrel dosing, the improvement
was only modest. In summary, although provocative, these data
are inconclusive.
• For patients who are receiving clopidogrel and doing well, clopi-
dogrel therapy should be continued and genetic testing is probably
not warranted.
• For patients undergoing PCI, genetic testing is not recommended
unless there is clinical suspicion of clopidogrel hyporesponsive-
ness as evidenced by breakthrough clinical events during dual
antiplatelet therapy.
• For secondary stroke prevention, genetic testing could be con-
sidered before use of clopidogrel since a good alternative ther-
apy (aspirin plus extended-release dipyridamole) is available for
nonresponders.
• For nonresponders without a good alternative therapy, it is not
clear whether simply increasing the clopidogrel dosage sufficiently
inhibits platelets.
Nongenetic factors, including drug-drug interactions from
lipophilic statins, calcium antagonists, and PPIs, may also
affect clopidogrel activation and the antiplatelet response. PPIs,
 49 Treatment and Prevention of Arterial Thrombosis
especially omeprazole, have been implicated in a lack of response
to clopidogrel, possibly through inhibition of the hepatic CYP2C19
activation of the drug to the active metabolite. In summary, the lab-
oratory evidence of reduced platelet inhibition when thienopyri-
dines are used during PPI therapy does not appear to translate into
clinically meaningful cardiovascular outcomes.
Although the main side effect of clopidogrel therapy is hem-
orrhagic complications, an additional adverse reaction includes
TTP. TTP is a rare, potentially fatal multisystem disease charac-
terized by thrombocytopenia, microangiopathic hemolytic ane-
mia, fever, neurologic deficits, and renal failure. Both clopidogrel
and ticlopidine have been shown to cause TTP in about 0.02% of
patients taking these drugs. In the original report of the associa-
tion, TTP developed in all but 1 of 11 patients within 14 days after
use of clopidogrel began. The pathophysiologic mechanism is
impaired proteolysis of von Willebrand factor by ADAMTS-13.
Immune-mediated ADAMTS-13 deficiency results in excess cir-
culating ultralarge von Willebrand factor multimers, which bind
platelets, resulting in widespread platelet microthrombi that char-
acterize the syndrome. Most patients respond to plasma exchange
therapy; however, multiple treatments may be required.
Prasugrel
In 2009, the FDA approved prasugrel, a third-generation thieno-
pyridine that irreversibly inhibits the ADP P2Y12 receptor.
Prasugrel requires hepatic metabolism for activation; however,
this occurs through CYP3A and CYP2B6 and is therefore not
limited by either PPI therapy or CYP2C19 polymorphism status.
In the TRITON–TIMI 38 trial, which compared prasugrel with
clopidogrel, the occurrence of the primary efficacy end point (car-
diovascular death, nonfatal MI, or nonfatal stroke) was decreased
in favor of prasugrel regardless of clinical presentation. Prasugrel
therapy was also associated with significantly lower rates of MI,
urgent target-vessel revascularization, and stent thrombosis.
Both major bleeding (2.4% vs 1.8%) and life-threatening bleed-
ing (1.4% vs 0.9%) were more frequent among patients receiving
prasugrel. Overall mortality did not differ between the 2 groups.
Because of the increased risk of bleeding, the FDA issued a black
box warning noting that this drug should not be used in patients
with active bleeding or a propensity to bleed, a history of TIA
or stroke, a body weight of less than 60 kg, or concomitant use
of anticoagulants. Prasugrel is not recommended for patients 75
years or older, except in high-risk situations, and prasugrel ther-
apy should be stopped at least 7 days before any surgery.
Ticagrelor
Ticagrelor is an oral ADP P2Y12 receptor inhibitor that is revers-
ible, unlike other thienopyridines. Compared with clopidogrel,
ticagrelor provides more rapid and complete receptor inhibi-
tion without an increase in major hemorrhage. The FDA has not
approved ticagrelor.
• Ticlopidine, clopidogrel, and prasugrel all require in vivo hepatic
transformation from prodrug to active metabolite.
• The combination of clopidogrel and aspirin has become standard
therapy in acute coronary syndromes and after stent implantation.
• Genetic testing for clopidogrel nonmetabolizer alleles (CYP2C19)
may be considered if there is clinical suspicion of clopidogrel hypo-
responsiveness or when a good alternative therapy is available.
• PPIs may lower in vitro measures of platelet inhibition by thienopyri-
dine agents but do not appear to affect clinical outcomes.
489
• TTP may be a complication of therapy with clopidogrel or
ticlopidine.
• Prasugrel therapy inhibits platelets more rapidly and completely
than clopidogrel therapy.
• Prasugrel therapy reduces the occurrence of thrombotic cardiovas-
cular events but is associated with more bleeding complications.
Phosphodiesterase Inhibitors
Dipyridamole
Phosphodiesterase is the principal enzyme responsible for the
metabolism of cAMP to AMP for the regeneration of ATP. Use
of dipyridamole inhibits phosphodiesterase and leads to an intra-
cellular accumulation of cAMP, which in turn inhibits platelet
activation (Box 49.3). For the prevention of stroke, the FDA has
approved oral extended-release dipyridamole (200 mg) in com-
bination with aspirin (25 mg) given twice daily. Metabolism is
primarily by the liver. This drug should be used with caution in
patients with liver failure, renal failure, or congestive heart fail-
ure. Available data are either insufficient or do not support the
use of dipyridamole (alone or in combination with aspirin) for
the prevention or treatment of arterial thrombosis in acute MI,
unstable angina, atrial fibrillation, mechanical heart prosthesis,
or venous thrombosis.
The ESPS-1 trial randomly assigned patients with a recent
stroke, TIA, or RIND of “atherosclerotic origin” to receive either
dipyridamole (75 mg) plus aspirin (325 mg) or placebo 3 times
daily. After 24 months of follow-up, the occurrence of the pre-
determined end point (stroke or death from any cause) was sig-
nificantly less in the treatment arm (15.8%) compared with the
placebo arm (22.6%). This study was criticized for not using
aspirin alone in the comparison.
In the ESPS-2 trial, aspirin (25 mg) plus extended-release
dipyridamole (200 mg) given twice daily was compared with
aspirin alone (50 mg daily) or dipyridamole alone (400 mg daily)
in patients who had stroke or TIA within the preceding 3 months.
The primary end points were stroke or death from all causes (or
both). Compared with placebo, aspirin alone reduced stroke risk
by 18%, dipyridamole alone by 16%, and combination therapy
by 37%. Combined risk of stroke or death was reduced by 13%
with aspirin alone, 15% with dipyridamole alone, and 24% with
the combination. From these data, it was concluded that the com-
bination of aspirin and dipyridamole provides additive protective
effects against stroke and TIA that are comparable to those with
clopidogrel therapy (similar to findings in the PROFESS trial).
GpIIb-IIIa Inhibitors
The final common pathway in arterial thrombosis, platelet
aggregation, involves the binding of fibrinogen by activated
Box 49.3. Dipyridamole
Phosphodiesterase inhibitor
Situation with the greatest benefit-to-risk ratio:
stroke prophylaxis (when combined with aspirin)
Risk of major hemorrhage: 1.8% per year (when
combined with aspirin)
No proven benefit for venous thrombosis
prophylaxis
490 V Aorta and Peripheral Vascular Disease
GpIIb-IIIa receptors on adjacent platelets. Receptor blockade,
by inhibiting fibrinogen binding, prevents aggregation regardless
of the initiating agonist. Since there are several potential ago-
nists, this strategy is particularly attractive. Numerous natural
and synthetic peptides have been evaluated for their potential
for blocking this receptor. These inhibitors can be subdivided
into anti– GpIIb-IIIa monoclonal antibodies, viper venoms, RGD
peptides, and nonpeptide analogues.
Abciximab
Abciximab (ReoPro) is the chimeric Fab fragment of a murine
monoclonal antibody against the activated GpIIb-IIIa receptor
(Box 49.4). Infusion of this antibody produces prolonged and
extensive receptor blockade with marked inhibition of plate-
let aggregation and bleeding times. For successful treatment
with abciximab, the antibody must occupy most (90%) of the
GpIIb-IIIa receptors. Clinical efficacy has been documented in
unstable angina, with a significant reduction in the occurrence of
ischemic events and angiographic improvement of the coronary
lesions. In combination with rtPA for the treatment of acute MI,
abciximab improved patency rates. Three trials (EPIC, EPILOG,
and EPISTENT) that involved patients undergoing high-risk
angioplasty, comprise the largest clinical experience with this
drug. In the EPIC trial, 2,099 patients were randomly assigned
to receive abciximab or placebo before undergoing high-risk
PCI. Among patients who received abciximab, the rate of the
combined primary end point decreased 35% (primary end points
were death, nonfatal MI, surgical revascularization, subsequent
PCI, unplanned implantation of a coronary stent, or insertion
of an intra-aortic balloon pump for refractory ischemia). Initial
trials in which abciximab was given in conjunction with stand-
ard doses of heparin and aspirin resulted in significantly more
bleeding complications and transfusions.
In the subsequent EPILOG study, which included patients
undergoing urgent or elective PTCA, the concomitant heparin
dosage was decreased in an attempt to improve clinical safety.
Abciximab in combination with a reduced dose of heparin
resulted in a 68% decrease in 30-day mortality and MI rates;
bleeding complication rates were similar to those with placebo.
Six-month follow-up data from the EPIC trial showed that
treatment with abciximab decreased the rate of clinical restenosis.
The mechanism behind this finding is unclear but may be related
Box 49.4. Abciximab
Antiplatelet glycoprotein IIb/IIIa complex murine
monoclonal antibody
Situations with the greatest benefit-to-risk ratio:
High-risk angioplasty and stenting
Percutaneous coronary intervention for acute
myocardial infarction
Not recommended as a fibrinolytic adjunctive
agent
Risk of major hemorrhage: 1.4%–3.3%
Thrombocytopenia limits its use in 1%–2% of
patients
Anti-abciximab antibody, which forms in a
minority of patients, is of unclear significance
to antibody cross-reactivity with a second receptor, αvβ3, the vit-
ronectin receptor. In addition to providing a scaffold for coagula-
tion factor–activating complex assembly, platelet storage granules
contain rich stores of growth factors. By limiting platelet content
at the site of injury, this agent may decrease the release of plate-
let growth factor and the generation of thrombin, both of which
have been implicated in the pathogenesis of restenosis. Subgroup
analyses of abciximab trials suggested that this benefit was par-
ticularly evident among diabetic patients undergoing PCI.
The TARGET trial assessed this hypothesis prospectively by
randomly assigning patients undergoing elective PCI to receive
either tirofiban or abciximab. Although diabetic patients had
more cardiovascular events than nondiabetic patients, compa-
rable event rates occurred among diabetic patients assigned to
either agent. These findings suggest that the non–GpIIb-IIIa
properties of abciximab do not translate into discernible long-
term clinical benefit for diabetic patients.
In the EPISTENT trial, patients were randomly assigned
to coronary stenting plus placebo, stenting plus abciximab, or
angioplasty plus abciximab. The rate of the primary end point
(combined death, MI, or need for urgent revascularization) was
lowest among the patients receiving abciximab, and the rate of
major bleeding complications was similar between groups.
The CAPTURE study assessed the efficacy of abciximab in
patients with refractory unstable angina undergoing PCI. Patients
were randomly assigned to receive intravenous abciximab or
placebo before PTCA. By 30 days, there was a significant reduc-
tion in the rates of ischemic events in the abciximab group, but
after 6 months of follow-up the 2 groups had equivalent rates of
death, MI, or need for another intervention.
In the larger CADILLAC study, STEMI patients were ran-
domly assigned to PTCA alone, PTCA plus abciximab, stent-
ing alone, or stenting plus abciximab therapy. At 6 months, there
were no significant differences among the groups in the rates of
death, stroke, or reinfarction. The rates of target vessel revas-
cularization were significantly less among patients undergoing
stenting with abciximab (5.2%) than among patients undergoing
PTCA alone (15.7%), but the rate of angiographically established
restenosis was nearly 50% lower for the patients receiving a stent.
The rates of reocclusion of the infarct-related artery were 11.3%
after PTCA and 5.7% after stenting; both results were independ-
ent of abciximab use.
The GUSTO IV-ACS trial randomly assigned 7,825 MI
patients to receive placebo, abciximab as a bolus plus 24-hour
infusion, or abciximab as a bolus plus 48-hour infusion in addi-
tion to standard therapy of heparin and aspirin. At 30 days, the
rate of the primary end point (composite of death and MI) was
similar between groups. Abciximab therapy, however, was asso-
ciated with a significantly higher risk of major hemorrhage (1.0%
vs 0.2%).
To address the hypothesis that the combination of GpIIb-
IIIa inhibitor and fibrinolytic therapy improves reperfusion,
the GUSTO-V investigators randomly assigned 16,588 STEMI
patients to receive 1) standard doses of reteplase or 2) half doses
of reteplase plus full doses of abciximab. At 30 days, there was
no difference in mortality between the 2 groups, but patients
who received reteplase alone had a less frequent need for urgent
revascularization, fewer major nonfatal ischemic complications
of MI, and less frequent major hemorrhage. In a similar study,
the ASSENT-3 investigators randomly assigned 6,095 patients
with acute MI to 1 of 3 regimens: full doses of tenecteplase
plus enoxaparin; half doses of tenecteplase plus abciximab
plus unfractionated heparin; or full doses of tenecteplase plus
 49 Treatment and Prevention of Arterial Thrombosis
unfractionated heparin. The efficacy outcomes were nearly iden-
tical for the enoxaparin and abciximab groups. On the basis of
these combined data, abciximab should not be used as an adjunc-
tive agent to fibrinolytic therapy.
Thrombocytopenia is a primary side effect of abciximab
therapy and may contribute to the hemorrhagic complications of
this drug. It is therefore recommended that a platelet count be
assessed 2 to 4 hours after initiating therapy. The platelet count
decrement is often mild, with a nadir above 75 × 103/μL. Marked
thrombocytopenia may require abrupt cessation of abciximab
therapy and platelet transfusion in 1% to 2% of patients. True
thrombocytopenia must be distinguished from pseudothrom-
bocytopenia, which results from platelet agglutination in vitro
when a blood sample is in EDTA. This accounts for one-third to
two-thirds of thrombocytopenia cases occurring after abciximab
administration. Distinguishing thrombocytopenia from pseudo-
thrombocytopenia is clinically important and must be completed
rapidly so as not to interrupt the delivery of antiplatelet therapy.
This can be accomplished by simply repeating the platelet count
with the use of citrate anticoagulant.
An additional potential side effect is the development of anti-
bodies to abciximab, which occurred in approximately 6% of
patients enrolled in the EPIC trial. These antibodies form against
the variable region of abciximab. The potential risk of abciximab
reexposure in these patients is unclear but could include anaphy-
laxis, drug neutralization, or thrombocytopenia.
• Abciximab should not be used as an adjunctive agent to fibrino-
lytic therapy.
• Thrombocytopenia is a primary side effect of abciximab therapy.
• The potential risk of abciximab reexposure in these patients with
antibody formation is unclear but could include anaphylaxis or
drug neutralization.
Tirofiban
The GpIIb-IIIa receptor recognition of the RGD amino acid
sequence on the fibrinogen molecule permits binding of the
receptor to this protein. By occupying these receptor sites, syn-
thetic RGD peptides competitively inhibit fibrinogen interaction
with the receptor and thereby prevent aggregation. Tirofiban
(Aggrastat) is a nonpeptide derivative of tyrosine, which selec-
tively inhibits the RGD binding site of the platelet GpIIb-IIIa
receptor (Box 49.5). Tirofiban is delivered intravenously and
has a rapid onset of action (5 minutes). In most patients, platelet
aggregation returns to pretreatment levels within 4 to 8 hours
after use of tirofiban is discontinued. The clearance of tirofiban
is primarily renal, and therefore a decreased dosage should be
considered for patients who have severe renal impairment (creat-
inine clearance <30 mL/min).
The RESTORE investigators assessed the efficacy of tirofiban
in decreasing the rates of the composite end points, including
Box 49.5. Tirofiban
Synthetic platelet glycoprotein IIb/IIIa inhibitor
Situation with the greatest benefit-to-risk ratio:
non–ST-segment myocardial infarction
Not recommended as a fibrinolytic adjunctive
agent
Risk of major hemorrhage: 0.4%–1.4%
491
all-cause mortality and recurrent ischemic events, among patients
undergoing PTCA. Within 2 days after the intervention, the
tirofiban group had a significant 38% relative risk reduction for
the composite end points; by 7 days, the relative risk reduction
was 27%; and at 30 days, it was 16%, still in favor of tirofiban
although not statistically significant. The rate of major bleed-
ing was not different between the 2 groups. In general, the rate
of major bleeding with tirofiban varied from 1.4% to 2.2% and
appeared to be higher for female and elderly patients.
The PRISM and PRISM-PLUS trials compared tirofiban with
heparin in nonSTEMI patients undergoing deferred coronary
angiography. At the completion of drug infusion, the rate of the
primary end point (death, MI, or refractory ischemia) was less
among patients receiving tirofiban than among patients receiv-
ing heparin (3.8% vs 5.6%); this benefit was maintained at 30
days. The incidence of major hemorrhage was the same for both
groups.
The PRISM-PLUS trial assigned patients to receive tirofiban
alone, tirofiban plus heparin, or heparin alone. At interim analy-
sis, the tirofiban-alone arm was discontinued by the Data Safety
and Monitoring Board owing to excess mortality. At the final
analysis, tirofiban plus heparin was associated with a significant
reduction in the rate of the primary composite end point (death,
MI, or refractory ischemia at 7 days) compared with heparin
alone (12.9% vs 17.9%). This benefit was maintained at 30 days
and 6 months. The rate of major bleeding was similar between
the tirofiban-plus-heparin group (1.4%) and the heparin-alone
group (0.8%).
To compare 2 different GpIIb-IIIa inhibitors in unstable
angina and MI, PCI patients were randomly assigned to receive
either tirofiban or abciximab before angiography. The primary
end point (composite of death, nonfatal MI, or urgent target
vessel revascularization at 30 days) occurred more frequently
among patients receiving tirofiban compared with abciximab
(7.6% vs 6.0%). The incidence of MI was significantly higher in
the tirofiban group compared with the abciximab group (6.9%
vs 5.4%). Rates of major bleeding complications were similar
between the 2 groups.
Eptifibatide
Eptifibatide (Integrilin) is a cyclic heptapeptide containing a
KGD sequence (Box 49.6). The KGD sequence provides more
inhibition specificity for the GpIIb-IIIa receptor compared with
other integrins containing the RGD sequence. Plasma half-life
is 2.5 hours, and the antiplatelet effects persist for 4 hours after
infusion is discontinued. Eptifibatide is cleared primarily by
the kidneys, and its use should be avoided in patients with renal
impairment (creatinine clearance <30 mL/min).
The IMPACT-II trial compared 2 doses of eptifibatide with
placebo in patients undergoing elective, urgent, or emergent
coronary intervention. With the smaller dose, the rates of early
Box 49.6. Eptifibatide
Synthetic platelet glycoprotein IIb/IIIa inhibitor
Situation with the greatest benefit-to-risk ratio:
non–ST-segment myocardial infarction
Not recommended as a fibrinolytic adjunctive
agent
Risk of major hemorrhage: 1.1%–3.0%
492 V Aorta and Peripheral Vascular Disease
abrupt closure and ischemic events at 30 days were less than
with placebo. This effect was not statistically significant with the
higher dose.
The ESPRIT trial assigned patients undergoing coronary
stenting to receive eptifibatide or placebo in addition to aspirin,
heparin, and a thienopyridine. The rate of the primary end point
(the composite of death, MI, urgent target vessel revasculariza-
tion, and thrombotic bailout GpIIb-IIIa inhibitor therapy within
48 hours after randomization) was lower in the treatment arm
than in the placebo arm (6.6% vs 10.5%). These benefits were
maintained at the 6-month follow-up.
The PURSUIT investigators tested the hypothesis that epti-
fibatide would be more effective than heparin and aspirin in
reducing adverse outcomes among patients with NSTEMI acute
coronary syndromes. In this study, patients received eptifibatide
or placebo, in addition to standard acute coronary syndrome ther-
apy. The primary end point, a composite of death and nonfatal
MI occurring in up to 30 days, had a lower rate in the eptifibatide
group than in the placebo group (14.2% vs 15.7%). The bene-
fit was apparent by 96 hours and persisted through 30 days. In
subgroup analysis, however, the beneficial effects of eptifibatide
were not evident in women.
Eptifibatide has also been assessed as an adjunct to thrombol-
ysis in MI. The INTRO-AMI investigators tested the hypothesis
that eptifibatide and a reduced dose of tPA enhance infarct artery
patency compared with a standard dose of tPA alone. Although
the early patency rates were improved for patients receiving epti-
fibatide, the incidence of death, reinfarction, or revascularization
at 30 days was similar between treatment groups.
• Tirofiban and eptifibatide are cleared primarily by the kidney, so
their use should be avoided in patients with renal impairment (cre-
atinine clearance <30 mL/min).
Thrombin Inhibitors
Thrombin, the most potent known platelet agonist, is directly
involved in both aggregation and thrombus propagation.
Thrombin has 2 sites that are important for enzymatic activity,
the anion-binding exosite, which provides substrate recognition
and interaction, and the catalytic active site, which cleaves the
substrate. DTIs bind thrombin and block its interaction with
substrates. Specific antidotes are not available to neutralize
these compounds. The parenteral thrombin inhibitors have been
licensed in the United States for limited indications: hirudin and
argatroban are approved for treatment of patients with HIT, and
bivalirudin is licensed as an alternative to heparin for patients
undergoing PCI.
Lepirudin
Lepirudin (a recombinant form of hirudin) is a specific and irre-
versible DTI (Box 49.7). Its amino-terminal domain interacts
with the active site of thrombin and its carboxyl-terminal tail
binds to exosite 1. Hirudin is renally excreted, and the dose must
be adjusted according to the creatinine clearance in patients with
renal impairment. Clot-bound thrombin, an important throm-
botic risk factor inaccessible to antithrombin-heparin, is effec-
tively inhibited by hirudin. Hirudin therapy is monitored with a
standard aPTT assay. The recommended aPTT target range for
patients receiving this drug is 1.5 to 2.5 times the baseline value.
Lepirudin should not be given as adjunctive therapy for patients
with acute MI unless there is concomitant HIT. Approximately
40% of patients treated with lepirudin form antibodies to the
Box 49.7. Lepirudin
A direct thrombin inhibitor that is a recombinant
form of hirudin
Situations with the greatest benefit-to-risk ratio:
Heparin-induced thrombocytopenia (FDA
approved)
Non–ST-segment myocardial infarction (not FDA
approved)
aPTT target range: 1.5–2.5 times baseline values
In approximately 40% of patients, antibodies
develop that reduce renal clearance
Risk of major hemorrhage: 1.2%–3.3%
Abbreviations: aPTT, activated partial thromboplastin time; FDA,
US Food and Drug Administration.
drug. These antibodies do not block the thrombin-inhibitory
activity of lepirudin but rather delay renal excretion. In vivo drug
concentrations may become dangerously high in these patients
and precipitate major hemorrhage. For this reason, the aPTT
must be followed closely with careful dose adjustment to keep
the aPTT within the therapeutic range.
• Lepirudin should not be given as adjunctive therapy for patients
with acute MI unless there is concomitant HIT.
Bivalirudin
Bivalirudin, a 20–amino acid synthetic polypeptide analogue
of hirudin, has a terminal half-life of 25 minutes after intrave-
nous injection, and only a fraction is excreted through the kid-
neys (Box 49.8). It has been evaluated in patients undergoing
PCI and as an adjunct to streptokinase in patients with STEMI.
Bivalirudin was approved for use in high-risk patients undergo-
ing PCI on the basis of data from several randomized clinical
trials; however, higher rates of severe bleeding complications
have been reported. For this reason, bivalirudin is not recom-
mended as routine adjunctive therapy with fibrinolytic agents for
the treatment of acute coronary syndromes.
• Bivalirudin is not FDA approved for use in patients with HIT.
Box 49.8. Bivalirudin
A direct thrombin inhibitor that is an engineered
hirudin
Situations with the greatest benefit-to-risk ratio:
PCI (FDA approval, with provisional glycoprotein IIb/
IIIa inhibitors)
PCI for unstable angina
Not recommended for routine use with fibrinolytic
agents
Risk of major hemorrhage: 3.7%–5.4%
Abbreviations: FDA, US Food and Drug Administration; PCI,
percutaneous coronary intervention.
 49 Treatment and Prevention of Arterial Thrombosis
Argatroban
Argatroban (Box 49.9) is a peptidomimetic arginine derivative
that binds noncovalently to the active site of thrombin and forms
a reversible complex. The drug is metabolized in the liver in a
process that generates several active intermediates. Although
this drug is safely used in patients with renal insufficiency, it
should be used very cautiously (if at all) in patients with hepatic
insufficiency. Relatively small trials have evaluated argatroban
for treatment of unstable angina, as an adjunct to thrombolysis
and as an alternative to heparin in patients undergoing coronary
angioplasty; none has shown definitive advantages of argatroban
over heparin.
Ximelagatran
Ximelagatran was the first orally administered DTI. Although
the efficacy of this drug was demonstrated for several indications,
hepatic toxicity prevented FDA approval, and further investiga-
tion and manufacturing of this drug were discontinued.
Dabigatran Etexilate
Dabigatran etexilate is an oral DTE that effectively inhibits
both free and clot-bound thrombin. This drug is a prodrug that
requires hepatic activation upon ingestion for full activity. The
onset of action is rapid, occurring 2 to 3 hours after ingestion. The
half-life of the active metabolite is approximately 15 hours, and
it is eliminated primarily through the kidney (80%). Phase III
randomized trial data exist for several indications, including
venous thromboembolism prevention (RE-NOVATE I and II,
RE-MODEL, RE-MOBILIZE), venous thromboembolism treat-
ment (RE-COVER), and atrial fibrillation (RE-LY). In each of
these large trials, dabigatran compared favorably against war-
farin (or LMWH; RE-NOVATE, RE-MODEL) with similar or
lower thromboembolic rates and similar or lower bleeding rates
depending on the drug dosage. In all of these trials, dabigatran
results for liver toxicity, defined as alanine aminotransferase lev-
els exceeding 3 times the upper limit of normal, compared favor-
ably with enoxaparin or warfarin. Dabigatran is now approved
by the FDA for stroke and peripheral embolism prevention in
nonvalvular atrial fibrillation. There are currently no available
antidotes for dabigatran.
Rivaroxaban
Rivaroxaban is an oral direct factor Xa inhibitor that is also in
phase III trial assessment. This drug is rapidly absorbed and has
Box 49.9. Argatroban
Synthetic direct thrombin inhibitor
Situations with the greatest benefit-to-risk ratio:
HIT (FDA approved)
Acute coronary syndrome (in patients with HIT or at
risk of HIT)
Not recommended for routine use with fibrinolytic
agents
Risk of major hemorrhage: 0.5%–2.3%
Abbreviations: FDA, US Food and Drug Administration; HIT,
heparin-induced thrombocytopenia.
493
a rapid onset of action, achieving full anticoagulant levels 1 to 4
hours after oral ingestion. The half-life is between 5 and 9 hours,
and it is hepatically metabolized to the inactive compound. Like
dabigatran, this drug has also been assessed for several indica-
tions in phase III randomized trials. In the RECORD 1-4 tri-
als, rivaroxaban was superior to enoxaparin for the prevention
of venous thrombosis after total hip and total knee replacement
surgery. In these studies, rates of venous thromboembolism were
lower with rivaroxaban without increased hemorrhagic side
effects. Treatment trials (EINSTEIN-DVT, EINSTEIN-PE) indi-
cate favorable comparison with LMWH and warfarin with very
low bleeding rates. The ROCKET AF investigators randomly
assigned 14,264 patients with nonvalvular atrial fibrillation to
receive rivaroxaban (20 mg daily) or warfarin in a non-inferiority
trial design. The primary end point (stroke or systemic embo-
lism) was non-inferior for rivaroxaban (1.7%/year) compared
with warfarin (2.2%/y). Major and non-major clinically relevant
bleeding were also similar for rivaroxaban (14.9%/y) and war-
farin (14.5%/y) groups. Intracranial (0.5% vs 0.7%) and fatal
hemorrhage (0.2% vs 0.5%) were lower in rivaroxaban-treated
patients. Major gastrointestinal bleeding was more common in
the rivaroxaban group (3.2% vs 2.2%). On the basis of these tri-
als, the FDA approved rivaroxaban for the indications of venous
thromboembolism prophylaxis after knee or hip replacement and
stroke and systemic embolism prophylaxis in nonvalvular atrial
fibrillation.
Apixaban
Apixaban is also an oral direct factor Xa inhibitor currently
undergoing trial assessment for several clinical indications. This
drug has rapid onset of action within 1 hour of ingestion and
has a half-life of 13 hours. This drug is metabolized by the liver
(CYP3A4) and is at least partially excreted through the kidneys.
In dose-finding studies, apixaban compared favorably with
LMWH or warfarin for the prevention of venous thromboembo-
lism after total knee replacement (APROPOS) and also for the
treatment of DVT (Botticelli-DVT). Phase III trials include the
efficacy and safety of apixaban in the setting of atrial fibrillation
(ARISTOTLE, AVERROES) and acute coronary syndromes
(APPRAISE-2). These studies have shown superiority of apixa-
ban both in terms of efficacy and safety compared with warfarin
(ARISTOTLE) in the setting of atrial fibrillation with similar
rates of major bleeding compared with aspirin (AVERROES). In
the setting of acute coronary syndromes (APPRAISE-2), apixa-
ban increased major bleeding events without improving ischemic
events.
• DTIs reduce the rate of reinfarction but not mortality among
patients with STEMI treated with fibrinolytic therapy. The major
benefit of direct thrombin inhibitors appears to be for patients
undergoing PCI, particularly after STEMI.
• Novel DTIs, such as dabigatran, compare favorably with warfarin
or LMWH in safety and efficacy.
• Novel DTIs such as rivaroxaban and apixaban also compare favor-
ably with traditional anticoagulants.
Heparinoids
Unfractionated Heparin
A mainstay of therapy for any thrombotic disorder, UFH has anti-
coagulant properties through its interaction with antithrombin
III (Box 49.10 and Figure 49.2). UFH induces a conformational
change in antithrombin III, enhancing its affinity for thrombin
494 V Aorta and Peripheral Vascular Disease
Box 49.10. Heparin
Inhibitory action through activation of
antithrombin III
Situations with the greatest benefit-to-risk ratio:
Acute coronary syndromes
Acute venous thromboembolism
Risk of major hemorrhage: 5%–7%
1,000-fold. The thrombin-antithrombin III complex is essentially
irreversible and inhibits thrombin as well as other factors of the
cascade (factors IXa, Xa, and Xia). UFH is a highly negatively
charged proteoglycan extracted from either porcine or bovine
intestinal mucosa. Nonspecific binding to various cells and plasma
proteins neutralizes the anticoagulant activity. UFH preparations
are heterogeneous, containing heparin chains of various lengths
with molecular weights from 5,000 to 50,000. Unpredictable vol-
ume of distribution and elimination kinetics require that therapy
with UFH be strictly monitored by serial measurements of aPTT
for both safety and efficacy. A weight-adjusted nomogram shows
that within 24 hours the therapeutic range is reached in only
82% of patients and they have a 5% to 7% risk of major hemor-
rhage. Despite these imperfections, the efficacy of UFH in acute
arterial occlusive syndromes has been well established. In acute
coronary syndromes, UFH plus aspirin reduces the risk of death
or MI compared with aspirin alone. In the prethrombolytic era,
UFH reduced the rates of reinfarction and mortality after acute
MI. As an adjuvant to thrombolytic therapy, UFH improves early
patency and may improve survival.
Low-Molecular-Weight Heparin
LMWH is produced by depolymerizing UFH either chemically or
enzymatically (Box 49.11, Figure 49.3). This yields a preparation
ATIII
ATIII activation
Heparin
binding
site
Reactive site
Heparin
Figure 49.2. Heparin Interaction With ATIII. Heparin binds ATIII through the interaction of a specific pentasaccharide sequence within the
molecule. This interaction transforms ATIII into an active inhibitor. Bound to ATIII, heparin forms a scaffold that promotes the interaction between
thrombin and ATIII. After the complex has been formed, heparin is free to participate in another round of inhibition. ATIII indicates antithrombin
III; TAT, thrombin-antithrombin III.
of small heparin fragments that have more uniform molecular
weights (4,000–6,000). LMWH requires antithrombin for activ-
ity, yet unlike its parent compound, it has a higher specificity for
factor Xa than for thrombin. Because of its uniformity of size
and less negative charge, it has much less nonspecific binding,
improved bioavailability, and more predictable pharmacology.
LMWH does not affect aPTT, and, when based on patient body
weight, dosing is safe and effective. Several potential anticoag-
ulant advantages also favor LMWH over UFH. PF4, secreted by
activated platelets, blocks the interaction between antithrombin
and heparin but not LMWH. In addition, factor Xa bound to the
platelet surface within the prothrombin activation complex is
inaccessible to the antithrombin-heparin complex, whereas the
LMWH-antithrombin complex can inhibit factor Xa under these
circumstances.
By far, the most experience with LMWH has been with DVT
prophylaxis and venous thromboembolism treatment. Studies
have provided efficacy data for the treatment of DVT, pulmonary
embolism, and arterial thrombosis. LMWH is at least as effec-
tive as standard heparin in the management of arterial occlusive
syndromes. The dose of LMWH is weight adjusted and given
once or twice daily subcutaneously depending on the formula-
tion. It does not require laboratory monitoring.
Heparin-Induced Thrombocytopenia
HIT is an antibody-mediated, adverse effect of heparin that is
important because of its strong association with venous and arte-
rial thrombosis (HITT). The neoepitopes recognized by HIT
antibodies are located on PF4 and are formed when PF4 binds
to heparin. HIT is a clinicopathologic syndrome, and diagnosis
requires both clinical and serologic assessment. Consequently,
HIT antibody seroconversion without thrombocytopenia or other
clinical sequelae is not considered HIT. In patients receiving
heparin, HIT should be suspected if the platelet count decreases
to less than 100 × 103/μL or decreases by more than 40% of the
baseline platelet count. This occurs in 1.3% of patients receiving
Thrombin TAT
A
inhibition complex
Thrombin
Thrombin
Heparin
 49 Treatment and Prevention of Arterial Thrombosis
Box 49.11. Low-Molecular-Weight Heparin
Inhibitory action through activation of
antithrombin III
Situations with the greatest benefit-to-risk ratio:
Acute coronary syndromes
Acute venous thromboembolism
Risk of major hemorrhage: 3%–5%
Weight-adjusted dosing
Renal excretion
No monitoring necessary
therapeutic doses of porcine heparin. Thrombocytopenia may
occur 3 to 15 days after the initiation of heparin therapy. In
patients with previous exposure, however, platelet counts can
begin to decrease within hours of treatment. HIT is more com-
mon with bovine heparin and less common with prophylactic
doses or LMWH preparations. The incidence of HITT ranges
from 0.2% to 20%; thrombosis may involve the arterial or venous
circulation and results in mortality rates as high as 30%.
Current recommendations of the ACCP Conference on
Antithrombotic and Thrombolytic Therapy for the recognition,
treatment, and prevention of HIT include platelet count moni-
toring (at least every other day) and, for patients with strongly
suspected (or confirmed) HIT or HITT, use of an alternative anti-
coagulant, such as lepirudin, argatroban, or bivalirudin. Patients
with strongly suspected (or confirmed) HIT should have rou-
tine ultrasonography of the lower limb veins for investigation
of DVT. All heparin administration must be discontinued and
ATIII
Heparin
binding
site
Reactive site
Fragmentation
LMWH
Figure 49.3. LMWH Interaction With ATIII. The enzymatic or chemical fragmentation of heparin yields fragments of uniform size, LMWH.
LMWH also binds ATIII through the interaction of a specific pentasaccharide sequence. This interaction transforms ATIII into an active inhibitor.
Unlike unfractionated heparin, LMWH cannot form a scaffold with thrombin; therefore, the specificity of ATIII is for factor Xa. ATIII indicates
antithrombin III; LMWH, low-molecular-weight heparin.
495
DTI therapy initiated. Vitamin K antagonist (coumarin) therapy
should begin with low maintenance doses only after the platelet
count has recovered. Administration of warfarin and DTI should
be overlapped for a minimum of 5 days or until the INR is ther-
apeutic (whichever is longer) before use of the DTI is stopped
(DTI therapy increases the INR). With argatroban therapy, the
therapeutic INR target is doubled to account for this interaction.
When the therapeutic target is achieved (INR 4–5, with an ulti-
mate target of 2.0–3.0), use of the argatroban can be stopped
and the INR repeated in 4 hours to ensure a therapeutic effect.
Warfarin therapy should be continued for 3 months to ensure res-
olution of the heparin-PF4 antibodies. These antibodies typically
circulate for 100 days.
For patients with a remote history of HIT who require car-
diac surgery, heparin can be given briefly for cardiopulmonary
bypass without excessive risk to the patient if the heparin-PF4
antibody ELISA is negative. When the surgery is completed,
heparin administration should be promptly discontinued. Owing
to significant cross-reactivity, LMWH is not an acceptable alter-
native in these patients. Fondaparinux, which does not bind to
PF4, should not cause HIT.
• HIT antibody seroconversion without thrombocytopenia or other
clinical sequelae is not considered HIT.
• Fondaparinux, which does not bind to PF4, should not cause HIT.
Fondaparinux
Fondaparinux can be thought of as ultra LMWH (Box 49.12).
This drug is a synthetic pentasaccharide analogue specific for the
sequence necessary for antithrombin III activation (Figure 49.4).
Fondaparinux is primarily excreted renally, so it should not be
used in patients with severe renal impairment (creatinine clear-
ance <30 mL/min). Fondaparinux therapy does not require
ATIII Thrombin
activation inhibition
Factor Xa
Factor Xa
Heparin
496 V Aorta and Peripheral Vascular Disease
Box 49.12. Fondaparinux
Synthetic pentasaccharide with actions through
antithrombin III
Situations with the greatest benefit-to-risk ratio:
ST-segment elevation myocardial infarction
Unstable angina
Non–ST-segment elevation myocardial infarction
Acute venous thromboembolism
Venous thromboembolism prophylaxis
Risk of major hemorrhage: 2%–3%
Renal excretion with long half-life (17–21 h)
No monitoring necessary
monitoring, and indeed there are no standardized anti-factor
Xa measures to determine either safety or efficacy. Routine
coagulation tests such as prothrombin time and aPTT are rel-
atively insensitive measures and are unsuitable for monitoring.
Because of its very low molecular weight and essentially neutral
net charge, this drug does not bind significantly to other plasma
proteins and specifically does not bind PF4. Therefore, the
risk of HIT is negligible. The net neutral charge, however, also
means that fondaparinux is not inhibited by protamine. There
is no specific antidote for bleeding while patients receive this
drug. Recombinant factor VIIa has been suggested as a possi-
ble therapy if major bleeding complicates fondaparinux therapy.
This therapy should be considered only for patients with life-
threatening bleeding and only as a last resort.
Fondaparinux has been evaluated for prevention and treat-
ment of venous thromboembolism and for treatment of arterial
thrombosis. It is currently approved for DVT prophylaxis for hip
fracture, hip replacement, or knee replacement surgery.
ATIII
Heparin
binding
site
Reactive site
Fondaparinux
Figure 49.4. Pentasaccharide Interaction With ATIII. A specific pentasaccharide sequence is required for ATIII activation, and new agents
exploiting this requirement have been introduced. These synthetic pentasaccharide molecules cannot form a scaffold with thrombin; therefore, the
specificity is for factor Xa inhibition. ATIII indicates antithrombin III.
The OASIS-5 trial was designed to compare the efficacy of
fondaparinux with that of enoxaparin in patients who had unsta-
ble angina or NSTEMI. The primary outcome events were sim-
ilar in the 2 groups; however, the rate of major bleeding was
significantly lower with fondaparinux than with enoxaparin.
Furthermore, mortality at 30 days and 180 days was signifi-
cantly less with fondaparinux therapy. The OASIS-6 trial dem-
onstrated similar efficacy of fondaparinux in patients who had
acute STEMI.
• Unlike UFH or LMWH, fondaparinux is not inhibited by
protamine.
• There is no specific antidote for bleeding while patients receive
fondaparinux.
Warfarin
Warfarin blocks the hepatic carboxylation of vitamin
K–dependent coagulation factors, thus inhibiting the activation
of the proenzyme to the enzyme (Box 49.13 and Figure 49.5).
Carboxylation is required for the calcium binding and the shape
reconfiguration that are necessary for incorporating the protein
into activation complexes on the phospholipid bilayer. With either
the inhibition of carboxylation or calcium sequestration (with cit-
rate, EDTA, etc), coagulation factor activation is brought to a
standstill. Relevant vitamin K–dependent proteins include both
procoagulant (factors II [prothrombin], VII, IX, and X) and anti-
coagulant (protein C and protein S) proteins. Each of these fac-
tors has considerably different terminal half-lives in vivo (Figure
49.6). As the use of warfarin is initiated, there is a theoretical
hypercoagulable state when protein C stores are depleted and
prothrombin (factor II) levels are normal. The INR is very sen-
sitive to factor VII depletion. The INR may increase, therefore,
despite adequate depletion of factor X and prothrombin levels.
For this reason, warfarin and heparin are given concurrently for
whichever is longer: 5 days or attainment of a therapeutic INR.
The clinical efficacy of warfarin in venous thromboembo-
lism and atrial fibrillation is universally acknowledged. Its role
ATIII Factor Xa
activation inhibition
Factor Xa
Factor Xa
 49 Treatment and Prevention of Arterial Thrombosis 497

• The role of warfarin anticoagulation in the secondary prevention

Box 49.13. Warfarin
Inhibitory action through blockade of hepatic
vitamin K–dependent γ-carboxylation
Situations with the greatest benefit-to-risk ratio:
Atrial fibrillation
Mechanical heart valve prophylaxis
Acute venous thromboembolism
Venous thrombosis prophylaxis
Risk of major hemorrhage: 1%–8% per year
in the treatment of acute and chronic coronary arterial occlusive
syndromes is less clear. In the prethrombolytic era, several post-
MI studies documented a significant reduction in the rates of
recurrent MI, stroke, and mortality among patients treated with
warfarin anticoagulation. The ASPECT trial randomly assigned
patients with recent MI to receive placebo or warfarin with an
INR goal of 2.8 to 4.8. At 37 months of follow-up, there was
a statistically significant reduction in the rate of recurrent MI
and stroke; however, no mortality benefit was shown between the
2 groups. The ATACS study compared warfarin plus aspirin to
aspirin alone in 214 patients with unstable angina or non–Q wave
MI. At the end of the trial, a trend favored the warfarin-plus-
aspirin group in decreasing the rate of recurrent angina, MI, or
death. The CARS investigators hypothesized that the addition of
a small, fixed dose of warfarin to aspirin would add the benefit of
an antithrombotic agent to an antiplatelet agent without increas-
ing either the risk of bleeding or the complexity of the treatment
regimen. The study was stopped prematurely by the data and
safety monitoring committee on the basis of similar efficacy of
treatment strategies.
of acute coronary syndromes is unclear, and warfarin anticoagula-
tion is rarely used now.
• In patients who have a large anterior wall MI, warfarin may be
used to reduce the risk of cardioembolic events.
• In many trials, warfarin provided effective prophylaxis against
stroke in atrial fibrillation, recurrent venous thromboembolism,
and both cardioembolic and valvular thrombotic complications in
patients with a mechanical heart valve prosthesis.
Bleeding complications have limited the enthusiasm for war-
farin use, especially in elderly patients. In the SPAF II study, at
an approximate INR of 2.0 to 4.5, major hemorrhage occurred
at a rate of 2.3% per year with warfarin compared with 1.1% per
year with aspirin. Age, increasing number of prescribed medica-
tions, and intensity of anticoagulation were independent risks for
bleeding. In patients younger than 75, the rate of major hemorrhage
was 1.7% per year, compared with 4.2% per year in older patients.
Other variables associated with excessive anticoagulation include
advanced malignancy, potentiating medications (such as antibiot-
ics or acetaminophen), anorexia, and diarrheal illnesses.
Fibrinolytic Agents
The mainstays of medical treatment for acute arterial thrombo-
sis are the plasminogen-activating agents: streptokinase, rtPA,
urokinase, and their derivatives. Unlike endogenous fibrinolysis,
which is marked by clot specificity, pharmacologic plasminogen
activation is indiscriminate in substrate preference and degrades
fibrin, fibrinogen, platelet receptors, and coagulation factors.
Streptokinase, unlike the endogenous activators, is not an enzyme
and therefore cannot activate plasminogen directly. Purified from
β-hemolytic streptococci, streptokinase promotes fibrinolysis by
inducing a conformational change in plasminogen, exposing the
enzymatic active site. The streptokinase-plasminogen complex
then cleaves a second plasminogen molecule to active plasmin.

γ-Carboxylation of coagulation proteins

Glu Gla
Vitamin K−
dependent proteins --N-C-C-- --N-C-C--
C C
C-COO OOC-C-COO

Vitamin K KH2 KO

Vitamin K K Vitamin K epoxide

reductase reductase

Warfarin

Figure 49.5. Mechanism of Warfarin Anticoagulant Effect. Posttranslational γ-carboxylation of vitamin K–dependent proteins occurs in the liver.
Warfarin inhibits vitamin K reduction, thus halting this process. Noncarboxylated proteins cannot bind calcium and cannot be activated and thus are
cleared more rapidly from the circulation. Warfarin anticoagulation represents a balance between this inhibition and exogenous vitamin K intake.
Exogenous vitamin K (KH2) is the reduced form. Gla indicates γ-carboxyglutamate residue; Glu, glutamate residue; KO, oxidized vitamin K.
498 V Aorta and Peripheral Vascular Disease

100 45% had primary intracranial hemorrhage, resulting in a 60%

mortality. Hemorrhagic conversion occurred in only 10% of the
Factor Activity, %

80 other stroke patients, with a 32% fatality rate. Advanced age,

prior cerebrovascular disease, and hypertension were significant
X PS predictors of intracranial hemorrhage. Severe or life-threatening
60
IX hemorrhage, defined as either intracranial hemorrhage or hemo-
Prothrombin dynamic compromise requiring treatment, occurred in 0.3% to
40
PC 0.5% of patients and was similar in all groups. Moderate hemor-
20 rhage occurred with an overall frequency of 5% to 6% and was
VII statistically less frequent among patients treated with rtPA.
0 The current recommendation is to administer any approved
0 1 2 3 4 5 fibrinolytic agent (streptokinase, anistreplase, alteplase,
Days reteplase, or tenecteplase) to patients who have had acute cor-
onary syndrome for less than 12 hours. For patients who have
Warfarin had symptoms for less than 6 hours, alteplase is preferred over
Heparin streptokinase. Alteplase, reteplase, or tenecteplase should be
used if patients have a known allergy or sensitivity to strepto-
kinase. Fibrinolytic therapy is contraindicated in patients with a
Figure 49.6. Dynamics of Vitamin K–Dependent Factor Depletion
With Warfarin Initiation. Effective anticoagulation is achieved when
history of intracranial hemorrhage, closed head trauma, or ische-
coagulation factor levels decrease to approximately 20%. Despite deple- mic stroke within the past 3 months.
tion of factors VII and IX, thrombosis may still occur if prothrombin
• Streptokinase, unlike the endogenous activators, is not an enzyme
and factor X levels are normal. PC indicates protein C; PS, protein S.
and therefore cannot activate plasminogen directly.

Abbreviations
Although streptokinase does not possess a fibrin binding site,
ACCP American College of Chest Physicians
this process is accelerated in the presence of fibrin and is some-
ADAMTS a disintegrin and metalloprotease with throm-
what clot specific. bospondin motif
APSAC is a complex of streptokinase already bound to plas- ADP adenosine diphosphate
minogen. This complex has increased specificity for fibrin and is AMP adenosine monophosphate
not inhibited by endogenous inhibitors of plasminogen systems. APSAC anisoylated plasminogen streptokinase activa-
APSAC has a plasma half-life 2 to 3 times longer than streptoki- tor complex
nase (70 minutes vs 25 minutes) and therefore can be given as a aPTT activated partial thromboplastin time
single bolus rather than as a prolonged infusion. Since these are ATP adenosine triphosphate
foreign proteins, repetitive streptokinase or APSAC use is hin- cAMP cyclic adenosine monophosphate
dered by neutralizing antibodies, which limit their efficacy. COX cyclooxygenase
CYP cytochrome P450
Staphylokinase, a protein produced by Staphylococcus
DTI direct thrombin inhibitor
aureus, activates plasminogen in much the same manner as DVT deep vein thrombosis
streptokinase. Staphylokinase is fibrin specific; inhibition by ELISA enzyme-linked immunosorbent assay
α2-antiplasmin is reduced. Although effective in the treatment of FDA US Food and Drug Administration
patients with acute MI, it has been associated with the induction GpIIb-IIIa glycoprotein IIb/IIIa complex
of high titers of neutralizing antibody formation. Like streptoki- HIT heparin-induced thrombocytopenia
nase, its use is therefore limited to a single infusion. HITT heparin-induced thrombocytopenia with throm-
Purified and recombinant forms of the endogenous plasmin- bosis
ogen activators have become widely used for local delivery and INR international normalized ratio
systemic fibrinolysis in various arterial and venous thrombotic KGD lysine-glycine-aspartic acid
LMWH low-molecular-weight heparin
disorders. These endogenous agents, including tPA, urokinase,
MI myocardial infarction
and scu-PA (also called prourokinase), have direct enzymatic NSTEMI non–ST-segment elevation myocardial infarction
activity toward plasminogen. The debate continues as to agent PAD peripheral artery disease
superiority and clinical indication; however, in vivo plasminogen PAI plasminogen activator inhibitor
activation is likely comparable, with minimal differences in clot PCI percutaneous coronary intervention
specificity between agents. In contrast to bacterial proteins, these PF4 platelet factor 4
agents are nonimmunogenic and therefore can be reinstituted. PGI2 prostacyclin
Although effective in thrombolysis, endogenous PAIs, such as PPI proton pump inhibitor
PAI-1, have the richest concentrations of platelet α-granules. PTCA percutaneous transluminal coronary angioplasty
This may explain the 20% to 50% failure rate of pharmacologic RGD arginine-glycine-aspartic acid
RIND reversible ischemic neurologic deficit
fibrinolytic therapy.
rtPA recombinant tissue plasminogen activator
One of the most devastating complications of thrombolytic scu-PA single-chain urokinase plasminogen activator
therapy is stroke. The overall risk in the GUSTO trial was 1.4%, STEMI ST-segment elevation myocardial infarction
but risk of stroke was lower for patients treated with streptoki- TIA transient ischemic attack
nase (1.19%) than with rtPA (1.55%); 45% of all strokes were tPA tissue plasminogen activator
fatal and 31% were disabling. Of the patients who had stroke, TTP thrombotic thrombocytopenic purpura
 49 Treatment and Prevention of Arterial Thrombosis 499

TXA2 thromboxane A2 GUSTO-V Global Use of Strategies to Open Occluded

UFH unfractionated heparin Coronary Arteries V
HOPE Heart Outcomes Prevention Evaluation
IMPACT-II Integrilin to Minimize Platelet Aggregation and
Coronary Thrombosis II
Names of Clinical Trials INTRO-AMI Integrilin and Low-Dose Thrombolysis in Acute
ACTIVE A Atrial Fibrillation Clopidogrel Trial With Myocardial Infarction
Irbesartan for Prevention of Vascular Events— ISAR-CHOICE Intracoronary Stenting and Antithrombotic
Aspirin Regimen: Choose Between 3 High Oral Doses
ACTIVE W Atrial Fibrillation Clopidogrel Trial With for Immediate Clopidogrel Effect
Irbesartan for Prevention of Vascular Events— ISAR-REACT Intracoronary Stenting and Antithrombotic
Warfarin Regimen: Rapid Early Action for Coronary
APPRAISE-2 Apixabam for Prevention of Acute Ischemic Treatment
Events 2 ISIS-2 Second International Study of Infarct Survival
APROPOS Apixibam PROphylaxis in Patients Undergoing MATCH Management of Atherothrombosis With
Orthopedic Surgery Clopidogrel in High-Risk Patients With Recent
ARISTOTLE Apixibam for Reduction in Stroke and Other Transient Ischemic Attacks or Ischemic Stroke
Thromboembolic Events in Atrial Fibrillation OASIS-5, -6 Organization to Assess Strategies in Acute
ASPECT Anticoagulants in the Secondary Prevention of Ischemic Syndromes 5, 6
Events in Coronary Thrombosis PCI-CLARITY Percutaneous Coronary Intervention—Clopi-
ASSENT-3 Assessment of the Safety and Efficacy of a New dogrel as Adjunctive Reperfusion Therapy
Thrombolytic Regimen 3 PCI-CURE Clopidogrel in Unstable Angina to Prevent
ATACS Antithrombotic Therapy in Acute Coronary Recurrent Events—Percutaneous Coronary
Syndromes Intervention
AVERROES Apixabam Versus Acetylsalicylic Acid [ASA] PRISM Platelet Receptor Inhibition for Ischemic
to Prevent Stroke in Atrial Fibrillation Patients Syndrome Management
Who Have Failed or Are Unstable for Vitamin PRISM-PLUS Platelet Receptor Inhibition in Ischemic Syn-
K Antagonist Therapy drome Management in Patients Limited by
CADILLAC Controlled Abciximab and Device Investigation Unstable Signs and Symptoms
to Lower Late Angioplasty Complications PROFESS Prevention Regimen for Effectively Avoiding
CAPRIE Clopidogrel Versus Aspirin in Patients at Risk Second Strokes
of Ischemic Events PURSUIT Platelet Glycoprotein IIb/IIIa in Unstable
CAPTURE Chimeric 7E3 Antiplatelet Therapy in Unstable Angina: Receptor Suppression Using Integrilin
Angina Refractory to Standard Treatment Therapy
CARS Coumadin Aspirin Reinfarction Study RECORD 1-4 Regulation of Coagulation in ORthopedic
CHARISMA Clopidogrel for High Atherothrombotic Risk Surgery to Prevent Deep Venous Thrombosis
and Ischemic Stabilization, Management, and and Pulmonary Embolism
Avoidance RE-COVER Efficacy and Safety of Dabigatran Compared
CLARITY Clopidogrel as Adjunctive Reperfusion Therapy to Warfarin for 6-Month Treatment of Acute
COMMIT Clopidogrel and Metoprolol in Myocardial Symptomatic Venous Thromboembolism
Infarction Trial RE-LY Randomized Evaluation of Long-term Antico-
CREDO Clopidogrel for the Reduction of Events During agulant Therapy
Observation RE-MOBILIZE The Oral Thrombin Inhibitor Dabigatran
CURE Clopidogrel in Unstable Angina to Prevent Etexilate vs the North American Enoxaparin
Recurrent Events Regimen for the Prevention of Venous
EINSTEIN-DVT Oral Direct Factor Xa Inhibitor Rivaroxaban in Thromboembolism After Knee Arthroplasty
Patients With Acute Symptomatic Deep-Vein Surgery
Thrombosis Without Symptomatic Pulmonary RE-MODEL Oral Dabigatran Etexilate vs Subcutaneous
Embolism Enoxaparin for the Prevention of Venous
EINSTEIN-PE Oral Direct Factor Xa Inhibitor Rivaroxaban in Thromboembolism After Total Knee
Patients With Acute Symptomatic Pulmonary Replacement
Embolism With or Without Symptomatic Deep- RE-NOVATE Dabigatran Etexilate Versus Enoxaparin for
Vein Thrombosis Prevention of Venous Thromboembolism After
EPIC Evaluation of IIb/IIIa Platelet Receptor Anta- Total Hip Replacement
gonist 7E3 in Preventing Ischemic Compli- REPLACE-2 Randomized Evaluation in PCI Linking
cations Angiomax to Reduced Clinical Events 2
EPILOG Evaluation of PTCA to Improve Long-term RESTORE Randomized Efficacy Study of Tirofiban for
Outcome by c7E3 GP IIb/IIIa Receptor Outcomes and Restenosis
Blockade ROCKET AF Rivaroxaban Once Daily Oral Direct Factor Xa
EPISTENT Evaluation of Platelet IIb/IIIa Inhibition for Inhibition Compared With Vitamin K Antago-
Stenting nism for Prevention of Stroke and Embolism
ESPRIT Enhanced Suppression of the Platelet IIb/IIIa SPAF II Stroke Prevention in Atrial Fibrillation II
Receptor With Integrilin Therapy STIMS Swedish Ticlopidine Multicenter Study
ESPS-1, -2 European Stroke Prevention Study 1, 2 TARGET Do Tirofiban and ReoPro Give Similar Efficacy
GUSTO Global Utilization of Streptokinase and tPA for Trial
Occluded Coronary Arteries TRITON–TIMI 38 Trial to Assess Improvement in Therapeutic
GUSTO IV-ACS Global Utilization of Strategies to Open Outcomes by Optimizing Platelet Inhibition
Occluded Coronary Arteries IV—Acute Coro- With Prasugrel—Thrombolysis in Myocardial
nary Syndromes Infarction 38
 50
Venous and Lymphatic Disorders
RAYMOND C. SHIELDS, MD
Introduction
Venous and lymphatic diseases are frequently encountered in
patients who have cardiac or arterial disorders. Understanding
the evaluation and management of these problems is an impor-
tant part of modern cardiovascular practice.
Venous Thromboembolism
VTE is the third most common cardiovascular disease, after acute
coronary syndromes and stroke. The annual incidence exceeds
1 per 1,000, with more than 300,000 new cases occurring in
the United States annually. The Virchow triad factors (stasis,
hypercoagulability, and vascular endothelial damage) contrib-
ute in various degrees to the development of venous thrombosis.
Independent acquired risk factors for VTE include immobility,
paralysis, recent surgery (patients undergoing total hip or knee
replacement surgery have a 40%-60% risk of VTE without pro-
phylactic therapy), trauma, malignancy, advanced age, prior his-
tory of superficial thrombophlebitis, central venous catheter or
transvenous pacemaker, pregnancy, and estrogen use (Box 50.1).
Up to 80% of patients with confirmed VTE have an identifiable
risk factor.
Pregnancy-Related VTE
The risk of VTE during pregnancy (0.5–1 per 1,000) is increased
5-fold compared with the risk for nonpregnant women of similar
age. More importantly, pulmonary embolism is the leading cause
of pregnancy-related death in the United States. Congenital
thrombophilias are associated with an increased risk of VTE
during pregnancy. Generalized screening for thrombophilia has
Abbreviations are expanded at the end of this chapter.
500
not been supported; however, pregnant women with a personal
or family history of VTE should undergo screening along with
genetic counseling.
Unfractionated heparin and low-molecular-weight heparin,
which do not cross the placenta, are recommended for treatment
of VTE during pregnancy. Traditionally, coumarin derivatives
have been contraindicated during pregnancy because of associ-
ated embryopathy, and package inserts for warfarin specifically
state that it is contraindicated during pregnancy. However, cou-
marin derivatives may be considered during certain stages of
pregnancy, particularly for patients who have a mechanical valve
prosthesis (a more complete discussion is presented in Chapter 80,
“Pregnancy and the Heart”). For nonpregnant women of child-
bearing age who require oral anticoagulation therapy, adequate
and appropriate contraceptive therapy must be ensured.
• The risk of VTE is increased 5-fold during pregnancy compared
with the risk for nonpregnant women.
• Pulmonary embolism is the leading cause of pregnancy-related
death in the United States.
• Coumarin derivatives are generally contraindicated during preg-
nancy because of associated embryopathy.
• Screening for thrombophilia is warranted for patients with a per-
sonal or family history of VTE.
Clinical Evaluation and Management
of Patients With VTE
A complete patient and family history, physical examination,
and general laboratory evaluation are indispensable in the initial
evaluation of patients who have acute VTE. Typical symptoms
of DVT include pain, redness, and swelling of a limb, although
many patients are asymptomatic at presentation. Signs of DVT
include pitting edema, warmth, erythema, tenderness, and a
 50 Venous and Lymphatic Disorders
Box 50.1. Acquired Risk Factors for Venous Thrombosis
Advanced age
Prior thrombosis
Immobilization
Major and orthopedic surgery
Malignancy
Oral contraceptives
Hormone replacement therapy
Central venous catheter
Antiphospholipid syndrome
Myeloproliferative disorders
Nephrotic syndrome
Paroxysmal nocturnal hemoglobinuria
dilated superficial venous pattern in the involved extremity.
Although lower limbs are the most common site of DVT, upper
extremity DVT may occur, especially in patients with a central
venous catheter or transvenous permanent pacemaker. Extensive
DVT involving an entire extremity may lead to venous gangrene
(phlegmasia cerulea dolens), most commonly in association with
an underlying malignancy. DVT in an unusual site (eg, cere-
bral, mesenteric, or renal vein) raises the probability of a hyper-
coagulable state. The clinical diagnosis of DVT is neither sensitive
(60%-80%) nor specific (30%-72%), and three-fourths of patients
who present with suspected acute limb DVT have other causes of
leg pain such as cellulitis, leg trauma, muscular tear or rupture,
postthrombotic syndrome, or Baker cysts. Objective noninvasive
tests generally are required to establish a diagnosis of DVT.
Venography is the criterion standard for the diagnosis of
DVT and is highly accurate for both proximal and calf DVT;
however, it is invasive, expensive, and technically inadequate in
about 10% of patients, and it may precipitate DVT in about 3% of
patients. Noninvasive tests for diagnosing DVT are accurate for
diagnosing proximal but not calf DVT. If the results of noninva-
sive testing are nondiagnostic or are discordant with the clinical
assessment, venography is indicated.
The assay for D-dimer, a degradation product of cross-linked
fibrin, is sensitive but not specific for acute DVT (ie, a normal
D-dimer level is unusual in patients with acute DVT). However,
when clinical suspicion for DVT is high, negative D-dimer results
alone are usually considered inadequate to rule out acute DVT.
Continuous-Wave Doppler Examination
A useful study for the diagnosis of proximal DVT is bedside
Doppler assessment, which evaluates each limb systematically
for spontaneous venous flow, phasic flow with respiration, aug-
mentation with distal compression, and venous competence
with the Valsalva maneuver and proximal compression. Doppler
examination is relatively insensitive to calf DVT.
Impedance Plethysmography and Strain-Gauge
Outflow Plethysmography
Both impedance plethysmography and strain-gauge outflow
plethysmography reliably detect occlusive thrombi in the prox-
imal veins (popliteal, femoral, and iliac veins), but they are less
reliable in detecting nonocclusive thrombi and are relatively
insensitive to calf DVT.
501
Venous Duplex Ultrasonography
Ultrasonography of the veins should include transverse gray-
scale images with and without compression ultrasonography
(in which venous noncompressibility is diagnostic of DVT and
venous compressibility excludes DVT) and spectral Doppler
waveforms. These elements of ultrasonography are highly sen-
sitive and specific for detecting proximal DVT when compared
with venography (Figures 50.1 and 50.2).
Treatment of DVT
Acute DVT should be treated with an initial course of unfraction-
ated heparin or low-molecular-weight heparin overlapping with
at least 5 days of oral vitamin K antagonist therapy (eg, warfarin
sodium). This initial course is followed by at least 3 months of
standard warfarin therapy with an international normalized ratio
goal of 2 to 3. A longer period of anticoagulation therapy (ie, 12
months) has been suggested for idiopathic or unprovoked DVT.
Catheter-directed thrombolytic therapy should be considered in
selected cases, such as younger patients with acute ileofemoral
DVT. Vena cava filter placement is indicated when there is a con-
traindication to—or a complication of—anticoagulation therapy;
however, anticoagulation therapy should be started as soon as
possible after filter placement. Catheter-based or surgical throm-
bolysis or thrombectomy may be indicated for selected patients,
but routine use is not advocated owing to the hemorrhagic risks
associated with treatment, the cost of therapy, and the frequent
need for reintervention.
An integral part of the initial therapy for acute DVT is to reduce
limb edema by elevating the leg or using a woven elastic bandage
(or both). When the edema has been reduced, a graduated com-
pression elastic support stocking (30–40 mm Hg) is recommended,
both to control edema and to reduce the risk of postthrombotic syn-
drome with subsequent venous stasis and ulceration. Additional
details on the diagnosis, treatment, and prophylaxis of DVT are
included in Chapter 78 (“Pulmonary Embolism”).
Superior Vena Cava Syndrome
SVC syndrome is most commonly caused by lung carci-
noma, mediastinal lymphadenopathy, and primary mediastinal
Figure 50.1. Compression Ultrasonography for a Patient With
Suspected Deep Vein Thrombosis. Noncompressibility of the right
superficial femoral vein is evidence of intraluminal thrombus. An echo-
genic thrombus is visible in the superficial femoral vein.
502 V Aorta and Peripheral Vascular Disease
Figure 50.2. Doppler Evaluation of Acute Occlusive Thrombus in
the Common Femoral Vein. No flow is shown.
malignancy, although indwelling central venous catheters, car-
diac pacing, and cardiac defibrillator leads are increasingly asso-
ciated with this syndrome. Mediastinal fibrosis, granulomatous
disease, and prior radiotherapy are also important contributors.
Patients who have SVC syndrome classically present with
head and neck fullness that is exacerbated by the supine position
or head dependency. Venous hypertension often results in dizzi-
ness, visual disturbance, headache, and syncopal spells. Patients
may also report impaired mentation, dyspnea, orthopnea, and
cough. Computed tomography effectively shows the location and
extent of obstructing structures and collateral venous routes.
Standard anticoagulation therapy is advocated for acute
thrombotic obstruction of the SVC. Thrombolytic therapy may
be considered for patients who have benign acute SVC syndrome.
Endovascular stenting of the SVC is often recommended as pri-
mary therapy for SVC obstruction due to malignant disease.
Stenting results in prompt relief of symptoms with a high rate of
prolonged efficacy.
• Clinical manifestations of SVC syndrome include dizziness, visual
disturbance, headache, and syncopal spells.
• Endovascular stenting of the SVC is often recommended in
selected patients.
Postthrombotic Syndrome
Postthrombotic syndrome occurs in as many as 80% of patients
who have DVT and is associated with considerable economic
costs and physical disability. Postthrombotic syndrome encom-
passes the clinical and pathologic findings of chronic venous
insufficiency as a result of DVT. Chronic venous insufficiency
commonly results in swelling, pain, fatigue, and heaviness in the
involved extremity. Secondary varicose vein formation, hyper-
pigmentation, brawny induration, and cutaneous ulceration can
occur if untreated. Venous claudication with previous iliofemoral
or vena caval thrombosis causes discomfort, fullness, tiredness,
and aching of the extremity during exercise. Unlike patients with
intermittent claudication, patients with venous claudication pre-
fer to sit down and elevate the extremity for relief. Postthrombotic
syndrome due to chronic deep venous incompetence is fre-
quently misdiagnosed as recurrent DVT. The correct diagnosis
is suggested by the clinical findings of chronic venous insuffi-
ciency and confirmed by exclusion of new thrombus formation
by compression ultrasonography. Side-by-side comparison of
current ultrasonographic studies with previous ultrasonographic
or venographic studies is invaluable in documenting new venous
thrombosis. Treatment of postthrombotic syndrome initially
includes aggressive efforts to reduce edema reduction with the
use of a woven elastic bandage and pumping devices, followed
by the use of graduated compression elastic support stockings
(30–40 mm Hg). Periodically elevating the leg during the day
and losing weight (for obese patients) is also beneficial.
Familial Thrombophilia
Routine screening for thrombophilia with an episode of VTE
should be avoided. The recommended indications for thrombo-
philia testing include idiopathic or recurrent VTE, the first VTE
episode before the age of 50, a first-degree relative with a history
of VTE before the age of 50, venous thrombosis in an unusual
territory (eg, cerebral, mesenteric, hepatic, or renal vein), neona-
tal purpura fulminans, and warfarin-induced skin necrosis.
Protein C Deficiency
Protein C deficiency is characterized by recurrent venous throm-
bosis and is inherited in an autosomal codominant fashion.
Episodes of thrombosis are generally spontaneous and usually
begin before the age of 30 years. Patients with this disorder
typically have protein C levels that are about 50% of normal.
Treatment requires lifelong oral anticoagulation, and there is a
potential risk of warfarin necrosis. Acquired protein C deficiency
can develop in patients postoperatively and in patients who have
liver disease or disseminated intravascular coagulation. Purpura
fulminans occurs in persons homozygous for this condition.
Protein S Deficiency
Protein S deficiency also causes recurrent venous thrombosis and
is inherited as an autosomal codominant trait. Episodes usually
begin before the age of 35 years and are generally spontaneous.
In affected persons, protein S levels are usually about 50% of
normal. Treatment requires lifelong oral anticoagulation, and
there is not an increased risk of warfarin necrosis. Acquired pro-
tein S deficiency can occur in association with the nephrotic syn-
drome, warfarin therapy, pregnancy, antiphospholipid antibody
syndrome, and disseminated intravascular coagulation.
Antithrombin III Deficiency
Antithrombin III deficiency, characterized by recurrent venous
and arterial thrombosis, is also inherited in an autosomal dom-
inant fashion. The first thrombotic episode usually occurs after
age 20 years and is often provoked by infection, trauma, surgery,
or pregnancy. Antithrombin III levels are typically 40% to 60%
of normal. An acquired form of antithrombin III deficiency can
occur in patients with nephrotic syndrome or severe liver disease
and in those receiving estrogen therapy.
Factor V Leiden Mutation
Factor V Leiden is the most common inherited thrombophilia in
white persons. Its heavy chain is resistant to the normal proteo-
lytic cleavage and inactivation by protein C (ie, activated protein
C resistance) owing to glutamine replacing arginine in position
506 of factor V. This mutation may occur in 5% to 10% of the
general population and is found in 20% to 60% of white patients
 50 Venous and Lymphatic Disorders 503

who have VTE. The risk of VTE is increased 3- to 8-fold for het-
erozygous carriers and 50- to 80-fold for homozygous carriers.

Prothrombin G20210A Mutation

Heterozygous carriers of the autosomal codominant prothrombin
G20210A mutation are at an increased risk of DVT and cere-
bral vein thrombosis. In these carriers, the plasma prothrombin
level is higher than the reference range. The prevalence is 0.7% to
6.5% in white persons and rare in other populations.

Hyperhomocysteinemia
Hyperhomocysteinemia has acquired and inherited causes and
is an independent risk factor for premature atherosclerosis and
recurrent DVT. However, in randomized trials with vitamin B
supplementation, the rate of recurrent VTE was not significantly
less among patients who had increased homocysteine levels com-
pared with patients who had normal levels. Routine screening
for hyperhomocysteinemia in patients with VTE is currently not
recommended.

JAK2 Mutations
JAK2 mutations are common in patients with myeloprolifera-
tive disorders such as polycythemia vera and essential throm- Figure 50.3. Young Woman With Lymphedema Praecox of Right
bocythemia, which are associated with an increased risk of Lower Extremity. The edema involves the toes.
thrombosis. Routine screening for JAK2 mutations is not rec-
ommended in the absence of splanchnic venous thrombosis or a grafts). Less frequently, HIT is complicated by stroke, myocar-
known myeloproliferative disorder. dial or mesenteric infarction, adrenal gland infarction, and skin
• Factor V Leiden is the most common inherited thrombophilia in necrosis. Among patients who have HIT, mortality is about 20%
white persons. and the limb loss rate is about 10%.
• Protein S deficiency is not associated with warfarin necrosis.
HIT is best managed by immediately stopping use of all
forms of heparin followed by alternative anticoagulation therapy.
• Routine screening for hyperhomocysteinemia in patients with VTE
Although the incidence of HIT associated with low-molecular-
is not recommended.
weight heparin is much less, known cross-reactivity with HIT
antibodies prohibits its use as an alternative to unfractionated
Heparin-Induced Thrombocytopenia
Thrombocytopenia is a known complication of heparin therapy.
HIT is generally categorized as either type I or type II. HIT type
I, also known as heparin-associated thrombocytopenia, is not
immune mediated and results in a transient decrease in the plate-
let count, usually within the first 2 days after beginning hepa-
rin use. The decreased platelet count usually normalizes despite
continued administration of heparin and is otherwise of minimal
clinical significance.
In HIT type II, antibodies to complexes of heparin and platelet
factor 4 appear to further augment platelet activation by binding
to the platelet Fc IIa receptor. Additional procoagulant platelet
microparticles are then released, stimulating endothelial cells,
which can lead to increased thrombin generation.
HIT type II is characterized by a 50% decrease in the plate-
let count beginning 5 to 10 days after heparin administration
is started in 70% of patients with HIT. Rapid-onset HIT (<24
hours after heparin reexposure) occurs in 30% of HIT patients
who have received heparin within the preceding 3 months.
Although HIT is a clinical diagnosis, laboratory confirmation is
generally recommended. The criterion standard for HIT is the
14
C-serotonin release assay; the positive predictive value of this
test nears 100%, and the negative predictive value is about 20%.
Thrombosis develops in almost 50% of patients who have HIT;
venous manifestations include DVT, pulmonary embolism, limb
gangrene, and cerebral sinus thrombosis. Arterial thrombosis is Figure 50.4. Inflammatory Lymphedema and Cellulitis. In this man,
most common in peripheral arteries (particularly in synthetic chronic tinea pedis served as a portal of entry for bacterial infection.
504 V Aorta and Peripheral Vascular Disease

Table 50.1. Characteristic Clinical Features of Regional or preeclampsia. Laboratory criteria consist of medium to high
Edema titers of IgG or IgM anticardiolipin antibodies, elevated levels of
β2-glycoprotein I IgG or IgM antibodies, or the presence of lupus
Feature Venous Edema Lymphedema Lipedema anticoagulant. These results require confirmation on 2 or more
Bilateral involvement Occasionally Maybe Always occasions at least 12 weeks apart. Of note, acquired antiphos-
Foot involved Yes Yes No pholipid antibodies secondary to certain infections (human
Toes involved No Yes No immunodeficiency virus, adenovirus, or Klebsiella infection;
Thickened skin No Yes No Lyme disease; rubella; varicella; syphilis) and medications (pro-
Stasis changes Yes No No cainamide, hydralazine, chlorpromazine, quinidine, isoniazid,
methyldopa) are generally not associated with thrombosis.

heparin. Two direct thrombin inhibitors, lepirudin and arga-
troban, are approved for management of HIT. Warfarin can be
administered when adequate anticoagulation has been achieved
with a thrombin-specific inhibitor and the platelet count is more
than 100 × 109/L. Fondaparinux, a pentasaccharide, may be con-
sidered for HIT prevention but is not approved for management
of HIT.
• HIT type II is immune mediated.
• Thrombosis develops in 50% of patients with HIT.
• Stopping use of all heparin and administering a thrombin-specific
inhibitor is recommended for management.
Antiphospholipid Syndrome
Manifestations of antiphospholipid syndrome include throm-
bosis and the presence of antiphospholipid antibodies, which
are a group of heterogeneous autoantibodies (including anti-
cardiolipin and lupus anticoagulant antibodies) against phos-
pholipid-binding proteins. In this syndrome, clotting occurs
more commonly in the deep veins of the legs; up to 50% of
the patients also have pulmonary emboli. Arterial thrombotic
events are comparatively less common, but up to 50% are stroke
and transient ischemic attacks. Coronary occlusions account
for 23%; the remaining 27% involve other vascular regions.
Cardiac manifestations of antiphospholipid syndrome also
include valvular vegetations, intracardiac thrombi, and non-
bacterial thrombotic endocarditis (also called Libman-Sacks
endocarditis).
The diagnosis of antiphospholipid syndrome requires that
1 clinical criterion and 1 laboratory criterion be met. Clinical cri-
teria are 1) objectively confirmed arterial, venous, or small-vessel
thrombosis and 2) pregnancy morbidity consisting of recurrent
fetal loss (≥3) before 10 weeks of gestation, an unexplained fetal
death at 10 weeks or more of gestation, or premature birth before
34 weeks of gestation due to placental insufficiency, eclampsia,
• The clinical manifestation of antiphospholipid syndrome includes
venous and arterial thrombosis.
• Diagnosis requires 1 clinical criterion and 1 laboratory criterion.
• Acquired antiphospholipid antibodies are generally not associated
with thrombosis.
Lymphedema
Lymphedema can be primary (idiopathic) or secondary to an
underlying disorder. Primary lymphedema, such as lymphedema
praecox, usually affects young women (9 times more frequently
than men) and begins before the age of 40 years (often before
age 20 years). In women, lymphedema often appears with men-
arche or the first pregnancy. Edema is bilateral in about half the
cases and is usually painless (Figure 50.3). The initial evaluation
of a young woman with lymphedema should include a complete
history and physical examination (including pelvic examination
and Papanicolaou test) and computed tomography of the pelvis to
exclude a neoplastic cause of lymphatic obstruction.
Secondary lymphedema is broadly classified into obstructive
(postsurgical, postradiation, neoplastic) and inflammatory (infec-
tious) types. Obstructive lymphedema due to neoplasm typically
begins after the age of 40 years and is caused by pelvic neoplasm
or lymphoma. The most frequent cause in men is prostate cancer.
Infection-related lymphedema frequently occurs as a result
of chronic or recurring lymphangitis or cellulitis. The portal
of entry for infection is usually dermatophytosis (tinea pedis)
(Figure 50.4). The diagnosis of lymphedema can be confirmed
by lymphoscintigraphy. Medical management of lymphedema
includes edema reduction therapy followed by daily use of cus-
tom-fitted, graduated compression (usually 40–50 mm Hg com-
pression) elastic support. Antifungal treatment is essential if
dermatophytosis is present. Weight reduction for obese patients
is also beneficial. Surgical treatment of lymphedema (lymphat-
icovenous anastomosis, lymphedema reduction surgery) is indi-
cated in highly selected patients.

Table 50.2. Clinical Features of the 4 Most Common Types of Leg Ulcer
Feature Venous Stasis Arterial Arteriolar Neuro trophic
Onset Traumaa Trauma Spontaneous Trauma
Course Chronic Progressive Progressive Progressive
Pain Absent (unless infected) Present Present Absent
Location Medial surface of leg Toe, heel, foot Lateral or posterior surface of leg Plantar
Surrounding skin Stasis changes Atrophic Normal Callus
Ulcer appearance
Edges Shaggy Discrete Serpiginous Discrete
Base Healthy Eschar, pale Eschar, pale Healthy or pale
a
Trauma may or may not be the cause.
 50 Venous and Lymphatic Disorders 505

A B

Figure 50.5. Types of Leg Ulcer. A, Venous stasis. B, Arterial. C, Arteriolar. D, Neurotrophic.

Edema Suggested Reading

Lower extremity edema is commonly encountered in cardiology Antithrombotic Therapy and Prevention of Thrombosis, 9th ed:
practice. Aside from edema due to cardiac disease, other causes American College of Chest Physicians Evidence-Based Clinical
of regional edema usually can be identified from characteristic Practice Guidelines Chest 2012 February;141:(2 Suppl)
clinical features (Table 50.1).

Abbreviations
Leg Ulcers
DVT deep vein thrombosis
The cause of lower extremity ulceration can usually be deter- HIT heparin-induced thrombocytopenia
mined by clinical examination. Clinical features of the 4 most JAK2 Janus kinase 2
common types of leg ulcer are summarized in Table 50.2 and SVC superior vena cava
Figure 50.5. VTE venous thromboembolism
 51

Vasculitis
PAUL W. WENNBERG, MD

Definitions and Pathogenesis unintentional weight loss, and night sweats are all common.
New-onset or markedly worsened vasospasm (Raynaud phenom-
By definition, vasculitis is inflammation of a vessel. Practically,
enon) in this setting is common and should raise suspicion for
vasculitis is a diverse group of diseases that are as difficult to
an inflammatory process. If an underlying disease is present,
classify as they are to diagnose and treat. Large, medium, and
such as rheumatoid arthritis or fibromyalgia, the symptoms may
small arteries, capillaries, and venules may be involved. The pre-
be attributed to this condition rather than to a vasculitis. Jaw or
senting symptoms vary by the organ system or systems affected
limb claudication is often overlooked by a patient because of an
(Table 51.1). Both primary and secondary causes of vasculitis
unconscious adaptation. Ischemic ulceration, a painful limb, or
exist (Table 51.2), and several conditions can mimic vasculitis
visceral pain may occur later in the disease course. Alternative,
(Box 51.1). This chapter discusses types of vasculitis grouped by
more common diagnoses are often made until more specific
vessel size, with cardiovascular pathology emphasized whenever
signs are present.
appropriate.
A thorough physical examination should be done, with care-
In general, the pathophysiologic mechanism of vasculitis is
ful attention to the vessels, skin, and mucous membranes. A rash,
a humoral- or a cell-mediated immune response to a noxious
localized pain over an organ, hemoptysis, or blood in the urine
stimulus. Often, the initial injury is nonspecific, due to immune
should be noted.
complex deposition from a nonvascular event. At other times a
Specific signs, such as a pulseless limb or ischemic wound,
direct immune response affects the vessel, such as the response
may allow for consideration of vascular involvement initially.
directed against the basement membrane in Henoch-Schönlein
Pain directly over an artery (eg, pain over the temporal arteries
purpura. The inflammation may be contained within the ves-
in giant cell arteritis) is described as a classic sign in textbooks
sel wall or may penetrate the vessel wall. In either situation, the
but is rarely seen. Localized abdominal tenderness may indicate
immune response is inappropriately sustained, and the inflam-
mesenteric artery involvement, especially if gastrointestinal tract
matory response by the vessel continues.
symptoms have been present.
• Vasculitis is commonly classified by the anatomic size of the
vessel(s) affected. Signs and Symptoms
• The pathogenesis of vasculitis is a sustained, inappropriate
humoral- or cell-mediated immune response. • Unexplained systemic illness with fever, malaise, and weight loss
are due to the systemic inflammatory syndrome.
• New-onset claudication, rash, Raynaud phenomenon, arterial tender-
General Principles of Evaluation
ness, or decrease in or loss of pulses are suggestive of vasculitis.
Symptoms and Signs
Nonspecific systemic symptoms usually precede the diagno- Testing
sis of a vasculitis by weeks to months. Fatigue, malaise, fevers, There is no single “best test” for vasculitis. The goal of testing is
2-fold: to determine the extent of involvement (ie, which arterial
Abbreviations and acronyms are expanded at the end of this chapter. segments and organ systems are involved) and to establish the
506
 51 Vasculitis 507

Table 51.1. Vessel Segment Involvement and Clinical Location of Vasculitis Syndromesa
Medium-Sized Arteries Small Arteries
Large and Great Vessels (named major, coronary, (small named, CNS, Arterioles, Capillaries,
(major branches, carotids, mid-distal limb, renal, digital, unnamed Venules (skin, mucosa,
Aorta subclavians, iliacs) mesenteric) intraorgan) alveoli, glomeruli)
Takayasu arteritis Takayasu arteritis Takayasu arteritis
Temporal arteritis Temporal arteritis Temporal arteritis
Cogan syndrome Cogan syndrome Cogan syndrome Cogan syndrome
Kawasaki disease Kawasaki disease Kawasaki disease
Behçet disease Behçet disease Behçet disease
Polyarteritis nodosa Polyarteritis nodosa
TAO TAO
Microscopic polyangiitis Microscopic polyangiitis
Churg-Strauss syndrome Churg-Strauss syndrome
Wegener granulomatosis Wegener granulomatosis Wegener granulomatosis
Arteriopathy of CTD Arteriopathy of CTD
Hypersensitivity Vasculitides
Henoch-Schönlein purpura Henoch-Schönlein
purpura
Leukocytoclastic vasculitis Leukocytoclastic
vasculitis
Abbreviations: CNS, central nervous system; CTD, connective tissue disease; TAO, thromboangiitis obliterans.
a
The disease processes most commonly affecting the segment are in bold.

diagnosis. Testing to determine the extent of involvement should Biopsy of an affected vessel is the most specific test for making
be guided by physical examination and symptoms (Box 51.2). the diagnosis. In some cases, such as in temporal arteritis, the ves-
Serologic evaluation typically begins at the same time and in sel is easily accessible, but this is more the exception than the rule.
many cases is diagnostic (Box 51.3). Both avenues of testing are
typically needed to make the diagnosis.
Treatment
Imaging of the arterial system is typically needed to deter-
mine the extent and severity of the arterial involvement. Treatment of vasculitis is primarily with immunosuppressive
Ultrasonography may be useful, especially for the carotid arter- agents, most often corticosteroids. Dosing and duration of treat-
ies or if an aortic or mesenteric aneurysm is suspected. CT ment are variable. Additional agents such as methotrexate, cyclo-
and MR angiography may be indicated and adequate in some sporine, azathioprine, and mycophenolate mofetil may be used
cases, but they lack the fine resolution of conventional contrast in conjunction with or in place of corticosteroids. Long-term
angiography. Angiography must be carefully planned to include treatment is best managed by personnel with more experience;
selective injections of the involved, and sometimes uninvolved,
arterial segments.
Box 51.1. Vasculitis Mimics
Cholesterol emboli
Table 51.2. Causes of Secondary Vasculitis
Bacterial endocarditis
Connective Tissue Diseases Infections Atrial myxoma
Scleroderma Parvovirus B19
Rheumatoid arthritis Streptococcus Pernioa
Systemic lupus erythematosus Staphylococcus Leprosy
Sjögren syndrome Hepatitis B
Systemic sclerosis HIV Vasoconstrictor use (eg, ergot, cocaine)
Antiphospholipid antibody Malaria Insect venom (eg, brown recluse spider)
syndrome
Dermatomyositis Tuberculosis Thoracic outlet syndrome
Cryoglobulinemia Mycoplasma Complex regional pain syndrome (reflex
Drugs Other sympathetic dystrophy)
Penicillin Inflammatory bowel disease Malignancy
Sulfa-based drugs Foreign proteins (immunizations)
Hydroxyurea Insecticides Cutaneous lymphoma
Methotrexate Organ transplant
Basal cell carcinoma
Allopurinol Malignancy-related vasculitis
Dilantin Sneddon syndrome
Tetracycline
Factitious
Quinidine
Interferon α a
Some consider pernio to be a limited form of vasculitis.
Abbreviation: HIV, human immunodeficiency virus.
508 V Aorta and Peripheral Vascular Disease

• Arterial or organ biopsy is the best test for diagnosis.

Box 51.2. Work-up of Vasculitis: Extent of Involvement • Conventional contrast angiography is frequently required for
Renal tests diagnosis and determining the extent and severity of arterial
compromise.
Urinalysis, creatinine value, biopsy
Respiratory tests Large Vessel Arteritis
PFTs, chest radiographs, CT The 2 most commonly occurring large vessel arteritides are
Cardiac tests temporal arteritis (often referred to as giant cell arteritis) and
Takayasu arteritis. In both of these forms, multinucleated giant
ECG, Holter monitoring, echocardiography
cells are seen on histopathologic analysis. Temporal arteritis and
Muscular tests Takayasu arteritis overlap substantially in pathology and distribu-
Creatinine kinase value, aldolase value, EMG
tion of disease; in practice they present as a spectrum of disease.

CNS tests
Temporal Arteritis
EMG, CSF analysis, MR imaging, CT
Epidemiology
Vascular tests
Temporal arteritis by definition occurs in persons older than
Segmental pressure measurements, conventional 50 years. Frequency increases with age, and women are affected
angiography, CT angiography, MR angiography, twice as often as men. The highest prevalence is in white per-
duplex ultrasonography sons, specifically those of northern European ancestry. There is a
strong association with polymyalgia rheumatica, with up to 50%
Abbreviations: CNS, central nervous system; CSF, cerebrospinal of patients with temporal arteritis having had a previous diagno-
fluid; CT, computed tomography; ECG, electrocardiography; EMG, sis of polymyalgia.
electromyography; MR, magnetic resonance; PFTs, pulmonary
function tests.
Anatomy and Pathology
The ophthalmic, temporal, vertebral, and carotid arteries are the
however, immediate treatment with high-dose oral or intra- segments most commonly involved. The other intracranial arteries
venous corticosteroids must be started in situations of critical are typically spared. Disease involvement is patchy, with normal seg-
organ or limb ischemia. Such early treatment is most notably ments present between affected areas. Involvement of arteries below
required to prevent visual loss in suspected giant cell arteritis. the diaphragm is rare. Pathologic analysis shows granulomatous
Investigations of antibody-based and receptor-modulating changes with disruption of the internal elastic lamina, T-lymphocyte
therapies are ongoing. infiltrates, and multinucleated giant cells (Figure 51.1).
• Testing should define the location of arterial stenoses and occlu-
sion and the involved organ systems. Symptoms and Signs
• Serologic testing may be diagnostic, but not in all cases. Even if Headache is the most common symptom at presentation.
serologic testing is diagnostic, the extent and severity of arterial
Nonspecific fatigue, weight loss, and myalgias are also com-
compromise must be defined.
mon. Jaw claudication, scalp tenderness, facial pain, and earache
occur with decreasing frequency. Ischemic events most com-
monly present as transient ocular changes, although permanent
Box 51.3. Work-up of Vasculitis: Serologic Assessment blindness can occur. Coronary artery involvement is uncommon
ANA (with extractable antigens) but should be considered if typical angina occurs in the setting of
a high ESR and other symptoms suggestive of temporal arteritis.
ANCA
Rheumatoid factor
Cryoglobulins
Anti-Scl-70 antibodies
Anticentromere antibodies
SPEP/UPEP
CH50 assay
HIV
Hepatitis serology
Parvovirus B19
Blood cultures

Abbreviations: ANA, antinuclear antibody; ANCA, antineutrophil

cytoplasmic antibody; HIV, human immunodeficiency virus; SPEP/ Figure 51.1. Giant Cell Arteritis of Left Temporal Artery.
UPEP, serum protein electrophoresis/urine protein electrophoresis. Granulomatous change disrupting the internal elastic lamina (arrow)
with lymphocytic infiltration. Multinucleated giant cell (arrowhead).
 51 Vasculitis
Tenderness over the temporal arteries is commonly present, but
absence does not exclude the diagnosis. The temporal arteries
are best palpated at the proximal portion, just anterior to the
pinnae of the ear.
Diagnosis and Treatment
An ESR of 50 mm/h or greater is expected, but frequently an ESR
greater than 100 mm/h is found. Rarely, patients with temporal
arteritis have a normal ESR (1%–2% of cases). If the ESR is nor-
mal and the rest of the clinical picture is suggestive of temporal
arteritis, the diagnosis should still be considered. C-reactive pro-
tein level is usually increased. Bilateral biopsy of the temporal
arteries provides definitive diagnosis. A 2- to 3-cm segment is
taken and multiple sections examined. Histopathologic changes
do not begin to normalize for several days after initiation of treat-
ment, so confirmatory biopsy is warranted.
Moderate- to high-dose corticosteroids (40–60 mg/d or
greater) should be initiated in an effort to minimize permanent
visual loss. If severe ocular symptoms or blindness is present,
high-dose intravenous corticosteroid therapy is recommended.
The ESR is usually a good marker of disease activity, although a
small percentage of patients may have a normal ESR. Surgical or
endovascular intervention is rarely indicated in the acute setting
but may be required late.
• Temporal arteritis occurs in persons older than 50 years, and
women are affected more often than men.
• Prior history of polymyalgia rheumatica is present in about 50%
of patients.
• Temporal artery biopsy is diagnostic. It should be done bilater-
ally, within several days of starting corticosteroids, with multiple
sections examined for highest sensitivity.
• Treatment should begin as soon as diagnosis is considered. High-
dose intravenous corticosteroids are appropriate if ocular symp-
toms are present.
• ESR is increased in more than 95% of cases and serves as a good
marker for disease activity.
Takayasu Arteritis
Epidemiology
Like temporal arteritis, Takayasu arteritis affects women much
more often than men, with studies showing female:male ratios
ranging from 2:1 to 9:1. Women are typically of childbearing
age. Those of Asian and Indian ancestry appear to be at greater
risk than those of European ancestry.
Anatomy and Pathology
The aorta is affected in all patients with Takayasu arteritis. The
great vessels, subclavian arteries, renal arteries, and mesenteric
arteries are frequently involved. The histopathologic characteris-
tics are varied, with granulomatous changes and multinucleated
giant cells early, followed by a diffuse inflammatory infiltrate,
then fibrosis with accelerated atherosclerotic changes. The
aggressive inflammation can later result in early stenosis and
occlusion and aneurysmal degeneration (Figure 51.2).
Symptoms and Findings
The classic finding of Takayasu arteritis on physical examina-
tion is an absent pulse or discrepant blood pressure in a young
woman. Symptoms may or may not be present, depending on the
509
Figure 51.2. Takayasu Arteritis of Right Carotid Artery in a 29-Year-
Old Woman. Accelerated atherosclerosis with thickened intima (arrow)
and vessel obliteration.
activity level of the patient. Nonspecific symptoms such as fever,
fatigue, and weight loss may occur. Specific symptoms reflect
the anatomic site(s) of involvement. Arm or leg claudication and
cerebrovascular symptoms (transient ischemic attack, visual
disturbances, and syncope) are the most common symptoms at
presentation.
Cardiac manifestations of Takayasu arteritis can be serious.
Angina, arrhythmia, and dyspnea are common, due to coronary
artery inflammation in the acute setting and accelerated athero-
sclerosis in the chronic setting. Up to 10% of patients in several
series had congestive heart failure due to acute aortic insuffi-
ciency with or without valve rupture. Pulmonary symptoms are
common and include dyspnea, hemoptysis, or pleurisy, caused
by pulmonary artery involvement. Renal involvement usually
results in hypertension. Assessment of blood pressure can be
problematic because stenosis may be present in the great vessels.
All 4 limbs should be assessed and the highest pressure used to
guide antihypertensive therapy.
Diagnosis and Treatment
Diagnosis of Takayasu arteritis is made by imaging in most
cases. Histologic analysis is rarely required. CT and MR angiog-
raphy are usually adequate to define the location of disease and
degree of stenosis. Segmental pressure measurements and duplex
ultrasonography are useful for assessment of chronic disease.
Echocardiography should be performed to assess presence and
severity of aortic insufficiency. Conventional angiography or
high-resolution CT or MR angiography is needed when coronary
artery involvement is suspected. In 10% to 30% of cases, ESR is
not increased. C-reactive protein level also may not be increased.
Other markers such as interleukin 6 are used in some centers
but are not well studied and the tests are not widely available.
Because of the variability in markers for disease activity, symp-
toms and imaging studies must be relied on in many cases for
long-term assessment.
Treatment is based on corticosteroids, frequently with the
addition of cyclosporine. Antiplatelet therapy is indicated in all
patients. Aggressive treatment of hypertension and optimization
of lipid levels to decrease atherosclerosis are reasonable meas-
ures. Revascularization of symptomatic arteries by endovascular
or surgical techniques is appropriate in chronic disease. In the
setting of acute inflammation, manipulation of the vessel should
be done only in the presence of severe cerebral, coronary, or
organ ischemia or critical limb ischemia.
510 V Aorta and Peripheral Vascular Disease
• Takayasu arteritis affects women of childbearing years.
• Nonspecific symptoms may be present for months before arterial
symptoms arise.
• Unilateral weakness or loss of pulse, discrepant blood pressures, or
onset of claudication in a young woman with systemic symptoms is
suggestive of Takayasu arteritis.
• Cardiac involvement is common.
• Coronary arteries may be affected directly.
• Aortic regurgitation may occur acutely or as a late complication.
• Occasional acute revascularization is needed for cerebral, cardiac,
or limb ischemia.
• Aggressive atherosclerosis and aneurysmal changes are common
in the late or end-stage phase.
Cogan Syndrome
Cogan syndrome is a rare disease of young adults that presents
with keratitis, uveitis, and vestibular symptoms. It often follows
an upper respiratory infection. Vertigo, nausea, and hearing
changes similar to those in Meniere disease may occur. The ESR
is increased, and 15% of patients have a vasculitis presenting as
aortitis or carditis. The most common cardiac manifestation is
aortic regurgitation due to aortic cusp involvement. Treatment
is high-dose corticosteroids.
Medium Vessel Arteritis
Kawasaki Disease
Epidemiology
Kawasaki disease, also known as mucocutaneous lymph node
syndrome, affects children younger than 5 years, most com-
monly before age 1 year. The cause is unclear but is thought to
be infectious. Children of Asian ancestry are affected more com-
monly, even when living outside Asia.
Anatomy and Pathology
Kawasaki disease affects medium-sized vessels primarily but not
exclusively. The coronary arteries are most commonly affected
(Figure 51.3), with aneurysm formation in 25% of untreated
children. Myocardial infarction occurs occasionally but typically
with good functional recovery. Myocarditis and valvulitis may
also be seen.
Figure 51.3. Left Anterior Descending Artery Thrombus in a
15-Year-Old Boy with Kawasaki Disease and Sudden Cardiac Death.
Thrombotic occlusion (arrow) of the vessel.
Symptoms and Findings
Criteria for diagnosis of Kawasaki disease have been set by the
American Heart Association and reflect the clinical symptoms.
Fever of more than 5 days’ duration despite antibiotics is found
in all patients. Nonsuppurative conjunctivitis; pleomorphic rash;
cervical lymphadenopathy; dry, cracked, and red mucous mem-
branes; and brawny edema of the hands and feet with desquama-
tion late are the other findings of the syndrome.
Diagnosis and Treatment
Intravenous immunoglobulin has been useful in the acute setting
for shortening the duration of symptoms and decreasing coro-
nary artery aneurysm formation. Aspirin alone is not effective in
decreasing symptom duration or aneurysm formation. Chronic
warfarin therapy is suggested for those with large aneurysms.
Long-term survival is good.
• Kawasaki disease affects children younger than 5 years.
• Cardiac manifestations are common.
• Coronary artery aneurysm is common.
• Myocardial infarction may occur.
• Treatment with intravenous immunoglobulin may decrease the
duration and complications of Kawasaki disease.
• Aspirin alone is not effective therapy.
Behçet Syndrome
Epidemiology
Behçet syndrome is a rare disorder found most commonly in the
Middle East, specifically Turkey, and along the ancient Silk Road
trading route between East Asia and the Mediterranean.
Anatomy and Pathology
Behçet syndrome is characterized by an autoimmune response
within the vaso vasorum that affects large vessels. Aneurysms,
stenosis, and occlusion of large and medium-sized arteries and
veins may occur. Thrombophlebitis of superficial and deep veins,
including cerebral venous thrombosis, occurs. The coronary
arteries and valve cusps may be affected.
Symptoms and Findings
Oral aphthous ulcers, genital ulcers, uveitis, and skin lesions are
the most common presentation of Behçet syndrome.
Treatment
Immunosuppressant therapy is less useful than in other
vasculitides.
Polyarteritis Nodosa
Epidemiology
PAN may occur in any age group. Hepatitis B is present in 10%
to 20% of patients.
Anatomy and Pathology
Muscular arteries of any organ may be involved with PAN, rarely
including the coronary arteries. The mesenteric and renal arteries
are the most frequently involved vessels. New-onset or worsened
 51 Vasculitis
Box 51.4. ANCAs in Vasculitisa
Large Vessel Vasculitis
None
Medium and Small Vessel Vasculitis
Churg-Strauss syndrome
p-ANCA positive ≈70%
Wegener granulomatosis
c-ANCA positive ≈90%
Microscopic polyangiitis
p-ANCA positive ≈80%
a
The presence of antineutrophil cytoplasmic antibodies (ANCAs) in
a perinuclear (p) or cytoplasmic (c) or atypical pattern is supportive,
not diagnostic.
hypertension is common in patients with renal involvement.
Microaneurysm formation is common and occurs as a result of
inflammatory destruction of the media. In the kidney, aneurysms
are often intraparenchymal, but glomerulonephritis is uncom-
mon. Patients are negative for ANCA (Box 51.4).
Symptoms and Findings
Systemic illness including fever, malaise, arthralgia, and myal-
gia are common in PAN. Localizing symptoms including men-
tal status changes, abdominal pain, flank pain, angina, and
peripheral neuropathy may be present. Mononeuritis multiplex
is present in at least two-thirds of patients. Skin changes from
livedo to vesicular bullae are present in up to half of patients.
Orchitis is common. Cardiac involvement is uncommon but usu-
ally presents as congestive heart failure. Angiography or biopsy
is diagnostic.
Treatment
Corticosteroids, with cyclophosphamide begun immediately or
soon after, is the usual treatment. Relapse is uncommon.
• Arthralgias, mononeuritis multiplex, skin, and gastrointestinal
tract symptoms are common in PAN.
• Cardiac involvement is uncommon but when present manifests as
congestive heart failure.
• No glomerulonephritis or pulmonary involvement is present.
• Patients are ANCA negative; up to 20% are hepatitis positive.
• Mesenteric angiography is the best diagnostic test.
• Recurrence of PAN is uncommon.
Small Vessel Arteritis
Microscopic Polyangiitis
Microscopic polyangiitis has recently been distinguished from
PAN. Like PAN, microscopic polyangiitis is a necrotizing vascu-
litis of small vessels with little or no immune complex deposition.
Mononeuritis multiplex, arthralgias, and systemic symptoms
including fever occur. Unlike PAN, necrotizing glomerulone-
phritis and pulmonary involvement are common, and p-ANCA
is positive in 80% of patients.
511
Churg-Strauss Syndrome
Churg-Strauss syndrome is rare, presenting as fever, asthma,
increased eosinophil count, and vasculitis. The histologic appear-
ance is a granulomatous inflammation with an eosinophil-rich
vasculitis of small to medium-sized vessels, most often involv-
ing the respiratory tract. p-ANCA is positive in 70% of patients.
Involvement of the gastrointestinal tract, kidneys, heart, and
central nervous system occur and greatly affect the prognosis.
Myocardial involvement can occur (Figure 51.4).
Wegener Granulomatosis
Wegener granulomatosis is a granulomatous vasculitis of small
to medium-sized arteries, capillaries, and venules of the res-
piratory tract and kidneys. Necrotizing glomerulonephritis is
common, occurring in 20% of patients at presentation and in
up to 80% during the course of the disease. The sinus mucosa
and nasal septum are frequently involved and lead to the classic
“saddle-nose” deformity. c-ANCA is positive in 90% of patients.
Treatment is usually cyclophosphamide with corticosteroids.
Hypersensitivity Vasculitis
Hypersensitivity vasculitis is a group of vasculitides with sev-
eral variations (eg, allergic vasculitis, leukocytoclastic vasculi-
tis, cutaneous necrotizing vasculitis) all affecting small vessels,
with cellular debris primarily on the venular side. The process
is immune mediated. Drug hypersensitivity, urticarial vasculitis,
mixed cryoglobulinemia, and serum sickness are common clini-
cal syndromes in this class. Skin findings include shallow painful
ulceration, purpura, petechia, urticaria, papules, and vesicles.
Henoch-Schönlein Purpura
Henoch-Schönlein purpura is a small vessel hypersensitiv-
ity vasculitis with IgA-dominant immune complexes typically
involving the gut, skin, and glomeruli. It affects children younger
than 10 years but can be seen in adults.
• Cardiac involvement is uncommon in small vessel arteritis.
° Occurs in 10% to 20% of cases.
° When present, myocarditis and coronary artery involvement are
most common.
• Symptoms of congestive heart failure, angina, dyspnea, and
arrhythmias in the setting of a suspected or known small vessel
vasculitis warrant further investigation.
• ANCA is positive in microscopic polyangiitis, Churg-Strauss syn-
drome, and Wegener granulomatosis.
Figure 51.4. Eosinophilic Myocarditis in Churg-Strauss Syndrome.
Eosinophilic infiltrate (arrow).
512 V Aorta and Peripheral Vascular Disease
Nonvasculitic Arteritis
Thromboangiitis Obliterans (Buerger Disease)
Epidemiology
TAO, or Buerger disease, is caused by use of or continued expo-
sure to tobacco products. Although TAO was historically thought
to be almost exclusive to men, more recent studies have shown
that the male-to-female ratio reflects smoking patterns in the pop-
ulation. The prevalence of TAO appears to be higher in India and
Asia than in Europe and North America. Age at presentation must
be younger than 45 years by definition and is frequently much
younger.
Anatomy and Pathology
Medium and small vessels are affected, most commonly the
distal limb vessels, but coronary, cerebral, and splanchnic
arteries also may be affected. Involvement of the aorta and
iliac and pulmonary arteries is rare. The histopathologic char-
acteristics are unique in that the vessel wall is not directly
involved. Rather, an intraluminal inflammatory thrombus is
present with an intact internal elastic lamina. Three histologic
phases have been defined. Acutely, an inflammatory thrombus,
highly cellular, is present, with microabscesses and giant cells
present. The intermediate phase shows further organization of
thrombus. Finally, the chronic phase demonstrates resolution
of inflammation and fibrosis. Superficial thrombophlebitis is
common and has the same histologic pattern as the arterial
findings.
Symptoms and Findings
The presenting symptom of TAO is usually a painful ulcer, typ-
ically on 1 or more digits in a young smoker. Both the upper
and lower extremities may be affected. Claudication may be
present. New-onset vasospasm (Raynaud phenomenon) is com-
mon. Concurrent or recent symptoms of superficial thrombo-
phlebitis are also common. Examination confirms an ischemic
ulcer, decreased or absent pulses, and classically palpable vas-
cular cords consistent with recent phlebitis. The Allen test is
frequently positive, even if upper extremity symptoms are
not present. Embolic causes, hypercoagulable states such as
antiphospholipid antibody syndrome, other forms of vascu-
litis, and arteriopathy due to connective tissue diseases must
be excluded. The ESR is not increased in TAO. If increased,
it could be due to an infected ulcer, but another cause must be
carefully considered.
Angiography is confirmatory. The affected limb(s) should be
imaged, as well as the contralateral limb and an unaffected upper
or lower extremity limb. The classic finding is a normal aorta
and iliac and proximal femoral arteries, with segmental stenosis
or occlusion of the distal leg and arm arteries. Corkscrew intra-
arterial “collaterals” are frequently seen.
Treatment
Smoking cessation and avoidance of secondhand tobacco expo-
sure is a must. More than 90% of those who quit smoking will
avoid amputation. The amputation rate in those who continue
to smoke is greater than 40%. Thrombolytic therapy or short-
term intravenous vasodilators such as iloprost are beneficial in
some, but not all, cases. Long-term prostanoids have not been
proved to be effective. Surgical and endovascular revasculariza-
tion, although beneficial late, are not indicated in the acute phase
except for limb salvage, with only guarded results expected.
Oral vasodilators and sympathectomy are helpful in the chronic
setting.
• Affects tobacco users before age 45 years.
• Vasculitis, embolism, and connective tissue disease must be
excluded.
• Histopathology shows an inflammatory thrombus, not vessel wall
inflammation.
• Medium and small arteries and veins are affected.
• Angiography is diagnostic; both upper and lower limbs—clinically
affected and unaffected—should be imaged.
• Amputation rate with continued tobacco use is greater than 40%.
Pernio
Chronic pernio, or chilblains, is a small vessel vasculitis of the
toes occurring during cold-weather months and resolving during
warm-weather months. Patients were classically thought to have
a past history of frostbite, but in practice it is more common for
the patient to have grown up in a northern location or to have par-
ticipated in a cold-weather sport. The symptoms begin decades
after the suspected period of exposure. Debate exists as to the
pathophysiologic mechanism, but many consider pernio a type of
limited vasculitis. Blisters over an erythematous base at the toe
tips leading to shallow ulcers and pitting is the usual presenta-
tion. Mild to moderate pain is common. Symptoms occasionally
occur after β-blockade is started for cardiac disease. Calcium
channel blockers or α-blockers are used to treat symptoms but do
not replace the use of good warm footwear during cold-weather
months.
• A limited vasculitis affecting the toe tips.
• Patients have a previous history of cold exposure.
• Appears in cold-weather months, resolves in warm months.
• Vasodilators (calcium channel and α-blockers) are helpful.
Connective Tissue Diseases
Several connective tissue diseases may result in vasculitis,
often in the form of an arteriopathy. The vessels affected are
very distal in the periphery, such as the digital arteries in the
upper and lower extremities. The initial presentation is vaso-
spasm (ie, Raynaud phenomenon) and cold intolerance. When
new-onset vasospasm is present in an adult (third decade or
later) and not related to other medical conditions or medica-
tions such as β-blockade, a connective tissue disease should be
considered.
Scleroderma, CREST syndrome, systemic lupus erythemato-
sus, and mixed connective tissue disease are some of the possible
diseases that cause vasospasm. If just 1 or 2 digits are affected,
an embolic event should be considered (Box 51.1).
Cardiac involvement is uncommon, but when present may
manifest as either myocarditis or pericarditis. Scleroderma often
results in pulmonary hypertension, most commonly with exer-
tional dyspnea as the main problem. Right-sided heart failure is
seen in extreme cases. Myocarditis is the most common presen-
tation in myopathies such as polymyositis.
 51 Vasculitis 513

• Late-onset vasospasm (Raynaud phenomenon) is a common initial Vascular manifestations of systemic autoimmune diseases. Asherson
presentation of a connective tissue disease. RA, Cervera R, Abramson SB, Piette J-C, Triplett DA, editors. Boca
Raton (FL): CRC Press; c2001.
• Embolism should be considered if sudden ischemia or vasospasm
of 1 or 2 digits is present.
Abbreviations
• Scleroderma is associated with pulmonary hypertension.
ANCA antineutrophil cytoplasmic antibody
CT computed tomography
ESR erythrocyte sedimentation rate
Suggested Reading
MR magnetic resonance
Langford CA. 15. Vasculitis. J Allergy Clin Immunol. 2003 Feb;111(2 PAN polyarteritis nodosa
Suppl):S602–12. TAO thromboangiitis obliterans
 52
Marfan Syndromea
NASER M. AMMASH, MD AND HEIDI M. CONNOLLY, MD
Marfan syndrome is the most common inherited multisystem
disorder of connective tissue. This autosomal dominant condi-
tion, first described by Antoine Marfan in 1896, has a reported
incidence of 2 to 3 per 10,000 persons, and it has no particular
sex, racial, or ethnic predilection. Early identification and appro-
priate management improve the outcomes among patients with
Marfan syndrome, who are prone to life-threatening cardiovas-
cular complications.
Genetics
Marfan syndrome is caused by a mutation of the fibrillin-1 gene
located at chromosomal locus 15q21.1. Fibrillin-1 protein is an
extracellular matrix glycoprotein that is an important component
of the connective tissue elastic microfibrils and is essential to
normal fibrinogenesis. Fragmentation and disorganization of the
elastic fibers in the aortic media, so-called medial degeneration,
are histologic markers of Marfan syndrome. These histologic
changes make the aorta stiffer and less distensible than the nor-
mal aorta. Recently, endothelial dysfunction also was shown to
be a feature of the aorta in Marfan syndrome.
The penetrance of the fibrillin mutation is high, and the pheno-
typic expression is extremely variable. To date, more than 1,000
different mutations involving the fibrillin-1 gene have been iden-
tified, but no genotype-phenotype correlations have been identi-
fied. In approximately 75% of cases, a person inherits the disorder
from an affected parent. The remaining 25% of cases result from
de novo mutation. Little prognostic information is provided by
a
Portions previously published in Ammash NM, Sundt TM, Connolly
HM. Marfan syndrome: diagnosis and management. Curr Probl Cardiol.
2008 Jan;33(1):7–39. Used with permission.
Abbreviations are expanded at the end of this chapter.
514
the detection of a mutation beyond the available information
from a patient’s own family history. In the revised Ghent criteria
for the diagnosis of Marfan syndrome, genetic testing (although
not mandatory) has a greater role in the diagnostic assessment
of patients suspected of having Marfan syndrome. In addition, it
facilitates prenatal or presymptomatic family screening.
With improved molecular techniques in gene testing, less than
5% of mutations that cause classic Marfan syndrome are missed
by conventional screening methods, which cost approximately
$2,500. Limitations to genetic testing include the following:
1. The mutation in the fibrillin-1 gene can cause conditions other than
Marfan-like disorder, such as the MASS syndrome (a heritable dis-
order of connective tissue involving the mitral valve, aorta, skeleton,
and skin) and familial annuloaortic ectasia
2. None of the current methods used to find mutations in the fibrillin-1
gene identify all mutations that cause Marfan syndrome
3. Family members with the same mutation causing Marfan syn-
drome can have wide variation in timing of onset and severity of
complications
Many manifestations of Marfan syndrome are now recog-
nized as resulting from excessive activation of transforming
growth factor β, which is a potent stimulator of inflammation,
fibrosis, and activation of certain matrix metalloproteinases.
• Marfan syndrome is an autosomal dominant genetic disorder
caused by a mutation of the fibrillin-1 gene.
• Marfan syndrome is a multisystem disorder involving the cardio-
vascular and skeletal systems and the eyes, lungs, skin, and dura
mater.
Cardiovascular Manifestations
Although Marfan syndrome is a multisystem disorder involving
the cardiovascular and skeletal systems and the eyes, lungs, skin,
 52 Marfan Syndrome 515

and dura mater, aortic root aneurysm and ectopia lentis are the Table 52.1. Scoring of Systemic Features
cardinal clinical features. The clinical features have been codi-
fied into the newly revised Ghent diagnostic nosology (Box 52.1 Feature Pointsa
and Table 52.1). The cardiovascular manifestations of Marfan Wrist and thumb sign 3
syndrome include the following: Wrist or thumb sign 1
1. Dilatation of the ascending aorta at the level of the sinuses of Pectus carinatum deformity 2
Valsalva, and, less commonly, the descending thoracic aorta, with Pectus excavatum or chest asymmetry 1
an associated increased risk of aortic valve incompetence Hindfoot deformity 2
2. Aortic dissection (dissection can involve the coronary artery ostia Plain pes planus 1
and result in myocardial infarction) Pneumothorax 2
3. Mitral valve prolapse with or without mitral valve regurgitation Dural ectasia 2
Protrusio acetabuli 2
Aortic root disease leading to aortic root aneurysm is progres- Decreased ratio of upper segment to lower segment and 1
sive and is the main cause of morbidity and mortality among increased ratio of arm span to height and no severe scoliosis
patients with Marfan syndrome. It is present in 50% to 60% of Scoliosis or thoracolumbar kyphosis 1
adults and in 50% of children with Marfan syndrome and can Reduced elbow extension 1
be readily detected with echocardiography (Figure 52.1). Both Facial features (3 of these 5: dolichocephaly, enophthalmos, 1
aortic diameter and aortic stiffness are independent predictors of downslanting palpebral fissures, malar hypoplasia,
progressive aortic dilatation, which predisposes to aortic dissec- retrognathia)
tion. Aortic root disease results from degeneration of the medial Skin striae 1
Myopia >3 diopters 1
layer, with associated fragmentation, disarray, and loss of elastic
Mitral valve prolapse (all types) 1
lamina. Dissection involves the ascending aorta in 90% of cases
(Figure 52.2), and 10% of dissections are distal to the left subcla- a
Maximum total: 20 points; score of 7 or more points indicates systemic
vian artery (ie, a type B dissection) (Figure 52.3). involvement.
Previously published. See “Credit Lines” section.
Marfan syndrome should be suspected in any patient younger
than 40 years with aortic dissection. Recent data suggest that
Marfan syndrome is present in 50% of patients with aortic dissec-
tion who present when younger than 40 years and in only 2% of undergoing an echocardiographic examination. In Marfan syn-
patients with dissection who are older. Many patients with Marfan drome, the mitral valve is characterized by thin, elongated leaflets
syndrome and aortic dissection have a family history of dissection. and anterior or bi-leaflet prolapse. An estimated 25% of patients
The 2 most important determinants of dissection risk are the max- with Marfan syndrome and mitral valve prolapse have progres-
imal aortic dimension and a family history of aortic dissection. sive regurgitation that could be due to worsening degeneration
Mitral valve prolapse with or without mitral valve regurgita- of the valve or spontaneous rupture of the chordae tendineae.
tion is found in 28% to 50% of patients with Marfan syndrome Tricuspid valve prolapse with or without regurgitation can occur
with or without mitral valve prolapse.
Pulmonary artery enlargement, and left ventricular dilatation
or dysfunction, beyond that explained by aortic or mitral regurgi-
Box 52.1. Revised Ghent Criteria for Diagnosis of tation has been reported among patients with Marfan syndrome.
Marfan Syndrome However, these features are not included in the scoring system
In the absence of family history used to confirm the diagnosis of Marfan syndrome.
Ao (z ≥ 2) and EL = MFSa
Ao (z ≥ 2) and FBN1 = MFS
Diagnosis
Ao (z ≥ 2) and Syst (≥7 pts) = MFSa The diagnosis of Marfan syndrome requires a careful history,
including information about any family members who may have
EL and FBN1 known Ao = MFS
the disorder or who had unexplained early or sudden unexpected
In the presence of family history death. A thorough physical examination, eye examination by an
EL and FH of MFS (as defined above) = MFS ophthalmologist, genetics evaluation, radiography, and echocar-
diography are routinely recommended. The goal is to determine
Syst (≥7 pts) and FH of MFS (as defined above) = MFSa
whether the diagnosis can be established clinically. The Ghent
Ao (z ≥ 2 if 20 y or older; z ≥ 3 if younger than 20 y) criteria, proposed in 1996 and revised in 2010, allow a uniform
and FH of MFS (as defined above) = MFSa approach to the diagnosis of Marfan syndrome and will decrease
the risk of premature diagnosis or misdiagnosis of Marfan syn-
Abbreviations: Ao, aortic diameter at the sinuses of Valsalva greater drome. A comprehensive multidisciplinary approach involving
than the indicated z score or aortic root dissection; EL, ectopia lentis; cardiac, orthopedic, ophthalmologic, and genetic consultations is
FBN1, fibrillin-1 mutation; FBN1 with known Ao, FBN1 mutation
that has been identified in a person with aortic aneurysm; FH, family
warranted to confirm or exclude the diagnosis. In the absence of
history; MFS, Marfan syndrome; pts, points; Syst, systemic score a conclusive family history of Marfan syndrome, the diagnosis is
(see Table 52.1); z, z score. made in 4 scenarios (Box 52.1):
a
Caveat: without discriminating features of Shprintzen-Goldberg 1. The presence of an aortic root aneurysm (with a z score ≥2 when
syndrome, Loeys-Dietz syndrome, or vascular form of Ehlers-Danlos standardized to age and body size) or aortic dissection and ectopia
syndrome and after TGFBR1/2, collagen biochemistry, COL3A1 lentis
testing if indicated. Other conditions or genes will emerge with time. 2. The presence of an aortic root aneurysm (with a z score ≥2 when
Previously published. See “Credit Lines” section. standardized to age and body size) or aortic dissection and the iden-
tification of the fibrillin-1 gene mutation
516 V Aorta and Peripheral Vascular Disease

Figure 52.1. Aortic Root Dilatation. Parasternal long-axis view on transthoracic echocardiogram shows aortic root dilatation in a patient with
Marfan syndrome. LA indicates left atrium; LV, left ventricle; RV, right ventricle; Vs, ventricular septum.

3. The presence of an aortic root aneurysm (with a z score ≥2 when stan-
dardized to age and body size) or aortic dissection and the presence
of systemic features with a score of 7 or more points (Table 52.1)
4. The presence of ecopia lentis and identification of the fibrillin-1 gene
mutation previously associated with aortic disease
For a patient with a positive family history of Marfan syn-
drome, the diagnosis can be established in the presence of 1)
ectopia lentis, 2) systemic features with a score of 7 or more
points, or 3) aortic root dilatation (with a z score ≥2 for adults
20 years or older or a z score ≥3 for patients younger than
20 years).
Aortic root diameter measurements should be made on trans-
thoracic echocardiography in a plane that is parallel to that of the
aortic valve and perpendicular to the axis of blood flow. These

Figure 52.2. Ascending Aortic Dissection. Long-axis view on transthoracic echocardiogram shows ascending aortic dissection with an intimal
flap in the aortic root (arrows). LA indicates left atrium; LV, left ventricle; RV, right ventricle.
 52 Marfan Syndrome
Figure 52.3. Type B Aortic Dissection. Magnetic resonance imag-
ing with gadolinium shows type B aortic dissection distal to the left
subclavian artery.
measurements should be standardized for age and body size (with
the z score) or compared with normal values based on age and
body surface area (Figure 52.4). The aortic dimension is usually
maximal at the sinuses of Valsalva. Other imaging techniques
such as transesophageal echocardiography, CT, angiography,
and MRI also may be helpful. Echocardiography, however, can
assess other cardiovascular manifestations of Marfan syndrome
and is therefore routinely used in establishing the cardiovascular
diagnosis and monitoring patients during follow-up.
The skeletal manifestations of Marfan syndrome necessitate
a thorough physical examination (Table 52.1). Occasionally, pel-
vic radiography may be helpful for diagnosing protrusio acetab-
uli, and lumbosacral MRI or CT may be helpful for diagnosing
dural ectasia. However, MRI and CT are performed only when
patients are symptomatic or occasionally when the findings
would be required for the revised Ghent criteria for the diagnosis
of Marfan syndrome. The most important skeletal manifesta-
tions can be easily assessed with a full skeletal examination that
includes evaluation for pectus deformity (Figure 52.5), arachno-
dactyly (Figure 52.6), hindfoot deformity with medial rotation
of the medial malleolus causing pes planus (flatfoot), character-
istic facial features, scoliosis, reduced extension of the elbows,
an increased ratio of arm span to height (>1.05), and a decreased
ratio of upper segment to lower segment (<0.85).
Lens dislocation is an important manifestation that is pres-
ent in 60% of patients with Marfan syndrome. The lens is best
assessed by an ophthalmologist in a slit-lamp examination with
pupillary dilation. Lens dislocation usually occurs early in life,
but annual ophthalmologic evaluation is recommended because
of the recognized risk of early severe myopia, retinal detach-
ment, glaucoma, and cataract formation in Marfan syndrome.
The other manifestations of Marfan syndrome are summarized
in Table 52.1.
Young patients in whom Marfan syndrome is suspected
but who do not meet the Ghent diagnostic criteria should have
517
evaluations in preschool, before puberty, and at age 18 years
because some of the clinical manifestations of Marfan syndrome
become more evident with age. Serial follow-up is recommended
when the aorta is enlarged irrespective of diagnostic criteria.
The differential diagnosis of Marfan syndrome includes the
following:
1. Homocystinuria shares several skeletal features (tall stature, over-
growth of long bones) and ocular features (ectopia lentis) with Marfan
syndrome, in addition to mitral valve prolapse. However, in contrast
to Marfan syndrome, affected patients often have mental retarda-
tion and are predisposed to thromboembolism. Homocystinuria is an
autosomal recessive disease that is characterized by increased uri-
nary homocysteine excretion, and it can be diagnosed by measuring
the total homocysteine value.
2. MASS phenotype includes myopia, mitral valve prolapse, mild aortic
enlargement, and nonspecific skin and skeletal features.
3. Ehlers-Danlos syndrome type IV includes skin laxity, scars, easy
bruising, and marked joint hypermobility as prominent features.
4. Stickler syndrome is characterized by retinal detachment (not ecto-
pia lentis), cleft palate, and hearing loss.
5. Congenital contractural arachnodactyly (Beals syndrome) is an
autosomal dominant disease that manifests with joint contractures,
scoliosis, crumpled ear malformation, and marfanoid appearance.
6. Familial thoracic aortic aneurysm or aortopathy is a condition in
which affected patients do not show any other systemic manifesta-
tions of Marfan syndrome.
7. Congenital bicuspid aortic valve disease with associated aortopa-
thy is often characterized by dilatation of the ascending aorta in the
mid section rather than at the aortic root level.
8. Loeys-Dietz syndrome includes unique features of hypertelorism, a
broad bifid uvula, and vascular involvement characterized by arte-
rial tortuosity and aneurysms with an increased risk of dissection
throughout the arterial tree, often at small arteries.
• A comprehensive multidisciplinary approach involving car-
diac, orthopedic, ophthalmologic, and genetic consultations
is warranted to confirm or exclude the diagnosis of Marfan
syndrome.
• The 2 most important diagnostic cardiovascular manifestations
of Marfan syndrome are dilatation of the aortic root and the
associated increased risk of aortic dissection.
Medical Management
The management of Marfan syndrome should involve a multi-
disciplinary approach, and treatment should be tailored to indi-
vidual manifestations because patients have variable degrees
of organ involvement. Early diagnosis and treatment are ben-
eficial. Patients of all ages should be encouraged to undergo
at least annual evaluation with clinical history, examination,
and echocardiography. Genetic counseling should be provided
initially to aid in the diagnosis and should also be provided to
potential parents because of the recognized 50% chance of trans-
mission from an affected parent to the child. Phenotypic vari-
ability, pregnancy counseling, and the availability of prenatal
diagnostic testing should be discussed. Annual ophthalmologic
examination should include screening and treatment for myopia,
retinal detachment, glaucoma, and cataracts. Significant skeletal
involvement is best managed by an orthopedic specialist.
Histologic abnormalities in the aorta reduce its distensibility
and compliance; as a result, the aorta becomes stiffer and has
excessive dilatation with age. β-Blockers can increase aortic dis-
tensibility and reduce aortic stiffness, in addition to decreasing
the heart rate and left ventricular ejection force, so these medi-
cations have been the treatment of choice and should be con-
sidered for all patients with Marfan syndrome. A randomized
518 V Aorta and Peripheral Vascular Disease

B 4.2

A 3.2 y = 1.02 + 0.98x

4.0 y = 0.97 + 1.12x
SEE = 0.18 SEE = 0.24

Sinuses of Valsalva, cm
3.0 3.8

Sinuses of Valsalva, cm
r = 0.93 r = 0.71
2.8 P < .0005 P = .0005
3.6
2.6
3.4
2.4
2.2 3.2
2.0 3.0
1.8
2.8
1.6
1.4 2.6
1.2 2.4

2.2
0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Body Surface Area, m2 1.4 1.6 1.8 2.0 2.2 2.4
Body Surface Area, m2
C 4.4
y = 1.92 + 0.74x
4.2 SEE = 0.37
r = 0.40
4.0 P = .0005
Sinuses of Valsalva, cm

3.8

3.6

3.4

3.2

3.0

2.8

2.6

2.4

1.4 1.6 1.8 2.0 2.2 2.4

Body Surface Area, m2

Figure 52.4. Normal Values of Aortic Root Diameter. Diameter is measured on transthoracic echocardiography at the level of the sinuses of
Valsalva. Normal values are based on age and body surface area. A, Normal infants and children. B, Adults younger than 40 years. C, Adults 40 years
or older. SEE indicates standard error of the estimate. (Previously published. See “Credit Lines” section.)

trial of propranolol treatment in adolescents and young adults
with Marfan syndrome showed a reduced rate of aortic dilatation
and fewer aortic complications in the treatment group. Patients
with Marfan syndrome were treated with propranolol; at an aver-
age follow-up of 10 years, the mean slope of the regression line
for aortic root dimensions was significantly lower in the pro-
pranolol group than the control group. In another study involv-
ing 44 patients with Marfan syndrome who were followed for
almost 4 years, the patients who took a β-blocker (or calcium
channel blocker if intolerant to β-blocker) showed a slower abso-
lute annual aortic growth rate of 0.9 mm compared with 1.8 mm
after adjustment for age and body size. In addition, prophylac-
tic medical treatment may be most effective in patients with an
aortic diameter of less than 4.0 cm. Patients who responded to
β-blockade tended to have a smaller aortic diameter (<4.0 cm),
and the data suggest that a reduction in the rate of aortic dilata-
tion with β-blockade is greatest in young patients with a small
aorta. Therefore, β-blockade should be considered in all patients
with Marfan syndrome, including children, and in those with an
aortic root diameter of less than 4 cm, unless contraindicated.
The dose of β-blocker should be adjusted to maintain a resting
heart rate of 60 to 70 beats per minute or a heart rate of 100 beats
per minute after submaximal exercise.
Although β-blocker therapy is the standard of care for Marfan
syndrome, alternative treatment options have emerged, with
primary interest focused on angiotensin-receptor blockers. In
a mouse model of Marfan syndrome, the angiotensin-receptor
blocker losartan blocked transforming growth factor β recep-
tor, suppressing apoptosis, promoting cellular proliferation, and
preserving the proximal aortic elastic fiber histology and overall
aortic diameter. Because of these encouraging findings, a mul-
ticenter study comparing losartan and atenolol was initiated to
address the question of optimal medical treatment in Marfan syn-
drome. A small retrospective study reported beneficial results of
losartan in young Marfan patients with progressive dilatation of
the aorta despite β-blocker therapy. The angiotensin-converting
 52 Marfan Syndrome
Figure 52.5. Severe Pectus Excavatum. This patient with Marfan
syndrome underwent aortic root replacement.
enzyme inhibitors enalapril and perindopril have also been
shown to reduce aortic growth rate.
A main concern in Marfan syndrome is the increased risk of
aortic dissection, which is inherent to the histologic abnormalities
of the aorta. Patients with Marfan syndrome should be counseled
to seek urgent medical evaluation for any acute chest pain, back
pain, abdominal pain, syncope, or sudden change in vision.
Treatment with β-blockers, although beneficial in most
patients with Marfan syndrome, does not protect against aortic
dissection. In 1 study, 20% of patients with Marfan syndrome
who received treatment with β-blockers or calcium channel
blockers had major cardiovascular complications requiring oper-
ation over a 4-year period. The risk factors for aortic dissection
in Marfan syndrome include 1) an aortic diameter of more than 5
cm, 2) aortic dilatation extending beyond the sinuses of Valsalva,
3) a rapid rate of aortic dilatation (>5% per year or >5 mm per
year in adults), and 4) a family history of aortic dissection.
In adults, a yearly echocardiographic measurement of the aor-
tic root diameter is recommended as long as the diameter is less
than 4.5 cm and no major change in aortic dimension has been
noted recently. Twice-yearly aortic surveillance is recommended
when the aortic root diameter is more than 4.5 cm. Alternatively,
Figure 52.6. Arachnodactyly Demonstrated With 2 Signs. Left, Steinberg thumb sign (the entire thumbnail projects around the ulnar border of
the hand). Right, Walker-Murdock wrist sign (the thumb and fifth finger overlap around the wrist).
519
CT or MRI of the chest can be performed. These imaging tech-
niques provide a complete assessment of the thoracic aorta,
including the descending segment that might not be optimally
visualized with echocardiography. These imaging methods com-
plement each other. Periodic CT or MRI of the descending tho-
racic and abdominal aorta is recommended for all patients with
Marfan syndrome, especially before aortic root replacement,
because of the recognized risk of aneurysm formation in other
parts of the aorta.
Patients should be educated to avoid smoking and to moni-
tor their blood pressure because nicotine use and hypertension
are believed to increase the risk of dilatation of the aorta. The
goal blood pressure is less than 120/80 mm Hg, according to the
Seventh Report of the Joint National Committee on Prevention,
Detection, Evaluation, and Treatment of High Blood Pressure.
Patients with Marfan syndrome should avoid strenuous iso-
metric exercise such as weight lifting. Static exercise is associ-
ated with a significant increase in peripheral blood pressure and
proximal aortic stress. In addition, competitive sports such as ice
hockey, full-court basketball, surfing, and scuba diving should be
avoided because of the increased risk of body collision. Impact
sports also may cause ocular complications in patients with
Marfan syndrome. A moderate level of dynamic or aerobic activ-
ity is safe for most Marfan patients. In general, unless they have
had prior aortic root or valve replacement, patients with Marfan
syndrome can participate in recreational exercise categorized as
low to moderate intensity, such as the following: golfing, boating,
skating, snorkeling, brisk walking, using a treadmill or stationary
bicycle, hiking, or playing doubles tennis. Individual assessment
is suggested for other activities thought to be of intermediate
risk, such as singles tennis, soccer, touch football, baseball, and
skiing.
• β -Blocker therapy and possibly angiotensin-receptor blocker ther-
apy should be considered for all patients with Marfan syndrome,
including children and patients with aortic root diameter <4 cm,
unless contraindicated.
• The risk factors for aortic dissection in Marfan syndrome include
an ascending aortic diameter >5 cm, aortic dilatation extending
beyond the sinuses of Valsalva, a rapid rate of aortic dilatation
(>5% per year or >5 mm per year in adults), and a family history
of aortic dissection.
520 V Aorta and Peripheral Vascular Disease
• Patients with Marfan syndrome should be educated to avoid smok-
ing and to monitor their blood pressure, with a goal blood pressure
<120/80 mm Hg.
• Patients with Marfan syndrome should avoid strenuous isomet-
ric exercise, competitive and contact sports, and exercising to the
point of exhaustion.
Surgical Management
On the basis of several comparative studies, there is general
agreement that overall outcome is better among patients who
undergo early elective aortic root surgery, in addition to con-
tinuing medical treatment. Emergency aortic replacement in
patients who have Marfan syndrome is associated with mark-
edly increased 30-day mortality (12%). In contrast, the operative
mortality among patients undergoing elective repair at experi-
enced centers is very low (<1%). Prophylactic surgery is rec-
ommended when the diameter of the aortic sinuses of Valsalva
reaches 5.0 cm or the aortic size index is 4.25 cm/m2 or more.
The need for surgery at an aortic sinus dimension of 4.5 cm or
less may be indicated by other factors, such as family history of
aortic dissection, personal history of descending aortic dissec-
tion, severe aortic valve regurgitation with associated symptoms
or progressive ventricular dilatation or dysfunction, the possibil-
ity of a valve-sparing operation, the contemplation of pregnancy,
and rapid rate of aortic dilatation (aortic root growth >5 mm per
year). The aortic root diameter should be plotted serially against
body surface area, and an operation should be considered if the
diameter begins to increase rapidly from a stable percentile even
if the absolute measurement is less than 5.0 cm. An increase in
aortic dimension of more than 1.0 cm per year is regarded as
rapid progression in a child, whereas an annual increase of 5%
or more or an annual increase of more than 5 mm in an adult is
considered clinically significant; accordingly, regular aortic root
and valve surveillance is needed.
The severity of aortic valve regurgitation due to root dilata-
tion is a major determinant of the kind of surgical intervention
offered when aortic root replacement is required. When clini-
cally significant aortic valve regurgitation or valve distortion
is present, aortic valve replacement is indicated. The original
operation developed for patients with Marfan syndrome was the
Bentall procedure, which uses a composite graft. This procedure
includes aortic root and valve replacement with either a biologic
valve or a mechanical valve and requires coronary artery reim-
plantation. The composite aortic graft is a good choice for older
children and adults, especially in the presence of ascending aortic
dissection or clinically significant aortic valve regurgitation. One
study reported that 8-year survival was 90% after the Bentall
procedure and freedom from reoperation was 96%. There was
however, an increased risk of thromboembolic events due to the
mechanical valve prosthesis.
In the absence of clinically significant aortic valve regurgita-
tion, a valve-sparing aortic root replacement can be considered.
There are 2 kinds of operations: 1) the Yacoub operation, in
which the aortic conduit is attached to a cuff of a native aorta
just beyond the aortic valve and the valve is conserved (remod-
eling), and 2) the David procedure, in which the native aortic
valve is reimplanted into the aortic graft, which is attached to
the left ventricular outflow tract (the reimplantation technique).
Several modifications of these approaches have been described.
The mortality risk is low when the valve-sparing operation
is performed by experienced surgeons. Studies have reported
that the survival rate at 8 years was 100%. Mild or no aortic
regurgitation may be present in up to 75% of patients for as long
as 10 years. The risk of the patient needing aortic valve replace-
ment for severe aortic valve regurgitation is estimated to be 6%
to 10% at 10 years. The need for an aortic valve–sparing opera-
tion is an indication for early surgery, and therefore an aortic
root diameter less than 5.0 cm with preserved aortic valve func-
tion is an indication to consider surgical repair. In addition, the
valve-sparing operation or the use of a biologic prosthesis is
recommended for a woman who wants to become pregnant and
for other patients with relative contraindications for long-term
anticoagulation.
Mitral valve repair for severe mitral regurgitation with associ-
ated symptoms or progressive left ventricular dilatation or dys-
function is associated with a very low operative risk (<1%). In a
study of 23 patients with Marfan syndrome who had mitral valve
repair, the 10-year survival rate was 79% and the 10-year rate of
freedom from mitral regurgitation was 87%.
• Prophylactic surgery is recommended when the diameter of the
aortic sinuses of Valsalva reaches 5.0 cm.
Postoperative Cardiovascular Care
The long-term medical treatment and lifelong surveillance
required for patients with Marfan syndrome, even after aortic
root surgery, are major commitments for both patient and phy-
sician. β-Blocker therapy or angiotensin-receptor blocker ther-
apy (or both) should be resumed postoperatively and continued
indefinitely unless contraindicated. Long-term aspirin therapy
and endocarditis prophylaxis are recommended for all patients.
Long-term anticoagulation with warfarin is recommended for
patients who have a mechanical prosthesis or other indications,
such as atrial fibrillation. At least annual cardiovascular and
ophthalmologic evaluations with a clinical history, examination,
and transthoracic echocardiography are recommended, and the
descending thoracic and abdominal aorta should be imaged peri-
odically. As patients with Marfan syndrome age, reoperation is
often needed if vascular complications develop elsewhere in the
arterial system. After patients with Marfan syndrome undergo
aortic root replacement, the most frequent cause of death is dis-
section or rupture of the residual aorta. Therefore, periodic imag-
ing of the entire aorta is recommended indefinitely after initial
aortic repair, and monitoring can be accomplished with MRI or
CT angiography. The rate of change of the aortic diameter should
influence follow-up intervals. Indications for replacement of an
enlarged or dissected segment of the descending aorta should
include any of the following:
1. A rapid increase in aortic dimension of more than 5 mm per year
2. A diameter of 55 mm or more
3. An affected segment diameter 2-fold greater than the adjacent
segment
4. Symptoms related to aortic dilatation or dissection
5. Malperfusion syndrome
Mitral valve replacement or repair may be needed in up to
10% of patients requiring aortic root operation. More than 60%
of patients with Marfan syndrome require multiple operations
during their lifetimes, and therefore lifelong comprehensive mul-
tidisciplinary follow-up is recommended.
An additional late complication of composite and valve-spar-
ing operations is the development of coronary ostial aneurysms.
These aneurysms develop at the site of coronary artery reimplan-
tation as a result of the perioperative stretch of the weakened wall
of the coronary ostium and are not likely to progress.
 52 Marfan Syndrome
Pregnancy in Marfan Syndrome
Pregnancy is possible in women who have Marfan syndrome.
However, 2 major issues need to be discussed with the patient and
family: the risk of cardiovascular complications in the affected
mother and the 50% risk of transmission of Marfan syndrome
to the fetus. Because the disorder is autosomal dominant, each
offspring of an affected parent has a 50% chance of inheriting
the genetic mutation. Genetic counseling should be offered to all
patients with Marfan syndrome. Mutation detection or linkage
can be used for prenatal diagnosis.
The risk of aortic dissection in pregnancy is increased and
may be caused by the inhibition of collagen and elastin deposi-
tion in the aorta by estrogen and the hyperdynamic hypervolemic
circulatory state of pregnancy. Gestational hypertension and
preeclampsia also may have a role. Reports of pregnancies in
patients with Marfan syndrome have shown a complication rate
of approximately 11%, mostly related to aortic rupture and endo-
carditis. The overall risk of death during pregnancy is around
1%. The risk of aortic root complication is increased when the
aortic root diameter is more than 4 cm at the start of pregnancy,
and the risk is further increased when the aorta is more than
4.5 cm or if it dilates rapidly during pregnancy. The risk of fur-
ther dilatation of the aorta during pregnancy is lowest in the first
trimester and greatest in the third trimester, during labor, and in
the first few weeks post partum.
Patients who become pregnant should continue β-blocker
therapy throughout pregnancy and have serial follow-up echocar-
diography to assess the change in the size of the aorta during
pregnancy. Surgery should be considered during pregnancy in
patients with progressive aortic dilatation or before the aortic
root diameter reaches 55 mm.
A planned cesarean delivery is generally preferred for patients
who have Marfan syndrome and aortic dilatation. Assisted vagi-
nal delivery can be considered when the aortic root diameter is
521
less than 4 cm, the aorta has not changed during pregnancy, and
there is no associated severe cardiovascular disease. Antibiotic
prophylaxis administered around the time of delivery is appropri-
ate for patients with clinically significant valvular regurgitation
or prior root and valve replacement surgery. Postpartum uterine
hemorrhage should be anticipated as a complication of Marfan
syndrome and has been reported in nearly 40% of women.
• The risk of aortic root complication during pregnancy is increased
when the aortic root diameter is >4 cm at the start of pregnancy.
Prognosis
The life expectancy of untreated patients with Marfan syndrome
is considerably shortened; an early study reported the lifespan
to be about 32 years. However, with medical advances such as
β-blocker therapy, angiotensin-receptor blockers, and elective
surgical repair, the median lifespan has increased gradually to
more than 72 years.
Summary
A marked advance in understanding of the cause of Marfan syn-
drome, early recognition of the disorder, and subsequent institu-
tion of medical and surgical therapy have resulted in a dramatic
improvement in the prognosis during the past few decades.
Further medical advances, focused primarily on the genetic basis
of Marfan syndrome, are expected to allow continued therapeu-
tic improvements with associated prognostic implications in the
future.
Abbreviations
CT computed tomography
MRI magnetic resonance imaging
This page intentionally left blank
 Section VI

Coronary Artery Disease Risk Factors

This page intentionally left blank
 53
Coronary Artery Disease Epidemiology
THOMAS G. ALLISON, PHD, MPH
Natural History and Pathophysiology
CAD develops as a consequence of decreased blood flow to the
myocardium due to coronary atherosclerosis. It begins as fatty
streaks in the coronaries and other arterial beds early in life.
Fatty streaks and some raised lesions have been identified in
various autopsy studies in men in their late teens and early 20s
and women by their late 20s and early 30s. The development of
atherosclerosis appears to follow a complex pathway:
1. Endothelial dysfunction is caused by a number of factors, such as
cigarette smoking, hypertension, and hyperlipidemia. This dysfunc-
tion permits entry of various blood components into the arterial inti-
mal layer. The components ordinarily roll along the endothelial layer
and do not damage the artery.
2. Infiltration of leukocytes, lipids (carried by LDL-C particles), and
macrophages takes place, and these blood cells accumulate within
the intimal layer of the artery.
3. Inflammation occurs, and lipid-rich foam cells form as macrophages
ingest LDL-C particles. The foam cells accumulate and grow into
fatty streaks, which eventually bulge into the arterial lumen. The
disease may still be reversible at this stage if LDL-C levels in the
blood are decreased, HDL-C particles are increased, and endothelial
function is restored.
4. Proliferation and migration of smooth muscle cells from the medial
layer form a fibrous cap over the fatty lesion. This complex lesion is
not entirely reversible. Proliferation of the vasa vasorum provides
the lesion with its own blood supply.
5. Continued plaque progression is characterized by growth and even-
tual necrosis of the lipid core, calcification, hemorrhage within the
plaque, and surface erosion with formation of nonobstructive clots.
The external elastic lamina may stretch to accommodate this plaque
growth without the development of ischemia, but eventually the arte-
rial lumen may narrow to the extent that ischemia develops during
periods of physical or psychological stress. This ischemia may be
silent or cause angina.
Abbreviations and acronyms are expanded at the end of this chapter.
6. Thinning and weakening of the fibrous cap due to the action of
matrix metalloproteinases released from macrophages, coupled
with the shear stress of blood flow over the luminal surface of the
plaque, may cause acute plaque rupture. Precipitating factors, such
as nicotine use, excessive physical stress, and psychological stress,
also appear to have a role in the rupture of atherosclerotic plaque.
Plaques that are less than 70% obstructive appear to be more likely
to undergo rupture, perhaps because of their greater lipid content,
thinner fibrous cap, and more irregular configuration with the pres-
ence of distinct “shoulders” where shear forces concentrate.
Clinical manifestations of plaque rupture with subtotal or
total occlusion of the affected artery, now called acute coronary
syndrome, include acute myocardial infarction, unstable angina,
and sudden death due to ventricular fibrillation. Other manifes-
tations of CAD include stable or effort angina, which occurs
when plaque growth leads to subtotal occlusion of the artery, and
ischemic cardiomyopathy. The latter may be secondary to thin-
ning and dilatation of the myocardium in the first few weeks after
a large infarction or it may develop more gradually as a result of
repeated, smaller myocardial infarctions and chronic ischemia.
Congestive heart failure is a frequent problem in patients with
ischemic cardiomyopathy. Heart failure is essentially an end-
stage process characterized by exercise intolerance; pulmonary
or peripheral edema, or both; and a high mortality rate.
Figures 53.1 through 53.6 trace the pathophysiology of CAD
from fatty streaks to plaque rupture, with total occlusion of the
artery leading to acute myocardial infarction and sudden death.
Prevalence and Incidence
of CAD and Its Trends
The AHA provides an annual statistical report on CAD.
Current data suggest that there are more than 17.6 million cases
of CAD in the United States and that, among persons older
than 20 years, 9.1% of men and 7.0% of women have CAD. The
525
526 VI Coronary Artery Disease Risk Factors
Figure 53.1. Fatty Streaks in the Aorta of a 19-Year-Old Man.
incidence of myocardial infarction is about 935,000 cases per
year (610,000 fi rst events and 325,000 recurrent attacks), and
approximately 309,000 out-of-hospital CAD-related deaths
occur per year. At age 40 years, a man in the United States
has a 49% lifetime risk of CAD versus a 32% lifetime risk for
a woman at age 40 years. The age-adjusted death rates from
CAD (per 100,000) by race are shown in Table 53.1. Of note,
the rates of CAD death for men are higher than for women in
each racial group.
From an international perspective, CAD rates in the United
States are intermediate between the low rates in eastern Asia
(one-third to one-fifth of the US rate) and the high rates in the
former socialist countries of Eastern Europe (2 or 3 times the
US rate).
Age-adjusted CAD rates in the United States decreased by
42% between 1970 and 2000. Rates of CAD for whites declined,
with the rate of decline greater for men than women, while rates
for blacks and Mexican Americans actually increased. For the
first 15 years of this period (1970–1985), investigators have
determined that behavior changes—particularly, the millions
of US men who discontinued cigarette smoking—accounted for
most of the decrease in CAD rates, whereas changes in medical
practice (ie, programs teaching cardiopulmonary resuscitation,
911 emergency coverage, coronary care units and defibrillators,
emergent coronary angioplasty, and improved medical therapy)
Figure 53.2. Advanced Lesion With Large Lipid Core. Lesion is
nonocclusive because of remodeling of artery.
Figure 53.3. Advanced Lesion With Plaque Hemorrhage. This event
is causing severe narrowing of the lumen and may be causing ischemia
and effort angina.
accounted for the greatest part of the decrease in the second
15 years (1985–2000) (Figures 53.7 through 53.9).
Despite the decline in age-adjusted rates, the number of actual
cases of CAD has remained stable in men and increased slightly
in women over the past 30 years, due principally to the aging of
the US population. Studies from Olmsted County, Minnesota,
confirm this shift in the CAD epidemic from middle-aged men
toward women and elderly persons. Internationally, the CAD epi-
demic is gradually moving from countries with established mar-
ket economies to developing countries.
Many cardiovascular specialists are concerned that CAD
rates in the United States have already reached or will soon reach
a low plateau and then begin to increase again. Cited reasons for
the end of the decline and the subsequent resurgence of the CAD
epidemic include the failure of smoking rates to decline further
than the 20% to 22% level at which they have been stuck for
several years and the dramatic increases in obesity, metabolic
syndrome, and type 2 diabetes mellitus. A recent study look-
ing at autopsy data from people who died of unnatural causes in
Rochester, Minnesota, suggests that CAD prevalence began to
increase after the period 1993 to 1996, after steady declines from
the period 1981 to 1984. There is, of course, a lag time between
the increased prevalence of subclinical CAD and the increased
death rate from CAD, so it may be a few more years before age-
adjusted death rates actually show an increase.
Figure 53.4. Intermediate Lesion With Incorporated Thrombus.
This condition appears to be due to plaque erosion rather than frank
plaque rupture and may be an asymptomatic event.
 53 Coronary Artery Disease Epidemiology 527

Table 53.1. Age-Adjusted Death Rates From CAD by

Race
CAD Death
Rate/100,000 Persons

Race Males Females

Black 206.4 130.0
White 176.3 101.5
Hispanic or Latino 132.8 85.4
Native American or Alaskan Native 122.4 76.8
Asian or Pacific Islander 101.3 58.9
Abbreviation: CAD, coronary artery disease.

Figure 53.5. Plaque Rupture With Torn Fibrous Cap. Exposure of
collagen to platelets causes increased platelet adhesiveness and aggre-
gation and likely serves as a platform for thrombus formation of the
affected artery. This event was probably symptomatic.
Risk Factors
Risk factors for CAD were first identified in the Framingham
Heart Study, which began in 1948, and have been confirmed in
numerous subsequent investigations. Although the pathophysiol-
ogy of CAD suggests that it is primarily a lipid disorder, other
risk factors also have important roles. Risk factors that have long
been recognized include nonmodifiable factors, such as age,
male sex, and family history of premature CAD (age of onset
<55 years for primary male relative and <65 years for primary
female relative). So-called major modifiable risk factors include
increased LDL-C level, low HDL-C level, hypertension, and dia-
betes. Obesity, lack of regular exercise, and psychological stress
have also been recognized as CAD risk factors, but they are fre-
quently termed minor risk factors. Reasons for relegating these
risk factors to a less important status may include a weaker or
less consistent association with CAD incidence or prevalence in
epidemiologic studies, difficulty in accurately defining the level
of the factor, and lack of successful interventions by the medical
community.
The impact of a risk factor is related to both the strength of
its association with CAD—often described in terms of relative
risk—and its prevalence in the population. Table 53.2 summa-
rizes these aspects of various CAD risk factors as they apply to
Although lack of regular exercise is a lower relative risk than
the major risk factors listed in Table 53.2, its high prevalence
suggests that increasing the physical activity level in the United
States could have a greater impact than more aggressive treat-
ment of lipid levels and blood pressure. Indeed, biobehavioral
variables—physical activity, diet, and cigarette smoking—appear
to predict more than 75% of CAD deaths worldwide, according
to the multinational INTERHEART study.
The 4 major risk factors for CAD—hyperlipidemia (includ-
ing both low HDL-C and high LDL-C levels), hypertension, dia-
betes, and cigarette smoking—have been shown to consistently
and strongly predict CAD incidence and prevalence in a large
number of observational studies, both cross-sectional and longi-
tudinal. Randomized clinical trials to reduce LDL-C level and
blood pressure have demonstrated conclusively that lowering the
level of these risk factors reduces CAD risk. Data for the effect
of blood glucose control on CAD risk in diabetes are less exten-
sive to date, and randomized clinical trials of smoking cessation
and CAD risk have not been conducted, nor are they likely to
ever be, but incontrovertible evidence for diabetes and smoking
as risk factors has been established from numerous observational
studies.
“Optimal” levels for each of these CAD risk factors continue
to be redefined by new epidemiologic studies and, more impor-
tantly, the results of large, randomized clinical trials, such as the
ICD8: 410-413 ICD9: 410-414
600
Mortality Rate per 100,000

the US population. 500

400
US whites
300
MSP
residents MEN
200

150
US whites
MSP
residents WOMEN
100
80
1970 1975 1980 1985 1990 1995
Year
Figure 53.7. Coronary Artery Disease Trends in the United
States and in Minneapolis and Saint Paul, Minnesota, From 1970
through 1997. ICD8 and ICD9 indicate eighth and ninth revisions of
Figure 53.6. Advanced Lesion Occluded With Thrombus. This International Classification of Diseases; MSP, Minneapolis and Saint
stage is the end of the atherosclerotic process for this patient. Paul. (Previously published. See “Credit Lines” section.)
528 VI Coronary Artery Disease Risk Factors

160 agencies. Prevention of obesity is a national priority in the United

140
Mortality Rate per 100,000

Out-of-hospital
120
100
80
In-hospital
60
40
Out-of-hospital
In-hospital
20
15
1985 1988 1991 1994
Year
Figure 53.8. Comparison of In-Hospital and Out-of-Hospital
Coronary Artery Disease Death Trends. The greater rate of decline for
in-hospital deaths suggests that medical care improvements are predom-
inantly responsible for the reduction in coronary artery disease death
rates after 1985. (Previously published. See “Credit Lines” section.)
Heart Protection Study of lipid reduction and the ALLHAT of
blood pressure control. For all these major modifiable risk fac-
tors, risk is either linearly or exponentially related to the level
of the risk factor (Figure 53.10, for serum cholesterol), but target
levels below which risk is acceptably low have been developed.
For prevention of CAD, the NCEP has determined that an LDL-C
level of less than 100 mg/dL is optimal for CAD prevention, and
a level of 70 mg/dL or less is now recommended for persons
with existing CAD plus other risk factors. Ideal blood pressure
was set at 120/80 mm Hg or less by the 7th report of the Joint
National Commission on Blood Pressure Awareness and Control.
According to the AHA and numerous other groups, smoking and
exposure to secondhand smoke should be avoided completely. A
normal blood glucose level is now defined as 100 mg/dL or less
by the American Diabetes Association, and the threshold blood
glucose level used to diagnose diabetes has been decreased to
126 mg/dL. For good cardiovascular health, moderately vigorous
physical activity should be performed for at least 30 minutes 5
or more days per week, according to the Surgeon General of the
United States and a number of professional medical groups and
States, and the nation has been advised to maintain weight within
10% of ideal, generally defined in terms of a BMI of 18.5 to 25
(calculated as weight in kilograms divided by height in meters
squared).
A meta-analysis of several large epidemiological trials—
MEN
encompassing more than 350,000 patients monitored for 21 to 39
years, with nearly 40,000 CAD deaths recorded—has shown that
fewer than 1 in 10 US adults has an optimal risk profile (ie, free
of any of the major risk factors with an LDL-C level <130 mg/dL;
an HDL-C level ≥45 mg/dL for men or ≥55 mg/dL for women;
WOMEN
blood pressure ≤120/80 mm Hg; being a nonsmoker; and having
no diabetes). However, persons with this ideal risk profile have a
CAD risk that is less than one-fifth of all others (ie, persons with
1997 at least 1 risk factor) combined and account for less than 2% of
all CAD deaths. More recent data published by Lloyd-Jones on
lifetime CAD risk for men and women at age 50 years indicate
that a man at that age with all optimally controlled risk factors
has a 40-year risk (ie, to age 90 years) of CAD death of only 5%
and a woman at age 50 years with all optimally controlled risk
factors has a 40-year risk of CAD death of 8%.
Outcome data from the Framingham Heart Study are used in
various equations to predict risk due to CAD for a given person
on the basis of the major risk factors (plus age and sex). In turn,
targets for LDL-C and, to some extent, blood pressure are indexed
to the Framingham Risk Score. A 10-year risk level greater than
20% is considered high risk; 10% to 19%, intermediate risk; and
less than 10%, low risk. Because of the strong relationship of
CAD risk to age, however, a Framingham Risk Score of 12%, for
example, is of much greater concern for a 35-year-old man than
for a 65-year-old man. For this reason, Framingham 30-year risk
estimates are now available from the Framingham Heart Study
website. The Framingham Risk Score conveniently provides low
and average scores for each sex by 5-year age-group, so a given
person’s risk score can be indexed against those norms. The
Web site for the NCEP (http://www.nhlbi.nih.gov/guidelines/
cholesterol/index.htm) provides pages that can be downloaded
to calculate 10-year risk of CAD according to the Framingham
Risk Score for men and women (using either total cholesterol or
LDL-C level) and an online calculator to estimate risk for adults
free of CAD or diabetes.
Criticisms of the Framingham Risk Score include 1) strong
reliance on age, thus making it fairly useless for calculating risk

600

500
Deaths per 100,000

400

300

200

100
ICD1 ICD2 ICD3 ICD4 ICD5 ICD6 ICD7 ICD8 ICD9 ICD10
0
1900 1908 1916 1924 1932 1940 1948 1956 1964 1972 1980 1988 1996 2004

Year
Figure 53.9. Coronary Artery Disease Death Rates for the US Population, Showing Continuing Decline Into the 21st Century. ICD1 through
ICD10 indicate International Classification of Diseases. (Previously published. See “Credit Lines” section.)
 53 Coronary Artery Disease Epidemiology
Table 53.2. CAD Risk Factors in the US
Population
Risk Factor Relative Risk
Hyperlipidemia 2.5
Cigarette smoking 3.0a
Hypertension 2.4
Diabetes mellitus 3.0b
Lack of exercise 1.9
Abbreviation: CAD, coronary artery disease.
a
Relative risk of CAD is greater in men (4.0) than women (2.0),
a difference likely related to the number of cigarettes smoked.
b
Relative risk of CAD is greater in women (4.0) than men (2.0).
in young adults (but which is now improved with publication of
30-year risk estimates); 2) 10-year risk projection, a problem in
using the score to stratify young adults by risk; 3) lack of con-
sideration of other well-established risk factors, including family
history, sedentary lifestyle, and impaired fasting glucose; and 4)
failure to include such newer risk factors as lipoprotein (a) and
C-reactive protein, which have been shown to predict risk inde-
pendent of the traditional risk factors. The newer risk factors are
discussed briefly herein and are the subject of a more extensive
discussion in a separate chapter.
CAD Epidemiology in Men Versus Women
The age-adjusted risk of CAD death or related events for women
is less than for men. For ages greater than 75 years, the rates of
CAD for women gradually approach (but do not catch up to) the
rates for men. However, younger men are at significantly greater
risk (up to 3-fold) than younger women. The cited reasons for
this risk variation include a lower-risk lifestyle for women (eg,
less smoking, a lower-fat diet), higher HDL-C levels (mostly
secondary to lower testosterone levels), and possible protective
effects of endogenous estrogens. However, diagnosing CAD
in women is somewhat more difficult than in men in that their
presentation is more varied in both symptoms and electrocar-
diographic abnormalities. When CAD is correctly diagnosed,
though, women seem to now be receiving appropriate care, at
least during the acute hospitalization. Current data suggest that
the gender gap in appropriate treatment of CAD between men
and women has largely or completely disappeared in recent years.
Long-term control of CAD risk factors in women may continue
to lag behind men, however.
6-Year CAD Deaths/1,000
14
12
10
8
6 Risk Score. Carotid intimal-medial thickness is likely more
4 appropriate for younger individuals, particularly women
2 (because it avoids radiation exposure), since calcification of cor-
0
153 175 187 198 208 216 226 238 253 290
Serum Cholesterol Deciles, mg/dL
Figure 53.10. Relationship of Serum Cholesterol Level to 6-Year
CAD Death Rate. Results are from the Multiple Risk Factor Intervention
Trial (MRFIT). CAD indicates coronary artery disease.
529
The rate of death after myocardial infarction is higher in
women than men, although most of that difference is due to
older age at presentation and more age-associated comorbidities.
Estimated US Women appear to be more susceptible to development of heart
Prevalence, %
failure than men, even after controlling for left ventricular func-
25 tion. Angina, by comparison, has a more favorable prognosis in
23 women than men. Finally, women may not have as favorable an
20 outcome after mechanical interventions for CAD, including angi-
7 oplasty, stenting, and coronary artery bypass surgery, because of
59 the smaller absolute size of their coronary arteries and a ten-
dency toward more diffuse CAD.
A New Epidemiology of CAD?
Although CAD death in individuals with none of the major risk
factors is rare, it is clear that the traditional Framingham risk fac-
tors do not fully explain the distribution of CAD in the US pop-
ulation. In the large meta-analysis of more than 350,000 patients
mentioned earlier, more than 90% of patients dying of CAD in
the 21- to 29-year follow-up period had at least 1 major modi-
fiable risk factor—but so did nearly 70% of those who did not
die of CAD over the same period. Traditional risk factors pre-
dict CAD more accurately cross-culturally than within a given
population, and predicting events in individual patients through
the use of traditional risk factors is even more difficult. As a
result of improved understanding of the biology of atheroscle-
rotic vascular disease, investigators have developed a number of
new markers of CAD risk. Some of these, such as lipoprotein
(a), appear to be causally related to CAD development, though
perhaps through mechanisms that are dependent and synergistic
with LDL-C, while others, such as C-reactive protein, may more
mark the biological steps—in this case, inflammation—in the
atherosclerotic process than directly cause it.
A revised risk factor score was recently published by Ridker
and his associates. Called the Reynolds Risk Score, it incor-
porates family history and C-reactive protein into a model
predicting the 10-year risk of heart attack, other CAD event,
or stroke for a person without prior CAD, stroke, or diabe-
tes. The Reynolds Risk Score also extends the maximum age
of the person for whom risk can be predicted from 65 years in
the Framingham Risk Score to 80 years. A simplified version
is available online (http://www.reynoldsriskscore.org/). A more
complicated version in the original publication includes lipopro-
tein (a), which drops out of the equation when the LDL-C level
is less than 100 mg/dL.
Still other novel risk factors serve as anatomical evidence for
atherosclerosis presence. Coronary calcium measured by com-
puted tomography and carotid intimal-medial thickness meas-
ured by ultrasonography represent 2 such noninvasive markers
for atherosclerosis that are being used clinically to predict CAD
risk and identify goals for LDL-C level. They have been shown
in several large epidemiologic databases to predict CAD events
independently of the traditional risk factors and the Framingham
onary lesions is uncommon for men until their sixth decade of
life and for women until their seventh decade. Soft, cholesterol-
rich plaques are likely present earlier in life, as studies of donor
hearts and autopsy studies of young people who died in wars or
motor vehicle crashes have shown, such as the Olmsted County
study described earlier and the PDAY study. Coronary calcium is
likely more strongly predictive of future CAD events.
530 VI Coronary Artery Disease Risk Factors

The availability of inexpensive generic drugs for risk fac- smoking, will always have the greatest potential, if successful, to
tor modification (ie, statins, metformin, and several classes of lower the incidence and prevalence of CAD.
antihypertensives) in primary prevention is also changing the
landscape of risk stratification. As cost of treatment decreases—
Suggested Reading
while the safety of treatment continues to be good over a progres-
sively longer observation period—the need for precision in risk American Heart Association Statistical Update. [cited 2012 Jan 18].
stratification decreases. Similarly, the added costs and, in some Available from: http://circ.ahajournals.org/cgi/reprint/CIRCULA-
cases, radiation exposure of noninvasive screening are less well TIONAHA.109.192667.
Behavioral Risk Factor Surveillance System home page. [cited 2012 Jan
justified than when expensive drugs with a short track record of
18]. Available from: http://apps.nccd.cdc.gov/brfss/page.asp?cat=
observation were the indication for risk reduction. The concept EX&yr=1999&state=US#EX.
of a polypill to prevent CAD, first proposed by Wald and Law National Cholesterol Education Program home page. [cited 2012 Jan 18].
in 2003, does not even require measurement of traditional risk Available from: http://www.nhlbi.nih.gov/guidelines/cholesterol/
factors. Instead, it requires solely the person’s age and sex by index.htm.
combining low doses of various agents to treat such risk fac- National Heart, Lung, and Blood Institute home page. [cited 2012
tors as hyperlipidemia and hypertension that are highly preva- Jan 18]. Available from: http://www.nhlbi.nih.gov/index.htm.
lent in the population for whom the polypill is designed (men
and women older than 55 years). To date, no US national body Abbreviations
(eg, NCEP, AHA, American College of Cardiology, American
Medical Association) has endorsed the polypill strategy, though AHA American Heart Association
various versions are being studied internationally. BMI body mass index
CAD coronary artery disease
It is not likely that laboratory tests for blood lipids and glu-
HDL-C high-density lipoprotein cholesterol
cose, measurement of blood pressure, and ascertainment of LDL-C low-density lipoprotein cholesterol
smoking status and physical activity levels will disappear from NCEP National Cholesterol Education Program
CAD risk prediction strategies, even as new markers for CAD
risk or atherosclerosis presence, or both, are being developed and
tested. A safe and inexpensive method of early detection of CAD Clinical Trials
or a good test to identify an impending CAD event would cer- ALLHAT Antihypertensive and Lipid-Lowering Treatment to
tainly improve prevention efforts. Yet, public health efforts to Prevent Heart Attack Trial
promote behavioral changes, such as diet, physical activity, and PDAY Pathological Determinants of Atherosclerosis in Youth

Metabolic Syndrome
THOMAS G. ALLISON, PHD, MPH
Definition of Metabolic Syndrome
Metabolic syndrome is a novel CV risk factor that has generated
widespread interest—and controversy—since its inclusion in the
ATP III of the NCEP (the Executive Summary was published in
May 2001). The NCEP lists 5 characteristics for consideration of
metabolic syndrome (Table 54.1).
Waist circumference and HDL-C cutoff values are sex-specific,
but the other 3 characteristics are the same for men and women.
Patients who have at least 3 of the 5 characteristics have meta-
bolic syndrome. In addition, the AHA/NHLBI scientific state-
ment document “Diagnosis and Management of the Metabolic
Syndrome” says that prothrombotic and proinflammatory states
generally accompany the metabolic syndrome even though they
are not specifically defined as additional characteristics.
Underlying factors that cause the metabolic syndrome—and
to some extent are included as defining characteristics—include
abdominal obesity, insulin resistance, lack of physical activity,
older age, and hormonal imbalance. Another contributing factor
is diet, particularly a diet that is high in calories (beyond daily
requirements) and simple carbohydrates.
Although metabolic syndrome is dichotomized as absent or
present on the basis of these characteristics, some large lipid-
lowering studies, such as the WOSCOPS, have shown that CV
risk increases progressively with the number of characteristics
that are present. Persons with 4 or 5 characteristics have approx-
imately 3 times the CV risk of those with none of the character-
istics (Figure 54.1). Studies from the United States and Finland
have identified similar relative risks for CV events among persons
with metabolic syndrome.
Abbreviations and acronyms are expanded at the end of this chapter.
54
Central obesity was associated with diabetes mellitus and dys-
lipidemia as early as 1967 in a publication by Avogaro and col-
leagues. Characterization of a cluster of risk factors associated
with insulin resistance under the term syndrome X was proposed
by Reaven in 1988. Following the system of Reaven, which iden-
tified insulin resistance as the central feature, the WHO defined
metabolic syndrome in 1999, 2 years before the NCEP publica-
tion. Although the NCEP definition could be largely considered a
system of counting non-Framingham or near risk factors for ath-
erosclerotic vascular disease (including impaired fasting glucose
level) that cluster in overweight and obese persons, the WHO def-
inition strictly requires type 2 diabetes mellitus, impaired fasting
glucose level, glucose intolerance, or demonstration of insulin
resistance by the clamp methods plus 2 or more of the following
4 criteria:
1. High BMI (calculated as weight in kilograms divided by height in
meters squared) (≥30) or high waist to hip ratio (men, >0.90; women,
>0.85)
2. High levels of serum triglycerides (≥1.7 mmol/L) or low levels of
HDL-C (men, <34 mg/dL; women, <39 mg/dL)
3. High blood pressure (systolic ≥140 mm Hg or diastolic ≥90 mm
Hg)
4. High urinary albumin excretion rate (>20 μg/min) or high albumin
to creatinine ratio (≥30 mg/g)
Although perhaps less mechanistic or physiologically consist-
ent than the WHO definition, the NCEP strategy for identifying
metabolic syndrome has the advantage of being somewhat eas-
ier to use clinically. Evidence to date suggests that the preva-
lence of metabolic syndrome is similar according to the NCEP
and WHO criteria, and both definitions are roughly equivalent
for predicting future CV risk. However, not all patients clas-
sified as having metabolic syndrome according to 1 definition
531
532 VI Coronary Artery Disease Risk Factors

Table 54.1. Characteristics of Metabolic Syndrome (as • Waist circumference: What is the technique for measuring waist
Defined by the ATP III of the NCEP) circumference?
• Treatment of risk factors: If blood pressure or the blood glucose,
Characteristic Level HDL-C, or triglyceride levels are treated below the cutoff values, do
they still count as metabolic syndrome characteristics?
Large waist circumference • Adjustment of definition for different ethnic groups: Various adjust-
Men >102 cm (40 in) ments have been proposed for different ethnic groups.
Women >88 cm (35 in) a. Chinese: waist circumference ≥90 cm (35.4 in) for men and ≥80
Low HDL-C cm (31.5 in) for women; HDL-C <38.6 mg/dL for both men and
Men <40 mg/dL women.
Women <50 mg/dL b. South Asians: waist circumference ≥90 cm (35.4 in) for men and
High triglycerides ≥150 mg/dL ≥80 cm (31.5 in) for women.
Elevated blood pressure ≥130 mm Hg systolic or ≥85 mm Hg c. African Americans: inherent genetic differences in lipid levels
diastolic between whites and African Americans have led some investiga-
Impaired fasting glucose level ≥110 mg/dL tors to suggest that the cutoff value for triglycerides should be
lowered to ≥110 mg/dL for African Americans and the cutoff for
Abbreviations: ATP III, Adult Treatment Panel III; HDL-C, high-density
lipoprotein cholesterol; NCEP, National Cholesterol Education Program. HDL-C should be increased.
Previously published. See “Credit Lines” section. d. Hispanics: the US Hispanic population has a higher prevalence
of hyperglycemia, low HDL-C, and high triglycerides but a
lower prevalence of high blood pressure compared with whites,
have it according to the other definition; the correspondence is although alterations in the cutoff values for Hispanics have not
80% to 85%. Therefore, this chapter uses the NCEP definition been specifically recommended.
of metabolic syndrome. The various definitions are compared in
Table 54.2. Relationship With CV Risk
• CV risk appears to increase continuously as the number of characteris-
Metabolic Syndrome Controversies tics increases rather than being dichotomous on the basis of 3 charac-
There have been well-publicized criticisms of the NCEP defini- teristics. This brings into question the value of giving the diagnosis of
metabolic syndrome to a patient with 3 or more characteristics.
tion of metabolic syndrome, some of which were discussed in a
• As a corollary, does the risk associated with metabolic syndrome
“critical appraisal” written by the American Diabetes Association differ from that attributable to the individual characteristics? Some
and the European Association for the Study of Diabetes (pub- of the potentially more important metabolic syndrome characteris-
lished in 2005). tics, particularly high blood pressure and low HDL-C, are already
included in the Framingham risk score for predicting CV risk.
• Some question whether metabolic syndrome is simply a more techni-
Clarification of Definition
cal way of measuring the effect of obesity on the risk of CV disease.
• Blood pressure: Is blood pressure elevated if either systolic or dias- But in women at least, obesity without metabolic syndrome has been
tolic pressure exceeds the cutoff value, or do both need to exceed shown to be relatively benign compared with the increased risk
their cutoff (systolic ≥130 mm Hg; diastolic ≥85 mm Hg)? How is among obese women with metabolic syndrome.
blood pressure to be measured (supine or sitting, or the average of a • Not all of the subset of metabolic syndrome characteristics may be
specific number of measurements)? equally predictive of CV risk. There are 10 possible combinations

12
4 or 5 factors
Participants With Event, %

10 3 factors

8
2 factors

1 factor
6

0 factors
4

0
0 1 2 3 4 5 6

Years
Figure 54.1. Cumulative risk of nonfatal myocardial infarction or death from coronary artery disease according to the number of metabolic syn-
drome factors (0–5) among participants in the WOSCOPS.
 54 Metabolic Syndrome 533

Table 54.2. Criteria for Clinical Diagnosis of Metabolic Syndrome According

to 3 Definitions
Clinical Measure Whoa,b NCEP ATP IIIc,d AHA/NHLBIe,f
Waist circumference ... Same as ATP III
Men ≥102 cm (40 in)
Women ≥88 cm (35 in)
BMI >30 ... ...
Triglycerides ≥150 mg/dL Same as WHO Same as WHO
HDL-C Same as ATP III
Men <35 mg/dL <40 mg/dL
Women <39 mg/dL <50 mg/dL
Blood pressure ≥140/90 mm Hg ≥130/85 mm Hg Same as ATP III
Glucose IGT, IFG, or T2D Fasting >110 mg/dL Fasting ≥100 mg/dL
Insulin resistance Yes No No
Microalbuminuria Yes No No
Abbreviations: AHA, American Heart Association; ATP III, Adult Treatment Panel III; BMI, body
mass index (calculated as weight in kilograms divided by height in meters squared); HDL-C, high-
density lipoprotein cholesterol; IFG, impaired fasting glucose; IGT, impaired glucose tolerance;
NCEP, National Cholesterol Education Program; NHLBI, National Heart, Lung, and Blood Institute;
T2D, type 2 diabetes mellitus; WHO, World Health Organization.
a
Alberti et al. Diabet Med. 1998 Jul;15(7):539–53.
b
WHO requires insulin resistance and 2 additional risk factors for diagnosis.
c
National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and
Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Circulation 2002 Dec
17;106(25):3143–421.
d
ATP III requires 3 of 5 risk factors for diagnosis.
e
Grundy et al. Circulation. 2005 Oct 25;112(17):2735–52. Erratum in: Circulation. 2005 Oct
25;112(17):e298.
f
AHA/NHLBI recommends that triglycerides, HDL-C, and blood pressure be considered abnormal
when drug treatment is prescribed.
Previously published. See “Credit Lines” section.

for 3 of 5 characteristics. Franco and colleagues tested several differ-
ent combinations in the Framingham Heart Study Offspring Study.
The most frequent triad—central obesity, high blood pressure, and
hyperglycemia—was present in 29.3% of the patients and yielded
significant hazard ratios of 2.36 for CV disease incidence and 3.09
for total mortality. The triad of low HDL-C, high blood pressure,
and high triglycerides was significant only for predicting CV disease
(hazard ratio, 1.94). None of the other possible triads were signifi-
cant predictors of either CV disease or total mortality in this patient
cohort.
Treatment Issues
• Does the treatment of metabolic syndrome differ substantially from
the treatment of the individual components? The answer to this ques-
tion appears to be “yes” according to the AHA/NHLBI scientific
statement “Diagnosis and Management of the Metabolic Syndrome,”
which recommends reducing global risk by treating metabolic syn-
drome patients with statins and aspirin, even though neither drug spe-
cifically addresses any of the 5 metabolic syndrome characteristics.
Metabolic Syndrome and Other CV Risk Factors
Whether C-reactive protein, another novel risk factor and marker
of inflammation, explains the excess CV risk associated with
metabolic syndrome also has been questioned. C-reactive protein
has been correlated with all the characteristics of metabolic syn-
drome. Analysis of the WOSCOPS data does show, however, that
metabolic syndrome and C-reactive protein level each predict
CV risk independently and that the excess CV risk among par-
ticipants with both metabolic syndrome and increased C-reactive
protein level was twice that conveyed by either of these 2 risk
factors alone. In other words, C-reactive protein level and meta-
bolic syndrome are additive to CV risk (Figure 54.2).
The relationship between metabolic syndrome and LDL-C
is complex. Increased LDL-C is not a characteristic of the
metabolic syndrome, and LDL-C levels are not strongly corre-
lated with the number of metabolic syndrome characteristics.
However, metabolic syndrome is associated with the presence
of small, dense LDL particles, and the number of metabolic syn-
drome characteristics correlated strongly with the total LDL
particle concentration in the Framingham Heart Study (Figures
54.3 and 54.4). LDL-C levels increased only in participants who
had 0 or 1 metabolic syndrome characteristics (Figure 54.3), but
LDL particle concentration progressively increased as the num-
ber of metabolic syndrome characteristics increased from 0 to 4
or 5 (Figure 54.4). Since each LDL particle carries 1 apo B-100
protein, the concentration of apolipoprotein B would also be
expected to increase as the number of metabolic syndrome char-
acteristics increases.
Other novel CV risk factors, such as leptin and adiponectin,
have been shown to be associated with metabolic syndrome.
• Metabolic syndrome and C-reactive protein level each predict CV
risk independently.
• Although LDL-C levels are not strongly correlated with the num-
ber of metabolic syndrome characteristics, metabolic syndrome is
associated with the presence of small, dense LDL particles, and the
number of metabolic syndrome characteristics correlates strongly
with the total LDL particle concentration.
Prevalence of Metabolic Syndrome
Epidemiologic investigations have shown that metabolic syn-
drome is widely prevalent in developed countries, especially
the United States. The NHANES data from 1988 to 1994 sug-
gest a prevalence of 26%, which is similar to that reported in
534 VI Coronary Artery Disease Risk Factors

14
MS present, CRP ≥3 mg/dL
Participants With Event, %

12

10
MS present, CRP <3 mg/dL
8 MS absent, CRP ≥3 mg/dL

6
MS absent, CRP <3 mg/dL
4

0
0 1 2 3 4 5 6

Years
Figure 54.2. Cumulative risk of nonfatal myocardial infarction or death from coronary artery disease according to the presence or absence of
metabolic syndrome (MS) and the C-reactive protein (CRP) level (<3 mg/dL or ≥3 mg/dL) among participants in the WOSCOPS.

the WOSCOPS, for which participants were recruited during
a similar time frame. Current estimates of the prevalence of
metabolic syndrome in the United States range as high as 35%.
The increase in metabolic syndrome is thought to be directly
related to the increasing prevalence of obesity. Although the
prevalence of metabolic syndrome increases with age, its pres-
ence in many adolescents in the United States is generally more
concerning to public health officials than its high prevalence in
older persons.
• The increasing prevalence of metabolic syndrome is thought to be
related to the increasing prevalence of obesity, especially in ado-
lescents and young persons.
Metabolic syndrome can be readily identified by nonphysician
health professionals using NCEP criteria. Nurses or other allied
health personnel working in the clinic setting can be trained to
measure waist circumference during measurement of weight and
blood pressure, and HDL-C, triglyceride, and fasting blood glu-
cose values can be determined from the laboratory record. The
patient’s data and the number of metabolic syndrome character-
istics should be made available to the physician for the history
and physical examination.
The ATP III includes metabolic syndrome as 1 of several risk
modifiers that can be used to lower targets for LDL-C treatment
among appropriate patients. The ATP III strategy for treating
metabolic syndrome itself is to encourage lifestyle modifica-
tions (diet, exercise, and weight loss) and then consider focused
treatment of the risk factors still not controlled by the lifestyle
changes. Use of strategies generally recognized to lower overall
CV risk—including lowering the LDL-C level with statin drugs,

160
Women
Men
140

120
LDL-C, mg/dL

100

80

60

40

20

0
0 1 2 3 4 5

No. of Metabolic Syndrome Characteristics

Figure 54.3. Average low-density lipoprotein cholesterol (LDL-C) levels according to the number of metabolic syndrome characteristics for men
and women in the Framingham Heart Study.
 54 Metabolic Syndrome
160
LDL Particle Concentration, nmol/L
Women
Men
140
120
100
80
60
40
20
0
0 1
No. of Metabolic Syndrome Characteristics
Figure 54.4. Average low-density lipoprotein (LDL) particle concentrations according to the number of metabolic syndrome characteristics for
men and women in the Framingham Heart Study.
low-dose aspirin, and fish oil—may be considered, depending on
the age and estimated CV risk of the patient.
No completed trial to date has specified metabolic syndrome
as an entry criterion. However, retrospective analyses of some
lipid-lowering trials, such as 4S, have shown that statin therapy
reduces risk for the subset of patients with metabolic syndrome
to an equal or slightly greater extent compared with patients
without metabolic syndrome.
Weight loss registries have shown that persons who lose more
than 10% of body weight and maintain the majority of the lost
weight long term generally have 3 characteristics: 1) they have
developed multiple behavior strategies for controlling calorie
intake, 2) they average more than 60 minutes of physical activity
per day, and 3) they weigh themselves regularly.
Patients with metabolic syndrome should be instructed in these
principles and advised to make gradual and permanent modifica-
tions in their lifestyles rather than resort to “crash” programs of
high-intensity exercise or fad diets. Reducing the intake of both
simple carbohydrates and saturated fats has been shown to pro-
mote weight loss. Targeting carbohydrate reduction while simul-
taneously encouraging increased saturated fat intake may produce
greater weight loss in the short term (ie, the first 6 months), but
the limited studies that are available suggest that this short-term
advantage is largely lost by 1 year. Further, many health profes-
sionals and agencies such as the AHA have raised concern about
the long-term health consequences of high-fat diets (fats being
what replace carbohydrates in some of the low-carbohydrate diet
plans). The degree of calorie reduction, the ability to maintain
the dietary changes long term, and high levels of physical activity
seem to be key factors in sustaining weight loss. Social support
and structure provided by various resources, including commer-
cial programs, also have been shown to be useful for long-term
maintenance of weight loss. Behavioral strategies, including stress
management, should be made available when possible.
Current guidelines for weight loss from groups such as
the American College of Sports Medicine now recommend
60 minutes or more of daily physical activity of at least mod-
erate intensity (such as brisk walking), which is more than the
535
2 3 4 5
30 minutes daily recommended for general CV health. Patients
may be counseled to develop a structured exercise program or
use strategies to increase overall daily physical activity. Wearing
a digital pedometer or personal activity monitor is a convenient
way of tracking overall physical activity and has been shown to
improve compliance with physical activity recommendations and
to increase weight loss. New and more sophisticated interactive
devices to measure and record caloric intake and physical activ-
ity are being introduced commercially.
Many physicians lack adequate training in nutrition and
dietetics or exercise science to actively prescribe dietary ther-
apy, exercise programs, and behavioral weight loss strategies to
patients in their office or hospital practice. Even if well quali-
fied, physicians generally are unable to devote adequate time to
patient education and counseling on these topics in the current
practice environment. However, the physician can and should
promote lifestyle change in the following manner:
1. Explain metabolic syndrome and its associated CV risk to affected
patients
2. Underscore the importance of lifestyle modification for treating
metabolic syndrome (having presumably identified patients who
have it)
3. Outline general strategies and help patients set realistic weight loss
goals
4. Provide materials such as pamphlets, lists of useful books and Web
sites, and information on services that might be available through
local hospitals and health clubs, reputable commercial weight loss
programs, and psychologists or counselors who have special interest
and expertise in weight loss
5. Arrange to have patients return at appropriate intervals to recheck
laboratory values and weight and to give positive reinforcement as
appropriate
Large group or hospital-based practices may even consider
it worthwhile to employ a dietitian or exercise specialist to pro-
vide advice and to have patients return at appropriate intervals
to review progress. Numerous studies on behavioral change have
identified feedback and frequent follow-up as factors that enhance
adherence to recommended lifestyle changes and, ultimately, are
536 VI Coronary Artery Disease Risk Factors
associated with outcomes such as weight loss, improved CV fit-
ness, or reduced blood pressure or blood glucose or lipid levels.
• Lifestyle changes, especially weight loss and regular exercise, are
key to reducing the CV risk associated with metabolic syndrome.
In the absence of satisfactory results with lifestyle modifica-
tion for weight control, bariatric surgery has been proved effec-
tive for resolving more than 95% of cases of metabolic syndrome.
Currently, however, only persons who are severely obese with a
BMI greater than 40, or persons with medically complicated obe-
sity and a BMI greater than 35, are generally considered for this
procedure. Studies on the effect of waist liposuction on metabolic
syndrome characteristics are limited and show conflicting results.
Weight loss drugs that are currently available, such as orlistat,
have limited benefit for improving the characteristics of meta-
bolic syndrome, although contemporary pharmacologic strate-
gies for weight loss lag far behind those for lipid management or
blood pressure control in terms of both efficacy and tolerability.
Pharmacologic control of the lipid variables of metabolic
syndrome (low HDL-C and high triglyceride levels) is possible
with drugs such as niacin or fibrates (or high-dose fish oil in the
case of high triglyceride levels), although data on risk reduction
with these agents, particularly in persons without documented
CV disease, are very limited. In the primary prevention Helsinki
Heart Study, gemfibrozil did not reduce mortality or CV event
rates in the cohort as a whole, although the subgroup with low
HDL-C and high triglyceride levels (ie, those most likely to have
metabolic syndrome) did experience a reduction in nonfatal CV
event rates. The lipid arm of the ACCORD trial showed simi-
larly favorable trends in treating diabetic patients who had low
HDL-C and high triglyceride levels with fenofibrate, but the
overall results were negative in terms of predefined end points.
In the FIELD trial, a study of fenofibrate therapy for patients
with type 2 diabetes mellitus, no overall benefit was shown. In
the VA-HIT, a secondary prevention trial in which 25% of par-
ticipants had diabetes and 25% had an impaired fasting glucose
level, gemfibrozil reduced the rates of CV events and strokes, but
no overall mortality benefit was found. The National Institutes of
Health AIM-HIGH study demonstrated improvement in HDL-C
and triglyceride levels but no incremental clinical benefit after
3 years of treatment with niacin. The use of niacin in the treat-
ment of hyperlipidemia is likely to be decreased as a result of the
findings of this study. The NCEP provides specific guidelines for
treatment of non–HDL-C, incorporating both HDL-C and tri-
glyceride abnormalities, but it recommends control of LDL-C as
the primary focus of lipid-lowering therapy and does not specifi-
cally favor one lipid-lowering drug over another for treatment of
non–HDL-C.
Treatment of an impaired fasting glucose level or prehyperten-
sive blood pressure with drugs is also controversial and generally
not included in current guidelines. Unlike with lipid treatment,
patients may become symptomatic if blood pressure or the blood
glucose level decreases below normal level; thus, treating border-
line levels of these characteristics pharmacologically may result
in discomfort to some patients. Drugs such as metformin and the
thiazolidinediones delay or prevent conversion from a state of
impaired fasting glucose to diabetes, although intensive behav-
ioral programs involving exercise and dietary changes generally
have been shown to be more effective.
The JNC-7 continues to use a systolic blood pressure of
more than 140 mm Hg or a diastolic blood pressure of more
than 90 mm Hg as an indication to start pharmacologic therapy.
Various guidelines advocate lower goals for blood pressure in
patients with diabetes and patients with CV disease. However, no
studies have addressed the benefits of treating prehypertensive
blood pressure in patients with metabolic syndrome. Diuretics
and β-blockers are 2 of the 5 classes of drugs listed by JNC-7 as
appropriate first-line therapy for stage 1 hypertension, although
there are some concerns about the use of these 2 drug classes as
initial therapy for patients with metabolic syndrome. Specifically,
both classes may increase blood glucose and triglyceride levels.
To summarize, metabolic syndrome as defined by the NCEP
is highly prevalent in the United States and other developed coun-
tries with market economies and conveys at least a moderate CV
risk. Metabolic syndrome can be diagnosed easily from common
laboratory measures, the usual vital signs, and waist circumfer-
ence. The proposed strategy for treating metabolic syndrome
is to promote lifestyle change (diet and exercise) and resultant
weight loss and then to consider individual treatment of specific
characteristics not controlled by lifestyle change. If the overall
CV risk is relatively high (eg, due to age, family history of CV
disease, and smoking), it may be appropriate to consider general
reduction of CV risk factors with a statin, low-dose aspirin, and
fish oil. Studies specifically evaluating various treatment strate-
gies for metabolic syndrome are in progress, but to date there are
no completed studies with published results.
Suggested Reading
Kahn R, Buse J, Ferrannini E, Stern M; American Diabetes Association;
European Association for the Study of Diabetes. The metabolic syn-
drome: time for a critical appraisal: joint statement from the American
Diabetes Association and the European Association for the Study of
Diabetes. Diabetes Care. 2005 Sep;28(9):2289–304.
National Cholesterol Education Program [database on the Internet].
[cited 2012 Feb 1]. Available from: http://www.nhlbi.nih.gov/about/
ncep/.
Abbreviations
AHA American Heart Association
ATP III Adult Treatment Panel III
BMI body mass index
CV cardiovascular
HDL-C high-density lipoprotein cholesterol
JNC-7 Seventh Report of the Joint National Committee on
Prevention, Detection, Evaluation, and Treatment of
High Blood Pressure
LDL low-density lipoprotein
LDL-C low-density lipoprotein cholesterol
NCEP National Cholesterol Education Program
NHANES National Health and Nutrition Examination Survey
NHLBI National Heart, Lung, and Blood Institute
WHO World Health Organization
Names of Clinical Trials
ACCORD Action to Control Cardiovascular Risk in Diabetes
AIM-HIGH Atherothrombosis Intervention in Metabolic Syndrome
With Low HDL/High Triglycerides: Impact on Global
Health
FIELD Fenofibrate Intervention and Event Lowering in Diabetes
4S Scandinavian Simvastatin Survival Study
VA-HIT Veterans Affairs High-Density Lipoprotein Intervention
Trial
WOSCOPS West of Scotland Coronary Prevention Study
 55

Pathogenesis of Atherosclerosis
JOSEPH L. BLACKSHEAR, MD, and BIRGIT KANTOR, MD, PHD

Arterial Anatomy, Physiology, and Pathology in elastin, are prominent. The abluminal boundary dividing the
media from the adventitia is the external elastic lamina.
Cross-sectional, histologic examination of arteries reveals
The arterial wall is normally nourished on the intimal side
three concentric regions: the intima, media, and adventitia
by oxygen diffusing through the endothelium into the subintima.
(Figure 55.1 A). The layers of the intima from the luminal to
In large species and in vessels with a wall thicker than 250 μm,
the abluminal boundaries include the endothelium (a surface
the diffusion of oxygen is insufficient to supply the media and
monocellular layer with extraordinary biochemical properties),
adventitia; to this end, vasa vasorum course into the adventitia.
the basement membrane, and the internal elastic lamina. The
Segmental heterogeneity is also present, with variable quantities
intima is normally extremely thin, with endothelial cells aligned
of networks of nutrient vasa vasorum, lymphatics, and nerves.
with the direction of flow, except in areas of flow disturbance,
Aside from vascular, lymphatic, or nerve cells, the adventitia
such as bifurcations or branches. Normal endothelium is inti-
comprises principally fibroblasts. Coronary artery vasa vasorum
mately involved in local vascular regulation. Endothelial cells
originate mostly from side branches of the artery; only approxi-
produce nitric oxide, which is antithrombogenic, antiprolifera-
mately 16% originate directly from the main lumen. Vasa vaso-
tive, and vasodilating. Normal endothelial cells can also syn-
rum may augment delivery of cellular elements to developing
thesize endothelin, which is a powerful vasoconstrictor and is
atherosclerotic lesions in response to chemical messages origi-
promitogenic, and the vasodilator prostacyclin. Numerous sur-
nating in the subintima. Although a high density of vasa vaso-
face receptors are present, including low-density lipoprotein
rum may reflect the increased blood supply needs of the growing
receptors.
plaque, other preclinical data indicate that vasa vasorum actually
In addition to the endothelial lining, normal intima may
initiate plaque growth.
include regions of “cushions” of smooth muscle cells more than
The prevalence of atherosclerotic lesions is highest in the
one or two layers thick. Cushions tend to occur at sites predis-
abdominal aorta, coronary arteries, femoropopliteal arteries,
posed to atherosclerosis, such as in branches, bifurcations, and
internal carotid arteries, and vertebrobasilar arterial regions.
the proximal segment of the left anterior descending coronary
Some arteries, such as the internal mammary artery or the radial
artery. These appear in infancy and are referred to as adaptive
artery, are rarely or never affected by atherosclerosis. Reasons
eccentric intimal thickening (Figure 55.2).
for the regional heterogeneity of the pathologic response to what
Atherosclerosis is a disease primarily involving pathologic
is a uniform systemic exposure to risk factors most likely include
changes in the intima, with reactive changes in the media and
a variation in local response to risk factor–mediated injury.
adventitia. The media is heterogeneous, being thin in conduct-
Hemodynamic stresses at branches and bifurcations may aug-
ance vessels, such as the aorta, and muscular in the coronary
ment mechanically mediated dysfunction of endothelium, and
arteries. Laminations of smooth muscle cells bounded by elas-
responses to this dysfunction may be different owing to differ-
tic membranes form the media, and interstitial spaces, also rich
ences in vasa vasorum density. Turbulent flow in such areas may
also decrease the binding of cells that might mediate vascular
Abbreviations and acronyms are expanded at the end of this chapter. repair, such as bone marrow–derived endothelial progenitor
537
538 VI Coronary Artery Disease Risk Factors

A B C

D E F

Figure 55.1. Elastic van Gieson Staining of Coronary Artery Cross Sections. A, Normal vessel from a 33-year-old woman. B, Stage II lesion with
subintimal cellular plaque. C, Stage III lesion with mostly acellular lipid pool. D, Complex, moderately obstructive grade IV plaque with evidence of
laminations in a 74-year-old woman. E, Grade V plaque with thin fibrous cap and >75% luminal compromise in a 69-year-old man. F, Total coronary
occlusion in a 62-year-old man. (Previously published. See “Credit Line” section.)

cells. Local differences in gene expression and inflammatory
reactions may affect the concentrations of growth factors and
adhesion molecules.
Cardiac Risk Factors
Major independent risk factors for the development of atheroscle-
rosis include elevated plasma concentrations of total cholesterol
and LDL-C, cigarette smoking, hypertension, diabetes mellitus,
advancing age, low plasma concentrations of HDL-C, and a fam-
ily history of premature coronary artery disease.
By consensus, the pathogenetic explanation for the develop-
ment of atherosclerosis is that cumulative exposure to risk fac-
tors over many years results in a cycle or series of episodes of
cyclical injury and repair, which culminate in clinical events,
including acute coronary thrombotic syndromes, angina,
intermittent claudication or acute arterial occlusion, ischemic
stroke, and development and rupture of abdominal aortic aneu-
rysm. This explanation is concordant with clinical experience
and is strongly implied by coronary artery pathology, in which
strata from event-and-repair cycles may occasionally be seen
as laminations within a single arterial segment (Figure 55.3).

Figure 55.2. Proliferation of Intimal Smooth Muscle Cells.

 55 Pathogenesis of Atherosclerosis
Acute coronary syndromes and exertional angina are the most
important clinical presentations of coronary artery disease. They
reflect two main aspects of the disease: 1) Unstable or ruptured
plaques are the main reasons for acute myocardial infarction and
unstable angina, although plaques responsible for acute clini-
cal syndromes are typically only moderately stenotic. 2) Stable
plaques cause exercise-induced chest pain if they occlude more
than 75% of the cross-sectional area; although they become
hemodynamically significant, they rarely lead to acute coronary
syndromes. Understanding these differences and identifying
patients or plaques at risk is currently the focus of intense clini-
cal and basic research.
In the ARIC study, the adjusted rate increases (due to being in
the highest vs the lowest quartile) of coronary artery disease and
stroke for the well-established, classical risk factors were 67%
for smoking, 65% for diabetes mellitus, 47% for hypertension,
and 59% for elevated LDL-C. Separate guidelines for manage-
ment of risk factors are described in detail in other chapters but
are reviewed briefly in this chapter because making therapeu-
tic decisions related to these risk factors is the main activity of
daily cardiovascular disease clinical practice. Several proposed
risk factors are under intense study but do not yet affect clinical
practice to a large degree.
Figure 55.3. Complex Coronary Plaque with Distinct Laminations
(Numbered 1–4), Suggesting Cycles of Progression.
539
Cholesterol
LDL-C can be clearly implicated in the pathogenesis of athero-
sclerosis. It is instructive to note that although human LDL-C
concentrations tend to cluster from 120 to 200 mg/dL and athero-
sclerotic disease is the main cause of death in humans, typical
nonhuman mammalian LDL-C levels are 10 to 60 mg/dL, and
atherosclerosis does not occur unless animals are fed high-fat
food. Epidemiologic human data link diets high in saturated fats
to the prevalence of atherosclerotic diseases in diverse human
populations and also support a direct relationship between cho-
lesterol levels and death from coronary artery disease. Relative
risk increases in a continuum as total cholesterol levels increase
from levels as low as 150 mg/dL. Unlike other risk factors (such
as diabetes, smoking, and hypertension), cholesterol is present
at the scene of the atherosclerotic plaque—mainly in the form
of oxidized LDL-C or cholesterol esters. In a series of pivotal
clinical trials, the use of HMG Co-A reductase inhibitors (ie,
statin drugs) were used in populations with or without prior cor-
onary artery disease and with or without pre-existing elevations
of LDL-C. Statin therapy reduced LDL-C by approximately 25%
to 40% and lowered the risk of coronary artery disease events by
approximately 30% (Figure 55.4). These trials have served as
final proof of the cholesterol hypothesis of atherogenesis, which
is now widely accepted.
The NCEP is responsible for digesting new trial data and
making therapeutic recommendations about treatment targets
for practitioners. Its most recently published guidelines support
manipulating LDL-C to less than 100 mg/dL in patients with
known coronary artery disease and now include an option to
manipulate LDL-C levels even lower (<70 mg/dL).
Hypertension
Each 20–mm Hg increase in systolic blood pressure or 10–mm
Hg increase in diastolic blood pressure doubles coronary artery
disease mortality and stroke mortality. Proposed mechanisms by
which hypertension promotes atherosclerosis include damage to
endothelial cells, which promotes activation and incorporation
of lipids into subintima, and stimulation of subintimal smooth
muscle cell proliferation. The Joint National Committee on
Prevention, Detection, Evaluation, and Treatment of High Blood
Pressure defines the following levels: normal blood pressure
(≤120/80 mm Hg), prehypertension (120–139/80–89 mm Hg),
stage I hypertension (140–159/90–99 mm Hg), and stage II
hypertension (>160 mm Hg systolic or <100 mm Hg diastolic).
Risk of adverse events from atherosclerosis is present even in
prehypertension. The treatment goal for most patients is less than
140/90 mm Hg, but more intensive blood pressure lowering is
usually desirable; a goal of less than 130/80 mm Hg is already
recommended in guidelines for patients with chronic kidney
disease or diabetes mellitus. Meta-analysis of treatment trials
suggests a 16% reduction in coronary artery events and mor-
tality attributable to treatment of mild or moderate high blood
pressure. This benefit is twice as strong in elderly as in younger
patients.
Diabetes Mellitus
In the NCEP report, diabetes mellitus is considered a coronary
artery disease equivalent, thereby elevating it to the highest risk
category. This classification is based on the observation that
patients with type 2 diabetes without prior history of myocardial
540 VI Coronary Artery Disease Risk Factors
LRC
10 11% LDL-C
19% CAD CARE
20%
24%
Risk reduction in CAD events, %
20
40
60
10
Decrease in serum concentration of LDL-C, %
Figure 55.4. Coronary Artery Disease (CAD) Risk Reduction in Four Placebo-Controlled Trials of Statin Drugs. CARE indicates Cholesterol
and Recurrent Events Study; LDL-C, low-density lipoprotein cholesterol; LRC, Lipid Research Clinics Trial; WOS, West of Scotland Study; 4S,
Scandinavian Simvastatin Survival Study.
infarction have the same risk of myocardial infarction (20%) and
coronary artery disease mortality (15%) as patients without dia-
betes who already had an infarction. Coronary atherosclerosis is
the cause of death in approximately three-fourths of diabetics.
Currently, 15 million Americans are known to have diabetes mel-
litus and this number is projected to double by 2030. For these
people, the risk of cardiac death and nonfatal myocardial infarc-
tion is increased 2 to 8 times, and the increase in risk appears to
occur even in the prediabetic phase. In addition to the increase
in mortality, complications associated with myocardial infarc-
tion, including postinfarction angina and heart failure, are more
likely. A possible contributory factor is that diabetic patients are
more likely to have multivessel disease and significantly fewer
coronary collateral vessels.
The prevalence of insulin resistance, which usually precedes
the onset of overt hyperglycemia, is expected to increase signif-
icantly as societal obesity continues to increase. The so-called
metabolic syndrome (ie, three of the following five compo-
nents are present: fasting hyperglycemia, hypertriglyceri-
demia, hypertension, waist circumference >88 cm in women or
>102 cm in men, and low HDL-C) is also increasingly preva-
lent and difficult to manage. The age-adjusted prevalence of the
metabolic syndrome is approximately 25%. In diabetic patients
without the metabolic syndrome, the prevalence of coronary
artery disease is 7.5%; in diabetic patients with the metabolic
syndrome, 19.2%.
The exact mechanisms underlying the accelerated progression
of atherosclerosis in diabetic patients are unclear. Patients with
type 2 diabetes have reduced myocardial flow reserve, a reflection
of impaired coronary vasodilator capacity and impaired angio-
genesis. Diabetes impairs endothelial function and enhances
monocyte adhesion to endothelium, one of the earliest events in
LDL-C
CAD WOS
26% LDL-C
31% CAD
4S
38% LDL-C
42% CAD
20 30 40
the genesis of atherosclerosis. Endothelial dysfunction appears
to be an inflammatory result of AGE formation activating the
proinflammatory transcription factor nuclear factor-κB, which
reduces nitric oxide availability. Although metformin and thi-
azolidinedione reduce endothelial dysfunction, it is unknown
whether such improvement leads to better outcomes.
In the early stages of diabetes, the number of vasa vasorum
significantly decreases even before atheromas develop. As plaque
size, inflammation, and lipid core size increase, vasa vasorum
rapidly proliferate.
Diabetes has several effects on platelet function and the coag-
ulation system that may contribute to coronary thrombosis. There
is increased platelet aggregation and activation and enhanced
binding of fibrinogen to the glycoprotein IIb/IIIa complex.
Diabetic patients also show an increase in the plasma concen-
tration of the cardiovascular risk factor fibrinogen. Fibrinolytic
activity is reduced, with increased plasma concentrations and
binding of PAI-1, which is also found locally in atherosclerotic
plaques of diabetic patients. Other aspects of plaque composition
may differ between diabetics and nondiabetics. Coronary artery
plaque from diabetic patients has greater amounts of lipid-rich
atheroma and more macrophage infiltration, both of which are
associated with a greater risk of plaque rupture.
There are several differences in the lipid profile between dia-
betics and nondiabetics that may contribute to the increase in
atherosclerosis and its complications. In type 1 diabetes, poor
glycemic control is characteristically associated with hypertrig-
lyceridemia and low serum HDL-C. Insulin resistance, relative
insulin deficiency, and obesity in type 2 diabetes are associated
with hypertriglyceridemia and high serum LDL-C and lipopro-
tein (a) values. HDL-C is decreased, reducing the capacity for
reverse cholesterol transport. This pattern can be detected even
 55 Pathogenesis of Atherosclerosis
before the onset of overt hyperglycemia. The LDL-C therapeutic
target for diabetic patents is less than 70 mg/dL, and the blood
pressure target is less than 130/80 mm Hg.
Levels of adiponectin, an adipocyte-derived peptide, cor-
relate with both insulin resistance and atherosclerosis. Low
adiponectin concentrations are associated with low HDL-C
concentrations. Human and animal studies have shown that
adiponectin improves insulin sensitivity, lipid profi les, and
levels of inflammatory markers, including CRP, TNF-α,
and IL-6.
Although good glycemic control reduces the risk of micro-
vascular complications and appears to be beneficial during acute
myocardial infarction, protection against macrovascular disease
has not been established in type 2 diabetes, suggesting that fac-
tors other than hyperglycemia are important.
Cigarette Smoking
Cessation of smoking reduces the risk of coronary artery disease
more than any other modifiable risk factor known. The risk is
reduced by slightly more than one-third compared with the risk
for subjects who continue smoking. Smoking at least one pack
of cigarettes per day increases the risk twofold to threefold, and
there is a synergistic increase in risk in women taking oral con-
traceptives. Smoking adversely affects endothelial function, pro-
motes platelet aggregation and monocyte adhesion, and reduces
endothelial nitric oxide production.
Other Cardiac Risk Factors
CRP, homocysteine, fibrinogen, D-dimer, lipoprotein (a), plasma
creatinine, intimal medial thickness, coronary calcium scores,
and systemic markers of infection have all been evaluated as pos-
sible potent atherosclerotic risk factors. Thus far, only CRP is
mentioned in published guidelines.
C-Reactive Protein
hsCRP is a systemic marker of inflammation in humans. It is
produced in the liver (also in atherosclerotic plaque and human
endothelium) and is augmented by the presence of macrophages
and IL-1 and IL-6. The concentration of hsCRP is increased in
type 2 diabetes, hypertension, sleep apnea, the metabolic syn-
drome, chronic kidney disease, and all inflammatory diseases.
It is an independent marker of risk of coronary events. At pre-
sent, measurement of hsCRP is recommended in patients with
intermediate risk of coronary events (ie, 10%-20% risk per
10 years) since an elevated level of hsCRP adds sufficient risk
to reclassify such patients as high risk and suggests a need for
a more aggressive lipid-lowering therapeutic target. In patients
with known coronary artery disease, hsCRP is also an independ-
ent prognostic marker for future myocardial infarction, death, or
restenosis after percutaneous coronary angioplasty and stenting.
Although considerable research has focused on interventions that
alter hsCRP—notably aspirin, peroxisome proliferator-activated
receptor γ agonists (thiazolidinediones), and statins—at pre-
sent, primary or secondary prevention therapies should not be
based on hsCRP levels but rather on 1) standard guidelines such
as those from the NCEP and the Joint National Committee on
Prevention, Detection, Evaluation, and Treatment of High Blood
Pressure and 2) the use of antiplatelet therapy based on standard
guidelines for the management of patients with cardiovascular
risk factors, stable angina, or recent acute coronary syndrome
541
after percutaneous coronary intervention or coronary artery
bypass grafting.
Early Processes
The earliest pathologic change of atherosclerosis is thickening
of the subendothelial intima. Smooth muscle cells in the intima
alter their phenotype and are distinguishable from the smooth
muscle cells in the medial layer. An early inflammatory reaction
of the endothelium (endothelial activation) is probably medi-
ated through classical risk factors, especially increased LDL-C
(Figure 55.5). Normally, circulating LDL-C binds to specific
receptors on the endothelium and transits through the intima and
media and finally into adventitial vasa vasorum or lymphatics.
Excess LDL-C accumulates in the subintima and binds to the
proteoglycan extracellular matrix in a process probably mediated
by apolipoprotein B. Circulating LDL-C resists oxidation, but
proteoglycan-bound LDL-C is exposed to oxidative enzymes.
Oxidized LDL-C products in the subintima are proinflammatory,
and inflammatory cascades are initiated in both the endothelium
and the subintima. Endothelial activation and LDL-C–receptor
binding density may also be enhanced by flow disturbance,
smoking, hypertension, and hyperhomocysteinemia.
Endothelial cells injured or “activated” by oxidized LDL-C
produce adhesion molecules, including VCAM-1 and chemo-
kines such as MCP-1, beginning the critical step of circulating
blood monocytes binding to the endothelium. Monocytes that are
attracted to and bind to endothelial cells are incorporated into
the subintima and differentiate into macrophages. Subintimal
macrophages scavenge oxidized LDL-C and may accumulate
cholesterol to the point that pathologic cytoplasmic droplets are
visible. Oxidized LDL-C also promotes further proliferation of
macrophages within the subintima. Coalescence of lipid droplets
within the macrophage cytoplasm results in the typical foam cell
of the early-stage atherosclerotic plaque (Figure 55.1 B and 55.6).
In this early stage, foam cells are capable of transporting choles-
terol back to the endothelial surface to be bound by circulating
high-density lipoprotein, which binds cholesterol and transports
it to the liver (ie, reverse cholesterol transport). However, if
the stimulus for atherogenesis is amplified, foam cells remain
in the subintima as repositories of cholesterol (Figure 55.1 C),
until they undergo apoptosis and release lipid into the extra-
cellular subintima, forming a mostly acellular subintimal lipid
pool (Figure 55.1 D). Amplification also occurs as macrophage-
derived inflammatory cytokines increase the numbers of endo-
thelial cell LDL-C receptors and increase the extracellular matrix
deposition and fibrosis of adjacent smooth muscle cells. Matrix
metalloproteinase enzymes are also promoted, and these digest
membranes, including internal and external elastic lamina, and
also stimulate a proliferative and angiogenic response from the
media and adventitia.
Experimental data suggest that adventitial fibroblast activity
is strikingly increased by endothelial or adventitial injury, stim-
ulating a migration from adventitia toward media and intima.
This migration is fostered by matrix metalloproteinases, which
are locally controlled by cytokines and growth factors, includ-
ing PDGF, bFGF, TGF-β, and plasmin. TGF-β importantly
influences synthesis or inhibition of other growth factors. In the
media, fibroblasts interact further with vascular smooth muscle.
Further migration of adventitial fibroblasts through the media to
a subendothelial location has been documented experimentally.
Apoptosis, cell proliferation, phenotypic alterations in medial
vascular smooth muscle cells, and scarring all appear to be
 Low LDL receptor affinity
Vessel
Dense Prolonged
lumen
LDL half-life Dense
LDL
Penetration

EC EC EC

Dense
LDL
Arterial
intima

Accelerated
oxidation

Proatherogenic factors
Macrophage Foam cell
uptake formation Proinflammatory factors
Prothrombogenic factors
Cholesterol
accumulation

Figure 55.5. Pivotal Role of Low-Density Lipoprotein Cholesterol (LDL-C) in the Genesis of Early Atherosclerosis Through Subendothelial
Conversion to Oxidized LDL-C with Macrophage Stimulation and Activation. EC indicates endothelial cell; LDL, low-density lipoprotein.

Figure 55.6. Foam Cells in Atherosclerosis.

 55 Pathogenesis of Atherosclerosis 543

controlled by the mixture of matrix growth factors, cytokines,
metalloproteinases, tissue inhibitors of metalloproteinases, and
oxidized LDL-C. The adventitial vascular layer simultaneously
thickens, fostering vasa vasorum growth into the expanding
atherosclerotic plaque.
Additional proinflammatory cytokines are numerous in
growing plaques. Endotoxin and heat shock protein bind to acti-
vated macrophages. T cells of the CD4+ type are recruited, and
the interaction between macrophages, CD4+ cells, and vascu-
lar smooth muscle cells results in elaboration of TNF, IL-1, and
interferon-γ. Activated T cells also stimulate angiogenesis, medi-
ated at least in part by VEGF, which is one of many signals for
vasa vasorum proliferation. Several key observations specifically
link CD4+ cells and angiogenesis. First, in mice in which the
CD4 system is knocked out, ligation of a hind limb artery is not
followed by the usual formation of collateral circulation. Second,
when apolipoprotein E knockout mice, which phenotypically
manifest hypercholesterolemia and accelerated atherosclerosis,
are mated with CD4 knockout mice, the atherosclerotic lesion
becomes smaller. Transfusion of the missing CD4+ cells results
in greater atherosclerotic lesion growth. Finally, polymorphisms
in the genes encoding the expected products of CD4+ activation,
such as TNF and interferon-γ, also result in altered phenotypic
expression of the atherosclerotic process.
Coronary calcification occurs in vesicles within the
extracellular matrix. Calcification occurs in collagen- and
noncollagen-associated organic matrix. Noncollagenous bone-
associated proteins that foster calcification, such as osteopontin,
are expressed by macrophages in human intima, but expression
has also been noted in smooth muscle cells and adventitia. One
other mechanism for calcium deposition is intraplaque hemor-
rhage, possibly from immature vasa vasorum with degradation of
phosphorus and calcium-rich membranes derived from red blood
cells. Accumulation of calcium is typical for less vascularized
lesions with small or no lipid cores.
Arterial remodeling occurs in two forms in response to ath-
erosclerotic plaques: 1) Positive remodeling, called the Glagov
effect, is a compensatory increase in local vessel size in response
to a plaque burden not exceeding 40% of the luminal area.
2) Negative remodeling refers to the local shrinkage of vessel
size, which can occur after balloon angioplasty without stenting.
Vasa Vasorum, Angiogenesis,
and Neovascularization
It is not fully understood why a coronary lesion progresses from
an asymptomatic fibroatheromatous plaque to a lesion at high
risk of rupture with a thin, fibrous cap and necrotic lipid-rich
core. Recent research suggests that the vasa vasorum and neovas-
cularization have roles during this development (Figure 55.7 A).
At sites of plaque formation there is proliferation of vasa
vasorum from the adventitial side into the abluminal side of the
plaque. In regions in which new vessels penetrate into the plaque,
there are collections of extravascular macrophages, T lympho-
cytes, and erythrocytes. Macrophages ingesting erythrocyte
membrane are also noted.
Some data indicate that the quantity of vasa vasorum may
be associated with symptomatic disease and more macrophages.
There is a clear correlation between an increased number of vasa
vasorum and plaque size, similar to the relationship between ves-
sel wall thickness and vasa vasorum density. Recent data also
demonstrate a close correlation between vasa vasorum density
and the size of the inflammatory infiltrate. In ruptured or unsta-
ble plaques, the necrotic core occupies 30% to 50% of the total

Figure 55.7. Microscopic Computed Tomographic Images of Coronary Arteries with Vasa Vasorum (arrows). A, Vasa vasorum of a normal coro-
nary artery run longitudinally and across the main artery lumen (L). B, Vasa vasorum of a coronary artery during early-stage diabetes before athero-
sclerotic plaques develop. The number and density of vasa vasorum are markedly diminished in response to hyperglycemia. Artery diameter, 2 mm.
544 VI Coronary Artery Disease Risk Factors
plaque area, whereas it occupies less than 20% in the majority of
stable plaques. This suggests that progressive expansion of the
necrotic core precedes plaque rupture.
As the plaque enlarges, subsequent hypoxia and inflammatory
cell infiltration promote neovascularization of vasa vasorum,
which further adds to the total plaque volume. Exposure to this
abnormal environment stimulates rapid and abnormal vascular
development of the microvessels characterized by disorganized
branching and formation of loops with “leaky,” imperfect endo-
thelial linings. The entrance of vasa vasorum into plaques occurs
at points adjacent to the necrotic plaque core through defects in
the medial layer. Intra-adventitial and intramedial vasa vasorum
are relatively mature; some contain cuffs of smooth muscle cells,
but proliferating vasa vasorum closest to the arterial lumen are
usually thin, immature, and tortuous. Although endothelialized,
they typically lack smooth muscle cells. These immature blood
vessels are inherently leaky and permit extravasation of eryth-
rocytes. This hemorrhage into the plaque yields rapid changes
in plaque size and composition and may promote the transition
from a stable lesion to an unstable lesion. Indeed, autopsy studies
indicate that intraplaque hemorrhage is more frequent in patients
dying of ruptured plaques than in patients with stable athero-
sclerotic lesions (Figure 55.8). Erythrocyte membranes are also
a potent source of additional cholesterol, particularly in hyperc-
holesterolemic patients. Extravascular erythrocytes in the plaque
region further stimulate chemoattraction of macrophages that
digest erythrocyte-bound cholesterol, which further increases the
size of the lipid pool after intraplaque hemorrhage has occurred.
Histologic Typology and Contents
The Committee on Vascular Lesions of the Council on
Arteriosclerosis of the American Heart Association has gener-
ated a histologic classification of atherosclerotic lesions. These
build on an earlier classification by Stary (Figure 55.9). The
early type I lesion is characterized by pathologic intimal thick-
ening and subintimal scattered foam-cell macrophages. Type II
lesions also incorporate lipid-laden smooth muscle cells in addi-
tion to foamy macrophages. These lesions are macroscopically
visible as arterial fatty streaks. Type III lesions have an increase
in smooth muscle cells and volumes of surrounding connective
tissue matrix plus collections of extracellular lipid droplets or
small lipid pools. The main molecular constituents of plaques in
Figure 55.8. A, High-power (×50) view of severely stenotic plaque with thin fibrous cap and intraplaque hemorrhage. B, Lower-power view of
intraplaque hemorrhage in a human right coronary artery.
types I through III are collagen, proteoglycans in the extracel-
lular matrix, crystalline cholesterol, cholesterol esters and other
phospholipids, and cellular components and remnants, including
macrophages, T lymphocytes, and smooth muscle cells. In more
complex plaque, components of thrombus, including platelets,
fibrin, and red blood cells, may be present, suggesting an impor-
tant role for repetitive stimuli of either erosion of the plaque sur-
face with hemorrhage or hemorrhage from vasa vasorum within
the plaque in the intermediate-term progression of the athero-
sclerotic process (Figures 55.1 E, 55.7 B, and 55.9).
In type IV lesions, extracellular lipid droplets have pooled
to create a large confluent extracellular lipid core (Figures 55.1
E, 55.9, and 55.10). When such a core is surrounded by a thick
fibrous cap, the lesion is referred to as type V (Figure 55.11); if
heavy calcification is present, type Vb; and if calcium is largely
absent, type Vc. Type VI lesions demonstrate either a disruption
of the fibrous cap with fissure (Figures 55.12 and 55.13) or hema-
toma (Figure 55.8 B) or a thrombus (Figure 55.14). Spontaneous
disruption of the cap of an unstable plaque is probably the culmi-
nating event causing acute myocardial infarction. Intense investi-
gation into the risk factors, mechanisms, and possible therapeutic
approaches to this specific occurrence occupy a considerable
space in the national effort to combat cardiovascular disease.
Vulnerability and Rupture of Plaque
With expansion of the lipid core of type IV plaques and accu-
mulation of macrophages, especially at the luminal edges of the
plaque, risk of rupture of the fibrous cap and exposure of the
thrombogenic contents of the plaque are increased. Comparison
of the fibrous caps in plaques that have ruptured and those
that have not ruptured indicates that fibrous cap thickness less
than 65 μm and the presence of more than 25 macrophages per
high-power field (0.3-mm diameter) increase the risk of plaque
rupture. This vulnerability appears to be more likely in moder-
ately stenotic lesions than in severely stenotic lesions, at least
as assessed by serial coronary angiography, and theoretically
there may be stronger physical forces at the edges of a moder-
ately stenotic plaque than anywhere near the surface of a subtotal
occlusion (Figure 55.15). In subjects dying of acute myocardial
infarction, the degree of infiltration of plaque by inflammatory
cells is not limited to the lesion causing the fatal infarct—rather,
greater inflammation is seen in arterial lesions elsewhere in
 55 Pathogenesis of Atherosclerosis 545

BIF

Atheroma

Stary IV

LAD 2
Stary III

LAD 3
Stary I-II

Figure 55.10. Highly Stenotic Coronary Plaque With a Vulnerable,

Relatively Thin Fibrous Cap Covering a Lipid Pool. This is a coronary
Figure 55.9. Stary stages of early atherosclerosis: I, focal intimal plaque considered at high risk for spontaneous rupture.
thickening due to smooth muscle cell proliferation as seen in human
infants; II, fatty streaks composed of subendothelial foam cells with
most lipid droplets being intracellular; III, more space is occupied by a
mostly extracellular lipid pool; IV, mixture of cellular and acellular lipid
pool with a fibrous cap of variable thickness. BIF indicates bifurcation;
LAD, left anterior descending coronary artery. (Previously published.
See “Credit Line” section.)

the coronary tree of these subjects, suggesting a diffuse effect

of risk factors on the vascular target. In addition to thinness of
the cap and degree of inflammatory infiltrate, vasa vasorum are
increased approximately fourfold in ruptured plaques and two-
fold in those with vulnerability defined by cap thinness and
inflammatory density over plaques that were not associated with
acute clinical presentation or were severely stenotic and associ-
ated with stable angina.

Thrombosis
Rupture of plaque or superficial erosion exposes the bloodstream
to thrombogenic stimuli. Superficial erosions or disruptions of
fibrous plaque surface attract platelet adhesion and formation of
a monolayer of platelet thrombus, with platelets binding to col-
lagen and wall-bound von Willebrand factor. Platelet phospho-
lipids on the newly damaged arterial surface lead to activation
of circulating coagulation factors. The activation and amplifica-
tion of coagulation factors foster binding of additional platelets
to the propagating thrombus mass (Figure 55.16). This process Figure 55.11. Highly Stenotic Plaque in a Human Left Anterior
involves the platelet glycoprotein IIb/IIIa receptor (subject to Descending Coronary Artery. Although a large lipid pool is present,
pharmacologic intervention in percutaneous coronary interven- there is a very thick fibrous cap. This is a coronary plaque considered at
tion and unstable angina) and fibrinogen. Endogenous inhibitors low risk for spontaneous rupture.
546 VI Coronary Artery Disease Risk Factors
Figure 55.12. Acute Plaque Rupture from Center of Plaque.
balancing the propagating thrombus include the fibrinolytic sys-
tem, antithrombin III, proteins C and S, and the tissue factor
inhibitor pathway. In superficial vascular injury, thrombin and
erythrocyte incorporation into the thrombus occur only after
total occlusion by the platelet plug. The process of occlusion may
be dynamic, with stuttering cycles of total or near-total occlusion,
Figure 55.13. Acute Plaque Rupture from “Shoulder” of Plaque.
distal embolization of platelet plugs by flowing blood, and refor-
mation of thrombus at the site of plaque rupture. The clinical
manifestations may include 1) sudden death due to ventricular
arrhythmias, bradyarrhythmias, or acute circulatory failure or
2) chest pain and classical ST-segment elevation myocardial
infarction, non–ST-elevation myocardial infarction, or unstable
angina. Such events may be clinically silent or misinterpreted in
as many as one-fourth of patients. If, instead of superficial ero-
sion, a deep arterial fissure or injury occurs, the thrombogenic
stimulus is accelerated by the presence of fatty core plaque com-
ponents, such as tissue factor, and an acute occlusive thrombus
may develop.
Stabilization and Regression
of Atherosclerotic Plaque
Therapeutic placebo-controlled studies and comparative studies
with different dosages and agents, mainly statins, have tested dif-
ferential intensities of lipid-lowering therapy in human subjects
who have acute coronary syndromes or manifestations of chronic
atherosclerotic disease. With therapy, events continue to occur
during follow-up; this suggests the need for additional therapies
and risk factor modification for stabilization or regression of the
atherosclerotic process.
In the PROVE-IT trial, patients with acute coronary syndromes
(acute myocardial infarction or unstable angina pectoris) were
randomly allocated to high-dose statin therapy with atorvastatin
or moderate-dose therapy with pravastatin. Mean LDL-C levels
at baseline were 106 mg/dL; at 4 months, 60 mg/dL in the ator-
vastatin group and 97 mg/dL in the pravastatin group. Intensive
therapy resulted in a 28% relative risk reduction in both early and
 55 Pathogenesis of Atherosclerosis 547

A B

Figure 55.14. Occlusive Coronary Thrombi. A, In mildly stenotic underlying plaque. B, In severely stenotic plaque.

late adverse cardiac events, and the differences in the cumulative
incidence of subsequent events continued to widen from 30 days
through 24 months after entry. Trial data have not yet defined
the lower limit of optimal therapy nor described an increase in
risk from therapy. In general, other trials support the concept
that more intensive statin therapy is likely to reduce long-term
adverse events (Table 55.1).
In a study of different patients but with similar different inten-
sities of statin therapy, intracoronary ultrasonography was used
to measure changes in coronary plaque volume over time. After
intensive therapy (atorvastatin, 80 mg per day; LDL-C decreased
from 150 to 79 mg/dL), plaque volume stabilized over 18 months;
after moderate-dose therapy (pravastatin, 40 mg per day; LDL-C
decreased from 150 to 110 mg/dL), plaque volume continued to

49% 29% 5%

13% 4%
Figure 55.15. Frequency of Sites of Tearing in Ruptured Plaques. Brown indicates fibrous tissue; tan, calcification; yellow, lipid pool.
548 VI Coronary Artery Disease Risk Factors

Figure 55.16. Physiology of Platelet Activation and Aggregation. ADP indicates adenosine diphosphate; GP, glycoprotein; vWF, von Willebrand
factor. (Previously published. See “Credit Line” section.)

increase. Thus, most clinicians seek to lower LDL-C to less than
100 mg/dL for patients at risk and to less than 70 to 75 mg/dL for
those with known disease. Mechanisms proposed for this con-
tinuum of reduction of clinical events include reduction in accu-
mulation of subendothelial oxidized LDL-C, with subsequent
reduction in signaling for monocyte adhesion and incorporation
into plaque. Reduced macrophage proteolytic activity allows for
increased interstitial collagen formation that promotes stability
of the plaque fibrous cap.
An additional means of assessing plaque stabilization or
regression under active investigation involves the use of recom-
binant APoA-I Milano-phospholipid complex. A spontaneous

Table 55.1. Comparative Statin Trials and Frequency of Cardiovascular Events

Trial Setting Therapies Outcome
A to Z ACS Simvastatin (40–80 mg/d vs 20 mg/d) Nonsignificant trend of fewer CV events
TNT Stable CAD Atorvastatin (80 mg/d vs 10 mg/d) 22% risk reduction in CAD death, nonfatal MI, stroke,
or resuscitated cardiac arrest
PROVE-IT—TIMI 22 ACS Atorvastatin (80 mg/d) vs pravastatin (40 mg/d) 16% risk reduction in death and major cardiac events
IDEAL Prior MI Atorvastatin (80 mg/d) vs simvastatin (20 mg/d) Nonsignificant trend of fewer CAD deaths, nonfatal MIs,
and resuscitated cardiac arrestsa
Abbreviations: ACS, acute coronary syndromes; A to Z, Aggrastat to Zocor; CAD, coronary artery disease; CV, cardiovascular; IDEAL, Incremental Decrease
in Endpoints Through Aggressive Lipid Lowering; MI, myocardial infarction; PROVE IT—TIMI 22, Pravastatin and Atorvastatin Evaluation and Infection
Therapy—Thrombolysis in Myocardial Infarction 22; TNT, Treating to New Targets.
a
Significant decrease in individual events, including nonfatal MI and revascularization.
 55 Pathogenesis of Atherosclerosis 549

A B

Figure 55.17. A, Fatty streaks in the abdominal aorta. B, Severe ulcerocalcific atherosclerosis of the abdominal aorta. C, Aneurysmal disease of
the aortoiliac trunk.
550 VI Coronary Artery Disease Risk Factors

single amino acid mutation in apolipoprotein A-I appears to the most severe bulky plaque formation. The most frequent site
convey protection from atherosclerosis and is associated with of atherosclerotic aneurysmal disease is the infrarenal abdomi-
longevity. In a multicenter double-blind randomized trial, exog- nal aorta (Figure 55.17 A–C). An association of abdominal aor-
enous administration of this compound after five weekly treat- tic aneurysm with diseases such as polycystic kidney disease
ments resulted in a significant regression of atheromatous plaque and chronic obstructive pulmonary disease suggests that matrix
volume as assessed by intravascular ultrasonography. The theo- degeneration is a common feature. In the coronary arteries,
retical basis for regression cited by the authors was enhancement the laminated structure of the media is preserved despite ath-
of reverse cholesterol transport from plaque to liver by means of erosclerotic plaque development, and in the abdominal aorta
the apolipoprotein A-I complex. there is nearly complete destruction of the media and loss of
The PPAR is a nuclear receptor present in endothelium, smooth the collagen and elastin in the internal elastic lamina. Higher
muscle cells, macrophages, and T lymphocytes as well as in liver, matrix metalloproteinase activity in response to the inflamma-
skeletal muscle, and adipose tissue. Stimulation of PPAR recep- tory atherosclerotic stimulus is suspected as the cause of media
tors improves insulin-mediated glucose disposal. Agonists of destruction. In the late 1960s, Wolinsky and Glagov observed
PPAR were initially introduced as therapies in diabetic patients that the abdominal aorta in humans lacks vasa vasorum in its
to improve insulin sensitivity. The thiazolidinediones are PPAR-γ outermost aspects and suggested that this may be one of the
agonists and are in clinical use. In patients with type 2 diabetes, reasons that the abdominal aorta is particularly vulnerable to
and even in nondiabetic patients, however, the agents appear to atherogenesis.
reduce markers of inflammation and increase HDL-C levels. In
subjects with hypertension or hypercholesterolemia but without
Abbreviations
diabetes, pioglitazone (a PPAR-γ agonist) reduced blood insu-
lin levels, improved insulin sensitivity, and improved endothelial AGE advanced glycation end-product
function, all changes that would be expected to favorably affect the bFGF basic fibroblast growth factor
risk of complications of atherosclerosis. The sole large clinical trial CMP-1 chemoattractant protein-1
evaluating outcomes, the PROactive study, evaluated pioglitazone CRP C-reactive protein
HDL-C high-density lipoprotein cholesterol
in type 2 diabetics with prior evidence of coronary or peripheral
hsCRP high-sensitivity CRP
atherosclerosis. Over nearly 3 years of follow-up, death, myocar- IL interleukin
dial infarction, or stroke was reduced on average by 16%. Therapy LDL-C low-density lipoprotein cholesterol
is probably most beneficial in patients with type 2 diabetes who NCEP National Cholesterol Education Program
have evidence of atherogenic dyslipidemia (small dense LDL-C PAI-1 plasminogen activator inhibitor-1
and low total HDL-C). PPAR-a agonists are more potent, and PDGF platelet-derived growth factor
dual-receptor PPAR agonists are being investigated clinically. PPAR peroxisome proliferator-activated receptor
Finally, new research is investigating a role for bone marrow– TGF-β transforming growth factor β
derived circulating endothelial progenitor cells in vascular heal- TNF tumor necrosis factor
ing. These cells may bind to injured arterial vessels and promote VCAM-1 vascular cell adhesion molecule-1
VEGF vascular endothelial growth factor
reestablishment of a healthy nonthrombogenic and antiathero-
genic endothelium. Regulatory processes affecting these cells
include aging and conventional cardiac risk factors, including
Names of Clinical Trials
diabetes and hyperlipidemia.
ARIC Atherosclerosis Risk in Communities
PROactive Prospective Pioglitazone Clinical Trial in Macrovascular
Atherosclerotic Aneurysms Events
The abdominal aorta is the target for the earliest manifestations PROVE-IT Pravastatin and Atorvastatin Evaluation and Infection
of systemic atherosclerosis—the fatty streak—and the site of Therapy
 56

Dyslipidemiaa
JOSEPH G. MURPHY, MD, THOMAS G. ALLISON, PHD, MPH,
and R. SCOTT WRIGHT, MD

Dyslipidemia According to National Lipid Association recommendations,

other patients potentially benefiting from advanced lipid test-
Abnormal cholesterol and lipoprotein metabolism, together with
ing include patients who develop atherosclerosis at a young
endothelial injury are the major progenitors of atherosclerosis,
age, patients who develop atherosclerosis with normal standard
the leading cause of obstructive coronary artery and vascular
lipid levels, and patients who develop recurrent coronary events
diseases worldwide. Cigarette smoking, hypertension, diabetes,
despite lipid levels which are well controlled according to cur-
and dyslipidemia are considered to be the major risk factors for
rent National Cholesterol Association guidelines. Lp(a) lipopro-
atherosclerotic cardiovascular disease, with dyslipidemia consid-
tein is also thought to play an important role in the development
ered an essential risk factor because atherosclerotic plaques are
of atherosclerosis, and its role in screening in selected patients,
formed in large part from cholesterol. Asian countries such as
particularly those with a family history of premature coronary
Japan and, until the introduction of the Western diet, China have
artery disease is endorsed by groups such as the National Lipid
long had low rates of coronary artery disease (despite high rates
Association and the European Atherosclerosis Society.
of hypertension and smoking), which were attributed to the low
In epidemiologic studies, a low serum LDL-C of less than
levels of serum cholesterol in the population.
75 mg/dL (≈ 2 mmol/L), typically found in vegetarians and those
Mean total serum cholesterol in a population is a good pre-
on non-Western style diets, protects against atherosclerotic car-
dictor of the overall incidence of cardiovascular disease in that
diovascular disease. Dietary saturated fat and cholesterol gen-
population, but individual patient management should always be
erally raise the level of LDL-C in the blood and proportionally
based on measurement of lipid subfractions including LDL-C,
raise the coronary risk. The related factors of obesity, metabolic
HDL-C, and triglycerides. Measurement by nuclear magnetic res-
syndrome, and type 2 diabetes may have only modest effects
onance spectroscopy of LDL particle number or measurement of
on LDL-C levels, but they increase coronary risk by increasing
apolipoprotein B concentration may add further clinically impor-
triglycerides and LDL particle concentration, while lowering
tant information in selected patient subsets, including patients
HDL-C and causing the production of small, dense LDL par-
with type 2 diabetes and patients with metabolic syndrome.
ticles. This condition is often called diabetic dyslipidemia or
high-risk dyslipidemia, as opposed to hyperlipidemia referring
a to high levels of LDL-C. An increase in carbohydrates and poly-
Portions previously published in Garza CA, Montori VM,
McConnell JP, Somers VK, Kullo IJ, Lopez-Jimenez F. Association unsaturated fats in the diet will lower LDL-C but generally will
between lipoprotein-associated phospholipase A2 and cardiovascular raise triglycerides and lower HDL-C. Monounsaturated fats have
disease: a systematic review. Mayo Clin Proc. 2007 Feb;82(2):159–65 a more neutral effect on HDL-C and triglycerides while lowering
and Singh RB, Niaz MA. Serum concentration of lipoprotein(a) LDL-C in comparison to high saturated fat intake.
decreases on treatment with hydrosoluble coenzyme Q10 in patients Screening of both symptomatic patients with cardiovascular
with coronary artery disease: discovery of a new role. Int J Cardiol. disease (secondary prevention) and asymptomatic individuals
1999 Jan;68(1):23–9. Used with permission. (primary prevention) are important strategies to reduce morbidity
Abbreviations and acronyms are expanded at the end of this chapter. and mortality of cardiovascular disease. The NCEP recommends
551
552 VI Coronary Artery Disease Risk Factors

that all adults over the age of 20 years be screened with a fasting
lipoprotein profile (total cholesterol, HDL-C, LDL-C, and trig-
lycerides) at least every 5 years.
Lipid levels are classified in Table 56.1 according to the
NCEP’s ATP-III.
The following diseases are considered equivalent to symp-
tomatic coronary disease for the purposes of dyslipidemia evalu-
ation and treatment.
• Diabetes mellitus—type 2.
• Symptomatic carotid artery disease.
• Symptomatic peripheral artery disease.
• Abdominal aortic aneurysm or significant aortic atherosclerosis.
• Multiple CAD risk factors that result in a 10-year Framingham risk
estimate of ≥10%.
Conditions not currently included in the list of CAD risk
equivalents presented in ATP-III but appear to convey a risk
equivalent to symptomatic CAD include:
• Chronic kidney disease (glomerular filtration rate less than
60 mL/min ≈ creatinine >1.5 mg/dL).
• End-stage renal disease (chronic kidney disease stage 5) patients
on dialysis.
• High level of coronary calcification identified on CT (no specific
guidelines given, but an absolute Agatston score of ≥300 or a score
>75th percentile for age might be considered to indicate the pres-
ence of significant atherosclerosis).
Conventional Risk Factors for Atherosclerotic
Cardiovascular Disease
• Cigarette smoking.
• Hypertension (BP ≥140/90 or on antihypertensive medication).
• LDL-C ≥130 mg/dL (≈ 3.4 mmol/L).
• Low HDL-C <40 mg/dL (≈ 1.0 mmol/L).
• Diabetes.
• Family history of premature CAD, defined as a first-degree male
relative with age at disease onset ≤55 years or female relative ≤65
years.
Blood pressure between 120/80 to 140/90, LDL-C between
100 and 130 mg/dL, and fasting blood glucose between 100 and
125 mg/dL are classified as prehypertension, near-optimal LDL-
C, and impaired fasting glucose, respectively, and warrant close
supervision, usually with lifestyle modification. Metabolic syn-
drome is not equivalent to type 2 diabetes in terms of coronary
risk, but would also fall into the category of conditions that war-
rant close supervision with lifestyle modification. ATP-III also
lists metabolic syndrome as a risk factor that warrants lowering
the goal for LDL-C from 100 to 70 mg/dL in patients with CAD.
Primer on Lipoprotein Metabolism
• Lipids are essential for human metabolism (energy storage and
production, cell membrane formation and steroid hormone and
bile salt production), but in their native state lipids are insoluble
in plasma.
• The primary neutral lipids involved in human metabolism are
unesterified cholesterol and triglycerides. These neutral lipids are
transported in plasma bound to proteins called apolipoproteins
which form soluble lipoproteins.
• Lipoproteins are a large family of particles that fulfill different
functions in lipid metabolism. They differ radically in their athero-
genic potential.
• The major lipoproteins are LDL, HDL, IDL, VLDL, and
chylomicrons.
LDL Subfractions
LDL-C is the body’s main delivery vehicle for transporting cho-
lesterol to body tissues where it is incorporated into biological
processes. The delivery of cholesterol is regulated primarily by
the expression of the LDL-r which binds the apolipoprotein B in
LDL and allows the exchange of cholesterol from the LDL sub-
fractions into tissue. The upregulation and downregulation of the
LDL-r is a highly conserved process that is an extremely efficient
mechanism for regulating cholesterol delivery into the body.
The liver also expresses the LDL-r as a mechanism to remove
cholesterol from the body.
An increased concentration of LDL is a well-known fac-
tor for development and progression of CAD. Large epidemio-
logical studies initially suggested an association between LDL
levels in populations and the risk of coronary heart disease. A
large number of clinical trials have also demonstrated that sta-
tins, which lower LDL, also reduce the risk of CAD and the

Table 56.1. Lipid-Lowering Therapy Recommendations From the National Cholesterol Education Program
Adult Treatment Panel III
Risk Category LDL-C Goal Initiate TLC Consider Drug Therapy
High risk: CAD or CAD risk equivalents a
<100 mg/dL ≥100 mg/dL ≥100 mg/dL (≤100 mg/dL optional) b
(10-y risk >20%) Optional goal: <70 mg/dL
Moderately high risk: ≥2 risk factorsc <130 mg/dL ≥130 mg/dL ≥130 mg/dL (100–129 mg/dL optional)
(10-y risk 10%–20%) Optional goal: <100 mg/dL
Moderate risk: ≥2 risk factors <130 mg/dL ≥130 mg/dL ≥160 mg/dL
(10-y risk <10%) Optional goal: <100 mg/dL
Lower risk: 0 or 1 risk factor 160 mg/dL ≥160 mg/dL ≥190 mg/dL (160–190 mg/dL optional)
Abbreviations: CAD, coronary artery disease; LDL-C, low-density lipoprotein cholesterol; TLC, therapeutic lifestyle changes.
a
CAD risk equivalents include history of peripheral vascular disease, abdominal aortic aneurysm, history of stroke, or diabetes mellitus.
b
Patients with recent myocardial infarction (MI), smoking and previous MI, diabetes mellitus and previous MI, or an acute coronary
syndrome.
c
Risk factors include cigarette smoking, hypertension, decreased high-density lipoprotein cholesterol (<40 mg/dL), family history of
premature CAD (first-degree male relative <55 years; female <65 years), and age (men 45 years or older; women 55 years or older). If the
triglyceride (TG) level is 200–499 mg/dL, consider fibrate or niacin after low-density lipoprotein-lowering therapy. If TG is ≥500 mg/dL,
consider fibrate or niacin before low-density lipoprotein-lowering therapy. Consider use of omega-3 fatty acids as adjunct therapy for high TG
levels.
 56 Dyslipidemia 553

Percent With CAD Event

PROVE IT
25
4S
HPS Secondary
20
Mixed
3.0
15 PROSPER Primary
LIPID 2.5
CARE
10 WOSCOPS 2.0
TNT
ASCOT-LLA
1.5 25
5
JUPITER 1.0 45
AFCAPS
0 0.5 65 HDL-C
50 70 90 110 130 150 170 190 210 85 (mg/dL)
0
LDL-C (mg/dL) 100 160 220
LDL-C (mg/dL)
Figure 56.1. Major statin trials showing the reduction in cardiovas-
cular risk according to reduction in low-density lipoprotein cholesterol Figure 56.2. Relationship of coronary artery disease risk to high-
(LDL-C). density lipoprotein (HDL) cholesterol stratified by level of low-density
lipoprotein cholesterol. (Previously published. See “Credit Lines”
section.)
morbidity and mortality of CAD in those who already suffer
from it (Figure 56.1).
LDL is a heterogenous collection of lipoproteins and can be The removal of plaque from vascular tissues by HDL subfrac-
separated into subtypes of different size and density. However, tions is called RCT. Studies designed to exploit HDL’s role in
not all LDL particles have the same atherosclerotic potential. RCT continue to be an ongoing area of clinical interest.
LDL can be classified into three classes: large fluffy or buoy- HDL can also bind to cholesterol released during lipolysis of
ant LDL, intermediate LDL, and small dense LDL. LDL size triglycerides and chylomicron remnants. This allows for a stable
is inversely related to atherosclerotic risk. The major difference repository for neutral lipids like cholesterol and highlights the
among these classes is the ratio of cholesterol molecules to apoli- body’s efficient storage and manipulation mechanisms.
poprotein B. Large buoyant LDL-C has the greatest cholesterol–
• An HDL level of 50 mg/dL (≈ 1.3 mmol/L) or greater in a man or
apolipoprotein B ratio, and small dense LDL-C has the lowest
60 mg/dL (≈ 1.6 mmol/L) in a woman is associated with a reduced
cholesterol–apolipoprotein B ratio. risk of atherosclerosis.
• LDL particle size is inversely related to atherosclerotic risk. • An HDL level of over 75 mg/dL (≈ 2 mmol/L) or greater in a man
or woman is associated with a very low risk of atherosclerosis.
• An HDL of less than 40 mg/dL (≈ 0.8 mmol/L) in a man and less
High-Density Lipoprotein
than 50 mg/dL (≈ 1 mmol/L) in a woman increases the risk of
HDL is synthesized in both the liver and the intestine as small atherosclerosis.
nascent HDL particles that contain minimal cholesterol. HDL • Low HDL levels appear as an independent risk factor for CV events
contains apolipoprotein A-I as its major surface protein compo- when combined with high LDL levels.
nent; some forms of HDL contain apolipoprotein E. • The JUPITER trial suggested that statin treatment in primary
HDL in the blood exists in small, intermediate, and large- prevention trials that reduced to LDL levels to <60 mg/dL was
sized particles, designations which reflect their cholesterol associated with no additional risk with low HDL values. This
content. The smallest form of HDL is a discoid shaped particle observation needs to be replicated in other ongoing studies before
often called pre-Beta HDL. Pre-Beta HDL has the most affinity it can be widely applied to patient populations.
for cholesterol and removes cholesterol from body tissue through
binding the ABC A-1 and ABC G-1 receptors. The uptake of
cholesterol results in an increase in the size of the HDL particles
such that they seek to transfer the cholesterol out and be recycled
into the pre-Beta form. The HDL particles typically bind the
CETP or the scavenger receptor B-1 on hepatocytes (SRB-1) and
transfer the cholesterol off. The binding of HDL to CETP results
in a transfer of neutral lipids to atherogenic particles such as
LDL, IDL, and VLDL. This recycling of cholesterol is a highly
preserved mechanism which allows most mammals to exist with-
out eating exogenous cholesterol. It is estimated that ∼30% of
LDL is maintained through this pathway. The binding of HDL to
the SRB-1 receptor allows free cholesterol to be removed from
the body and excreted via the biliary pathway. HDL itself can be
removed via the kidney as well. The ability of HDL to remove
cholesterol from body tissues like vascular plaque has led many
to hypothesize that HDL has a protective effect against athero-
sclerosis. A large body of epidemiological data certainly sug-
gests an inverse relationship between HDL levels in populations
and risk of coronary heart disease, and that this risk is indepen- Figure 56.3. Trials studying high-density lipoprotein cholesterol
dent of LDL cholesterol (Figures 56.2. and 56.3). levels.
554 VI Coronary Artery Disease Risk Factors

• Apolipoprotein A-I Milano is a naturally occurring mutant of apo- Diabetes Mellitus

lipoprotein A-I, found in a fortunate Italian family. Individuals with
this mutation have very low HDL cholesterol levels, but paradoxi-
cally are protected against atherosclerosis likely due to the enhanced
ability of the mutant HDL to promote cholesterol efflux from cells.
• CETP is a plasma protein that facilitates the exchange of cholesteryl
esters and triglycerides between the lipoproteins. Triglycerides from
LDL are exchanged for a cholesterol moiety from HDL. This pro-
cess is proatherogenic as HDL-C is decreased, while the LDL parti-
cles become smaller and denser with increased cholesterol content.
Lp(a) Lipoprotein
Lp(a) lipoprotein is composed of LDL (apolipoprotein B-100)
and a low-molecular-weight glycoprotein called apolipoprotein
a, which shows genetic size polymorphism. It is attached by a
disulfide bond. Thirty-four isoforms of apolipoprotein a have
been identified, and studies have indicated that the size of apo-
lipoprotein a is inversely associated with the Lp(a) lipoprotein
levels in plasma. Studies indicate that Lp(a) lipoprotein is an
independent risk factor of premature CAD. Serum levels of Lp(a)
lipoprotein have been shown to correlate well with the presence,
severity, and lesion score with coronary angiography as well as
with the occurrence and recurrence of myocardial infarction and
cardiac death. Also, levels of Lp(a) lipoprotein are increased
in response to pregnancy, in advanced malignancy, and in end-
stage renal disease. Apolipoprotein a has considerable homol-
ogy with plasminogen and interferes with fibrinolysis and may
predispose to thrombosis. Serum Lp(a) is largely unaffected by
environmental factors, and maximal levels are reached early in
infancy, which predisposes individuals to CAD at a younger age.
Additionally, it is important to screen for increased Lp(a) lipo-
protein in patients in whom statin treatment does not lower the
LDL-C to the desired target level because Lp(a) lipoprotein can
inhibit LDL clearance.
A significant component of the risk associated with type 2 diabetes
may be related to its characteristic lipid triad profile of increased
levels of small dense LDL particles, decreased levels of HDL, and
increased levels of triglycerides. When type 2 diabetes mellitus is
diagnosed, many patients already have early or advanced athero-
sclerosis. This observation suggests that the atherogenic effect of
metabolic dysregulation starts many years before glucose levels
become significantly elevated. Type 2 diabetes mellitus leads to
a higher risk of cardiovascular events in women than in men, and
it has a multiplicative effect with smoking. The INTERHEART
study has confirmed the independent nature of type 2 diabetes as
a significant risk factor for the development of myocardial infarc-
tion. The incidence of diabetes is increasing due to the Western
lifestyle, especially decreased physical activity and increased
carbohydrate intake leading to weight gain.
Metabolic Syndrome
Metabolic syndrome defines a complex metabolic disturbance
characterized by a clustering of risk related to decreased HDL-C
levels and increased levels of one or more of the following: insu-
lin, fasting blood glucose, visceral fat, triglycerides, and blood
pressure (Table 56.2). It is estimated that up to 40% of the US
adult population has metabolic syndrome. Subjects with this con-
dition are two to three times more likely to develop CAD or to
die of a cardiovascular cause; the risk is similar to smoking two
packs of cigarettes a day and significantly higher than each of
the metabolic syndrome criteria alone. Thus, there appears to be
synergy between components of metabolic syndrome in causing
cardiac and vascular disease. Metabolic syndrome also precedes
type 2 diabetes mellitus. Recent data show that metabolic syn-
drome is highly prevalent and associated with a worse prognosis
in patients with an acute myocardial infarction.

Table 56.2. American Heart Association Criteria for Clinical Diagnosis of Metabolic
Syndrome
Criterion (Any 3 Constitute
Diagnosis of Metabolic Syndrome) Defi nition
Increased waist circumferencea,b ≥102 cm (≥40 inches) in men
≥88 cm (≥35 inches) in women
Increased triglycerides 150 mg/dL (1.7 mmol/L) or
Receiving drug treatment for increased triglyceridesc
Decreased HDL-C <40 mg/dL (<1.03 mmol/L) in men
<50 mg/dL (<1.3 mmol/L) in women or
Receiving drug treatment for decreased HDL-Cc
Increased blood pressure 130 mm Hg systolic blood pressure or
85 mm Hg diastolic blood pressure or
Receiving antihypertensive drug treatment with a history of
hypertension
Increased fasting glucose 100 mg/dL or
Receiving drug treatment for increased glucose
Abbreviation: HDL-C, high-density lipoprotein cholesterol.
a
To measure waist circumference, locate top of right iliac crest. Place a measuring tape in a horizontal plane
around abdomen at level of iliac crest. Before reading tape measure, ensure that tape is snug but does not
compress the skin and is parallel to floor. Measurement is made at the end of a normal expiration.
b
Some U.S. adults of non-Asian origin (eg, white, black, and Hispanic) with marginally increased waist
circumference (eg, 94–101 cm [37–39 inches] in men and 80–87 cm [31–34 inches] in women) may have a
strong genetic contribution to insulin resistance and should benefit from changes in lifestyle habits, similar to
men with categorical increases in waist circumference. A lower waist circumference cutoff (eg, ≥90 cm [35
inches] in men and ≥80 cm [31 inches] in women) appears to be appropriate for Asian Americans.
c
Fibrates and nicotinic acid are the most commonly used drugs for increased triglycerides and decreased
HDL-C. Patients taking one of these drugs are presumed to have high triglyceride levels and low HDL-C.
 56 Dyslipidemia
The mechanisms linking metabolic syndrome to cardiovascu-
lar disease have not been fully elucidated. By definition, patients
with metabolic syndrome are more likely to have an atherogenic
lipid profile because elevated levels of triglycerides and low levels
of HDL-C are two of the five diagnostic criteria. However, some
evidence suggests that the cardiovascular disease mechanisms
of metabolic syndrome go beyond the coexistence of hyperten-
sion or the atherogenic lipid profile. For example, patients with
metabolic syndrome have impaired fibrinolysis and increased
systemic inflammation, which are also characteristics of OSA.
Metabolic syndrome is discussed in detail in another chapter in
this text.
Novel Cardiovascular Risk Factors
C-Reactive Protein
Inflammation has a key role in the pathogenesis of cardiovascu-
lar disease, acute atherothrombotic events, and atherosclerosis.
Inflammation also regulates the production of the acute-phase
proteins such as CRP. High-sensitivity CRP is an independent
predictor of atherosclerosis, cardiovascular events, atherothrom-
bosis, hypertension, and myocardial infarction, even after con-
sidering other cardiovascular risk factors such as age, smoking,
obesity, diabetes, hypercholesterolemia, and hypertension. CRP
may be a causal factor and not just a marker of risk.
The current ACC/AHA recommendations for the use of CRP
in the prediction of the vascular risk state that patients with
an intermediate 10-year cardiac risk (between 10% and 20%)
using the Framingham score (Table 56.3) can be better stratified
using CRP. CRP and novel risk factors are discussed in detail in
another chapter in this text.
Homocysteine
Homocysteine is a non-essential sulfur-containing amino acid.
Several epidemiologic studies have linked hyperhomocysteine-
mia with an increased risk of CAD, although the findings are
inconsistent. Levels of homocysteine and cardiovascular disease
have also been linked to folate status. Hyperhomocysteinemia
may result in direct endothelial injury and a predisposition to
a prothrombotic state. Many patients with hyperhomocysteine-
mia and atherosclerotic vascular disease are also deficient in
Table 56.3. Framingham Risk Scorea,b
Risk Factor Points (Men) Points (Women)
Age group −9 to 13 −7 to 16
Total cholesterol 0 to 11 0 to 13
Smoking status 0 to 8 0 to 9
HDL-C −1 to 2 −1 to 2
Systolic blood pressure and 0 to 3 0 to 6
treatment status
Diabetes (yes/no) 0 to 2 0 to 2
Abbreviations: CAD, coronary artery disease; HDL-C, high-density lipoprotein
cholesterol; LDL-C, low-density lipoprotein cholesterol.
a
This update of the 1991 Framingham coronary prediction algorithm provides
estimates of total CAD risk (risk of developing one of the following: angina
pectoris, myocardial infarction, or death from coronary disease) over 10 years.
Separate score sheets are used for men and women. The factors used to estimate
risk include age, blood cholesterol (or LDL-C), HDL-C, blood pressure,
cigarette smoking, and diabetes mellitus. Relative risk of CAD is estimated by
comparison with low-risk Framingham participants.
b
This range of estimated 10-year risk by summing points is <1% to ≥30%.
555
vitamin B6 or B12. However, despite multiple epidemiologic and
basic research studies in patients with mild to moderate hyper-
homocysteinemia, results from recent clinical trials do not sup-
port supplementation with vitamin B6 and folic acid in patients
who have an elevated cardiovascular risk, such as patients with
a history of cardiovascular disease (SEARCH trial) or stroke.
Studies specifically targeting hyperhomocysteinemia have not
been conducted.
Lipoprotein-Associated Phospholipase A2
Lp-PLA2 has been proposed as a predictor of cardiovascular dis-
ease events. This enzyme is a member of the phospholipase A2
superfamily and is produced by monocytes, T lymphocytes, and
mast cells. In plasma, about 80% of Lp-PLA2 is bound to LDL, and
the remaining 20% is linked to HDL and very low-density lipopro-
teins. Lp-PLA2 has a role in the hydrolysis of oxidized LDL and
the resulting formation of lysophosphatidylcholine, a proathero-
genic and inflammatory mediator. Lysophosphatidylcholine
is an important chemoattractant for macrophages and T cells;
it induces migration of vascular smooth muscle cells, affects
endothelial function, and increases the expression of adhesion
molecules and cytokines. Also, several studies support an anti-
inflammatory function of Lp-PLA2. The protein has been shown
to have a role in the hydrolysis of the platelet-activating factor and
to manifest a possible antiatherogenic effect when high levels of
Lp-PLA2 are associated with HDL in mice.
Several epidemiologic studies have investigated the asso-
ciation between plasma Lp-PLA2 levels and the risk of subse-
quent cardiovascular disease events. Pooled evidence shows that
plasma Lp-PLA2 levels predict cardiovascular disease beyond the
predictions of traditional risk factors.
OSA and Ischemic Heart Disease
OSA activates multiple disease mechanisms associated with myo-
cardial ischemia and infarction. Nocturnal ST-segment changes
consistent with myocardial ischemia are evident even in patients
with OSA who do not have clinically significant coronary artery
disease. OSA may contribute to nocturnal angina, and ST-segment
depression during sleep appears to be related to the severity of
oxygen desaturation. Treatment with continuous positive airway
pressure attenuates nocturnal ST-segment depression.
Multiple cohort studies support the notion that OSA is associ-
ated with cardiovascular disease. The observational data provide
a sound basis for suspecting a causal relationship between OSA
and cardiovascular outcomes because the major causality corol-
laries are met: 1) there is an association between OSA and the
presence of coronary artery disease; 2) the association is graded
based on the severity of OSA; 3) there is a temporal relation-
ship; and 4) the risk seems to be attenuated after treatment of
OSA. Details of the association between OSA and cardiovascu-
lar disease are discussed in another chapter in this text.
Treatment Strategies
Treatment Strategies and Goals in Dyslipidemia
The recommendations for treatment of dyslipidemia are based on
the ATP-III guidelines published in 2001 and updated in 2004.
The management of dyslipidemia is based on the underlying risk
of CAD. The principal target of treatment is the LDL-C level,
after secondary causes of hypercholesterolemia have been ruled
out (Table 56.4). The secondary target is non-HDL-C, which is
556 VI Coronary Artery Disease Risk Factors

Table 56.4. Secondary Causes of Hyperlipidemia
Hypertriglyceridemia
Excessive alcohol or simple sugars
Contraceptives, estrogens, pregnancy
Obesity
Type 2 diabetes mellitus
Chronic renal failure
Cushing disease, corticosteroid therapy
NCEP ATP III Guidelines for LDL Therapy
Hypercholesterolemia LDL-C <160 for 1 or no risk factors
Excessive dietary cholesterol or LDL-C <130 for 2+ risk factors
saturated fats (or both) <100 is a therapeutic option
Hypothyroidism
LDL-C <100 for CAD and CAD equivalents
Obstructive liver disease
<70 is option for very high-risk patients
Nephrotic syndrome
Multiple myeloma or 1. CAD + multiple risk factors, especially diabetes
dysglobulinemia 2. CAD + severe or poorly controlled risk factor(s)
Progestational agents and 3. CAD + metabolic syndome
anabolic steroids
4. Acute coronary syndrome
5. CAD event despite baseline LDL-C <100

calculated simply by subtracting HDL-C from total cholesterol.
Non-HDL-C becomes the secondary target when triglyceride
levels are 200 mg/dL or more, otherwise LDL-C is the only tar-
get of therapy under current guidelines. HDL-C and triglycerides
are not separate targets. There is overwhelming evidence that
treating hypercholesterolemia in patients with and without CAD
improves survival and reduces the incidence of cardiovascular
events.
The goal LDL-C level depends on the underlying cardiovas-
cular risk. Table 56.5 and Figure 56.4 present the ATP-III guide-
lines for management of LDL-C according to the original 2001
paper and the 2004 addendum.
The 2004 addendum also suggested that treatment of patients
with CAD or other atherosclerotic disease with statin therapy
was reasonable, no matter what the level of LDL-C, and recom-
mended lowering LDL-C at least 30% from baseline.
The goal of less than 100 mg/dL also applies to patients with
CAD risk equivalents, such as presence of peripheral vascular
disease, abdominal aortic aneurysm, history of stroke, or dia-
betes mellitus. For patients with CAD or CAD risk equivalents
who are considered to be at “very high risk” (patients with other
modifiable cardiovascular risk factors that have not been fully
controlled, such as current smokers or patients with uncontrolled
diabetes mellitus), it is reasonable to have an LDL-C level less
than 70 mg/dL as well as aggressive modification of all coexis-
tent risk factors.
Among patients without CAD or CAD risk equivalents, the
Framingham risk score can determine whether they have a mod-
erate or low risk of CAD. Further details of LDL-C goals accord-
ing to the underlying cardiovascular disease risk are shown in
Table 56.1.
The goals for non-HDL-C are simply formulated by adding
30 mg/dL to the corresponding goal for LDL-C. Again, the sec-
ondary target for non-HDL-C is appropriate for patients with
triglycerides of at least 200 mg/dL.
Epidemiologic research suggests that patients with an
increased HDL-C level and a slight elevation of LDL-C have a
Table 56.5. Goals For Low-Density Lipoprotein Cholesterol
(LDL-C) According to Adult Treatment Panel III
Risk Category LDL-C Goal Consider Drug Therapy
CAD or CAD risk equivalent <100 mg/dL ≥130 mg/dLa
≥2 Risk factors
10-yr risk 10–20% <130 mg/dL ≥130 mg/dL
10-yr risk 10% <130 mg/dL ≥160 mg/dL
<2 Risk factors <160 mg/dL ≥190 mg/dL
Abbreviations: CAD, coronary artery disease; yr, year.
Figure 56.4. National Cholesterol Education Program (NCEP) 2004
addendum to the 2001 Adult Treatment Panel III (ATP-III) guidelines
for treatment of low density lipoprotein (LDL) cholesterol (LDL-C).
CAD indicates coronary artery disease.
CAD risk comparable to that of patients with a normal LDL-C
level, and subgroup analyses from clinical trials have shown that
the clinical benefit of statins is reduced in patients with a low
LDL-C/HDL-C ratio. However, the current ATP-III guidelines
do not base any recommendations on LDL-C/HDL-C ratios.
Patients with an elevated LDL-C level would require pharmaco-
logic treatment even if they have a high HDL-C level, as long as
they meet criteria based on the Framingham risk score or have
CAD or a CAD risk equivalent. Conversely, patients with a rel-
atively normal level of LDL-C (100–130 mg/dL) and a low level
of HDL-C may require lipid-lowering therapy if they have other
major cardiovascular risk factors resulting in a high Framingham
risk score. In both circumstances—high HDL-C and high
LDL-C or normal LDL-C and low HDL-C—patients should
have target LDL-C values based on the overall Framingham risk
score or on the presence of CAD. Patients are a special chal-
lenge if they have more than one risk factor for CAD that is not
accounted for in the Framingham score, such as family history
of CAD in first-degree relatives, central obesity, OSA, elevated
Lp(a) lipoprotein, a high apolipoprotein B/apolipoprotein A-I
ratio, or chronic renal failure. In those situations, clinicians
should exercise their judgment to determine how low they want
to target LDL-C. As mentioned in the beginning of this chapter,
the current definition of “normal” cholesterol may be too con-
servative; therefore, a more aggressive approach in patients with
increased CAD risk—with use of either traditional risk scores or
novel risk factors—might be of clinical benefit.
Patients with hypercholesterolemia should receive dietary
recommendations and increase their level of physical activ-
ity. The optimal dietary recommendations for the management
of hypercholesterolemia include limiting fat intake, restricting
calories if obesity is present, and increasing the intake of veg-
etables and fruit. The focus on dietary modification should be
on reducing consumption of saturated fat, trans-fatty acids, and
cholesterol while allowing “good” fats like monounsaturated and
polyunsaturated fatty acids. Patients following these recommen-
dations can achieve a decrease in the LDL-C level of up to 20%.
Some subgroups of patients may have a higher reduction in the
level of LDL-C with dietary changes alone, especially those who
are fully compliant with the dietary recommendations and those
without a genetic trait for dyslipidemia.
Patients at high risk of cardiovascular disease may benefit
from receiving lipid-lowering therapy and making therapeutic
lifestyle changes at the same time. Patients with low or moderate
 56 Dyslipidemia 557

Table 56.6. Factors Influencing Levels of High-Density HDL high-density lipoprotein

Lipoprotein Cholesterol HDL-C high-density lipoprotein cholesterol
IDL intermediate-density lipoprotein
Relatively High Levels Relatively Low Levels LDL low-density lipoprotein
LDL-C low-density lipoprotein cholesterol
Females Males LDL-r low-density lipoprotein receptor
US blacks US whites Lp-PLA2 Lipoprotein-associated phospholipase A 2
Exercise Diabetes NCEP National Cholesterol Education Program
Estrogen Hypertriglyceridemia OSA obstructive sleep apnea
Alcohol High-carbohydrate diet RCT reverse cholesterol transport
Weight reduction Obesity SRB1 scavenger receptor class B membrane 1
Nicotinic acid Smoking VLDL very low-density lipoprotein
Fibric acid derivatives Progesterone
Chlorinated hydrocarbons Antihypertensive drugs
Familial hyperalphalipoproteinemia Sedentary lifestyle
Insulin
Names of Clinical Trials
4S Scandanavian Simvastatin Survival Study
ACCORD Action to Control Cardiovascular Risk in Diabetes
risk may start with therapeutic lifestyle changes for 3 to 6 months AFCAPS Air Force Coronary Atherosclerosis Prevention
and start lipid-lowering therapy if the target LDL-C level is not Study
achieved. Patients need to be aware that a high intake of satu- AIM-HIGH Atherothrombosis Intervention in Metabolic
rated fat is a risk factor independent of serum lipids; therefore, Syndrome With Low HDL/High Triglycerides:
they should modify their diet even if, with cholesterol medica- Impact on Global Health Outcomes
ASCOT-LLA Anglo-Scandanavian Cardiac Outcomes Trial:
tions, the goal LDL-C level is achieved.
Lipid-Lowering Arm
Patients with isolated low levels of HDL-C represent a special BIP Bezafibrate Infarction Prevention Study
challenge. The priority should be to maintain LDL-C within tar- CARE Cholesterol and Recurrent Events
get values, increase the level of physical activity, and allow some FIELD Fenofibrate and Event Lowering in Diabetes
monounsaturated and polyunsaturated fatty acids in the diet. HERS Heart and Estrogen/Progestin Replacement Study
When patients have a very low HDL-C (<25 mg/dL), it would HHS Helsinki Heart Study
be unrealistic to expect normalization of the HDL level with HPS Heart Protection Study
lifestyle modification alone. In that situation, medications like ILLUMINATE Investigation of Lipid Level Management to
fibrates or nicotinic acid can be considered after initial therapy Understand Its Impact in Atherosclerotic Events
with statins, particularly in patients with CAD or a CAD risk INTERHEART A Study Of Risk Factors For First Myocardial
Infarction In 52 Countries And Over 27,000
equivalent. Factors associated with relatively high or relatively
Subjects
low levels of HDL-C are listed in Table 56.6. JUPITER Justification for the Use of Statins in Prevention:
Patients with isolated hypertriglyceridemia have a good An Intervention Trial Evaluating Rosuvastatin
response rate with dietary changes, especially if triglyceride lev- LIPID Long-term Intervention With Pravastatin in
els are not higher than 400 mg/dL. Patients eating diets with lim- Ischemic Disease
ited refined carbohydrates have shown significant improvement PROSPER Prospective Study of Pravastatin in the Elderly at
in fasting triglyceride levels. Patients with very high triglyceride Risk
levels (>500 mg/dL) require immediate treatment to prevent the PROVE IT Pravastatin or Atorvastatin Evaluation and Infection
development of pancreatitis. Therapy
SEARCH Study Of The Effectiveness Of Additional Reduc-
tions In Cholesterol And Homocysteine
Abbreviations
TNT Treating to New Targets Study
ATP-III Adult Treatment Panel III
VA-HIT Veterans Affairs High-Density Lipoprotein
CAD coronary artery disease
Cholesterol Intervention Trial
CETP cholesterol ester protein complex
CRP C-reactive protein WOSCOPS West of Scotland Coronary Prevention Study
 57
Lipid-Lowering Clinical Trials and Medications
JOSEPH G. MURPHY, MD, R. SCOTT WRIGHT, MD,
and THOMAS G. ALLISON, PHD, MPH
Lifestyle modification with diet and exercise, along with avoid-
ance of cigarette smoking and second-hand cigarette smoke, is
the foundation of all primary and secondary risk factor modi-
fication for atherosclerotic CV disease. An LDL-C level that is
consistently below 70 mg/dL either through diet or through lipid-
lowering medication, will generally prevent the development of
atherosclerosis or halt its progression in patients with pre-existing
disease. Modification of other atherogenic risk factors that dam-
age endothelial function, specifically smoking, hyperglycemia,
and hypertension, is also important to optimize the protective
effect for the vascular endothelium.
Diet
The beneficial effect of a Mediterranean diet, which is rich in
α-linoleic acid and low in saturated fats, is strongly supported by
evidence from the Lyon Diet Heart Study. This study, conducted
in France, showed a dramatic reduction in future CV events in
post-MI patients. The Lyon Diet Heart Study showed a reduc-
tion in the combined CV study end point by 76% (P < .0001) in
an interim analysis performed at 2 years. At the 4-year analy-
sis point, patients on the Mediterranean diet continued to have
a lower incidence of CV events compared with control diet
patients, by about 70% (2.59 vs 9.03 events per 100 patients per
year). Other diet interventions, including the DART trial and
the Moradabad Heart Study (both of which focused somewhat
more specifically on fish consumption rather than the more gen-
eral approach of the Lyon Heart Study), showed similar though
somewhat less dramatic results. All of these studies were done
in the pre-statin era before effective lipid-lowering drugs were
available.
Abbreviations and acronyms are expanded at the end of this chapter.
558
The NCEP developed its “Step 2 Plus” diet with the
Mediterranean diet in mind but also advocated the addition of
soluble fiber and sitostanol esters to enhance cholesterol lower-
ing. Of course there is no “official” Mediterranean diet, because
food choices and diet vary somewhat among the countries border-
ing the Mediterranean Sea; however, there are some constants:
multiple servings of fresh fruits and vegetables, tomato-based
sauces, whole grains, fish, olives and olive oil, nuts, and (in non-
Moslem parts of the Mediterranean) regular consumption of
wine. Conversely, the Mediterranean diet generally includes only
limited quantities of red meat and very few, if any, processed
foods.
A 2003 study involving 22,043 healthy Greek adults followed
for 44 months showed that 2 single nutrients in the Mediterranean
diet predicted CAD death. There was an 18% reduction in CAD
death for every 200 g/day of fruits and nuts in the diet and a 14%
reduction for every increase of 0.5 in the monounsaturated to sat-
urated fat ratio. For every 2-point increase (of a total 10 points)
in Mediterranean diet score there was a 25% reduction in total
mortality, a 33% reduction in CAD mortality, and a 24% reduc-
tion in cancer mortality.
The NCEP Step 2 diet is more specific in classifying nutrient
content of the diet, with 25% to 35% of calories coming from
fat, but no more than 7% from saturated fat. Polyunsaturated
fats may go up to 20% of total calories, with up to 10% of calo-
ries for monounsaturated fats. Daily cholesterol intake should be
<200 mg. Carbohydrate intake (preferably complex rather than
simple carbohydrates) should constitute 50% to 60% of calories,
while protein should make up the balance of ∼15% of calories.
Current recommendations for soluble fiber consumption are 10
to 25 g/day, whereas 2 g/day of phytosterol intake is recom-
mended. Additionally, the TLC diet recommends that salt intake
be limited to 2,400 mg/day.
 57 Lipid-Lowering Clinical Trials and Medications 559

Yet another recommended dietary strategy for CAD preven- powerful agents for reducing total cholesterol and LDL-C; these
tion is the polymeal, which consists of the following 6 foods: benefits are combined with a moderate triglyceride-lowing effect
and a small HDL-C-raising effect seen in most, but not all, sta-
• Wine (150 mL/day) tins. The statins act by inhibiting the synthesis of cholesterol in
• Fish (114 g 4 times/week) the liver and promoting increased uptake and degradation of
• Dark chocolate (100 g/day) LDL-C from the blood (Figure 57.1). Statins are competitive
• Fruit and vegetables (400 g/day) inhibitors of HMG-CoA reductase, the key enzyme in the rate-
• Garlic (2.7 g/day) limiting step in cholesterol biosynthesis. Reduced hepatic choles-
terol biosynthesis leads to an increase in LDL receptor turnover
• Almonds (68 g/day)
and an increase in hepatic LDL receptor cycling.
The combined effect of the polymeal is estimated to be a 63% In most patients, hypercholesterolemia can be managed with
to 84% reduction in CAD risk. diet and a statin agent alone. These drugs are generally very safe
An alternative strategy of a diet extremely low in fat (10% of and cost-effective and reduce both coronary-related mortality
calories) and cholesterol ( < 5 mg/day) is provided by the “Ornish and total mortality in patients with CAD. In addition, statins
diet,” shown to reduce the progression of atherosclerosis in the reduce the risk of future ischemic stroke in patients with a prior
small, “randomized invitational” Lifestyle Heart Trial. history of stroke or transient ischemic attack.
Since the development of effective drugs for lowering LDL- The ultimate target of all lipid-lowering therapy is the vul-
C, dietary recommendations have shifted more toward providing nerable atherosclerotic plaque; coronary plaque rupture is the
healthy nutrients, such as antioxidants, that might theoretically final common pathway for all the unstable coronary syndromes.
protect the heart with less emphasis on the cholesterol-lowering Statin therapy, in addition to its effect on LDL-C metabolism,
potential of the diet. may stabilize the vulnerable atherosclerotic plaque and convert
lipid-rich plaques that are at high risk of rupture into more stable
Lipid-Lowering Medications fibrotic plaques. The reduction in CV events in all the statin trials
occurred earlier than would be expected from coronary plaque
The principal drugs used in the treatment of hyperlipidemia are regression alone due to pure lipid reduction. This suggests an
statins, niacin, fibric acid derivatives, the cholesterol absorp- additional beneficial effect of statins on vulnerable plaques.
tion inhibitor ezetimibe, and bile acid–binding sequestrants Statin therapy may stabilize the vulnerable plaque through a
(Table 57.1). reduction in macrophages and extracellular lipid accumulation
in the plaque region, by an increase in the collagen content of
Statins the extracellular plaque matrix, by a reduction in calcification
and neovascularization in the intima of the plaque, and through
Statins (HMG-CoA reductase inhibitors) are the most commonly an inhibitory role on the coagulation and inflammatory cascades
prescribed lipid-lowering medications in the world and all are that accompany plaque rupture.
The most common side effects seen in patients treated with
statins are muscle cramps, myositis, and an asymptomatic incre-
Table 57.1. Drug Treatment of Hyperlipidemia ase in hepatic transaminase enzyme values. These side effects are
Typical Expected Effectsa largely reversible with discontinuation of the therapy. Monitoring
of hepatic transaminase enzyme levels before starting statins and
Cholesterol Effect periodically thereafter was once considered important because
Triglyceride of the rare occurrence of hepatotoxicity. A February 28, 2012,
Agent Daily Dosage ↓ LDL, % ↑ HDL, % Effect, % FDA advisory has dropped that recommendation.
Cholestyramine 12–24 g 15–20 0–2 5–10 ↑ The range of skeletal muscle myopathies associated with sta-
Colestipol 15–30 g 15–20 0–2 5–10 ↑ tins includes mild muscle aches without muscle enzyme elevation
Sitostanol esters 1.5–3.3 g 10 0–2 4–8 ↓ or muscle weakness, muscle weakness and clinical myositis with
Niacin 1.5–6 g 20–30 20 30–40 ↓ elevations in serum creatine kinase, and, rarely, life-threatening
Gemfibrozil 600–1,200 mg 10 20 50–60 ↓ overt rhabdomyolysis that can precipitate acute renal failure.
Fenofibrate 67–200 mg 10–15 5–20 40–50 ↓ When myopathies occur, they usually begin within weeks of
Lovastatin 20 mg 25–30 0–10 0–6 ↓ starting statin therapy and usually resolve, together with ele-
80 mg 35–40 0–10 25 ↓ vated creatine kinase concentrations, within weeks of stopping
Pravastatin 20 mg 25–30 0–10 10 ↓
the therapy. Susceptibility to statin-associated myopathy and
40 mg 25–35 0–10 25 ↓
Simvastatin 10 mg 25–30 0–10 0–5 ↓
myositis is increased in patients with renal failure, obstructive
80 mg 40–50 0–10 25–40 ↓ liver disease, or hypothyroidism and in patients taking erythro-
Atorvastatin 10 mg 35–40 0–10 20 ↓ mycin, cyclosporine, azole antifungal agents, nicotinic acid, or
40 mg 40–60 0–10 35 ↓ gemfibrozil.
80 mg 60 0–10 35–45 ↓ The February 2012 FDA advisory noted above also mentions
Rosuvastatin 5 mg 28 3 21 ↓ a small risk of diabetes in patients with impaired glucose toler-
10 mg 45 8 37 ↓ ance who are taking statins. This was seen first in the JUPITER
20 mg 31 22 37 ↓ trial and later reported in the WHI observational cohort. The
40 mg 43 17 43 ↓ risk of diabetes with statins is small (1 case of diabetes for every
Ezetimibe 10 mg 18 1 8↓ 3 heart attacks prevented), confined to those with an elevated
Abbreviations: HDL, high-density lipoprotein; LDL, low-density fasting glucose at baseline, more likely to occur in older female
lipoprotein. patients and to be dose-related. The risk of diabetes is surpris-
a
Arrows indicate increase (↑) or decrease (↓). ing in that one of the first statin trials—WOSCOPS (discussed
560 VI Coronary Artery Disease Risk Factors
Liver
Intestine
Figure 57.1. Overview of Cholesterol Transport. Acetyl CoA indicates acetyl coenzyme A; C, cholesterol; HDL, high-density lipoprotein; IDL,
intermediate-density lipoprotein; LDL, low-density lipoprotein; R, receptor; SR-BI, scavenger receptor class B type I; VLDL, very-low-density
lipoprotein.
below)—actually showed a significant reduction in the incidence
of type 2 diabetes in men with high cholesterol but no history of
MI or coronary revascularization (a small percentage of subjects
did report angina). The treatment arm in WOSCOPS received
pravastatin 40 mg/day.
The advisory also noted reports of cognitive dysfunction with
statins, but this (unlike the diabetes risk) has never been con-
firmed in clinical trials. Initial observations from a large clinical
database in the United Kingdom suggested a negative relationship
between statin use and the risk of dementia, but this could have
been due to a bias against treating patients with dementia with
statins. A randomized clinical trial called PROSPER evaluated
the impact of statin (40 mg of pravastatin) on risk of dementia
and cognitive function but found no positive or negative effects.
A small excess of new cancers was reported in PROSPER, but
this has not been seen in other statin trials. Meta-analyses of all
trials involving pravastatin and all trials with any statin have
shown no significant differences between statin and placebo in
terms of cancer incidence or cancer mortality.
Statins also have multiple secondary beneficial effects, which
include a reduced risk of osteoporotic fractures (particularly
in older patients) and a mild blood pressure–lowering effect in
hypertensive patients. Statins appear to have pleiotropic actions
on atheromatous plaques and atherogenesis beyond lowering cho-
lesterol levels and may have a beneficial effect when used early
after an acute MI. Though pleiotropic effects appear to be feasi-
ble from a statistical standpoint (ie, greater benefit from statins
than would be expected solely from their lipid-lowering effect)
and from mechanistic considerations (eg, reduced inflammation),
it is important to recognize that these pleiotropic effects have not
been confirmed experimentally in humans where LDL-C is held
constant by combined cholesterol feeding plus statin therapy and
event rates compared with placebo-treated subjects.
Extrahepatic
tissues
Nicotinic Acid
Nicotinic acid (niacin) is a widely used, nonprescription, inex-
pensive member of the water-soluble vitamin B complex family.
In high doses (minimum 500 mg/day vs the 40 mg/day dose as a
vitamin supplement), niacin is used in the treatment of hyperlipi-
demia. Niacin moderately lowers total cholesterol, LDL-C, and
triglyceride levels but also increases HDL-C. Nicotinic acid’s
biochemical action on LDL-C is through the inhibition of hepatic
synthesis of VLDL and LDL and decreased mobilization of free
fatty acids from adipose tissue. It also raises HDL-C levels by
about 30% by inhibiting lipid transfer between HDL and VLDL
and by delaying HDL clearance. Another favorable property of
nicotinic acid is a reduction in plasma fibrinogen levels, a known
risk factor for atherosclerosis progression and arterial thrombus
formation. Nicotinic acid is the most potent medication currently
approved for increasing low HDL-C values and is particularly
helpful in patients with metabolic syndrome–associated lipid
abnormalities typically characterized by small, dense LDL par-
ticles, elevated triglycerides, and low HDL-C levels.
Nicotinic acid often decreases levels of lipoprotein (a), a very
atherogenic LDL variant, while statins have little direct effect on
lipoprotein (a) levels. Lipoprotein (a) levels are diet independent
and primarily determined by genetic polymorphisms at the apo-
lipoprotein (a) gene locus of the LPA gene. Statins are still ben-
eficial in patients with lipoprotein (a)–induced atherosclerosis to
reduce total LDL-C levels, which should be reduced aggressively
to less than 70 mg/dL.
The major side effects of nicotinic acid are pruritus, flushing,
gastrointestinal tract distress, glucose intolerance, rash, provoca-
tion of gout, and liver toxicity. Aspirin taken 30 to 60 minutes
before nicotinic acid can reduce the flushing. Patient tolerance of
the medication is increased with the sustained-release formula-
tion of nicotinic acid. Atrial arrhythmia and ocular maculopathy
 57 Lipid-Lowering Clinical Trials and Medications
are rare side effects that may occur with nicotinic acid. When used
concomitantly with a statin, nicotinic acid somewhat increases
the risk of myopathy and hepatotoxicity, but with a much better
lipid profile that can generally be achieved with a statin alone
because of its greater effect on raising HDL-C levels.
Nicotinic acid is available in 3 forms: immediate-release
(or crystalline) niacin, generic sustained-released niacin, and
prescription-only, extended-release niacin (Niaspan). The imme-
diate-release form of nicotinic acid is the most potent for rais-
ing HDL-C levels and lowering triglycerides and also carries
the lowest risk of liver function abnormalities. Unfortunately,
it is also the most poorly tolerated, with high rates of flushing.
Immediate-release niacin was used in several trials, including the
Coronary Drug Project outcome trial and regression trials includ-
ing CLAS, UCSF-SCOR trial, FATS, and HATS. The HATS pro-
tocol indicated starting treatment with sustained-release niacin
up to 2 g/day but advancing to higher doses of immediate-release
niacin in order to achieve the HDL-C goal. Whereas sustained-
release and extended-release forms of niacin can be used up to a
dose of 2 g/day with reasonable safety, immediate-release nia-
cin has been employed in doses of 3 to 6 g/day in clinical trials.
Sustained-release and extended-release niacin are associated with
less flushing compared with immediate-release niacin, but unfor-
tunately the flushing rate is still moderately high. Niacin is the
most poorly tolerated of the lipid-lowering drugs, with reported
discontinuance rates in clinical databases of 50% and higher.
Several forms of “no-flush” niacin are marketed. These
include inositol and niacinamide. Their no-flush status is due to
a lack of bioavailability; hence “no flush” equates to no benefit
on lipid levels.
The Oxford Niaspan Study evaluated carotid artery wall
thickness using magnetic resonance imaging at 12 months
after treatment with extended-release nicotinic acid in patients
with documented CV disease and an HDL-C below 40 mg/dL
(≈1.0 mmol/L) who were already taking a statin. Patients in the
nicotinic acid group had a greater decrease in carotid artery wall
thickness compared with the placebo group.
The AIM-HIGH trial tested the benefits of extended-release
niacin 1,500 to 2,000 mg/day compared with placebo (100 to
200 mg of niacin daily) in patients with established CV disease
and low HDL-C and elevated triglyceride levels. All patients
received a statin (simvastatin 40–80 mg/day) plus ezetimibe
10 mg/day, if needed, to maintain an LDL-C level of 40 to
80 mg/dL. The study found that in spite of significant improve-
ments in HDL cholesterol and triglyceride levels in the niacin
treatment arm, there was no additional benefit to extended-
release niacin compared with statin alone.
Fibric Acid Derivatives
Fibric acid derivatives, or “fibrates,” activate (PPAR-α) that stim-
ulate the synthesis of enzymes of fatty acid oxidation. PPAR-αs
are hormone receptors located in the cell nucleus (nuclear tran-
scription factors) that modify the expression of several genes
responsible for lipoprotein expression.
Fibrates also exert other favorable effects on lipids including:
• The promotion of a shift from small, dense LDL particles into a
smaller number of larger, less dense LDL particles that are both
less atherogenic and more easily removed from the circulation.
• Stimulation of lipoprotein lipase activity, which results in enhanced
triglyceride clearance.
• Reduced hepatic secretion of VLDL.
561
Fibrates decrease serum triglycerides (by 30%–50%) and
increase serum HDL-C levels (by 15%–25%). The fibrates have
a variable effect on serum lipoprotein (a). Currently available
fibric acid derivatives include fenofibrate, gemfibrozil, and ben-
zafibrate (available outside the United States). The earliest ver-
sion of fibric acid derivative, clofibrate, was widely studied as
an agent for reducing CAD incidence and was available for use
for several years in clinical practice, but it was withdrawn from
the market due to increased mortality in the WHO Cooperative
Trial of primary prevention of ischemic heart disease published
in 1984. Deaths came from a variety of causes, the mechanisms
of which have not been fully explained.
• Fenofibrate and benzafibrate decrease fibrinogen levels, but gemfi-
brozil has no demonstrable effect.
• Fibrates have a modest effect on LDL-C levels but exert a ben-
eficial effect in many dyslipidemic patients, particularly those
with metabolic syndrome or diabetes by increasing the particle
size away from small dense LDL particles; LDL particle size is
inversely related to atherogenicity.
• Similar to niacin, fibrates potentiate the effects of warfarin by
decreasing protein binding.
The older of the 2 commercially available fibrates in the
United States, gemfibrozil has been widely used for many years
in the treatment of hypertriglyceridemia, mixed hyperlipidemias,
which are characterized by an increase in both serum triglycer-
ide and LDL-C (a common occurrence in diabetic or metabolic
syndrome patients), and isolated low HDL-C level. Gemfibrozil,
in particular, and fibrates, in general, have little effect on LDL-C
levels. In some cases, fibrates may actually raise the level of
LDL-C in the blood by enhancing triglyceride clearance, thus
moving VLDL along the metabolic cascade to LDL. The main
side effects of gemfibrozil are gastrointestinal tract intolerance
and a possible increased risk of cholelithiasis. Gemfibrozil can
cause muscle toxicity, particularly when used in combination
with the high-potency statins, especially those employing the
PY450 3A4 pathway for degradation. Cerivastatin was with-
drawn from the market in August 2001, due to 39 cases of fatal
rhabdomyolysis; 12 of those cases occurred when cerivastatin
was combined with gemfibrozil.
Fenofibrate is the other commercially available fibrate. It is
available in both generic form, with a maximum daily dose of
200 mg, and in the micronized form (Tricor), with a maximum
daily dose of 145 mg. Fenofibrate is approved for the treatment
of hypertriglyceridemia. The main side effects of fenofibrate are
rash and gastrointestinal tract upset.
Fibrates decrease the LDL-C level less than statins (if at all)
but may increase the HDL-C level more than statins, particu-
larly if the triglycerides are elevated. This is true with gemfibro-
zil especially, but largely also with fenofibrate. Fibrates lower
triglyceride levels 30% to 50%, an effect much greater than
that seen with statins, and are indicated to prevent pancreati-
tis in patients with very high triglyceride levels. Use of fibrates
decreased markedly with the introduction of statins, though their
use was revived with the publication of the successful secondary
prevention trial VA-HIT, which showed benefit of gemfibrozil in
lowering CV events in a secondary prevention population with
high rates of diabetes and metabolic syndrome. More recently,
enthusiasm for fibrates has dropped markedly in the wake of two
negative trials of fenofibrate in patients with type 2 diabetes,
specifically FIELD and ACCORD-Lipid. A report on all fenofi-
brate trials though FIELD identified a small but consistent trend
toward increased total mortality, significant in meta-analysis,
562 VI Coronary Artery Disease Risk Factors
versus placebo, but mortality was not increased in the fenofibrate
arm of the ACCORD-Lipid trial.
Bile Acid–Binding Sequestrants
The anionic resins, or bile acid–binding sequestrants (choles-
tyramine, colestipol, and colesevalem) are safe and moderately
effective therapy for hyperlipidemia. These agents decrease total
cholesterol and LDL-C levels by binding positively charged bile
acids in the gut to interrupt the enterohepatic circulation of bile
acids. This stimulates new bile acid production and a secondary
increase in hepatic LDL receptors, which in turn remove LDL-C
from the circulation. Resins usually have a small positive effect
on HDL-C level and generally raise triglyceride levels modestly,
although rarely they may increase triglycerides dramatically.
As expected from their mode of action, the main side effects
associated with the resin agents are related to gastrointestinal
tract intolerance: gas, bloating, constipation, nausea, and esopha-
geal reflux. The side effects cause about 50% of patients to dis-
continue therapy at 1 year. These agents decrease LDL-C, and, if
the dose is increased slowly, resin agents can be reasonably toler-
ated by many patients. They are excellent adjunctive agents in
severe hyperlipidemia when used in combination with statins or
nicotinic acid. Bile acid–binding sequestrants are relatively con-
traindicated in patients with significant hypertriglyceridemia.
A clinical problem with resins is their effect on the absorption
of vitamin K, especially in patients receiving warfarin. Resins
also inhibit the absorption of digoxin, warfarin, thyroxine, sta-
tins, and diuretics if given concomitantly with these agents.
Resins were used extensively in older trials (eg, the LRC-
CPPT) of lipid lowering with some limited success and were
combined with niacin and later statins for enhanced lipid lower-
ing in trials such as FATS. In the pre-statin era, lipid manage-
ment with dietary restrictions on saturated fat and cholesterol
intake, niacin, fibrates, and resins was difficult to manage, poorly
tolerated, and of limited efficacy.
Because the older agents, cholestyramine and colestipol, had
to be mixed with water, juice, or a soft, semiliquid food like
applesauce to be taken, they have largely disappeared from clini-
cal use in the United States. Colesevalem (Welchol), on the other
hand, comes in pill form (though the standard dose requires
ingestion of 3 large pills twice daily) and has less tendency to
bind to other drugs. For these reasons, it is the usual form of resin
used today, though overall use of resins in clinical practice is
low. Ezetimibe, another type of intestinally acting lipid-lowering
agent, has largely driven resins from the formulary due to greatly
enhanced convenience and far fewer side effects with similar
efficacy for lowering LDL-C level.
Cholesterol Absorption Inhibitor–Ezetimibe
Ezetimibe has a unique mechanism of action that is different
from that of all other classes of lipid-lowering medications in
that it electively inhibits the absorption of cholesterol at the brush
border of the small intestine by about 50%. Intestinal cholesterol
is derived from two sources: dietary cholesterol derived from
animal sourced foods and hepatic cholesterol secreted into the
bile in the form of bile salts that aids in fat absorption. Hepatic
cholesterol is derived from 3 sources: de novo hepatic synthesis
of cholesterol, cholesterol removed from serum lipoproteins, and
cholesterol absorbed by the small intestine (a combination of diet
and bile salt cholesterol).
Ezetimibe does not inhibit synthesis of cholesterol in the liver
or increase excretion of bile acid cholesterol but rather inhibits
the intestinal absorption of cholesterol, thereby interrupting the
enterohepatic circulation of cholesterol, which leads to a decrease
in the delivery of intestinal cholesterol to the liver. This in turn
causes a reduction of hepatic cholesterol stores and an increase in
the clearance of cholesterol from the blood.
• Ezetimibe decreases levels of total serum cholesterol, LDL-C,
apo B, and triglycerides, although it only slightly increases serum
HDL-C.
• Ezetimibe has a beneficial effect on cholesterol metabolism by
blocking dietary cholesterol absorption and interrupting the
enterohepatic recirculation of cholesterol through biliary secretion
and intestinal reabsorption.
Ezetimibe (10 mg daily) decreases serum LDL-C by about
18% when used as a sole lipid-lowering agent. When ezetimibe
is added to a statin incrementally, it lowers LDL-C overall by
a slightly smaller amount (about 15%) in addition to the statin-
induced lowering of LDL-C level. However, in patients who do
not respond favorably to a statin as a fi rst agent (not achieving
their goal for LDL-C), the addition of ezetimibe will gener-
ally result in a much greater reduction in LDL-C, averaging
25% (with some patients responding even more dramatically).
It is thought that these patients with a more modest response
to statins represent “high absorbers” of intestinal choles-
terol, whereas statin hyper-responders are “high producers” of
LDL-C.
While statins are the first drug of choice for pharmacologic
reduction of LDL-C levels (because of their potency in lower-
ing LDL-C plus their added pleiotropic action on atheromatous
plaques and atherogenesis), ezetimibe is a valuable additional
agent in patients who do not meet NCEP cholesterol goals with
statin therapy alone, those who are either intolerant of a high
statin dose or any statin dose because of side effects (generally
myalgia or increased levels of liver enzymes), and those with
familial hypercholesterolemia who require a maximal lipid-
lowering effect.
Ezetimibe is generally well tolerated when administered
alone, and the incidence of either myopathy or liver serum
transaminase elevations is similar to the incidence with pla-
cebo, though creatine kinase–negative myalgias do occur in rare
patients. When ezetimibe is administered in conjunction with a
statin, the incidence of serum transaminase elevation and other
side effects is similar to that with statin therapy alone.
Ezetimibe comes as a single agent with a single recommended
dose of 10 mg/day but is also available in combination with var-
ious doses of simvastatin (ie, Vytorin). There is a possible drug
interaction between ezetimibe and warfarin, and gemfibrozil
may increase ezetimibe levels.
Sitostanol Plant Esters, Estrogens, Alcohol,
Fish Oil, and Other Antiatherogenic Agents
Sitostanol Plant Esters
Sitostanol plant esters are derivatives of naturally occurring
plant esters that reduce the absorption of cholesterol from the
gastrointestinal tract and are incorporated into some margarine
and other food products. Several recent trials have examined the
cholesterol-lowering effect of dietary substitution of sitostanol
esters for soybean-enriched margarine. A daily consumption
of 1.5 to 3.3 g of sitostanol ester was associated with an 8% to
13% decrease in total cholesterol and LDL-C values and a 1%
to 2% increase in the HDL-C value. Additionally, triglyceride
levels were reduced 4% to 12%. One trial has demonstrated that
 57 Lipid-Lowering Clinical Trials and Medications 563

sitostanol esters can be used in combination with statins to gain
additive benefit. No major side effects have been reported.
Hormone Replacement Therapy
Hormone replacement therapy in postmenopausal women does
not reduce future CV events. The HERS study of hormone ther-
apy in secondary prevention and the WHI study in primary pre-
vention both had negative findings.
Alcohol Consumption
Light alcohol consumption (ie, 1 drink/day for women, 1–2
drinks/day for men) decreases the risk of coronary disease by
about 50% and all-cause mortality by 18%. Moderate alcohol
consumption also decreases the risk of ischemic stroke and sud-
den death. The beneficial effects of alcohol are lost at high con-
sumption rates. Alcohol probably mediates its effect through an
increase in protective HDL-C.
Fish Oils
Individuals with high intake of Ω-3 polyunsaturated fatty acids,
usually associated with a high fish and low red meat diet, have
low rates of heart disease. Fish oil concentrates reduce triglycer-
ide levels and increase LDL particle size, a beneficial effect when
administered in high doses (6–12 g/day).
Serum CRP and Lipid-Lowering Medications
Epidemiologic evidence suggests that inflammation is an impor-
tant mediator of vascular atherogenesis and plaque rupture. Serum
CRP is a marker of inflammation that is statistically linked to an
increased risk of future CV events. CRP is considered a surro-
gate marker of vascular inflammation and atherosclerotic risk.
Measurement of high-sensitivity CRP levels in conjunction with
lipid profile analysis improves patient risk stratification. Statins
reduce CRP levels independently of their effect on cholesterol
levels, suggesting that part of the beneficial effects of statins may
lie in their anti-inflammatory effects.
• High-sensitivity CRP is a useful independent marker of vascular
inflammation and CV risk in patients with stable or unstable CV
disease.
• Determination of CV risk by high-sensitivity CRP testing in
patients without known CV disease is based on high-sensitivity
CRP values as follows: < 1 mg/L, low risk; 1–3 mg/L, average
risk; and >3 mg/L, high risk.
• Among patients with known CV disease, a cutoff value of >3 mg/L
is recommended for predicting outcomes among patients with sta-
ble CV disease, and a threshold of >10 mg/L is recommended for
patients with unstable coronary syndromes.
Major Lipid-Lowering Trials
There is unequivocal clinical evidence from multiple random-
ized clinical trials that treatment of hypercholesterolemia with
statins reduces future CV events in persons with or without
clinically evident ischemic heart disease. In general, patients at
the highest risk—those with established CAD or its risk factor
equivalent benefit the most with risk factor modification, includ-
ing pharmacological lipid lowering. Primary prevention trials
show lower absolute CV and mortality benefits when compared
with secondary prevention or mixed trials over a similar time
period (Figure 57.2).
• The timeline is longer for primary prevention to show benefit, and
the benefit is smaller in absolute terms than in secondary preven-
tion trials.
• The risk-benefit ratio of statins in secondary prevention is much more
favorable than in primary prevention trials, but this should be seen
in the overall very low risk of serious side effects with statins and
the high lifetime risk of CV events in patients with dyslipidemia.

30

25 HPS
4S
% With CAD Event

20 Secondary
Primary
15 PROSPER Mixed
LIPID
10 WOSCOPS
CARE
5
AFCAPS/TexCAPs
0
50 70 90 110 130 150 170 190 210
LDL-C (mg/dL)
Figure 57.2. Comparison of Outcomes Among Trials Studying Use of Statins for Primary Prevention, Secondary Prevention, or Mixed Purpose.
4S indicates Scandinavian Simvastatin Survival Study; AFCAPS/TexCAPS, Air Force/Texas Coronary Atherosclerosis Prevention Study; CARE,
Cholesterol and Recurrent Events Study; HPS, Heart Protection Study; LIPID, Long-term Intervention With Pravastatin in Ischemic Disease;
PROSPER, Pravastatin in Elderly Individuals at Risk of Vascular Disease; WOSCOPS, West of Scotland Coronary Prevention Study. (Previously
published. See “Credit Lines” section.)
564 VI Coronary Artery Disease Risk Factors
Primary Prevention of CV Disease
Primary prevention is the reduction in future CV events in sub-
jects without documented CV disease. Major trials studying
primary prevention of CV disease are reviewed below.
WOSCOP Study (1995)
WOSCOP study examined a patient cohort with a diet very high
in saturated fats; it tested the hypothesis that primary preven-
tion with pravastatin would reduce mortality and nonfatal MIs
in patients with hyperlipidemia who had not had a prior MI. The
study randomly assigned patients to receive pravastatin (40 mg
daily) or placebo and included nearly 6,600 middle-aged men
with hypercholesterolemia and fasting LDL-C values of more
than 252 mg/dL who did not respond adequately to diet after
4 weeks (LDL-C still greater than 155 mg/dL). Pravastatin
decreased LDL-C by an average of 26% and decreased the
following study end points:
1. All-cause mortality risk by 22% (P = .05)
2. All coronary events by 31% (P < .001)
3. Nonfatal MI by 31% (P < .001)
4. Death from all CV causes by 33% (P = .033)
5. Myocardial revascularization (CABG or PTCA) by 37%
In addition, there was no increase in noncardiac mortality.
There were 2 major caveats with the WOSCOP study, namely,
female patients were excluded and a very high proportion of
patients (78%) were smokers or ex-smokers.
• The WOSCOP study established that primary prevention with
pravastatin in middle-aged men with hyperlipidemia decreased
coronary events by about a third and all-cause deaths by one-fifth
in the 5 years of study participation.
AFCAPS/TexCAPS Trial (1998)
The AFCAPS/TexCAPS trial extended the benefit of primary
prevention to patient populations with “average” cholesterol
values. The investigators randomly assigned just over 6,600
patients (including almost 1,000 women, all without clinical
CAD) to receive lovastatin or placebo for a mean of 5.2 years.
Baseline mean serum levels were as follows: triglycerides,
221 mg/dL; LDL-C, 150 mg/dL; and HDL-C, 36 mg/dL for men
and 40 mg/dL for women. Treatment with lovastatin decreased
LDL-C levels by 25% and increased HDL-C levels by 6%. There
was a 37% risk reduction in the occurrence of the composite end
point (fatal or nonfatal MI, sudden death, or unstable angina)
with lovastatin therapy, and coronary revascularization proce-
dures decreased by 32%.
• The AFCAPS/TexCAPS trial demonstrated for the first time a
treatment benefit favoring statin therapy in a population without
known CAD and with a previously identified “average” cholesterol
value.
• Taken together, the WOSCOP and AFCAPS/TexCAPS trials
established that statin treatment of hyperlipidemia has a beneficial
effect in the primary prevention of CAD.
ASCOT-LLA Trial (2003)
The ASCOT-LLA trial was a European, multicenter, random-
ized-controlled trial that evaluated atorvastatin in the prevention
of coronary and cerebrovascular events in high-risk hyperten-
sive patients who had normal or mildly raised cholesterol lev-
els. Just over half of the hypertensive patients had a total serum
cholesterol concentration of 250 mg/dL or less (≤6.5 mmol/L)
and were randomly assigned to receive atorvastatin (10 mg daily)
or placebo. These 10,300 patients formed the lipid-lowering arm
of the study. Follow-up was initially planned for 5 years, but the
trial was stopped early after a median follow-up of 3.3 years
because of a significant benefit in patients receiving atorvastatin.
The primary end point of the study was a combination of non-
fatal MI or CAD death, which occurred in 1.9% of the atorvas-
tatin patients and 3% of the placebo patients (hazard ratio, 0.64;
P < .001). Atorvastatin also significantly reduced the stroke risk,
total CV events, and all coronary events, but the trial did not
show a statistically significant reduction in all-cause mortality
or CV mortality.
• The ASCOT-LLA Study established that low-dose atorvasta-
tin reduced CV events and stroke risk in high-risk hypertensive
patients with normal or slightly elevated cholesterol levels.
Collaborative Atorvastatin Diabetes Study (2004)
CARDS was a European study that evaluated the effect of either
atorvastatin (10 mg daily) or placebo on primary prevention of
CV events in patients with type 2 diabetes mellitus and no known
CV disease. The trial enrolled just over 2,800 diabetic patients
who had a serum LDL-C concentration of 160 mg/dL or less,
a fasting triglyceride concentration of 600 mg/dL or less, and
at least 1 of the following high-risk features: retinopathy, albu-
minuria, active smoker, or hypertension.
The trial was terminated 2 years earlier than planned because
of a substantial benefit in the atorvastatin group. CV events were
decreased by 37%, and it was estimated that treatment would
prevent 37 major CV events per 1,000 patients treated during
a 4-year period. Subgroup analysis showed that atorvastatin
decreased the stroke risk by 48%. Atorvastatin decreased the
death rate by 27%, but the difference was not statistically sig-
nificant (P = .059).
• The CARDS trial concluded that atorvastatin, 10 mg daily, reduced
the risk of first CV disease event, including stroke, in patients with
type 2 diabetes mellitus and mild to moderately elevated LDL-C
levels.
• The CARDS trial found no scientific justification for a particular
threshold level of LDL-C to be the sole criterion for when diabetic
patients should receive statins.
JUPITER Trial (2008)
The JUPITER trial tested the hypothesis that asymptomatic indi-
viduals with elevated high-sensitivity CRP levels but relatively
normal lipid levels would benefit from statin treatment. It was
previously know that increased levels of the inflammatory bio-
marker high-sensitivity CRP predict CV events and that statins
lower levels of high-sensitivity CRP in addition to cholesterol.
This trial randomly assigned nearly 18,000 healthy individuals,
men aged ≥50 years and women ≥60 years with an LDL-C level
below 130 mg/dL (3.4 mmol/L) and a C-reactive protein level
of at least 2.0 mg/L, to treatment with a potent statin (rosuva-
statin 20 mg daily) or placebo. The trial was stopped early after
a median follow-up of 1.9 years because of a strong beneficial
effect seen in the statin treatment arm.
The primary end point of a first major CV event, includ-
ing nonfatal MI or nonfatal stroke, hospitalization for unstable
angina, arterial revascularization procedure (PCI or CABG) or
confirmed death from CV causes. Rosuvastatin reduced all-cause
mortality by 20% from 1.25 to 1.0 deaths (hazard ratio, 0.80;
 57 Lipid-Lowering Clinical Trials and Medications
95% CI, 0.67–0.97) per 100 person-years and the combined end
point by 44% from 1.36 to 0.77 CV events (hazard ratio, 0.56;
95% CI, 0.46–0.69) per 100 person-years.
• The conclusion of the study was that apparently healthy persons
without hyperlipidemia but with elevated high-sensitivity CRP lev-
els have reduced the incidence of major CV events when treated
with a potent statin.
Secondary Prevention of CV Disease
Secondary prevention refers to a reduction in future CV events in
subjects with known CV disease. Major trials studying second-
ary prevention of CV events are reviewed below.
Scandinavian Simvastatin Survival Study (1994)
The 4S was a pivotal cardiology trial from Scandinavia which
proved for the first time that statins could reduce mortality in
patients with established CAD. The 4S tested the hypothesis that
secondary prevention with simvastatin in patients with known
CAD would reduce mortality and nonfatal infarctions in patients
with hyperlipidemia. The study randomly assigned patients to
receive either simvastatin (20-40 mg daily) or placebo. The study
enrolled 4,444 patients (which followed the 4S theme), who had
total cholesterol values between 200 and 300 mg/dL. The study
excluded high-risk patients: patients who had congestive heart
failure or recent MI or who required revascularization (CABG
or PTCA).
Simvastatin decreased total cholesterol levels by 25% and
LDL-C by 35%, with a modest increase (about 8%) in HDL-C.
Simvastatin decreased the following:
1. All-cause mortality (by 30%)
2. Coronary events (by 34%)
3. The risk of coronary death (by 42%)
4. Myocardial revascularization (CABG or PTCA) (by 37%)
In addition, there was no increase in noncardiac mortality.
The sickest patients with CAD were excluded from the study.
Of note female, diabetic, and elderly patients were included in
the 4S.
• The 4S established that there was a clear benefit for secondary
prevention treatment with simvastatin in patients with established
CAD and hypercholesterolemia.
The mortality in the female placebo group in the 4S was lower
than 50% of the mortality for men; thus, the study was under-
powered to draw statistical conclusions about the mortality ben-
efit of lipid-lowering treatment in women. With regard to major
coronary events, women had a decrease that directly paralleled
that seen in men, who in turn showed a mortality benefit with
lipid-lowering treatment. Patients older than 60 years had both
an improved all-cause survival and a decreased incidence of
major coronary events, similar to that for patients younger than
60 years.
CARE Study (1996)
The CARE study tested the hypothesis that secondary prevention
with pravastatin (40 mg daily) instead of placebo would reduce
mortality and cardiac events in patients with “average” choles-
terol levels. CARE investigators randomly assigned nearly 4,200
patients (21–75 years old) who had a history of an MI in the pre-
vious 2 years and who had a total cholesterol value of less than
240 mg/dL, an LDL-C value between 115 and 174 mg/dL, and a
565
triglyceride value of less than 350 mg/dL. Patients were excluded
if they had symptomatic congestive heart failure or a low ejection
fraction (<25%). The median follow-up period was 5.0 years.
Pravastatin decreased LDL-C by 32%. In addition, pravasta-
tin decreased the following:
1. All-cause mortality (by 9%; not statistically significant)
2. Coronary events (by 24%)
3. Nonfatal MI (by 23%)
4. Risk of coronary death (by 19%)
5. Stroke (by 31%)
6. Myocardial revascularization (CABG or PTCA) (by 27%)
The primary end point of CARE was the combined rate of
fatal coronary events and nonfatal MI. The group randomized
to pravastatin experienced a 10.2% primary end point event
rate, while the placebo group had a 13.2% primary end point
event rate, with an ARR of 2% (NNT = 50) and a RRR of 24%,
P = .003. Other clinical end points were also reduced.
• The need for subsequent CABG was reduced from 10.0% in the
placebo group to 7.5% in the treated group (ARR, 2.5%; NNT = 40;
RRR, 26%; P = .005).
• The need for PCI was reduced from 10.5% to 8.3% (ARR 2.2%;
RRR, 23%; NNT = 45; P = .01).
• Stroke was reduced 31% (P = .03).
There was no increase in noncardiac mortality. Women were
included in the study and benefited from pravastatin more than
men. Women had a more dramatic reduction in fatal and nonfatal
MI with pravastatin than men (45% vs 19%). The CARE study
showed a clear nonmortality benefit with secondary prevention
treatment with pravastatin in patients with known CAD and
“average” cholesterol levels. Overall, the CARE study showed
no significant decrease in all-cause mortality, and no significant
benefit (mortality or otherwise) occurred in the patient subgroup
with a baseline LDL-C of 125 mg/dL or less.
The benefit with pharmacologic cholesterol lowering is illus-
trated in a plot of the degree of LDL-C lowering compared with
the decrease in CAD relative risk in the LRC-CPPT, CARE,
WOSCOP, and 4S trials. The greater the percentage of LDL-C
lowering, the greater the percentage decrease in CV events.
Atorvastatin Versus Revascularization Treatment (1999)
The AVERT trial was a randomized trial that examined the inci-
dence of future cardiac ischemic events in a direct comparison of
medical therapy (atorvastatin, 80 mg daily) with interventional
therapy (PTCA) and usual care in patients who had stable CAD
and class I or II angina, all of whom had been recommended
to have PTCA. The study included 341 patients who had angio-
graphically proven CAD with stenosis of 50% or more in at least
1 vessel recommended for angioplasty, an LDL-C value of at
least 115 mg/dL, and class I or II angina. Patients were excluded
if there was electrocardiographic evidence of severe ischemia at
less than 4 minutes on a Bruce protocol treadmill test. LDL-C
was decreased by 46% in the atorvastatin group and by 18% in
the angioplasty group.
At 18 months, there was a 36% decrease in ischemic events in
the atorvastatin group compared with the PTCA group. Among
patients who received atorvastatin, those who had a reduction in
LDL-C of 40% or more had significantly fewer ischemic events
than those who had a reduction of less than 40%. Most medi-
cally treated patients (87%) who received atorvastatin were able
to continue medical therapy for 18 months without a cardiac
ischemic event.
566 VI Coronary Artery Disease Risk Factors
This study is important because it highlights the importance
of aggressive lipid lowering in all patients recommended to
undergo coronary intervention. Current cardiology practice is at
variance with the randomization strategy in the AVERT trial and
favors early PCI in patients with a combination of angina, flow-
limiting coronary stenosis, and a coronary anatomy suitable for
PCI. Aggressive lipid reduction and PCI should be complemen-
tary rather than competitive strategies for the treatment of CAD.
LIPID Trial (1998)
The LIPID trial examined 9,014 patients with known CAD
who were randomly assigned to treatment with pravastatin or pla-
cebo and observed for 6.1 years. Death from CAD was decreased
from 8.3% in the placebo group to 6.4% in the pravastatin group
(P < .001). All-cause mortality was decreased from 14.1% to
11.0% (P < .001). In addition, the risk reduction was 19% for
stroke, 20% for coronary revascularization, and 29% for MI.
TNT Trial (2005)
The TNT trial randomly assigned 10,000 patients with stable
CAD to receive atorvastatin in a low dose (10 mg daily) or a
high dose (80 mg daily) with LDL-C goal values of 100 mg/dL
(with low doses) or 75 mg/dL (with high doses). Patients were
required to have a baseline LDL-C value between 130 mg/dL and
250 mg/dL and to achieve an LDL-C value less than 130 mg/dL
with low-dose atorvastatin (10 mg daily) during an initial open-
label run-in period. The primary end point of the TNT trial was a
major CV event—coronary death, nonfatal MI, cardiac arrest, or
stroke—during the median follow-up period of 4.9 years.
High-dose atorvastatin had beneficial effects over low-dose
atorvastatin and significantly lowered the occurrence of the pri-
mary end point (any major CV event) from 10.9% in the low-dose
group to 8.7% in the high-dose group. This represented an abso-
lute decrease of 2.2% and a relative decrease of 22% in CV risk
(hazard ratio, 0.78; P < .001). Overall mortality was not differ-
ent between the 2 treatment groups. High-dose atorvastatin also
reduced the risk of MI, fatal or nonfatal stroke, and CV death.
The high-dose treatment group had a greater occurrence of per-
sistently elevated liver enzymes (1.2% vs 0.2%).
Heart Protection Study (2002)
The HPS was a very large United Kingdom trial that included
over 20,000 adults aged 40 to 80 years who had CAD, arterial
vascular disease, treated hypertension, or diabetes mellitus. The
subjects were randomly assigned to receive simvastatin (40 mg
daily[average compliance, 85%]) or a matching placebo (aver-
age nonstudy statin use, 17%). In 33% of the patients, the base-
line LDL-C values were less than 116 mg/dL; in 25%, 116 to
135 mg/dL; and in 42%, greater than 135 mg/dL. Patients in all
3 baseline cholesterol groups benefited equally from simvastatin.
The average follow-up was 5.5 years. Simvastatin reduced all-
cause mortality by 13%, with an 18% decrease in deaths from
any CV cause and a 24% decrease in major CV events. There
was a 25% reduction in the risk of first ischemic stroke (not hem-
orrhagic stroke). Participants with diabetes had a larger benefit
than participants without diabetes, with a 28% decrease in the
incidence of MI and stroke. The proportional reduction in the
event rate was similar (and significant) in each subcategory of
participant studied, including participants without diagnosed
CAD, participants with cerebrovascular disease, participants
with peripheral artery disease, and participants with diabetes;
men and women; subjects aged either younger than 70 or older
than 70 years at entry; and, most notably, participants who pre-
sented with an LDL-C less than 116 mg/dL or a total cholesterol
less than 193 mg/dL. The benefits of simvastatin were additional
to those of all other cardioprotective treatments.
• Adding simvastatin to existing treatments safely produces substan-
tial additional cardioprotective benefits for a wide range of high-
risk patients, irrespective of their initial cholesterol concentrations
or specific form of vascular disease.
• For high-risk individuals as studied in the HPS, 5 years of simvas-
tatin therapy would prevent 70 to 100 people per 1,000 subjects
from having a major vascular event.
• The size of the 5-year benefit with simvastatin depends chiefly on
the individual’s overall risk of major vascular events rather than on
serum lipid concentrations alone.
• Clear and compelling evidence demonstrates that the rate of non-
fatal cholesterol events, deaths from CV disease, and total mortal-
ity can be reduced with aggressive lipid lowering in patients with
known CAD. This is especially important for patients who have
had a recent MI; the CARE study demonstrated benefit in as little
as 2 months after initiation of treatment.
IDEAL Study (2005)
The IDEAL study was a Northern European study that directly
compared high-dose atorvastatin (80 mg daily) with usual-dose
simvastatin (20–40 mg daily) in 8,888 patients 80 years or younger
who had a history of acute MI. The main end point was any major
coronary event—coronary death, nonfatal MI, or cardiac arrest
with resuscitation. During treatment, the mean LDL-C levels
were 104 mg/dL in the simvastatin group and 81 mg/dL in the
atorvastatin group. A major coronary event occurred in 10.4% of
the simvastatin patients and in 9.3% of the atorvastatin patients
(hazard ratio, 0.89; P = .07). Aggressive lowering of LDL-C did
not result in a significant reduction in the primary outcome of all
major coronary events but did reduce the risk of other composite
secondary end points and nonfatal acute MI. There were no dif-
ferences in CV or all-cause mortality between the groups.
Statins in ACSs or Post-ACS Studies
A number of studies have examined LDL-C-lowering strategies
with statin agents during hospitalization or following discharge
in patients with ACSs.
CARE Trial (1996)
The CARE trial, described earlier in this chapter, tested whether
pravastatin was superior to placebo in 4,159 patients who had
suffered an acute MI and had survived to discharge. Patients
were randomized to pravastatin (40 mg) or placebo approxi-
mately 3 months following hospital discharge and were followed
for nearly 5 years.
• The CARE trial was the first study to demonstrate a benefit of sta-
tin therapy soon after hospitalization for acute MI.
MIRACL Trial (2001)
The MIRACL trial tested if the early use of high-dose atorvas-
tatin would reduce myocardial ischemia in patients hospitalized
with unstable angina or MI. A total of 3,086 patients with total
cholesterol values <270 mg/dL were randomized to atorvastatin
(80 mg) or placebo daily on average 5 days after presentation
 57 Lipid-Lowering Clinical Trials and Medications
with an ACS event. The primary end point was a combination
of death, nonfatal MI, cardiac arrest requiring resuscitation, and
hospitalization for recurrent ischemia. The patients random-
ized to atorvastatin experienced fewer primary end point events
(14.8%) than did the group randomized to placebo (17.4%) (ARR,
2.6%; RRR, 16%; NNT = 38; P = .048). The data from MIRACL
supported earlier initiation of statin therapy and affected clinical
practice significantly.
A to Z Trial (2004)
The A to Z trial compared a strategy of moderate and delayed
LDL-C reduction with simvastatin (20 mg daily) to intense,
immediate LDL-C reduction with simvastatin (80 mg daily) fol-
lowing hospitalization with ACS. Patients were randomized if
their total cholesterol was <250 mg/dL. Patients were followed
for up to 24 months and the primary end point was the com-
bination of CV death, nonfatal MI, stroke, and readmission for
ACS. Patients randomized to the high-dose simvastatin arm
experienced lower primary end point event rates (14.4%) com-
pared with those randomized to the lower-dose simvastatin arm
(16.7%) (ARR, 2.3%; RRR, 11%; P = .14).
PROVE-IT TIMI 22 (2005)
The PROVE-IT TIMI 22 trial tested whether moderate-intensity
LDL-C reduction with pravastatin (40 mg) would be superior
to intense LDL-C reduction from atorvastatin (80 mg) therapy
in 4,162 patients who had been hospitalized with an ACS up to
10 days prior to randomization and provided their total choles-
terol was <240 mg/dL. The primary end point was a combined
end point of all-cause mortality, recurrent MI, unstable angina
requiring hospitalization, or stroke by 2 years. The group ran-
domized to atorvastatin experienced fewer primary end point
events (22.4%) than did the group randomized to pravastatin
(26.3%) (ARR, 1.9%; RRR, 16%; NNT = 53; P = .005). The sur-
vival curves separated by 30 days and remained separated during
the 2.5 years of follow-up. The data from PROVE-IT TIMI 22
established that more aggressive LDL-C reduction was benefi-
cial in the ACS patient population and that clinical benefit would
accrue much earlier than previously observed. An important
finding of this study was that patients who have low CRP levels
after statin therapy have better clinical outcomes than those with
higher CRP levels, after adjusting for LDL-C. Patients who had
LDL-C levels of less than 70 mg/dL and CRP levels of less than
1 mg/L after statin therapy had the lowest rate of recurrent CV
events.
Coronary Artery Regression Studies
Several coronary angiographic trials have examined the effect
of lipid reduction on the regression of CAD lesions. These
angiographic trials used aggressive lipid management, gener-
ally decreasing the LDL-C value by 25% to 40%, with a relative
decrease in MI and coronary mortality of 25% to 45%. Some evi-
dence of CAD regression was demonstrated in 14% to 20% of the
patients, but the most important finding was that CAD progres-
sion was slowed by about 50%. The MAAS trial (simvastatin vs
placebo), the REGRESS trial (pravastatin vs placebo), and the
PLAC-I study (pravastatin vs placebo) all demonstrated signifi-
cant reductions in the rate of progression of angiographic CAD for
patients in whom statin treatment significantly decreased serum
LDL-C values. The reductions in coronary events and death that
567
were observed in the WOSCOP study, the 4S, and the CARE
study all occurred more quickly than would have been predicted
on the basis of coronary atherosclerotic regression. Evidence
now suggests that the primary beneficial effect of statin lipid-
lowering therapy may be atherosclerotic plaque stabilization.
GAIN Trial (2001)
The GAIN trial was an innovative study that used intravascu-
lar ultrasonography to compare the effects of atorvastatin on
plaque volume and composition in 130 patients with established
CAD. After 12 months of atorvastatin therapy, the progression
of plaque volume and thickness decreased and the plaque echo-
genicity increased, indicative of a change in plaque composition
(putatively from a lipid-rich plaque to a fibrotic and calcified
plaque) that would be consistent with an increase in plaque sta-
bility with less risk of precipitation of a future unstable coronary
syndrome.
ASTEROID Trial (2006)
The ASTEROID trial tested the effect of very high-intensity
statin therapy (rosuvastatin, 40 mg daily) on the regression of
coronary atherosclerosis in 507 patients, as determined by IVUS
imaging at baseline and after 24 months of treatment. This study
was a prospective, open-label, blinded end points trial in which
patients were required to have undergone coronary angiogra-
phy for a standard clinical indication, generally chest pain or an
abnormal cardiac stress test. Patients were selected on the basis
of an angiographic stenosis of greater than 20% narrowing in at
least 1 coronary vessel and a target artery suitable for IVUS with
narrowing of no more than 50%. The primary study end point
was change in atheroma volume in the 10-mm subsegment with
the greatest disease severity at baseline.
A subset of 158 patients (30%) were not included in the final
IVUS analysis for several reasons, including withdrawal of
consent (32 patients), adverse events (63 patients), and inade-
quate IVUS images (33 patients). After 24 months, 349 patients
had evaluable serial IVUS examinations. Mean LDL-C levels
decreased by 53%, from 130 mg/dL to 61 mg/dL; HDL-C levels
increased by 15% from 43 to 49 mg/dL. Most patients (78%) had
atheroma regression with a mean percentage decrease in ather-
oma volume of 6.1 mm3 (or 8.5%) in the 10-mm subsegment with
the greatest disease severity.
The study authors concluded that very high-intensity sta-
tin therapy that decreases LDL-C levels to less than currently
accepted guidelines, when accompanied by significant HDL-C
increases, can cause regression of atherosclerosis in CAD
patients. Further studies are needed to determine the clinical
effect of these salutary atheroma volume changes.
ARBITER-6 (2009)
The ARBITER-6 study evaluated a strategy of additional LDL-C
reduction on top of statin therapy with ezetimibe versus a strat-
egy of LDL-C reduction and HDL-C increase with niacin on top
of existing statin therapy. Patients were enrolled who had CAD
or a CAD risk equivalent and LDL-C values <100 mg/dL and
HDL-C values <50 mg/dL. All were on statin therapy prior to
randomization to extended release niacin (2,000 mg/day) or
ezetimibe (10 mg/day). The patients randomized to statin plus
niacin had reductions in mean and maximal carotid intimal
thickness (P = .001) compared with the ezetimibe group. While
568 VI Coronary Artery Disease Risk Factors
not powered to test clinical end points, the group randomized to
statin plus niacin had fewer clinical events (1% vs 5%, P = .04).
The patients on ezetimibe had an increase in carotid intimal-
medial thickness despite evidence of better LDL-C reduction.
Combination Therapy in Refractory
Hyperlipidemia or Mixed Hyperlipidemia
Combination therapy is frequently necessary to treat refractory
hyperlipidemia. Advantages of combination therapy are that it
often allows for the use of lower drug doses, it may reduce the
incidence of medication side effects, it often decreases LDL-C
levels and increases HDL-C levels, and it lowers triglyceride val-
ues very effectively.
Statin With Ezetimibe
The use of a statin and ezetimibe is an excellent, complemen-
tary combination that blocks both hepatic LDL-C production
and intestinal cholesterol absorption, resulting in a decrease
in LDL-C beyond that seen with statin therapy alone. The risk
of statin hepatotoxicity and myositis increases slightly when
ezetimibe is used in combination with a statin.
There are little clinical trial data available comparing the
addition of ezetimibe to statin versus statin alone.
SHARP Trial (2011)
The SHARP trial randomized 9,270 patients with chronic kid-
ney disease to simvastatin (20 mg) with ezetimibe (10 mg) or to
placebo and followed them for CV events for nearly 5 years. The
group randomized to simvastatin with ezetimibe had fewer clini-
cal events than did the placebo group. The primary end point was
time to a first major atherosclerotic event defined as CAD death,
nonfatal MI, nonhemorrhagic stroke, or any arterial revascular-
ization procedure. The group randomized to simvastatin with
ezetimibe had an 11.3% event rate versus 13.4% in the placebo
group (ARR, 2.3%; RRR, 17%; NNT = 43; P = .0021). These
data are among the first to demonstrate a benefit of the combina-
tion of simvastatin with ezetimibe over placebo.
Statin With Bile Acid–Binding Resin
The combination of a statin agent and a bile acid–binding resin is
very effective when either agent alone does not reduce the LDL-C
level to the target goal. This combination is particularly effec-
tive for lowering LDL-C in type IIa hyperlipidemia. With careful
upward titration of the resin dose, the gastrointestinal tract side
effects that are frequently observed with resins may be avoided.
Statin With Fibric Acid Derivative
The combination of pravastatin and gemfibrozil is excellent for
decreasing both triglyceride and LDL-C levels but at an increased
risk of myositis. One approach to treatment of mixed hyperlipi-
demia is to estimate the percentage reduction needed to bring the
LDL-C and triglyceride values into the desired target range.
Monotherapy with high-dose simvastatin or atorvastatin
decreases triglyceride values 25% to 40% and LDL-C values
by 30% to 60%. Combination therapy with a statin and feno-
fibrate or gemfibrozil should be considered if the triglyceride
value needs to be less than the value typically obtained with sta-
tin monotherapy. Fenofibrate can be used to decrease LDL-C by
10% to 20%. If a greater reduction in LDL-C is the goal, use of
gemfibrozil in combination with a statin is necessary. The use
of fibrates alone is occasionally associated with myopathy, and
the combined use of a statin with a fibrate increases the risk of
myopathy substantially. Combination therapy must be instituted
at low doses for both agents, with careful upward titration until
the desired lipid reduction is achieved. Careful monitoring for
muscle and liver toxicity is important.
Treatment of an Isolated Low HDL-C Value
An isolated low HDL-C value is a strong risk factor for CV
disease and may be clinically difficult to treat. Niacin, gemfibro-
zil, and fenofibrate all increase an isolated low HDL-C value by
10% to 15%, whereas most statins increase the HDL-C level by
only 5% to 8%. Niacin, fenofibrate, and gemfibrozil all may be
effective in treating the combination of a low HDL-C level and
an increased triglyceride level. This lipid pattern is often found
in patients with diabetes mellitus or obesity with insulin resist-
ance. Therapy with a statin in combination with niacin, gemfi-
brozil, or fenofibrate effectively treats patients who have a lipid
profile characterized by both a low HDL-C and a high LDL-C
with or without increased triglycerides.
Cholesteryl Ester Transfer Protein
CETP is a plasma protein that mediates the transfer of cholesteryl
esters from HDL to LDL and VLDL particles in exchange for
triglycerides. CETP appears to create a more atherogenic plasma
in that it decreases HDL-C and increases atherogenic small
dense LDL particles. Its physiological function in man is unclear
but it may be a remnant pathway for LDL synthesis in vegetarian
animals. Anacetrapib and dalcetrapib are investigational inhibi-
tors of CETP currently being studied in large, randomized trials
in patients with low HDL-C.
Side Effects and Drug Interactions
of Lipid-Lowering Drugs
The two most common side effects of statins and fibrates are
myositis and an increase in hepatic transaminase enzymes. Both
of these side effects are uncommon and usually disappear when
drug therapy is discontinued. There is an increased incidence of
myositis among patients receiving statins in combination with
high-dose niacin or fibrates. The use of cyclosporine in com-
bination with statins may increase the serum concentration of
the statin and in turn increase the incidence of drug-related side
effects. It is important to decrease the dose of the statin by 50%
when a patient begins receiving cyclosporine. The dose of statin
may be slowly increased with time if the serum cholesterol val-
ues necessitate upward titration. Pravastatin is the best statin to
use with cyclosporine because it is associated with a lower inci-
dence of drug-drug interactions. Table 57.2 summarizes interac-
tions and side effects of lipid-lowering agents.
Common Pitfalls in Lipid Treatment
A major pitfall in the treatment of patients with hyperlipidemia
is a failure to achieve the desired LDL-C target goal, despite a
maximal dose of the administered drug. This is often the result
of patient noncompliance with pharmacologic and, more fre-
quently, nonpharmacologic lipid treatment strategies (ie, diet
and exercise). It is important to also consider an exacerbation of
a coexisting illness such as diabetes mellitus, hypothyroidism,
renal failure, or excess alcohol consumption. Additionally, the
lipid-lowering agent should be taken at bedtime, because the liver
synthesizes cholesterol predominantly during the sleep cycle.
 57 Lipid-Lowering Clinical Trials and Medications 569

Table 57.2. Interactions and Side Effects of Lipid-Lowering Agents

Agent Interactions and Side Effects
Cholestyramine and colestipol Other medications should be given 1 h before or 4–6 h after the resin agent. May impair absorption of vitamins A, D,
E, and K and cause vitamin K deficiency. May alter absorption of digoxin, warfarin, thiazide diuretics, propranolol,
tetracycline, penicillin G, estrogens and progestins, and thyroid supplements. May produce or worsen constipation,
nausea, abdominal pain, flatulence, vomiting, and anorexia
Niacin Patients may experience flushing, itching, tingling, feelings of warmth, headache, rash, upset stomach, hypotension.
May worsen fasting hyperglycemia and cause atrial arrhythmias
Gemfibrozil Contraindicated in renal failure (creatinine >2.0 mg/dL), hepatic dysfunction (including primary biliary cirrhosis), and
preexisting gallbladder disease. Incidence of rhabdomyolysis is increased when agent is used with lovastatin
Fenofibrate Contraindicated in renal failure (creatinine clearance < 50 mL/min) and hepatic dysfunction. Doses of warfarin
must be reduced and prothrombin time monitored frequently. Dose may need to be adjusted when given with
cyclosporine. May increase risk of gallbladder disease
All statins May increase aspartate aminotransferase. Monitor at initiation of therapy and with any change in dose. Rarely
associated with myositis and rhabdomyolysis. Should not be used in pregnant or nursing women. Should not be used
with oral or intravenous antifungal agents. May be used with bile acid–binding resins. No effect on β-adrenergic
blockers, angiotensin-converting enzyme inhibitors, and diabetic agents
Lovastatin Should not be used with gemfibrozil. May cause rhabdomyolysis when used with cyclosporine plus itraconazole or with
erythromycin. Widely studied, with good primary and secondary prevention data
Pravastatin Most widely studied agent, with good primary and secondary prevention data. Little evidence of myopathy when used
with cyclosporine, niacin, or gemfibrozil
Simvastatin Widely studied, with good secondary prevention data. Dose should be decreased when given with cyclosporine. May
potentiate effect of digoxin and warfarin
Atorvastatin A potent agent with a good side-effect profile. Excellent long-term mortality data for primary and secondary prevention
effect
Rosuvastatin Very potent statin
Ezetimibe Blocks cholesterol absorption from the intestine

The following are possible treatment strategies for persistently
increased LDL-C despite seemingly adequate therapy with lipid-
lowering statin medication:
1. Check the lipoprotein (a) value to be certain that there is no interac-
tion from this highly atherogenic lipoprotein fraction.
2. Use a more potent statin, such as atorvastatin, simvastatin, or rosu-
vastatin. Both simvastatin and atorvastatin can be used safely at high
doses (80 mg daily) with additional lipid-lowering benefit.
3. Add ezetimibe 10 mg daily to block intestinal cholesterol absorption.
4. Add a bile acid–binding resin to the statin.
5. Consider the addition of fenofibrate or gemfibrozil to potentiate the
effect of triglyceride lowering in addition to further LDL-C lower-
ing. This strategy increases the risk of side effects.
6. Consider referral to a specialized center that performs LDL apheresis
for the persistently increased LDL-C value despite maximal pharma-
cologic therapy. LDL apheresis removes apolipoprotein B–containing
lipoproteins directly from the blood by extracorporeal circulation
through adsorption columns. It is indicated in the treatment of refrac-
tory hypercholesterolemia despite the use of maximally tolerated
lipid-lowering drug therapy and intense dietary modification.
Special Populations With Hyperlipidemia
Women With Heart Disease
Heart disease is the leading cause of death among women,
although it develops at an older age in nonsmoking women than
in men. The benefit of aggressive lipid-lowering therapy with sta-
tin agents in women with hyperlipidemia is now well established.
Hormone replacement therapy causes a decrease in LDL-C of
about 15%, an increase in HDL-C of about 15%, and an increase
in triglycerides of 25%, but hormone replacement therapy had no
proven CV benefit in the WHI and HERS studies.
Elderly Patients
Elderly patients derive significant primary and secondary CV
prevention benefit from pharmacologic LDL-C reduction, and
the decision to start or not start therapy with cholesterol-lowering
medication should not be based on chronologic age. Elderly
patients derive a greater benefit in absolute terms than younger
patients because the occurrence of CV events is much higher
in the elderly. The following lipid-lowering randomized trials
enrolled older subjects: 4S, CARE, LIPID, and PROSPER.
The 4S, a secondary prevention study of simvastatin included
approximately 1,000 patients older than 65 years who had hyper-
lipidemia and established CAD. Simvastatin caused similar
reductions in serum lipids in elderly and younger subjects and
decreased all-cause mortality and CV events by about one-third
in the elderly.
The CARE trial, a secondary prevention trial of pravasta-
tin, included almost 1,300 patients aged 65 and older. Findings
included a reduction in coronary events by about one-third in
elderly patients, which was approximately a 2-fold greater per-
centage benefit than that seen in younger patients.
The LIPID trial, a secondary prevention trial with pravastatin,
included over 3,500 patients between the ages of 65 and 75 years
who had mild to moderate hyperlipidemia and established CV
disease. Pravastatin reduced CV events and all-cause mortality
by a similar percentage in older and younger patients, but the
absolute benefit to the elderly was greater.
Patients in the PROSPER trial were aged 70 to 82 years
and had a history of CV disease or risk factors for vascular
disease. This study randomly assigned over 5,800 patients
to receive pravastatin (40 mg daily) or placebo. Pravastatin
decreased serum LDL-C by 34% and significantly reduced the
combined CV end point of coronary death, nonfatal MI, and
stroke (hazard ratio, 0.81; P = .014). Mortality from coronary
disease decreased by 24% (P = .043). Overall stroke risk was
unaffected by therapy, but pravastatin decreased the hazard
ratio for transient ischemic attack (hazard ratio, 0.75; P = .051).
There was no significant reduction in all-cause mortality, but the
decrease in cardiac deaths was offset by a significant increase
in new diagnoses of cancer, a finding considered spurious since
570 VI Coronary Artery Disease Risk Factors

a meta-analysis of all major statin trials (including PROSPER) HDL high-density lipoprotein
showed no statistical link between statins and cancer. HDL-C high-density lipoprotein cholesterol
HMG-CoA 3-hydroxy-3-methylglutaryl coenzyme A
• Elderly patients with hyperlipidemia are markedly undertreated IVUS intravascular ultrasonography
with statins, are less compliant with medication, have more statin- LDL low-density lipoprotein
associated side effects, and probably have more drug interactions LDL-C low-density lipoprotein cholesterol
with statins. MI myocardial infarction
NCEP National Cholesterol Education Program
NNT number needed to treat
Patients With Diabetes Mellitus PCI percutaneous coronary intervention
Diabetes mellitus is an important risk factor for the development PPAR-α peroxisome proliferator-activated receptor-α
of CV disease. Most patients with diabetes die of complica- PTCA percutaneous transluminal coronary angio-
plasty
tions from CAD, stroke, or peripheral vascular disease. Diabetic
RRR relative risk reduction
patients without cardiac symptoms or a diagnosis of known CAD TLC Therapeutic Lifestyle Changes
also have a high mortality from CV causes, similar to that of UCSF-SCOR University of California, San Francisco Special-
patients with known CV disease, suggesting that much diabetes- ized Center of Research in Arteriosclerosis
associated CAD is clinically silent or unrecognized and also pro- VLDL very-low-density lipoprotein
viding the rationale for treating all diabetic patients as having a WHO World Health Organization
coronary artery equivalent condition from a lipid management
and CV risk prevention perspective. Hypertriglyceridemia is a
strong predictor of CAD in diabetics. Names of Clinical Trials
Aggressive primary prevention of CAD is important in dia- 4S Scandinavian Simvastatin Survival Study
betic patients with hyperlipidemia. Goal values are LDL-C less A to Z Aggrastat to Zocor
than 70 mg/dL, HDL-C more than 40 mg/dL, and triglyceride ACCORD-Lipid Action to Control Cardiovascular Risk in
less than 200 mg/dL. It can be reasonably argued that all adult Diabetes
diabetic patients should receive a statin medication for primary AFCAPS/TexCAPS Air Force/Texas Coronary Atherosclerosis Pre-
prevention of CV disease unless there is a strong contraindica- vention Study
tion, such as previous hepatotoxicity or myopathy. Pharmacologic AIM-HIGH Niacin Plus Statins to Prevent Vascular Events
therapy—with a statin or with a statin in combination with ARBITER Arterial Biology for the Investigation of
Treatment Effects of Reducing Cholesterol
fibrate—should be initiated in conjunction with nonpharmaco-
ASCOT-LLA Anglo-Scandinavian Cardiac Outcomes Trial—
logic measures, including strict diabetes control, diet, exercise, Lipid Lowering Arm
and an angiotensin-converting enzyme inhibitor (for its cardiore- ASTEROID A Study to Evaluate the Effect of Rosuvastatin
nal protective effect). on Intravascular Ultrasound-Derived Coronary
Atheroma Burden
AVERT Atorvastatin Versus Revascularization Treatment
Recipients of Organ Transplants
CARDS Collaborative Atorvastatin Diabetes Study
Recipients of heart and other organ transplants often manifest CARE Cholesterol and Recurrent Events
an accelerated form of coronary and peripheral vascular disease CLAS Cholesterol Lowering Atherosclerosis Study
(posttransplant vasculopathy) several years after transplant. The DART Dilation Versus Ablation Revascularization
cause of this vasculopathy is probably multifactorial: low-grade Trial
FATS Familial Atherosclerosis Treatment Study
transplant rejection, the untoward effects of chronic treatment
FIELD Fenofibrate (Tricor) Intervention and Event
with immunosuppressive agents (eg, cyclosporine, steroids), Lowering in Diabetes
hypertension, diabetes, and hypercholesterolemia. Although GAIN German Atorvastatin Intravascular Ultrasound
posttransplant vasculopathy is not strongly linked to serum HATS HDL-Atherosclerosis Treatment Study
LDL-C levels, it is recommended that all transplant patients HERS Heart and Estrogen/Progestin Replacement Study
be treated according to targets established by the National HPS Heart Protection Study
Cholesterol Education Program for patients with known CAD. IDEAL Incremental Decrease in End Points Through
This approach often requires pharmacologic therapy with a sta- Aggressive Lipid Lowering
tin agent. Pravastatin is considered the statin of choice for post- LIPID Long-term Intervention With Pravastatin in
transplant patients when the antirejection drug cyclosporine Ischemic Disease
LRC-CPPT Lipid Research Clinics Coronary Primary Pre-
also is being administered. The use of cyclosporine in conjunc-
vention Trial
tion with a statin increases the risk of myopathy significantly, MAAS Multicenter Anti-Atheroma Study
but probably less so with pravastatin than with the other potent PLAC-I Pravastatin Limitation of Atherosclerosis in the
statins. Coronary Arteries
PROSPER Pravastatin in Elderly Individuals at Risk of
Vascular Disease
Abbreviations
REGRESS Regression Growth Evaluation Statin Study
ARR absolute risk reduction SHARP Subcutaneous Heparin and Angioplasty Reste-
CABG coronary artery bypass graft nosis Prevention
CAD coronary artery disease TNT Treating to New Targets
CETP cholesteryl ester transfer protein VA-HIT Veterans’ Affairs High-Density Lipoprotein
CRP C-reactive protein Cholesterol Intervention Trial
CV cardiovascular WHI Women’s Health Initiative
FDA US Food and Drug Administration WOSCOP West of Scotland Coronary Prevention

Novel Biomarkers of Coronary Artery Disease Risk
IFTIKHAR J. KULLO, MD, FACC, FAHA
Assessment of cardiovascular risk is necessary to inform pre-
ventive and therapeutic interventions. Typically, the 10-year
probability of CAD is estimated based on the presence of
‘conventional’ risk factors using several available algorithms
(Figure 58.1). Predictive models based on conventional risk
factors have less than desired accuracy in predicting CAD risk
in an individual, in part because of the widespread prevalence
of these risk factors in the general population. Although most
patients who suffer a cardiovascular event will have one or more
of the conventional risk factors, so do many adults who do not
have CAD. Consequently, there is interest in new biomarkers
Imaging/
function
Novel risk
factors
Family history
Metabolic syndrome
Framingham risk score
Figure 58.1. The Cardiovascular Risk Stratification Pyramid.
(Previously published. See “Credit Lines” section.)
Abbreviations and acronyms are expanded at the end of this chapter.
58
that might improve cardiovascular risk stratification (Tables 58.1
and 58.2). Examples include CRP, fibrinogen, NT-proBNP, Lp
(a), LDL particle size and number, Lp-PLA2, and genetic sus-
ceptibility variants. This chapter provides a brief update on these
‘novel’ risk markers (Figure 58.2).
Clinical Utility of a Novel Biomarker
Clinical utility refers to the potential of a biomarker to improve
outcomes. An initial step in determining the clinical utility of a
biomarker is to ask whether the biomarker better discriminates
between individuals with or without CAD compared to standard
risk profiling. Typically the c-statistic is calculated to assess
the discriminatory power of a biomarker, but newer statistical
approaches are increasingly being used to quantify the incre-
mental predictive value of a biomarker (Table 58.3). Additional
Table 58.1. The Need For Studying Novel Biomarkers of
CAD Risk
Current predictive models have less than desired accuracy in predicting
CAD risk in an individual patient.
Conventional risk factors explain <50% of the variability in quantitative
measures of atherosclerotic vascular disease (assessed by coronary
angiography or electron beam computed tomography).
The risk of CAD varies among different ethnic groups and ‘novel’ risk
markers may partly explain the variation. An example is the elevated
CAD risk of South Asians for a given level of conventional risk factors.
In this ethnic group, ‘novel’ risk markers such as insulin resistance and
Lp (a) may be important in causation of atherosclerosis.
Identifying new risk markers may lead to new preventive and therapeutic
approaches.
571
572 VI Coronary Artery Disease Risk Factors

Table 58.2. Novel Circulating Biomarkers of CAD Risk

Risk Marker Mechanisms Mediating Risk
CRP Unclear whether CRP has pro-atherogenic
effects. It has been reported to
increase expression of tissue factor
and plasminogen activator inhibitor-1,
activate complement, and decrease
nitric oxide production.
Fibrinogen Increased plasma viscosity
Increased platelet aggregability
Vascular smooth muscle proliferation
NT-proBNP Marker of hemodynamic stress
Lipoprotein (a) Impaired fibrinolysis
Delivery of cholesterol at sites of arterial
injury
Vascular smooth muscle cell proliferation
Endothelial dysfunction
LDL particle size Smaller LDL particles can more easily
and number cross into the subintimal space, bind
to intimal proteoglycans, and are
susceptible to oxidation. They also have
reduced affinity for the LDL receptor.
Lp-PLA2 Hydrolyzes oxidized phospholipids in
LDL, leading to the generation of
lysophosphatidylcholine, oxidized
nonesterified fatty acids, as well as other
active phospholipids and inflammatory
mediators
Assays
Immunoassays
(high sensitivity)
Immunoassay
(clotting-time based assay)
Immunoassay
Gel electrophoresis; immunoassays,
lipoprotein electrophoresis
Analytic ultracentrifugation
Density ultracentrifugation
Non-denaturing gradient gel
electrophoresis
Polyacrylamide gel electrophoresis
Nuclear magnetic resonance
spectroscopy
Enzyme-linked sandwich
immunosorbent assay (ELISA)
Treatment
Statins, fibrates, and fish oil lower CRP levels;
drugs that selectively lower CRP are not
currently available.
Fibrates lower fibrinogen levels. Estrogens lower
fibrinogen levels in contrast to their effects
on raising CRP. Drugs that specifically lower
fibrinogen levels are not currently available.
Elevated levels may indicate the need for
more aggressive risk factor modification,
eg, optimizing blood pressure control in
hypertensive individuals.
Niacin neomycin, gemfibrozil, omega-3 fatty
acids, and estrogens have been reported to
lower Lp (a) levels. Apheresis is the most
effective modality but impractical for most
patients. Statins do not appear to have a
significant effect on Lp (a) levels.
Fibrates, niacin, and peroxisome proliferator-
activated receptor (PPAR) agonists increase
LDL particle size. Dual PPAR agonists (α
and γ) may have even more favorable effects
on LDL particle size. Statins have little
effect on LDL particle size and primarily
reduce the number of LDL particles.
Lipid-lowering therapies reduce Lp-PLA2.
Drugs under development that specifically
inhibit Lp-PLA2 activity have been shown
to lower Lp-PLA2 activity and inflammatory
markers.

Abbreviations: CAD, coronary artery disease; CRP, C-reactive protein; Lp-PLA 2, lipoprotein-associated phospholipase AII; NT-proBNP, N-terminal pro-brain type
natriuretic peptide.

questions that need to be answered to understand clinical utility ACC/AHA Guidelines on Novel
include: What are the added benefits and harms for using a bio- Plasma Biomarkers
marker compared to other diagnostic tests? What are the added
According to the ACC/AHA guidelines for cardiovascular risk
benefits and harms for using a biomarker compared to other risk
assessment published in November 2010, measurement of CRP
assessment tools?
should be considered in men 50 years of age or older or women
60 years of age or older with LDL cholesterol <130 mg/dL
(Class IIa, Level of Evidence: B) who may be candidates for
statin therapy. Measurement of CRP may be “reasonable” in
Primary prevention Secondary prevention asymptomatic intermediate-risk men 50 years of age or younger
• Family history of • Early-onset CAD or women 60 years of age or younger for cardiovascular risk
early-onset CAD assessment (Class IIb, Level of Evidence: B). Measurement of
• CAD in absence of
• Intermediate 10-year conventional risk CRP for cardiovascular risk assessment is not recommended in
CAD risk factors asymptomatic high-risk adults, and in low-risk men younger than
• “Aggressive” disease 50 years of age or women 60 years of age or younger. The guide-
lines consider measurement of Lp-PLA2 as ‘reasonable’ for car-
diovascular risk assessment in intermediate-risk asymptomatic
adults. Although there are robust data showing that NT-proBNP
Assess novel and Lp (a) provide incremental information for assessing CAD
risk markers risk, measurement of these biomarkers was not recommended by
the guidelines.
• Motivate patient to make lifestyle changes
Family History and Genetic Markers
• Aggressive treatment of conventional risk factors
CAD has a significant heritable component as exemplified by
• Pharmacologic/nonpharmacologic treatment
twin and family studies. The familial clustering of CAD can
Figure 58.2. A Potential Algorithm for the use of Novel Risk be partly explained by heritable variation in known CAD risk
Markers. CAD indicates coronary artery disease. factors such as hypertension and diabetes, but there is evidence
 58 Novel Biomarkers of Coronary Artery Disease Risk 573

to suggest that family history contributes to an increased risk
of CAD independently of the known risk factors. Family his-
tory has been described as a “. . . free, well-proven, personalized
genomic tool that captures many of the genes and environmental
interactions and can serve as the cornerstone for individualized
disease prevention.” A history of early CAD in a first-degree rel-
ative approximately doubles the risk of CAD.
To date, GWASs have identified ∼25 susceptibility variants
for CAD. The susceptibility variants increase the risk mod-
estly, typically by 10%–40% per risk allele. However, because
many of the alleles are frequent in the population, the population
attributable risk is significant. The most well-known CAD sus-
ceptibility variant is at the 9p21 locus, which is present in 45%
of non-Hispanic whites. Homozygosity for the risk allele results
in ∼50% increase in CAD risk. Many more such loci of weak
effects will likely be identified by GWAS of large sample sizes
and also as a result of meta-analyses that combine results of sev-
eral studies. As additional susceptibility variants are discovered,
a proportion of the population will be identified as harboring
several of the risk-mediating alleles, and knowledge of such
alleles in an individual may yield predictive information that is
incremental to conventional risk factors. Although genetic test-
ing may improve accuracy of risk profiles and thereby aid in
early detection of CAD, it remains to be established whether
genetic testing will improve outcomes. Several aspects need fur-
ther investigation, including how to communicate risk to patients
and the effect of genetic risk communication on motivating life-
style changes.
ACC/AHA Guidelines on Family History
and Genetic Markers
ACC/AHA guidelines give a class I indication for obtaining fam-
ily history of atherosclerotic disease for assessing cardiovascular
risk in asymptomatic adults. The AHA guidelines do not, at pres-
ent, recommend genetic testing for CAD risk.

Table 58.3. Approaches to Assessing Incremental Predictive Summary and Conclusion

Value of a Biomarker Individualized medicine has the potential to significantly improve
Discrimination. Evaluation of diagnostic or predictive tests by receiver the health of the population, by enabling early detection of
operator characteristic (ROC) analysis uses the area under curve (AUC) disease, targeted screening, and personalized drug therapy. With
as a measure of the test’s ability to discriminate individuals with disease advances in genomics and proteomics, ‘individualized’ cardio-
from those without disease. Distributions of biomarkers in individuals vascular risk stratification and targeted therapy may become fea-
with and without CAD typically overlap a great deal, which provides sible. Ideally, a new marker should add to risk prediction beyond
one explanation for the modest increase in AUC seen in most clinical the traditional risk factors, have a standardized and reproducible
biomarker studies of CAD risk. assay with established cutoff points to guide interpretation of the
Net Reclassification Index (NRI) refers to the proportion of persons results, and have available a therapeutic intervention that leads to
who change risk categories when prediction models incorporate new
a reduction in CAD events.
biomarkers. Reclassification refers to the ability of new biomarkers
to move people between discrete risk categories, so that some low
risk individuals may be reclassified as high risk and vice versa. For Abbreviations
CAD, reclassification is meaningful and clinically relevant given the
widespread use of low-, intermediate-, and high-risk categories of ACC American College of Cardiology
10-year CAD risk. If risk categories are defined according to cut points AHA American Heart Association
used to indicate type or intensity of interventions, as is true for CAD, CAD coronary artery disease
reclassification can impact clinical management. CRP C-reactive protein
Integrated Discrimination Improvement (IDI) is another method of GWAS genome-wide association study
assessing the discriminative value of a biomarker beyond known risk LDL low-density lipoprotein
factors. It can be regarded as a continuous version of NRI that uses Lp (a) lipoprotein a
probability differences instead of categories. Lp-PLA2 lipoprotein-associated phospholipase AII
NT-proBNP N-terminal pro-brain type natriuretic peptide

Diabetes Mellitus and Coronary Artery Disease
ROBERT L. FRYE, MD, and DAVID R. HOLMES JR, MD
The prevalence of diabetes mellitus has reached epidemic pro-
portions and continues to increase. Worldwide, the prevalence of
diabetes in all age groups is projected to increase from 2.8% in
2000 to 4.4% in 2030. The total number will accordingly increase
from 171 million to 366 million diabetic patients in 2030. There
is a striking racial distribution (Table 59.1 and Figure 59.1). As
can be seen in the age-adjusted rates of Americans 20 years
or older, the prevalence of diabetes is markedly higher among
blacks and Mexican Americans than whites. Mortality rates
are also higher among blacks, Mexican Americans, and Native
Americans than others. The increase in the prevalence of diabe-
tes has been accompanied by a marked increase in the prevalence
of obesity (Figure 59.2).
Type 2 diabetes has been estimated to account for up to 90%
to 95% of all diagnosed cases of diabetes. In addition to the
staggering number of diagnosed cases, a U.S. Department of
Health and Human Services study reported that approximately
40% of U.S. adults aged 40 to 71 years have prediabetes. This
condition, which is silent and may not be identified by the
patient, results in increasing risk of cardiovascular disease and
in the development of type 2 diabetes. The increasing propor-
tion of younger patients with impaired glucose tolerance or type
2 diabetes is a critical public health issue. It has profound future
implications for an increasing prevalence of cardiovascular dis-
ease in younger age groups after decades have been spent try-
ing to delay the development of cardiovascular disease in the
elderly.
Adverse effects of diabetes on short-, intermediate-, and
long-term outcomes have been the focus of intense investigation.
There are societal costs as well as individual adverse outcome
measures. It has been estimated that the direct cost of heart
Abbreviations and acronyms are expanded at the end of this chapter.
574
59
disease related to type 2 diabetes is $174 billion (in US in 2007).
As the prevalence increases, this cost will increase accordingly.
Heart disease is the leading cause of diabetes-related mortal-
ity. Almost two-thirds of diabetics die of heart or blood vessel
disease, and death rates from heart disease are two to four times
higher among diabetic patients than among nondiabetic patients.
In addition, diabetes is associated with an increase in the risk of
stroke, with a relative risk that ranges from 1.6 to nearly 6.0. In
part, this increased risk is related to the increased prevalence of
hypertension, which is seen in almost 75% of diabetics and is
documented in recent National Institutes of Health statistics.
There are substantial differences between men and women
in the outcome of diabetes. Women with diabetes have a twofold
increase in the age-adjusted prevalence of major cardiovascular
disease compared with women without diabetes. Of particular
concern is evidence that the cardiovascular mortality rate among
patients with diabetes is increasing, particularly among elderly
women. Cardiologists must be aware of these trends and under-
stand the influence of type 2 diabetes on CAD.
Most patients with type 2 diabetes are obese, but the link
between obesity and type 2 diabetes is complex. Abdominal or
visceral obesity provokes insulin resistance, which is the funda-
mental problem in people with the metabolic syndrome, defined
by a cluster of risk factors identifying patients at high risk of
type 2 diabetes. As insulin resistance increases, hyperglyce-
mia persists in spite of hyperinsulinemia, leading to the clinical
consequences of type 2 diabetes. Hyperglycemia-induced tissue
damage is provoked in part by the glycation of proteins with the
generation of advanced glycation end products, which have toxic
effects, including excess oxidative stress, endothelial dysfunc-
tion, impaired fibrinolysis, and proinflammatory gene expression.
Although insulin has beneficial effects in glycemic control, it has
other complex actions. It exhibits both anti-atherosclerotic and
 59 Diabetes Mellitus and Coronary Artery Disease 575

Table 59.1. U.S. Demographic Data for Diabetes Mellitusa

Prevalence of Prevalence of Incidence of Mortality Hospital
Physician-Diagnosed Undiagnosed Prevalence of Diagnosed (Diabetes) Discharges
Population Group Diabetes (2003) Diabetes (2003) Prediabetes (2003) Diabetes (2003) b (2003) Cost (2003)
Total 14,100,000 (6.7%) 6,000,000 (2.9%) 14,700,000 (7.0%) 1,500,000 73,965 597,000 $132 billion
Total males 7,000,000 (7.2%) 3,000,000 (3.0%) 8,600,000 (8.9%) ... 35,257 (47.7%) c 286,000 ...
Total females 7,100,000 (6.3%) 3,000,000 (2.7%) 6,100,000 (5.4%) ... 38,748 (52.4%) c 314,000 ...
NH white males 6.2% 3.0% 8.6% ... 28,765 ... ...
NH white females 4.7% 2.7% 4.6% ... 30,189 ... ...
NH black males 10.3% 1.3% 8.3% ... 5,401 ... ...
NH black females 12.6% 6.1% 5.9% ... 7,419 ... ...
Mexican-American 10.4% 3.5% 8.7% ... ... ... ...
males
Mexican-American 11.3% 1.8% 7.2% ... ... ... ...
females
Hispanics or Latinosd 8.6% ... ... ... ... ... ...
Asiansd 6.5% ... ... ... ... ... ...
American Indians/ 12.2% ... ... ... ... ... ...
Alaska Nativesd

Abbreviations: Ellipses indicate data not available; NH, non-Hispanic.

a
People with undiagnosed diabetes had fasting blood glucose ≥126 mg/dL but did not report being told they had diabetes by a health care provider.
Patients with prediabetes had fasting blood glucose of 100 to <126 mg/dL (impaired fasting glucose). Patients with prediabetes also had impaired
glucose tolerance. Percentages for racial/ethnic groups are age-adjusted for Americans 20 years or older. Mortality data represent underlying cause
of death only; data for white and black males and females include Hispanics. Hospital discharge data include people discharged alive or dead. Data
are from the following: Centers for Disease Control/National Center for Health Statistics; National Health Interview Survey; National Health and
Nutrition Examination Survey; National Heart, Lung, and Blood Institute; and National Hospital Discharge Survey.
b
Preliminary.
c
These percentages represent the proportion of total diabetes mellitus mortality that is males vs females.
d
Data are estimates for Americans 18 years or older.

pro-atherosclerotic effects. In addition, in many studies hyperin-
sulinemia has been associated with a worse clinical outcome.
Cardiologists need to understand that fat cells (adipocytes)
are active metabolically and are the source of several cytokines
and other substances that not only influence insulin sensitivity
and resistance but have profound effects on the cardiovascular
system. An example is leptin, which causes sympathetic stim-
ulation, oxidative stress, and enhancement of platelet aggrega-
tion while influencing insulin responsiveness in a tissue-specific
manner. Leptin has been demonstrated to be associated with
more extensive angiographic evidence of CAD. Another example
is adiponectin, which has anti-inflammatory and anti-atheroscle-
rotic effects; a low level has been correlated with the presence of
CAD and, consequently, insulin resistance with type 2 diabetes.
14
% of Group Population
Complex central nervous system relationships have been identi-
fied and studied. For example, cerebral pathways may be involved
in the susceptibility to obesity. The importance of these has been
emphasized in recent reports of inhibitors of cannabinoid recep-
tors, which are involved in significant weight loss and a decrease
in insulin resistance.
Symptoms typically associated with myocardial ischemia or
myocardial infarction in patients without type 2 diabetes may be
absent or muted in patients with type 2 diabetes. Recognizing
myocardial ischemia as the cause of an “angina equivalent” of
dyspnea or other stress-related symptoms is essential in analyz-
ing clinical symptoms in patients with type 2 diabetes. Silent
infarction is more common in patients with type 2 diabetes, and
infarctions have been found to be larger among those with dia-
betes than among those without diabetes. Congestive heart fail-
ure in diabetic patients may be due to ischemia or infarction or
to diabetic cardiomyopathy without epicardial coronary artery

12 flow-limiting disease. In these patients, the influence of myocar-

dial insulin resistance may have a role along with lipotoxicity in
10
more obese patients.
8 Clinical outcomes are worse among patients with type 2 dia-
6 betes than among those without diabetes regardless of the clinical
4 event. However, in acute STEMI, patients with diabetes receive
short-term benefit from thrombolysis and primary angioplasty, as
2
do patients without diabetes. With longer-term follow-up, those
0 with diabetes have higher event rates. This is true for patients
NHW AA NHB H
with other acute coronary syndromes. The original DIGAMI
Ethnic Group trial documented the survival benefit of aggressive control of
Figure 59.1. Age-adjusted Prevalence of Physician-diagnosed blood glucose concentrations with insulin in acute myocardial
Diabetes Mellitus in Americans 20 years or Older by Race/Ethnicity infarction. Other studies in medical intensive care units have
(2007–2009). NHW indicates non-Hispanic white; AA, Asian confirmed the importance of controlling hyperglycemia in criti-
American; NHB, non-Hispanic black; H, Hispanic. cally ill patients regardless of the presence of diabetes.
576 VI Coronary Artery Disease Risk Factors
40
Overweight
30
Percent
20
Obese
10
Extremely obese
0
1960-1962 1971-1974 1976-1980
Figure 59.2. Trends in Overweight, Obesity, and Extreme Obesity Among Adults Aged 20–74 Years: United States, 1960–2008.
Outcomes with revascularization in patients with type 2 dia-
betes have been of great interest, particularly after the unex-
pected observation from the original BARI trial: patients with
treated diabetes who were randomly assigned to receive PTCA
had worse 5-year and 10-year survival than those randomly
assigned to receive CABG. This was true only if patients hav-
ing CABG received at least one internal mammary graft. The
difference in survival specifically related to cardiac mortality.
In the BARI registry, patients with type 2 diabetes fared better
with PTCA; it appeared that the higher-risk patients received
CABG. All patients in the BARI trial had multivessel CAD
that was selected on the basis of suitability for both proce-
dures; thus, patients with the most severe anatomical disease
were excluded. As a result of the BARI trial, CABG with at
least one internal mammary graft was recommended as the
preferred therapy for patients with characteristics of those in
the randomized trial. However, much has changed since the
BARI trial, with the introduction of bare-metal and drug-
eluting stents, use of clopidogrel, glycoprotein IIb/IIIa platelet
inhibitors, and widespread use of statins. All these interven-
tions have improved outcomes for patients with type 2 diabetes
undergoing PCI.
The SYNTAX trial has been reported with important obser-
vations comparing outcomes after CABG and PCI with drug-
eluting stents. While controversial in drawing final conclusions
regarding noninferiority of PCI it is clear the SYNTAX angi-
ographic score classifies extent of myocardial ischemia, and
such stratification demonstrates similar outcomes with PCI
and CABG early on in those with lower SYNTAX scores (less
extensive disease) while a clear advantage of CABG is observed
in those with the highest SYNTAX scores (greater ischemia).
These generalizations appear to apply to those with diabetes
mellitus as well.
The results of the BARI 2D trial have been reported as
well, with no clear advantage of insulin sensitization therapy
for control of glycemia and no advantage of early revascu-
larization compared to optimal medical therapy as an initial
strategy. However, in those with the most extensive disease
selected for CABG a reduction in myocardial infarction was
observed. FREEDOM trial will provide a comparison of cur-
rent state-of-the-art PCI and CABG in patients with diabetes
mellitus.
1988-1994 1999-2000 2007-2008
2003-2004
New data from various trials clarify issues in the medical
management of patients with type 2 diabetes and CAD in rela-
tion to the following:
1. Glycemic control
2. Blood pressure control
3. Effects of TZDs and metformin
4. Therapy for hyperlipidemia
5. Use of ACE inhibitors and ARBs
6. Antiplatelet therapy
The UKPDS asked the following question: Does intensive
glucose control decrease the risk of macrovascular or micro-
vascular complications? At 10 years, hemoglobin A1c was, on
average, 7.0% in patients receiving intensive therapy with insulin
and sulfonylureas and 7.9% in patients receiving dietary therapy.
Patients were free of evidence of CAD at entry. There was a sig-
nificant reduction in microvascular complications with intensive
glucose control and a reduction in myocardial infarction rates,
but it was not statistically significant. In addition, blood pres-
sure control, even though it did not achieve current target levels,
resulted in dramatically lower rates of congestive heart failure,
myocardial infarction, and stroke. In a substudy of the UKPDS,
the use of metformin in obese patients significantly reduced the
occurrence of myocardial infarction. A more recent study of the
Medicare database of acute myocardial infarction also showed
that in patients with diabetes the combination of a TZD and
metformin is associated with lower mortality with myocardial
infarction than either drug alone or only a sulfonylurea. The
combination of TZD and metformin reduces insulin resistance.
The results of the ACCORD study provided a shock for those
believing that aggressive glycemic control would enhance sur-
vival. Indeed, the trial was stopped early because of excess mor-
tality among those randomized to aggressive glycemia control.
However it was noted that in this group there was a reduction in
rates of MI.
TZD drugs, which increase insulin sensitivity, have now fallen
into some disrepute as a result of a meta-analysis that suggested
an elevated risk of myocardial infarction in diabetic patients uti-
lizing these drugs. In spite of the well documented advantageous
effects of TZD on secondary endpoints related to atherosclerosis
and restenosis of stents black box warnings and reports of poten-
tial adverse cardiovascular effects have drastically limited their
 59 Diabetes Mellitus and Coronary Artery Disease 577

current use. Sulfonylurea drugs, which are used for glycemic diabetes. Advantages, if any, of the newer antiplatelet drugs in
control because they stimulate β cells to release insulin, are of patients with diabetes mellitus have not been demonstrated.
great interest to cardiologists. These drugs stimulate the adeno- Cardiologists must also be aware of the profound importance
sine triphosphate–dependent potassium ion (K ATP) channels of of lifestyle changes in managing CAD in association with type
the β cells, but to some degree they also stimulate these channels 2 diabetes. Indeed, type 2 diabetes mellitus can be prevented by
in the heart. Variation between sulfonylurea drugs in relation to major lifestyle intervention. A major challenge is losing weight,
the extent of binding to K ATP cardiac channels is important, since and some patients may benefit from bariatric surgery. Bariatric
those with high cardiac binding may have profound effects on surgery may eliminate or greatly reduce the severity of type 2
the heart, including inhibition of ischemic preconditioning, and diabetes.
perhaps on the cardiac rhythm. In observational studies, mortal-
ity from primary angioplasty in STEMI has been higher among Suggested Reading
patients with diabetes receiving sulfonylurea drugs.
Treatment strategies for control of hyperlipidemia should be Bypass Angioplasty Revascularization Investigation (BARI) [database
on the Internet]. Available from: http://www.edc.pitt.edu/bari/.
aggressive in patients with diabetes and CAD. Use of statins has
Future Revascularization Evaluation in Patients With Diabetes Mellitus:
been shown to provide similar benefit between patients with or Optimal Management of Multivessel Disease (FREEDOM) [data-
without diabetes. In addition, patients with diabetes frequently base on the Internet]. Available from: http://www.clinicaltrials.gov/
have a pattern of hypertriglyceridemia and low levels of high- ct/show/NCT00086450.
density lipoprotein cholesterol (as discussed for metabolic syn-
drome). These patients remain a challenge as the Lipid Substudy
Abbreviations
of ACCORD showed that there was no advantage of adding feno-
fibrate to simvastatin, and the AIM-HIGH Study was stopped ACE angiotensin-converting enzyme
prematurely because of no advantage of the combination of nia- ARB angiotensin-receptor blocker
cin plus a statin over the use of a statin alone and there was an CABG coronary artery bypass grafting
CAD coronary artery disease
increase in ischemic stroke in those assigned to niacin therapy.
CETP cholesterol ester transfer protein complex
We must await the results on ongoing trials assessing the out- PCI percutaneous coronary intervention
come of CTEP inhibition as an approach to elevating HDL cho- PTCA percutaneous transluminal coronary angioplasty
lesterol and hopefully reducing cardiovascular events. STEMI ST-segment elevation myocardial infarction
In diabetic patients, screening is necessary to detect microal- TZD thiazolidinedione
buminuria, which is a prognostic marker not only for progression
to renal failure but for cardiac events. Fortunately, several large
trials have documented that probability of progression to renal Names of Clinical Trials
failure is reduced with ACE inhibitors as well as with ARBs. ACCORD Action To Control Cardiovascular Risk in Diabetes
Current guidelines recommend that all patients with type 2 dia- AIM-HIGH Atherothrombosis Intervention in Metabolic Syndrome
betes and microalbuminuria receive an ACE inhibitor or ARB with Low HDL Cholesterol/High Triglyceride and
regardless of blood pressure. In most patients with diabetes and Impact on Global Health Outcomes
hypertension, more than one antihypertensive drug is required BARI Bypass Angioplasty Revascularization Investigation
and all classes of drugs may be helpful. Diuretics, sometimes BARI 2D Bypass Angioplasty Revascularization Investigation in
feared because of adverse metabolic effects, have been shown to Type 2 Diabetics
be safe and effective. DIGAMI Diabetes Mellitus Insulin-Glucose Infusion in Acute
Myocardial Infarction
Antiplatelet therapy is crucial. Use of aspirin is recommended
FREEDOM Results of the Future Revascularization Evaluation in
in all patients with diabetes and CAD, and the beneficial effects Patients With Diabetes Mellitus: Optimal Management
of clopidogrel before or at PCI have been well established, with of Multivessel Disease
the recommendation for continued use for 6 to 12 months after the SYNTAX Synergy Between Percutaneous Coronary Intervention
procedure. Glycoprotein IIb/IIIa inhibitors have also been shown With Taxus and Cardiac Surgery
to reduce major cardiac events after PCI in patients with type 2 UKPDS United Kingdom Prospective Diabetes Study

Hypertension
ADRIAN J. B. BRADY, BSc, MD, R. SCOTT WRIGHT, MD,
and JOSEPH G. MURPHY, MD
Definition
Hypertension is the clinical condition of sustained high blood
pressure. Untreated, hypertension promotes atherosclerotic
cerebral and peripheral arterial vascular disease, carotid, and
coronary artery disease. Directly, it causes hypertensive left
ventricular hypertrophy and heart failure, renal failure, and reti-
nal damage. It is the most important risk factor for hemorrhagic
stroke and causes hypertensive encephalopathy.
The detection and treatment of hypertension is the single
most important and cost effective intervention in CV medicine
worldwide, according to WHO. While detection of hypertension
is straightforward, and effective drugs are both inexpensive and
readily available, hypertension remains one of the most impor-
tant chronic conditions seen by all physicians, with a patient life-
time risk of developing hypertension well in excess of 50% in
all developed countries. More than 90% of patients with hyper-
tension have primary, or essential, hypertension where the basic
cause is fundamentally unknown, but a family history is common
and unhealthy lifestyle choices (obesity, lipid rich diet and lack
of exercise) are frequent important cofactors. A few individuals
have secondary hypertension caused by a structural or endocrine
abnormality, and specific interventions may be curative.
Traditionally, the upper limits of blood pressure used in the
United States have been 139 mm Hg for systolic blood pressure
and 89 mm Hg for diastolic blood pressure. JNC 7 recognized
the continuous and consistent relationship (which is independent
of other risk factors) between blood pressures that exceed 119/79
mm Hg and CV events. Hence, because blood pressures greater
than this level carry sufficient risk for complications, they are no
longer considered normal. JNC 8 is due to report during 2012, but
the definitions of stages of hypertension are unlikely to change.
Abbreviations and acronyms are expanded at the end of this chapter.
578
60
Blood Pressure Staging
There is a an increased and escalating risk of CV events for all
individuals with a sustained blood pressures in excess of 119/79
mm Hg. As a consequence of this elevated blood pressure–CV
risk relationship at pressures greater than 119/79 mm Hg, the
JNC 7 staged blood pressures across four ranges (Table 60.1).
With the blood pressure–CV risk relationship now well recog-
nized for systolic blood pressures between 120 and 139 mm Hg
and diastolic blood pressures between 80 and 89 mm Hg, a level
previously considered borderline, blood pressures within these
ranges must now be considered as greater than desirable. Because
the previous designation of high normal for these pressure ranges
failed to carry sufficient impact to bring about action related to
long-term risk, the JNC 7 classified blood pressures within this
range as prehypertension.
The benefit of treatment in reducing CV and cerebrovascular
deaths when blood pressure is in the ranges of 140 to 159 mm Hg
(systolic) and 90 to 99 mm Hg (diastolic) prompted the JNC 7 to
describe these blood pressures as stage 1 hypertension. Differing
from the previous report, JNC 6, the current document classi-
fies pressures greater than 160/100 mm Hg as stage 2 hyperten-
sion. Decreasing the classification from six to four stages should
simplify decision treatment options for health care providers and
provide guidance for hypertension trials. Older patients may typ-
ically have high systolic BP with normal or even low diastolic BP,
termed systolic hypertension.
US and Global Demographics
With hypertension defined as blood pressure of 140/90 mm Hg
or greater, it is estimated that approximately 50 million indi-
viduals within the United States meet this criterion for the
disease. Worldwide, 20 times that many are hypertensive. For
 60 Hypertension 579

Table 60.1. JNC 7 Blood Pressure Staging (Figures 60.3–60.5); their conclusion is that many patients fail to
receive optimal treatment for an easily diagnosed condition for
Blood Pressure, mm Hg which numerous effective treatments are readily available.
Stage Systolic Diastolic
Normal <120 and <80 BP Measurement
Prehypertension 120–139 or 80–89
Stage 1 hypertension 140–159 or 90–99 Accurate BP measurement is the first step in the evaluation and
Stage 2 hypertension ≥160 or ≥100 treatment of the hypertensive patient. The individual should
be seated comfortably for 5 minutes, the back fully supported
Abbreviation: JNC 7, Seventh Report of the Joint and the brachial artery maintained at heart level. The positive
National Committee on Prevention, Detection, results of hypertension interventional trials reflect measurements
Evaluation, and Treatment of High Blood Pressure.
using a cuff with a bladder that encompasses at least 80% of the
Previously published. See “Credit Lines” section.
upper arm.
Stimulants, such as caffeine and smoking, should be avoided
a 55-year-old normotensive individual, this can be interpreted for the hour prior to blood pressure measurement. This is an
as a 90% lifetime risk of hypertension. The new JNC 8 defini- important caveat for patients monitoring blood pressure at home.
tions will increase the number of individuals at risk of the con- Crucially, individuals must not talk during BP measurement;
sequences of abnormal blood pressure: CV, cerebrovascular, and talking increases BP by up to 20 mm Hg.
renal diseases. Office BP measurement is a snapshot of an individual’s blood
Worldwide, WHO has determined that hypertension is the pressure and use of ABPM gives a much greater insight into the
single greatest modifiable risk factor for CV disease, causing person’s actual BP levels (Figure 60.6). A normotensive indi-
more disability and premature death than smoking, dyslipidemia, vidual’s average blood pressure does not exceed in a sustained
obesity, diabetes and the other classical risk factors for heart dis- fashion 135/85 mm Hg in the awake state or 125/75 mm Hg
ease (Figure 60.1). while asleep. Ambulatory blood pressure monitoring provides
Modern combination therapy for hypertension with achieve- valuable insight into the management of resistant hypertension,
ment of target BP early in the course of the disease will reduce suspected medication-related hypotensive symptoms, stress-
an individual’s risk of stroke by 50%, MI by at least 30%, and related (white coat) increases in blood pressure, and autonomic
substantially protect against hypertensive cardiac and renal fail- dysfunction. The circadian rhythm in blood pressure in normo-
ure (Figure 60.2). tensive individuals (“nocturnal dippers”) is maintained in most
Hypertension is both underdiagnosed and, when diagnosed, patients who have hypertension even though the pressures are
is often undertreated. Using the older 140/90 mm Hg as the higher throughout the 24-hour cycle. The failure of the blood
lower limit of hypertension, the NHANES has monitored pressure to decrease during sleep (“nondippers”) correlates with
hypertension awareness, treatment, and control over 25 years an increased risk of CV events.

Attributable deaths (thousands)

0 2000 4000 6000 8000

High blood pressure

Smoking
High cholesterol
Child underweight for age
Unsafe sex
Low fruit and vegetable intake
Overweight and obesity
Physical inactivity
Alcohol abuse
Indoor air pollution from solid fuels
Unsafe water, sanitation, and hygiene
Zinc deficiency
Urban air pollution
Vitamin A deficiency East Asia and Pacific
Iron deficiency anemia Europe and Central Asia
Latin America and Caribbean
Contaminated healthcare injections
Middle East and North Africa
Illicit drug use
South Asia
Unmet contraception need Sub-Saharan Africa
Child sexual abuse High-income

Figure 60.1. Global Risk Factors for Health, World Health Organization, 2006.
580 VI Coronary Artery Disease Risk Factors

A B
Prospective Cohort Studies Collaboration Asia Pacific Cohort Studies Collaboration
(n=958,074) (n=425,325)
Age at risk
(years)
Age at risk
256 80-89 64 (years)
(floating absolute risk and 95% CI)

128 ≥70
70-79 32
60-69

Fatal and non fatal stroke

64
60-69 16 <60
Stroke mortality

32

HR (95% CI)
50-59 8
16
4
8
2
4
1
2

1 0.5

0.5 0.25
110 130 150 170 190 110 130 150 170 190
Usual systolic blood pressure (mm Hg) Usual systolic blood pressure (mm Hg)

Figure 60.2. Blood Pressure and Stroke Risk—Global and Asian-Pacific Data.

Baseline Investigations in the

Self-measurement of blood pressure (HBPM) is an alterna-
tive means of monitoring blood pressure. In general, arm cuff
machines are preferred to wrist monitors, but it is vital to use a
validated BP monitor (Figure 60.7).
The long-term PAMELA Italian follow-up study by Mancia et
al has shown that office, ABPM, and HBPM are equally predic-
tive of CV risk over 10 years, so likely all three methods of BP
determination are valid. Day-to-day variability in BP has been
suggested as a novel risk factor among hypertensive patients, but
has yet to be tested prospectively.
Hypertensive Patient
All patients should have baseline renal function, fasting glucose,
and lipids, thyroid function, uric acid, and hematology values
determined. An ECG should be performed as well as echocar-
diography to evaluate for LVH or hypertensive diastolic dysfunc-
tion. A dipstick test for proteinuria, together with measurement
of urinary albumin-creatinine ratio, is mandatory and a 24-hour
urinary protein and creatinine clearance if there is evidence of
proteinuria on the screening dipstick.

100 1999-2000
2001-2002
2003-2004
80 2005-2006
2007-2008

60
Percent

40

20

0
Age 18-39 Age 40-59a Age 60 and Non-Hispanic Non-Hispanic Mexican
oldera whitea blacka American
a
Statistically significant in trend: P≤.05

Figure 60.3. Age-specific and age-adjusted awareness of high blood pressure among US adults with high blood pressure: 1999–2000 through
2007–2008. (Previously published. See “Credit Lines” section.)
 60 Hypertension 581

100 1999-2000
2001-2002
2003-2004
80 2005-2006
2007-2008

Percent 60

40

20

0
Age 18-39a Age 40-59 Age 60 and Non-Hispanic Non-Hispanic Mexican
oldera whitea blacka Americana
a
Statistically significant in trend: P≤.05

Figure 60.4. Age-specific and age-adjusted treatment of high blood pressure among US adults with high blood pressure: 1999–2000 through
2007–2008. (Previously published. See “Credit Lines” section.)

If the diagnosis of hypertension is in doubt a 24-hour ABPM
is carried out.
In younger patients < 50 years of age renin/aldosterone levels
are measured to screen for primary aldosteronism; 24-hour urine
analysis for fractionated metanephrines and catecholamines to
screen for catecholamine secreting tumors (pheochromocytoma),
MRI renal angiography to examine the renal vasculature (renal
artery stenosis and fibromuscular dysplasia) and kidney paren-
chyma, together with identification of adrenal gland masses.
Important clinical signs in hypertension are the radial-femoral
delay characteristic of coarctation of the aorta and renal bruits
in renal artery stenosis. Absence of these clinical signs does not
exclude these diagnoses and aortic imaging should be considered
in young hypertensive patients.
Primary Prevention of Hypertension
Lifestyle choices are major contributors to hypertension—
sodium intake, alcohol consumption, excess weight, and lack
of exercise. Even modest weight loss results in a significant
decrease in systolic and diastolic blood pressures in patients with
established hypertension while the incidence of new hyperten-
sion over nearly a decade can be decreased by almost one-third
with optimal lifestyle choices. The Nurses’ Health Study in the
US showed that at least 1 mm Hg of blood pressure is lost for
each kilogram of weight loss, a valuable pointer for lifestyle
intervention. With the pandemic of obesity across the world,
the importance of weight management cannot be overstressed.
Exercise training lowers blood pressure and also makes treat-
ment of hypertension more responsive to prescribed therapy.
Thirty minutes of exercise 5 times per week is recommended by
many authorities.
Changes in diet can also benefit hypertensive patients. Diets
high in fruits and vegetables and low in saturated fats can
decrease blood pressure while the addition of sodium restrictions
increases the benefit. As with weight loss, sodium intake reduc-
tion has been associated with significantly lower normal blood
pressures in meta-analyses. Moderation in alcohol intake and

60 1999-2000
2001-2002
2003-2004
50
2005-2006
2007-2008
40
Percent

30

20

10

0
Age 18-39a Age 40-59a Age 60 and Non-Hispanic Non-Hispanic Mexican
oldera whitea blacka Americana
a
Statistically significant in trend: P≤.05

Figure 60.5. Age-specific and age-adjusted control of high blood pressure among US adults with high blood pressure: 1999–2000 through
2007–2008. (Previously published. See “Credit Lines” section.)
582 VI Coronary Artery Disease Risk Factors

24h NIBP: Profile representation

[mmHg] [bpm]
(___) (___)
300
280
260 210
240 200
220 190
200 180
180 170
160 160
140 150
120 140
100 130
80 120
60 110
40 100
20 90
0 80
70
M 60

16 17 18 19 20 21 22 23 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Time
- Day - Night upper limits (syst., diast.) M - manually released

Figure 60.6. Illustration of Ambulatory Blood Pressure Monitoring.

regular exercise also decrease blood pressure in individuals con-
sidered normotensive. Sustained long term use of NSAIDs causes
sodium retention and can raise blood pressure. Furthermore,
NSAIDs by the same action reduce the efficacy of ACEIs, ARBs,
and diuretics, so prescription of anti-inflammatory drugs should
be minimized in hypertensive patients.
table is crucial. Recognition of the importance of lifestyle mod-
ification to hypertension is central in all the JNC 7 treatment
recommendations.
The correlations between specific behavioral changes and
blood pressure reductions are outlined in Table 60.2. These chan-
ges require constant reinforcement at each patient encounter.

Pharmacologic Treatment of Hypertension

Lifestyle Modification as Treatment
Just as with primary prevention, lifestyle modification can result
in significant blood pressure reduction in the hypertensive pop-
ulation. Correcting each of the lifestyle contributors should be
the foundation and mainstay of therapy and must be included in
any treatment regimen that the practitioner chooses. Avoidance
of added dietary salt, either in processed foods or added at the
Figure 60.7. Illustration of Home Blood Pressure Monitoring.
In an attempt to simplify clinical decision making, the JNC 7
created a treatment algorithm incorporating the result of large
treatment trials that had been published since the JNC 6. The
algorithm (Figure 60.8) focuses on compelling indications as key
decision points in the choice of pharmacologic agents. American
guidelines have been substantially influenced by the very large
ALLHAT trial, published in 2002. It compared the benefits of
ACEIs, dihydropyridine CCBs, and β-blockers with the ben-
efits of diuretics in controlling hypertension and positively
affecting outcomes of diseases associated with hypertension.
α2-Adrenergic receptor blockers were removed from the study
before it was completed because of concerns about this class
of drug and a propensity for congestive heart failure in patients
receiving them. The results showed equal efficacy in controlling
blood pressure among the agents studied for most patients with
hypertension, and similar rates of MI and stroke. JNC 7’s inter-
pretation of ALLHAT was that except when compelling indica-
tions determine alternative first-line therapy, the cost, ease of
administration, and efficacy favor diuretic therapy for the major-
ity of patients with essential hypertension. A ten-year follow up
study of the original ALLHAT cohort was published in 2009,
with essentially unchanged scientific findings.
Since the publication of JNC 7 in 2003 there have been many
large outcome trials in hypertension, together with meta-analyses
of large hypertension databases. There continues to be much
debate over whether one class of drug is better than another
 60 Hypertension 583

Table 60.2. Lifestyle Modification Combination Therapy for Hypertension

Approximate Reduction in
Modification Systolic Blood Pressure, mm Hg
Weight reduction 5–20 (per 10-kg loss)
Adopt DASH eating plan 8–14
Dietary sodium reduction 2–8
Physical activity 4–9
Moderation of alcohol consumption 2–4
Abbreviation: DASH, Dietary Approaches to Stop Hypertension.
Previously published. See “Credit Lines” section.
(Figure 60.9). What all authors agree on is that lowering blood
pressure is the most important thing; which drugs are chosen is
important, but not as critical as reducing the BP.
Two recent large meta-analyses summarize the findings well.
The Blood Pressure Treatment Trialists Collaboration in 2008
incorporated data on 190,606 individuals from 31 trials and con-
cluded that all drug classes were broadly equivalent in their pro-
tection against major events. Law et al, in 2009, studying data on
464,000 hypertensive patients came to similar conclusions, with
two additional findings:
1. CCBs have a small advantage in protecting against stroke.
2. β-Blockers have a particular benefit in preventing recurrent MI in
the early years following a first event.
The reality is that many or indeed most patients with hyper-
tension will require more than one drug, but there are compel-
ling indications for particular drug types in certain conditions,
discussed below.
Because most hypertensive patients require more than one agent
to control their blood pressure, knowledge of appropriate combi-
nation therapy is important. Multiple-drug therapy often affords
an excellent level of control without the side effects that may be
encountered with single-drug therapy at maximal dose.
Recent trials have examined different combinations of therapy
to identify whether there might be benefits. The European ASCOT
trial concluded in their study of >19,000 patients that an ACEI
plus CCB was better than a β-blocker plus diuretic. Jamerson
in the US compared ACEI plus CCB with ACEI plus diuretic,
finding the ACEI-CCB combination being superior. However, in
ASCOT patients on ACEI plus CCB had slightly lower BP, and
fewer developed diabetes; and in Jamerson’s trial a rather low
dose of diuretic was compared against a high-dose CCB.
In practical terms since the renin-angiotensin system is cen-
tral to BP control it seems sensible to include an ACEI or ARB
in combination therapy. Likewise since sodium retention prob-
ably contributes to sustained high blood pressure, inclusion of
chlortalidone or a thiazide diuretic is often needed. CCBs are
powerful agents for lowering BP, so the UK and New Zealand in
their current 2012 guidelines suggest ACEI (or ARB) plus CCB
plus diuretic as a target triple therapy. ACEIs and ARBs are par-
ticularly effective in younger patients who have high renin levels.
Individuals of Afro-Caribbean origin tend to exhibit low-renin
hypertension and may respond less well to ACEIs or ARBs as
initial monotherapy. However, either an ACEI or an ARB (but
very rarely both together) is important in combination therapy
for all individuals with hypertension requiring multiple drugs.
The very large ONTARGET trial of ramipril, telmisartan or the

Lifestyle modifications

Not at goal blood pressure (<140/90 mm Hg)

(<130/80 mm Hg with diabetes or chronic kidney disease)

Initial drug choices

Without compelling With compelling

indications indications

Stage 1 hypertension Stage 2 hypertension Drug(s) for the compelling

(SBP 140-159 mm Hg or DBP 90-99 mm Hg)
Thiazide-type diuretics for most
May consider ACEI, ARB, BB, CCB,
or combination
(SBP 160 mm Hg or DBP 100 mm Hg) indications
2-Drug combination for most Other antihypertensive drugs
(usually thiazide-type diuretic and (diuretics, ACEI, ARB, BB, or CCB)
ACEI, ARB, BB, or CCB) as needed

Not at goal
blood pressure

Optimize dosages or give additional drugs

until goal blood pressure is achieved
Consider consultation with hypertension specialist

Figure 60.8. Algorithm for treatment of hypertension. “Compelling indications” are listed in Table 60.4. ACEI indicates angiotensin-converting
enzyme inhibitor; ARB, angiotensin receptor blocker; BB, β-blocker; CCB, dihydropyridine calcium channel blocker; DBP, diastolic blood pressure;
SBP, systolic blood pressure. (Previously published. See “Credit Lines” section.)
584 VI Coronary Artery Disease Risk Factors

Blood pressure
difference (mm Hg) Coronary heart disease events

Systolic Diastolic No. of No. of Relative risk Relative risk

trials events (95% CI) (95% CI)

Thiazides v any other -1.4 0.2 15 4,229 0.99 (0.91 to 1.08)

ß-blockers v any other 1.4 0.6 10 2,182 1.04 (0.92 to 1.17)
Angiotensin-converting 0.9 0.4 21 6,026 0.97 (0.92 to 1.17)
enzyme inhibitors v
any other
Angiotensin receptor -0.4 0.1 10 2,744 1.04 (0.94 to 1.16)
blockers v any other
Calcium channel -0.4 -0.9 21 6,288 1.00 (0.91 to 1.10)
blockers v any other
0.7 1 1.4
Specified
drug better Specified
drug worse

Figure 60.9. Data From Meta-analysis of 464,000 Subjects in Hypertension Trials. (Previously published. See “Credit Lines” section.)

combination, showed no overall additional benefit and indeed
some harm from an ACEI plus ARB combination.
It remains to be seen whether JNC 8 will recommend specific
combinations.
Antihypertensive Drug Therapy
Most individual antihypertensive drugs lower blood pressure to
a similar degree when used as monotherapy, by around 7%–8%,
although there is wide individual variation. When a drug is
added as second or third line, the reduction in BP is usually less
marked. Two valuable exceptions are spironolactone and minoxi-
dil (see below).
Angiotensin-Converting Enzyme Inhibitors
These drugs act by blocking the conversion of angiotensin I to
angiotensin II, and are thus vasodilators. They potentiate bra-
dykinin by the same pathway. By their intrarenal action ACEIs
lower intraglomerular pressure. They are well tolerated but cause
cough, related to the bradykinin, in about 18% of patients, and
anaphylactic angioedema in about 1/1,000,000. There is abundant
evidence of their protective value, particularly for patients with
coronary heart disease, heart failure, diabetes, or prior stroke.
Renal function must be checked after initiation, and particu-
larly if there is concomitant renal illness. They are often valuable
for patients with mild-moderate renal impairment but their use
requires vigilance to guard against hyperkalemia.
Angiotensin Receptor Blockers
ARBs prevent angiotensin II acting on AT1 receptors and are
similar to ACEIs, but have no effect on bradykinin and do not
cause cough or bradykinin-related anaphylaxis. They are the best
tolerated of all antihypertensive drugs. ARBs have similar com-
pelling indications to ACEIs, except following MI, when ACEIs
are preferred. Valsartan is approved as an alternative to an ACEI
for MI patients.
principally by reducing renin production in the kidney, and so
are alternatives to ACEIs and ARBs. They are first line therapy
for patients with myocardial ischemia, previous MI, and (in the
case of carvedilol and bisoprolol) heart failure. β-Blockers may
cause a slight elevation in blood glucose, although whether this is
clinically important is debated.
Calcium Channel Blockers
CCBs are widely used, and are very effective blood pressure
lowering agents. They have a very large database of supportive
evidence and are particularly protective against stroke. CCBs act
by relaxing smooth muscle in arterioles but cause dose-related
troublesome edema in about 15% of patients by their effect on
capillary hydrostatic pressure. ARBs and ACEIs can reverse this
side effect to some extent. Most CCBs are also approved for use
in angina.
Short-acting dihydropyridine CCBs (eg, nifedipine, felo-
dipine) may provoke a reflex tachycardia and should always
be used in their long-acting modified release preparations.
Amlodipine is a longer acting drug and has a fairly neutral effect
on heart rate.
Non-dihydropyridine CCBs (verapamil and diltiazem) lower
heart rate as well as BP and can be of particular value in angina,
if β-blockers are contraindicated.
Diuretics
Thiazide, and particularly thiazide-like (chlortalidone and inda-
pamide), diuretics have the largest evidence base for CV protec-
tion of all antihypertensive drugs. They act by promoting sodium
excretion and may lower potassium by the increased delivery of
sodium for exchange in the distal tubule. Thiazides cause an ele-
vation in uric acid and may cause gout. Like β-blockers they also
cause a small rise in blood glucose but the pathological effect of
this is not certain. Diuretics are usually drugs of first choice for
older patients with systolic hypertension. The best data support a
thiazide-ACEI combination for this condition.

β-Adrenergic Blockers Other Drugs

β-Blockers are also valuable antihypertensive drugs, but cause α-Adrenergic blockers, principally doxazosin, are potent
more side effects than newer agents. In hypertension they act BP-lowering drugs with effects and side effects like CCBs.
 60 Hypertension
However, they do not have a large evidence base for CV pro-
tection. Indeed, the doxazosin group in the original ALLHAT
cohort was discontinued because of an excess of heart failure
symptoms, although in the European ASCOT trial post-hoc anal-
ysis suggested a safer profile for doxazosin.
Direct renin inhibitors are a new class of drugs. Aliskerin
is the only available agent from this new class. It blocks renin
release from the kidney, hence reducing angiotensin II. However,
in December 2011 the pivotal ALTITUDE outcome study of
aliskerin added to conventional therapy was discontinued pre-
maturely because of an excess of adverse events, so the benefit of
these new drugs remains unproven.
Amiloride and Spironolactone
Amiloride acts in the distal tubule to reduce sodium absorption
with a potassium-sparing effect. It has a valuable additional effect
on BP when added as fourth line therapy. Spironolactone, which
inhibits aldosterone action at its receptor, is even more potent, par-
ticularly beneficial in heart failure, but requires regular monitoring
of potassium and should not be used in renal failure. Spironolactone
may cause gynecomastia, and eplerenone may be used as a weaker
aldosterone inhibitor, but with less marked side effects.
Older Drugs
Methyldopa is a centrally acting alpha-adrenergic agonist
and is used mostly in pregnancy where it is safe for the fetus.
It may cause fatigue, dry mouth and depression in the mother.
Hydralazine is another vasodilator but causes a marked reflex
tachycardia so it must be used in combination with a β-blocker.
Minoxidil has a direct action on smooth muscle and is a pow-
erful antihypertensive. It is useful in resistant hypertension but
also causes reflex tachycardia. It must be used with concomi-
tant β-blocker and diuretic therapy, and interestingly its use is
re-emerging in recent years. Minoxidil has a unique side effect
in that it causes marked hair growth.
Table 60.3 presents the compelling indications and the drugs
of choice for each. For each compelling indication, the goal of
treatment is to prevent progression of the disease or syndrome
by reducing the contribution of hypertension. Although these
recommendations are appropriate across populations, treatment
must be individualized. Duration of action, side effects, and drug
interactions all bear consideration in the choice of antihyper-
tensive medication. Cost is now less of a concern since generic
agents from all classes of antihypertensive drug are available.
Table 60.3. Drug Choices for Compelling Indications in the
Treatment of Hypertension
Compelling Indication Diuretic BB ACEI ARB CCB Aldo ant
Heart failure + + + +
After MI + + +a
High-risk CAD or TIA + + +
Diabetes + +
CRI + +
Systolic hypertension + + + Table 60.4. Causes of Resistant Hypertension
Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; Aldo ant,
aldosterone antagonist; ARB, angiotensin receptor blocker; BB, β-blocker;
CAD, coronary artery disease; CCB, dihydropyridine calcium channel blocker;
CRI, chronic renal impairment; MI, myocardial infarction; TIA, transient
ischemic attack.
a
If left ventricular impairment present.
Previously published. See “Credit Lines” section.
585
The J-Shaped Curve for Hypertension Treatment
Hypertension specialists have debated for years about the risks
of lowering BP too much and where CV risks increase with very
low BP. In practice, low BP is most often seen with combination
therapy in heart failure and less often, angina patients. Light-
headedness, dizziness and collapse are common in patients with
postural hypotension. Checking supine and erect BP is a straight-
forward way of detecting BP which is overtreated and too low.
ABPM is also valuable.
Post-hoc analysis of achieved BP in lipid-lowering and acute
coronary syndrome trials has shown an increase in CV events
with diastolic BP < 70 mm Hg, but there may be many confound-
ing influences and in 2012 there is as yet no prospective data
on the existence of the J-shaped curve in hypertensive patients
without overt heart disease.
Resistant Hypertension
In a segment of any hypertensive population, blood pressure can-
not be controlled with what appears to be adequate therapy. The
JNC 7 defines adequate therapy as the use of three drugs, includ-
ing a diuretic, in recommended doses. Table 60.4 lists the com-
mon conditions that may contribute to resistant hypertension.
In the authors’ experience, poor medication compliance is fre-
quently to blame, and true resistant hypertension is quite uncom-
mon. Observing the individual taking their usual medication,
with BP recordings over the next 4 to 6 hours, is an inexpensive
and rapid way of assessing whether true resistant hypertension
exists.
Pseudoresistance is the apparent increase in blood pressure
brought about by noncompliant vessels, as often occurs in the
elderly population. A one-time correlation between an intra-
arterial blood pressure and a cuff measurement can serve as a
valuable guide to the strategy of a treatment program.
Poor adherence, or compliance, to therapy has several causes.
Cost of medications (especially among elderly patients receiv-
ing a fixed, small income), misunderstanding of directions for
use, and insufficient education are some of the reasons why
patients fail to comply with instructions as is the fear of drug
side effects.
One scientific approach to resistant hypertension is as follows
(Figure 60.10). Check for renal artery stenosis with MRI angiog-
raphy, and carry out the usual tests for secondary hypertension.
Then, with a patient established on three drugs, usually ARB
plus CCB plus diuretic, salt restriction and NSAID avoidance
confirmed, measure plasma renin. If renin is high, add a drug to
reduce renin-angiotensin activity, ie, a β-blocker or an ACEI. If
renin is low, add a second diuretic, eg, amiloride or spironolac-
tone. With intermediate renin concentrations add doxazosin with
or without a second diuretic. In every case, excessive sodium
and fluid retention must be considered, and adequate diuresis
achieved. If these measures are insufficient, consider adding
minoxidil or a long-acting isosorbide preparation as next add-on
therapy. Within the last year, RDN has now become an option.
Pseudoresistance
Nonadherence to therapy
Fluid volume overload
Drug-related causes
Sleep apnea
Previously published. See “Credit Lines” section.
586 VI Coronary Artery Disease Risk Factors
A+C+D
Measure plasma renin
High Low
(>100 mU/I) Normal (<20 mU/I)
β (or 2nd A) +α ΔD
Figure 60.10. Algorithm for Resistant Hypertension. A indicates
angiotensin receptor blocker; C, calcium channel blocker; D, diuretic.
Percutaneous RDN
In 2010 the first results of the safety and efficacy of percutaneous
RDN were presented at the American Heart Association meeting
and published in The Lancet. This procedure uses RFA, similar
to cardiac electrophyhsiological RFA for accessory pathways and
atrial fibrillation. A spiral series of burns is performed in the
proximal renal arteries which disrupt the sympathetic nerves to
each kidney, with a marked fall in BP (Figure 60.11). Indeed,
in Esler’s landmark trial the average fall was 33/11 mm Hg in
patients with severe, resistant hypertension, the benefit sustained
over 1 year. There was no demonstrable renal damage and the
procedure appears safe.
However, the effect may not be permanent, and the renal nerves
probably grow back over 5 years or so. So far unpublished case
reports suggest a loss of the marked early benefit after 3 to 4 years.
Our own early experience is that some patients have a profound
benefit from RDN, while others have very little improvement.
Older individuals with stiff, calcified vessels seem to respond less
well. However, RDN is a new and exciting development in hyper-
tension and its place remains to be refined and determined.
Figure 60.11. Percutaneous Renal Denervation.
Carotid Sinus Stimulation
It has been known for 60 years that carotid sinus stimulation low-
ers BP. In 2011 the first results of the Rheos Pivotal study were
released. A subcutaneous pulse generator similar to a permanent
pacemaker device is connected to bilateral electrical leads which
stimulate both carotid sinus baroreceptors. The brain interprets
these signals as a rise in BP and causes changes in heart, renal
and vascular function to lower BP. The surgery is not without
risk, and the effects on BP were variable in the trial. While this
procedure may become valuable, it is too early to say whether
use of this technology will become widespread.
Sleep Apnea and Hypertension
Breathing-related sleep disorders are associated with obesity and
hypertension and may be responsible for failure of drug ther-
apy to successfully control BP. Nocturnal oxygen desaturation
leads to a rise in plasma catecholamine levels that lasts many
hours out from the period of sleep apnea itself. Patients often
do not volunteer a history of fatigue or hypersomnolence, so the
clinician must ask about daytime fatigue, excessive snoring, or
observed interruption of breathing by the patient’s sleeping part-
ner. Screening for nocturnal oxygen desaturation using home
overnight oximetry is a valuable initial step in patients with sus-
pected sleep apnea, but sleep center polysomnography may be
needed to confirm the diagnosis and evaluate the effect of con-
tinuous positive airway pressure.
Secondary Hypertension
More than 90% of the hypertensive population has essential or
primary hypertension (ie, hypertension for which a single cause
cannot be identified). A few individuals, usually younger patients
with structural abnormalities of the renal arteries or aorta, or
endocrine conditions, have an identifiable cause. Atheromatous
 60 Hypertension 587

Table 60.5. Clues to Secondary Hypertension
Recent onset of hypertension
Loss of blood pressure control
Resistant hypertension
Evidence of peripheral vascular disease (increased serum creatinine
concentration with angiotensin blockade)
Unprovoked hypokalemia (inappropriately low-normal serum potassium
concentration)
renal artery stenosis in older patients is fairly common among
individuals with more widespread vascular disease.
Table 60.5 lists clues to the existence of secondary hyperten-
sion. Table 60.6 lists the common causes of secondary hyperten-
sion. If appropriate treatment is instituted for secondary causes
of hypertension there is often a marked improvement or cure of
the hypertension specifically in 1) renovascular hypertension,
2) mineralocorticoid excess states (eg, primary aldosteronism),
and 3) catecholamine-secreting neoplasms.
Renovascular Hypertension
Hypertension at a young age, poor BP control, evidence of renal
dysfunction, or worsening renal function with ACEIs and ARBs
all suggest the possibility of RAS. Likewise, the presence of an
abdominal bruit or coexistent peripheral vascular disease is an
example of physical findings that warrant a search for renal
artery stenosis. Patients with severe bilateral stenoses may pres-
ent with flash pulmonary edema.
In children and younger adults fibromuscular dysplasia of the
renal arteries causes sequential stenoses with marked hyperten-
sion. In older adults atheroma is the usual cause. Indeed, one
series of patients with peripheral vascular disease identified RAS
in 24% of patients.
MRI or CT angiography is the gold standard imaging modal-
ity for detecting RAS. Stenoses >70% are considered likely to
cause a reduction in renal blood flow and renovascular hyper-
tension. Currently in 2012, there is no definitive trial that has
established a clear benefit for renal artery angioplasty plus stent
over medical therapy alone. The ongoing NIH-funded CORAL
trial is addressing this deficiency in the literature. The European
STAR trial randomized patients with modest 50%–70% stenoses
and failed to show any conclusive benefit over medical therapy
alone.
nodules can result in abnormal aldosterone secretion and blood
pressure elevation. The most common form is a diffuse hyperac-
tivity of zona glomerulosa cells, rather akin to the thyroid gland
in thyrotoxicosis.
Unprovoked hypokalemia, hypokalemia with minimal diure-
tic therapy, and low-normal serum potassium concentrations
seen in conjunction with hypertension drugs usually associated
with elevated potassium concentrations (eg, ACEIs, ARBs, and
potassium-sparing diuretics) signal the possibility of primary
aldosteronism. The association of hypokalemia, evidence of vol-
ume expansion (upper-normal or elevated serum sodium concen-
trations), low plasma renin and raised aldosterone levels are the
hallmarks of the syndrome.
To determine which pathologic process is responsible for the
syndrome, MRI or CT imaging of the adrenals offers the best
diagnostic opportunity (Figure 60.12). In occasional circum-
stances, adrenal vein catheterization for measurement of the
aldosterone concentration may be necessary. Medical treatment
is with the use of the specific aldosterone antagonist spironolac-
tone, which often results in a marked improvement in blood pres-
sure. Tumors should be removed by adrenalectomy.
Pheochromocytoma
Much less common than either renovascular hypertension or
primary aldosteronism are neoplasms of the sympathetic ner-
vous system. The most common of these arise from the adrenal
medulla and bear the histologic diagnosis of pheochromocytoma
(Figure 60. 13). Classically, 10% are multiple, 10% exist out-
side the adrenals and 10% are malignant (the ten percent tumor).
Pathologists use the term paraganglioma to label those occur-
ring at other sites along the sympathetic chain. Clinically and
biochemically there is little to distinguish the two types except
that pheochromocytomas secrete epinephrine. Quantitation of
catecholamine production over a 24-hour period (24-hour urine
collection) has been the diagnostic test of choice for the past

Primary Aldosteronism
Aldosterone is the major mineralocorticoid of the adrenal cortex.
Autonomous secretion by the zona glomerulosa, a single neo-
plasm of the adrenal cortex (adenoma vs carcinoma), or multiple

Table 60.6. Causes of Secondary Hypertension

Sleep apnea
Drug-induced or drug-related hypertension
Chronic kidney disease
Primary aldosteronism
Renovascular disease
Chronic steroid therapy and Cushing syndrome
Pheochromocytoma
Coarctation of the aorta
Thyroid or parathyroid disease
Figure 60.12. Computed tomographic image showing aldosterone-
Previously published. See “Credit Lines” section. producing adenoma (arrow).
588 VI Coronary Artery Disease Risk Factors
Figure 60.13. Computed tomographic image showing adrenal pheo-
chromocytoma (arrow).
several decades. Elevations of plasma metanephrine concentra-
tions are also valuable in diagnosis.
After biochemical testing establishes the diagnosis of pheo-
chromocytoma or paraganglioma, the goal is to anatomically
localize the abnormal tissue. Because adrenal pheochromo-
cytomas are more common, CT of the adrenal glands should
be the first choice. In the absence of an unequivocal CT diag-
nosis, MR imaging of the abdomen and thorax is the study of
choice for examination of the sympathetic chain (Figure 60.14).
Metaiodobenzylguanidine (MIBG) is a radioiodinated com-
pound that can be incorporat ed into the catecholamine synthetic
pathway to identify overproduction. On occasion, somatostatin
receptors on the tumor may make it possible to use the imaging
agent octreotide to localize the pathologic tissue. Radionuclide
imaging offers little over MRI and is less practical to use except
in individuals in whom CT or MRI is not feasible. Treatment is
Figure 60.14. Magnetic resonance image showing mediastinal paraganglioma.
surgical, strictly in experienced centers only, with powerful peri-
operative α- and β-adrenergic blockade, to minimize the effects
of a massive pressor surge frequently seen at operation.
Hypertension in Special Populations
Chronic Renal Disease
Hypertension is an almost universal feature of the chronic renal
diseases. Hypertension in polycystic disease presents early;
in other conditions hypertension tends to mirror the progres-
sion of disease, and can fade in the final stages of renal failure.
Hypertension in itself is also a major cause of renal failure world-
wide. Data from the United States Renal Data System indicates
hypertension as the principal cause in about 30% of cases of
chronic renal impairment in the US.
Patients with renal impairment develop sodium and volume
overload, neuroendocrine activation and sympathetic nervous
system activation. Exposure of the glomeruli to sustained high
pressure damages the tissues and causes glomerulosclerosis with
hyperfiltration followed by loss of nephrons.
Specific investigation with autoimmune serology, quantifica-
tion of proteinuria, renal scintigraphy to determine differential
renal function, and MRI angiography are all required.
Treatment of Hypertension in Renal Disease
Combination therapy is usually required for hypertension in
renal disease. ACEIs reduce intraglomerular pressure by dilata-
tion of efferent arterioles, as well as lowering BP by their usual
actions, and are drugs of first choice. ARBs have a similar effect.
The combination of both ACEI and ARB remains controversial.
Diuretics are usually needed because of sodium and fluid over-
load. In more severe renal impairment thiazides become ineffec-
tive and loop diuretics are required. β-Blockers decrease renin
release and are often added. Doxazosin is a favorite choice of
many renal physicians. CCBs may preferentially dilate afferent
arterioles and are less often used if other agents are available.
Minoxidil may have a marked beneficial effect in more severe
renal impairment.
 60 Hypertension
Pregnancy
The National High Blood Pressure Education Program working
group classification for hypertension during pregnancy has four
categories: 1) preeclamsia-eclampsia, 2) preeclampsia super-
imposed on chronic hypertension, 3) chronic hypertension, and
4) gestational hypertension.
Preeclampsia is a multiorgan disease affecting not only the
vasculature but also renal, hepatic, pulmonary, and hematologic
function as manifested by proteinuria, hepatocellular injury, pul-
monary edema, and microangiopathic-hemolytic anemia usually
occurring late in pregnancy. This type of hypertension can prog-
ress to convulsions (eclampsia) and is a major risk for mater-
nal and fetal mortality. Delivery is the definitive treatment for
preeclampsia. Pending delivery, magnesium sulfate is the phar-
macologic treatment of choice.
Gestational hypertension refers to mild blood pressure
elevations that occur in the second half of pregnancy and are
unaccompanied by proteinuria. Blood pressures usually return to
normal within 12 weeks after delivery.
Given the limited time course of pregnancy, mild elevations
in blood pressure can be tolerated and initiation of pharmaco-
logic treatment is reserved for pressures that are consistently
higher than 150 mm Hg systolic, or 90–100 mm Hg diastolic.
Methyldopa is the drug of choice in the US for hypertension
during pregnancy. It is safe for mother and fetus but causes
drowsiness and lethargy. Nifedipine is unapproved for hyper-
tension in pregnancy in the US and has a pregnancy category
C rating because of potential problems seen in animal stud-
ies; it is used widely around the world for pregnancy related
hypertension. Labetalol is safe and well tolerated and is the
β-blocker of choice. ACEIs and ARBs are teratogenic and are
contra-indicated. Diuretics are usually avoided because of the-
oretical risks of exacerbating pre-eclampsia. In severe eclamp-
sia loop diuretics are often required during or after caesarian
section.
Postpartum, ACEIs, diuretics, and other drugs may be used
and ACEIs are the drugs of choice, provided the mother is not
breastfeeding.
Diabetes
There is an “unholy trinity” of diabetes—elevated blood pres-
sure, blood lipids, and blood glucose. Recent large trials have
emphasized that powerful BP lowering is probably the most cru-
cial of the three variables. Aggressive glucose lowering does not
yield the hoped-for benefits. Reducing cholesterol to low levels
offers marked protection from major CV events.
Coincident end-organ damage from diabetes and hyperten-
sion has prompted recommendations for more vigorous treat-
ment in diabetic hypertensive patients. The blood pressure goal
is <130/80 mm Hg.
In most instances, two or more drugs are necessary to achieve
the blood pressure goal. ARBs and ACEIs are the drugs of first
choice, lowering intraglomerular pressure as well as BP, slowing
progression of chronic renal disease and delaying the need for
dialysis or transplantation.
CCBs are well tolerated. The propensity for thiazide diuretics
to worsen glycemic control warrants note, but in small doses this
is not a common problem. β-Blockers may mask the CV symp-
toms of hypoglycemia in patients prone to this diabetic treatment
complication, limiting their use in this population.
589
Hypertensive Urgencies and Emergencies
Fulminant hypertension with pulmonary edema and visual and
neurological symptoms is rare but a true medical emergency.
Attempts at rapid reduction are attractive, but sudden and rapid
declines in blood pressure may offer a stroke risk at least as great
as extreme elevations.
Oral administration is the best approach for any blood pressure
level that is unaccompanied by severe symptoms. Theoretically
blockade of both α and β adrenoceptors is safest, certainly in
suspected cases of pheochromocytoma, but modest doses of
atenolol 25 mg as a starting dose to begin BP lowering, with fur-
ther doses later and additional agents the next day is often highly
effective without a precipitate drop in blood pressure.
Hypertensive emergencies, with acute CV or neurologic fea-
tures, warrant the administration of parenteral agents of rapid
onset and offset action. Careful monitoring of organ function
and achieving a safe blood pressure level should be the thera-
peutic goal. Intravenous nitroglycerin, familiar to all practicing
cardiologists, is safest, beginning at 1 mg/hour and titrating to
achieve a safer BP. A β-blocker is often important to reduce sym-
pathetic drive, but may be unwise early in therapy if heart failure
is present.
Seldom is it necessary to normalize blood pressure in emer-
gency circumstances. Indeed, during acute cerebrovascular
events, elevated blood pressure may be a result rather than a
cause of cerebral ischemia, and sudden lowering of the blood
pressure may worsen the cerebral damage.
Conclusions
Hypertension is the most important CV risk factor in global
terms. Good treatment will substantially protect against stroke,
renal disease and hypertensive heart failure, and offer impor-
tant protection from MI. Detection is straightforward and cur-
rent treatment follows scientific principles. Combination therapy,
lifestyle change and achievement of more aggressive targets are
the cornerstones of modern management.
Suggested Reading
ALLHAT Officers and Coordinators for the ALLHAT Collaborative
Research Group. The Antihypertensive and Lipid-Lowering Treatment
to Prevent Heart Attack Trial. Major outcomes in high-risk hyperten-
sive patients randomized to angiotensin-converting enzyme inhibi-
tor or calcium channel blocker vs diuretic: The Antihypertensive and
Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT).
JAMA. 2002 Dec 18;288(23):2981–97. Erratum in: JAMA 2003 Jan
8;289(2):178. JAMA. 2004 May 12;291(18):2196.
Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo
JL Jr, et al; National Heart, Lung, and Blood Institute Joint National
Committee on Prevention, Detection, Evaluation, and Treatment
of High Blood Pressure; National High Blood Pressure Education
Program Coordinating Committee. The Seventh Report of the Joint
National Committee on Prevention, Detection, Evaluation, and
Treatment of High Blood Pressure: the JNC 7 report. JAMA. 2003
May 21;289(19):2560–72. Epub 2003 May 14. Erratum in: JAMA.
2003 Jul 9;290(2):197.
Garovic V, Textor SC. Renovascular hypertension: current concepts.
Semin Nephrol. 2005 Jul;25(4):261–71.
Abbreviations
ABPM ambulatory blood pressure monitoring
ACEI angiotensin converting enzyme inhibitor
590 VI Coronary Artery Disease Risk Factors

ARB angiotensin receptor blocker RFA radiofrequency ablation

BP blood pressure WHO World Health Organization
CCB calcium channel blocker
CT computed tomography
CV cardiovascular
ECG electrocardiography
Names of Clinical Trials
HBPM home blood pressure monitoring ALLHAT Antihypertensive and Lipid-Lowering Treatment to
JNC 7 [6, 8] The Joint National Committee on Prevention, Detec- Prevent Heart Attack Trial
tion, Evaluation, and Treatment of High Blood Pres- ALTITUDE Aliskiren Trial in Type 2 Diabetes Using Cardio-
sure, Seventh Report [Sixth Report, Eighth Report] vascular and Renal Disease Endpoints
LVH left ventricular hypertrophy ASCOT Anglo-Scandinavian Cardiac Outcome Trial
MI myocardial infarction CORAL Cardiovascular Outcomes in Renal Atherosclerotic
MRI magnetic resonance imaging Lesions
NHANES National Health and Nutrition Examination Survey ONTARGET Ongoing Telmisartan Alone and in Combination with
NSAID nonsteroidal anti-inflammatory drug Ramipril Global Endpoint Trial
RAS renal artery stenosis PAMELA Pressioni Arteriose Monitorate e Loro Associazioni
RDN renal denervation STAR Stimulation Therapy for Apnea Reduction

Stroke and Cardiovascular Disease
JOSEPH G. MURPHY, MD, and MARGARET A. LLOYD, MD
Stroke is the third leading cause of death in western countries
and is the reported cause in 1 of every 18 deaths annually in the
United States. There are over 6 million stroke survivors in the
US, of whom about 20% require long-term institutional care.
• 800,000 people in the United States have a stroke each year of
whom about 140,000 will die of stroke.
• Aggressive risk factor modification can prevent about 80% of first
strokes.
• Black men have a 60% increased risk of stroke compared with
white men.
• Black women have a 40% increased risk of stroke compared with
white women.
Physiology of Autoregulation
of Cerebral Blood Flow
The brain is exquisitely sensitive to changes in cerebral blood
flow but much less sensitive to changes in brain perfusion blood
pressure because of a mechanism of cerebral blood flow auto-
regulation that, under physiological circumstances, maintains
cerebral blood flow at a nearly constant level. This autoregulation
mechanism is controlled by changes in cerebral vascular resis-
tance mediated through minute-to-minute changes in cerebral
vascular smooth muscle tone that in turn is altered by vascular
smooth muscle relaxation or contraction. Cerebral blood flow
autoregulation works well in the face of physiological changes
in brain perfusion pressure over a wide range of BPs (between
systolic BP of 85 up to 200 mm Hg), but fails either when systolic
pressure falls much below 85 mm Hg, leading to cerebral hypo-
perfusion and brain ischemia, or at a BP above 200 mm Hg when
Abbreviations and acronyms are expanded at the end of this chapter.
61
cerebral hyperperfusion and brain edema develop. Cessation of
cerebral perfusion for as short as 3 to 5 seconds, which frequently
occurs during periods of cardiac asystole, can lead to rapid loss
of consciousness.
Hypertensive Encephalopathy
Malignant hypertension is severe systemic hypertension associ-
ated with retinal hemorrhages, exudates, or papilledema, while
the closely related hypertensive encephalopathy is systemic
hypertension that overwhelms cerebral blood flow autoregulation
and manifests with clinical signs of cerebral edema. Hypertensive
encephalopathy is typically caused by sudden and steep rises in
systemic BP. Severe hypertension damages the cerebral arterio-
lar and capillary vascular endothelial wall, which loses vascular
integrity and allows plasma fibrinoid material to enter the vas-
cular wall, leading to fibrinoid necrosis of the cerebral arterioles
and capillaries.
The onset of hypertensive encephalopathy is variable and does
not occur at the same BP in all individuals. In previously normo-
tensive individuals, hypertensive encephalopathy can occur at
diastolic BPs as low as 100 mm Hg, particularly in patients who
develop rapid-onset severe acute hypertension, usually in the set-
ting of acute renal failure or pregnancy-associated hypertension
(preeclampsia). In contrast, patients with chronic hypertension
are relatively protected against hypertensive encephalopathy
because chronic arteriolar hypertrophy reduces the transmission
of BP to the capillary circulation with the result that diastolic
pressures up to 120 mm Hg may be well tolerated.
The emergency treatment of hypertensive crises is to lower
the diastolic pressure to about 105 to 100 mm Hg within 2 to 4
hours. The BP should not be lowered acutely more than 50% of
the difference between presentation and goal BP or more than
591
592 VI Coronary Artery Disease Risk Factors
25% of the initial elevated BP value. Excessive lowering of BP
may be counterproductive and lead to cerebral ischemia or cere-
bral infarction by lowering the BP below the patient’s autoregu-
latory range; typically it resets to a higher level in a chronically
hypertensive patient.
Hypertension and Stroke
Hypertension is the single most important, modifiable risk factor
for stroke. Treatment of hypertension to goal BP is the most cost
effective strategy for the preventing of both ischemic and hemor-
rhagic stroke.
• Systolic BP should be treated to a goal of <140 mm Hg and dia-
stolic BP to <90 mm Hg because these levels are associated with
a lower risk of stroke and cardiovascular events (Class I; Level of
Evidence A).
• In patients with hypertension with diabetes or renal disease, the BP
goal is <130/80 mm Hg (Class I; Level of Evidence A).
Stroke and TIA
A stroke is an abrupt loss of cerebral function due to an acute dis-
ruption in the blood supply to the brain. Strokes can be broadly
classified into five major types: hemorrhagic (about 20% of all
strokes), divided anatomically into intracerebral (10%) or suba-
rachnoid hemorrhage (10%); ischemic strokes (80%) divided
etiologically into thrombotic stroke, embolic stroke, or stroke
due to systemic hypoperfusion. The affected area of the brain
may permanently lose function when there is cerebral infarction,
often manifested as clinical neurological signs and symptoms,
or the infarction may be clinically silent and evident only on
neuroimaging.
Lacunar infarcts are small noncortical cerebral infarcts caused
by occlusion of penetrating end arteries that arise at acute angles
from the larger cerebral arteries. Lacunar infarcts typically occur
in association with hypertension, diabetes, and specific genetic
markers (eg, APOE e4 allele and ACE DD genotype).
A TIA is a brief episode of neurologic dysfunction resulting
from focal temporary cerebral ischemia not associated with a
neurological deficit or end-organ injury (cerebral infarction) as
visualized by neuroimaging studies.
• The old definition of a TIA, neurological dysfunction lasting less
than 24 hours, has now been discarded.
Systemic hypoperfusion is a cause of global cerebral ischemia
often associated with circulatory failure, cardiac arrest, or major
arrhythmias. Symptoms and signs of brain dysfunction are bilat-
eral and nonfocal, and range from agitation to impairment of
consciousness to deep coma.
Cardiac Arrest and Hypoxic-Ischemic
Brain Injury
Hypoxic-ischemic brain injury is often present in patients resus-
citated from cardiac arrest and is the commonest cause of death
after resuscitation from cardiac arrest, especially out-of-hospital
cardiac arrest. Induced hypothermia, in which brain temperature
is therapeutically lowered to 32°C to 34°C by systemic cooling
using externally applied surface cooling blankets, during the first
few hours after cardiac arrest reduces the risk of subsequent neu-
rologic injury; however, for it to be effective it must be started
less than 6 hours after the return of spontaneous circulation and
must be maintained for 12 to 24 hours.
Mild induced therapeutic hypothermia has proved to decrease
mortality and improve neurologic outcome in patients with out-
of-hospital cardiac arrest who present with either ventricular
fibrillation or pulseless ventricular tachycardia and have had
successful restoration of cardiac rhythm. It is unclear whether
induced hypothermia provides a similar benefit to patients who
present with asystole and pulseless electrical activity.
• Absence of somatosensory evoked potentials, which are measures
of the electrical responses seen in the central nervous system fol-
lowing peripheral somatosensory stimulation (eg, median nerve
electrical stimulation at the wrist), is predictive of poorer progno-
sis in hypoxic-ischemic brain injury.
• A markedly elevated level of serum NSE is predictive of poorer
prognosis in hypoxic-ischemic brain injury.
• Induced hypothermia makes the predictive value of somatosensory
evoked potentials and serum NSE less reliable in patients with
hypoxic-ischemic brain injury, especially in the first 24–48 hours
of therapy.
Cardioembolic strokes are an important source of cerebral
embolism. They typically have an abrupt onset with focal neu-
rological deficit that generally resolves, at least in part, over the
following 24 to 72 hours due to spontaneous lysis of the embolic
thrombus. Cardioembolic strokes will generally recur unless the
underlying cardiac condition is treated (Table 61.1).
Stroke and Atrial Fibrillation
Atrial fibrillation is associated with ischemic stroke through
its association with blood flow stasis in the left atrium and left
atrial appendage that leads to thrombus formation in a dilated
and noncontractile left atrium. Stroke risk is similar in patients
with chronic atrial fibrillation and those with paroxysmal atrial
fibrillation. Cerebral thromboembolism from left atrial thrombus
causes larger strokes than the typically smaller predominantly
platelet thrombi that embolize from carotid artery plaques. Atrial
fibrillation is a marker of a worse prognosis in patients who suffer
an ischemic stroke, with a higher mortality and greater residual
neurological deficit than ischemic stroke occurring in patients in
sinus rhythm.
• Stroke may be the initial clinical presenting feature of atrial
fibrillation.
• Atrial fibrillation is associated with about a 5-fold increased risk of
ischemic stroke due to cardioembolism, even when not associated
with valvular heart disease.
Table 61.1. Cardiac Conditions Strongly Linked to Card-
ioembolic Strokes
Atrial fibrillation (both paroxysmal and sustained)
Atrial flutter
Rheumatic mitral valve disease
Mechanical heart valves
Atrial or ventricular thrombi
Myocardial infarction with a low ejection fraction <30%
Heart failure with ejection fraction <30%
Infective endocarditis
Nonbacterial endocarditis
Libman-Sacks endocarditis associated with systemic lupus
Antiphospholipid syndrome
Marantic endocarditis often associated with malignancy
Cardiac tumors
Papillary fibroelastoma
Left atrial myxoma
 61 Stroke and Cardiovascular Disease 593

• Atrial fibrillation accounts for about 50,000 cardioembolic isch-
emic strokes in the US each year among the 2 million patients who
have either chronic or paroxysmal atrial fibrillation.
Warfarin anticoagulation with a target INR of 2 to 3 sig-
nificantly reduces the risk of stroke by about 64% in patients
with atrial fibrillation, while aspirin reduces the risk by 19%.
Dabigatran, a direct thrombin inhibitor, was superior to warfarin
when administered at a dose of 150 mg and equal to warfarin at
a dose of 110 mg for preventing ischemic stroke. Bleeding com-
plications were similar with high-dose dabigatran and warfarin
but much less with low-dose dabigatran. Choice of anticoagulant
should be individualized in light of concerns regarding higher
gastrointestinal bleeding rates in older patients and patients with
poor renal function. Other concerns with dabigatran include
lack of reversibility, greater cost, and a small increased risk of
myocardial infarction. Patients on anticoagulation therapy who
experience an ischemic stroke have a lower mortality and less
neurological deficit compared with stroke in atrial fibrillation
patients who are not on anticoagulation therapy (Tables 61.2
and 61.3).
Cryptogenic Stroke
Cryptogenic stroke is cerebral infarction that is not attributable
to definite cardioembolism, large artery atherosclerosis, or small
artery disease in spite of a comprehensive clinical, imaging, and
laboratory evaluation.
The causes of cryptogenic stroke include occult cardioembo-
lism secondary to paroxysmal atrial fibrillation or paradoxical
embolism through an atrial septal defect or patent foramen ovale.
Embolism from aortic atheromatous disease has been proposed
as a mechanism of cryptogenic stroke, including retrograde
embolism from blood flow reversal in the descending aortic arch
during diastole. Other causes include subclinical cerebrovascular
disease, infectious (eg, infective endocarditis) and inflammatory
diseases, and hypercoagulable hematologic states.
• Cryptogenic stroke is a diagnosis of exclusion based on the exclu-
sion of other likely stroke etiologies.
• Cryptogenic stroke accounts for about 30% of ischemic strokes.
Reperfusion in Ischemic Stroke
Fibrinolysis of the cerebral artery thrombus with recombinant
t-PA has been proven in randomized trials to improve functional
outcome from ischemic stroke when administered early in the
course of stroke. The NINDS Stroke trial reported in 1995 estab-
lished that benefit accrues in ischemic stroke provided fibrinolysis
is administered within 3 hours of stroke onset. The ECASS 3
trial reported in 2008 established that benefit also extended to
patients who receive fibrinolytic treatment within 4.5 hours of
stroke onset, but benefit decreases continuously over time from
stroke onset. Fibrinolysis does not significantly improve overall
long-term survival and was associated with a tenfold increase in
ICH. Symptomatic ICH was seen within 36 hours of treatment in
6.4% of patients given t-PA but only 0.6% of placebo patients in
the NINDS trial, but there was no overall increase in long-term
mortality at 3 months after stroke (17% in the t-PA group and
21% in the placebo group).
Patients with a new ischemic stroke that causes a measur-
able neurologic deficit are candidates for thrombolysis, provided
the onset of symptoms was less than 4.5 hours prior and ide-
ally 3 hours before beginning treatment. The main contraindica-
tions to fibrinolysis for stroke include a previous stroke or head
trauma in the previous 3 months, any history of intracranial
hemorrhage (including brain computed tomographic scan evi-
dence of intracranial hemorrhage), major surgery in the previous
14 days, gastrointestinal or urinary tract bleeding in the previous
21 days, or myocardial infarction in the previous 3 months. For
late thrombolytic stroke treatment between 3 to 4.5 hours, other
relative exclusions are age >80 years and/or a combination of
both previous stroke and diabetes mellitus. Other contraindica-
tions are a persistently high BP (systolic ≥185 mm Hg, diastolic
≥110 mm Hg), low platelet count (<100,000/mm3), warfarin anti-
coagulation with an INR >1.7, or the use of dabigatran within
48 hours of stroke onset.
Intracerebral Hemorrhage
An ICH is bleeding directly into the brain (parenchyma), usually
from arterioles or small arteries. In favorable circumstances a
localized brain hematoma forms, with further bleeding limited
by the surrounding brain tissue. Alternatively, bleeding may con-
tinue with hematoma enlargement until it enters the CSF, either
via the cerebral ventricular system or directly onto the surface
of the brain. A large intracranial hematoma may increase intrac-
ranial pressure causing cerebral ventricular shift and cerebral
edema and is associated with a high mortality.
Cardiovascular diseases associated with ICH include sys-
temic hypertension, vascular malformations, iatrogenic bleeding
disorders (often secondary to thrombolysis), antithrombin and
antiplatelet agents, vasculitis and amyloid angiopathy, and recre-
ational drug use (eg, amphetamines and cocaine). An ICH is often
characterized by progressive neurologic symptoms and signs that
are not maximal at symptom onset but develop in tandem with
hematoma formation over minutes or hours. In contrast, stroke

Table 61.2. Efficacy of Warfarin and Aspirin for Stroke Prevention in Atrial Fibrillation: Meta-
Analysis of Randomized Trialsa
Relative Risk Estimated NNT for
Comparison No. of Trials No. of Patients Reduction, 95% CI Primary Preventionb
Adjusted dose warfarin vs control 6 2,900 64% (49–74) 40
Aspirin vs control 7 3,990 19% (1–35) 140
Adjusted dose warfarin vs aspirin 9 4,620 39% (19–53) 90
Abbreviations: CI, confidence interval; NNT, number needed to treat.
a
Data from Hart et al. Ann Intern Med. 2007 Jun 19;146(12):857–67. Includes all strokes (ischemic and hemorrhagic).
b
NNT for 1 year to prevent 1 stroke, based on a 3.5% per year stroke rate in untreated patients with atrial fibrillation and
without prior stroke or TIA.
Previously Published. See “Credit Lines” section.
594 VI Coronary Artery Disease Risk Factors

Table 61.3. Stroke Risk Stratification Schemes for Patients bleeding continues. Symptoms of SAH begin abruptly, in con-
With Atrial Fibrillationa trast to the more gradual onset of neurological symptoms in ICH.
About one-third of SAH patients have a sudden severe warning
CHADS2 ACC/AHA/ESC 2006 Guidelinesb headache (ie, sentinel headache) due to minor hemorrhage that
Congestive heart failurec = 1 point High risk precedes a major SAH. Blood in the CSF induces vasoconstric-
Hypertension = 1 point Prior thromboembolism tion of intracranial arteries which can cause secondary cerebral
Age, 75 y = 1 point >2 moderate risk features ischemia. The treatment strategy of SAH is to rapidly identify
Diabetes = 1 point Moderate risk the cause of SAH through neuroimaging and expeditiously treat
Stroke/TIA/heart failure = 2 points Age >75 y the bleeding vessel to prevent further bleeding.
Hypertensiond
Risk scores range from 0–6 points Diabetes
Low risk = 0 points LVEF <36% or fractional Suggested Reading
shortening Goldstein LB, Bushnell CD, Adams RJ, Appel LJ, Braun LT,
Moderate risk = 1 point Low risk Chaturvedi S, et al; American Heart Association Stroke Council;
High risk = ≥2 points No moderate- or high-risk features Council on Cardiovascular Nursing; Council on Epidemiology and
Abbreviations: ACC/AHA/ESC, American College of Cardiology/American
Prevention; Council for High Blood Pressure Research; Council
Heart Association/European Society of Cardiology; LVEF, left ventricular on Peripheral Vascular Disease; and Interdisciplinary Council on
ejection fraction; and TIA, transient ischemic attack. Quality of Care and Outcomes Research. Guidelines for the primary
a
This scheme is identical to the stratification recommended by the American prevention of stroke: a guideline for healthcare professionals from the
College of Chest Physicians Evidence-Based Clinical Practice Guidelines American Heart Association/American Stroke Association. Stroke.
(8th edition). Singer et al. Chest. 2008 Jun;133(6 Suppl):546S-92S. 2011 Feb;42(2):517–84. Epub 2010 Dec 2. Erratum in: Stroke. 2011
b
Data from Fuster et al. Circulation. 2006 Aug 15;114(7):e257–354. Erratum in: Feb;42(2):e26.
Circulation. 2007 Aug 7;116(6):e138.
c
Recent heart failure exacerbation was used in original stratification, but
subsequently any prior heart failure has supplanted.
d
Abbreviations
History of hypertension; not specifically defined.
Previously Published. See “Credit Lines” section. BP blood pressure
CSF cerebrospinal fluid
ICH intracerebral hemorrhage
secondary to cardiac embolism usually presents with an abrupt INR international normalized ratio
onset of neurologic symptoms that are maximal soon after stroke NSE neuron-specific enolase
onset and improve gradually with time. SAH subarachnoid hemorrhage
TIA transient ischemic attack
t-PA tissue plasminogen activator
Subarachnoid Hemorrhage
SAH is bleeding into the CSF within the subarachnoid space that
surrounds the brain and is typically due to rupture of a saccular Names of Clinical Trials
arterial aneurysm at the base of the brain or bleeding from a ECASS 3 European Cooperative Acute Stroke Study III
vascular malformation near the brain surface. Rupture of a sac- NINDS National Institute of Neurological Disorders and Stroke
cular aneurysm may lead to rapid bleeding with a precipitate RE-LY Randomized Evaluation of Long-Term Anticoagulation
rise in intracranial pressure with frequent death or coma if the Therapy
 62

Heart Disease in Women

PATRICIA J. M. BEST, MD, and SHARONNE N. HAYES, MD

Cardiovascular disease is the leading cause of death in women, Sex Differences and Bias
outnumbering deaths from all other causes combined. Each year,
Clear sex differences have been identified in the epidemiology
more than 500,000 women experience a myocardial infarction
and presentation of disease, prevalence of risk factors, physiol-
and more than 250,000 die of coronary artery disease. Despite
ogy, and response to diagnostic tests and interventions (Box 62.1).
the national campaigns to increase the awareness of heart dis-
Also, several factors solely affect women, including menopausal
ease in women, including the Go Red campaign and the Red
status, hormone replacement therapy, oral contraceptives, and
Dress campaign, in a recent survey only 55% of women were
pregnancy-related heart disease. During the past 2 decades, sev-
aware that cardiovascular disease is the leading cause of death
eral studies have noted important sex differences in clinical out-
in women and fewer than 15% of women perceived it as a sig-
comes and in the use of diagnostic and therapeutic interventions
nificant risk to themselves. Furthermore, the prevalence of
during the evaluation and treatment of women with chest pain
cardiovascular disease in women, including coronary artery dis-
and myocardial infarction. Despite continued evidence of the
ease, congestive heart failure, stroke, and hypertension, exceeds
importance of heart disease in women, they are still evaluated
that in men in the population older than 55 years (Table 62.1).
less intensively, underreferred, and not treated as aggressively as
Because of the higher proportion of women in the aging popula-
men for comparable presentation and disease. Furthermore, in
tion, each year more women die of cardiovascular disease than
1 study, 10 times as many men with abnormal results of nuclear
men. Importantly, increasing prevalence of risk factors for heart
stress tests were referred for coronary angiography as women
disease such as obesity and diabetes, which affect women to a
with similar results, and women with abnormal test results were
greater extent than men, likely will make heart disease in women
more than 4 times as likely to have their symptoms attributed
more prevalent at an even younger age in the future. The mortal-
to psychiatric causes. Myocardial infarctions are still clinically
ity rate from cardiovascular disease in men has declined steadily
misdiagnosed at a much greater frequency in women than in men.
during the past 20 years. In women, unfortunately, this rate has
Although most of the differences and apparent bias found in some
remained relatively unchanged (Figure 62.1).
studies can be attributed to differences in baseline characteristics
Despite the magnitude of the problem, on average only 25%
of the patients, some investigators have been concerned that the
of subjects in most cardiovascular trials are women. Newer stud-
almost universal worse outcomes of cardiovascular disease in
ies have shown marked sex differences in response to therapy,
women cannot be explained solely by statistically controlling for
outcomes, and preventive strategies; these differences support
older age and comorbid conditions.
the need for more information about optimal primary and sec-
ondary prevention strategies, diagnostic methods, and response
to medical and surgical therapy in women. Coronary Artery Disease
Symptoms
On average, women with coronary artery disease present with
Abbreviations and acronyms are expanded at the end of this chapter. symptoms, cardiovascular events, or sudden cardiac death
595
596 VI Coronary Artery Disease Risk Factors

Table 62.1. Age-Adjusted Prevalencea of Coronary Heart Disease,b by Sex: Behavioral Risk Factor Surveillance System, United
States, 2006–2010
2006 2007 2008 2009 2010
P Value for % Change From
Characteristic % (95% CI) % (95% CI) % (95% CI) % (95% CI) % (95% CI) Linear Trend 2006 to 2010
Men 8.5 (8.3–8.8) 8.0 (7.8–8.2) 8.2 (8.0–8.4) 7.5 (7.3–7.7) 7.8 (7.6–7.9) <0.01 −8.2
Women 5.2 (5.0–5.4) 4.8 (4.7–5.0) 4.9 (4.7–5.0) 4.4 (4.2–4.5) 4.6 (4.5–4.7) <0.01 −11.5

Abbreviation: CI, confidence interval.

a
Weighted estimates, age-adjusted to the 2000 U.S. standard population.
b
Respondents were asked, “Has a doctor, nurse, or other health professional ever told you that you had angina or coronary heart disease?” and “Has a
doctor, nurse, or other health professional ever told you that you had a heart attack, also called a myocardial infarction?” Refused, don’t know, and
missing responses were excluded from analysis.
Previously published. See “Credit Line” section.

approximately 10 years later than men. Although the mechanism
for this delay has not been explained completely, it is likely due to
the protective effects of endogenous estrogen in premenopausal
women. Most men and women present with typical symptoms
of coronary artery disease. However, disproportionately more
women present with atypical symptoms, including prominent
dyspnea, fatigue, referred pain, indigestion, nausea, syncope,
or sweating. For women, advanced age, lower activity level, and
increased prevalence of diabetes and other comorbid conditions
often contribute to the more frequent occurrence of silent isch-
emia, dyspnea, and other nonclassic symptoms. Furthermore,
women typically present later than men to the emergency depart-
ment with symptoms of an acute myocardial infarction. Because
of this later presentation and less-typical symptoms, the diagno-
sis of heart disease in women can be missed and the delays in
therapy can result in ineligibility for therapy such as emergency
percutaneous coronary intervention or fibrinolysis.
Stress Testing
The noninvasive diagnosis of coronary artery disease in women
is challenging. Standard stress electrocardiographic testing
is less accurate in women than men. This difference has led
some practitioners to adopt a rather negative approach and to
treat women empirically without further investigation or use of
invasive testing. This approach is not warranted and could con-
tribute to poor outcomes. Numerous studies have examined the
results of exercise electrocardiography in women and found high
false-positive rates in women compared with those in men, per-
haps because of the lower prevalence of coronary artery disease
in women until the age of 70 years. Also, there is evidence that
the lower specificity is related to sex-specific autonomic and sex
hormone effects on electrocardiography. In older women, failure
to achieve an adequate stress level as a result of deconditioning
or orthopedic limitation may adversely affect the sensitivity of
an exercise test. Despite these limitations, normal findings on
stress electrocardiography at an adequate workload in a woman
are a good indicator that flow-limiting coronary artery disease
is unlikely.
Because of these limitations, imaging stress tests have gained
popularity for women. However, sex-specific artifacts and physi-
ologic responses have been described in studies of both nuclear
and echocardiographic stress tests. Historically, women have
been underrepresented in studies of imaging stress tests, and
the reported sensitivities and specificities have varied widely.
Therefore, drawing conclusions about the absolute incremen-
tal value of stress imaging over standard stress testing is not
possible.
Stress thallium scintigraphy improves diagnostic accuracy
in women compared with radionucleotide angiography (mul-
tiple gated acquisition scanning), but many of the available
clinical trials relied on planar thallium rather than the more
commonly used single-photon emission computed tomography.
Breast tissue attenuates radioactivity and may produce a false-
positive result because of artifactual defects in the anterior wall
and septum. The use of technetium Tc 99m sestamibi, a higher

550 Males
Deaths in Thousands

Females

500

450

400

1980 1985 1990 1995 2000 2005

Years
Figure 62.1. Mortality Trends for Cardiovascular Disease, by Sex, in the United States. (Previously published. See “Credit Line” section.)
 62 Heart Disease in Women
Box 62.1. Known Sex Differences That May Affect
Diagnosis, Treatment, and Outcomes of Heart Disease
Epidemiology and prevalence
Age at onset
Etiology
Presenting symptoms
Risk factor prevalence and strengths
Comorbid conditions, including diabetes mellitus,
obesity, and chronic kidney disease
Body and coronary artery size
Menopausal and hormonal status
Myocardial response to aging, blood pressure,
and volume overload
Accuracy of diagnostic tests
Physiologic response to exercise
Response to pharmacologic intervention
Psychosocial and economic factors
Communication style
energy radiotracer, reduces the breast tissue attenuation artifact,
and limited comparison studies have suggested that thallium and
sestamibi have similar test sensitivities but test specificity may
be enhanced by sestamibi imaging. The use of pharmacologic
agents such as adenosine or dobutamine has a similar diagnostic
accuracy for both men and women.
Cardiac positron emission tomography has a high sensitivity
similar to that of conventional stress sestamibi imaging, but it
has a higher specificity. Positron emission tomography is par-
ticularly beneficial for obese patients because it does not have
the attenuation artifacts present with sestamibi imaging. Even
in patients with no significant obstructive coronary disease on
angiography, abnormal myocardial blood flow and coronary flow
reserve found on positron emission tomography are associated
with increased mortality.
Exercise echocardiography improves the accuracy of exercise
testing for the diagnosis of heart disease in women and has been
proposed as a cost-effective initial approach to the evaluation of
chest pain. Also, it has the advantage of providing information
about other causes of chest pain, including valvular and myocar-
dial function. Dobutamine echocardiography is safe in women,
and studies support a similar diagnostic accuracy for men and
women.
As in all stress testing, the pretest probability of disease is
likely more important for determining diagnostic accuracy than
the specific type of test that is used. If the likelihood of coro-
nary artery disease is low, no stress test is very accurate. If the
goal is localization of ischemia or the resting electrocardiogram
is abnormal, imaging techniques should be used. Additionally, if
a woman cannot exercise adequately because of noncardiac fac-
tors, the initial stress test should be pharmacologic. If, however,
she can exercise adequately, this type of stress testing should
be used because of the added benefit of the exercise portion of
the test.
In women with an intermediate probability of coronary artery
disease and a normal resting electrocardiogram, standard stress
electrocardiography has acceptable sensitivity and specificity.
597
If the results are normal, there is high negative predictive value
regarding the absence of coronary artery disease and the prog-
nosis is good. There are not enough data to demonstrate a clearly
superior imaging technique in women, so when this approach
is chosen, the type of study should depend on local expertise,
patient characteristics, and cost.
In women with worrisome symptoms and a high pretest prob-
ability of coronary artery disease, an argument can be made to
proceed directly to coronary angiography to define the anatomy.
Coronary angiography is safe in women, and most studies have
found that despite sex differences in rates of referral to angiogra-
phy, after the anatomy is defined, women are revascularized at a
similar rate to men.
Electron-beam computed tomography is a sensitive method
for detecting coronary calcium and is similarly predictive of
obstructive coronary artery disease in men and women. Women
have a significantly lower coronary calcium score than men at
all ages, even when coronary artery disease is present. However,
the specificity and both positive and negative predictive values
of electron beam computed tomography are similar for men and
women.
Medical Therapy
Overall, aspirin, β-blockers, hydroxymethylglutaryl coenzyme
A reductase inhibitors (statins), and angiotensin-converting
enzyme inhibitors are underutilized in eligible patients, espe-
cially women and the elderly, even when coronary artery disease
is documented or after a myocardial infarction. More commonly,
calcium channel blockers, which have no documented survival
benefit, have been used for treatment in women. Greater use of
proven effective therapies is needed in female patients.
Aspirin
Multiple studies have shown that the use of aspirin has equal ben-
efit in men and women after a myocardial infarction. Recently,
Ridker and colleagues studied the use of aspirin for primary
prevention in 3,876 healthy women older than 45 years. They
received 100 mg of aspirin every other day or a placebo and were
followed for 10 years. All women had a significant benefit in the
prevention of strokes (relative risk, 0.83). Alternatively, there
was no impact on the prevention of myocardial infarction in the
overall study. However, the 4,097 women older than 65 years had
a significant reduction in the risk of myocardial infarctions (rela-
tive risk, 0.66) and a reduction in major adverse cardiovascular
events (relative risk, 0.74). Thus, in women, aspirin is effective
for secondary prevention and for primary prevention of stroke in
those older than 45 years and for myocardial infarction in those
older than 65 years.
β-Blockers
The ISIS-1 and ISIS-2 trials showed improved survival in women
receiving β-blockers and aspirin, comparable to that in men, and
the greatest benefit was in patients at highest risk. Other studies
of the use of β-blockers have shown a similar benefit.
Hydroxymethylglutaryl Coenzyme A Reductase
Inhibitors (Statins)
The most current studies show an added benefit for greater
reduction of low-density lipoprotein levels to less than 70 mg/dL
for secondary prevention. Interestingly, in these studies and
598 VI Coronary Artery Disease Risk Factors
prior studies of secondary prevention, women had generally a
greater benefit from reduction of low-density lipoprotein levels
with statin therapy. In studies such as the REVERSAL trial, the
women who received high-dose statin therapy had a reduction
in atheroma volume on intravascular ultrasonography compared
with men, who had only slowing of disease progression.
Angiotensin-Converting Enzyme Inhibitors
Angiotensin-converting enzyme inhibitors generally have ben-
eficial effects in women but often less than those in men. This
difference may reflect the fact that women typically have less
reduction in left ventricular function after a myocardial infarc-
tion and less benefit from the treatment of hypertension. Despite
these sex differences, angiotensin-converting enzyme inhibitors
still should be used in women who meet the appropriate criteria.
Coronary Angiography
Several studies have found that female sex is an independent pre-
dictor of a lower likelihood of receiving coronary angiography,
even when baseline factors and severity of disease are consid-
ered. When women are referred for angiography, they tend to
be later in the course of their disease. Because coronary angiog-
raphy is a prerequisite for catheter- or surgical-based revascu-
larization, women in these studies had a de facto lower rate of
revascularization. Among men and women who undergo coro-
nary angiography, there appears to be little difference in the
subsequent use of percutaneous coronary revascularization and
coronary artery bypass grafting, a suggestion that once the anat-
omy is defined, subsequent decisions are based primarily on the
severity of disease and not on sex. Although the appropriate use
of coronary angiography is still debated, these studies showed
that appropriate cardiac catheterization was associated with low
mortality in both men and women and the relative underuse of
coronary angiography in women is associated with adverse long-
term outcomes.
Percutaneous Coronary Intervention
Early in the interventional era, the procedural success rate for
balloon angioplasty in women was lower than that in men. In
the National Heart, Lung, and Blood Institute registry of per-
cutaneous transluminal coronary angioplasty alone, the angio-
graphic success rate was 60.3% in women and 66.2% in men,
and the clinical success rate was 56.6% in women and 66.2% in
men. Additionally, women also had a higher rate of complica-
tions, including a higher rate of intimal tears, which in the bal-
loon angioplasty era was associated with an increased need for
emergency coronary artery bypass grafting. Angioplasty was
associated with a six-fold increase in short-term mortality in
women compared with men. Many of the increased procedural
complications and lower success rate in women were attributed
to large and nonsteerable catheters and balloons in the generally
smaller coronary arteries of women. Subsequent to 1985, sex dif-
ferences in outcomes after percutaneous coronary revasculariza-
tion disappeared.
With current technology and equipment, no sex difference
in procedural success or restenosis rates has been documented.
Women undergoing coronary angiography and percutaneous
coronary revascularization continue to have increased vascular
complications compared with men. Additionally, women have
a greater risk of bleeding associated with adjuvant therapies,
including glycoprotein IIb/IIIa inhibitors, although their use
generally has a similar benefit in women. In both men and
women, stenting has decreased the mortality associated with
percutaneous coronary intervention, decreased morbidity, and
allowed sicker patients with more complex disease to be treated.
Women have slightly smaller coronary arteries, which contribute
to adverse outcomes, but in the stenting era and now drug-eluting
stent era, these differences are minimal and sex differences, after
adjustment for comorbid conditions, are eliminated. Long-term
outcomes after percutaneous coronary intervention are similar
for both men and women. There is some suggestion that women
may actually have less angiographic restenosis and fewer sub-
sequent myocardial infarctions, repeat percutaneous coronary
interventions, or coronary artery bypass graftings after the initial
procedure.
Nearly all reported series show that women have more severe
angina and more concomitant illness, including diabetes melli-
tus, hypertension, and heart failure, at the time of intervention.
When these baseline characteristics are considered, there are
minimal or no sex differences in short- or long-term survival
rates or rates of myocardial infarction or coronary artery bypass
grafting whether the interventional procedures were performed
for unstable angina, acute myocardial infarction, or electively for
stable angina.
Women are more likely than men to have residual angina
and to take anti-anginal medications after percutaneous coro-
nary intervention. These differences also occur after coronary
artery bypass grafting and have not been explained completely.
They may be related to greater microvascular disease and abnor-
malities in coronary flow reserve associated with left ventricular
hypertrophy or diabetes mellitus.
Currently, percutaneous coronary intervention should be
offered to women who have appropriate indications for revascu-
larization and suitable anatomy without specific concerns about
sex. An appropriate referral for coronary angiography is neces-
sary in these women to document the anatomy and lead to subse-
quent revascularization.
Coronary Artery Bypass Grafting
According to early studies, women who had coronary artery
bypass grafting experienced greater operative and short-term
mortality than men. Various explanations for this difference
include technical factors related to smaller body size, more
advanced disease at the time of operation, women more often
presenting on an urgent or emergency basis, and referral bias.
However, population studies and long-term results from the
CASS registry and BARI trial indicate similar graft patency
and long-term survival benefit in men and women after surgical
revascularization. The rates of perioperative death and myocar-
dial infarction are greater for women, but this disparity disap-
pears when baseline factors (eg, age) are considered.
Women have a greater risk of heart failure, external wound
infections, longer hospital stays, longer postoperative intubation,
more blood transfusions, and postoperative congestive heart fail-
ure. Women are more likely to have residual angina requiring
therapy after bypass grafting. Importantly, in studies such as
BARI, there was no difference in acute and 5-year outcomes in
bypass grafting between men and women, after adjustment for
differences in baseline characteristics women had a lower mor-
tality at 5 years with an odds ratio for survival of 0.60. Women
with diabetes also had similar outcomes regardless of whether
they were treated with percutaneous coronary intervention or
 62 Heart Disease in Women
coronary artery bypass grafting, and the differences in diabetic
patients occurred primarily in men. Thus, women should be
referred for coronary artery bypass grafting when appropriate,
and concerns about increased mortality should not influence
referral.
Myocardial Infarction
Women consistently have a nearly 50% greater mortality rate
after myocardial infarction than men. Much of this increased
mortality can be attributed to comorbid conditions, including
hypertension, diabetes, heart failure, chronic kidney disease,
and advancing age. Debate remains whether there is any mor-
tality difference in women and men after adjustment for these
factors, especially in older patients. However, younger women
(<50 years) in the National Registry of Myocardial Infarction-2
had a marked increase in mortality compared with their male
counterparts (odds of death 7%, for every 5 years of decreasing
age in women). This unusual association of younger age with
increased mortality has been observed in several study popula-
tions, but the specific cause of this effect remains unknown.
Women are different in other aspects of their presentation.
They have a greater delay to presentation to the emergency
department and have a greater delay to treatment after their first
electrocardiogram. Increased public awareness of the risk of
heart disease in women likely will improve the delays women
have to presentation, but the persistent delays in hospital therapy
of women suggest that physicians still need to better recognize
the risk and symptoms of heart disease in women. Furthermore,
in contemporary studies, acute myocardial infarctions are still
misdiagnosed more commonly in women. When women present
with an acute coronary syndrome, they more commonly present
with unstable angina and have a greater increase in B-type natri-
uretic peptide than men.
Fibrinolytic Therapy
Fibrinolysis is a highly effective therapy for the reduction of
mortality after a myocardial infarction. Women and men seem to
derive a similar reduction in mortality with fibrinolytic therapy;
the risk reduction in most studies is 25% to 30%. The absolute
mortality in women receiving fibrinolytic therapy is still higher
than that in men, given the even higher mortality in women with-
out reperfusion therapy. Women have angiographic reperfusion
rates similar to those in men, but they have higher bleeding com-
plications, including hemorrhagic stroke and major bleeding. The
increase in hemorrhagic stroke seems to be particularly higher
in women older than 70 years. However, the increase in bleed-
ing complications may be at least partially explained by higher
activated partial thromboplastin times. Reinfarction also seems
to be more than twice that in men after fibrinolytic therapy, a dif-
ference that raises the possibility that women may require more
aggressive angiography after fibrinolytic therapy. Despite the
marked benefit of fibrinolytic therapy in women, there is a lower
utilization of this therapy in women, and those who receive the
therapy have a greater time from initial electrocardiography to
treatment.
Percutaneous Coronary Revascularization
In women, percutaneous coronary intervention for an acute
ST-elevation myocardial infarction reduces mortality. In the
PAMI study, the unadjusted mortality was 3.3 times higher with
599
fibrinolytic therapy than with percutaneous coronary interven-
tion. Importantly, percutaneous coronary intervention is asso-
ciated with a lower incidence of intracranial hemorrhage than
fibrinolytic therapy, and given the higher incidence of intracra-
nial hemorrhage in women with fibrinolytic therapy, the overall
risk:benefit ratio favors percutaneous coronary intervention for
reperfusion therapy in women more than in men. Women still
have approximately a three-fold higher vascular access compli-
cation rate with percutaneous coronary intervention than men.
In other studies, such as that by Mulller and colleagues, women
treated as aggressively with primary percutaneous coronary
intervention for acute myocardial infarctions had even lower mor-
tality after a myocardial infarction than men. Other studies with
a similar goal of aggressive percutaneous coronary intervention
in both women and men have shown no difference in mortality
from myocardial infarction. However, retrospective data suggest
that women still have a higher mortality after primary percutane-
ous coronary intervention. Exclusion criteria in studies, delays in
therapy, and biases in the use of proven strategies to reduce mor-
tality in women with myocardial infarctions may explain some of
these differences between the trials and real-world practices.
Adjuvant Medical Therapy
Although no prospective studies have been designed to evaluate
the role of adjuvant medical therapy in women, substudies, meta-
analyses, and retrospective data suggest a similar benefit for
women and men with aspirin, β-blockers, and thienopyridines.
All efforts should be made to provide these therapies to women
according to the current guidelines of the American College of
Cardiology/American Heart Association.
Postmenopausal Hormone Therapy
Our current understanding of postmenopausal hormone therapy
highlights the importance of prospective studies performed in
women at risk for cardiovascular disease or who have established
coronary artery disease. Multiple observational studies sug-
gested a 40% to 50% reduction in cardiovascular events with
the use of hormone therapy for both primary and secondary pre-
vention. Furthermore, estrogen is well established to have many
mechanisms with potential cardiovascular benefit. These include
a 10% to 20% reduction in low-density lipoprotein cholesterol, a
10% to 30% increase in high-density lipoprotein cholesterol, and
a 25% to 50% reduction in lipoprotein (a) level. In the 1990s,
these findings led to guidelines to consider estrogen as the pri-
mary therapy for secondary prevention in women with hyperlipi-
demia. However, in 1998, when results of HERS were presented
by Hulley and colleagues, the beneficial effects of estrogen were
questioned. In this secondary prevention study of 2,763 women
younger than 80 years who received continuous combined hor-
mone therapy (conjugated equine estrogen plus medroxypro-
gesterone) and followed for 4.1 years, there was no reduction in
the incidence of cardiovascular events with hormone therapy.
Additionally, there were increased thrombotic events, including
pulmonary emboli, with hormone therapy, particularly during
the first 2 years of follow-up. Hormone therapy had no significant
effect on stroke.
The next important study was the ERA study by Herrington
and colleagues, which randomized 309 women with coronary
artery stenosis more than 30% to therapy with conjugated equine
estrogen, conjugated equine estrogen plus medroxyprogesterone,
or placebo. Angiographic follow-up was performed at 3.2 years,
600 VI Coronary Artery Disease Risk Factors
and there was no change in luminal diameter or new lesion devel-
opment with hormone therapy.
The WHI was the largest of these prospective hormone ther-
apy studies. This study included 16,608 women 50 to 79 years
old (mean age, 63.3 years) randomized to conjugated equine
estrogen plus medroxyprogesterone or placebo with a planned
follow-up of 8.5 years. However, this study was terminated at
5.2 years because there was no evidence for cardiovascular ben-
efit and there was excessive risk. This study found a hazard ratio
of 1.4 for stroke, 2.1 for pulmonary emboli, 1.3 for breast can-
cer, and 1.3 for heart disease. In the treated women, there was
an excess of stroke of 8 events per 10,000 woman-years. There
was, however, a significant reduction in colorectal cancer and hip
fractures.
In this landmark study of primary prevention in a low-risk
population with a wide age range, there was no cardiovascular
disease benefit to continuous combined hormone therapy. This
study led to multiple questions, such as the timing and dose of
estrogen replacement therapy and the progestin utilized. In the
estrogen-alone trial, women who had previously had a hysterec-
tomy were randomized to receive conjugated equine estrogen or
placebo. This trial was terminated early after 6.8 years of follow-
up (planned, 8.5 years). There was no difference in the primary
outcome of myocardial infarction or coronary death in those
assigned to estrogen therapy compared with placebo (hazard
ratio, 0.95; range, 0.79–1.16). This hazard ratio tended to favor
the use of estrogen in women 50 to 59 years (0.63; 0.36–1.08),
but it did not reach statistical significance. In women 50 to 59
years old, there was less coronary revascularization with estro-
gen therapy (hazard ratio, 0.55; range, 0.35–0.86). Additionally,
the risk of venothromboembolism in the conjugated equine
estrogen plus medroxyprogesterone study was significantly less
with conjugated equine estrogen alone and was associated with
a risk of 2.2 women per 1,000 patient-years compared with pla-
cebo (hazard ratio, 1.32; range, 0.99–1.75).
One criticism of WHI and HERS has been that participants
were postmenopausal for many years before receiving hormone
therapy. In animal studies, it seems that there is a window of
opportunity during which estrogen therapy may be benefi-
cial if initiated in the perimenopausal period but not if started
later. Prospective studies in women are under way and include
KEEPS. Currently, the US Food and Drug Adminstration has a
black box label on estrogen products stating an increased risk for
heart disease, myocardial infarction, stroke, and breast cancer.
Furthermore, it advises that estrogen not be used for prevention
of heart disease and that it should be used in the lowest effective
dose and for the shortest duration possible.
Microvascular Dysfunction and
Endothelial Dysfunction
With publication of the CASS data, it became clear that 50% of
women and 17% of men referred for coronary angiography with
chest pain did not have significant epicardial coronary artery dis-
ease. With the WISE study (sponsored by the National Heart,
Lung, and Blood Institute), greater emphasis has been placed on
the impact of microvascular dysfunction in women (Box 62.2).
Women presenting with an acute coronary syndrome also have
a lower incidence of significant epicardial artery occlusions.
Microvascular abnormalities and other markers of age-related
arterial stiffness, including increased pulse pressure, can be used
to predict ischemic heart disease outcomes in women. These
correlations have not consistently been found to exist in men.
Box 62.2. Vascular Abnormalities in Women With
Heart Disease Compared With Those in Men
Decreased epicardial and microvascular size
Increased arterial stiffness suggesting greater
fibrosis and altered remodeling
More diffuse atherosclerotic disease
More endothelial dysfunction
More smooth muscle dysfunction
Microvascular dysfunction is associated with a worse progno-
sis and may also help explain abnormal results on stress tests
and anginal symptoms in women without evidence of obstructive
disease in the epicardial arteries. In the WISE study, endothelial
dysfunction was associated with decreased functional capacity
and worse outcomes.
Endothelial dysfunction is also an important component of
coronary disease in women. Sex hormones exert effects on vas-
cular reactivity through the endothelium and smooth muscle
cells. These hormonal effects have influenced nitric oxide syn-
thase and L-type voltage-gated calcium; activated calcium chan-
nels; and affect vascular repair. Endothelial dysfunction also
may result in coronary artery spasm that can be demonstrated
by provocative challenges to the endothelium in the cardiac cath-
eterization laboratory with agents such as acetylcholine or meth-
ergine. Patients with endothelial dysfunction are at increased risk
of cardiac mortality and need aggressive management of their
cardiovascular disease risk factors and symptoms.
Heart Failure
Epidemiology
Heart failure is the most common cause of hospitalization in both
men and women. Both the incidence of heart failure and, more
dramatically, the prevalence of heart failure are increasing, in
part due to the improved survival of patients with risk factors
for heart failure such as coronary artery disease. Men have a
slightly higher incidence of heart failure at all ages, but, because
women represent a greater proportion of the elderly population,
more than half (51%) of the people living with heart failure are
women. Women have a 20% risk for development of heart failure
in their lifetime. Generally, men have a greater mortality from
heart failure. In the Framingham Study, the median survival
after diagnosis was 1.7 years for men and 3.2 years for women.
Still, 63% of all heart failure deaths in the United States occur
in women.
Risk Factors
The most common risk factors for heart failure are coronary
artery disease, hypertension, and valvular heart disease in both
men and women. Diabetes, obesity, nicotine use, and age are also
common risk factors. However, the relative roles of the risk fac-
tors differ by sex. Hypertension, diabetes, tobacco use, and left
ventricular hypertrophy are more potent risk factors in women
than men, and physical inactivity is a risk factor in women but not
in men. Diabetes is associated with an increase in heart failure
in women, even when coronary artery disease is not present. Up
to 91% of men and women with heart failure have hypertension,
 62 Heart Disease in Women
and after the age of 55 years women are more likely to have heart
failure than men. After a myocardial infarction, women are more
likely to have heart failure, even if the overall ejection fraction
is preserved.
Systolic Versus Diastolic Heart Failure
Preserved systolic function is present in 40% to 60% of patients
hospitalized for heart failure. In the Framingham Study, women
represented 65% of the patients with diastolic heart failure but
only 25% of those with systolic heart failure. Importantly, most
trials have been conducted in patients with left ventricular systolic
dysfunction, and therefore optimal management and outcomes of
treatment have not been fully evaluated in this population. The
prevalence of diastolic heart failure increases with age, and the
mortality in diastolic heart failure is lower than that in systolic
heart failure (annual mortality 8%-9% in diastolic heart failure,
and 15%-19% in systolic heart failure).
There may be clear sex differences in the response to pressure
overload which may account for the differences in the incidence
of diastolic heart failure. In women, there is a 30% increase in
the end-diastolic volume in heart failure which does not occur in
men, a difference suggesting that women rely on Frank-Starling
mechanisms to increase cardiac output, whereas men rely more
on increases in contractility. Sex hormones may contribute to
some of these differences. Estrogen decreases renin activity,
smooth muscle cell growth, and collagen deposition and inhibits
cardiac fibroblasts and monocytes. Testosterone increases renin
activity, cardiac hypertrophy, and cardiac fibrosis.
Treatment
Angiotensin-converting enzyme inhibitors are currently a fun-
damental part of heart failure regimens. In a meta-analysis of
5 large trials by Flather and colleagues, angiotensin-converting
enzyme inhibitors were beneficial in women (odds ratio for death,
0.85), but this benefit was less than in men (0.79). β-Blockers also
are effective therapy in women; in a pooled analysis of 4 of the
β-blocker studies, women had a benefit from β-blockers similar
to that in men (relative risk of mortality was 0.69 in women and
0.66 in men). Aldosterone antagonists also seem to be equally
effective regardless of sex. However, special consideration is
needed when digoxin is used in women. In the DIG trial, digoxin
resulted in a reduction in hospitalizations for heart failure in both
men and women, and in the overall trial and in men this result was
without an increase in mortality. In women, mortality was higher
with digoxin therapy (33.1% vs 28.9%, P<0.05) and associated
with an increased peak digoxin level. Thus, if digoxin is used in
women, careful monitoring of the digoxin level is warranted.
Despite the benefit of heart failure therapy to women, women
are treated with these medications less frequently than men.
This difference may be due to multiple factors, including the
increased age in women and more preserved left ventricular
function. Additionally, women are less likely to be referred to a
cardiologist for care, which also may impede optimal diagnostic
testing and management.
Outcomes
Women with heart failure are generally more symptomatic than
men, having greater complaints of dyspnea and edema, a greater
reduction in exercise capacity, and more overt signs of heart fail-
ure (edema, increased jugular venous distention, rales, and third
601
heart sound). Women also have a lower quality of life with heart
failure than men, are more likely to be hospitalized with heart
failure, and have longer hospital stays. Despite the increased
symptoms in women, the mortality in women is lower (1-year
mortality in the Framingham Study was 24% for women and
28% for men).
Arrhythmias and Sudden Cardiac Death
Clear sex differences in normal cardiac electrophysiology exist,
including a higher resting heart rate in women, even after auto-
nomic blockade, and prolongation of the QT interval. The QT
interval is similar in boys and girls, but at puberty the QT inter-
val shortens in boys as a result of androgens. Boys develop a
typical male pattern of ventricular repolarization that is charac-
terized by higher amplitude of the J-point, a shorter and steeper
ST-segment course, a steeper ascent, and a higher amplitude of
the T wave. Further sex differences and potential proarrhythmic
effects of medications are caused by the competitive metabolism
of estrogen and other agents by the cytochrome P-450 enzyme
system. Many of these differences are still not well understood.
Paroxysmal Supraventricular Tachycardia
Atrioventricular nodal reentry tachycardia is twice as common
in women as men, whereas atrioventricular reentry through an
accessory pathway is twice as common in men. Hormonal vari-
ations, specifically in the luteal phase of the menstrual cycle,
may play a role in triggering these arrhythmias in women with
a history of paroxysmal supraventricular tachycardia, but the
specific cause has not been elucidated. Possible contributors
include variations in estrogen or progesterone level, increased
catecholamine level, increased body temperature, or altered
calcium channel activity. Despite the sex differences in the fre-
quency of these arrhythmias, radiofrequency ablation is equally
efficacious.
Atrial Fibrillation
Atrial fibrillation is the most common arrhythmia in both men
and women. Although the incidence is higher in men at all ages,
because of the larger number of women older than 75 years the
prevalence is higher in women. In the Framingham Heart Study,
atrial fibrillation also was associated with a higher mortality in
women (adjusted odds ratio for death of 1.9 in women and 1.5
in men). Women have faster ventricular rates with atrial fibrilla-
tion, a higher incidence of QT prolongation with antiarrhythmic
use, and a greater frequency of torsades de pointes. The treat-
ment of atrial fibrillation is similar in men and women, but cau-
tion should be used with antiarrhythmic therapy because of QT
prolongation.
Sudden Cardiac Death
In the 28-year follow-up of the Framingham Study, sudden death
was more frequent in men than women at all ages. This differ-
ence may be partially related to the delayed onset of coronary
artery disease in women. However, even in patients without coro-
nary artery disease, men still have a higher incidence of sudden
cardiac death. In recent years, sudden cardiac death has been
declining in all age groups except women between the ages of
35 and 44 years, in whom the incidence has increased 21%. The
specific reason for this increase is not known. Prevention of
602 VI Coronary Artery Disease Risk Factors

sudden death includes the prevention of coronary artery disease, DIG Digitalis Investigation Group
medical therapy and revascularization for coronary artery dis- ERA Estrogen Replacement and Atherosclerosis
ease, and internal cardiac defibrillators. Although women were HERS Heart and Estrogen/Progestin Replacement Study
underrepresented in most of the trials of internal cardiac defibril- ISIS-1,2 International Study of Infarct Survival, First and
Second
lators, women seem to have equal benefit as men when a device
KEEPS Kronos Early Estrogen Prevention Study
is indicated. PAMI Primary Coronary Angioplasty for Acute Myocardial
Infarction
REVERSAL Reversal of Atherosclerosis With Aggressive Lipid
Names of Clinical Trials Lowering
BARI Bypass Angioplasty Revascularization Investigation WHI Women’s Health Initiative
CASS Coronary Artery Surgery Study WISE Women’s Ischemic Syndrome Evaluation
 63
Heart Disease in the Elderly
IMRAN S. SYED, MD, JOSEPH G. MURPHY, MD,
and R. SCOTT WRIGHT, MD
Older patients have more heart disease than younger patients;
they also benefit more from aggressive medical therapy but at the
cost of a higher incidence of adverse side effects.
Age-Related Changes in Cardiac
Anatomy and Physiology
Ventricle
Heart weight increases about 1 g per year between ages 30 and
90 years, probably owing to left ventricular hypertrophy second-
ary to an age-related increase in systolic blood pressure. This
results in an increase in both left ventricular mass and wall thick-
ness. In addition to myocardial cell mass, intercellular collagen
also increases with age.
Ventricular hypertrophy leads to significant changes in car-
diac function with age, including 1) an increase in left ventricu-
lar stiffness, 2) reduced filling during early diastole, and 3) a
prolonged diastolic isometric relaxation phase. The left ventricle
shortens in length, from apex to base, with the development of an
S-shaped, or sigmoid-shaped, septum in some patients.
Conduction System
The conduction system undergoes marked changes with age,
including the loss of 50% to 75% of the pacemaker cells of the
sinoatrial node and fibrosis of the specialized conduction tissue
of the bundle of His. Interestingly, some parameters of cardiac
function change little with age, including resting cardiac output,
stroke volume, and ejection fraction.
• Systolic function is well preserved in the elderly, but diastolic func-
tion is impaired.
Abbreviations and acronyms are expanded at the end of this chapter.
Atrial Fibrillation
The occurrence of atrial fibrillation increases steadily with age
and is present in about 5% of subjects older than 65 years. Atrial
fibrillation in the elderly is frequently asymptomatic and is asso-
ciated with an increased long-term mortality primarily due to
stroke. In the absence of anticoagulation, the overall stroke risk
with nonrheumatic atrial fibrillation is approximately 5% per
year. The incidence is much higher in patients older than 80 years
and in those with rheumatic valvular heart disease–associated
atrial fibrillation. Independent risk factors for a thromboembo-
lic event in patients with atrial fibrillation include age, diabetes,
hypertension, a history of heart failure, a prior transient ischemic
attack or stroke, an enlarged left atrium, and poor left ventricular
function.
Correctly dosed warfarin (international normalized ratio, 2–3)
reduces the risk of stroke by about two-thirds. Age is an inde-
pendent risk factor for hemorrhagic complications with warfarin
therapy, as are poorly controlled hypertension and excessive anti-
coagulation. Older patients benefit more than younger patients
from anticoagulation, and, overall, the benefits of warfarin
therapy outweigh the bleeding risks in the elderly population. In
general, warfarin is indicated for the prophylaxis of thromboem-
bolic events in most elderly patients in atrial fibrillation. Aspirin
provides limited protection from stroke in elderly patients when
warfarin anticoagulation is strongly contraindicated. A strategy
of rhythm control in which an antiarrhythmic drug is used to
lower the risk of recurrent atrial fibrillation does not reduce the
stroke risk more than a strategy of rate control alone; anticoagu-
lation is the key to stroke prevention in most elderly patients with
atrial fibrillation.
The diagnosis of atrial fibrillation may be subtle in the elderly,
especially when combined with conduction system disease that
may produce a relatively normal ventricular rate. Atrial fibril-
lation can worsen or precipitate heart failure and angina in the
603
604 VI Coronary Artery Disease Risk Factors
elderly patient, and the worsening or new onset of these symp-
toms should merit a search for arrhythmias, including atrial
fibrillation. Mental status changes, stroke, or a transient ischemic
attack also may be the presenting symptoms for new-onset atrial
fibrillation. Patients with occult hyperthyroidism, silent myo-
cardial infarction, hypokalemia due to use of a diuretic, alco-
holism, or digoxin toxicity may present with atrial fibrillation.
Pulmonary disease, including pneumonia, pulmonary embo-
lism, and chronic obstructive lung disease, may precipitate atrial
fibrillation, especially if beta-agonists are used.
• Atrial fibrillation occurs in about 5% of subjects older than 65
years.
• Age is an independent risk factor for hemorrhagic complications
with warfarin therapy in the elderly.
• Warfarin reduces the stroke risk by about two-thirds.
Bradycardias
Aging is associated with an increased occurrence of conduction
system fibrosis within the sinus node, atrioventricular node, and
bundle branches. Sympathetic and parasympathetic neural influ-
ence on the conduction system decreases. Maximal heart rate
decreases with age, and sinus bradycardia is common in the
elderly even in the absence of cardiac disease. The elderly are
more dependent than younger patients on atrial systole to com-
plete late ventricular diastolic filling.
• Maximal heart rate decreases with age.
Coronary Artery Disease
Dyspnea is a more common presenting symptom of coronary
artery disease in the elderly patient than in the younger patient.
A fourth heart sound and a soft mitral regurgitation murmur are
frequently present in elderly patients and are poor predictors of
the presence of coronary artery disease. The treatment of coro-
nary artery disease in the elderly is similar to that in younger
patients. Both coronary artery bypass and PCI are very effective
in the elderly but are associated with a somewhat higher morbid-
ity and mortality rate.
• A fourth heart sound and a soft mitral regurgitation murmur are
poor predictors of the presence of coronary artery disease in the
elderly.
Myocardial Infarction
Myocardial infarction is associated with a higher mortality rate,
a higher incidence of congestive cardiac failure, and a higher re-
infarction rate in the elderly patient than in the younger patient.
Fewer elderly patients with myocardial infarction are eligible
for thrombolysis because of contraindications (such as stroke,
transient ischemic attack, severe hypertension, and bleeding),
and in those without contraindications, thrombolytics are used
less often in part because of the higher occurrence of late and
atypical presentations of myocardial infarction. The diagnosis of
myocardial infarction is more difficult in the elderly; dyspnea and
pulmonary edema are the most common symptoms at presenta-
tion. Electrocardiography frequently is nondiagnostic because of
baseline electrocardiographic abnormalities, including left bun-
dle branch block. Non–Q-wave myocardial infarction, cardio-
genic shock, cardiac rupture, and death due to electromechanical
dissociation are more common. The size of a first infarct does
not increase with age, and death from ventricular fibrillation is
less common than in the younger patient with infarction. In the
elderly, both primary PCI and thrombolysis are beneficial for the
treatment of acute STEMI, but PCI is associated with a lower
rate of stroke than thrombolysis. PCI is the treatment of choice
for elderly patients with STEMI. In the absence of a timely access
to PCI, outcomes with thrombolytic therapy are still better than
with placebo in older individuals. The cost, though, especially
in patients older than 75 years, is increased mortality and bleed-
ing complications and a lower survival than in younger patients.
The risk of stroke following thrombolysis is approximately 3%
in patients 85 years or older. The indications for adjunctive PCI
after myocardial infarction in the elderly are similar to those in
younger patients and include the occurrence of exercise-induced
or spontaneous myocardial ischemia.
• Elderly patients with myocardial infarction are less often candi-
dates for thrombolysis.
• Primary PCI is associated with a lower stroke risk that thrombolyt-
ics in the elderly.
Stress Testing
Treadmill exercise stress testing is less useful in the elderly popu-
lation because of the increased occurrence of resting ST-segment
changes, increased use of digoxin, and the increased incidence
of peripheral vascular, orthopedic, and lung diseases that limit
exercise capacity. The predictive accuracy of a negative stress
test is less in the elderly patient because of the increased occur-
rence of occult coronary artery disease. Stress testing with car-
diac imaging is the preferred option for noninvasive evaluation of
suspected coronary artery disease in the elderly.
• The predictive accuracy of a negative stress test is less in the
elderly patient.
Lipid Management
Despite their proven benefit, lipid-lowering drugs are markedly
underused in elderly patients. Large numbers of elderly patients
with hyperlipidemia, both with and without clinical heart disease,
have participated in multiple clinical trials of cholesterol-lowering
medications, including the 4S, CARE, LIPID, HPS, and PROSPER
trials. These studies have consistently demonstrated significant
benefits in the elderly, comparable to those in younger patients in
relative terms, with an approximately 30% to 40% reduction in
all-cause mortality, major coronary events, and number of revas-
cularization procedures. Benefit was present both for patients with
established coronary artery disease and for those with hypercho-
lesterolemia but without clinically overt heart disease.
In general, the benefits and side effects of lipid lowering with
statins are similar to in older and younger patients. Secondary
prevention of cardiovascular events in elderly patients with estab-
lished coronary disease by statin medication is almost always
beneficial. Primary prevention in elderly patients with dys-
lipidemia without overt coronary disease is beneficial in older
patients with a reasonable life expectancy.
Older patients may benefit more than younger patients in
absolute terms with preventive cardiology strategies because of
their higher prevalence of coronary disease. Secondary causes
of hyperlipidemia, including diabetes and hypothyroidism, are
more common in the elderly. The Women’s Health Initiative
and HERS randomized trials showed that hormone replacement
 63 Heart Disease in the Elderly
therapy in postmenopausal women is not protective against coro-
nary atherosclerosis.
Hypertension in the Elderly
Hypertension is common in persons older than 65 years, with
an overall incidence of approximately 50% and a lifetime risk
of about 80% in subjects surviving into their 80s. Systolic pres-
sure increases and diastolic pressure decreases after age 60 in
both untreated hypertensive and non hypertensive subjects.
Isolated systolic hypertension is a common form of hyperten-
sion in the elderly and is caused primarily by arterial stiffness
and diminished vascular compliance. Pulse pressure increases
with age because of an increase in systolic pressure, usually
with little or no change in diastolic blood pressure. Elevated
systolic blood pressure and pulse pressures are strong predic-
tors of cardiovascular events in the elderly. Multiple random-
ized controlled trials have conclusively proven the benefit of
treating hypertension in elderly patients, including the very
elderly (specifically over the age of 80 in the HYVET trial).
This benefit applies both to patients with isolated systolic
hypertension as well as diastolic hypertension. β-Blockers are
not as beneficial as other antihypertensive drugs in the elderly
in regard to stroke prevention.
Isolated systolic hypertension is an elevated systolic blood
pressure (>140–160 mm Hg) with a diastolic pressure less than
90 mm Hg. Isolated systolic hypertension increases all-cause
mortality by a factor of 2 to 3 and is strongly associated with an
increased risk of myocardial infarction, left ventricular hyper-
trophy, renal dysfunction, and stroke Treating isolated systolic
hypertension in the elderly significantly reduces morbidity and
mortality as demonstrated by the SHEP trial, in which subjects
received a thiazide diuretic and, if needed, a β-blocker. This
benefit extends to the primary end point of stroke and to the
secondary end point of myocardial infarction. Antihypertensive
medication has also been shown to lower the incidence of multi-
infarct and vascular dementia in the elderly in the Syst-Eur trial.
Thiazide diuretics should be administered in low doses (equiva-
lent to 12.5–25 mg of hydrochlorothiazide) to minimize the met-
abolic complications of high-dose thiazide use, which includes
depletion of potassium and magnesium, elevation of serum uric
acid, and mild elevations of plasma glucose and cholesterol. In
general, systolic pressure should be decreased gradually in the
elderly to avoid postural hypotension. The systolic pressure goal
is 1) 20 mm Hg less than the baseline level if the initial value
was between 160 and 180 mm Hg or 2) less than 160 mm Hg
if the initial value was greater than 180 mm Hg. To attain the
target systolic pressure, diastolic blood pressure should not be
reduced to less than 65 mm Hg because excess reduction in dia-
stolic blood pressure may increase the risk of cardiovascular
events.
JNC 7 provided the following key recommendations for
hypertension management:
1. In middle-aged and elderly patients, a systolic blood pressure greater
than 140 mm Hg is a stronger predictor of future cardiovascular
events than diastolic blood pressure.
2. The blood pressure goal is less than 140/90 mm Hg for most patients
but is less than 130/80 mm Hg for patients with diabetes or chronic
kidney disease.
3. Prehypertension is defined as systolic blood pressure of 120 to
139 mm Hg or diastolic blood pressure of 80 to 89 mm Hg. To pre-
vent the development of cardiovascular disease, aggressive lifestyle
modifications are required.
605
4. Thiazide-type diuretics, either alone or combined with drugs from
other classes, should be used in drug treatment of most patients with
uncomplicated hypertension.
Valvular Heart Disease
Calcific aortic stenosis, usually due to degenerative changes in
a tricuspid aortic valve, is the most common valvular heart dis-
ease in the elderly. The classic physical signs of aortic stenosis
seen in younger patients, including the pulsus parvus and pulsus
tardus waveforms, may be absent because of increased arterial
stiffness. Benign systolic murmurs due to aortic sclerosis with-
out stenosis are frequent in older patients. AVR is superior to
balloon aortic valvuloplasty in all but moribund patients with
severe aortic stenosis: TAVI is an emerging alternative to surgi-
cal AVR.
Mitral regurgitation due to ischemic papillary muscle dys-
function and myxomatous degeneration of the mitral valve appa-
ratus occur frequently in the elderly. Mitral stenosis is usually
the late result of rheumatic fever. The opening snap of mitral
stenosis may be absent in the elderly patient because of valve cal-
cification and rigidity. The intensity of the first heart sound also
may be decreased for similar reasons. Balloon mitral valvulo-
plasty is less successful in the elderly patient with mitral stenosis
because of the increased occurrence of valvular and subvalvular
calcification.
• The classic physical signs of aortic stenosis may be absent in the
elderly.
• The opening snap of mitral stenosis may be absent in the elderly.
Congestive Heart Failure
Heart failure occurs in up to 10% of patients older than
80 years, and in many cases it is due to diastolic ventricular
dysfunction with preserved systolic ventricular function often
in association with long standing systemic hypertension. The
elderly are relatively more dependent on the Frank-Starling
stretch response and less dependent on heart rate to increase
cardiac output in response to exercise. The impaired ability of
aged kidneys to excrete excess sodium and water explains both
the pressor effect of excess dietary sodium and similarly the
beneficial antihypertensive efficacy of sodium restriction in the
elderly. Restriction of dietary sodium intake to 100 mEq/day
or less is specifically important in all elderly heart failure and
hypertensive patients. Factors that result in ventricular diastolic
dysfunction and heart failure in the elderly include impaired
ventricular relaxation and increased myocardial stiffness, lead-
ing to an increase in left ventricular diastolic filling pressure.
Treatment of diastolic ventricular dysfunction is primarily
with angiotensin-converting enzyme inhibitors or angiotensin-
receptor blockers.
• The elderly are more dependent on the Frank-Starling stretch
response and less dependent on heart rate to increase cardiac out-
put than younger individuals.
Cardiac Drugs
Compared with younger individuals, the elderly have a smaller
lean body mass, decreased serum protein levels, decreased
glomerular filtration rate, and decreased hepatic microsomal
606 VI Coronary Artery Disease Risk Factors

oxidation. Renal clearance of digoxin, quinidine, and procain- STEMI ST-segment elevation myocardial infarction
amide is decreased, and drug toxicity occurs more easily than TAVI transcutaneous aortic valve insertion
in younger patients. Adverse drug reactions are at least doubled in
the elderly, and patient compliance with drug regimens is poorer.
The elderly have blunted baroreceptor reflexes and diminished β Names of Clinical Trials
receptor responsiveness. 4S Scandinavian Simvastatin Survival Study
CARE Cholesterol and Recurrent Events
• Adverse drug reactions are at least doubled in the elderly. CAST Cardiac Arrhythmia Suppression Trial
HERS Heart and Estrogen/Progestin Replacement Study
HPS Heart Protection Study
Abbreviations
HYVET Hypertension in the Very Elderly Trial
AVR aortic valve replacement LIPID Long-Term Intervention With Pravastatin in Ischemic
JNC 7 Seventh Report of the Joint National Committee on Disease
Prevention, Detection, Evaluation, and Treatment of PROSPER Prospective Study of Pravastatin in the Elderly
High Blood Pressure SHEP Systolic Hypertension in the Elderly Program
PCI percutaneous coronary intervention Syst-Eur Systolic Hypertension in Europe
 Section VII

Myocardial Infarction
This page intentionally left blank
 64
Cardiac Biomarkers
BRIAN P. SHAPIRO, MD, LUCIANO BABUIN, MD, PHD,
and ALLAN S. JAFFE, MD
Introduction
The need for a rapid and reliable diagnosis of MI is crucial. Initial
efforts focused on clinical assessment and electrocardiography,
but over time it became clear that they are often nondiagnostic.
Therefore, clinicians now rely heavily on cardiac biomarkers.
Guidelines call for the routine assessment of cardiac biomarkers
for the diagnosis of MI (Box 64.1).
The interpretation of cardiac biomarkers may be difficult. To
be successful, clinicians must have an understanding of the kinet-
ics of the biomarkers they use, the mechanisms that can cause
elevation, situations in which values can be falsely elevated, and
the accuracy of these assays.
Cardiac Troponin
The cTn complex comprises 3 subunits: 1) cTnC binds calcium,
2) cTnI inhibits actin-myosin interaction, and 3) cTnT binds tro-
pomycin. These subunits are confluent with the actin filament
and have a key role in the regulation of calcium-dependent car-
diac contraction. Although the bulk of cTn exists within the myo-
cardial contractile apparatus, traces are present in what has been
termed the cytosolic, or early releasable, pool (6% for cTnT and
3% for cTnI). It is thought that the first molecules detected in the
bloodstream after myocardial injury are unbound cTn from the
early releasable pool.
It is probable, although unproved and at times controver-
sial, that the integrity of cell membranes must be compro-
mised to allow cTn to exit the cell. Subsequently, portions of
the contractile apparatus are involved and the myofibril-bound
portion of cTn enters the interstitium and eventually the circula-
tion. Newer-generation assays, if used with the sensitive cutoff
Abbreviations and acronyms are expanded at the end of this chapter.
suggested, document most cTn elevations by 2 to 3 hours after
the onset of MI and nearly 100% by 6 hours. cTn values typi-
cally peak at 24 hours and can remain elevated for 1 to 2 weeks
(Table 64.1). cTnT stays elevated slightly longer than cTnI.
Owing to its enhanced sensitivity and nearly perfect cardiac
specificity, cTn is preferred over CK-MB for detecting cardiac
injury. If one is astute in the use of cTn values, the use of CK-MB
should be unnecessary. Newer assays are extremely sensitive and
detect even minute traces of myocardial damage. This enhanced
sensitivity is due in large part to a “release ratio” (the amount of
marker depleted from myocardium that arrives in the circulation)
that is greater than that of CK-MB and to the prolonged window
during which cTn remains elevated. Multiple studies confirm
that for patients with acute coronary syndrome, the use of the
99th percentile of a healthy population maximizes the ability to
determine risk. With the use of these recommended cutoff val-
ues, other testing (eg, myoglobin or CK isoforms) to provide an
earlier diagnosis is no longer needed.
Measured cTn is highly specific for myocardial tissue owing
to the presence of unique genes that encode cTnI and cTnT.
Although some troponin exists in skeletal muscle, including
some fetal isoforms of cTnT, newer assays use antibodies that
are highly specific and selectively bind only to cardiac forms
of troponin. Contemporary assays may, on occasion, detect the
fetal forms of cTnT in developing skeletal muscle. Although
elevated cTn levels indicate damage to the myocardium, they do
not necessarily define the mechanism (Table 64.2). Any stimu-
lus that damages the myocyte will result in release. Thus, clini-
cians should be aware of the various causes of cTn elevations not
related to acute ischemic heart disease.
• Although elevated cTn levels indicate damage to the myocardium,
they do not necessarily define the mechanism.
609
610 VII Myocardial Infarction
Box 64.1. ESC/ACCF/AHA/WHF Consensus Defini-
tion of Myocardial Infarction
Detection of increase and decrease in levels of
cardiac biomarkers (preferably troponin) that
exceed the 99th percentile of the upper reference
limit
Evidence of myocardial ischemia with ≥1 of the
following:
1. Ischemic symptoms
2. Electrocardiographic changes suggestive of new
ischemia
3. New pathologic Q waves
4. Imaging evidence of myocardial damage
Pathologic findings of acute myocardial
infarction
Abbreviations: ACCF, American College of Cardiology Foundation;
AHA, American Heart Association; ESC, European Society of
Cardiology; WHF, World Heart Federation.
Previously published. See “Credit Lines” section.
• Most cTn elevations occur by 2–3 hours after the onset of MI and
nearly 100% by 6 hours.
• cTn values typically peak at 24 hours after MI and can remain
elevated for 1 to 2 weeks.
Creatine Kinase
CK, a protein located within muscle cells, is essential in adeno-
sine triphosphate generation. Three isoenzymes and a mitochon-
drial form exist. The cytosolic forms are composed of M and B
chains. Thus, the isoenzymes are 1) CK-MB (predominantly in
myocardium, but 1%–7% is in skeletal muscle, and traces are
in the small intestine, tongue, diaphragm, uterus, and prostate);
2) CK-BB (in brain and kidney); and 3) CK-MM (in skeletal
muscle). Assays for total CK activity and CK-MB mass and
activity are widely available and were considered the biomark-
ers of choice before the use of cTn. Much like cTn, CK may be
elevated within 3 to 6 hours after onset of symptoms and peak
at 24 hours. However, unlike cTn, CK typically decreases to the
baseline value by 36 to 48 hours. Older CK-MB assays used gel
electrophoresis, which was time-consuming, expensive, and
imprecise. Newer assays analyze the CK-MB concentration with
an enzyme-linked immunosorbent assay. This technique is more
Table 64.1. Cardiac Biomarkers of Myocardial Injury
Molecular
Biomarker Weight, Da
Frequently used in clinical practice
CK-MB 86,000
cTnI 23,500
cTnT 33,000
Infrequently used in clinical practice
Myoglobin 17,800
CK-MB tissue isoform 86,000
CK-MM tissue isoform 86,000
Abbreviations: CK, creatine kinase; cTnI, cardiac troponin that inhibits actin-myosin interaction; cTnT,
cardiac troponin that binds tropomycin.
Previously published. See “Credit Lines” section.
accurate and rapid, and it is the preferred method for CK-MB.
Most assays for total CK are activity-based assays.
Various criteria have been used to define elevations in total
CK and CK-MB. Some authors have advocated a total CK value
greater than twice the upper limit of the reference range with a
concurrent increase in CK-MB. However, increases in CK-MB
despite a normal CK level are known to have adverse prognostic
significance in patients with ischemic heart disease. Some have
suggested the use of a “CK-MB mass index,” which is the ratio of
CK-MB to total CK. A value of 2.5 or more enhances specificity
for cardiac injury but at the expense of sensitivity. In addition, it
is inaccurate if the total CK is extremely elevated or if there is
concomitant skeletal muscle injury that results in reexpression of
the B chain gene as part of the reparative process.
The CK-MB assay cannot distinguish between cardiac muscle
damage and skeletal muscle damage. In contrast, cTn is highly
specific for the heart. Since variable amounts of CK-MB are
found in skeletal and smooth muscle and these muscles contain
more CK per gram than is found in the heart, muscle damage
can cause variable increases in CK and CK-MB. Thus, false-
positive results from extracardiac tissue damage are common
(Box 64.2).
Advantages of cTn Over CK-MB
Levels of both cTn and CK-MB start to increase and peak at sim-
ilar time intervals. However, cTn is markedly better for indicat-
ing myocardial necrosis because of its greater release ratio, which
results in cTn having a higher concentration in the bloodstream
than CK-MB (Figure 64.1). For this reason, there is an enhanced
signal-to-noise ratio, which enables detection of even minute myo-
cardial damage. Thus, one-third of patients with AMI have an ele-
vated cTn level even though they have a normal CK-MB level.
Since the half-life of cTn is longer, this biomarker provides
a longer period for detecting MI. Newer cTn assays detect up to
80% of MIs in 2 to 3 hours. In addition, cTn levels are highly
specific for myocardial damage, as opposed to CK-MB, which is
often elevated with injuries involving skeletal muscle.
One potential advantage of CK-MB was thought to be that it
did not persist after an ischemic insult. This would allow one to
estimate the time of onset of MI and to diagnose reinfarction and
infarct extension more effectively. However, recent studies sug-
gest that cTn may be just as effective as CK-MB in this regard.
For all these reasons, the use of CK-MB is diminishing.
• One-third of patients with acute MI have an elevated cTn level even
though they have a normal CK-MB level.
Mean Time to
Time to Initial Peak Elevation Time to Return to
Elevation, h (Nonreperfused) Normal Range
3–12 24 h 48–72 h
3–12 24 h 5–10 d
3–12 12 h to 2 d 5–14 d
1–4 6–7 h 24 h
2–6 18 h Unknown
1–6 12 h 38 h
 64 Cardiac Biomarkers 611

Table 64.2. Nonthrombotic Causes of Elevated Troponin

Levels Box 64.2. Causes of CK-MB Elevation Other Than
Myocardial Infarction
Diagnosis Mechanism
Cardioversion or defibrillation, especially high
Demand ischemia dose (>400 J)
Sepsis or systemic inflammatory Myocardial depression or supply- Myocardial contusion
response syndrome demand mismatch
Resuscitative chest compressions
Hypotension Decreased perfusion pressure
Hypovolemia Decreased filling pressure or output Skeletal muscle injury or disorders (eg, Duchenne
Supraventricular tachycardia or Supply-demand mismatch muscular dystrophy)
atrial fibrillation Reye syndrome
Left ventricular hypertrophy Subendocardial ischemia Hypothyroidism
Myocardial ischemia Childbirth
Coronary vasospasm Prolonged ischemia with
myonecrosis
Frequent binge drinking and delirium tremors
Intracranial hemorrhage or stroke Imbalance of autonomic nervous Moderately severe exercise
system Severe hypokalemia
Ingestion of sympathomimetic Direct adrenergic effects
Gangrene or severe ischemia or extremities
agents
Direct myocardial damage Abbreviation: CK, creatine kinase.
Cardiac contusion Traumatic
Previously published. See “Credit Lines” section.
Direct-current cardioversion Traumatic
Cardiac infiltrative disorders Myocyte compression
Chemotherapy Cardiac toxicity
Myocarditis Inflammatory stratified into quartiles according to levels of cTnT: mortality was
Pericarditis Inflammatory higher in the highest quartile than in the lowest.
Cardiac transplant Inflammatory or immune-mediated Angiographic data suggest that coronary artery disease is
Myocardial strain more extensive, thrombi are more numerous, and Thrombolysis
Congestive heart failure Myocardial wall stretch in Myocardial Infarction flow grades are reduced when cTn is
Pulmonary embolism Right ventricular stretch elevated. Thus, cTn may be helpful in determining which patients
Pulmonary hypertension or Right ventricular stretch need aggressive therapeutic measures. Several large trials have
emphysema shown that a strategy of using glycoprotein IIb/IIIa platelet
Strenuous exercise Ventricular stretch inhibitors, low-molecular-weight heparin, and early revascular-
Chronic renal insufficiency Unknown ization is more efficacious in patients with elevated cTn and MI.
Previously published. See “Credit Lines” section. Previously published. See This is clearly the case for spontaneous, so-called type 1 AMIs
“Credit Lines” section. (Box 64.3). However, there can also be infarction due to fixed
coronary artery disease or endothelial dysfunction with increases
in myocardial oxygen demand such as that seen with tachycardia,
Diagnosis of AMI
hypotension, or hypertension. These infarctions do not require
The diagnosis of AMI per guidelines group is predicated on the the same sort of aggressive interventional care as type 1 AMIs.
use of cTn. In a clinical situation where ischemia is likely, a ris- cTn values correlate well with infarct size as measured by thal-
ing and/or falling pattern of cTn values confirms the diagnosis lium, sestamibi, or magnetic resonance imaging. For cTnT, mea-
of AMI. The guidelines advocate the use of the 99th percentile surements at 72 to 96 hours correlate best, but for cTnI, the peak
of a normal reference population as the appropriate metric for value may be better.
defining an elevation in cTn. The definition of a changing pat-
tern is unique to the specific assay being used and, at lower lev- Troponin Elevation After PCI
els, will require a greater degree of change than at higher levels. and Cardiac Surgery
(The change necessary to know that a significant change has
Any elevation of cTn after PCI suggests myocardial injury.
occurred is roughly 3 times the standard deviation of the values
However, it is often difficult to know whether this increase was
measured.) For elevated values, this is a 20% change, which is
related to the MI or to the intervention. Recent data indicate that
the criterion suggested for the diagnosis of reinfarction. These
when one accounts for the prognostic significance of baseline
calculations can easily be accomplished for each assay by clini-
elevations, the importance of post-PCI values vanishes for both
cal laboratories; they are much more complex with novel, very
cTn and CK-MB. In the absence of baseline elevations, increases
high-sensitivity assays.
are modest and also do not manifest prognostic importance.
Elevations of cTn are ubiquitous after cardiac surgery. Studies
Use of Cardiac Biomarkers to Estimate suggest that the higher the cTn value after cardiac surgery, the
Prognosis and Infarct Size larger the amount of aggregate damage and the worse the prog-
nosis. According to a recent consensus document, cTn elevations
Data support the use of cardiac biomarkers to estimate progno-
5 times normal should be considered abnormal and suggestive of
sis and the size of infarction after MI (Figure 64.2). Numerous
a postoperative MI.
studies have confirmed the prognostic importance of elevations
of CK-MB and cTn. For example, the GUSTO-III trial studied
approximately 12,000 patients who had ST-segment elevation
Cardiac Biomarkers and Renal Dysfunction
MI. By 30 days, more patients had died with an elevated cTnT The interpretation of cardiac biomarkers may be difficult in
than without (16% vs 6%). In GUSTO-IV, patients were further renal dysfunction. Elevations in cTn and CK-MB are common
 50

Multiples of Upper Reference Limit

Cardiac troponin after
“classic” AMI
20 CK-MB after AMI
Cardiac troponin after
10 “microinfarction”
Upper reference limit

0
0 1 2 3 4 5 6 7 8

Days After Onset of AMI

Figure 64.1. Time Course of the Appearance of Various Markers in the Blood After Acute Myocardial Infarction (AMI). Shown are the time
concentrations/activity curves for troponin after large and small infarctions and for creatine kinase (CK)-MB. For some patients, cardiac troponin
curves have a second peak. (Previously published. See “Credit Lines” section.)

but so is the incidence of coronary artery disease in this
population. In 1 study, cTn levels were measured in 733 dial-
ysis patients who had no cardiac symptoms (cTnI ≥0.1 μg/L
and cTnT ≥0.01 μg/L were considered elevated). cTnT was high
in 82% of these patients, and cTnI was high in 6%. The rea-
sons for this difference are unclear. Pathologic studies have
confi rmed that the mechanism for cTn elevations is cardiac
injury, but it need not be from AMI. Often the abnormalities
were related to myocytolysis. Associations between cTn eleva-
tions and left ventricular hypertrophy, endothelial dysfunction,
and acute left ventricular stretch have been reported. Total CK
and CK-MB are also highly inaccurate in these patients (>50%

Clinical Trials
Antman et al
N Engl J Med. 1996 Oct 31;335(18):1342-9
Christenson et al (GUSTO-IIa)
Clin Chem. 1998 Mar;44(3):494-501
Heeschen et al (PRISM)
Lancet. 1999 Nov 20;354(9192):1757-62
Luscher et al (TRIM)
Circulation. 1997 Oct 21;96(8):2578-85
Summary

Cohort Studies
Brscic et al
Am J Cardiol. 1998 Oct 15;82(8):971-3
Hamm et al
N Engl J Med. 1997 Dec 4;337(23):1648-53
Janorkar et al
Indian Heart J. 1999 Jan-Feb;51(1):31-4
Meyer et al
Cardiology. 1998;90(4)286-94
Summary

0.01 0.1 1 10 100

Less Risk More Risk
OR
Figure 64.2. Odds ratio (OR) for Increased Mortality With a Positive Troponin I. Data are displayed for each study with at least 1 death in both
positive and negative troponin T subgroups. The summary OR assumes random effects. Error bars represent 95% confidence intervals. (Previously
published. See “Credit Lines” section.)
 64 Cardiac Biomarkers
Box 64.3. Clinical Classification of Different Types
of Myocardial Infarction
Type 1
Spontaneous myocardial infarction related to
ischemia due to a primary coronary event such
as plaque erosion and/or rupture, fissuring or
dissection
Type 2
Myocardial infarction secondary to ischemia due
to either increased oxygen demand or decreased
supply, eg, coronary artery spasm, coronary
embolism, anemia, arrhythmias, hypertension, or
hypotension
Type 3
Sudden unexpected cardiac death, including
cardiac arrest, often with symptoms suggestive of
myocardial ischemia, accompanied by presumably
new ST elevation, or new LBBB, or evidence of fresh
thrombus in a coronary artery by angiography and/
or at autopsy, but death occurring before blood
samples could be obtained, or at a time before the
appearance of cardiac biomarkers in the blood
Type 4a
Myocardial infarction associated with PCI
Type 4b
Myocardial infarction associated with stent
thrombosis as documented by angiography or at
autopsy
Type 5
Myocardial infarction associated with CABG
Abbreviations: CABG, coronary artery bypass grafting; LBBB, left
bundle branch block; PCI, percutaneous coronary intervention.
Previously published. See “Credit Lines” section.
false-positive rate). In this study, 2-year survival was signif-
icantly worse among dialysis patients with high cTn values
(Figure 64.3). Even with minor elevations, there was a 2-fold to
5-fold decrease in survival.
• If the cTn value is increasing by 20% or more, an acute event is
likely. If the cTn value is increased but unchanged from previous
values, an MI is far less likely.
Causes of an Elevated cTn Level in the
Absence of Acute Ischemia
Although the cTn assay is highly specific for detecting myocar-
dial damage, it does not necessarily point to an exact mechanism,
since anything that damages myocytes will cause an elevated
value (Table 64.2).
Mechanical or Electrical Injury
Cardiac trauma from surgery or biopsy and from penetrating or
nonpenetrating trauma to the chest are common causes for cardiac
and skeletal muscle injury. Although CK-MB is often elevated
613
when there is damage to the skeletal muscle, any elevation of cTn
suggests cardiac muscle damage. Elevations are often mild, but
in certain circumstances they may indicate coronary artery or
myocardial trauma. Likewise, electrical cardioversion (including
implantable cardioverter-defibrillator firings), ablation, or cardiac
arrest may also cause modest elevations in cTn. Any elevation
should alert the clinician that myocardial damage is present.
Toxins
An elevated cTn level occurs in more than 50% of patients
with sepsis. Although the exact mechanism is unclear, experts
theorize that underlying ischemic heart disease, hypotension,
microemboli, cytokine activation, and myocardial depression
probably have a role. Elevations of cTn invariably have prognos-
tic significance since values typically correlate with severity of
illness and the degree of left ventricular dysfunction. Toxins such
as chemotherapeutic agents, snake venom, and other vasoactive
substances also have the potential to cause cardiac damage.
Myocardial Inflammation
Inflammation of the myocardium or pericardium, as occurs
in infiltrative cardiomyopathy and in infectious, neoplastic, or
inflammatory myocardial involvement, often causes cTn eleva-
tions. In symptomatic patients, this inflammation may be con-
fused with AMI and is the most common cause of elevated cTn
in patients with possible AMI who have angiographic coronary
arteries. Cardiac biomarker elevation often occurs soon after
onset of symptoms but decreases with disease remission. In cases
of infiltrative cardiomyopathy, such as amyloidosis, cTn elevation
often correlates with disease severity and clinical outcome.
Demand Ischemia or Myocardial Strain
Although often related to ischemia due to epicardial atheroscle-
rosis, elevated cardiac biomarkers can result from other causes
related to the coronary arteries. For example, in left or right ven-
tricular hypertrophy, increased wall stress causes a mismatch
between supply and demand. Therefore, in hypertrophied hearts
(eg, hypertrophic obstructive cardiomyopathy, concentric hyper-
trophy, pulmonary hypertension, or heart failure) or in situations
with significant increases in wall stress (eg, congestive heart
failure), elevations of cTn or CK-MB may reflect myocardial
damage and cell death. In pulmonary embolism, acute right ven-
tricular strain and pulmonary hypertension cause cTn elevation
that is often more transient than elevations with acute ischemic
heart disease and usually resolve in less than 2 days. However,
elevations are linked to worse outcomes and are often indica-
tive of hemodynamic instability. Subendocardial ischemia and
endothelial dysfunction are also common causes for cTn eleva-
tion due to supply-demand mismatch.
• The mechanism for cTn elevations in these situations is cardiac
injury, but it need not be from acute infarction.
Summary
cTn is the preferred biomarker and is highly specific for myocar-
dial damage. It can be helpful in confirming an MI as well as in
estimating infarct size and prognosis. Although the most com-
mon cause for an elevated value is AMI, cTn and CK-MB may
be elevated in several disease states. Guidelines caution against
the use of cardiac biomarkers for routine screening, especially for
patients who have a low pretest probability of MI, since an ele-
vated value may lead to misdiagnosis. However, careful follow-up
614 VII Myocardial Infarction

A 100
cTnT <0.01 μg/L
90

Cumulative Survival, %
cTnT ≥0.01 to <0.04 μg/L
80

70

60 cTnT ≥0.04 to <0.10 μg/L

50
cTnT ≥0.10 μg/L
40

30
0 0.5 1.0 1.5 2.0 2.5 3.0     

Time Since Blood Test, y

No. at Risk
cTnT <0.01 μg/L 132 106 25 12
cTnT ≥0.01 to <0.04 μg/L 214 166 41 15
cTnT ≥0.04 to <0.10 μg/L 239 180 63 18
cTnT ≥0.10 μg/L 148 93 20 8

B 100

90
Cumulative Survival, %

80

70 cTnl <0.1 μg/L

60

50
cTnl ≥0.1 μg/L
40

30
0 0.5 1.0 1.5 2.0 2.5 3.0     

Time Since Blood Test, y

No. at Risk
cTnl <0.1 μg/L 688 514 120 51
cTnl ≥0.1 μg/L 45 31 6 2
Figure 64.3. Kaplan-Meier Curves for Survival According to Quartile of Troponin Elevation Among Patients With Renal Failure. Survival was
less among patients in the highest quartile than among those in the lowest quartile. A, Troponin T (cTnT). B, Troponin I (cTnI). (Previously published.
See “Credit Lines” section.)

is mandatory because in most situations such elevations are indic- Names of Clinical Trials
ative of structural cardiac disease and an adverse prognosis. GUSTO-IIa Global Use of Strategies to Open Occluded Coronary
Arteries IIa
GUSTO-III Global Use of Strategies to Open Occluded Coronary
Abbreviations Arteries III
AMI acute myocardial infarction GUSTO-IV Global Use of Strategies to Open Occluded Coronary
CK creatine kinase Arteries IV
cTn cardiac troponin PRISM Platelet Receptor Inhibition in Ischemic Syndrome
MI myocardial infarction Management
PCI percutaneous coronary intervention TRIM Thrombin Inhibition in Myocardial Ischemia

Acute Coronary Syndromes
R. SCOTT WRIGHT, MD, STEPHEN L. KOPECKY, MD,
and JOSEPH G. MURPHY, MD
The leading cause of death in western societies is an ACS.
ACS occurs when there is acute destabilization of a previously
stable atherosclerotic plaque within the coronary circulation,
usually due to plaque rupture or plaque erosion that results in
acute myocardial ischemia. ACS classification includes four fre-
quently overlapping conditions: UA, STEMI, NSTEMI and SCD
(Figure 65.1).
The diagnosis of ACS is primarily a clinical diagnosis of
symptoms of myocardial ischemia, supported by electrocar-
diographic or cardiac imaging evidence of ischemia or infarc-
tion and frequently by biomarker evidence of myocyte necrosis.
Unstable angina, by definition, is a clinical history of myocardial
ischemic symptoms, generally associated with ischemic ECG
changes including ST-segment depression, transient ST elevation
and T-wave inversion, but no detectable rise or subsequent fall in
troponin levels. UA and NSTEMI share a common pathophysiol-
ogy and may be indistinguishable at initial presentation, as the
requisite rise in serum troponin may not appear for several hours
in NSTEMI. SCD is the sudden cessation of cardiac function
with hemodynamic collapse, typically due to sustained ventricu-
lar tachycardia or ventricular fibrillation which are triggered by
acute myocardial ischemia.
The clinical symptoms of ACS are chest pain or chest pres-
sure, chest burning that may simulate indigestion, unexplained
shortness of breath, and atypical discomfort that radiates into the
neck, jaw, arm, or back. Other atypical symptoms are nausea,
diaphoresis, and unexplained vomiting, particularly in women or
the elderly. Most patients with ACS progress to AMI, either a
STEMI or NSTEMI. In as many as 30% of patients, SCD outside
of a health-care setting may be the initial presentation of ACS
after a short prodrome of chest pain, indigestion, or dyspnea.
Abbreviations and acronyms are expanded at the end of this chapter.
65
The diagnosis of MI is supported by an abnormal rise and
subsequent fall in a cardiac biomarker such as troponin or the
less frequently measured CK-MB. An elevated troponin level
indicates cardiac myocyte injury, but not always ischemic injury.
Other causes of an acute troponin rise include heart failure, myo-
carditis, pulmonary embolism or acute toxin-induced injury (see
Table 65.1).
A recent working group from the ACC/ESC established the
criterial for diagnosis of AMI as a rapid rise or fall in a cardiac
necrosis biomarker, ideally troponin plus one additional factor:
1. Symptoms of ischemia
2. ECG changes consistent with ischemia (ST elevation, ST depression
≥ 1 mm or a new LBBB
3. New pathological Q waves (≥0.04 mm wide and/or 1/3 the height of
the QRS)
4. Imaging evidence of a new wall motion abnormality or a new loss of
viability in an area of myocardium
AMI should not be diagnosed solely on the basis of an elevated
troponin level, and in the absence of all of the above criteria
another source of biomarker elevation should be sought.
AMI is also classified into mechanistic subtypes, and while
most episodes of AMI are related to acute plaque destabiliza-
tion and coronary thrombosis, AMI may be attributed to a coro-
nary intervention such as PCI-related thromboembolization,
side branch occlusion, or injury sustained during the process of
cardiac surgery. The final pathophysiologic mechanism remains
prolonged acute myocardial ischemia, regardless of the trigger-
ing mechanism.
Traditional AMI, or AMI related to acute plaque destabili-
zation, is classified as type I AMI. Other mechanisms include
a supply demand mismatch (type II), SCD where the inciting
mechanism may be suspected but not verifiable short of autopsy
(type III) or AMI associated with a coronary procedure. It is
615
616 VII Myocardial Infarction

Acute Coronary Syndrome Classification

Symptoms suggestive of ACS

ST-segment elevation?
Yes No

STEMI NSTE-ACS

Evaluate for reperfusion Cardiac biomarker

therapy elevation?

Yes No
NSTEMI Unstable angina

Assess risk

Figure 65.1. Acute coronary syndrome (ACS) classification. NSTE-ACS, non–ST-segment elevation ACS; NSTEMI, non–ST-segment elevation
myocardial infarction; STEMI, ST-segment elevation myocardial infarction.

important to realize that when AMI occurs in the setting of PCI
or CABG, the cause and effect may not be easily established
so the new classification scheme simply identifies a temporal
relationship.
Classification of MI
MI classified clinically according to the presumed proximate
cause of the myocardial ischemia:
• Type 1: MI consequent to a pathologic process in the wall of the cor-
onary artery (eg, plaque erosion/rupture, fissuring, or dissection)
• Type 2: MI consequent to increased oxygen demand or decreased
supply (eg, coronary artery spasm, coronary artery embolus, ane-
mia, arrhythmias, hypertension or hypotension)
• Type 3: Sudden unexpected cardiac death before blood samples for
biomarkers could be drawn or before their appearance in the blood
• Type 4a: MI associated with percutaneous coronary intervention
• Type 4b: MI associated with stent thrombosis
• Type 5: MI associated with coronary artery bypass graft surgery
Pathophysiology of Unstable
Coronary Syndromes
Myocardial ischemia occurs when there is a mismatch between
myocardial supply and demand for blood flow. In chronic stable
angina there is a relatively fixed regional myocardial, blood flow
with changes in myocardial demand precipitated by changes
in heart rate, myocardial oxygen consumption, and wall stress

Table 65.1. Likilhood of Significant CAD

Feature High Likelihood Intermediate Likelihood Low Likelihood
History Symptoms same as prior angina Absence of high-likelihood features but Absence of high- or intermediate-
History of CAD, MI, sudden death may have any of the following: likelihood features but may have the
Variant angina (pain with reversible Chest or left arm pain as primary symptom following:
ST-segment elevation) Chest pain that is probably not angina Chest pain not consistent with angina
Transient hemodynamic or ECG changes in patients with DM or in non-DM Recent cocaine use
during pain patients with ≥2 other risk factors (high One risk factor but no DM
cholesterol, hypertension, and smoking)
Male 70 years or older
Female 60 years or older
Examination Hypotension, diaphoresis, pulmonary Extracardiac vascular disease Chest discomfort reproduced with
edema, transient MR palpation
ECG ST-segment depression ≥1 mm from Q waves T-wave flattening or inversion of ≤1 mm
baseline ST-segment depression <0.5 mm in leads with dominant R waves
Marked symmetrical T-wave inversion in T-wave inversion >1 mm in leads with Normal ECG
multiple leads dominant R waves
Cardiac markers Elevated TnT, TnI, CK-MB Normal Normal
Abbreviations: CAD, coronary artery disease; CK-MB, creatine kinase muscle and brain subunits; DM, diabetes mellitus; ECG, electrocardiogram; MI,
myocardial infarction; MR, mitral regurgitation; TnI, troponin I; TnT, troponin T.
Previously published. See “Credit Lines” section.
 65 Acute Coronary Syndromes
(exercise, emotion etc.) In acute coronary syndromes there is
usually a precipitous decrease or an abrupt cessation in regional
coronary blood flow with resultant acute myocardial ischemia.
Rare nonatherogenic causes of ACS include coronary artery
spasm, spontaneous coronary dissection, which is typically seen
in young women during the peripartum period or as a late conse-
quence of Kawasaki disease (an infantile coronary vasculitis that
causes childhood coronary aneurysm formation and late coro-
nary stenoses during adulthood in about 20% of patients), or a
reversible stress induced acute myocardial dysfunction typically
seen in elderly women whose exact pathophysiological mecha-
nism remains elusive.
Less commonly, a marked increase in myocardial demand
may precipitate a “watershed” type MI as is typically observed in
perioperative and periprocedural MI, or high cardiac output state
include hyperthyroidism, anemia, fever, pheochromocytoma,
arteriovenous fistula, and hypertensive emergency. Rarely, ACS
is caused by nonocclusive CAD in which epicardial or microvas-
cular spasm decreases myocardial blood supply.
Pathologically an ACS develops when an atherosclerotic
plaque destabilizes with concurrent fissuring or rupture of a lipid
rich core which then comes into direct contact with blood in the
vessel lumen: coronary vasospasm and thrombosis follow. This
“rupturing” of a previously stable plaque exposes a previously
encased lipid-rich necrotic plaque core, itself a highly thrombo-
genic mass to continuity with coronary blood flow. Alternatively,
there can be an erosion of the endothelium that covers an ath-
erosclerotic plaque without exposure of a plaque core. Rarely,
a calcific nodule present in an atherosclerotic plaque may erode
through to the vessel lumen. Plaque rupture is associated with
an intense inflammatory response with macrophage and lym-
phocyte infiltration in proximity to the site of the rupture in the
atherosclerotic plaque fibrous cap. Other factors that contribute
to plaque rupture include intra-plaque hemorrhage from vasa
vasorum, release of matrix metalloperoxidase enzymes and
local mechanical stresses. There is formation of a platelet-rich
thrombus overlying the culprit lesion (Figure 65.2); this abruptly
decreases myocardial blood supply which results in chest pain
and ECG changes indicative of ischemia and, if prolonged, in
myocardial necrosis and troponin biomarker release. Plaque
Figure 65.2. Acute plaque rupture with apparent atheroembolism.
617
erosion is thought to result from blood flow turbulence and apop-
tosis that lead to loss of endothelial cells that overlie the ath-
erosclerotic plaque with minimal inflammatory reaction and no
rupture of a fibrous cap or exposure of a lipid-rich necrotic core.
Smoking is strongly linked to plaque erosion especially in pre-
menopausal women.
• ACS is usually caused by unstable plaque rupture or plaque erosion
that leads to coronary thrombus formation.
• ACS can also be associated with a coronary oxygen supply demand
mismatch, percutaneous coronary revascularization and coronary
artery bypass grafting surgery.
Initial Clinical Assessment
When clinical symptoms suggest an ACS, the goal of the medical
evaluation is to answer two principle questions:
1. What is the likelihood (high, intermediate, or low) that the patient’s
symptoms represent myocardial ischemia (Table 65.1)?
2. What is the patient’s risk category—Low, intermediate, or high? We
cannot emphasize enough that initial clinical evaluation is critical
to the differentiation of ischemic symptoms from nonischemic, but
equally life-threatening, mechanisms.
What are the first steps to clinical evaluation?
We suggest the following first steps:
1. Obtain an accurate, goal directed history
2. Perform a limited physical examination
3. Review the patient’s ECG within 10 minutes of arrival in the emer-
gency department
4. Obtain and review the cardiac biomarker profile if available
How can one differentiate ischemic chest pain or its equiva-
lent from conditions that mimic MI?
History of Symptoms
First, obtain a history of the symptoms and any precipitating
activities. Most episodes of UA are associated with a visceral
discomfort such that the patient suspects something is gravely
wrong. Sometimes the discomfort is mild but disconcerting
enough to seek medical evaluation. Finally, be sure and rule out
muscular injury or strain as a precipitant.
Pearls from the Clinical History
• Chest pain characteristics that suggest acute myocardial ischemia
include constricting-type chest pain, radiation to the left arm or jaw;
chest pain worse with exertion; associated diaphoresis, nausea, or
vomiting; chest pain similar to a prior MI; older age; and male sex.
• Chest pain that is pleuritic, positional, sharp, reproducible with
pressure, inframammary, and not associated with exertion
decreases the likelihood of AMI.
• Atypical symptoms alone do not exclude the possibility of ACS,
especially in older patients and women who tend to present with
atypical symptoms and may delay seeking medical care.
• Women with ACS derive similar benefits as men from aggressive
medical therapy and have a similar improvement in mortality.
Imaging
Second, it is critical to obtain an ECG within the first ten min-
utes of patient arrival into an emergency department. The timing
delay to ECG performance and review in chest pain patients is
now a “quality marker” in most hospital systems.
618 VII Myocardial Infarction

ECG Pearls
• If ST elevation is present in two contiguous ECG leads, then the
presumptive diagnosis is STEMI, although other diagnoses such as
pericarditis should also be considered.
• A diagnosis of STEMI warrants emergency consideration of reper-
fusion therapy either primary PCI or thrombolysis, a decision that
should be made within 30 minutes of presentation.
• If the ECG shows ST depression, consider the patient at risk for a
NSTEMI or UA depending on the subsequent laboratory data.
Laboratory Evaluation
Third, you must initiate an acute laboratory evaluation with car-
diac biomarkers, a random (non-fasting) plasma glucose, serum
creatinine, sodium, potassium, and a blood count. The ultimate
diagnosis of ACS will be a clinical decision integrating the find-
ings of the history and examination, data from the ECG and the
presence or absence of a biomarker elevation.
Clinical Laboratory Pearls
• Any degree of renal insufficiency (CKD classes III, IV and V) are
associated with an adverse prognosis in ACS
• RPG values > 200 mg/dL establish the diagnosis of diabetes mel-
litus regardless of the patient’s prior history
• Any RPG > 160 at time of ACS is associated with an adverse
prognosis
• Anemia is a frequently mimicker of unstable angina and a negative
prognostic marker in true ACS
• Thrombocytopenia may limit use of dual or triple anti-platelet
therapy
Guidelines
According to the ACC/AHA guidelines for UA–NSTEMI,
several factors from the initial clinical history, physical
examination, ECG, and cardiac biomarkers are associated with a
high likelihood that symptoms reflect acute myocardial ischemia
(Table 65.1). A high likelihood of significant CAD is defined as
more than an 85% chance of having a flow-limiting coronary
artery obstruction, an intermediate likelihood is defined as a
15% to 85% chance, and a low likelihood is defined as less than
a 15% chance.
After estimating the likelihood that the symptoms represent
myocardial ischemia, it is important to determine the risk for the
patient since this affects decisions for triage. The ACC/AHA risk
table (Table 65.2) provides an estimation of the short-term risk of
death or nonfatal MI in patients with NSTEMI-ACS and assists in
the initial triage of these patients. Patients at high or intermediate
risk should be hospitalized in a monitored bed for intensive man-
agement. Low-risk patients may be observed in a chest pain obser-
vation unit. If the patient does not have recurrent pain and the
follow-up 12-lead ECG and serum cardiac biomarker results are
negative for ischemia and infarction respectively after 6 to 8 hours
of observation, the patient may undergo stress testing. This testing
may be done before discharge from the chest pain unit or within
72 hours after discharge. If the patient has recurrent chest pain
consistent with myocardial ischemia, if a 12-lead ECG is consis-
tent with ischemia, or if serum cardiac biomarkers are elevated,
the patient should be admitted and treated for NSTEMI-ACS.
Classification of UA
UA is defined (and differs from stable angina) by the duration and
intensity of angina as graded by CCS classification (Table 65.3).
There are three classical presentations of unstable angina: 1) rest
angina (lasting >20 minutes), 2) new-onset angina (at least CCS
III intensity), and 3) accelerated angina (angina with activity that
is occurring earlier, categorized in a more intense CCS class, or
with increased duration). Other classification include post infarct
angina and post revascularization angina. An initial diagnosis
of unstable angina may later be changed to NSTEMI if initial or
serial levels of cardiac biomarkers become elevated.

Table 65.2. Factors Increasing Short-Term Risk of Death or Nonfatal MI in NSTE-ACS

Intermediate Risk Low Risk
High Risk (No High-Risk Features and (No High- or Intermediate-Risk Features
Feature (At Least 1 of the Following) At Least 1 of the Following) and May Have 1 of the Following)
History Acceleration of ischemic symptoms over Prior MI, extracardiac vascular disease,
past 2 days CABG, prior aspirin use
Pain characteristics Prolonged, ongoing (>20 min) rest pain Prolonged (>20 min) rest now resolved New-onset or progressive CCS class III or
IV angina in the past 2 weeks without
prolonged (>20 min) rest pain
Clinical findings Pulmonary edema, new or worsening Older than 70 years
MR murmur, S3 or new/worsening
rales, hypotension, bradycardia,
tachycardia
Age >75 years
ECG Angina at rest with transient ST-segment T-wave inversions >0.2 mV Normal or unchanged ECG during an
changes >0.05 mV Pathologic Q waves episode of chest discomfort
Bundle branch block, new
Sustained VT
Biomarkers Elevated (eg, TNT or TnI >0.1 ng/mL) Slightly elevated (eg, TnT >0.01 ng/mL Normal
but <0.1 ng/mL
Abbreviations: ACS, acute coronary syndrome; CABG, coronary artery bypass grafting; CCS Canadian Cardiovascular Society; ECG, electrocardiogram; MI,
myocardial infarction; MR, mitral regurgitation; NSTE, non–ST-segment elevation; S3, third heart sound; TnI, troponin I; TnT, troponin T; VT, ventricular
tachycardia.
Previously published. See “Credit Lines” section.
 65 Acute Coronary Syndromes 619

Table 65.3. Canadian Cardiovascular Society Classification
of Unstable Angina
Class Activity Provoking Angina
I Prolonged exertion
II Walking >2 blocks
III Walking ≤2 blocks
IV Minimal or rest
• Unstable angina has three classical presentations: prolonged rest
angina, new-onset angina, and accelerated angina.
• NSTEMI typically manifests as rest angina and is differentiated
from unstable angina by an elevation of cardiac biomarkers.
Principles of Management of NSTEMI and UA
Step 1: Risk Stratification
Estimation of short-term CV risk is a key component of the initial
assessment of patients with ACS: CV risk in this setting refers
to death or a further CV event (new or recurrent MI, or severe
recurrent ischemia requiring revascularization) within 30 days if
the patient is managed medically.
Why Perform Early Risk Stratification?
The 2011 ACC/AHA UA/NSTEMI guideline recommendations
mandate early risk stratification in the assessment of patients
with ACS. All subsequent recommendations for stabilization and
management flow from this initial CV risk assessment. Several
large randomized clinical trials support this strategy.
Risk Stratification in Practice
Several clinical tools are used to assess risk and guide both the
intensity of medical treatment and the need for, and timing of
coronary angiography and invasive therapies in ACS patients.
Risk stratification can occur using one of several risk scores
including the TIMI score (Figure 65.3), the GRACE score, the
Mayo score or by using clinical criteria such as the presence of
ST-segment depression, evidence of acute pulmonary edema or
comorbid clinical conditions like diabetes mellitus or chronic
renal insufficiency.
50
Death, MI, or ischemia requiring
revascularization at 14 days, %
A widely used score to risk stratify patients with ACS is the
TIMI risk score derived from the TIMI 11B trial and validated in
the ESSENCE, PRISM-PLUS, and TACTICS-TIMI-18 trials as
Limits to Normal Activity well as the National Registry of Myocardial Infarction. The risk
None score awards one point for each predictor variables as follows:
Slight 1. Age 65 years or older
Marked 2. Three or more risk factors for CAD (family history of CAD, hyper-
Severe tension, hypercholesterolemia, diabetes, or being a current smoker
3. Prior coronary artery stenosis of 50% or more
4. ST-segment deviation on presenting ECG
5. Two or more anginal events within the prior 24 hours
6. Use of aspirin within 7 days
7. Elevated sernlum cardiac markers
The sum of the seven-point risk score predicts the risk of
developing an adverse outcome of death, infarction, reinfarction,
or recurrent ischemia requiring revascularization within 14 days
after randomization.
The subsequent risks faced by patients with ACS divide nicely
into three broad categories with the TIMI score: low, intermedi-
ate and high risk. The low risk patients typically score in the
range of 0- 2 on the TIMI score calculator and this is associated
with a 14 day CV risk of less than 5%. Patients with a TIMI score
of 3 to 4 are deemed intermediate risk and have a CV event rates
of 13% to 20%. Patients with a TIMI score of 5 or greater are
classified as high risk and have CV event risks in the 25%–40%
range.
The GRACE score, as well as other scores, can also be used to
classify patients into low, intermediate and high risk subgroups.
The GRACE score is based on patient
• Age
• Killip class at presentation (Table 65.4).
• Systolic blood pressure
• Presence of ECG ST-segment deviation
• Cardiac arrest before or during presentation
• Serum creatinine concentration
• Elevation in serum cardiac biomarkers
• Heart rate at presentation
The precise score utilized is less important compared with
an intentional effort to risk stratify every patient who presents
with ACS.
Risk Stratification Using Clinical Examination Findings
Risk stratification can also occur outside of a formal scoring sys-

Low risk
Intermediate risk 40.9 tem. You can easily and quickly stratify the risk of patients by a
40 High risk number of clinical criteria. The Killip classification is an easy
bedside assessment of risk based upon clinical exam findings of
30 26.2 heart failure in patients presenting with AMI (Figure 65.4). There
19.9 is an association between higher Killip class scores in STEMI
20
13.2
10 8.3
4.7 Table 65.4. Killip Classification
0
0/1 2 3 4 5 6/7
Class Examination Findings
TIMI score I No signs of heart failure
II Third heart sound
Figure 65.3. Thrombolysis in Myocardial Infarction (TIMI) scores Elevated jugular pulse
for predicting risk of adverse outcomes. Scores of 0–2 indicate low risk Rales in <50% of posterior lung fields
(green); 3 or 4, intermediate risk (yellow); and 5–7, high risk (red). III Overt pulmonary edema
MI, myocardial infarction. (Previously published. See “Credit Lines” IV Cardiogenic shock
section.)
620 VII Myocardial Infarction
25 Killip class
I
20 II
Mortality rates, %
III/IV
15
10
5 the potential bleeding risks of dual antiplatelet therapy are low.
0
30 days 6 months
Figure 65.4. Mortality rates for Killip classes I (green), II (yellow),
and III/IV (red) at 30 days and 6 months in patients with non–ST-seg-
ment elevation acute coronary syndrome. (Previously published. See
“Credit Lines” section.)
(GUSTO study) and NSTEMI (Mayo CCU Registry) and ele-
vated short and long-term mortality risks in patients with ACS.
Clinical Pearls
Established clinical markers of elevated risk include:
1. Hypotension or impending hemodynamic collapse
2. Significant ST depression (≥2 mm) on 2 or more ECG leads
3. Troponin elevation at time of presentation
4. Acute pulmonary edema
5. CKD stages III-V
6. Diabetes mellitus
7. Elevated glucose
Lesser well known markers of elevated clinical risk:
1. Peripheral vascular disease
2. Significant chronic obstructive pulmonary disease grade 2 or
greater mitral regurgitation
3. Impaired right ventricular function
4. Recent aspirin use
What Value Is Risk Stratification in ACS Patients?
Several randomized controlled studies have demonstrated that
patients presenting at intermediate or high risk benefit signifi-
cantly from an early invasive strategy, while low-risk patients
do equally well with an intensive medical therapy or an early
invasive strategy.
Low-risk ACS patients can be managed in an ECG telemetry
unit or on overnight observation while patients at intermediate
or high risk require admission to a cardiac care unit. Multiple
randomized studies (FRISC II, TACTICS, TIMI-18, ICTUS,
EARLY ACS) support early angiography of patients at inter-
mediate or high risk. The TIMACS trial supports immediate
catheterization in patients with a markedly elevated GRACE
score.
• Early assessment and identification of intermediate and high-risk
ACS patients enables the clinician to determine both the appro-
priate intensity of medical therapy and the timing of cardiac
catheterization.
• Higher TIMI risk score, higher Killip class, and individual markers
of increased age, troponin elevation, ST-segment depression, and
renal insufficiency all portend a worse outcome in NSTEMI-ACS.
Step 2: Initiate Immediate Medical Stabilization
The ACC/AHA 2011 UA/NSTEMI guidelines recommend
immediate medical stabilization in patients presenting with
known or suspected ACS.
Immediate medical stabilization includes the following:
1. Initiation of dual oral antiplatelet agents (aspirin 162 or 325 mg
and a platelet P2Y12 inhibitor such as clopidogrel or ticagrelor when
Prasugrel is also very effective in ACS but is associated with an
increased bleeding risk especially in low weight patients (<60 kg)
and older patients (age ≥75 years) and is best reserved for patients
with suspected clopidogrel resistance and patients undergoing PCI.
2. Initiation of β-blocker therapy when appropriate and not contrain-
dicated (high-grade atrioventricular block, heart rate <55 bpm,
hypotensive, acute pulmonary edema, asthma)
3. Initiation of nitrates, topically or intravenously, to alleviate angina
4. Initiation of antithrombotic therapy with heparin, low-molecular-weight
heparin or a direct thrombin inhibitor like bivalirudin or fondaparinux
5. Statin therapy within the first day of hospitalization
Clinical Pearls
• The 2011 ACC/AHA Guidelines recommend dual oral antiplatelet
therapy in nearly all ACS patients. The dual therapy can be aspirin
+ P2Y12 inhibitor (clopidogrel or ticagrelor) or aspirin + GP IIb/
IIIa antagonist (eptifibatide or tirofiban)
• Prasugrel within the P2Y12 class should be reserved for those
undergoing PCI or patients with suspected clopidrogel resistance
• Abciximab within the GP IIb/IIIa class should be reserved for
those undergoing PCI
• Recent trial data (EARLY ACS) suggests a higher bleeding risk in
patients who receive GP IIb/IIIa therapy upstream from the cath-
eterization laboratory
Step 3: Decide on Early Invasive Versus
Conservative Management Strategy Based
Upon ACS Risk Stratification
Low-Risk Patients
Low-risk ACS patients fare well with a strategy of intense medi-
cal management and stress testing for ischemia prior to hospi-
tal dismissal, with the caveat that patients who develop further
ischemic symptoms while in the hospital need coronary angiog-
raphy (Figure 65.5). It is appropriate to maintain dual antiplatelet
therapy for 1 year following hospital dismissal. Left ventricular
function should be measured during hospitalization, usually by
echocardiography, as well as assessment of any concurrent val-
vular heart disease.
Intermediate-Risk and High-Risk Patients
The 2011 ACC/AHA UA/NSTEMI guidelines recommend an
early invasive strategy in most patients at intermediate or high
risk after initial ACS assessment (Figure 65.6). Medical therapy
described earlier should be intensified in these subgroups of ACS
patients.
Intensified medical therapy includes:
1. Initiation of dual antiplatelet therapy immediately and escalation
to triple antiplatelet therapy during PCI
2. Early initiation of aggressive antithrombotic therapy including
unfractionated heparin or low-molecular-weight-heparin or a
potent antithrombin agent such as bivalirudin or fondaparinux.
3. Aggressive statin therapy, preferably ahead of PCI
 65 Acute Coronary Syndromes 621

Early Invasive vs Conservative Management

Elevated troponin
New ST segment depression
Recurrent CP with CHF
Recurrent CP at rest/minimal activity
Recurrent CP on antianginal therapy
LV dysfunction (EF <40%)
Hemodynamic instability
Sustained VT
Prior PCI within 6 months
Prior CABG

Any 1 present None present

Noninvasive strategy
Coronary
or
angiography
coronary angiography

Figure 65.5. Clinical factors that determine the use of early invasive or early conservative therapy according to the American College of
Cardiology/American Heart Association recommendations. CABG, coronary artery bypass grafting; CHF, congestive heart failure; CP, chest pain;
EF, ejection fraction; LV, left ventricular; PCI, percutaneous coronary intervention; VT, ventricular tachycardia.

Issues in Antiplatelet Therapy in ACS
Most patients with ACS should receive dual anti platelet ther-
apy for at least one year following hospitalization, regardless of
whether they received PCI.
The 2011 ACC/AHA guidelines recommend that all ACS
patients be started on two antiplatelet agents soon after hospi-
talization but grant some latitude to clinicians with regard to
the selection of a second antiplatelet agent. The use of dual oral
antiplatelet agents is preferred over aspirin plus a GP IIb/IIIa
antagonist in most patients as it is associated with a lower risk
of bleeding.
Oral P2Y12 Inhibitors and the CYP2C19 Pathway. There
are three oral platelet P2Y12 inhibitors approved by the FDA for
use in ACS: clopidogrel, prasugrel and ticagrelor. Clopidogrel
and ticagrelor are approved for initiation outside of the cath
eterization laboratory while prasugrel is optimally used in asso-
ciation with PCI. Clopidogrel and prasugrel require activation
before they are effective in vivo.
Clopidogrel activation is significantly influenced by the
CYP2C19 enzyme pathway while prasugrel activation is inde-
pendent of this pathway.
CYP2C19 is a member of the cytochrome P-450 mixed-
function oxidase system; it is important as there are three major
polymorphisms, two of which encode a loss of function allele.
There are also ethnic differences with regard to expression of
these loss of function alleles. It is estimated that 50% of Asians,
34% of African Americans and at least 25% of Caucasians carry
at least one copy of the loss of function allele. These loss of func-
tion alleles prompted the FDA to issue a “black box” warning
about the use of clopidogrel and its diminished effectiveness in
some patients. There is currently no consensus on genetic test-
ing of patients for platelet function polymorphisms prior to use
of clopidogrel or whether testing after initiation of drug therapy
improves survival.
Clinical Pearls
• CYP2C19 accounts for 45% of the activation of clopidogrel in
patients
• There is ethnic variation with regard to the loss of function allele
associated with CYP2C19
• Chinese patients seem to have the highest risk of carrying at least
one loss of function allele
• Pretreatment genetic testing or on therapy platelet function testing
have not been conclusively demonstrated to improve outcomes
ACS patients with diabetes have increased activation of plate-
lets and possibly reduced responsiveness to clopidogrel, that is
not closely related to glycemic control, but rather appears to be
part of a systemic inflammatory response
Prasugrel. Prasugrel offers a pharmacological advantage over
clopidogrel in that it does not require activation through the
CYP2C19 pathway. Prasugrel was compared with clopidogrel in
the TRITON TIMI 38 trial and demonstrated to be clinically
superior with regard to reducing a combined primary CV end
point (CV death, nonfatal MI, and nonfatal stroke). The benefit
observed with prasugrel was offset by a higher major bleeding
risk compared with clopidogrel. The FDA approved prasugrel for
use in ACS populations undergoing PCI but suggested restricting
622 VII Myocardial Infarction

  Diagnosis of UA/NSTEMI is likely or definite 

ASA (class I, LOE: A)

Clopidogrel if ASA intolerant (class I, LOE: A)

Select management strategy

Initial conserative strategy or unknown Invasive strategy

Initiate anticoagulant therapy (class I, LOE: A)
Acceptable options include
• Enoxaparin or UFH (class I, LOE: A)
• Fondaparinux (class I, LOE: B)
• Enoxaparin or fondaparinux preferred
over UFH (class IIa, LOE: B)
Initiate clopidogrel (class I, LOE: B)
Initiate anticoagulant therapy (class I, LOE: A)
Acceptable options include
• Enoxaparin or UFH (class I, LOE: A)
• Bivalirudin (class I, LOE: B)
PCI: add 2nd antiplatelet agent (class I, LOE: A)
• Clopidogrel (class I, LOE: B) or
• GP IIb/IIIa inhibitor (class I, LOE:A)
(IV eptifibatide or tirofiban preferred)
Next step per triage decision at angiography

CABG: PCI: class I: Rx: D/C GP IIb/IIIa

maintenance • Clopidogrel (if not begun precath) (LOE:A) or inhibitors if begun
ASA (class I, • Prasugrel (LOE: B) or and give clopidogrel
LOE: A) • Selectively, GP IIb/IIIa inhibitor (if not begun per conservative
precath) (LOE: A) strategy

Figure 65.6. Class I and Class IIa Recommendations for Initial Management of UA or NSTEMI. ASA indicates American Society of
Anesthesiologists; CABG, coronary artery bypass grafting; D/C, discontinue; GP, glycoprotein; IV, intravenous; LOE, level of evidence; NSTEMI,
non–ST-segment elevation myocardial infarction; PCI, percutaneous coronary intervention; Rx, prescription; UA, unstable angina; UFH, unfraction-
ated heparin. (Previously published. See “Credit Lines” section.)

its use in patients who are elderly (age > 75), have a prior history
of TIA or stroke and/or have additional risk factors for bleeding
such as those with a body weight < 60 kg.
Clinical Pearls
• Prasugrel is a more powerful antiplatelet agent than clopidogrel,
but is overall associated with more bleeding events.
• Presugrel is the thienopyridine antiplatelet agent of choice in
patients with a history of stent thrombosis, clopidogrel treatment
failure, or in those suspected of having laboratory clopidogrel
resistance.
• Prasugrel should not be used in certain patients subgroups, includ-
ing patients with prior TIA or stroke, those who are over age 75
and those with increased bleeding risks, such as those who weigh
< 60 kg
Ticagrelor. Ticagrelor is a reversible and direct-acting P2Y12
inhibitor. Unlike prasugrel and clopidogrel, it does not require
intestinal and hepatic activation to work. Studies also show that it
achieves greater and more complete inhibition of platelet activity
than clopidogrel in a faster time period.
Ticagrelor was tested against clopidogrel in the PLATO trial.
All patients were on baseline aspirin therapy. The primary end
point of the PLATO trial was the occurrence of the composite
end point of of death from any vascular cause, recurrent AMI or
new stroke at 12 months; this was significantly reduced in those
randomized to ticagrelor compared to those on clopidogrel.
Ticagrelor was superior to clopidogrel in patients only when
a lower dose of aspirin (<100 mg) was used. There is no clear
explanation for this observation but the FDA has suggested
reducing the aspirin dose to less than 100 mg daily if concurrent
ticagrelor therapy is used.
Ticagrelor was associated with a higher rate of major non-
CABG bleeding in the PLATO trial as well as unexplained higher
rates of dyspnea as well as slight increases in serum creatinine
and uric acid compared to those randomized to clopidogrel.
Clinical Pearls
• Ticagrelor is a reversible inhibitor of P2Y12
• Ticagrelor use had slightly higher rates of non-CABG associated
major bleeding compared with clopidogrel
• Use of ticagrelor may be associated with a sensation of dyspnea
and with elevations in serum creatinine and uric acid
GP IIb/IIIa Antagonist Therapy in ACS. There are three
FDA-approved intravenous GP IIb/IIIa blockers available for use
 65 Acute Coronary Syndromes
in patients with ACS. Abciximab, the oldest drug in this group
is a platelet glycoprotein IIb/IIIa receptor antagonist. Its use is
restricted to catheterization laboratory use at the time of PCI.
Eptifibatide and tirofiban have been tested outside of the cath lab
in the PURSUIT and PRISM-PLUS studies respectively. Both
are associated with a marginal improvement in clinical outcome,
largely driven by biomarker associated AMI. Both are also used
in the catheterization laboratory with PCI. All of these agents
were approved for use prior to the widespread use of clopidogrel
and to date no definitive study has established a benefit with GP
IIb/IIIa antagonist therapy in ACS, when added to current dual
antiplatelet inhibition with aspirin and clopidogrel.
The EARLY ACS and ACUITY trials tested a strategy of
routine, upstream use of GP IIb/IIIa antagonist therapy against
selective drug use in the catheterization laboratory, as an adjunct
to PCI in 9,492 patients with ACS. All patients in both studies
were on aspirin and clopidogrel, mimicking real-world condi-
tions. The EARLY ACS study was a superiority trial evaluat-
ing a 30 day primary end point of death, recurrent MI, recurrent
ischemia requiring PCI, and PCI-related thrombosis. The
ACUITY trial evaluated 9,207 patients with ACS using a non-
inferior design with a primary end point of the combination of
death, recurrent MI, or unplanned revascularization at 30 days.
Neither trial demonstrated a benefit for the upstream, routine
use of GP IIb/IIIa antagonist therapy compared to selective use
in the catheterization laboratory. Indeed, the ACUITY trial dem-
onstrated a lower rate of major bleeding in the group where GP
IIb/IIIa antagonist therapy was used selectively in the catheter-
ization laboratory.
The 2011 ACC/AHA guidelines suggest that the routine use of
a GP IIb/IIIa antagonist upstream from the catheterization labo-
ratory is of no benefit in low risk ACS patients and increases the
risk of bleeding. The guidelines recommend use of these antago-
nists as the second of two antiplatelet agents in patients who are
at intermediate or high risk prior to PCI, if a second oral anti-
platelet agent has not been chosen. It is preferable to use an oral
P2Y12 inhibitor instead of a GP IIb/IIIa antagonist unless there
is a strong possibility that the patient will need urgent cardiac
surgery in which case an intravenous GP IIb/IIIa antagonist with
a short half -life may be preferable to an oral P2Y12 inhibitor with
a long half-life, in order to minimize perioperative bleeding.
The use of GP IIb/IIIa antagonist drugs during PCI should
be reserved for those at highest risk, diabetic subjects, and those
with evidence of a significant intra coronary clot burden at the
time of PCI. The use of three anti platelet agents should only
continue for 12 to 18 hour following PCI, as the bleeding risks do
not justify continuation of therapy beyond this time.
Clinical Pearls
• Abciximab is reserved for use at the time of PCI
• Eptifibatide and tirofiban are best utilized only at the time of PCI
but can substitute for an oral P2Y12 inhibitor if there is a strong
possibility of immediate cardiac surgery.
• Use of a gGP IIb/IIIa antagonist outside of the catheterization
laboratory is associated with a significantly elevated risks of major
bleeding and is rarely justified when the clinical benefit is com-
pared with the bleeding risks of therapy.
Double-agent Antiplatelet Therapy Versus Triple
Antiplatelet Therapy. The ACC/AHA guidelines suggest
escalation to triple antiplatelet therapy at the time of PCI in the
highest-risk subset of patients with ACS. While there is always
a concern about bleeding and hemorrhagic complications with
623
an intravenous GP IIb/IIIa antagonist, such concerns should not
preclude their use in appropriate high risk patients including dia-
betic patients with ACS as well as in those who have a large
thrombotic burden prior to PCI.
Clinical Pearls
• TAPT should be reserved for only those at high risk and concur-
rently with PCI
• The use of an intravenous GP IIb/IIIa antagonist upstream, prior
to PCI is associated with an increased risk of major bleeding with-
out significant clinical benefit
Role and Timing of Invasive Therapy
The use of invasive evaluation and percutaneous coronary revas-
cularization improves outcome in ACS and reduces the risk
of recurrent ischemia in the majority of patients, especially in
patients who are risk-stratified as intermediate- or high-risk at
the time of presentation with ACS.
ACC/AHA guidelines recommend emergency angiography
for all hemodynamically unstable patients including patients with
cardiogenic shock, severe left ventricular dysfunction or pulmo-
nary edema, recurrent or persistent rest angina not responsive to
intensive medical therapy, and patients with recurrent sustained
ventricular tachycardia or ventricular fibrillation provided the
patients are potential candidates for PCI or CABG.
In hemodynamically stable patients there are two general
strategies for timing of invasive management of patients with
NSTEMI-ACS, and these strategies are based largely on patient
risk. Multiple large randomized trials (TIMACS, ABOARD,
EARLY ACS, ICTUS, FRISC II, TACTICS-TIMI 18, RITA 3)
showed a benefit to an early invasive strategy in NSTEMI-ACS
while two older randomized trials (TIMI 11B and VANQWISH)
found minimal to no benefit. The balance of the current scien-
tific data favors an early invasive strategy. We can summarize the
2011 ACC/AHA guidelines recommendations with the following
principles:
1. ICTUS and other studies support strongly the use of aggressive
medical stabilization in all patients prior to or on the way to the
catheterization laboratory. Medical stabilization includes potent anti-
thrombotic therapy, dual oral anti-platelet therapy and appropriate
adjunctive medical therapy. Table 65.5 highlights these therapies.
2. The early invasive strategy is the recommended approach to treat-
ment of patients at high or intermediate risk. Patients treated with
this strategy should generally undergo coronary angiography within
24–48 hours with angiographically directed revascularization, PCI
or CABG as needed. These patients often benefit from administra-
tion of a P2Y12 antagonist prior to or at the time of angiography and
those at highest risk also benefit from use of a GP IIb/IIIa inhibitor
at the time of percutaneous coronary revascularization.
3. The timing of revascularization has been carefully studied in the
ABOARD and TIMACS trials with regard to whether invasive ther-
apy can occur mainly during normal working hours or should be
escalated to immediate care similar to STEMI.
The ABOARD trial evaluated whether randomization to immedi-
ate catheterization and PCI would reduce troponin biomarker release
associated with the ACS. It was a small study and did not demon-
strate any benefit to immediate revascularization. The TIMACS trial
evaluated over 3,000 patients with ACS and demonstrated no advan-
tage to immediate invasive therapy with regard to the combined end
point of death, recurrent AMI and stroke. The TIMACS trial did
demonstrate a benefit for immediate invasive treatment with regard
to a secondary end point which included death, MI and refractory
ischemia. The TIMACS trial also demonstrated a benefit to immedi-
ate invasive therapy in those at highest risk, based upon the GRACE
risk score at baseline.
624 VII Myocardial Infarction

Table 65.5. Drug Therapy in NSTE-ACS

Drug Action Dose Use in NSTE-ACS
Aspirin Cyclooxygenase-1 inhibitor 325 mg chewed, then 81–162 mg Start immediately and continue
daily indefinitely
ADP inhibitors ADP-receptor blocker Clopidogrel—300 mg PO loading Start immediately and continue for
Clopidogrel dose, then 75 mg daily up to 9 mo in medically and PCI-
Ticlopidine manage patients
Hold if catheterization within 24–48 h
or CABG possible
GP IIB/IIIa inhibitors Inhibits interaction of fibrinogen Example: eptifibatide—180 μg/ Start at admission or at PCI in patients
Abciximab with the GP IIb/IIIa receptor kg, then continuous IV infusion who may need revascularization
Eptifibatide of 2 μg/kg per minute (up to 72
Tirofiban h); ↓ infusion to 1 μg/kg per
minute if serum Ct >2 mg/dL
UFH Binds to and enhances the activity 60 units/kg bolus, then 12 units/ Start immediately and continue for
of antithrombin III kg per hour target to aPTT of 2–7 d as clinically indicated
1.5–2.5
LMWH Factor Ca inhibitor 1 mg/kg q 12 h SC with Ct Cl Alternative to UFH; start immediately
Enoxaparin >30 mL/min and continue for 2–7 d as clinically
indicated
β-Blockers Decrease cardiac workload (↓ HR, Example: metoprolol 5 mg IV q Start immediately and continue
Cardioselective agents ↓ BP, ↓ contractility) 5 min × 3 doses, then 50 mg po indefinitely
preferred (atenolol, BID, first dose 15–20 min after
esmolol, metoprolol) last IV dose
Nitrates Decrease cardiac workload (↓ 0.04 mg SL q 5 min × 3 doses; if Start immediately and continue with
preload through venodilation) pain persists, 5–100 μg/min IV IV infusion if pain persists
Morphine sulfate Venodilation decreases preload; 2–4 mg IV q 5–15 min (should not Start after β-blocker and nitrates have
opioid analgesic exceed 25 mg in 24 h) been used and pain persists, or
sooner if anxiety present
Abbreviations: ADP, adenosine diphosphate; aPTT, activated partial thromboplastin time; BID, twice daily; BP, blood pressure; CABG, coronary artery bypass
grafting; Cl, clearance; Ct, creatinine; GP, glycoprotein; HR, heart rate; IV, intravenous; LMWH, low-molecular-weight heparin; NSTE-ACS, non–ST-segment
elevation acute coronary syndrome; PCI, percutaneous coronary intervention; PO, per os; q, every; SC, subcutaneous; SL, sublingual; UFH, unfractionated heparin.

4. It is appropriate to use a conservative strategy in those at low risk
after initial risk stratification. Such a strategy does not preclude ulti-
mate invasive evaluation but instead relies on noninvasive evaluation
after a period of observation, with catheterization and revasculariza-
tion reserved for patients with evidence of recurrent ischemia at rest
or with provocative stress testing. Low-risk ACS patients should be
treated with aspirin, β-blockers, heparin, nitrates and either clopi-
dogrel or ticagrelor. In this strategy, GP IIb/IIIa inhibitors should
not be routinely used except where a previously stable low-risk ACS
patient has ongoing ischemia and now become high-risk, in whom
an immediate transfer to the catheterization laboratory and probable
PCI is planned.
Indications for CABG in ACS. Most patients who present with
NSTEMI-ACS undergo coronary angiography. Among patients
with NSTEMI-ACS who undergo coronary angiography, 10% to
20% have normal or insignificant CAD, 5% to 10% have signifi-
cant left main CAD, 20% to 25% have three-vessel disease, 25%
to 30% have two-vessel disease, and 30% to 35% have single-
vessel disease. Patients with significant coronary artery stenosis
(ie, >70% stenosis of the left anterior descending, circumflex, or
right coronary artery or more than 50% stenosis of the left main
coronary artery) are candidates for revascularization.
If cardiac catheterization shows significant left main CAD or
three-vessel disease with reduced left ventricular function, the
patient should be considered for CABG. In general, patients with
proximal two-vessel disease (particularly proximal left anterior
descending artery disease or with diabetes mellitus), and reduced
left ventricular function should be considered for CABG. PCI-
based revascularization is also an option for patients with com-
plex CAD including left main stenosis.
SYNTAX Trial. The SYNTAX trial was a large, randomized-
control trial of 1,800 stable angina patients with severe coronary
disease (three-vessel or left main disease) assigned to either
CABG or PCI with a drug-eluting stent. At 12 months of fol-
low-up the composite primary end point (all death, stroke, MI,
or repeat revascularization) was much higher in the PCI group
(17.8% vs 12.4%), driven primarily by the need for more frequent
revascularization with PCI. The stroke rate was significantly
higher with CABG group (2.2% vs 0.6%), but CABG provided
more complete myocardial revascularization and a small but sta-
tistically meaningful decrease in anginal frequency compared
with PCI.
• The 2011 ACC/AHA guidelines recommend coronary angiography
in NSTEMI-ACS patients who have new ST-segment depression,
a troponin elevation, recurrent chest pain, left ventricular dysfunc-
tion, or other high-risk features.
• Generally, CABG is favored over PCI in patients with left ven-
tricular systolic dysfunction, severe left main CAD or three-vessel
disease, two-vessel disease with severe proximal stenosis of the
LAD, or diabetes.
Step 4: Late Risk Stratification and Long-Term
Follow-up in the ACS Patient Population
Low- to intermediate-risk patients (by ACC/AHA criteria) who
have been treated with an early conservative strategy and have
not had recurrent ischemia may undergo stress testing for further
risk stratification. In most cases, testing should be done within
72 hours after presentation. The choice of the stress test depends
 65 Acute Coronary Syndromes 625

Table 65.6. Guidelines for Stress Method in Risk Assessmenta

Dobutamine
Exercisea Dipyridamoleb Adenosineb (Thallium, Sestamibi,
Patient Factors (Treadmill, Cycle, Arm) (Thallium, Sestamibi) (Thallium, Sestamibi) RNA, Echocardiography)
Carotid bruits
Without symptoms Yes Yes Yes Yes
With recent symptoms Yes Noc Yesd Noc
Lung disease
Mild/moderate COPD Yes Yes Yes Yes
Severe COPD/asthma Yes No No Yes
Theophylline Yes No No Yes
LBBB No Yes Yes No
β-Blockers Yes Yes Yes No
Dipyridamole Yes Yes No Yes
PPM No Yes Yes No
Poorly controlled HTM No Yes Yes No
Significant ventricular ectopy Yes Yes Yes No
Abbreviations: COPD, chronic obstructive pulmonary disease; HTN, hypertension; LBBB, left bundle branch block; PPM, non–rate-responsive
permanent pacemaker; RNA, radionuclide angiography.
a
If the patient has a normal electrocardiogram and can walk, use exercise treadmill. If possible, exercise the patient to ≥5 metabolic equivalents or chest
discomfort.
b
Withhold caffeine for 12 hours.
c
Because of possible high or low blood pressure response.
d
Graduated influsion, with blood pressure monitoring suggested.

on the patient’s resting ECG, the ability of the patient to perform
exercise, and the available methods and local expertise.
Exercise treadmill testing is the standard mode of stress test-
ing in patients with a normal ECG who are able to exercise.
Conditions precluding accurate interpretation of the stress ECG
include digoxin therapy, widespread resting ST-segment depres-
sion (≥1 mm), left ventricular hypertrophy, LBBB, significant
interventricular conduction delay, and preexcitation syndrome.
In patients with these conditions, an imaging modality such as
a radionuclide imaging or exercise echocardiography should be
considered (Table 65.6). Patients who cannot exercise because
of general debility, chronic obstructive pulmonary disease,
peripheral vascular disease, or orthopedic limitations should
undergo pharmacologic stress testing with an imaging modality
(Table 65.7).
Patients who are able to exercise to a reasonable workload
(ie, ≥5 METs) without demonstrable ischemia have a good prog-
nosis and may continue to be managed medically. Patients who
have evidence of ischemia at a low workload (ie, <5 METs), on
the basis of clinical symptoms, ECG changes, or imaging abnor-
malities, should be considered for coronary angiography.
Patients at low risk on exercise stress testing have a pre-
dicted average cardiac mortality of less than 1% per year,
compared with at least 4% per year for those at high risk. All
forms of exercise testing are less accurate in women than in
men; however, it is still reasonable to use noninvasive testing

Table 65.7. Guidelines for Imaging Modality in Risk Assessmenta

Feature Treadmill Only Echocardiography Thallium Sestamibi RNA
Goal of test
EF No Yes No Yes Yes
Screening Yes Yes Yes Yes Yes
Low cost Yes No No No No
Post-MI viability Yes Yes Yes Yes Yes
Patient factors
Large chest Yes No Yes Yes Yes
Obese Yes Yes No Yes Yes
COPD Yes No Yes Yes No
ECG factors
LBB No Yes Yes Yes Yes
Nonspecific ST-T–wave changes due to No No Yes Yes No
digoxin, WPW, MVP, LVH, PPM
Irregular rhythm (AF, frequent PVCs) No Yes Yes Yes No
Cannot exercise No Yes Yes Yes No
Abbreviations: AF, atrial fibrillation; COPD, chronic obstructive pulmonary disease; ECG, electrocardiographic; EF, ejection
fraction; LBBB, left bundle branch block; LVH, left ventricular hypertrophy; MI, myocardial infarction; MVP, mitral valve
prolapse; PPM, non-rate-responsive permanent pacemaker; PVC, premature ventricular contraction; RNA, radionuclide
angiography; WPW, Wolff-Parkinson-White syndrome.
a
Local expertise and availability are extremely important in choosing imaging modality.
626 VII Myocardial Infarction

in women for risk stratification in the early conservative treat- TIA transient ischemic attack
ment strategy. UA unstable angina
It is important to measure plasma lipids and left ventricu-
lar function in patients with ACS prior to hospital discharge. Names of Clinical Trials
Patients with evidence of systolic dysfunction should be re-
evaluated for reversible causes of reduced systolic function and ABOARD Angioplasty to Blunt the Rise of Troponin in
Acute Coronary Syndromes Randomized for
should undergo PCI or surgical revascularization as appropri-
an Immediate or Delayed Intervention
ate. Patients with left ventricular ejection fractions persistently EARLY ACS Early Glycoprotein IIb/IIIa Inhibition in
below 30% to 35% at 6 weeks after discharge are candidates for Non–ST-Segment Elevation Acute Coronary
ICD implantation. Syndromes
ESSENCE Efficacy and Safety of Subcutaneous
Enoxaparin in Unstable Angina and Non-
Clinical Pearls
Q-Wave MI
• Patients treated with the early conservative strategy should have FRISC II FRagmin and Fast Revascularization during
stress testing done before hospital discharge. InStability in coronary artery disease
GRACE Global Registry of Acute Coronary Events
• An exercise capacity of ≥5 METs without ischemia indicates a
GUSTO Global Utilization of Streptokinase and
good long term prognosis.
Tissue Plasminogen Activator for Occluded
• Low-risk patients have <1% annual cardiac mortality. Coronary Arteries
• High-risk patients have ≥4% annual cardiac mortality. ICTUS Invasive Versus Conservative Treatment in
• Measure left ventricular ejection fraction prior to hospital dis- Unstable Coronary Syndromes
missal in all patients PLATO Ticagrelor Compared with Clopidogrel by
Geographic Region in the Platelet Inhibition
• Refer all patients to a cardiac rehabilitation program more than and Patient Outcomes
40 days after MI PRISM-PLUS Platelet Receptor Inhibition in Ischemic
• Consider referral for ICD therapy in patients with an EF < 30%–35% Syndrome Management in Patients Limited
by Unstable Signs and Symptoms
PURSUIT Platelet Glycoprotein IIb/IIIa in Unstable
Abbreviations Angina: Receptor Suppression Using
ACE angiotensin-converting enzyme Integrilin Therapy
ACC American College of Cardiology RITA-3 Randomized trial of a Conservative Treatment
ACS acute coronary syndrome Strategy Versus an Interventional Treatment
AMI acute myocardial infarction Strategy in Patients With Unstable Angina
CABG coronary artery bypass grafting SYNTAX Synergy Between Percutaneous Coronary
CAD coronary artery disease Intervention With Taxus and Cardiac
CCS Canadian Cardiovascular Society Surgery
CK-MB creatine kinase MB TACTICS-TIMI 18 Treat angina with Aggrastat and deter-
CV cardiovascular mine Cost of Therapy with an Invasive or
ECG electrocardiographic, electrocardiography Conservative Strategy-Thrombolysis in
ESC European Society of Cardiology Myocardial Infarction 18
FDA US Food and Drug Administration TIMACS Timing of Intervention in Patients With Acute
GP glycoprotein Coronary Syndromes
LBBB left bundle branch block TIMI Thrombolysis in Myocardial Infarction
MET metabolic equivalent TIMI-11B Thrombolysis in Myocardial Infarction
MI myocardial infarction Phase 11B
NSTEMI non–ST-segment elevation myocardial infarction TRITON TIMI 38 Trial to Assess Improvement in Therapeutic
PCI percutaneous coronary intervention Outcomes by Optimizing Platelet Inhibition
RPG random plasma glucose with Prasugrel
SCD sudden cardiac death VANQWISH Veterans Affairs (VA) Non-Q-Wave Infarction
STEMI ST-segment elevation myocardial infarction Strategies In-Hospital
 66
Chronic Stable Angina
KALKIDAN G. BISHU, MD, and FRANK V. BROZOVICH, MD, PHD
Introduction
An estimated 16 million Americans over 20 years of age have
CAD. The lifetime risk of CAD after the age of 40 years is 49%
for men and 32% for women. It is associated with high morbid-
ity and mortality, causing 1 out of every 6 deaths in the United
States in 2007. Half of the patients with CAD present with
chronic stable angina. William Heberden gave 1 of the earliest
descriptions of angina pectoris: “They who are afflicted with it
are seized while they are walking with a painful and most dis-
agreeable sensation in the breast.”
Signs and Symptoms
Character
Angina pectoris typically manifests as a retrosternal discomfort
in the chest and adjacent areas and can vary from being “con-
stricting,” “crushing,” or “squeezing” to mild and pressure-like,
with an occasional associated numbing or burning sensation.
The discomfort often radiates down the ulnar surface of the left
arm, but it can also radiate on the outer surfaces of both arms.
Belching or epigastric discomfort in conjunction with chest pain
is not uncommon. Other associated symptoms such as dyspnea,
fatigue, and syncope are particularly common in the elderly.
Duration
Typical angina is usually characterized by a crescendo increase
in the intensity of pain over minutes, neither lasting hours nor
occurring in brief spurts of only a few seconds. It is often precipi-
tated by exercise, exposure to cold, or carbohydrate-rich meals
Abbreviations and acronyms are expanded at the end of this chapter.
and relieved within minutes by rest or nitroglycerin (or both). In
some patients, exertion with intermittent periods of rest can lead
to greater levels of exertion without symptoms. This warm-up
phenomenon is hypothesized to result from either ischemic pre-
conditioning or recruitment of coronary collateral circulation. If
the character of anginal symptoms is unchanged in 60 days, the
angina is considered stable.
If the pain is not relieved within 5 to 10 minutes after rest or
nitroglycerin use, it is unlikely to be secondary to myocardial
ischemia. However, for prolonged chest pain, alternative diagno-
ses must be considered, including severe ongoing ischemia, such
as in the case of unstable angina or acute MI.
Grading of Severity
Grading of angina according to the CCS is a modification of
the NYHA functional classification and has since been widely
accepted because the CCS classification allows for more spe-
cific categorization of patients according to their activity level
(Table 66.1). Other classification schemes include the Specific
Activity Scale, developed by Goldman and colleagues, and the
anginal score by Califf and associates. Any functional classifica-
tion is subject to variability in activity tolerance as perceived by
patients, and hence its reproducibility is variable. Reproducibility
reaches 73% with either the CCS criteria or the Specific Activity
Scale, but the latter seems to correlate better with objective
measures by treadmill exercise.
Physical Examination
Brief episodes of angina can result in transient left ventricular
and papillary muscle dysfunction, characterized by a third heart
sound, a loud fourth heart sound, pulmonary rales, and apical
627
628 VII Myocardial Infarction

Table 66.1. Three Methods of Assessing Cardiovascular Disability

New York Heart Association Functional
Class Classification
I Patients with cardiac disease but without
resulting limitations of physical activity
Ordinary physical activity does not cause
undue fatigue, palpitation, dyspnea, or
anginal pain
II Patients with cardiac disease resulting in
slight limitation of physical activity
Patients are comfortable at rest
Ordinary physical activity results in fatigue,
palpitation, dyspnea, or anginal pain
III Patients with cardiac disease resulting in
marked limitation of physical activity
Patients are comfortable at rest
Less than ordinary physical activity causes
fatigue, palpitation, dyspnea, or anginal
pain
IV Patients with cardiac disease resulting in
inability to carry on any physical activity
without discomfort
Symptoms of cardiac insufficiency or of the
anginal syndrome may be present even
at rest
If any physical activity is undertaken,
discomfort is increased
Abbreviation: MET, metabolic equivalent task.
Previously published. See “Credit Lines” section.
Canadian Cardiovascular Society
Classification
Ordinary physical activity (eg, walking,
climbing stairs) does not cause angina
Angina with strenuous, rapid, or prolonged
exertion at work or recreation
Slight limitation of ordinary activity
Angina with walking or climbing stairs
rapidly; walking uphill; walking or stair
climbing after meals, in cold, in wind,
under emotional stress, or only during the
few hours after awakening; walking >2
blocks on the level; and climbing >1 flight
of ordinary stairs at a normal pace and in
normal conditions
Marked limitation of ordinary physical
activity
Angina with walking 1–2 blocks on the
level and climbing >1 flight in normal
conditions
Inability to carry on any physical activity
without discomfort—anginal syndrome
may be present at rest
Specific Activity Scale
Patients can perform to completion any activity
requiring ≥7 METs (eg, carry 24 lb up
8 steps; carry objects that weigh 80 lb; do
outdoor work [shovel snow, spade soil]; do
recreational activities [ski; play basketball,
squash, or handball; jog or walk 5 mph])
Patients can perform to completion any
activity requiring ≥5 METs (eg, have sexual
intercourse without stopping, garden, rake,
weed, roller skate, dance, fox trot, walk at 4
mph on level ground) but cannot and do not
perform to completion activities requiring
≥7 METs
Patients can perform to completion any activity
requiring ≥2 METs (eg, shower without
stopping, strip and make bed, clean windows,
walk 2.5 mph, bowl, play golf, dress without
stopping) but cannot and do not perform to
completion any activities requiring ≥5 METs
Patients cannot or do not perform to completion
activities requiring ≥2 METs
Cannot carry out activities listed above (Specific
Activity Scale class III)

systolic murmurs. A displaced ventricular impulse is a sign of
a dyskinetic left ventricle, and ischemia can give rise to delayed
left ventricular contraction, resulting in the paradoxic split of a
second heart sound. A midsystolic click with late systolic mur-
mur, as in mitral valve prolapse, can occur in patients with CAD.
On peripheral vascular examination, any evidence of peripheral
vascular disease, such as a decreased ankle-brachial index or
early carotid disease on ultrasonography, is strongly associated
with CAD. It should be understood, however, that patients with
angina often have normal physical examination findings.
both sympathetic and vagal afferent fibers. Sympathetic affer-
ent impulses converge with somatic sensory fibers from tho-
racic structures and travel to the thalamus and frontal cortex.
Sympathetic activation is responsible for the perception of
referred cardiac pain. Failed transmission of afferent impulses
from the thalamus to the frontal cortex on positron emission
tomographic scan has been postulated to cause silent ischemia
in patients with autonomic neuropathy. Vagal afferent fibers syn-
apse in the medulla and innervate the upper cervical spinotha-
lamic tract, which gives rise to pain in the neck and jaw.

Differential Diagnoses Pathogenesis of Angina Pectoris

All the following conditions may cause pain that mimics the
symptoms of chronic angina: acute MI; aortic dissection; pul-
monary hypertension with right ventricular ischemia; pulmonary
embolism; acute pericarditis; gastroesophageal reflux and disor-
ders of esophageal motility, including diffuse spasm, nutcracker
esophagus, and achalasia; biliary disorders; cholecystitis; costo-
sternal chondritis or Tietze syndrome; cervical radiculitis; and
shoulder bursitis or tendonitis.
Pathophysiology
Neuromechanisms of Cardiac Pain
Myocardial ischemia leads to the activation of chemorecep-
tors and mechanoreceptors, which in turn causes the release
of bradykinin and adenosine. These neurochemicals stimulate
Myocardial ischemia can be due to either increased myocardial
oxygen requirements or decreased myocardial oxygen supply.
Angina precipitated by increased myocardial oxygen requirements
is sometimes termed demand angina or fixed-threshold angina,
whereas angina secondary to a transiently decreased oxygen supply
is sometimes termed supply angina or variable-threshold angina.
Angina Due to Increased Myocardial
Oxygen Requirement
In demand angina, the physiologic responses to physical exer-
tion, mental or emotional stresses, and conditions such as fever,
hypoglycemia, and thyrotoxicosis trigger the release of norepi-
nephrine, which increases myocardial oxygen requirements.
Hence, in demand angina with few dynamic (vasoconstrictor)
 66 Chronic Stable Angina
components, the amount of physical activity that precipitates
angina is relatively constant.
Angina Due to Decreased Myocardial Oxygen Supply
In supply angina, similar to unstable angina, dynamic stenosis
can further decrease the myocardial oxygen supply in the pres-
ence of a fixed organic stenosis. Platelets and leukocytes can
elaborate vasoconstrictors such as thromboxane A2 and sero-
tonin, which, along with the already damaged endothelium and
decreased nitric oxide production secondary to coronary athero-
sclerosis, promote vasoconstriction. Typically, organic coronary
luminal stenosis of less than 70% of the cross-sectional area does
not cause ischemia or angina since the microvasculature can
dilate to increase perfusion. However, if the stenosis is more than
70%, maximally dilated microvasculature does not improve per-
fusion. CAD is typically associated with impaired endothelium-
dependent vasodilation. This dynamic component superimposed
on fixed organic obstruction contributes to variable threshold
angina. Other features of supply angina include circadian varia-
tion, in which angina occurs more often in the morning, and cold
temperature–induced coronary vasoconstriction. A high-car-
bohydrate diet can redistribute coronary blood flow away from
stenotic vessels and precipitate postprandial angina.
• The warm-up phenomenon is hypothesized to result from either
ischemic preconditioning or recruitment of coronary collateral
circulation.
• Angina precipitated by increased myocardial oxygen requirements
is termed demand angina or fixed-threshold angina.
•
Angina secondary to a transiently decreased oxygen supply is
termed supply angina or variable-threshold angina.
Diagnostic Testing
Noninvasive Testing
Biochemical Tests
Initial biochemical evaluation should include a fasting lipid
profile and a fasting blood glucose determination to screen for
dyslipidemias and insulin resistance. Other markers of increased
atherogenicity, such as Lp(a) lipoprotein, small dense LDL, and
apolipoprotein B, appear to add to traditional tests of total cho-
lesterol and LDL in better predicting the future cardiovascular
event risk. Measurement of high-sensitivity CRP can be used
in secondary prevention to prognosticate patients with estab-
lished CAD. Levels of BNP and the inactive NT-proBNP have
been shown to be strong predictors of morbidity and mortality
among patients with heart failure and acute coronary syndromes.
Two prospective observational studies from Denmark and
Germany showed that in patients with chronic stable angina, an
NT-proBNP level in a higher quartile correlated with decreased
survival after adjusting for conventional risk factors (age, pres-
ence or absence of diabetes mellitus, smoking status, left ven-
tricular ejection fraction, presence or absence of suspected heart
failure, and severity of angiographic coronary artery disease)
(Figure 66.1). Another study from Australia and New Zealand
showed that in patients with stable CAD (NYHA class II or III),
BNP and NT-proBNP showed a strong negative predictive value
in ruling out severely reduced left ventricular ejection fraction
and directly correlated with all-cause mortality or worsening
heart failure after 1-year follow-up. Taken together, these data
suggest that BNP and NT-proBNP offer additional prognostic
information for patients with stable CAD.
629
100
75
Survival, %
50
First quartile
25 Second quartile
Third quartile
Fourth quartile
0
0 2 4 6 8 10 12
Years
Figure 66.1. Adjusted estimates of overall survival among patients
with stable coronary artery disease, according to quartiles of levels of
the N-terminal fragment of brain natriuretic peptide (NT-proBNP). The
survival estimates have been adjusted for age, presence or absence of
diabetes mellitus, smoking status, left ventricular ejection fraction, pres-
ence or absence of suspected heart failure, and severity of angiographic
coronary artery disease. The NT-proBNP levels were as follows: first
quartile, <64 pg/mL; second quartile, 64–169 pg/mL; third quartile,
170–455 pg/mL; and fourth quartile, >455 pg/mL (log-rank test for the
overall comparison among the groups, P<.001). (Previously published.
See “Credit Lines” section.)
Resting Electrocardiogram
A resting 12-lead ECG should be obtained for all patients with
suspected angina. However, approximately half of the patients
with chronic stable angina have a normal resting ECG. The most
common ECG findings in chronic CAD include nonspecific ST-T
wave changes with or without abnormal Q waves. Q waves are
relatively specific but not sensitive indicators for previous MI.
During an anginal episode, the ECG becomes abnormal in about
50% of the patients with a normal baseline ECG.
Noninvasive Stress Testing
Noninvasive stress testing is used for diagnosis and to further
assess prognosis during the initial evaluation of patients with
suspected angina. However, appropriate stress testing should be
directed on the basis of the estimate of the pretest probability
(prevalence of disease) of CAD in the particular patient popula-
tion (Table 66.2). Diagnostic stress testing is valuable when the
pretest probability of CAD is intermediate (10%–90%). Patients
with a high pretest probability have high false-negative rates and
those with low pretest probability have high false-positive test
results. Stress testing modalities vary according to the type of
stressor used (exercise, dobutamine, or vasodilators such as dipyr-
idamole or adenosine) and the method used to detect ischemia
(ECG changes, perfusion defects on myocardial perfusion imag-
ing, and wall motion abnormalities on echocardiography).
1. Exercise ECG—For patients who have a moderate probability of
CAD and angina, exercise ECG is particularly helpful, provided
they have a normal resting ECG and achieve an adequate workload.
According to the ACC/AHA guidelines on exercise testing, mean
sensitivity is 68% and specificity is 77%. If one corrects for the post-
test referral bias, however, sensitivity decreases further to 45% to
50%, but specificity increases to 85% to 90%. In part, the low sen-
sitivity of exercise testing in this population includes the inability of
630 VII Myocardial Infarction
Table 66.2. Pretest Likelihood of Coronary Artery Disease in Symptomatic
Patients According to Age and Sexa
Nonanginal Chest Pain Atypical Angina
Age, y Men Women Men
30–39 4 2 34
40–49 13 3 51
50–59 20 7 65
60–69 27 14 72
a
Each value is the percentage of patients who have significant coronary artery disease at
coronary angiography.
Previously published. See “Credit Lines” section.
many patients to reach at least 85% of the maximally predicted heart
rate, either because of physical inability or the effects of antianginal
medication.
2. Stress myocardial imaging—For patients who have abnormal rest-
ing ECGs (eg, from repolarization abnormalities, left ventricular
hypertrophy, electronically paced ventricular rhythm, preexcitation
syndromes, left bundle branch block, or digitalis effect), exercise
perfusion imaging is superior to exercise ECG alone in localizing
diseased vessels, identifying multivessel disease, and detecting
areas of ischemia or infarction. Pharmacologic vasodilators such
as adenosine or dipyridamole can be used in elderly patients who
cannot exercise because of comorbid pulmonary or peripheral vas-
cular diseases. The accuracy of exercise echocardiography is simi-
lar to that of stress myocardial perfusion imaging. Newer imaging
modalities such as contrast echocardiography and harmonic imag-
ing have significantly improved endocardial border definition, which
enhances the visualization of ischemic myocardium.
Invasive Testing
Coronary Angiography
Cardiac catheterization and coronary angiography are needed to
definitively diagnose CAD and precisely evaluate the severity and
anatomical distribution of the disease. Coronary angiography is
recommended for risk stratification for patients who have angina
that interferes with their lifestyle despite maximal tolerable medi-
cal therapy, for patients considered high risk after noninvasive
testing, and for patients who have angina and signs and symptoms
of heart failure. If clinical symptoms suggest angina in a patient
with a high pretest probability of severe CAD, a direct referral for
coronary angiography may be cost-effective. Among patients who
have chronic stable angina and are referred for coronary angiogra-
phy, about 25% each have single-, double-, or triple-vessel disease
(>70% luminal diameter narrowing); 5% to 10% have obstruction
of the left main coronary artery; and 15% do not have detectable
critical obstruction. Intravascular ultrasonography allows assess-
ment of the cross-luminal dimensions and determination of the
plaque composition. Angiography also allows visualization of
coronary collateral circulation, coronary artery ectasia or aneu-
rysms, and myocardial bridging and assessment of left ventricular
function with biplane contrast angiography.
• Levels of BNP and the inactive NT-proBNP are strong predictors
of morbidity and mortality among patients with heart failure and
acute coronary syndromes.
• Diagnostic stress testing is valuable when the pretest probability of
CAD is intermediate (10%–90%).
• If clinical symptoms suggest angina in a patient with a high pretest
probability of severe CAD, a direct referral for coronary angiogra-
phy may be cost-effective.
Typical Angina
Women Men Women
12 76 26
22 87 55
31 93 73
51 94 86
Prognosis and Risk Stratification
Older data from the Framingham Study demonstrated that the
annual average mortality among patients with chronic stable
angina pectoris was 4%. With the advent of pharmacotherapy
with drugs such as aspirin and β-blockers and aggressive modi-
fication of risk factors, the annual mortality rate for middle-aged
men with CAD has decreased to 1.7% to 3%. Left ventricu-
lar function is the strongest predictor of long-term survival in
CAD. The second strongest predictor of long-term outcomes is
the anatomical extent and severity of coronary artery involve-
ment. To predict the presence of severe CAD (triple-vessel or
left main CAD), a 5-point clinical score is assigned according
to the following clinical variables: male sex, typical angina,
history and ECG evidence of MI, diabetes mellitus, and use of
insulin (Figure 66.2). For patients who have CAD and receive
medical therapy, the 5-year survival is based on the number of
diseased vessels and the severity and location of the obstruction
(Figure 66.3).
Therapy
The goals of treatment of chronic stable angina are 1) to alleviate
symptoms and ischemia, 2) to delay or prevent the progression
of CAD, and 3) to prevent MI and death. Aggressive medical
management that includes risk factor reduction should gener-
ally be the cornerstone of treatment of chronic stable angina.
Myocardial revascularization is considered for patients who
remain symptomatic despite maximal medical therapy and for
those who have increased long-term risk if they receive medical
therapy alone.
Medical Management
Treatment of Associated Diseases
Conditions that can exacerbate previously stable angina include
anemia, thyrotoxicosis, fever, infection, tachycardia, and use of
drugs that activate the sympathetic nervous system. Treating
these conditions will reduce myocardial oxygen demand and
increase oxygen delivery, thus alleviating anginal symptoms.
Reduction of Coronary Risk Factors
1. Hypertension—The risk of ischemic heart disease doubles for every
20–mm Hg increase (range, 115–185 mm Hg) among persons aged
40 to 70. Pharmacologic treatment of mild or moderate hypertension
results in a 16% reduction in CAD events. β-Blockers in combina-
tion with ACE inhibitors or ARBs should be considered initially,
with calcium channel blockers added if blood pressure is still not
 66 Chronic Stable Angina 631

1.0 3. Tobacco smoking—Smoking increases myocardial oxygen demand,

Score causes coronary vasospasm, and decreases the efficacy of antiangi-
0 3
Predicted Probability

1 4 nal drugs. Among patients with documented CAD, smokers have a

0.8 higher 5-year risk of sudden death, MI, and all-cause mortality than
2 5
those who have quit smoking. Hence, smoking cessation is one of
0.6 the most powerful and cost-effective approaches in the prevention of
CAD progression.
4. Increased levels of LDL cholesterol—Clinical trials, including 4S,
0.4 CARE, LIPID, SAGE, and HPS, have convincingly shown that in
patients with atherosclerotic vascular disease, statins significantly
0.2 reduce subsequent cardiovascular events (eg, cardiovascular mortal-
ity, rate of MI, or the need for a CABG). Statins have also been
shown to improve endothelial function, lower the circulating level of
0.0 CRP, reduce thrombogenicity, and stabilize atherosclerotic plaques.
30 35 40 45 50 55 60 65 70 75 80 The ASTEROID and REVERSAL trials showed that high-intensity
statin therapy led to regression of atherosclerotic disease in patients
Age, y with established CAD. Taken together, these data provide the basis
Figure 66.2. Probability of severe (triple-vessel or left main) coro- for more aggressive cholesterol-lowering therapy among patients
nary artery disease based on a 5-point clinical score assigned on the with established CAD. According to the 2007 ACC/AHA guideline
basis of the following clinical variables: male sex, typical angina, update, LDL levels should be less than 100 mg/dL and a value less
history and electrocardiographic evidence of myocardial infarction, than 70 mg/dL is reasonable in very high-risk patients (diabetes mel-
diabetes mellitus, and use of insulin. (Previously published. See “Credit litus, continued smoking, metabolic syndrome, acute coronary syn-
Lines” section.) drome). Decreasing LDL cholesterol levels to target levels should
always be the primary goal of lipid-lowering treatment. However,
ezetimibe, a new agent that reduces intestinal absorption of cho-
controlled or if β-blockers are not tolerated. Thiazide diuretics can lesterol and further decreases LDL levels when given with statins
be added for blood pressure control. A target blood pressure of less has yet to show additional benefit in reducing definitive clinical
than 140/90 mm Hg in patients with hypertension and chronic stable outcomes.
angina is reasonable. 5. High TG levels or low levels of HDL cholesterol (or both)—In add-
2. Diabetes mellitus—Blood glucose levels should be controlled to ition to diet and exercise, pharmacotherapy with niacin or fibrates
maintain hemoglobin A1c levels at about 7% or less for long-term may decrease TG and increase HDL levels. Among patients with
reduction of microvascular and macrovascular complications in all TG levels greater than 200 mg/dL, non-HDL cholesterol levels
adults with diabetes mellitus. However, less stringent goals may be should be less than 130 mg/dL, and further reduction to less than
necessary for patients with a history of severe hypoglycemia, lim- 100 mg/dL is reasonable. The CDP study showed a 27% reduction
ited life expectancy, and advanced microvascular or macrovascular in nonfatal reinfarction at 5 years and a 9% reduction at 15 years
complications. with niacin treatment. In a study of 208 CAD patients who were

1 Vessel 93

2 Vessels 88

1 Vessel, >95% in proximal LAD 83

Extent of CAD

2 Vessels, >95% in proximal LAD 79

3 Vessels 79

3 Vessels, >95% in at least 1 73

3 Vessels, 75% in proximal LAD 67

3 Vessels, >95% in proximal LAD 59

0 20 40 60 80 100

Percent
Figure 66.3. Angiographic extent of coronary artery disease (CAD) and subsequent 5-year survival with medical therapy. A gradient of mortal-
ity risk is based on the number of diseased vessels and the presence and severity of disease of the proximal left anterior descending coronary artery
(LAD). (Previously published. See “Credit Lines” section.)
632 VII Myocardial Infarction
receiving statin therapy, both niacin and ezetimibe decreased serum
LDL levels, but only niacin was associated with increased HDL lev-
els, decreased TG, regression of carotid intima media thickness, and
fewer major cardiovascular events at 1 year. A common side effect
with niacin treatment is facial flushing. In the VA-HIT study, gemfi-
brozil increased HDL levels by 6% and decreased the rate of death,
nonfatal MI, or stroke by 24%. Fibrates may cause muscle toxicity,
especially when given with statins. Novel agents currently under
study include cholesteryl ester transfer protein inhibitors.
6. Exercise—The conditioning effect of exercise lowers the heart rate
and increases the cardiac output at any given level of myocardial
oxygen consumption. Several small randomized trials have shown
improved effort tolerance, oxygen consumption, and quality of life
in patients who had chronic stable CAD and were undergoing exer-
cise training.
• Aggressive medical management that includes risk factor reduc-
tion should be the cornerstone of treatment in chronic stable
angina.
Vasculoprotective Agents to Prevent MI
Antiplatelet Agents
Aspirin. Meta-analysis of 140,000 patients from the Antiplatelet
Trialists’ Collaboration showed that aspirin (75–325 mg daily)
decreased the rate of subsequent MI, stroke, or death by 22%
among patients with a history of angina pectoris, MI, CABG sur-
gery, or stroke. In the SAPAT study, aspirin (75 mg daily) in con-
junction with the β-blocker sotalol conferred an additional 34%
decrease in the rates of acute MI and sudden death among men
and women with chronic stable angina. Aspirin also improves
endothelial function and, when used in high dose (300 mg daily),
decreases the circulating levels of CRP. For secondary preven-
tion, a dosage of 75 to 150 mg daily appears to be comparable
in efficacy to a dosage of 160 to 325 mg daily. The lower dosage
also reduces the bleeding risk.
Clopidogrel. In the CAPRIE trial, a randomized compar-
ison between clopidogrel and aspirin showed that clopidogrel
resulted in an 8.7% relative decrease in the risk of vascular
death, ischemic stroke, or MI among patients with established
atherosclerotic vascular disease. However, no data support the
use of dual antiplatelet therapy for primary prevention. In the
CHARISMA trial, dual antiplatelet therapy with clopidogrel and
aspirin was not significantly more effective than aspirin alone
in reducing the rate of MI, stroke, or cardiovascular death in
patients with established vascular disease or at high risk of vas-
cular disease. Clopidogrel may be an alternative for patients who
are allergic to aspirin.
Angiotensin-Converting Enzyme Inhibitors
ACE inhibitors reduce left ventricular hypertrophy, vascular
hypertrophy, atherosclerotic progression, plaque rupture, and
thrombotic risk. ACE inhibition also enhances coronary endo-
thelial vasomotor function and decreases inflammatory changes
in animal models of atherosclerosis. Thus, ACE inhibitors pro-
mote favorable myocardial oxygen supply-demand relationships
and lessen sympathetic activity. The HOPE trial showed that
the ACE inhibitor ramipril (10 mg daily) reduced the relative
risk of cardiovascular death, MI, or stroke in patients who had
atherosclerotic vascular disease or diabetes mellitus and at least
1 other CAD risk factor. The EUROPA study provided further
support for the beneficial effect of ACE inhibitors to decrease
subsequent cardiac events among patients who had stable CAD
without heart failure. However, in the PEACE trial, trandolapril
did not provide any mortality benefit among patients who had
stable CAD and preserved left ventricular function. A meta-
analysis of the HOPE, EUROPA, and PEACE trials showed
that ACE inhibitors reduced mortality, MI, stroke, heart failure,
and CABG rates. Therefore, according to the 2007 ACC/AHA
guideline update on the management of chronic stable angina,
ACE inhibitors should be given indefinitely to all chronic angina
patients who have impaired ventricular function, hyperten-
sion, diabetes mellitus, or chronic kidney disease. ACE inhibi-
tors should also be given to chronic stable angina patients who
have normal ventricular function and cardiovascular risk factors
that are not well controlled and in whom revascularization has
not been performed. If patients cannot tolerate ACE inhibitors,
ARBs may be used.
Antianginal and Anti-ischemic Therapy
Nitrates
Nitrates induce endothelium-independent vasodilatation. They
relax systemic arteries (including coronary arteries) and veins,
but the venodilator effect predominates. The venodilator effect
reduces ventricular preload, which in turn decreases the myo-
cardial wall tension and oxygen requirement. In the coronary
circulation, nitrates also dilate epicardial stenoses with eccen-
tric lesions and alleviate endothelial dysfunction, thus improv-
ing blood flow across obstructed regions. However, if patients
are also taking phosphodiesterase inhibitors, the use of nitrates
within 24 hours may result in severe hypotension.
Nitroglycerin Tablets. Sublingual nitroglycerin is the treat-
ment of choice for acute anginal episodes. Sublingual admin-
istration avoids first-pass hepatic metabolism. The half-life
of nitroglycerin is brief, and within 30 to 60 minutes hepatic
breakdown has abolished the hemodynamic and clinical effects.
Sublingual nitroglycerin, when taken prophylactically before
physical activities, can prevent angina for up to 40 minutes.
Nitroglycerin Ointment. Nitroglycerin ointment (15 mg/inch)
can be applied to the chest in 0.5- to 2.0-inch strips. The onset of
action is delayed about 30 minutes, but the clinical effect lasts for
4 to 6 hours. Cutaneous permeability of the ointment is enhanced
with increased hydration and plastic coverings. Ointment prepa-
rations are most commonly used in chronic severe angina and
nocturnal angina.
Nitroglycerin Transdermal Patches. Impregnated with
polymer matrix or silicone gel, nitroglycerin is absorbed 24 to
48 hours after application. The release rate varies from 0.1 to
0.8 mg per hour. Transdermal administration of nitroglycerin
helps patients prolong exercise duration and provides antiangi-
nal effects for 12 hours without significant nitrate tolerance or
rebound phenomena.
Isosorbide Dinitrate. Isosorbide dinitrate undergoes rapid
hepatic metabolism and thus has low bioavailability after oral
administration. One of its subsequent metabolites exerts potent
vasodilatory effect and is cleared less rapidly (urinary excretion).
Partial or complete nitrate tolerance develops with isosorbide
dinitrate administered at a dosage of 30 mg 3 or 4 times daily.
Hence, a dosage schedule that includes a 10- to 12-hour nitrate-
free interval should be implemented to prevent tolerance.
Isosorbide 5-Mononitrate. Isosorbide 5-mononitrate, the
active metabolite of dinitrate, does not undergo first-pass hepatic
 66 Chronic Stable Angina
metabolism. The complete bioavailability of the 5-mononitrate
preparation increases the efficacy for treating chronic stable
angina. Isosorbide 5-mononitrate reaches its peak level between
30 minutes and 2 hours after ingestion and has a plasma half-
life of 4 to 6 hours. The sustained-release form is given once
daily. Nitrate tolerance has not been demonstrated with once-
a-day dosing intervals but can occur with twice-daily dosing at
12-hour intervals.
β-Adrenergic Receptor Blocking Agents
β-Blockers, through the competitive inhibition of β-adrenoceptors,
antagonize the effect of circulating and neuronally released cat-
echolamines. β-Blockers slow heart rate and increase the dias-
tolic portion of the cardiac cycle, which allows more time for
coronary perfusion. By blocking the surges of sympathetic
activity during exercise, β-blockers reduce contractility and left
ventricular wall tension, resulting in lower myocardial oxygen
demand. Many studies have shown that β-blockers decrease the
frequency of anginal episodes and increase the anginal thresh-
old. Other salutary properties of β-blockade include decreasing
mortality and reinfarction among post-MI patients and decreas-
ing mortality among patients with heart failure. Abrupt with-
drawal of β-blockade can precipitate unstable angina and even
MI in patients who have chronic CAD. Although there has been
concern of worsening bronchoconstriction and peripheral vascu-
lar disease, β-blockers can be used safely in many patients who
have chronic obstructive pulmonary disease and peripheral arte-
rial disease.
Calcium Channel Blockers
Calcium channel blockers consist of 3 subclasses: dihydropyri-
dines (eg, nifedipine), phenylalkylamines (eg, verapamil), and the
modified benzothiazepines (eg, diltiazem). The antianginal effect
of calcium channel blockers derives primarily from their ability to
reduce myocardial oxygen demand and enhance myocardial oxy-
gen supply. The latter effect is particularly useful in patients with
vasospastic components of angina pectoris, including variable-
threshold angina, Prinzmetal angina (also called variant angina),
and angina of small coronary arteries with impaired vasodilator
reserve. This vasodilatory effect can be seen in both systemic and
coronary arterial beds. Inhibition of calcium entry into cardiac
myocytes can have serious negative inotropic effects on the heart
and potentially lead to worsening heart failure in patients with
clinically significant left ventricular dysfunction. Verapamil and
diltiazem decrease conduction through the atrioventricular node,
and care should be taken when combining with β-blockers owing
to the risk of atrioventricular block.
Calcium channel blockers have also been postulated to have
antiatherosclerotic properties. The PREVENT study showed
slowing of atherosclerotic progression in carotid but not in coro-
nary vasculatures. A randomized, double-blind study of 7,665
patients with stable symptomatic CAD showed that long-acting
nifedipine had no effect on major cardiovascular event–free
survival but did reduce the need for coronary angiography and
CABG. Given to patients before undergoing PTCA, amlodipine
reduced major cardiovascular end points (death, MI, CABG, and
subsequent PTCA) in the CAPARES trial.
Other Antianginal Agents
Ranolazine. Ranolazine is a late sodium current inhibi-
tor, approved by the US Food and Drug Administration for the
633
treatment of chronic angina. It is thought to reduce myocardial
tension and oxygen consumption without changing hemody-
namic parameters, including heart rate and blood pressure. In
the CARISA trial, ranolazine increased exercise capacity and
provided antianginal relief in patients who had severe chronic
angina and were receiving standard antianginal therapy, includ-
ing β-blockers and calcium channel blockers.
Ivabradine. Ivabradine is a new selective heart-rate lowering
agent that inhibits the If sinoatrial node pacemaker currents and
reduces heart rate at rest and during exercise. Randomized con-
trolled trials have demonstrated a dose-dependent improvement
in exercise tolerance and reduction of ischemia that is compa-
rable to the benefit seen with β-blockers. In the BEAUTIFUL
trial, ivabradine did not improve the cardiovascular death rate or
the hospital admissions rates for acute MI or acute heart failure
overall among patients with chronic stable angina and ventricular
systolic impairment. Among the subgroup of patients with rest-
ing heart rates greater than 70 beats per minute, ivabradine did
decrease the rates of hospital admissions for MI, unstable angina,
and coronary revascularization. Ivabradine has not been approved
for use in the United States, but it may have a role in the future as
an antianginal agent in patients who cannot tolerate β-blockers.
Pharmacologic Therapy No Longer
Recommended in the Treatment
of Chronic CAD
Estrogen Replacement
Randomized, controlled secondary prevention trials, such as
HERS, HERS II, and WHI, have suggested that hormone replace-
ment therapy does not reduce cardiovascular events or mortality.
Current recommendations do not support the use of hormone
therapy to reduce the risk of heart disease. Unanswered ques-
tions include the following: 1) Were these results from the dos-
age or regimen of hormone used? 2) Is there a small window of
opportunity after menopause during which initiation of hormone
therapy will reduce risk? Current studies and recommendations
are discussed in CHAPTER 62 “HEART DISEASE IN WOMEN.”
Antioxidants
The Heart Protection Study Collaborative Group enrolled more
than 20,000 patients who had atherosclerotic vascular disease or
diabetes mellitus. The group conclusively showed that there was
no reduction in all-cause mortality, MI, or other vascular events
with the regimen of vitamin C, vitamin E, and β-carotene.
Nonpharmacologic Antianginal Interventions
Spinal Cord Stimulation
Patients with refractory angina not amenable to coronary revas-
cularization may benefit from spinal cord stimulation, which
involves inserting specific electrodes into the epidural space
and using neuromodulation to reduce painful stimuli. Several
observational studies have reported success rates of up to 80% in
decreasing anginal frequency and severity. However, more data
are still needed; therefore, spinal cord stimulation should be con-
sidered only when other treatment options have failed.
Enhanced External Counterpulsation
EECP is another alternative therapy for refractory angina.
Most data are from observational studies, which have reported
634 VII Myocardial Infarction

improved exercise tolerance and reduced anginal frequency and
nitroglycerin use among patients treated with EECP. EECP has
been postulated to decrease myocardial oxygen demand, enhance
myocardial collateral flow by means of increased transmyocar-
dial pressure, and improve endothelial function.
Myocardial Revascularization
Approaches to revascularize the myocardium include PCI (ie,
balloon angioplasty and stenting) and CABG surgery. Patients
with chronic stable angina should be considered for PCI or
CABG surgery if significant symptoms persist despite intense
medical therapy. Revascularization should also be considered for
patients at increased risk with medical therapy alone (eg, patients
with left main CAD, multivessel disease, high-risk features on
stress testing, or left ventricular dysfunction).
Percutaneous Coronary Intervention
Technical success rates with PCI are usually high among younger
patients (younger than 70) with single-vessel disease (single lesion)
and no evidence of congestive heart failure (ejection fraction
>40%). Owing to increased operator experience and widespread
use of stents, PCI now has an overall procedural success rate of
90% with an expected periprocedural mortality of less than 1%.
No randomized trial comparing PCI with medical ther-
apy has demonstrated improved survival among patients with
chronic stable angina. The COURAGE trial, which involved
2,287 patients who had stable angina or ischemia, did not find
a difference in all-cause death or nonfatal MI between patients
who underwent PCI with bare metal stenting and patients who
received pharmacologic therapy over a mean follow-up of 4.6
years (Figure 66.4). However, PCI was associated with fewer

A 1.0 B 1.0
Survival Free of Death
From Any Cause and

PCI
Myocardial Infarction

Medical therapy

Overall Survival
0.9 0.9
PCI
Medical therapy
0.8 0.8

0.7 0.7
Hazard ratio, 1.05; Hazard ratio, 0.87;
95% CI, 0.87-1.27; P=.62 95% CI, 0.65-1.16; P=.38
0.6 0.6
0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
Years Years
No. at Risk No. at Risk
Medical 1,138 1,017 959 834 638 408 192 30 Medical 1,138 1,073 1,029 917 717 468 302 38
therapy therapy
PCI 1,149 1,013 952 833 637 417 200 35 PCI 1,149 1,094 1,051 929 733 488 312 44

C 1.0 D 1.0
of Myocardial Infarction

Medical therapy
Survival Free of ACS

Medical therapy
0.9 PCI 0.9 PCI
Survival Free

0.8 0.8

0.7 0.7
Hazard ratio, 1.07; Hazard ratio, 1.13;
95% CI, 0.84-1.37; P=.56 95% CI, 0.89-1.43; P=.33
0.6 0.6
0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
Years Years
No. at Risk No. at Risk
Medical 1,138 1,025 956 833 662 418 236 127 Medical 1,138 1,019 962 834 638 409 192 120
therapy therapy
PCI 1,149 1,027 957 835 667 431 246 134 PCI 1,149 1,015 954 833 637 418 200 134

Figure 66.4. COURAGE Trial Kaplan-Meier Survival Curves. A, The estimated 4.6-year rate of the composite primary outcome of death from
any cause and nonfatal myocardial infarction was 19.0% in the percutaneous coronary intervention (PCI) group and 18.5% in the medical therapy
group. B, The estimated 4.6-year rate of death from any cause was 7.6% in the PCI group and 8.3% in the medical therapy group. C, The estimated
4.6-year rate of hospitalization for acute coronary syndrome (ACS) was 12.4% in the PCI group and 11.8% in the medical therapy group. D, The
estimated 4.6-year rate of acute myocardial infarction was 13.2% in the PCI group and 12.3% in the medical therapy group. (Previously published.
See “Credit Lines” section.)
 66 Chronic Stable Angina
revascularization rates overall. A subgroup of 314 patients in
the COURAGE trial underwent serial rest/stress myocardial
perfusion studies. Among the patients in that subgroup who had
demonstrable ischemia, there was a trend toward lower risk-
adjusted rates of death or nonfatal MI with PCI compared with
optimal medical therapy alone. These results underscore a poten-
tial role for stress testing in risk stratification and determining
the initial choice of treatment. Of note, high-risk patients with
left main CAD, markedly positive stress tests, or a left ventricular
ejection fraction of less than 30% were excluded from the study.
In the BARI 2D trial, mortality at 5 years was similar for
2,368 patients with stable CAD and type 2 diabetes mellitus when
comparing invasive therapy (PCI or CABG) with optimal phar-
macologic therapy. Additional revascularization rates, however,
have been consistently higher in the medical treatment groups
in most trials to date. A recent meta-analysis that included the
COURAGE trial similarly found no difference in death, MI, and
CABG surgery rates for patients who had chronic stable angina
and received either bare metal stents or medical therapy, whereas
medical therapy was associated with more frequent additional
revascularization rates. In the COURAGE and BARI 2D trials,
respectively, 93% and 94% of the patients received antiplatelet
agents, 86% and 88% received β-blockers, 78% and 92% received
ACE inhibitors or ARBs, and 93% and 95% received statins with
good rates of compliance with guideline-recommended pharma-
cologic therapy. However, both studies enrolled patients who had
stable CAD, low to moderate risk, and anatomical characteriza-
tion with coronary angiography before randomization.
Although flow-limiting lesions in chronic stable angina cause
ischemia, they may or may not be the lesions predisposing to acute
coronary syndrome or death. CAD appears to be a systemic proin-
flammatory and proatherosclerotic disease associated with multiple
vulnerable plaques that are not necessarily obstructive or angio-
graphically prominent, and a systemic approach should be used to
stabilize them. Hence, aggressive medical management is gener-
ally the initial treatment of patients with chronic stable angina.
CABG Surgery
A CABG may be an option for patients with refractory angina
and surgically approachable vessels. A CABG procedure is less
invasive with the advent of OPCAB and MIDCAB. The avoid-
ance of CPB reduces the risk of bleeding, systemic thromboem-
bolism, renal insufficiency, myocardial stunning, stroke, and
neurologic damage. Other advances involve the use of femoral-
femoral CPB in a TECAB. Cardioplegic techniques have also
improved substantially with the use of blood and other substrates
such as glutamate to facilitate myocardial aerobic metabolism
and lower lactate production. Retrograde cardioplegia through
the coronary sinus provides more uniform distribution of car-
dioplegic solution and is currently used in conjunction with the
more traditional antegrade delivery.
Selection of Patients for CABG Surgery. Patients with left
main CAD, multivessel disease, proximal LAD disease, and left
ventricular dysfunction appear to derive more survival benefit
from a CABG procedure than from medical therapy. Benefits are
also greatest among those with severe symptoms or abnormal
stress testing. In the BARI 2D trial, patients who had type 2 dia-
betes mellitus and stable CAD either underwent revasculariza-
tion with a CABG procedure or received medical therapy. The
rate of major cardiovascular events (death, MI, or stroke) was
lower in the CABG group, which had more frequent 3-vessel dis-
ease and proximal disease of the LAD. This effect was mainly
635
driven by lower nonfatal MI rates with the CABG procedure.
However, overall mortality was similar between the CABG and
optimal medical treatment groups.
A CABG procedure is highly effective in providing com-
plete relief of angina and other symptoms. Data from a group of
patients with saphenous vein grafts showed that 90% of patients
were angina free at 1 year; in the subsequent 4 years, the recur-
rence rate was approximately 3% per year and 5% per year there-
after. Angina-free rates were 78% at 5 years, 52% at 10 years,
and 23% at 15 years.
PCI Versus CABG. Earlier observational studies that compared
PCI with a CABG procedure were limited largely to the use of
PTCA. Over 1 to 5 years, the rates of mortality and nonfatal MI
did not differ significantly between patients revascularized with
a CABG procedure or with PTCA. However, 1 year after PTCA,
recurrent symptoms or the need for subsequent procedures was
approximately 40%. In addition, subgroup analysis showed that
patients who had left ventricular dysfunction or proximal LAD
stenosis greater than 70% derived more survival benefit from a
CABG procedure than from PTCA.
The BARI trial compared 10-year outcomes between CABG
surgery and PTCA in 1,829 patients who had stable multivessel
CAD. The 10-year survival and angina rates were similar in the
2 groups, but the PTCA group had higher revascularization rates.
The subgroup of patients with treated diabetes mellitus, however,
had a larger survival benefit with CABG surgery. The ARTS trial
also found no difference in survival between CABG surgery and
stenting in 1,205 patients. However, subsequent revasculariza-
tions were less frequent after CABG surgery, and the subgroup
of patients with diabetes mellitus had better results with CABG
surgery. In a systematic review of 23 randomized controlled tri-
als in which 5,019 patients were assigned to PCI and 4,944 to
CABG surgery, there was no survival difference at 10 years.
Subsequent revascularization was more common with PCI, and
procedure-related stroke was more common with CABG sur-
gery. In summary, long-term survival rates are similar between
PCI and CABG surgery; however, CABG surgery is associated
with less frequent revascularization and more complete relief
of angina. For diabetic patients, CABG surgery is the preferred
revascularization strategy.
CABG surgery is currently the preferred strategy in the man-
agement of 3-vessel or left main CAD. In the SYNTAX trial, 1,800
patients with 3-vessel or left main CAD were randomly assigned
to CABG surgery or PCI with drug-eluting stents and followed
for 1 year. Revascularization rates were higher in the PCI group,
whereas stroke was more likely to occur with CABG surgery.
However, the rates of death or MI were similar in the 2 groups.
• Patients with chronic stable angina should be considered for PCI
or CABG surgery if significant symptoms persist despite intense
medical therapy.
• Revascularization should be considered for patients at increased
risk with medical therapy alone (eg, patients with left main CAD,
multivessel disease, high-risk features on stress testing, or left
ventricular dysfunction).
• Although flow-limiting lesions in chronic stable angina cause
ischemia, they may or may not be the lesions predisposing to acute
coronary syndrome or death.
• Long-term survival rates are similar between PCI and CABG
surgery; however, CABG surgery is associated with less frequent
revascularization and more complete relief of angina.
• For diabetic patients, CABG surgery is the preferred revascular-
ization strategy.
636 VII Myocardial Infarction

Abbreviations BEAUTIFUL Morbidity-Mortality Evaluation of the If Inhibitor

Ivabradine in Patients With Coronary Artery Disease
ACC/AHA American College of Cardiology and American Heart and Left Ventricular Dysfunction
Association CAPARES Coronary Angioplasty Amlodipine Restenosis Study
ACE angiotensin-converting enzyme CAPRIE Clopidogrel Versus Aspirin in Patients at Risk of
ARB angiotensin receptor blocker Ischemic Events
BNP brain natriuretic peptide CARE Cholesterol and Recurrent Events
CABG coronary artery bypass graft CARISA Combination Assessment of Ranolazine in Stable
CAD coronary artery disease Angina
CCS Canadian Cardiovascular Society CDP Coronary Drug Project
CPB cardiopulmonary bypass CHARISMA Clopidogrel for High Atherothrombotic Risk and
CRP C-reactive protein Ischemic Stabilization, Management, and Avoidance
ECG electrocardiogram COURAGE Clinical Outcomes Utilizing Revascularization and
EECP enhanced external counterpulsation Aggressive Drug Evaluation
HDL high-density lipoprotein EUROPA European Trial on Reduction of Cardiac Events With
LAD left anterior descending coronary artery Perindopril in Stable Coronary Artery Disease
LDL low-density lipoprotein HERS Heart and Estrogen/Progestin Replacement Study
MI myocardial infarction HERS II Heart and Estrogen/Progestin Replacement Study
MIDCAB minimally invasive direct coronary artery bypass Follow-up
NT-proBNP N-terminal fragment of brain natriuretic peptide HOPE Heart Outcomes Prevention Evaluation
NYHA New York Heart Association HPS Heart Protection Study
OPCAB off-pump coronary artery bypass LIPID Long-term Intervention With Pravastatin in Ischemic
PCI percutaneous coronary intervention Disease
PTCA percutaneous transluminal coronary angioplasty PEACE Prevention of Events With Angiotensin-Converting
TECAB totally endoscopic coronary artery bypass Enzyme Inhibition
TG triglyceride PREVENT Prospective Randomized Evaluation of the Vascular
Effects of Norvasc Trial
Names of Clinical Trials
REVERSAL Reversal of Atherosclerosis With Aggressive Lipid
4S Scandinavian Simvastatin Survival Study Lowering
ARTS Arterial Revascularization Therapies Study SAPAT Swedish Angina Pectoris Aspirin Trial
ASTEROID A Study to Evaluate the Effect of Rosuvastatin on SAGE Study Assessing Goals in the Elderly
Intravascular Ultrasound-Derived Coronary Atheroma SYNTAX Synergy Between PCI With Taxus and Cardiac
Burden Surgery
BARI Bypass Angioplasty Revascularization Investigation VA-HIT Veterans Affairs High-Density Lipoprotein Interven-
BARI 2D Bypass Angioplasty Revascularization Investigation tion Trial
2 Diabetes WHI Women’s Health Initiative

Right Ventricular Infarction
JOSEPH G. MURPHY, MD, and R. SCOTT WRIGHT, MD
Isolated RV MI is rare, but an estimated 30% of patients with
acute inferior MI have clinical or echocardiographic evidence of
RV dysfunction or infarction (Figures 67.1 and 67.2).
RV infarction is frequently underdiagnosed and is associated
with an increased in-hospital mortality compared with inferior
wall MI in patients with preserved RV function. Many studies
have shown that patients with RV infarction are a unique sub-
group with a distinctive clinical presentation, treatment, and
prognosis.
Compared with the LV, the RV is partially protected from
ischemia because it pumps at a lower pressure and has lower wall
stress and oxygen demand and a much smaller muscle mass. In
addition, coronary perfusion in the RV occurs throughout systole
and diastole. Thus, the RV usually recovers from RV infarction
provided the patient survives the initial event.
RV Infarction Culprit Coronary Lesion
RV infarction is commonly associated with acute LV inferior
wall MI. Postmortem and angiographic data have shown that
most RV infarctions usually result from occlusion of the proximal
RCA; mid and distal RCA occlusion usually spares enough RV
branches of the RCA that RV infarction does not occur. Inferior
RV involvement may also occur 1) with occlusion of a distal left
anterior descending coronary artery that has an anatomical dis-
tribution that wraps around the LV apex or 2) with occlusion of a
dominant left circumflex coronary artery. The proximal location
of the RCA occlusion is critical in the development of RV infarc-
tion because it compromises RV branch vessels that supply the
RV free wall. The magnitude of RV dysfunction correlates with
the extent of impairment of RCA marginal branch blood flow;
Abbreviations and acronyms are expanded at the end of this chapter.
67
failure to restore marginal branch blood flow is associated with
poor recovery of RV function, persistent systemic hypotension,
low cardiac output, and high mortality. In about 40% of cases of
deaths from inferior MI, necropsy studies found evidence of RV
infarction with evidence of a common pathologic triad of infarc-
tion involving the LV inferoposterior wall, the shared septum,
and the posterior RV free wall.
• RV infarction is commonly observed with inferior LV infarction.
• Occlusion of the proximal RCA is the most common cause of RV
infarction.
• Rapid restoration of RV branch vessel perfusion is key to preven-
tion of RV infarction.
Clinical Presentation
In 1930, a patient was described who had the classic triad of
hypotension, increased jugular venous pressure, and clear lung
fields and postmortem evidence of extensive RV necrosis but
minimal LV damage. The clinical findings in RV infarction show
a distinct hemodynamic syndrome of severe right heart failure,
clear lung fields, and low cardiac output in the presence of pre-
served LV systolic function.
In addition to this classic triad, the Kussmaul sign (an increase
in jugular venous pressure with inspiration), pulsus paradoxus
(a decrease in systolic blood pressure >10 mm Hg with inspira-
tion), and a positive hepatojugular reflux test may be observed
in patients with RV infarction. In patients with an inferior MI,
the Kussmaul sign is suggestive of RV infarction. A thin-walled
RV responds to ischemia by dilating to fill the available pericar-
dial space, thus mimicking the above clinical signs of pericar-
dial constriction. In spite of the RV dilatation that occurs with
RV infarction, an RV heave on clinical examination is rare. On
637
638 VII Myocardial Infarction

Figure 67.1. Acute Right Ventricular Infarct With Right Ventricular

Dilatation.

cardiac auscultation, RV failure may be evidenced by an RV third

or fourth heart sound that is distinguishable from an LV third or
fourth heart sound by its augmentation on inspiration. Tricuspid
regurgitation due to RV and tricuspid dilatation may also result
in a holosystolic murmur.
• The triad of hypotension, increased jugular venous pressure, and
clear lung fields in patients with acute inferior MI suggests an RV
infarction.
• The differential diagnosis of RV infarction clinical signs includes
pulmonary embolus, pericardial tamponade, and constrictive
pericarditis.

Diagnosis of RV Infarction
Electrocardiography
Figure 67.2. Healed Right Ventricular Infarct With Right Ventricular
In a patient with an acute inferior or inferoposterior MI, Wall Fibrosis.
ST-segment elevation that is greater in lead III than in lead II is
suggestive of acute RV infarction. Right precordial leads, specifi-
cally V4R, significantly improve ECG accuracy for the diagnosis • The combination of RAP ≥10 mm Hg and RAP/PCWP ≥0.86 sug-
of RV infarction (88% sensitivity and 78% specificity). gests RV infarction.

• ST-segment elevation greater in lead III than in lead II is sugges- Echocardiography

tive of acute RV infarction.
• Right precordial lead V4R is the most sensitive and specific for the
ECG detection of RV infarction.
• All patients with an inferior wall ST-segment elevation MI should
be evaluated with a right precordial V4R ECG lead, particularly if
they are hemodynamically compromised.
Hemodynamic Assessment
The clinical findings of increased jugular venous pressure, clear
lung fields, and hypotension are the classic manifestation of the
hemodynamic compromise resulting from RV infarction. RV
infarction decreases the compliance of the RV, leading to an ele-
vated RAP (≥10 mm Hg) and an increased RV filling pressure,
especially during inspiration. An RAP of at least 10 mm Hg and
an RAP/PCWP ratio of at least 0.86 suggest RV infarction. Right
atrial systolic dysfunction may complicate the hemodynamics of
RV infarction, while patients with a small-amplitude a wave tend
to fare worse clinically. Volume depletion exacerbates hypoten-
sion but may mask the classic jugular venous pulse findings of
RV infarction.
Echocardiography is the imaging modality of choice for diag-
nosis of RV infarction. Typical findings are RV dilatation, RV
free wall hypokinesia or akinesia, and abnormal interventricu-
lar septal motion caused by a reversal of the LV-RV transseptal
pressure gradient due to the increased RV end-diastolic pressure.
Decreased descent of the RV base and plethora of the inferior
vena cava may also occur.
• Echocardiographic evidence of abnormal RV free wall motion and
RV dilatation is suggestive of RV infarction.
• Short-axis echocardiographic views have demonstrated high sensi-
tivity and specificity for detection of hemodynamically significant
RV infarction.
• Hemodynamically compromised patients with inferior ST-segment
elevation MI should undergo urgent echocardiography to screen
for RV infarction or cardiac tamponade.
Pulmonary Embolism
Pulmonary embolism is often in the differential diagnosis of
RV infarction. Patients with either condition may present with
 67 Right Ventricular Infarction
hypotension, hypoxia, and clear lung fields. Serum troponin lev-
els can increase in both conditions. Two studies are useful: 1) the
ECG often shows ST-segment elevation in the right precordial
leads in RV infarction, and 2) echocardiography shows regional
wall motion abnormalities that frequently spare the RV apex in
pulmonary embolism.
Complications of RV Infarction
Low Cardiac Output
The hemodynamic complications of RV infarction can be pro-
found. RV cardiac output is dependent on RV preload, which in
turn is very sensitive to patient fluid status. LV function in RV
infarction is “preload deprived,” resulting in an overall low car-
diac output. The low cardiac output in RV infarction is not solely
the result of a decrease in RV systolic performance; dilatation of
the RV compromises LV filling because of a leftward displace-
ment of the interventricular septum during diastole.
Bradyarrhythmias
The ischemic RV has a relatively fixed stroke volume; thus, car-
diac output becomes closely heart-rate dependent. Even in the
absence of AV dyssynchrony, bradycardia may have a profound
negative effect on cardiac output and hemodynamic effects.
High-degree AV block has been reported to occur in as many as
50% of patients with RV infarctions.
Tachyarrhythmias
Compared with patients without RV involvement, patients with
RV infarction have a greater incidence of ventricular arrhyth-
mias, especially patients with unsuccessful myocardial reper-
fusion. In up to one-third of patients with RV infarction, atrial
fibrillation is precipitated by concomitant atrial infarction or
right atrial dilatation. The loss of atrial contraction can have
profound hemodynamic consequences in RV infarction because
the noncompliant RV becomes exquisitely dependent on atrial
contraction for adequate preload.
Mechanical Complications
Severe dilatation of the right atrium and RV in combination with
diastolic pressure elevation may result in severe tricuspid valve
regurgitation. This in turn further impairs RV output and exac-
erbates the overall low cardiac output state. The increase in right
atrial pressure in relation to left atrial pressure can promote right-
to-left shunting across a patent foramen ovale or atrial septal
defect, resulting in systemic hypoxemia or paradoxical emboli,
a complication that should always be considered in MI patients
who have hypoxemia that is nonresponsive to oxygen therapy.
• The hemodynamic complications of RV infarction can be pro-
found, resulting in low cardiac output.
• High-degree AV block has been reported to occur in as many as
50% of patients with RV infarctions.
• In up to one-third of patients with RV infarction, atrial fibrilla-
tion is precipitated by concomitant atrial infarction or right atrial
dilatation.
• An uncommon mechanical complication of RV infarction is sig-
nificant right-to-left shunting across a patent foramen ovale or
atrial septal defect that results in systemic hypoxemia or paradoxi-
cal emboli.
639
Treatment of RV Infarction
The treatment of RV infarction is early reperfusion of the cul-
prit artery (with thrombolysis or percutaneous coronary inter-
vention), maintenance of RV preload, inotropic support of the
ventricular septum (which has a dual blood supply from the RCA
and the left anterior descending coronary artery) and the dys-
functional RV, reduction of RV afterload, and maintenance of
atrioventricular synchrony.
Reperfusion
In patients with RV infarction, successful reperfusion of the
RCA significantly improves RV function and reduces in-hospital
mortality. Successful thrombolysis imparts a survival benefit to
patients who have RV involvement, and failure to restore infarct
artery patency is associated with persistent RV dysfunction and
increased mortality. The RCA may reocclude with a higher inci-
dence than other coronary arteries after thrombolysis. Primary
angioplasty is more likely to result in successful recanalization
of the RCA, with resulting benefits for RV performance and clin-
ical outcomes. Preinfarction angina predicts a lower incidence of
RV infarction in patients with acute inferior MI, possibly because
of ischemic preconditioning of the RV.
Fluid Resuscitation
In RV infarction, the RV is very preload dependent while the LV
is preload deprived. Optimization of ventricular preload is criti-
cal in the treatment of RV infarction. In patients with low cardiac
output, central venous pressure should be increased with admin-
istration of isotonic saline. Placement of a Swan-Ganz catheter is
frequently indicated to guide fluid administration and optimize
hemodynamics. Fluid expansion beyond a PCWP of more than
15 mm Hg is unlikely to improve the hemodynamic profile fur-
ther. Volume loading may not always produce an increase in car-
diac output; therefore, it is important to quantitate the effect of
volume loading on the stroke volume and cardiac output in order
to guide the use of additional fluid therapy.
Inotropic Therapy
Inotropic therapy for RV infarction may improve hemodynam-
ics by causing hyperdynamic septal contraction, which may
compensate for RV free wall hypokinesia. This may be enough
to maintain RV cardiac output even in the presence of marked
RV free wall hypokinesis. Dopamine infusion in RV infarction
patients can result in a significant increase in the cardiac index,
stroke volume, and RV ejection fraction.
Reduction of RV Afterload With LV Dysfunction
Arterial vasodilators are generally not helpful in RV infarction
and will exacerbate hypotension. Mechanical support with intra-
aortic balloon counterpulsation, while not directly improving RV
performance, may do so through increases in myocardial perfu-
sion pressure. RV assist devices may also provide bridging sup-
port in patients who have severe hemodynamic deterioration that
is refractory to other therapy.
Specific Therapy to Avoid
Because of their dependency on adequate RV preload, patients
with extensive RV infarction show exquisite sensitivity to nitrates,
which can cause profound hypotension. In patients with inferior
640 VII Myocardial Infarction

MI, hypotension should suggest possible RV involvement. Drugs and 26% among patients with TIMI risk scores of 0 or 1, 2 or 3,
that slow conduction through the AV node, such as β-blockers and 4 or more, respectively.
and calcium channel blockers, may increase the risk of serious Resolution of RV dysfunction is common among patients who
bradyarrhythmias and should be used cautiously in patients with survive RV infarction. These patients appear to have no more
inferior wall MI or RV infarction. additional mortality risk in long-term follow-up than patients
with acute inferior wall MI uncomplicated by RV infarction.
Chronic right heart failure due to RV infarction is rare. Initial
Maintenance of AV Synchrony
TIMI risk score analysis at presentation has also been shown
In patients with RV infarction, the RV is dependent on atrial con- to be predictive of long-term mortality in a small retrospective
traction to optimize RV filling and maintain RV cardiac output; cohort of patients with RV infarction.
thus, loss of chronotropy or AV synchrony can have devastating
consequences. Atrial or dual chamber pacing should be consid- • In-hospital and 30-day morbidity and mortality are high with RV
ered if ventricular pacing becomes necessary because of brady- infarction.
cardia. Every effort should be made to keep the patient in sinus • Morbidity and mortality are decreased among patients who have
rhythm. RV infarction and successful, prompt revascularization.
• Initial TIMI risk score analysis provides important prognostic data
• The treatment of RV infarction involves early reperfusion, main- on both short- and long-term outcomes.
tenance of RV preload, inotropic support of the dysfunctional RV,
reduction of RV afterload, and maintenance of AV synchrony.
Abbreviations
AV atrioventricular
Prognosis for Patients With RV Infarction ECG electrocardiographic
RV infarction, which can often result in hemodynamic shock, LV left ventricular
may be accompanied by high-degree AV block because of occlu- MI myocardial infarction
sion of the AV nodal artery, which is a distal branch of the RCA; PCWP pulmonary capillary wedge pressure
RAP right atrial pressure
this particularly detrimental combination results in high in-hos-
RCA right coronary artery
pital morbidity and mortality. When culprit vessel reperfusion RV right ventricular
is achieved, in-hospital and short-term morbidity and mortality
decrease significantly. Clinical assessment by TIMI risk score
analysis is predictive of in-hospital and 30-day mortality among Name of Clinical Trial
patients with RV infarction. In-hospital mortality was 7%, 13%, TIMI Thrombolysis in Myocardial Infarction
 68
Adjunctive Therapy in Acute Myocardial Infarction
R. SCOTT WRIGHT, MD, IMRAN S. SYED, MD,
and JOSEPH G. MURPHY, MD
Adjunctive therapies in AMI are best considered as specifi-
cally targeting high-risk features pertaining to the risk triad of
the unstable atherosclerotic plaque, patient characteristics, and
plasma characteristics that are associated with poor clinical out-
come (Figure 68.1).
Adjunctive Therapy for AMI in the
ED or Prehospital Setting
Adjunctive therapy for AMI should be initiated immediately
once the diagnosis is suspected—in many cases before the diag-
nosis confirmation or definitive reperfusion therapy is instigated.
Ideally it should begin in the ED or prehospital setting, absent
contraindication, and include the following:
1. Aspirin
2. Oxygen
3. Morphine
4. Intravenous UFH or subcutaneous LMWH
5. β-Blockade, initially intravenously and then orally
6. Nitroglycerin, except in patients with hypotension or suspected right
ventricular infarction
Early risk stratification of high-risk AMI patients (score ≥5)
with either the TIMI risk score or the Mayo Clinic risk score
is important to facilitate an aggressive therapeutic approach for
patients at highest risk. The TIMI score requires knowledge
of serum cardiac biomarkers and prior coronary anatomy, but
the Mayo Clinic risk score is based solely on patient data that
are generally available earlier during admission (Tables 68.1
and 68.2).
Abbreviations and acronyms are expanded at the end of this chapter.
With either TIMI or Mayo Clinic risk scoring, mortality risk
is classified as low (score, 0-2), intermediate (3 or 4), or high
(≥5) (Figure 68.2).
• There is no evidence to support the prophylactic use of intravenous
lidocaine to prevent venticular tachycardia/ventricular fibrillation
in the AMI patient.
• There is no evidence that calcium channel blockers reduce mortal-
ity in acute AMI patients and their use is generally contraindicated
with exception of AMI associated with refractory coronary spasm.
Reflex sympathetic activation is a problem with some short-acting
calcium channel blockers, particularly immediate-release nifedi-
pine, which is contraindicated in AMI.
Aspirin
Aspirin inhibits both platelet aggregation and activation
through inhibition of the platelet cyclooxygenase pathway.
Additionally, aspirin has a beneficial anti-inflammatory effect
on the unstable atheromatous plaque, both acutely and long
term (Figure 68.3).
There is substantial evidence documenting the clinical effec-
tiveness of aspirin in all types of AMI and unstable angina AHA/
ACC class I evidence. Aspirin (162–325 mg) should be adminis-
tered immediately to all patients with suspected AMI, and there-
after given daily at a dose of 162–325 mg once daily; the aspirin
dose is generally reduce to 81 mg daily when dual antiplatelet
therapy is administered. Aspirin therapy should be withheld
only in patients with substantive active bleeding or a documented
aspirin allergy, in which case clopidogrel or an alternative anti-
platelet agent should be given immediately in place of aspirin.
Aspirin should be given initially as a chewed dose of 162–325 mg
in the ED or prehospital setting.
641
642 VII Myocardial Infarction

Figure 68.1. Atherosclerotic plaque biology. ACE, angiotensin-converting enzyme; A-CH, acetylcholine; ADR, adrenergic; agg’n; aggregation;
AT, angiotensin receptor; Ca++Ch Inh, calcium channel inhibitor; IFN, interferon; IL, interleukin; LAM, lymphocyte adhesion molecule; L-ARG,
L-arginine; MCP, membrane cofactor protein; N-Pr, natriuretic peptide receptor; PAI, plasminogen activator inhibitor; PLT, platelet; SMC, smooth
muscle cell; Stim, stimulates; TNF, tumor necrosis factor; tPA, tissue plasminogen activator.

• Aspirin should be given daily at a dose of 162-325 mg. • Oxygen therapy should be initiated in all AMI patients in the pre-
• The aspirin dosage should be reduced to 81 mg daily when given in hospital or ED setting.
conjunction with clopidogrel. • Oxygen therapy should be titrated to a resting arterial saturation
• It is important to discontinue all nonsteroidal anti-inflammatory of ≥90%.
drugs (except aspirin) in AMI patients due to their association with
an increased risk of CV events Morphine
Intravenous morphine sulfate is the drug of choice for relief of
Oxygen chest pain and anxiety in AMI. The initial dose is 2 to 4 mg, with
All patients presenting with AMI require supplemental oxygen,
Table 68.2. Mayo Clinic Risk Score
preferably in the prehospital setting. For patients with docu-
mented hypoxemia, oxygen therapy should be titrated to increase Factor Score
the arterial oxygen saturation to at least 90%. Oxygen therapy can
Age >80 y 2
be discontinued after the initial day of stabilization in patients
Sex—female 3
with a resting saturation of more than 90% on room air.
SBP ≤140 mm Hg 3
Creatinine >1.4 mg/dL 1
ST-segment depression
Table 68.1. TIMI Risk Score
1–2 μV 1
Age ≥65 >2 μV 3
Presence of ≥3 conventional risk factors for coronary artery disease QRS duration ≥100 ms 1
Prior coronary stenosis ≥50% Killip class >I 3
Presence of ST-segment deviation on initial admission electrocardiogram MI location—anterior 1
At least 2 anginal episodes in prior 24 hours
Elevated serum cardiac biomarkers Abbreviations: MI, myocardial infarction; SBP,
Aspirin use in prior 7 days systolic blood pressure.
Previously published. See “Credit Line” section.
 68 Adjunctive Therapy in Acute Myocardial Infarction
Figure 68.2. Thirty-day mortality by risk score in training and vali-
dation sets. (Previously published. See “Credit Line” section.)
increments of 2 to 8 mg repeated at intervals of 5 to 15 minutes
if needed. Watch for respiratory depression.
Heparin
The most commonly used heparin in AMI is UFH. LMWHs are
increasingly being administered to AMI patients, especially to
those with NSTEMI. Heparin works by inhibiting thrombin gen-
eration and arterial thrombosis. Heparin augments the fibrinolytic
action of intravenous fibrinolytic agents and must be given concur-
rently with the newer fibrin-specific fibrinolytics. Heparin reduces
mortality and the risk of re-infarction in patients with AMI.
Figure 68.3. Therapy for acute myocardial infarction (MI) and its effect on mortality. ACE-I, angiotensin-converting enzyme inhibitor; ACS,
acute coronary syndrome; EF, ejection fraction.
643
UFH administration should precede or be given concurrently
with fibrin-specific fibrinolytics in all circumstances. We recom-
mend weight-adjusted UFH administration as a bolus of 60 U/
kg (maximum, 4,000 U) followed by an infusion of 12 U/kg per
hour (maximum, 1,000 U/h) adjusted to a partial thromboplas-
tin time of 1.5 to 2.0 times the control. There is significant evi-
dence that over-anticoagulation with UFH can increase the risk
of death among STEMI patients after fibrinolysis. When a non-
selective fibrinolytic such as streptokinase is used, UFH can be
given selectively to patients with a high risk of systemic emboli,
including large myocardial infarcts (particularly anterior wall
infarcts), patients with atrial fibrillation or known left ventric-
ular thrombus, and those with severely reduced left ventricular
ejection fraction (<30%).
We recommend a weight-adjusted bolus dose of UFH of 50 to
70 U/kg accompanied by an infusion of 12 U/kg per hour (maxi-
mum, 1,000 U/h) in patients treated with primary PCI revascu-
larization for STEMI.
Enoxaparin, one of several LMWHs on the market, can also
be administered in patients with STEMI at the time of intrave-
nous fibrinolytic therapy or prior to PCI. It should be admin-
istered as a single 30 mg IV bolus followed immediately by a
1 mg/kg subcutaneous dose. Further doses are 1 mg/kg sub-
cutaneously every 12 hours for at least 48 to 72 hours in those
receiving fibrinolytic therapy or until appropriate to discontinue
following primary PCI. Consider dosing adjustments in patients
over age 75. Other LMWHs have not been approved for use with
intravenous fibrinolysis in the US.
The use of LMWH is an acceptable alternative to UFH for
patients with NSTEMI, and it should be dose adjusted for those
with significant renal dysfunction. The most commonly used
agents are enoxaparin and dalteparin. Both have been demon-
strated to be effective in NSTEMI.
644 VII Myocardial Infarction
The duration of heparin therapy during hospitalization for
AMI has not been well studied and there are few data from which
to extrapolate recommendations. We recommend routine use of
heparin therapy after fibrinolytic therapy for about 48 hours
or until the time of coronary angiography in patients without
another indication for heparin. The continued infusion of hepa-
rin following a primary PCI procedure is generally not indicated
and its continued use may delay vascular sheath removal.
• Heparin therapy should be initiated very early in the initial stabili-
zation of patients with AMI.
• UFH should be dosed in a weight-adjusted manner for patients
receiving intravenous fibrinolytic therapy: 60 U/kg (maximum,
4,000 U) followed by an infusion of 12 U/kg per hour (maximum,
1,000 U/h) and continued for about 48 hours unless coronary
angiography is performed earlier.
• The role of LWMH in NSTEMI is well established.
• The use of enoxaparin in STEMI is supported by clinical trial data
and is FDA approved.
• Heparin therapy may be discontinued after coronary angiography
or PCI unless there is another reason to initiate anticoagulation.
β-Blockers
β-Blockers are important in the initial stabilization of patients
with AMI. They block the β-adrenergic receptors on the myo-
cardium and reduce myocardial oxygen demand, enhancing
ventricular electrical stability and weakly inhibiting platelet
aggregation.
Pooled data from numerous randomized clinical trials, includ-
ing pooled Swedish trial data and the GUSTO-1 trial, demon-
strated that patients given β-blocker therapy early for AMI had
lower risks of death, heart failure, electrical instability (including
ventricular arrhythmias), and high-grade heart block.
The ACC/AHA AMI guidelines state that oral administration
of β-blocker therapy in the initial hours of hospitalization for
AMI has level 1 evidence for nearly all patients, except those with
severe heart failure and cardiogenic shock. The use of β-blocker
therapy in patients with advanced heart failure can be initiated
slowly after initial stabilization and treatment of the heart fail-
ure. The guidelines suggest that there is level 2a evidence for the
use of intravenous β-blockers in the early hours of hospitaliza-
tion for AMI: early clinical trials demonstrated the benefit, while
later trials have been inconclusive.
β-Blockers should not be administered to patients with pul-
monary edema or to those with bradycardia (heart rate <50 beats
per minute), hypotension, cardiogenic shock, PR-interval 0.24
seconds or longer, and those with second or third-degree AV
block. Additionally, β-blockers should be withheld from those
with a history of significant asthma. β-Blockers can be admin-
istered to patients with COPD when initiated slowly and titrated
carefully. It is our practice to administer intravenous β-blocker
therapy in the ED or prehospital setting whenever possible in
patients with AMI. Patients are given follow-up oral β-blocker
therapy within a few hours after the initial intravenous dosing,
and this therapy is continued to hospital discharge and beyond
if tolerated.
• Oral β-blocker therapy should be administered in the initial hours
of hospitalization (level 1 evidence) unless specifically contraindi-
cated in all AMI patients.
• Intravenous β-blocker therapy provides additional benefit when
administered very early in the ED or prehospital setting.
• β -Blocker therapy should be withheld in patients with pulmonary
edema until they are medically stabilized.
• β -Blocker therapy should not be given to those with severe brady-
cardia, conduction system disease as evidenced by significant first
degree AV block, second or third degree AV block, and patients
with known or suspected reactive airways disease (asthma).
• β -Blockers can be safely given to those with nonreactive COPD.
• A specific exception to the general use of beta blockers in AMI
is in patients with suspected AMI or chest pain due to acute
cocaine intoxication, in which β receptor blockade is contraindi-
cated because of the possibility of further coronary artery vaso-
constriction due to loss of the beta receptor mediated coronary
vasodilatation.
Nitroglycerin
Sublingual nitroglycerin at a dose of 0.4 mg every 5 minutes for a
total of three doses is indicated in patients presenting with chest
pain consistent with myocardial ischemia. Intravenous nitroglyc-
erin is indicated in AMI patients with persistent ischemic chest
pain, pulmonary edema, systemic hypertension, or heart failure.
Nitrates are contraindicated in patients with hypotension (systolic
blood pressure <90 mm Hg or ≥30 mm Hg below the patient’s
baseline blood pressure), marked bradycardia or tachycardia,
right ventricular infarction, hypertrophic cardiomyopathy, severe
aortic stenosis, or patients who have used of a phosphodiesterase
inhibitor for erectile dysfunction or pulmonary hypertension
within the previous 36 hours.
Although no single trial’s data prove that mortality is reduced
with early nitrate use in myocardial infarct patients, trial meta-
analysis is suggestive of benefit, and many patients experience sig-
nificant symptomatic improvement after nitrate therapy. Nitrates
mediate their benefit by reducing both cardiac preload (venous
dilation) and afterload (peripheral arterial dilation) and by direct
vasodilation of the coronary vasculature by a mechanism inde-
pendent of endothelial function, which is especially important
in patients with endothelial dysfunction and atherosclerosis; this
in turn improves coronary blood flow and myocardial oxygen
delivery. Nitrates also dilate coronary collateral vessels, potentially
creating a favorable subendocardial-epicardial blood flow ratio.
Nitrates are harmful in patients with hypotension, bradycar-
dia, or suspected right ventricular infarction as well as in those
who have received a phosphodiesterase inhibitor for erectile dys-
function within the preceding 36 hours. We recommend an ini-
tial nitrate dose of 0.4 mg administered sublingually followed by
the initiation of intravenous nitroglycerin or administration of
oral isosorbide dinitrate. Nitrates should generally be continued
beyond the initial 24 hours of hospitalization only to aid in the
symptomatic treatment of ischemia or heart failure.
• Nitrates reduce coronary ischemia by reducing cardiac preload and
afterload and by vasodilation of the coronary vasculature.
• Nitrates may reduce myocardial ischemia by dilating the coronary
collateral vasculature.
• Nitrates provide symptomatic relief of angina but have not been
conclusively demonstrated to reduce mortality in AMI.
• Nitrates should not be continued beyond the initial day of hospi-
talization except to aid in the symptomatic treatment of angina or
heart failure.
P2Y12 Antiplatelet Therapy
There are three P2Y12 agents which are FDA approved for
use in patients with STEMI or NSTEMI who are undergoing
 68 Adjunctive Therapy in Acute Myocardial Infarction
PCI: clopidogrel, prasugrel and ticagrelor. Two of the agents,
clopidogrel and ticagrelor, are approved for use in NSTEMI with
or without PCI.
Clopidogrel
Clopidogrel, a potent antiplatelet agent, is a thienopyridine com-
pound that inhibits platelet aggregation through modification
of the platelet adenosine diphosphate pathway. The CURE trial
randomly assigned patients with unstable angina or NSTEMI to
receive aspirin or aspirin plus clopidogrel, with a primary end
point of CV death, myocardial infarction, or stroke. CURE dem-
onstrated a significant reduction in the combined primary end
point with clopidogrel plus aspirin versus aspirin alone (9.3% vs.
11.4 %), primarily from fewer myocardial infarctions but at the
cost of more serious bleeding complications (3.7% vs. 2.7%).
The use of clopidogrel in NSTEMI is widespread in the
United States, in large part because of the widespread use of
coronary angiography and percutaneous coronary revasculari-
zation. Clopidogrel or prasugrel is mandated in all patients who
receive a coronary stent at the time of primary PCI and should be
continued for a minimum of 6 – 8 weeks in patients treated with
bare metal stents, and there are data to justify extending therapy
for up to 1 year. The use of clopidogrel or prasugrel following
PCI with DESs should be for at least 12 months; the duration of
therapy may be increased if the patient is considered at high risk
of stent thrombosis.
Clopidogrel can also be administered concomitantly with
intravenous thrombolysis, where it has been demonstrated to
reduce the risk of death as well as the combined end point of
death, nonfatal stroke, and nonfatal recurrent re-infarction
Clopidogrel dosing is based upon the clinical situation being
treated. It should be administered as a loading 300-mg dose in
patients with NSTEMI and dosed at 75 mg per day thereafter for
up to one year. It should be administered as a 75-mg initial dose
at the time of initiation of intravenous fibrinolytic therapy and
continued daily at 75 mg/day for at least one year. The dosing
with PCI is more controversial. It is approved at a 300-mg load-
ing dose followed by 75 mg/day. It is the practice of many inter-
ventional cardiologists to administer a loading dose of 600 mg
of clopidogrel at the time of PCI in order to achieve more rapid
and complete platelet inhibition followed by 75 mg/day following
stent implantation.
• Clopidogrel is widely used after implantation of a bare metal coro-
nary stent for at least 6-8 weeks after PCI, and may be extended
for up to 1 year.
• Clopidogrel should be administered for at least 12 months after
implantation of any DES.
• Clopidogrel 75 mg should be administered concurrently with intra-
venous fibrinolytic therapy and continued at 75 mg/day for up to
one year.
• Clopidogrel should be loaded as 300 to 600 mg at the time of PCI
and continued as 75 mg daily after stent implantation.
Prasugrel
Prasugrel is a second FDA approved P2Y12 agent for AMI. It is
similar to clopidogrel in structure but is converted more quickly
to its active metabolite and generally achieves greater platelet
inhibition. Moreover, the drug does not need hepatic activa-
tion, which might work better for patients with genetic variants
regarding the enzyme CYP2C19.
645
Prasugrel was tested in the TRITON- TIMI 38 trial in patients
with STEMI and NSTEMI who were undergoing planned PCI.
Prasugrel was superior to clopidogrel with regard to a reduction
in a combined primary CV end point (CV death, non-fatal myo-
cardial infarction and non-fatal stroke). The benefit observed
with prasugrel was offset by a higher major bleeding risk com-
pared with clopidogrel. The FDA approved prasugrel for use in
acute coronary syndrome populations undergoing PCI but sug-
gested restricting its use in patients who are elderly (age > 75),
have a prior history of transient ischemic attack or stroke and/
or have additional risk factors for bleeding such as those with a
body weight < 60 kg.
Ticagrelor
Ticagrelor is a reversible and direct acting P2Y12 inhibitor. Like
prasugrel and clopidogrel ticagrelor blocks the platelet ADP
receptors, subtype P2Y12, but in contrast to the other antiplate-
let drugs, ticagrelor has a binding site different from other ADP
inhibitors, making it an allosteric antagonist. Its metabolism is
also different in that it does not require intestinal and hepatic
activation to work. Studies also show that it achieves greater and
more complete inhibition of platelet activity than clopidogrel in
a faster time.
Ticagrelor was tested against clopidogrel in the PLATO trial.
All patients were on baseline aspirin therapy. The primary end
point of the PLATO trial was the occurrence of the compos-
ite end point of death from any vascular cause, recurrent AMI
or new stroke at 12 months; this end point was significantly
reduced in those randomized to ticagrelor compared to those on
clopidogrel.
Ticagrelor was superior to clopidogrel in patients only when
a lower dose of aspirin (<100 mg) was used. There is no clear
explanation for this observation but the FDA has suggested
reducing the aspirin dose to less than 100 mg daily if concurrent
ticagrelor therapy is used.
Ticagrelor was associated with a higher rate of major non-
coronary artery bypass grafting bleeding in the PLATO trial
as well as unexplained higher rates of dyspnea as well as slight
increases in serum creatinine and uric acid compared with those
randomized to clopidogrel.
Renin-Angiotensin-Aldosterone
Antagonist Therapy
The use of inhibitors of the renin-angiotensin-aldosterone axis
is an important component of managing the AMI patient. These
agents reduce mortality, risk of reinfarction, and risk of heart
failure (Figure 68.4).
ACE Inhibitors
ACE inhibitors block the conversion of angiotensin I to angio-
tensin II by inhibiting the ACE. This results in less sodium
retention and less arterial vasoconstriction. ACE inhibitors also
reduce circulating levels of PAI-1, a known prothrombotic sub-
stance in the plasma (Figure 68.5).
These agents have been widely tested in patients with AMI,
and overall they significantly reduce the risks of death, reinfarc-
tion, and heart failure. The absolute risk reduction in these end
points is most evident in AMI patients with reduced left ventric-
ular ejection fractions (<40%), evidence of systolic heart failure,
or with an anterior AMI. The benefit of routine ACE inhibitor
646 VII Myocardial Infarction

Figure 68.4. Medical management of acute myocardial infarction. HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein
cholesterol; TG, triglyceride.

Figure 68.5. Renin-angiotensin-aldosterone system. ACE, angiotensin-converting enzyme; AT, antithrombin.

 68 Adjunctive Therapy in Acute Myocardial Infarction
therapy in all patients with coronary artery disease is less well
established but has been in most AMI patients by the ACC/AHA
guidelines.
The greatest benefit of ACE inhibitor use in AMI patients
when the agents are initiated very early during hospitaliza-
tion, but hypotension should be avoided. There is no proven
clinical difference among the available ACE inhibitors with
regard to short-term or long-term efficacy following AMI
(Table 68.3).
• ACE inhibitors should be initiated as early as possible in patients
with AMI, absent shock or hypotension
• Avoid hypotension when titrating ACE inhibitors.
• Most AMI patients should continue to receive ACE inhibitors
indefinitely.
Angiotensin Receptor Blockers
Much of the endogenous production of angiotensin II in humans
occurs through non-ACE pathways. The largest trial of patients
who had AMI complicated by heart failure or left ventricu-
lar dysfunction was the VALIANT trial, which enrolled more
than 16,000 patients and randomly assigned them to receive
captopril, valsartan, or a combination of these agents. The
results demonstrated that valsartan was equivalent to captopril
for all outcomes—survival and freedom from heart failure or
re-infarction. This trial conclusively established that there was
a role for ARB antagonists in patients with AMI and that this
class of agent could be used interchangeably with ACE inhibi-
tors with fewer side effects, principally cough and hyperkalemia
(Figure 68.6).
• ARB agents should be used in AMI patients who are intolerant of
ACE inhibitors.
• ARB agents reduce the risks of mortality and heart failure equiva-
lent to outcomes with ACE inhibitors in patients with heart failure
or reduced left ventricular function.
• Valsartan and candesartan are currently FDA-approved ARB
agents for use in post-AMI patients with clinical heart failure or
reduced left ventricular function.
• Olmesartan, an intestinally activated ARB, has recently been
linked to a rare side effect in a small number of patients. This
condition has diagnostic features that resemble the clinical and
intestinal biopsy findings of celiac enteropathy but is tissue trans-
glutaminase antibody negative and unresponsive to a gluten-free
diet; however it does respond to discontinuation of olmesartan.
Aldosterone Antagonists
The use of aldosterone inhibition therapy is also an important
adjunctive medical therapy in post-AMI patients. Agents that
block aldosterone include spironolactone and eplerenone. The
EPHESUS trial randomly assigned post-AMI patients to receive
placebo or eplerenone; eplerenone therapy significantly reduced
the risk of all-cause death, sudden cardiac death, and rehos-
pitalization due to a CV mechanism (Figure 68.7). The ACC/
AHA guidelines for AMI suggest that long-term aldosterone
blockade should be prescribed for post-STEMI patients with-
out significant renal dysfunction (creatinine ≤2.5 mg/dL in
men and ≤2.0 mg/dL in women) or hyperkalemia (potassium
≤5.0 mEq/L) who are already receiving therapeutic doses of an
ACE inhibitor, have a left ventricular ejection fraction of 40%
or less, and have either symptomatic heart failure or diabetes
(level of evidence, A).
647
• Long-term aldosterone blockade should be prescribed for post-
STEMI patients without significant renal dysfunction (creatinine
≤2.5 mg/dL in men and ≤2.0 mg/dL in women) or hyperkalemia
(potassium ≤5.0 mEq/L) who are already receiving therapeutic
doses of an ACE inhibitor, have a left ventricular ejection fraction
≤40%, and have either symptomatic heart failure or diabetes.
Statin Agents and Other
Lipid-Lowering Therapies
The use of statin agents in AMI is well established and safe.
Many patients arrive at the hospital already receiving statin ther-
apy, which should be continued. Statins primarily decrease LDL
cholesterol and may very modestly increase HDL cholesterol,
but they also likely secondarily stabilize atherosclerotic plaques,
reverse endothelial dysfunction, reduce arterial wall inflamma-
tion, and decrease plasma concentrations of inflammatory CRP.
The cardioprotective benefit of statin therapy after AMI is well
proven from the CARE, 4S, and LIPID trials.
Results of recent lipid trials, including MIRACL (atorvas-
tatin vs pravastatin) and PROVE-IT (atorvastatin), suggested
an early benefit from aggressive high-potency statin therapy
(atorvastatin) initiated early (within 24-96 hours) in patients
with AMI and unstable angina, long before the established
long-term benefits of LDL lowering were likely operative. The
A to Z trial (simvastatin) did not reproduce this early benefit
until a second analysis was published which looked at LDL and
CRP values attained by one month from hospital discharge.
The PROVE-IT TIMI 22 and the A to Z trials both demon-
strated the best outcomes in patients where the LDL and CRP
values one month following discharge were <70 mg/dl and
2 mg/L respectively. It is our practice to initiate in-hospital,
early, potent statin administration in almost all patients with
AMI, with a treatment goal of LDL at 70 mg/dL or less and
CRP at 2 mg/L or less.
The use of other lipid-lowering therapies in patients with AMI
has less clinical evidence of benefit than the use of statin therapy.
The best studied classes of agents are the fibric acid deriva-
tives, such as gemfibrozil and fenofibrate and the class of drugs
containing nicotinic acid. All non-statin lipid-lowering agents,
including niacin, the fibrates, ezetimibe, and bile acid–binding
resins, should be used to augment statin therapy rather than
replace statins unless the patient is statin intolerant (Figure 68.8
and 68.9).
• Statin therapy should be initiated in all patients with AMI as early
as possible during hospitalization and preferably on admission.
• The goal of statin therapy is LDL ≤70 mg/dL and CRP ≤2 mg/L
after hospital discharge.
• Other lipid-lowering therapy should be used only to achieve LDL
≤70 mg/dL, HDL ≥40 mg/dL, and CRP <2 mg/L when statin ther-
apy is insufficient or not tolerated.
Management of Diabetes Mellitus
It is imperative to treat patients with diabetes and AMI very
aggressively. We recommend moderate glycemic control dur-
ing the initial period of hospitalization (fasting morning glucose
values 140-200 mg/dL) and aggressive diabetic treatment after
discharge from the hospital. All diabetic patients should receive
a potent statin and some may benefit from an added fibrate if nec-
essary to increase the level of HDL cholesterol.
The choice of which oral agent to use in treatment of type 2
diabetes is somewhat controversial. The data on metformin are
Table 68.3. Summary of Adjunctive Medical Therapies in Acute Myocardial Infarction
Agent When to Initiate When to Stop Initial Dose Dose Goal Intended Action Common Side Effects
Aspirin Prehospital or in ED Only with aspirin allergy 162–325 mg 62–325 mg Antiplatelet and GI bleeding
or intolerance or bleeding antiinflammatory actions Dyspepsia
Oxygen Prehospital or in ED After 24 h and when 2 L/min O2 saturation >90% Improve oxygenation Sore or dry nose, epistaxis
O2 saturation >90%
Morphine Prehospital or in ED Free of pain and anxiety Initially 2–4 mg, with Titrate to response Pain and anxiety relief Respiratory depression
increments of 2–8 mg Hypotension
at intervals of 5–15 min
Nitroglycerin Prehospital or in ED Hypotension Sublingually, 0.4 mg; repeat Titrate to response Dilatation of coronary arteries Hypotension
every 5 min up to 3 times and arterioles Hemodynamic collapse
if needed Venous dilatation with
IV, 5–10 μg/min; gradually decreased preload
increased Relief of coronary spasm
Heparin ED arrival 24-48 h 60 U/kg bolus 12 U/kg/h Anti-thrombotic action Bleeding
β-Blocker (metoprolol) ED arrival Give only 1 time 5–15 mg IV HR 50–60 Slow HR Bradycardia
IVa Lower BP Hypotension
Orally ~6 h after IV dose Only with intolerance 25 mg BID HR 50–55 at rest Treat ischemia Low BP, lightheadedness
Reduce or eliminate ischemia Erectile dysfunction
and angina
ACE inhibitor (lisinopril) First 12 h Only with intolerance 2.5 mg 10–20 mg Lower BP Low BP, lightheadedness
or Reduce cardiac remodeling Erectile dysfunction
ARB (valsartan) Lower PAI-1 values Low BP, lightheadedness
80 mg BID 160 mg BID Lower BP Erectile dysfunction
Reduce cardiac remodeling
Aldosterone antagonist Days to weeks Only with intolerance 12.5–25 mg 50 mg Reduce cardiac remodeling Low BP
(spironolactone) Reduce risk of heart failure Hyperkalemia
Renal failure
Statin (atorvastatin) ED arrival to first day Only with intolerance 40 mg LDL-C <70 mg/dL Reduce risk of recurrent AMI Muscle pain
CRP <2 mg/L Lower risk of death LFT elevation
Lower LDL Rarely rhabdomyolysis
Lower CRP
Lower PAI-1
Clopidogrel After PCI 1 mo—bare metal stent; 300–600 mg 75 mg Reduce risk of stent thrombosis Rash
6 mo—DES and reinfarction Bleeding, bruising

Abbreviations: ACE, angiotensin-converting enzyme; AMI, acute myocardial infarction; ARB, angiotensin II receptor blocker; BID, twice daily; BP, blood pressure; CRP, C-reactive protein; DES, drug-
eluting stent; ED, emergency department; GI, gastrointestinal tract; HR, heart rate; IV, intravenously; LDL-C, low-density lipoprotein cholesterol; LFT, liver function test; PAI-1, plasminogen activator
inhibitor-1; PCI, percutaneous coronary intervention.
a
The benefits of ACE inhibition, ARB blockade, aldosterone antagonism, statins, and non–intrinsic sympathomimetic activity β-blockade are class effects, with multiple alternative medications available
in all classes.
 68 Adjunctive Therapy in Acute Myocardial Infarction 649

Acute infarction Infarct expansion Global remodeling

(hours) (hours to days) (days to months)

Figure 68.6. Left ventricular remodeling after myocardial infarction.

strong, and this agent should be used as a first line therapy if
at all possible. The second line of therapy should be a dipepti-
dyl peptidase-4 (DPP-4) antagonist of a glucagon-like peptide-1
(GLP-1) agonist in most patients. The sulfonylurea compounds
are also appropriate to use in diabetic patients with AMI. They
are well established but have a theoretical concern with regard to
promoting coronary vasoconstriction.
There was some evidence that the thiazolidinediones, which
are insulin sensitizers, may benefit the post-AMI patient, espe-
cially when given in combination with metformin, but now this
class has fallen out of favor. These agents promote moderate
weight gain and should be avoided in the early post-AMI popu-
lation in most circumstances. They should be used with caution
in patients with mild heart failure, and should not be used at all
in patients with advanced heart failure. Most patients with renal
dysfunction and heart failure should probably receive insulin
therapy.
In addition to glucose-modifying therapies, diabetic patients
should be encouraged to exercise regularly and maintain ideal
body weight. Their LDL cholesterol values should be aggres-
sively decreased to less than 70 mg/dL.
• Hypoglycemic therapy should be initiated to maintain hemoglobin
A1c at less than 7% (level of evidence, B).
• Thiazolidinediones should not be used in patients recovering from
STEMI who have New York Heart Association class III or IV heart
failure (level of evidence, B).
Long-term Adjunctive Therapy
Sublingual nitroglycerin should be prescribed for all post-AMI
patients to use at home for recurrent ischemic pain. All AMI
patients should receive long-term aspirin and β-blockers unless
contraindicated. The dose of aspirin may need to be adjusted if
clopidogrel is used concurrently.
• β -Blocker therapy should be continued indefinitely in nearly all
patients with AMI.

Figure 68.7. Rate of sudden death from cardiac causes in the Eplerenone Post–Acute Myocardial Infarction Heart Failure Efficacy and Survival
Study (EPHESUS). CI, confidence interval; RR, relative risk. (Previously published. See “Credit Line” section.)
650 VII Myocardial Infarction

Figure 68.8. Survival among patients who received statin therapy initiated within 48 hours of hospitalization versus no statin therapy. (Previously
published. See “Credit Line” section.)

Figure 68.9. Cumulative incidence of recurrent myocardial infarction (MI) or death from coronary causes, according to the achieved levels of
low-density lipoprotein cholesterol (LDL chol) and C-reactive protein (CRP). (Previously published. See “Credit Line” section.)
 68 Adjunctive Therapy in Acute Myocardial Infarction 651

• Aspirin should be continued indefinitely in all patients with AMI acute myocardial infarction
AMI unless contraindicated by aspirin allergy or major bleeding ARB angiotensin-receptor blocker
complications. AV atrioventricular
COPD chronic obstructive pulmonary disease
• Heparin and nitroglycerin therapies can be discontinued during the
CRP C-reactive protein
initial days of hospitalization for AMI.
DES drug-eluting stent
ED emergency department
Lifestyle Risk Factor Management FDA US Food and Drug Administration
HDL high-density lipoprotein
Lifestyle modification is at least as important as medications LDL low-density lipoprotein
in the long-term treatment of AMI. Moderate alcohol intake LMWH low-molecular-weight heparin
appears to have a cardioprotective effect on the heart, but heavy NSTEMI non–ST-segment elevation myocardial infarction
alcohol use has an adverse effect. The AHA defines moderate PAI-1 plasminogen activator inhibitor 1
alcohol consumption as an average of one to two drinks per day PCI percutaneous coronary intervention
for men and one drink per day for women. (A drink is 12 oz of STEMI ST-segment elevation myocardial infarction
beer, 4 oz of wine, 1.5 oz of 80-proof spirits, or 1 oz of 100-proof UFH unfractionated heparin
spirits.) Alcohol in large doses elevates triglycerides and blood
pressure, but alcohol in lower doses increases cardioprotective Names of Clinical Trials
HDL cholesterol. Flavonoids in red wine may also be beneficial.
A to Z Aggrastat to Zocor
In patients who do not drink alcohol, red grape juice or cranberry
CARE Cholesterol and Recurrent Events
juice is indicated. CURE Clopidogrel in Unstable Angina to Prevent
Recurrent Events
• Patients with AMI who smoke must be counseled to immediately EPHESUS Eplerenone Post–Acute Myocardial Infarction
stop all smoking and avoid second hand smoke exposure. Heart Failure Efficacy and Survival Study
• The use of nicotine replacement therapy and bupropion therapy to GUSTO-1 Global Utilization of Streptokinase and TPA
aid in smoking cessation is safe and effective in AMI patients. for Occluded Arteries
• All AMI patients should receive rehabilitation and exercise LIPID Long-term Intervention With Pravastatin in
counseling. Ischemic Disease
MIRACL Myocardial Ischemia Reduction With
• Alcohol in moderation is beneficial after AMI.
Aggressive Cholesterol Lowering
• Patients who are obese or overweight at the time of AMI should PLATO Ticagrelor Compared With Clopidogrel by
receive intensive instruction in lifestyle modification. Geographic Region in the Platelet Inhibition
• Morbidly obese patients may be candidates for bariatric surgery and Patient Outcomes
for refractory obesity management at least 6 months after AMI. PROVE-IT TIMI 22 Pravastatin and Atorvastatin Evaluation and
Infection Therapy
TIMI Thrombolysis in Myocardial Infarction
TRITON-TIMI 38 Trial to Assess Improvement in Therapeutic
Abbreviations
Outcomes by Optimizing Platelet Inhibition
ACC/AHA American College of Cardiology/American With Prasugrel
Heart Association VALIANT Valsartan in Acute Myocardial Infarction
ACE angiotensin-converting enzyme Trial
ADP adenosine diphosphate 4S Scandinavian Simvastatin Survival Study
 69
Arrhythmia Complications of Acute
Myocardial Infarction
JOSEPH G. MURPHY, MD, and MARGARET A. LLOYD, MD
Arrhythmias after MI are common, and about 80% of hospi-
talized patients will have an arrhythmia during the peri-infarct
period, of which about 10% will be life-threatening.
Supraventricular Arrhythmias After MI
Sinus Bradycardia
Sinus bradycardia is the most common arrhythmia occurring
within hours after MI and may occur in up to 40% of inferior
and posterior infarctions. Bradycardia may be related to auto-
nomic imbalance or to atrial and sinus node ischemia (or to
both). Profound bradycardia may predispose the patient to ven-
tricular ectopy. This arrhythmia usually resolves spontaneously,
and treatment is reserved for patients with hemodynamically
symptomatic arrhythmias or bradycardia-dependent ventricular
arrhythmias. Atropine is often successful in treating sympto-
matic bradycardia, but it may cause transient rebound tachycar-
dia. Temporary pacing is rarely required (Table 69.1).
Sinus Tachycardia
Sinus tachycardia may occur in up to one-third of patients in the
peri-infarct period, especially in those with anterior MI. The
ischemic left ventricle may have a relatively fixed stroke volume;
thus, augmentation of cardiac output is primarily dependent on
an increase in heart rate. Sinus tachycardia may also occur as
a result of sympathetic stimulation from locally released and
circulating catecholamines, concurrent anemia, hypovolemia
or hypervolemia, hypoxia, pericarditis, inotropic drugs, pain,
or fear. Treatment includes optimizing hemodynamics and oxy-
genation, correction of anemia and electrolyte and acid-base
Abbreviations are expanded at the end of this chapter.
652
abnormalities, pain control, and anxiolytic agents. β-Blockers
are indicated for patients without evidence of significant LV
dysfunction or hypovolemia. Persistent sinus tachycardia as
an early manifestation of heart failure is an indicator of poor
prognosis.
Premature Atrial Contractions
Premature atrial contractions may be present in up to one-half of
patients with MI and may be due to atrial or sinus node ischemia,
atrial infarction, pericarditis, anxiety, or pain. The combination
of atrial asystole and a rapid ventricular rate markedly decreases
cardiac output and increases myocardial oxygen demands.
Premature atrial contractions have no prognostic significance
after MI.
Atrial Fibrillation
New atrial fibrillation occurs in about 10% to 20% of patients
with MI and is usually transient. It may be due to atrial or sinus
node ischemia, associated RV infarction, pericarditis, heart
failure, or increased atrial pressures. It usually occurs in older
patients, more often in those with a history of hypertension,
mitral regurgitation, and larger left atria. New atrial fibrilla-
tion in the peri-infarct period is associated with a higher infarct
mortality.
Atrial systole may contribute up to one-third of the cardiac
output in patients with an ischemic left ventricle. Patients with
persistent or refractory atrial fibrillation in the peri-infarct per-
iod have higher PCWPs and lower ejection fractions and are in
a lower Killip class overall than patients who maintain sinus
rhythm.
 69 Arrhythmia Complications of Acute Myocardial Infarction
Table 69.1. Indications for Temporary Pacing in the
Peri-Infarct Period
Sinus bradycardia with hypotension, bradycardia-dependent ventricular
arrhythmias, angina, syncope/presyncope, or congestive heart failure
and refractory to atropine
Accelerated idioventricular rhythm with symptomatic rate <40 bpm
Prolonged (>3 s) sinus pauses
Atrial fibrillation with inadequate ventricular rate
Asystole
Mobitz II second-degree block
Third-degree (complete heart) block
New or progressive bifascicular block
Abbreviation: bpm, beats per minute.
Ventricular Arrhythmias After MI
Ventricular Fibrillation
Primary VF is early VF within 48 hours of infarction that is
directly related to the infarct mechanism. Many studies have
reported that the incidence of primary VF in MI is about
5%-10% among patients in whom a rhythm other than asystole
is documented. It is important to note that VF occurs with-
out antecedent-warning arrhythmias, such as frequent PVCs,
in over 50% of these patients. The true incidence of primary
VF is probably much higher because an estimated one-half of
all patients with coronary artery disease die of sudden cardiac
death, many out of hospital, presumably due to VF. Factors
associated with an increased incidence of VF include a history
of current or past smoking, male sex, ST-elevation MI, LBBB,
larger infarcts, hypotension, and hypokalemia. Patients with
anterior MI and VF have a worse long-term prognosis than those
with VF associated with inferior MI. VF may occur at the time
of initial arterial occlusion by thrombus or later at the time of
culprit artery reperfusion following successful thrombolysis or
PCI. Preinfarction angina is protective against VT, possibly due
to collateral vessel formation and ischemic myocardial precon-
ditioning. Primary VF predicts an increased in-hospital mortal-
ity but no increase in later mortality in patients who survive
their acute infarct to hospital discharge. Thus, early peri-infarct
VF is not an indication for AICD placement.
Late VF beyond 48 hours of infarction is associated with
an increased risk of both in-hospital and late mortality. It is
associated with ongoing MI, failure of reperfusion, and heart
failure. Patients with late VF should be considered for AICD
implantation.
• β -Blockers decrease the incidence of lethal ventricular arrhyth-
mias, including VF, in the peri-infarct period.
Ventricular Tachycardia
NSVT occurs in up to 40% of patients in the first 48 hours of
infarction: it is usually transient and benign. Sustained mono-
morphic VT (lasts 30 seconds or more at a heart rate of >100
beats per minute) occurs in 2% of patients with ST-elevation MI
and 1% of patients with non–ST-elevation MI within 48 hours
of infarction. Early sustained VT, in contrast to early NSVT or
early VF, may be a marker of long-term risk.
Late VT is associated with transmural infarction, LV dysfunc-
tion, hemodynamic deterioration, and a markedly higher mortal-
ity, both in-hospital and long term.
653
Accelerated Idioventricular Rhythm
AIVR is an ectopic ventricular rhythm consisting of three or
more consecutive ventricular beats with a rate that is faster than
the normal ventricular escape rate of 30 to 40 beats per minute
but slower than VT (>100–120 beats per minute). Onset and off-
set usually are gradual, and isorhythmic dissociation is often
present. AIVR has been reported in about 40% of cases of MI,
especially with early reperfusion. The incidence is equal among
patients with inferior or anterior infarcts and is not related to
infarct size. The presence of AIVR during the peri-infarct period
is not correlated with increased mortality or incidence of VF.
AIVR may also be seen with digitalis toxicity, myocarditis, and
cocaine use. Symptoms may be related to loss of AV synchrony
or slow ventricular rates (or both).
Premature Ventricular Complexes
PVCs occur frequently during MI, but their significance in pre-
dicting VT and VF is unclear. Treatment of PVCs in the peri-
infarct period has not been shown conclusively to decrease the
incidence of malignant ventricular arrhythmias or to improve
mortality; in fact, in randomized trials in which PVCs were
treated prophylactically in the peri-infarct period with lidocaine,
pooled results showed an increased mortality. β-Blockers may
be the best option for treating PVCs and preventing malignant
ventricular arrhythmias.
Miscellaneous Considerations
Reperfusion Arrhythmias
Typically, AIVR has been credited with being a marker for reper-
fusion. However, any arrhythmia (or no arrhythmia) may occur
with reperfusion; conversely, AIVR may occur without reper-
fusion. Other clinical factors should be considered when decid-
ing whether reperfusion has occurred, such as resolution of
chest pain, improved hemodynamics, and normalization of ECG
changes. The occurrence of reperfusion arrhythmias is related to
size of infarct, length and severity of ischemia, rate of reperfu-
sion, heart rate, extracellular potassium concentration, and the
presence of congestive heart failure or LV hypertrophy (or both).
Asystole and Electromechanical Dissociation
Asystole and electromechanical dissociation (pulseless electrical
activity) occur in a small fraction of patients with MI and are
usually associated with large infarcts. The prognosis is extremely
poor, even with aggressive therapy. Defibrillation should be
attempted in patients with apparent asystole because the rhythm
may actually be fine VF.
Electrical Storm
An uncommon complication of MI is an electrical storm in which
repeated episodes of VF or hemodynamically unstable VT occur
within a 24-hour period. Treatment consists of alleviation of
underlying ischemia, monitoring for electrolyte and drug toxici-
ties, and intravenous β-blockade and amiodarone therapy.
T-Wave Alternans
T-wave alternans is a transient ECG finding usually seen with
ischemia and is most pronounced in leads overlying the affected
myocardium.
654 VII Myocardial Infarction

Other ECG Findings to atropine, administered intravenously in increments of 0.5 to 1 mg

to a total of 3 mg. Symptomatic bradycardia in anterior infarction is
Patients with regional pericarditis, which sometimes occurs after generally atropine resistant. Atropine does not increase infranodal
Q-wave infarctions, may present with PR depression but more conduction, may paradoxically slow conduction and the ventricular
commonly with atypical ST-segment and T-wave changes. These rate in Mobitz type II block, is ineffective in transplanted hearts, and
changes typically consist of gradual premature reversal of ini- rarely precipitates VF.
tially inverted T waves or persistent or recurrent ST-segment 4. First-degree heart block and second-degree AV type I block
elevation (or both). Persistent ST-segment elevation after infarct (Wenckebach block) are usually AV nodal and are frequently tran-
may be due to continuing ischemia or aneurysm formation, or it sient, whereas second-degree AV type II block and complete heart
may herald free-wall rupture. block are usually due to injury to the infranodal conduction system
and may be permanent.
LV free-wall rupture occurs in approximately 10% of cases of
5. Patients with second-degree AV block and anterior infarction can
fatal transmural MI. ECG findings include failure of the charac- progress to high-grade AV block very rapidly and pacing should
teristic evolution of the ST segment or T wave, or both. Persistent, begin at an early stage, whereas inferior infarctions usually progress
progressive, or recurrent ST-segment elevation in the absence of in a stepwise fashion and patients can be observed without tempo-
recurrent ischemia may be seen. Failure of the T wave to invert rary pacing if clinically stable.
or initial inversion followed by reversion to the upright may also 6. Inferior infarctions usually are paced for symptomatic bradycardia
be seen. Abrupt bradycardia responsive to atropine may occur and hypoperfusion, whereas anterior infarctions are paced early on
and is believed to mark the time of rupture. the basis of ECG criteria alone, even if asymptomatic.
Other ECG findings may include U waves with hypokalemia
and tall peaked T waves with hyperkalemia. First-Degree Heart Block
Approximately 5% to 10% of patients with MI have first-degree
Conduction Disturbances after MI heart block sometime during the peri-infarct period. Almost all
Conduction abnormalities after acute MI result from autonomic have supra-Hisian conduction abnormalities. First-degree heart
disturbances or interruption of the blood supply to the conduc- block may be associated with drugs that prolong AV conduction.
tion system. In most individuals, the RCA supplies blood to the
sinoatrial node (60% of cases) and the AVN (90% of cases); in Second-Degree AV Block (Mobitz
the other individuals, the LCX is the blood supply (sinoatrial Type I Block, Wenckebach Block)
node, 40%; AVN, 10%). The bundle of His is supplied from the
Wenckebach block may occur in up to 10% of cases of MI, typ-
AV branch of the RCA with a small contribution from the sep-
ically in inferior infarcts, and is due to increased vagal tone or
tal perforator of the LAD. After the His bundle divides into the
ischemia. Usually the conduction defect is in the AVN and, early
right and left bundles, the septal perforators of the LAD sup-
in the course of an MI, responds to atropine. Resolution usu-
ply the right bundle with collaterals from the right circumflex
ally occurs after 48 to 72 hours. Initial treatment is with atro-
artery and LCX. The left bundle, in turn, divides proximally into
pine and, rarely, temporary pacing for symptomatic bradycardia.
the left anterior and posterior fascicles. The left anterior fasci-
Late-occurring Wenckebach block is less sensitive to atropine
cle is supplied from the LAD, and the proximal portion of the
and may be due to recurrent ischemia. Very rarely, Wenckebach
left posterior fascicle receives a dual blood supply from the AV
block will progress to higher grades of block that require perma-
nodal artery, generally a branch of the RCA, and the LAD. The
nent pacing. The rhythm in the peri-infarct period has no effect
distal portion of the posterior fascicle is also supplied from two
on long-term prognosis (Figure 69.1).
sources: the anterior and posterior septal perforating arteries.
• Conduction in the left anterior fascicle (blood supply from the Second-Degree AV Block (Mobitz Type II Block)
LAD perforators) is very sensitive to ischemia.
Mobitz type II block occurs in 1% of cases of MI and is more
• The overall incidence of high-degree AV block in patients receiv-
common after anterior MI. There is a high risk of progression to
ing thrombolysis is about 10% with inferior infarcts and 3% with
anterior infarcts. higher degrees of block, including sudden complete heart block
• Heart block complicating inferior infarction increases in-hospital
mortality but not long-term mortality after hospital dismissal.
The principles of the treatment of heart block complicating
MI are as follows:
1. Anterior infarctions cause heart block because of septal injury
that leads to necrosis of the infranodal conduction system. Inferior
infarctions usually cause heart block because of activation of abnor-
mal cardiovascular reflexes or transient ischemic injury of the AVN.
Necrosis of the AVN is rare in inferior infarction because of the
presence of collateral vessels from the LAD in addition to the nor-
mal blood supply from the AV nodal artery, supplied by the RCA in
85% of cases and by the LCX in the remaining 15% of patients (left
dominant coronary circulation).
2. Inferior infarcts complicated by high-grade AV block are generally
associated with narrow QRS complex escape rhythms with ventricu-
lar rates between 40 and 60 beats per minute. Prognosis for recovery
of conduction within a week is good and mortality is low.
3. Symptomatic bradycardia within the first 24 hours of inferior infarc- Figure 69.1. Edema of the atrioventricular node after inferior wall
tion is often due to excess vagal activity (vagotonia) and may respond myocardial infarction.
 69 Arrhythmia Complications of Acute Myocardial Infarction
with ventricular asystole. Patients should have a temporary pac-
ing wire placed prophylactically at the first sign of Mobitz type II
block in the peri-infarct period. The conduction defect is more
likely to be infranodal than in Mobitz type I block, and most
patients should be treated with permanent pacing. If it is uncer-
tain whether permanent pacing is indicated, electrophysiologic
evaluation should be performed before hospital dismissal to
assess the integrity of the infranodal conduction system. Long-
term prognosis is related primarily to the size of the infarct rather
than to the conduction abnormality.
Third-Degree Heart Block
(Complete Heart Block)
Complete heart block may occur with either an anterior or an
inferior MI. With inferior infarcts, the conduction defect is likely
to be in the AVN, with escape rhythms exceeding 40 beats per
minute and exhibiting a narrow QRS complex. With an anterior
MI, the conduction defect is infranodal and the escape rhythm
(if present) is usually less than 40 beats per minute with a wide
QRS complex. Typically, complete heart block seen with ante-
rior MI is preceded by progressive fascicular, bundle, or Mobitz
type II block.
Temporary pacing may be required for complete heart block
in association with inferior MI if the patient is hemodynamically
unstable. Temporary pacing should always be used in patients
with anterior infarcts if progressive or complete heart block is
present. Permanent pacing is almost always required for high-
grade block with anterior MI; the prognosis is poor because of the
large amount of myocardium involved. Electrophysiologic evalu-
ation before hospital dismissal should be considered for patients
with anterior MI and transient complete heart block to assess the
integrity of the infranodal conduction system. Transient com-
plete heart block with inferior MI rarely requires permanent pac-
ing and usually resolves spontaneously.
Mechanisms and Significance of AV Block
• Anterior infarction: extensive necrosis of infra-His conduction system
• Multiple mechanisms in inferior MI
° Increased parasympathetic tone
° Ischemic “stunning” of AV node
° Increased local potassium concentration due to infarction
° Increased local release of adenosine
• Autopsy studies show that infarction of the AV node does not occur
Table 69.2. Short-term Mortality in Patients
With AV Block
Mortality, %
Trial No AV block AV block
TIMI-II 2.2 9.9
TAMI 4 20
GUSTO-I 6 21
Abbreviations: AV, atrioventricular; GUSTO-I, Global
Utilization of Streptokinase and Tpa for Occluded
Arteries; TAMI, Thrombolysis and Angioplasty in
Myocardial Infarction; TIMI-II, Thrombolysis in
Myocardial Infarction-II.
655
Table 69.3. Mortality Rates for
Patients With BBB
Mortality, %
Trial BBB No BBB
TAMI-9 8.7 3.5
GUSTO-I 18 11
Abbreviations: BBB, bundle branch block;
GUSTO-I, Global Utilization of Streptokinase
and tPA for Occluded Arteries; TAMI-9,
Thrombolysis and Angioplasty in Myocardial
Infarction-9.
• AV block is a marker for larger infarction and failure of reperfusion
• Short-term mortality is increased among patients with AV block
(Table 69.2)
Bundle Branch Block
New BBB has been reported in about 15% of cases of MI and
is associated with an increased risk of complete heart block,
congestive heart failure, cardiogenic shock, ventricular arrhyth-
mias, and sudden death. Most common is RBBB, with LBBB
and alternating BBB being less common. This may be related
to the discrete anatomical size of the right bundle compared
with the broad, fan shape of the left bundle. The correlation
between the infarct-related artery and the presence of BBB is
strong, with more than half of all BBBs occurring in infarcts
involving the LAD. Progressive infra-Hisian block indicates a
significant risk of sudden complete heart block and asystole, and
patients with progression should have temporary pacing wires
placed. Persistent BBB confers a significantly higher mortality
because of the large amount of myocardium that must be involved
in the infarct to include the bundle branches. Thrombolytic ther-
apy and catheter-based early reperfusion appear to decrease the
incidence of BBB in the peri-infarct period.
• Mortality rates for patients with BBB are significantly increased
(Table 69.3)
• Patients with BBB that develops in the hospital have higher mortal-
ity rates than those who present with BBB.
• Patients with blocks that are transient have mortality rates similar
to those of patients without blocks.
Intraventricular Block
New isolated left anterior hemiblock occurs in 3% to 5% of
patients with MI; new isolated left posterior hemiblock occurs in
1% to 2% of patients with acute MI. Anatomically, left posterior
hemiblock is larger; hence, a larger infarct is required to produce
the block. Mortality is greater among these patients. Left anterior
hemiblock in combination with a new R BBB is also indicative of
a larger infarct and higher subsequent mortality.
• BBBs and fascicular blocks are markers for larger infarctions.
• In up to 22% of patients with a new BBB, the block progresses to
a high grade AV block.
• A new bifascicular block with or without PR prolongation has the
highest likelihood of developing into a complete heart block.
• In approximately one-fourth of patients, the conduction abnormal-
ities are transient.
The blood supply to the cardiac conduction system is outlined
in Table 69.4.
656 VII Myocardial Infarction

Table 69.4. Blood Supply to the Cardiac Conduction System Abbreviations

and Papillary Muscles AICD automatic implantable cardioverter-defibrillator
SA node RCA 70% AIVR accelerated idioventricular rhythm
LCX or dual blood supply 30% AV atrioventricular
AV node RCA 80% AVN atrioventricular node
LCX or dual blood supply 20% BBB bundle branch block
Bundle of His LAD 80% ECG electrocardiogram, electrocardiographic
AV bundle RCA or dual blood supply 20% LAD left anterior descending coronary artery
Right bundle branch LAD LBBB left bundle branch block
Left bundle LAD
LCX left circumflex artery
LV left ventricular
anterior fascicle
MI myocardial infarction
Left bundle LAD and RCA
NSVT nonsustained ventricular tachycardia
posterior fascicle
PCWP pulmonary capillary wedge pressure
Abbreviations: AV node, atrioventricular node; LAD, left anterior descending PVC premature ventricular contraction
coronary artery; LCX, left circumflex coronary artery; RCA, right coronary RBBB right bundle branch block
artery; SA node, sino-atrial node. RCA right coronary artery
RCA is the dominant blood supply to the SA and AV nodes. RV right ventricular
LAD is the dominant blood supply to the bundle of HIS and the bundle VF ventricular fibrillation
branches.
VT ventricular tachycardia
The anterolateral papillary muscle has a dual arterial blood supply from the left
anterior descending coronary artery and either its diagonal branch or a marginal
branch of the left circumflex coronary artery.
The posteromedial papillary muscle has a single arterial blood supply from
either the left circumflex or right coronary artery, depending on coronary
dominance, which makes it particularly susceptible to ischemic injury.
 70

Mechanical Complications of Acute

Myocardial Infarction
JOSEPH G. MURPHY, MD, and MARGARET A. LLOYD, MD

Cardiogenic shock and heart failure are the most common causes Table 70.2. Forrester Classification and Invasive Monitoring
of death among patients hospitalized with acute MI. Findings
Class Finding PCWP, mm Hg CI, L/min per m2
Hemodynamic Classification of MI
I Normal hemodynamics ≤18 2.2
Patients with acute MI can be classified into four hemody-
II Good cardiac output, >18 ≥2.2
namic subsets on the basis of the cardiac examination (Killip
pulmonary congestion
classes I-IV) (Table 70.1) or invasive monitoring (Forrester clas- III Low cardiac output, no ≤18 <2.2
sification I-IV) (Table 70.2). Although the two classifications pulmonary congestion
overlap, they are not interchangeable for either prognosis or man- IV Low cardiac output, >18 <2.2
agement. Table 70.3 provides data on hemodynamic patterns in pulmonary congestion
cardiovascular disease.
Abbreviations: CI, cardiac index; PCWP, pulmonary capillary wedge pressure.

Cardiogenic Shock fluid replacement, whereas cardiogenic shock is associated with

Cardiogenic shock is persistent hypotension conventionally
defined as a systolic pressure of less than 80 to 90 mm Hg for more
than 30 minutes in the absence of hypovolemia. Hypovolemia
results in hypotension due to inadequate LVEDP, which is typi-
cally measured clinically through its surrogate—PCWP of less
than 12 mm Hg. Hypotension due to hypovolemia responds to
Table 70.1. Killip Class and Clinical
Examination Findings
Class Finding
I No S3 or rales
II Rales in less than half of lung field
III Rales in more than half of lung field
IV Cardiogenic shock
Abbreviation: S3, third heart sound.
Abbreviations and acronyms are expanded at the end of this chapter.
an elevated LVEDP and PCWP and is refractory to fluid chal-
lenge. Cardiogenic shock frequently is associated with oliguria
or anuria, metabolic acidosis, peripheral hypoperfusion, and
cerebral hypoxia. The cardiac index in cardiogenic shock is usu-
ally less than 2.0 to 2.2 L/min per m2.
The most common causes of cardiogenic shock include 1) large
LV infarct (usually >40% of the left ventricle), present in about
80% of shock patients; 2) RV infarct in 10% of shock patients;
and 3) mechanical complications of MI (ventricular septal defect,
acute mitral regurgitation, tamponade) in 10% of shock patients.
Cardiogenic shock also affects the viable perfused myocardium
surrounding the infarct zone by rendering it more prone to ische-
mic necrosis due to hypotension and poor perfusion. In patients
who develop cardiogenic shock, the shock develops early and is
evident at the time of hospital admission in only 10% to 20% of
cases, while in most patients shock is delayed several hours to
days after the initial myocardial insult.
The incidence of cardiogenic shock is about 8% of all patients
with acute MI, and mortality is about 80% with conservative
657
658 VII Myocardial Infarction

Table 70.3. Typical Hemodynamic Patterns in Cardiovascular Diseasea

Pressure Pulmonary
Arteriolar
Resistance
Condition RA PA PCW CI Index
Normal <6 mm Hg <28/12 mm Hg <18 mm Hg >2.4 L/min per m 2 <2 Wood unitsb
Tamponade High Variable Low Low Normal
Right ventricular infarction High Low-normal Low Low Normal
Acute pulmonary embolus High Normal-high Low Low Normal-high
Left ventricular failure Normal Normal-high High Low-normal Normal
High-output heart failure High Normal-high High High Normal
Right ventricular failure High Variable Low-normal Low-normal Normal
Cardiogenic shock High Normal-high High Low Normal
Septicemia Low Low-normal Low High Low
Chronic pulmonary hypertension High High Normal Low-normal High
Hypovolemia Low Low-normal Low Low Low
Abbreviations: CI, cardiac index; PA, pulmonary artery; PCW, pulmonary capillary wedge; RA, right atrial.
a
The values and descriptions reflect typical clinical scenarios, but there is significant patient-to-patient variation.
b
Multiply by 80 to convert Wood units to dynes/s per cm5.

management. Thrombolytics are generally ineffective after hypo-
tension has become established because the perfusion pressure
delivering the thrombolytic agent to the culprit vessel is low and,
even if effective, vessel reocclusion is common. Primary PCI is
generally the treatment of choice for cardiogenic shock while
inotropic support from an intra-aortic balloon pump or positive
inotropic drugs is also valuable. The clinical management and
results of randomized trials in cardiogenic shock are discussed
in detail in chapter 71 on cardiogenic shock.
Differential Diagnosis of Hypotension After MI
• Cardiogenic shock due to a large LV infarct
• RV infarction
• Papillary muscle rupture
• Ventricular septal rupture
• Free wall rupture
• Bradycardia and conduction abnormalities
• Ventricular tachycardia
• Hypovolemia
• Pulmonary embolus
• Dynamic LV outflow tract obstruction
Definition of Cardiogenic Shock
Cardiogenic shock is defined as decreased cardiac output with
evidence of insufficient tissue perfusion in the presence of
adequate intravascular volume (Killip class IV). Clinical signs
include oliguria; cool, cyanotic extremities; and altered mental
status. Hemodynamic criteria include sustained hypotension
(systolic blood pressure <80 mm Hg for >30 min), pulmonary
artery wedge pressure >18 mm Hg, and cardiac index <2.2 L/min
per m2 (Forrester class IV).
• LV outflow tract obstruction (aortic stenosis or hypertrophic
obstructive cardiomyopathy)
• Ruptured chordae
• LV inflow tract obstruction (mitral stenosis or myxoma)
• Septic shock with myocardial depression
• Pulmonary embolus
• Aortic dissection with acute aortic regurgitation or tamponade
Hemodynamic Monitoring in Acute MI
Generally accepted indications for invasive pulmonary artery
pressure monitoring in acute MI are as follows:
1. Cardiogenic shock
2. RV infarction
3. Hypotension unrelated to bradycardia and unresponsive to fluids
4. Combined hypotension and heart failure
5. Diagnostic assessment of suspected mechanical complications of
acute MI
6. To optimize fluid management and the use of inotropic drugs in
patients with unstable hemodynamics
Indications for IABP Counterpulsation in MI
1. Cardiogenic shock, low cardiac output, or unresponsive hypotension
2. Intractable angina, myocardial ischemia, ventricular arrhythmia, or
heart failure not responsive to conventional therapy
3. In conjunction with high-risk PCI or cardiac surgery
More powerful alternative hemodynamic support devices,
including left ventricular assist devices and cardiopulmonary
assist devices, may be more appropriate than IABP for selected
patients with cardiogenic shock.
An IABP is a short-term hemodynamic support device and
should only be used as part of a more definitive strategy to reverse
the cause of cardiac injury.

Causes of Noninfarct-Related Hemodynamic Effects of IABP

Cardiogenic Shock Hemodynamic effects are increased diastolic arterial blood pres-
• Myocarditis
sure, with augmented coronary diastolic blood flow and cardiac
output, increased or decreased systolic arterial blood pressure
• End-stage cardiomyopathy
and reduction in LV afterload with lower impedance to LV
• Myocardial contusion ejection, and both of the aforementioned results in a reduction
 70 Mechanical Complications of Acute Myocardial Infarction
in myocardial oxygen consumption, diminished heart rate, and
increased urinary output.
Complications of IABP
1. Vascular complications (related to insertion site, aortic wall, or dam-
age from multiple balloon inflations)
2. Hematologic problems (hemolysis or systemic emboli)
3. Balloon dependence (unable to wean from support)
4. Vascular complications are increased in elderly female patients of
small stature (with smaller-caliber vessels) and in patients with dia-
betes mellitus or peripheral vascular disease
Contraindications of IABP
1. Patient not a candidate for aggressive revascularization
2. Aortic incompetence
3. Severe peripheral vascular disease
4. Aortic aneurysm (thoracic or descending aorta)
5. Aortic dissection
RV Infarction
RV infarction should be suspected in any patient with an infe-
rior MI complicated by hypotension. Other hemodynamic
features of RV infarction include an increased RA pressure
(>12 mm Hg) and normal or decreased RV systolic and PA sys-
tolic pressures. In general, central venous pressure, RA pres-
sure, and RV diastolic pressure are all increased, whereas RV
systolic pressure, PA systolic pressure, and cardiac output are
decreased in RV infarction. A hemodynamic pattern that sug-
gests constrictive pericardial physiology (steep RA y descent,
square root sign, increased jugular venous pulse, and, rarely, a
positive Kussmaul sign) may occur in RV infarction because
of acute RV dilation within a fixed pericardial volume. A clear
lung field on chest radiography in a hypotensive patient is a
hallmark of RV infarction. RV infarction is more common in
patients with existing RV hypertrophy, a finding usually associ-
ated with chronic lung disease or congenital heart disease. In
these patients, RV infarction can also infrequently occur with-
out flow-limiting epicardial coronary artery disease. RA infarc-
tion may accompany RV infarction and is usually clinically
manifest by atrial arrhythmias. The diagnosis and management
of RV infarction is discussed in detail in another chapter in this
text (Figure 70.1).
Figure 70.1. Old right ventricular infarct with thinning of the infe-
rior and right ventricular wall.
659
• Significant RV infarction is associated with hypotension, an
increased jugular venous pressure, and clear lung fields.
• Significant RV infarction rarely occurs in the absence of evidence
of an inferior wall infarction.
• Always consider either a pulmonary embolus or a new right-to-left
shunt across a patent foramen ovale in a patient with marked arte-
rial desaturation complicating an inferior wall infarction.
• The hemodynamic findings associated with RV infarction are low
cardiac output, low pulmonary wedge pressure, and increased RA
pressure.
• RV infarction may be complicated by tricuspid regurgitation due to
tricuspid annular dilatation.
• The differential diagnosis of RV infarction is pulmonary embo-
lism, constrictive pericarditis, pericardial tamponade, and cardio-
genic shock due to other causes.
• True posterior MI often complicated by RV infarction is the only
non–ST-segment elevation MI for which thrombolytics should be
administered.
• Patients with hypotension or decreased urinary output due to RV
infarction should have moderate volume loading with pharmaco-
logic inotropic support (dopamine to augment septal contraction)
to achieve a pulmonary wedge pressure of 15–18 mm Hg.
• Avoid “pushing” fluids beyond above parameters. RV overdisten-
tion can increase RV myocardial oxygen consumption and thereby
decrease cardiac output by increasing intrapericardial pressure and
limiting LV filling.
• Maintenance of AV synchrony is important for maintenance of RV
filling. Temporary pacing should be used in high-grade AV block.
Atrial fibrillation or flutter should be promptly cardioverted.
• RV infarction complicating inferior infarction increases in-hospital
mortality but not long-term mortality among patients discharged
alive from the hospital.
Failed Reperfusion
Failed thrombolysis is characterized by persistent or worsening
chest pain, persistent or worsening ST-segment elevation, or
hemodynamic instability. Absence of these clinical indicators
of ongoing myocardial ischemia is not a completely reliable
predictor of successful reperfusion for all patients. The success
of thrombolytic therapy in patients with an ST-segment eleva-
tion MI is dependent on complete restoration of normal infarct-
related artery blood flow (grade 3 TIMI flow) (Table 70.4).
Restoration of grade 3 TIMI flow improves LV function and
survival among MI patients, but restoration of lesser grades of
blood flow—namely, TIMI flow of grade 2 or less—does not
reduce mortality.
Angiographic “no-reflow” or “slow flow” is a special case of
nonreperfusion at the time of primary PCI for acute MI, in which
Table 70.4. Coronary Artery Angiographic Classification
of TIMI Flow
TIMI flow grade Feature
0 Absence of any antegrade flow beyond a coronary
occlusion
1 Faint antegrade coronary flow beyond the occlusion,
although filling of the distal coronary bed is
incomplete
2 Delayed or sluggish antegrade flow with complete
filling of the distal territory
3 Normal flow that fills the distal coronary bed
completely
660 VII Myocardial Infarction

there is a failure to re-establish normal myocardial perfusion in

the absence of a significant residual epicardial coronary stenosis.
It is characterized microscopically by swollen endothelial cells
that result in capillary plugging by red cells, neutrophils, platelets,
and fibrin thrombi, which, in turn, obstruct capillary blood flow.
This phenomenon is a predictor of future cardiac events, includ-
ing mechanical complications of MI, and a poorer prognosis.

Infarct Extension and Expansion

Infarct extension or recurrent infarction is an increase in myo-
cardial necrosis remote from the original infarct site and usually
occurs between day 2 and day 10 after the initial MI. It is asso-
ciated with cardiogenic shock, subendocardial infarct, female
sex, and previous infarctions. A secondary elevation in CK-MB
beyond the first 24 hours and a new Q wave on the ECG are sug-
gestive of the diagnosis: serum troponins are less useful diag-
nostically in this situation since they remain elevated longer after
infarction than CK-MB enzyme.
Infarct expansion or ventricular remodeling refers to thinning
and dilatation of the infarcted myocardium due to the pulsatile
hemodynamic pressure without an increase in myocardial necro-
sis. Infarct expansion stretches the myocyte bundles and reduces
the density of cardiac myocytes in the infarcted wall (Figure 70.2).
Infarct expansion typically occurs with large anterior wall infarcts
and may result in ventricular aneurysm formation, heart failure,
and refractory ventricular arrhythmias (Figure 70.3).
True and False Ventricular Aneurysms
A true LV aneurysm is a discrete, thinned segment of the left
ventricle that protrudes during both systole and diastole and has a
broad neck. In contrast, a false LV aneurysm has a narrow neck,
a result of a prior LV free wall rupture, and is contained solely
by the adherent pericardium. The presence of hypertension and
the use of corticosteroids and NSAIDs may promote aneurysm
formation (Figure 70.4–70.6).
• True LV aneurysms rarely rupture, whereas rupture is not uncom-
mon with false LV aneurysms.
Dynamic LV Outflow Tract Obstruction
Dynamic LV outflow tract obstruction results from hyperdy-
namic contraction of the basal portions of the left ventricle in
response to catecholamine stimulation, usually after an ante-
rior wall infarction. The cross-sectional area of the outflow
Figure 70.3. Thinning and scarring of the inferior wall after a large
inferior wall myocardial infarction.
tract is reduced, resulting in a decreased cardiac output, often
associated with arterial desaturation and a new systolic ejec-
tion murmur. In addition, there may be systolic anterior motion
of the mitral valve leaflets toward the septum because of a
Venturi-like effect leading to further outflow tract obstruction
and mitral regurgitation. The diagnosis is made with echocar-
diography, and treatment is with intravenous fluids, cessation or
reduction in administered inotropic drugs, and cautious use of
β-blockers.
Myocardial (Cardiac) Rupture
Myocardial rupture after MI encompasses rupture of the inter-
ventricular septum and the LV free wall, both of which share
a similar cellular pathophysiology. Rupture typically occurs in
an area of infarction without myocardial reperfusion with a per-
sistently occluded infarct-related coronary artery. Early throm-
bolysis or primary PCI decreases the risk of cardiac rupture,
provided successful reperfusion is achieved. Unsuccessful or late
thrombolysis probably does not increase the overall risk of car-
diac rupture but may accelerate the process of cell necrosis and
lead to earlier ventricular wall rupture, usually within 24 hours
of administration. Late PCI (>12 hours after presentation) prob-
ably reduces the risk of ventricular rupture. Rupture typically
occurs at the junction of normal and infarcted myocardium
(Figures 70.7 and 70.8).

Figure 70.2. “Wavy myocytes” after myocardial infarction. Figure 70.4. Very large apical aneurysm with thrombus.
 70 Mechanical Complications of Acute Myocardial Infarction
Figure 70.5. False aneurysm (contained left ventricular rupture)
after myocardial infarction.
Rupture of the Ventricular Free Wall
Patients with rupture of the ventricular free wall usually present
catastrophically with either sudden death, usually due to electro-
mechanical dissociation, or tamponade with cardiogenic shock.
Rarely, they present with subacute ventricular rupture manifested
by pericardial pain, ECG evidence of pericarditis, and a peri-
cardial rub. Rupture typically occurs within 4 days after acute
infarction. Significant predisposing factors for early ventricular
rupture in the TIMI-9b trial were old age (older than 70 years;
odds ratio, 5.0) and female sex (odds ratio, 3.6). Other commonly
identified risk factors for cardiac rupture include hypertension,
absence of ventricular hypertrophy, previous infarction, poor
collateral flow, and lateral wall MI. Possible additional risk fac-
tors include the use of corticosteroids and anticoagulation.
There are three types of myocardial ruptures:
1. A slit-like tear that occurs early after infarction and is associated
with single-vessel coronary disease without any thinning of the LV
wall and with good preservation of LV function (most common type
of rupture)
2. A rupture that results from a subacute process with localized necro-
sis of myocardium
3. Rupture that is preceded by the development of myocardial thin-
ning, with rupture in the center of the thinned area (Figures 70.9
and 70.10)
Late ruptures are associated with multivessel disease and
occur days to weeks after infarction. Rupture usually occurs in
Figure 70.6. Apical aneurysm after myocardial infarction.
661
Figure 70.7. Myocardial rupture after hemorrhagic infarction and
thrombolysis.
the left ventricle (8 times more often than in the right ventricle),
in the terminal distribution of the LAD (anterior wall rupture),
or in diagonal branches (lateral wall rupture), usually at the junc-
tion of normal and infarcted myocardium.
The treatment of rupture of the ventricular free wall is
emergency cardiac surgery. Rarely, the rupture may be walled
off to produce an LV false aneurysm or pseudoaneurysm.
Echocardiography is the diagnostic imaging method of choice,
and cardiac surgery is almost always required for pseudoaneu-
rysm because they rupture without warning. Pseudoaneurysms
are also associated with heart failure caused by loss of myo-
cardial power and systemic thromboembolism (Figures 70.11
and 70.12).
• Predisposing factors for ventricular free wall rupture are old age
(>70 years) and female sex.
• Elderly women are also at higher risk of rupture of the interven-
tricular septum.
• A common clinical challenge is the diagnosis and management
of an MI complicated by a new systolic murmur. The differential
diagnosis includes papillary muscle dysfunction or rupture, inter-
ventricular septal rupture, dynamic LV outflow tract obstruction,
and new tricuspid regurgitation due to RV infarction or massive
pulmonary embolus.
• Rarely, aortic dissection may manifest as an inferior wall
MI because of extension of the dissection into the ostium of
the RCA.
Figure 70.8. Hemorrhagic myocardial infarct after thrombolysis.
662 VII Myocardial Infarction
Figure 70.9. Junction of infarcted (left) and normal (right) myocardium.
Rupture of the Ventricular Septum
Rupture of the ventricular septum is similar pathologically
to rupture of the ventricular free wall in that it almost always
occurs within a few days of acute infarction, is frequently a
serpiginous tract rather than a discrete septal defect, and is usu-
ally associated with transmural infarction. Patients with ventric-
ular septal rupture usually present abruptly with hypotension,
acute RV failure, and a new pansystolic murmur frequently
associated with a systolic precordial thrill. Echocardiography
is the diagnostic imaging method of choice. Inferior infarctions
cause septal rupture in the basal inferior septum, whereas ante-
rior infarctions cause rupture in the apical septum. Treatment
of rupture of the ventricular septum is emergency cardiac
surgery or percutaneous defect closure in selected patients
(Figure 70.13).
• Patients with ventricular septal rupture present with hypotension,
acute RV failure, and a new pansystolic murmur frequently associ-
ated with a systolic thrill.
• Patients with acute ventricular septal rupture usually lie flat
because of systemic hypotension; patients with acute mitral valve
rupture typically cannot lie flat because pulmonary edema devel-
ops very rapidly.
• Septal rupture from inferior wall MI has a worse prognosis than
that associated with anterior wall infarction.
Figure 70.10. Inflammatory cell response 72 hours after myocardial
infarction.
Figure 70.11. Cardiac free wall rupture after myocardial infarction.
Acute Mitral Regurgitation
About 3% of patients with an acute MI have severe mitral regur-
gitation with a 30-day mortality of 25% and a one-year mortality
of 50%. A further 15% of patients have mild to moderate mitral
regurgitation, which is often transient and usually has a good
prognosis. In patients with cardiogenic shock, about 40% will
have at least moderate and often severe mitral regurgitation; this
doubles the mortality at one year compared with patients without
mitral regurgitation.
Acute mitral regurgitation after MI can be due to papillary
muscle dysfunction (also called papillary muscle displacement)
caused by fibrosis or ischemia of the papillary muscle, papillary
muscle rupture (either partial or complete), or mitral annular
dilatation associated with LV failure. The blood supply to the
posteromedial papillary muscle (derived only from the poste-
rior descending artery) is more tenuous than that to the anter-
olateral papillary muscle (derived from both the LAD and the
LCX). Consequently, 90% of papillary muscle ruptures involve
the posteromedial papillary muscle. This is fortuitous because
the posteromedial papillary muscle usually has multiple heads
(in contrast to the single head of the anterolateral papillary mus-
cle), and rupture of an individual head is frequently survivable,
at least in the short term. Complete transection of a LV papillary
muscle is usually fatal because of massive sudden mitral regur-
gitation. Infarctions associated with papillary muscle rupture are
Figure 70.12. Pericardial tamponade from left ventricular free wall
rupture and hemopericardium.
 70 Mechanical Complications of Acute Myocardial Infarction
Figure 70.13. Septal rupture after myocardial infarction.
usually small, and frequently there is only single-vessel coronary
disease. Patients with papillary muscle rupture usually present
up to 1 week after MI with acute pulmonary edema. The loud-
ness of the mitral regurgitation murmur is variable and may not
correlate with the degree of mitral regurgitation. The murmur
may be completely absent in some patients with severe mitral
regurgitation. A thrill is rarely present in acute mitral regurgita-
tion. Large V waves are present in the pulmonary wedge tracing,
and, more rarely, the regurgitant jet may be transmitted through
the pulmonary vasculature to lead to increased oxygen saturation
in the PA; this may suggest an erroneous diagnosis of ventricu-
lar septal rupture. Echocardiography is the diagnostic imaging
method of choice (Figures 70.14 and 70.15). Afterload reduction
with sodium nitroprusside and placement of an intra-aortic bal-
loon pump will temporarily improve patient hemodynamics by
increasing forward cardiac output and decreasing the regurgi-
tant fraction. Emergency cardiac surgery is generally indicated
with an expected mortality of about 25%: ideally the mitral valve
Figure 70.14. Papillary muscle rupture complicating acute inferior
myocardial infarction; magnified transverse four-chamber view. The
ruptured head of the posteromedial papillary muscle (arrow) prolapses
freely into the left atrium (LA); the posterior mitral valve leaflet (arrow-
head) is flail. LV, left ventricle; RA, right atrium. (Previously published.
See “Credit Line” section.)
663
should be repaired if there is no papillary muscle necrosis, but
mitral valve replacement is often needed. CABG at the time of
mitral valve surgery improves outcome and in patient who sur-
vives surgery, long-term prognosis is often good.
• In 90% of papillary muscle ruptures, the posteromedial papillary
muscle ruptures.
• A precordial thrill is rarely present in acute mitral regurgitation.
• The loudness of the mitral regurgitation murmur is a poor guide to
the severity of acute mitral reguritation
• The classic systolic murmur of mitral regurgitation may be absent
in acute severe mitral regurgitation. This occurs because turbulent
blood flow is often minimal in the setting where mitral valve func-
tion is severely compromised when the left atrium and left ventri-
cle are functionally almost one chamber
Acute mitral regurgitation may be differentiated clinically
from ventricular septal rupture by the following features:
1. An increase in oxygen saturation between the RA and PA is present
in ventricular septal rupture but not in acute mitral regurgitation
2. Large V waves in the pulmonary capillary wedge and PA tracings
are present in acute mitral regurgitation but not in ventricular septal
rupture
3. A precordial thrill is common in about 50% of patients with ventric-
ular septal rupture but is uncommon in acute mitral regurgitation
4. The systolic murmur of ventricular septal rupture is loud and best
heard in the lower left sternal area and posteriorly
5. The murmur of acute mitral regurgitation is frequently soft and usu-
ally heard best at the apex with radiation to the axilla, or it may be
absent entirely
Pericardial Effusion and Pericarditis
Early pericarditis and pericardial effusion occur in about 20%
of patients that do not receive reperfusion therapy for acute MI.
In contrast only 5% of patients who receive reperfusion therapy
will develop this complication. Postinfarct pericarditis results
from a localized area of pericardial inflammation, usually over
the site of a transmural infarct, and thus the development of an
early pericardial rub with the first few days after MI is associated
with a larger infarct, a lower ejection fraction, and a higher early
and late mortality. Pericarditis may be associated with nonis-
chemic chest pain and a pericardial rub and is more common
in patients with anterior infarcts, transmural infarcts, and heart
failure. Treatment is with aspirin, 650 mg every 4 to 6 hours.
Corticosteroids are generally contraindicated because of their
effect on fluid retention, the risk of precipitation of heart failure,
and because of their adverse effect on myocardial wound healing
and association with myocardial aneurysm formation and myo-
cardial rupture. Antiplatelet agents should not be discontinued in
the setting of postinfarct pericarditis.
Postinfarction Syndrome (Dressler Syndrome)
Dressler syndrome (which occurs in about 3% of MI patients) is
one of the post–cardiac injury syndromes that is thought to result
from injury to mesothelial pericardial cells that release cardiac
antigens and subsequently generate an inflammatory response in
the pericardium, pleura, and lungs. The clinical scenario consists
of pleuropericardial chest pain often associated in the early stages
with an auscultatory friction rub, fever, arthralgia, leukocytosis,
and pulmonary infiltrates. It typically occurs several weeks after
infarction and may be recurrent. Histologically it is character-
ized by a localized fibrinous pericarditis and neutrophil infiltra-
tion and has been associated with the presence of antibodies to
664 VII Myocardial Infarction

Figure 70.15. Ruptured posteromedial mitral papillary muscle in acute myocardial infarction.

cardiac tissue. Treatment is with aspirin, 650 mg every 4 to 6 LV Mural Thrombus

hours, or NSAIDs, provided that ventricular and renal function
Mural thrombus formation on the endocardial surface of the left
are well preserved. Colchicine may also be useful in treatment
ventricle is an important complication of large MIs, because of
or in prophylaxis when a post–cardiac injury syndrome may be
its association with peripheral thromboembolism in the peri-
anticipated, such as after elective cardiac surgery. Corticosteroid
infarct period, particularly non-hemorrhagic stroke. Patients at
treatment should be avoided.
risk include those with large anterior wall infarcts, low ejection
fractions, multiple regional wall abnormalities by echocardiogra-
phy and well-developed Q waves on the ECG. Peripheral embo-
lization may occur in about 10% of patients with mural thrombi;
embolism risk decreases with time since infarction, as the throm-
bus undergoes resolution and fibrosis and warfarin anticoagula-
tion reduces both the incidence of LV thrombi and subsequent
embolization (Figure 70.16).

Abbreviations
CABG coronary artery bypass grafting
CK-MB creatine kinase-MB
ECG electrocardiogram, electrocardiographic
LAD left anterior descending coronary artery
LCX left circumflex artery
LV left ventricular
LVEDP left ventricular end-diastolic pressure
MI myocardial infarction
NSAID nonsteroidal anti-inflammatory drug
PA pulmonary artery
PCI percutaneous coronary intervention
PCWP pulmonary capillary wedge pressure
RA right atrial
RCA right coronary artery
RV right ventricular

Names of Clinical Trials

TIMI Thrombolysis in Myocardial Infarction
Figure 70.16. Left ventricular mural thrombus (arrows) after TIMI 9B Thrombolysis and Thrombin Inhibition in Myocardial
myocardial infarction. Infarction 9B
 71
Cardiogenic Shock
JOSEPH G. MURPHY, MD
Shock is defined as a physiological state of inadequate blood flow
to vital organs that leads to a global reduction in tissue perfusion
with subsequent widespread cellular dysfunction. Three types of
shock are recognized:
1. Hypovolemic shock, usually due to dehydration, bleeding, or fluid
“third-spacing,” is associated with a reduced cardiac preload, low
cardiac output as a consequence of inadequate ventricular filling,
decreased mixed venous oxygen saturation, and increased systemic
vascular resistance as the body tries to compensate for hypotension.
2. Distributive shock, usually due to sepsis, is associated with mark-
edly reduced systemic vascular resistance and normal or high car-
diac output.
3. Cardiogenic shock is end-organ hypoperfusion due to cardiac dys-
function in which the pump function of the left ventricle, the right
ventricle, or both, is inadequate to maintain perfusion of vital
organs. It is typically associated with a systolic blood pressure less
than 90 mm Hg, a cardiac index less than 2.0 L/min/m 2, peripheral
vasoconstriction, cerebral ischemia, and renal oliguria.
• Shock is reversible in the early stages, but overall carries a high
mortality rates estimated at 40% to 80% for cardiogenic shock
and 30% to 50% for septic shock.
Causes of Cardiogenic Shock
The most common causes of cardiogenic shock are acute MI
(including extensive left ventricular injury), right ventricular
infarction, ruptured papillary muscle, postinfarct septal rupture,
ventricular free wall rupture with tamponade, and, uncommonly,
severe ventricular stunning in association with the apical balloon
(takotsubo) syndrome. Other causes of cardiogenic shock are
end-stage cardiomyopathy, advanced myocarditis from any
cause, critical valvular heart disease (typically acute mitral or
aortic regurgitation and critical aortic stenosis or mitral stenosis),
Abbreviations and acronyms are expanded at the end of this chapter.
severe hypertrophic cardiomyopathy, and severe ventricular dys-
function as a complication of cardiopulmonary bypass surgery.
This chapter addresses primarily cardiogenic shock as a compli-
cation of acute MI.
Cardiogenic shock is the leading cause of death among hos-
pitalized patients following acute MI. It is important to recog-
nize that the majority of patients who develop cardiogenic shock
do so in the hospital and that it is uncommon for patients to be
in frank cardiogenic shock at the time of their initial healthcare
presentation.
Incidence and Mortality of Cardiogenic Shock
Cardiogenic shock occurs in about 45,000 patients in the United
States annually, which equates to about 6% of patients hospi-
talized with STEMI and about 3 % of patients with NSTEMI.
Prior to the current strategy of PCI reperfusion for STEMI,
cardiogenic shock carried a mortality of about 90%. Current
in-hospital mortality from cardiogenic shock is much better,
but remains high (about 40%–50%) overall for all age groups
combined), and is significantly higher among patients older than
75 years (about 60%–70%) than among younger patients (about
30%–40%). Interestingly, the high mortality seen in cardiogenic
shock is almost all within the index hospitalization (during the
first week), and long-term survival data from the SHOCK trial
reported 3- and 6-year survival rates of about 40% and 30%,
respectively, in patients who received early revascularization.
Predictors of Incidence and Mortality
in Cardiogenic Shock
Risk factors for the development of cardiogenic shock in MI
patients include larger infarcts (STEMI, anterior MI, left
bundle-branch block), prior ischemic heart disease (MI, angina,
heart failure), older age, and diabetes mellitus. Successful
665
666 VII Myocardial Infarction
reperfusion within 3 hours of symptom onset by either primary
PCI or thrombolysis is largely protective against the later devel-
opment of cardiogenic shock.
Patient age has a strong effect on mortality in cardio-
genic shock and 30-day mortality risk was shown to increase
about 1.5-fold for every 10-year increment in patient age in
the GUSTO-1 thrombolysis database. In a German registry of
patients undergoing primary PCI in the setting of cardiogenic
shock, mortality was 30% among patients younger than 55 years
and 63% among patients older than 75 years, with a gradation of
mortality between these two age extremes
The other factor that appears to be strongly associated with
higher in-hospital mortality is delay between symptom onset and
hospital admission of the patient. It is uncommon for patients
to have established cardiogenic shock at presentation; with
increasing delay between symptom onset and admission, the
risk of development of cardiogenic shock increases, presumably
because of progressive and ultimately unsalvageable, left ven-
tricle myocardial injury. Even when patients undergo myocardial
revascularization, those who present later than 6 to 12 hours after
symptom onset have a 25% higher in-hospital mortality than
patients who present within 6 hours of symptom onset.
• Strong echocardiographic predictors of poorer survival in cardio-
genic shock are moderate to severe mitral regurgitation and poor
ventricular function (ejection fraction <30%).
• Hemodynamic predictors of poorer survival in cardiogenic shock
are a low mean arterial blood pressure (<60 mm Hg) or low cardiac
index (<2.0 L/min/m2) at presentation.
• Delay in initiation of reperfusion in STEMI or failure of reperfu-
sion are predictors of an increased incidence of cardiogenic shock
and poorer prognosis when cardiogenic shock does develops.
• Cardiogenic shock occurs twice as often in ST elevation MI com-
pared with NSTEMI, but has similar mortality when it occurs in
both types of infarction.
• Cardiogenic shock associated with left main coronary occlusion
has the worst prognosis; right coronary artery occlusion has the
best prognosis while left anterior descending and circumflex artery
occlusion have an intermediate prognosis.
• Oliguria (<25 mL of urine per hour) or an elevated serum creati-
nine (>2 mg/dL) predict poorer survival in cardiogenic shock.
Diagnosis of Cardiogenic Shock
It is helpful to define cardiogenic shock both with clinical cri-
teria and with invasive hemodynamic criteria. Clinically, car-
diogenic shock is defined by systolic blood pressure of less than
80 to 90 mm Hg or systolic blood pressure of 90 mm Hg or
more maintained only by pharmacologic or mechanical support
in association with a severely reduced cardiac output (cardiac
index ≤2.0 L/min/m2 in the setting of adequate cardiac filling
pressures). Some studies use a definition of a cardiac index
< 1.8 L/min/m2 without support and cardiac index < 2.2 L/min/m2
with inotropic or mechanical support.
For a diagnosis of cardiogenic shock, hypotension needs to
persist for at least 30 minutes and cannot be reversed with flu-
ids. Hemodynamic parameters of cardiogenic shock are gen-
erally accompanied by evidence of end-organ hypoperfusion,
specifically cool clammy extremities and decreased urine output
(<25 mL per hour). Significant bradyarrhythmia or tachyarrhyth-
mias may also result in adverse hemodynamic changes and need
to be excluded before a diagnosis of primary cardiogenic shock
is made. A heart rate of 60 beats per minute or more should be
present for a diagnosis of cardiogenic shock.
It is important to recognize, from a practical standpoint, that
clinical criteria are sufficient to make an initial diagnosis of car-
diogenic shock and to initiate clinical resuscitative measures.
Hypovolemic shock due to hemorrhage or dehydration may simu-
late cardiogenic shock but is associated with a low (<15 mm Hg)
pulmonary capillary wedge pressure. Septic shock is associated
with a low peripheral vascular resistance and a normal or high
cardiac index.
• The invasive hemodynamic criteria for cardiogenic shock include
a cardiac index of less than 2.0 L/min/m2 of body surface area and
adequate left ventricular filling with a pulmonary capillary wedge
pressure of >15 mm Hg.
• This definition of cardiogenic shock excludes hypovolemic shock,
which is characterized by low filling pressures and distributive
shock characterized by a normal or high cardiac output.
Cardiogenic shock usually occurs in the setting of extensive
left ventricular damage (Figure 71.1) and ventricular dysfunc-
tion. In cardiogenic shock patients, it is important to urgently
exclude a surgically treatable cause of shock—in particular,
mechanical rupture, such as ventricular free wall rupture, ven-
tricular septal rupture (Figure 71.2), and papillary muscle rup-
ture (Figure 71.3). Aortic dissection can occasionally present
as an acute MI—typically a right coronary artery occlusion;
the takeoff of the right coronary artery from the outer curva-
ture of the aorta has a higher wall tension (law of Laplace) and
results in a higher dissection rate compared with the left coro-
nary artery. Aortic dissection may also lead to cardiogenic shock
due to aortic rupture and hemorrhage with or without pericardial
tamponade. Finally, hemorrhagic shock due to bleeding needs to
be excluded, particularly since many cardiogenic shock patients
will have received fibrinolytic therapy or dual platelet inhibitors
following primary PCI. The differential diagnosis of cardiogenic
shock is summarized in Table 71.1.
Right Ventricular Infarction and
Cardiogenic Shock
Cardiogenic shock can occur in the setting of inferior MI com-
plicated by right ventricular infarction (Figure 71.4). Right ven-
tricular infarction causes about 5% of cases of cardiogenic shock
and requires a different treatment approach than isolated left
ventricular infarction complicated by shock. Right ventricular
infarction rarely occurs in isolation, almost always in association
Figure 71.1. Cardiogenic shock with large acute myocardial infarction.
 71 Cardiogenic Shock
Figure 71.2. Ventricular septal rupture after myocardial infarction.
with a left ventricular inferior wall MI. Cardiogenic shock due to
right ventricular infarction also carries high mortality and is rec-
ognized clinically by the classical triad of hypotension, elevated
jugular venous pressure, and clear lung fields. A combination of
right-sided electrocardiographic leads, echocardiography, and
invasive hemodynamic monitoring may be used to establish the
diagnosis of RVMI.
The pathophysiology of cardiogenic shock due to right ven-
tricular infarction is different to left ventricular shock and typi-
cally is associated with a high right ventricular end-diastolic
pressure that may result in a bulging of the ventricular septum
toward the left ventricular cavity, with consequent mechanical
impairment of left ventricle filling, that leads to left ventricular
systolic dysfunction and low cardiac output.
Practical Approach to Management
of Cardiogenic Shock
Faced with a patient with acute infarction complicated by signs
of shock and/or congestive heart failure, urgent diagnosis and
Figure 71.3. Partial mitral valve papillary muscle rupture.
667
Table 71.1. Differential Diagnosis of Cardiogenic Shock
Secondary to Myocardial Infarction Involving Predominantly
the Left Ventricle
Shock secondary to right ventricular infarction (usually with inferior
infarction)
Hypovolemia—dehydration, hemorrhagic shock
Myocardial rupture—free wall rupture, ventricular septal rupture,
papillary muscle rupture
Acute aortic dissection (DeBakey type A)
Pulmonary embolism
treatment are required. Allowing that significant bradycardia or
tachycardia has been excluded, the physician is left with three
important differential diagnoses:
1. Acute pulmonary edema
2. Hypovolemia
3. Cardiogenic shock (low cardiac output)
Acute Pulmonary Edema
Acute pulmonary edema can be easily recognized with the
clinical findings of rales on chest examination, elevated jugu-
lar venous pressure, and radiologic evidence of pulmonary con-
gestion. Immediate treatment includes intravenous furosemide,
morphine, and oxygen therapy. Decompensated respiratory fail-
ure requires endotracheal intubation and mechanical ventilation.
If the systolic blood pressure is adequate (>100 mm Hg), nitro-
glycerin, initially sublingual and then by intravenous infusion,
should be administered. However, if the systolic blood pressure
is less than 100 mm Hg, intravenous dopamine should be admin-
istered (see below).
Hypovolemia
In a hypotensive patient who does not have clinical evidence of
pulmonary edema, intravenous fluid can be administered care-
fully while causes of hypovolemic shock (including bleeding,
dehydration or “third spacing” of interstitial fluid) are evaluated.
Cardiogenic Shock
In shock patients, blood pressure of less than 90 mm Hg needs
to be confirmed with repeated measurements. Regardless of
Figure 71.4. Acute right ventricular myocardial infarction.
668 VII Myocardial Infarction
whether there are early signs of hypoperfusion, dopamine
(5–15 mcg/kg per minute) should be given by intravenous infu-
sion. If systolic blood pressure is critically low (<70 mm Hg)
with evidence of hypoperfusion, norepinephrine should be
given immediately as an intravenous infusion at a dose of 8 to
12 mcg/min and then titrated to 2 to 4 mcg/min.
Once the diagnosis of a shock state has been established, the
first consideration is immediate bedside resuscitation, includ-
ing fluid administration to rectify hypervolemia, correction of
arrhythmias, and initiation of vasopressor support. These mea-
sures are temporary, and it is important to expeditiously deter-
mine the exact pathophysiology of the patient’s shock using
appropriate diagnostic modalities (echocardiography, cardiac
catheterization, etc), in order that definitive treatment can begin.
Clinical examination is always important in cardiogenic shock
but is insufficient to exclude a mechanical complication of MI as
a cause of cardiogenic shock.
• Emergency echocardiography in cardiogenic shock may show car-
diac free wall, papillary muscle, or septal rupture.
• Emergency coronary angiography is almost always warranted in
cardiogenic shock unless the patient’s general medical condition
(eg, terminal malignancy or severe dementia) would make inter-
vention futile or it is against the wishes of the patient.
• Medical management with inotropic or vasopressor medication
alone is inadequate treatment in cardiogenic shock and is unlikely
to improve the survival of the patient.
• Clinical examination and transthoracic echocardiography may
occasionally miss acute severe mitral regurgitation, and further
studies with pulmonary capillary wedge pressure measurement,
transesophageal echocardiogram, or left ventriculography are
sometimes required.
Role of Invasive Monitoring, Intra-Aortic Balloon
Pumps, and Angiography
The most successful treatment strategy for cardiogenic shock
associated with MI is myocardial reperfusion and early surgical
correction of mechanical complications of MI where relevant.
For cardiogenic shock patients who have not received reperfusion
therapy, immediate arrangements should be made for transfer to
a cardiac catheterization laboratory for coronary angiography
and possible PCI. This applies to both to patients with STEMI
and NSTEMI. In hospitals without PCI capability, emergency
patient transfer to a PCI-capable facility is urgently required. If
shock develops in a patient who has recently received a fibrin-
olytic agent, emergency cardiac catheterization is indicated to
assess the patency of the culprit vessel, and rescue PCI can be
performed if there has been a failure of reperfusion.
• Fibrinolysis is generally ineffective when cardiogenic shock has
developed in STEMI and primary PCI is the preferred reperfusion
strategy.
An emergency echocardiogram is important in all cardio-
genic shock patients with a suspected infarct-related mechanical
complication. Swan-Ganz catheter should generally be placed in
all patients who have incipient or established cardiogenic shock,
particularly in hypotensive patients that are unresponsive to a
fluid bolus or in whom fluid administration is contraindicated (ie,
in pulmonary edema).
Intra-arterial pressure monitoring (eg, with a radial artery
line) should be placed in patients with cardiogenic shock, par-
ticularly in those with persistent hypotension or in patients who
require inotropic support to maintain their blood pressure within
the normal range (class I indication per ACCF/AHA guidelines).
In patients who are about to undergo coronary angiography or
who have recently returned from the cardiac catheterization lab-
oratory, arterial pressure monitoring can be performed through
the arterial sheath.
An intra-aortic balloon should be placed as soon as pos-
sible in cardiogenic shock patients in conjunction with coronary
diagnostic and resuscitative procedures. Intra-aortic balloon
counterpulsation decreases afterload, reduces cardiac work, and
increases coronary perfusion. It is important to recognize that
systolic pressure measured during normal intra-aortic balloon
counterpulsation is often lower than when counterpulsation is
not used.
• The benefit of intra-aortic balloon counterpulsation is largely from
its augmentation of diastolic blood pressure with a resultant overall
increase in mean arterial pressure and coronary artery blood flow.
Placement of an intra-aortic balloon pump is also a class I recom-
mendation in the ACCF/AHA guidelines for cardiogenic shock
patients.
An intra-aortic balloon pump is often particularly helpful in
patients with mechanical complication of MI in whom cardiac
surgery is planned. It is generally advisable to place the intra-
aortic balloon pump immediately before coronary angiography
in the cardiac catheterization laboratory. If coronary angiogra-
phy or PCI is performed before placing an intra-aortic balloon
pump in patients with severe hypotension, their clinical condi-
tion can deteriorate rapidly such that resuscitation is not possible.
In our practice, an intra-aortic balloon pump is placed in the
femoral artery contralateral to that used for coronary angiog-
raphy, although placement of the intra-aortic balloon pump in
either femoral artery followed by angiography and PCI through a
brachial or radial approach is also acceptable.
A summary of the ACCF/AHA recommendations for inva-
sive interventions in patients with cardiogenic shock is shown in
Table 71.2.
Coronary Angiography and Revascularization
All patients in cardiogenic shock should be considered for emer-
gency coronary angiography in anticipation of primary PCI or
for confirmation of patency of the culprit artery after fibrinolytic
therapy. Patients who have already undergone successful pri-
mary PCI but in whom shock subsequently develops should be
returned to the cardiac catheterization laboratory without delay
to reconfirm the patency status of the culprit vessel. The estab-
lishment of a patent culprit coronary artery is the number one
priority in all cardiogenic shock patients after initial cardiopul-
monary resuscitation and this is true for patients with either left
ventricular or right ventricular infarction.
Data from various PCI registries and randomized trials,
including the SHOCK trial, have shown that immediate or early
Table 71.2. Cardiogenic Shock Class I Indications
PCI is recommended for patients with acute myocardial infarction
who develop cardiogenic shock and are suitable candidates (level of
evidence: B)
A hemodynamic support device is recommended for patients with
cardiogenic shock after STEMI who do not quickly stabilize with
pharmacological therapy (level of evidence: B)
Abbreviations: PCI, percutaneous coronary intervention; STEMI, ST-elevation
myocardial infarction.
Previously published. See “Credit Lines” section.
 71 Cardiogenic Shock
revascularization is superior to medical treatment alone in car-
diogenic shock patients.
The importance of restoring normal coronary artery blood
flow has been demonstrated in a registry of the nonrandomized
SHOCK population. Among patients in whom TIMI coronary
artery blood flow was 0 or 1 at the conclusion of PCI, mortality
was in excess of 80%. Among patients with residual TIMI grade
2 blood flow, mortality was approximately 50% to 60%, and
among those who had normal (TIMI grade 3) coronary artery
blood flow, mortality was lowest (approximately 35%).
• Restoration of normal coronary artery blood flow is a strong deter-
minant of patient survival in cardiogenic shock
Randomized clinical trials, particularly the SHOCK trial,
have established that early revascularization reduces mortality in
patients with cardiogenic shock compared with intensive medi-
cal therapy (Figure 71.5). The benefits of revascularization com-
pared with intensive medical therapy in the SHOCK trial were
more evident in the younger patient cohort (<75 years old) than in
the older patient cohort ≥75 years old. The lack of a demonstrable
survival benefit for the elderly in the SHOCK trial was likely due
to poorer ventricular function in the baseline ejection fraction
between the treatment groups. Many elderly patients with cardio-
genic shock will benefit from emergency PCI, as demonstrated
in the SHOCK registry; but clinicians need to individualize care
to those older patients for whom revascularization is medically
appropriate (Table 71.2).
In the majority of shock patients, PCI (including stenting) of
the culprit vessel should be performed. Interventional treatment
is controversial for lesions in nonculprit, but critical stenosed
coronary vessels in patients with multivessel disease and cardio-
genic shock. In the absence of cardiogenic shock, PCI of noncul-
prit stenosis in MI patients is contraindicated. The AHA/ACCF
guidelines do not state whether CABG surgery is preferable to
PCI in cases of cardiogenic shock.
SHOCK Trial
The SHOCK Trial was a landmark randomized trial of 302
patients in cardiogenic shock complicating acute STEMI
(also included infarction with new left bundle branch block)
that tested the hypothesis that emergency revascularization
(CABG or PCI) was better than initial medical stabiliza-
tion including thrombolysis and placement of an intra-aortic
balloon pump.
100
P=0.11 P=0.04 P<0.03
80
64% 66%
Mortality, %
60 56% 53%
47% 52%
40
20
0
30 days 6 months 1 year
Figure 71.5. Mortality at 30 days, 6 months, and 1 year in the
SHOCK (Should We Emergently Revascularize Occluded Coronaries
for Cardiogenic Shock?) trial, which compared early revascularization
(blue bars) and intensive medical therapy (green bars).
669
The main conclusions of the SHOCK Trial were
• Conformation of the strong correlation between patient age and
mortality in cardiogenic shock.
• The benefit of early revascularization was only evident in the
patients younger than 75 years in the randomized SHOCK trial.
• The SHOCK registry suggested benefit from emergency PCI in
selected elderly cardiogenic shock patients chosen on the basis of
expert clinician judgment.
Inotropic and Vasopressor Agents
in Cardiogenic Shock
Inotropic and vasopressor agents are frequently important in
the management of a patient with newly diagnosed cardiogenic
shock and should be used for initial resuscitation and to maintain
adequate blood pressure during the early hours or days of car-
diogenic shock.
Inotropic agents are “double-edged swords” in the manage-
ment of cardiogenic shock in that while they improve hemody-
namics in the short-term, they also increase myocardial oxygen
consumption. In this context, pharmacological agents are an
adjunctive, temporary measure and not a substitute for aggressive
myocardial reperfusion and mechanical circulatory support.
• Inotropic and vasopressor agents should be used at the lowest doses
for the shortest time period, and patients should be weaned rapidly
when hemodynamically feasible.
Inotropic agents increase myocardial contractility and heart
rate and result in varying degrees of systemic vasodilation or
vasoconstriction. Dopamine is generally considered to be the
first-line agent for patients with hypotension, particularly if there
is evidence of shock. Dobutamine should not be used when there
are clinical signs of hypoperfusion and shock. Dobutamine is
reserved for patients with systolic blood pressure above 90 mm
Hg and is used to increase cardiac output; its additional vasodi-
lator properties also provide afterload reduction. In patients in
whom there is severe reduction in systolic blood pressure (gener-
ally ≤70 mm Hg), norepinephrine should be used.
In general, inotropic agents should be used for as short a time
as possible; their use is often limited by their toxicity, particu-
larly arrhythmias. Inotropic agents should never be considered
as definitive therapeutic for cardiogenic shock.
• Newer techniques of mechanical support of the circulation such
as left ventricular assist device should be considered in all cardio-
genic shock patients who require continued inotropic support after
reperfusion to maintain an acceptable blood pressure.
Circulatory Assist Devices in Cardiogenic Shock
A number of circulatory assist devices are available for poten-
tial use in patients with cardiogenic shock. The theory behind
the use of temporary mechanical circulatory support in car-
diogenic shock is to interrupt the pathophysiological vicious
cycle of hypotension, myocardial ischemia, and left ventricu-
lar dysfunction that characterizes cardiogenic shock and buy
time for recovery of noninfarcted but stunned and hibernating
myocardium.
1. The intra-aortic balloon pump is the most commonly used circu-
latory assist device and works by balloon counterpulsation within
the descending aorta, inflating during diastole and deflating dur-
ing systole. It is indicated in persistent hypotension or cardiogenic
shock, particularly in association with mechanical complications of
MI such as acute mitral regurgitation or ventricular septal rupture
670 VII Myocardial Infarction
pending emergency cardiac surgery. It is also indicated for recurrent
myocardial ischemia or hemodynamic instability due to poor left
ventricular function. Contraindications to intra-aortic balloon pump
use include aortic regurgitation, abdominal aortic aneurysm or aor-
tic dissection, severe peripheral artery disease, sepsis, or bleeding
disorders. Risk factors for complications with an intra-aortic bal-
loon pump include older age, female sex, diabetes, small stature, and
peripheral vascular disease.
• An intra-aortic balloon pump improves most hemodynamic
parameters by about 20% to 25% in cardiogenic shock patients,
including cardiac output, left ventricular end-diastolic pres-
sure, and diastolic blood pressure but typically decreases sys-
tolic blood pressure.
2. Extracorporeal membrane oxygenation is indicated primarily for
patients with refractory respiratory failure when used in venovenous
bypass mode (blood is extracted from the right atrium or vena cava
and returned to the right atrium) but can be used in veno-arterial
bypass mode (blood is extracted from the right atrium and returned
on the arterial system), completely bypassing the heart and lungs, to
support patients in cardiogenic shock.
3. Tandem Heart is a percutaneous left ventricular assist device which
extracts blood from the left atrium through a transseptal puncture
and returns blood to the iliac artery and has been used for short-term
stabilization of cardiogenic shock patients pending cardiac surgery.
4. Impella is an axial flow pump that works on the principle of an
Archimedes screw, in which the blood is extracted from the left ven-
tricle through a catheter placed retrograde across the aortic valve
and is pumped into the ascending aorta. Blood flow with axial flow
pumps is nonpulsatile which, if prolonged, can lead to end-organ
dysfunction, including renal failure. (This phenomenon is poorly
understood.) Other complications include traumatic hemolysis and
a systemic inflammatory response syndrome (SIRS).
General Management Issues
in Cardiogenic Shock
In addition to a strategy of reperfusion with mechanical and
pharmacologic support of the patient circulation, general medi-
cal issues are frequently important in the management of patients
with cardiogenic shock.
The workload and oxygen requirement of the left ventricle
should be reduced as much as possible; patients should receive
adequate analgesia and sedation to treat pain and any associated
anxiety. Judicious use of morphine is generally excellent in this
regard. Severe anemia should be corrected cautiously with red
blood cell transfusion.
Oxygen status should be monitored continuously and supple-
mental oxygen provided; patients who have respiratory failure
(either manifest or incipient) should be promptly intubated and
ventilated. Early and aggressive treatment of arrhythmias is
important in patients who have cardiogenic shock, particularly
treatment of ventricular tachycardia, atrial fibrillation, severe
bradycardia, or heart block and maintenance of atrioventricular
synchrony.
Failed Treatment in Cardiogenic Shock
There is a high mortality with cardiogenic shock, even with
invasive treatment, and it is not uncommon for patients to be
hemodynamically unresponsive to aggressive intervention. The
overall mortality rate is approximately 50%—somewhat lower in
younger patients and higher in older patients. In selected patients,
consideration of mechanical support of the circulation with a left
ventricular assist device or orthotopic cardiac transplantation
may be indicated. Important considerations for this decision are
the wishes of the patient and family, and the presence or absence
of significant end-organ damage, such as renal failure, liver fail-
ure, significant lung injury, or anoxic brain injury.
Inflammatory State Accompanying
Cardiogenic Shock Physiology
The classic understanding of the pathophysiology of cardio-
genic shock due to MI starts with an initial major insult to the
left ventricular myocardium that decreases myocardial con-
tractility leading to a reduced cardiac output and hypotension,
which in turn results in decreased coronary blood flow, which
further exacerbates myocardial ischemia. In this model of car-
diogenic shock, the vicious cycle of decreased coronary blood
flow, myocardial ischemia, left ventricular “pump failure,”
and hypotension is compensated, to some extent, by systemic
vasoconstriction observed hemodynamically by measurement
of high systemic vascular resistance. In patients who succumb
to cardiogenic shock after acute MI, autopsy studies show that
generally at least 40% of the left ventricular myocardium was
infarcted, although this may represent a combination of new and
old infarcts in some patients.
Recent observations have challenged this classic under-
standing of the pathophysiology of cardiogenic shock. The
observation that systemic vascular resistance is not consis-
tently elevated supports an alternative hypothesis of cardio-
genic shock; instead, a wide range of resistance measurements
has been found. The left ventricular ejection fraction is typi-
cally about 30%, a value often seen in many well compensated
patients with dilated cardiomyopathy. Evidence of a systemic
inflammatory response syndrome (SIRS) analogous to that seen
in other forms of shock has been observed in many patients
with cardiogenic shock. This is most commonly associated with
sepsis, trauma, burns, and pancreatitis, and it is characterized
by abnormal temperature (>38°C or <36°C), heart rate (>90
beats per minute), respiratory rate (>20 breaths per minute),
and an abnormal white blood cell count (>12,000 × 106 /L or
<4,000 × 106 /L). Hemodynamically, these patients, as in some
patients with cardiogenic shock, have no objective evidence of
systemic vasoconstriction—indeed, they have low systemic
vascular resistance.
The underlying mechanism is uncertain, but an overproduc-
tion of nitric oxide, specifically the iNOS isoform, has been
postulated. Excessive levels of iNOS have been shown to inhibit
myocardial contractility, to decrease mitochondrial respiration,
to be associated with reperfusion injury, to have adverse effects
on glucose metabolism, and to be proinflammatory. High levels
of iNOS may also lead to a decreased responsiveness to cate-
cholamines and induce systemic vasodilation.
Markers of inflammation including elevated levels of cytokine
IL-6 are generally slightly elevated in patients who have large
acute MIs; however, if cardiogenic shock develops, IL-6 levels
increase even higher. In cardiogenic shock patients in whom
multiorgan failure occurs, extremely high IL-6 levels are often
present similar in magnitude to levels associated with septic
shock, which is typically associated with profound vasodilatory
and decreased systemic vascular resistance.
The TRIUMPH trial that evaluated the nitric oxide inhibi-
tor tilarginine in cardiogenic shock was terminated because of
a lack of efficacy; this agent had little selectivity for iNOS, and
its failure may have been due, in part, to inhibition of the other
NOS isoforms.
 71 Cardiogenic Shock 671

Abbreviations Names of Clinical Trials

ACCF American College of Cardiology Foundation GUSTO-1 Global Utilization of Streptokinase an tPA for Occluded
AHA American Heart Association Arteries
CABG coronary artery bypass graft SHOCK Should We Emergently Revascularize Occluded Coro-
IL-6 interleukin-6 naries for Cardiogenic Shock
MI myocardial infarction TIMI Thrombolysis in Myocardial Infarction
NSTEMI non–ST-elevation myocardial infarction TRIUMPH Tilarginine Acetate Injection in a Randomized
PCI percutaneous coronary intervention International Study in Unstable MI Patients With
STEMI ST-elevation myocardial infarction Cardiogenic Shock
 72
Reperfusion Strategies for ST-Segment Elevation
Myocardial Infarction
JOSEPH G. MURPHY, MD
The cause of STEMI is rupture or erosion of an atheromatous
plaque that leads to formation of an occlusive intracoronary
thrombus. In animal models of acute myocardial infarction,
ischemic myocardial necrosis proceeds in a “wave front” man-
ner, spreading from the subendocardium to the epicardium. This
process begins 20 minutes after acute coronary occlusion and
involves most of the myocardial wall within 6 hours. In humans,
the time to complete myocardial infarction is variable and can
be much longer than the 6 hours seen in the animal model due
to the presence of collateral vessels to the ischemic territory,
ischemic preconditioning of the myocardium, and the occurrence
of intervening periods of spontaneous reperfusion secondary to
a dynamic pathologic process that may cycle periods of plaque
thrombosis followed by spontaneous clot lysis.
• In-hospital mortality for STEMI has declined for about 25% in the
1960s to the current 5% mortality.
PCI in STEMI
Primary PCI is the treatment of choice for all consenting patients
with EKG evidence of new STEMI or its equivalent of new left
bundle branch block in association with clinical symptoms of
myocardial infarction, provided primary PCI can be performed
in a timely manner by an experienced team, generally considered
to be within 90 minutes from medical presentation. New isolated
posterior wall myocardial infarction or suspected circumflex cor-
onary artery occlusion, both of which may not display classical
ST-elevation EKG findings also warrant emergency coronary
angiography and PCI when occlusion is confirmed.
Abbreviations and acronyms are expanded at the end of this chapter.
672
STEMI is a medical emergency, comparable to major trauma
in which a rapid, abbreviated clinical assessment trumps a com-
prehensive, but slower, clinical evaluation. Delays (measured in
minutes) in the initiation of appropriate emergency treatment
significantly increase mortality, and all STEMI patients should
have a fast-track diagnostic assessment with prompt transfer to a
cardiac catheterization laboratory for primary PCI or, if PCI is
not available in a timely manner, consideration of fibrinolysis.
Primary PCI is superior to fibrinolysis as a reperfusion strat-
egy because it opens more occluded arteries at a much lower risk
of serious complications, particularly intracranial hemorrhage.
Primary PCI, precisely because it restores normal coronary
blood flow (TIMI 3 flow) more often than fibrinolysis, is associ-
ated with an improved left ventricular function, smaller infarct
size, and better long-term prognosis compared with fibrinolysis.
In situations where primary PCI is not available within a
90-minute therapeutic window, fibrinolysis is the treatment of
choice for most patients with STEMI, provided strong contra-
indications to its use are not present (Table 72.1).
The Evidence for Primary PCI
The benefits from primary PCI can be attributed to improved
myocardial salvage because of higher initial rates of TIMI grade
3 coronary blood flow and, secondarily, less culprit vessel resid-
ual stenosis, better plaque healing, and less post-PCI coronary
blood flow turbulence that leads to a lower risk of reinfarction
and reocclusion. Primary PCI is also safer than thrombolysis,
with a lower risk of bleeding complications, particularly hemor-
rhagic stroke and left ventricular rupture, that is typically associ-
ated with intramyocardial hemorrhage. At hospitals with 24-hour
PCI availability or non-PCI hospitals with rapid transfer capabil-
ity, primary PCI has become the preferred reperfusion strategy
 72 Reperfusion Strategies for ST-Elevation Myocardial Infarction
Table 72.1. Absolute and Relative Contraindications for
Fibrinolysis
Absolute contraindications
Any prior intracranial hemorrhage
Known structural cerebrovascular lesion
Known malignant intracranial neoplasm
Ischemic stroke within the past 3 months (except for acute stroke within
3 hours)
Suspected aortic dissection
Active bleeding or bleeding diathesis (excluding menses)
Significant closed-head or facial trauma within 3 months
Relative contraindications
History of chronic, severe, poorly controlled hypertension
Severe uncontrolled hypertension on presentation (systolic pressure
>180 mm Hg or diastolic pressure >110 mm Hg)
History of prior ischemic stroke >3 months previously, dementia,
or known intracranial pathology not covered in absolute
contraindications
Traumatic or prolonged CPR (>10 minutes) or major surgery (within past
3 weeks)
Recent internal bleeding (within past 2–4 weeks)
Noncompressible vascular punctures
Pregnancy
Active peptic ulcer
Current use of anticoagulants: the higher the INR, the higher the risk of
bleeding
For streptokinase/anistreplase: prior exposure (>5 days previously) or
prior allergic reaction to these agents
CPR, cardiopulmonary resuscitation; INR, international normalized ratio.
for all STEMI patients, except for the rare exception in which
vascular access is not possible or the patient or family requests
conservative management only.
An invasive cardiac catheterization approach to STEMI, even
when PCI is not performed, has several advantages in the follow-
ing situations:
1. Identification of coronary artery anatomy suitable for surgical revas-
cularization (critical left main or three-vessel disease)
2. Identification of mechanical complications of myocardial infarction
requiring emergency cardiac surgery, namely acute mitral regurgita-
tion, ventricular septal defect or free wall rupture, or (rarely) aortic
dissection associated with an inferior wall STEMI
3. Shock requiring hemodynamic support with intra-aortic balloon
pump or left ventricular assist devices
4. Exclusion of coronary artery occlusion as the cause of the patient’s
symptoms and identification of mimics of myocardial infarction that
cause ST-segment elevation (namely pericarditis, myocarditis, ven-
tricular aneurysm, takotsubo or apical balloon syndrome, coronary
bridging or spasm, and incidental left bundle branch block)
Early trials that first demonstrated the benefits of primary
angioplasty in STEMI achieved rapid reperfusion with DTB time
of 90 minutes or less.
The GUSTO IIb substudy demonstrated a direct relationship
between time from study randomization to angioplasty and sub-
sequent 30-day mortality as follows: 60 minutes or less, 1.0%
mortality; 61 to 75 minutes, 3.7% mortality; 76 to 90 minutes,
4.0% mortality; 91 minutes or more, 6.4% mortality (P = .001).
Furthermore, the highest 30-day mortality (14.1%) was
observed in the patient subset that did not undergo angioplasty.
Analysis of a prospective observational registry of 27,080 patients
in NRMI 2 showed that the multivariate-adjusted odds of mor-
tality were 40% to 60% higher if the DTB time was longer than
2 hours.
673
• Time-to-reperfusion treatment is an important predictor of both
early and late mortality for all reperfusion strategies, be it primary
PCI or fibrinolysis.
Primary PCI vs Fibrinolysis in STEMI
Multiple randomized trials have compared the safety and effi-
cacy of primary PCI with intravenous thrombolytic therapy in
acute STEMI, and pooled analysis has established that PCI is
the superior strategy in terms of reduced mortality (7% vs 9%),
nonfatal reinfarctions (3% vs 7%), and strokes (1% vs 2%) com-
pared with thrombolysis, and that this early benefit also persists
in long-term follow-up studies.
The relative benefits of primary PCI can be attributed to
the more frequent achievement of normal coronary blood flow
(TIMI grade 3), which translates into better myocardial sal-
vage, ventricular function, and long-term survival. Moreover,
the significantly lower risk of intracranial hemorrhage and
other bleeding complications makes PCI the safer means of
reperfusion.
PCI-Related Delay and STEMI
A caveat in the management of STEMI by PCI is that PCI must
be performed by an experienced operator in a timely fashion,
traditionally interpreted as within 90 to 120 minutes of health-
care presentation or an additional 60 to 90 minutes compared
with the time of alternate fibrinolysis administration. However,
since many hospitals worldwide do not have 24-hour primary
PCI capability, fibrinolysis will remain an important reperfu-
sion strategy for many patients with STEMI. The time from first
medical contact to the option for administration of fibrinolytic
therapy is known as the DTN. The “PCI-related delay time”
is the difference between these times (DTB–DTN) and is an
important decision point for non-PCI hospitals when deciding
whether to transport an STEMI patient to a PCI-capable hospi-
tal or administer fibrinolysis locally, known as the “drip versus
ship” decision. PCI-related time delay diminishes the advan-
tages of PCI over fibrinolysis and a long PCI delay is associ-
ated with a worse clinical outcome than local fibrinolysis. The
PCI-fibrinolysis equipoise point is the PCI-related time delay
that exactly offsets the advantage of PCI over fibrinolysis and is
generally considered to be about 60 to 90 minutes for patients
who present within a three hour time period after chest pain
onset (Tables 72.2 and 72.3).
• Current guidelines suggest that 60–90 minutes of additional
transport time (PCI-related delay) is acceptable in order to per-
form primary PCI, instead of fibrinolysis when a patient pres-
ents within the fi rst three-hour period after chest pain onset in
STEMI
• Acceptable PCI-related treatment delay is shorter for anterior
infarcts than for nonanterior infarcts and increases with patient
age (>65 years) and later presentation (>2 hours after symptom
onset). The rationale behind this is because patients with anterior
wall infarcts have a higher mortality and thus more to gain by ear-
lier reperfusion that non-anterior wall infarcts; older patients are
at higher risk of bleeding complications after fibrinolysis and fibri-
nolysis is less effective on more older stabilized occlusive thrombi
that have been present for two hours or more, while PCI is less
affected by time related thrombus stabilization.
• Anticipated PCI-related delay should take into account real-world
conditions such as weather, traffic, and PCI personnel availability
674 VII Myocardial Infarction
Table 72.2. Guidelines for Selecting Fibrinolysis or Pri-
mary PCI
If onset of symptoms is <3 hours and there is no delay for an invasive
strategy, neither fibrinolysis nor primary PCI is preferred
Fibrinolysis is generally preferred in the following situations:
Early presentation (onset of symptoms <3 hours) and anticipated delay
for an invasive strategy
Catheterization laboratory is occupied or not available
Vascular access difficulties
Lack of access to a skilled, high-volume PCI facility
Prolonged transport delay for primary PCI
Door-to-balloon time minus door-to-needle time >60 minutes
Total door-to-balloon time >90 minutes
Primary PCI is generally preferred in the following situations:
Skilled PCI facility with on-site surgical backup door-to-balloon time
minus door-to-needle time <60 minutes
Total door-to-balloon time <90 minutes
High clinical risk, including cardiogenic shock (age <75) or Killip
class ≥3
Contraindication to fibrinolysis or high bleeding risk
Late presentation with onset of symptoms >3 hours
Diagnosis is in doubt; differential diagnosis includes pericarditis,
myocarditis, aneurysm, and tako-tsubo apical ballooning
PCI, percutaneous coronary intervention.
• PCI should always be performed as soon as possible, and the con-
cept of an acceptable PCI delay is only useful as a decision point
in the evaluation of immediate fibrinolysis versus a delayed PCI
scenario. It is not a threshold time for myocardial damage.
• Fibrinolysis prevents about 3 early deaths per 100 patients with
STEMI treated within 6 hours of symptom onset and higher risk-
patients will derive an even greater absolute benefit.
• Thrombolysis is not an effective treatment in patients with an
NSTEMI and is contraindicated in this setting.
• Patients with occlusion of a large circumflex coronary artery may
show no EKG changes or only subtle EKG changes in the limb
leads I and aVL; if circumflex occlusion is suspected on the basis
of EKG or echo evidence, emergency coronary angiography is
indicated.
• All non steroidal anti-inflammatory drugs, including selective
COX-2 inhibitors but excluding aspirin, increase the risk of death,
reinfarction, and cardiac rupture and should be stopped in all
STEMI patients.
Primary PCI is generally superior to thrombolysis for the fol-
lowing reasons:
1. About 25% of patients with STEMI, particularly the elderly are not
candidates for thrombolysis because of strong contraindications.
Primary PCI is rarely contraindicated in a consenting patient.
Table 72.3. Framework for Selecting
Reperfusion Strategy
Duration From Onset of
Symptoms, h
Transport time, min <3 >3
<30 Primary PCI Primary PCI
30–60 Fibrinolysis Primary PCI
>60 Fibrinolysis Primary PCI
or
Fibrinolysis
PCI, percutaneous coronary intervention.
2. Primary PCI achieves complete normalization of infarct artery
blood flow (TIMI 3 flow) in over 95% of patients compared with
about 60% with best case thrombolysis.
3. Intracranial hemorrhage, a devastating complication with a mortal-
ity rate in excess of 50% occurs at least 10 times more often after
thrombolysis than after primary PCI.
Intracranial Hemorrhage With Reperfusion
The overall risk of intracranial hemorrhage after thromboly-
sis is about 1% but rises to over a 4% risk in high-risk patients
including those with multiple risk factors (elderly ≥75 years,
female sex, black race, prior stroke or TIA, systolic hyperten-
sion ≥160 mm Hg, relatively low weight [≤ 65 kg for women or
≤ 70 kg for men], and patients taking warfarin with a high INR
[>4]) compared with about 0.1% of patients overall after primary
PCI, a rate that is also less sensitive to the above risk factors.
• Primary PCI is the treatment of choice for STEMI complicated
by cardiogenic shock as fibrinolysis is much less successful in this
situation, probably because of lack of adequate perfusion pres-
sure for the fibrinolytic agent to penetrate the occlusive throm-
bus and the high occurrence of infarcted artery reocclusion, also
likely due to low coronary blood flow consequent on systemic
hypotension.
• Primary PCI achieves a higher infarct-related artery patency
(>95%) than thrombolysis (about 60%), at a much lower risk of
intracranial hemorrhage.
• There are relatively high rates of recurrent ischemia (25%), reoc-
clusion (10%) and reinfarction (5%) after successful fibrinolysis.
Reinfarction significantly increases short-term and long-term
mortality.
Late Reperfusion (Symptom Onset
From 12 to 48 Hours)
The BRAVE-2 trial randomly assigned 365 STEMI patients
who presented 12 to 48 hours after symptom onset to an inva-
sive strategy or a conservative noninvasive strategy. The inva-
sive approach of coronary stenting with adjunctive abciximab
decreased infarct size more than a conservative strategy (8% vs
13%; P<.001).
Late presenters account for 10% to 30% of all STEMI
patients, and the BRAVE-2 trial suggested that myocardium
may remain viable in some individuals for extended periods
of time (even after more than 12 hours of ischemia) and may
thus be successfully salvaged with late PCI. This finding could
be explained by ischemic preconditioning, presence of collat-
eral circulation, and stunned or hibernating myocardium in the
infarct zone.
• Primary PCI, should also be considered in patients who present
later than 12 hours of symptom onset, if there is ongoing chest pain
and/or EKG evidence of ongoing ischemia.
• Multiple randomized trials in patients presenting later than
12 hours after onset of symptoms in STEMI demonstrated that
PCI revascularization resulted in better left ventricular function
and long-term clinical outcome than a conservative noninvasive
strategy, especially if there was evidence of ongoing ischemia.
• Administration of the platelet glycoprotein IIb/IIIa inhibitor abcix-
imab upstream of a planned primary PCI procedure for STEMI did
not increase the patency of the infarct-related artery and had no
advantage compared to administration in the catheterization labo-
ratory at the time of primary PCI.
 72 Reperfusion Strategies for ST-Elevation Myocardial Infarction
Clinical Detection of Reperfusion
and Rescue PCI
Clinical assessment of fibrinolytic reperfusion is not an exact
science but is an important clinical decision point because of
the failure of fibrinolytic reperfusion in about 40% of patients
with STEMI and the potential for rescue PCI in these patients.
Ongoing ischemic chest pain and <50% resolution in ST-segment
elevation in the lead(s) with the highest ST-segment elevations
60 to 90 minutes after the start of fibrinolytic therapy is the gen-
erally accepted clinical standard for diagnosing failed fibrino-
lytic reperfusion.
Confounding factors in the diagnosis of failed fibrinolytic
reperfusion are the observations that ischemic chest pain may
resolve with narcotics or with myocardial denervation as a
consequence of myocardial necrosis, while the characteristic
ST-segment elevation may return to preinfarct levels as a result of
the natural EKG evolution after infarction.
Rescue PCI is emergency PCI performed within 12 hours fol-
lowing an initial failed strategy of fibrinolytic reperfusion for
STEMI, in which the culprit coronary artery remains occluded.
Rescue PCI is indicated in the setting of failed fibrinolytic
reperfusion for STEMI and has been demonstrated to reduce sub-
sequent heart failure and reinfarction with a trend towards lower
all-cause mortality compared with a conservative strategy.
• Rescue PCI is associated with a higher risk of hemorrhagic stroke
and bleeding complications compared to primary PCI
• Rescue PCI should be considered for patients with reperfusion fail-
ure after fibrinolysis with ongoing chest pain and or <50% resolu-
tion of maximal ST-segment elevation at 90 minutes.
The REACT trial evaluated 427 patients with STEMI who
had failed fibrinolytic reperfusion, defined as resolution of the
ST-segment elevation of less than 50% at 90 minutes; patients
were randomly assigned to additional fibrinolysis, conservative
therapy, or PCI. Event-free survival was highest in those who
were treated with PCI (84.6%), followed by conservative ther-
apy (70.1%), while repeated fibrinolysis was associated with the
worst prognosis (68.7%). Thus, considering the possible benefits
of rescue angioplasty, a heightened awareness of the potential
of reperfusion failure after fibrinolysis is needed to risk stratify
STEMI patients and to implement rescue PCI.
Cardiac Catheterization Assessment
of Coronary Blood Vessel Flow
and Vascular Bed Perfusion
Vessel patency and myocardial vascular bed perfusion is graded
in the cardiac catheterization laboratory using the TIMI semi-
quantitative, visual assessment of contrast flow following cor-
onary catheter injection and is used to assess both epicardial
coronary blood flow and vascular bed filling (bluish) as a mea-
sure of microvascular perfusion (Table 72.4). It is important to
note that absence of completely normal perfusion (TIMI grade 3)
Table 72.4. TIMI Classification of Coronary Blood Flow
TIMI grade 0: no antegrade flow
TIMI grade 1: partial penetration of contrast past the point of occlusion
TIMI grade 2: opacification but delayed filling of the distal vessel
TIMI grade 3: normal flow
Abbreviation: TIMI, Thrombolysis in Myocardial Infarction.
675
after primary PCI is associated with poor left ventricular recov-
ery and increased late mortality.
Multiple trials have demonstrated that the long-term clinical
benefits of primary PCI or fibrinolytic therapy (decreased mor-
tality, smaller infarct size, and improved LV function) accrue
only with the restoration of normal coronary blood flow (TIMI
grade 3) and that lower degrees of reperfusion (TIMI grade 2
flow or less) are generally inadequate to save substantial amounts
of ischemic myocardium.
• Grade 0 is the complete absence of antegrade contrast flow through
the occlusion in the infarct-related artery.
• Grade 1 is where there is some antegrade penetration of contrast
material beyond the point of obstruction in the culprit artery, but
without normal perfusion of the distal coronary bed.
• Grade 2 is intact antegrade perfusion of the entire infarct vessel
distally into its vascular bed, but with slow contrast flow.
• Grade 3 is intact antegrade contrast flow in the culprit infarct
artery and normal vascular bed perfusion.
Early evaluation of the patient with a suspected STEMI or its
equivalent should focus on three issues:
1. Establishing the correct diagnosis—usually made by clinical symp-
toms of myocardial ischemia in association with EKG finding of
ST elevation in two contiguous EKG leads or new-onset left bundle
branch block
2. Rapid assessment of comorbidities that may influence the treatment
choice, including a history of stroke, TIA, vascular disease, bleeding
disorder, diabetes, renal failure, medications, procedural history and
drug allergies
3. Expedited consent to proceed with treatment
Principles of Myocardial Reperfusion
Three important principles guide the approach to myocardial
reperfusion:
1. Time is myocardium—Early reperfusion of the infarct-related artery
within the first few hours of artery occlusion limits infarct size,
increases myocardial salvage, preserves left ventricular function,
and in turn improves patient survival. The time from coronary artery
occlusion to human cardiac myocyte loss is uncertain, and no safe
threshold time between vessel occlusion and myocyte loss has been
established. Consequently, the culprit artery should be open by PCI
as soon as possible.
2. Myocardial ischemic time is difficult to assess. The time of initial
artery occlusion can be deduced from the onset of chest pain, but
many patients have stuttering symptom onset or atypical infarct
symptoms. In many patients there is a period of unstable angina pre-
ceding complete vessel occlusion with repeated vessel opening and
closing as opposing thrombotic and intrinsic fibrinolytic cascades
duel for supremacy. Ischemic preconditioning due to prior exposure
of the myocardium to ischemia alters multiple biochemical pathways
that are currently imperfectly understood in favor of prolonged myo-
cyte protection from the cellular consequences of ischemia. Finally,
collateral vessels may supply enough blood flow retrogradely to the
occluded vascular bed to maintain myocardial viability. Cessation
of contraction saves the cardiac myocyte about 80% of its adenos-
ine triphosphate energy requirements, such that even modest collat-
eral blood flow can sustain myocyte metabolism enough to maintain
cells in a hibernating or stunned but potentially viable state pending
reperfusion. The presence of continuing ischemic chest pain and the
absence of Q waves on the EKG suggests at least partial viability of
ischemic myocardium.
3. Open artery hypothesis–Improved long-term patient prognosis
attributable to successful reperfusion and maintenance of patency of
the infarct-related artery either by primary PCI or by thrombolysis is
well supported by multiple studies. Late mechanical opening of the
676 VII Myocardial Infarction
infarct related artery at 12 to 36 hours after artery occlusion is less
certain of benefit but may reduce infarct size by salvaging stunned
and hibernating myocardium kept viable by collateral vessels. Other
possible benefits of an open artery are an improvement in ventric-
ular diastolic function, favorable left ventricular remodeling, and
improved electrophysiological stability.
Facilitated Reperfusion
On the basis of current trial data and guidelines, there are two
proven reperfusion strategies for patients with STEMI—namely,
full-dose fibrinolysis or primary PCI. A third strategy that of
facilitated PCI (Table 72.2) in which pharmacologic reperfu-
sion, either in the form of usual or low dose fibrinolysis or a
platelet glycoprotein IIb/IIIa inhibitor is administered en route
to PCI as a bridge strategy to counter the PCI-induced treatment
delay.
Facilitated PCI, while attractive in theory as a way to improve
coronary artery patency rates prior to planned PCI, has failed in
multiple clinical trials to demonstrate clinical benefit.
The ASSENT–4 PCI trial enrolled STEMI patients with a
symptom duration of less than 6 hours and an expected delay
of 1 to 3 hours for primary PCI. Transfer after the admin-
istration of full-dose tenecteplase was compared with usual
transfer for primary PCI. The tenecteplase group (compared
with the standard PCI group) had higher in-hospital mortality
(6% vs 3%; P = .0105), a higher number of strokes (1.8% vs
0%; P<.0001), and a higher rate of reinfarction (6% vs 4%;
P = .0279). The study concluded that facilitated PCI with
fibrinolytic therapy should not be performed as part of routine
practice.
In the FINESSE trial, 2,452 patients with an STEMI were
randomly assigned to either early administration of abciximab
and half-dose reteplase prior to PCI or, alternatively, abcix-
imab administered at the time of PCI. There was no significant
difference between the two groups in the composite primary
90-day end point of death from all causes, ventricular fibrilla-
tion occurring more than 48 hours after randomization, cardio-
genic shock, or heart failure. In addition, the combined therapy
was associated with a trend toward an increase rate of intra-
cranial hemorrhage and a significant increase in TIMI major
and minor bleeding. The third arm of the FINESSE trial, that
of early abciximab without reteplase given in the emergency
room, was also inferior to abciximab given at the time of PCI.
• Facilitated PCI with fibrinolytic therapy should be avoided in
STEMI patients because of significant increases in mortal-
ity, nonfatal reinfarction, urgent target lesion revascularization
and stroke, and a trend toward a higher rate of major bleeding.
Facilitated fibrinolytic therapy also increases the risk of free wall
rupture.
• Facilitated PCI with reduced dose fibrinolytic therapy and glyco-
protein IIb/IIIa inhibitor does not improve outcomes compared to
glycoprotein IIb/IIIa inhibitor at the time of PCI and is not a rec-
ommended strategy in STEMI.
• Advances in STEMI treatment over the past 2 decades have
decreased 30-day hospital mortality to 6% or less among those
treated with fibrinolysis and to 4% or less among those treated with
primary PCI.
Fibrinolytic Drugs
All clinical fibrinolytic drugs generate plasmin, a natu-
rally occurring lytic peptide enzyme formed from inactive
plasminogen that in turn promotes lysis of intravascular throm-
bus. Streptokinase is a polypeptide molecule originally isolated
from the β-hemolytic streptococcus bacteria, that was the first
fibrinolytic agent introduced into clinical practice for the treat-
ment of STEMI. It works by binding to plasminogen, forming
a new enzyme complex that in turn cleaves other plasminogen
molecules to form active plasmin. The effectiveness of streptoki-
nase was initially validated in the GISSI-1 trial published in 1986
and the ISIS-2 trial published in 1988; it remains the standard of
care for fibrinolysis in many countries.
The tPA (alteplase) is a recombinant version of naturally
occurring tPA produced by vascular endothelial cells. It is highly
thrombus-specific and binds firmly to fibrin on the clot surface
where it activates plasminogen to plasmin. The GUSTO trial val-
idated the superiority of accelerated tPA (alteplase) in combina-
tion with heparin over streptokinase.
TNK-tPA (tenecteplase) and rPA (reteplase) are genetically
mutated forms of native tPA that are highly clot specific with a
longer half-lives than tPA, and can be more easily administered
by single bolus (TNK-tPA) or double bolus (rPA) injection that
the continuous infusion required for tPA (Table 72.5).
Prehospital Fibrinolysis
The strategy of prehospital fibrinolysis is based on the obser-
vation that STEMI patients who receive fibrinolysis within the
first 60 to 90 minutes of symptom onset have the greatest mor-
tality benefit among patients receiving fibrinolytic reperfusion.
Prehospital fibrinolysis is attractive for many non-PCI capable
rural communities where hospital transfer times are long (greater
than 90 minutes), but implementation requires a significant
investment in emergency medical services training to correctly
analyze and transmit off-site EKGs and appropriately administer
fibrinolytic therapy where indicated.
In circumstances where there is a short hospital transfer time
there was no real advantage to prehospital fibrinolysis. The MITI
trials evaluated the outcomes of 360 patients randomly assigned
to prehospital or in-hospital fibrinolysis with aspirin and alteplase.
All patients received fibrinolysis early within 3 hours of symp-
tom onset, while the prehospital fibrinolysis saved on average
33 minutes of treatment delay and almost all of the prehospital
patients received fibrinolysis within 1 hour of emergency care
request. The MITI study found that there were no significant dif-
ferences in left ventricular ejection fraction, infarction size, or
all-cause mortality between the treatment groups.
The MITI trial may have been biased against real-world
conditions as pre hospital EKG interpretation was done on all
patients, such that in-hospital fibrinolysis occurred very early
after hospital arrival in the in-hospital treatment patients, thus
lessening the real-world time saving of prehospital fibrinolysis.
The GREAT was conducted in rural Britain and randomly
assigned 311 patients with a suspected STEMI to prehospital or
in-hospital fibrinolytic therapy. The GREAT reported a reduction
of more than 50% in 1-year mortality (10.5% vs 21.5%; P<.01)
and a time savings of about 2 hours when at-home fibrinolysis
was compared with in-hospital fibrinolysis.
The EMIP Trial was a larger randomized trial of nearly 5,500
patients randomized to prehospital treatment with anistreplase
compared with in-hospital fibrinolysis. Cardiovascular mortality
at 30 days decreased from 9.7% to 8.3% (P = .05) with prehos-
pital fibrinolysis, and the time savings was about 1 hour for drug
administration.
 72 Reperfusion Strategies for ST-Elevation Myocardial Infarction 677

Table 72.5. Comparison of Approved Fibrinolytic Agents

Streptokinase Alteplase Reteplase Tenecteplase-t PA
Dose 1.5 MU in Up to 100 mg in 90 min 10 U × 2 each 30–50 mg based
30–60 min (based on weight) over 2 min on weight
Bolus administration No No Yes Yes
Allergic reactions Yes No No No
(especially hypotension)
Approximate 90-min 50 75 75 75
patency rates, %
TIMI grade 3 flow, % 32 54 60 63
Abbreviations: TIMI, Thrombolysis in Myocardial Infarction; t PA, tissue plasminogen activator.

The CAPTIM Trial compared prehospital fibrinolysis with
primary PCI for patients within 2 hours of symptom onset and
showed a trend towards improved mortality (2.2% vs 5.7%;
P = .058) and a significantly decreased risk of cardiogenic shock
with prehospital thrombolysis.
• Prehospital thrombolysis is beneficial when hospital transfer times
are long (typically greater than 60–90 minutes), but in circum-
stances where there is a short hospital transfer time there was no
real advantage to prehospital fibrinolysis.
Advantages of Fibrinolysis
Fibrinolysis is indicated for STEMI within the first 12 hours after
symptom onset. The main benefits of using fibrinolysis are the
following:
1. It is widely available in rural and community hospitals.
2. It is easily administered, with typical DTN time <30 minutes.
3. It requires fewer personnel and less equipment than 24-hour access
to an onsite cardiac catheterization laboratory.
The Achilles heel of fibrinolysis is that TIMI grade 3 blood
flow is achieved in only about 60% of patients treated at 90
minutes; furthermore, up to 30% to 40% of patients have abso-
lute or relative contraindications to fibrinolytic therapy because
of bleeding risk (Table 72.1). Hemorrhagic stroke associated
with fibrinolysis occurs in about 1% of patients but remains a
catastrophic complication, with a mortality of more than 50%
in individual patients. In addition, in 10% to 20% of patients,
the infarct vessel reoccludes after successful thrombolytic
reperfusion.
door-to-balloon time of less than 90 minutes (Table 72.1). Benefit
may still occur with reperfusion therapy given within 12 hours
after the onset of symptoms. However, the greater the time dura-
tion from the onset of infarct symptoms the lower the benefit
with reperfusion therapy. The absolute benefit of fibrinolytic
therapy is less for inferior STEMI, except for patients with an
inferior STEMI associated with a right ventricular infarction
(or anterior ST-segment depression indicative of a larger terri-
tory at risk). Therefore, in low-risk patients at increased risk of
bleeding complications with fibrinolytic therapy, the risk-benefit
ratio needs to be carefully assessed. Factors associated with a
higher risk of intracranial hemorrhage are older age, low body
mass index, hypertension, and female sex. Recommendations for
treatment with fibrinolytic therapy and contraindications for use
have been determined by the American College of Cardiology
and American Heart Association. Fastidious use of these guide-
lines would result in more patients treated, since currently there
is likely significant underutilization of reperfusion therapy, espe-
cially fibrinolytic therapy, in non-PCI-capable hospitals.
Mortality Trials of Fibrinolytics in Acute
Myocardial Infarction
Several landmark studies (GISSI-1 and ISIS-2) showed con-
vincing beneficial effects of fibrinolytic therapy in patients
Fibrinolytic Control/placebo
Time to
treatment, h better better
0-1

Principles of Fibrinolysis >1-2

The single most important predictor of benefit for patients >2-3
treated with fibrinolytic therapy is the time from the vessel occlu-
sion to the restoration of normal coronary TIMI 3 blood flow >3-6
(Figure 72.1). Fibrinolytic therapy is equally effective for both
sexes and elderly individuals, resulting in a significant reduction >6-12
in the higher mortality attendant in the elderly, but at an increased
risk of intracranial hemorrhage. Patients who benefit the most >12-24
from fibrinolytic therapy have anterior wall STEMI, diabetes, or
sinus tachycardia at the time of presentation. Fibrinolytic therapy
that is administered early, especially within 1 to 3 hours of the 0.5 1.0 1.5
onset of symptoms, produces the greatest benefit.
Odds ratio
All patients with acute STEMI presenting within 12 hours
of the onset of chest pain should be considered for reperfusion Figure 72.1. Proportional effect of fibrinolytic therapy on 35-day
therapy, ideally primary PCI if cardiac catheterization facilities mortality according to treatment delay in a meta-analysis of 22 trials
and trained interventional staff are available within an expected (50,246 patients).
678 VII Myocardial Infarction
with acute myocardial infarction. These beneficial effects
were also found in multiple subpopulations at higher risk,
including the elderly, diabetic patients, and patients with a pre-
vious myocardial infarction. In spite of the increased fibrino-
lysis complication risk in these patients, overall they have an
equivalent or greater benefit with fibrinolytic therapy. Diabetic
patients, who have increased risks associated with myocardial
infarction, derive greater benefit overall than patients without
diabetes.
The Fibrinolytic Therapy Trialists’ Collaborative Group
presented the data of nine trials of fibrinolytic therapy, which
included 58,600 patients who received therapy. Overall mor-
tality was 10.5%, with a 1% risk of stroke and a 0.7% risk of
major noncerebral bleeding. Overall, there was an 18% reduction
in mortality with the use of fibrinolytic therapy and an approx-
imately 25% reduction in mortality for patients who presented
with STEMI or left bundle branch block.
Patients with STEMI or new-onset left bundle branch block
benefit the most from fibrinolysis. Patients with non-STEMI
and unstable angina do not benefit from fibrinolysis and there
is a trend toward harm. Patients with cardiogenic shock respond
poorly to fibrinolysis, probably because of poor penetration of
the fibrinolytic agent into the occlusive thrombus and the lack
of adequate coronary perfusion pressure (in the setting of severe
hypotension) that is needed to prevent recurrent thrombosis and
maintain vessel patency. Cardiogenic shock is a strong indication
for primary PCI.
A select number of the most important clinical fibrinolytic
trials that have advanced the treatment of acute myocardial
infarction are reviewed below. Fibrinolytic trials are frequently
classified by sponsoring organizations (eg, GISSI, ISIS-2, TIMI,
ECSG, TAMI, GUSTO-1, and MITI).
GISSI-1
The GISSI trial (or GISSI-1 trial) was the first study that
reported the benefit of intravenous streptokinase treatment
within 12 hours of symptom onset. It included 11,816 patients
with chest pain accompanied by ST-segment elevation or
depression of 1 mm or more in any limb lead or of 2 mm or
more in one or more of the precordial leads. There was a 17%
reduction in mortality at 21 days and at 1 year among patients
with STEMI treated with streptokinase within 6 hours of infarc-
tion. Early administration resulted in better outcomes, with a
mortality reduction at 21 days of 47% if the patient was treated
within 1 hour after the onset of chest pain, 23% if within 1 to 3
hours, and 17% if within 3 to 6 hours. At 1 year, there was still
a significant reduction in mortality with earlier treatment (64%,
15%, and 17%, respectively). The overall 21-day mortality was
10.7% in the group treated with streptokinase and 13% in the
placebo group. The 1-year mortality was 17.2% in the streptoki-
nase group and 19% in the placebo group (P<.01). A follow-up
to this study demonstrated that the initial mortality benefit was
maintained at 10 years. There was no significant mortality ben-
efit in patients with ST-segment depression or in those treated
later than 6 hours.
GISSI-2
The GISSI-2 trial was one of the important head-to-head trials
of streptokinase and tPA. It compared use of intravenous strep-
tokinase with tPA with or without heparin in 12,490 patients
admitted within 6 hours after the onset of chest pain. There
were no significant differences in the mortality rate, the rate of
reinfarction, the stroke rate, or the incidence of postinfarction
angina. Subcutaneous heparin had no added benefit on mortal-
ity, heart failure, or ventricular function, but it increased bleeding
complications.
ISIS-2
The ISIS-2 trial compared aspirin and streptokinase and the
combination in a 2 by 2 design. It tested four regimens (strep-
tokinase 1.5 million units over 1 hour, aspirin 162.5 mg daily
for 1 month, a combination of the two, or placebo) in more than
17,000 patients 24 hours or less after the onset of a myocardial
infarction. Aspirin and streptokinase independently reduced
mortality and had a synergistic benefit without increasing the
stroke risk when used together, and this benefit was still present
at 1 year. The combination of streptokinase and aspirin reduced
mortality by 53% when administered within 4 hours of symptom
onset and by 38% when administered within 12 to 24 hours of
the infarction.
TIMI
The TIMI trials were a large series of trials that initially looked
at the efficacy of tPA (TIMI 1–4) and later hirudin (TIMI 5–9B),
tenecteplase (TIMI 10A and 10B), enoxaparin (TIMI 11A, 11B,
23, 25, and 28), glycoprotein IIb/IIIa inhibitors (TIMI 12, 14–16,
18, 20, and 24), and prehospital fibrinolysis (TIMI 19). TIMI
investigators also determined the classification of coronary blood
flow status by angiography (Table 72.4).
TIMI 1
After myocardial infarction, use of rtPA resulted in more rapid
reperfusion (assessed angiographically) than use of streptoki-
nase. There was no added benefit in mortality or ejection frac-
tion with rtPA than with streptokinase. Bleeding complications
were similar in both groups. At 90 minutes after thrombolysis,
TIMI flow grade 2 or 3 coronary reperfusion rates were 60%
with rtPA and 35% with streptokinase (P<.001). At 21 days, mor-
tality was 4% with rtPA and 5% with streptokinase (not statisti-
cally significant).
TIMI 2
In patients who received rtPA, heparin, and aspirin for an acute
myocardial infarction, an invasive strategy of PTCA within 18
to 48 hours of the infarction was of no added benefit compared
with a conservative strategy of only PTCA in patients with spon-
taneous or exercise-induced myocardial ischemia. There was no
difference at 6 weeks or 1 year in mortality or reinfarction rates.
Mortality and nonfatal reinfarction rates at 1 year were 14.7%
and 15.2%, respectively, in the invasive and conservative treat-
ment groups (P not significant).
TIMI 2A
The TIMI 2A trial differed from the TIMI 2 trial by the addition
of an immediate PTCA group (<2 hours after rtPA administra-
tion). This approach was compared with the delayed invasive
strategy of PTCA within 18 to 48 hours of rtPA administration
and a conservative treatment arm of PTCA for spontaneous or
 72 Reperfusion Strategies for ST-Elevation Myocardial Infarction
exercise-induced myocardial ischemia. The conservative and
invasive strategies were equally effective when judged according
to the predismissal vessel patency, ejection fraction, and 1-year
mortality and reinfarction rates.
GUSTO-I
The GUSTO-I trial randomly assigned 41,021 patients to differ-
ent thrombolysis regimens to test the hypothesis that early and
sustained patency of the infarct-related vessel would improve
survival for patients with an evolving myocardial infarction.
This study reported that rapid restoration of coronary blood
flow with tPA was associated with an improved survival and
a 14% reduction in mortality at 30 days when compared with
survival and mortality for patients who received streptokinase
and intravenous or subcutaneous heparin. Accelerated use of
tPA, as compared with use of streptokinase, was also associated
with a significantly higher incidence of hemorrhagic stroke.
For overall benefit, as assessed from the combined end point of
total mortality and disabling stroke, tPA was significantly bet-
ter than streptokinase; higher-risk patients received the great-
est benefit. The summary of the GUSTO-I trial suggested that
accelerated use of tPA and intravenous heparin resulted in a
lower mortality than use of streptokinase combined with either
subcutaneous or intravenous heparin. A secondary fi nding was
that patients with cardiogenic shock benefited from emergency
PTCA. The angiographic substudy of GUSTO-I showed that
the 1-year mortality rates favored tPA (9.1% vs 10.1% for strep-
tokinase) combined with either intravenous or subcutaneous
heparin (P≤.01).
GUSTO-IIB Angiographic Substudy Trial
The GUSTO-IIB angiographic substudy trial was a head-to-
head comparison of primary PTCA and fibrinolysis. Primary
PTCA was found to be an excellent alternative to thrombolysis in
skilled hands and had a small short-term advantage over thromb-
olysis with tPA.
GUSTO-III
The GUSTO-III trial was a head-to-head trial of two forms of
rtPA, alteplase and reteplase (a longer-acting mutant variety of
alteplase). This study included more than 15,000 patients with
STEMI or new left bundle branch block myocardial infarction
who presented within 6 hours of symptom onset. The results
with both agents were not significantly different. Mortality at
30 days was 7.22% with alteplase and 7.43% with reteplase,
and the incidence of hemorrhagic stroke was 0.88% and 0.91%,
respectively.
Adjunctive Therapy With Fibrinolysis
Aspirin
The ISIS-2 study provided the strongest evidence that aspirin
reduces mortality in STEMI patients. This benefit of aspirin
therapy was additive to the benefit of fibrinolytic therapy and
did not significantly increase the risk of stroke or intracranial
hemorrhage. In an acute STEMI, aspirin reduces mortality by
23% and the reocclusion rate by 50%. Therefore, all patients
presenting with STEMI, not previously taking aspirin should
679
be given 162 to 325 mg of non–enteric-coated chewable aspirin.
Subsequently, they should take 75 to 162 mg of aspirin daily.
P2Y12 Receptor Blockers
All patients with STEMI should receive a platelet P2Y12 recep-
tor blocker (clopidrogel, ticagrelor, prasugrel) as soon as pos-
sible after healthcare presentation, irrespective of whether the
patient is destined for fibrinolysis, primary PCI, or conservative
management with no reperfusion.
The CLARITY-TIMI 28 trial was a large trial of nearly 3,500
patients with STEMI, randomized to clopidrogel or placebo in
addition to standard fibrinolysis. Clopidrogel significantly reduced
the primary combined short-term end point of an angiographi-
cally occluded infarct-related artery, death or reinfarction by 36%
(15.0% vs 21.7 % with placebo) with no increase in major bleed-
ing events or intracranial hemorrhage. A platelet P2Y12 receptor
blocker is standard of care for all patients receiving primary PCI
for STEMI and significantly decreases stent thrombosis.
Best Reperfusion Strategy for STEMI
in a Specific Patient Group
Cardiogenic Shock
• Primary PCI is the treatment of choice for cardiogenic shock
patients as fibrinolysis is much less effective in this setting and
PCI offers the opportunity for circulatory support.
Early (<3 hours) Presentation Following
Symptom Onset
• Patients who present less than 3 hours from symptom onset will
benefit substantially with excellent myocardial salvage and signif-
icant mortality reduction from prompt reperfusion by either pri-
mary PCI or fibrinolysis. These patients are very time sensitive and
delays in reperfusion strongly influence long-term survival and sur-
rogate markers of survival including residual ventricular function
(ejection fraction) and infarct size (Figures 72.2 and 72.3).
• Primary PCI is the reperfusion strategy of choice provided it can
be delivered within 90 minutes of healthcare presentation because
of its ability to achieve higher culprit vessel patency at lower risk.
In practice this translates into a maximum acceptable PCI related
treatment delay of 60 minutes or less. (PCI treatment delay is DTB
time minus the expected DTN time for fibrinolytic therapy; fibrin-
olysis can generally be administered within 30 minutes of health-
care presentation.)
• Fibrinolysis is strongly indicated for patients who cannot receive
primary PCI within the 90 minutes window from healthcare pre-
sentation which translates into an expected PCI treatment delay of
more than 60 minutes.
• Every 30-minute delay from symptom onset to reperfusion is asso-
ciated with an 8% increase in relative mortality at 1 year.
Presentation Between 3 and 12 Hours
Following Symptom Onset
• Patients who present between 3 and 12 hours from symptom onset
derive less benefit from reperfusion than patients who present
within 3 hours of symptom onset and are less time sensitive that
early presenters.
• Fibrinolysis is much less effective at opening occluded arteries
after 3 hours from vessel occlusion due to clot stabilization, which
makes it more resistant to fibrinolysis.
680 VII Myocardial Infarction

100

Mortality reduction, %
80
Modifying factors
60
• Collaterals
• Ischemic preconditioning
40 • MVO2

20

Extent of salvage (% of area at risk)

0
0 3 6 12
Hours
Time to treatment is critical Opening the artery
is primary goal
Figure 72.2. Relationship Between Mortality Reduction and Myocardial Salvage. MVO2, myocardial oxygen consumption.

• Primary PCI is the treatment of choice in this setting which should
be performed as soon as possible, but longer anticipated PCI-
related transport delays are acceptable (up to 90 to 120 minutes)
before fibrinolysis should be considered.
Presentation Between 12 and 24 Hours
Following Symptom Onset
• Fibrinolysis is not indicated in late presenters because the bleeding
risks outweigh the likely benefits.
• Primary PCI is the treatment of choice for later presenters if there is
ongoing ischemic chest pain or hemodynamic or electrical instabil-
ity, particularly if new Q waves have not developed on the EKG.
Failed Thrombolysis

• Patients who present between 12 and 24 hours following symptom • Rescue PCI is indicated for patients with clinically suspected failed
onset derive much less benefit, in general, from reperfusion than fibrinolysis (ongoing ischemic chest pain or failure of resolution of
patients who present earlier; however, extended myocardial viabil- maximal ST elevation by 50% or more).
ity can occur up to 24 hours after symptom onset in the presence of
collateral circulation to the ischemic myocardium. Abbreviations
DTB door-to-balloon
DTN door-to-needle
EKG electrocardioraphic, electrocardiogram
STEMI INR international normalized ratio
NSTEMI non–ST-segment elevation myocardial infarction
PCI percutaneous coronary intervention
PTCA percutaneous transluminal coronary angioplasty
STEMI ST-segment elevation myocardial infarction
TIA transient ischemic attack
Yes PCI capable No tPA tissue plasminogen activator
and available? rtPA recombinant tissue plasminogen activator

Names of Clinical Trials

ASSENT-4 Assessment of the Safety and Efficacy of a New
Treatment Strategy With Percutaneous Coronary
Intervention
No Eligible for Yes BRAVE-2 Beyond 12 Hours Reperfusion Alternative Evaluation
thrombolysis? Trial
CAPTIM Comparison of Angioplasty and Prehospital Throm-
bolysis in Acute Myocardial Infarction
CLARITY Clopidogrel as Adjunctive Reperfusion Therapy
DANAMI[-2] Danish Multicenter Randomized Study on Fibrinolytic
Therapy Versus Acute Coronary Angioplasty in Acute
Primary PCI See Table 3 Myocardial Infarction[-2]
EMIP European Myocardial Infarction Project
Figure 72.3. Selection of Reperfusion Strategy. PCI, percutane- ECSG European Cooperative Study Group
ous coronary intervention; STEMI, ST-segment elevation myocardial FINESSE Facilitated Intervention With Enhanced Reperfusion
infarction. Speed to Stop Events
 72 Reperfusion Strategies for ST-Elevation Myocardial Infarction 681

FTT Fibrinolytic Therapy Trialists’ Collaborative Group ISIS International Study of Infarct Survival
GISSI Gruppo Italiano per lo Studio della Streptochinasi LATE Late Assessment of Thrombolytic Efficacy
nell’Infarto Miocardico MITI Myocardial Infarction Triage and Intervention
GREAT Grampian Region Early Anstreplase Trial NRMI 2 Second National Registry of Myocardial Infarction
GUSTO Global Utilization of Streptokinase and Tissue Plas- REACT Rescue Angioplasty Versus Conservative Treatment
minogen Activator for Occluded Coronary Arteries or Repeat Thrombolysis
GUSTO IIb Global Use of Strategies to Open Occluded Arteries in TAMI Thrombolysis and Angioplasty in Myocardial Infarction
Acute Coronary Syndromes TIMI Thrombolysis in Myocardial Infarction
 73

Cardiac Emergencies
JOSEPH G. MURPHY, MD, BARRY A. BOILSON, MD,
and MARGARET A. LLOYD, MD

This chapter outlines the general principles of managing cardiac 2010 Guidelines for Adult Advanced
emergencies in adults. It is deliberately not an exhaustive review. Cardiovascular Life Support
Sudden cardiac arrest is a leading cause of death in developed
American Heart Association guidelines for CPR and emer-
nations and is frequently associated with VF. Resuscitation for
gency cardiovascular care algorithms for response to and man-
cardiac arrest is most successful if defibrillation is performed
agement of cardiac arrest (Figures 73.1 and 73.2), bradycardia
within the first 5 minutes after collapse.
(Figure 73.3), and tachycardia (Figure 73.4) are reviewed below.

Figure 73.1. American Heart Association Algorithm for Response to Adult Cardiac Arrest. CPR indicates cardiopulmonary resuscitation; ET,
endotracheal; IO, intraoral; IV, intravenous; PEA, pulseless electrical activity; PETCO2, partial pressure of end-tidal carbon dioxide; VF, ventricular
fibrillation; VT, ventricular tachycardia. (Previously published. See “Credit Lines” section.)

Abbreviations are expanded at the end of this chapter.

682
 683

Figure 73.2. American Heart Association Algorithm for Post–Cardiac Arrest Care. AMI indicates acute myocardial infarction; FIO2, fraction of
inspired oxygen; IO, intraoral; IV, intravenous; PETCO2, partial pressure of end-tidal carbon dioxide; SBP, systolic blood pressure; SpO2, oxygen sat-
uration; STEMI, ST-elevation myocardial infarction. (Previously published. See “Credit Lines” section.)

Figure 73.3. American Heart Association Algorithm for Treatment of Adult Bradycardia (With Pulse). ECG indicates electrocardiography;
IV, intravenous. (Previously published. See “Credit Lines” section.)
684 VII Myocardial Infarction
Figure 73.4. American Heart Association Algorithm for Treatment of Adult Tachycardia (With Pulse). CHF indicates congestive heart failure;
ECG, electrocardiography; IV, intravenous; NS, normal saline; VT, ventricular tachycardia. (Previously published. See “Credit Lines” section.)
Polymorphic Ventricular Tachycardia
For patients presenting with PVT with prolonged QT interval
(torsades de pointes), use the following approach:
1. Initiate unsynchronized DC cardioversion if the patient is hemody-
namically unstable (150–200 J biphasic or 360 J monophasic).
2. Administer magnesium sulfate: 1 to 2 g IV in 5% dextrose in water
given over 5 to 60 minutes, even if the serum level of magnesium is
within the reference range.
3. Shorten the QT interval and eliminate triggering a long-short R-R
interval by increasing the heart rate to 100 to 120 beats per minute
with either of the following:
a. Isoproterenol infusion: start at 1 mcg/min up to 20 mcg/min
b. Temporary pacemaker
4. Correct underlying electrolyte abnormalities, such as hypokalemia
and hypomagnesemia.
5. Stop treatment with all QT-prolonging agents.
a. Antiarrhythmics: quinidine, procainamide, disopyramide, sotalol,
ibutilide, amiodarone.
b. Antipsychotics: haloperidol, thioridazine, droperidol.
c. Antibiotics: erythromycin.
d. Antivirals and antiprotozoals: amantadine, pentamidine.
e. Antifungal agents: ketoconazole, itraconazole.
f. Antihistamines: terfenadine, astemizole.
g. Propulsive agents: cisapride.
h. Hypolipidemic agents: probucol.
6. Note: β-blockers may be effective in preventing torsades de pointes
in congenital long-QT syndrome.
For patients presenting with PVT with normal QT interval,
use the following approach:
1. In hemodynamically unstable patients, initiate DC cardioversion
(100–200 J biphasic).
2. In hemodynamically stable patients, administer amiodarone and lido-
caine or correct myocardial ischemia or electrolyte disturbances.
a. Amiodarone: 150 mg IV given over 10 minutes, then 1 mg/min
for 6 hours followed by 0.5 mg/min.
b. Lidocaine: 1 to 1.5 mg/kg IV followed by 1 to 4 mg/min, supple-
mented by boluses of 0.5 to 0.75 mg/kg every 5 to 10 minutes to
a maximum of 3 mg/kg.
Electrical (or arrhythmic) storm is defined as drug-refractory,
recurrent, hemodynamically destabilizing, sustained PVT or VF
(VT storm) that may occur after acute MI related to uncontrolled
ischemia or increased sympathetic tone, requiring recurrent
cardioversion or defibrillation. For patients presenting with elec-
trical storm, use the following approach:
1. Initiate sympathetic blockade using
a. Propranolol: 0.15 mg/kg given IV over 10 minutes and then 3 to
5 mg every 6 hours to maintain sinus rhythm unless the heart rate
is less than 45 beats per minute, or
b. Esmolol: 500 mcg/kg IV over 1 minute and then 50 to 250 mcg/
kg/min; titrate maintenance infusion rate at 5- to 10-minute
intervals up to 250 mcg/kg/min, or
c. Left stellate ganglionic blockade (for refractory arrhythmia), or
 73 Cardiac Emergencies
d. Full sedation with mechanical ventilatory support
e. Amiodarone: 150 mg IV over 10 minutes; then 1 mg/min for 6
hours followed by 0.5 mg/min.
2. Consider intra-aortic balloon pump.
3. Perform emergency revascularization.
Atrial Flutter or Atrial Fibrillation With Rapid
Ventricular Response
Indications for Emergency Cardioversion
The following are indications for emergency cardioversion in
patients presenting with atrial flutter or atrial fibrillation with
rapid ventricular response:
1. Cardiogenic shock with hypoperfusion
2. Uncontrolled ongoing myocardial ischemia
3. Pulmonary edema
4. Pre-excitation syndrome, because of the risk of precipitation of VF.
Achieving Rate Control
There are several pharmacologic treatment options for achieving
rate control of a rapid ventricular response.
1. Verapamil: 2.5 to 5 mg IV over 2 to 3 minutes; repeat 5 to 10 mg
every 15 to 30 minutes to a total dose of 20 mg, or
2. Diltiazem: 0.25 mg/kg IV over 2 minutes; repeat 0.35 mg/kg in
15 minutes, and then give 5–15 mg/h titrated according to the heart
rate, or
3. Esmolol: 0.5 mg/kg IV over 1 minute; then 50 to 200 mcg/kg/min; or
4. Atenolol: 5 mg IV over 5 minutes to a total of 10 mg in 10 to
15 minutes, or
5. Metoprolol: 5 mg IV every 5 minutes to a total of 15 mg over
15 minutes, or
6. Propranolol: 0.1 mg/kg by slow IV push divided into 3 equal doses
at 2- to 3-minute intervals.
7. Digoxin: 0.25 mg IV every 2 hours, until ventricular rate control
is achieved up to a total maximum dose of 1.5 mg within 24 hours.
Reduce total dose to 1.0 mg for elderly or low lean body mass
patients.
Restoring Sinus Rhythm
There are 2 basic treatment options to restore sinus rhythm: elec-
trical or pharmacologic cardioversion. Further details are pro-
vided below:
1. Electrical: elective synchronized DC cardioversion.
2. Pharmacologic.
a. Ibutilide: 1 mg over 10 minutes, and repeat once after 10 min-
utes; watch for torsades de pointes and correct hypokalemia or
hypomagnesemia before administration, or
b. Amiodarone: 150 mg IV over 10 minutes, followed by an infusion
of 1 mg/min for 6 hours and then 0.5 mg/min over 18 hours.
General Management Principles for Atrial
Flutter and Atrial Fibrillation
Adenosine is not an effective therapy for atrial fibrillation or
atrial flutter. Heparin should be administered to all patients not
therapeutically anticoagulated, and anticoagulation should be
continued after cardioversion. Transesophageal echocardiog-
raphy is valuable prior to elective cardioversion to exclude left
atrial thrombus. Consider the use of an antiarrhythmic agent to
prevent a recurrence of atrial fibrillation or atrial flutter, such as
propafenone (with a structurally normal heart), sotalol (can be
used with ischemic heart disease), or amiodarone (often valuable
in the setting of left ventricular dysfunction).
685
Paroxysmal Supraventricular Tachycardia
For patients presenting with PSVT, asses hemodynamic status
and use the approach outlined below:
1. For patients with unstable hemodynamic status
a. Initiate synchronized DC cardioversion (sedate first if conscious)
2. For patients with stable hemodynamic status
a. Initiate vagal maneuvers (Valsalva maneuver or carotid sinus
massage)
b. Administer adenosine: 6 mg IV given rapidly followed by saline
flush; repeat 12 mg in 1 to 2 minutes and then once more if patient
does not convert
c. Administer verapamil, diltiazem, or esmolol
i. Verapamil: 2.5 to 5 mg IV over 2 to 3 minutes; repeat 5 to
10 mg every 15 to 30 minutes to a total dose of 20 mg, or
ii. Diltiazem: 0.25 mg/kg IV over 2 minutes; repeat 0.35 mg/kg
in 15 minutes and then 5 to 15 mg/h titrated according to the
heart rate, or
iii. Esmolol: 500 mcg/kg IV over 1 minute; then 50 to
200 mcg/kg/min.
d. Consider elective synchronized DC cardioversion (50–100 J).
Reciprocating Tachycardia Complicating
Wolff-Parkinson-White Syndrome
For patients presenting with reciprocating tachycardia compli-
cating Wolff-Parkinson-White syndrome, use the approach out-
lined below:
1. Maintain airway, assist breathing, and give oxygen.
2. Assess hemodynamic status.
a. In unstable patients
i. Initiate synchronized DC cardioversion (sedate patient first if
conscious).
b. In stable patients
i. Administer adenosine IV for patients with orthodromic con-
duction (narrow complex QRS) (may convert from recipro-
cating tachycardia to atrial fibrillation).
a) Atrial fibrillation with a rapid ventricular response may
require emergency synchronized DC cardioversion.
ii. Administer amiodarone or procainamide for conversion to
sinus rhythm.
a) Amiodarone: 150 mg IV over 10 minutes; repeat as
needed to maximum of 2.2 g daily, or
b) Procainamide: 1 g IV over 30 minutes followed by 1 to
4 mg/min (watch for hypotension)
iii. Note: For atrial fibrillation that complicates Wolff-Parkinson-
White syndrome, adenosine is contraindicated; also avoid AV
node–blocking agents (diltiazem, verapamil, or digoxin) that
can accelerate anterograde conduction over the bypass tract
and increase the ventricular rate.
iv. Consider DC cardioversion if pharmacologic treatment fails.
Multifocal Atrial Tachycardia
For patients presenting with multifocal atrial tachycardia, use the
approach outlined below:
1. Treat the underlying disease (frequently, lung disease and hypoxemia).
2. Stop treatment with provocative agents such as β-agonists or
theophylline.
3. Avoid β-blockers if there is underlying bronchospastic lung disease.
4. Note: magnesium or potassium administered IV may be effective
even if the serum levels of magnesium and potassium are within the
reference ranges
a. Magnesium sulfate: 1–2 g IV over 10 minutes; then 1–2 g/h over
5 hours
b. Potassium if hypokalemic
5. Note: verapamil (given orally or IV) is usually the drug of choice
(5–10 mg IV; repeat 5–10 mg for a total of 20 mg).
686 VII Myocardial Infarction
6. Note: digoxin or cardioversion is not effective in treating this
arrhythmia due to abnormalities in impulse initiation.
7. Always suspect digoxin toxicity in paroxysmal atrial tachycardia
with heart block, and treat accordingly.
Pacemaker-Mediated Tachycardia
PMT is a form of reentrant tachycardia that is observed in patients
with a dual-chamber pacemaker and retrograde ventriculoatrial
conduction. Ventricular depolarization (paced beat or PVC) ret-
rogradely activates the atria through the AV conduction system.
Atrial depolarization is sensed by the atrial lead and then trig-
gers the pacemaker to pace the ventricle, which is then again
retrogradely conducted to the atria to create a reentrant loop that
repeats the sequence, resulting in incessant tachycardia at or near
the programmed upper rate limit of the pacemaker (commonly
set at a paced rate of 120).
1. Increase retrograde AV conduction time or induce AV node block.
a. Vagal maneuvers (carotid sinus massage, Valsalva maneuver).
b. Adenosine, verapamil, or β-blocker, or
2. Make atrial lead insensitive to the retrograde P wave.
a. Apply a magnet to disable atrial tracking (especially when pace-
maker model is not known or pacemaker programmer is unavail-
able), or
b. Increase the PVARP, or
c. Reprogram pacemaker mode to DVI or VVI.
New pacemakers have the capability to detect PMT and initi-
ate PMT intervention. This is done by automatically prolonging
the PVARP (PVARP extension) for the beat after a ventricular-
sensed event that is not preceded by atrial pacing (such as a PVC)
or dropping a paced ventricular beat when the pacemaker is pac-
ing at the upper tracking rate for a specified period or shortening
the AV interval for a single beat to induce a retrograde AV block
and terminate the tachycardia.
Acute Coronary Syndromes
For patients with unstable angina or NSTEMI, use the approach
outlined below:
1. Administer oxygen: 4 L/min; maintain oxygen saturation greater
than 90%.
2. Administer aspirin: 160 to 325 mg orally.
3. Administer nitroglycerin: 3 doses (sublingual or aerosol) at 3
to 5 minutes, followed by IV dosage (5 to 10 mcg/min; increase
by 10 mcg/min every 3 to 5 minutes to a maximal dose of
200 mcg/min). Avoid in patients with low systolic blood pressure
(<90 mm Hg) or in those who have received a phosphodiesterase
inhibitor for erectile dysfunction within 24 hours.
4. Initiate sedation and analgesia (morphine sulfate: 2–4 mg IV every
5–10 minutes as tolerated).
5. Administer β-blocker: orally or IV unless contraindicated.
a. Metoprolol: 5 mg IV; repeat 3 times and then 50 to 100 mg orally
twice daily, or
b. Esmolol: 0.5 mg/kg/min IV initially and then titrate to 50 to
200 mcg/kg/min.
6. Consider clopidogrel: 300 mg orally.
7. Administer heparin.
a. Unfractionated heparin: 60 U/kg IV bolus followed by 12 U/kg/h
infusion to keep aPTT around 60 to 70 seconds, or
b. Low-molecular-weight heparin.
8. Consider glycoprotein IIb/IIIa inhibitor.
9. Use intra-aortic balloon pump to stabilize the condition of selected
patients before coronary artery bypass graft or PCI.
10. Initiate emergency coronary angiography and intervention for
refractory ischemia (persistent chest pain, recurrent ST-segment
deviation, hemodynamic instability, VT, heart failure).
11. Administer an ACE inhibitor or angiotensin receptor blocker.
12. Administer an HMG-CoA reductase inhibitor (statin).
For patients with STEMI or with new left bundle branch block
use the approach indicated below:
1. Administer oxygen: 4 L/min; maintain oxygen saturation greater
than 90%.
2. Give aspirin: 160 to 325 mg orally.
3. Administer nitroglycerin: sublingual or aerosol followed by IV
(5–10 mcg/min; increase by 10 mcg/min every 3–5 minutes to a
maximal dosage of 200 mcg/min). Avoid systolic blood pressure
less than 90 mm Hg.
4. Initiate sedation and analgesia (morphine sulfate: 2–4 mg every
5–10 minutes as tolerated).
5. Administer β-blocker: given orally or IV unless contraindicated.
a. Metoprolol: 5 mg IV; repeat 3 times and then 50 to 100 mg
orally twice daily, or
b. Esmolol: 0.5 mg/kg/min initially and then titrate to 50 to
200 mcg/kg/min.
6. Avoid calcium channel blockers, if possible.
7. Consider clopidogrel: 300 mg orally.
8. Administer heparin.
a. Unfractionated heparin: 60 U/kg IV bolus followed by 12 U/kg/h
infusion to keep aPTT around 60 to 70 seconds, or
b. Low-molecular-weight heparin.
9. Time from chest pain onset: 12 hours or less.
10. Initiate reperfusion therapy: fibrinolysis (goal: door-to-drug
≤30 minutes) or emergency PCI (goal: door-to-balloon inflation
≤90 minutes).
a. Criteria for considering fibrinolysis.
i. Early presentation (<3 hours from symptom onset)
ii. Emergency PCI is not an option or would be delayed
(>90 minutes)
iii. No contraindications to fibrinolysis.
b. Absolute contraindications for fibrinolysis
i. Presence of structural cerebral lesion or intracranial
neoplasm
ii. Any prior intracranial hemorrhage
iii. Ischemic stroke within 3 months, except when used to treat
an acute stroke within 3 hours of onset
iv. Suspected aortic dissection
v. Significant closed head or facial trauma within 3 months
vi. Major surgery or trauma (including traumatic CPR) within
3 weeks
vii. Known active bleeding or bleeding diathesis
c. Fibrinolytic agents
i. Streptokinase: 1.5 million IU IV over 1 hour
ii. Alteplase (tissue plasminogen activator): 15-mg bolus IV,
followed by 0.75 mg/kg (up to 50 mg) over 30 minutes; then
0.5 mg/kg (up to 35 mg) infusion over 1 hour
iii. Tenecteplase: single bolus of 0.5 mg/kg.
11. Administer inotropic agents for cardiogenic shock.
a. Dobutamine: 2 to 15 mcg/kg/min.
b. Dopamine: 0.5 to 10 mcg/kg/min (increases renal blood flow at
<2 mcg/kg/min only).
c. Amrinone: 0.75 mg/kg followed by 5 to 10 mcg/kg/min.
12. Initiate intra-aortic balloon pump for selected patients with hypo-
tension, or emergency PCI for cardiogenic shock.
13. Administer fluids given IV for right ventricular infarction (avoid
nitroglycerin or diuretics).
14. Lidocaine for treatment of ventricular arrhythmias but not for prophy-
laxis–loading dose of 100 mg IV followed by a second loading dose of
100 mg IV 30 minutes later followed by an infusion at 2–4 mg/min.
15. Avoid calcium channel blockers.
 73 Cardiac Emergencies
16. Administer magnesium sulfate for patients with low serum levels of
magnesium.
17. Administer an ACE inhibitor unless contraindicated—captopril:
start with 6.25 mg orally and increase to 50 mg orally twice daily as
tolerated.
18. Administer an HMG-CoA reductase inhibitor (statin).
Right Ventricular Ischemia or Infarction
For patients presenting with right ventricular ischemia or infarc-
tion, use the following approach.
1. Maintain right ventricular preload with volume loading (IV saline).
a. Avoid nitrates and diuretics because they decrease preload.
b. Maintain AV synchrony, and initiate AV pacing for sympto-
matic high-grade AV block unresponsive to atropine.
c. Note: Prompt cardioversion is important for hemodynamically
significant supraventricular tachycardia (atrial fibrillation may
occur in up to one-third of patients).
2. Initiate inotropic support
a. Administer IV dobutamine if cardiac output fails to improve
after infusion of 1 to 2 L of saline solution.
b. Inotropic agents can increase cardiac output in right ventric-
ular infarction by making septal wall motion hyperdynamic
even in the presence of severe right ventricular free wall
hypokinesia.
Pulmonary Edema
For patients presenting with pulmonary edema, use the approach
indicated below.
1. Identify and treat precipitating factors, if possible.
2. Place the patient in a sitting position to improve oxygenation.
3. Give supplemental oxygen by mask or nasal cannula to maintain
oxygen saturation greater than 90%.
4. Intubate and mechanically ventilate the patient if hypoxia or hyper-
capnia cannot be corrected adequately.
5. Give morphine sulfate: 2 to 5 mg IV; repeat every 10 to 25 minutes
if needed.
6. Give a loop diuretic such as furosemide: 20 to 80 mg IV.
7. Start an infusion of nitroglycerin.
a. Nitroglycerin is a venodilator that predominantly affects cardiac
preload.
b. Nitroglycerin: initially 5 mcg/min IV; increase by 5 to 10 mcg/
min every 3 to 5 minutes to a maximal dosage of 200 mcg/min
as tolerated.
8. Give nitroprusside.
a. Nitroprusside is a mixed arterial and venous vasodilator that
decreases both cardiac preload and afterload and is indicated
specifically in pulmonary edema due to systemic hypertension
and acute mitral or aortic insufficiency.
b. Nitroprusside: initially 0.3 mcg/kg/min IV; titrate up to 10 mcg/
kg/min as needed.
9. Note: Inotropic agents (eg, dopamine, dobutamine) may be helpful
in pulmonary edema due to hypotension and cardiogenic shock.
a. Dopamine: 0.5 to 10 mcg/kg/min.
b. Dobutamine: 2 to 15 mcg/kg/min.
10. Consider an intra-aortic balloon pump in selected patients.
Hypertensive Emergencies
For patients presenting with hypertensive emergency, begin treat-
ment using the agents indicated below.
1. Nitroprusside: for hypertensive encephalopathy, cerebral hemor-
rhage, clonidine withdrawal syndrome, and aortic dissection (in
combination with a β-blocker)
687
a. Administer only in an intensive care setting and with intra-arte-
rial blood pressure monitoring.
b. Initially 0.3 mcg/kg/min IV; titrate slowly up to 10 mcg/kg/min.
2. Nitroglycerin: 10 to 400 mcg/min IV
a. Use after MI and in left ventricular failure
b. Captopril (see below) is another option for these indications.
3. Esmolol: initially 200 mcg/kg IV for 1 minute; then 25 to 300
mcg/kg/min.
4. Labetalol: initially 20 mg IV; repeat with 20 to 40 mg every
10 minutes or 0.5 to 2.0 mg/min continuous infusion, up to 300 mg.
a. Use for an adrenergic crisis (eg, cocaine overdose, monoamine
oxidase inhibitor–induced hypertensive crisis, and pheochro-
mocytoma).
b. Phentolamine (see below) is another option for these indications.
5. Captopril or enalapril
a. Captopril: 6.25 to 50 mg orally at hourly intervals
i. Use after MI and in left ventricular failure
ii. Nitroglycerin (see above) is another option for these
indications.
b. Enalapril: 0.625 to 1.25 mg IV every 6 hours.
6. Phentolamine IV: 5–20 mg
a. Use for an adrenergic crisis (cocaine overdose, monoamine
oxidase inhibitor–induced hypertensive crisis and pheochro-
mocytoma).
b. Labetalol (see above) is another option for these indications.
Structural, Metabolic, or Toxic Emergencies
Aortic Dissection
For patients presenting with dissection of the ascending aorta,
use the approach indicated below:
1. In an emergency: treat surgically or invasively if possible.
2. In dissection of the ascending aortic arch, monitor closely for signs
of complications such as acute aortic valve incompetence, inferior
MI due to extension of the dissection into the right coronary artery,
neurologic signs due to dissection into the arch vessels, and ische-
mia of the upper limbs.
For patients presenting with dissection of the descending
aorta, use the approach indicated below:
1. For most patients, treat surgically or invasively.
2. For patients without organ ischemia, treat medically.
Blood Pressure Management
In the setting of aortic dissection, control blood pressure using
the following principles:
1. Reduce systemic systolic blood pressure to a level of 90 to 100 mm
Hg (measured with an arterial line) or to the lowest blood pressure
that permits adequate organ perfusion (monitored with urinary out-
put) and cerebral function.
2. Avoid increasing shear forces on the intimal flap of the aorta.
3. Administer nitroprusside in combination with esmolol.
a. Nitroprusside: initially 0.3 mcg/kg/min; increase up to 10
mcg/kg/min.
b. Esmolol: initially 500 mcg/kg IV for 1 minute; titrate to 250 to
300 mcg/kg/min to achieve a target heart rate of 60 to 70 beats
per minute.
Pericardial Tamponade
For patients presenting with nontraumatic cardiac tamponade,
the treatment of choice is pericardiocentesis under echocardio-
graphic guidance. For those presenting with traumatic cardiac
tamponade or tamponade due to aortic dissection, emergency
688 VII Myocardial Infarction
thoracotomy should be performed. Volume expansion with
saline, blood, or plasma is valuable pending pericardiocentesis
or surgery. Avoid mechanical ventilation and β-blockade.
Diuretics and nitrates are contraindicated.
Pulmonary Thromboembolism
For patients presenting with pulmonary thromboembolism, initi-
ate supplemental oxygen and anticoagulation.
1. Start full-dose heparin anticoagulation along with infusion of a lytic
agent.
a. Heparin: bolus of 7,500 to 10,000 U IV; maintenance dosage is
20 U/kg/h.
b. Keep aPTT between 60 to 80 seconds.
2. Give a thrombolytic agent IV for hemodynamically unstable or pro-
foundly hypoxemic patient with evidence of massive pulmonary
embolism even without overt hemodynamic compromise (unless
contraindicated, as in a postoperative state).
a. Streptokinase: bolus of 250,000 U; maintenance dosage of
100,000 U/h for 24 to 72 hours.
b. Urokinase: bolus of 4,400 U/kg; maintenance dosage of 4,400
U/kg/h for 12 hours.
c. Alteplase: bolus of 15 mg; then 50 mg during 30 minutes; then
35 mg during the next hour.
3. Administer aggressive fluid resuscitation and inotropic or pressor
agents if hypotension develops.
4. Consider surgical thrombectomy only with severe hemodynamic
compromise, when thrombolytic therapy is contraindicated or
unsuccessful.
Hyperkalemia
For patients presenting with hyperkalemia, use the following
treatment approach:
1. Initiate continuous ECG monitoring.
2. Administer calcium gluconate or calcium chloride: 10 mL of 10%
solution in 10 minutes.
3. Give glucose with insulin: 50 mL of 50% dextrose IV, with 5 to 10 U
regular insulin sodium bicarbonate: 50 mEq IV in 5 minutes.
5. Give a β2-adrenergic agonist—albuterol: 10 to 20 mg nebulized in
15 minutes or IV.
6. Initiate diuresis with furosemide: 40 to 80 mg IV.
7. Administer sodium polystyrene sulfonate (Kayexalate) (20–30 g
orally with equal amount of sorbitol) or retention enema (50–100 g
in 200 mL of water and sorbitol).
8. Note: Dialysis may be required for persistent hyperkalemia.
9. Correct underlying defect (eg, renal failure, potassium-sparing
diuretics, or adrenal insufficiency).
Digoxin Toxicity and Overdose
In patients presenting with digoxin toxicity, specific cardiac
arrhythmias include VT (especially bidirectional VT), sinus
bradycardia, heart block, and paroxysmal atrial tachycardia with
block. Nausea, vomiting, and drowsiness may occur.
For treatment of digoxin overdose, use the principles outlined
below.
1. Induce emesis or perform gastric lavage, and administer charcoal to
decrease further absorption of the drug.
2. Administer digoxin-immune Fab: give 10 mcg to test for hypersensi-
tivity initially; then give up to 20 vials (dose according to calculated
digoxin body load if known) through a 0.22-mcm membrane filter
for the following:
a. VT or VF
b. High-grade AV block not responding to atropine or hyperkalemia
c. Acute ingestion of 10 mg of digoxin or more
d. Serum level of 10 ng/mL or greater or severe hyperkalemia.
3. Treat hyperkalemia in the usual way with insulin, glucose, and bicar-
bonate, but avoid calcium gluconate because it potentiates digoxin
toxicity.
4. Treat high-grade AV block with atropine and temporary pacing if
needed.
5. Ventricular arrhythmias
a. Lidocaine; phenytoin (100 mg IV in 5 minutes and repeat 100 mg
in 5 minutes up to a full loading dose of 18 mg/kg); esmolol
(500 mcg/kg IV, and then 50–200 mcg/kg/min IV); or magne-
sium (1–2 g IV in 10 minutes).
b. If hemodynamically unstable, use synchronized DC cardiover-
sion beginning at 10 J and increasing by 10 J to 50 J; then use
100 J, 200 J, 300 J, and 360 J if necessary.
6. Correct hypokalemia or hypomagnesemia.
β-Blocker Toxicity and Overdose
For patients presenting with β-blocker toxicity, the first priority is
emergency care to stabilize the airway, breathing, and circulation.
1. Monitor ECG.
2. Perform gastric lavage and administer charcoal, but avoid emesis
because of vagal side effects.
3. Treat hypotension with saline and isoproterenol (initially 2–20
mcg/min IV; increase up to 200 mcg/min).
4. Administer glucagon: 50 to 150 mcg/kg IV in 1 minute; then 1 to
5 mg/h in 5% dextrose solution.
5. Give calcium chloride or calcium gluconate (10 mL of 10% solution
IV in 10 minutes) may be useful for myocardial depression.
6. Treat bronchospasm with β-adrenergic agonists (administered sys-
temically or by inhalation) or with theophylline.
7. Treat heart block with atropine (up to 2 mg IV) or temporary pacing.
8. Treat hypoglycemia with glucose and glucagon given IV.
9. Note: Sotalol overdose may cause torsades de pointes.
Calcium Channel Blocker Toxicity
and Overdose
For patients presenting with calcium channel blocker toxicity,
as with β-blocker toxicity, the goal of treatment is to decrease
absorption of the drug and increase perfusion to the critical organ
system.
1. Perform gastric lavage and administer charcoal, but avoid emesis
because of vagal side effects.
2. Treat hypotension and bradycardia.
a. Administer saline infusion and IV calcium chloride: bolus of 1 g
(10 mL of 10% solution in 10 minutes); then a continuous infu-
sion (20–50 mg/kg/h).
b. Note: Calcium may have deleterious effects in patients taking
digoxin.
3. Treat high-grade heart block with atropine (up to 2 mg IV) or tempo-
rary pacing.
4. Consider treatment with glucagon (50–150 mcg/kg IV in 1 minute
followed by 1–5 mg/h) for hypotension and heart block.
5. Use dobutamine or dopamine to treat heart failure.
6. Use norepinephrine to treat hypotension.
Abbreviations
ACE angiotensin-converting enzyme
aPTT activated partial thromboplastin time
AV atrioventricular
CPR cardiopulmonary resuscitation
DC direct current
 73 Cardiac Emergencies 689

ECG electrocardiogram PVARP postventricular atrial refractory period

IV intravenous PVC premature ventricular contraction
MI myocardial infarction PVT polymorphic ventricular tachycardia
NSTEMI non–ST-segment elevation myocardial infarction STEMI ST-segment elevation myocardial infarction
PCI percutaneous coronary intervention VF ventricular fibrillation
PMT pacemaker-mediated tachycardia VT ventricular tachycardia
 74

Risk Stratification After Myocardial Infarction

RANDAL J. THOMAS, MD, MS

Background for instance, patients were considered to be at high risk of

Risk stratification after MI is the process of estimating the risk
of future CVD events in survivors of MI. This process helps
health care providers tailor short- and long-term management
plans, with the goal of applying cost-effective treatments that
are most appropriate for the risk level of the patient and that
are most likely to optimize the patient’s quality and quantity
of life.
Risk assessment strategies are applicable to patient care at
several times during and after MI, including at the assessment
by emergency medical technicians, on arrival at the emergency
department, at admission to the intensive care unit, and at the
assessment before and soon after hospital discharge. Although
each of these settings is vitally important in risk-stratified treat-
ment approaches to MI, this review focuses on principles that are
particularly important just before or just after hospital discharge
after MI.
Risk-stratified treatment approaches continue to be impor-
tant, even though mortality rates for patients with CVD have
decreased approximately 50% in the United States over the past
4 decades. Evidence suggests that this decrease in mortality is
due to at least 2 factors: 1) a reduction in the incidence of CVD
events and 2) a reduction in case fatality after a CVD event. The
incidence of MI has decreased in the United States by approxi-
mately 20% to 30% during the past 4 decades. In addition, 1-year
mortality after MI decreased from approximately 20% in 1970 to
approximately 5% in 2000.
As improvements in MI-related outcomes have occurred
over the past 4 decades, definitions have changed for low risk,
moderate risk, and high risk. In the era before thrombolytics,
Abbreviations and acronyms are expanded at the end of this chapter.
future CVD events if their estimated 1-year mortality risk
was greater than 10%; at moderate risk, if 5% to 10%; and
at low risk, if less than 5%. Today, because most patients
have a much smaller risk after MI than did patients in previ-
ous decades, 1-year mortality risks for high-risk patients are
greater than 3%; for moderate-risk patients, 1% to 3%; and for
low-risk patients, less than 1%. With new, more effective, and
more costly treatment strategies emerging, the risk-stratified
approach to cost- effective and evidence-based risk reduction is
as important as ever.
Basic Principles
To incorporate risk stratification into the management of MI,
several key principles should be considered. In MI survivors,
future CVD events are generally due to 1 or more of the follow-
ing high-risk conditions:
1. Recurrent myocardial ischemia
2. Left ventricular dysfunction
3. High-grade arrhythmias
4. Other complications after MI (renal failure, stroke, etc)
To help identify people with these high-risk conditions and to
estimate a patient’s risk of future CVD events after MI, various
patient characteristics can be used, including symptoms, physical
findings, and test results. After a patient’s post-MI risk is assessed,
that risk can be used to identify and deliver an evidence-based,
risk-stratified approach to treatment.
• Several treatments are appropriate for nearly all post-MI patients,
no matter what their level of risk, and can greatly reduce the risk
of future CVD events, with a low risk of side effects and at low to
moderate cost. These include the ABCs of treatment (Box 74.1).

690
 74 Risk Stratifi cation After Myocardial Infarction
Box 74.1. ABCs of Secondary Prevention After
Myocardial Infarction
A Antiplatelet agent
Angiotensin-converting enzyme inhibitor
(for left ventricular dysfunction)
Avoid tobacco
B β-Blockers
Blood pressure control
C Cholesterol-lowering medications
D Dietary control
Diabetes control
E Exercise
F Fish oil (ω-3 fatty acids)
• Some invasive therapeutic options, such as coronary revascular-
ization procedures, can reduce CVD risk significantly, especially
among patients at higher risk, but the risks and costs are relatively
high compared with the ABCs of treatment.
• The risk-benefit ratio of the more invasive therapeutic options is
generally better for patients at higher risk of future CVD events
than for patients at lower risk of future CVD events.
Risk stratification helps guide various parts of care after hos-
pital discharge, including cardiac rehabilitation and return-to-
work recommendations. Patients at highest risk generally have
a more gradual and monitored progression of post-MI therapy
and recuperation than do post-MI patients at lower risk of future
CVD events.
Methods of Risk Estimation
Several published reports and consensus statements have sug-
gested frameworks by which risk stratification can be carried
out after MI (Table 74.1). Published experience from Olmsted
County, Minnesota, suggests the following:
1. Risk-stratification scoring systems, such as the one from the
PREDICT trial, appear to perform equally well for STEMI and for
non-STEMI, particularly if they include information about comor-
bid conditions.
2. LVEF is a powerful predictor of future CVD events in post-MI
patients and should be included in risk-stratification strategies.
Assessment of LVEF to guide the use of ICDs after MI may be more
accurate when performed at least 6 weeks after MI than within the
first 6 weeks after MI.
Published guidelines for risk-stratification strategies after
STEMI were written by a joint panel from the American College
of Cardiology and the American Heart Association. Highlights
from those guidelines are shown in Figures 74.1 and 74.2. On the
basis of these guidelines, an approach to risk stratification after
MI could include the following 4 steps.
1. Assess Patient for High-Risk Symptoms,
Signs, and Comorbid Conditions
All post-MI patients should be assessed for high-risk symptoms
and signs, including those that suggest ischemia (angina), left
ventricular dysfunction (dyspnea, rales, and ventricular gal-
lop), and high-grade arrhythmias (documented ventricular
tachycardia or fibrillation, history of cardiac arrest, syncope or
presyncope or both). Comorbid conditions that add prognostic
information to risk stratification include age, prior MI, pre-MI
691
angina, hypertension, stroke, coronary artery bypass grafting
or cardiac arrest, and renal dysfunction. Patients with high-
risk findings should be considered for more aggressive evalu-
ation (eg, coronary angiography) and treatment options (eg,
revascularization).
2. Assess Left Ventricular Function
After MI, LVEF is a powerful predictor of future CVD events.
For post-MI patients with an LVEF less than 30%, 6-month
mortality is approximately 15%; for those with an LVEF more
than 50%, 6-month mortality is 1% to 2%. LVEF can be mea-
sured with echocardiography, nuclear imaging, or coronary
angiography.
3. Assess With Exercise Testing
Exercise testing, with or without echocardiographic or nuclear
imaging, has been a long-standing, noninvasive approach to
risk stratification of post-MI patients. In addition to helping to
identify inducible ischemia, an important high-risk condition
after MI, exercise testing helps assess exercise capacity, an even
more powerful predictor of future CVD events. Exercise test-
ing also provides information about heart rate and blood pres-
sure responses to exercise, factors that can add to risk prediction
for CVD events. Information about exercise capacity also helps
tailor the exercise prescription during cardiac rehabilitation ses-
sions and helps guide return-to-work recommendations for the
post-MI patient.
4. Assess for Arrhythmia Risk
A small percentage of MI patients have high-grade ventricular
arrhythmias during hospitalization and are candidates for fur-
ther electrophysiologic testing and possible treatment with an
ICD. For other patients, further tests for significant arrhythmias
have relatively low yield and are generally not recommended,
although methods for assessing autonomic tone may have a role
in risk stratification. Recent studies (including the MADIT-II
study) suggest, however, that post-MI patients with an LVEF
less than 30% derive significant survival benefit from ICD ther-
apy. Patients are considered possible candidates for ICD place-
ment if their LVEF is less than 30% at least 1 month after MI.
Given the significant cost of an ICD, however, there is consid-
erable debate on the cost-effectiveness of such an approach to
risk reduction.
Summary
Several strategies are available to help risk-stratify patients after
MI, according to their risk of future CVD events. The strate-
gies include various patient characteristics and test results that
can help risk-stratify patients at every point along the health
care continuum, from the initial evaluation by emergency med-
ical technicians to the evaluation visit in the physician’s office a
month after hospital discharge.
Risk-stratification methods are designed to help identify
the patients at increased risk of future CVD events who would
most likely benefit from more aggressive management strategies
(ie, coronary angiography or revascularization, or both). More
importantly, these methods help to identify lower-risk patients in
whom aggressive treatments would be less beneficial and would
instead result in additional adverse risks and costs.
Table 74.1. Features Used for Scoring in Selected Clinical Trials for Post-MI Risk Stratification
Feature TIMI-STEMI GUSTO-STEMI TIMI-NSTEMI PURSUIT-NSTEMI PREDICT CCP
Historical data Age 65–74 y, ≥75 y Age Age ≥65 y Age Age Older age
Diabetes mellitus/hypertension Prior MI Pre-MI angina Sex Prior MI, pre-MI angina, CABG
or angina Prior coronary stenosis ≥50% Pre-MI angina or cardiac arrest, hypertension,
Aspirin in prior 7 d stroke
≥3 CAD risk factors
Hemodynamics Systolic blood pressure In-hospital CHF Systolic blood pressure Shock CHF/pulmonary edema or
<100 mm Hg HR HR CHF cardiomegaly
HR >100 bpm LVEF CHF LVEF
Killip class II-IV
ECG Left bundle branch block/anterior ST-segment deviation ST-segment depression ECG severity score
ST-segment elevation
Comorbidity Weight <67 kg Charlson index Body mass index <20 kg/m 2
Renal function Renal function
Urinary incontinence
Assisted mobility
Peripheral vascular disease
Other Time to treatment >4 h Elevated biomarkers
Abbreviations: bpm, beats per minute; CABG, coronary artery bypass grafting; CAD, coronary artery disease; CHF, congestive heart failure; ECG, electrocardiogram; HR, heart rate; LVEF, left ventricular ejection
fraction; MI, myocardial infarction.
Previously published. See “Credit Lines” section.
 74 Risk Stratifi cation After Myocardial Infarction 693

STEMI

Primary invasive Fibrinolytic No reperfusion

strategy therapy therapy

Cath No cath
performed performed

EF >40% EF ≤40% EF ≤40% EF >40%

No high-risk High-risk High-risk No high-risk

features features features features

Cath and
Revascularization
revascularization
as indicated
as indicated

Functional
evaluation

ECG ECG
interpretable uninterpretable

Able to exercise Unable to exercise Able to exercise

Pharmacologic
stress

Submaximal Symptom-limited Adenosine or Exercise

Dobutamine Exercise
exercise test exercise test before dipyridamole nuclear
echo echo
before discharge or after discharge nuclear scan scan

Cath and Clinically significant No clinically

revascularization Medical therapy
ischemia significant ischemia
as indicated

Figure 74.1. Risk-stratified management approaches to identify patients with ST-segment elevation myocardial infarction (STEMI) who would
most likely benefit from more aggressive management strategies, including coronary angiography and revascularization. Cath indicates catheteriza-
tion; ECG, electrocardiogram; echo, echocardiography; EF, ejection fraction. (Previously published. See “Credit Lines” section.)
694 VII Myocardial Infarction

No spontaneous VF or sustained VT
>48 h after STEMI
EF measured ≥1 mo after STEMI

Path A Path B Path C

EF <30% EF =31%-40% EF >40%

Is there additional
Class IIa evidence of electrical No
LOE: B instability (eg, NSVT)?

Class IIb
LOE: B Class III
Yes LOE: B

ICD Positive EPS Negative No ICD

Class I Class IIb
LOE: B LOE: B

Figure 74.2. Algorithm for management of ST-segment elevation myocardial infarction (STEMI) to help select patients for implantable cardio-
verter-defibrillator (ICD) therapy. EF indicates ejection fraction; EPS, electrophysiologic studies; LOE, level of evidence; NSVT, nonsustained ven-
tricular tachycardia; VF, ventricular fibrillation; VT, ventricular tachycardia. (Previously published. See “Credit Lines” section.)

A simplified, evidence-based approach to risk stratification Names of Clinical Trials

for MI patients at hospital discharge (or soon after) includes the CCP Cooperative Cardiovascular Project
following: GUSTO-STEMI Global Use of Strategies to Open Occluded
Coronary Arteries—ST-Segment Elevation
1. Assessment of high-risk symptoms, signs, or comorbid conditions Myocardial Infarction
2. Assessment of left ventricular function MADIT-II Multicenter Automatic Defibrillator Implan-
3. Exercise testing tation Trial II
4. Assessment of arrhythmia risk PREDICT Predilatation vs Direct Stenting in Coronary
Treatment
PURSUIT-NSTEMI Platelet Glycoprotein IIb/IIIa in Unstable
Angina: Receptor Suppression Using Integrilin
Abbreviations Therapy—Non–ST-Segment Elevation
CVD cardiovascular disease Myocardial Infarction
ICD implantable cardioverter-defibrillator TIMI-NSTEMI Thrombolysis in Myocardial Infarction—Non–
LVEF left ventricular ejection fraction ST-Segment Elevation Myocardial Infarction
MI myocardial infarction TIMI-STEMI Thrombolysis in Myocardial Infarction—
STEMI ST-segment elevation myocardial infarction ST-Segment Elevation Myocardial Infarction
 75
Cardiac Rehabilitationa
THOMAS G. ALLISON, PHD, MPH
Definition of CR
Cardiac rehabilitation is defined by the World Health Organi-
zation as the sum of activities required to ensure patients the
best possible physical, mental and social conditions so that they
may resume and maintain as normal a place as possible in the
community.
The US Public Health Service defines cardiac rehabilitation
as follows:
Cardiac rehabilitation services are comprehensive, long-
term programs involving medical evaluation, prescribed
exercise, cardiac risk factor modification, education, and
counseling. These programs are designed to limit the phys-
iologic and psychological effects of cardiac illness, reduce
the risk for sudden death or reinfarction, control cardiac
symptoms, stabilize or reverse the atherosclerotic process,
and enhance the psychosocial and vocational status of
selected patients.
CR services are prescribed for patients who
1. have had a recent MI
a •
Portions previously published in King ML, Williams MA, Fletcher
GF, Gordon NF, Gulanick M, King CN, et al; American Heart
Association; American Association for Cardiovascular and Pulmonary
Rehabilitation. Medical director responsibilities for outpatient cardiac
rehabilitation/ secondary prevention programs: a scientific statement
from the American Heart Association/American Association for
Cardiovascular and Pulmonary Rehabilitation. Circulation. 2005 Nov
22;112(21):3354–60 and Cardiac Rehabilitation Guideline Panel.
Cardiac rehabilitation: AHCPR Clinical Practice Guidelines, No. 17.
Report No.: 96–0672. Rockville (MD): Agency for Health Care Policy
and Research (AHCPR); October 1995. Used with permission.
Abbreviations are expanded at the end of this chapter.
2. have had recent coronary bypass surgery
3. have chronic stable angina pectoris
4. have had recent PCI
5. have had a recent cardiac transplant, or
6. have had recent valve replacement or repair.
The services are provided in 3 phases beginning during hos-
pitalization (phase I), followed by a supervised ambulatory
outpatient program of a maximum of 36 monitored exercise ses-
sions that are to be completed within 1 year of the index event
(phases II and III), and continuing in a lifetime maintenance
stage in which physical fitness and risk factor reduction are
accomplished in a minimally supervised or unsupervised set-
ting (phase IV).
Exclusion criteria for CR include the following conditions:
• unstable angina
• class IV heart failure
• uncontrolled sustained tachyarrhythmias or bradyarrhythmias,
• severe and symptomatic aortic or mitral stenosis,
• hypertrophic obstructive cardiomyopathy,
• severe pulmonary hypertension
other conditions that could be aggravated by exercise such as
° resting systolic blood pressure > 200 mm Hg or
° resting diastolic blood pressure > 110 mm Hg,active or suspected
myocarditis or pericarditis, thrombophlebitis
° and recent significant systemic or pulmonary embolus.
Many authors feel that CR may be helpful for patients with
chronic heart failure, peripheral vascular disease, or a history of
stroke. As this chapter is being written, however, patients with
these 3 diagnoses are not eligible for CR reimbursement. Many
cardiac rehabilitation programs accommodate these patients in a
self-pay phase IV program setting.
695
696 VII Myocardial Infarction
CR has also been endorsed by the professional medical
community. The American Heart Association and American
College of Cardiology Joint Guidelines on Management of
Patients with Acute Myocardial Infarction (1999 update) states
the following:
The majority of patients need to modify their lifestyle after
acute MI. Typical recommendations require a change in
previous behavior. . . . Achievement of these goals is often
complicated by denial of the significance of the event,
physical deconditioning that may reflect a lifelong history
of sedentary behavior, and emotional distress . . . (and) may
be facilitated through participation in a formal cardiac
rehabilitation program.
CR has been proposed as one of the components of “pay for
performance” for patients hospitalized with a diagnosis appro-
priate for cardiac rehabilitation, and this proposal has been
endorsed by the American College of Cardiology.
Historical Perspective
To understand the history of CR, it is important to understand the
management of CAD in the 1950s and 1960s. CR developed in
large part to deal with the effects of those management practices
on exercise capacity, quality of life, and survival of patients. In
general, CAD patients at that time were younger and more likely
to be male than today; age-adjusted rates for CAD have been
decreasing since 1967—for men especially, in part due to the
large reduction in smoking. Management of acute MI or unsta-
ble angina from 1950 to 1960 largely involved treating symp-
toms with nitroglycerin and morphine. Procedures to improve
or restore blood flow, such as percutaneous coronary interven-
tion, reperfusion therapy, or coronary artery bypass grafting,
were unknown. There was not a good understanding of risk fac-
tors in the pathogenesis of CAD. Drugs with survival benefit—
statins, angiotension-converting enzyme inhibitors, β-blockers,
and aspirin—were either nonexistent or not yet known to have a
protective effect.
It is instructive to review the treatment schedule devised by
Dr. Paul Dudley White for President Dwight Eisenhower’s heart
attack in 1955 (Table 75.1). After remaining in the hospital for
7 weeks, Eisenhower still was not considered capable of climbing
Table 75.1. Schedule of Events for President Dwight
Eisenhower’s Medical Care Postmyocardial Infarction
Date in 1955 Event
September 24 Myocardial infarction; bedrest prescribed
October 11 First allowed to see a cabinet member
October 22 Sitting up in a chair for a few hours each day
and holding daily conferences about his
presidential duties
November 7 Walking and starting to climb stairs
November 11 Returned to Washington, DC, after delaying the
trip 1 month so that Eisenhower would not be
seen being wheeled to the airplane and lifted
on board; after landing in Washington, DC,
goes to his farm in Gettysburg, Pennsylvania.
December 26 Eisenhower doubts whether he should run for a
second term with that “sword of Damocles”
over his head and suggests that Vice President
Nixon should run instead
Figure 75.1. President Eisenhower leaving Denver to return to
Washington, DC, 7 weeks postmyocardial infarction.
the steps into the airplane for his return to Washington, DC, from
Denver, Colorado where he had the heart attack (Figure 75.1).
Interestingly, White was criticized for mobilizing the President
too rapidly.
As a result of the lack of myocardium-sparing therapies and
reperfusion strategies to control ischemia in patients with acute
coronary syndromes, transmural infarction, ongoing ischemia,
and death were common. Exercise remained one of the few
options for control of symptoms and improvement of functional
capacity, although exercise also carried greater risk because of
uncontrolled ischemia and reduced left ventricular function.
Early physician pioneers in CR included Herman Hellerstein
at Case-Western, Nanette Wenger at Emory, and Terence
Kavanagh in Toronto. The emphasis in early CR programs was
largely on aerobic training. Education was a lesser component,
and risk factor management was an even smaller part of the
effort. Vigorous exercise was pursued not only for its anti-angi-
nal (and, as later determined) survival benefits but also, to some
degree, to educate the cardiology community, demystify CAD,
and to free patients from draconian restrictions, social stigma,
and inappropriate fear and anxiety over their condition. Exercise
physiologists, such Philip Wilson in La Crosse, Wisconsin, were
instrumental in developing CR programs at a time when exercise
programs were not available in hospitals and the medical com-
munity in general considered exercise to be of greater risk than
benefit to CAD patients. Currently, appropriate physical activity
is recommended for all patients with CAD, including patients
with reduced left ventricular function and compensated heart
failure.
Improvements in the management of CAD and MI and atti-
tudes toward physical activity for CAD patients have greatly
shaped the nature and content of CR programs. Exercise is not
only more generally recognized to be important in long-term risk
reduction in CAD, but exercise is also safer because of better
care during the acute event and aggressive use of revasculariza-
tion and drugs to control ischemia and prevent disease progres-
sion. The duration of inpatient programs gradually decreased as
the length of hospital stays decreased markedly. Education and
counseling gradually became more important since activity was
much less restricted—partly because of changing attitudes but
 75 Cardiac Rehabilitation
largely because of myocardial salvage and aggressive manage-
ment of ischemia. Outpatient programs largely have moved into
hospital settings, where patients exercise on individual ergome-
ters and are continuously monitored by telemetry. Patients begin
outpatient (phase II) CR much sooner after the acute event; reim-
bursement for programs is limited to 12 weeks, although many
patients complete their programs much sooner and resume their
work, travel, and other usual activities. Recommendations are
also routinely provided for exercise to be done at home or at a
community facility on non-class days, and weight training along
with nontraditional activities such as tai chi, yoga, and balance
training are frequently included in outpatient CR.
CR was initially promoted by the American College of Sports
Medicine, and certification programs for medical professionals
working in CR were developed by that organization. In 1985, a sep-
arate organization, the American Association of Cardiovascular
and Pulmonary Rehabilitation (AACVPR), devoted specifically
to cardiac (and pulmonary) rehabilitation was organized.
Structure of CR
Phase I
Phase I, inpatient CR, lasts 3 to 7 days. Initially, the phase I
structured activity program was conducted by nurses or physical
therapists (for example, Nanette Wenger’s 14-Step program), but
shortened hospital stays have changed phase I programs into a
more purely educational focus and to refer patients into phase II
programs.
Phase II
Phase II, outpatient CR, consists of a maximum of 36 monitored
exercise sessions to be completed within 36 weeks of referral,
though, most commonly, completed within 3 months. Patients
generally are referred to and begin CR within a week or 2 of
their qualifying event, though current rules allow up to 1 year
after the qualifying event to start CR. CR is conducted in a hos-
pital setting. Patients exercise 1 to 3 times per week, generally
on ergometers, but they walk or run sometimes around tracks in
hospitals that feature a comprehensive wellness center. Interval
training is now included as a standard component of CR, as stud-
ies have demonstrated more rapid improvement with interval
versus continuous training without increased risk.
Patients are monitored by telemetry. Commercially available,
dedicated databases can interface with telemetric systems and
facilitate electronic record-keeping. Exercise includes not only
cardiovascular conditioning, such as treadmill walking, but it
should also include light resistance exercise (weight training) and
other activities to improve flexibility and balance. Educational
programs are generally offered, and various forms of counsel-
ing—dietary, psychologic, and occupational—may be provided
as part of the program. To comply with Medicare guidelines,
physician availability during exercise classes is mandated, and
there must be a physician-directed, patient-specific treatment
plan that prioritizes goals and outlines intervention strategies for
exercise training and cardiovascular disease risk reduction, and
a follow-up plan that reflects progress toward goals and guides
long-term secondary prevention strategies.
Exercise testing—standard exercise electrocardiography, car-
diopulmonary exercise testing, and 6-minute walks—is used to
evaluate progress and revise the exercise prescription. Phase II
CR programs are also required to track specific patient outcomes,
as discussed below in the “Current Practices” section.
697
Phase III
Phase III, extended CR, may be requested for patients who prog-
ress slowly for various reasons or who had severely impaired
functional status initially may require an extended course of
CR before returning to usual activity in the community. These
patients may continue in hospital-based programs under close
supervision and perhaps further telemetric monitoring.
Phase IV
Phase IV, lifetime CR, lasts for months or years. Many hospitals
offer a long-term program for patients to continue to exercise
(but generally not be continuously monitored) in a medically
supervised setting on the hospital campus after they have com-
pleted phase II (and possibly phase III) CR. The hospital may
contract with a community agency or private health club to pro-
vide the facilities, with hospital personnel conducting the actual
classes. These programs are generally paid for by the patient,
although some employers subsidize participation. Patients not
qualifying for reimbursement under current Medicare guidelines
(such as peripheral vascular disease patients) may participate in
CR through these phase IV programs. Some programs may also
accept high-risk patients who do not have a diagnosis of cardio-
vascular disease (patients with diabetes are the prime example).
Current Practices
Consistent with the shift in the epidemiology of CAD and in
the vastly improved care of the patient with acute coronary syn-
drome, CR today serves patients across a wide range of ages,
degrees of cardiovascular impairment and risk, and numbers
and severity of comorbidities. Young male smokers with recent
infarcts still make up a large portion of CR patients, but their
left ventricular function and exercise tolerance are generally well
preserved and the patients are usually free of ongoing ischemia
and are able to return to work within weeks (if not days) after the
acute event. Thus, the CR course is considerably accelerated, and
ensuring smoking cessation may be the most important task for
the CR team. At the other end of the spectrum, elderly patients
with recurrent CAD—kept alive by medical and technologic
progress—present to CR with severe impairments in functional
capacity due to extensive cardiac disease and comorbidities asso-
ciated with aging and obesity. Their progress is obviously much
slower, and the burden of care for the CR team is much higher.
Although the ability to actually modify the disease process or
prolong survival in these patients is limited, there is a great
potential to reduce the economic burden of illness by prevent-
ing rehospitalization for exacerbation of symptoms that could be
adequately managed on an outpatient basis.
Participation rates for CR are lower for women than for men,
but increasing numbers of women now participate in CR; many
of them are older and present with multiple comorbidities and
poor functional capacity. Often an added challenge is the patient’s
lack of previous physical activity and attitudes and preferences
unfavorable to exercising.
As mentioned above, CR programs are organized around
the provision of telemetrically monitored cardiovascular condi-
tioning. This is generally provided in a small class with 4 to 8
patients exercising at a time. Generally, a session lasts 1 hour and
is performed one to three times per week, depending on patient
availability and the need for close supervision and monitoring.
In addition to the cardiovascular conditioning classes, resistance,
flexibility, and balance training may be offered. Education and
counseling of various forms are also generally provided.
698 VII Myocardial Infarction
Screening for conditions that may interfere with progress
and CR or decrease long-term prognosis is recommended. The
PHQ-9 is a free, convenient, validated tool used to screen for
depression in CR:
http://www.depression-primarycare.org/clinicians/toolkits/materials/
forms/phq9/
Similarly, the Berlin Questionnaire can be used as an initial
tool to identify sleep disordered breathing:
http://www.swclab.com/images/PDFS/Berlin-Questionnaire.pdf
In 1995, the AHCPR published Clinical Practice Guidelines
(No. 17) for CR (Web site address provided in “Suggested
Reading” Section). These guidelines recommended that CR pro-
grams facilitate and track eight specific outcomes (Table 75.2).
To carry out these recommendations, a case management
system has been instituted in many CR programs. Each mem-
ber of the CR team is assigned a specific group of patients and
is then responsible for evaluating these patients on the basis of
these recommendations, making appropriate referrals for inter-
vention (such as dietary counseling, smoking cessation counsel-
ing, and psychologic counseling) and tracking progress. We have
published data on our case management system at Mayo Clinic
showing improved 3-year medication use compared to usual
care: statins, 91% versus 44%; beta blockers 78% versus 48%,
and angiotensin converting inhibitors 78% versus 43%. Rates of
risk factor control at 3-years was high: not smoking 95%, systolic
and diastolic blood pressure control 73%, low density lipoprotein
cholesterol < 100 mg/dL 74%, and 30 minutes or more of daily
physical activity 57%. There have been many other studies show-
ing equivalence or superiority to case management of chronic
disease compared to usual care.
Benefits of CR
The 1995 AHCPR Support Clinical Practice Guidelines consider
that evidence exists for the following outcomes in CR partici-
pants compared with nonparticipants:
1. Improvement in exercise tolerance
2. Improvement in symptoms
3. Improvement in blood lipid levels
4. Improvement in psychosocial well-being and reduction of stress
5. Reduction in cigarette smoking
6. Reduction in mortality
In an examination of mortality benefit in 2003, the Cochrane
Library performed a meta-analysis of secondary prevention ben-
efits of CR programs. Analysis of outcomes from 8,440 CAD
patients in exercise-based rehabilitation programs showed the
following:
1. 27% reduction in all-cause mortality
2. 31% reduction in CAD mortality
3. No evidence of reduction in nonfatal CAD
CR programs offering only exercise training seemed to show
a stronger benefit than comprehensive programs, but that finding
was most likely an artifact of the time frame of the study because
earlier programs were less comprehensive, probably exercised
patients more vigorously, and enrolled patients at higher risk.
For example, an early CR program in Finland showed signifi-
cant mortality benefit in a relatively small number of participants
owing to the exceptionally high risk of cardiac death (10-year
risk was nearly 50% in control patients; see Figure 75.2).
Mayo Clinic observational data were used to compare survival
of participants in CR after MI with survival of nonparticipants
Table 75.2. Recommendations for Cardiac Rehabilitation
1. Smoking
A combined approach of education, counseling, and behavioral
interventions in cardiac rehabilitation results in smoking cessation and
relapse prevention and is recommended for cardiac risk reduction.
2. Lipids
Intensive nutritional education, counseling, and behavioral interventions
improve dietary fat and cholesterol intake. Education and counseling
about nutrition along with behavioral interventions—with or
without pharmacologic lipid-lowering therapy—result in significant
improvement in blood lipid levels and are recommended as components
of cardiac rehabilitation.
3. Body weight
Multifactorial rehabilitation that combines dietary education,
counseling, and behavioral interventions designed to reduce body
weight can help patients lose weight. With education as a sole
intervention, patients are unlikely to achieve and maintain weight loss.
These multifactorial cardiovascular risk-reduction interventions are
recommended as components of comprehensive cardiac rehabilitation.
4. Blood pressure
Expert opinion supports education as an important component of a
multifactorial approach to the management of hypertension, including
education, counseling, behavioral intervention, and pharmacology. This
approach is documented to be effective in nonrehabilitation populations
and should also be included in cardiac rehabilitation. Education,
counseling, and behavioral interventions alone have not been shown to
control elevated blood pressure levels.
5. Exercise tolerance
Cardiac rehabilitation education, counseling, and behavioral
interventions without exercise training are unlikely to improve exercise
tolerance and are not recommended for that purpose.
6. Symptoms
Cardiac rehabilitation education, counseling, and behavioral
interventions are recommended alone or as components of multifactorial
cardiac rehabilitation to reduce symptoms of angina.
7. Return to work
Education, counseling, and behavioral interventions have not been
shown to improve rates of return to work, which are contingent on
many social and policy issues. In selected patients, formal cardiac
rehabilitation vocational counseling may improve return-to-work rates.
8. Stress and psychologic well-being
Education, counseling, and psychosocial interventions—either alone
or as components of multifactorial cardiac rehabilitation—result in
improved psychologic well-being. Education, counseling, and behavioral
interventions are recommended to complement the psychosocial benefits
of exercise training.
Previously published. See “Credit Lines” section.
in 1982 to 1998 (Figure 75.3). Among eligible men, participation
rates were 71% (vs. 40% for women). Overall, participants had
26% lower mortality than nonparticipants. When these figures
were adjusted with a propensity score for participation in CR, the
benefit increased to 43%.
A subsequent study at Mayo Clinic evaluated the benefits of
CR after PCI using similar methods. A total of 2,395 consecu-
tive patients who underwent percutaneous coronary intervention
in Olmsted County, Minnesota, from 1994 to 2008 were studied
with respect to outcomes in patients who did or did not partic-
ipate in CR. Participation in CR, noted in 40% (964 of 2395)
of the cohort, was associated with a significant decrease in all-
cause mortality by 3 different statistical techniques (hazard ratio,
0.53 to 0.55; P<0.001). A trend toward decreased cardiac mortal-
ity was also observed in CR participants; however, no effect was
observed for subsequent MI or revascularization. The associa-
tion between CR participation and reduced mortality rates was
 75 Cardiac Rehabilitation 699

50

Cumulative cardiac mortality, %

40
Control

30
Intervention

20

10 Mantel-Cox test
P=0.03

0
0 2 4 6 8 10
Years
Figure 75.2. Finnish cardiac rehabilitation study that demonstrated a significant mortality benefit with cardiac rehabilitation among high-risk
patients.

similar for men and women, for older and younger patients, and
for patients undergoing elective or nonelective PCI.
Future of CR
CR has a long history of adjustment to improve CAD manage-
ment strategies, to change CAD epidemiology, and to meet new
regulations and guidelines. Challenges for cardiac rehabilitation
for the present and near future include the following:
1. Increasing patient referrals and participation rates. In contrast to
the Mayo Clinic data cited above, national referral rates have been
estimated at only 13% to 41% for men and 7% to 22% for women.
Many barriers to participation still exist apart from the availability
of a program and insurance coverage. Greater distance from a CR
center, more non-cardiac comorbidities, female sex, older age, liv-
ing alone, obesity, and smoking appear to be significant deterrents
to participation. Continuing medical education may be important,
along with public education and continued research into CR benefits.
Appropriate physician behavior can be encouraged with financial
incentives. Referral to CR is being tracked as a performance meas-
ure in the outpatient Physician Quality Reporting System and the
American College of Cardiology PINNACLE database. Cardiologists
who utilize CR for their patients receive increased reimbursement
for their services.

100

Participation
80
Survival, %

60 No participation

40

20

0
0 1 2 3 4 5
Years after myocardial infarction
Figure 75.3. Observational study of survival after myocardial infarction among participants and nonparticipants in a cardiac rehabiliation pro-
gram. (Previously published. See “Credit Lines” section.)
700 VII Myocardial Infarction
2. Expanding the range of eligible patients. Adding heart failure,
peripheral vascular disease, and stroke as acceptable diagnoses
would further expand referral bases. Not only would more patients
benefit from CR, but programs could more readily survive financially
with higher patient volumes and not be forced into ever-increasing
per-session charges to meet expenses.
3. Professional training. As CR programs are required to provide
more extensive services and administer them to a broader range of
patients, public agencies and professional organizations will need
to upgrade continuing education and certification or licensing pro-
grams to ensure that CR professionals possess the knowledge and
skills necessary to provide the level of service required.
Abbreviations
AHPCR Agency for Health Care Policy and Research
CAD coronary artery disease
CR cardiac rehabilitation
MI myocardial infarction
PCI percutaneous coronary intervention
PINNACLE Practice, Innovation, and Clinical Excellence
Suggested Reading
AHA/AACVPR medical director responsibilities for outpatient cardiac
rehabilitation/secondary prevention programs. Available from: http://
circ.ahajournals.org/cgi/content/full/112/21/3354
AHA scientific statement: exercise and physical activity in the preven-
tion and treatment of atherosclerotic cardiovascular disease. Available
from: http://circ.ahajournals.org/cgi/content/full/107/24/3109.
AHCPR supported clinical practice guidelines: cardiac rehabilita-
tion. Clinical Guideline No. 17. (AHCPR Publication No. 96–0672,
October 1995.) Available from: http://www.ncbi.nlm.nih.gov/books/
bv.fcgi?rid=hstat2.chapter.6677.
American Association of Cardiovascular and Pulmonary Rehabilitation.
Available from: http://www.aacvpr.org/.
American College of Sports Medicine. ACSM’s guidelines for exercise
testing and prescription. 6th ed. Philadelphia: Lippincott Williams &
Wilkins; c2000.

Coronary Artery Bypass Surgery
THORALF M. SUNDT III, MD
Background
Surgical coronary revascularization is the gold standard therapy
for cardiac ischemia. Even with scientific and procedural devel-
opments that narrow the gap between surgical and percutaneous
revascularization strategies, surgical revascularization contin-
ues to demonstrate superior durability and greater long-term
cost-effectiveness at the expense of greater invasiveness for the
patient. Accordingly, cardiologists should understand the surgi-
cal perspective on coronary revascularization, particularly the
surgeon’s view of preoperative assessment and certain aspects of
intraoperative decision making and postoperative care.
Preoperative Assessment
Preoperative assessment of the candidate for coronary revascu-
larization includes a determination of the risk-benefit ratio for
the patient. Several tools are available for the assessment of pre-
operative risk, the most sophisticated of which is provided by
the Society of Thoracic Surgeons National Database. This data-
base, which includes data on hundreds of thousands of cardiac
surgical cases, from coronary artery disease to valvular disease,
provides a Web-based risk-stratification tool that is available
to anyone (http://www.sts.org). At present, the risk algorithm
itself is proprietary. Nonetheless, the essential elements that help
determine risk are well recognized and have been demonstrated
in numerous publications (Table 76.1).
Unfortunately, there is no statistical tool that substitutes for
the “foot of the bed” test. It is common surgical lore that a suc-
cessful coronary artery bypass operation demands 3 elements:
adequate left ventricular function, a suitable conduit, and
Abbreviations and acronyms are expanded at the end of this chapter.
76
graftable targets. Optimal results are obtained when all 3 are
present. Good results can generally be obtained when 2 of the 3
are acceptable. Extreme caution should be exercised if only 1 of
these 3 critical elements is of reasonable quality. Other elements,
including general physiologic condition and good function of
other organ systems—most importantly, renal function—enter
into the assessment as well. The perioperative risk of death is
dramatically increased among patients who have perioperative
renal failure. Indeed, patients with marginal renal function who
progress to renal failure perioperatively fare worse than those
who are dependent on dialysis preoperatively. Furthermore,
the risk of significant morbidity must be considered, since lit-
tle is accomplished by a technically “successful” operation in a
debilitated patient.
Preoperative Management
Perioperative risk can be modified to some degree by preop-
erative preparation. Patients fare better if they are not in acute
congestive failure at the time of their procedure. The risk of cor-
onary bypass is also elevated in the first 2 weeks after an acute
myocardial infarction. While one may argue about the time point
at which statistical significance is lost, in nearly every study
performed the trend is the same: the closer to the episode of
infarction, the higher the perioperative risk. Often a patient at
extraordinarily high risk who has had a large infarct resulting in
severely depressed ventricular function will have an uneventful
operation if it can be delayed 2 to 6 weeks while ventricular func-
tion recovers. Right ventricular performance should be evaluated
in patients who have had a recent inferior wall myocardial infarc-
tion. A notorious trap is proximal right coronary artery occlu-
sion and an associated right ventricular infarction unrecognized
with left ventriculography. Echocardiographers, if not focused on
701
702 VII Myocardial Infarction

Table 76.1. Relative Mortality Risk and Core Coronary Artery Bypass Graft Variables for
Operative Mortality
Variable NNE VA STS NYS CC AGH

Patients, No. 3,055 12,712 332,064 57,187 4,918 1,567

Year of publication 1992 1992 1997 1994 1997 1996
Years included 1987–1989 1987–1990 1990–1994 1989–1992 1993–1995 1991–1992
Relative Mortality Risk
Database variables
Age 1.04 1.04 1.1 1.0 1.05 ...
Female 1.2 ... 1.5 1.5 1.63 1.48
Prior heart surgery 3.6 3.2 3.0 3.7 1.72 1.39
Left main coronary artery ... ... 1.3 1.43a ... ...
disease (>70% stenosis)
Diseased coronary vessels, No.
1 1 ... 1.0 ... ... ...
2 1.3 ... 1.0 ... ... ...
3 1.6 ... 1.2 ... ... ...
Urgency of operation
Elective 1 1.0 1.0 1.0 1 1
Urgent 2.1 2.4 1.2 1.42 ... 3.5
Emergent 4.4 3.8 2.0 4.0 5.07 7.14
Salvage ... ... 6.7 ... ... 29.9
Ejection fraction, %
50–59 ... ... ... ... ... ...
40–49 ... ... ... 1.0b ... ...
30–39 ... ... ... 1.6 ... ...
20–29 ... ... ... 2.2 ... 2.89c
<20 ... ... ... 4.1 ... ...
Abbreviations: AGH, Allegheny General Hospital; CC, Cleveland Clinic; NNE, Northern New England Cardiovascular
Disease Study Group; NYS, New York State Cardiac Surgery Reporting System; STS, Society of Thoracic Surgeons
National Database; VA, US Department of Veterans Affairs cardiac surgical database.
a
Stenosis >90%.
b
Ejection fraction >40%.
c
Ejection fraction <30%.
Previously published. See “Credit Lines” section.

right ventricular function, may miss this finding. If a patient is
brought to the operating room with an unrecognized right ven-
tricular infarction, the patient will not be weaned from cardio-
pulmonary bypass.
Pharmacologic measures that may be used preoperatively to
reduce perioperative risk include the administration of β-blockers
and statins. Data supporting the use of β-blockers are quite solid
even if only 1 dose can be administered preoperatively. The
effect of preoperative administration of statins is less certain.
Mounting evidence supports lower risk among patients receiving
statins, possibly related to improved endothelial cell function.
Although experimental data suggest that several weeks of ther-
apy are required to achieve this effect, from a practical stand-
point it seems reasonable to initiate statin therapy as soon as a
decision has been made to proceed with surgical intervention.
Intraoperative Management
and Decision Making
On-Pump or Off-Pump Surgery
Once a decision has been made to proceed with surgical interven-
tion, one must choose between on-pump surgery and off-pump
surgery as well as whether to use all-arterial conduits. Some
of the earliest coronary artery bypass procedures performed
in the 1950s and 1960s were performed without cardiopulmo-
nary bypass. Of late, there has been a resurgence of interest in
off-pump approaches in hopes of reducing neurocognitive defi-
cits after coronary bypass. The term pump head was coined
to describe alterations in mental status thought to be related to
cardiopulmonary bypass, and some data have supported the
notion that these neurocognitive deficits may persist. It is logical,
therefore, that coronary artery bypass performed without cardio-
pulmonary bypass may obviate these concerns.
Unfortunately, the data supporting an improvement in neu-
rocognitive outcomes with the use of off-pump techniques have
been weak at best. Several nonrandomized studies suggested
significant advantages to off-pump surgery. Those studies were
followed by several prospectively randomized studies that dem-
onstrated trends toward less bleeding and fewer transfusions
as well as improved renal function with off-pump surgery, but
there was no convincing difference in neurocognitive outcome.
This paradox can be understood by reviewing more recent pub-
lications that reported neurocognitive outcomes among patients
after surgery of all types and among patients who had coronary
artery disease and underwent angioplasty or medical manage-
ment compared with patients who underwent coronary artery
bypass graft surgery. In these studies, apparent neurocogni-
tive deficits were readily detectable after all manner of surgical
procedures in young and old as well as after general anesthe-
sia or spinal anesthesia. This suggests that some neurocognitive
alterations may be related simply to the stress of surgery. One
study tracked perceived neurocognitive deficits to indexes of
 76 Coronary Artery Bypass Surgery 703

psychologic depression. Additionally, most studies showed that distal anastomosis between the coronary artery and the ITA.
neurocognitive function returned to baseline within 3 months Devascularization of the sternum due to loss of the ITA blood
after surgery. In short, even though there is undeniable anecdotal supply to the sternum may increase the risks of wound complica-
experience with the occasional patient who has significant mem- tions, particularly among diabetic patients, and as the population
ory loss or personality changes (or both) after cardiopulmonary of diabetic patients undergoing coronary bypass increases, the
bypass, most patients appear to have no adverse effects from use concerns for sternal wound infection will increase. Some older
of the pump. data suggest that the use of bilateral ITAs significantly increases
Beyond neurocognitive issues, an argument can still be made the risk of sternal wound infection among persons with diabe-
for off-pump surgery. An axiom of surgery is that the simpler one tes; however, the more recent use of skeletonized ITAs results
can make a procedure the better, and if the same operation can in less devascularization of the sternum and appears to obviate
be performed off-pump instead of on-pump, why not do so? The this concern.
principal debate has centered around late graft patency and com- A third common conduit is the radial artery. The radial artery
pleteness of revascularization. Although several studies reported was used in the early days of coronary bypass and abandoned
by leaders in the field of off-pump surgery have reported similar because of apparent early failures. In the past decade, with the
indexes of revascularization regardless of the method used, typ- recognition that some patients who had “early string sign” (atre-
ically fewer graft operations are performed in off-pump groups sia) of the radial artery had patent radial artery grafts on sub-
than in on-pump groups. Similarly, some experts have reported sequent angiography many years later, it has been suggested
excellent graft patency; however, other studies have shown alarm- that the radial artery is a better conduit than originally thought.
ing rates of graft occlusion. This is hotly debated but remains Accordingly, there has been a resurgence in interest in the radial
unresolved. At this time, it is fair to say that off-pump surgery is artery. It has the advantages of being harvested simultaneously
a reasonable option. There is no clear mandate, however, to per- with the ITA and resulting in less sternal devascularization. In
form procedures without bypass and there is no hard evidence to addition, the radial artery is somewhat longer than the ITA; this
support the superiority of this approach. It is still reasonable for may be advantageous, particularly when performing complex
surgeons and cardiologists to elect either on-pump or off-pump and complete arterial revascularizations. In multiple studies,
coronary bypass. including a prospectively randomized trial, the patency of the
radial artery was superior to that of the saphenous vein but less
than that of the ITA. These patency data, however, need to be
Conduit Choice viewed in light of the effect of both target vessel stenosis and
A second and likely more critical decision in the long run is the target vessel identity. As shown in Figure 76.2, the radial artery
choice of conduits to be used in the revascularization strategy. The performs particularly poorly in a stenosis that is less than critical,
principal clinical outcome difference between coronary bypass and involution has been observed in a nicely patent radial artery
and percutaneous coronary intervention is durability. Although
percutaneous coronary intervention is by far less invasive, and
the target vessel revascularization failure gap has been narrowed
with the use of drug-eluting stents, coronary artery bypass con-
tinues to provide a more durable result. It is also the procedure
that has been more solidly associated with improvement in sur-
vival. The durability of the procedure, however, depends largely
on graft patency.
Graft patency depends not only on a technically perfect anas-
tomosis but also on the characteristics of the conduit used. A
saphenous vein is the traditional conduit for coronary revascu-
larization. It is readily harvested and is relatively easy to work
with from a technical standpoint. Unfortunately, solid data dem-
onstrate a significant occlusion rate. One can expect at least 10%
of saphenous vein grafts to be occluded at 1 year and 50% to
be occluded at 10 years. It has been hypothesized, although not
proved, that vein graft attrition largely results from subjecting a
venous structure to arterial hemodynamic forces. Still, in many
situations it has proved to be a satisfactory conduit.
The superiority of the ITA is, however, unquestionable. The
ITA has become the gold standard for durability (Figure 76.1). In
some of the earliest coronary bypass procedures, use of the ITA
was held dear by a small group of enthusiasts until it became
apparent in the 1980s that use of an ITA graft to the left anterior
descending artery actually improved patient survival. In large
measure, this is most likely secondary to the 10-year patency rate
of 90% to 95%. The reason for this patency may relate in part to
the unfenestrated internal elastic lamina of the ITA, which inhib-
its migration of smooth muscle cells to the subintimal space; the Figure 76.1. Use of Internal Thoracic (Mammary) Artery Grafts for
minimal degree of medial muscularity; or its increased basal Coronary Revascularization. LCA indicates left coronary artery; LITA, left
excretion of nitric oxide. The disadvantages of the ITA are its internal thoracic artery; OM 1, obtuse marginal branch of circumflex artery;
fragility and the increased level of difficulty in performing a PDA, posterior descending artery; RITA, right internal thoracic artery.
704 VII Myocardial Infarction
100
Cumulative Patency, %
75
50
25
0
0 12
Months
Figure 76.2. Radial Artery Graft Patency in Coronary Artery Revascularization. In Cox multivariate analysis of risk factors, compared with a refer-
ence stenosis of 90% or more, a stenosis of 71% to 89% had a relative risk of 1.80 (P = .12) and a stenosis of 70% or less had a relative risk of 1.69 (P<.001).
Cx, circumflex artery; LAD, left anterior descending coronary artery; RCA, right coronary artery. (Previously published. See “Credit Lines” section.)
graft after angioplasty of stenoses proximal to the target lesion. It
is now widely accepted that the radial artery should be used only
for targets with a stenosis greater than 80%. In addition, target
identity is an important factor determining patency regardless of
the conduit used. All conduits have the worst patency to the right
coronary artery and its branches and intermediate patency to the
circumflex artery.
Mitral Regurgitation
MR is frequently observed soon after myocardial infarction, and
delay of surgery until some element of ventricular recovery has
occurred may permit resolution of functional MR. Acute MR
due to papillary muscle rupture, however, is a surgical emer-
gency that requires prompt intervention. In some instances, the
mitral valve may be repaired by suturing the ruptured papillary
muscle to the left ventricular wall or adjacent papillary muscle.
More often the valve is replaced. Either treatment is acceptable.
Many advocate that functional MR due to annular dilatation may
be repaired with simple “undersized” annuloplasty that uses a
complete or partial ring with a rigid or flexible design. There is
remarkably little consensus on the optimal approach. Regardless
of approach, the rate of recurrent MR is higher than that after
repair of mitral valve with degenerative disease. Furthermore,
the effect of repair on late outcome is unclear.
Postoperative Management
Hemodynamic Factors
Postoperative care of the coronary bypass patient is like that of
any other patient undergoing a cardiac surgical procedure. Of
note, patients undergoing off-pump surgery receive much less
fluid intraoperatively and require less aggressive diuresis post-
operatively. In coronary artery bypass patients, the cardiac index
should be maintained at more than 2.2 L/min per square meter.
A cardiac index less than this standard should be managed
LAD
Cx
RCA
24 36 48
according to a thoughtful algorithm, beginning with assessment
of the rate and rhythm. Cardiac surgical patients are particularly
sensitive to a loss of sinus rhythm, possibly from perioperative
diastolic dysfunction after the ischemic insult and edema in the
myocardium. After ensuring an adequate rate and a sinus rhythm,
one should consider the filling pressures. A right atrial pressure
of 8 to 12 mm Hg and a pulmonary artery diastolic pressure of
16 to 20 mm Hg should be adequate in most instances. A patient
who has adequate hemodynamics but lower filling pressures,
however, does not need fluid administration just to “fix the num-
bers.” Unnecessary administration of fluids will cause hemodilu-
tion and increase the need for transfusion. If filling pressures are
adequate or elevated owing to left ventricular failure or right ven-
tricular failure, inotropic support should be initiated. The first-
line drugs are often dopamine and dobutamine, with epinephrine
an increasingly popular choice among cardiovascular anesthe-
siologists. Milrinone or other phosphodiesterase inhibitors can
also be quite helpful.
An important concept in the postoperative care of patients
is their passage through stages of recovery, including an early
requirement for additional fluids as the heart recovers, shifting
later to a requirement for active diuresis when myocardial function
has been restored. This is the rationale behind the apparent para-
dox that low urine output postoperatively is treated with boluses
of fluid, which are countered with furosemide the next day.
Pulmonary Function
Patients may be extubated soon after coronary artery bypass
surgery. In some centers, patients are extubated in the operat-
ing room if their core body temperature has returned to normal.
During the cardiopulmonary bypass process, some surgeons
actively cool the patient’s body, and almost all patients have
some core body cooling in the operating room owing to ambi-
ent heat loss. Intubation and mechanical ventilation help to
offset the increased oxygen consumption caused by shivering
 76 Coronary Artery Bypass Surgery
during rewarming. The philosophy of waiting for extubation
until the following morning, however, which was popular a
decade ago, has given way to fast-track protocols that empha-
size early extubation. This is accomplished principally with
the use of appropriate short-acting anesthetic agents. Most cor-
onary artery bypass patients can be extubated within 6 hours
postoperatively.
Postoperative pulmonary dysfunction can result from cardio-
pulmonary bypass, although this is uncommon. Other causes
of respiratory insufficiency include pleural effusions, which are
relatively common and are usually manageable with a simple
pleural tap. Occasionally, postoperative pleural effusions recur,
requiring several taps. Pleurodesis or insertion of a pleural chest
tube is seldom required. One reason to minimize blood trans-
fusions is that patients may experience transfusion-related acute
lung injury, which may progress to acute respiratory distress
syndrome.
Another potential cause of respiratory dysfunction postopera-
tively is phrenic nerve injury. The phrenic nerve enters the tho-
racic space adjacent to the ITA, and during harvest of the ITA the
nerve could be damaged. It is uncommon to transect it, but use
of electrocautery near the nerve can cause a nerve palsy, which is
usually transient. The use of intrapericardial ice is becoming less
common; however, in centers where it is still used, phrenic nerve
injury can occur simply from the cold.
Sternal Wound
Sternal wound complications are uncommon after coronary
artery bypass. In most institutions, the rate of deep sternal wound
infection is 1% or less. The earliest sign of deep sternal wound
infection is often excessive sternal pain. The sternotomy itself
should not be particularly painful after the first day. Increasing
sternal pain, particularly in the first days postoperatively, should
not be discounted as postpericardiotomy syndrome without a
rigorous evaluation for sternal wound infection, including physi-
cal examination to assess sternal stability as well as a computed
tomographic scan to determine the integrity of the sternal bone
itself. Occasionally substernal air will be apparent, although this
is most often simply from the recent presence of a chest tube. It is
common to see soft tissue thickening behind the sternum, so the
computed tomographic scan may not be helpful in that regard.
Drainage from the inferiormost portion of the wound is common
but can be a harbinger of deep sternal wound problems. Drainage
from the wound should be considered a serious problem. There
are no clear guidelines for antibiotic use, but most surgeons
administer antibiotics liberally because of the gravity of sternal
wound infections.
The sternum itself should heal in 6 to 12 weeks. During that
time, the patient should not engage in activities that stress the
sternum, including swinging a golf club or bowling. If both
ITAs have been harvested, sternal wound healing is slowed
accordingly.
Postoperative Bleeding
Postoperatively, some output from the chest tube is expected.
The usual criteria for return to the operating room are 400 mL of
chest tube output in 1 hour, 300 mL per hour for 2 consecutive
705
hours, or 200 mL per hour for 3 hours. These criteria are not
rigid and, of course, the decision will be affected by specific sur-
gical factors.
Before being returned to the operating room, the patient
should have a complete evaluation of coagulation studies,
including platelet count, prothrombin time, and partial throm-
boplastin time. Mild elevation of the prothrombin time (14–16
seconds) is expected postoperatively; however, if the prothrom-
bin time is markedly prolonged, administration of fresh fro-
zen plasma should be considered. The platelet count should be
higher than 100 × 109 /L. A prolongation of the partial throm-
boplastin time in the early postoperative period may be from
the incomplete reversal of heparin. The additional admin-
istration of 25 to 50 mg of protamine may be helpful in this
circumstance. Patients with preexisting renal dysfunction are
particularly prone to bleeding, and cryoprecipitate may be
beneficial for the correction of acquired platelet dysfunction.
Administration of clotting factors, however, is not a substitute
for a prompt return to the operating room if surgical bleeding
is the culprit.
Dark blood from the chest tubes is most often venous and
can almost always be controlled nonoperatively with increased
positive end-expiratory pressure unless the volume of bleeding is
excessive. Bright red blood is more problematic and may indicate
that surgical reexploration will be required.
Patients who have experienced excessive chest tube output are
at risk of postoperative pericardial tamponade. The cardiologist
should recognize that postoperative tamponade can be due to a
localized clot, in contrast to the tamponade due to pericardial
effusion more often seen with echocardiography. A localized
clot compressing the right atrium but not encasing the ventricles
can cause hemodynamic instability, as can a localized throm-
bus behind the left atrium. Echocardiographic criteria for early
postoperative tamponade are, therefore, different from those for
chronic pericardial effusion. Unexplained hemodynamic deteri-
oration should prompt echocardiographic evaluation.
Weaning Patients From Inotropic Drips
The rate at which patients are weaned from inotropes is as much
a matter of judgment as science. A reasonable approach is to
use a rate dictated by the half-life of the drug itself. Therefore,
phosphodiesterase inhibitors may require a slower weaning than
β-adrenergic agents. Regardless, weaning should be guided by
maintaining the cardiac index at more than 2.2 L/min per square
meter in the presence of acceptable filling pressures.
Abbreviations
ITA internal thoracic artery
MR mitral regurgitation
Names of Clinical Trials
BHACAS 1 Beating Heart Against Cardioplegic Arrest Study 1
BHACAS 2 Beating Heart Against Cardioplegic Arrest Study 2
PRAGUE Primary Angioplasty in Patients Transferred From
General Community Hospitals to Specialized PTCA
Units With or Without Emergency Thrombolysis
SMART Surgical Management of Arterial Revascularization
Therapies
This page intentionally left blank
 Section VIII

Diseases of the Heart, Pericardium,

and Pulmonary Circulation
This page intentionally left blank

Pericardial Diseases
ROWLENS M. MELDUNI, MD
Anatomy and Function of the Pericardium
The pericardium is a fibromembranous inelastic sac less than
2 mm in thickness that comprises 2 distinct layers. The outer
inelastic fibrous layer (fibrous pericardium) anchors the heart
in the mediastinum with attachments anteriorly to the manu-
brium and xiphoid process, posteriorly to the vertebral column,
and inferiorly to the diaphragm. The inner serous double layer
(serous pericardium) is divided into the visceral pericardium
(epicardium) and the parietal pericardium. The visceral peri-
cardium wraps around the heart and proximal great vessels and
is reflected back so that it lines the inner surface of the fibrous
pericardium, which together form the parietal pericardium
(Figure 77.1). The serous pericardium secretes a small amount
of clear serous fluid (normally about 25–50 mL) in the pericar-
dial cavity, referred to as the pericardial reserve volume, which
lubricates the surface of the heart and reduces friction within
the pericardial space as the heart moves and twists within the
pericardial cavity.
Superiorly, a few centimeters of the proximal portions of the
great vessels are surrounded by the parietal pericardium; the
outer fibrous layer fuses with the external adventitial layer of
the great vessels so that the heart is “suspended” within the peri-
cardial space. This anatomical arrangement explains the develop-
ment of hemopericardium related to proximal aortic dissection.
The pericardium serves as a protective nondistensible mecha-
nical barrier for the heart by containing and protecting it from
acute overdistention. The pericardium also affects the atria and
ventricles hemodynamically. Normally, intrapericardial pressure
is equal to intrapleural pressure and is transmitted uniformly
throughout the fluid-filled intrapericardial space. Ventricular
Abbreviations and acronyms are expanded at the end of this chapter.
77
volume is greater at any given ventricular filling pressure with
the pericardium removed than with the pericardium intact. The
pericardium also contributes to diastolic coupling between the
2 ventricles: the distention of 1 ventricle alters the filling of
the other, an effect that is important in the pathophysiology of
cardiac tamponade and constrictive pericarditis. Ventricular
interdependence becomes more marked at high ventricular fil-
ling pressures.
• The normal thickness of the pericardium is <2 mm.
• The pericardium protects and lubricates the moving surface of the
heart.
• The pericardium contributes to diastolic coupling of the right and
left ventricles, an effect that is important in cardiac tamponade and
constrictive pericarditis.
• Proximal portions of the great vessels reside in the pericardial sac.
Congenital Absence of the Pericardium
Complete absence of the pericardium is very rare and is usually
asymptomatic. Most common is the absence of a small portion
of the pericardium, usually on the left side. Rarely with extreme
cardiac shift to the left, the patient may experience left-sided
nonexertional stabbing chest pain, trepopnea (the presence of
dyspnea when the patient is lying on 1 side but not the other), or
prominent, displaced cardiac pulsation.
This condition usually is diagnosed incidentally on chest
radiography and displays a marked levoposition (left-sided shift)
of the heart without tracheal deviation. Lung tissue is present
between the aorta and the main pulmonary artery and between
the inferior border of the heart and the left hemidiaphragm. The
left ventricular contour is flattened (left upper border) and elon-
gated, resembling the dog “Snoopy” (Figure 77.2A and 77.2B).
709
710 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Figure 77.1. Cross Section of the Myocardium and Pericardium. The distinct layers are shown. Note the pericardial reflection, which lines the
point of folding where the surfaces of visceral and parietal pericardium meet each other. Ao indicates aorta; IVC, inferior vena cava; PA, pulmonary
artery; PV, pulmonary vein; SVC, superior vena cava.

Lung tissue appears as a lucent area between the aorta and the Pericardial Cyst
main pulmonary artery and between the inferior border of the
A pericardial cyst is a rare benign structural abnormality of the
heart and the left hemidiaphragm. The traditional echocardio-
pericardium that usually is detected as an incidental mass lesion
graphic windows demonstrate predominance of the right-sided
on chest radiography or echocardiography in an asymptomatic
cardiac chambers and may lead to an erroneous diagnosis
person (Figure 77.3A). Most frequently, it is located at the right
of right ventricular volume overload and atrial septal defect.
costophrenic angle, but it may also be found at the left costo-
Cardiac motion, especially of the posterior wall of the left ven-
phrenic angle, hilum, or superior mediastinum. The differential
tricle, is exaggerated on echocardiography. All cardiac structures
diagnosis is malignant tumors, cardiac chamber enlargement,
are shifted to the left, resulting in prominent visualization of the
and diaphragmatic hernia. Two-dimensional echocardiography,
right ventricular cavity and abnormal ventricular septal motion
CT, or MRI may be used to differentiate pericardial cysts from
(Figure 77.2C).
other solid tumors (Figure 77.3B and 77.3C). Rarely, compression
Congenital absence of the pericardium is associated with
of adjacent structures occurs, resulting in persistent cough, dys-
atrial septal defect, bicuspid aortic valve, and bronchogenic cysts.
pnea, chest pain, or hemodynamic compromise. The vast majority
Rarely, herniation of cardiac chambers through a partial defect
of pericardial cysts are in asymptomatic patients and require no
of the pericardium may cause sudden death, presumably because
treatment. For symptomatic patients for whom there is no clear
of marked ischemia from compression of the coronary artery.
alternative explanation, aspiration may be performed to investigate
Closure of the pericardial defect is necessary in symptomatic
whether symptoms are related to the cyst. The rate of recurrence
patients.
is high after aspiration. However, if symptoms improve, surgical
• Congenital absence of the pericardium gives a “Snoopy” dog car- resection or aspiration and sclerotherapy may be beneficial.
diac silhouette on a chest radiograph.
• Pericardial cyst is usually benign and located at the right costo-
• In congenital absence of the pericardium, chest radiography shows phrenic angle.
a lucent area between the aorta and the main pulmonary artery
and between the inferior border of the heart and the left hemidi- • Pericardial cyst is usually asymptomatic and requires no treatment.
aphragm from lung interposition.
• Congenital absence of the pericardium is associated with atrial Acute Pericarditis
septal defect, bicuspid aortic valve, and bronchogenic cysts.
The causes of acute pericarditis are numerous (Figures 77.4–
• Partial absence of the pericardium has been linked to sudden death. 77.6). Acute pericarditis usually is self-limited unless caused by
 77 Pericardial Diseases 711

Figure 77.2. Congenital Absence of the Pericardium. A, Characteristic posteroanterior chest radiograph. B, Magnetic resonance imaging showing
marked leftward displacement of the heart. Arrows indicate area of absent pericardium. C, Characteristic 4-chamber view of the heart on 2-dimen-
sional echocardiography showing lateral displacement of the apex and appearance of an enlarged right ventricle (RV) in the presence of a normal-
sized right atrium (RA). LA indicates left atrium; LV, left ventricle. (B and C, Previously published. See “Credit Lines” section.)

malignancy or other systemic disease. Occasionally, acute peri-
carditis may undergo a transient constrictive phase. The most
prominent symptom of acute pericarditis is pleuropericardial
chest pain. Because the visceral pericardium is devoid of pain
fibers, the parietal pericardium must be inflamed to cause chest
pain. Characteristically, the pain is sharp, stabbing, and pleuritic
and radiates to the scapula and back. Pericarditic pain may mimic
anginal pain, and clinical differentiation may be difficult on the
basis of the medical history alone. It is not uncommon for patients
with acute pericarditis to undergo urgent coronary angiography
because of a concern that the condition is an ST-segment eleva-
tion MI. In patients with acute pericarditis, pericardial effusion
may develop to the point of causing hemodynamic compromise,
resulting in dyspnea, hypotension, tachycardia, and heart failure.
On physical examination, a typical finding is a pericardial fric-
tion rub, which is characterized by scratchy high-pitched sounds
with 3 distinct components (coincidental with rapid ventricular
filling, ventricular contraction, and atrial contraction). In a sub-
set of patients, however, the pericardial friction rub may have
only 1 component. A pericardial rub usually is heard best at the
left sternal border, with the patient leaning forward during held
expiration. It is common for the rub to disappear when a peri-
cardial effusion develops. A pericardial knock does not occur in
acute pericarditis.
712 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

A B
Liver

RV
LV RA *

Figure 77.3. Pericardial Cyst. A, Chest radiograph showing typical findings in an asymptomatic patient with pericardial cyst, which is usually
located in the right side of the heart. B, Subcostal echocardiographic view showing a large echo-free cystic structure (*) adjacent to the right atrium
(RA) and typical of a pericardial cyst. C, Computed tomographic scan of the heart showing a pericardial cyst (*) in the same patient as in B. LV
indicates left ventricle; RV, right ventricle.

Diagnostic Evaluation of Acute Pericarditis
The ST-segment elevation in acute pericarditis is different from
that in acute MI (Figure 77.7). In pericarditis, ST-segment eleva-
tion is more diffuse, involving both limb and precordial leads,
and it is concave upward and associated with upright T waves.
After several days of pericarditis, the ST segment returns to
baseline and the T wave flattens and, later, becomes inverted.
Another ECG characteristic of pericarditis is depression of the
PR segment because of atrial involvement (Figure 77.7). This
happens within several days after the onset of pericarditis.
Chest radiographs usually are normal unless the patient
has a large amount of pericardial effusion. The most sensi-
tive diagnostic technique for detecting pericardial effusion is
echocardiography, which shows an echo-free space around the
heart. The absence of pericardial effusion on echocardiography
does not exclude the diagnosis of acute pericarditis.
Treatment of Acute Pericarditis
In most patients, acute pericarditis resolves gradually, and treat-
ment is with NSAIDs, usually aspirin (650 mg every 4 hours) or
indomethacin (25–75 mg 3 times daily for 7–10 days), with grad-
ual tapering. Rarely, recurrent chest pain develops, for which
more intense treatment with NSAIDs or colchicine (0.6 mg
twice daily), or both, should be considered. Treatment of peri-
carditic pain with a corticosteroid may lead to steroid depen-
dency. Corticosteroid treatment should be considered only when

Figure 77.4. Fibrinous Bread-and-Butter Pericarditis. Figure 77.5. Epicardial Fat Necrosis of the Heart in Pancreatitis.
 77 Pericardial Diseases
Figure 77.6. Healed Pericarditis Due to Systemic Lupus Erythematosus.
pericarditic pain does not respond to combinations of NSAIDs.
Colchicine has been used to treat recurrent pericarditic pain, but
its benefit is much less if the patient is already receiving cortico-
steroid treatment. If the pain continues to limit the patient’s
activity and lifestyle, pericardiectomy may be required, even if
there is no hemodynamic embarrassment.
Transient Constrictive Phase of Acute
Pericarditis
About 7% to 10% of patients with acute pericarditis have a tran-
sient constrictive phase. These patients usually have a moderate
amount of pericardial effusion, and as the pericardial effusion
disappears, the pericardium remains inflamed, thickened, and
noncompliant, resulting in a constrictive hemodynamic con-
dition. The patient presents with dyspnea, peripheral edema,
increased jugular venous pressure, and sometimes ascites, as
in patients with chronic constrictive pericarditis. This transient
constrictive phase may last 2 to 3 months before it gradually
resolves either spontaneously or with treatment with anti-inflam-
matory agents. When hemodynamic conditions and findings typ-
ical of constriction develop in patients with acute pericarditis,
initial treatment is indomethacin for 2 to 3 weeks and, if there is
no response, corticosteroids for 1 to 2 months after confirming
Figure 77.7. Electrocardiogram Typical of Acute Pericarditis. Arrow indicates PR-segment depression in lead I. Circle indicates typical concave
upward ST-segment elevation.
713
that the pericarditis is not caused by bacterial infection, includ-
ing tuberculosis. A constrictive hemodynamic condition can be
diagnosed readily with Doppler echocardiography (see below);
resolution of constrictive physiology can be documented clini-
cally and by follow-up echocardiography.
• The classic pericardial rub has 3 components, coincidental
with rapid ventricular filling, ventricular contraction, and atrial
contraction.
• Acute pericarditis may cause PR-segment depression on the ECG
because of inflammation of the atrial wall.
• Among patients with acute pericarditis, 7%-10% may have a tran-
sient constrictive phase.
Pericardial Effusion and Cardiac Tamponade
Pericardial inflammation from any cause may result in a large
amount of fluid collecting in the pericardial sac. The pericardium
may be filled with blood (hemopericardium) because of cardiac
rupture (from injury, an iatrogenic cause, or acute MI), proxi-
mal aortic dissection, or cardiac bypass surgery (Figure 77.8).
Pericardial effusion may be related to underlying heart failure
or to an abnormality in lymphatic drainage (chylous effusion).
When a pericardial effusion develops gradually, so that it does
not impair pericardial compliance, the patient may remain
asymptomatic, even with a massive amount of pericardial effu-
sion. If the rate of accumulation is rapid, even a small amount
(50–100 mL) of fluid or blood in the pericardium can cause car-
diac tamponade. Cardiac tamponade is the result of a critical ele-
vation of intrapericardial pressure produced by an accumulation
of pericardial fluid.
Hemodynamics of Cardiac Tamponade
When intrapericardial pressure is increased, atrial pressure is
also increased, resulting in impaired venous return. This, in turn,
results in systemic venous congestion and reduction in cardiac
output (Figure 77.9). On physical examination, jugular venous
pressure is increased, with a prominent systolic x descent and a
blunted diastolic y descent (Table 77.1). With pericardial effu-
sion, the precordium is quiet and cardiac sounds are diminished.
The classic findings of the Beck triad consist of 1) a decrease
in systolic pressure, 2) an increase in systemic venous pressure,
and 3) a quiet heart. They occur in only 10% to 40% of patients.
714 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

30

Paradoxus,
24

Pulsus
18

%
12
6
0

Brachial Artery
170 Systolic

Pressure,
mm Hg
130 Expiration
90 Inspiration
Diastolic
50

11

Cardiac
Output,
L / m in
9
7
5

Heart
Rate,
bpm
103
93

Volume,
100

Stroke

mL
80
60
40
RVEDP
Pressure,

100 Pericardial
mm Hg

80
60
40

Figure 77.8. Postoperative Organizing Hemopericardium.
The reduction in cardiac output produces a narrow pulse pres-
sure, and systemic venous congestion may cause hepatomegaly,
peripheral edema, and ascites. In patients with cardiac tampon-
ade, the intrapericardial pressure is increased critically and does
not vary with intrapleural pressure. Normally, intrapericardial
pressure does change with fluctuations in intrapleural pressure
such that with inspiration, intrapleural pressure decreases by 5
to 7 mm Hg and similar changes occur in intrapericardial pres-
sure. However, in cardiac tamponade, intrapericardial pressure
is increased to the level of ventricular diastolic pressures. Both
ventricular diastolic pressures equalize with the pericardial
pressure. Therefore, left atrial, right atrial, and right ventricu-
lar end-diastolic pressure, pulmonary end-diastolic pressure, and
pulmonary capillary wedge pressure are equalized within 5 mm
Hg of one another.
Patients who are severely hypovolemic may have “low pres-
sure cardiac tamponade,” whereby comparatively low to normal
intrapericardial pressure results in cardiac chamber compression
due to low filling pressure. Ventricular filling and stroke volume
are affected by relatively normal pressures. Characteristic clini-
cal findings such as jugular venous distention may be absent.
Pulsus Paradoxus
Pulsus paradoxus is a decrease (>10 mm Hg) in systolic blood
pressure during inspiration and is due to the underlying mech-
anism of cardiac tamponade. With increased intrapericardial
Pericardial
Fluid, mL
Present 1,000 800 600 400 200 0
Figure 77.9. Hemodynamic Changes During Pericardial Fluid
Withdrawal in a Patient With Cardiac Tamponade. In the second frame
from the top, note the disappearance of pulsus paradoxus. bpm indicates
beats per minute; RVEDP, right ventricular end-diastolic pressure.
pressure, normal pressure transmission from the intrapleural to
the intrapericardial cavity does not occur. Thus, on inspiration,
the driving blood pressure across the pulmonary vascular bed
decreases as the lungs expand. Pulmonary arteriolar pressure
decreases, while left atrial and left ventricular pressures remain
relatively fixed. Thus, the decrease in pulmonary venous return to
the left heart during inspiration translates into a decrease in left
ventricular stroke volume, which is detected clinically as pulsus
paradoxus. Pulsus paradoxus is characteristic of cardiac tampon-
ade, but it also occurs in other conditions in which there is a sig-
nificant decrease in forward stroke volume with inspiration, such
as acute cor pulmonale (pulmonary embolism), chronic obstruc-
tive lung disease, right ventricular infarction, and asthma.
Echocardiographic Diagnosis of Pericardial
Effusion and Cardiac Tamponade
Chest radiography may show cardiomegaly with a globu-
lar “water bottle,” sac-like appearance. The best way to detect
pericardial effusion and cardiac tamponade is with echocardi-
ography. A small amount of pericardial fluid appears as an echo-
free space. As pericardial effusion increases, movement of the
parietal pericardium decreases. When there is a large volume
 77 Pericardial Diseases 715

Table 77.1. Comparison of Cardiac Tamponade and Con- fixed with cardiac tamponade, reciprocal changes occur in the
strictive Pericarditis right chambers so that tricuspid inflow velocity increases with
inspiration and decreases with expiration. With a decrease in fi l-
Feature Cardiac Tamponade Constrictive Pericarditis ling to the right chambers during expiration, there is significant
Pulsus paradoxus Present Present flow reversal in the hepatic vein with expiration during diastole
Kussmaul sign Absent May be present (Figures 77.11–77.13).
Pericardial knock Absent May be present
Systolic x descent Large Normal
Diastolic y descent Small or absent Large
Treatment of Cardiac Tamponade
The only effective treatment for cardiac tamponade is the removal
of pericardial fluid. The best way to perform pericardiocentesis
of pericardial effusion, the heart may swing in the pericardial is with echocardiographic guidance, which helps in locating the
cavity (Figure 77.10), causing the ECG manifestation of car- optimal site for the puncture, in determining the depth of the
diac tamponade, electrical alternans. However, the swinging pericardial effusion, in measuring the distance from the punc-
motion may be absent in cardiac tamponade. Other M-mode or ture site to the effusion, and in monitoring the results of the
2-dimensional echocardiographic findings of cardiac tamponade pericardiocentesis.
include the following:
• Pulsus paradoxus is classically seen in cardiac tamponade.
1. Inspiratory decrease in the E-F slope of the mitral valve • Elective removal of pericardial fluid should always be guided
2. Early diastolic collapse of the right ventricle echocardiographically to reduce complications.
3. Late diastolic collapse of the right atrial free wall
4. Plethora of the inferior vena cava with a blunted respiratory change
5. Abnormal ventricular septal motion Pericardial Effusion Due to Malignancy
In acute myocardial rupture, clotted blood in the pericar- Pericardial effusion due to malignancy is a poor prognostic sign.
dial sac is highly suggestive of hemopericardium. If air is If cytologic examination demonstrates malignant cells in the
present in the pericardial sac (pneumopericardium), echocar- pericardial effusion, the patient’s prognosis is grim regardless
diographic imaging may be difficult. The Doppler findings in of the underlying type of malignancy. Infrequently, pericardial
cardiac tamponade are based on the hemodynamic pathophysi- effusion may be the initial manifestation of an underlying malig-
ology described for pulsus paradoxus. With inspiration, the driv- nancy. The tumors that spread most frequently to the pericar-
ing pressure gradient to the left cardiac chamber is decreased dium are those of the lung and breast, followed by lymphoma and
so that mitral inflow velocity decreases with inspiration and leukemia. Patients with angiosarcoma, a primary cardiac tumor,
increases with expiration. Because cardiac volume is relatively present with pericardial effusion and pericarditis. The pericardial

Figure 77.10. Pericardial Effusion. Upper, Parasternal long-axis views showing a large amount of pericardial effusion and swinging motion of the
heart. Lower, With the swinging motion of the heart, the QRS voltage direction alternates, producing electrical alternans.
716 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

A
Normal Normal Mitral Flow Velocity
Change in Intrathoracic Pressure

10
With Respiration, mm Hg

Lung
Exp PC PV 1 m/s
5 Insp
LA

0
LV
-5
-10
Exp
Insp

Tamponade
20 Exp Tamponade
Change in Intrathoracic Pressure

PV 1 m/s Mitral Flow Velocity

PC Lung
With Respiration, mm Hg

15 Insp LA
A
10

5 LV E A

0
-5
Exp

Insp
Figure 77.11. Intrathoracic and Intracardiac Pressure Changes With
Respiration. Upper, Normal physiology. Lower, Cardiac tamponade.
The shaded areas indicate left ventricular (LV) filling pressure gradi-
ents (difference between pulmonary capillary wedge pressure and LV B Tamponade Hepatic Venous Flow Velocity
diastolic pressure). At the bottom of each drawing is a schematic mitral
inflow Doppler velocity profile reflecting LV diastolic filling. In car-
diac tamponade, a decrease in LV filling occurs after inspiration (Insp)
AR
because the pressure decrease in the pericardium and LV is smaller than
in the pulmonary capillaries (PC). LV filling is restored after expiration
(Exp). LA indicates left atrium; PV, pulmonary vein. (Previously pub-
lished. See “Credit Lines” section.)
Exp
2 m/s D
S

fluid in malignancies is usually bloody, but a bloody effusion is

not specific for malignancy. Recurrent pericardial effusion can Insp
be treated with creation of a pericardial window, with repeated Hepatic Venous Flow Velocity
Normal
pericardiocentesis, and sometimes with a pigtail catheter left in
place for several days for continuous or intermittent drainage of
reaccumulated fluid. Instillation of a sclerosing agent into the
pericardium is also effective but is painful and rarely used. VR AR

Pericarditis in Acute MI
Pericardial effusion occurs in about 20% of patients with acute
transmural MI and is usually associated with a large anterior
wall MI. The chest pain, which is different from that of ischemic
chest pain, has a pleuritic component and may be associated with
a pericardial rub. The presence of a pericardial effusion with or
without pericarditic pain after MI is not a contraindication for
intravenous treatment with heparin. Hemopericardium can occur
after myocardial rupture as a complication of acute MI and most
frequently is associated with a lateral MI (Figure 77.14). In most
patients with myocardial rupture, electromechanical dissociation
develops, and they do not survive. A subgroup of patients may
have subacute cardiac rupture and present with nausea, vomit-
ing, restlessness, and persistent ECG changes. Rarely, a pseudo-
aneurysm develops, in which the hemopericardium is contained
D
S 2 m/s
Exp
Insp
Figure 77.12. Flow Velocity Profiles. A, Mitral inflow velocity pro-
files typical of a healthy person (upper) and a patient with cardiac tam-
ponade (lower). B, Hepatic venous flow velocity profiles from a patient
with cardiac tamponade (upper) (same patient as in lower panel of A)
and a healthy person (lower). A indicates mitral A-wave velocity at
atrial contraction; AR, pulmonary vein atrial reversal flow; D, pulmo-
nary vein diastolic forward flow; E, peak mitral early filling velocity;
Exp, expiration; Insp, inspiration; S, pulmonary vein systolic forward
flow; VR, ventricular reversal flow. (Previously published. See “Credit
Lines” section.)
 77 Pericardial Diseases
Figure 77.13. Pericarditis With Pericardial Effusion.
by the adjacent structures. Although pseudoaneurysm of the left
ventricle was previously considered a surgical emergency, review
of the data now suggests that rupture of chronic pseudoaneurysm
is rare or occurs at a low rate.
Dressler Syndrome
In some patients, pericarditis develops several weeks after MI
(Dressler syndrome). It is thought to be immune mediated.
Usually it is treated initially with NSAIDs, but corticoster-
oid therapy may be needed for a small number of patients with
refractory chest pain.
Postcardiotomy Syndrome
Postcardiotomy syndrome is similar to Dressler syndrome, and
an autoimmune response to cardiac antigens has been impli-
cated. Postcardiotomy syndrome occurs in about 5% of patients
who have a cardiac surgical procedure, with symptoms occur-
ring 3 weeks to 6 months postoperatively. It most likely is related
to surgical trauma and irritation by blood in the mediastinum and
Figure 77.14. Pathology Specimen From a Patient Who Died of
Cardiac Rupture and Hemopericardium. (See “Credit Lines” section.)
717
pericardium. The initial treatment of this syndrome is NSAIDs,
and only for refractory symptoms should systemic corticoster-
oids be used. Pericardiectomy is rarely required. A study has
suggested that postcardiotomy syndrome can be prevented with
prophylactic use of colchicine in the early postoperative period.
Cardiac Tamponade Related
to Aortic Dissection
Cardiac tamponade or hemopericardium may occur with prox-
imal aortic dissection. After cardiac tamponade is recognized
as being a result of aortic dissection, urgent surgical repair of
the aortic dissection and tamponade is needed. In patients with
cardiac tamponade and aortic dissection, pericardiocentesis may
increase blood pressure and cause rupture of the dissected aorta.
A study has shown 60% early mortality for patients with an aor-
tic dissection complicated by cardiac tamponade. All patients
who underwent pericardiocentesis died shortly thereafter.
Hemopericardium due to aortic dissection has a characteristic
appearance on echocardiography.
Constrictive Pericarditis
Constrictive pericarditis is characterized by restrictive ventric-
ular filling due to a pericardium that is thickened or calcified,
or both. The pericardium usually contains calcified fibrous scar
tissue from an inflammatory process, and in the advanced stage,
the scarring may involve the epicardium. The causes of constric-
tion are several, including acute pericarditis, collagen vascular
disease, coronary artery bypass surgery (Figures 77.15 and 77.16),
and tuberculosis (Box 77.1). However, in many patients, the
cause may not be identified. In the modern era, the most common
cause of constrictive pericarditis is a previous cardiac surgery
(accounting for one-third of cases at Mayo Clinic). In develop-
ing countries, where tuberculosis is still common, it is a frequent
cause of constriction, but it is rare in the developed world.
The main clinical features at presentation are dyspnea,
peripheral edema, marked systemic venous congestion with
hepatomegaly, and ascites. Frequently, patients are evaluated for
primary liver disease and may undergo a liver biopsy before con-
strictive pericarditis is diagnosed. The most prominent findings
on physical examination are related to systemic venous conges-
tion, such as distention of the jugular vein (Box 77.2). Venous
Figure 77.15. Noncalcific Constrictive Pericarditis 4 Years After
Coronary Artery Bypass Grafting.
718 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Box 77.2. Signs and Symptoms of Constrictive

Pericarditis
In >95% of patients
Increased jugular venous pressure
Hepatomegaly
In >75% of patients
Dyspnea
Edema
Abdominal swelling (ascites)
Pleural effusion
Severe fatigue
In >25% of patients
Pulsus paradoxus
Figure 77.16. Constrictive Pericarditis After Coronary Artery Palpitations
Bypass Grafting. Cough
Abdominal pain
pressure often increases with inspiration (Kussmaul sign) Orthopnea
because of the inability of the right side of the heart to accept In <25% of patients
the increased cardiac input with inspiration. Unlike patients with
Pericardial knock
cardiac tamponade with x descent, patients with constrictive
pericarditis have rapid y descent, which reflects the early dias- Muscle wasting
tolic decrease in right ventricular pressure. Nausea and vomiting
With high atrial pressure, rapid filling of the ventricle is acceler- Dizziness
ated, and this generates the third heart sound (pericardial knock), In <5% of patients
which usually occurs 80 to 120 ms after aortic valve closure Finger clubbing
(Figure 77.17), corresponding to the nadir of the y descent and the
end of early diastolic filling. The differential diagnosis of a dias-
tolic gallop occurring 80 to 120 ms after aortic valve closure also Pericardial Calcification
includes opening snap of the mitral valve in mitral stenosis (which
is followed by a diastolic rumble), tumor plop from atrial myxoma, When constrictive pericarditis is suspected clinically, chest
and a third heart sound related to left ventricular failure. radiographs, including a left lateral projection, should be
The correct diagnosis of constrictive pericarditis is crucial reviewed to look for pericardial calcification (Figures 77.18 and
because most of the symptoms can be reversed with pericardiec- 77.19). Pericardial calcification was commonly seen in patients
tomy. However, the symptoms and clinical findings mimic those with tuberculous pericarditis, but currently it is seen most often
of restrictive cardiomyopathy, which is a progressive disease in patients with idiopathic constrictive pericarditis. If a patient
with no effective treatment. presents with ascites and other clinical evidence of significant
systemic venous congestion and a pericardial knock, chest radio-
graphic findings of pericardial calcification make constrictive
pericarditis the leading diagnosis, and the patient requires surgi-
Box 77.1. Causes of Constrictive Pericarditis
cal exploration and pericardiectomy. No single ECG abnormality
Unknown (idiopathic) is diagnostic of constrictive pericarditis.
Acute pericarditis of any cause
Cardiac surgery
ECG
Uremia
Connective tissue disease (systemic lupus S1 S2
erythematosus, scleroderma, rheumatoid arthritis) LSB
Trauma
Drugs (procainamide, hydralazine, methysergide) x
JUG
Radiation
Neoplastic pericardial disease (melanoma, y
mesothelioma)
Tuberculosis, fungal infections (histoplasmosis,
coccidioidomycosis), parasitic infections
Myocardial infarction
Dressler syndrome Figure 77.17. Pericardial Knock. Simultaneous electrocardiogram
(ECG), phonocardiogram (LSB), and jugular venous pressure (JUG)
Purulent pericarditis tracings showing the timing of the pericardial knock (arrow). S1 indi-
Pulmonary asbestosis cates first heart sound; S2, second heart sound. (Previously published.
See “Credit Lines” section.)
 A

Figure 77.18. Pericardial Calcification. A, Lateral chest radiograph showing calcified pericardium (arrows). B, Computed tomographic scan of
the chest showing egg shell–like calcification of the pericardium (arrow).

A B
Figure 77.19. Pericardial Calcification. Chest radiographs showing calcified pericardium (arrows) in constrictive pericarditis. A, Posteroanterior
view. B, Lateral view.
720 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
Echocardiography in Constrictive Pericarditis
Echocardiography is helpful in diagnosing constrictive pericar-
ditis. The characteristic M-mode or 2-dimensional echocardio-
graphic findings include abnormal ventricular septal motion
(Figure 77.20), increased pericardial thickness or calcification,
dilated inferior vena cava with no significant changes on inspi-
ration, and flattening of the left ventricular posterior wall dur-
ing diastole. However, these findings are neither sensitive nor
specific.
Although the underlying pathologic mechanism of constric-
tion is different from that of cardiac tamponade, the hemo-
dynamic events of respiratory variation during left and right
ventricular filling are similar in the 2 conditions. The thickened
pericardial layer prevents full transmission of intrapleural pres-
sure changes with respiration to the pericardial and intracar-
diac cavity, creating respiratory variation in the left-side filling
pressure gradient (the pressure difference between the pulmo-
nary vein and the left atrium). Therefore, the mitral inflow and
pulmonary venous diastolic flow velocities decrease immedi-
ately after the onset of inspiration and increase with expiration
(Figure 77.21). Reciprocal changes occur in tricuspid inflow and
hepatic venous flow velocity because of the relatively fixed car-
diac volume. With decreased filling of the right cardiac cham-
bers on expiration, there are exaggerated diastolic flow reversals
and decreased diastolic forward flow in the hepatic vein with the
onset of expiration. In contrast, hepatic venous flow reversals are
more prominent with inspiration in restrictive cardiomyopathy.
However, it is not unusual to see significant diastolic flow rever-
sals in the hepatic vein during both inspiration and expiration
in patients with advanced constriction or combined constriction
and restriction.
A representative Doppler spectrum of constrictive pericar-
ditis is shown in Figure 77.22. A subgroup of patients with
constrictive pericarditis may not show the characteristic respi-
ratory variation and Doppler velocities. Therefore, the absence
of respiratory variation in patients with clinical evidence of sig-
nificant systemic venous congestion does not exclude the diag-
nosis of constrictive pericarditis, and additional studies should
be performed. The typical respiratory variation and Doppler
RV
VS
LV
Figure 77.20. M-Mode Echocardiogram With Simultaneous
Respirometer Recording in Constrictive Pericarditis. Upward deflec-
tion indicates passive inspiration, and downward deflection the onset
of expiration. The ventricular septum (VS) moves toward the left ven-
tricle (LV) with inspiration (red arrow) and toward the right ventricle
(RV) with expiration (blue arrow). The underlying hemodynamics are
explained in the text.
A S D
HV PV
LA LA
R
A R
A
LV LV
RV Thickened RV
pericardium
Inspiration Expiration
PCW
IP
B
Figure 77.21. Thickened Pericardium. A, Diagram of the heart with
a thickened pericardium illustrates the respiratory variation in ven-
tricular filling and the corresponding Doppler features of the mitral
valve, tricuspid valve, pulmonary vein (PV), and hepatic vein (HV).
These changes are related to discordant pressure changes in the ves-
sels in the thorax, such as pulmonary capillary wedge (PCW) pressure
and intrapericardial (IP) and intracardiac pressures. Hatched areas
under the curves indicate the reversal of flow. Thicker arrows indi-
cate greater filling. B, Simultaneous pressure recordings from the LV
and the PCW together with mitral inflow velocity on a Doppler ech-
ocardiogram. The onset of the respiratory phase is indicated at the
bottom. With the onset of expiration, PCW pressure increases much
more than LV diastolic pressure, creating a large driving pressure gra-
dient (large arrowhead). With inspiration, however, PCW pressure
decreases much more than LV diastolic pressure, with a very small
driving pressure gradient (3 small arrowheads). These respiratory
changes in the LV filling gradient are well reflected by the changes
in the mitral inflow velocities recorded on Doppler echocardiogra-
phy. D indicates diastolic flow; Exp, expiration; Insp, inspiration; LA,
left atrium; LV, left ventricle; RA, right atrium; RV, right ventricle;
S, systolic flow. (Previously published. See “Credit Lines” section.)
velocities also can occur in other conditions, such as chronic
obstructive lung disease, right ventricular infarct, sleep apnea,
asthma, and pulmonary embolism. In these situations, pulsed
wave Doppler recording of the superior vena cava demonstrates
a marked increase in systolic forward flow with inspiration,
whereas the increase is less marked (<20 cm/s) in constrictive
pericarditis.
A relatively new echocardiographic technique, tissue Doppler
imaging, records the velocity of the myocardium and has consid-
erable diagnostic value in the diagnosis of constriction. The Ea
correlates well with the status of myocardial relaxation. Since
 77 Pericardial Diseases 721

Mitral 1 m/s Tricuspid

E 0.6 m/s
E
A A

Insp Exp Insp Exp

Pulmonary Vein Hepatic Vein

0.6 m/s 0.4 m/s
DR
SR

Insp Exp Insp D Exp

Figure 77.22. Constrictive Pericarditis. Composite of mitral valve, tricuspid valve, pulmonary vein, and hepatic venous flow velocities that are
typically seen. Mitral inflow velocity decreases (single arrowhead) after inspiration and increases (double arrowheads) after expiration. Tricuspid
inflow velocity shows the opposite changes. Diastolic forward pulmonary venous flow decreases (single arrowhead) after inspiration and increases
(double arrowheads) after expiration. The hepatic vein shows a marked decrease in diastolic forward flow and an increase in diastolic flow reversals
(DR) after expiration. A indicates late diastolic velocity; D, diastolic flow; E, early diastolic velocity; Exp, expiration; Insp, inspiration; S, systolic
flow; SR, systolic flow reversal. (Previously published. See “Credit Lines” section.)

myocardial relaxation is reduced in myocardial diseases, Ea is
reduced in restrictive cardiomyopathy (normally, Ea >10 cm/s; in
cardiomyopathy, Ea <6 cm/s) but is preserved or even increased
in constrictive pericarditis. Therefore, if Ea is more than 8 cm/s
in a patient with clinical evidence of heart failure, constrictive
pericarditis should be considered.
Pericardial Thickness in Constrictive Pericarditis
CT or MRI is best for determining pericardial thickness
(Figure 77.23). Demonstration of increased pericardial thickness
on CT or MRI in patients with significant systemic venous con-
gestion generally indicates constrictive pericarditis. Pericardial
thickness has been assessed with transesophageal echocardiog-
raphy, and the findings correlate well with those from CT scan.
By itself, though, this finding is not sensitive or specific for
constrictive pericarditis.
However, normal pericardial thickness on an imaging study
cannot be used to exclude the diagnosis of constriction. In a
study from Mayo Clinic, 18% of patients with surgically con-
firmed constrictive pericarditis had pericardium with a thickness
of 2 mm or less. When the pericardial thickness is not increased,
the pericardium is usually adhered to the heart, causing constric-
tive hemodynamic abnormalities. Therefore, the diagnosis of
constriction should be based on hemodynamic abnormalities in
addition to anatomical abnormalities.
BNP in Constrictive Pericarditis
The BNP level is markedly elevated in patients with heart fail-
ure due to systolic dysfunction, and it is less elevated in patients
with heart failure due to diastolic dysfunction in the presence of
normal ejection fraction. BNP is even less elevated or normal
in patients with constrictive pericarditis even though the filling
pressure may be similarly elevated.
Hemodynamic Findings in Constrictive
Pericarditis
The hemodynamic findings in constrictive pericarditis include
an increase in right atrial pressure and a dip-and-plateau con-
figuration of the right and left ventricular diastolic pressure
tracings (Figure 77.24). Because there is no restriction of early
ventricular filling, the y descent is quite prominent, correspond-
ing to a prominent early diastolic dip of the ventricular pressure
tracing. Right ventricular systolic pressure is usually less than
50 mm Hg, but this finding is not sensitive or specific and cannot
be used to differentiate constrictive pericarditis from restrictive
722 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

• A subset of patients with constrictive pericarditis may not show the

typical respiratory changes in Doppler velocities.
• A patient with heart failure and normal ejection fraction may have
constriction.

Restriction Versus Constriction

Figure 77.23. Thickened Pericardium. Computed tomographic scan
showing thickened pericardium (arrow) along with pleural effusion due
to constrictive pericarditis.
cardiomyopathy (Figure 77.25). The concept of ventricular inter-
dependence and the reciprocal pressure changes in the right and
left ventricles with respiration can be used in hemodynamic
assessment. Simultaneous left and right ventricular pressure
tracings show a discordant direction of pressure changes with
respiration (Figure 77.24). With inspiration, left ventricular pres-
sure decreases and right ventricular pressure increases. Opposite
changes occur with expiration. Also, simultaneous left ventric-
ular diastolic pressure and pulmonary capillary wedge pressure
show a significant reduction in the pressure difference between
the pulmonary capillary wedge pressure and the left ventricular
diastolic pressure with inspiration in comparison with the differ-
ence during expiration (Figure 77.21).
• There is no characteristic ECG abnormality that is diagnostic of
constrictive pericarditis.
• A patient who has nonspecific findings on liver biopsy may have
pericardial constriction.
The clinical and hemodynamic profiles of restriction and con-
striction are similar (Figure 77.24) even though these conditions
have distinctly different pathophysiologic mechanisms. Both
are caused primarily by diastolic filling abnormalities, with
preserved global systolic function. The diastolic dysfunction in
restrictive cardiomyopathy results from a stiff and noncompli-
ant ventricular myocardium, whereas in constrictive pericarditis
it results from a thickened and noncompliant pericardium. Both
disease processes limit diastolic filling and result in diastolic
heart failure (Figure 77.26). Pathologically, restriction and con-
striction may appear similar, with normal-sized ventricles and
enlarged atria, but the pericardium is thickened in constriction.
Infiltrative Cardiomyopathy
Infiltrative cardiomyopathy has typical 2-dimensional echocar-
diographic findings and biochemical abnormalities. A proto-
typical example is cardiac amyloidosis, which is characterized
by increased ventricular wall thickness, a granular or spar-
kling myocardial appearance on echocardiography, and typical
amyloid deposits in fat in myocardial biopsy specimens. Also,
patients usually (but not always) have monoclonal gammopathy
on serum protein electrophoresis. The ECG shows a low voltage
despite increased left ventricular wall thickness.
Noninfiltrative Restrictive Cardiomyopathy
Noninfiltrative restrictive cardiomyopathy is difficult to diag-
nose. The myocardium becomes noncompliant with fibrosis and
scarring, but systolic function is usually maintained. With lim-
ited diastolic filling and increased diastolic pressure, the atria
become enlarged. In contrast, myocardial compliance usually is
not decreased in patients with constrictive pericarditis. The thick-
ened and sometimes calcified pericardium limits diastolic filling,
resulting in hemodynamic characteristics that are different from

RCM Constriction
100 Insp
Exp
LV 150
Pressure, mm Hg

LV

100
50
RV
RV
50

0 Insp Exp 0

Figure 77.24. Ventricular Pressure Tracings. Simultaneous left ventricular (LV) and right ventricular (RV) pressure tracings in restrictive cardi-
omyopathy (RCM) and constrictive pericarditis. In both conditions, the tracings show equalization of diastolic pressures and a dip-and-plateau con-
figuration (arrowheads). There are concordant changes in LV and RV systolic pressures with respiration in RCM (both LV and RV pressures increase
with inspiration), whereas there are discordant pressure changes (LV systolic pressure decreases and RV systolic pressure increases with inspiration)
in constrictive pericarditis (slopes of arrows show the concordance and discordance). Exp indicates expiration; Insp, inspiration.
 A Feature Constriction Restriction

LVEDP – RVEDP, mm Hg ≤5 >5

RVSP, mm Hg ≤50 >50

RVEDP/RVSP ratio ≥0.33 <0.33

B LVEDP – RVEDP RVSP RVEDP/RVSP ratio
30 100 0.8

80
Pressure, mm Hg

Pressure, mm Hg
20 0.6

60

Ratio
10 0.4
40

0 0.2
20

-10 0 0.0
Constriction Restriction Constriction Restriction Constriction Restriction
Figure 77.25. Comparison of Hemodynamic Features of Constriction and Restriction. LVEDP indicates left ventricular end-diastolic pressure;
RVEDP, right ventricular end-diastolic pressure; RVSP, right ventricular systolic pressure. (B, Previously published. See “Credit Lines” section.)

Normal Constriction Restriction

ECG

i e i e i e
Resp
E
A

MV
DT≥150 ms DT<150 ms

Mitral
Annulus
Velocity

SR DR
HV
D or
S

Figure 77.26. Mitral Valve (MV) Inflow, Mitral Annulus Velocity, and Hepatic Vein (HV) Features in Normal Conditions, Constriction, and
Restriction. A indicates late diastolic filling; D, diastole; DR, diastolic reversal; DT, deceleration time; e, expiration; E, early diastolic filling; ECG,
electrocardiogram; i, inspiration; Resp, respiration; S, systole; SR, systolic reversal.
724 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
those of restrictive cardiomyopathy. Atrial enlargement is less
prominent in constriction than in restrictive cardiomyopathy.
When restrictive cardiomyopathy affects both ventricles, clin-
ical signs due to abnormalities of right-sided heart failure are
apparent, with increased jugular venous pressure and peripheral
edema. An early diastolic gallop sound is the rule and, in restric-
tion, is often difficult to distinguish from a pericardial knock.
ECG and chest radiographic findings are nonspecific, except that
a calcified pericardium indicates the possibility of constrictive
pericarditis.
Echocardiographically, it is difficult to distinguish between
restriction and constriction only on the basis of M-mode and
2-dimensional findings. The features of both conditions include
normal left ventricular systolic function, enlarged atria, and a
dilated inferior vena cava. An increase in ventricular wall thick-
ness, a thickening of the valves, and a small amount of pericar-
dial effusion are typical of cardiac amyloidosis. In constrictive
pericarditis, the most striking finding is ventricular septal motion
abnormalities, which can be explained on the basis of respira-
tory variation in ventricular filling. The pericardium usually is
thickened, but this may not be obvious on transthoracic echocar-
diography. Transesophageal echocardiographic measurement of
pericardial thickness correlates well with that measured by elec-
tron-beam CT. However, constriction cannot be distinguished from
myocardial disease on the basis of pericardial thickness alone.
Diagnostic Strategy to Differentiate Restrictive
Cardiomyopathy From Constrictive Pericarditis
The following diagnostic strategy is recommended for differenti-
ating restrictive cardiomyopathy from constrictive pericarditis:
1. The findings of pulsus paradoxus, calcification of the pericardium
(seen on chest radiography), and pericardial knock favor the diagno-
sis of constrictive pericarditis. Decreased voltage on the ECG may
indicate cardiac amyloidosis.
2. Two-dimensional echocardiographic findings of increased left ven-
tricular wall thickness and normal septal motion in conjunction with
enlarged atria suggest restrictive cardiomyopathy. A thickened or
calcified pericardium and ventricular septal motion favor the diag-
nosis of constrictive pericarditis.
3. In constriction, there is a typical respiratory variation in ventric-
ular filling (decreased filling of the left ventricle with inspiration,
increased filling with expiration, and significant hepatic venous flow
reversal with expiration because of decreased filling on the right
side). Restrictive cardiomyopathy is characterized by restrictive
Doppler physiology with increased early diastolic filling (E-wave)
velocity, decreased late diastolic filling (A-wave) velocity, E/A ratio
greater than 2, and shortened deceleration time of E wave. Hepatic
venous diastolic flow reversals occur with inspiration instead of
expiration. A subgroup of patients with constrictive pericarditis may
have similar Doppler findings without respiratory variation. In such
cases, the Doppler examination should be repeated after an attempt
has been made to reduce preload (head-up tilt position or diuretic
therapy). Respiratory Doppler studies may be difficult to perform
with patients who have atrial fibrillation, but these patients should
have abnormal septal motion and hepatic venous flow velocity
changes on Doppler echocardiography.
If the diagnosis is still uncertain after a careful clinical exam-
ination, review of laboratory data, and 2-dimensional Doppler
echocardiographic evaluation, additional studies are needed.
These would include CT or MRI (to examine pericardial thick-
ness) and cardiac catheterization (to look for characteristic
discordant respiratory changes in the left and right ventricular
pressure tracings).
Suggested Reading
Bull RK, Edwards PD, Dixon AK. CT dimensions of the normal peri-
cardium. Br J Radiol. 1998 Sep;71(849):923–5.
Connolly HM, Click RL, Schattenberg TT, Seward JB, Tajik AJ.
Congenital absence of the pericardium: echocardiography as a diag-
nostic tool. J Am Soc Echocardiogr. 1995 Jan-Feb;8(1):87–92.
Imazio M, Brucato A, Ferrazzi P, Rovere ME, Gandino A, Cemin
R, et al; COPPS Investigators. Colchicine reduces postoperative
atrial fibrillation: results of the Colchicine for the Prevention of the
Postpericardiotomy Syndrome (COPPS) atrial fibrillation substudy.
Circulation. 2011 Nov 22;124(21):2290–5. Epub 2011 Nov 16.
Melduni RM, Oh JK, Bunch TJ, Sinak LJ, Gloviczki P. Reconstruction
of occluded thoracic duct for treatment of chylopericardium: a novel
surgical therapy. J Vasc Surg. 2008 Dec;48(6):1600–2.
White CS. MR evaluation of the pericardium. Top Magn Reson Imaging.
1995 Fall;7(4):258–66.
Abbreviations
BNP brain natriuretic peptide
CT computed tomography
Ea early diastolic velocity of the mitral annulus
E/A ratio ratio of early diastolic filling velocity to late diastolic
filling velocity
ECG electrocardiographic
MI myocardial infarction
MRI magnetic resonance imaging
NSAID nonsteroidal anti-inflammatory drug

Pulmonary Embolism
VIVEK IYER, MD
Introduction
PE and DVT are essentially 2 clinical manifestations of the same
disease process and can be grouped as VTE. VTEs can be either
unprovoked (idiopathic or primary) or provoked (secondary).
Provoking factors include inherited and acquired factors such
as the factor V Leiden mutation, cancer, surgery, trauma, and
immobility. Morbidity and mortality from PE depend not only on
the clot burden but also on the presence and extent of preexisting
cardiopulmonary disease and the response of the cardiovascular
system to the PE.
Epidemiology of PE
Findings from the Rochester Epidemiology Project and other
large epidemiologic studies confirm that VTE is a major health
problem with an age- and sex-adjusted annual incidence rate
of 1:1,000, which is comparable to that of stroke. These stud-
ies also showed that in both sexes, VTE risk increases with age
and increases sharply after age 70. Compared with residents in
the community, hospitalized patients have a 100-fold increased
risk of VTE and account for about 30% of the total commun-
ity burden of VTE. Patients recently dismissed from the hospital
account for an additional 30% of VTE cases, and nursing home
residents account for another 20%.
The frequency of PE depends on the diagnostic setting. About
half of hospitalized patients who have VTE—but only one-third
of outpatients who have VTE—present with a PE.
In a large registry of VTE patients from Italy, 42% of the
patients had PE. The PE-related mortality rate at 3 months was
4%. About 50% of these PE-related deaths had occurred by the
5th day, and about 75% had occurred by the 12th day. CTEPH,
Abbreviations and acronyms are expanded at the end of this chapter.
78
a complication of PE, occurs in 0.5% to 4% of patients present-
ing with acute PE and is a cause of long-term morbidity and
mortality.
• In 40%–50% of patients who have symptoms of DVT but no clini-
cal symptoms of PE, imaging studies show a PE.
• After MI and stroke, PE is the third most common cause of death
due to vascular causes.
• PE is diagnosed more often in hospitalized patients than in
outpatients.
Etiology of PE
PE typically occurs as a result of embolization from a prox-
imal lower extremity DVT. Most lower extremity DVTs origi-
nate in calf vessels distal to the popliteal vein. From there, they
may extend proximally to the popliteal vein and beyond, with
an increased potential to embolize and produce a PE. DVTs
that remain isolated distal to the popliteal vein (ie, distal DVT)
rarely cause PE. Other sources of PE include clots formed in the
veins of the upper extremities, clots formed in the RV, and clots
formed in response to indwelling venous catheters.
The pathogenesis of DVT is intimately related to the Virchow
triad: venous stasis, endothelial injury, and a procoagulant state.
Every VTE can be traced to 1 or more of these factors. Box 78.1
shows a classification of hypercoagulable states according to
whether they are inherited, acquired, or mixed-origin disorders.
Cancer and VTE
An association between venous thrombosis and occult malig-
nancy was observed by Trousseau in 1868. Active cancer is one
of the strongest risk factors for VTE, increasing the risk by 5- to
6-fold. Patients with active cancer account for about 20% of the
community burden of VTE. In addition, the presence of VTE
725
726 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
Box 78.1. Classification of Hypercoagulable States
Inherited
Loss of function
Antithrombin deficiency
Protein C deficiency
Protein S deficiency
Gain of function
Factor V Leiden
Prothrombin factor II G20210A
Elevated level of factor VIII, IX, or XI
Acquired
Previous venous thromboembolism
Obesity
Cancer
Pregnancy, puerperium
Drug-induced
Heparin-induced thrombocytopenia
Prothrombin complex concentrates
L-Asparaginase
Hormonal therapy
Postoperative
Myeloproliferative disorders
Mixed
Hyperhomocysteinemia
Previously published. See “Credit Lines” section.
independently portends a poorer prognosis from the underly-
ing cancer. Rates of UEDVTs are also increased independently
of the presence of peripherally inserted central catheter lines.
Similarly, rates of abdominal vein DVTs are increased in active
cancer (the Trousseau sign).
Cancer is sometimes diagnosed after a VTE episode. Incident
rates for cancer in the year after an unprovoked VTE are high
(7%–10%). However, the benefits of cancer screening for all
patients with an unprovoked VTE are uncertain. A single under-
powered randomized trial from Italy showed no mortality bene-
fit, but the issue is far from settled.
Thrombophilic Disorders and VTE
The search for a thrombophilic disorder is typically undertaken
in younger patients who have an idiopathic VTE. In about 25% of
patients with VTE, the VTE episodes are unprovoked; an identi-
fiable genetic predisposition to thrombosis (ie, hereditary throm-
bophilia) is identified in one-third to one-half of these patients.
In more than 90%, the identified defect is either a heterozygous
factor V Leiden or prothrombin factor II G20210A mutation.
Homozygous factor V Leiden or G20210A mutation or double
heterozygous (factor V Leiden/G20210A) mutations carry an
even higher risk. The risk of VTE is also higher with antithrom-
bin, protein C, and protein S deficiencies compared with hetero-
zygous factor V Leiden or prothrombin gene mutations.
VTE often occurs when another risk factor (eg, pregnancy,
oral contraceptive pills, trauma, surgery) is present in addition
to the underlying thrombophilia. Factor V Leiden mutation in
isolation is not a significant VTE risk factor for patients undergo-
ing hip or knee replacement surgery. Studies have shown a low
annual incidence of spontaneous DVT in asymptomatic carri-
ers of the factor V Leiden mutation. Management guidelines
for VTE in most thrombophilia situations are no different from
routine management guidelines. Thus, routine screening for
thrombophilia is not recommended and should be pursued only
if the results of such testing will alter anticoagulation decisions.
Testing for the patient (and for family members) should be indi-
vidualized according to the clinical situation and patient values
and preferences.
Pregnancy and VTE
Pregnant women have about a 5-fold increased risk of VTE when
compared with a matched cohort of nonpregnant women. In
about half of these patients, thrombophilia is identified. DVTs
are more common during pregnancy, whereas PEs seem to be
more common in the postpartum period. The left lower extrem-
ity accounts for 75% of DVTs; mechanical factors related to the
enlarging uterus may be partly responsible. Some studies sug-
gest an equal VTE risk in each trimester, whereas others suggest
a slightly increased VTE risk in the second and third trimes-
ters. The greatest risk of VTE is in the postpartum period, with
a roughly 10-fold increase in risk compared with the period of
pregnancy.
PE From Other Causes
PE can also result from embolization from other venous sites,
such as the upper extremities, neck veins, or right-sided cardiac
valves. In addition to blood clots, other foreign material such as
talc, drugs, tumors, and medical equipment (embolization coils)
can embolize to the pulmonary circulation.
Tumors can also involve the pulmonary vasculature through
embolization, lymphatic invasion, lymphangitic carcinomato-
sis, or a combination of these methods. Tumors such as hepa-
tocellular and renal cell carcinoma can invade the IVC and
subsequently dislodge to cause massive PE. Microscopic tumor
emboli leading to subacute or chronic cor pulmonale have been
described with gastric, ovarian, gall bladder, and breast cancers
among others.
Lower Extremity DVT
Although most cases (approximately 90%) of PE arise from
thrombi of the proximal lower extremity veins, among patients
with fatal PE, DVT has been clinically identified in only about
50% of cases. Approximately 45% of femoral and iliac vein
DVTs embolize to the lungs.
• Calf-only DVTs: 20% propagate to the thigh and iliac veins.
• Superficial thrombophlebitis cases: 10%–25% are complicated
by VTE.
• Femoral and iliac DVTs: approximately 50% embolize to the lungs.
• About 60% of patients with PE have DVT of the lower extremities.
Upper Extremity DVT
UEDVT accounted for 11% of all ultrasonographically confirmed
DVTs in a large cross-sectional study. UEDVTs are associated
with the presence of a CVC in about 50% of cases. The presence
of a CVC increases the odds of a UEDVT 7-fold. Cancer also
 78 Pulmonary Embolism
appears to be more prevalent in patients with a CVC-associated
UEDVT. PE appears to occur less frequently in UEDVT than in
lower extremity DVT.
The incidence of DVT without adequate prophylaxis in var-
ious clinical circumstances is as follows: major abdominal sur-
gery, 14% to 33%; thoracic surgery, 25% to 60%; post-MI, 20%
to 40%; congestive heart failure, 70%; stroke with paralysis,
50% to 70%; postpartum period, 3%; and trauma, 20% to 40%.
Diagnosis of DVT
Physical examination findings are helpful when positive for
DVT, but they should not be relied on to make the diagnosis of
DVT. The diagnosis of DVT requires a standardized and algo-
rithmic approach that incorporates clinical prediction models
(eg, the Wells Criteria) and high sensitivity D-dimer testing. The
Homans sign (pain and tenderness on dorsiflexion of the ankle)
and the Moses sign (pain on pressing the calf muscle) are classic
but infrequently seen signs that are present in less than half of
patients with proven DVT.
D-dimer is a degradation product released into the circulation
when cross-linked fibrin undergoes endogenous fibrinolysis. In
patients with a low clinical probability of DVT, a normal sen-
sitive D-dimer level effectively rules out DVT and may obviate
the need for any further testing. However, further testing is rec-
ommended for all patients with an intermediate or high clinical
probability of DVT, regardless of D-dimer test results. For those
with positive results on D-dimer tests and negative findings on
initial ultrasonographic evaluation, follow-up ultrasonography
in a week may be recommended. Subsequent ultrasonographic
examinations are not required for patients who have a clinically
suspected DVT and who have had negative findings on an initial
ultrasonographic examination and negative results with the sen-
sitive D-dimer test.
Duplex ultrasonography has become the first-line test for
DVT diagnosis, replacing contrast venography, which has long
been considered the reference test. Duplex ultrasonographic
imaging involves the use of both manual compression ultra-
sonography and Doppler assessment of blood flow to determine
vessel patency. Additional features such as increased vessel
size and wall thickness may also indicate the presence of a
thrombus.
Impedance plethysmography and contrast venography are
infrequently performed today because they have several limita-
tions. CT and MRI venography are also useful for DVT diagno-
sis, but they are performed infrequently because of the ease of
performing duplex ultrasonography.
• Diagnosis of DVT should follow a standardized algorithm.
• The incorporation of clinical prediction models and high sensitiv-
ity D-dimer testing is essential.
• Physical signs alone lack sensitivity and reliability for DVT
diagnosis.
• Duplex ultrasonography is the most reliable and most commonly
used test for DVT diagnosis.
Clinical Features of PE
PE has no pathognomonic clinical signs or symptoms. Hypoxia,
tachycardia, dyspnea, chest pain or discomfort, syncope, cough,
and fever are all signs and symptoms associated with PE. Patients
with PE can also be clinically asymptomatic. About 40% to 50%
of patients with DVT with no specific clinical signs or symptoms
727
of PE have positive results of studies for PE if imaging is done
during the acute DVT episode. The differential diagnosis of PE
is essentially a laundry list of cardiac, pulmonary, musculo-
skeletal, neurologic, and gastrointestinal tract diseases, ranging
from life-threatening diseases such as MI and aortic dissection
to more benign causes such as esophageal spasm and anxiety.
Working systematically through a broad differential diagnosis
will help avoid diagnostic errors.
Diagnosis of PE
The most important points to be made for the diagnosis of PE
are 1) consider the diagnosis and 2) choose the appropriate diag-
nostic test. The choice of test is based on the patient’s clinical
situation and the presence or absence of comorbidities such as
renal dysfunction or hemodynamic instability. It is also impor-
tant to realize that diagnostic tests (eg, echocardiography, CT)
can provide prognostic information and have been incorporated
into several prognostic models for acute PE. Figure 78.1 presents
a diagnostic cascade for use when evaluating a patient for possi-
ble PE. Commonly used diagnostic tests for PE are summarized
below.
CT Pulmonary Angiography
Multidetector row CTPA has become the first-line diagnostic
test for PE. Multidetector row scanners (64, 128, 256, or 320
rows) allow for ultrafast scanning (in a single breath hold) dur-
ing injection of contrast material into a peripheral or central vein
(Figures 78.2 and 78.3). In 10% to 15% of cases, CTPA also
uncovers other important thoracic and upper abdominal disease.
The sensitivity of CTPA is excellent for main, lobar, or segmen-
tal vessels, but sensitivity is less for subsegmental vessels , espe-
cially if there is motion artifact. Image quality is also affected by
the timing of the contrast bolus; early or delayed contrast injec-
tion can result in suboptimal scan quality. Image quality can also
be affected by factors such as the timing of the contrast bolus and
patient motion. A study has shown that negative CTPA findings,
in conjunction with a low-probability clinical diagnostic algo-
rithm and a negative D-dimer test result, can effectively rule out
PE without the use of compression ultrasonography.
Echocardiography
Echocardiography has both diagnostic and prognostic utility
in PE. It is especially useful in the hemodynamically unsta-
ble patient, in whom transport for CT may be hazardous.
Echocardiography can, in rare cases, provide direct confirmation
of PE by identifying a thrombus in transit through the right-sided
cardiac chambers, IVC, or proximal pulmonary arteries. In most
cases, it provides indirect evidence for PE by demonstrating RV
dilatation or systolic dysfunction with or without flattening of the
intraventricular septum resulting in a D-shaped left ventricle. A
reduction in RV free wall contractility with preserved RV apical
function (the McConnell sign) has also been found in patients
with acute PE, but the sign is relatively nonspecific and should
not be relied on to make or exclude the diagnosis. Prognostic
information can also be obtained by quantifying RV dilatation
by measuring the largest RV and LV dimensions and calculating
the RV/LV ratio. A value of 0.5 or 0.6 is compatible with RV
enlargement; larger values are associated with worse outcomes.
Echocardiography also allows for rapid identification of compet-
ing diagnoses such as cardiac tamponade, LV dysfunction, and
valvular problems.
728 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Suspected PE
New or worsening dyspnea, chest pain, or sustained
hypotension without another obvious cause

Clinical probability assessment

Hemodynamically stable Hemodynamically unstable

Low or intermediate High clinical Critically ill and high

clinical probability probability Not critically ill
clinical probability

Multidetector Multidetector
D-dimer testing
CT available CT not available

Multidetector Transthoracic or
Normal Elevated
CT transesophageal echocardiography

PE PE Right ventricular No right ventricular

Negative
ruled out confirmed dysfunction dysfunction

Search for
alternative diagnosis

Figure 78.1. Diagnostic Workup for Pulmonary Embolism (PE). The initial assessment of the clinical probability of PE is based on either clinical
judgment or clinical decision rules. Patients are considered to be hemodynamically unstable if they are in shock or have a systolic blood pressure of
less than 90 mm Hg or a decrease in pressure of more than 40 mm Hg for more than 15 minutes (in the absence of new-onset arrhythmia, hypovo-
lemia, and sepsis). If multidetector computed tomography (CT) is not available or if patients have renal failure or allergy to contrast dye, the use
of ventilation-perfusion scanning is an alternative. If patients have a high clinical probability and an elevated D-dimer level but negative findings
on multidetector CT, venous ultrasonography should be considered. Among critically ill patients with right ventricular dysfunction, thrombolysis is
an option; multidetector CT should be performed when the patient’s condition has been stabilized if doubts remain about clinical management. If
patients are candidates for percutaneous embolectomy, conventional pulmonary angiography can be performed to confirm the diagnosis of PE imme-
diately before the procedure, after the finding of right ventricular dysfunction. (Previously published. See “Credit Lines” section.)

Serum Markers: Troponin and D-Dimer

Both troponin T and troponin I are highly sensitive and specific
markers for myocardial cell injury. Elevated levels of troponin
are not specific for PE, but they are markers for increased short-
term morbidity and mortality from PE.
The assay for D-dimer, a specific fibrin degradation product,
is a very sensitive but relatively nonspecific test for VTE. Most
currently available D-dimer tests are quantitative and highly sen-
sitive and have replaced older semiquantitative tests. Advanced
age, pregnancy, trauma, cancer, critical illness, and prior DVT or
PE are all associated with increased D-dimer levels. The value
of D-dimer testing is in its high negative predictive value, espe-
cially for patients who have a low pretest probability of VTE.
For those patients, negative D-dimer test results may be enough
to confidently exclude VTE. In contrast, negative D-dimer test
results from patients who have an intermediate or high risk of
VTE does not exclude the diagnosis and further testing is war-
ranted. Thus it is very important to integrate both the clinical
probability of PE with available diagnostic tests to confidently
make or exclude a diagnosis of PE. It is important to consider
 78 Pulmonary Embolism
Figure 78.2. Ultrafast Computed Tomographic Scan. Large pul-
monary emboli are occluding major pulmonary arteries bilaterally
(arrows).
together the clinical probability of PE, the results of concomitant
imaging studies, and the levels of serum markers in a diagnostic
algorithm of PE.
• Most D-dimer tests in current use are quantitative with very high
sensitivity.
• A negative D-dimer test result for a patient with a low probability
of VTE effectively rules out the diagnosis.
Pulmonary Angiography
Pulmonary angiography is no longer used as a first-line diagnos-
tic test for PE and has been replaced by CTPA. Still used in some
diagnostic algorithms for evaluating patients for pulmonary
thromboendarterectomy, pulmonary angiography is an invasive
procedure with a mortality rate of 0.5% and a complication rate
Figure 78.3. Ultrafast Computed Tomographic Scan. The lobar and
segmental pulmonary arteries are occluded bilaterally (arrows).
729
of 1% to 2%. False-negative results occur in 1% to 2% of pul-
monary angiograms. In the PIOPED study, experts disagreed on
the presence of PE in 8% of patients and on the absence of PE in
17%, and they rated 3% of the angiograms as nondiagnostic.
• Rates of major and minor complications with pulmonary angiog-
raphy are 1%–2%.
• CTPA has replaced pulmonary angiography as the standard initial
test for PE diagnosis.
. .
V/Q Lung Scan
. .
The V/Q scan has also been replaced by CTPA as the front-line
test for PE. It is still useful, though, in situations such as contrast
dye allergy and renal dysfunction, .in.which contrast nephropathy
would be a serious concern. The V/Q scan is used to diagnose a
PE by detecting areas of the lung that are ventilated but not per-
fused (ie, a mismatch). Scans are categorized as high, interme-
diate, or low probability according to revised PIOPED . . criteria.
In the original PIOPED study, which compared V/Q scanning
with pulmonary
. . angiography, 13% of patients had high-proba-
bility V/Q scans, with a high specificity (97%) but low sensitivity
(41%) for PE diagnosis; 39% of patients had intermediate-proba-
bility scans. Of all the patients who underwent angiography, 33%
received a diagnosis of PE on the basis of the angiogram. A low-
probability lung scan (34% of scans) excluded the diagnosis of
PE in approximately 85% of patients and a normal/near normal
lung scan excluded
. . PE in nearly 95% of patients. Thus, the clini-
cal utility of V/Q scanning is limited by the fact that the majority
of scans are categorized as intermediate or low probability, and
additional testing is necessary to definitively diagnose PE.
. .
• A high-probability V/Q scan equates to a 97% probability of PE.
• A normal lung scan excludes PE in nearly 95% of patients.
• The majority of scans are clinically nondiagnostic and require fur-
ther testing.
Magnetic Resonance Imaging
A study on the use of MRI (MRA with MRV) for the diagnosis
of PE (PIOPED III study) was published in 2010. The routine use
of MRI faces several technical challenges. MRA images were
technically inadequate for PE diagnosis in 25% of patients in
this study. The subset of technically adequate MRA studies had
a sensitivity of 78% and a specificity of 99%, whereas the combi-
nation of adequate MRA and MRV had a sensitivity of 92% and
a specificity of 96%. Currently, the role of MRA and MRV in
the diagnosis of PE is limited owing to several factors, including
cost, access, scan time (approximately 30 minutes), claustropho-
bia, ferromagnetic implants, image quality, and concerns about
nephrogenic systemic fibrosis in patients with renal dysfunction.
Electrocardiography
The most common ECG abnormalities in patients with PE are
nonspecific changes (noted in 80% of patients), ST-segment and
T-wave changes (in 65%), the pattern of S wave in lead I and Q
wave in lead III (S1Q3 pattern) (in 15%), right bundle branch block
(in 12%), and left-axis deviation (in 12%). Nonspecific T-wave
inversion in the precordial leads is commonly seen and is the
ECG sign best correlated with the severity of PE. ECG changes
by themselves are relatively nonspecific, but in the appropriate
context, they can add confidence to the clinical diagnosis of PE.
• The classic pattern of S wave in lead I and Q wave in lead III (S1Q3
pattern) is seen in only 15% of patients with PE.
730 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

• T-wave inversion in the precordial leads is the ECG sign best cor-
related with the severity of PE.
Chest Radiography
The chest radiograph is frequently abnormal in patients with PE,
but the findings are relatively nonspecific. Classic signs such as
Westermark sign (focal oligemia) and Hampton hump (peripheral
wedge-shaped opacity adjacent to the diaphragm) are infrequent.
Pleural effusions, cardiomegaly, pulmonary artery enlargement,
atelectasis or infiltrates and diaphgramatic elevation have also
been reported in the acute setting.
Arterial Blood Gases
Frequently, there is no relation between oxygenation and clot
burden and a normal arterial blood gas does not exclude the diag-
nosis of PE. Indeed, in the PIOPED study, about 20% of patients
with angiographically documented PE had a normal PAO2-PAO2
(≤20 mm Hg). Many patients tend to show some degree of
hypocapnia due to hyperventilation resulting from dead space.
Prognostication in PE
Morbidity and mortality from PE depend not only on the clot
burden but also on the presence and extent of preexisting car-
diopulmonary disease and the response of the cardiopulmonary
system to the presence of clot in the pulmonary vasculature. This
understanding has led to a robust body of literature on prognos-
tication in PE. Several scores have been developed and validated
to classify patients on a risk continuum. These scores include the
PESI (Table 78.1), the Geneva rule, and the PREP study score.
Patients classified as low risk have the potential to be safely man-
aged as outpatients without being hospitalized.
Treatment of PE
When PE is diagnosed, optimal management requires imme-
diate and continued assessment of the patient’s hemodynamic
status and other clinical parameters. As mentioned earlier,
select patients with a low clot burden and low risk stratifica-
tion may be safely managed through outpatient anticoagula-
tion with the use of LMWH and a period of warfarin overlap.
This approach is currently being validated in controlled clin-
ical trials.
The majority of patients, however, require inpatient manage-
ment with immediate initiation of anticoagulation with either
UFH or LMWH. Warfarin therapy can be initiated simulta-
neously if instability, bleeding, or invasive procedures are not
planned. An overlap period with at least 2 days of a therapeu-
tic INR (INR, 2–3) is required before the use of heparin (UFH
or LMWH) may be discontinued. Anticoagulation management
of PE and DVT is similar. UFH (80 U/kg) is administered as a
bolus, followed by a maintenance dose (18 U/kg per hour intra-
venously). The dose should be adjusted to maintain an aPTT
greater than 1.5 times the control value. In a large meta-analysis,

Table 78.1. Baseline Patient Characteristics and Risk Class Distributiona,b

All Samples Perrier et al, Perrier et al, Perrier et al, Roy et al,
Patient Characteristics (N = 899) 2001c (N = 117) 2004d (N = 221) 2005e (N = 194) 2006f (N = 367) P Value
Age >65 y 576 (64) 74 (63) 131 (59) 119 (61) 252 (69) .10
Male sex 382 (42) 48 (41) 100 (45) 83 (43) 141 (38) .13
Cancer 109 (12) 18 (15) 35 (16) 23 (12) 33 (9) .06
Heart failure 134 (15) 41 (35) g 26 (12) 10 (5) 57 (16) <.001
Chronic lung disease 75 (8) 9 (8) 19 (9) 18 (9) 29 (8) .94
Temperature <36°C 24 (3) 2 (2) 8 (4) 2 (1) 12 (3) .29
Pulse ≥110 beats per minute 173 (19) 22 (19) 31 (14) 27 (14) 93 (25) .001
Systolic blood pressure 25 (3) 1 (1) 4 (2) 7 (4) 13 (4) .32
<100 mm Hg
Respiratory rate ≥30 breaths 112 (12) 15 (13) 24 (11) 21 (11) 52 (14) .42
per minute
Altered mental statush 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) ...
Arterial oxygen saturation <90% 132 (15) 9 (8) i 13 (6) 41 (21) 69 (19) <.001
Risk class distribution
Class I 184 (20) 20 (17) 54 (24) 44 (23) 66 (18) .18
Class II 242 (27) 29 (25) 60 (27) 55 (28) 98 (27) .92
Classes I and II combined 426 (47) 49 (42) 114 (52) 99 (51) 164 (45) .17
Class III 248 (28) 34 (29) 62 (28) 56 (29) 96 (26) .88
Class IV 132 (15) 19 (16) 25 (11) 26 (13) 62 (17) .27
Class V 93 (10) 15 (13) 20 (9) 13 (7) 45 (12) .15
a
Overall, 0.1% of patients had unknown values for heart failure and chronic lung disease, 7.6% for temperature, 0.6% for pulse, 1.7% for systolic blood
pressure, 10.3% for respiratory rate, and 18.2% for arterial oxygen saturation. For calculating the frequency of baseline patient characteristics, unknown
values were assumed to be normal and were included in the denominator of all samples.
b
Data are presented as number (percentage) of patients.
c
Perrier et al. Ann Intern Med. 2001 Jul 17;135(2):88–97.
d
Perrier et al. Am J Med. 2004 Mar 1;116(5):291–9.
e
Perrier et al. N Engl J Med. 2005 Apr 28;352(17):1760–8.
f
Roy et al. Ann Intern Med. 2006 Feb 7;144(3):157–64.
g
Because information about heart failure was not available, the presence of any heart disease was used as a surrogate marker in this study sample.
h
Because information about mental status was not available, mental status was assumed to be normal in all patients.
i
Because oxygen saturation was not available in this study sample, we used a partial oxygen pressure of less than 60 mm Hg as a surrogate marker.
Previously published. See “Credit Lines” section.
 78 Pulmonary Embolism
fixed-dose subcutaneous LMWH was as effective and safe
as dose-adjusted intravenous heparin for the treatment of PE.
Caution should be used with LMWH in certain clinical situa-
tions, such as pregnancy, severe obesity (body mass index >50,
calculated as weight in kilograms divided by height in meters
squared), and severe renal failure; plasma anti-Xa levels can be
monitored in these situations if required.
• In acute DVT and PE, treatment with heparin and warfarin can be
initiated simultaneously.
• Heparin dosage: bolus, 80 U/kg; maintenance, 18 U/kg per hour.
• LMWH is preferred over UFH in the treatment of nonmassive PE
owing to ease of use.
The current American College of Chest Physicians treatment
guidelines for duration of treatment are as follows:
• Patients with the first episode of DVT due to a transient risk factor:
3 months.
• Patients with the first episode of idiopathic DVT: 3 months, with
subsequent reassessment for long-term anticoagulation according to
bleeding risk.
• Patients with DVT and cancer: LMWH for 3–6 months followed by
warfarin or continued LMWH.
• Patients with recurrent DVT: indefinite treatment.
Bleeding Complications
Patients receiving long-term warfarin therapy tend to have fewer
complications when they are cared for by an experienced antico-
agulation clinic. In spite of best practices, major bleeding rates
can exceed 1% at 1 year and 2% at 3 years. Presence of malig-
nant disease at initiation of warfarin therapy is associated with
a risk of major hemorrhage. Among patients with PE treated
with thrombolytic agents, the risk of intracranial hemorrhage is
about 1%.
Drugs that prolong the effect of warfarin include salicylates,
amiodarone, heparin, estrogen, antibiotics, clofibrate, quini-
dine, and cimetidine. Drugs that decrease the effect of warfarin
include rifampin and barbiturates. This is only a partial list of
drugs that interfere with warfarin metabolism, and it is wise to
review possible interactions when a new drug is administered to
a patient receiving warfarin.
Newer Agents: Direct Thrombin and
Factor Xa Inhibitors
Several new anticoagulant agents are now being assessed for
use in VTE treatment and prophylaxis. The 2 major classes of
drugs under investigation are factor Xa inhibitors (eg, rivar-
oxaban, apixaban) and DTIs (eg, bivalent: bivalirudin, lep-
irudin ; univalent: argatroban, dabigatran). These agents have
already shown efficacy for management of acute coronary syn-
drome, VTE prophylaxis, and thromboprophylaxis in atrial
fibrillation. Lepirudin and argatroban are currently used for
anticoagulation in heparin-induced thrombocytopenia, but cur-
rently no medications are approved by the US Food and Drug
Administration for outpatient anticoagulation management of
VTE. This situation will change soon with emerging data about
the efficacy and safety of these agents for management of acute
VTE. The advantages of these agents are convenient oral dos-
ing, no interaction with food, and lack of need for monitoring,
as compared with warfarin. The disadvantages are cost and
the lack of quick and easy reversibility in the event of a major
bleeding episode.
731
IVC Interruption
IVC plication is largely of historical interest and is no longer
routinely performed. An IVC filter allows for continued blood
flow through the IVC while blocking the transmigration of large
emboli. Although there are no evidence-based guidelines for
their use, these devices are often used after a major PE when
continued anticoagulation is either contraindicated or likely to be
frequently interrupted for tests or procedures. Another common
indication is after a major PE if the patient will likely be unable
to sustain another embolic event. The routine use of an IVC fil-
ter in patients with cancer and DVT or PE is not recommended.
The PREPIC study showed no mortality benefit from IVC filter
use but did show decreased risk of PE along with an increased
risk of DVT. IVC filters can be either permanent or temporary.
Temporary retrievable IVC filters have become very popular and
allow for immediate PE risk mitigation and avoidance of long-
term filter-related complications. Most IVC filters are inserted
through the internal jugular vein. Multiple types of filters are
available in the United States (Figure 78.4).
• Recurrent PE can occur even after filter insertion.
• IVC filter placement did not change mortality in the PREPIC trial.
• IVC filters can be either permanent or temporary (retrievable).
• IVC filters reduce the risk of recurrent PE while increasing the risk
of recurrent DVT.
Massive PE
The term massive PE does not describe the extent of the clot bur-
den or the presence of a saddle embolus. Rather, it refers to a PE
that causes hemodynamic instability (ie, shock or tissue hypoper-
fusion). Most patients with massive PE, however, typically have
embolus in the proximal pulmonary arteries (Figure 78.5). Both
massive PE and submassive PE (discussed below) show evidence
of RV strain, but hemodynamic instability and organ hypoperfu-
sion occur only with massive PE. Whether a PE will be hemody-
namically significant depends both on the actual clot burden and
on the extent of underlying cardiopulmonary disease. Massive
PE results in obstructive cardiogenic shock with RV failure, tis-
sue hypoperfusion, and death if untreated (Figure 78.6). Massive
PE is a rapidly fatal condition that accounted for 4% to 5% of PEs
in the ICOPER. Mortality is approximately 10% within 1 hour
and up to 85% in the first 6 hours. In the appropriate setting,
Figure 78.4. Greenfield Inferior Vena Caval Filter With Occluding
Thrombus After Pelvic Irradiation for Uterine Cancer.
732 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

A B

Figure 78.5. Massive Pulmonary Embolism in a 42-Year-Old Woman. A, Reformatted coronary non-electrocardiographic (ECG)-gated contrast-
enhanced computed tomographic (CT) scan at level of carina depicts central bilateral pulmonary emboli. B, Reformatted 4-chamber non-ECG-
gated contrast-enhanced CT scan shows dilatation of right ventricle (RV) and septal bowing, with paradoxical concavity toward RV (arrowheads).
C, Reformatted short-axis non-ECG-gated contrast-enhanced CT scan shows severe septal bowing (arrowheads). (Previously published. See “Credit
Lines” section.)

the sudden onset of severe dyspnea, hypoxia, and hemodynamic
collapse with or without syncope should raise the possibility of
massive PE. If available, the echocardiogram is the most useful
and immediate test for massive PE. It shows a dilated RV that is
under severe strain and has poor systolic function. As expected,
the LV is often hyperdynamic and underfilled and may appear
D-shaped due to septal shift as a result of high RV pressures”
(Figure 78.5). The echocardiogram also allows for rapid simulta-
neous assessment of valvular and pericardial pathology in clin-
ically unstable patients. The ECG often shows tachycardia, an
S1Q3 pattern, or a new partial or total right bundle branch block.
Right atrial pressure is acutely increased and may result in open-
ing of a patent foramen ovale, with a resultant right-to-left shunt
and worsening hypoxia. A rare but serious complication is par-
adoxical embolization into the systemic circulation through the
patent foramen ovale.
The treatment of choice, in the absence of contraindications,
is rapid institution of thrombolysis. Catheter-directed or surgi-
cal thrombectomy is a well-described option if thrombolysis is
absolutely contraindicated.
The use of intravenous fluids and pressors alone rarely fully
reverses the shock state. Heparinization should be started as
quickly as possible, stopped during thrombolysis, and resumed
thereafter.
In desperate cases (eg, ongoing cardiopulmonary resusci-
tation, profound shock) in which PE is strongly suspected but
impossible to confirm, thrombolysis should be administered
if the clinical and echocardiographic data fit the diagnosis.
Alteplase (total dose 100 mg over 2 hours, with or without an ini-
tial loading dose of 10–15 mg over 1–2 minutes) is approved by
the US Food and Drug Administration for this indication. Other
thrombolytic agents are streptokinase (loading dose of 250,000
international units infused over 30 minutes, followed by a main-
tenance dose of 100,000 U hourly for up to 24 hours) and uro-
kinase (loading dose of 4,400 IU/kg, infused over 10 minutes,
followed by continuous infusion of 4,400 IU/kg per hour for
12–24 hours). Heparin infusion is begun or resumed if aPTT is
less than 1.5 times the upper limit of normal after thrombolytic
therapy. The contraindications to thrombolytic therapy are listed
in Box 78.2.
 78 Pulmonary Embolism
Figure 78.6. Fatal Massive Pulmonary Embolus Occupying the
Right and Left Pulmonary Trunks. The pulmonary valve and pulmo-
nary outflow tract are seen inferiorly, below the right pulmonary artery
(long arrow) and the left pulmonary artery (short arrow).
Box 78.2. Contraindications for Thrombolytic
Therapy in Venous Thromboembolism
Absolute
Intracranial disorders (neoplasms, vascular
accidents, hemorrhagic stroke, aneurysm)
Intracranial or intraspinal surgery or trauma within
the preceding 2 mo
Operation, obstetric delivery, or organ biopsy within
the preceding 10 d
Active internal bleeding within the preceding 6 mo
Bleeding diathesis
Severe hypertension (systolic blood pressure
>200 mm Hg or diastolic blood pressure
>110 mm Hg)
Recent trauma (including cardiopulmonary
resuscitation)
Relative
Pregnancy
Infective endocarditis
Pericarditis
Diabetic hemorrhagic retinopathy or other
hemorrhagic ophthalmic conditions
Previously published. See “Credit Lines” section.
733
Submassive PE
Patients with submassive PE show evidence of RV strain or tro-
ponin leak (or both) but do not have hemodynamic instability
or tissue hypoperfusion. These patients have increased mortality
and morbidity, with in-hospital mortality rates of 2% to 10%.
The role of thrombolysis in submassive PE is controversial and
cannot be routinely recommended at this time.
• Massive PE results in obstructive cardiogenic shock with acute RV
failure and tissue. hypoperfusion
• Mortality in massive PE: 10% within 1 hour and up to 85% in
6 hours.
• In the absence of contraindications, thrombolytic agents should
be administered without delay in hemodynamically significant
(massive) PE.
• Heparin therapy is necessary after thrombolytic therapy.
• Bedside echocardiography should be used as soon as the diagnosis
is suspected.
Chronic Thromboembolic Pulmonary
Hypertension
CTEPH is probably more common than previously thought and
occurs in about 3% of patients who have PE. Many patients with
the initial diagnosis of idiopathic pulmonary hypertension are
subsequently found to have CTEPH. Factors that increase the
risk of CTEPH after PE include recurrent VTE, younger age,
a larger perfusion defect, and idiopathic PE at presentation. In
a subset of patients, CTEPH develops even though recurrent PE
cannot be documented. In some patients, multiple emboli may
not be the cause of CTEPH, but rather the culprit may be recur-
rent in situ thrombosis after an initial PE.
Less than 50% of patients with CTEPH have a history com-
patible with a previous episode of DVT or PE. Lupus anticoag-
ulant is present in about 10% of patients with CTEPH, and 1%
have deficiencies of antithrombin III, protein C, or protein S.
CTEPH develops in many patients without an underlying coagu-
lopathy because of a delayed or missed diagnosis of PE or inad-
equate anticoagulation for previously documented DVT or PE.
Lung biopsy specimens from patients with CTEPH show plex-
iform lesions, suggesting that factors other than obstructive PE
may be important in the onset of CTEPH. Indeed, the degree
of angiographic obstruction correlates poorly with the degree of
pulmonary hypertension.
• Recurrent PE leads to development of CTEPH in about 3% of
patients.
• Chronic recurrent in situ thrombosis initiated by the first PE may
be responsible for many cases of CTEPH.
• Plexiform histologic lesions are found at lung biopsy in CTEPH.
Many patients remain asymptomatic—in a “honeymoon”
period—despite extensive PE. When symptoms develop, they
are similar to those of idiopathic pulmonary hypertension.
An uncommon clinical finding is the presence of low-pitched
flow murmurs or bruits, systolic or diastolic, in areas of partial
branch pulmonary artery stenosis caused by intra-arterial clots
and fibrotic bands, which are present in up to 20% of patients.
Chest radiographic findings are usually unremarkable. The dif-
fusing capacity of the lung for carbon monoxide may either be
normal or decreased; a normal value does not exclude CTEPH.
A mild to moderate restrictive defect is noted in 20% of patients.
The PAO2-PAO2 is usually widened, and PAO2 decreases with exer-
cise in most patients. In almost all patients with CTEPH who
734 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
. .
undergo V/Q scanning, at least 1 segmental or larger perfusion CT computed tomography
defect (with corresponding ventilation mismatch) is observed. CTEPH chronic thromboembolic pulmonary hyper tension
. . CTPA computed tomographic pulmonary angiography
However, the V/Q scan often underestimates the extent of central
pulmonary vascular obstruction. CVC central venous catheter
DTI direct thrombin inhibitor
• In many patients, a “honeymoon” (asymptomatic) period occurs DVT deep vein thrombosis
despite extensive PE. ECG electrocardiographic
INR international normalized ratio
• Low-pitched flow murmurs or bruits, systolic or diastolic, are IVC inferior vena cava
heard over lung fields in up to 20% of patients. LMWH low-molecular-weight heparin
PTEA is the treatment of choice in symptomatic patients, LV left ventricle
whether symptomatic at rest or during exercise. The mean pul- MI myocardial infarction
MRA magnetic resonance angiography
monary vascular resistance in patients undergoing surgery is 800
MRI magnetic resonance imaging
to 1,000 dynes · s · cm −5. Selection criteria for PTEA include MRV magnetic resonance venography
thrombi that are surgically accessible, acceptance of surgical PAO2-PAO2 alveolar-arterial gradient in partial pressure of
risk by patient, acceptable patient age and comorbidity, and oxygen
the absence of severe underlying lung disease, either obstruc- PE pulmonary embolism
tive or restrictive. However, none of these criteria are absolute. PESI Pulmonary Embolism Severity Index
Complications of PTEA include reperfusion pulmonary edema, PTEA pulmonary thromboendarterectomy
RV failure due to residual pulmonary hypertension, aortic dis- RV right ventricle
section, cerebrovascular accident, mediastinal hemorrhage, UEDVT upper extremity deep vein thrombosis
and intraoperative cardiac arrest. Short-term mortality is less UFH
. . unfractionated heparin
V/Q ventilation-perfusion
than 10% in well-selected cases, and perioperative mortality
VTE venous thromboembolism
is approximately 5%. Although a small number of patients do
not show an immediate positive response to PTEA, sympto-
matic improvement may occur over a prolonged period of up to Names of Clinical Trials
12 months. Long-term anticoagulation and IVC filter placement
are indicated for all patients who undergo PTEA. Lung trans- EINSTEIN-DVT Oral Rivaroxaban Versus Standard Therapy
plant may be considered for patients who are not candidates for in the Initial Treatment of Symptomatic Deep
Vein Thrombosis and Long-Term Prevention of
PTEA or for those who have inadequate results.
Recurrent Venous Thromboembolism
• PTEA: overall mortality, 5%–10%; perioperative mortality, 5%. ICOPER International Cooperative Pulmonary Embolism
Registry
• PTEA selection criteria: thrombi accessible through a surgical PIOPED Prospective Investigation of Pulmonary Embolism
approach, acceptance of surgical risk by patient, and acceptable Diagnosis
age and comorbidities, including coexisting lung disease. PIOPED III Prospective Investigation of Pulmonary Embolism
• Lifelong anticoagulation is indicated after PTEA. Diagnosis III
PREP Prognostic Factors for Pulmonary Embolism
PREPIC A Clinical Trial of Vena Caval Filters in the
Abbreviations Prevention of Pulmonary Embolism in Patients
aPTT activated partial thromboplastin time With Proximal Deep Vein Thrombosis

Pulmonary Hypertension
RACHEL J. LE, MD and GARVAN C. KANE, MD, PHD
Definition
The normal pulmonary circulation has low-pressure, low-resis-
tance flow through compliant, thin-walled vessels. PH is defined
as an mPAP greater than 25 mm Hg at rest. Clinically significant
PH is usually associated with substantially higher PA pressures.
Although a suspicion of PH can be based on echocardiographic
estimates, the diagnosis cannot be established without right heart
catheterization.
The pressures used to define PH are based on a general con-
sensus that this degree of PA pressure elevation is usually clin-
ically significant. Similarly, an mPAP greater than 30 mm Hg
with exercise is considered abnormal, but there has not been a
rigorous assessment of its clinical implications. The historical
diagnosis of “exercise-induced PH” has even been removed from
current classifications of the disease. Before determining that PH
is present, one should consider that a PA systolic pressure greater
than 40 mm Hg (when measured by echocardiography) is present
in 6% of otherwise healthy persons older than 50 years and in
5% of persons, regardless of age, with a body mass index greater
than 30 (calculated as weight in kilograms divided by height
in meters squared). Athletic men also have a higher resting PA
systolic pressure.
The following equation relates PA pressure to pulmonary
venous pressure (PV), PA blood flow (Qp), and pulmonary
vascular resistance (Rp):
Qp (L/min per m 2 ) = PA − PV (mm Hg)
Rp (units/m 2 )
Abbreviations and acronyms are expanded at the end of this chapter.
79
Symptoms and Signs
More than 90% of patients with PH present with dyspnea, a
defining feature at diagnosis. Approximately 20% of patients
have syncope during the course of their disease, 20% to 30%
have chest pain, and 30% to 50% have lower extremity edema.
The cause of syncope in PH is not clearly defined, but it seems
to be related to a decrease in right ventricular cardiac output
rather than to a primary arrhythmic mechanism. Angina in PH
may be the result of several mechanisms, including elevated right
ventricular myocardial oxygen demand due to high wall stress,
decreased systolic blood flow in the coronary branches to the
right ventricle, and compression of the left main coronary artery
by the dilated central pulmonary artery. Concomitant atheroscle-
rotic disease should not be forgotten when assessing for the cause
of angina in PH patients.
The clinical signs that indicate the presence and severity of
PH are highlighted in Tables 79.1 through 79.3.
Diagnostic Classification
While PH is defined as the presence of elevated pulmonary pres-
sure, regardless of mechanism, the type of PH is classified into
groups on the basis of underlying etiologic and pathophysiologic
factors (Box 79.1). The clinical classification groups guide the
clinician in determining prognosis and management.
Group 1 PH: PAH
PAH describes a group of various pulmonary hypertensive
diseases that have similar histopathology, clinical manifesta-
tions, and responses to pulmonary vascular–targeted therapy.
These include idiopathic and heritable causes, PH related to
735
736 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Table 79.1. Physical Signs That Indicate Clinically Significant
Pulmonary Hypertension
Sign
Accentuated pulmonary component
of second heart sound (audible
at apex in >90% of patients)
Early systolic click
Midsystolic ejection murmur
Left parasternal lift
Right ventricular fourth heart
sound (in 38% of patients)
} High right ventricular pressure and
Implication
High pulmonary pressure increases
force of pulmonary valve closure
Sudden interruption of opening of
pulmonary valve into high-
pressure artery
Turbulent transvalvular pulmonary
outflow
hypertrophy
Table 79.3. Physical Signs That Indicate Possible Underlying
Cause or Associations of Pulmonary Hypertension
Sign Implication
Central cyanosis Hypoxemia, right-to-left shunt
Clubbing Congenital heart disease, pulmonary
venopathy
Cardiac auscultatory findings, Congenital or acquired heart or
including systolic murmurs, valvular disease
diastolic murmurs, opening
snap, and gallop
Rales, dullness, or decreased Pulmonary congestion or effusion
breath sounds
Fine rales, accessory muscle
use, wheezing, protracted
(or both)
Pulmonary parenchymal disease

Increased jugular a wave High right ventricular filling
pressure
systemic-to-pulmonary shunts, connective tissue diseases, por-
tal hypertension, HIV infection, exposure to certain drugs or
dietary products, and persistent PH of the newborn. In addi-
tion, pulmonary venoocclusive disease and pulmonary capillary
hemangiomatosis are included in the category of PAH because
they also are characterized by arteriopathy and by similar risk
factors and possibly genetic substrates.
Idiopathic
PH of undetermined cause was previously referred to as IPAH. It
is a rare disease, with a prevalence of less than 0.2%. It is more
common in women than in men, the female preponderance is
greatest among blacks, and the mean age at onset is 35 years.
The clinical course of untreated IPAH is generally an inex-
orable progression toward death. Among patients who do not
undergo treatment with an effective pulmonary vascular–tar-
geted agent or lung transplant, survival is 68% to 77% at 1 year
and 22% to 38% at 5 years.
• IPAH should be diagnosed only when all other potential causes of
PAH have been excluded.
Table 79.2. Physical Signs That Indicate Severity of Pul-
monary Hypertension
Sign Implication
Moderate to Severe Pulmonary Hypertension
Holosystolic murmur that increases
with inspiration
Increased jugular v wave
Pulsatile liver
Diastolic murmur
}
Tricuspid regurgitation
expiration, productive cough
Obesity, kyphoscoliosis, enlarged Risk factors for disordered ventilation
tonsils
Sclerodactyly, arthritis, rash Connective tissue disorder
Peripheral venous insufficiency or Possible venous thrombosis
obstruction
Previously published. See “Credit Lines” section.
Heritable
HPAH shares the clinical phenotype of IPAH but includes
patients with a family history of PAH and patients who may
have de novo mutations with heritable risk. The primary form
of HPAH is a result of a mutation in the BMPR2 gene, present
in 70% of HPAH patients. The BMPR2 gene is on chromosome
2q31–32 and codes for bone morphogenetic protein receptor 2,
a member of the transforming growth factor β superfamily of
signaling molecules. BMPR2 gene mutation predisposes a person
to increased proliferation and decreased apoptosis of vascular
smooth muscle cells, thereby promoting constrictive pulmonary
vascular lesions. A pattern of autosomal dominant inheritance
has been observed in families with 2 or more members having
PAH and no other underlying or associated condition.
The phenotypic expression of the genetic abnormality is var-
iable and incomplete, with disease penetrance for all known
BMPR2 mutations of 15% to 20% in most families. Persons
known to have the genetic mutation may not have PH, although
they may still transmit the disease to their offspring. Thus, the
siblings or children of HPAH patients have an overall 50% risk
of inheriting the gene; with 20% penetrance, their risk of clinical
disease is 10%. There is a female preponderance and a tendency
for HPAH to develop at earlier ages in subsequent generations
within a family (genetic anticipation). Between 10% and 40% of

Pulmonary regurgitation
patients with apparently sporadic IPAH have the BMPR2 genetic
mutation and should be considered to have HPAH. Likewise,
about 30% of families with PAH do not have mutations in the
BMPR2 gene. Compared with patients who have IPAH, the
Hepatojugular reflux High central venous pressure patients who have HPAH perhaps have more severe disease and
Advanced Pulmonary Hypertension With Right Ventricular Failure a smaller chance of having vasoreactivity.
Right ventricular third heart sound Right ventricular dysfunction Although the genetic risk sets the stage for the development of
(in 23% of patients) PAH, a second environmental or coexisting genetic condition is
Marked distention of jugular veins generally thought to trigger the clinical expression of the disease.
Hepatomegaly
Peripheral edema (in 32% of
patients)
Ascites
Low blood pressure, diminished
}Right ventricular dysfunction or
tricuspid regurgitation (or both)

Reduced cardiac output, peripheral

pulse pressure, cool extremities
Mutations of the BMPR2 gene have been detected in patients
with PAH associated with fenfluramine derivatives but not in
patients with PAH in the scleroderma spectrum of disease or
with HIV-associated PAH. Mutation of the ALK1 receptor gene
vasoconstriction
(endoglin) confers susceptibility to PH in some patients with or
without hereditary hemorrhagic telangiectasia. These appear to
 79 Pulmonary Hypertension 737

be a less common cause of HPAH. ALK1 is also a member of the

Box 79.1. Updated Clinical Classification of Pulmo- transforming growth factor β family.
nary Hypertension (Dana Point Classification, 2008) a
• Genetic testing should be offered to patients with IPAH, particu-
1. Pulmonary arterial hypertension larly those with a family history of PAH and to relatives of patients
1.1. Idiopathic with HPAH.
1.2. Heritable
1.2.1. BMPR2 Associated With Connective Tissue Disease
1.2.2. ALK1, endoglin (with or without hereditary
hemorrhagic telangiectasia) PAH has been reported to occur with all the connective tissue
1.2.3. Unknown diseases but most frequently (15%–20%) with systemic scle-
1.3. Drug- and toxin-induced
rosis (typically limited scleroderma or the CREST syndrome).
PAH is the cause of death in up to 50% of patients who die
1.4. Associated with
of scleroderma-related complications. Isolated PAH is rela-
1.4.1. Connective tissue diseases
tively uncommon in diffuse scleroderma; it most often occurs
1.4.2. HIV infection
in association with the antinucleolar antibody anti-U3 ribonu-
1.4.3. Portal hypertension
cleoprotein. PAH is especially associated with the presence of
1.4.4. Congenital heart diseases autoantibodies (including anticentromere antibodies and anti-
1.4.5. Schistosomiasis nucleolar antibodies) in patients with long-standing limited
1.4.6. Chronic hemolytic anemia scleroderma.
1.5. Persistent pulmonary hypertension of the PAH is the most common cause of death in mixed connec-
newborn tive tissue disease (an overlap syndrome with features of scle-
1′. Pulmonary venoocclusive disease or pulmonary roderma, SLE, and polymyositis associated with the anti-U1
capillary hemangiomatosis (or both) ribonucleoprotein antibody). PAH is less frequent in SLE, rheu-
2. Pulmonary hypertension due to left heart matoid arthritis, and polymyositis. Anticardiolipin antibodies
disease are associated with PAH in up to 68% of patients with SLE.
2.1. Systolic dysfunction Non-PAH PH also occur in connective tissue diseases as a result
of fibrotic lung disease.
2.2. Diastolic dysfunction
2.3. Valvular disease • All patients with limited scleroderma should have periodic echo-
cardiographic screening for PH.
3. Pulmonary hypertension due to lung disease or
hypoxia (or both) • All patients with unexplained PAH should have clinical and auto-
antibody testing for connective tissue disease.
3.1. Chronic obstructive pulmonary disease
3.2. Interstitial lung disease
3.3. Other pulmonary diseases with mixed restrictive Associated With HIV Infection
and obstructive pattern Patients with HIV infection have an increased risk of PAH,
3.4. Sleep-disordered breathing which may manifest at any viral load. The overall prevalence of
3.5. Alveolar hypoventilation disorders PAH in HIV has been reported to be as high as 1:200. The clin-
3.6. Chronic exposure to high altitude ical, hemodynamic, and histologic characteristics are similar to
3.7. Developmental abnormalities those of IPAH. The mechanism remains unclear. Interestingly,
compared with IPAH, HIV-associated PAH is highly responsive
4. Chronic thromboembolic pulmonary to PAH-specific therapy.
hypertension
• All patients with unexplained PAH should be tested for HIV
5. Pulmonary hypertension with unclear
infection.
multifactorial mechanisms
5.1. Hematologic disorders: myeloproliferative
disorders, splenectomy Associated With Toxic Exposure
5.2. Systemic disorders: sarcoidosis, PAH has been associated with the ingestion of appetite suppres-
pulmonary Langerhans cell histiocytosis: sants (which are molecularly similar to amphetamine), includ-
lymphangioleiomyomatosis, neurofibromatosis, ing aminorex and derivatives of fenfluramine. Dexfenfluramine
vasculitis increases the likelihood of PAH developing by a factor of
5.3. Metabolic disorders: glycogen storage disease, more than 20 times in patients who took the drug longer than
Gaucher disease, thyroid disorders 3 months. PAH can also occur in users of illicit methampheta-
5.4. Others: tumoral obstruction, fibrosing mine and is an increasing health problem with widening abuse
mediastinitis, chronic renal failure on dialysis of the drug. Other toxic associations include contaminated rape-
seed oil (an epidemic affecting 20,000 people occurred in Spain
Abbreviations: ALK1, activin receptor-like kinase type 1; BMPR2, bone in 1981; 2.5% had clinically significant PAH and 20% died).
morphogenetic protein receptor type 2; HIV, human immunodeficiency Contaminated tryptophan has caused eosinophilic myalgia syn-
virus. drome; PAH was a severe complication in some patients. Other
a
Main modifications to the previous Venice classification are possible causes, which require additional study to define, include
underlined. cocaine, phenylpropanolamine-containing diet drugs, St John’s
Previously published. See “Credit Lines” section. wort, chemotherapeutic agents, and selective serotonin reuptake
inhibitors. Although previously considered to be contributing
738 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
factors, estrogens, oral contraceptives, and cigarette smoke are
unlikely to be associated with PAH development.
Systemic-Pulmonary Shunts
Congenital systemic-to-pulmonary (left-to-right) shunts initially
lead to a period of high pulmonary flow but low pulmonary
resistance. Over time, however, the high vascular shear stress
related to elevated blood flow leads to endothelial damage and
irreversible pulmonary vascular remodeling. Eisenmenger syn-
drome is an irreversible state of PAH that is mediated through
arterial shunts causing pulmonary vasculopathy and ultimately
resulting in a right-to-left or bidirectional shunt.
The likelihood of PAH developing usually depends on the site
and severity of the defect (Box 79.2). A VSD will cause PAH
more commonly than will an ASD, which will cause PAH more
commonly than will a PDA. In VSD or PDA, Eisenmenger syn-
drome tends to develop earlier than in ASD; in complex anoma-
lies, such as atrioventricular septal defects or truncus arteriosus,
PAH tends to develop early and be present before the right-to-left
shunt characteristic of Eisenmenger syndrome. Rarely, PAH may
develop even after the defect is corrected.
PAH may also occur in other forms of left-right shunts,
including large peripheral arteriovenous malformations and even
iatrogenic arteriovenous fistulae placed for dialysis.
Compared with a patient who has IPAH, a patient who has
Eisenmenger syndrome and a comparable degree of PH has a
greater probability of longer survival. The degree of disability
due to hypoxemia may be considerable, however.
• Hemodynamic catheterization is critical for distinguishing flow-
mediated PH from shunt-mediated PAH.
Associated With Portal Hypertension
Among patients with portal hypertension who undergo evaluation
for orthotopic liver transplant, 10% to 20% have a PA systolic
pressure of 30 to 50 mm Hg or more. These patients are often
asymptomatic, but the possibility of PAH should be considered
and investigated since the mortality of liver transplant patients
is demonstrably increased if mPAP exceeds 35 mm Hg. Patients
with portopulmonary hypertension have a spectrum of hemody-
namic characteristics but frequently have a hyperdynamic circu-
lation with high cardiac output.
• All patients with cirrhosis should have periodic echocardiographic
screening for PH.
Box 79.2. Causes of Pulmonary Hypertension Due to
Left-to-Right Shunts
Extracardiac shunts
Patent ductus arteriosus
Aortopulmonary window
Rupture of aortic sinus
Peripheral arteriovenous fistula
Hemodialysis shunts
Intracardiac shunts
Ventricular septal defect
Atrial septal defect
Associated With Schistosomiasis
While rarely seen in the United States, schistosomiasis is the
leading cause of PAH worldwide. This PH was previously thought
to be a form of thromboembolic PH. It is now recognized that
schistosomiasis may lead to a pulmonary vascular inflammatory
remodeling process with clinical and pathophysiologic features
similar to other forms of PAH.
• Schistosomiasis is the leading cause of PAH worldwide.
Persistent PH of the Newborn
Persistent PH of the newborn, occurring in almost 2 newborns
per 1,000 live births, is present when pulmonary vascular resist-
ance remains high after birth and results in arterial hypox-
emia due to a right-to-left shunt through the fetal circulatory
pathways.
Pulmonary Venoocclusive Disease and Pulmonary
Capillary Hemangiomatosis
Pulmonary venoocclusive disease and pulmonary capillary
hemangiomatosis are very rare entities that have similarities in
pathologic features, clinical presentation, and response to ther-
apy. Although they share many pathophysiologic and clinical
types with PAH, they are classified as slightly different from
PAH since, in addition to having similar histologic changes in
the small pulmonary arteries, they also have a concomitant pul-
monary venopathy or microvasculopathy. Mortality is high, and
patients have a variable response to pulmonary vascular-targeted
therapy. Lung transplant is the only successful therapy. Clues to
the antemortem diagnosis include the combination of PAH with
evidence of pulmonary edema or pulmonary edema occurring in
response to acute or chronic pulmonary vasodilator challenge or
therapy. Surgical lung biopsy is required to make the diagnosis,
but this procedure carries a high, often prohibitive, risk to the
severely ill patient.
Group 2 PH: PH Due to Left Heart Disease
Any disorder of the left heart that increases pulmonary venous
pressure may lead to increases in PA pressure, initially from
passive reflection of venous pressures into the arterial bed. The
transpulmonary pressure gradient and the pulmonary vascular
resistance are within a normal range, but direct measures of left
atrial pressure and left ventricular end-diastolic pressure are ele-
vated. With time, PA remodeling can occur and be associated
with an increase in the transpulmonary gradient resulting from
high pulmonary vascular resistance. Diagnosis of this type of PH
is based on not only the pulmonary capillary wedge pressure but
also on an integration of clinical examination findings, cardiac
imaging and functional assessment, and hemodynamic measure-
ments reflecting the underlying left heart problem.
Group 3 PH: PH Due to Lung Disease
or Hypoxia (or Both)
The predominant cause of PH associated with lung disease or
hypoxia is inadequate oxygenation of arterial blood as a result
of diseases of the lung parenchyma, including ventilatory dis-
orders, or prolonged exposure to high altitude. As a result of
these diseases, hypoxic vasoconstriction causes increased
impedance at the level of pulmonary arterioles. The increase
in mPAP tends to correlate with the severity of the hypoxia and
 79 Pulmonary Hypertension
underlying lung disease. Most patients with group 3 PH have
rather modest elevations in mPAP (25–35 mm Hg). However,
some patients with chronic obstructive pulmonary disease have
severe PH, even with a relatively minimal reduction in the FEV1/
FVC ratio. The diagnostic strategy includes evaluation for pos-
sible obstructive or restrictive pulmonary disease, pulmonary
fibrosis, hypoxic sleep disorders, and other neurologic hypoven-
tilation syndromes.
• Sleep apnea alone rarely causes severe elevations in PA pressures.
Group 4 PH: CTEPH
In up to 4% of patients who survive an acute pulmonary embo-
lism, evidence of PH develops within 2 years. However, many
patients with documented CTEPH do not have a recognized
history of an acute venous thromboembolic event. Since many
cases of proximal CTEPH are amenable to surgical intervention,
further investigation is important.
For all patients with unexplained PH, ventilation-perfusion
scanning should be performed to rule out CTEPH; a normal scan
effectively excludes a diagnosis of CTEPH. Normal findings on
contrast-enhanced CT or MRI do not exclude the diagnosis of
CTEPH and should not be used to screen for CTEPH. When a
ventilation-perfusion scan is suggestive of CTEPH, pulmonary
angiography is required for accurate diagnosis and best ana-
tomical definition to assess operability for surgical pulmonary
thromboendarterectomy.
• All patients with unexplained PH should undergo ventilation-per-
fusion scanning to exclude CTEPH.
• Patients with CTEPH should be evaluated at an experienced PH
clinic and surgical center for potentially curative thromboen-
darterectomy.
Group 5 PH: PH With Unclear Multifactorial
Mechanisms
Several forms of PH occur in a small proportion of patients who
have various underlying diseases. These diseases that affect
the pulmonary vasculature and cause PH include miscellane-
ous disorders that cause inflammatory processes or mechanical
obstruction. The mechanism for PH in these diseases is often
multifactorial or unclear.
PH has been associated with multiple hematologic diseases,
including chronic myeloproliferative disorders, polycythemia
vera, essential thrombocytosis, and myelofibrosis with mye-
loid metaplasia accompanying chronic myeloid leukemia or the
myelodysplastic syndrome. Likewise, patients who have had a
splenectomy (autosplenectomy or surgical splenectomy) have
been reported to have an 11.5% prevalence of clinically signifi-
cant PAH after 4 to 32 years. Findings of PAH have been reported
with certain rare genetic diseases including type 1a glycogen
storage disease (von Gierke disease) and Gaucher disease.
Metabolic disorders, including hyperthyroidism, hypothy-
roidism, and hyperuricemia, are associated with higher preva-
lences of PH than expected. Other causes include sarcoidosis,
Langerhans cell histiocytosis, and lymphangiomatosis. Persons
at risk of PH also include those who have neurofibromatosis
type 1, antineutrophil cytoplasmic autoantibodies–associated
vasculitis, or end-stage renal disease and are undergoing long-
term hemodialysis. Extrinsic compression of pulmonary vessels
(arterial or venous) by adenopathy, malignancy, or mediastinal
fibrosis may also lead to PH.
739
Pathology
All forms of PH have medial hypertrophy of muscular and elastic
arteries, dilatation and intimal atheromas of elastic pulmonary
arteries, and right ventricular hypertrophy due to pressure over-
load. PAH specifically is characterized by constrictive and com-
plex arterial lesions.
The plexiform lesion is a focal proliferation, often located at
the branches of preacinar and intra-acinar pulmonary vessels of
the endothelial channels lined by myofibroblasts, smooth mus-
cle cells, and connective tissue matrix (Figure 79.1). It is typi-
cally associated with IPAH or HPAH (although the prevalence
is unclear) or with Eisenmenger syndrome, but it virtually never
occurs in PAH associated with connective tissue diseases. The
endothelial cells in plexiform lesions of IPAH have been reported
to be monoclonal in origin and express vascular endothelial
growth factor; they are thought to perhaps be part of a neoplastic-
like disordered angiogenesis. Dilatation lesions are venous-like
vessels sometimes observed downstream from plexiform lesions.
Other pathologic findings include intimal hyperplasia, adven-
titial thickening, and arteritis. Enlargement and hypertrophy of
the right ventricle (Figure 79.2A) and right atrium (Figure 79.2B)
are often prominent in severe PH.
Mechanisms
Several mechanisms and genetic factors appear to contribute to
the initiation and progression of PH. Pulmonary vasoconstriction
is believed to be an early component of the process. Mechanisms
promoting vasoconstriction in PH include abnormal function or
expression of potassium channels, reduced production of vaso-
dilators such as nitric oxide and prostacyclin, and overexpres-
sion of vasoconstrictors such as endothelin-1. All these factors
are also involved in stimulating vascular remodeling. The endo-
thelial production of prostacyclin, an endogenous pulmonary
vasodilator and platelet inhibitor, is decreased in PH patients. In
clinically apparent PH, vascular remodeling appears to be the
main abnormal process contributing to PH.
The level of lung and circulating endothelin-1 is increased
in patients with various types of PH and probably contributes
to both abnormal vasoconstrictive and proliferative abnormali-
ties. Reduced expression of nitric oxide synthase in endothelial
cells decreases production of nitric oxide from the substrate
amino acid arginine. This decreases the conversion of guano-
sine triphosphate to cGMP, facilitating entry of calcium ions into
Figure 79.1. Plexiform Lesion in Pulmonary Arteriole in Pulmonary
Hypertension.
740 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
A
B of the peripheral pulmonary arteries and obliteration of the ret-
Figure 79.2. Enlarged Right Ventricle and Atrium in Pulmonary
Hypertension. A, Right ventricular hypertrophy and dilatation in pul-
monary hypertension. B, Right atrial dilatation and normal left atrium.
vascular smooth muscle cells and promoting vasoconstriction
and cellular proliferation.
In patients with PAH, elevated circulating serotonin levels
and depressed platelet stores may directly increase pulmonary
vascular resistance through vasoconstriction and cellular pro-
liferation. Increased serotonin levels may be a mechanism by
which appetite suppressants such as fenfluramine increase the
risk of PAH. The appetite suppressants also block potassium
channels in pulmonary vascular smooth muscle cells, resulting
in increased entry of calcium ions into the cell and consequent
vasoconstriction.
Inflammatory mechanisms may be important in some forms
of PAH, such as those related to connective tissue diseases and
HIV infection. Elevated circulating levels of the proinflammatory
cytokines interleukin 1 and 6 have been observed. Thrombotic
processes and platelet dysfunction are involved in PAH, both by
promoting intravascular thrombus formation and by contributing
to the structural remodeling of the blood vessel itself. In situ PA
thrombosis is present in a large percentage of patients.
Diagnosis: Essential Tests
A general algorithm for the assessment of patients with PH is
shown in Figure 79.3.
Electrocardiography
If PAH is suspected, ECG should be performed to screen for a
spectrum of cardiac anatomical and arrhythmic abnormalities.
While ECG lacks sufficient sensitivity to serve as an effective
screening tool for PAH, it contributes prognostic information for
patients known to have PAH. ECG characteristically shows evi-
dence of right ventricular hypertrophy and right atrial enlarge-
ment. Up to 13% of patients with severe PAH have unremarkable
ECG findings. A P-wave amplitude in lead II of 0.25 mV or more
is associated with a greater risk of death (Figure 79.4).
Chest Radiography
If PAH is suspected, chest radiography should be performed to
look for features supportive of a diagnosis of PAH and to screen
for other cardiopulmonary diseases. Chest radiographs in PAH
show central pulmonary artery and right ventricular enlarge-
ment. Findings specific to advanced PAH include a prominent
pulmonary trunk and hilar pulmonary arteries with “pruning”
rosternal clear space by the enlarged, anteriorly situated right
ventricle (Figure 79.5).
Transthoracic Doppler Echocardiography
Two-dimensional and Doppler echocardiography should be per-
formed as a noninvasive screening tool in the assessment for PH
in suspected symptomatic patients or high-risk asymptomatic
patients. Estimated pulmonary arterial systolic, diastolic, and
mean pressures can be obtained with Doppler echocardiographic
criteria. Systolic pressure can be extrapolated from right ven-
tricular pressure in the absence of right ventricular outflow tract
obstruction or pulmonary stenosis. Right ventricular pressure is
determined from continuous-wave Doppler echocardiographic
signals showing the retrograde velocity across the tricuspid valve
(Figure 79.6). Careful Doppler examination by experienced
sonographers yields quantifiable tricuspid regurgitant signals in
70% to 85% of patients.
The tricuspid pressure gradient is derived from the modified
Bernoulli equation: ΔPTV = 4(VTR)2, where ΔPTV indicates max-
imal systolic pressure gradient across the tricuspid valve; and
VTR, peak Doppler signal velocity across the tricuspid valve. The
addition of estimated RAP yields RVSP: ΔPTV + RAP = RVSP.
The RAP should be estimated from the ultrasonographic charac-
teristics of the inferior vena cava and pulsed wave Doppler evalu-
ation of hepatic vein flow.
End-diastolic pulmonary regurgitant velocity can be measured
with continuous-wave Doppler echocardiography (Figure 79.7).
PADP corresponds to the regurgitant pressure gradient plus the
RAP (ie, the right ventricular end-diastolic pressure):
ΔPPV = 4(VPR)2 and ΔPPV + RAP = PADP,
where ΔPPV indicates end-diastolic pressure gradient across the
pulmonary valve; and VPR, regurgitant Doppler signal velocity
across the pulmonary valve.
Echocardiography also provides a critical assessment of right
ventricular size and function, right atrial size, degree of tricuspid
valve regurgitation and pericardial effusion and features sug-
gestive of a left-sided cause of PH, such as left-sided valvular
lesions, ventricular systolic or diastolic dysfunction, or intra-
cardiac shunt. A D-shaped but normally contractile left ventricle
on 2-dimensional echocardiography supports the diagnosis of
 RVE, RAE, RVSP
Left-sided heart disease Emphysema
Valvular heart disease ILD Sleep
Congenital heart disease Thoracic abnormality disorder

Chest
Echocardiography PFTs Sleep study
radiography

Ventilation-perfusion
Liver function tests
scan
and clinical evidence HIV test Autoantibody tests Contrast computed
of cirrhosis and portal
tomography
hypertension
Angiography

HIV Scleroderma
Portopulmonary SLE
hypertension Chronic PE
Rheumatoid arthritis
Vasculitis

• Functional test
• RH catheterization
• Vasodilator test

Figure 79.3. Flowchart for Evaluation of Suspected Pulmonary Hypertension (PH). The solid arrows indicate progression of tests for evalua-
tion of PH. The course of evaluation may vary depending on specific results of tests (eg, right heart catheterization may be performed directly after
echocardiography). Echocardiographic findings need to be clarified or confirmed before proceeding to evaluation of potential underlying causes of
PH. The dashed arrows point to some potential risk factors or characteristics of PH that may be elucidated by the associated test. The diagnoses shown
are not all inclusive. HIV indicates human immunodeficiency virus; ILD, interstitial lung disease; PE, pulmonary embolism; PFT, pulmonary func-
tion test; RAE, right atrial enlargement; RH, right heart; RVE, right ventricular enlargement; RVSP, right ventricular systolic pressure; SLE, systemic
lupus erythematosus. (Previously published. See “Credit Lines” section.)

aVR V1 V4

aVL V2 V5

V6

aVF V3

Figure 79.4. Electrocardiogram From a 28-Year-Old Woman With Primary Pulmonary Hypertension. The electrocardiogram shows right atrial
enlargement, right ventricular hypertrophy, and a right ventricular strain pattern.
742 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

A B

Figure 79.5. Chest Radiographs From a Patient With Primary Pulmonary Hypertension. A, Lateral view. B, Posteroanterior view.

2.52 x 4 = 25 mm Hg

2.92 x 4 = 34 mm Hg
Tricuspid
regurgitation
Doppler
signal

4.32 x 4 = 74 mm Hg

5.52 x 4 = 121 mm Hg

Figure 79.6. Continuous-Wave Doppler Echocardiographic Signals Showing Various Degrees of Tricuspid Regurgitant Velocity. The retrograde
velocity across the tricuspid valve is used to estimate right ventricular systolic pressure.
 79 Pulmonary Hypertension 743

Figure 79.7. Continuous-Wave Doppler Echocardiographic Signals of Pulmonary Regurgitation. The regurgitant velocity at end-diastole (Ved)
is used to estimate pulmonary artery diastolic pressure. An end-diastolic velocity of 0.9 m/s corresponds to an estimated pressure gradient of
4 × (0.92) = 3.24. If the right atrial pressure is 14 mm Hg, the pulmonary artery diastolic pressure is 3 + 14 = 17 mm Hg.

a pressure-overloaded and enlarged right ventricle (Figure 79.8).
Emerging echocardiographic techniques with 3-dimensional
echocardiography and strain imaging provide increasingly accu-
rate and reproducible measures of right ventricular size and
function.
Left atrial enlargement in the absence of mitral valve disease
suggests that an elevated left-sided filling pressure may be
contributing to pulmonary pressure elevation even if left ventric-
ular systolic function is normal.
Echocardiography provides one of the best evaluations of
congenital heart disease. Although the diagnosis of congenital
heart disease often precedes the discovery of PH, if PH is discov-
ered in a patient without a specific causal diagnosis, a thorough
echocardiographic study for evidence of intracardiac shunting is
warranted. Anomalous pulmonary venous return or pure left-to-
right shunts may be missed with transthoracic echocardiography,
and transesophageal echocardiography may be warranted for
best anatomical definition.
Anatomical evaluation with echocardiography should also
be completed for patients with suspected or documented PH.
The purpose is to look for left ventricular systolic and diastolic
dysfunction, left-sided chamber enlargement, and valvular heart
disease.

Cardiac MRI
Cardiac MRI is another way to assess ventricular structure and
function and is a useful alternative to transesophageal echocar-
diography for assessing systemic-pulmonary shunts, particularly
those that are extracardiac (eg, anomalous pulmonary veins).

Figure 79.8. Two-Dimensional Echocardiogram of Pressure-
Overloaded, Enlarged Right Ventricle. The normally contractile,
D-shaped left ventricle supports the diagnosis.
Radionuclide Studies
Ventilation-perfusion lung scanning is recommended for screen-
ing for CTEPH. The high sensitivity of the test (90%–100%)
for CTEPH means that a negative or very low-probability result
essentially rules out CTEPH (Figure 79.9). A segmental perfusion
defect visualized on ventilation-perfusion scanning should be
evaluated for definitive assessment with pulmonary angiography.
744 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Idiopathic Pulmonary Hypertension

R L L R

A Anterior Posterior

Thromboembolic Pulmonary Hypertension

R L L R

B Anterior Posterior

Figure 79.9. Ventilation-Perfusion Lung Scans. A, Diffuse inhomogeneity of perfusion, especially apically, suggests idiopathic pulmonary hyper-
tension. B, Segmental and subsegmental defects consistent with chronic major vessel thromboembolic pulmonary hypertension. L indicates left,
R, right.

The ventilation-perfusion scan correlates poorly with severity of
obstruction.
Pulmonary Function Testing
Pulmonary function testing is used to diagnose and quantify
underlying airway or parenchymal lung disease and should be
performed for all patients who have PH. The DLCO is reduced in
patients with PAH and tends to correlate with the severity of the
disease. A patient with PAH would be expected to have normal
results for spirometry and lung volumes. In patients with sys-
temic sclerosis, pulmonary function testing with DLCO should be
performed periodically (every 6–12 months) to improve detec-
tion of pulmonary vascular or interstitial disease. The DLCO is
decreased slightly, to approximately 60% to 80% of the predicted
value, in PAH. In all forms of PAH, desaturation during exer-
cise is primarily related to the inability of the right ventricle to
augment cardiac output, resulting in further depression of mixed
venous oxygen saturation.
Screening Overnight Oximetry
Overnight oximetry can screen for clinically significant obstruc-
tive sleep apnea or hypopnea during sleep. In the evaluation of
patients with PH, an assessment of sleep-disordered breathing
is recommended. Even in the absence of sleep apnea, noctur-
nal hypoxia may occur in PAH even with normal resting oxygen
saturation.
Essential Blood Tests
ANA titer is used to screen for connective tissue disease.
Although 40% of patients with IPAH have positive but low ANA
titers (≥1:80 dilutions), patients with a substantially elevated
ANA titer or suspicious clinical findings require further sero-
logic assessment and rheumatology consultation. HIV serology
should be performed unless exposure can be confidently excluded
because of the history. All patients should have a complete blood
cell count with platelet count and liver function tests.
 79 Pulmonary Hypertension 745

Assessment of Exercise Capacity 3. Congenital heart disease and systemic-to-pulmonary shunts

4. Portal hypertension and evaluation for orthotopic liver transplant
An objective assessment of exercise capacity is important for
several reasons: Patients with other potential PAH risk factors do not warrant
routine screening in the absence of symptoms because of the
1. To eliminate alternative reasons for dyspnea
infrequency of diagnosis or low likelihood of instituting treat-
2. To determine maximal exercise tolerance
3. To characterize a patient’s comfortable activity level ment at a presymptomatic stage.
4. To provide prognostic information
5. To establish exercise capacity at baseline and during therapy CT of the Chest
6. In some cases, to assess the hemodynamic response to exercise as a
possible cause of symptoms in patients with unremarkable resting High-resolution CT of the chest assists in the diagnosis of pul-
hemodynamics monary fibrosis. Signs of venopathy include smooth thickening
of interlobular septa, peribronchovascular cuffing, and alveolar
The 6MWT distance is predictive of survival for patients who
ground-glass opacification. In addition, CT of the chest may
have PH and correlates inversely with the severity of the WHO
provide further supportive evidence of CTEPH if the screening
functional status. Arterial oxygen desaturation of more than 10%
ventilation-perfusion scan is suggestive. However, negative find-
during the 6MWT increases the mortality risk 2.9 times over a
ings on a CT scan when there is a high suspicion of PH should not
median follow-up of 26 months. Observations from cardiopul-
be a reason to defer pulmonary angiography. A mosaic pattern of
monary exercise testing indicate that the mechanisms of exercise
lung attenuation in a noncontrast CT scan raises the possibility
limitations in PAH include ventilation-perfusion mismatching,
of chronic thromboembolism. Contrast-enhanced CT pulmonary
lactic acidosis at a low work rate, arterial hypoxemia, and inabil-
angiography can be used to visualize central chronic pulmo-
ity to adequately increase stroke volume and cardiac output.
nary thromboemboli (Figure 79.10). The CT features of CTEPH
are complete occlusion of pulmonary arteries, eccentric filling
Right Heart Catheterization defects consistent with thrombi, recanalization, and stenoses
RHC is mandatory to confirm the diagnosis of PH and should be or webs. The sensitivity of CT for detecting central pulmonary
completed on all patients before initiating therapy. Cardiac out- embolism is from more than 85% to 90%. Diagnostic accuracy
put, pulmonary vascular resistance, PA pressure, and pulmonary for CTEPH is improving with new-generation scanners, particu-
capillary wedge pressure should all be determined. Intracardiac larly in the hands of experienced readers.
shunting should be ruled out.
A vasoreactivity study should be performed when PAH is Invasive Pulmonary Angiography
discovered or confi rmed during RHC in patients for whom
Invasive pulmonary angiography is still often required to con-
treatment is contemplated. Patients who may benefit from CCB
firm CTEPH and assess candidacy for surgical pulmonary
treatment can be identified by a decrease in mPAP of at least
thrombectomy in patients who have a positive ventilation-perfu-
10 mm Hg to a pressure of 40 mm Hg or less in response to
sion scan or suggestive contrast-enhanced CT images. Chronic
short-acting vasodilators (intravenous epoprostenol, inhaled
thrombi appear different from acute thrombi on the angiogram
nitric oxide, or adenosine) without a decrease in cardiac out-
and occur in highly variable locations, often incorporated into
put. The absence of a vasodilatory response indicates a lack of
and retracting the vessel wall. Obstructions can take the form
acute vasoresponsiveness but does not indicate that a patient
of bands or webs, sometimes with poststenotic dilatation. An
will not respond to chronic pulmonary vascular–targeted vaso-
irregular intimal surface, rounded or pouch-like termination of
dilator therapy. The abrupt development of pulmonary edema
(occurring in <5% of patients) during acute vasodilator testing
suggests the presence of pulmonary venopathy or pulmonary
microvasculopathy and is a contraindication to long-term vaso-
dilator treatment.
Although RHC is considered the gold standard for PA hemo-
dynamics, the values obtained must be interpreted with care. PA
pressures are dynamic and fluctuate with time of day, physical
and emotional stress, hypoxia, and volume status. Hence, a single
result obtained at catheterization may not entirely represent the
pathophysiology of the PH patient. Attention to detail is required
to ensure that technical errors are avoided such as the use of an
inaccurate pulmonary capillary wedge pressure or an inappro-
priate thermodilution technique for cardiac outputs, particularly
with shunts or severe tricuspid regurgitation.

Diagnosis: Contingent Tests

Presymptomatic Screening
Screening for PH with Doppler echocardiography is warranted
when a patient has any of the following:
1. A known genetic mutation associated with PAH or a first-degree Figure 79.10. Pulmonary Thromboembolism. Cardiac computed
relative in a family with HPAH tomographic scan shows thrombus in the left main pulmonary artery
2. Scleroderma spectrum of disease (arrow).
746 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
Figure 79.11. Chronic Pulmonary Embolism. In this pulmonary
angiogram, the arrow points to a thrombus.
segmental branches, luminal narrowing of the central vessel, and
odd-shaped pulmonary arteries all may indicate the presence of
chronic pulmonary embolism (Figure 79.11).
Contingent Blood Tests
Arterial blood gas or oximetry measurements showing desatu-
ration may suggest abnormal gas exchange, right-to-left shunt-
ing, ventilation-perfusion mismatching, interstitial fibrosis, other
parenchymal lung disease, or hypoventilation. The degree of
arterial hypoxemia at rest in IPAH is usually mild to moderate;
severe hypoxemia should suggest another contributing prob-
lem. Failure of blood gases to normalize with a high fraction of
inspired oxygen suggests the presence of a component of direct
pulmonary-to-systemic circulation that resulted from intracar-
diac or extracardiac shunts or intrapulmonary shunting.
Exercise desaturation suggests that supplemental oxygen may
be useful to improve exercise capacity. Right-to-left shunting
through a patent foramen ovale during exercise may contribute
to desaturation in PAH.
Overnight oximetry may disclose nocturnal desaturation
associated with sleep apnea. Nocturnal hypoxemia occurs in
more than 75% of IPAH patients.
Hyperuricemia occurs with high frequency in patients with
PH and correlates with hemodynamic abnormalities (eg, RAP)
and mortality in IPAH.
The brain natriuretic peptide level is elevated in right ven-
tricular pressure overload and correlates with severity of right
ventricular dysfunction and mortality in PAH.
Polysomnography
If indicated from the history or screening overnight oximetry
results, a sleep study should be considered to assess a possible
contributory role for obstructive sleep apnea in PH. Up to 27%
of patients with sleep apnea syndromes have a degree of PH that
is usually mild and correctable by treatment with continuous
positive airway pressure.
Lung Biopsy
Lung biopsy is risky in patients with PH and should be limited to
circumstances in which histopathologic findings are required to
direct treatment, such as excluding or establishing a diagnosis of
active vasculitis, granulomatous pulmonary disease, pulmonary
venopathy, pulmonary microvasculopathy, interstitial lung dis-
ease, or bronchiolitis. In general, routine performance of a lung
biopsy to establish a diagnosis of PAH or to determine its cause
is discouraged.
Assessment of Severity and Prognosis
As will be discussed in the next section, therapy for PAH is
complex and expensive and needs to be tailored to each patient.
Key components needed for therapeutic decisions are a deter-
mination of the group (or type) of PH and an assessment of
disease severity. Factors that indicate advanced stages of the
disease and that are associated with a worse prognosis typi-
cally favor a more aggressive and complicated early therapeutic
approach.
The New York Heart Association functional classification for
heart failure was adapted to reflect the symptoms that limit PH
patients (dyspnea, fatigue, and presyncope) and is referred to as
the WHO functional classification. In PAH, syncope is associ-
ated with poor survival and is considered by many specialists to
be equivalent to WHO functional class IV disease. Rapid pro-
gression of symptoms is also a concerning feature indicative of
a high-risk patient.
Apart from symptoms, several clinical, echocardiographic,
and hemodynamic features portend a poor prognosis. Patients
with PH associated with connective tissue disease or portal
hypertension tend to do worse than those with other forms of
PAH. A short 6MWT distance (<300 m) is a negative factor as
is a disproportionately elevated level of brain natriuretic peptide.
The degree of right atrial or right ventricular enlargement on
echocardiography or cardiac MRI correlates well with the sever-
ity of disease and mortality. Other echocardiographic factors
that indicate a poor prognosis are markers of right ventricular
dysfunction, including the right ventricular index of myocardial
performance, reduced Doppler-derived cardiac output, reduced
longitudinal contractility (by tricuspid annular plane systolic
excursion, tissue Doppler, or strain imaging), degree of infe-
rior vena cava dilatation, and the presence of severe tricuspid
valve regurgitation or pericardial effusion (or both). At RHC,
the RAP, cardiac output, and PA capacitance all predict out-
come. Pulmonary pressures in isolation correlate less well with
outcome in patients with PAH because although PA pressures
tend to increase as the disease progresses, in severe disease the
PA pressure may plateau and then decrease as the right ventricle
fails. An integrative assessment of clinical, echocardiographic,
and invasive hemodynamic data provides prognostic information
that helps guide patient counseling and clinical decision making.
Several prognostic scores (eg, REVEAL risk score) are useful in
this assessment.
• Right ventricular function determines outcome in patients who
have PAH.
• PA pressures need to be integrated with right ventricular function
to determine severity.
• Prognosis is best determined for patients who have PAH by inte-
grating etiologic factors and functional class with clinical, echo-
cardiographic, and hemodynamic factors.
 79 Pulmonary Hypertension
Approach to Treatment in All Groups of PH
Hypoxemia is a potent pulmonary vasoconstrictor and can con-
tribute to the development or progression of PH. If possible,
supplemental oxygen should be used to maintain an oxygen
saturation of more than 90% at all times, although considerable
shunting may preclude achieving this goal. In patients who have
obstructive sleep apnea and PH, treating obstructive sleep apnea
with positive airway pressure therapy should cause pulmonary
pressures to decrease (although they may not normalize), par-
ticularly when PH is more severe.
A cardiopulmonary rehabilitation program should be consid-
ered for PH patients who have limited exercise capacity.
Diuretics are required in patients who have right ventricular
failure that causes peripheral edema or ascites. Digitalis is war-
ranted for modest right ventricular inotropic support and treat-
ment of atrial arrhythmias in patients who have refractory right
ventricular failure.
In situ microscopic thrombosis has been documented in some
patients with IPAH, and numerous prothrombotic abnormalities
have been identified. Patients who have right ventricular failure
and venous stasis are at increased risk of pulmonary throm-
boembolism. Improved survival has been reported with oral
anticoagulation in patients with IPAH. The target international
normalized ratio in patients who have IPAH treated with war-
farin is approximately 1.5 to 2.5. Patients with known CTEPH
should receive lifelong warfarin with a target international nor-
malized ratio of 2.0 to 3.0.
All PH patients should be immunized against influenza annu-
ally and against Streptococcus pneumoniae at diagnosis and at
age 65 or 5 years later, whichever is later. Patients with severe PH
requiring surgery should undergo a careful anesthesiology evalu-
ation to determine the risk and appropriate choice of anesthesia.
Approach to Treatment of Group 1 PH: PAH
In addition to the general approach discussed above, dual con-
traceptive therapy or surgical sterilization should be impressed
upon all women of childbearing potential after a PAH diagno-
sis is made. The risks associated with pregnancy include a 50%
maternal mortality for women with Eisenmenger syndrome.
Similarly, in 50% of pregnant women, preterm delivery and fetal
growth retardation may occur.
Calcium Channel Blockers
As previously described (see section on RHC), patients who
respond to the short-term administration of a short-acting vaso-
dilator may be treated cautiously with oral CCBs and should be
monitored closely to determine both the efficacy and the safety
of the therapy. CCBs with a strong negative inotropic effect, such
as verapamil, should be avoided; nifedipine, diltiazem, or amlod-
ipine are used most frequently and are titrated to the highest tol-
erated dose. Clinical improvement with CCBs is rarely marked
and is primarily seen only in IPAH patients. Replacement or
combination with additional medications should be considered in
symptomatic patients. There is no clear proven efficacy of CCBs
to lower pulmonary pressures in other forms of PH; indeed,
inappropriate use of CCBs in PH may be harmful.
Pulmonary Vascular–Targeted Therapy
Although there is no cure for PAH, since the 1990s the introduc-
tion of therapies that target pathways central to the pathologic
747
pulmonary vascular remodeling in PAH have transformed its
management, improving mortality for many patients and mor-
bidity for most. The 3 mechanistic pathways central to PAH
development that are targets for therapy are highlighted in
Figure 79.12. The nitric oxide and prostacyclin pathways are
vasodilatory and antiremodeling, and the endothelin pathway
promotes vasoconstriction and abnormal vascular remodeling.
Prostanoids
The FDA has approved 3 prostacyclin analogues for use in symp-
tomatic patients with PAH. This class of medications is expen-
sive and associated with numerous side effects.
Epoprostenol Sodium
Epoprostenol sodium is a synthetic prostacyclin used to replace
the prostacyclin that is deficient in patients with PH. It thereby
provides vasodilation, inhibition of platelet aggregation, and
antiproliferative effects. The mechanism is through the cyclic
adenosine monophosphate pathway. Treatment with epopros-
tenol is complex since it has a half-life of only several minutes
and therefore must be administered by continuous intravenous
infusion through a portable pump and indwelling central venous
catheter. The short half-life (6 minutes) also accounts for its most
feared adverse effect: the rebound exacerbation of PH with inter-
ruption of the infusion. The dose must be carefully titrated and
adjusted.
Epoprostenol treatment reduces symptoms of dyspnea and
improves exercise capacity, pulmonary hemodynamics, and sur-
vival. The hemodynamic improvement is relatively modest but
may become more notable with increasing duration of treatment.
The early effects probably result from vasodilation, whereas the
later benefits may result from reversal of vascular remodeling
due to antiproliferative mechanisms (Figure 79.13).
Common side effects of epoprostenol therapy include head-
ache, flushing, jaw pain with initial mastication, diarrhea, nau-
sea, a blotchy erythematous rash, and musculoskeletal aches and
pain (predominantly involving the legs and feet). Long-term use
of a high dosage can lead to the development of a hyperdynamic
state. Other complications of long-term intravenous therapy with
epoprostenol include infections related to the infusion catheter
line, catheter-associated venous thrombosis, thrombocytopenia,
and ascites. Central venous catheters used for prostacyclins are
more prone to promoting infection than central catheters used for
other disease states.
Treprostinil
Treprostinil is a prostacyclin analogue with a serum half-life of
4.5 hours. It can be administered by continuous subcutaneous
or intravenous infusion or (at lower dose) inhaled 4 times daily
during waking hours. Like epoprostenol, it improves symptoms
and modestly improves hemodynamics, but a definite survival
benefit has not been convincingly demonstrated. In addition,
adverse events, including headache, diarrhea, flushing, jaw pain,
and foot pain, are common. An additional side effect, frequent
pain and erythema at the infusion site, occurs with subcutaneous
infusion. Because of the longer half-life of treprostinil, infusion
interruptions due to catheter dislodgment or pump malfunction
tend to be less serious. An inhaled preparation of treprostinil
now allows less invasive administration of moderate doses
4 times a day.
748 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Nitric oxide pathway Prostacyclin pathway Endothelin pathway

Arginine AA Big-ET

Endothelial
eNOS PS ECE

cell
NO PGI2 ET1

PGI2 ETRA ETRA

ETA ETB
GTP ATP

muscle cell
Smooth
GC AC
5 GMP Ca++
PDE5
PDE5i cGMP CCBs
cAMP

Vasodilation, antiproliferation Vasoconstriction, proliferation

Figure 79.12. The 3 Mechanistic Pathways Known to Be Disturbed in Patients With Pulmonary Arterial Hypertension (PAH). The short, thick,
black arrows depict aberrations observed in these pathways in patients with PAH. The points at which drug treatment affects these mechanistic pro-
cesses are shown in gray circles. AA indicates arachidonic acid; CCB, calcium channel blocker; ETRA, endothelin receptor antagonist (eg, bosentan
[dual], ambrisentan and sitaxsentan [receptor A selective]); PDE5i, phosphodiesterase 5 inhibitor (eg, sildenafil).
Left, In the nitric oxide (NO) pathway, NO is created in endothelial cells by type III (ie, endothelial) NO synthase (eNOS), which in pulmonary
arterial smooth muscle cells (PASMCs) induces guanylate cyclase (GC) to convert guanosine triphosphate (GTP) to cyclic guanosine monophosphate
(cGMP). Cyclic GMP is a second messenger that constitutively maintains PASMC relaxation and inhibition of PASMC proliferation by ultimately
reducing the inward flux of calcium ions (Ca++). Cyclic GMP is removed by the PDE5 enzyme to yield the inactive product 5′-guanosine monophos-
phate (5′GMP). Patients with PAH have reduced expression and activity of eNOS.
Middle, In the prostacyclin pathway, the production of prostaglandin I2 (PGI2 [ie, prostacyclin]) is catalyzed by prostacyclin synthase (PS) in
endothelial cells. In PASMCs, PGI2 stimulates adenylate cyclase (AC), thus increasing production of cyclic adenosine monophosphate (cAMP) from
adenosine triphosphate (ATP). Cyclic AMP is a second messenger that constitutively maintains PASMC relaxation and inhibition of PASMC prolif-
eration. Patients with PAH have reduced expression and activity of PS.
Right, In the endothelin (ET) pathway, big-ET (ie, pro-ET) is converted in endothelial cells to ET1 (a 21–amino acid peptide) by endothelin-
converting enzyme (ECE). ET1 binds to PASMC ETA and ETB receptors, ultimately leading to PASMC contraction, proliferation, and hypertro-
phy. Endothelin 1 also binds to endothelial cell ET B receptors (not illustrated). Patients with PAH have increased expression and activity of ECE.
(Previously published. See “Credit Lines” section.)

Inhaled Iloprost potent vasoconstrictor and smooth muscle mitogen, at endothe-

Iloprost is a prostacyclin analogue with a serum half-life of 20 to
25 minutes. It is approved in the United States for administration
as an aerosolized inhalant. An advantage of an inhaled agent is
that it is selectively distributed to ventilated areas of the lungs,
so that its local vasodilatory effect results in optimal ventilation-
perfusion matching. The clinical benefits and adverse effects are
similar to those of the other prostanoids, with the addition of
cough. A unique disadvantage, however, is that, because of the
relatively short duration of action, it must be inhaled 6 to 9 times
a day. The hemodynamic effects disappear within 30 to 90 min-
utes after inhalation but may be extended by coadministration of
other medications, such as sildenafil.
Endothelin Receptor Antagonists
Endothelin receptor antagonists address the excess effect of
endothelin that has been observed in patients with PH.
Bosentan
Bosentan is an orally administered nonselective endothelin
receptor antagonist that blocks the action of endothelin-1, a
lin A and B receptors. It therapeutically decreases vasoconstric-
tion and pulmonary vascular hypertrophy caused by increased
plasma levels of endothelin-1 in patients with PAH and medi-
ated predominantly through endothelin A receptors. The demon-
strable clinical vasodilatory effect of the drug is quite modest in
patients with established PAH, but clinical studies have demon-
strated an augmented 6MWT distance (compared with placebo)
and improved WHO functional classification over 16 weeks.
Some of its benefit may be related to antiproliferative and anti-
fibrotic effects that stabilize the disease process and promote
remodeling. The medication is administered as an oral pill twice
daily. The costs are slightly less than for prostacyclin analogues.
Adverse effects associated with bosentan include dose-depen-
dent elevation of transaminases, reflecting hepatic toxicity for
which up to 5% to 10% of patients need to discontinue use of
the medication. Liver function tests must be monitored monthly
through the duration of treatment. Bosentan is also associated
with typically small decreases in hemoglobin levels. Syncope
and flushing occur in some patients. Drug interactions are rec-
ognized with sildenafil, glyburide, cyclosporine, ketoconazole,
statins, and warfarin; bosentan may interfere with the action of
hormonal contraceptives. The drug is highly teratogenic.
 79 Pulmonary Hypertension
100 Observed
Expected
80
Survival, %
60
40
20
0 6 12 18 24 30 36
Months
Figure 79.13. Observed Survival of Patients Treated With
Epoprostenol Compared With Expected Survival Predicted From
Baseline Hemodynamic Parameters. Asterisks indicate P<.001 at 1, 2,
and 3 years. (Previously published. See “Credit Lines” section.)
Ambrisentan
Ambrisentan is a selective endothelin A receptor antagonist with
actions similar to those of bosentan. This medication is admin-
istered orally once daily. The risks of hepatotoxicity are signifi-
cantly reduced; only episodic liver function test monitoring is
needed.
Phosphodiesterase Type 5 Inhibition
Phosphodiesterases are enzymes that hydrolyze the cyclic
nucleotides, including cGMP, to inactive monophosphates
(5′-guanosine monophosphate). Drugs that inhibit cGMP-spe-
cific phosphodiesterases increase the pulmonary vascular effect
of the nitric oxide pathway, resulting in vasodilation and antipro-
liferative effects.
Sildenafil Citrate
Sildenafil citrate is a potent and specific orally administered phos-
phodiesterase type 5 inhibitor that is a safe and effective treat-
ment of erectile dysfunction. The drug is approved by the FDA for
use in PAH at a dosage of 20 mg 3 times daily. The relatively few
minor side effects include headache, nasal congestion, epistaxis,
and heartburn. Rare adverse effects of acute visual or auditory
loss have been reported. This medication acutely improves pul-
monary hemodynamics and may have sustained hemodynamic
benefit. Clinically, functional capacity and 6MWT distances are
improved with sildenafil use, and the actions appear additive to
other classes of medication used.
Tadalafil
Tadalafil has a longer half-life than sildenafil, allowing a once-
daily dosing regimen.
Choice of Therapy
In general, choice of therapy is guided by disease severity. All
PH patients should be offered diuretics, digoxin, and oxygen as
indicated. Oral anticoagulation should be used in IPAH, HPAH,
and CTEPH and considered for associated PAH. In the absence of
749
right heart failure, patients who have a favorable acute response
to a vasodilator should be considered candidates for a trial of
therapy with an oral CCB. CCBs should not be used empirically
to treat PH in the absence of demonstrated acute vasoreactivity.
Symptomatic patients who have PAH WHO functional class II
and who are not candidates for CCB therapy, or who have not had
an adequate response to a trial of CCB therapy, may benefit from
further treatment. Drug therapy with endothelin receptor antago-
nists or phosphodiesterase type 5 inhibitors, may be considered.
Enrollment in clinical trials is encouraged.
Endothelin receptor antagonists, phosphodiesterase type
5 inhibitors or prostanoids should all be considered for WHO
functional class III disease. First-line treatment with parenteral
prostaglandins is favored for WHO class III patients who have
poor prognostic markers and for those who have WHO class IV
disease (Figure 79.14).
All patients should have frequent reassessment, and optimiza-
tion or a combination of therapy is recommended in the absence
of disease control. Patients who have severe, refractory disease
should be considered early for atrial septostomy or transplant
(or both) as described below.
Procedures and Surgery for Treatment of PAH
Atrial Septostomy
Creation of an interatrial right-to-left shunt by graded balloon
pullback techniques (atrial septostomy) appears to decompress
the failing right ventricle and increase left ventricular preload.
The result, improved left-sided filling, is exchanged for arterial
desaturations. Patients who have severe PAH refractory to medi-
cal therapy and signs of impaired left ventricular filling, includ-
ing syncope, should be considered for septostomy. The procedure
is contraindicated in patients with severely elevated RAP or poor
baseline arterial oxygen saturations.
Lung and Heart-Lung Transplant
Heart-lung, single lung, and bilateral lung transplants have been
used in the treatment of patients with PAH, who compose about
24% of heart-lung recipients, 2% of single lung recipients, and
8% of bilateral lung recipients. Each approach has advantages
and disadvantages. Heart-lung transplant is hampered by a short-
age of available organs, although it has the advantages of requir-
ing only 1 airway anastomosis, having a very low rate of vascular
complications, and generally yielding the best hemodynamic out-
comes. Single lung transplant is easier to perform and requires
less operative, ischemic, and cardiopulmonary bypass time,
but the potential exists for ventilation-perfusion mismatch and
reperfusion injury. Bilateral lung transplant may produce better
hemodynamics, less ventilation-perfusion mismatch, fewer early
complications, better immediate overall lung function, and pos-
sibly improved long-term survival, but the operation is longer
and more difficult to perform.
Survival of patients with IPAH after heart-lung or lung trans-
plant is approximately 70% at 1 year, 40% at 5 years, and 25%
at 10 years. Among operative survivors, most patients have early
improvement in mPAP (preoperative mPAP, 60–70 mm Hg; post-
operative mPAP, 20–25 mm Hg) and concomitant improvements
in cardiac index and pulmonary vascular resistance. Among
long-term survivors of lung or heart-lung transplant, over 80%
have no limitation in activity at 1 year and 5 years after trans-
plant, and 40% to 50% work at least part-time.
PAH patients with progressive disease or WHO functional
class IV symptoms should be referred to a transplant center
750 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

General care
Oral anticoagulants ± diuretics ± oxygen ± digoxin

Acute vasoreactivity testing

Positive

Negative
Oral CCB

Lower risk Higher risk

Sustained
response

ETRAs or PDE5 inhibitors (oral) Epoprostenol or treprostinil (IV)

No Iloprost (inhaled) Iloprost (inhaled)
Epoprostenol or treprostinil (IV) ETRAs or PDE5 inhibitors (oral)
Yes Treprostinil (subcutaneous) Treprostinil (subcutaneous)

Continue Investigational protocols Atrioseptostomy

CCB Combination regimens Lung transplant

Clinical reassessment; consider additional

therapy if treatment goals are not met

Determinants of Risk Lower Risk Higher Risk

Clinical evidence of RV failure No Yes

Progression Gradual Rapid
WHO class II, III IV
6-min walk distance test Longer (>400 m) Shorter (<300 m)
Brain natriuretic peptide level Minimally elevated Very elevated
Echocardiographic Minimal RV Pericardial effusion;
dysfunction substantial RV dysfunction
Hemodynamics Normal or near-normal High right arterial pressure,
right arterial pressure low cardiac index
and cardiac index

Figure 79.14. Algorithm for the Pharmacologic Management of Pulmonary Arterial Hypertension. Assessment of risk for each patient includes
clinical variables as outlined in the table below the algorithm. Patients deemed to be at highest risk on the basis of clinical assessment should be
considered for early intravenous (IV) therapy. Those at lower risk are candidates for oral therapy. Patients should be followed up closely, and their
response to therapy should be assessed within several months. If treatment goals are not met, addition of a second agent may be warranted. CCB indi-
cates calcium channel blocker; ETRA, endothelin receptor antagonist; PDE5, phosphodiesterase type 5; RV, right ventricular; WHO, World Health
Organization. (Previously published. See “Credit Lines” section.)

for evaluation and listing for lung or heart-lung transplant. In
patients with PAH who are undergoing transplant, the procedure
of choice is bilateral lung transplant. In adult patients with PAH
and simple congenital heart lesions, bilateral lung transplant with
repair of the cardiac defect is the procedure of choice. In adult
patients with PAH and complex congenital heart disease who are
undergoing transplant, heart-lung transplant is the procedure of
choice.
Liver Transplant in Portopulmonary Hypertension
Liver transplant is the therapy of choice for patients who have
portopulmonary hypertension; however, it should be consid-
ered only if and when the PH is controlled with therapy. The
degree of PAH in this group that is accepted for the patient to be
a transplant candidate is variable. In general, contraindications
are hemodynamic findings of mPAP greater than 35 mm Hg and
pulmonary vascular resistance greater than 3.125 Wood units,
which are associated with a high mortality after orthotopic liver
transplant.
Approach to Treatment in Group 2 PH:
PH Due to Left Heart Disease
Typically, with pulmonary venous hypertension, PH occurs in
proportion to the severity of the disease and responds to thera-
pies that lower pulmonary venous pressure. When PH is present,
it reflects the severity of the left heart disease, but there is little
 79 Pulmonary Hypertension 751

evidence that direct targeting of PH in left heart disease is war-

ranted or beneficial. The approach to management always favors
addressing the left heart disease. In some instances, the severity
of the PH appears to exceed that which would be expected from
the degree of left heart disease. In these instances, it is unclear
whether the chronicity of pulmonary venous hypertension has led
to a secondary pulmonary vasculopathy. Furthermore, in these
instances, whether there is a role for pulmonary vascular–targeted
therapy is also unclear and understudied. Phosphodiesterase type
5 inhibitors currently being tested for heart failure with preserved
ejection fraction may have a therapeutic role in selected cases
and are the focus of ongoing investigation. These cases are com-
plex, and empirical use of pulmonary vascular therapies outside
of clinical trials should be considered with great caution.
Patients with PH secondary to left heart disease may be
Figure 79.15. Surgically Excised Fibrotic Obstructive Tissue.
eligible for heart transplant alone. The PH in this disease is
typically neurohormonally mediated and not associated with
significant structural changes of the pulmonary vasculature. have persistent PH postoperatively. This intervention improves
In these patients, proof of reversibility of pulmonary vascular cardiac output, PA pressure, right ventricular function, symp-
resistance to less than 4 Wood units with the use of inotropes is tomatic status, and survival in appropriately selected patients.
mandatory before listing for transplant. Another criterion often Regardless of the surgical approach or outcome, patients who
considered exclusionary for transplant is a transpulmonary gra- have CTEPH should receive lifelong warfarin anticoagulation
dient (mPAP minus mean pulmonary capillary wedge pressure) and should be considered for permanent inferior vena cava filter
greater than 15 mm Hg. Either of these factors predicts failure of placement.
the donor right ventricle in the immediate postoperative period.
Heart transplant in eligible patients is typically curative for left
heart–related PH. Suggested Reading
http://www.phaonlineuniv.org/
Approach to Treatment in Group 3 PH: PH Due
to Lung Disease or Hypoxia (or Both)
Abbreviations
As with patients who have group 2 PH, the degree of PH in con-
comitant lung disease is typically mild and tends to correlate with ALK1 activin receptor-like kinase type 1
the severity of the underlying disease. However, some patients ANA antinuclear antibody
ASD atrial septal defect
have PH that is proportionately more severe than expected. As in
BMPR2 bone morphogenetic protein receptor type 2
left heart disease, the presence and severity of PH in lung disease CCB calcium channel blocker
has significant prognostic implications; however, the therapeu- cGMP cyclic guanosine monophosphate
tic implications are less clear. The main driver of PH in lung CT computed tomography
disease is hypoxia, and the major therapeutic focus should be CTEPH chronic thromboembolic pulmonary hypertension
directed toward improving or normalizing oxygenation. The goal DLCO diffusing capacity of lung for carbon monoxide
should be to maintain oxygen saturations greater than 90% at ECG electrocardiography
rest, with exercise, and with sleep. To date, studies involving the FDA US Food and Drug Administration
use of pulmonary vascular–targeted agents in group 3 PH have FEV1 forced expiratory volume in the first second of expiration
resulted in a null effect or harm. Although empirical treatment FVC forced vital capacity
HIV human immunodeficiency virus
of PH in often untreatable underlying lung disease is tempting,
HPAH heritable pulmonary arterial hypertension
one must be cautious because of the potential for exacerbating a IPAH idiopathic pulmonary arterial hypertension
ventilation-perfusion mismatch and exacerbating hypoxia with mPAP mean pulmonary arterial pressure
systemic vasodilator therapy. Typically oxygen and transplant MRI magnetic resonance imaging
are the only proven therapeutic options. 6MWT 6-minute walk test
PA pulmonary arterial
PADP pulmonary arterial diastolic pressure
Approach to Treatment in Group 4 PH: CTEPH PAH pulmonary arterial hypertension
All patients who have otherwise unexplained PH should undergo PDA patent ductus arteriosus
a diagnostic evaluation for possible thromboembolic disease. If PH pulmonary hypertension
the anatomy is conducive to surgery, outcome can be markedly RAP right atrial pressure
REVEAL Registry to Evaluate Early and Long-term Pulmonary
improved through surgical excision of the fibrotic obstructive tis-
Arterial Hypertension Disease Management
sue (Figure 79.15). Pulmonary endarterectomy should be per- RHC right heart catheterization
formed in an experienced surgical center. The use of pulmonary RVSP right ventricular systolic pressure
vascular–targeted therapy in patients with CTEPH should be SLE systemic lupus erythematosus
reserved for those without central (main, lobar, or segmental) dis- VSD ventricular septal defect
ease (ie, not a surgical candidate) or the minority of patients who WHO World Health Organization
 80
Pregnancy and the Hearta
HEIDI M. CONNOLLY, MD
Approximately 2% of pregnancies occur in women who have
heart disease. Congenital heart disease is the predominant form
of heart disease among pregnant women in developed countries,
whereas rheumatic heart disease predominates in developing
countries. Heart disease does not preclude successful pregnancy
but increases the risk for both mother and baby and requires
special management.
This chapter discusses the normal hemodynamic changes
that occur during pregnancy and the resultant effect on common
cardiovascular diseases.
Physiology
Hemodynamic Changes During Normal
Pregnancy
Substantial hemodynamic changes occur during normal preg-
nancy, including a 20% to 30% increase in red blood cell mass
and a 30% to 50% increase in plasma volume. As a result,
there is an increase in total blood volume with relative anemia
(Figure 80.1). Heart rate increases about 10 beats per minute,
and systemic and pulmonary vascular resistance decrease. Blood
pressure generally decreases slightly during pregnancy. These
hemodynamic changes result in a steady increase in cardiac
output during pregnancy until the 32nd week, when cardiac
output plateaus at 30% to 50% above the prepregnancy level.
The pregnant uterus can require up to 18% of cardiac output.
a
Portions previously published in Connolly HM. Pregnancy in women
with congenital heart disease. Curr Cardiol Rep. 2005 Jul;7(4):305–9
and Connolly HM. Pregnancy in women with coarctation of the
thoracic aorta. ACC Curr J Rev. 1997 May-Jun;6(3):55–7. Used with
permission.
Abbreviations and acronyms are expanded at the end of this chapter.
752
Oxygen consumption increases steadily throughout pregnancy
and reaches a level of approximately 30% above the prepregnant
level by the time of delivery. This increase is due to the metabolic
needs of both mother and fetus. During the last half of pregnancy,
cardiac output is significantly affected by body position, because
the enlarging uterus decreases venous return from the lower
extremities. The left lateral position minimizes this reduction in
venous return. Normally, the hemodynamic changes that occur
during pregnancy are well tolerated by the mother. Heart disease
may be manifested initially during pregnancy because cardiac
output is increased or because minor preexisting symptoms are
exacerbated.
Cardiac Examination in Normal Pregnancy
During normal pregnancy, there is a brisk and full carotid upstroke,
and jugular venous pressure is normal or slightly increased, with
prominent a and v waves. The left ventricular impulse is dis-
placed laterally and is enlarged. The first heart sound is louder
than normal. The pulmonic second sound may be prominent, and
there often is persistent splitting of the second heart sound. A
third heart sound is audible in more than 80% of healthy preg-
nant women (Figure 80.2). An early peaking ejection systolic
murmur is audible in more than 90% of healthy pregnant women
and is caused by a pulmonary outflow murmur. Venous hums and
mammary continuous murmurs are common but without signif-
icance. Peripheral edema and venous varicosities are common.
Abnormal physical findings include a fourth heart sound, a loud
systolic murmur (grade 3/6 or louder), and a diastolic murmur or
fixed splitting of the second heart sound. These do not occur dur-
ing normal pregnancy in the absence of heart disease.
• Normal physical findings during pregnancy may be misinterpreted
as abnormal.
 80 Pregnancy and the Heart 753

Plasma decreases it further. Approximately 500 mL of blood is lost at

Increase in Blood Volume, %
50
vaginal delivery, and approximately 1,000 mL is lost during a
normal cesarean delivery.
40
Whole blood
30 Hemodynamic Changes Post Partum
After delivery, venous return increases because of relief from
20 Erythrocytes fetal compression on the inferior vena cava. Spontaneous diure-
sis occurs during the first 24 to 48 hours after delivery; however,
10 it takes about 2 to 4 weeks for hemodynamic values to return
to baseline after vaginal delivery, and it takes even longer after
0 cesarean delivery.
0 10 20 30 40
• Cardiac output increases by 30% to 50% during normal pregnancy.
Gestational Age, wk
• Cardiac output increases to about 80% above baseline during labor
Figure 80.1. Hemodynamic Changes During Normal Pregnancy. and delivery.
Total blood volume increases with relative anemia. (Previously pub- • Hemodynamic values return to baseline 2 to 4 weeks after vaginal
lished. See “Credit Lines” section.) delivery and up to 6 weeks after cesarean delivery.

Cardiac Disease in Pregnancy

Cardiac Studies in Pregnancy
On chest radiographs, the cardiac silhouette is enlarged, with
increased vascular markings. On echocardiography, there is a
small increase in right and left ventricular volumes. The electro-
cardiogram shows an increase in heart rate, with a leftward shift
of the QRS and T-wave axes because of the upward and horizon-
tal displacement of the heart by the pregnant uterus.
Hemodynamic Changes During Labor
and Delivery
With uterine contractions, an additional 300 to 500 mL of blood
enters the circulation. This increase in blood volume in conjunc-
tion with increased blood pressure and heart rate during labor
increases cardiac output. At delivery, cardiac output increases as
much as 80% above the prepregnancy level (and may be as great
as 9 L/min). Administration of epidural anesthesia decreases car-
diac output to about 8 L/min, and the use of general anesthesia
Principles of Management
Antepartum management of heart disease should include an
anatomical and hemodynamic assessment of any cardiac abnor-
mality to determine the maternal and fetal risks of pregnancy.
Doppler echocardiography is ideal for the evaluation of pregnant
women with heart disease. In evaluation of women with cardio-
vascular disease, the safety of the mother is always the highest
priority. If a specific cardiovascular condition poses an unaccept-
able risk of death to either the mother or the baby, pregnancy
should be avoided.
Fetal growth and development are monitored with ultrasonog-
raphy. Fetal cardiac ultrasonography is recommended for women
with a personal history of congenital heart disease, a prior preg-
nancy resulting in an offspring with congential heart disease, or
abnormal fetal cardiac anatomy noted on routine fetal ultraso-
nography. The woman’s hemodynamic response to pregnancy
requires special attention.

Wide,
loud, split S1, 88%
MC TC A2 P2
Systolic murmur, 96% Diastolic “flow”
murmur, 18%

S4
Occasional S3
Loud, 84%

Figure 80.2. Normal Auscultatory Findings During Pregnancy. Percentages indicate the frequency of the finding among healthy pregnant women.
A2 indicates aortic second sound; MC, mitral valve closure; P2, pulmonic second sound; S1, first heart sound; S3, third heart sound; S4, fourth heart
sound; TC, tricuspid valve closure.
754 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
The time and route of delivery should be planned before spon-
taneous labor to facilitate the delivery and to intervene as appro-
priate. With few exceptions, vaginal delivery with a facilitated
second stage (forceps delivery or vacuum extraction) is preferred
for women with heart disease. Cesarean delivery is indicated for
obstetric reasons and for a patient who is fully anticoagulated
with warfarin, because of the risk of fetal intracranial hemorrhage
(Box 80.1). In addition, cesarean delivery should be considered
for patients with critical or symptomatic, fixed cardiac obstruc-
tive lesions; a markedly enlarged or enlarging aorta during preg-
nancy; or pulmonary hypertension. The optimal anesthesia and
analgesia and the administration of prophylactic treatment for
endocarditis should be considered before pregnancy. Invasive
hemodynamic monitoring is occasionally recommended for
severe maternal heart disease. Maternal postpartum care should
include early ambulation, attention to neonatal concerns, and
consideration of contraception if appropriate.
• Antepartum anatomical and hemodynamic assessment of cardiac
status is imperative to determine the maternal and fetal risks of
pregnancy.
• Vaginal delivery is the preferred mode of delivery for most women
with heart disease.
Maternal Prognosis
Maternal prognosis during pregnancy is strongly related to func-
tional class. NYHA class greater than II is the strongest predic-
tor of cardiovascular complication during pregnancy. Additional
factors that contribute to the risk of cardiovascular complications
during pregnancy include 1) a prior history of heart failure, tran-
sient ischemic attack, stroke, or arrhythmia; 2) the presence of
cyanosis; 3) left heart obstruction with either mitral stenosis with
a mitral valve area less than 2 cm2 or aortic stenosis with a peak
left ventricular outflow tract gradient greater than 30 mm Hg
and an aortic valve area less than 1.5 cm2 on echocardiographic
Doppler examination; and 4) reduced systemic ventricular func-
tion (EF < 40%).
Pregnant women in NYHA class I or II should limit strenuous
exercise, have adequate sleep and rest, maintain a low-salt diet,
avoid anemia (maintain hemoglobin >11 g/dL), have frequent
prenatal examinations (both obstetric and cardiovascular), and
be monitored for arrhythmias. In symptomatic women (NYHA
class >II) and in those with systemic ventricular dysfunction
(ejection fraction <40%), pregnancy should be avoided because
of the risk of pregnancy-related complications. The option to
continue or interrupt the pregnancy should be discussed with the
patient. If the patient opts to continue her pregnancy, bed rest is
often required during part of the pregnancy, and close cardiac
and obstetric monitoring are mandatory. Treatment of conges-
tive heart failure is more challenging in pregnant women than in
nonpregnant women. Conservative measures are very important;
Box 80.1. Indications for Cesarean Delivery in
Women With Cardiovascular Disease
Obstetric reasons
Anticoagulation with warfarin
Fixed, severe obstructive cardiac lesions
Pulmonary hypertension
Dilated or unstable aorta
Box 80.2. Cardiac Features of a High-Risk
Pregnancy
Prosthetic valves
Obstructive lesions, including uncorrected
coarctation of the aorta
Marfan syndrome
Hypertrophic obstructive cardiomyopathy
Cyanotic congenital heart disease
Pulmonary hypertension
Systemic ventricular dysfunction (ejection
fraction <40%)
Significant uncorrected congenital heart disease
however, pharmacologic therapy may be required, and safe
medical options are limited owing to adverse effects of angio-
tensin-converting enzyme inhibitors and angiotensin receptor
blockers on the fetus. These medications should be avoided dur-
ing pregnancy.
With the hemodynamic changes that occur during pregnancy,
fixed obstructive cardiac lesions or those associated with pul-
monary hypertension generally are poorly tolerated, primarily
because of the inability to increase cardiac output. In contrast,
regurgitant lesions are relatively well tolerated because of the
pregnancy-related decrease in systemic vascular resistance.
Cardiac features of a high-risk pregnancy include 1) prosthetic
valves; 2) obstructive lesions, including uncorrected coarctation
of the aorta; 3) Marfan syndrome; 4) hypertrophic obstructive
cardiomyopathy; 5) cyanotic congenital heart disease; 6) pulmo-
nary hypertension; 7) systemic ventricular dysfunction (ejection
fraction <40%); or 8) significant uncorrected congenital heart
disease (Box 80.2).
Cardiac Contraindications to Pregnancy
There are certain cardiac conditions in which pregnancy should
be avoided or, if pregnancy occurs, termination should be con-
sidered (Box 80.3). These include severe primary or secondary
pulmonary hypertension (pulmonary artery pressure ≥75% sys-
temic pressure), including Eisenmenger syndrome. Two other
situations in which pregnancy is contraindicated are cardiomy-
opathy with NYHA class III or IV congestive heart failure and
left ventricular dysfunction (ejection fraction <40%) with signif-
icant symptoms. Any form of severe obstructive cardiac lesion,
such as aortic stenosis, mitral stenosis, pulmonary stenosis,
coarctation, or hypertrophic obstructive cardiomyopathy, may
result in important limitations during pregnancy. Intervention
before pregnancy is the preferred management option. Women
with Marfan syndrome with an aortic root of 40 mm or more
should be strongly counseled to avoid pregnancy because of
the unpredictable risk of aortic dissection and rupture. Severe
cyanosis is a relative contraindication to pregnancy, primarily
because of adverse fetal outcome. A patient with a history of
peripartum cardiomyopathy should be counseled to avoid addi-
tional pregnancies. The risk of recurrence is approximately
50%, and recurrence carries a risk of progressive heart failure
and death.
• Pregnancy should be avoided if the systemic ventricular ejection
fraction is <40% or the NYHA class is more than II.
 80 Pregnancy and the Heart 755

syndrome, or hypertrophic cardiomyopathy). Occasionally famil-

Box 80.3. Cardiac Contraindications to Pregnancy
Severe pulmonary hypertension
Eisenmenger syndrome
Cardiomyopathy with NYHA class III or IV
congestive heart failure
Systemic ejection fraction <40%
Severe obstructive cardiac lesions
Marfan syndrome with aortic root ≥40 mm
Severe cyanosis
History of peripartum cardiomyopathy
Abbreviation: NYHA, New York Heart Association.
Congenital Heart Disease in Pregnancy
For patients with repaired complex congenital heart disease,
some uncertainties still remain about the ability to conceive, the
effects of pregnancy on maternal heart disease, and the effects of
heart disease on the fetus.
An increasing number of women with congenital heart disease
are reaching childbearing age and considering pregnancy. This
is primarily because congenital heart disease is being diagnosed
and managed earlier and women are surviving to childbearing
age.
Patients should be counseled about pregnancy and the increased
risk of congenital heart disease in the fetus. The incidence of con-
genital heart disease in the general population is about 1%. The
offspring of women with congenital heart disease have a 5% to
6% incidence of congenital heart disease. Usually, the lesion in
the offspring is not the same as in the mother, except for syn-
dromes in which the incidence of recurrence with each preg-
nancy may be up to 50% (eg, autosomal dominant disorders such
as the 22q11 microdeletion in conotruncal abnormalities, Marfan
ial left-sided obstructive lesions and atrial septal defects occur.
Bicuspid aortic valve is now recognized to be inherited, and
screening of first-degree relatives is recommended.
Fetal cardiac ultrasonography is used routinely in women
with congenital heart disease to detect congenital heart disease
in the fetus.
Endocarditis prophylaxis is generally recommended around
the time of delivery in high-risk patients who have congenital
heart disease (for specific recommendations, SEE CHAPTER 83,
“INFECTIVE ENDOCARDITIS”).
Cyanosis inhibits fetal growth and development (Figure 80.3).
Pregnancy is generally contraindicated in women with severe
cyanosis. Surgical repair of the underlying cardiac anomaly
should be considered before pregnancy if possible (eg, Ebstein
anomaly with cyanosis from a right-to-left shunt related to an
atrial septal defect).
Rarely, device intervention is indicated during pregnancy,
such as for closure of an atrial septal defect or balloon dilata-
tion of a valve. However, the procedure can usually be postponed
until after delivery.
• Pregnant cyanotic women have a high risk of fetal loss, and cya-
nosis is a recognized handicap to fetal growth, resulting in low
birth weight.
• Congenital heart disease is the most common form of structural heart
disease that affects women of childbearing age in the United States.
• Congenital heart disease has important implications for both the
mother and the fetus.
• The incidence of congenital heart disease is increased in the
offspring of women with congenital heart disease.
Marfan Syndrome and Other Aortopathies in
Pregnancy
The risk for pregnant patients with Marfan syndrome relates
to the underlying medical changes affecting the aorta. The

4
Birth Weight, kg

Normal mean
1

0
0 10 20 30 40 50
Gestational Age, wk
Figure 80.3. Maternal Cyanosis. Severity of maternal cyanosis, as indicated by hemoglobin level, is related directly to fetal loss (gestational age
<20 weeks), prematurity, and infant birth weight. Blue rectangle indicates term by weight and age. Circles and squares indicate maternal hemoglobin
level (red circles, ≤15.0 g/dL; blue circles, 15.1–18.0 g/dL; and green squares, >18.0 g/dL). (Previously published. See “Credit Lines” section.)
756 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
hormonal and physiologic changes that occur during pregnancy
may adversely affect an abnormal aorta. This results in an unpre-
dictable maternal risk of aortic dissection and rupture. Patients
with Marfan syndrome who have an aortic dimension of 40 mm
or more are usually counseled against pregnancy because of the
unpredictable risk of a cardiovascular event.
If a patient with Marfan syndrome is considering pregnancy,
careful prepregnancy cardiovascular evaluation and counsel-
ing are recommended. Genetic counseling is also vital before
pregnancy and should include a discussion of the inheritance of
Marfan syndrome, the variability of the disorder, and the possi-
bility of prenatal diagnosis. Pregnant patients with Marfan syn-
drome should be observed closely with regular aortic imaging
by echocardiography. Treatment with a β-blocker during preg-
nancy is recommended. Fetal cardiac ultrasonography is often
performed to assess the aorta and the cardiac status of the fetus.
Rare cases of severe Marfan syndrome may be diagnosed in the
fetus. The risk of inheriting Marfan syndrome is 50% for each
offspring of an affected parent. Near term, a facilitated delivery
should be planned to avoid excessive strain on the aorta. If the
aortic dimension is 40 mm or more or if it has enlarged during
pregnancy, cesarean delivery should be considered. Endocarditis
prophylaxis should be administered around the time of delivery
as indicated. The postpartum period requires special monitoring.
The risk of aortic dissection persists during this time and there is
also an increased risk of postpartum hemorrhage.
Patients with bicuspid aortic valves, coarctation of the aorta,
and other aortopathies have a predisposition for aortic dissection
and aneurysm formation, and this risk may be increased during
pregnancy. Prepregnancy aortic assessment of these patients is
imperative. In pregnancy, β-blockers may decrease the rate of
dilatation of the aortic root and should be considered for all preg-
nant patients with aortopathy. Regular aortic follow-up with ech-
ocardiography when feasible is recommended during pregnancy
in patients with aortopathy.
Peripartum Cardiomyopathy
Peripartum cardiomyopathy is defined as congestive heart fail-
ure that occurs late in pregnancy or during the early postpar-
tum period (the last trimester or up to 6 months post partum) in
the absence of congenital, coronary, or valvular heart disease or
another recognized cause of heart failure. Most commonly, it is
diagnosed during the first month post partum. The incidence in
the United States ranges from 1:1,300 to 1:15,000 live births and
is higher in certain parts of the world, such as Africa and Haiti.
Peripartum cardiomyopathy occurs more frequently in multifetal
pregnancies, multiparous women, women older than 30 years,
black women, women with gestational hypertension or pre-
eclampsia, and women treated with tocolytic therapy. The cause
is unknown, and the prognosis varies. Peripartum cardiomyop-
athy is a major cause of pregnancy-related deaths in the United
States, with a maternal mortality of 10% to 15%. Mortality is
increased in patients with persistent left ventricular dysfunction
more than 6 months after delivery. Maternal death is related to
profound left ventricular dysfunction with heart failure, throm-
boembolic events, or arrhythmias.
Improvement in left ventricular function within 6 months
after delivery is expected in 50% of women who have peripar-
tum cardiomyopathy. The recommended management is sup-
portive and includes delivery of the baby when cardiomyopathy
is identified before parturition and standard medical treatment
for congestive heart failure. Use of intravenous immune globulin,
pentoxifylline, and bromocriptine should be considered. Because
of the risk of thromboembolism, anticoagulation should also be
considered when the left ventricular ejection fraction is less than
35%. Referral for a ventricular assist device or transplant should
be considered for select patients.
Recurrence with subsequent pregnancies is common, and
patients with a history of peripartum cardiomyopathy should
avoid additional pregnancies. Subsequent pregnancy is often
associated with a decrease in ventricular function, clinical heart
failure, and death; the risk of these complications is significantly
greater for women with persistent left ventricular dysfunction at
subsequent pregnancy.
• Within 6 months after delivery, 50% of women with peripartum
cardiomyopathy have improvement in left ventricular function.
• Because recurrence of peripartum cardiomyopathy is common,
repeated pregnancy is contraindicated.
Cardiovascular Drugs in Pregnancy
The US Food and Drug Administration categorizes drugs accord-
ing to their potential to cause birth defects. The pregnancy cat-
egories depend on the reliability of documentation of fetal risk
and the potential risk-to-benefit ratio (Box 80.4).
Pharmacologic Management of Congestive
Heart Failure During Pregnancy
The treatment of congestive heart failure is more difficult in preg-
nant women than in nonpregnant women owing to the hemody-
namic changes associated with pregnancy and the limited number
of safe treatment options available. Conservative measures such
as salt restriction and limitation of activity are extremely impor-
tant. Pharmacologic therapy may be required.
Digoxin and Diuretics
Digoxin can be safely administered during pregnancy.
Box 80.4. US Food and Drug Administration Preg-
nancy Categories for Drugs
Class A—No fetal risks have been documented.
Class B—Animal studies suggest risk, but it is
unconfirmed in controlled human studies (eg,
methyldopa, thiazides, dipyridamole).
Class C—Animal studies have demonstrated
adverse fetal effects, but there have been no
controlled human studies (eg, propranolol,
digoxin, hydralazine, heparin, furosemide,
quinidine, procainamide, verapamil). These drugs
should be given only if the potential benefits
justify the risk.
Class D—Evidence of human fetal risk. These
drugs should be given only in a life-threatening
situation or for a serious disease for which safer
drugs either cannot be used or are ineffective.
Informed consent is advised when administering
these agents during pregnancy (eg, phenytoin,
captopril).
Class X—Documented fetal abnormalities; the
drug is contraindicated during part or all of
pregnancy (eg, warfarin).
 80 Pregnancy and the Heart
Diuretics impair uterine blood flow and placental perfu-
sion, but no teratogenic effects of diuretics have been described.
Cases of neonatal thrombocytopenia, jaundice, hyponatremia,
and bradycardia have been reported with the use of thiazides,
although no single diuretic is clearly contraindicated. The most
experience has been with thiazide diuretics and furosemide.
Continuing diuretic therapy that began before conception
does not seem unfavorable, but routine initiation of diuretic med-
ications during pregnancy is not generally recommended. Use
of diuretics should be limited to the treatment of symptomatic
congestive heart failure with clear evidence of elevated central
venous pressure.
Maternal use of furosemide during pregnancy has not been
associated with toxic or teratogenic effects, although metabolic
complications have been observed. Neonatal hyponatremia and
fetal hyperuricemia have been reported.
ACE Inhibitors and Angiotensin II Blockers
The use of ACE inhibitors and angiotensin II receptor block-
ers is contraindicated during pregnancy. Maternal-fetal trans-
fer of captopril has been documented; in animals, exposure to
ACE inhibitors during pregnancy has produced prolonged fetal
hypotension and death. Fetal exposure to ACE inhibitors during
the first trimester of pregnancy increased the risk of congeni-
tal cardiovascular and central nervous system malformations.
In addition, use of ACE inhibitors during the second trimes-
ter of pregnancy increases the risk of early delivery, low birth
weight, oligohydramnios, or neonatal anuria and renal failure (or
a combination of these). These agents can be used safely during
lactation.
• Administration of ACE inhibitors and angiotensin II inhibitors is
contraindicated during pregnancy, but these agents can be used
during breast-feeding.
• Digoxin and hydralazine are considered safe during pregnancy and
breast-feeding.
Nitrates
The use of organic nitrates during pregnancy has been reported
in the treatment of hypertension; however, in 1 case, the decrease
in blood pressure with nitroglycerin was associated with fetal
heart rate decelerations. Therefore, treatment with nitrates
requires further evaluation for the management of pregnancy-
related hypertension and congestive heart failure. Nitrates are
excreted in breast milk and may cause methemoglobinemia in
infants.
Management of Arrhythmias During Pregnancy
Most cardiovascular drugs cross the placenta and are secreted
in breast milk. Therefore, the risk to benefit ratio must be con-
sidered when administering any medications during pregnancy.
Cardiac arrhythmias during pregnancy should be evaluated the
same as for a nonpregnant patient and the underlying disease or
precipitating factors treated if possible.
Direct Current Cardioversion
Direct current cardioversion may be used safely during preg-
nancy. This is the treatment of choice for arrhythmias causing
hemodynamic compromise. For less urgent situations, pharma-
cologic management of arrhythmias may be required.
757
Digoxin and Quinidine
Digoxin is thought to be safe for treating arrhythmias except for
an increased risk of prematurity and intrauterine growth retar-
dation. Quinidine is an alternate antiarrhythmic medication.
Adverse fetal effects have not been reported with quinidine
given at a therapeutic dose, but toxic doses may induce prema-
ture labor. Limited information is available on the use of pro-
cainamide, disopyramide, and propafenone during pregnancy,
but no adverse fetal effects have been reported. Flecainide and
sotalol have been safely used during pregnancy to treat fetal and
maternal arrhythmias.
Amiodarone and Verapamil
The use of amiodarone during pregnancy has been reported
in several cases and may result in fetal hypothyroidism. The
manufacturer recommends against using amiodarone during
pregnancy. Amiodarone use should be limited to patients with
refractory life-threatening arrhythmias, and serum amiodarone
levels should be kept as low as possible. Fetal electrocardio-
graphic monitoring should be performed before, during, and after
birth, and neonatal thyroid function should be monitored at birth
and continued during the exposure to amiodarone. Amiodarone
and its active metabolite have been found in human breast milk
in significant concentrations; therefore, the use of amiodarone
is not recommended in women who are breast-feeding their
infants.
Verapamil has been used in pregnancy to manage supraven-
tricular arrhythmias, and no adverse effects have been reported.
However, it has been recommended that verapamil therapy be
discontinued at the onset of labor to prevent dysfunctional labor
or postpartum hemorrhage.
β-Blockers
The use of β-blockers during pregnancy has been reported to
cause intrauterine growth retardation, apnea at birth, fetal brady-
cardia, hypoglycemia, and hyperbilirubinemia. Large studies
have not confirmed these concerns, and β-blocking agents have
been used in many pregnant women without adverse effects.
β2-Blockers now are thought to be relatively safe and may be
used in the treatment of arrhythmias, hypertrophic cardiomy-
opathy, and hyperthyroidism during pregnancy if clinically
indicated. All available β-blockers cross the placenta and are
present in breast milk. These agents can reach significant lev-
els in the fetus or newborn. Therefore, if they are used during
pregnancy, it is appropriate to monitor the fetal and newborn
heart rate and blood glucose level and respiratory status after
delivery.
Adverse fetal effects have been associated with the use of
atenolol during pregnancy, especially when atenolol was admin-
istered early in the pregnancy, according to a retrospective anal-
ysis of pregnancies complicated by hypertension. Babies born
to mothers in the atenolol group weighed significantly less than
babies of mothers who used other antihypertensive monotherapy.
Among mothers using atenolol, there was also a trend for early
delivery and infants who were small for their gestational age,
especially when atenolol was administered early in pregnancy.
The World Health Organization considers atenolol unsafe
during breast-feeding because it concentrates in breast milk,
resulting in a pharmacologically significant dose to the breast-
fed infant with an associated risk for hypoglycemia and brady-
cardia. Metoprolol should be considered as an alternative.
758 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
Adenosine
Adenosine has been used successfully to treat supraventricular
tachycardia during pregnancy. To date, no adverse fetal or mater-
nal effects related to adenosine have been reported.
• Adenosine has been used safely to treat acute supraventricular
tachycardia in pregnancy.
• β -Blockers and calcium channel blockers can be used for supraven-
tricular tachycardia prophylaxis in pregnancy, but their use should
be discontinued near the time of delivery.
• Atenolol should be avoided during pregnancy and lactation.
Management of Coronary Artery Disease
During Pregnancy
Coronary artery disease is occasionally encountered in the preg-
nant population and carries a high risk. Careful evaluation is
necessary to exclude other causes of chest pain or acute coronary
syndromes such as aortic or coronary artery dissection. Treatment
depends on the presentation and the available resources. Acute
myocardial infarction has a high risk of death during pregnancy
and should be treated aggressively with acute percutaneous inter-
vention when feasible. Abdominal shielding will limit the radia-
tion exposure of the fetus.
Heparin, low-dose aspirin, nitrates, and β-blockers are all
routinely and safely used during pregnancy. The role and safety
of thrombolytic agents, glycoprotein IIb/IIIa inhibitors, and the
thienopyridines such as clopidogrel have not been established.
These agents should be reserved for emergency management.
The management of hyperlipidemia during pregnancy
requires review of the benefits and risks. Select reports have sug-
gested that statins may be teratogenic, but this was not confirmed
in other studies.
Management of Valvular Heart
Disease in Pregnancy
Fixed obstructive cardiac valve lesions, particularly mitral and
aortic valve stenosis, generally are tolerated poorly during preg-
nancy because of the blood volume expansion and the result-
ant increase in stroke volume and cardiac output. In contrast,
regurgitant valve lesions are tolerated relatively well because of
the decrease in systemic vascular resistance. The effect of preg-
nancy on a patient with valve disease depends on the specific
valve lesion, ventricular function, pulmonary artery pressures,
and NYHA class. A careful cardiovascular evaluation should
be performed before pregnancy for patients with valvular heart
disease.
Percutaneous Valve Intervention
Interventional procedures are effective alternatives to surgery in
several cardiac disorders that occur during pregnancy. Preliminary
reports are favorable; however, no large series have reported on
the safety of cardiac interventions during pregnancy.
Percutaneous mitral commissurotomy is the strategy of
choice for pregnant women who have severe mitral stenosis
and symptoms that cannot be controlled with medical therapy.
Marked relief of symptoms and excellent maternal and fetal out-
comes have been reported. This procedure should be attempted
only in centers that have extensive experience with percutaneous
mitral procedures and also have surgical backup. Special con-
siderations for balloon valvuloplasty in the gravid state include
radiation exposure and pregnancy outcome. No increase in the
incidence of reported congenital malformations or abortions
has been reported with fetal radiation exposure of less than 5
rads, which can be achieved by shielding the gravid uterus and
minimizing fluoroscopy time. Transesophageal or intracardiac
echocardiographic guidance has also been used during the pro-
cedure to reduce radiation exposure. Percutaneous balloon aortic
valvuloplasty has been reported as a safe and effective pallia-
tive procedure during pregnancy. Pulmonary balloon valvulo-
plasty is rarely required during pregnancy but has been reported.
These procedures should be considered alternatives to surgery in
patients with severe, symptomatic native valve stenosis identified
during pregnancy.
Cardiac Surgery During Pregnancy
Cardiac surgery should be reserved for patients with a refractory
response to medical management in whom further delay would
prove detrimental to maternal and fetal health. Cardiopulmonary
bypass can adversely affect both the mother and the fetus. High-
flow, high-pressure, normothermic perfusion appears safest for
the fetus. Fetal heart rate monitoring is recommended during car-
diopulmonary bypass. When cardiac surgery is necessary during
pregnancy, the optimal time is between weeks 24 and 28 of preg-
nancy, and the duration of cardiopulmonary bypass should be as
short as possible. A multidisciplinary team approach is required
to optimize maternal and fetal outcomes.
Prosthetic Heart Valves
The best type of heart valve prosthesis to use in women of child-
bearing age who have critical valvular heart disease is still being
debated. Some data have suggested that premature valve deteri-
oration occurs in bioprosthetic valves during pregnancy, but this
has not been documented conclusively or demonstrated experi-
mentally. One report has suggested that reoperation (required for
most patients with bioprosthetic valves) carries a higher risk of
morbidity and mortality than the risk of anticoagulation during
pregnancy. Recent surgical data suggest that the risk of reopera-
tion may not be substantially higher than the risk of the initial
operation, depending on patient characteristics. Pregnant women
with a mechanical heart valve have an increased risk of pros-
thetic valve thrombosis or other life-threatening complication
developing during pregnancy. The complication rate depends on
the anticoagulation regimen used. The best anticoagulation reg-
imen for a pregnant woman with mechanical valve prosthesis is
controversial.
Management of Anticoagulation
During Pregnancy
Hematologic changes that occur during normal pregnancy
include an increase in clotting factor concentrations, an increase
in platelet adhesiveness, and a decrease in fibrinolysis and pro-
tein S activity. These changes result in an overall increased risk
of thrombosis or embolism.
The American Heart Association/American College of
Cardiology and the American College of Chest Physicians have
provided recommendations for anticoagulation during preg-
nancy. In the first trimester, adjusted-dose unfractionated hep-
arin or LMWH can be used. Warfarin can also be used during
the first trimester, albeit at a slightly increased risk to the fetus.
During the second trimester, adjusted-dose unfractionated
 80 Pregnancy and the Heart
heparin, LMWH, or warfarin can be used. During the third tri-
mester, LMWH or warfarin therapy should be continued until
approximately week 37 of pregnancy. In anticipation of delivery,
patients should be hospitalized and treatment with intravenous
unfractionated heparin should be started. Unfractionated hepa-
rin is the drug of choice near the time of delivery since its use can
be stopped or its effects reversed should bleeding occur. The acti-
vated partial thromboplastin time should be more than 2.5 times
the control value.
Labor and delivery are particularly high risk for patients who
require anticoagulation during pregnancy. Delivery should be
planned and intravenous heparin treatment stopped peripartum.
Cesarean delivery should be performed if spontaneous labor
occurs during warfarin anticoagulation because of the risk of
fetal intracranial hemorrhage with vaginal delivery. Heparin
should be resumed 4 to 6 hours after cesarean or vaginal deliv-
ery in the absence of bleeding.
Thus, there is no consensus on the single best anticoagulation
regimen during pregnancy for patients with mechanical valve
prostheses (Box 80.5). Informed discussion with the patient and
her partner and meticulous monitoring of the chosen anticoagu-
lation regimen are mandatory.
Box 80.5. Anticoagulation Recommendations for
Pregnant Patients Who Have a Mechanical Heart
Valvea
Before pregnancy to week 6 of pregnancy
Warfarin
Weeks 6 to 12 of pregnancy
UFH (IV or SC) or
LMWH (SC) or
Warfarin (increased fetal risk)
Weeks 13 to 37 of pregnancy
UFH (IV or SC) or
LMWH (SC) or
Warfarin (increased fetal risk)
Week 37 of pregnancy to delivery
Stop use of SC UFH, LMWH, or warfarin
Start continuous use of IV UFH
Plan delivery
After delivery
Resume use of warfarin when bleeding is controlled
Continue use of IV UFH until INR is therapeutic
Anticoagulation monitoring
UFH—aPTT at least twice the control value
LMWH—anti-Xa 0.7–1.2 U/mL (4 h after LMWH dose)
Warfarin—INR 3 (range, 2.5–3.5)
Abbreviations: aPTT, activated partial thromboplastin time; INR,
international normalized ratio; IV, intravenously; LMWH, low-
molecular-weight heparin; SC, subcutaneously; UFH, unfractionated
heparin.
a
Recommendations are from the American Heart Association/
American College of Cardiology and the American College of Chest
Physicians.
759
Unfractionated Heparin
Unfractionated heparin does not cross the placenta. The primary
concern with heparin use during pregnancy is the increased risk
of thromboembolic complications, including fatal valve throm-
bosis, in high-risk pregnant women given subcutaneous unfrac-
tionated heparin. The efficacy of adjusted-dose subcutaneous
heparin has not been established, and for high-risk patients (those
with caged-ball or the Björk-Shiley tilting-disk mitral prosthe-
ses), this form of anticoagulation should be used only between
the 6th and 12th weeks of pregnancy and around delivery. The
heparin dose for high-risk patients should be adjusted so that the
activated partial thromboplastin time is at least 2.5 to 3.5 times
the control value 6 hours after the dose is administered. Lower-
level anticoagulation may be appropriate for patients with other
prostheses.
Prolonged unfractionated heparin therapy (intravenous or
subcutaneous) can result in thrombocytopenia, osteoporosis, and
alopecia. Erratic absorption of subcutaneously delivered heparin
may occur, so the activated partial thromboplastin time must be
monitored frequently to ensure therapeutic anticoagulation.
Warfarin
Because warfarin has a low molecular weight, it crosses the pla-
centa and results in fetal anticoagulation. The effect of warfa-
rin on the fetus is greater than that on the mother because there
are fewer vitamin K–dependent factors in the fetal liver. Fetal
anticoagulation increases the risk of spontaneous abortion, pre-
maturity, fetal deformity, and stillbirth. Retroplacental hemor-
rhage and fetal intracranial hemorrhage are additional risks to
the fetus. However, warfarin use throughout pregnancy, until
near term, provides the lowest risk of maternal thromboembolic
events, complications, and death.
Warfarin Embryopathy. Historical reports describe a 30%
risk of embryopathy with administration of warfarin during the
first trimester (weeks 6–12). More recent data suggest that the
incidence of warfarin embryopathy is less than 10%. The mater-
nal dose of warfarin during the first trimester appears to be
important, and the risk of warfarin embryopathy appears to be
low with a warfarin dose less than 5 mg daily. Warfarin embry-
opathy results in bone and cartilaginous abnormalities with
chondrodysplasia, nasal hypoplasia, optic atrophy, microphthal-
mia, blindness, minor neurologic dysfunction, reduced intelli-
gence quotient, and seizures. Warfarin does not enter breast milk
and, thus, can be administered safely to women who breast-feed
their infants.
Low-Molecular-Weight Heparin
LMWH has been used in recent years as an anticoagulation
method for patients with mechanical heart valves during preg-
nancy. No teratogenic effects have been reported with LMWH,
which does not cross the placenta. LMWH is used regularly in
patients requiring anticoagulation during pregnancy for causes
other than mechanical valves.
Concern about the safety of LMWH is based in part on an
unpublished South African study that was stopped early because
of 2 maternal deaths in the LMWH group. Although anti-Xa
levels were measured and were low in both deceased patients,
the dose of LMWH was not adjusted when the levels were low.
Subsequently, a literature review reported on 81 pregnancies in
760 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
75 women who had mechanical prostheses and were treated with
LMWH. Valve thrombosis occurred in 8.6% of these pregnan-
cies. However, when 51 pregnancies were reviewed in which anti-
Xa levels were monitored and the LMWH dose was adjusted,
only 1 thromboembolic event occurred. Another study demon-
strated that LMWH requirements increase during pregnancy and
that weight-based administration of LMWH is inadequate during
pregnancy, so the dose should be adjusted according to anti-Xa
levels. The ideal peak anti-Xa level 4 hours after administration
is about 1.0 U/mL.
Antiplatelet Agents
Low-dose aspirin (81 mg) is safe to use during pregnancy. It is
recommended for patients with intracardiac shunts (eg, those with
atrial septal defect), cyanosis, or a valve prosthesis. However, the
antiplatelet effect has not been proved.
Dipyridamole should not be used during pregnancy. Limited
data are available on the effects of ticlopidine or clopidogrel dur-
ing pregnancy; these agents are not recommended. Information is
limited on administration of glycoprotein IIb/IIIa inhibitors dur-
ing pregnancy. Thrombolytic therapy has been used in pregnancy
and may be the recommended treatment of choice for pregnant
patients with mechanical valve thrombosis. Operation is recom-
mended when patients have contraindications for thrombolytic
therapy depending on the valve involved and thrombus burden.
Thrombolytic therapy should be considered in the critically ill
patient with acute coronary syndrome when urgent percutaneous
intervention is not available or feasible.
• At approximately week 37 of pregnancy, patients with mechanical
valve prostheses should be hospitalized for adjusted-dose intrave-
nous unfractionated heparin; treatment is interrupted around the
time of delivery.
• Use of warfarin during the first trimester of pregnancy is associ-
ated with an increased risk of miscarriage and warfarin embryopa-
thy but may be the safest method of anticoagulation for the mother,
especially with older mechanical mitral prostheses.
• Adjusted-dose unfractionated heparin administered between
weeks 6 and 12 of pregnancy decreases the risk of fetal complica-
tions but doubles the risk of maternal thromboembolism and death
compared with warfarin.
• LMWH adjusted to a therapeutic anti-Xa level is a recognized
method of anticoagulation in pregnant women with mechanical
prostheses.
• Considerable controversy exists about the best method of antico-
agulation during pregnancy.
Endocarditis Prophylaxis
The American Heart Association does not recommend endocar-
ditis prophylaxis for patients expected to have an uncomplicated
cesarean or vaginal delivery. However, standard prophylactic
treatment with antibiotics given intravenously or intramuscularly
is recommended for the placement of a urinary catheter in the
presence of urinary tract infection and for vaginal delivery in the
presence of vaginal infection.
Mayo Clinic’s recommendation, which is more conservative,
is to give antibiotics intravenously for endocarditis prophylaxis
(using the gastrointestinal or genitourinary regimen) to all high-
risk cardiac patients because of the risk of undiagnosed infec-
tions and the patient morbidity and mortality that result from
infective endocarditis. Antibiotic therapy should be adminis-
tered 30 to 60 minutes before delivery is expected and repeated
8 hours later.
Contraception in Patients With Heart Disease
More than 50% of teenagers are sexually active and 10% of the
women in the United States who are 15 to 19 years old have
unplanned pregnancies.
Use of the estrogen-containing oral contraceptive pill, or
“combination pill,” increases the risk of thromboembolic events,
pulmonary embolism, and fluid retention; therefore, it should be
prescribed with caution to women with severe structural heart
disease. Alternative methods include the progesterone-only
pill, or “mini pill,” for patients with pulmonary hypertension,
right-to-left shunts, or a prosthetic valve. The failure rate of the
progesterone-only pill is higher than that of the combination pill
and is similar to the failure rate of barrier methods. Also, the
progesterone-only pill must be taken at the same time each day.
Breakthrough bleeding is common and, when this occurs, con-
traceptive coverage is not reliable. Barrier methods also have a
high failure rate (18% per year) and should be used with cau-
tion in patients in whom pregnancy is absolutely contraindicated.
An intrauterine device is generally not suggested for women
with heart disease, because of the potential risk of infection.
Medroxyprogesterone acetate is a synthetic progestogen that is
administered as an intramuscular injection every 3 months. It is
generally well tolerated in patients with cardiovascular disease.
Tubal ligation should be reserved for women in whom pregnancy
is absolutely contraindicated and transplant is not possible, since
successful pregnancy has been reported in women after heart-
lung transplant. The Essure endovaginal tubal ligation is another
option for high-risk cardiac patients. This procedure can be
performed with low risk in high-risk cardiac patients and is an
effective form of pregnancy prevention.
Abbreviations
ACE angiotensin-converting enzyme
LMWH low-molecular-weight heparin
NYHA New York Heart Association
 81
Adult Congenital Heart Disease
NASER AMMASH, MD, and CAROLE A. WARNES, MD
Introduction
Remarkable improvement in outcomes for patients with CHD
has occurred over the past half century owing to better diag-
nostic tools leading to early recognition and enhanced surgical
repair. Now it is estimated that 85% of all newborns with CHD
will reach adulthood. Given the growing population of adults
with CHD (increasing by 5% per year), there are more adults
than children with CHD. These patients are divided into 2 broad
categories: natural selection and unnatural selection.
Natural selection includes patients who have mild congeni-
tal cardiac lesions that did not require surgery during infancy
or childhood. Examples are bicuspid aortic valve, ASD, pulmo-
nary valve abnormalities, and Ebstein anomaly. Also included
are patients who have inoperable CHD (eg, Eisenmenger
syndrome).
Unnatural selection includes a larger group of patients who
have had prior palliative or complete surgical repair of the CHD
and who require lifelong follow-up to identify the residua and
sequelae of surgery, including any of the following:
1. Progression of a previously insignificant lesion
2. Degeneration of a prosthetic valve, intracardiac conduit, or intracar-
diac repair
3. New surgical options related to advancements in congenital cardiac
surgery
4. Postoperative medical complications, including infective endocardi-
tis, ventricular dysfunction, and arrhythmias
Several syndromes are associated with CHD (Box 81.1). Adult
CHD survivors have challenging problems and special needs that
require expert care. It is imperative to recognize the importance
of this patient population and to deliver optimal health care
Abbreviations and acronyms are expanded at the end of this chapter.
(Box 81.2). This chapter discusses the most common types of
CHD, associated hemodynamic disturbances, and complications
before and after repair.
Atrial Septal Defect
ASDs are common congenital heart defects in adults who may
present with dyspnea, exercise intolerance, atrial arrhythmias,
paradoxical embolism, or cardiomegaly. The majority of ASDs
are ostium secundum defects (65%–75%); others are ostium
primum defects (partial atrioventricular canal) (15%–20%) and
sinus venosus defects (5%–10%). Coronary sinus ASD is rare.
All isolated ASDs result in a left-to-right shunt across the atrial
septum. The ensuing increase in pulmonary blood flow can be
as high as 3 to 4 times the systemic blood flow, with associated
secondary pulmonary artery enlargement and hypertension.
However, Eisenmenger syndrome is an uncommon complication
of isolated ASD.
Secundum ASD
The secundum type of ASD is the most common adult con-
genital heart defect after bicuspid aortic valve. The defect is in
the central portion of the atrial septum and is associated with
left-to-right shunting and right ventricular volume overload.
Adults are often asymptomatic, and the murmur may be heard
incidentally on physical examination. The natural history of
ASD depends on the size of the shunt created by the defect and
on associated anomalies, which include anomalous pulmonary
venous connection, VSD, and PS. In patients who have unre-
paired ASD and who are in their 40s or 50s, atrial fibrillation
develops in association with tricuspid regurgitation due to annu-
lar dilatation and often right ventricular failure. Death is usually
761
762 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

deviation. In addition, a notch (crochetage) is sometimes noted

Box 81.1. Syndromes Associated With CHD
Down syndrome: atrioventricular septal defects
(atrioventricular canal, primum ASD), ventricular
septal defect
Turner syndrome: coarctation of the aorta,
bicuspid aortic valve
Holt-Oram syndrome: secundum ASD
Marfan syndrome: aortic dilatation, dissection,
and rupture; mitral valve prolapse
Noonan syndrome: pulmonary stenosis and
hypertrophic cardiomyopathy
Williams syndrome: supravalvar aortic or
pulmonary stenosis, coarctation of the aorta,
renal artery stenosis
Shone syndrome: serial left-sided obstructive
CHD, including coarctation of the aorta,
subaortic stenosis, parachute mitral valve, or
supravalvar mitral ring
Abbreviations: ASD, atrial septal defect; CHD, congenital heart disease.
from heart failure or thromboembolic stroke, although some
patients survive into old age.
Physical Examination
The jugular venous pressure is often normal, and the a and v
waves may be equal in amplitude. Other findings are a right ven-
tricular lift, an ejection systolic murmur from the pulmonary
area (always less than grade 3/6), fixed splitting of the second
heart sound, and a tricuspid diastolic flow rumble if the shunt is
large (Qp/Qs >2.5:1).
in the QRS complex in leads II and III.
Chest Radiography
A prominent pulmonary artery, right ventricular enlarge-
ment, and pulmonary plethora are found on chest radiographs
(Figures 81.1 and 81.2). Left atrial enlargement is not typically
present unless the patient is older than 40 years or has atrial
fibrillation. If left atrial enlargement is present in a young person
with an ASD in sinus rhythm, consider a primum ASD in the
differential diagnosis or coincidental mitral valve disease.
Evaluation and Management
The diagnosis of ASD can be made with transthoracic echocar-
diography, but if the image quality is suboptimal, transesopha-
geal echocardiography should be performed because it improves
visualization of the atrial septum and enhances diagnostic
accuracy. Echocardiography is also used to assess the severity
of the shunt and the presence and severity of associated CHD,
pulmonary hypertension, and tricuspid regurgitation. Cardiac
catheterization is usually unnecessary to confirm an ASD unless
coexistent coronary artery disease is suspected or pulmonary
vascular resistance and reversibility of pulmonary hypertension
need to be assessed before committing to ASD repair.
If right ventricular volume overload has been identified
by examination, chest radiography, or echocardiography, the
ASD should be closed to prevent right heart failure, paradoxi-
cal embolus, and atrial arrhythmia. (Closing an ASD later in
life, although still beneficial, is associated with an increased
risk of atrial arrhythmias during late follow-up.) Percutaneous
device closure of an isolated secundum ASD is preferred treat-
ment in experienced hands, as long as the defect has an adequate
rim. Surgical repair should be considered for a large defect
(>3 cm), multiple defects, hemodynamically significant tricuspid

Electrocardiography
Typical electrocardiographic findings include an RSR′ pattern
and partial right bundle branch block, often with right axis

Box 81.2. Indications for Endocarditis Prophylaxis

Before Dental Procedures (American Heart Association
Recommendations)
Prosthetic cardiac valves, including bioprosthetic
and homograft valves and use of prosthetic
material for cardiac valve repair
Previous infective endocarditis
Unrepaired and palliated cyanotic CHD,
including CHD in patients with surgically
constructed palliative shunts and conduits
Within 6 months after complete repair of CHD with
prosthetic material or device by either surgery or
catheter intervention
Inhibited endothelialization after repair of CHD
that has residual defects at the site or adjacent to
the site of a prosthetic patch or prosthetic device Figure 81.1. Chest radiograph showing mild cardiomegaly with
prominence of the pulmonary artery (arrow). There are increased pul-
Abbreviation: CHD, congenital heart disease. monary vascular markings from a left-to-right shunt consistent with a
significant atrial septal defect.
 81 Adult Congenital Heart Disease
Figure 81.2. Chest radiograph from a 35-year-old woman with atrial
septal defect. There is severe pulmonary vascular obstructive disease
with prominence of the main pulmonary artery and decreased periph-
eral pulmonary vascularity. (Previously published. See “Credit Lines”
section.)
regurgitation, anomalous pulmonary venous connection, or
atrial arrhythmias. Symptoms and right ventricular size usually
improve after closure.
• The secundum type of ASD is the most common adult congenital
heart defect after bicuspid aortic valve.
• Cardiac catheterization is usually unnecessary to confi rm an ASD
unless coexistent coronary artery disease is suspected or severe
pulmonary hypertension is present.
• The presence of symptoms, atrial arrhythmias, or signs of right
ventricular volume overload are indications for repair.
• Some secundum ASDs are best repaired surgically.
Primum ASD
The primum type of ASD involves the lower portion of the atrial
septum and is actually a defect in the atrioventricular septum.
Both atrioventricular valves are on the same anatomical level
and are congenitally structurally abnormal. Classically, the
mitral valve is cleft, and 4% of defects may be associated with a
double-orifice mitral valve. The mitral valve, therefore, has vari-
ous degrees of regurgitation and may occasionally be stenotic.
The septal leaflet of the tricuspid valve often is deficient, with
various degrees of tricuspid regurgitation. The findings on
physical examination in primum ASD are the same as those for
secundum ASD, with the addition of various degrees of mitral
regurgitation.
Electrocardiography
Electrocardiography shows left axis deviation with right bun-
dle branch block. First degree atrioventricular block occurs in
approximately 75% of patients. Left axis deviation is a charac-
teristic of this defect.
763
Evaluation and Management
Primum ASD is diagnosed with echocardiography, and cardiac
catheterization is usually unnecessary. The elongated ventricu-
lar outflow tract results in the so-called gooseneck deformity on
angiography. Subaortic stenosis is a recognized association.
Patch closure is used for a primum ASD. Repair of the associ-
ated cleft mitral valve is indicated if symptoms, atrial arrhyth-
mias, right ventricular volume overload, or hemodynamically
significant mitral regurgitation is present.
• The findings on physical examination in primum ASD are the same
as those for secundum ASD, with the addition of various degrees of
mitral regurgitation.
• Primum ASD should be considered when a patient with ASD has
left axis deviation on electrocardiography or cleft mitral valve on
echocardiography.
Sinus Venosus ASD
Sinus venosus ASD is a defect in the superior fatty portion of the
atrial septum near the superior vena cava–right atrial junction. It
is commonly associated with an anomalous pulmonary vein con-
nection. Classically, the right upper pulmonary vein connects to
the superior vena cava. Occasionally, the right middle pulmonary
vein connects anomalously to the right atrium.
Evaluation and Management
The diagnosis usually can be made with echocardiography.
Often transesophageal echocardiography is needed to delineate
the defect and to assess an anomalous pulmonary vein connec-
tion. Computed tomography and magnetic resonance imaging
could be considered for assessing an anomalous pulmonary vein
connection. Surgical repair should include closing the defect and
redirecting the anomalous pulmonary vein connection into the
left atrium.
• Eisenmenger syndrome develops in approximately 5% of patients
with ASD.
• The presence of pulmonary hypertension or severe volume over-
load of the right ventricle in association with a small ASD should
raise suspicion for an anomalous pulmonary vein connection,
which has important implications for optimal management.
Ventricular Septal Defect
Defects can occur at many different areas in the ventricular sep-
tum, the most common being in the membranous and muscular
parts of the septum. Muscular defects can be single or multiple.
A small VSD, as in Roger disease, often produces a loud murmur
but has little or no clinical significance apart from the need for
periodic assessment. Larger membranous or muscular defects
may close spontaneously up to age 20 years. In contrast, subaortic
defects (also referred to as supracristal or subpulmonary defects)
and canal-type defects never close spontaneously (Figure 81.3).
Subaortic defects are located more anteriorly and are commonly
associated with progressive aortic regurgitation due to prolapse
of the right or left coronary cusp into the VSD. Canal-type or
inlet defects are near the mitral and tricuspid valves, which are
typically structurally normal.
Physical Examination
Small defects may be associated with a thrill at the left sternal
edge, usually in the fourth interspace. The murmur is usually
764 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
Figure 81.3. Chest radiograph from a 22-year-old man with ven-
tricular septal defect complicated by pulmonary vascular obstruc-
tive disease. Note the nearly normal cardiac size with the apex tilted
upward, the marked dilatation of the central pulmonary arteries, and the
decreased peripheral pulmonary vascularity. (Previously published. See
“Credit Lines” section.)
holosystolic, but it may be shorter if it is in the muscular septum
because the defect may be occluded during late systole. Large
defects also may produce a mitral diastolic flow rumble at the
apex (particularly with Qp/Qs >2.5:1). A diastolic murmur of
aortic regurgitation may be heard with subaortic defects. A sys-
tolic ejection murmur at the left sternal border should raise suspi-
cion for pulmonary stenosis (valvular or subvalvular).
Electrocardiography
Electrocardiography shows various degrees of left ventricular
enlargement, depending on the volume of the shunt and the asso-
ciated aortic regurgitation.
Evaluation and Management
Echocardiography is the diagnostic method of choice for assess-
ing VSD, associated defects, and hemodynamic disturbances.
Cardiac catheterization is recommended for hemodynamically
significant pulmonary hypertension to confirm severity and
assess reversibility. The presence of unexplained left ventricular
volume overload is the most common indication for VSD repair.
Although percutaneous device closure of membranous and mus-
cular VSD has been reported, surgical repair is the treatment of
choice in adults. Late postoperative heart block has been reported
after VSD repair.
• Large VSDs result in pulmonary hypertension and Eisenmenger
syndrome in the absence of right ventricular outflow obstruction
such as pulmonary stenosis.
• VSD is the most common congenital heart defect to produce
Eisenmenger syndrome.
• The typical holosystolic murmur of VSD disappears in Eisenmenger
syndrome. It is replaced by a loud, sometimes palpable, pulmo-
nary closure sound and often a diastolic murmur of pulmonary
regurgitation.
Patent Ductus Arteriosus
PDA, associated with maternal rubella, is a congenital abnor-
mality in which a vessel connects the left pulmonary artery to
the proximal descending aorta just distal to the origin of the left
subclavian artery. A small PDA may be of little or no hemody-
namic consequence and is consistent with a normal life span. A
PDA commonly calcifies in adult life. A moderately sized PDA
leads to volume overload of the pulmonary arteries, left atrium,
and left ventricle. A large unrepaired PDA can cause pulmonary
hypertension and Eisenmenger syndrome.
Physical Examination
Because PDA produces an arteriovenous fistula, the pulse pres-
sure is usually wide, with a prominent left ventricular impulse
and a continuous machinery murmur enveloping the second heart
sound. This murmur usually is audible beneath the left clavicle
in the second intercostal space. In Eisenmenger syndrome, the
murmur disappears owing to pressure equalization between the
systemic and pulmonary circulations. However, the patient may
show differential cyanosis of the lower extremities owing to the
right-to-left shunt in the proximal descending aorta distal to
the origin of the left subclavian artery.
Evaluation and Management
PDA can be readily identified with transthoracic echocardiog-
raphy and should be considered whenever a machinery or con-
tinuous murmur is present. The differential diagnosis of PDA
includes pulmonary atresia with systemic collateral vessels in a
patient with cyanosis, aortopulmonary window, and a VSD with
concomitant aortic regurgitation. Other fistulae, such as coronary
and pulmonary fistulae, may also produce continuous murmurs
and, rarely, coarctation. All these cardiovascular abnormalities
could also be assessed by a comprehensive echocardiographic
examination. However, when coronary fistulae are suspected,
computed tomographic imaging or coronary angiography (or
both) is recommended.
Elective repair of PDA is indicated for unexplained left ven-
tricular volume overload. Percutaneous device closure of the duc-
tus is the most common therapy; surgical ligation and division is
an alternative if the ductus is not amenable to device therapy.
Pulmonary Stenosis
PS may be secondary to bicuspid valve or associated with
Noonan syndrome, in which case the valve is frequently dysplas-
tic. Unless the PS is severe, it usually produces few or no symp-
toms of dyspnea, fatigue, chest pain, syncope, or arrhythmias.
The valve usually remains pliable until well into middle age;
thus, an ejection click is common until later life.
Physical Examination
The following features suggest PS: prominent a wave in the jugu-
lar venous pulse, right ventricular lift, ejection click (the earlier
the click, the more severe the stenosis), systolic murmur, and
delayed P2 (absent in severe pulmonary stenosis). The ejection
 81 Adult Congenital Heart Disease
click of pulmonary valve stenosis is louder in expiration (there is
more systolic excursion of the leaflets).
Electrocardiography
In severe PS, the electrocardiogram typically shows right ven-
tricular hypertrophy. In mild or moderate PS, the electrocardio-
gram may be normal.
Chest Radiography
When the stenosis is at the level of the valve, the main and
left pulmonary arteries have poststenotic dilatation, which is a
typical radiographic appearance (Figure 81.4). Lung fields are
oligemic only in severe PS.
Evaluation and Management
The diagnosis is made from the findings on clinical examina-
tion and echocardiography. Right ventricular enlargement should
raise suspicion for concomitant pulmonary regurgitation or an
intracardiac shunt such as ASD. In severe PS, the gradient is usu-
ally more than 50 mm Hg (with right ventricular pressure more
than two-thirds of the systemic pressure). Most patients have
good results from balloon valvuloplasty. Often, coexistent sub-
pulmonary stenosis is due to hypertrophy of the infundibulum,
which usually regresses with time after pulmonary valvuloplasty.
Mid-term results appear to be comparable to those with surgical
commissurotomy; hence, balloon valvuloplasty is now the proce-
dure of choice for treatment of PS. Various degrees of pulmonary
regurgitation have been noted after valvuloplasty, and therefore,
long-term follow-up is recommended.
• Pulmonary stenosis may be associated with Noonan syndrome, in
which case the valve is frequently dysplastic.
• Patients are asymptomatic until isolated PS is severe, when the
gradient is usually >50 mm Hg.
Figure 81.4. Chest radiograph showing severe pulmonary stenosis.
Note the poststenotic dilatation of the main and left pulmonary arteries
(arrow), indicating that this is valvular pulmonary stenosis.
765
• Balloon valvuloplasty is the procedure of choice for treatment
of PS.
Coarctation of the Aorta
CoA is usually a discrete infolding or narrowing of the descend-
ing aorta just distal to the origin of the left subclavian artery.
Occasionally an elongated, narrowed thoracic aorta involves the
arch. Patients with CoA can present at any age, but CoA is most
often diagnosed in childhood because of an incidental heart mur-
mur or hypertension. CoA is much more common in males than
females and often is associated with a bicuspid aortic valve. It
is the most common anomaly associated with Turner syndrome.
CoA may be associated with VSD, Shone syndrome, and cerebral
aneurysms in the circle of Willis.
The major complications of CoA include aortic rupture or
dissection, coexistent aortic valve disease, left ventricular fail-
ure, stroke (due to either systemic hypertension or rupture of a
cerebral aneurysm), endocarditis, and endarteritis. Pregnancy in
patients with CoA is associated with an increased risk of aor-
tic dissection and rupture. The risk of aortic dissection and rup-
ture also is increased in Turner syndrome, even in the absence
of CoA.
Physical Examination
Since the aortic narrowing in CoA is often distal to the left
subclavian artery, the narrowing causes systemic hyperten-
sion of the upper extremities and decreased blood pressure and
delayed pulsation below the coarctation (in the legs). Therefore,
the findings on physical examination include radiofemoral or
brachiofemoral delay and a decrease in blood pressure between
the upper and lower extremities. An ejection systolic murmur
is heard in the second space at the left sternal edge. The mur-
mur occasionally extends into diastole, producing a continuous
murmur when the coarctation is severe. Collateral murmurs
may be audible and palpable over the thorax, particularly over
the scapulae. An ejection click may be heard when there is an
associated bicuspid aortic valve. The aortic valve component of
the second heart sound may be loud, and a fourth heart sound
may be heard if systemic hypertension is present. Sometimes
patients present because of signs of hypertension on retinal
examination.
Electrocardiography
Electrocardiography shows various degrees of left ventricular
hypertrophy depending on the severity of hypertension or aor-
tic regurgitation (or both) in association with a bicuspid aortic
valve.
Chest Radiography
Chest radiographs may show the classic figure 3 sign beneath the
aortic knob, which represents a dilatation of the aorta above the
coarctation and a dilatation below the coarctation. This is uncom-
mon, however. Rib notching is a variable feature (Figure 81.5).
Evaluation and Management
Coarctation of the aorta is diagnosed from findings on physi-
cal examination and Doppler echocardiography (Figure 81.6).
In hemodynamically significant CoA, Doppler echocardiog-
raphy of the abdominal aorta shows decreased and delayed
 A B

Figure 81.5. A, Coarctation of the aorta with the figure 3 sign (arrow) along the upper aspect of the left cardiac silhouette; indentation just below
the aortic arch represents the coarcted segment with a poststenotic dilatation below it. B, No other characteristics of aortic coarctation are present
except notching (arrow) beneath the undersurface of the ribs, evident on the left side. (Previously published. See “Credit Lines” section.)

3.5 m/s 5 m/s

Figure 81.6. Continuous wave Doppler echocardiographic recordings down the descending aorta in a patient with coarctation of the aorta. Left,
Resting recording with a peak systolic velocity of 3.5 m/s indicates a maximal instantaneous gradient of approximately 49 mm Hg. Note the persis-
tence of high velocity (approximately 1 m/s) in diastole (arrow). Right, With exercise, the peak velocity increases to 5 m/s (maximal instantaneous
gradient, 100 mm Hg) and the diastolic flow increases to 2 m/s. These measurements are consistent with severe coarctation. (Previously published.
See “Credit Lines” section.)
 81 Adult Congenital Heart Disease
systolic flow and increased continuous diastolic flow. In con-
trast, Doppler echocardiography of the proximal descending
thoracic aorta at rest may show a systolic and diastolic gradi-
ent. Exercise Doppler echocardiography may improve the diag-
nostic accuracy. If the CoA is not well visualized, the gradient
must be interpreted with caution because collateral vessels may
reduce the gradient even when significant CoA is present. If
Doppler echocardiographic imaging is not satisfactory, con-
sider magnetic resonance imaging, computed tomography, or
aortography.
In experienced hands, therapy with percutaneous balloon
angioplasty and stent placement is successful in appropriately
selected patients with discrete noncalcified CoA. However, per-
cutaneous repair is less successful in patients with higher gra-
dients or significant arch hypoplasia, and it may be complicated
by dissection, rupture, and recoarctation. Surgical repair, often
resection with end-to-end anastomosis by left lateral thora-
cotomy, is a proven and accepted treatment. Recoarctation can
also occur after surgical repair. There is a high incidence of
systemic hypertension (75% at 30 years) after CoA repair espe-
cially when performed in adults. Even after successful repair of
coarctation, the aorta is still abnormal and patients are still at
increased risk of dying of dissection and rupture. They also die
of premature coronary artery disease, heart failure, or stroke.
The younger the patient’s age at repair, the less chance of sys-
temic hypertension and related complications. In a Mayo Clinic
series, patients who had an operation when they were younger
than 14 years had a 20-year survival rate of 91%, and patients
who had an operation at age 14 years or older had a 20-year
survival rate of 79%.
• Coarctation is much more common in males than females and
commonly is associated with a bicuspid aortic valve.
• Coarctation should be easily suspected by physical examination
and should be considered in the presence of hypertension.
• The major complications of CoA include aortic rupture or dissec-
tion, coexistent aortic valve disease, left ventricular failure, stroke
(due to either systemic hypertension or rupture of a cerebral aneu-
rysm), endocarditis, and endarteritis.
• In the presence of concomitant coronary artery disease, valvular
disease, or ascending aortic aneurysm, placing an ascending-to-
descending aortic bypass and repairing the concomitant cardiac
abnormalities is an alternative surgical option.
I aVR
II aVL
III aVF
Figure 81.7. Electrocardiogram from a 15-year-old boy with Ebstein anomaly. Note the prominent P waves, prolonged atrioventricular conduc-
tion, and delay in right ventricular conduction.
767
Ebstein Anomaly
Ebstein anomaly of the tricuspid valve results from failure of
1 or more tricuspid valve leaflets to completely delaminate or
separate from the right ventricular myocardium during embryo-
genesis. As a result, the major abnormality in Ebstein anomaly is
inferior apical displacement of the septal and posterior tricuspid
valve leaflets into the right ventricle, producing an “atrialized”
right ventricle above and a small, often dysplastic, right ventricle
below. The degree of displacement is variable, as is the degree of
abnormality of the tricuspid valve: the septal leaflet is variably
deficient or absent, the posterior leaflet often is deficient, and the
anterior leaflet is large and sail-like. Among patients with Ebstein
anomaly, 50% have either a patent foramen ovale or a secundum
ASD, and 25% have 1 or more accessory atrioventricular conduc-
tion pathways (Wolff-Parkinson-White syndrome). The anomaly
is thought to be associated with maternal lithium ingestion.
Physical Examination
Physical examination findings include a low-volume pulse with
cool extremities and sometimes peripheral cyanosis reflecting
low cardiac output. Central cyanosis may be present if there is
an atrial communication. A v wave may be present in the jugu-
lar venous pulse, although this is uncommon even with hemo-
dynamically significant tricuspid regurgitation because the large
right atrium can contain the regurgitant volume. There is a subtle
right ventricular lift. The loud tricuspid valve component of the
first heart sound is produced by the sail-like anterior leaflet of the
tricuspid valve. The holosystolic murmur of tricuspid regurgita-
tion is often associated with 1 or more systolic clicks in Ebstein
anomaly.
Electrocardiography
Right atrial enlargement produces tall P waves, which may be
larger than those in any other anomaly (Himalayan P waves)
(Figure 81.7 Right bundle branch block often is present, or there
may be evidence of preexcitation.
Chest Radiography
Various degrees of cardiomegaly with marked right atrial enlarge-
ment are seen on chest radiographs (Figure 81.8). In contrast to
V1 V4
V2 V5
V3 V6
768 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
Figure 81.8. Chest radiograph from a patient with severe Ebstein
anomaly, severe right atrial enlargement, and huge cardiomegaly with
a very narrow pedicle.
cardiomyopathy, in Ebstein anomaly the pedicle is very narrow
(because the pulmonary artery is small). Usually the lung fields
are normal or oligemic. A large pericardial effusion could cause
similar radiographic findings.
Evaluation and Management
The diagnosis of Ebstein anomaly is made with echocardiog-
raphy. Cardiac catheterization is unnecessary unless coexistent
coronary artery disease is suspected. Indications for surgical
repair are the following:
1. Patients have functional limitation or deteriorating exercise capacity
on stress testing
2. Progressive right ventricular enlargement or dysfunction is demon-
strated on echocardiography or magnetic resonance imaging
3. An atrial communication is present and the patient is cyanotic (risk
of stroke)
4. The patient has severe tricuspid regurgitation
Surgical repair of Ebstein anomaly should also be considered
when early signs of left ventricular dysfunction are noted. If an
accessory bypass tract is present, mapping and radiofrequency
ablation should be performed preoperatively or intraoperatively
(or at both times).
Surgical repair consists of closing the atrial communication,
repairing the tricuspid valve if there is sufficient mobility to the
anterior leaflet, and plicating the atrialized right ventricle. If the
valve is tethered and immobile, tricuspid valve replacement may
be necessary.
• Among patients with Ebstein anomaly, 50% have either a pat-
ent foramen ovale or a secundum ASD, and 25% have 1 or more
accessory atrioventricular conduction pathways (Wolff-Parkinson-
White syndrome).
• Globular cardiomegaly on chest radiography with narrow pedi-
cle and normal or oligemic lung fields should raise suspicion for
Ebstein anomaly.
• Right ventricular dysplasia occurs in all patients with Ebstein
anomaly and causes various degrees of right ventricular dysfunc-
tion. Magnetic resonance imaging is a reliable technique used for
assessment of right ventricular size and function.
Congenitally Corrected Transposition
of the Great Arteries
Congenitally corrected transposition of the great arteries
(levotransposition) is an anomaly in which there is atrioven-
tricular discordance and ventriculoarterial discordance. That is,
blood from the right atrium passes through a mitral valve into
the morphologic left ventricle, which ejects blood into the pul-
monary artery. Blood from the left atrium passes through a tri-
cuspid valve into the morphologic right ventricle, which ejects
blood into the aorta. Thus, the blood flows correctly (hence,
the term corrected transposition) but through the wrong cham-
bers. The morphologic right ventricle is the systemic ventricle,
and the morphologic left ventricle is the subpulmonic ventricle.
Atrioventricular valves enter the appropriate ventricle; the flimsy
tricuspid valve sits at the mouth of the right ventricle in the sys-
temic circulation. The coronary artery pattern is also reversed.
The commonly associated anomalies are VSD, PS, dextrocar-
dia, and abnormalities of the left atrioventricular valve (tricuspid
valve), which is usually regurgitant (as in Ebstein anomaly). If
VSD and PS are present, the patient will be cyanotic.
Complete heart block is a common association. Up to 30%
of patients need permanent pacing. Patients may survive to their
50s or 60s if they have congenitally corrected transposition and
no associated anomalies, but problems often occur because the
morphologic right ventricle supports the systemic circulation and
is prone to failure. With an associated defect, particularly left
atrioventricular valve regurgitation, patients usually present ear-
lier in life with atrial fibrillation and left-sided heart failure. The
presence of PS and VSD produces various degrees of cyanosis
(depending on the severity of the pulmonary stenosis), and the
clinical features may resemble those of TOF.
Electrocardiography
Electrocardiographic findings include the absence of Q waves in
the left precordial leads and the presence of Q waves in the right
precordial leads (QR pattern in leads II, III, and V1). Complete
heart block may be present; its incidence increases with age.
Chest Radiography
A straight left aortic border is seen on chest radiographs (because
the aorta does not ascend on the right and the pulmonary trunk
does not form a border on the left). The ventricle may show a
humped contour in the region of the left atrial appendage.
Dextrocardia may be apparent (Box 81.3).
Evaluation and Management
Congenitally corrected transposition of the great arteries has
characteristic echocardiographic features, including disconti-
nuity between the aortic and tricuspid valves on the parasternal
long-axis view and an anteriorly displaced aortic valve to the left
of the posterior pulmonary valve on the parasternal short-axis
view. The apical view shows the tricuspid valve in continuity
 81 Adult Congenital Heart Disease
Box 81.3. Conditions Associated With Dextrocardia
With Cardiac Apex on 1 Side and Gastric Bubble on the
Other on Radiographic Evaluationa
Congenitally corrected transposition of the great
arteries
Single ventricle
a
If both the cardiac apex and the gastric bubble are on the right, the
cardiac anatomy may be normal (ie, mirror-image reversal).
with the left atrium and hypertrophied morphologic right ven-
tricle. Various degrees of tricuspid regurgitation occur.
If a VSD is present, it should be surgically closed. Relief of
pulmonary stenosis may be difficult because of access problems
and the danger of producing heart block or damage to the right
coronary artery. Therefore, a conduit is often necessary. The left
atrioventricular valve (tricuspid valve) cannot be repaired. If
regurgitation is moderate or severe, the valve must be replaced
before the vulnerable systemic right ventricle deteriorates.
Cyanotic Heart Disease
Many patients with cyanotic CHD reach adulthood without hav-
ing had any intervention or surgery. These natural survivors
include patients with TOF, anatomical variants with a very large
interventricular shunt (single ventricle), pulmonary stenosis with
ASD, Ebstein anomaly with ASD, or Eisenmenger syndrome.
Tetralogy of Fallot
One of the so-called conotruncal abnormalities, TOF consists of
a large subaortic VSD and obstruction to the pulmonary outflow,
usually at the infundibular level and often at the pulmonary valve
level also. This produces right ventricular hypertrophy. In addi-
tion, the aorta overrides the VSD. It can also be associated with
ASD, right-sided aortic arch, and anomalous coronary artery.
In TOF, the right ventricular pressure is the same as the left
ventricular pressure because of the large VSD. The obstruc-
tion to pulmonary blood flow causes desaturated blood to be
diverted into the aorta, which is often large; thus, the degree of
right ventricular obstruction determines the degree of cyanosis.
Hence, in childhood, the so-called acyanotic TOF or pink TOF
occurs when there is little obstruction to pulmonary blood flow
and patients do not have cyanosis. The infundibular hypertro-
phy, however, tends to be progressive; thus, cyanosis occurs and
increases with advancing age.
Most patients have repair in childhood, but occasionally
they reach adulthood without surgical intervention. These adult
patients do not have right ventricular failure until they are at least
40 years old, unless they have a superimposed arrhythmia.
Physical Examination
Findings on physical examination include various degrees of
cyanosis and clubbing, a right ventricular lift, a thrill at the left
sternal edge if the pulmonary obstruction is severe, a long sys-
tolic murmur in the pulmonary area, and absent P2.
Adult patients who have not had an operation may have aortic
regurgitation because the aorta is large and the cusps prolapse
into the defect. The degree of aortic regurgitation may vary. If
769
the aortic regurgitant jet enters the right ventricle, right ventricu-
lar failure may develop.
Electrocardiography
TOF is associated with right ventricular hypertrophy, usually
right axis deviation, and tall, peaked P waves.
Chest Radiography
TOF produces a characteristic coeur en sabot or boot–shaped
heart on chest radiography. Findings include right aortic arch
in approximately 25% of patients (Box 81.4), right ventricular
enlargement, concave pulmonary bay, and, possibly, pulmonary
oligemia.
Evaluation and Management
The diagnosis can usually be made with echocardiography. If
surgical correction is being contemplated, the coronary artery
anatomy should be determined because of the increased inci-
dence of anomalies, commonly with the left coronary artery aris-
ing from the right coronary artery. If the size of the pulmonary
arteries is adequate, surgical repair involves closing the VSD
and relieving the right ventricular outflow obstruction. In simple
cases, this involves resecting the infundibular muscle, but if the
pulmonary annulus is small, it may involve a pulmonary val-
votomy or valvectomy, a right ventricular outflow patch (which
could be transannular), or a conduit from the right ventricle to
the pulmonary artery.
The most common long-term problem after surgical repair
of TOF is pulmonary regurgitation, which occurs more often
when the pulmonary annulus has been patched or after pulmo-
nary valvectomy. Pulmonary regurgitation leads to progressive
right ventricular enlargement and dysfunction and ultimately
to exercise limitation and fatigue. Progressive right ventricular
enlargement has also been associated with progressive QRS pro-
longation and ventricular arrhythmias. The presence of atrial
or ventricular arrhythmias after repair of TOF should always
prompt a search for an underlying hemodynamic abnormality
that could be a residuum or sequela from a previous repair. If
severe pulmonary regurgitation is present, surgical or percutane-
ous pulmonary valve replacement should be performed before
irreversible right ventricular dysfunction occurs. Magnetic reso-
nance imaging has emerged as an important complementary
imaging technique for assessing right ventricular size and func-
tion. Other problems that could occur after TOF repair include
residual pulmonary stenosis, a right ventricular aneurysm at the
site of the surgical patch, residual VSD, and aortic dilatation and
aortic regurgitation. Sudden death may result from ventricular
arrhythmias and is more common in patients who have residual
hemodynamic abnormalities and ventricular dysfunction and in
patients who had late surgical repair.
Other patients with TOF may survive into adulthood because
of earlier palliative shunts that improve pulmonary blood flow
Box 81.4. Conditions Associated With Right Aortic
Arch
Pulmonary atresia
Truncus arteriosus
Tetralogy of Fallot
770 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
and help pulmonary arteries to grow. Types of shunts include
Blalock-Taussig (a subclavian artery–to–pulmonary artery anas-
tomosis, which can be used on either the right side or the left
side); Waterston shunt between the ascending aorta and the right
pulmonary artery; Potts shunt (descending aorta–to–left pulmo-
nary artery shunt); and a central shunt constructed with a polytef
graft. Problems with palliative shunts include distortion of the
pulmonary arteries (which may kink, thrombose, or occlude)
and, when the shunt is too large, pulmonary vascular disease.
Patients with large shunts are at risk of left ventricular volume
overload and, ultimately, ventricular failure with pulmonary
vascular disease. These patients are not accepted for heart-lung
transplant because of the lateral thoracotomy scar and the pro-
found risk of bleeding.
• TOF consists of a large subaortic VSD, obstruction to the pul-
monary outflow, override of the aorta, and right ventricular
hypertrophy.
• Palliative repair involves placing a systemic-to-pulmonary shunt,
which could lead to distortion of the pulmonary arteries (which
may kink, thrombose, or occlude) and, when the shunt is too large,
pulmonary vascular disease.
• Although long-term survival after complete repair of TOF is very
good, residua and sequelae may necessitate reoperation.
• The most common indication for reoperation after repair of TOF is
residual pulmonary regurgitation or stenosis (or both).
Pulmonary Atresia With VSD
Pulmonary atresia with VSD is another conotruncal abnormal-
ity. It has the same intracardiac anatomy as TOF, except that
the right ventricular outflow tract is blind or atretic. Pulmonary
blood flow often comes from collateral vessels that arise directly
from the descending aorta or from a PDA, bronchial collateral
vessels, and coronary collateral vessels. Collateral vessels may
be end arteries feeding into the lung tissue directly, or 1 or more
collateral vessels may enter into central pulmonary arteries. A
right aortic arch (Box 81.4) is present in 40% of patients with
pulmonary atresia. Unobstructed collateral blood flow into the
pulmonary circulation can lead to pulmonary hypertension. This
complex CHD is surgically repaired in childhood. When possi-
ble, the VSD is closed and a right ventricle–to–pulmonary artery
conduit is placed. Progressive conduit stenosis and, less often,
regurgitation require continued follow-up in adulthood.
Truncus Arteriosus
In truncus arteriosus, the pulmonary arteries arise from the
aorta; the intracardiac anatomy is the same as in pulmonary atre-
sia. The pulmonary arteries are usually not stenosed, so the clini-
cal features are the same as in Eisenmenger syndrome. Truncal
regurgitation is common.
Transposition of the Great Arteries
By the time they reach adulthood, virtually all patients with
transposition of the great arteries have had surgery—either an
atrial baffle procedure (Mustard or Senning operation) or an
arterial switch procedure (Jatene operation). Long-term com-
plications of an atrial baffle procedure are hemodynamically
significant because the right ventricle still supports the systemic
circulation; hence, right ventricular failure and tricuspid regur-
gitation are common. Atrial arrhythmias (particularly junctional
rhythm and atrial flutter) and, less commonly, baffle stenosis and
leak are present at follow-up. Late complications after an arterial
switch operation include pulmonary artery stenosis, neo-aortic
root enlargement, aortic regurgitation, and proximal coronary
artery obstruction.
Tricuspid Atresia
Nearly all patients with tricuspid atresia have had an operation
by the time they reach adulthood. The tricuspid valve is absent,
so systemic blood flows from the right atrium through an ASD
to the left atrium. It then enters the left ventricle, flows through a
VSD into a hypoplastic right ventricle, and reaches the pulmonary
artery. This pattern occurs when the great arteries are normally
related. In most patients, pulmonary blood flow is decreased
because the VSD is small or because there is an associated PS.
If the VSD is large, though, and there is no PS, unobstructed
pulmonary blood flow causes pulmonary hypertension. The
presence of a small, restrictive VSD or subpulmonary stenosis
protects against pulmonary hypertension. However, the patient
would still be cyanotic with reduced pulmonary blood flow that
might require a palliative systemic-to-pulmonary artery shunt
such as a Blalock-Taussig, Waterston, or central shunt.
Single Ventricle
Many forms and combinations of abnormalities involve a single-
ventricle heart. The most common type in adulthood is a dou-
ble-inlet left ventricle with PS. In this congenital heart anomaly,
both mitral and tricuspid valves connect to the left ventricle. The
small, rudimentary right ventricle connects to the left ventricle
and pulmonary valve through a VSD. PS protects against pul-
monary hypertension. Patients with this abnormality, therefore,
have cyanosis due to reduced pulmonary blood flow caused by
the PS, with left ventricular hypertrophy on the electrocardio-
gram and signs of PS on examination. Because they have PS
and reduced pulmonary blood flow, some of these patients might
benefit from placement of a systemic-to-pulmonary artery shunt
to improve the cyanosis. Without PS, patients who have a single
ventricle reach adulthood with severe pulmonary hypertension
due to unobstructed pulmonary blood flow.
The Fontan Procedure
Most adult patients with a single ventricle and PS have had a pre-
vious repair. The most common is the Fontan procedure, which
is considered if certain hemodynamic conditions are met, includ-
ing 1) normal pulmonary artery pressure and resistance and
2) normal left ventricular systolic and diastolic function. Many
technical variations to the original operation can create “Fontan
circulation,” which allows the systemic venous return from the
inferior and superior venae cavae to be baffled directly into 1 or
more pulmonary arteries and bypass the right ventricle, which
is often hypoplastic. This procedure usually corrects the cyano-
sis unless a residual shunt is present. Unfortunately, late serious
complications after Fontan operations are not uncommon. These
complications include obstruction of the Fontan pathway, right
atrial or pulmonary artery thrombus, ventricular dysfunction,
pulmonary venous obstruction, protein-losing enteropathy, and
atrial arrhythmias.
Eisenmenger Syndrome
Infants born with a large VSD or PDA have a large left-to-right
shunt in early childhood, with increased blood volume and
 81 Adult Congenital Heart Disease
pressure transmitted to the pulmonary circulation. The result
is pulmonary hypertension and subsequent pulmonary vas-
cular disease, which may become established within the first
2 years of life. Rarely, other intracardiac shunts may also result
in Eisenmenger physiology. Reversal of the left-to-right shunt
causes cyanosis, and the original CHD then becomes inoperable.
In some infants with large shunts, the pulmonary vascular resis-
tance never decreases and pulmonary vascular disease is pres-
ent from an early age. The right-to-left shunting associated with
irreversible pulmonary hypertension is called Eisenmenger syn-
drome. Rarely, it occurs with secundum ASD (<5% of patients),
usually later in life.
Physical Examination
Physical examination findings include the following: cyanosis
and clubbing, jugular venous pressure that may be normal or
with a slightly prominent a wave, a right ventricular lift, an ejec-
tion click from the dilated pulmonary artery, little or no murmur
(pressure in both ventricles is equal), loud P2 (may be palpable),
and a variable murmur of pulmonary regurgitation. A holosys-
tolic murmur can be heard in the presence of tricuspid regurgita-
tion due to annular dilatation. There may be differential cyanosis
between the limbs if the patient has a PDA.
Electrocardiography
The electrocardiogram shows evidence of right atrial enlarge-
ment and ventricular hypertrophy. Premature beats are not
uncommon.
Chest Radiography
Chest radiography shows prominent central pulmonary arteries
(they may be calcified and are sometimes mistaken for lymph-
adenopathy), right ventricular contour, and peripheral pulmo-
nary artery pruning.
Evaluation and Management
The diagnosis can be made from the findings on physical
examination and echocardiography. Rarely, the shunt is missed
because the pressure is equal in both chambers and the VSD is
overlooked. A PDA may be difficult to detect because there is
little blood flow through the ductus. The differential diagno-
sis of Eisenmenger syndrome is primary pulmonary hyperten-
sion, but, in comparison, patients with Eisenmenger syndrome
have a much better long-term survival. Eisenmenger syndrome
is a multisystem disorder that causes not only cardiac prob-
lems but also hematologic, renal, orthopedic, and pulmonary
abnormalities.
Hematologic Abnormalities
Patients with cyanosis have an increased concentration of eryth-
rocytes (but it is not polycythemia), and management of second-
ary erythrocytosis may be difficult. Patients with high degrees
of erythrocytosis (hemoglobin >20 g/dL; hematocrit >65%) may
experience symptoms of hyperviscosity (poor concentration,
headache, and fatigue). This condition is uncommon with lower
hemoglobin levels (unless the patient is dehydrated). Patients
therefore should not have therapeutic phlebotomy unless the
hemoglobin value is more than 20 g/dL. Frequent phleboto-
mies, in particular, should be avoided, because they destabilize
771
erythropoiesis and produce a rebound response from the bone
marrow and, ultimately, iron deficiency anemia. Iron-deficient
microcytes not only cause a deterioration in exercise capac-
ity but also paradoxically increase the risk of stroke (because
iron-deficient red cells are less deformable than normal red
cells). Phlebotomy should never be performed in patients with-
out concomitant fluid replacement, particularly in patients with
Eisenmenger syndrome, who may experience hypotension and
even sudden death.
Although patients with cyanotic heart disease have a slightly
increased risk of stroke, they also have hemostatic problems and
are at increased risk of bleeding. These hemostatic problems
include prolonged prothrombin time, prolonged activated par-
tial thromboplastin time, decreased levels of coagulation factors,
decreased platelet count, and abnormal platelet function. Thus,
patients with cyanotic CHD should never receive anticoagulation
therapy unless there is a very strong indication to do so, and,
ideally, the international normalized ratio should be kept on the
low side of the therapeutic range. If a patient with cyanosis will
undergo surgery and the hemoglobin value is more than 20 g/dL,
therapeutic phlebotomy with fluid exchange may normalize the
hemostatic problems.
Renal Abnormalities
Adults with cyanotic CHD frequently have abnormal renal func-
tion with a reduced glomerular filtration rate, proteinuria, and
hyperuricemia. The high uric acid levels are due to low fractional
uric acid excretion and overproduction of urate from red cell
turnover. The increased red cell turnover rate predisposes to gall-
stone and gout. Hyperuricemia is particularly important when
cyanotic patients have cardiac catheterization, and they should
not be dehydrated near the time of the procedure, particularly
because they may require a large amount of imaging contrast
material, which may induce acute renal failure. Intravenous fluid
hydration is indicated for very cyanotic patients, with meticulous
attention to fluid balance and good urine output.
Orthopedic Abnormalities
Scoliosis is much more common in patients with cyanotic heart
disease (even without a lateral thoracotomy). In addition, patients
may have a painful arthropathy due to hypertrophic changes in
the long bones.
Additional Concerns and Prognosis
Patients with Eisenmenger syndrome are at particular risk of
hemoptysis (which can be life threatening and may result from
pulmonary hemorrhage, pulmonary embolus, or in situ pulmo-
nary infarction). In addition, they are vulnerable to vasodilata-
tion, which may be fatal. Any decrease in blood pressure (such as
that produced by vasodilators) may cause increased right-to-left
shunting, cerebral hypoxia, and sudden death; thus, patients
should not be given injudicious vasodilator therapy. Extreme
caution must be used when patients with Eisenmenger syndrome
undergo noncardiac surgery; even relatively minor procedures
such as appendectomy may be fatal.
Death in Eisenmenger syndrome may occur from acute
hypoxia, sudden ventricular arrhythmia, or massive hemopty-
sis. Patients frequently experience symptomatic deterioration in
their 40s but may survive to their 60s. Causes of vasodilata-
tion should be avoided (eg, hot tubs and vasodilator therapy).
772 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Patients should be followed up for progressive right ventricular • Patients with Eisenmenger syndrome are also at particular risk of
dilatation and tricuspid regurgitation, which may herald right cholelithiasis, scoliosis, acne, endocarditis, heart failure, arrhyth-
ventricular failure. They may become extremely symptomatic mias, and hemoptysis that could be fatal.
with the onset of atrial arrhythmias, and sinus rhythm should be • Eisenmenger syndrome is an absolute contraindication for preg-
maintained whenever possible. Treatment options for primary nancy, with a reported 50% maternal mortality.
pulmonary hypertension have been applied to patients with
Eisenmenger syndrome and may improve symptoms. These
include prostacyclin analogues, endothelin receptor antagonists, Abbreviations
and phosphodiesterase inhibitors. Other options, rarely per-
ASD atrial septal defect
formed, are heart-lung transplant or single-lung transplant with
CHD congenital heart disease
closure of the defect. CoA coarctation of the aorta
P2 pulmonic valve component of the second heart sound
• Patients with Eisenmenger syndrome have a much better long-term
PDA patent ductus arteriosus
survival than patients with primary pulmonary hypertension.
PS pulmonary stenosis
• Adults with cyanotic CHD frequently have abnormal renal func- Qp/Qs ratio of pulmonary blood flow to systemic blood flow
tion with a reduced glomerular filtration rate, proteinuria, hyper- TOF tetralogy of Fallot
uricemia, and gout. VSD ventricular septal defect
 82
HIV Infection and the Heart
JOSEPH G. MURPHY, MD, and ZELALEM TEMESGEN, MD
Cardiac involvement in HIV infection is common, typically
presenting clinically as pericardial diseases. HIV cardiomyop-
athy may present as clinical heart failure, be clinically silent but
evidenced as an asymptomatic fall in left ventricular ejection
fraction on echocardiography, or be detected only at autopsy as
focal myocarditis. Clinical detection of HIV-associated cardiac
disease significantly underestimates the true incidence of disease
detected by cardiac imaging studies or at autopsy Additionally,
as HIV-infected patients now live longer on HAART, with an
estimated annual mortality of 1%–2%, conventional risk-factor,
age-associated cardiovascular problems such as hypertension
and coronary atherosclerosis have become more prominent.
HIV cardiac disease is likely due both to the direct effects
of HIV infection and to the indirect effects of HAART treat-
ment, as well as the contribution of conventional well-established
causes of cardiac disease in other populations such as smoking,
hyperlipidemia and hypertension. Long-term HIV therapy is
associated with many metabolic disturbances, particularly lipid
and glucose metabolic derangement which may increase the risk
of atheromatous cardiovascular disease.
Pericarditis
Pericarditis, often associated with pericardial effusion, is the
most common cardiac finding in patients with HIV infection.
Many autopsy and echocardiographic series estimate the inci-
dence of HIV-associated pericardial disease at about 1% of
HIV patients overall. The spectrum of pericardial disease in
HIV-infected patients ranges from asymptomatic effusions inci-
dentally detected by echocardiography to potentially fatal tam-
ponade. Most asymptomatic pericardial effusions in HIV patients
do not have a specific identifiable cause, while a specific cause
can generally be established in about two-thirds of symptomatic
pericardial effusions. The etiologic spectrum of HIV-related
pericardial disease is wide and includes infectious organisms and
malignancy. Among the infectious causes of pericarditis, myco-
bacterial organisms, including Mycobacterium tuberculosis and
atypical mycobacteria, are often isolated from pericardial effu-
sions in HIV patients—specifically, in 34% of 66 published cases
of cardiac tamponade. Other infectious causes of pericardial dis-
ease in HIV-infected patients include Staphylococcus aureus,
Streptococcus pneumoniae, Nocardia asteroides, Listeria
monocytogenes, Chlamydia species, Histoplasma capsulatum,
Cryptococcus neoformans, herpes simplex, CMV, and coxsackie-
virus. The neoplastic causes of pericardial effusions in patients
with AIDS are commonly Kaposi sarcoma and lymphoma.
HIV-infected patients with small, asymptomatic pericardial
effusions should be managed conservatively with follow-up echo-
cardiography at regular intervals but without specific diagnostic
or therapeutic interventions. Symptomatic patients and those
with large effusions require pericardiocentesis and fluid analysis
for cytology, culture, and biochemical studies. Pericardial biopsy
increases the diagnostic yield, particularly for the diagnosis of
tuberculosis.
Cardiac tamponade may occur with large or moderately sized,
but rapidly accumulating, pericardial effusions and may be the
presenting feature of patients with pericarditis. Cardiac tampon-
ade requires urgent pericardial catheter drainage and, in many
cases, a surgically created pericardial window (through a sub-
xiphoid pericardiotomy) for refractory effusions. Percutaneous
balloon pericardiotomy is generally ineffective and gets rapidly
obstructed. Bacterial or fungal pericarditis should be treated
with appropriate drugs. Patients with large refractory pericardial
effusions for which no specific cause has been established may
require empirical antituberculous chemotherapy. Corticosteroids
used as an adjunct to antituberculous therapy in patients with
773
774 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

HIV infection and tuberculous pericarditis are probably Table 82.1. Selected Drugs That Have the Potential to Cause
beneficial. Pericarditis due to lymphoma may respond in the Cardiac-Related Toxicity and Are Used for the Treatment of
short term to radiotherapy and chemotherapy. Human Immunodeficiency Virus Infection or Its Complications
• In approximately two-thirds of AIDS patients with symptomatic Drug Effect
pericarditis, the pericarditis has an identifiable and potentially
treatable cause. Nucleoside analogue reverse Mitochondrial toxicity,
transcriptase inhibitors cardiomyopathy
• Mycobacterial infections are a common infectious cause of tam- Non-nucleoside reverse transcriptase Dyslipidemia
ponade in AIDS patients. inhibitors
• In patients with suspected tuberculous pericarditis, surgical peri- Protease inhibitors Dyslipidemia, increased risk of
cardial biopsy is the diagnostic test of choice if pericardial fluid coronary artery disease, insulin
microscopy is negative. resistance
Trimethoprim-sulfamethoxazole Torsades de pointes
Pentamidine Ventricular arrhythmia, torsades
Myocarditis and Cardiomyopathy
de pointes
The spectrum of myocardial involvement in HIV infection is Anabolic steroids Increased risk of coronary artery
wide and includes inflammatory myocarditis, dilated cardiomyo- disease
pathy, and infiltrative neoplastic disease (Figure 82.1). Interferon Hypertension, hypotension,
The prevalence of myocarditis in HIV infection is variable, tachycardia, cardiomyopathy
depending on the population selected and the diagnostic meth- Doxorubicin Cardiomyopathy
ods used. Autopsy studies of HIV-infected adults have reported Vinblastine Increased risk of coronary artery
disease
myocarditis in up to 52% of cases. Echocardiographic series
have identified cardiomyopathy in 30% to 40% of HIV-infected
patients. A higher incidence of left ventricular dysfunction cor- associated with cardiac-related toxicities. These are listed in
relates with lower CD4 counts in HIV patients, but cardiomyopa- Table 82.1. Finally, recreational drugs, including cocaine and
thy may precede the development of advanced HIV disease. The methamphetamine, are directly cardiotoxic.
etiologic spectrum of myocarditis in HIV-infected individuals is Echocardiography is the main diagnostic tool to documenting
wide. HIV has been detected in cardiac tissue by culture, immu- cardiac dysfunction while chest radiographs may show cardio-
nohistochemistry, and molecular techniques. However, a direct megaly; electrocardiographic findings are generally nonspecific.
causal link between the presence of HIV in the myocardium and Myocardial biopsy is specific but lacks sensitivity for the diag-
the induction of myocarditis has not been established. The lack nosis of HIV myocarditis and a specific cause is found on biopsy
of CD4 cell receptors on myocardial cells argues against the only in a minority of cases (20%). Histology generally shows
direct role of HIV in causing myocarditis. only small focal collections of mononuclear cells, usually with-
Several mechanisms through which HIV may indirectly cause out evidence of myocardial necrosis.
myocardial disease have been proposed, including the release of The treatment of HIV-associated myocardial disease is
inflammatory cytokines, endothelial dysfunction and autoimmu- directed toward a treatable etiologic agent if one has been iden-
nity. Infectious agents and infections that have been implicated tified. The routine use of immunosuppressive therapy to treat
in HIV-associated myocardial disease include toxoplasmosis, myocarditis is not indicated. Treatment of congestive heart fail-
tuberculosis, atypical mycobacterial infections, cryptococco- ure is the same as for HIV-negative patients. Digoxin, diuretics,
sis, histoplasmosis, CMV, coxsackievirus, and Chagas disease angiotensin-converting enzyme inhibitors, angiotensin receptor
in Central and South America (Figure 82.2). Several nutritional blockers, and β-blockers can all be used.
deficiencies, including deficiences of selenium, L-carnitine, and Mortality in HIV-infected patients with cardiomyopathy
vitamin B, have also been associated with myocardial dysfunc- is higher than in HIV-infected patients without cardiomyop-
tion and cardiomyopathy. Several drugs, used for the treatment athy or in non–HIV-infected patients with idiopathic dilated
of either HIV infection itself or its complications, have been cardiomyopathy.

Figure 82.1. Aspergillus myocarditis. Figure 82.2. Cytomegalovirus myocarditis.

 82 HIV Infection and the Heart 775

Pulmonary Vascular Disease was responsible for 5% to 20% of hospital admissions and for
and Pulmonary Hypertension 5% to 10% of total deaths in intravenous drug users with HIV
infection. The clinical presentation was similar to that observed
Right ventricular dysfunction is a well-documented cardiac
in HIV-negative patients. The clinical outcome depends more on
complication of HIV infection and occurs in about 10% of AIDS
the affected valve and the causative organism than on the HIV
patients with cardiovascular disease. It is seen more frequently
serostatus of the patient. In intravenous drug users, the most
in intravenous drug users and patients with a history of chronic
common valve involved is the tricuspid valve, and the most com-
pulmonary infections, including Pneumocystis jiroveci infection.
mon causative organism is S aureus. The microbiologic spec-
The cause is probably multifactorial.
trum for infective endocarditis in HIV-infected non–intravenous
Many cases of HIV-related pulmonary hypertension have no
drug users is wide and includes unusual organisms such as
additional risk factors for pulmonary hypertension other than the
Salmonella, Aspergillus, Cryptococcus, and Candida species in
HIV infection itself. The estimated prevalence of this entity is
addition to the usual bacteria causing endocarditis in the general
about 0.5%, which is 2,500 times greater than the prevalence
population.
of primary pulmonary hypertension in the general population.
No correlation between low CD4 cell counts or the presence of
opportunistic infection and the development and progression of Cardiac Tumors
pulmonary hypertension has been observed. The pathophysiol-
Kaposi sarcoma, as seen in patients with AIDS, can involve the
ogy of this condition is unclear and there is no evidence that HIV
myocardium and pericardium and classically presents with peri-
directly infects pulmonary artery endothelial cells. Attempts to
cardial effusion or, less commonly, cardiac tamponade. Patients
detect HIV or its proteins in lung tissue by electron microscopy
with primary cardiac lymphoma, a rare malignancy associated
or by immunohistochemical or molecular techniques have not
with AIDS, present with heart failure or ventricular arrhyth-
been successful. An indirect role for HIV through the release
mias due to diffuse infiltration of the ventricular wall or, less
of cytokines (interleukins 1 and 6, tumor necrosis factor α, and
commonly, with mechanical obstruction of valve function due
endothelin-1) has been proposed.
to localized nodules or intracavitary masses. Surgery, chemo-
The histopathology of HIV-associated pulmonary hyperten-
therapy, and radiotherapy are generally palliative (Figures 82.3
sion is similar to that of primary pulmonary hypertension in the
and 82.4).
general population and includes plexogenic pulmonary arteriolar
lesions, in situ thrombotic pulmonary arteriopathy, and pulmo-
nary veno-occlusive disease. Symptoms, clinical features, and Thromboembolism
findings from diagnostic studies are not different from those HIV infection is associated with increased risk of deep venous
reported for non–HIV-infected patients with primary pulmonary thrombosis, pulmonary embolism and arterial embolism, proba-
hypertension. Progressive dyspnea is the most common present- bly on the basis of the generation of a pro thrombotic state.
ing symptom. Chest radiographs show cardiomegaly and pul-
monary artery prominence; right ventricular hypertrophy, right
atrial abnormality, and right axis deviation are commonly noted Coronary Artery Disease
on electrocardiographic tracings. Echocardiographic findings With continued use of HAART and longer survival of HIV-
usually consist of right heart enlargement, tricuspid regurgita- infected patients, several metabolic complications of HIV
tion, and paradoxical septal motion, whereas Doppler echocardi- infection and its treatment have been observed. These include
ography and cardiac catheterization show increased pulmonary dyslipidemias, insulin resistance, hyperglycemia, and body
artery pressures in the setting of normal left heart pressures. composition changes (lipodystrophy and lipoatrophy). The
The treatment of HIV-associated pulmonary hypertension is appreciation of these metabolic disorders associated with anti-
similar to that of primary pulmonary arterial hypertension. The retroviral therapy has led to a growing concern about a possible
effect of antiretroviral therapy on the course of HIV-related pul- increased risk of cardiovascular disease. Dyslipidemia is com-
monary hypertension is controversial. Some studies have shown mon in HIV-infected patients and can be an isolated entity, an
a benefit; others have not. Worsening of the clinical course with HIV-induced metabolic syndrome, or a complication of HAART
antiretroviral therapy has also been reported.
The prognosis of HIV-related pulmonary hypertension is
poor, with a median survival of 6 months from the diagnosis of
pulmonary hypertension to death.

Endocardial Disease
A common incidental finding at autopsy in patients dying of
AIDS is marantic endocarditis, a nonbacterial thrombotic endo-
carditis in which sterile valvular vegetations occur without an
infectious cause. Systemic embolization is an uncommon clin-
ical presentation of marantic endocarditis; valve destruction or
clinical valve dysfunction is rare.
Infective endocarditis in HIV-infected patients is uncom-
mon and occurs almost exclusively in intravenous drug users. In
a retrospective review of infective endocarditis in HIV-infected
patients between 1979 and 1999 at a tertiary-care hospital, only 8
out of 599 cases of infective endocarditis were diagnosed in non–
intravenous drug users. In another review, infective endocarditis Figure 82.3. Primary cardiac lymphoma.
776 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

exercise, and modification of other risk factors (eg, smoking),

should generally be attempted first before instituting drug thera-
pies. Pharmacologic therapy of dyslipidemias usually includes
treatment with statins with or without fenofibrates. In general,
PIs are substrates as well as inhibitors of cytochrome P-450.
The primary route of metabolism for most statins is also the
cytochrome P-450 system. Thus, a significant potential for
drug interaction exists. Simvastatin and lovastatin should not
be used in patients taking PIs or the non NRTI, delavirdine.
Atorvastatin, fluvastatin, pravastatin, and rosuvastatin appear to
be safe for use with PIs. Fibrates are metabolized by glucuroni-
dation and thus do not present a significant potential for drug
interaction.

Long-QT Syndrome in HIV Disease

Figure 82.4. Lymphoma cells in the myocardium.
QT prolongation can occur with HIV infection, even without drug
therapy. Pentamidine may cause torsades de pointes both directly
therapy. HIV patients often exhibit a low total cholesterol—both and indirectly by magnesium wasting. HIV protease inhibitors
HDL and LDL are low with a high plasma triglyceride levels, can cause the long-QT syndrome by blocking the HERG chan-
independent of HAART exposure. HAART therapy is associated nel. Foscarnet, a drug often used to treat CMV retinitis in HIV
with endothelial dysfunction, hypertension, and insulin resis- patients, has been associated with QT prolongation and torsades
tance. There is no definitive proof that HIV infection itself is a de pointes.
risk factor for cardiovascular disease, but the weight of evidence
suggests that there is a link between antiretroviral therapy, par- Lactic Acidosis and HIV Drugs
ticularly the use of PIs, and an increased risk of coronary artery
disease. NRTIs, in addition to their intended target binding to HIV
In an analysis of data from a cohort of 5,672 outpatients viral DNA polymerase, can also variably inhibit human mito-
with HIV-1 at nine US HIV clinics, the use of PIs was associ- chondrial DNA polymerase gamma, an essential enzyme in
ated with an increased risk of myocardial infarction. In a study mitochondrial DNA function. Long-term use (>6 months) may
of 19,795 HIV-infected French men receiving a PI-based regi- result in cellular mitochondrial dysfunction which presents clin-
men, morbidity ratios were greater with longer exposure to PIs. ically as myopathy, lipoatrophy, peripheral neuropathy, hepatic
Similarly, receipt of a PI-based regimen was associated with MI dysfunction, and lactic acidosis.
after adjustment for age in the Frankfurt HIV Cohort. The DAD There are four important clinical points with regard to hyper-
trial, a large international collaborative observational study, lactatemia and lactic acidosis with NRTI drugs, namely, 1) it is a
evaluated the incidence of MI in 23,400 HIV-infected patients type B lactic acidosis (no systemic hypoperfusion); 2) symptoms
from 11 cohorts in Europe, Australia, and the United States. The are nonspecific and include nausea, vomiting, abdominal pain
risk of MI increased progressively with the number of years the and severe fatigue; 3) severe lactic acidosis generally requires
patients received combination antiretroviral therapy (adjusted liver dysfunction for lactate accumulation, since a normally func-
relative risk, 1.16 per year of exposure; 95% confidence inter- tioning liver clears lactate; and 4) lactic acidosis in this setting
val, 1.09–1.23). Increased PI exposure was associated with an may be fatal.
increased risk of MI, which is partly explained by dyslipidemia.
Age, male sex, a past history of cardiovascular disease, smoking, Abbreviations
elevated total cholesterol level at baseline, and the presence of
diabetes mellitus at baseline were also independent predictors for AIDS acquired immunodeficiency syndrome
MI in HIV-infected patients.. CMV cytomegalovirus
HAART highly active antiretroviral therapy
The risk of HAART-related cardiovascular disease is out-
HIV human immunodeficiency virus
weighed by the benefits of antiretroviral therapy and should HDL high-density lipoprotein
not be a reason to withhold therapy. Although consideration of LDL low-density lipoprotein
dyslipidemia and cardiovascular risk is appropriate in the con- MI myocardial infarction
struction of an antiretroviral regimen, virologic suppression is NRTI nucleoside reverse transcriptase inhibitor
the overriding goal of antiretroviral therapy. Current guidelines PI protease inhibitor
recommend that HIV-infected adults undergo evaluation and
treatment on the basis of the National Cholesterol Education
Program Adult Treatment Panel guidelines for dyslipidemia. Name of Clinical Trial
Nonpharmacologic interventions, such as dietary counseling, DAD Data Collection on Adverse Events of Anti-HIV Drugs
 83
Infective Endocarditisa
NANDAN S. ANAVEKAR, MB, BCH, M. RIZWAN SOHAIL, MD,
and JOSEPH G. MURPHY, MD
IE is an infection on the endocardial surface of the heart, usually
affecting the heart valves, a congenital heart shunt, or an intra-
vascular device. Acute endocarditis has a fulminant course with
high fever, systemic toxicity, and death within days to weeks if
left untreated; subacute and chronic endocarditis have an indolent
course with low-grade fever, night sweats, weight loss, embolic
and immunologic phenomena, and death within weeks to months
if left untreated. Acute endocarditis is caused by Staphylococcus
aureus, Streptococcus pyogenes, Streptococcus pneumoniae, or
Neisseria gonorrhoeae, and the subacute and chronic forms are
usually due to viridans streptococci.
Epidemiology
Approximately 15,000 new cases of IE are diagnosed annually in
the United States. The mean age of patients with IE has increased
gradually—from less than 30 years in the pre-antibiotic era to
more than 60 years currently. Men are affected both at a younger
age (by about 6 years) and more frequently than women (male to
a
Portions previously published in Bonow RO, Carabello BA, Chatterjee
K, de Leon AC Jr, Faxon DP, Freed MD, et al; 2006 Writing Committee
Members; American College of Cardiology/American Heart Association
Task Force. 2008 Focused update incorporated into the 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): endorsed by the Society of Cardiovascular
Anesthesiologists, Society for Cardiovascular Angiography and
Interventions, and Society of Thoracic Surgeons. Circulation. 2008
Oct 7;118(15):e523–661. Epub 2008 Sep 26. Used with permission.
Abbreviations and acronyms are expanded at the end of this chapter.
female ratio, approximately 2:1). In children, endocarditis is rare
and is associated with underlying congenital heart disease.
Health care–associated endocarditis is an important noso-
comial infection associated with implantable cardiovascular
devices and indwelling vascular catheters. In patients with native
valve endocarditis and no history of injection drug use, more
than a third have health care–associated endocarditis.
Rheumatic heart disease is the major predisposing factor
for IE in developing nations (affecting the mitral valve in most
cases) but is associated with less than 5% of cases in developed
countries. In the United States, mitral valve prolapse with mitral
regurgitation and degenerative or bicuspid aortic valve disease are
the leading cardiac conditions that predispose to IE in adults.
The heart valve most commonly involved in IE is the mitral
valve (Figure 83.1), followed by the aortic valve, tricuspid valve
(especially in intravenous drug users and patients with indwell-
ing catheters), and, rarely, the pulmonary valve. Isolated mitral
valve endocarditis is more common in women, and isolated aor-
tic valve endocarditis, often associated with a congenitally bicus-
pid aortic valve, is more common in men.
Congenital heart disease is responsible for about 15% of IE
cases; specific lesions include patent ductus arteriosus, ventricu-
lar septal defect, and coarctation of the aorta. Surgical closure of
a ventricular septal defect reduces the risk of IE, provided there
is no residual shunt. Patients with hypertrophic cardiomyopathy
are at increased risk of IE, usually due to mitral valve trauma
associated with systolic anterior motion, especially in those with
hemodynamically severe forms of the disease (high peak systolic
pressure gradient and markedly symptomatic). IE is very rare
in patients with secundum atrial septal defects because of the
absence of significant blood flow turbulence.
Endocarditis may follow an atypical course in immuno-
suppressed patients, both in terms of clinical presentation
777
778 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

creating a sterile vegetation. Transient bacteremia occurs when-

Figure 83.1. Infective Endocarditis With Vegetations on the Mitral
Valve (arrow).
and underlying microbiologic features. The prevalence of IE
is increasing among patients with HIV infection, including
the unusual organism Bartonella quintana. Intravenous drug
users are at the highest risk for recurrent IE and polymicrobial
infection.
Pathogenesis
Endocarditis infection requires that the valve surface be dam-
aged by either turbulent blood flow or the offending organism
itself to allow bacterial adherence and colonization (Figure 83.2).
Aggregation and deposition of platelets and fibrin follow,
ever a mucosal surface, heavily colonized with bacteria (such as
the gingival margin), is traumatized, an event allowing bacterial
seeding of the vegetation. Hemodynamic turbulence contributes
to the localization of vegetations usually downstream from a
regurgitant jet, characteristically the atrial surface of the mitral
valve and the ventricular surface of the aortic valve. Most bacter-
emias are rapidly cleared from the circulation, and only organ-
isms that are resistant to the complement-mediated bactericidal
activity of serum survive and cause endocarditis. Bacterial colo-
nies are found beneath the surface of the vegetation surrounded
by platelets and fibrin, and these create a localized sanctuary of
impaired host resistance.
IE results in stimulation of both humoral and cellular immu-
nity, as evidenced by hypergammaglobulinemia, splenomegaly,
and the appearance of macrophages in the peripheral blood.
Rheumatoid factor (anti-immunoglobulin G and M antibodies)
develops in about 50% of patients with IE of longer than 6 weeks
in duration. Antinuclear antibodies also occur in IE and may
be pathogenic in the musculoskeletal manifestations, low-grade
fever, and serositis. Patients with IE and circulating immune
complexes are at risk for development of glomerulonephritis,
in which immune complexes with complement components are
deposited subepithelially along the kidney glomerular basement
membrane.
Organ System Manifestations
Heart
Vegetations are usually located along the line of closure of a
valve leaflet, on the atrial surface of the atrioventricular valves,

Valvular or vascular endothelium Mucous membranes

Trauma or other colonized tissue
Turbulence Trauma
Metabolic changes Intravenous drug use
Intravascular lines
Endothelial damage
Platelet, fibrin deposition (biofilm)
Bacteremia
Nonbacterial thrombotic endocarditis
Adherence

Colonization
Bacterial division
Fibrin deposition
Platelet aggregation
Protection from neutrophils
Protection from antibiotics
Mature infected vegetation

Valve ring abscess Valvular incompetence Valvular perforation

Figure 83.2. Pathogenesis of Infective Endocarditis.
 83 Infective Endocarditis 779

or on the ventricular surface of the semilunar valves. Vegetations

seen in acute endocarditis are generally larger, more friable, and
associated with more necrosis than those in subacute or chronic
cases. Untreated, IE results in local destruction of cardiac struc-
tures, including leaflet perforation, destruction of the valvular
apparatus, development of valve ring abscesses with formation
of fistulas in the myocardium or pericardial sac, and formation of
aneurysmal dilatation at the sinus of Valsalva. Large vegetations
may create a hemodynamic effect of valvular stenosis by obstruc-
tion to flow across a valve. Myocarditis, myocardial infarction,
and pericarditis are frequently found at autopsy. Myocardial
abscesses are found in up to 20% of autopsy cases and are gen-
erally associated with acute staphylococcal infection. Embolic
phenomena are common in IE; the clinical embolic event rate
is 15% to 35%, and the pathologic embolic event rate at autopsy
is 45% to 65%. Emboli most frequently involve the cerebral,
renal, splenic, and coronary circulations. Emboli and immune
complex deposition contribute to the extracardiac manifestations
of IE, including Osler nodes, Janeway lesions, and Roth spots
(Table 83.1).

Kidney
Endocarditis may cause 3 pathologic processes in the kidney:
septic abscess formation, infarction, and glomerulonephri- Figure 83.3. Gross Appearance of Kidney in Staphylococcus aureus
Endocarditis Showing Multiple Small Areas of Infection on the Kidney
tis (Figures 83.3 and 83.4). Abscesses are uncommon in IE,
Surface.
infarctions have been reported in about 50% of autopsy cases,
and focal glomerulonephritis is found in 48% to 88% of cases.
Between 10% and 15% of the patients with IE have an immune

Table 83.1. Extracardiac Pathologic Findings in Infective

Endocarditis
Location Manifestation
Central nervous system Cerebral emboli
Cerebral infarction
Arteritis
Abscess
Mycotic aneurysm
Intracerebral or subarachnoid hemorrhage
Cerebritis
Meningitis
Spleen Splenic infarct
Splenic abscess
Splenic enlargement a
Lung Pulmonary embolib
Pleural effusion
Empyema
Skin Petechiae
Osler nodesc
Janeway lesionsd
Eye Roth spotse
a
Pathologic findings include hyperplasia of lymphoid follicles, proliferation of
reticuloendothelial cells, and scattered focal necrosis.
b
Associated with right-sided infective endocarditis.
c
Consist of arteriolar intimal proliferation with extension to venules and
capillaries, which may be accompanied by thrombosis and necrosis; diffuse
Table 83.1 (continued)
perivascular infiltrate consisting of neutrophils and monocytes surrounds
dermal vessels; immune complexes may be seen in dermal vessels.
d
Consist of bacteria, neutrophilic infiltration, necrosis, and subcutaneous
hemorrhage; secondary to septic emboli; subcutaneous abscesses on histologic
examination.
e
Consist of lymphocytes surrounded by edema and hemorrhage in nerve fiber Figure 83.4. Microscopic Appearance of Kidney in Staphylococcus
layer of retina. aureus Endocarditis.
780 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

complex glomerulonephritis, akin to that in systemic lupus ery-

thematosus. This form of glomerulonephritis is associated with
hypocomplementemia and positive rheumatoid factor.

Mycotic Aneurysms
Mycotic aneurysms are aneurysmal dilatations of an artery and
usually present catastrophically in cases of acute IE, but on
occasion they may be detected months or years after success-
ful treatment (Figure 83.5). They are more common with viri-
dans streptococci and are found in 10% to 15% of autopsy cases.
Responsible pathogenic mechanisms include direct invasion of
the arterial wall with microabscess formation and rupture, sep-
tic or bland embolic occlusion of the vasa vasorum followed by
arterial wall weakening and aneurysm formation, or immune
complex deposition within the arterial wall and inflammation-
mediated aneurysmal dilatation.
Mycotic aneurysms tend to occur at arterial bifurcation sites
and are most commonly found in the cerebral circulation; other Figure 83.6. Brain Abscess Caused by Aspergillus Species. Patient
sites include the abdominal aorta, the sinus of Valsalva, and the had Aspergillus infective endocarditis (arrow) (same patient as in
splenic, coronary, pulmonary, and superior mesenteric arteries. Figures 83.14 and 83.15).

Central Nervous System
Cerebral emboli are the most common serious complications of
IE. They are clinically evident in up to 30% of patients with IE,
but the actual cerebral embolic rate may be much higher because
the emboli are clinically silent. The middle cerebral artery and
its branches are most commonly affected. Cerebral infarction,
arteritis, abscess (Figure 83.6), mycotic aneurysms, intracerebral
or subarachnoid hemorrhage, encephalomalacia, cerebritis, and
meningitis have been reported. Hemorrhagic transformation of
an ischemic infarct due to septic emboli is the most common
mechanism leading to fatal intracerebral hemorrhage in IE.
Roth spots are a retinal manifestation of IE (Figure 83.7).
Pathologically, these lesions comprise lymphocytes surrounded
by edema and hemorrhage within the nerve fiber layers of the
retina.
Spleen
Splenomegaly is common in IE and is usually reactive in nature
as a result of hyperplasia of the lymphoid follicular network.
Splenic infarcts have been reported in up to 44% of autopsy
series but are usually clinically silent. Splenic abscess is uncom-
mon in IE and typically presents with fever, left upper quadrant
pain, and leukocytosis.
Lung
Right-sided IE may cause pulmonary embolism with or without
infarction, acute pneumonia, pleural effusions, or empyema.
Skin and Mucous Membranes
Skin involvement in IE is the result of septic emboli or immune
complex deposition. Petechiae are found in 20% to 40% of cases
(Figures 83.8 and 83.9). Osler nodes are white, painful nodularites
usually found on the pads of the fingers or the toes; microscopi-
cally, they show arteriolar intimal proliferation with extension
to venules and capillaries and may be accompanied by throm-
bosis and necrosis (Figure 83.10). Immune complexes may be
present in the dermal vessels. Janeway lesions are erythematous,

Figure 83.5. Mycotic Aneurysm (arrow). Figure 83.7. Roth Spots.

 83 Infective Endocarditis 781

Figure 83.10. Osler Node (arrow).

endocarditis but may be absent with right-sided endocarditis.

More than 90% of patients in whom new valvular incompetence
develops also have development of CHF, and this condition is
Figure 83.8. Palatal Petechiae. now the leading cause of death in IE. Pericarditis is rare, but
when present it is usually accompanied by myocardial abscess
macular, painless lesions occurring on the palmar surface of the formation as a complication of staphylococcal infection.
hands or plantar aspect of the feet; they originate from septic
emboli (Figure 83.11).

Clinical Manifestations
High-grade bacteremia may lead to IE symptoms within hours
to days; however, the time between symptoms to the diagnosis
of subacute IE is often delayed (median time, approximately
5 weeks). The clinical features of IE arise from 1) local car-
diac effects from infection on the valve that result in conduction
abnormalities, creation of intracardiac shunts, valvular insuffi-
ciency or obstruction, or development of hemodynamic instabil-
ity and congestive heart failure; 2) distal organ manifestations of
bland or septic embolization and immune complex deposition;
and 3) continuous bacteremia with metastatic foci of infection.
Fever is the most common clinical manifestation of IE (95%
of cases), but it may be absent in the setting of CHF, renal fail-
ure, terminal illness, or recent prior antibiotic use or in elderly
patients. New or changing heart murmurs are typical in left-sided

Figure 83.9. Conjuctival Petechiae. Figure 83.11. Janeway Lesions on the Sole of the Foot.
782 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
Major embolic events as a group are second to CHF as a com-
plication of IE and occur in at least a third of patients with IE.
Splenic emboli may result in infarction and can present with left
upper quadrant pain with radiation to the left shoulder. Renal
infarctions may be associated with microscopic or gross hema-
turia. Retinal artery emboli are rare, occurring in less than 2%
of cases, and may present with sudden loss of vision. Coronary
artery emboli usually arise from vegetations on the aortic valve
and can cause myocardial infarction. Major vessel emboli are
common in fungal endocarditis.
Neurologic complications are a frequent presenting feature
of IE, whereas the development of new clinical neurologic signs
during IE is associated with a twofold to fourfold increase in
mortality. Mycotic aneurysms in the cerebral circulation occur in
2% to 10% of cases of IE and may cause fatal subarachnoid hem-
orrhage. Other neurologic manifestations of IE include seizures,
severe headache, choreathetoid movements, mononeuropathy,
and cranial nerve palsy. Toxic encephalopathy ranging from a
mild personality change to frank psychosis may occur, especially
in the elderly.
Among intravenous drug users, acute infections of any kind
account for 60% of hospital admissions, and IE accounts for a
quarter of these. A history of cocaine use should heighten the
suspicion for IE. The most reliable predictors of IE in febrile par-
enteral drug users are visualization of vegetations on echocar-
diography and the presence of embolic phenomena. The tricuspid
valve is the most common valve involved in this setting. Of the
patients with tricuspid valve endocarditis, 30% have pleuritic
chest pain and pulmonary findings may dominate the clinical
picture; radiographic evidence of pulmonary infiltrates, effu-
sions, or septic emboli are present in the vast majority (>80%) of
patients. Most patients presenting with IE in the setting of intra-
venous drug use are 20 to 40 years old, and the male to female
ratio is 4:1. In this setting S aureus is the most common etiologic
organism. IE is more common among injection drug users with
HIV or AIDS. Conversely, in HIV-infected patients in whom
AIDS develops, but who are not intravenous drug users, IE is
rare, and when it occurs it has equal frequency of right- and left-
sided IE.
Table 83.2. Ancillary Blood Tests in Infective Endocarditis
Estimated Frequency
Laboratory Result in Endocarditis
Anemia 70%–90%
Thrombocytopenia 5%–15%
Leukocytosis 20%–30%
Leukopenia 5%–15%
Increased ESR 60%
Hypergammaglobulinemia 20%–30%
Positive rheumatoid factor 40%–50%
Hypocomplementemia 5%–15%
Abnormal urinalysis 70%
Circulating immune complexes Frequently present
Increased C-reactive protein level >90%
Abbreviations: CHF, congestive heart failure; ESR, erythrocyte sedimentation rate; IE, infective endocarditis; IV, intravenous.
Investigations
Finding the Organism
Blood cultures are the most important laboratory tests performed
in the diagnostic evaluation of IE, and the first 2 blood cultures
yield the etiologic agent in more than 90% of cases when ongoing
bacteremia is present. Some unusual organisms, such as Brucella,
Nocardia, and members of the HACEK group are slow-growing
and may require cultures to be held for extended (up to 2 weeks)
incubation. Tissue cell culture may be useful for isolating obli-
gate intracellular bacteria (Coxiella burnetii, Chlamydia species)
from blood, vegetations, or valvular tissues. Blood culture results
may be negative in up to 50% of cases of fungal endocarditis.
Serologic studies are useful for the diagnosis of Q fever, murine
typhus, bartonellosis, brucellosis, legionellosis, and psittacosis.
Gram stain of the resected valve specimens may be helpful for
the diagnosis and confirmation of IE.
PCR can be used to detect organisms that grow poorly or not
at all using conventional culture techniques. There are 2 basic
approaches to PCR for the diagnosis of IE. The first detects a
genetic target unique to a specific genus or species of microorgan-
ism. The second broadly detects nucleic acid common to a large
group of organisms based on a conserved genetic target. A com-
mon target used for broad-range detection of bacteria using PCR
is 16S ribosomal DNA, which is present in all bacteria but not in
humans. PCR is a very sensitive technique; a positive PCR result
must always be interpreted in the correct clinical context. Even
several months after therapy for IE, PCR results may remain
positive. PCR does not provide information about antimicrobial
susceptibility and is complementary to standard blood culture.
Ancillary Blood Tests
The estimated frequency of abnormal results of laboratory inves-
tigations and their significance is summarized in Table 83.2.
Imaging
The objectives of echocardiography when IE is suspected are
summarized in Box 83.1. TTE should be performed in all
Comments
Normochromic and normocytic, with low iron-binding capacity and low serum
iron concentration
Presence may indicate splenomegaly
In the absence of renal failure, CHF, or disseminated intravascular coagulation, a
normal ESR is strong evidence against IE
Especially common with a duration of illness >6 weeks
Predictor of immune complex deposition in the kidneys
Gross or microscopic hematuria, pyuria, bacteriuria, and proteinuria are common
Can also be detected in up to a third of patients who have septicemia without IE.
May be particularly useful in the setting of culture-negative IE or in patients
with a history of IV drug use. Can also be used to follow response to therapy or
predict relapse and treatment failures
May also be useful for monitoring the course of illness
 83 Infective Endocarditis
Box 83.1. Objectives of Echocardiographic Evalua-
tion When Infective Endocarditis Is Suspected
Detection of presence, location, and size of
vegetations
Evaluation of valve function (eg, valvular
regurgitation)
Identification of infected valve anatomy and
other possible companion diseases
Assessment of consequences of valvular
dysfunction (eg, left ventricular function,
pulmonary hypertension)
Detection of other intracardiac complications
(eg, myocardial abscess)
patients in whom IE is being considered, but it should not be
used to screen febrile patients in whom the diagnosis of IE is
unlikely. A negative result on TTE does not exclude the diag-
nosis of IE, especially when the clinical suspicion is high.
Visualization of valvular vegetations on TTE identifies patients
at a greater risk for development of IE complications. Serial
echocardiograms often show the persistence of vegetations after
successful therapy. The risk of vegetation embolization depends
on the infecting microorganisms; it is more common with viri-
dans streptococci. Mitral valve vegetations, especially those
located on the anterior leaflet, regardless of size, have the high-
est propensity to embolize.
TEE is more sensitive (95%) than TTE (60%–65%) for the
detection of intracardiac vegetations, particularly in the setting
of prosthetic valve endocarditis. TEE is the method of choice
for the detection of perivalvular extension of infection. TEE is
superior to TTE for the detection of myocardial abscesses, val-
vular perforations, lead vegetations in cardiac device recipients,
eustachian valve IE, and IE in the elderly. TEE should be consid-
ered for patients in whom IE is suspected and who have negative
results on TTE. However, as with TTE, a negative result on TEE
does not exclude the diagnosis of IE. Reasons for false-negative
Box 83.2. Modified Duke Criteria for Diagnosis of Endocarditis
Major Criteria
Blood culture positive for infective endocarditis with
typical organismsa from 2 separate blood cultures
Blood cultures persistently positive for any organismb
Findings on echocardiography consistent with infective
endocarditis, including vegetation, myocardial abscess,
or new partial dehiscence of a prosthetic valve
a
Typical microorganisms consistent with infective endocarditis include viridans streptococci, Streptococcus bovis, HACEK group, Staphylococcus
aureus; or community-acquired enterococci, in the absence of a primary focus; or single blood culture positive for Coxiella burnetii or antiphase
1 immunoglobulin G antibody titer more than 1:800.
b
At least 2 positive cultures of blood samples drawn more than 12 hours apart or all of 3 or a majority of 4 separate cultures of blood (with first and last
sample drawn at least 1 hour apart).
783
results on TEE include small vegetations and previous emboliza-
tion of vegetations. If the clinical suspicion for IE remains high
despite a negative results on TEE, TEE performed again in 7 to
10 days is a reasonable diagnostic strategy.
Multislice computed tomography may be useful for demon-
strating the anatomy of complex intracardiac complications of
IE. It is comparable to TEE with regard to detection of large
vegetations, abscess, and pseudoaneurysm, but it is inferior to
TEE for detection of valve perforations less than 2 mm and
vegetations less than 4 mm.
Diagnostic Criteria
The diagnosis of IE is based on a combination of clinical, bacterio-
logic, and echocardiographic findings incorporated in the modified
Duke criteria as proposed in 2000 (Box 83.2 and Figure 83.12).
Microbiology
Streptococci
The majority of patients in whom streptococcal endocarditis
develops (>80%) have underlying heart disease. Endocarditis in
young women with isolated mitral valve disease is almost always
caused by viridans streptococci. Approximately 20% of strepto-
coccal IE cases come to clinical attention because of emboli, and
cure rates exceed 80% to 90% with nonenterococcal streptococci
IE; complication rates remain high at approximately 30%. The
association of bacteremia due to Streptococcus bovis with colon
cancer and other gastrointestinal lesions warrants colonoscopy if
this organism is isolated from blood cultures.
In the pre-antibiotic era, S pneumoniae was responsible
for approximately 10% of cases of IE, but this percentage has
decreased to 1% to 3% currently. The clinical course is usu-
ally fulminant (mortality rate >50%) and is classically associ-
ated with perivalvular abscess formation or pericarditis or both.
Left-sided infection is the rule, and there is a strong predilection
for the aortic valve. Many adults with pneumococcal IE have a
background significant for alcohol abuse, and there is concurrent
development of meningitis in about 70% of cases.
Minor Criteria
Predisposing heart condition or injection drug use
Fever ≥38°C
Vascular phenomena: major arterial emboli, septic
pulmonary infarcts, mycotic aneurysm, intracranial
hemorrhage, conjunctival hemorrhages, and Janeway lesions
Immunologic phenomena: glomerulonephritis, Osler nodes,
Roth spots, and rheumatoid factor
Positive blood culture that does not meet a major criteria
or serologic evidence of active infection with organism
consistent with infective endocarditis
784 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
2 major criteria
Definite 1 major & 3 minor
criteria
5 minor criteria
Findings consistent with IE
Possible falling short of definite but
not meeting “rejected”
Firm alternative
diagnosis
Rejected Resolution of syndrome
≤4 days
No pathologic evidence of
IE after ABx for ≤4 days
Figure 83.12. Classification of Endocarditis Cases According to
Duke Criteria. ABx indicates antibiotics; IE, infective endocarditis.
Group B streptococci are normal inhabitants of the mouth,
vagina, and anterior urethra in 5% to 12% of the general pop-
ulation. Risk factors for group B streptococcal sepsis and IE in
adults include diabetes mellitus, malignancy, alcoholism, hepatic
failure, elective abortion, and intravenous drug use. Group B
streptococcal IE is associated with large friable vegetations and
increased frequency of major systemic embolic events. The mor-
tality rate is high (50%). Group A streptococci remain a rare
cause of IE and are associated with a high complication rate.
Enterococci
Enterococci are normal inhabitants of the gastrointestinal tract
and occasionally of the anterior urethra. Enterococcus organ-
isms are responsible for 5% to 18% of IE cases and the incidence
seems to be increasing. Most enterococcal bacteremias are noso-
comial in origin, often polymicrobial and are generally associated
with other underlying comorbid conditions. The disease usually
runs a subacute course and affects older men after genitourinary
manipulation or younger women after obstetric procedures. Over
40% of patients have no underlying structural heart disease and
the vast majority (>95%) develop a cardiac murmur during the
course of the illness. Cure can be difficult due to antimicrobial
resistance and mortality is high.
Staphylococci
S aureus is a major cause of IE in the industrialized world,
probably explained by the growing importance of health care
contact as a risk factor for S aureus bacteremia. IE develops
in about 10% of all patients who have staphylococcal bacter-
emia. The organism seeds normal valves in up to a third of
cases (Figure 83.13). The course is frequently fulminant when
Figure 83.13. Gram Stain of Staphylococcus aureus Vegetation
From a Patient With Infective Endocarditis.
it involves the mitral or aortic valve, and widespread metastatic
infection and death occur in 40% of cases.
Staphylococcal IE is associated with serious complica-
tions. Myocardial abscess, purulent pericarditis, and valve-ring
abscesses are more common in staphylococcal endocarditis than
in any other form of IE. Peripheral foci of suppuration involving
the brain, spleen, kidney, and lung are common, affecting more
than a third of patients, and their presence may serve as an early
indicator of the presence of underlying IE, especially in intrave-
nous drug users. Mortality rates can exceed 50% in the elderly.
Staphylococcus epidermidis is recognized as an important
agent of prosthetic valve IE and more recently has emerged as an
important cause of native valve endocarditis. Health care contact
is the primary risk factor for S epidermidis–associated IE. The
finding of S epidermidis in blood cultures typically represents an
uncomplicated bacteremia in more than 50% of cases, but dis-
tinction of bacteremia from IE is extremely important because S
epidermidis IE is associated with high mortality.
Gram-Negative Bacilli
Gram-negative aerobic bacilli are a rare cause of IE (2%) and
are commonly associated with injection drug use and health
care contact. Other risk factors include prosthetic heart valves,
implanted endovascular devices, and cirrhosis.
In uncomplicated gram-negative septicemia, the bloodstream
is usually cleared with appropriate antibiotic therapy; in con-
trast, in IE due to gram-negative bacilli, persistent bacteremia
is common, even with appropriate antimicrobial therapy. CHF
frequently develops and the prognosis is frequently poor, but
prompt surgical intervention can improve outcomes.
Salmonella species have an affinity for abnormal cardiac
valves. Valvular perforation and destruction, atrial thrombi,
myocarditis, and pericarditis are common, as is the development
of endarteritis and aneurysms of major vessels.
Pseudomonas IE occurs in injection drug users and usually
affects normal valves. Major embolic phenomena, neurologic
complications, ring and annular abscesses, splenic abscesses,
bacteremic relapses, and rapidly progressive CHF are common.
HACEK Organisms
HACEK organisms are normal inhabitants of the human orophar-
ynx and are reported to cause endocarditis in patients who have
dental infections or a history of dental procedures and in injec-
tion drug users who have “cleaned” the injection site with saliva.
 83 Infective Endocarditis
HACEK endocarditis is characterized by a lengthy (2 weeks-6
months) course before diagnosis, large friable vegetations, fre-
quent emboli, and the development of CHF with eventual need
for valve replacement. HACEK organisms are fastidious and
may require weeks for isolation in blood cultures.
Endocarditis in Intravenous Drug Users
The organisms responsible for causing IE in intravenous drug
users are different than those in other patients. In one series, the
relative frequencies of IE in injection drug users were as fol-
lows: 38%, S aureus; 14.2%, Pseudomonas aeruginosa; 13.8%,
Candida species; 8.2%, enterococci; 6%, viridans group of
streptococci; 1.7%, S epidermidis; 1.7% to 15%, gram-nega-
tive aerobic bacilli; 2.2%, other bacteria; 1.3%, polymicrobial
infections; and 12.9%, culture-negative IE. S aureus IE tends
to be less severe in drug users than in other patients, and the
mortality rates are 2% to 6%. The lower observed mortality is
related to the increased frequency of tricuspid valve involve-
ment, which is more responsive to medical management. HIV
infection may predispose injection drug users to IE due to
unusual pathogens.
Fungi
Fungal IE is rare, causing less than 1% of IE cases. Fungal IE
is usually due to Candida or Aspergillus species. Injection drug
use and health care contact are major risk factors. The cure rate
in fungal IE is poor. This result may, in part, be related to the
presence of large, bulky vegetations with widespread systemic
emboli; tendency for fungal invasion of the myocardium; poor
penetration of antifungal agents into the vegetation; low toxic to
therapeutic ratio of available antifungal agents; and usual lack
of fungicidal activity of these compounds (Figures 83.14 and
83.15). As a result a cure is almost impossible without surgical
intervention. There appears to be an increasing occurrence of
fungal endocarditis due to increased number of immunocompro-
mised patients, drug users, the extensive use of broad-spectrum
antimicrobial agents, and the use of indwelling central venous
catheters or parenteral nutrition.
Others
C burnetii, the etiologic agent of Q fever, can cause IE. Q fever is
usually a self-limited respiratory illness caused by the inhalation
Figure 83.14. Gross Appearance of Heart From a Patient With
Aspergillus Infective Endocarditis. Arrow points to vegetations.
785
Figure 83.15. Microscopic Appearance of Brain Abscess. (Figures
83.6, 83.14, and 83.15 are from the same patient.)
of infected aerosols. Most cases of C burnetii IE are chronic and
occur in males with preexisting heart disease. Risk factors may
include exposure to parturient cats or rabbits. The aortic valve is
involved in more than 80% of cases. Hepatosplenomegaly, hepa-
titis, and hematuria may be present in patients with C burnetii
IE. An immune complex deposition–mediated glomerulonephri-
tis occurs in up to 25% of cases.
Culture-Negative Endocarditis
Culture-negative IE accounts for a small proportion of cases
(<5%). It may occur because of recent administration of anti-
microbial agents before blood cultures are drawn; slow growth
of fastidious organisms; fungal endocarditis; endocarditis caused
by nonculturable intracellular organisms such as Bartonella spe-
cies, Chlamydia species, or Tropheryma whipplei; or the pres-
ence of noninfectious (marantic) endocarditis.
Clues to the diagnosis of culture-negative IE are summarized
in Box 83.3.
Therapeutic Approach
The response to antimicrobial therapy for IE is unique among
infections. Although the organisms may have in vitro sus-
ceptibility to antibiotics, complete eradication takes weeks to
achieve and relapse is not unusual. Parenteral antibiotics are
recommended over oral drugs in most circumstances because
of the importance of sustained antibacterial activity. There is
an emphasis on prolonged treatment (4–6 weeks). Bacteriostatic
agents are generally ineffective, and antibiotic combinations are
used to produce a rapid bactericidal effect. Serial blood cultures
should be obtained during the early phase of therapy to ensure
eradication of the bloodstream infection and in patients with per-
sistent or recurrent fever during therapy. The use of anticoagu-
lants during therapy for native valve IE has been associated with
fatal subarachnoid hemorrhage and other bleeding complications
and, as such, is considered contraindicated. In cases of IE local-
ized to mechanical prosthetic valves, many clinicians maintain
anticoagulation within therapeutic limits, provided there is no
evidence of major vascular emboli.
Patients with left-sided IE should receive initial management
in a facility in which cardiothoracic surgery can be performed.
The most common cause of persistent or recurrent fevers despite
appropriate antimicrobial selection is the presence of extensive
valve ring or paravalvular infection. Almost a third of patients
786 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
Box 83.3. Clues to the Diagnosis of Culture-Negative
Endocarditis
Epidemiologic clues
Travel to endemic areas—Coxiella burnetii, Brucella
species
Exposure to animals or their products—C burnetii,
Chlamydia psittaci, Brucella species, Bartonella
henselae
Risk factors for fungal endocarditis
Travel to areas with endemic fungi
Injection drug abuse—fungi, Corynebacterium
species
Homelessness, chronic alcoholism, human
immunodeficiency virus—Bartonella species
Underlying immunocompromised host—Listeria
species, Corynebacterium species, Legionella
species
Poor dental hygiene—HACEK group, Granulicatella
adiacens, Abiotrophia defective
Echocardiographic clues
Large vegetations—HACEK group, fungi
Vegetations with fingerlike projections—Chlamydia
species
Clinical clues
Periodontal disease, emboli—HACEK group,
Granulicatella adiacens, Abiotrophia defective
Underlying neoplasm (atrial myxoma,
adenocarcinoma, lymphoma, rhabdomyosarcoma,
carcinoid tumor)—noninfective endocarditis
Underlying autoimmune disease (rheumatic heart
disease, systemic lupus erythematosus)—
Libman-Sacks endocarditis, antiphospholipid
syndrome, polyarteritis nodosa, Behçet disease,
noninfective endocarditis
Postvalvular operation—noninfectious process
(eg, thrombus, sutures, other postvalvular surgical
change)
Miscellaneous conditions associated with
noninfective endocarditis (eg, eosinophilic heart
disease, ruptured mitral chordae, myxomatous
degeneration)
Abbreviation: HACEK, Haemophilus species, Actinobacillus actino-
mycetemcomitans, Cardiobacterium hominis, Eikenella corrodens,
and Kingella species.
with left-sided IE require surgery during the acute stages of the
illness for either valve replacement or metastatic infection.
Medical Therapy
Antimicrobial regimens, based on AHA guidelines for the man-
agement of endocarditis are summarized in the Appendixes.
Surgical Therapy
CHF and hemodynamic compromise are the strongest indica-
tions for surgery in IE. The hemodynamic status of the patient
at the time of valve replacement surgery is the main determi-
nant of operative mortality, and early surgical intervention needs
to be weighed against the advantage of “valve sterilization.” If
metastatic infection of other organs is present, surgery should be
delayed, if possible, to avoid relapse of infection of the prosthetic
valve caused by these sites of metastatic infection. Indications
for surgery in patients with native valve IE are summarized in
Box 83.4. Embolic neurologic complications during IE are asso-
ciated with a twofold to fourfold increase in mortality. Delay of
valve replacement for 2 or 3 weeks, however, is recommended in
embolic infarcts and for at least 1 month in intracerebral hemor-
rhages, if possible. In right-sided IE, persistent infection is the
usual indication for surgery. Most patients with right-sided IE
are injection drug users with endocarditis caused by organisms
that are difficult to eradicate with antimicrobial therapy alone.
Currently, tricuspid valvulectomy or resection of the vegetation
with valvuloplasty is the procedure of choice for refractory right-
sided IE.
Box 83.4. Indications for Cardiac Surgery in Patients
With Native Valve Infective Endocarditisa
Left-sided endocarditis
Accepted indications
Acute aortic regurgitation or mitral regurgitation with medically
uncontrolled heart failure
Acute aortic regurgitation with tachycardia and early closure of
the mitral valve
Fungal endocarditis
Evidence of valve dysfunction and persistent infection after
a prolonged period (7–10 days) of appropriate antimicrobial
therapy, as indicated by fever, leukocytosis, and bacteremia,
provided there are no noncardiac causes of infection
Relative indications
Evidence of abscess (annular or aortic) or aneurysm (aortic sinus
or aortic true aneurysm or pseudoaneurysm)
Recurrent emboli after appropriate antibiotic therapy
Infection with gram-negative organisms or organisms with a
poor response to antimicrobials in patients with evidence of
valve dysfunction
Right-sided endocarditis
Uncontrolled sepsis despite adequate antimicrobial
treatment
Intractable right heart failure despite appropriate
medical treatment
Paravalvular abscess or fungal endocarditis
Very large (>20 mm) vegetations (some authors use
the criteria of vegetation size >10 mm and persisting
fever as an indication for surgery)
a
Criteria also apply for repair of mitral and aortic allograft or
autograft valves.
Previously published. See “Credit Lines” section.
 83 Infective Endocarditis
Prosthetic Valve Endocarditis
PVE occurs in up to 6% of patients during the lifetime of the
prosthesis. Median age of patients with PVE is 65 years (range,
50–75 years). For mechanical prostheses, the incidence peaks
in the first few weeks after valve replacement and decreases to
a stable low incidence rate during subsequent months to years.
The risk of infection with mechanical and bioprosthetic valves
is similar, and there is no difference in the risk of endocarditis
between mitral prostheses and aortic prostheses. The cumulative
risk of PVE is highest within the initial 12 months after valve
replacement, and the peak is during the first 2 months. This dif-
ference in time to presentation has led to an arbitrary classifica-
tion of PVE as early when patients present in the first 60 days
after implantation and as late when patients present more than
60 days after surgery.
• PVE occurs in up to 6% of patients during the lifetime of the
prosthesis.
• PVE has been classified arbitrarily as early when it occurs within
the first 60 days after implantation and as late when it occurs more
than 60 days after surgery.
Pathogenesis
Early S epidermidis PVE is thought to result from valve contam-
ination during the perioperative period. This may occur at the
time of surgery or in the immediate postoperative period when
the prosthetic valve and sewing ring are not yet endothelialized
and are susceptible to microbial adherence. Nosocomial bactere-
mia, especially in patients with intravascular devices or hemodi-
alysis, is an important risk factor for PVE. In contrast to native
valves, injection drug use is not a frequent predisposing factor
for PVE.
The pathogenesis of late PVE is similar to that of native valve
endocarditis, with microorganisms from a transient bacteremia
localizing on a prosthesis or area of damaged endothelium.
Size and type of the valvular vegetation partly depend on
causative agents. S aureus, a highly virulent organism, is mostly
associated with small vegetations but a high likelihood of perival-
vular extension. In contrast, streptococci are slow-growing; they
result in large vegetations but milder destruction of surround-
ing tissues. Fungal vegetations on prosthetic valves are bulky
and may partially occlude the orifice or embolize and occlude
medium-sized arteries.
• Early PVE results from valve contamination during the periopera-
tive period.
• Late PVE results more often from transient bacteremia.
Valve-Ring Abscess
Valve-ring abscess is a serious complication of PVE and occurs
with both mechanical and bioprosthetic valves. Valve-ring
abscesses occur when infection involves the sutures used to
secure the sewing ring to the periannular tissue; this may result
in dehiscence of the valve and manifest as an echocardiographic
appearance of a “dancing” prosthesis. The clinical finding of a
new perivalvular leak in a patient with PVE is presumptive evi-
dence of a valve-ring abscess. Extension of the abscess beyond
the valve ring may result in myocardial abscess formation, sep-
tal perforation, aneurysm formation, or purulent pericarditis. In
addition to valve-ring abscesses, PVE of the bioprosthesis may
cause leaflet destruction, with resulting valvular incompetence
787
and clinical manifestations of heart failure. Large vegetations
occasionally obstruct blood flow and lead to functional valvu-
lar stenosis or a combination of stenosis and insufficiency. This
complication seems to be more common in mitral PVE than in
aortic disease.
Echocardiography
TTE is less accurate for the diagnosis of PVE than for the diag-
nosis of native valve endocarditis because the echoes generated
by the prosthesis may mask subtle abnormalities such as small
vegetations. TEE is more sensitive than TTE for the detection
of vegetations, periprosthetic tissue destruction with prosthetic
dehiscence, myocardial abscesses, fistulas, pseudoaneurysms,
and perivalvular abscesses.
Because infective endocarditis involving mechanical prosthe-
ses usually starts at the prosthetic ring, the search for vegetations
must focus on the prosthetic ring. In contrast, infective endo-
carditis of biologic prostheses involves both the ring and the val-
vular leaflets. For patients with negative findings on TEE and an
intermediate probability of PVE, a second evaluation with TEE
is recommended after a week, especially if an aortic prosthesis
could be involved.
Hematogenous seeding of the prosthetic valves is reported in
up to 50% of the patients with S aureus bacteremia and 40%
of those with coagulase-negative staphylococcal bloodstream
infection. Therefore, TEE should be performed in all patients
with prosthetic valves and staphylococcal bacteremia to exclude
possibility of PVE.
• TEE is the preferred imaging method for the diagnosis of PVE.
• Computed tomography or magnetic resonance imaging (or both)
of the head is indicated in patients with PVE and neurologic
symptoms.
• All patients with prosthetic valves and S aureus bacteremia should
have TEE to exclude PVE.
Diagnostic Criteria
The currently accepted diagnostic criteria for PVE are outlined
in Box 83.2 and Figure 83.12.
• Staphylococci are the most common cause of PVE in the first post-
operative year.
Treatment
Antimicrobial therapy is based on laboratory identification of
the etiologic microorganism and in vitro susceptibility testing.
Bactericidal antimicrobials with high intrinsic activity and
synergistic action are necessary. Recommended antimicro-
bial regimens are listed in the Appendixes at the end of this
chapter.
Because of rifampin’s ability to penetrate the biofilm, it (in
combination with a β-lactam agent or vancomycin) is recom-
mended for treatment of PVE due to staphylococci. However,
rifampin should not be used for native valve endocarditis or for
PVE due to nonstaphylococcal organisms. Aminoglycosides
(such as gentamicin) are recommended for the initial 2 weeks of
treatment for PVE due to staphylococci and for the entire treat-
ment course (up to 6 weeks) in cases of enterococcal PVE.
Fungal PVE usually requires combined medical and surgical
therapy. For Candida endocarditis, the combination of high doses
of amphotericin B given intravenously with oral flucytosine is
788 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
often used. Alternative considerations include use of caspofungin
or the combination of fluconazole and oral flucytosine.
• For methicillin-resistant staphylococcal PVE, treatment with
a combination of vancomycin, rifampin, and gentamicin is
recommended.
Indications for cardiac surgery in patients with PVE are
listed in Box 83.5. In selected patients with PVE, the results of
treatment with antimicrobial agents alone are comparable to
the results of combined surgical and medical therapy. Included
in this subgroup are patients with late-onset PVE (≥12 months
postoperatively) who are infected with less virulent organisms
(viridans streptococci, enterococci, and fastidious gram-negative
coccobacilli) and who do not have complicated endocarditis. For
these patients, medical therapy is recommended.
• Medical therapy alone is appropriate for selected patients with
PVE.
The timing of cardiac surgery in patients with PVE must be
individualized. The hemodynamic status of the patient is the
most important consideration. As in patients with native valve
endocarditis, the likelihood of those with PVE surviving valve
replacement is inversely related to the severity of the patient’s
heart failure at the time of operation. Thus, although in theory it
may be desirable to control infection with antimicrobial therapy
preoperatively, this must not be attempted at the expense of pro-
gressive destruction of perivalvular tissue and further deterio-
ration in the patient’s hemodynamic status. A cerebral embolic
event is not a contraindication for surgery if there is no hemor-
rhagic transformation and if the time between embolic event and
surgery is short (<72 hours).
• The patient’s hemodynamic status is the most important consider-
ation in determining the timing of operation.
Closely monitored anticoagulation therapy is recommended
for patients with mechanical PVE, which usually warrants
maintenance of anticoagulation. When surgery is anticipated
(Box 83.5), heparin should be used for anticoagulation instead
of warfarin. Also, antiplatelet therapy should be discontinued at
initial admission while the patient is being evaluated for the need
for surgical intervention and the possibility of cerebral emboli.
Prognosis
The mortality associated with PVE is 30% to 80% in the early
form and 20% to 40% in late postsurgical endocarditis, and a
worse prognosis is associated with advanced age (>75 years),
new or worsening heart failure, persistent bacteremia despite
use of appropriate antimicrobials, myocardial abscess, renal
insufficiency, S aureus as the causative agent, and neurologic
complications.
Prophylaxis for Infective Endocarditis
In 2007, the AHA issued sweeping new guidelines for antimicro-
bial prophylaxis for bacterial endocarditis. These changes rec-
ognized that there was minimal evidence that dental procedures
lead to bacterial endocarditis cases and that, even if prophylaxis
was 100% effective, it would prevent only a small number of
endocarditis cases. Emphasis has now shifted from antibiotic
prophylaxis to good oral health.
In the new AHA guidelines, the use of prophylactic antibiot-
ics on the basis of a patient’s lifetime risk for acquiring bacterial
Box 83.5. Indications for Surgery for Prosthetic
Valve Endocarditisa
≤2 months
Early prosthetic valve endocarditis (≤
after surgery)
Heart failure with prosthetic valve dysfunction
Fungal endocarditis
Staphylococcal endocarditis not responding to
antimicrobial therapy
Evidence of paravalvular leak, annular or
aortic abscess, aortic sinus or aortic true or
false aneurysm, fistula formation, or new-onset
conduction disturbances
Infection with gram-negative organisms with a
poor response to antimicrobials
Persistent bacteremia after prolonged course
(7–10 days) of appropriate antimicrobial therapy
without noncardiac causes of bacteremia
Recurrent peripheral embolus despite therapy
a
Criteria exclude repaired mitral valves and aortic allograft or
autograft valves.
Previously published. See “Credit Lines” section.
endocarditis is no longer recommended. Instead, prophylaxis
focuses on patients with the highest risk for poor outcomes from
endocarditis. Candidates for antimicrobial prophylaxis for endo-
carditis, on the basis of the 2007 AHA guidelines, are listed in
Box 83.6.
Recommendations for use of endocarditis prophylaxis for
specific procedures are listed in Box 83.7. The prophylactic regi-
mens for IE that are used before dental procedures are listed in
Table 83.3.
Infections Related to Cardiovascular Devices
The risk of a cardiovascular device becoming infected depends
on the type of the device and the host characteristics (Table 83.4).
When infection involves intravascular or endovascular por-
tions of a device, clinical manifestations resemble those of IE
or endarteritis. Fever is often present, as are embolic events,
which involve either the pulmonary or the systemic vasculature
(depending on the site of the device). The majority of cardiovas-
cular device–related infections involve S aureus and coagulase-
negative staphylococci.
Cardiovascular Implantable Electronic Devices
(Pacemakers and Implantable Cardioverter-
Defibrillators)
The reported rate of CIED infection, including permanent pace-
makers and implantable cardioverter-defibrillators, ranges from
0.13% to 19.9%. Implantable cardioverter-defibrillators have a
several-folds higher rate of infection than permanent pacemak-
ers. Several risk factors for CIED infection have been reported,
and these include temporary pacing leads before implantation,
lack of antibiotic prophylaxis at the time of implantation, fever
within 24 hours of implantation, presence of a tunneled cen-
tral venous catheter, operator inexperience, previous history of
 83 Infective Endocarditis
Box 83.6. Candidates for Antimicrobial Prophylaxis
for Endocarditis
Only patients with underlying conditions or
situations that place them at the highest risk
for poor outcomes from IE should receive
prophylaxis. These high-risk conditions include
Prosthetic heart valves
Previous history of IE
CHD that meets the following specific criteria:
Unrepaired cyanotic CHD, including palliative shunts and
conduits
Congenital heart defect completely repaired with prosthetic
material or a prosthetic device placed either during initial
surgery or by catheter intervention during the first 6 months after
the procedure
CHD repair with residual defects at the site of or adjacent to the
site of a prosthetic patch or prosthetic device
Development of cardiac valvulopathy after
cardiac transplant
Abbreviations: CHD, congenital heart disease; IE, infective
endocarditis.
Previously published. See “Credit Lines” section.
multiple revisions or infection, multiple transvenous leads, use
of corticosteroids, anticoagulation (higher risk of pocket hema-
toma), and comorbid conditions (diabetes mellitus, heart failure,
renal dysfunction, chronic skin conditions, and malignancy).
Microbial contamination of the device generator or leads with
skin flora at the time of implantation is the predominant mech-
anism of CIED infection.
Diagnosis of CIED infection is self-evident when patients
present with inflammatory findings at the generator pocket
(swelling, pain, erythema, skin erosion, and purulent drain-
age). However, bacteremia alone in the absence of pocket find-
ings may pose a diagnostic challenge. This is especially true in
patients with S aureus bacteremia, in whom the risk of under-
lying CIED infection is up to 50%. All such patients should
have echocardiography to evaluate for seeding of device leads
or cardiac valves. TEE is more sensitive for detecting lead or
valvular vegetations than TTE and should be performed in all
cases of S aureus bacteremia or blood cultures positive for other
organisms when no other obvious focus of infection is present.
Persistent S aureus bacteremia in the setting of an implanted
device, even in the absence of clinical evidence of CIED infec-
tion, should be managed with complete device removal and
antistaphylococcal antibiotics for 4 weeks from the day of first
negative blood cultures.
The approach to management of CIED infection and appro-
priate timing of reimplantation are outlined in Figures 83.16 and
83.17. These guidelines were initially proposed by Mayo Clinic
and were endorsed by the AHA in its 2010 scientific statement on
the management of CIED infections.
• Staphylococci are the most common cause of infections related to
CIEDs.
• Optimal treatment of CIED infections includes complete removal of
hardware along with antibiotics against the causative pathogens.
789
Box 83.7. Endocarditis Prophylaxis for Specific
Proceduresa
Dental procedures (prophylaxis is directed
against viridans group streptococci)
Prophylaxis is recommended for any procedures
that involve manipulation of gingival tissue or the
periapical region of the teeth or perforation of the
oral mucosa
Prophylaxis is not required for routine anesthetic
injections through noninfected tissue, dental
radiographs, placement of removable
prosthodontics or orthodontic appliances,
adjustment of orthodontic appliances, and
placement of orthodontic brackets
Respiratory procedures
It may be reasonable to give one of the
prophylactic regimens recommended
for dental procedures (Table 83.3) before
an invasive procedure (eg, tonsillectomy,
adenoidectomy) involving the respiratory tract that
necessitates incision or biopsy of the respiratory
mucosa
Prophylaxis is not recommended for bronchoscopy
unless the procedure involves incision of the
respiratory tract mucosa
Gastrointestinal or genitourinary procedures
Prophylaxis solely to prevent endocarditis is no
longer recommended
For patients scheduled for an elective urinary tract
manipulation who also have an enterococcal
urinary tract infection or colonization, it may be
reasonable to administer antibiotic therapy to
eradicate enterococci from the urine before the
procedure
If the urinary tract procedure is not elective, it may
be reasonable to administer an antimicrobial
regimen that contains an agent active against
enterococci
Amoxicillin or ampicillin is the preferred agent
for enterococcal coverage; vancomycin may be
administered to patients who cannot tolerate
ampicillin
Procedures involving infected skin, skin structure,
or musculoskeletal tissue
Regimen administered for treatment of the
infection should contain an agent active against
staphylococci and β-hemolytic streptococci
An anti-staphylococcal penicillin product or
cephalosporin is preferable; vancomycin may
be administered to patients unable to tolerate
a β-lactam or who are known or suspected to
have an infection caused by methicillin-resistant
staphylococcus
a
Prophylactic antibiotics should be administered only to patients with
the high-risk conditions listed in Box 83.6.
790 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Suspected CIED infection

Blood and generator-pocket cultures

Positive blood cultures or prior antibiotic treatment Negative blood cultures

TEE Pocket Generator/

infection lead erosion

Valve vegetation Lead vegetation Negative TEE

Treat with Treat with
antibiotics antibiotics
for 14 daysa for 7-10 daysa
Follow AHA Complicated Uncomplicated Other Staphylococcus
guidelines for (such as aureus
treatment of septic venous
infective thrombosis,
endocarditisa osteomyelitis) Treat with Treat with
antibiotics antibiotics
for 2 weeksa for 4 weeksa
Treat with 4-6
weeks of
antibioticsa

Figure 83.16. Approach to the Management of Adults With Cardiovascular Implantable Electronic Device Infection. AHA indicates American
Heart Association; CIED, cardiovascular implantable electronic device; TEE, transesophageal echocardiography. aDuration of antibiotics should be
counted from the day of device explantation. (Previously published. See “Credit Lines” section.)

Implantation of new CIED

Blood cultures + Blood culture + Generator-pocket infection/

TEE + TEE - generator or lead erosion

Repeat blood cultures after Repeat blood cultures after Negative admission blood
device removal device explantation cultures for 72 hours

Valve Lead Implant if repeat blood cultures Implant once adequate

vegetation vegetation are negative for at least 72 hours débridement is achieved

Implant Implant if
device after repeat blood
14 days of cultures are
first negative negative for
blood culture 72 hours

Figure 83.17. Guidelines for Implantation of New Device in Patients With Cardiovascular Implantable Electronic Device Infection. CIED indi-
cates cardiovascular implantable electronic device; TEE, transesophageal echocardiography; +, positive; −, negative. (Previously published. See
“Credit Lines” section.)
 83 Infective Endocarditis 791

Table 83.3. Prophylactic Regimens for Infective Endocarditis Before Dental Proceduresa,b
Adult Prophylaxis
Clinical Situation (Choose One Option)
Oral regimen Amoxicillin 2 g oral
Unable to take oral medication Ampicillin 2 g IM or IV
or
cefazolin 1 g IM or IV
or
ceftriaxone 1 g IM or IV
Allergy to penicillin or ampicillin Cephalexinc,d 2 g oral
(oral regimen) or
clindamycin 600 mg oral
or
azithromycin 500 mg oral
or
clarithromycin 500 mg oral
Allergy to penicillin or ampicillin Cefazolind 1 g IM or IV
(unable to take oral medication) or
ceftriaxoned 1 g IM or IV
or
clindamycin 600 mg IM or IV
Pediatric Prophylaxis
(Choose One Option)
Amoxicillin 50 mg/kg oral
Ampicillin 50 mg/kg IM or IV
or
cefazolin 50 mg/kg IM or IV
or
ceftriaxone 50 mg/kg IM or IV
Cephalexinc,d 50 mg/kg oral
or
clindamycin 20 mg/kg oral
or
azithromycin 15 mg/kg oral
or
clarithromycin 15 mg/kg oral
Cefazolind 50 mg/kg IM or IV
or
ceftriaxoned 50 mg/kg IM or IV
or
clindamycin 20 mg/kg IM or IV

Abbreviations: IM, intramuscularly; IV, intravenously.

a
Give single dose 30 to 60 minutes before procedure.
b
If the antibiotic is inadvertently not administered before the procedure, it may be administered up to 2 hours
after the procedure.
c
Or another first- or second-generation cephalosporin in equivalent dose.
d
Do not use cephalosporin in patients with a history of anaphylaxis, angioedema, or urticaria with penicillin
or ampicillin.
Previously published. See “Credit Lines” section.

Left Ventricular–Assist Devices healing at the driveline exit site, blood transfusion, prolonged
hospitalization, poor nutritional status of the host, and comor-
Infection is a frequent complication of left ventricular–assist
bid conditions (heart failure, renal failure, diabetes mellitus, and
devices, and reported rates range from 25% to 70%. Purported
obesity). Infection most commonly occurs when the device has
risk factors for infection related to left ventricular–assist devices
been in place for at least 2 weeks.
include length and type of surgery required for device implanta-
Infections related to left ventricular–assist devices can be
tion, postoperative hemorrhage, device revision, delayed wound
divided into 3 different syndromes. Infection involving the drive-
line is the most common type, typically with local inflamma-
Table 83.4. Nonvalvular Cardiovascular Device–Related tory changes and drainage at the cutaneous exit site. The second
Infections syndrome involves infection at the left ventricular–assist device
pocket, which causes local inflammatory changes. The third, and
Type of Device Rate of Infection, % relatively infrequent, syndrome is endocarditis due to infection
involving either the valve or the internal lining (ie, parts in con-
Intracardiac
tact with the blood) of the device or both. Patients can have more
Permanent pacemaker 0.13–19.9
Implantable cardioverter-defibrillator 0–3.2
than 1 type of infection at the same time.
Left ventricular–assist devices 13–80 Staphylococci are the main bacteria isolated, followed by gram-
Ventriculoatrial shunts 2.4–9.4 negative bacilli (P aeruginosa and Escherichia coli), Enterococcus
Pledgets (cardiac suture line) Rare species, Corynebacterium species, and Candida species. Because
Patent ductus arteriosus occlusion devices Rare most of these infections are hospital-acquired, antimicrobial
Atrial septal defect closure devices Rare resistance is common. Antimicrobial therapy should be directed
Conduits Rare toward the causative organism and administered for 3 to 4 weeks.
Patches Rare Limited débridement of an infected driveline exit site or pocket
Arterial may help to control localized infection in some patients. For unre-
Peripheral vascular stents 0.05–4 solved infection, the device may need to be removed. Importantly,
Vascular grafts, including hemodialysis 1–6 left ventricular–assist device infection, including persistent bacte-
Intra-aortic balloon pumps ≤5–27 remia or fungemia, is not a contraindication to cardiac transplant.
Angioplasty- or angiography-related bacteremias <1
Vascular closure devices 0.0–5.1
Coronary artery stents Rare Coronary Artery Stents
Patches 1.8 Infections involving intracoronary stents are rare but often fatal.
Venous They are due to contamination of a stent at delivery or subsequent
Vena caval filters Rare
transient bacteremia. Reported pathogens include S aureus and
Previously published. See “Credit Lines” section. P aeruginosa. Associated findings consist of local myocardial
792 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

abscess, aneurysm formation, coronary artery perforation, sup- device deployment to presentation with infection is 1 week. The
purative pericarditis, and sepsis. Computed tomography arteri- majority of the patients present with local inflammatory find-
ography or cardiac magnetic resonance imaging may be helpful ings at the device deployment site, with or without associated
for making the diagnosis when TEE results are inconclusive. bloodstream infection. Mycotic pseudoaneurysm formation is
Surgical débridement and resection of the infected stent are reported in up to 50% of the cases. S aureus is the predomi-
needed to cure the infection. Reported mortality is up to 50%. nant pathogen in these cases. Treatment requires surgical débri-
dement followed by 3 to 4 weeks of parenteral antimicrobial
Intra-aortic Balloon Pumps therapy.
Infections involving intra-aortic balloon pumps are rare. Most
cases of bacteremia are due to spread from a colonized or infected
insertion site. Risk factors include contamination of the femoral Abbreviations
area, especially in obese patients, and insertions performed in an
AHA American Heart Association
intensive care unit especially on an emergency basis (compared CHF congestive heart failure
within an operating room or cardiac catheterization suite). Risk CIED cardiovascular implantable electronic device
of infection increases with duration of intra-aortic balloon pump HACEK Haemophilus species, Actinobacillus actinomycetemcomi-
therapy. Treatment consists of appropriate antimicrobial therapy tans, Cardiobacterium hominis, Eikenella corrodens, and
and local wound care in addition to removal of the intra-aortic Kingella species
balloon pump, if possible. HIV human immunodeficiency virus
IE infective endocarditis
Vascular Closure Devices PCR polymerase chain reaction
PVE prosthetic valve endocarditis
Infectious complications are reported in up to 5% of recipients TEE transesophageal echocardiography
of vascular closure devices. Median incubation period from TTE transthoracic echocardiography
 83 Infective Endocarditis 793

Appendix

Appendix 83.1. Therapy for Native Valve Endocarditis Caused by Highly Penicillin-Susceptible Viridans Group Streptococci and
Streptococcus bovis (MIC ≤0.12 mcg/mL)
Regimen Dosagea and Route Duration Comments

Penicillin G 12–18 million units per day IV either 4 wk Preferred options in most patients >65 y or with
or continuously or in 4–6 equally divided doses impaired 8th cranial nerve function or impaired renal
ceftriaxone 2 g IV or IM q 24 h 4 wk function

Penicillin G 12–18 million units per day IV, either 2 wk The 2-week regimen is not intended for patients
or continuously or in 6 equally divided doses with known cardiac or extracardiac abscess or for
ceftriaxone 2 g IV or IM q 24 h 2 wk those with ClCr <20 mL/min, impaired 8th cranial
(1 of the above 2 agents) plus nerve function, or infection with Abiotrophia,
Granulicatella, or Gemella spp
gentamicinb 3 mg/kg IV or IM q 24 h 2 wk Use nomogram for once-daily dosing of gentamicin

Vancomycinc 30 mg/kg every 24 h IV in 2 equally divided 4 wk Use vancomycin for patients unable to tolerate penicillin
doses not to exceed 2 g q 24 h unless or ceftriaxone
serum levels are inappropriately low Adjust dosage to obtain peak (1 hour after infusion)
serum level of 30–45 mcg/mL and trough level of
10–15 mcg/mL
Abbreviations: ClCr, creatinine clearance; IM, intramuscularly; IV, intravenously; MIC, minimal inhibitory concentration; q, every.
a
Recommended dosages are for adult patients with normal renal function.
b
Other potentially nephrotoxic drugs (eg, nonsteroidal anti-inflammatory drugs) should be used with caution in patients receiving gentamicin therapy.
c
Vancomycin dosages should be infused over at least 1 hour to reduce the risk of histamine release (red man syndrome).
Previously published. See “Credit Lines” section.

Appendix 83.2. Therapy for Native Valve Endocarditis Caused by Relatively Penicillin-Resistant Strains of Viridans Group
Streptococci and S bovis (MIC >0.12 mcg/mL to ≤0.5 mcg/mL)
Regimen Dosagea and Route Duration Comments

Penicillin G 24 million units per day IV either continuously 4 wk Use an enterococcal endocarditis regimen for patients
or or in 4–6 equally divided doses with endocarditis caused by penicillin-resistant
ceftriaxone 2 g IV or IM q 24 h 4 wk strains (MIC >0.5 mcg/mL)
(1 of the above 2 agents) plus
gentamicin 3 mg/kg IV or IM q 24 h 2 wk

Vancomycinb 30 mg/kg every 24 h IV in 2 equally 4 wk Use vancomycin only for patients unable to tolerate
divided doses not to exceed 2 g q 24 h penicillin or ceftriaxone
unless serum levels are inappropriately low
Abbreviations: IV, intravenously; MIC, minimal inhibitory concentration; q, every.
a
Recommended dosages are for adult patients with normal renal function.
b
Adjust vancomycin dosage to obtain a peak (1 hour after infusion) serum level of 30 to 45 mcg/mL and a trough level of 10 to 15 mcg/mL.
Previously published. See “Credit Lines” section.
794 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Appendix 83.3. Therapy for Endocarditis of Prosthetic Valves or Other Prosthetic Material Caused by Viridans Group Streptococci
and Streptococcus bovis
Regimen Dosagea and Route Duration Comments

Penicillin-Susceptible Strain (MIC ≤0.12 mcg/mL)

Penicillin G 24 million units per day IV either continuously 6 wk Use of either penicillin with gentamicin or ceftriaxone
or or in 4–6 equally divided doses with gentamicin has not demonstrated superior cure
2 g IV or IM q 24 h rates compared with monotherapy with penicillin
ceftriaxone 6 wk or ceftriaxone for patients with a highly susceptible
(1 of the above 2 agents) plus strain
optional addition of 3 mg/kg IV or IM q 24 h
gentamicin 2 wk Do not use gentamicin in patients with ClCr <30 mL/min

Vancomycinb 30 mg/kg every 24 h IV in 2 equally 6 wk Use vancomycin only for patients unable to tolerate
divided doses penicillin or ceftriaxone

Penicillin-Resistant (Relatively or Fully) Strain (MIC >0.12 mcg/mL)

Penicillin G 24 million units per day IV either continuously 6 wk

or or in 4–6 equally divided doses

ceftriaxone 2 g IV or IM q 24 h 6 wk
(1 of the above 2 agents) plus 6 wk
gentamicin 3 mg/kg IV or IM q 24 h

Vancomycinb 30 mg/kg every 24 h IV in 2 equally 6 wk Use vancomycin only for patients unable to tolerate
divided doses penicillin or ceftriaxone
Abbreviations: IV, intravenously; MIC, minimal inhibitory concentration; q, every.
a
Recommended dosages are for adult patients with normal renal function.
b
Adjust vancomycin dosage to obtain a peak (1 hour after infusion) serum level of 30 to 45 mcg/mL and a trough level of 10 to 15 mcg/mL.
Previously published. See “Credit Lines” section.

Appendix 83.4. Therapy for Endocarditis Caused by Staphylococci in the Absence of Prosthetic Materials
Regimen Dosagea and Route Duration Comments

Oxacillin-Susceptible Strains

Nafcillinb 12 g/24 h IV in 4–6 equally divided doses 6 wk Use 6 weeks for complicated right-sided IE and for
or left-sided IE; or use 2 weeks for uncomplicated
oxacillinb 12 g/24 h IV in 4–6 equally divided doses 6 wk right-sided IE
(1 of the above 2 agents)
plus optional addition of
gentamicinc 3 mg/kg every 24 h IV or IM in 3–5 days Clinical benefit of aminoglycosides has not been
2–3 equally divided doses established
For penicillin-allergic Consider skin testing for oxacillin-susceptible
(non-anaphylactoid type) patients: staphylococci and with a questionable history of
immediate-type hypersensitivity to penicillin
cefazolin 6 g/24 h IV in 3 equally divided doses 6 wk Avoid cephalosporins in patients with anaphylactoid-
plus optional addition of type hypersensitivity to β-lactams; use
vancomycind instead
gentamicinb 3 mg/kg every 24 h IV or IM in 3–5 days Clinical benefit of aminoglycosides has not been
2–3 equally divided doses established

Oxacillin-Resistant Strains

Vancomycind 30 mg/kg every 24 h IV in 2 equally 6 wk Adjust vancomycin to achieve 1-hour peak serum
divided doses level of 30–45 mcg/mL and trough level of
10–15 mcg/mL
Abbreviations: IE, infective endocarditis; IM, intramuscularly; IV, intravenously.
a
Recommended dosages are for adult patients with normal renal function.
b
Use penicillin G 24 million units per day in place of nafcillin or oxacillin for penicillin-susceptible strain (minimal inhibitory concentration <0.1 mcg/mL).
c
Administer gentamicin in close temporal proximity to vancomycin, nafcillin, or oxacillin dosing.
d
Adjust vancomycin dosage to obtain a peak (1 hour after infusion) serum level of 30–45 mcg/mL and a trough level of 10–15 mcg/mL.
Previously published. See “Credit Lines” section.
 83 Infective Endocarditis 795

Appendix 83.5. Therapy for Prosthetic Valve Endocarditis Caused by Staphylococci

Regimen Dosagea and Route Duration Comments

Oxacillin-Susceptible Strains

Nafcillin 12 g/24 h IV in 6 equally divided doses ≥6 wk Use penicillin G 24 million units per day instead of nafcillin or
or oxacillin if strain is penicillin-susceptible
(MIC ≤0.1 mcg/mL) and does not produce β-lactamase
oxacillin 12 g/24 h IV in 6 equally divided doses ≥6 wk Use vancomycin in patients with immediate-type hypersensitivity
(1 of the above 2 agents) plus reactions to β-lactam antibiotics
rifampin 900 mg/24 h IV or oral in 3 equally ≥6 wk Substitute cefazolin for nafcillin or oxacillin in patients with
and plus divided doses nonimmediate-type hypersensitivity reactions to penicillin
gentamicinb 3 mg/kg every 24 h IV or IM in 2 wk
2–3 equally divided doses

Oxacillin-Resistant Strains

Vancomycin 30 mg/kg every 24 h IV in 2 equally ≥6 wk Adjust vancomycin to achieve 1-hour peak serum level of
plus divided doses 30–45 mcg/mL and trough level of 10–15 mcg/mL
rifampin 900 mg/24 h IV or oral in 3 equally ≥6 wk
and plus divided doses
gentamicin 3 mg/kg every 24 h IV or IM in 2 wk
2–3 equally divided doses
Abbreviations: IM, intramuscularly; IV, intravenously; MIC, minimal inhibitory concentration.
a
Recommended dosages are for adult patients with normal renal function.
b
Adjust gentamicin dosage to achieve a peak serum level of 3 to 4 mcg/mL and a trough level of less than 1 mcg/mL.
Previously published. See “Credit Lines” section.

Appendix 83.6. Therapy for Native Valve or Prosthetic Valve Enterococcal Endocarditis Caused by Strains Susceptible to
Penicillin, Gentamicin, and Vancomycina
Regimen Dosageb and Route Duration Comments
Ampicillin 12 g/24 h IV in 6 equally divided doses 4–6 wk Native valve: use 4-week therapy for patients with
or symptoms of illness lasting ≤3 months and
penicillin G 18–30 million units per day IV either 4–6 wk 6-week therapy for patients with symptoms lasting
(1 of the above 2 agents) plus continuously or in 6 equally divided doses >3 months
gentamicinc 3 mg/kg every 24 h IV or IM in 3 equally 4–6 wk Prosthetic valve or other prosthetic cardiac material:
divided doses use 6-week minimum therapy
Vancomycind 30 mg/kg every 24 h IV in 2 equally 6 wk Use vancomycin only for patients unable to tolerate
plus divided doses penicillin or ampicillin
gentamicinc 3 mg/kg every 24 h IV or IM in 3 equally 6 wk Use 6 weeks of vancomycin therapy because of
divided doses decreased activity against enterococci
Abbreviations: IM, intramuscularly; IV, intravenously.
a
For strains resistant to gentamicin and susceptible to streptomycin, substitute streptomycin IV or IM 15 mg/kg every 24 h in 2 divided doses. See full-text
article of Baddour (Credit Lines) for management of enterococcal infective endocarditis strains that are penicillin-resistant and management of strains resistant
to penicillin, aminoglycosides, and vancomycin.
b
Recommended dosages are for adult patients with normal renal function.
c
Adjust gentamicin dosage to achieve a peak serum level of 3 to 4 mcg/mL and a trough level of less than 1 mcg/mL. Patients with creatinine clearance less than
50 mL/min should be treated in consultation with an infectious diseases specialist.
d
Adjust vancomycin dosage to obtain a peak (1 hour after infusion) serum level of 30 to 45 mcg/mL and a trough level of 10 to 15 mcg/mL.
Previously published. See “Credit Lines” section.
796 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Appendix 83.7. Therapy for Both Native Valve and Prosthetic Valve Endocarditis Caused by HACEKa Microorganisms
Regimen Dosageb and Route Duration Comments
Ceftriaxonec 2 g IV or IM q 24 h 4 wk Substitute cefotaxime or another third- or fourth-
or generation cephalosporin for ceftriaxone
ampicillin-sulbactam 12 g every 24 h IV in 4 equally divided doses 4 wk
or
ciprofloxacinb,c,d 1,000 mg every 24 h oral or 800 mg every 4 wk Use fluoroquinolone therapy only for patients unable to
24 h IV in 2 equally divided doses tolerate a cephalosporin and ampicillin; levofloxacin or
moxifloxacin may be substituted; fluoroquinolones not
generally recommended for patients <18 y
Abbreviations: IM, intramuscularly; IV, intravenously.
a
Haemophilus species, Actinobacillus actinomycetemcomitans, Cardiobacterium hominis, Eikenella corrodens, and Kingella species.
b
Recommended dosages are for adult patients with normal renal function.
c
Patients should be informed that IM injection of ceftriaxone is painful.
d
Fluoroquinolones are highly active in vitro against HACEK microorganisms. Published data on use of fluoroquinolone therapy for endocarditis caused by
HACEK are minimal.
Previously published. See “Credit Lines” section.
 84

Systemic Disease and the Heart

GRACE LIN, MD, MARIAN T. MCEVOY, MD,
and JOSEPH G. MURPHY, MD

Many systemic diseases affect the heart either directly or indi- Duchenne Muscular Dystrophy
rectly. This chapter summarizes the systemic diseases most fre-
Duchenne muscular dystrophy is an X-linked recessive dis-
quently encountered in clinical practice that affect the heart. An
order characterized by the absence of the protein dystrophin
important issue is how the systemic disease modifies either the
normally found on the sarcolemma of muscle cells. The disor-
diagnosis or the management of the cardiac problem and how the
der affects all muscle types, and patients typically present with
cardiac disease, in turn, modifies the approach to the systemic
weakness at 2 to 3 years of age. The incidence of dilated car-
disease.
diomyopathy is high because of extensive fibrosis selectively
involving the posterior wall of the left ventricle, sometimes
Neurologic Disease including the posterolateral papillary muscle and resulting in
mitral regurgitation. Typical ECG findings are tall R waves
Cardiac involvement in neurologic disease is unusual. The most
in lead V1 and Q waves in inferolateral leads I, aVL, V5, and
important examples are summarized below.
V6. Patients may be asymptomatic because they cannot exer-
cise; periodic cardiac imaging is recommended. Heart block,
Friedreich Ataxia inappropriate sinus tachycardia, and atrial and ventricular
Friedreich ataxia is an autosomal recessive neuromuscular disor- arrhythmias are common.
der characterized by ataxia, muscle weakness, and sensory loss.
Associated cardiac abnormalities include variant hypertrophic
cardiomyopathy, which differs from the classic type in that sep- Becker Muscular Dystrophy
tal myofibrillary disarray is absent and left ventricular systolic
Becker muscular dystrophy is also an X-linked recessive disorder
and diastolic function are relatively normal. Concentric left ven-
caused by mutation of the dystrophin gene. Clinical manifesta-
tricular hypertrophy is more common than asymmetric septal
tion is similar but milder than Duchenne muscular dystrophy,
hypertrophy. In most cases, the course is relatively benign; how-
with weakness occurring at a later age. Dilated cardiomyopa-
ever, if dilated cardiomyopathy with clinical heart failure devel-
thy is prevalent; however, right ventricular involvement occurs
ops, patients have a poor prognosis. Malignant arrhythmias are
before left ventricular involvement. Conduction system disease,
rare, but conduction abnormalities may be present in about 10%
including atrioventricular block and bundle branch block, can
of patients. The most common ECG finding is diffuse T-wave
also occur and may progress to complete heart block.
inversion.
• Periodic cardiac imaging is recommended for Duchenne muscular
• Hypertrophic cardiomyopathy seen with Friedreich ataxia is more dystrophy and Becker muscular dystrophy.
benign than the classic form.
• Angiotensin-converting enzyme inhibitor and β-blocker therapy
may be beneficial for treating dilated cardiomyopathy in Becker
Abbreviations and acronyms are expanded at the end of this chapter. muscular dystrophy and Duchenne muscular dystrophy.

797
798 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Myotonic Muscular Dystrophy and sodium restriction. Cardiovascular changes in acromegaly

Myotonic muscular dystrophy is an autosomal dominant neuro-
muscular disorder characterized by myotonia, delayed relaxation
of skeletal muscles after contraction, and clinical manifestation
in adolescence. The ECG is often abnormal, with deep Q waves
in the absence of myocardial infarction; atrioventricular block
and bundle branch block can also occur, sometimes progressing
to complete heart block. Although atrial arrhythmias are more
common, patients are also at risk of ventricular arrhythmias
and sudden cardiac arrest. In most cases, cardiac involvement
is limited to conduction disturbances, but dilated cardiomyop-
athy, regional wall motion abnormalities, and heart failure may
develop. Screening ECGs are recommended.
Kearns-Sayre Syndrome
Kearns-Sayre syndrome is a genetic disease transmitted through
mitochondrial DNA and characterized by progressive external
ophthalmoplegia, pigmentary retinopathy, and heart block that
frequently requires permanent pacemaker implantation.
• Regardless of symptoms, patients with neuromuscular disorders
and high-grade, second-degree, or complete heart block have a
class I indication for permanent pacemaker because of concern for
unpredictable progression of atrioventricular block.
may respond to suppression of growth hormone concentrations
with octreotide.
Thyroid Disease
In hyperthyroidism, the heart is hyperdynamic, with an increase
in heart rate, systolic blood pressure, cardiac output, and stroke
volume and a decrease in systemic vascular resistance. Atrial
fibrillation occurs in 25% of patients and may be associated
with heart failure. Since cardioversion may be unsuccessful,
β-blockers or calcium channel blockers are the drug of choice
for rate control pending definitive treatment of hyperthyroidism.
About 40% of patients have spontaneous reversion from atrial
fibrillation to sinus rhythm when they become euthyroid. It is
important to check for occult hyperthyroidism in patients with
unexplained new-onset atrial fibrillation. Ventricular tachycardia
can occur in uncontrolled hyperthyroidism (called thyroid crisis
or thyroid storm).
Hypothyroidism is associated with a decrease in heart rate,
ventricular contractility, and cardiac output but an increase in
systemic vascular resistance and systemic blood pressure. It may
also be associated with pericardial effusion and an abnormal
lipid profile (an increase in low-density lipoprotein cholesterol
and triglyceride levels).

Guillain-Barré Syndrome Cushing Syndrome

Guillain-Barré syndrome, a nonhereditary demyelinating neu- Cushing syndrome results from glucocorticoid excess due to
ropathy associated with autonomic dysfunction, can result in corticotropin-producing adenomas of the pituitary, ectopic
labile blood pressure and sinus tachycardia. Bradyarrhythmias corticotropin-producing tumors, or primary adrenal tumors.
and pauses are usually the result of increased vagal activity. Iatrogenic Cushing disease may be caused by the therapeu-
Ventricular arrhythmias and, rarely, sudden death may also tic use of corticosteroids (Figure 84.1). Cushing syndrome is
occur. Neurogenic stunned myocardium, presumed to result associated with hyperkalemia, systemic hypertension, and left
from catecholamine toxicity causing left ventricular dysfunction, ventricular hypertrophy. Hypertension in Cushing syndrome is
usually resolves with resolution of the clinical illness. relatively resistant to conventional antihypertensive medications
but may respond to ketoconazole because of its inhibitory effect
on adrenal enzymes.
Cerebral Hemorrhage
Head injuries and cerebral hemorrhage may be associated with
marked ECG changes, including QT prolongation, prominent U
waves, ST-segment elevation or depression, and deep symmetri-
cal T-wave inversion in the precordial leads. Neurogenic stunned
myocardium has resulted in an apical ballooning phenomenon,
usually with subsequent improvement in left ventricular systolic
function.
• Neurogenic stunned myocardium can occur with Guillain-Barré
syndrome or cerebral hemorrhage and often resolves with improve-
ment in left ventricular ejection fraction.

Endocrine and Metabolic Disease

Acromegaly
Acromegaly, a growth hormone disorder, is associated with
cardiomegaly, hypertension, focal myocardial fibrosis, lympho-
cytic myocarditis, and premature atherosclerosis. Acromegalic
cardiomyopathy is characterized by myocardial fibrosis with
degeneration of myofibrils and by left ventricular hypertrophy
with an increased left ventricular mass index due to cham-
ber dilatation, which may result in high-output heart failure.
Hypertension in acromegaly results from a low level of renin due
to plasma volume expansion, which responds well to diuretics Figure 84.1. Exogenous Cushing Disease.
 84 Systemic Disease and the Heart
Addison Disease
Addison disease and other causes of adrenal insufficiency, includ-
ing abrupt cessation of corticosteroid medication, are associated
with arterial hypotension, postural hypotension, and syncope.
Specific ECG changes are low-voltage amplitudes, prolonged QT
interval, and sinus bradycardia.
Infiltrative Cardiomyopathies
and Storage Diseases
Fabry Disease
Fabry disease is an X-linked recessive lysosomal storage dis-
order that results in glycosphingolipid infiltration of the heart,
skin, brain, and kidneys. Infiltration of cardiac myocytes results
in ventricular hypertrophy and dysfunction, myocardial ischemia
and infarction, endothelial dysfunction, cardiac valve thickening
and incompetence, and arrhythmias.
Glycogen Storage Diseases
In type II glycogen storage disease (Pompe disease) and type III
disease, cardiomyopathy occurs because of myocardial deposi-
tion of abnormal glycogen, leading to cardiomyopathy.
Amyloidosis
Amyloid infiltration of the heart may occur in the senile, pri-
mary, and familial forms of amyloidosis and lead to restrictive
Figure 84.2. Cardiac Amyloidosis. A, Cardiac magnetic resonance images showing diffuse delayed enhancement abnormality (arrows).
B, Echocardiographic images showing granular, sparkling 2-dimensional appearance. Left panels, Apical 4-chamber view. Right panels, Parasternal
long-axis view.
799
cardiomyopathy. The prognosis is poor, especially in the primary
form; patients with senile cardiac amyloidosis have a better prog-
nosis. Cardiac transplant may be appropriate in selected cases.
Although primary amyloidosis affects multiple organs, senile
amyloidosis often exclusively affects the heart or manifests with
minimal noncardiac involvement. Amyloid deposition in the
myocardium leads to left ventricular hypertrophy and results in
a typical sparkling appearance on 2-dimensional echocardiogra-
phy, abnormal ventricular strain imaging, progressive diastolic
dysfunction, and valvular and pericardial disease. Cardiac mag-
netic resonance imaging shows diffuse delayed enhancement
abnormality (Figure 84.2). In contrast to the ECG for patients
with other diseases resulting in left ventricular hypertrophy,
amyloidosis causes low-voltage amplitudes (Figure 84.3).
Sarcoidosis
Sarcoidosis, characterized by noncaseating granulomas in
affected organs, can involve multiple organs, including the lungs,
skin, eyes, and reticuloendothelial system. Cardiac involvement
occurs in up to 20% to 30% of cases. Infiltration of the conduc-
tion system and myocardium causes complete heart block, ven-
tricular arrhythmias, sudden cardiac arrest, and left ventricular
dysfunction or pulmonary hypertension and right heart failure
due to pulmonary sarcoidosis. Typical echocardiographic fea-
tures include a dilated left ventricle with regional wall motion
abnormalities; aneurysms may occur, often in the posterobasal
region. Cardiac magnetic resonance imaging shows a patchy
delayed enhancement abnormality (Figure 84.4).
800 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

V1

V5
Figure 84.3. Cardiac Amyloidosis. Electrocardiogram shows low-voltage QRS complexes and a pseudoinfarction pattern with absent R waves in
leads V1 through V3.

Rheumatologic Diseases Routine screening with ultrasonography during pregnancy is

Rheumatoid Arthritis
Rheumatoid arthritis may involve all cardiac structures, includ-
ing the pericardium, valves, myocardium, conduction system,
coronary arteries, aorta, and pulmonary circulation. Rheumatoid
arthritis can cause both granulomatous and nongranulomatous
inflammation of the cardiac valve leaflets, rarely leading to
severe mitral or aortic valve incompetence. The myocardium
may be involved by an inflammatory myocarditis or, rarely, by
the deposition of amyloid. Pericarditis in rheumatoid arthritis is
characterized by a low glucose concentration and complement
depletion in the pericardial fluid. Constrictive pericarditis can
also occur. Rheumatoid nodules may be deposited in the conduc-
tion system, leading to all degrees of heart block. Aortitis and
pulmonary hypertension due to pulmonary vasculitis are rare
complications.
• Pericarditis in rheumatoid arthritis is characterized by a low glu-
cose concentration and complement depletion in the pericardial
fluid.
Systemic Lupus Erythematosus
Systemic lupus erythematosus (Figure 84.5) can involve all
cardiac structures, but most commonly it involves the pericar-
dium, resulting in pericarditis and a pericardial effusion in up to
two-thirds of patients. Special features of cardiac involvement
in systemic lupus erythematosus include nonbacterial endocar-
ditis, antiphospholipid antibody syndrome, and congenital heart
block in the offspring of mothers with subacute cutaneous lupus
erythematosus because maternal anti-Ro and anti-La antibod-
ies are transferred across the placenta. Although congenital
heart block occurs rarely, treating the mother with corticoster-
oids may be beneficial if fetal complete heart block is detected.
recommended.
• The offspring of mothers who have anti-Ro and anti-La antibodies
are at risk of congenital heart block.
Polymyositis and Dermatomyositis
Polymyositis is associated with myocarditis, pericarditis, and
conduction system disease (Figure 84.6). Myocarditis is rare
in the absence of systemic myositis but may respond to sys-
temically administered corticosteroids; hence, endomyocardial
biopsy is important if myocarditis is suspected in polymyositis.
All degrees of conduction system disease may be seen with poly-
myositis. Pericarditis, rarely leading to constrictive pericarditis,
may occur.
Scleroderma
Pericardial disease (pericarditis, pericardial effusions, tampon-
ade, or constrictive pericarditis) and pulmonary hypertension are
the classic cardiac complications associated with diffuse scle-
roderma, although conduction system disease and myocardial
fibrosis also occur. Patients with CREST syndrome (calcinosis
cutis, Raynaud phenomenon, esophageal dysfunction, sclerodac-
tyly, and telangiectasia) are at higher risk of cardiac complica-
tions, particularly pericarditis, than patients with uncomplicated
scleroderma (Figure 84.7).
Spondyloarthropathies
Two of the spondyloarthropathies, ankylosing spondylitis and
reactive arthritis (Reiter syndrome), are associated with aorti-
tis, aortic regurgitation, conduction system disease, and peri-
cardial disease. More rarely, mitral valve prolapse and mitral
 84 Systemic Disease and the Heart
Figure 84.4. Cardiac Sarcoidosis. Cardiac magnetic resonance images. Left panels, Normal. Right panels, Patchy delayed enhancement abnormality.
regurgitation occur. Histologically there is an infiltrate of lym-
phocytes and plasma cells in the aortic wall and around the vasa
vasorum, with resultant shortening and thickening of the aortic
valve leaflets and aortic root dilatation.
Relapsing Polychondritis
Relapsing polychondritis is an autoimmune disease that is associ-
ated with systemic vasculitis in up to 25% of patients. Vasculitis
may involve vessels of all sizes, including medium-sized vessels
such as the coronary arteries. Aneurysms of the aorta, including
the ascending aorta and the aortic root, that cause aortic regur-
gitation may result from inflammation of the cartilaginous sup-
porting tissues. The presence of systemic vasculitis is associated
with a poor prognosis.
Vasculitis
Churg-Strauss Syndrome
Classically, Churg-Strauss syndrome is an allergic granulom-
atous disease characterized by asthma, allergic rhinitis, and
eosinophilia. Vasculitis involves small and medium-sized ves-
sels; eosinophilic infiltration of the myocardium and pericar-
dium can occur, resulting in restrictive cardiomyopathy and heart
failure, myocardial infarction, arrhythmias, and pericarditis.
801
Cardiac involvement is a major cause of death in Churg-Strauss
syndrome. Cutaneous lesions may be present (Figure 84.8).
Polyarteritis Nodosa
Polyarteritis nodosa is a nongranulomatous disease involv-
ing medium-sized arteries, including the coronary arteries.
Myocarditis and pericarditis may also occur, and heart failure
can occur from either hypertension (due to renal involvement) or
vasculitis of the coronary arteries.
Behçet Disease
Behçet disease is associated with aneurysms of the arch vessels
and abdominal aorta, aortitis, and aortic incompetence.
Hematologic and Oncologic Diseases
Anemia, Thalassemia, and Sickle Cell Disease
Chronic anemia leads to a compensatory increase in cardiac out-
put owing to increased venous preload and decreased systemic
vascular resistance. Venous return and systemic catecholamines
increase, and, rarely, high-output cardiac failure occurs. Left
ventricular dilatation and hypertrophy are common, and flow
murmurs can also occur.
802 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

A B

C D

Figure 84.5. Cutaneous Manifestations of Systemic Lupus Erythematosus. A, Malar erythema. B, Subacute cutaneous lupus erythematosus.
C, Neonatal lupus. D, Livedo reticularis.

Thalassemia is frequently associated with systemic iron
overload and myocardial iron deposition due to extravascular
hemolysis and multiple blood transfusions leading to heart fail-
ure. Recurrent pericarditis and pericardial effusions resulting
in pericardial tamponade occur rarely. Heart block may occur
because of iron deposition in the atrioventricular node. Chelation
agents reduce the occurrence of transfusion-associated cardiac
dysfunction.
Sickle cell disease is associated with chronic anemia, heart
failure, myocardial infarction, and pulmonary infarction.
Sickling is aggravated by high oxygen extraction by the myo-
cardium. Papillary muscle infarction is a well-recognized com-
plication of sickle cell disease. Pulmonary infarction can result
from thrombosis in situ or from pulmonary emboli, and it can
predispose patients to recurrent pulmonary infections. Cardiac
complications due to iron overload are less common in sickle cell
disease than in thalassemia.
Hemochromatosis
Hemochromatosis is an autosomal recessive genetic disorder of
increased iron absorption that results in an increase in the serum
levels of iron and ferritin and in a decrease in total iron-binding
capacity. Deposition of iron in the myocardium leads to cardiac
manifestations, including atrial arrhythmias and ventricular
arrhythmias, heart block, dilated cardiomyopathy with restric-
tive physiology, and heart failure. Repeated phlebotomy is pro-
tective against the cardiac complications of hemochromatosis
and may partially reverse end organ dysfunction in patients who
have left ventricular dysfunction.
Carcinoid
Carcinoid heart disease is caused by carcinoid plaques due
to serotonin and bradykinin released from carcinoid tumors
in patients who have liver metastases. The result is valvular
disease. The valvular disease is usually right sided because
these substances are inactivated in the lung, but the left-sided
valves may be involved in patients who have intracardiac shunts
or pulmonary metastases. Valvular disease is characterized by
thickening and retraction of valve leaflets, resulting in limited
mobility, which leads to tricuspid regurgitation and pulmonary
stenosis (Figure 84.9). In some cases, tumors may occur in the
myocardium.
 84 Systemic Disease and the Heart
Figure 84.6. Heliotrope Periorbital Rash in Dermatomyositis.
Hypereosinophilic Syndrome
Hypereosinophilic syndrome results from overproduction of
eosinophils and resultant eosinophilic-mediated myocardial
injury. Cardiac involvement, although variable, is a major cause
of morbidity and death and is characterized by 3 distinct phases:
acute necrotic, intermediate, and fibrotic. The acute necrotic
phase is characterized by myocardial infiltration, which may be
clinically silent with normal echocardiographic findings. During
the intermediate phase, thrombus formation occurs at the ven-
tricular apices and can lead to progressive obliteration of the
cavity (Figure 84.10). Endomyocardial fibrosis occurs during
the fibrotic stage, resulting in diastolic dysfunction and restric-
tive cardiomyopathy. Valvular involvement can also occur, most
commonly involving the posterior mitral valve leaflet.
Renal Failure
Patients with chronic renal failure have long-standing hyper-
tension, which results in left ventricular hypertrophy and dias-
tolic dysfunction; they also have calcification and sclerosis of
the valves and mitral annulus. Echocardiographic findings may
mimic cardiac amyloidosis, with a sparkling appearance of the
myocardium on 2-dimensional imaging and concentric hypertro-
phy (Figure 84.11). However, in contrast to cardiac amyloidosis,
which is typically associated with an ECG of low-voltage ampli-
tude, the ECG in renal failure has features that are more consist-
ent with those of left ventricular hypertrophy.
Figure 84.7. Systemic Scleroderma With Ischemic Digital Ulceration.
803
Figure 84.8. Churg-Strauss Granulomas. Subcutaneous nodules
occurring on extensor surfaces of elbows.
Radiation-Induced Cardiac Disease
Radiation damage occurs after irradiation of the mediastinum
and can affect any cardiac structure, depending on the field of
irradiation. Initial presentation is often delayed and may not
occur until more than a decade after irradiation. Valvular disease
results from thickening and fibrosis of the valve leaflets, with left-
sided valves more frequently involved than right-sided valves.
Pericarditis may occur in any form, including acute and chronic
pericarditis with or without effusion, effusive-constrictive peri-
carditis, and constrictive pericarditis, which carries a poor prog-
nosis even after pericardiectomy.
Cardiac irradiation can damage coronary artery endothelium,
leading to nonatherosclerotic coronary artery disease. Restrictive
cardiomyopathy can result from irradiation of the myocardium.
Irradiation during childhood and the concomitant use of anthra-
cycline chemotherapeutic agents increase the risk of myocardial
damage.
Drug-Induced Cardiac Disease
Chemotherapy
The anthracyclines, including daunorubicin and doxorubicin, are
the major causes of chemotherapy-induced cardiomyopathy. The
probability of cardiomyopathy developing is dependent on the
cumulative dose of medication administered: the risk of cardio-
myopathy is less than 1% at a dose of less than 400 mg/m2, 7%
at 550 mg/m2, and more than 15% at 700 mg/m2. The risk of
cardiomyopathy increases with preexisting cardiac disease, con-
comitant use of cyclophosphamide, previous chest irradiation, or
age older than 70 years.
Before beginning anthracycline chemotherapy, all patients
should have a baseline evaluation of ventricular function.
Chemotherapy should not be initiated if the ejection fraction is
less than 30%. Chemotherapy may be administered to patients
who have high-risk malignancies and an ejection fraction of 30%
to 50%. If the ejection fraction is 50% or more, the risk of cardio-
myopathy is low. However, the ejection fraction should be deter-
mined again after dose levels of 300 mg/m2 and 400 mg/m2 and
after each subsequent dose. Chemotherapy should be discontinued
if the ejection fraction decreases 10% or more from baseline or to
less than 50% in patients who had a normal ejection fraction.
Other agents that can cause cardiotoxicity include cyclophos-
phamide, paclitaxel, fluorouracil, and monoclonal antibodies,
including trastuzumab (Herceptin). Fluorouracil is the second
804 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

RVOT
RV
LV

LA RA
PA

A B

Figure 84.9. Echocardiographic Images From Patient With Carcinoid Heart Disease. A, Apical 4-chamber view showing thickened retracted
leaflets of tricuspid valve. LA indicates left atrium; LV, left ventricle; RA, right atrium; RV, right ventricle. B, Parasternal view showing thickened,
fixed leaflets of pulmonary valve. PA indicates pulmonary artery; RVOT, right ventricular outflow tract.

most common cause of cardiotoxicity; the drug causes coronary Cocaine

vasospasm, resulting in angina, myocardial infarction, myocar-
ditis, or arrhythmias. Cyclophosphamide toxicity, resulting in
acute cardiomyopathy or hemorrhagic myopericarditis, is asso-
ciated with high-dose protocols and is unrelated to the cumula-
tive dose. Paclitaxel may result in heart block or cardiomyopathy
when combined with an anthracycline. Trastuzumab, used in the
treatment of HER-2 overexpressing breast cancer, can cause car-
diomyopathy with reduced ejection fraction.
Ergot Alkaloids, Anorectic Drugs, Pergolide
Use of ergot alkaloids for migraine headaches, anorectic drugs,
and pergolide can result in valvular disease with echocardio-
graphic features that resemble carcinoid or rheumatic valvu-
lopathy; any valve may be involved. The anterior mitral leaflet
is typically thickened and retracted with a relatively fixed and
immobile posterior leaflet because of tethering from shortened
and thickened chordae tendineae. On pathologic evaluation, the
valves are intact but covered with fibrous plaque. When patients
have used anorectic drugs, diagnosis requires echocardiographic
evidence of at least mild aortic regurgitation or moderate mitral
regurgitation.
Cocaine causes numerous adverse cardiovascular effects; many
are related to its sympathomimetic effect of blocking the reuptake
of catecholamines in sympathetic nerve terminals. Cocaine has
been associated with acute myocardial infarction, sudden cardiac
death, chest pain, myocarditis, and cardiomyopathy. Hypertension
is often a prominent feature because of systemic vasoconstriction.
Other effects include accelerated atherosclerosis, subarachnoid
hemorrhage, and aortic dissection. Supraventricular and ven-
tricular arrhythmias are common. β-Blockers should be avoided
because of the risk of further vasoconstriction. In acute toxicity,
benzodiazepines and nitrates may be useful, and phentolamine
can be used if refractory hypertension is present.
Specific Syndromes With Cardiac Involvement
Kartagener Syndrome
Kartagener syndrome is caused by an autosomal recessive
genetic disorder that affects ciliary motility. Clinically, it is char-
acterized by the triad of bronchiectasis, sinusitis, and situs inver-
sus, including dextrocardia. Abnormal sperm and cilia result in
sterility.

RV

LV LV
RV

A B

Figure 84.10. Echocardiographic Images From Patient With Hypereosinophilic Syndrome. A, Apical 4-chamber view showing thickening at the
ventricular apices. LV indicates left ventricle; RV, right ventricle. B, Parasternal short-axis view at the apical level showing cavity obliteration.
 84 Systemic Disease and the Heart 805

Marfan Syndrome
Marfan syndrome is an autosomal dominant disorder character-
ized by musculoskeletal, cardiovascular, and ocular abnormali-
ties. Cardiac involvement includes mitral valve prolapse leading
to mitral regurgitation, aortic root dilatation leading to aortic
regurgitation, aortic aneurysm, aortic dissection, and rupture.
This syndrome is caused by a genetic mutation in the fibrillin
gene. The risk of aortic dissection and rupture increases signif-
icantly during pregnancy, especially in women who have aortic
root dilatation that is moderate or greater before pregnancy. All
patients who have Marfan syndrome should receive β-blockers,
which decrease the rate of aortic root expansion and the risk of
aortic rupture. Aortic replacement should be considered in asymp-
tomatic patients with aortic root diameters of 5 cm or more.
Figure 84.11. Echocardiographic Image From Patient With End-
stage Renal Disease. Parasternal long-axis view showing concentric left
ventricular wall thickening. Ehlers-Danlos Syndrome
Ehlers-Danlos syndrome is an autosomal dominant disorder
associated with hyperextensile joints, spontaneous pneumotho-
A races, scoliosis, arthritis, and cardiovascular abnormalities that
include mitral and tricuspid valve prolapse, aortic root dilatation
and rupture, and dissection and rupture of other major arteries.
Marked variation in the risk of cardiovascular abnormalities is
found among the different forms of this syndrome, of which at
least 15 are known (Figure 84.12).

Figure 84.13. Pseudoxanthoma Elasticum. A, Angioid streak in ret-

Figure 84.12. Ehlers-Danlos Syndrome. A and B, Cutis hyperelastica. ina (arrows). B, Pseudoxanthoma elasticum.
806 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
Pseudoxanthoma Elasticum
Pseudoxanthoma elasticum is associated with yellow skin pap-
ules, angioid streaks in the retina, and cardiovascular disease
(Figure 84.13). Coronary arteries, peripheral arteries, cardiac
valves, and the cardiac conduction system may all be affected.
The risk of atherosclerosis is high even in the absence of tradi-
tional risk factors.
Osteogenesis Imperfecta
Osteogenesis imperfecta is characterized by blue sclera, increased
bony fragility, and hearing loss. It is associated with mitral valve
prolapse and aortic incompetence.
Noonan Syndrome
Noonan syndrome is an autosomal dominant disorder charac-
terized by impaired mental abilities, facial dysmorphism, and
cardiac involvement. Pulmonary valve stenosis is the most com-
mon defect. Other cardiac abnormalities can include pulmonary
artery hypoplasia, infundibular stenosis, hypertrophic cardiomy-
opathy, patent ductus arteriosus, atrial septal defect, or ventricu-
lar septal defect.
Williams Syndrome
Williams syndrome is characterized by mental impairment, elfin
facies, hypercalcemia, and supravalvular aortic stenosis. Other
cardiac abnormalities can include pulmonary artery stenosis,
mitral valve prolapse, pulmonary valve stenosis, and ventricular
septal defect.
Abbreviation
ECG electrocardiographic
 85
Cardiac Tumors
JOSEPH G. MURPHY, MD, and R. SCOTT WRIGHT, MD
Primary tumors of the heart are exceedingly rare, accounting
for less than 5% of all cardiac tumors; the remaining 95% are
metastatic tumors to the heart. The most common primary car-
diac tumors in adults are myxomas, usually occurring in the left
atrium (Figure 85.1), followed by lipomas and fibroelastomas
(Box 85.1). The most common malignant cardiac tumor in adults
is angiosarcoma. A malignant lymphoma may occasionally
develop in the adult heart. In children, the most common cardiac
tumor is rhabdomyoma and the most common malignant tumor
is rhabdomyosarcoma.
• Primary cardiac tumors are 5% of all cardiac tumors.
• Metastatic cardiac tumors are 95% of all cardiac tumors.
• The most common primary cardiac tumors in adults are myxomas.
• Among patients dying of cancer, 20% have myocardial or pericar-
dial metastases.
Presentation of Cardiac Tumors
• Asymptomatic—incidental discovery on echocardiography, com-
puted tomography, or magnetic resonance imaging of the chest.
• Pericardial effusion with or without cardiac tamponade (usually
due to pericardial metastases).
• Obstruction to myocardial filling or emptying.
• Obstruction to valve opening or closing.
• Atrial or ventricular arrhythmias including heart block.
• Thromboembolism to systemic or pulmonary circulation.
• Systemic symptoms (fever, weight loss).
Cardiac Myxomas
Cardiac myxoma may result in systemic symptoms, largely
due to embolic phenomena or to secretion of interleukin 6 by
the myxoma. Constitutional symptoms include fever and weight
loss. Embolic phenomena are due to tumor fragmentation and
thromboembolism from the tumor surface. These embolic epi-
sodes may mimic systemic vasculitis or infective endocarditis.
Right-sided cardiac tumors may result in recurrent pulmonary
emboli. The most common presenting symptoms of left-sided
cardiac myxomas are dyspnea on exertion, paroxysmal noctur-
nal dyspnea, and fever, but sudden death and hemoptysis also
may occur. The most common physical finding with a left atrial
myxoma is a mitral diastolic murmur (similar to mitral stenosis
but without the opening snap) (Table 85.1) or a mitral systolic
murmur due to mitral incompetence. Other features include an
added heart sound or tumor plop, atrial fibrillation, clubbing, and
Raynaud phenomenon. Left atrial tumors may mimic mitral ste-
nosis or incompetence, endocarditis, or vasculitis (Figure 85.2).
Right atrial tumors may mimic Ebstein anomaly, atrial septal
defect, or constrictive pericarditis (Figure 85.3). Left ventricular
tumors may mimic aortic stenosis or hypertrophic obstructive
cardiomyopathy, and right ventricular tumors may mimic pul-
monary stenosis, pulmonary hypertension, or pulmonary emboli.
The locations of cardiac myxomas are listed in Table 85.2.
• Cardiac myxoma results in systemic symptoms, largely due to its
secretion of interleukin 6.
• Right atrial tumors may mimic Ebstein anomaly, atrial septal
defect, or constrictive pericarditis.
Familial Cardiac Tumors
The familial cardiac myxoma syndromes constitute approx-
imately 10% of myxomas and have an autosomal dominant
transmission.
In Carney syndrome, myxomas arise in noncardiac locations
(usually breast or skin), pigmentation of the skin occurs (usually
807
808 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Table 85.1. Features Differentiating Left Atrial Myxoma

From Mitral Valve Disease
Mitral Valve
Feature Myxoma Disease
History Acute Chronic
Duration Present Absent
Associated constitutional
symptoms
Syncope Occasionally Rarely
Symptoms Occasionally episodic Progressive
Physical examination
Sound Tumor plop Opening snap
Murmur Varies with position Constant
Associated valve disease Unusual Common
Electrocardiogram Sinus rhythm Atrial fibrillation
Chest radiograph
Calcification Tumor Valve
Left atrium size Small Enlarged
Echocardiogram Characteristic findings Characteristic
findings
Previously published. See “Credit Line” section.

noncardiac tumors in approximately two-thirds of cases and with

Figure 85.1. Lateral Chest Radiograph Demonstrating Calcified Left endocrine tumors in one-third of cases.
Atrial Myxoma. (Previously published. See “Credit Line” section.) Cardiac rhabdomyomas are associated with the tuberous scle-
rosis syndrome, which is characterized by hematomas in multi-
ple organs, epilepsy, mental deficiency, and adenoma sebaceum.
lentigines or pigmented nevi), and there may be endocrine Cardiac fibromas are benign connective tissue tumors that occur
tumors (including pituitary adenomas), adrenocortical disease, predominantly in children. Imaging of cardiac tumors is usu-
or testicular tumors. Other syndromes include the NAME syn- ally with echocardiography, Imatron computed tomography,
drome (nevi, atrial myxoma, myxoid neurofibromas, and eph- or magnetic resonance imaging. Atrial myxomas may recur in
elides) and the LAMB syndrome (lentigines, atrial myxoma, and approximately 5% of patients. Malignant cardiac tumors have a
blue nevi). uniformly poor prognosis.
Familial cardiac myxoma syndromes differ from nonfamilial
(sporadic) myxomas in that they occur in younger patients, have • NAME syndrome— nevi, atrial myxoma, myxoid neurofibromas,
less of a female preponderance, are frequently multiple, recur and ephelides.
more frequently postoperatively, and occur more often in a ven- • LAMB syndrome—lentigines, atrial myxoma, and blue nevi.
tricular site. Familial myxomas are associated with freckling and
Lipomatous Hypertrophy of the Atrial Septum
Lipomatous hypertrophy of the atrial septum, although not a true
Box 85.1. Common Types of Primary Tumors of the tumor, is the accumulation of nonencapsulated adipose tissue
Heart
Benign (75%)
Myxoma
Rhabdomyoma
Fibroma
Lipoma and lipomatous hypertrophy of the atrial
septum
Atrioventricular node tumor
Papillary fibroelastoma
Hemangioma
Malignant (25%)
Angiosarcoma
Rhabdomyosarcoma
Fibrosarcoma

Previously published. See “Credit Line” section. Figure 85.2. Left Atrial Myxoma. (Previously published. See “Credit
Line” section.)
 85 Cardiac Tumors 809

Figure 85.3. Multicentric Right Atrial Myxomas. Transesophageal

echocardiography in the modified basal transverse plane shows 2 of 3
right atrial myxomas (T). The larger has a well-defined slender stalk
(arrow) attaching the tumor to the lateral wall of the right atrium (RA).
The smaller myxoma has a broad-based attachment to the posterior
atrial septum. AV indicates aortic valve; LA, left atrium. Figure 85.4. Lipomatous Hypertrophy of the Atrial Septum.

(both fetal and adult type) within the atrial septum (Figures 85.4 is a proven etiologic agent. Treatment is largely palliative and
and 85.5). This can lead to massive atrial septal hypertrophy that prognosis for cure is very poor.
may protrude into the right atrium. It is more common in elderly,
obese women. Lipomatous hypertrophy of the atrial septum has
been variably associated with supraventricular arrhythmias. The Papillary Fibroelastoma
diagnosis is usually made with echocardiography. Papillary fibroelastomas are benign tumors that arise from the
cardiac valves (Figure 85.8). They may cause valvular incompe-
Cardiac Sarcomas tence, coronary obstruction if located on the arterial side of the
aortic valve, and thromboembolic complications. Surgical exci-
Cardiac sarcomas are the most common primary malignant sion is curative (Figure 85.9).
tumor of the heart and include several histologic variations,
including angiosarcomas derived from malignant vascular form-
ing cells and rhabdomyosarcomas usually derived from the ven-
tricular walls. The prognosis is poor, with median survival of
less than 1 year. Rarely, surgical treatment has been successful.
Cardiac transplant is of unproven benefit when combined with
complete heart removal for cardiac tumor.

Mesothelioma of the Atrioventricular Node

Mesothelioma of the atrioventricular node is a cystic tumor usu-
ally diagnosed at autopsy (Figures 85.6 and 85.7). It is a rare
cause of sudden death due to complete heart block, ventricular
fibrillation, or cardiac tamponade.
Malignant mesothelioma of the pericardium usually manifests
with features of cardiac tamponade if acute or with features of
pericardial constriction if chronic. Its relation to asbestos expo-
sure is unproven, unlike pleural mesothelioma, in which asbestos

Table 85.2. Locations of Cardiac Myxomas

Figure 85.5. Lipomatous Hypertrophy of the Atrial Septum.
Location Cases, % Transesophageal echocardiography in the transverse plane, 4-chamber
Left atrium 75 view, shows extensive fatty tissue accumulation within the atrial septum
Right atrium 15 (arrowheads) that spares the fossa ovalis membrane (arrow), imparting
Right ventricle 5 the typical dumbbell appearance of lipomatous atrial septal hypertro-
Left ventricle 5 phy. LA indicates left atrium; LV, left ventricle; RA, right atrium; RV,
right ventricle. (Previously published. See “Credit Line” section.)
810 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Figure 85.6. Mesothelioma of the Atrioventricular Node Associated

With Sudden Death.

Figure 85.9. Papillary Fibroelastoma of the Mitral Papillary Muscle.

Metastatic Cardiac Tumors

Metastatic tumors of the heart, predominantly from carcinoma
of the lung or breast, malignant melanoma (Figure 85.10), and
the leukemias and lymphomas, constitute the majority of cardiac
tumors, outnumbering primary cardiac tumors approximately 20
to 1. Carcinoma of the bronchus or breast spreads to the heart pri-
marily through the lymphatics but also by direct extension and,
more rarely, through the pulmonary veins (bronchogenic carci-
noma). Carcinoma of the testis and kidney may spread through
the venous system and lead to intracardiac metastasis.
Figure 85.7. Mesothelioma With Pericardial Constriction. Pericarditis from pericardial metastasis is the most com-
mon symptom of metastatic heart disease. Less commonly, the

Figure 85.8. Papillary Fibroelastoma of the Tricuspid Valve (TV).

Transesophageal echocardiography in the transverse plane shows a
broad-based mass (arrow) attached to the atrial surface of the septal leaf-
let of the TV. The “shimmering” mobility of its fronds on real-time exam-
ination is typical of papillary fibroelastoma. LA indicates left atrium; Figure 85.10. Metastatic Melanoma Affecting the Epicardial
RA, right atrium. (Previously published. See “Credit Line” section.) Surface of the Heart.
 85 Cardiac Tumors
Table 85.3. Tumors That Cause Neoplastic Pericarditis
Tumor Cases, %
Lung carcinoma 40
Breast carcinoma 20
Hodgkin disease, leukemia, lymphomas 15
Other carcinoma 10
Melanoma 5 A hemorrhagic pericardial effusion may be due to extramedul-
Sarcoma 5 lary intrapericardial hematopoiesis in chronic myeloid leukemia
Others 5 and myelomonocytic leukemic blast crisis. Small, nonprogres-
patient may have an asymptomatic pericardial effusion detected
on chest radiography or echocardiography. If the pericardial
fluid collects rapidly, the patient may present with pericardial
tamponade. Pericardiocentesis under echocardiographic guid-
ance allows diagnosis of pericardial metastases in 70% to 80%
of cases.
Patients with myocardial metastasis may be clinically asymp-
tomatic or they may present with nonspecific ST-T wave changes,
cardiac arrhythmias, heart block, or myocardial dysfunction.
Echocardiography is the most commonly used imaging method
in suspected cardiac metastatic disease, but magnetic resonance
imaging and computed Imatron tomography imaging are also
valuable.
Myocardial metastasis is frequent in patients dying of wide-
spread carcinomatosis, but it is rarely diagnosed before death.
Endomyocardial and valvular metastases may mimic primary
cardiac tumors, but they are rare.
Neoplastic Pericarditis
Approximately 10% of patients dying of malignancy have per-
icardial involvement, and 5% have myocardial metastases.
Figure 85.11. Carcinoid Heart Disease Causing a Combination of
Pulmonary Valve Stenosis and Incompetence.
811
Table 85.3 lists tumors that can cause neoplastic pericarditis.
Nephroblastoma (Wilms tumor) and neuroblastoma are addi-
tional causes of neoplastic pericarditis in children.
Primary pericardial tumors are very rare. They include meso-
thelioma (possibly associated with asbestos exposure), pheo-
chromocytoma, and sarcomas (fibrosarcoma, liposarcoma, and
angiosarcoma).
sive, asymptomatic pericardial effusions may occur in 50% of
patients with breast cancer, probably as a result of lymphatic
obstruction.
Treatment of pericardial metastasis is usually palliative, but
radiotherapy and chemotherapy are valuable. Malignant peri-
cardial effusion can be treated in the short term through an
indwelling drainage catheter and in the long term through a peri-
cardial window. Surgical pericardiectomy may be required if the
pericardial window becomes obstructed.
Acquired Immunodeficiency Syndrome
In patients with acquired immunodeficiency syndrome (AIDS),
cardiac tumors may be due to non-Hodgkin lymphoma or meta-
static Kaposi sarcoma.
• Cardiac tumors associated with AIDS include Kaposi sarcoma and
primary lymphoma of the heart.
Carcinoid Heart Disease
Carcinoid heart disease (Figures 85.11 and 85.12) is character-
ized by plaque-like deposits of fibrous tissue on the endocardial
surface of valve cusps and leaflets, usually on the right heart
valves (90% of cases) but less commonly on the left heart valves
(10% of cases) if there is a patent foramen ovale or the carci-
noid tumor is a bronchial carcinoid. Carcinoid tumor cells are not
found in the heart, but remote secretion of seratonin is considered
the mechanism of carcinoid heart disease.
Figure 85.12. Carcinoid Heart Disease Causing a Combination of
Tricuspid Valve Stenosis and Incompetence

Sleep Apnea and Cardiac Diseasea
TOMAS KARA, MD, PHD, TOMAS KONECNY, MD,
and VIREND SOMERS, MD, PHD
Sleep is often a salutary and restorative time occupying one-third
of our lives, but in patients with sleep apnea (also called sleep-
disordered breathing), sleep can be a period of hemodynamic,
respiratory, and electrochemical turmoil that contributes to the
initiation, progression, and worsening of outcomes of various
forms of cardiovascular disease. This topic is gaining particu-
lar relevance with the strikingly high and growing prevalence of
sleep apnea in the general population and in cardiac patients.
Normal Sleep
Normal sleep can be divided into REM sleep and NREM sleep.
NREM sleep is further divided into stages I, II, III, and IV
(in the current guidelines of the American Academy of Sleep
Medicine, stage I is called N1, stage II is called N2, and stages
III and IV together are called N3). Each stage, from I through
IV, represents progressively deeper sleep, with gradual decreases
in sympathetic activity and consequent decreases in heart rate,
stroke volume, cardiac output, systemic vascular resistance,
and blood pressure. Simultaneously, ventilation and metabolic
rate also decrease. During stage IV (N3), sympathetic activity,
heart rate, blood pressure, and ventilation are usually at their
nadir. Conversely, parasympathetic or vagal activity increases in
NREM sleep, particularly in stage IV (N3).
REM sleep, by contrast, is a state of neural activation charac-
terized by sporadic REMs and the occurrence of dreams. During
REM sleep, sympathetic activation is marked, with intermittent
a
Portions previously published in Wolk R, Kara T, Somers VK. Sleep-
disordered breathing and cardiovascular disease. Circulation. 2003
Jul 8;108(1):9–12. Used with permission.
Abbreviations and acronyms are expanded at the end of this chapter.
812
86
surges in blood pressure and heart rate. Surprisingly, the level of
sympathetic activity is about 2-fold the levels recorded during
quiet wakefulness, and blood pressure and heart rate are simi-
lar to those during wakefulness. Occasional bursts of parasym-
pathetic activity also occur during REM and can manifest as
bradyarrhythmias.
Healthy adults cycle through NREM and REM sleep in
approximately 90-minute intervals and overall spend more than
75% of their sleep time in NREM sleep. Sleep apnea can seri-
ously disrupt the structured autonomic and hemodynamic pro-
files of both NREM and REM sleep.
Definition of Sleep Apnea
In sleep apnea, sleep is characterized by periods of attenuated or
absent ventilation, which result in decreased oxygen saturation
and repetitive arousals. The 2 types of sleep apnea are central
and obstructive. CSA is caused by diminution or cessation of
thoracoabdominal respiratory movement due to decreased cen-
tral respiratory drive. CSA occurs primarily in people with CHF,
although it occasionally occurs in healthy people, in people at
high altitude, and in people with central nervous system lesions.
By contrast, OSA is caused by upper airway collapse during
inspiration and is accompanied by strenuous breathing efforts
(ie, increased respiratory drive) (Figure 86.1). OSA with more
than 5 episodes of apnea or hypopnea per hour of sleep is rela-
tively common, affecting 24% of middle-aged men and 9% of
middle-aged women.
• CSA is caused by diminution or cessation of thoracoabdominal
respiratory movement due to decreased central respiratory drive.
• OSA is caused by upper airway collapse during inspiration and is
accompanied by strenuous breathing efforts.
 86 Sleep Apnea and Cardiac Disease 813

Air flow

Tongue

Nasal
passage

Soft palate
Uvula
Normal

Hypopnea Apnea
Figure 86.1. Obstructive Sleep Apnea. Upper, Normal ventilation during sleep. Lower left, Partial airway obstruction resulting in hypopnea.
Lower right, Complete airway obstruction leading to apnea. (Previously published See “Credit Lines” section.)

Obstructive Sleep Apnea
Acute Responses
Repetitive apneic episodes can induce nocturnal oxygen desat-
uration to levels as low as 40% to 50%, at which point stand-
ard oximetry measures become inaccurate. The hemodynamic
responses to severe hypoxemia in patients with OSA resemble
the diving reflex, an intriguing adaptive mechanism that allows
oxygen conservation during times of limited oxygen supply (the
diving reflex is particularly well developed in sea mammals
and elite human divers). In the diving reflex, hypoxemic stress
elicits increased peripheral sympathetic activity (with diffuse
vasoconstriction except in the brain and heart) and simultane-
ously increases vagal cardiac activation (Figure 86.2). This com-
bination of responses may help preserve oxygen by limiting its
consumption (both by increasing peripheral vascular resistance
and by bradycardia-related lower cardiac output). Thus, it is not
surprising that in about 10% of sleep apnea patients, nocturnal
apneic episodes may be associated with severe bradyarrhythmias,
including atrioventricular block and occasional sinus arrest.
When an episode of OSA is released, compensatory hyper-
ventilation (ie, hyperpnea) ensues, during which the R-R interval
shortens, cardiac output increases, and systemic blood pres-
sure increases to surprisingly high values (240/130 mm Hg or
more) (Figure 86.2). Severe sleep apnea may also acutely trig-
ger the release of several vasoactive substances, including cat-
echolamines, atrial natriuretic peptide, and endothelin.
Also implicated in the acute effects of OSA are the hemody-
namic and cardiac structural consequences of inspiration against
a closed airway, known as the Müller maneuver. The negative
intrathoracic pressure generated during OSA, which can reach
−60 mm Hg or more, may elicit increases in ventricular after-
load, rapid changes in left atrial size, and elevations in left atrial
and even left ventricular transmural pressure with subsequently
increased cardiac wall stress. These hemodynamic effects may
be particularly important in OSA patients who have coexisting
heart disease.
• Diving reflex: hypoxemic stress elicits increased peripheral sym-
pathetic activity (with diffuse vasoconstriction except in the brain
and heart) and simultaneously increases vagal cardiac activation.
• Müller maneuver: inspiration against a closed airway, creating
negative intrathoracic pressure as in OSA.
Sustained Effects of OSA Persisting
Into Daytime Wakefulness
There is considerable evidence that the acute pathophysiologic
mechanisms activated by repeated apneic episodes at night may
carry over into daytime disease processes. Even during nor-
moxic wakefulness, patients with OSA have higher sympathetic
activity, faster heart rates, diminished heart rate variability, and
increased blood pressure variability. High sympathetic drive may
in part be due to tonic chemoreflex activation, since chemoreflex
deactivation by 100% oxygen lowers sympathetic drive, slows
heart rate, and lowers blood pressure in OSA patients. Left
atrial size may be chronically increased in patients with OSA
(Figure 86.3).
814 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

CPAP Therapy During

Awake REM Sleep
Sympathetic
Nerve Activity

Respiration
150 150
Blood 100 100
Pressure, 50 50
mm Hg 0 0
10 s

OSA During REM Sleep

Sympathetic
Nerve Activity

Respiration OSA OSA

250
200
Blood 150
Pressure, 100
mm Hg 50
0

Figure 86.2. Neural and Circulatory Changes in Obstructive Sleep Apnea. Recordings of sympathetic nerve activity, respiration, and intra-
arterial blood pressure in the same patient when awake, with obstructive sleep apnea (OSA) during rapid eye movement (REM) sleep, and with elim-
ination of OSA episodes by continuous positive airway pressure (CPAP) therapy during REM sleep. Sympathetic nerve activity is very high during
wakefulness but increases even more in OSA during REM sleep. Blood pressure increases from 130/65 mm Hg when the patient is awake to 256/110
mm Hg at the end of the apneic episode. Elimination of apneic episodes with CPAP therapy results in decreased sympathetic activity and prevents
blood pressure surges during REM sleep. (Previously published. See “Credit Lines” section.)

Hypoxemia Intrathoracic pressure

Reoxygenation Arousals

Hypercapnia Obstructive Sleep deprivation

sleep apnea

Sympathetic activation Endothelial dysfunction

Metabolic dysregulation
Disease Systemic inflammation
mechanisms
Left atrial enlargement Hypercoagulability

Systemic
Hypertension Associated Renal disease
Pulmonary cardiovascular
Heart failure disease Stroke
Arrhythmias Myocardial infarction

Sudden cardiac death

Figure 86.3. Suspected Pathophysiologic Links Between Obstructive Sleep Apnea and Cardiovascular Morbidity. (Previously published See
“Credit Lines” section.)
 86 Sleep Apnea and Cardiac Disease 815

OSA and Cardiovascular Disease Conditions
Hypertension
Although many reports have shown an association between both
pulmonary and systemic hypertension and OSA, the studies from
the Wisconsin Sleep Cohort first provided the prospective evi-
dence of a dose-response association between the severity of
OSA and the likelihood of systemic hypertension developing in
subjects who were normotensive at baseline. This association
was independent of other known risk factors, such as baseline
blood pressure, body mass index and habitus, age, sex, alcohol
use, and cigarette use. No single causal factor responsible for
the occurrence of incident hypertension in OSA has been iden-
tified. The mechanisms are probably multifactorial and include
activation of the sympathetic nervous system, endothelial dys-
function (including that caused by systemic inflammation),
and increased endothelin levels, all of which would potentiate
vasoconstriction.
Treatment of OSA may induce decreases in blood pressure not
only at night but also during the day. Blood pressure reduction
with treatment of OSA may be especially evident in patients who
have more severe apnea and hypertension. Sleep apnea should
be considered especially in patients with resistant hypertension,
particularly if they are obese. Sleep apnea should also be sus-
pected in patients who are “nondippers” (ie, those in whom blood
pressure does not decrease appropriately during the night).
The most recent Joint National Committee on Prevention,
Detection, Evaluation, and Treatment of High Blood Pressure
guidelines list sleep apnea as an important identifiable cause of
hypertension.
Atherosclerosis and Coronary Artery Disease
Epidemiologic evidence suggests that OSA may be etiologically
linked to the development of atherosclerosis. There is a high
prevalence of OSA among patients with coronary artery disease,
and several case-control or prospective studies implicate OSA as
an independent predictor of coronary artery disease. The exact
mechanisms of any atherogenic effects of OSA have not been
established. However, reports linking OSA to inflammation lend
support to the concept that systemic or local inflammation may
have a direct role in atherogenesis in persons with OSA. The
role of oxidative stress in the vascular pathophysiology of OSA
is controversial. OSA-induced hypertension may contribute to
endothelial damage and thereby to atherosclerosis.
The presence of OSA has been linked to an increased likeli-
hood of coronary artery calcification, and prospective observa-
tional studies confirm that OSA is independently associated with
the incidence of new coronary artery disease; this effect seems
to be present more in men than women (Figure 86.4). In persons
with existing coronary artery disease, OSA may trigger acute
nocturnal cardiac ischemia and even possibly MI. Nocturnal

A 1.00 B 1.00
Disease-Free Survival
Disease-Free Survival

Probability of
Probability of

0.90 0.90

0.80 0.80

AHI categories <5 5-15 AHI categories <5 5-15

15-30 >30 15-30 >30
0.70 0.70
0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
Follow-up, y Follow-up, y
C 1.00 D 1.00
Disease-Free Survival

Disease-Free Survival
Probability of

Probability of

0.90 0.90

0.80 0.80

AHI categories <5 5-15 AHI categories <5 5-15

15-30 >30 15-30 >30
0.70 0.70
0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
Follow-up, y Follow-up, y
Figure 86.4. Obstructive Sleep Apnea and Cardiovascular Outcomes. Obstructive sleep apnea was independently associated with worse cardio-
vascular outcomes in men but not in women in a prospective study of 1,927 men and 2,495 women. Coronary artery disease–free survival is shown
according to apnea-hypopnea index (AHI) category for men (A) and women (B). Congestive heart failure–free survival is shown according to AHI
category for men (C) and women (D). The AHI is a measure of sleep apnea severity. (Previously published See “Credit Lines” section.)
816 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
exacerbation of ischemic heart disease by OSA may result from
oxygen desaturation, high sympathetic activity, increased cardiac
oxygen demand (due to tachycardia and increased systemic vas-
cular resistance), and a prothrombotic state. These mechanisms
could theoretically lead to coronary plaque rupture and onset of
an acute coronary event. In patients with OSA, compared with
those without OSA, a higher proportion of MI events tend to
occur during the night. Indeed, if an MI wakes a person from
sleep, there is a high likelihood that the person has OSA. From
the clinical perspective, OSA should be considered in the differ-
ential diagnosis for patients with evidence of cardiac ischemic
events triggered at nighttime. Furthermore, OSA may indicate a
poor prognosis in patients with coronary artery disease; 5-year
mortality among those with untreated OSA has been reported to
be 38%, compared with 9% among patients without OSA.
Observational studies suggest that among patients with sta-
ble coronary artery disease, treatment of OSA by CPAP may
decrease the incidence of new cardiovascular events. Given the
relatively benign profile of CPAP treatment and the encouraging
results from observational studies of CPAP treatment in cardiac
patients, diagnosis and appropriate treatment of OSA should be
routinely considered for patients with cardiovascular diseases,
especially MI, since these patients have a strikingly high prev-
alence of undiagnosed OSA (Figure 86.5). However, definitive
standard-of-care guidelines cannot be written until randomized
controlled studies evaluate the outcomes among these patients.
Arrhythmias and Sudden Cardiac Death
For an unselected population of atrial fibrillation patients pre-
senting for cardioversion, the risk of OSA (assessed with the
Berlin Questionnaire) is approximately 50%; for a general car-
diology clinic population, the risk is 30%. Patients with OSA
who undergo coronary artery bypass grafting have an increased
Potential underdiagnosis
70
68
60 66
57
56
50
Patients, %
40
30
20
10 12
14
0 stroke patients should be pursued, and, if indicated, appropriate
A B C R1 R2 R3
Figure 86.5. Potential Underdiagnosis of Obstructive Sleep Apnea.
Discordance exists between the high prevalence rates and low degree
of documented diagnosis or inhospital investigation of obstructive
sleep apnea (OSA) among patients hospitalized for acute myocardial
infarction (MI). Bar A, Documented diagnosis or investigation of OSA
in patients with MI. Bar B, Documented diagnosis or investigation of
OSA in consecutive MI patients who were also recruited for overnight
polysomnography. Bar C, Actual prevalence of OSA in the same con-
secutive MI patients recruited for polysomnography. Bars R1, R2, and
R3, For comparison, the prevalence of OSA is shown from 3 studies
of patients with acute coronary syndromes. (Previously published See
“Credit Lines” section.)
likelihood of postoperative atrial fibrillation. Furthermore,
among atrial fibrillation patients undergoing cardioversion,
observational data suggest that the presence of untreated OSA
was associated with a 2-fold greater risk of 1-year recurrence
of atrial fibrillation (82%); the recurrence risk for OSA patients
receiving appropriate therapy was 42%.
Several pathophysiologic mechanisms may contribute to the
association between OSA and atrial fibrillation. OSA-related
hypoxia, atrial stretch, sympathetic activation, acute blood pres-
sure surges, and changes in atrial refractoriness may all reason-
ably be expected to predispose patients to an increased risk of
atrial fibrillation. Evidence from recent studies of dogs showed
that atrial fibrillation can be induced by simulated episodes of
OSA and that ablation of the right pulmonary artery ganglion-
ated plexi can ameliorate the effect of OSA on inducing atrial
fibrillation.
Bradyarrhythmias frequently associated with OSA are sinus
arrest, sinoatrial block, and atrioventricular block, all of which
may lead to ventricular asystole. The mechanism of these bra-
dyarrhythmias is usually mediated by the diving reflex (see
above).
Arrhythmias or cardiac ischemia may contribute to the
increased likelihood of sudden cardiac death during the night-
time among patients with OSA. In contrast to sudden cardiac
death among patients without OSA, which is more likely to occur
in the morning (between 6 AM and noon), about 50% of sudden
cardiac deaths among patients with OSA occur at night (between
10 PM and 6 AM).
Stroke
The association between OSA and atrial fibrillation may in
part explain the increased stroke risk for persons with OSA.
Atherosclerosis, vascular inflammation, hypertension, and pro-
coagulant effects of OSA are other important risk factors for
stroke. Furthermore, the decreased cardiac output and increased
intracranial pressure during acute OSA episodes, with a conse-
quent decrease in cerebral blood flow, may have an important
role. However, the nature of the association between stroke and
OSA is unclear. It is known that patients with stroke have a high
prevalence of OSA (>40%). What is uncertain is whether OSA
is a causal factor in stroke or whether OSA is a consequence of
stroke and results from stroke-induced impairment of respira-
tory and muscle tone control. There is controversy about whether
patients who have transient ischemic attacks and OSA have a
greater risk of stroke than those without OSA. Nevertheless, pro-
spective observational data showed that severe OSA is an impor-
tant predictor of risk of new stroke. Among stroke survivors, OSA
could further compromise physical and cognitive functions and
worsen the prognosis. Therefore, the diagnosis of sleep apnea in
therapy should be instituted.
Congestive Heart Failure
The acute hemodynamic and cardiac structural effects of OSA,
together with adrenergic activation, systemic inflammation,
increased endothelin, endothelial dysfunction, hypertension, and
ischemic heart disease, are some of the mechanisms that may
underlie the association between OSA and CHF and the associa-
tion between OSA and acute exacerbations of CHF. By virtue of
soft tissue edema, especially of the airway, CHF may predispose
to OSA. Although it is not clear whether the prevalence of OSA
 86 Sleep Apnea and Cardiac Disease
among CHF patients is greater than the prevalence of OSA in the
general population, OSA should be suspected in CHF patients
who are obese. Treatment of OSA may significantly improve
clinical status. Among obese men and women in the Framingham
study, the very high risk of CHF over 20-year follow-up could
partly be explained by occult sleep apnea.
Hypertrophic Cardiomyopathy
OSA is highly prevalent among patients with hypertrophic car-
diomyopathy, and recent evidence supports the notion that OSA
may contribute to morbidity in these patients. Hypertrophic car-
diomyopathy patients with comorbid OSA have a higher likeli-
hood of atrial fibrillation, increased left atrial size, and decreased
exercise capacity. A randomized trial is needed to reliably show
that treatment with CPAP improves morbidity or mortality
among patients with hypertrophic cardiomyopathy. Several case
reports have suggested that this improvement may be possible.
• There is a dose-response association between the severity of OSA
and the likelihood of systemic hypertension developing. Sleep
apnea should be considered especially in patients with resistant
hypertension, particularly if they are obese.
• OSA may be etiologically linked to the development of athero-
sclerosis.
• In patients with OSA, compared with those without OSA, a higher
proportion of MI events tend to occur during the night.
• Patients with stroke have a high prevalence of OSA.
• Hypertrophic cardiomyopathy patients with comorbid OSA have a
higher likelihood of atrial fibrillation, increased left atrial size, and
decreased exercise capacity.
Diagnosis and Treatment of OSA
OSA should always be considered in patients who have any of
the following:
1. Risk factors for OSA (especially obesity, older age, and male sex)
2. Symptoms suggestive of a sleep disorder, including daytime sleep-
iness, fatigue, snoring, and witnessed cessation of breathing during
sleep
3. Specific clinical features, such as resistant hypertension (especially
the “nondipping” pattern), resistant CHF with frequent nocturnal
exacerbations, cardiac ischemia during sleep, recurrent atrial fibril-
lation, and stroke
The risk of OSA can be assessed by the use of overnight oxim-
etry (severe and repetitive oxygen desaturations being sugges-
tive of OSA) and by the use of questionnaires (eg, the Berlin
Questionnaire) that contain scoring systems indicating the prob-
ability of OSA. However, the definitive diagnosis can be made
only by polysomnography, which is usually performed in a
specialized sleep center.
The severity of OSA is often attenuated by behavioral and
lifestyle modifications, such as weight loss, avoidance of seda-
tives and alcohol, and avoidance of sleeping on the back. In
selected patients, surgical procedures designed to increase the
diameter of the upper airway (such as uvulopalatopharyngo-
plasty and laser-assisted uvuloplasty) can be used, but beneficial
effects on OSA may be transient and unpredictable. Other treat-
ment options include dental appliances that minimize posterior
displacement of the mandible during sleep; these appliances may
be helpful in less severe OSA.
The treatment of choice is CPAP, which should be used in
patients with moderate to severe OSA and perhaps in those
817
with mild OSA. CPAP therapy is associated with a decrease in
OSA severity and OSA-related symptoms (leading to a mark-
edly improved quality of life), and it can also favorably affect
the risk and the course of OSA-related cardiovascular disease.
Specifically, among OSA patients with hypertension, effective
CPAP treatment significantly reduces daytime and nocturnal
blood pressure. Nocturnal ST-segment depression and noctur-
nal angina might be significantly reduced after CPAP therapy as
well. Observational data suggest that CPAP treatment of patients
with severe OSA is accompanied by a reduction in acute cardio-
vascular events. CPAP may also reduce the recurrence of atrial
fibrillation in patients undergoing cardioversion. Treatment of
OSA in CHF has been associated with improvements in ejection
fraction, lower blood pressure, and slower heart rate. However,
final answers await the completion of randomized controlled
studies to determine whether treatment of OSA results in fewer
hospitalizations, fewer cardiovascular events, or mortality ben-
efit in CHF or in any other OSA-related cardiovascular condi-
tions, or whether treatment prevents incident or recurrent atrial
fibrillation.
An important, unresolved question is how to treat OSA
patients who do not adhere to CPAP therapy or in whom CPAP
does not effectively alleviate OSA severity. In patients with life-
threatening OSA who cannot tolerate CPAP, tracheostomy should
be considered. Better understanding of the biology of OSA may
help in developing new pharmacologic and nonpharmacologic
therapies that act on specific pathophysiologic pathways in the
development of OSA or OSA-associated cardiovascular disease.
• Treatment of choice for OSA is CPAP.
• Among OSA patients with hypertension, effective CPAP treatment
significantly reduces daytime and nocturnal blood pressure.
• CPAP may also reduce the recurrence of atrial fibrillation.
Central Sleep Apnea
Acute and Chronic Effects of CSA
Many of the acute and sustained effects of CSA are similar to
those of OSA as described earlier. These effects are related
primarily to chemoreceptor activation by hypoxia and lead to
elevated sympathetic drive and an increase in circulating cate-
cholamines. Similarly, frequent apneic episodes causing arou-
sal and sleep fragmentation can result in sleep deprivation and
daytime fatigue and somnolence.
However, in contrast to OSA, CSA is not accompanied by
airway occlusion and strenuous respiratory efforts; thus, hemo-
dynamic effects related to changes in intrathoracic and cardiac
transmural pressures are modest, and hypoxemia is usually less
marked.
Association Between CSA and
Cardiovascular Disease
The main clinical association of CSA is with CHF, although it
also occurs in healthy persons with hypobaric hypoxia at high
altitude. With CHF, CSA occurs as Cheyne-Stokes respiration
(also called periodic breathing), which is characterized by repet-
itive episodes of apnea or hypopnea followed by hyperpnea (ie,
hyperventilation) and manifested as crescendo-decrescendo
changes in tidal volume (a pattern of waxing and waning).
Although the pathophysiology of CSA in CHF is not fully
understood, its etiology is probably multifactorial (Figure 86.6)
and includes instability in the respiratory rhythm control (such
818 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Heart failure
Supine posture
Sympathetic
Circulatory ↑ Leptin activation
delay

↑ Cardiac
filling pressure ↑ CO2 sensitivity

Upper airway Adrenergic modulation

Lung of chemoreflex
edema and congestion
narrowing

Altered ventilatory control

Central sleep apnea

Figure 86.6. Heart Failure and Central Sleep Apnea. Complex pathophysiologic links have been postulated for interactions between heart failure
and central sleep apnea. CO2 indicates carbon dioxide. (Previously published See “Credit Lines” section.)

as heightened chemoreceptor drive and fluctuating arterial car-
bon dioxide levels), upper airway narrowing, pulmonary venous
congestion (resulting in reflex afferent stimulation of pulmonary
mechanoreceptors), and prolonged circulation time, which leads
to a delay in sensing changes in arterial blood gases by chemore-
ceptors. It occurs predominantly during NREM sleep stages I
(N1) and II (N2), when ventilation is regulated by the levels of
arterial carbon dioxide.
The prevalence of CSA in CHF has been estimated to be 33%
to 70% among patients with stable systolic CHF, although it
depends on various factors, including the cause of heart failure,
sex (CSA is more common in men), age, ejection fraction, and
hemodynamic status. CSA frequently occurs not only in patients
with advanced CHF but also in those with asymptomatic left
ventricular dysfunction.
The clinical significance of CSA in CHF is 2-fold. First, CSA
is associated with the severity of CHF. The nature of this asso-
ciation is complex and not fully understood. On one hand, CHF
patients with CSA have a lower exercise capacity, a lower ejec-
tion fraction, increased left ventricular volumes, elevated pul-
monary capillary wedge pressure, and a higher prevalence of
cardiac arrhythmias, indicating that the presence of CSA may be
merely an index of CHF severity. On the other hand, CSA con-
ceivably leads to the progression of CHF through several mecha-
nisms, such as neuroendocrine effects (elevated catecholamines),
hypoxia, blood pressure and heart rate fluctuations, and car-
diac arrhythmias. In fact, even in patients with asymptomatic
left ventricular dysfunction, CSA is accompanied by impaired
cardiac autonomic control and increased cardiac arrhythmias,
suggesting that perhaps CSA precedes the development of overt
heart failure.
Second, irrespective of the mechanisms involved, CSA is an
important risk factor and its presence worsens the prognosis in
CHF. The effect is independent of other known risk factors, such
as left ventricular ejection fraction, hemodynamic parameters,
and peak oxygen consumption. For example, among clinically
stable patients with CHF, CSA was an independent predictor of
cardiac death and transplant during 2-year follow-up.
• CSA frequently occurs not only in patients with advanced CHF but
also in those with asymptomatic left ventricular dysfunction.
• In clinically stable patients with CHF, CSA has been shown to be
an independent predictor of cardiac death and transplant.
• In CSA, the hemodynamic effects related to changes in intratho-
racic and cardiac transmural pressures are modest, and hypoxemia
is usually less marked than in OSA.
Treatment
The primary goal of treating CHF patients with CSA is to opti-
mize pharmacologic heart failure therapy. Decreased CHF sever-
ity and hemodynamic improvement of CHF are often associated
with a decrease in CSA. More aggressive treatment of CSA may
be indicated in persistent CSA and refractory CHF. Theophylline,
nocturnal oxygen supplementation, and acetazolamide have each
been suggested to decrease the severity of CSA, but their long-
term effects on outcome are not known. Pilot studies suggest
that CPAP therapy may improve transplant-free survival, but the
 86 Sleep Apnea and Cardiac Disease 819

CANPAP trial showed no mortality benefit from CPAP treat- risk factors for sleep apnea. Treating sleep apnea may help in car-
ment of CSA in CHF patients. However, the efficacy of CPAP in diovascular disease management. The diagnosis and treatment
reducing CSA was limited, with suboptimal patient adherence to of sleep apnea is important even in the absence of any clinically
therapy, which may help explain the absence of benefit. An ongo- overt cardiovascular disease because sleep apnea may contribute
ing multinational study (SERVE-HF) is retesting the effect of to its development. Randomized trials examining the cardiovas-
airway pressure on outcome among CHF patients with CSA who cular effects of the treatment of sleep apnea are lacking, however,
are using an adaptive servoventilation technology, which more and whether treatment of sleep apnea truly prevents cardiovascu-
effectively corrects CSA. lar events and has a mortality benefit is unknown.
Another new, potentially promising therapeutic strategy is
cardiac resynchronization therapy. It may have beneficial hemo-
Abbreviations
dynamic effects in CHF and decrease the severity of CSA.
CHF congestive heart failure
CPAP continuous positive airway pressure
Conclusions CSA central sleep apnea
Both OSA and CSA are associated with cardiovascular disease. MI myocardial infarction
NREM non–rapid eye movement
Even though a causal relationship between sleep apnea and car-
OSA obstructive sleep apnea
diovascular disease has not been definitively proved, the coex- REM rapid eye movement
istence of sleep apnea with cardiovascular disease exacerbates
symptoms and may accelerate progression. The diagnosis of sleep
apnea should always be considered in cases of refractory CHF, Name of Clinical Trial
resistant hypertension, transient ischemic attacks, and nighttime CANPAP Canadian Continuous Positive Airway Pressure for
cardiac ischemia or arrhythmias, especially when patients have Patients with Central Sleep Apnea and Heart Failure
 87
Cardiovascular Trauma
JOSEPH G. MURPHY, MD, and R. SCOTT WRIGHT, MD
Cardiovascular trauma is a significant cause of death, particu-
larly among young men in our society. Cardiovascular trauma
is classified into penetrating injuries; blunt, nonpenetrating
trauma; and medical injuries to the heart sustained during an
invasive cardiovascular procedure, medical device implantation,
or cardiopulmonary resuscitation.
Penetrating cardiac injury is due primarily to knife or gunshot
injuries, whereas blunt cardiac injury is usually due to automo-
bile or motorcycle accidents or industrial incidents. Iatrogenic
cardiac trauma also may occur from cardiopulmonary resusci-
tation, endomyocardial biopsy, or the use of intravascular cath-
eters, including Swan-Ganz catheters (Box 87.1).
• The most common recipients of cardiac trauma are males between
the ages of 15 and 35 years.
• Nonpenetrating cardiac trauma usually results from automobile
or industrial injuries, whereas penetrating cardiac injuries usually
result from knife or gunshot wounds.
• Among patients with traumatic penetrating cardiac injury, 50% die
rapidly, usually before hospitalization.
Penetrating Cardiac Injury
Penetrating cardiac trauma most commonly affects the right ven-
tricle, followed by (in decreasing order of frequency) the left ven-
tricle, right atrium, and left atrium. Cardiac injury may occur
from direct injury or from indirect injury (through rib fractures
that puncture the cardiac chambers). The principal consequences
of perforating cardiac injury are cardiac tamponade and exsan-
guinating hemorrhage, both of which lead to death rapidly if
not treated on an emergency basis. The development of cardiac
tamponade depends on the chamber penetrated, the size of the
penetration, and the presence or absence of lacerated pericar-
dium. The left ventricle is usually capable of sealing a small hole
820
because of the thickness of the surrounding muscle, whereas a
perforation of the right atrium or right ventricle usually leads to
rapid hemopericardium. If the pericardium also is opened by the
initial injury, tamponade usually will be prevented and the bleed-
ing will manifest as a hemothorax. Occasionally, the pericardial
tear acts as a flap valve and prevents blood drainage into the
pleural space, leading to tamponade. The signs and treatment of
pericardial tamponade are discussed in Chapter 77 (“Pericardial
Disease”).
• Cardiac injury may occur from direct injury or from indirect injury
(through rib fractures that puncture the cardiac chambers).
Blunt Cardiac Injury
Nonpenetrating cardiac trauma may result directly from a force
on the chest wall or indirectly from pressure on the abdomen
displacing a large volume of blood suddenly into the heart
(Figure 87.1). Both forms of injury are frequent in automobile
accidents. Nonpenetrating cardiac trauma may result in myocar-
dial contusions; chamber or vessel lacerations; rupture of chor-
dae tendineae, papillary muscles, or cardiac valves; pericarditis;
pericardial lacerations; and, rarely, the late development of con-
strictive pericarditis.
Pericardial Injury
Pericardial injury may result from blunt or penetrating chest
injuries and may lead to rapid death due to cardiac tamponade or
slowly progressive pericardial inflammation and fibrosis leading
to late pericardial constriction. Delayed pericardial tamponade
and localized pericardial tamponade are variants of pericardial
injury that are often difficult to diagnose.
 87 Cardiovascular Trauma 821

Box 87.1. Traumatic Cardiac Lesions

Pericardial
Hemorrhagic pericarditis
Pericardial laceration
Tamponade
Purulent pericarditis due to associated rupture of
esophagus
Constrictive pericarditis
Intrapericardial diaphragmatic hernia
Myocardial
Contusion
Ischemic infarction due to traumatic occlusion of
coronary artery
Myocardial hematoma
Myocardial laceration
Figure 87.1. Cardiac Trauma and Fatal Hemopericardium Resulting
Myocardial rupture
From a Motor Vehicle Accident.
Aneurysm
Pseudoaneurysm
Medical Cardiac Injury
Diffuse calcification (“myocarditis ossificans”)
Cardiopulmonary resuscitation frequently results in nondisplaced
Endocardial—thrombus
rib fractures and may also lead to rupture of the left ventricle
Valvular or right ventricle if performed too vigorously, especially after
Atrioventricular valves (chordal rupture, papillary a recent myocardial infarction has softened the myocardium.
muscle rupture, torn leaflet) Other complications include rupture of the papillary muscles that
support the tricuspid valve, resulting in severe tricuspid regurgi-
Semilunar valves (avulsion of cusp, avulsion of
tation, and, more rarely, rupture of the aorta. Penetrating cardiac
commissure, torn cusp, intimal tear in adjacent
injuries caused by medical trauma occur during endomyocardial
aorta with cusp displacement, sinus of Valsalva
biopsy of the right ventricle or after perforation with a tempo-
aneurysm with cusp displacement)
rary pacemaker wire. Dissection of the aorta or coronary ostia
Coronary artery (laceration, arteriovenous fistula) may occur during coronary angiography. In rare cases, coronary
Aorta and pulmonary artery angioplasty has resulted in coronary artery rupture and tampon-
ade. Intra-aortic balloon counterpulsation also may cause aortic
Rupture
dissection, but it is more likely to cause thromboembolic com-
Aneurysm plications as a result of pulsation against atheromatous plaques
in the aorta. Indwelling venous catheters may migrate and perfo-
Previously published. See “Credit Line” section. rate the pulmonary arteries, and improper use of balloon-tipped
pulmonary artery catheters may lead to branch pulmonary artery
rupture and intrapulmonary hemorrhage (Figure 87.2).

Myocardial Contusion
Myocardial contusion is primarily a pathologic diagnosis, and
there are no definitive clinical methods to establish it. Although
late complications from blunt cardiac trauma may occur in a
small number of patients, in the majority of patients with car-
diac trauma who present with stable vital signs, a short period
of electrocardiographic monitoring (about 24 hours) is usually
sufficient to determine whether arrhythmia or heart failure will
develop. Sudden cardiac death due to ventricular fibrillation may
occur with low-energy impact to the chest wall (such as from a
baseball) if the blow coincides with a narrow window of time
during cardiac repolarization. Damage to the coronary artery
may result in occlusion, laceration, or, more rarely, fistula forma-
tion. Rupture of the intraventricular septum, myocardial aneu-
rysm, and pseudoaneurysm formation also have been reported.
• Damage to the coronary artery may result in occlusion, laceration,
or, more rarely, fistula formation. Figure 87.2. Catheter Perforation of Right Atrium With Tamponade.
822 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Diagnosis of Cardiac Injury
Cardiac trauma must always be suspected after blunt or pene-
trating trauma to the chest or abdomen. A rapid assessment of
the patient is mandatory to look for clues of tamponade or hem-
orrhagic shock through evaluation of the airway, breathing, and
circulation (ABC); neck veins; and extremities. Myocardial con-
tusion is frequently unrecognized. It may lead to regional wall
motion abnormalities and to associated hemorrhagic infiltration
and myocyte necrosis, apparent on histologic examination. Acute
heart failure and ventricular arrhythmias may occur, but they
usually resolve within hours or days. Cardiac injury can occur in
the absence of sternal or rib fractures or other significant chest
injuries. In cases of blunt cardiac injury, the electrocardiogram
may show 1) nonspecific ST-T wave changes, 2) electrocardio-
graphic changes of acute pericarditis, or 3) pathologic Q waves.
An increase in the troponin level confirms the presence of car-
diac injury.
Echocardiography is the imaging method of choice for identi-
fication of cardiac injury (Figure 87.3). The findings include per-
icardial contusion, pericardial tamponade, regional wall motion
abnormalities, chamber enlargement, valvular incompetence,
and the presence of intracardiac shunts. An adequate transtho-
racic echocardiographic examination is not possible in up to 30%
of trauma victims, and transesophageal echocardiography may
be needed. Transesophageal echocardiography may not be possi-
ble in patients with cervical, maxillary, or mandibular injuries.
• Cardiac injury can occur in the absence of sternal or rib fractures
or other significant chest injuries.
• Echocardiography is the imaging method of choice for identifica-
tion of cardiac injury.
Treatment of Cardiac Injury
Emergency pericardiocentesis for cardiac tamponade may be
lifesaving if the patient is hemodynamically unstable. Echocar-
diographically guided pericardiocentesis is preferred if imme-
diately available. Emergency thoracotomy is the treatment of
choice for severe hemorrhage due to cardiac trauma. Lesser
degrees of cardiac contusion may be managed conservatively.
For hypotensive patients who have multiple-injury trauma and do
not respond to fluids, consider cardiac tamponade or ventricular
hypokinesia—both conditions are easily diagnosed with echocar-
diography. Inotropic agents and intra-aortic balloon counterpul-
sation (provided there is no aortic injury) may be beneficial for
ventricular hypokinesia due to myocardial contusion. Traumatic
rupture of the atrial or ventricular septum or major valve injury
generally requires surgical repair. Traumatic cardiac rupture due
to blunt trauma occurs most commonly in the right or left ventricle.
Late cardiac rupture may occur from contusion complicated by
intramyocardial hemorrhage, necrosis, and softening. Emergency
operation is the treatment of choice for cardiac rupture. Patients
who present with pseudoaneurysm formation should have surgi-
cal repair because future rupture is unpredictable.
Damage to Intracardiac Structures
The aortic valve is the most frequently damaged valve in nonpen-
etrating chest injuries. Patients with underlying valvular heart
disease are considered to be at higher risk than those without
preexisting disease. Aortic or mitral incompetence due to valve
leaflet tears usually manifests early and worsens with time.
Tricuspid valve injury is unique in that it may be recognized only
years after the original injury. Aortic incompetence may result
from a combination of damage to the aortic wall and damage to
the valve leaflets, and it may improve when perivalvular edema
and hemorrhage subside.
Sudden traumatic obstruction to left ventricular outflow dur-
ing systole may result in papillary muscle or mitral valve rup-
ture. The risk of cardiac valve injury is dependent on the time at
which the injury occurs during the cardiac cycle. Injury during
systole damages the mitral valve, whereas injury during diastole
damages the aortic valve. Severe abdominal injury, even in the
absence of chest trauma, may result in tricuspid valve or right
ventricular papillary muscle rupture. Definitive treatment for
significant valve injury is valve replacement or repair.
Injury to the coronary arteries from blunt or penetrating trauma
may lead to coronary occlusion and myocardial infarction. Left
ventricular pseudoaneurysm or aneurysm formation may result
from coronary artery trauma. Atrioventricular fistula formation
is a rare complication of penetrating trauma and most commonly
affects the right coronary artery. The fistula may extend from the
right coronary artery into the coronary sinus, the great cardiac
vein, the right ventricle, or the right atrium (Figure 87.4).

Figure 87.3. Traumatic Right Ventricular Contusion and Tricuspid

Valve Disruption. Modified right ventricular inflow view at 0°, multiplane
imaging, shows the right ventricle (RV) markedly enlarged and nearly aki-
netic on real-time examination. The anterior tricuspid leaflet (large arrow)
is unsupported and flail; both the anterior and the septal (small arrow)
leaflets are dwarfed by profound tricuspid annular dilatation, which is
responsible for an expansive gap between the noncoapting leaflets. RA
indicates right atrium. (Previously published. See “Credit Line” section.) Figure 87.4. Traumatic Rupture of Right Coronary Artery (×5).
 87 Cardiovascular Trauma
Figure 87.5. Traumatic Rupture of the Descending Thoracic Aorta
After a Motor Vehicle Accident. Transesophageal echocardiography in
transverse plane shows a large rent (arrow) in the aorta (Ao) commu-
nicating with an adjacent para-aortic space (arrowheads); there is also
a hematoma formation (H). (Previously published. See “Credit Line”
section.)
Figure 87.6. Aortic Transection Just Distal to the Ligamentum
Arteriosum Resulting From a Motor Vehicle Accident.
823
Figure 87.7. Chest Radiograph Showing Wide Mediastinum in
Acute Aortic Dissection With Rupture and Hemopericardium.
• The risk of cardiac valve injury is dependent on the time at which
the injury occurs during the cardiac cycle.
• Atrioventricular fistula formation is a rare complication of penetrat-
ing trauma and most commonly affects the right coronary artery.
Injury to the Aorta and Great Vessels
Rupture of the aorta is the most frequent nonpenetrating injury to
the great vessels and occurs in about 8,000 cases annually in the
United States. Traumatic rupture of the aorta occurs after rapid
deceleration injuries such as with falls from a height or automobile
accidents (Figure 87.5). Traumatic aortic rupture is present in about
20% of patients who die of injuries from motor vehicle accidents.
The site of the aortic tear is usually the junction of the aortic arch
and the descending aorta where the descending aorta is fixed to the
spine by the intercostal arteries, just distal to the origin of the left
subclavian artery (Figure 87.6). This injury is fatal in 80% to 90%
of patients, but survival with emergency cardiac operation has been
reported. Partial rupture of the aorta is associated with increased
arterial pressure in the upper extremities, decreased arterial pres-
sure in the lower extremities, and evidence of widening of the supe-
rior mediastinum on chest radiography or computed tomography
(Figure 87.7). Pseudoaneurysm of the aorta tends to expand and
rupture, but it also may contain thrombus that embolizes to distant
sites. Fistulas may form to adjoining structures. Transesophageal
echocardiography, computed tomography, and aortic angiography
are the imaging methods of choice in cases of suspected aortic
injuries. The most common angiographic findings are an intimal
flap and a pseudoaneurysm. With aggressive surgical intervention,
about 80% of patients who reach a hospital survive; without sur-
gery, in 2% to 5% of patients a chronic pseudoaneurysm develops.
• The site of the aortic tear is usually the junction of the aortic arch
and the descending aorta.
Electrical Injury to the Heart
Electrical injury to the heart may occur from a lightning strike
(about 100 deaths annually in the United States), from an indus-
trial or residential electrical accident, or from an electroshock
gun. Ventricular fibrillation is the usual cause of death. Drug
intoxication with cocaine or phencyclidine may increase the
lethality of an electrical injury by increasing catecholamine lev-
els and predisposing to fibrillation.

Acute Brain Injury and the Heart
NANDAN S. ANAVEKAR, MB, BCH, SARINYA PUWANANT, MD,
and KRISHNASWAMY CHANDRASEKARAN, MD
Nervous system and cardiovascular system functions are closely
intertwined: Cerebral perfusion is dependent on cardiac per-
formance, and much of cardiac function is regulated by higher
brain centess. Dysfunction in either organ system can precipitate
dysfunction in the other.
Acute brain injury may result from vascular injury (eg, stroke
or subarachnoid or parenchymal hemorrhage), trauma (eg, closed
head injury), or inflammation (eg, encephalitis or meningitis).
Cardiac sequelae may manifest as fluctuations in blood pressure
and hemodynamics, electrocardiographic changes, arrhythmias,
and elevations in levels of cardiac biomarkers.
Cardiac Innervation
The autonomic nervous system (Figure 88.1) strongly influences
the electrical and mechanical activities of the heart. It is com-
posed of 2 broad categories of efferent pathways, namely the
parasympathetic nervous system and the sympathetic nervous
system. These efferent pathways are modulated by higher brain
centers that constitute a functional unit, the central autonomic
network. Neurons in the cerebral cortex, basal forebrain, hypo-
thalamus, midbrain, pons, and medulla participate in autonomic
control. The central autonomic network integrates visceral,
humoral, and environmental information to produce coordinated
autonomic, neuroendocrine, and behavioral responses to external
or internal stimuli.
Parasympathetic Innervation of the Heart
The heart is innervated by both arms of the autonomic nervous
system. The parasympathetic preganglionic neurons originate in
Abbreviations and acronyms are expanded at the end of this chapter.
824
88
the nucleus ambiguus of the medulla and synapse with intracar-
diac ganglia by way of the vagus nerve. From these ganglia, short
postganglionic parasympathetic neurons innervate the myocar-
dial tissue. The parasympathetic innervation of the heart is par-
ticularly abundant in the sinus node and AV conduction system.
Parasympathetic innervation of the heart is mediated entirely
through the vagus nerve. The right vagus nerve innervates the
sinoatrial node and, when stimulated excessively, predisposes
the heart to sinus node bradyarrhythmias. The left vagus nerve
innervates the AV node and, when stimulated excessively, pre-
disposes the heart to AV blocks. The parasympathetic postgan-
glionic neurons release acetylcholine, which activates type 2
muscarinic receptors in the heart, resulting in slowing the heart
rate, reducing the contractile forces of the atrial cardiac mus-
cle, and slowing the conduction velocity through the AV node.
Vagal stimulation has no effect on ventricular muscle function.
Excessive vagal tone during emotional stress, which is usually
a parasympathetic overcompensation to strong sympathetic acti-
vation during stress, can cause syncope because of a sudden
decrease in blood pressure and heart rate. Patients with bulimia
and anorexia or spinal cord injury may have high vagal activity,
which is associated with the cardiac arrhythmias that often occur
in these patients.
Sympathetic Innervation of the Heart
The sympathetic innervation of the heart originates from the
intermediolateral column of the thoracic spinal cord and syn-
apses with the sympathetic postganglionic neurons in the supe-
rior, middle, and inferior cervical ganglia. The postganglionic
sympathetic neurons innervate the sinoatrial and AV nodes, the
conduction system, and myocardial fibers, most prominently in
the ventricles. The postganglionic sympathetic nerves release
 88 Acute Brain Injury and the Heart 825

Ciliary
III
Midbrain Eye
Pterygopalatine
VII Lacrimal gland
VII Mucous membrane of
Medulla Submaxillary nose and palate
IX Submaxillary gland
X Sublingual gland
Cervical Otic Oral mucosa
Superior Parotid gland
cervical
ganglion
Heart

Thoracic Larynx

Trachea

Greater Bronchi
splanchnic Celiac
ganglion Esophagus

Stomach
Lesser
splanchnic Abdominal vessels

Liver and ducts

Superior Pancreas
mesenteric
ganglion
Lumbar Adrenal

Inferior Small intestine

mesenteric
ganglion

Large intestine
Sacral

Rectum

Pelvic splanchnic nerves Kidney

Bladder

Sexual organs

External genitalia

Figure 88.1. Anatomy of the Autonomic Nervous System. (Previously published. See “Credit Lines” section.)
826 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
norepinephrine, which, primarily through β-adrenergic receptor
stimulation, increases heart rate, conduction velocity through the
AV node, and contractility of the heart. Sympathetic outflow to
the peripheral circulation causes vasoconstriction through acti-
vation of α-adrenergic receptors. Activation of these receptors
can increase the metabolic requirements of the heart, which can
manifest clinically as myocardial ischemia.
The autonomic outflow to the heart and peripheral vasculature
fluctuates moment to moment. It is regulated by various reflexes,
which are initiated by arterial baroreceptors and chemoreceptors
and by various intracardiac receptors. The autonomic innerva-
tion of the heart can also be influenced by pathologic events in
the central nervous system that alter the balance between par-
asympathetic and sympathetic outflow. These alterations can
ultimately lead to disturbances in cardiac function and hemo-
dynamics. Thus, cardiac innervation serves as the common link
between acute brain injury and its cardiac complications.
Cardiovascular Complications
of Cerebrovascular Accident
Patients who have a stroke are predisposed to cardiac distur-
bances. It is often difficult to ascertain in an individual patient
whether the cerebrovascular event was initiated by a primary
cardiac disturbance or whether the stroke caused a cardiac dis-
turbance, since the prevalence of cardiac morbidity in stroke
patients is high.
Electrocardiographic Abnormalities
ECG abnormalities are present in up to 90% of patients
presenting with acute stroke. Typical ECG changes are large,
upright T waves and prolonged QT intervals (Figure 88.2). These
ECG changes are also seen in SAH, transient ischemic attacks,
and nonvascular cerebral lesions. The ECG changes have been
postulated to arise from subendocardial ischemia as a result of
I aVR
II aVL
III aVF
II
V1
Figure 88.2. Electrocardiogram From a Patient With Acute Ischemic Stroke. Features include classic, tall, upright T waves (closed arrows) in
leads V3 through V5 and a prolonged corrected QT interval of 471 milliseconds. Occasional ventricular extrasystole (open arrow) is also seen.
increased centrally mediated catecholamine release in hypotha-
lamic hypoperfusion.
Repolarization Abnormality
ECG repolarization abnormality manifesting as QT prolonga-
tion is seen in about 38% of stroke patients. QT prolongation,
which is seen especially with right middle cerebral artery stroke,
increases the vulnerable period in the cardiac cycle for arrhyth-
mias and sudden death. In the absence of hypokalemia in stroke
patients, QT prolongation identifies high-risk patients vulnerable
to arrhythmia-related sudden death.
ST-Segment Abnormality
Nonspecific ST-segment changes are seen in over 20% of patients
presenting with acute stroke; the changes commonly include
ST-segment depression, a feature more frequently seen with left
middle cerebral artery stroke. Dynamic ST-segment changes
may also be indicative of true myocardial ischemia: ST-segment
changes due to stroke are generally transient and paradoxically
improve with brain death.
Q Waves
New Q waves occur in up to 10% of patients with acute stroke.
The Q waves may be a transient feature of the ECG, or they may
proceed through the typical evolutionary changes seen in myo-
cardial infarction. The Q waves seen in acute stroke do not reflect
myocardial ischemic damage.
U Waves
U waves are a common finding in acute stroke. They are usually
unrelated to any electrolyte abnormality and may be found alone
or with T-wave changes or prolonged QT intervals.
V1 V4
V2 V5
V3 V6
 88 Acute Brain Injury and the Heart 827

Arrhythmias Neurogenic Left Ventricular Dysfunction

Cardiac arrhythmias frequently occur after an acute stroke,
even when not present on the admission ECG. All types of
dysrhythmia can be seen, including ventricular extrasystoles,
atrial extrasystoles, supraventricular tachycardia, and ventric-
ular tachycardia. Only ventricular arrhythmia is associated
with increased mortality among stroke patients. A neurogenic
cause of these dysrhythmias is considered likely because many
of the patients have normal cardiac function. The pathophysi-
ologic mechanism is postulated to be alterations in the central
autonomic outflow to the heart. The type of arrhythmia can be
predicted from the type and location of the stroke. Stroke in the
right insular region tends to result in bradycardia and vasode-
pressor effects, and stroke in the left insular region tends to
result in tachycardia and hypertension. Restoration of normal
autonomic tone can take up to 6 months after the acute cerebro-
vascular event. During this period, the patient has an increased
risk of sudden cardiac death.
Cardiac Biomarkers
In acute stroke, patients may have increased levels of cardiac bio-
markers, including troponin T, creatine kinase, and myoglobin.
Typically, the magnitude of the increase is small. Usually the
troponin T level becomes elevated in stroke patients only when
they have coronary artery disease; this elevation is associated
with left ventricular dysfunction and a poor prognosis. Increased
autonomic sympathetic discharge increases myocardial oxygen
demand and may result in myocardial stunning and elevated lev-
els of cardiac biomarkers. Classic myocardial infarction occurs
in up to 6% of patients with acute stroke and presents a therapeu-
tic dilemma since many myocardial infarct therapies increase the
risk of intracerebral bleeding.
Neurogenic stunned myocardium describes neurologically medi-
ated cardiac injury that is reversible and clinically manifested by
ventricular dysfunction with or without hemodynamic instabil-
ity. Commonly it is accompanied by ECG changes, arrhythmias,
and release of cardiac biomarkers. This phenomenon is unrelated
to any underlying coronary artery disease and frequently occurs
in stroke patients, more commonly in patients with left insular
stroke. Pathologic findings include a characteristic pattern of pete-
chial subendocardial hemorrhage and “contraction band necrosis.”
The pathogenic mechanism is thought to be excessive autonomic
sympathetic discharge and catecholamine release. Catecholamine
blockade is protective against neurogenic cardiac injury.
Takotsubo syndrome (apical ballooning syndrome), a condi-
tion whose exact etiology remains undetermined, can mimic the
clinical presentation of acute anterior wall myocardial infarction.
Some authorities postulate that the syndrome arises from stress-
associated catecholamine release. Takotsubo syndrome is asso-
ciated with a characteristic pattern of abnormal ventricular wall
motion, with hypokinesis of the cardiac apex and mid ventricle
and relative sparing of the cardiac base (Figure 88.3). Prognosis
for recovery is excellent.
Subarachnoid Hemorrhage
SAH is associated with significant patient morbidity and mor-
tality and accounts for 10% of all strokes. It usually results from
rupture of a saccular intracerebral aneurysm, but other causes
include trauma, arteriovenous malformation, and use of illicit
drugs, including cocaine and amphetamines. Almost all the
changes described above for the ECG, cardiac biomarkers, and
left ventricular dysfunction can occur in SAH; however, certain
characteristics are typically seen.

A B

Figure 88.3. Left Ventriculograms From a Patient With Apical Ballooning Syndrome. End-diastolic (A) and end-systolic (B) frames show aki-
nesis of the mid ventricle and dyskinesis of the apical regions of the left ventricle, with hyperdynamic contraction of the basal left ventricle. This
phenomenon may occur with any acute brain injury but is most common with subarachnoid hemorrhage and ischemic stroke involving the thalamus
and brainstem.
828 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

The acute ECG changes noted in more than 50% of patients
with SAH are classically described as deep T-wave inversions
or “cerebral T waves” (Figure 88.4). ECG changes may be seen
up to 2 weeks after the acute bleeding episode. Prolongation of
the QT interval and QT dispersion are seen in more than 70% of
patients presenting with SAH, a finding that predisposes them to
ventricular arrhythmia and sudden death. A prolonged corrected
QT interval of more than 440 milliseconds identifies severe head
trauma patients at risk of ventricular arrhythmias and sudden
death.
Elevations in cardiac enzyme levels are common after SAH;
the putative mechanism is excessive sympathetic discharge. The
greater the increase in the troponin level, the worse the clinical
outcome.
Neurogenic left ventricular dysfunction is much more com-
mon in SAH than in ischemic stroke, with an incidence of about
10%. The pathophysiology of left ventricular dysfunction is
similar to that in stroke, namely cardiac myocyte injury from a
catecholamine surge due to increased sympathetic discharge. In
some patients with SAH, the apex and the base of the left ventri-
cle contract normally and the mid ventricle is akinetic, a condi-
tion that Japanese authors call “panic myocardium.”
output; with concomitant cardiac dysfunction, the neurologic
outcome is worsened.
Brain Death and the Heart
Patients with brain death after head injury, ischemic stroke, or
SAH are potential organ donors for cardiac transplant. Hearts
with dysfunction from neurogenic mechanisms generally recover
as excess autonomic sympathetic activity subsides, which may
take from 72 hours to 1 week. Myocardial recovery depends on
the maintenance of adequate mean arterial pressure to ensure
coronary perfusion. β-Blockers can protect the myocardium
from the toxic effects of catecholamines, and glucocorticoids
decrease cardiac dysfunction after brain death.
Head Injury and Anticoagulation
In patients who are receiving anticoagulation, head injury
adversely affects survival by increasing the incidence of intracra-
nial hemorrhage. An international normalized ratio greater than
3.3 is associated with an increased incidence of intracerebral
hemorrhage after a head trauma and with an adverse outcome.

Cardiac Complications of Head Trauma

Closed head trauma leads to 175,000 deaths and 500,000 hospi-
talizations per year, with a peak incidence in men aged 15 to 24
years. Head trauma is associated with cardiac complications that
can negatively affect clinical outcomes, including cardiac rhythm
and conduction disturbances. ECG findings include diffuse, tall,
upright, or deeply inverted T waves; prolonged QT intervals;
ST-segment depression or elevation; and U waves. Dysrhythmias
in head-injured patients often resolve when intracranial pressure
is decreased. QT prolongation and associated fatal arrhythmias
are more commonly associated with intracerebral hemorrhage
with increased intracranial pressure. Reduction of ischemic
cerebral injury depends on the preservation of adequate cardiac
Cardiac Manifestations in Seizures
Epileptic Seizures
Cardiovascular manifestations of epileptic seizures are common,
and symptoms and signs can occur in the preictal, ictal, or post-
ictal period. The typical cardiovascular effects of alterations in
autonomic function include changes in heart rate, blood pressure,
and ECG. ECG manifestations of epilepsy include ST-segment
depression, ST-segment elevation, and T-wave inversion. In
a seizure, ECG changes can occur without changes in cardiac
rhythm. Sinus tachycardia occurs in more than 60% of patients,
and when it accompanies peripheral vasoconstriction and an
increase in blood pressure, sinus tachycardia is associated with

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

Figure 88.4. Electrocardiograms From a Patient With Subarachnoid Hemorrhage. A, The ST-segment elevation and T-wave inversion (arrows)
in leads V2 and V3 may suggest ST-segment elevation anteroseptal infarction. However, the absence of reciprocal changes and marked prolongation
of the QT interval are common in subarachnoid hemorrhage and help differentiate it from ST-segment elevation myocardial infarction. B, Follow-up
electrocardiogram shows persistence of ST-segment elevation and changes in the T waves. The T waves are large and symmetrically inverted (arrows)
in leads V1 through V5, with a markedly prolonged corrected QT interval.
 88 Acute Brain Injury and the Heart 829

I aVR V1 V4

II aVL V2 V5

III aVF V3 V6

II

V1

Figure 88.4. (Continued)

hypothalamic lesions. Bradyarrhythmias occur in less than 5% Cardiac Complications of Encephalomyelitis

of seizures and include sinus bradycardia, sinus arrest, AV block,
and prolonged asystole. Other cardiac arrhythmias that occur
with seizures include acceleration and deceleration of heart rate,
enhanced sinus arrhythmia, atrial premature beats, sinus pauses,
AV block, nodal escape, paroxysmal supraventricular tachycar-
dia, and ventricular ectopy.
Electroconvulsive Therapy
ECT artificially induces a seizure that is often associated with an
increase in catecholamine levels and autonomic discharge, par-
ticularly parasympathetic discharge. ECG changes occur fairly
commonly during ECT and are similar to changes in primary
seizure disorders. There has been only 1 prospective study ana-
lyzing the effects of ECT on left ventricular systolic function. It
showed that ECT-associated left ventricular systolic dysfunction
is a common but transient early phenomenon. Multiple ECT ses-
sions do not have a cumulative effect on ventricular function, and
tolerance to shocks appears to develop after multiple treatments.
Encephalomyelitis is an inflammatory disorder of the central
nervous system, dorsal root ganglia, and autonomic nerves due
to an infectious or noninfectious cause. Noninfectious causes are
often associated with vasculitis or paraneoplastic syndromes.
Encephalomyelitis may be associated with cardiac manifesta-
tions owing to profound disturbances in autonomic function that
are common in this condition. These findings include hyperten-
sion, tachycardia, and high plasma catecholamine levels that are
consistent with a hypersympathetic state. The plethora of car-
diac manifestations in brainstem encephalitis is explicable by the
density of autonomic fibers that traverse the brainstem and by the
location of the primary vasomotor area.
Abbreviations
AV atrioventricular
ECG electrocardiographic
ECT electroconvulsive therapy
SAH subarachnoid hemorrhage
 89
Noncardiac Anesthesia in Patients With
Cardiovascular Disease
LAURENCE C. TORSHER, MD
Introduction
The perioperative period stresses the cardiovascular system
due to hemodynamic and neuroendocrine physiologic changes
induced by surgical trauma, underlying disease, fluid shifts,
inflammation, blood loss, as well as anesthetic medications and
interventions.
Preoperative Preparation
Screening and Risk Mitigation
Preoperative assessment for noncardiac surgery in patients with
heart disease is covered in detail in Chapter 4. Preoperative
assessment stratifies patients into risk groups so that physicians
and patients can make informed decisions about the surgery.
Preoperative assessment should also provide information to the
perioperative medical personnel that will allow them to opti-
mize care of the patient. In the past, risk scoring systems like the
Goldman and Detsky systems would allow medical personnel to
identify patients as high or low risk but did not provide informa-
tion as to how to optimize the care. The American Society of
Anesthesiologists classification (Table 89.1) is an extremely
simple global risk stratification scheme that is based solely on
functional status. In spite of its simplicity it has proven to be as
sensitive as many more sophisticated schemes.
Preoperative evaluation, in addition to giving information
about the perioperative risk, should also provide information that
will affect perioperative management decisions (eg, the deci-
sion to forgo or modify a proposed surgical procedure, delay a
procedure to optimize the patient’s medical conditions, choice
of intraoperative and postoperative monitors, modification of
Abbreviations are expanded at the end of this chapter.
830
perioperative medical therapy, disposition of the patient post-
operatively (floor, ICU, outpatient) and even location of care
(outpatient surgicenter, small community hospital, tertiary-care
center).
With respect to coronary artery disease, emphasis should
be placed on identifying the extent of the disease, whether it is
stable or evolving, and left ventricular performance. The impact
of the disease and its impact on outcome have been formal-
ized in the 2007 ACC/AHA guideline entitled “Perioperative
Cardiovascular Evaluation for Noncardiac Surgery.” The surgi-
cal procedure being proposed is classified as high, intermediate,
or low risk (Table 89.2). Then a stepwise approach, as outlined
in Figure 89.1, is utilized to determine whether to immediately
proceed to surgery or proceed with further testing. Patients with
active cardiac conditions (Box 89.1) should have all but emer-
gency surgery postponed until the conditions are optimized,
although these active conditions are issues that should be dealt
with whether the patient were coming for surgery or not. The
recurring theme throughout the guidelines is an assessment
of functional status (as a clinical reflection of left ventricular
performance) as reflected by exercise tolerance. Patients with
no clinical risk factors (Box 89.2) with moderate or excellent
exercise capability (>4 METs) may proceed directly to surgery
without further workup.
Aortic stenosis is the most important of the cardiac valve
lesions to be identified preoperatively because of the associated
incidence of sudden death as well as the ineffectiveness of cardiac
massage during a cardiac arrest. Identification will allow discus-
sion of correction of the lesion or at the least communication
to the care givers who will concentrate on avoidance of tachy-
cardia and maintaining preload, contractility, and afterload.
Patients with mitral stenosis are at increased risk for pulmo-
nary edema. Perioperative management goals are control of heart
 89 Noncardiac Anesthesia in Patients With Cardiovascular Disease
Table 89.1. ASA Physical Status Classification Systema
Classification Description
1 Healthy patient with no systemic disease
2 Mild systemic disease, no functional limitations
3 Moderate to severe systemic disease, some
functional limitations
4 Severe systemic disease, incapacitating, and a
constant threat to life
5 Moribund patient, not expected to survive >24 h
without surgery
6 Brain-dead patient undergoing organ harvest
E Added when the case is emergent
a
The American Society of Anesthesiologists (ASA) physical status
classification, developed in 1941, is used for risk stratification in outcome
studies.
Previously published. See “Credit Line” section.
rate, maintenance of sinus rhythm (with a low threshold to move
to DC cardioversion if atrial fibrillation occurs), and maintaining
preload and afterload. Regurgitant lesions pose far lower periop-
erative risk than stenotic valve lesions.
Patients with congenital heart disease and Eisenmenger-type
physiology (pulmonary hypertension and right to left shunting)
are at significant perioperative risk with noncardiac surgery.
Pulmonary hypertension due to any cause increases peri-
operative risk. HCM is associated with sudden death but more
commonly exacerbation of heart failure in the perioperative per-
iod. HCM’s physiological derangements may become more sig-
nificant in the perioperative period since large fluid shifts may
lead to ventricular underfilling and exaggeration of symptoms.
Also, surgical stress and pain may lead to increased contractility
and tachycardia that shortens diastole more than systole and thus
reduces ventricular filling. All of these factors increase the intra-
ventricular gradient of HCM and may exacerbate symptoms.
Hypertension, unless severe, has not been shown to be
an independent risk factor for perioperative cardiac events.
However, preoperative hypertension has been noted to be asso-
ciated with increased intraoperative lability of blood pressure.
Unless the patient is showing evidence of end-organ damage
from the hypertension or it is severe, it is reasonable to proceed
with surgery. Antihypertensive medications should be continued
throughout the perioperative period.
Table 89.2. Risk Stratificationa
Risk Stratification Procedure Examples
High (reported cardiac risk Aortic and other major vascular surgery
often more than 5%) Peripheral vascular surgery
Intermediate (reported cardiac Intraperitoneal and intrathoracic
risk generally 1% to 5%) surgery
Carotid endarterectomy
Head and neck surgery
Orthopedic surgery
Prostate surgery
Lowb (reported cardiac risk Endoscopic procedures
generally less than 1%) Superficial procedure
Cataract surgery
Breast surgery
Ambulatory surgery
a
Combined incidence of cardiac death and nonfatal myocardial infarction.
b
These procedures do not generally require further preoperative cardiac testing.
Previously published. See “Credit Line” section.
831
Preoperative Medications
Perioperative fasting guidelines have not changed recently.
Currently clear fluids are allowed up to 2 hours before surgery,
so this need not be an excuse for withholding oral medications.
Diuretics should be held on the day of surgery to avoid exa-
cerbation of hypovolemia in a fasting patient and electrolyte
disturbances.
Anticoagulation, including aspirin, should be evaluated on
a case-by-case basis, taking into account the surgery proposed
and indication for the anticoagulation. This may require bridging
therapy with heparin or other agents.
In the case of aspirin or clopidogrel that is being given as
thrombus prophylaxis for coronary stents, there is significant
risk for stent thrombosis if antiplatelet agents are stopped prior
to one year after stent placement. Unless surgery is emergent,
consideration should be given to postponing surgery until the
one-year anniversary of stent placement before discontinuing
antiplatelet agents. Patients with bare metal stents are at much
lower risk of stent thrombosis, and essential but non-emergency
surgery can performed 4-6 weeks after stent implant on aspirin
alone. Elective surgery should ideally be postponed for 3 months.
Patients treated with PTCA without stenting can have surgery
2 weeks after stent implant on aspirin alone. Frank discussion
with the surgeon may reveal that many cases can be done with
antiplatelet agents in place. Aspirin should not be discontinued in
the perioperative period in stented patients.
β-Blockers should be continued throughout the perioperative
period in any patient who normally takes them. In the postoper-
ative period the β-blockers should be given intravenously if the
patient is not able to take them by mouth. If the patient is not
already on a β-blocker and there is no contraindication, current
ACC/AHA guidelines suggest that the patients undergoing inter-
mediate or high risk surgery that have one or more clinical risk
factors may benefit from perioperative β-blockade. β-Blockade
should be started prior to surgery and continue postoperatively.
There is evidence to suggest that treating everyone in the periop-
erative period may decrease MIs, but increase stroke rate as well
as 30-day death rates in addition to clinically significant episodes
of bradycardia and hypotension. There is insufficient evidence to
state definitively which β-blocker to use, or how long to continue
therapy postoperatively although it should probably be at least 7
and as long as 30 days.
Clonidine has been shown to decrease the risk of periop-
erative cardiac events in intermediate and high-risk patients
undergoing surgery when started preoperatively and continued
throughout the perioperative period in both β-blocked and non-
β-blocked patients. It may be an alternative for those patients
in whom β-blockade is contraindicated. If the patient is taking
clonidine already it must be continued in the perioperative per-
iod either orally, by patch, or intravenously to avoid catastrophic
rebound effects.
ACE inhibitors, taken preoperatively and in the immediate
perioperative period have been associated with refractory hypo-
tension as well as increased mortality in cardiac and vascular sur-
gery patients in small series. Meta-analysis and larger studies do
not show the same association. A practical strategy, given these
conflicting results, is to withhold ACE inhibitor therapy on the
day of surgery and restart in the postoperative period.
HMG-CoA-reductase inhibitors (statins) appear to decrease
the rate of cardiac events by stabilizing plaques and decreasing
systemic inflammation in at risk patients. In addition, in CABG
patients it appears that there may be some renal protective effects.
832 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Need for emergency Yes Perioperative surveillance

STEP 1 Operating room
noncardiac surgery? (Class I, LOE C) and postoperatve risk
stratification and risk
No factor management

Active cardiac Yes Evaluate and treat per Consider

STEP 2 conditions* ACC/AHA guidelines operating room
(Class I, LOE B)

No

Yes Proceed with

STEP 3 Low risk surgery planned surgery†
(Class I, LOE B)

No

Functional capacity
greater than or equal Yes Proceed with
STEP 4 planned surgery†
to 4 METs without (Class IIa, LOE B)
symptoms‡

No
STEP 5 or unknown

3 or more clinical 1-2 clinical No clinical

risk factors§ risk factors§ risk factors§

Vascular surgery Intermediate Vascular surgery Intermediate

risk surgery risk surgery Class I,
LOE B
Class IIa,
LOE B

Consider testing Proceed with planned surgery with HR control¶ (Class IIa, LOE B) Proceed
if it will change or consider noninvasive testing (Class IIb LOE B) with planned
management¶ if it will change management surgery†

Figure 89.1. Cardiac Evaluation and Care Algorithm for Noncardiac Surgery Based on Active Clinical Conditions, Known Cardiovascular
Disease, or Cardiac Risk for Patients 50 Years of Age or Greater. *See Table 2 in source for active cardiac conditions. †See Class III recommenda-
tions in Section 5.2.3 of source. Noninvasive Stress Testing. ‡See Table 3 in source for estimated MET level equivalent. §Noninvasive testing may
be considered before surgery in specific patients with risk factors if it will change management. Clinical risk factors include ischemic heart disease,
compensated or prior heart failure, diabetes mellitus, renal inssufficiency, and cerebrovascular disease. ¶ Consider perioperative β-blockade (See
Table 11 in source) for populations in which this has been shown to reduce cardiac morbidity/mortality. ACC/AHA indicates American College of
Cardiology/American Heart Association; HR, heart rate; LOE, level of evidence; and MET, metabolic equivalent. (Previously published. See “Credit
Line” section.)

Patients for whom long-term statin therapy would be indicated
appear to benefit from them in the perioperative period. Current
recommendations are to continue patients on their statin medica-
tions if they are otherwise indicated, but there is currently not
enough data to suggest starting them simply for the perioperative
period. Statins, like β-blockers, should be continued through the
postoperative period. There is evidence to suggest that with acute
withdrawal of statins an inflammatory rebound may occur.
Devices
Indications for pacemaker and ICD placement perioperatively
are no different than if the patient were not going to the OR.
Pacemakers should be interrogated for proper function prior
to going to the OR and if there are any rate enhancements func-
tions they should be turned off, since electrical noise from the
electrosurgical units (Bovie) may interfere with them. The
patient’s underlying rate, rhythm, and dependence on the device
should be documented. If the patient is pacemaker dependent, the
device should be set to an asynchronous mode (VOO or DOO).
Effect of a magnet on the particular device should be communi-
cated to the anesthesia team. ICD units should be disabled prior
to going to the OR. While the ICD is disabled, the patient will
require continuous monitoring. Electrical interference with the
device can be minimized by using bipolar rather than unipolar
cautery devices, using short bursts of cautery, maximizing the
distance between the cautery and the device, and finally if uni-
polar cautery must be used, placing the grounding or return pad
in a position so that the electrical pathway from the cautery wand
to the grounding pad does not cross the device. Both pacemakers
and ICDs should be reprogrammed as soon as the patient returns
to the recovery room.
Not every generator palpated in the chest is a pacemaker or
ICD. Pain stimulators, thalamic stimulators for Parkinson dis-
ease, phrenic nerve stimulators for diaphragmatic drive, or vagus
stimulators for seizure control may also be placed in the upper
 89 Noncardiac Anesthesia in Patients With Cardiovascular Disease
Box 89.1. Active Cardiac Conditions That Should
Postpone all but Emergency Surgeries Until Resolved
or Optimized
Ischemic heart disease
Heart failure
Diabetes mellitus
Renal failure
Stroke
Previously published. See “Credit Line” section.
chest. Some subcutaneous venous access devices may be mis-
taken for a pacemaker if simply palpated as well. Devices can
usually be identified by examination of the chext X-ray.
Intraoperative Management
Anesthetic Choice
There is no evidence to definitively show that a well-conducted
general anesthetic is any more or less successful than a regional
anesthetic in minimizing cardiac complications. Some have
Box 89.2. Clinical Risk Factors Modifying Decision
Making Using a Stepwise Approach to Preoperative
Evaluation
Unstable coronary syndromes
Acute or recent MIa with evidence of important
ischemic risk by clinical symptoms or noninvasive
study
Unstable or severeb angina (Canadian
Cardiovascular Society class III or IV)c
Decompensated heart failure
Significant arrhythmias
High-grade atrioventricular block
Mobitz II atrioventricular block
Third-degree atrioventricular heart block
Symptomatic ventricular arrhythmias in the
presence of underlying heart disease
Supraventricular arrhythmias with uncontrolled
ventricular rate
Newly recognized ventricular tachycardia
Severe valvular disease
Severe aortic stenosis
Symptomatic mitral stenosis
Abbreviation: MI, myocardial infarction.
a
According to the American College of Cardiology National Database
Library, a recent MI is one that occurred >7 days but <1 month (30
days) previously; an acute MI occurred within the previous 7 days.
b
May include “stable” angina in patients who are unusually sedentary.
c
Campeau L. Grading of angina pectoris. Circulation. 1976 Sep;54(3):
522–3.
Previously published. See “Credit Line” section.
833
suggested that regional anesthetics may be safer, but those tri-
als provided virtually no sedation to their regional patients, a
practice not universally used. Early literature suggests that there
is a lower incidence of thromboembolism in patients undergo-
ing regional anesthesia with lower extremity surgery. There is
some suggestion that this may also extend to slightly lower inci-
dence of graft failure in vascular patients as well, presumably by
decreasing the degree of surgical stress that triggers a periopera-
tive hypercoagulable state. Aggressive postoperative pain relief
with a multimodal approach to care, including regional anes-
thetic techniques, neuraxial narcotics, and parenteral and oral
nonnarcotic analgesics, in addition to traditional narcotic anal-
gesic management, may decrease sympathetic outflow, and sec-
ondarily decrease cardiac risk. Choice of anesthetic technique
should be left to the anesthesia provider.
General Anesthetic
A general anesthetic consists of medications that result in uncon-
sciousness, amnesia, analgesia, and relaxation. The patient may
breathe spontaneously or with positive pressure ventilation. The
airway may be secured with an endotracheal tube, laryngeal
mask, or the patient may be breathing simply through a mask.
Commonly used anesthetic drugs and their effects on cardio-
vascular physiology are outlined in Table 89.3.
Hypoventilation may occur because of loss of airway when a
breathing device is not in place or incorrectly placed in a sedated
or anesthetized patient or decreased drive to breathe due to medi-
cations. Hypoxemia may occur because of hypoventilation, V/Q
mismatch due to atelectasis, aspiration, pulmonary edema, or pre-
existing lung pathology. Cardiac disease may contribute to the
hypoxemia because of pulmonary congestion or right-to-left shunt.
All of these may be exacerbated in the anesthetized patient because
of supine positioning, positive pressure ventilation, impaired
hypoxic pulmonary vasoconstriction, and blunting of compensa-
tory autonomic nervous system mechanisms by medications.
Hypotension may arise from fluid shifts, or decreases in pre-
load, afterload or contractility from anesthetic agents. Increased
intrathoracic pressure associated with positive pressure ventila-
tion may also contribute to decreased preload.
Regional Anesthetic
Regional anesthesia may consist of a peripheral or a central
neuraxial block. Peripheral techniques include blockade of a
single extremity (eg, axillary plexus block for hand or forearm
surgery) or block of a single peripheral nerve (eg, femoral nerve
block for analgesia of the anterior thigh). Central neuraxial tech-
niques refer to those in which a needle is placed within the spinal
canal (eg, spinal or epidural injections). Regional anesthetics are
rarely used as the sole technique; various degrees of sedation are
usually used as well, with concomitant risks.
Contraindications to regional anesthesia include preexisting
neurological abnormality, coagulopathy, allergy to local anes-
thetic agents, systemic infection, local infection at needle insertion
site, or lack of patient cooperation. Patients are at risk for a failed
or patchy block which will require either conversion to a general
anesthetic, supplementation by the surgeon, or heavy sedation.
Patients are also at risk for LAST, which may manifest as
short, self limited seizure to life-threatening cardiac dysrhyth-
mias. LAST culminating in nonperfusing cardiac dysrhythmias
had frequently been fatal in the past. Recently, utilization of
intravenous lipid emulsion (20% Intralipid) has shown success in
834 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation

Table 89.3. Hemodynamic Effects of Common Drugs Used in General Anesthesia

Drug Heart Rate Preload Afterload Contractility Comment

Sedatives
Benzodiazepines ↓ — Slight ↓ —
Clonidine ↓ — ↓ —
Scopolamine Slight ↑ — — —
Induction drugs
Propofol ↓ ↓ ↓ ↓ Dose-dependent effect
Thiopental ↓ ↓ ↓ ↓
Etomidate — — Slight ↓ —
Dexmedetomidine ↓ — ↓ —
Ketamine ↑ — ↑ — Increases myocardial oxygen
consumption
Reversal agents
Naloxone ↑ — ↑ ↑ Sympathetic activation due to sudden
loss of narcotic modulation
Flumazenil ↑ — ↑ ↑ Sympathetic activation due to sudden
loss of sedative action
Narcotics
Fentanyl ↓ — — —
Sufentanil ↓ — — —
Alfentanil ↓ — — — Short-acting narcotic
Remifentanil ↓ — — Slight ↑ Ultrashort-acting narcotic
Morphine ↓ ↓ ↓ ↓ Vasodilation due to histamine release
Methadone ↓ — — — Torsades de pointes associated with
Meperidine ↑ — — ↓ high doses
Muscle relaxants
Succinylcholine ↓ — — — Transient hyperkalemia may lead to
Pancuronium ↑ — — — dysrhythmias
Vecuronium — — — —
Rocuronium — — — —
Atracurium — ↓ ↓ —
Cisatracurium — — — — Vasodilation due to histamine release
Inhaled agents
Isoflurane ↑ ↓ ↓ ↓
Sevoflurane ↓ ↓ ↓ ↓
Desflurane ↑ ↓ ↓ ↓
Nitrous oxide — ↑ ↑ ↑ Increases sympathetic tone
Xenon — — — —
Antiemetics
Droperidol — ↑ ↓ — Prolongs QT interval
Ondansetron — — — —

resuscitation of these patients, presumably by acting as a “lipid
sink,” sequestering free local anesthetic from the circulation.
The neuraxial techniques, spinal or epidural catheter, are
most commonly used. In addition to the general contraindi-
cations for regional anesthesia, elevated intracranial pressure
and spinal stenosis are other contraindications to these blocks.
Blockade is generally dermatomal. In addition to the sensory
and motor blockade, sympathetic blockade will lead to venodi-
lation (decreased preload) and arterial vasodilation (decreased
afterload) with subsequent hypotension. If the sympathetic
blockade extends above T4, it begins to impinge upon the car-
diac accelerator fibers of the sympathetic nervous system and
may result in unopposed parasympathetic tone to the heart
with profound bradycardia. The hypotension seen with central
neuraxial techniques is usually treated with IV fluids to sup-
plement preload as well as α-agonists, usually phenylephrine,
to reverse to vasodilation. Anticholinergics like atropine or
glycopyrrolate will reverse the bradycardia associated with the
high block. It has been noted that in cardiac arrest in patients
with spinal anesthetics, early use of epinephrine led to higher
survivability.
Bleeding from needle trauma within the spinal canal is
potentially catastrophic because pressure from a hematoma
impinging on the cord and roots may result in a devastating and
potentially irreversible deficit. Therefore, the anesthesiologist
is especially cognizant of coagulation abnormalities in these
patients both at the time of surgery and postoperatively. Patients
are at risk for bleeding with initial needle placement and place-
ment of indwelling catheter and also with catheter manipula-
tion. The American Society of Regional Anesthesia and Pain
Medicine (ASRA) drafted a consensus statement with recom-
mendations around the use of these techniques in the face of
anticoagulation.
1. Avoid fibrinolytic agents for 10 days after a spinal or epidu-
ral. In the event that a patient requires antifibrinolytic therapy
within that window, the patient should have frequent assess-
ment of his/her neurological function on a q 2 hourly basis.
 89 Noncardiac Anesthesia in Patients With Cardiovascular Disease
2. In using unfractionated heparin, subcutaneous heparin 5,000
U twice daily, is not a contraindication to neuraxial tech-
niques. Therapeutic levels of anticoagulation with heparin
infusion are a contraindication to spinal orepidural block.
The infusion should be off 4 hours before starting spinal or
epidural block. If a neuraxial technique is anticipated and
there will be intraoperative anticoagulation, this is accepta-
ble if there are no other concomitant coagulopathies and the
heparin is given more than one hour after spinal or epidural
block occurred. If there is a postoperative indwelling epidu-
ral or spinal catheter, heparin should be discontinued for 2 to
4 hours before it is removed (spinal hematoma may frequently
be associated with bleeding from catheter removal as well as
placement) and should not be restarted for at least an hour
after its removal. The indwelling catheter should minimize
the use of local anesthetics to avoid masking new neurologi-
cal deficits. Antiplatelet or oral anticoagulants in addition to
the heparin may increase the risk of bleeding.
3. In patients receiving prophylactic dosing of LMWH preoper-
atively, the last dose should be at least 12 hours prior to nee-
dle placement. Patients receiving treatment doses of LMWH
preoperatively should have a 24 hour window prior to nee-
dling. Patients starting on postoperative LMWH for throm-
boprophylaxis on a twice daily regimen should have the first
dose started at least 24 hours after needle placement and the
epidural catheter should be removed at least 2 hours prior to
the first dose. With once a day dosing, the first dose may be
given 6–8 hours after needle placement with the next dose
at least 24 hours after the first dose. An indwelling catheter
should be removed at least 10 to 12 hours after LMWH dosing
and the next dose of LMWH should be given at least 2 hours
after the catheter is discontinued. Antiplatelet or oral antico-
agulants in addition to the LMWH may increase the risk of
bleeding.
4. Coumadin should be discontinued at least 4 days prior to the
procedure and a normalized INR documented. If coumadin
is being started for perioperative thromboprophylaxis and the
first preoperative dose was taken more than 24 hours prior or a
second dose was taken preoperatively, a normal INR should be
documented prior to needle placement. If coumadin is being
taken while an indwelling catheter is in place, daily INRs
should be checked and the catheter discontinued if the INR
>1.5. If the INR>3.0 while an indwelling catheter is in place,
coumadin should be held to facilitate removal of the catheter.
5. In reviewing bleeding risks with antiplatelet agents, non-
steroidal antiinflammatory drugs by themselves pose no addi-
tional bleeding risk for neuraxial placement. However patients
should stop taking ticlopidine for 14 days and clopidogrel for
7 days prior to needle placement. For normal platelet function
to return requires 24 to 48 hours after abciximab and 4–8
hours after eptifibatide or tirofiban. One should avoid nee-
dling within that time. Other anticoagulants used in addition
to these agents will increase the risk of bleeding.
6. In patients receiving thrombin inhibitors (eg, desirudin, lep-
irudin, bivalirudin, and argatroban), current recommenda-
tions are to avoid neuraxial techniques in patients receiving
these medications because there is insufficient evidence to
comment on the safety and there is no reversal or antidote.
Because of these bleeding risk factors, if indwelling catheters
are used as part of the postoperative analgesia regimens, a mech-
anism should be in place to flag new orders for anticoagulation in
patients with indwelling catheters.
835
Ultimately, choice of anesthetic technique is influenced by
skills of the local anesthesiology and surgical departments.
Sedation or Monitored Anesthesia Care
Although one would expect that patients having only sedation or
sedation with local anesthesia would be the safest, in fact con-
fidential reporting of anesthetic complications in Australia has
shown that there is a similar risk of death with sedation com-
pared with other anesthesia techniques. This may be a result of
decreased vigilance because of the misperception they are safer,
inadequate block of procedural site resulting in escalating dosing
of sedation, different staffing models with less skilled caregivers
providing care, or the proceduralist attempting to simultaneously
take responsibility for sedation care.
Monitoring
The role of the pulmonary artery catheter for perioperative man-
agement is becoming far less clear than once thought. Numerous
trials both within the OR and the ICU environment have shown
no survival benefit. Fewer catheters are being placed in practice.
Management from clinical signs and symptoms and utilization
of TEE monitoring seem to be as successful as the pulmonary
artery catheter.
TEE has been used by anesthesiologists for many years
during cardiac surgery cases. As its use becomes a common
part of the training for anesthesiology residents, TEE’s role is
expanding beyond the cardiac OR. It is used to confi rm normal
cardiac anatomy as well as to provide an ongoing assessment
of cardiac performance. It is used in liver transplantation to
evaluate both right ventricular and left ventricular function and
filling. In the patient with significant myocardial dysfunction it
can be helpful for guiding therapy. Survey for patent foramen
ovale and air emboli is useful in patients with unusual posi-
tioning (eg, sitting neurosurgical patients or some orthopedic
cases).
A recent trend has been a move towards goal-directed fluid
therapy in the perioperative period, in which fluids are admin-
istered to meet a specific target, typically a change in cardiac
output. Various methods, both invasive and noninvasive, to deter-
mine cardiac output, or a surrogate for it, have been utilized.
Other
Maintaining perioperative normothermia has been noted to
decrease surgical infections, as well cardiac complications.
Presumably this is due to the shivering and increased systemic
vascular resistance noted in the hypothermic patient. Transfusion
triggers for the surgical patient continue to be debated. The ben-
efits of increased oxygen carrying capacity must be balanced
against the risks of microvascular impediments to flow, altered
inflammatory and immune response, and circulatory overload.
Postoperative Care
Disposition
Decisions must be made postoperatively whether to admit the
patient to a monitored setting or discharge to a regular floor.
Factors to take into account are the likelihood of myocardial
ischemia and associated rhythm disturbances and risk of other
rhythm abnormalities; perioperative risk of bleeding and fluid
shifts may also alter cardiac function. Finally, pain control and
836 VIII Diseases of the Heart, Pericardium, and Pulmonary Circulation
monitoring of marginal respiratory status with secondary car-
diac stress may also be indications for ICU admission.
Postoperative Analgesia
Poor postoperative pain control can lead to increased cardiac
and pulmonary events. Aggressive pain control can decrease
these complications. This may be achieved with careful titra-
tion of narcotics, often with a PCA pump, although this still
may require careful choice and alteration of dosing by care-
givers. Unfortunately, due to the increased risk of obesity
and sleep apnea in the American patient population, a strictly
narcotic-based analgesia strategy will lead to increased epi-
sodes of respiratory depression. Indwelling epidural catheters,
running narcotic with or without dilute local anesthetic, can be
very effective for lower extremity, abdominal, or thoracic pain.
Anticoagulant considerations, as outlined earlier in the chapter,
as well as increased sensitivity to respiratory depression from
concomitant use of additional narcotics or sedatives require that
these be thoughtfully managed. Peripheral nerve catheters run-
ning local anesthetic, in which an indwelling catheter is laid next
to a peripheral nerve covering the surgical site (eg, femoral nerve
for knee surgery), allow analgesia with few systemic side effects.
Finally, utilization of nonopioid agents as adjuncts (eg, nonsteroi-
dal anti-inflammatory drugs, acetaminophen, ketamine, gaba-
pentin, magnesium), may also help to achieve analgesia without
increasing the risk of respiratory depression.
Abbreviations
ACC American College of Cardiology
ACE angiotensin-converting enzyme
AHA American Heart Association
CABG coronary artery bypass grafting
HCM hypertrophic cardiomyopathy
ICD implantable cardioverter-defibrillator
ICU intensive care unit
INR international normalized ratio
LAST localized anesthetic toxicity
LMWH low-molecular-weight heparin
MET metabolic equivalent
OR operating room
PCA patient-controlled anesthesia
TEE transesophageal echocardiography
Suggested Reading
American College of Cardiology Foundation/American Heart
Association Task Force on Practice Guidelines; American Society
of Echocardiography; American Society of Nuclear Cardiology;
Heart Rhythm Society; Society of Cardiovascular Anesthesiologists;
Society for Cardiovascular Angiography and Interventions; Society
for Vascular Medicine; Society for Vascular Surgery, Fleisher LA,
Beckman JA, Brown KA, Calkins H, Chaikof EL, Fleischmann KE,
et al. 2009 ACCF/AHA focused update on perioperative beta block-
ade incorporated into the ACC/AHA 2007 guidelines on periopera-
tive cardiovascular evaluation and care for noncardiac surgery. J Am
Coll Cardiol. 2009 Nov 24;54(22):e13-e118.
These guidelines provide a detailed evidence based consensus state-
ment on preoperative assessment and management of the cardiac
patient presenting for noncardiac surgery.
Horlocker TT, Wedel DJ, Rowlingson JC, Enneking FK, Kopp SL,
Benzon HT, Brown DL, Heit JA, Mulroy MF, Rosenquist RW, Tryba
M, Yuan CS. Regional anesthesia in the patient receiving anti-
thrombotic or thrombolytic therapy: American Society of Regional
Anesthesia and Pain Medicine Evidence-Based Guidelines (Third
Edition). Reg Anesth Pain Med. 2010 Jan-Feb;35(1):64–101.
Evidence-based consensus guidelines for management of regional
anesthesia in the face of anticoagulation.
Neal JM, Bernards CM, Butterworth JF 4th, Di Gregorio G, Drasner K,
Hejtmanek MR, Mulroy MF, Rosenquist RW, Weinberg GL. ASRA
practice advisory on local anesthetic systemic toxicity. Reg Anesth
Pain Med. 2010 Mar-Apr;35(2):152–61.
These guideline describe the current knowledge of LAST, particularly
the cardiac components and recommendations for resucsictation.
 Section IX

Cardiomyopathy and Heart Failure

This page intentionally left blank
 90
Reflex and Humoral Control of the Circulation
ALFREDO L. CLAVELL, MD
Optimal regulation of the circulation is dependent on an integra-
tion of cardiovascular reflexes with local and circulating humoral
factors that regulate myocardial contractility, vascular tone, and
intravascular volume, while intravascular volume is regulated
primarily through renal sodium excretion. Under physiologic
conditions, cardiovascular reflexes affect short-term cardiovas-
cular control, whereas humoral mechanisms function as long-
term modulators of cardiovascular homeostasis.
Cardiovascular Reflexes
Two principal cardiovascular reflex arcs are involved in the reg-
ulation of blood pressure:
1. Arterial baroreceptors are located in the carotid sinus and aortic
arch; they respond with increasing neural discharge in response to
stretch caused by increases in arterial blood pressure.
2. Cardiopulmonary baroreceptors are located in the ventricular myo-
cardia and also in the atria and venoatrial junctions.
Normal Cardiac Function
The neurons of the arterial and cardiopulmonary reflexes dis-
charge during cardiac systole; their rate of discharge is directly
related to the force of myocardial contraction and to cardiac fil-
ling pressure. Afferent signals from both arterial and cardiopul-
monary receptors travel to the nucleus solitarius in the brainstem.
The principal functions of these receptors are 2-fold:
1. To inhibit efferent sympathetic neural outflow to the heart and vessels,
decreasing arterial blood pressure and systemic vascular resistance.
2. To augment efferent parasympathetic neural outflow to the heart,
slowing the activity of the sinus node and prolonging atrioventricu-
lar conduction.
When arterial pressure and cardiac filling pressures decrease
under physiologic conditions, the inhibitory discharge of these
Abbreviations and acronyms are expanded at the end of this chapter.
receptors decreases. Efferent sympathetic neural outflow
increases, increasing systemic vascular resistance, and efferent
parasympathetic outflow decreases, resulting in tachycardia.
Conversely, during increases in arterial blood pressure and car-
diac filling pressures, the inhibitory discharge of these recep-
tors is enhanced. Efferent sympathetic neural outflow decreases,
decreasing systemic vascular resistance, and parasympathetic
outflow increases, resulting in bradycardia.
Congestive Heart Failure
In chronic CHF, chronic reduction in arterial filling decreases
the inhibitory signaling to the cardiovascular reflex center and
increases systemic vascular resistance. Despite high cardiac
filling pressures due to ventricular dysfunction, the inhibitory
action of the cardiopulmonary baroreceptors is attenuated. The
dysfunction of cardiovascular reflexes in CHF enhances adren-
ergic activity with systemic vasoconstriction. Additionally,
sympathetic activation may have secondary actions and lead to
activation of local and neurohumoral systems (such as the renin-
angiotensin system) and to greatly increased sodium retention
due to increased sodium resorption by the kidney.
• Arterial baroreflexes are located in the carotid sinus and aortic
arch; they respond to increases in arterial blood pressure.
• Dysfunction of cardiovascular reflexes in CHF enhances adrener-
gic activity with systemic vasoconstriction.
Local and Circulating Humoral Systems
Vasodilatory, Natriuretic, and Antimitogenic
Systems
Natriuretic Peptides
The natriuretic peptide system encompasses a family of car-
diovascular peptides: ANP and BNP are of cardiac myocyte
839
840 IX Cardiomyopathy and Heart Failure
origin, whereas CNP is of endothelial cell origin. These pep-
tides are released in response to both acute and chronic atrial
stretch (ANP and BNP) and in response to numerous other
humoral stimuli (CNP). They have important actions on the
heart, through autocrine and paracrine mechanisms, and on
other organs such as the kidney, adrenal gland, and vascular
wall (Figure 90.1). Important biologic actions include modula-
tion of myocardial function and structure, natriuresis, inhibition
of the renin-angiotensin-aldosterone system, vasodilatation, and
an antimitogenic effect on vascular smooth muscle cells. CNP
is devoid of natriuretic actions but is a powerful vasodilatory
and antimitogenic peptide. Components of this important cardi-
ovascular humoral system activate specific particulate guanylate
cyclase receptors, which are involved in catalyzing the synthesis
of a second messenger, cyclic guanosine monophosphate. The
activity of this system is modulated by 2 pathways involved in
clearance and degradation of the natriuretic peptides, including
neutral endopeptidase and a unique receptor-based clearance
mechanism (Figure 90.2).
In chronic CHF, circulating levels of ANP and BNP are
increased. ANP and BNP have functional significance in the
overall regulation of the cardiovascular system in CHF as evi-
denced by their inhibition with unique receptor antagonists in
experimental animal models of heart failure. The result is a rapid
deterioration manifested by rapid activation of the renin-angio-
tensin-aldosterone system together with vasoconstriction and
sodium retention (Box 90.1).
The increased levels of natriuretic peptides in heart failure
are important for both prognosis and diagnosis of early asymp-
tomatic left ventricular dysfunction. In particular, BNP has been
ANP
BNP
ET inhibition
Vasodilation ANP
BNP
CNP
ANP
BNP
Antiproliferation
effect
Figure 90.1. Natriuretic Peptide System. ANP indicates atrial natriuretic peptide; BNP, brain natriuretic peptide; CNP, C-type natriuretic peptide;
ET, endothelin.
recognized as a marker for left ventricular dysfunction and hyper-
trophy (Box 90.2). Therapeutic strategies have emerged that
potentiate the endogenous natriuretic peptides through inhibiting
their degradation by neutral endopeptidase and through exoge-
nous intravenous and subcutaneous administration of natriuretic
peptides. However, a clinical trial that used a peptidase inhibitor
had disappointing results.
• In chronic CHF, ANP and BNP levels are increased.
Endothelium-Derived Relaxing Factor (NO)
In addition to the natriuretic peptides, the endothelial cell–
derived relaxing factor, NO, is involved in the activation of cyclic
guanosine monophosphate through stimulation of soluble gua-
nylate cyclase. This endogenous factor causes vasodilation and
natriuresis and inhibits vascular proliferation. Indeed, inhibition
of endogenous NO by unique inhibitors results in systemic, renal,
and pulmonary vasoconstriction and sodium retention. Long-
term inhibition of the endogenous NO system results in hyper-
tension and ventricular and vascular remodeling.
NO synthetases are responsible for NO production; several
isoenzymes have been identified. At the level of the endothe-
lium, the production and function of NO appear to be impaired in
CHF. Other factors such as cytokines, free radicals, and changes
in handling of cellular calcium contribute to the apparent dys-
function of the NO system in CHF. However, some studies sug-
gest that NO activity is enhanced in human and experimental
animal heart failure because inhibition of NO generation in heart
failure results in further ventricular dysfunction and systemic
vasoconstriction.
Sympathoinhibitory
Antifibrotic
Lusitropic
ANP
BNP
ANP
BNP
Aldosterone
inhibition
Natriuresis
Renin inhibition
 90 Refl ex and Humoral Control of the Circulation 841

NPR-A Guanylyl cyclase-A

ANP/BNP
GTP cGMP

NPR-B
Biologic
effects
CNP
GTP cGMP
Neutral
endopeptidase
NPR-C

ANP>CNP>BNP
Clearance
receptor

Figure 90.2. Natriuretic Peptide Hormone Binding and Clearance. ANP indicates atrial natriuretic peptide; BNP, brain natriuretic peptide;
cGMP, cyclic guanosine monophosphate; CNP, C-type natriuretic peptide; GTP, guanosine triphosphate; NPR, natriuretic peptide receptor.

• NO is involved in the activation of cyclic guanosine monophos- by the kidney. The sympathetic, renin-angiotensin-aldosterone,
phate through stimulation of soluble guanylate cyclase. and endothelin systems are 3 important vasoconstrictor, anti-
• The clinical significance of NO activity in CHF is unclear. natriuretic, and mitogenic systems that control cardiovascular
homeostasis and have a role in the pathophysiology of CHF.
Vasoconstrictor, Antinatriuretic, and
Mitogenic Systems Sympathetic Nervous System
Endocrine mechanisms modulate vascular tone, growth of car- Plasma catecholamines (norepinephrine and epinephrine) are
diac myocytes and vascular smooth muscle, and sodium excretion the circulating humoral counterparts of the sympathetic nerv-
ous system. Norepinephrine is released locally from sympa-
thetic nerve endings adjacent to myocardium and modulates
myocardial contractility. The adrenal medulla also releases
Box 90.1. Neurohumoral Mechanisms in Congestive both catecholamines in response to diverse stimuli and ampli-
Heart Failure fies the cardiovascular response to sympathetic nervous system
Vasodilatory, natriuretic, and antimitogenic
activation. The myocardium is rich in β-receptors, which are the
factors targets of these cardiovascular hormones.
In chronic CHF, the sympathetic nervous system is acti-
Natriuretic peptides
vated in response to the reduction in myocardial contractility
Kallikrein, kinins
Prostaglandin
Dopamine Box 90.2. Causes of Increased Levels of Brain
Natriuretic Peptide
Endothelium-derived relaxing factor (nitric oxide)
Left ventricular hypertrophy
Adrenomedullin
Myocarditis
Vasoconstrictive, antinatriuretic, and mitogenic
factors Cardiac allograft rejection
Renin-angiotensin-aldosterone system Kawasaki disease
Sympathetic nervous system Primary pulmonary hypertension
Vasopressin Renal failure
Thromboxane
Ascitic cirrhosis
Endothelin
Cushing disease
Cytokines
Primary hyperaldosteronism
Previously published. See “Credit Line” section. Advanced age
842 IX Cardiomyopathy and Heart Failure
and cardiac output. Although the resultant vasoconstriction and
increase in myocardial contractility are essential for maintain-
ing blood pressure, eventually this response becomes deleterious
and contributes to a further decline in myocardial function. In
the presence of chronically increased serum norepinephrine lev-
els, there is down-regulation of myocardial β-receptors, perhaps
as a protective mechanism. Circulating levels of norepinephrine
correlate with patient mortality in CHF.
β-Adrenergic blockade is an important strategy in the ther-
apeutic neurohumoral modulation of CHF. Studies have dem-
onstrated a paradoxic increase in left ventricular function with
β-blockers, improved clinical symptoms, and better prognosis in
heart failure, regardless of the cause of the CHF and in addi-
tion to angiotensin-converting enzyme inhibition. Additionally,
studies have suggested that nonselective β-blockade is superior
to selective β1-blockade in the management of CHF. In fact, the
mortality benefit from β-blockade appears to be superior to the
benefit from angiotensin-converting enzyme inhibition.
• CHF is accompanied by chronic activation of norepinephrine and
down-regulation of myocardial β-receptors.
• Treatment with β-blockade improves left ventricular function,
clinical symptoms, and prognosis in patients with CHF.
Renin-Angiotensin-Aldosterone System
Angiotensin II is one of the most potent vasoconstrictor and
mitogenic peptides that is produced both systemically and locally
in the heart, lung, kidney, and vascular endothelium because of
the abundance of angiotensin-converting enzyme (Table 90.1).
Angiotensin II and its specific receptor subtypes are responsible
for stimulation of norepinephrine release and sympathetic activa-
tion. Metabolism and growth in myocyte and nonmyocyte cells
are altered by circulating and locally generated angiotensin II,
which increases cellular proliferation and impairs myocyte
contractile activity. Additionally, aldosterone produced by the
adrenal gland is activated by angiotensin II and has effects on
nonmyocytes in addition to its sodium-retaining action in the
kidney.
Studies have suggested that aldosterone may be responsible
for cardiac fibrosis through specific receptors within the heart.
Table 90.1. Angiotensin II Sites and Actions
Target Action
Heart Positive inotropism, hypertrophy
Kidney Renin release, mesangial contraction,
sodium resorption
Adrenal body Aldosterone release
Brain Vasopressin release, thirst, increased
sympathetic outflow
Sympathetic nervous system Norepinephrine release
Vascular smooth muscle Vasoconstriction, hypertrophy
Previously published. See “Credit Line” section.
These 2 important hormones, angiotensin II and aldosterone,
have emerged as targets for pharmacologic inhibition in the treat-
ment of CHF; in severe human CHF, inhibition of angiotensin II
generation has improved mortality and morbidity. However,
chronic escape from angiotensin-converting enzyme inhibition
has been noted, and newer strategies relying on angiotensin II
receptors and aldosterone antagonists have proved useful in
managing refractory heart failure.
• Angiotensin II and its specific receptor subtypes are responsible for
stimulation of norepinephrine release and sympathetic activation.
Endothelin System
Endothelin is a 21-amino acid vasoconstrictor peptide that
is produced by the vascular endothelium. Endothelin is syn-
thesized as an inactive molecule, known as big ET-1, that is
initially cleaved to pro-ET and then, through several further
modifications, to the biologically active isoforms of endothelin:
ET-1, the predominant vascular isoform (which is a potent vaso-
constrictor); ET-2, present only in the kidney and intestine,
where it also functions as a vasoconstrictor; and ET-3, which
has minimal vasoconstrictor properties. Endothelins are rap-
idly cleared, short-acting vasoconstrictors that bind to 2 major
types of endothelin receptors, namely, ET-A and ET-B, that are
present on many cells throughout the vascular system.
Endothelin most likely maintains vascular tone and arterial
blood pressure. In CHF, it functions in a compensatory mechan-
ism to mediate vasoconstriction and augment inotropic function.
Myocardial responsiveness to endothelin may be preserved in
late heart failure when the myocardium has become refractory to
other endogenous inotropic agonists.
Like angiotensin II, endothelin has growth-promoting and
mitogenic potential and may contribute to cardiac and vascular
remodeling. Endothelin stimulates renin and aldosterone release
and augments activation of cardiac fibroblasts. Endothelin also
has potent renal vasoconstricting and sodium-retaining actions
in CHF. Studies have suggested that an increase in plasma has
may have adverse prognostic implications in CHF. However, in a
randomized study of CHF patients, chronic blockade of endothe-
lin receptors resulted in no clinical benefit. Selective blockade of
endothlin-A receptors has proved efficacious in managing pul-
monary hypertension.
• An increase in plasma endothelin has adverse prognostic implica-
tions in CHF.
• Blockade of endothelin receptors is a proven strategy in managing
pulmonary hypertension.
Abbreviations
ANP atrial natriuretic peptide
BNP brain natriuretic peptide
CHF congestive heart failure
CNP c-type natriuretic peptide
NO nitric oxide
 91
Systolic Heart Function
WAYNE L. MILLER, MD, PHD, and LYLE J. OLSON, MD
Cellular Aspects of LV Contraction
Microanatomy
The myocardium is composed of cardiac myocytes enveloped
in a dense extracellular matrix of collagen, the main structural
protein of the heart. Cardiac myocytes account for 70% to 75%
of the myocardium by cell volume but only 25% to 30% by cell
number. Cardiac myocytes contain myofibrils that are composed
of longitudinally repeating sarcomeres separated by Z bands
(thickened and invaginated portions of the surface membrane
called the sarcolemma). The sarcomeres occupy about 50%
of the mass of cardiac myocytes. Thin filaments, composed of
actin, are attached to each Z band and interdigitate with the thick
filaments, composed of myosin molecules. The thick and thin
myofilaments slide past one another in a ratchet-type mechan-
ism to generate force and shorten the myocyte. The myofilaments
maintain a fixed length throughout contraction. Mitochondria,
which compose about 20% of the cell volume, are the organelles
in which ATP is generated. They are located near the myofibrils
and just beneath the sarcolemma. Platelike folds, or cristae, pro-
ject inward from the surface membrane of the mitochondria and
contain the respiratory enzymes for energy production.
• Contractile sarcomeres occupy about 50% of the mass of cardiac
myocytes.
Excitation and Contraction Coupling
The coupling of cardiac excitation (an electrical event) and con-
traction (a mechanical event) is fundamentally molecular. The
sarcolemma is a thin phospholipid membrane that maintains
Abbreviations are expanded at the end of this chapter.
electrical polarization. The phospholipid bilayer acts as an ionic
barrier and maintains a relatively high intracellular concentra-
tion of potassium and low intracellular concentrations of sodium
and calcium (Figure 91.1).
Near the Z bands are wide invaginations (T tubules) of the sar-
colemma, the T system, which branch through the cell. Closely
coupled to but not continuous with the T system is the sarcoplas-
mic reticulum, a complex network of anastomosing membrane-
limited intracellular tubules that surround each myofibril and
have a critical role in the excitation-contraction coupling of the
heart muscle.
Troponin (which is composed of troponin C, I, and T) and
tropomyosin are regulatory proteins in the thin filaments. In the
absence of troponin and tropomyosin, the contractile proteins
actin and myosin are activated, requiring only the presence of
magnesium and ATP. When present, the regulatory proteins pre-
vent cross-bridge formation between myosin and actin. When
calcium binds to troponin C, the binding of troponin I to actin
is inhibited, causing a conformational change in tropomyosin,
so that tropomyosin enhances cross-bridge formation instead of
inhibiting it. Thus, calcium blocks an inhibitor of the interaction
between actin and myosin. The key element in the initiation of
contraction is the release of sarcoplasmic calcium. Depolarization
of the sarcolemma in the upstroke of the action potential opens
the ion channels that carry the inward calcium current, which in
turn triggers a release of the large stores of calcium in the sarco-
plasmic reticulum. With cellular depolarization, the myoplasmic
concentration of calcium increases and calcium is bound to tro-
ponin. After each cross-bridge sliding action is completed, the
myosin head releases its ATP breakdown products, binds another
ATP molecule, and detaches from the actin site. The myosin head
then returns to its original configuration and the cycle is repeated
(Figure 91.2).
843
844 IX Cardiomyopathy and Heart Failure

Figure 91.1. Regulation of Excitation-Contraction Coupling. The sarcolemma and sarcoplasmic reticulum modulate cytoplasmic calcium availa-
bility, and the troponin-tropomyosin complex regulates responsiveness to cytoplasmic calcium (Ca). AC indicates adenylate cyclase; ADP, adenosine
diphosphate; AMP, adenosine monophosphate; ATP, adenosine triphosphate; ATPase, adenosine triphosphatase; cAMP, cyclic adenosine mono-
phosphate; GI, guanine nucleotide-binding regulatory protein that inhibits adenylate cyclase; GS, guanine nucleotide-binding regulatory protein that
stimulates adenylate cyclase; H, hydrogen; K, potassium; Na, sodium; P, phosphorus. (Previously published. See “Credit Lines” section.)

Figure 91.2. Major Shifts of Calcium Ions During Myocyte Excitation-Contraction Coupling and Relaxation. The dots represent calcium ions, and
the positive and negative signs indicate the electrical charge across the membrane partitions. (Previously published. See “Credit Lines” section.)
 91 Systolic Heart Function 845

Relaxation is brought about by the active (ie, requiring ATP)

reuptake of calcium into the sarcoplasmic reticulum. Thus, cal- Translates
cium is essential to the excitation-contraction coupling, and when
the calcium concentration decreases to a critical point, contrac-
tion ceases.
• Troponin and tropomyosin are regulatory proteins in the thin
filaments.
• The key element in the initiation of cardiac contraction is the
release of sarcoplasmic calcium. Accordions
• The transmembrane calcium current does not directly cause car-
diac contraction but promotes release of sarcoplasmic calcium.

Mechanics of Contraction
The motion of the LV during contraction can be summarized in
the mnemonic TARTT. During systole, the LV translates (moves Rotates
from side to side), accordions (moves with the base and apex
approximating each other), rotates (moves about the long axis of
the LV), tilts (moves perpendicular to the long axis), and thick-
ens (Figure 91.3).
Myocardial fibers are spiraled around the central LV cavity.
The subendocardial and subepicardial fibers run largely parallel
to the long axis of the cavity, and the mid-wall fibers are mostly Tilts
perpendicular to the long axis (ie, circumferential). During ven-
tricular ejection, these fibers shorten and thicken, and as the size
of the LV cavity decreases circumferentially and longitudinally,
the inner surface shortens more than the external surface (as dic-
tated by geometry). Because the muscle mass remains constant,
an increase in wall thickness must occur.
During isovolumic LV contraction, the chordae tendineae Thickens
become tense, the mitral valve closes, and the ellipsoid LV
becomes more spherical. During LV ejection, with the opening
of the aortic valve, the longitudinal axis shortens by only about
10%, whereas the short-axis diameter shortens by about 25%, Figure 91.3. Mechanisms of Contraction and Motion of the Heart.
thus accounting for 80% to 90% of the normal SV. The motion of the left ventricle is summarized in the mnemonic TARTT.
Isovolumic contraction refers to the interval (about 50 ms)
between the onset of ventricular systole and the opening of the
semilunar (aortic and pulmonic) valves. For the valves to open, Preload
the LV pressure must exceed that in the aorta during diastole. Defining preload for the intact LV as the ventricular end-diastolic
Pressure in the aorta increases slightly just before the semilunar wall stress provides a direct analogy to the preload of the isolated
valves open, causing the incisura or dicrotic notch. Ventricular muscle strip, which in turn determines the resting length of the
ejection is the phase of ventricular systole (about 350–400 ms at sarcomeres. Increases in preload augment the SV as well as the
a normal heart rate) in which blood is ejected through the aortic extent and velocity of wall shortening. At a constant preload,
valve. The first phase (about 100 ms) is rapid, and then ejection there is an inverse relation between systolic wall stress and SV.
slows toward the end of ventricular systole. The increase in ven- Alterations in preload, operating through changes in end-
tricular pressure is more marked in the rapid phase. diastolic fiber length, are important determinants of the per-
formance of the intact ventricle and provide the basis for the
• Myocardial fibers are arranged spirally around the central LV
length-function curves of the intact ventricle. The ability to aug-
cavity.
ment preload provides a functional reserve to the heart in situ-
ations of acute stress or exercise. Preload is thus an important
Determinants of Contraction of the Intact LV factor in maintaining LV systolic performance in many disease
The mechanical determinants of cardiac function are preload, states (Figure 91.4).
afterload, contractility, and heart rate. When the intact heart is • The ability to augment preload provides a functional reserve to the
compared with isolated muscle, heart volume and pressure are heart in situations of acute stress or exercise.
analogous to muscle length and tension. The Starling law of the
heart describes a fundamental property of heart muscle: the
force of contraction at any given tension depends on the initial Afterload
muscle fiber length. This, in turn, depends on the ultrastruc- Afterload in the intact LV is the tension (force or wall stress) act-
tural disposition of thick and thin myofilaments within the sar- ing on the fibers of the LV after the onset of shortening. This is
comeres. After the classic experiments with isolated heart and primarily the arterial pressure and is a major determinant of SV.
muscle strips, preload, afterload, and contractility first became An abrupt increase in the impedance to LV ejection, when pre-
clinically useful terms. load is constant, causes a decrease in fiber shortening and LV SV.
846 IX Cardiomyopathy and Heart Failure

During normal ejection, the LV becomes smaller and its walls

thicken. Thus, despite an increase in aortic pressure during LV
ejection, the afterload or wall stress decreases during ejection. In
this situation, there are inverse relationships between afterload
(systolic pressure or wall stress) and SV, extent of wall short-
ening, and velocity of shortening. Normally, SV is maintained
despite a modest increase in arterial pressure by augmenting
LV end-diastolic pressure and volume; that is, the incremental
increase in afterload is met by an increase in preload. However,
in the diseased heart with little preload reserve (such as in heart
failure), the LV SV decreases. Also, even in the normal heart,
when there is relative hypovolemia (such as with sepsis or hem-
orrhage), the preload cannot increase sufficiently and an increase
in afterload reduces the SV (Figure 91.5). Table 91.1 shows LV
loading in disease states.

Contractility
The term contractility has been used synonymously with ino-
tropic state. It is difficult to define in a quantitative sense because
there is no clear-cut single measurement of contractility that
provides a numeric value that can be assigned to a given heart.
However, when loading (preload and afterload) conditions remain
constant, an improvement in contractility augments cardiac per-
formance, whereas a depression in contractility decreases car-
diac performance. Inotropic influences generally act through
altered calcium availability to the myofilaments or through

Figure 91.5. Factors Affecting Left Ventricular (LV) Afterload.

PVR indicates peripheral vascular resistance.

altered myofilament calcium sensitivity. Additional factors that

directly or indirectly affect contractility are sympathetic neural
activity and circulating catecholamines (Figure 91.6).
The LV pressure-volume relationship is a convenient assess-
ment framework for understanding the responses of LV contrac-
tion to alterations in preload, afterload, and contractility. In its
simplest sense, the average circumferential wall stress (σ, the
force per unit of cross-sectional area of wall) in the intact heart
is the product of intraventricular pressure (P) and the internal
radius of curvature of the chamber (a) divided by the thickness
of the muscle walls (h × 2). The Laplace law for a spherical
chamber is

σ = Pa/2h

Figure 91.4. Factors Affecting Myocardial Stretch and Left
Ventricular (LV) Preload. LVEDP indicates left ventricular end-
diastolic pressure; LVEDV, left ventricular end-diastolic volume.
Heart Rate
Increasing the frequency of contraction does not shift the ven-
tricular performance curve relating LV end-diastolic pressure
and stroke work, but it does increase stroke power at any given
level of filling pressure. Thus, increasing the heart rate improves
myocardial contractility because the systolic fraction of the car-
diac cycle is increased. The positive inotropic effect resulting
from an increase in heart rate is more prominent in the depressed
heart than in the normal heart. In the normal heart, an artificial
 91 Systolic Heart Function 847

Table 91.1. Left Ventricular Loading in Disease States

Condition Preload Afterload Contractility Therapy
Sepsis ↓ →↓ →↓ Fluids and antibiotics
Dehydration ↓ → → Fluids
Heart failure ↑ → ↓ Diuretics
Cardiogenic shock ↑ → ↓ Inotropes
Intra-aortic balloon pump
RV infarct ↓ → → Increase intravenous fluids to maintain high RV filling pressure
Acute mitral regurgitation ↑ → → Intra-aortic balloon pump
Surgery, if severe
Aortic stenosis → ↑ → Surgery, if severe
Systemic hypertension → ↑ → Antihypertensive medications
Abbreviations: ↓, decreased; ↑, increased; →, no increase or decrease; →↓, same or decreased; RV, right ventricular.

increase in heart rate (such as with a pacemaker) will not increase
cardiac output because venous return to the heart is reflexly
and metabolically stabilized. However, if the diastolic volume
of the LV is increased by increasing venous return, as during
exercise, tachycardia is important in increasing cardiac output.
This assumes, however, that the speed is increased not only in
contraction but also in relaxation. This effect requires preserva-
tion of both systolic and diastolic functions. Of course, when the
heart rate is too fast (generally >180 beats per minute), the short
duration of diastole impedes ventricular filling and cardiac out-
put decreases (Box 91.1).

Myocardial Infarction
Left ventricular contractility:
Infarction of 30% or more of the LV mass results in a decrease
• Synonymous with inotropic state, but there
in LVEF. Initially, cardiac output is depressed; when damage to
is no single measurement of its value
the LV is considerable, function may deteriorate further, leading
• Generally mediated through altered Ca2+ to hemodynamic compromise and death. However, usually with
availability or altered myofilament Ca2+ adequate reserve, the cardiac SV is augmented by increases in
sensitivity ventricular preload within hours of the infarction. This change
• By definition, contractility is independent of is generally accomplished with an increase in LV end-diastolic
loading conditions pressure and is a direct consequence of the Starling law of the
heart. Increases in afterload may also accompany these changes
Contributing factors: and thus offset the increases in SV brought about by increased
preload. There are limits to preload reserve, beyond which fur-
Circulating Force-frequency Loss of ther increases in cardiac output must be from increased heart
catecholamines relation myocardium rate. This situation also occurs in patients with dilated cardiomy-
opathy and congestive heart failure. In those circumstances, the
Anoxia use of agents to reduce afterload may be beneficial in augmenting
Sympathetic Hypercapnia
nerve Contractility Acidosis
LV SV (Box 91.2).
traffic
• Myocardial infarction of ≥30% of the LV mass results in a decrease
in LVEF.
Pharmacologic Digitalis
depressants and other
inotropes

Ventricular Performance Box 91.1. Effect of Heart Rate on LV Systolic

Function
Positive inotropic effect—normally, ventricular
pacing has little effect on stroke volume because
of reflex stabilization of venous return
In the overloaded LV (eg, in CHF), increases in
heart rate augment stroke volume (to a point)
In exercise, with increased venous return,
increases in heart rate are the major contributors
to increased cardiac output; normally, the speeds
of LV contraction and relaxation are increased,
facilitating accommodation of the increased
venous return (up to 180–220 beats per minute,
LVEDP but much lower rates in CHF)

Figure 91.6. Determinants of Left Ventricular Contractility. Ca2+ Abbreviations: CHF, congestive heart failure; LV, left ventricular.
indicates calcium ion; LVEDP, left ventricular end-diastolic pressure.
848 IX Cardiomyopathy and Heart Failure
Box 91.2. Effect of Loss of Myocardium on LV
Systolic Function
Infarction (≥30% of LV mass), fibrosis, infiltration,
and myopathies all reduce LV systolic
performance
Generally, preload reserve (Starling law) can
assist in augmentation of stroke volume
In some circumstances, reflex and intrinsic
regulatory humoral factors may pathologically
increase SVR (increase afterload); when preload
reserve is exhausted, there is an afterload-
preload mismatch
Abbreviations: LV, left ventricular; SVR, systemic vascular resistance.
LV Hypertrophy
LV hypertrophy may occur in conditions of pressure and vol-
ume overload. Acquired disorders associated with a pathologic
increase in LV preload include chronic aortic or mitral valvular
regurgitation, dilated cardiomyopathy, and, often, myocardial
infarction. Disorders associated with a pathologic increase in
LV afterload include severe aortic stenosis and chronic systemic
hypertension. Although the changes in the sarcomeres are differ-
ent for these 2 overload states, in both situations the overall mass
of the LV is increased. Nevertheless, the hypertrophic response,
which is an important adaptive process that enables the heart to
compensate for overloading, is a complex process that is both
beneficial and detrimental to LV performance. The hypertro-
phied cells are not necessarily normal, and abnormalities in ino-
tropic response and vascular reactivity have been identified. LV
hypertrophy increases myocardial oxygen demand. Along with
changes in ventricular loading (primarily afterload) and heart
rate, LV hypertrophy is a major cause of increased myocardial
oxygen consumption (Box 91.3).
Physiologic Measures of LV Systolic Function
Ejection Fraction
The most commonly available measure of LV systolic perfor-
mance or contraction is EF. It is simply a ratio of the LV SV to
the LV EDV. It can be determined with various imaging meth-
ods (Box 91.4). Many of these methods rely on the assumption
Box 91.3. Effect of LV Hypertrophy on LV Systolic
Function
LV hypertrophy is common in chronic
pressure and volume overloading, in dilated
cardiomyopathy, and after myocardial infarction
(if ≥20% of LV is involved)
LV hypertrophy may assist in normalizing LV wall
stress but is a major component of myocardial
oxygen demand and can be associated with
reduced myocardial flow reserve (mechanism of
angina in aortic stenosis) and altered inotropic
responsiveness
Abbreviation: LV, left ventricular.
Box 91.4. Clinical Methods of Measuring LV Systolic
Function
Ejection fraction (EF) can be determined with
available imaging tools, but be cautious of
methods that rely on assumptions of LV geometry;
an acute increase in preload or decrease in
afterload increases EF, and vice versa
The velocity of circumferential fractional
shortening (VCF) is a better index of contractility
than the actual amount of shortening; it is
relatively insensitive to acute changes in preload
and is difficult to calculate clinically
Peak LV systolic emptying rate (PER) is a load-
dependent index of systolic function; use
angiography, RNA, or cine-CT
Abbreviations: cine-CT, cine-computed tomography; LV, left ventri-
cular; RNA, radionuclide angiography.
that the LV shape can be approximated by an ellipsoid. However,
the LV can be distorted by various diseases, and thus the accu-
racy of any measure of EF is dependent on the completeness of
the measurements. For instance, use of simple formulas derived
from 2-dimensional echocardiographic measures of end-systolic
and end-diastolic dimensions and estimates of LV long-axis
length can be erroneous if there are regional LV wall abnormali-
ties of contraction, such as after infarction. In these instances,
accurate measures of EF can be derived with radionuclide ven-
triculography (which does not require assumptions about ven-
tricular shape) or directly through quantifying EDV and ESV
with magnetic resonance imaging or electron beam computed
tomography. Thus,
EDV ESV SV
EF = =
EDV EDV
• The most commonly applied and clinically available measure of
systolic performance or contraction is EF.
Maximal Elastance
Another measure of LV contractility is Emax, which is based on the
linear relationship between pressure and volume at end-systole.
Stated another way, all end-systolic pressure-volume intercepts
form a straight line on the pressure-volume curve for a given
degree of contractility. The slope of this line is called the Emax.
As contractility increases, the slope of Emax increases; as contrac-
tility decreases, the slope of Emax decreases. Calculation of Emax
requires construction of pressure-volume curves and manipula-
tion of either preload or afterload (Figure 91.7).
Myocardial Relaxation
During myocardial relaxation, the myocardium returns to its
initial length and tension relationship. Cardiac relaxation is an
energy-dependent process that consumes high-energy phos-
phates. At the cellular level, abnormalities of calcium reuptake
may account for LV diastolic abnormalities and impaired relax-
ation. Relaxation also depends on systolic and diastolic loads and
the passive elastic characteristics of the ventricle.
 91 Systolic Heart Function 849

300
End-systolic Increased Normal
and isovolumic contractility
pressure-volume
240 curve (Emax) Heart failure
LV Pressure, mm Hg

180

120

Diastolic
15 pressure-volume
curve

SV

0 25 50 75 100 125
LV Volume, mL
Figure 91.7. Left Ventricular (LV) Pressure-Volume Curves. Maximal elastance (Emax) is a sensitive measure of LV function and is derived from
LV pressure-volume loops. SV indicates stroke volume.

Relaxation may be simplistically regarded as occurring during
the isovolumic relaxation period and as part of the rapid filling
period. If the ventricle can fully and quickly complete relaxation,
the ventricle rapidly expands and a large portion of blood flows in
from the left atrium to the LV after the mitral valve opens. However,
if there is a delay in the rate and duration of relaxation, the ven-
tricle continues to expand slowly, even after the mitral valve opens.
Thus, there will be a decrease in the rate of early rapid filling.
The term myocardial stiffness is used to distinguish changes
in the stiffness of each unit of muscle from changes in overall
chamber stiffness. Thickening of ventricular walls from any
cause (eg, LV hypertrophy) tends to increase both myocardial and
chamber stiffness. An increased volume-to-mass ratio is often
associated with increased chamber stiffness, whereas increased
chamber stiffness may also occur with a normal volume-to-mass
ratio, implying increased myocardial stiffness.

Ventricular Compliance
In mid and late diastole, pressure and volume increase, and the Abbreviations
passive diastolic properties of the ventricle (namely, chamber
ATP adenosine triphosphate
stiffness or its inverse, chamber compliance) can be assessed. LV
EDV end-diastolic volume
compliance is the change in volume per unit pressure as the LV EF ejection fraction
fills with blood from the left atrium. Thus, a decrease in compli- Emax maximal elastance
ance results in less blood entering the LV for a given increase in ESV end-systolic volume
pressure. Myocardial fibrosis from any cause increases ventricu- LV left ventricular
lar stiffness because collagen fibers are very rigid and virtually LVEF left ventricular ejection fraction
nondistensible at normal pressures. SV stroke volume
 92
Heart Failure With Preserved Ejection Fractiona
BARRY A. BORLAUG, MD
Introduction
HF affects nearly 6 million Americans and represents the lead-
ing cause of hospitalization among patients 65 years of age and
older. Of the total HF population, approximately half of patients
with HF have HFpEF. The prevalence of HFpEF relative to
HFrEF, or “systolic” HF, is growing at an alarming rate of 1%
per year. With the aging population characteristics observed in
developed countries, HFpEF will become the most common
form of HF in the next two decades. In contrast to HFrEF, where
abundant trial evidence is available to guide diagnosis and treat-
ment, little evidence-based data are available regarding optimal
management of HFpEF. HFpEF was historically referred to as
“diastolic” HF based upon the premise that diastolic dysfunc-
tion was the exclusive causal mechanism, but recent studies
have identified multiple nondiastolic mechanisms in the path-
ophysiology, making HFpEF the preferred term. A host of car-
diovascular and noncardiovascular disorders (eg, valvular heart
disease, constrictive pericarditis, infiltrative cardiomyopathy,
hypertrophic cardiomyopathy, pulmonary disease, anemia, and
deconditioning) may be confused with HFpEF but must be dis-
tinguished because of their unique treatments. While many of
the aforementioned processes produce symptoms of HF in the
presence of normal EF, for the purposes of this chapter, use of
the term HFpEF is restricted to patients wherein these specific
diagnoses have been excluded.
a
Portions previously published in From AM, Borlaug BA. Heart
failure with preserved ejection fraction: pathophysiology and emerging
therapies. Cardiovasc Ther. 2011 Aug;29(4):e6–21. Used with
permission.
Abbreviations and acronyms are expanded at the end of this chapter.
850
Epidemiology
There are differing opinions as to whether HFpEF and HFrEF
represent two distinct HF phenotypes or are simply polar ends
of one continuous HF spectrum. Proponents of the single-
disease hypothesis point toward the common presence of both
systolic and diastolic dysfunction in each form of HF, with more
severe systolic abnormalities in HFrEF and more prominent
diastolic impairments in HFpEF. In addition, symptoms, clini-
cal presentation, and functional limitation are similar in HFpEF
and HFrEF. However, population-based studies have shown a
bimodal rather than unimodal distribution of EF among HF
patients; patterns of ventricular and cardiomyocyte remodeling
differ in HFpEF and HFrEF, and responses to treatments such
as vasodilators or angiotensin antagonists appear to be categor-
ically different. Each of these findings provide strong evidence
in support of distinct HF phenotypes (Table 92.1). It has also
been suggested that HFpEF may progress to HFrEF, but there is
little evidence for this natural history in the absence of interven-
ing myocardial infarction.
Patients with HFrEF are more often male and more fre-
quently have a history of coronary disease or antecedent myo-
cardial infarction and ventricular conduction abnormalities. In
contrast, the dominant risk factors for HFpEF are older age,
female sex, and history of hypertension, obesity and/or diabe-
tes mellitus (Table 92.1). Women outnumber men by nearly 2
to 1 in HFpEF. The reasons for the disparity are not known
but may be related to fundamental sex-specific differences in
ventricular remodeling and vascular stiffening, as women have
greater predilection for concentric LV remodeling and reduced
aortic compliance, particularly with aging. The majority of
HFpEF patients in community-based studies are >60 years of
age (median, ∼74 years), and the presence of HF symptoms in
 92 Heart Failure With Preserved Ejection Fraction 851

Table 92.1. Similarities and Differences in Heart Failure With Preserved and Reduced
Ejection Fraction
HFpEF HFrEF
Age Usually >60 yrs Any
Sex predominance Women Men
Common comorbidities Hypertension, obesity, diabetes Coronary artery disease
mellitus
Ventricular structure Usually concentric LV remodeling Usually eccentric remodeling
or hypertrophy or hypertrophy
Dyspnea & fatigue ++ ++
Exercise intolerance ++ ++
Systolic dysfunction + ++
Diastolic dysfunction ++ ++
Secondary pulmonary hypertension ++ ++
Benefit from ACEIs & ARBs − ++
Benefit from β-blockers ? ++
Benefit from aldosterone ? ++
Benefit from ICD/CRT ? ++

Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; CRT, cardiac
resynchronization therapy; HFpEF, heart failure with preserved left ventricular ejection fraction; HFrEF, heart
failure with reduced ejection fraction; ICD, implantable cardioverter-defibrillator.

a younger patient with normal EF, particularly without other
risk factors, should raise suspicion for another cause such as
hypertrophic cardiomyopathy, valvular disease or pulmonary
arterial hypertension.
Noncardiovascular comorbidities are common in HFpEF
and are important contributors to outcome. Recent studies have
shown that noncardiovascular mortality is more prevalent in
HFpEF compared with HFrEF, and this may confound the abil-
ity of drugs targeting cardiovascular derangements to improve
outcome in clinical trials of HFpEF if the specified primary end
point is death. The presence of comorbidities such as diabetes
and renal disease in HFpEF have been associated with differ-
ences in LV mechanical properties, and spirometric abnormali-
ties indicative of mild airway obstruction were recently found
to increase the risk of incident HFpEF. Pulmonary disease fre-
quently coexists with HFpEF, and it can be extremely difficult to
separate the two contributors, especially when one considers that
pulmonary abnormalities could also be related to HF by affect-
ing gas exchange, pulmonary vasculature, or airway congestion.
Obesity is common in HFpEF and may contribute directly to
disease pathogenesis via effects on LV form and function. It may
be challenging, however, to discern breathlessness due to obesity
per se from an obese patient with dyspnea due to HFpEF.
Earlier studies suggested that outcomes were better in HFpEF
compared with HFrEF, but several recent studies have shown
that morbidity and mortality are similar in both forms of HF
(Figure 92.1). Among hospitalized HFpEF patients, between one-
quarter and one-third will be readmitted within 2 to 3 months,
and 5-year mortality rates range between 55%–74%. Of particu-
lar concern, improvements in outcome noted in HFrEF have not
been observed for HFpEF over the past 30 years. Mortality rates
and mode of death in the randomized trial databases may dif-
fer considerably compared with community-based studies, due
to entry criterion-based attrition of patients with older age and
greater comorbidity burden. Predictors of increased risk of death
in HFpEF are, for the most part, similar to those observed in
HFrEF; higher natriuretic peptide levels, older age, renal dys-
function, diabetes mellitus, increased heart rate, atrial fibrilla-
tion and lower EF are associated with increased risk of death
in HFpEF. Body mass displays a U-shaped relationship with
mortality in HFpEF, similar to what has been observed in HFrEF
and in patients with coronary artery disease.
Pathophysiology
Diastolic Dysfunction
The typical cardiac structural changes noted in HFpEF are an
exaggerated version of what is often seen with aging and chronic
arterial hypertension: increased arterial stiffening contributing
to chronic pressure overload, secondary concentric LV chamber
remodeling, and left atrial enlargement due to elevated diastolic
LV filling pressures. The latter also may lead to elevation in sec-
ondary pulmonary hypertension due to both left atrial hyper-
tension and secondary pulmonary vascular remodeling, similar
to that observed in HFrEF. The central pathophysiologic mech-
anism considered to produce elevation in filling pressures and
symptoms in HFpEF is diastolic dysfunction characterized by
elevated passive chamber stiffness and/or delayed myocardial
relaxation during early diastole (Figure 92.2).
Abnormalities in relaxation and chamber stiffness conspire to
impair diastolic filling of the ventricle or, at a minimum, require
pathologically elevated ventricular filling pressures to achieve an
adequate filling volume (preload). Elevation in LV filling pres-
sure is a universal finding in HFpEF—present either at rest or
with stress. Chamber stiffness, elevated in HFpEF at rest, may
increase further during dynamic exercise to lead to greater filling
pressure elevation. LV chamber stiffness varies as a function of
ventricular volume and pressure, as can be seen based upon the
curvilinear DPVR. If the LV is operating in the flat portion of the
DPVR there is little change in pressure for any change in volume.
Thus in this territory, the “operant” stiffness is low. However,
if the same LV were to fill to larger volumes, the increase in
pressure for any increase in volume would be much greater, thus
higher “operant” stiffness.
LV relaxation delay becomes highly prevalent during normal
aging and is thus not specific for HFpEF. Indeed, the presence of
mild diastolic dysfunction becomes more the rule than the excep-
tion in older patients. In population-based studies, the prevalence
of any diastolic dysfunction increases from 25% among subjects
 1.0

0.8

EF ≥50%
0.6

Survival
0.4
EF <50%

0.2

0.0
0 1 2 3 4 5
Years
No. at risk
EF <50% 2,424 1,637 1,350 1,049 813 604
EF ≥50% 2,166 1,539 1,270 1,001 758 574

Figure 92.1. Survival is similarly poor in HFpEF (red) compared with HFrEF (black) in population-based studies. HFpEF indicates heart fail-
ure with preserved left ventricular ejection fraction; HFrEF, heart failure with reduced ejection fraction. (Previously published. See “Credit Lines”
section.)

B
A High

LVEDP
LV Pressure

HFpEF DPVR
Systole Normal

Diastole LVEDP
Left Ventricular Pressure

Normal DPVR

LV Volume

High operant
LV Pressure

stiffness
Prolonged (large ΔP/ΔV)
Relaxation
Normal
Relaxation
Low operant
stiffness
(small ΔP/ΔV)
Time
LV Volume
Figure 92.2. A, With diastolic dysfunction, the rate of pressure decline during isovolumic relaxation (onset of diastole) decreases, such that
ventricular pressure takes longer to reach its minimum (dotted line). B, Passive LV stiffness is described by the curvilinear slope of the diastolic
pressure-volume relationship (DPVR). In healthy humans, the LV can fill with large volume of blood with minimal increment in pressure (flat DPVR,
straight line). In HFpEF, passive diastolic stiffness is increased, such that the DPVR is shifted up and to the left (dotted line). This results in higher
filling pressure (LV end-diastolic pressure [LVEDP]) for any filling volume. C, Even within the same patient, LV stiffness varies as a function of
where in the DPVR the LV is operating. When volumes are low, there is minimal change in pressure per change in volume, and operant stiffness
is low. Conversely, when the LV is filled to the steeper portion of the DPVR, operant stiffness is high, and further increases in LV volume increase
LVEDP dramatically. HFpEF indicates heart failure with preserved left ventricular ejection fraction; LV, left ventricular.
 92 Heart Failure With Preserved Ejection Fraction
aged >45 years to 87% between 60 and 69 years and 96% above
70 years. Mechanistically, impaired relaxation has minimal effect
on LV end-diastolic pressure and pulmonary venous pressures at
normal heart rates but may contribute to filling pressure eleva-
tion and inadequate enhancement in cardiac output during tach-
ycardia. Enhanced LV filling during exercise in healthy humans
involves suction of blood from left atrium to LV apex, such that
left atrial pressures remain stable or increase only slightly. This
suction is created by elastic recoil which can be quantified by the
extent and velocity of LV untwisting during early diastole—both
of which are impaired in HFpEF. In the clinical catheterization
laboratory, LV volumes are usually not measured clinically to
measure stiffness, and relaxation is also not directly measured.
Thus, demonstration of high LV filling pressure (either pul-
monary capillary wedge pressure or LV end-diastolic pressure
>15 mm Hg at rest or >25 mm Hg with exercise) in a patient with
normal LV size and normal EF is considered to be sufficient evi-
dence of HFpEF.
In clinical practice, diastolic function is most often assessed
by echocardiography, principally based upon observation of the
pattern of Doppler mitral inflow (Figure 92.3). Under normal
circumstances, the majority of blood enters the LV during early
diastole, being “pulled” into the apex via suction effects (see
above). This results in a tall E wave that corresponds to the pos-
itive pressure gradient between LA and LV during early filling.
After E wave filling, pressure in the LA and LV equalizes and
there is no transmitral flow. This period is termed diastasis. The
electrocardiographic P wave initiates atrial contraction, which
then serves to further enhance LV filling at the end of diastole.
Normally this creates a lower-amplitude “A wave” in the trans-
mitral Doppler inflow pattern. With mild diastolic dysfunction,
usually characterized by prolonged relaxation, the decay in pres-
sure during isovolumic relaxation is prolonged (Figure 92.1).
This reduces the amount of blood flowing in during early diastole
and increases reliance on filling during atrial systole. This results
in reversal in the E/A ratio. This pattern is normal in adults older
than 60 years of age. If relaxation is quite prolonged, more time
is required for the E wave to decelerate and this can be quantified
by a prolongation in the mitral E wave deceleration time.
With further progression in diastolic dysfunction, LA pres-
sure rises further and LV compliance becomes reduced further.
Normal
Diastolic
Function
DT >140 ms
0.75 <E/A <2
2.0
Velocity, m/s
E
Mitral Inflow A
E/e' <10
Velocity, m/s
Doppler Tissue 0
Imaging of Mitral
Annular Motion 1.5 a'
e'
Figure 92.3. Doppler Echocardiographic and Tissue Doppler Echocardiographic Classification of Diastolic Function. (Previously published. See
“Credit Lines” section.)
853
At this point, the greater pressure gradient from LA to LV over-
whelms the tendency for reduced E wave filling, and the E wave
velocity increases above the A wave. This is often termed the
pseudonormal mitral inflow pattern but can be distinguished
from normal by reduction in TDE velocities (see below) or by
observation of the mitral inflow during the Valsalva maneuver,
wherein the E/A ratio reverts with the reduction in venous return
during Valsalva strain. As diastolic function further deteriorates,
the ratio of the E wave to the A wave becomes progressively
greater, and the mitral E wave deceleration becomes very short
(Figure 92.3). This pattern is often termed the restrictive mitral
inflow pattern and is often associated with markedly elevated LV
filling pressures and increased operant LV stiffness. Restrictive
mitral inflow patterns are commonly confused with the clini-
cal entity of restrictive cardiomyopathy, but any patient with HF
(with reduced or preserved EF) can display a restrictive mitral
inflow pattern when LV operant stiffness and/or filling pressures
are markedly elevated.
TDE measures the velocity of motion at the mitral annulus as
opposed to the velocity of blood flow. Thus, as opposed to E/A
ratio where there is a U-shaped relationship between normal to
mildly abnormal to severely abnormal function, TDE early dias-
tolic velocity (e′) decreases continuously as diastolic relaxation
becomes more impaired. The ratio of mitral inflow E to TDE
e′ (E/e′) is frequently used clinically as a surrogate marker of
LV filling pressures, where larger values indicate increased pres-
sure. Some authorities have advocated the use of an elevated E/e′
ratio (>15) to positively diagnose HFpEF, whereas others have
questioned this practice. Other echocardiographic patterns asso-
ciated with diastolic dysfunction include abnormal pulmonary
venous inflow, high estimates of right ventricular systolic pres-
sure, and LA enlargement. No single measure should be relied
upon to make a clinical diagnosis of HFpEF, but in a typical
patient with signs and symptoms of HF the presence of several of
these abnormalities greatly increases the confidence that HFpEF
is indeed the diagnosis (Table 92.2).
Other Pathophysiologic Mechanisms in HFpEF
EF is the most common measure of systolic function used in prac-
tice, but because EF varies inversely with afterload it is a rather
Mild Moderate Severe
Diastolic Diastolic Diastolic
Dysfunction Dysfunction Dysfunction
E/A ≤0.75 DT >140 ms DT <140 ms
0.75 <E/A <2 E/A >2
E/e' <10 E/e' ≥10 E/e' ≥10
854 IX Cardiomyopathy and Heart Failure
Table 92.2. Common Echocardiographic Findings in HfpEF
Concentric LV remodeling or LV Elevated estimated right ventricular
hypertrophy systolic pressure (marker of high
LA enlargement left heart pressure)
Abnormal transmitral Doppler Abnormal pulmonary vein Doppler
flow pattern (particularly flow pattern
“pseudonormal” or “restrictive” Dilated IVC with reduced
mitral inflow pattern) inspiratory collapse
Increased TDE E/e’ ratio
(>15 or more)
Abbreviations: HFpEF, heart failure with preserved left ventricular ejection
fraction; IVC, inferior vena cava; LA, left atrial; LV, left ventricle; TDE, tissue
Doppler echocardiography.
crude estimate of LV contractile function. A number of studies
have identified abnormalities in regional systolic function using
TDE in HFpEF patients, even in the presence of EF >50%–55%,
and others have shown that LV chamber and myocardial con-
tractility are reduced in HFpEF despite preserved EF, with the
extent of systolic dysfunction being related to worse outcome.
These subtle impairments in resting systolic function in HFpEF
may become extremely limiting when the cardiovascular system
is stressed, as with physical exercise. Indeed, a number of stud-
ies have indicated that while resting EF is normal, EF does not
increase with stress in HFpEF compared with normal controls
(even in the absence of ischemia).
LV and aortic stiffness increase with age. In HFpEF, this
age-associated ventricular-arterial stiffening is exaggerated,
with important effects on reserve capacity, myocardial oxygen
demand, blood pressure regulation, and responses to vasodila-
tors. For example, with any change in arterial afterload the
patient with HFpEF will display a relatively greater drop in blood
pressure than a normal control. This is in striking contrast to the
patient with HFrEF, where high doses of vasodilators may be tol-
erated with little or no drop in blood pressure. Since the majority
of HFpEF patients are elderly, great care is required in the use of
medicines which alter afterload (eg, vasodilators) or preload (eg,
diuretics or nitrates) because they may cause hypotension with
associated risk of falls and subsequent trauma.
Chronotropic incompetence is very common in HFpEF, even
after adjusting for β-blocker use and age. Heart rate reserve is
strongly related to exercise capacity in HFpEF, and may become
an even more critical reserve mechanism when diastolic and
systolic reserves become impaired. The causes of chronotropic
incompetence in HFpEF remain unclear but may relate to auto-
nomic abnormalities. β-blockers and calcium channel blockers
should be used with caution in patients with HFpEF and chrono-
tropic incompetence.
Patients with HFpEF display inadequate arterial vasodila-
tion during exercise, and this is associated with reduced exer-
cise capacity. Abnormal vasodilatation may be related in part
to endothelial dysfunction, which is present in HFpEF and
HFrEF. The LA serves an important, if under-appreciated, role
in HFpEF; it acts as both a reservoir to store blood during ven-
tricular systole, and enhances LV filling via atrial contraction at
end diastole, which can be very important in patients with severe
diastolic dysfunction. Patients with HFpEF frequently display
LA enlargement, with depressed LA pump function. LA dila-
tion creates the mechanical substrate for the development of
atrial fibrillation, which affects about 35%–40% of patients with
HFpEF and is associated with greater morbidity and mortality.
Finally, pulmonary hypertension and anemia are both common
in HFpEF; both are associated with worse outcomes and are cur-
rently being evaluated as potential therapeutic targets, though
currently the guidelines do not advocate for the use of pulmonary
vasodilators in HF. In summary, the pathophysiology of HFpEF
is complex and involves multiple coexisting abnormalities that
contribute to produce signs and symptoms of HF (Figure 92.4).
Clinical Features
Clinical presentation, history, and physical findings in HFpEF
are, for the most part, indistinguishable from those of HFrEF.
Cardinal symptoms are those of all patients with HF and vary
with the severity of disease: exertional dyspnea and fatigability
in the early stages, progressing to dyspnea with minimal exer-
tion, orthopnea, and paroxysmal nocturnal dyspnea with more
advanced HF. Jugular distention, gallop sounds, and periph-
eral edema are similarly distributed among both forms of HF.
As in HFrEF, HFpEF may present as acute pulmonary edema,
gradually progressive volume overload, or simply exertional
dyspnea. Pulmonary edema is characteristically noted in the set-
ting of uncontrolled hypertension in HFpEF and may respond
favorably to improved blood pressure control and modest diu-
resis. Precipitators for HF decompensation in HFpEF are simi-
lar to HFrEF and include diet or medication indiscretion, atrial
fibrillation or other arrhythmia, thyroid disease, infection, renal
dysfunction, or acute ischemia; the cause of destabilization is
unidentified in one-half of patients. Many patients with HFpEF
operate within a very narrow window between volume over-
load and apparent hypovolemia and azotemia. This is related
to increased systolic and diastolic ventricular stiffness, together
with increased aortic stiffness.
A number of diseases may be difficult to distinguish upon
clinical presentation from garden variety HFpEF (Table 92.3),
and these must be carefully ruled out before the diagnosis is made
because treatments may be vastly different. If signs of severe
systemic venous congestion (eg, ascites, pulsatile hepatomegaly,
severe peripheral edema) are present that are out of proportion to
the left-sided findings, one should consider other causes of severe
diastolic dysfunction such as constrictive pericarditis or infiltra-
tive cardiomyopathy. Patients with high-output HF may be mis-
diagnosed as having HFpEF but have a fundamentally different
pathophysiologic mechanism and treatment. Examination clues
favoring high-output HF include bounding pulses, tachycardia,
early systolic flow murmurs, and a thrill or venous hum if an
arteriovenous malformation or fistula is present. The diagno-
sis of cardiac amyloid should be considered in the patient with
ventricular hypertrophy on echocardiography but low or normal
voltage on electrocardiography. Clues to the diagnosis of hyper-
trophic cardiomyopathy include ECG abnormalities, hypertrophy
in the absence of hypertensive history (especially with preferen-
tial involvement of the septal wall), absence of systemic venous
congestion, and a dynamic outflow murmur that increases with
Valsalva maneuver or moving from a squat to standing position.
Pulmonary arterial hypertension can sometimes be difficult to
discern from HFpEF because both types of patient present with
dyspnea and normal EF, but clues supporting HFpEF include
older age, comorbidities such as hypertension and diabetes, and
LA enlargement.
Diagnosis
HF is a probability-based, clinical diagnosis—there is no uni-
versally agreed-upon single diagnostic test that can completely
 92 Heart Failure With Preserved Ejection Fraction 855

A B

LV pressure (mm Hg)

LV pressure (mm Hg)

160 160
EDV Preload
LV volume

Reserve 120 120

Impaired
Stroke
SV SV SV SV Volume 80 80
Inotropic Reserve
& Vascular
ESV Reserve 40 40

0 0

Rest Exercise Rest Exercise

C D E
30 8
P=.002
Peak VO2 (mL/min*kg)

r=0.7, P<.0001 P=.002

150

Peripheral Arterial
Tonometry Ratio
25
Heart Rate (bpm)

6
20
4
100 15
2
10

50 5 0
Rest 50% Peak 0 50 100 30 60 90 120 150 180 210 240
Change in Heart Rate (bpm) Time After Occlusion Release (sec)

Figure 92.4. The Complex Pathophysiology of HFpEF is Revealed by Examining Exercise Reserve Responses. A, While LV EF (SV/EDV) is
normal at rest, EF increases with exercise less in HFpEF (blue) than controls (red). This may be related to impairment in recruitment of preload
(less increase in end-diastolic volume [EDV]), impaired enhancement in contractility or blunted vasodilation (less reduction in end systolic volume).
These effects reduce the augmentation in stroke volume (SV) during stress. B, In well-compensated HFpEF patients, LV filling pressures are normal
at rest, but during low-level exercise, dramatic elevations in LVEDP (arrow) may be observed. C, Increases in heart rate are reduced in HFpEF (blue)
compared to controls (red), and D, the increase in heart rate is closely associated with exercise capacity (peak oxygen consumption, VO2). E, Increases
in arterial blood flow in response to reactive hyperemia are decreased in HFpEF (blue) compared to controls (red), consistent with endothelial dys-
function. This may contribute to symptoms of dyspnea and fatigue and impair dynamic vasodilation during stress in HFpEF. EF indicates ejection
fraction; HFpEF, heart failure with preserved left ventricular ejection fraction; LV, left ventricular; LVEDP, left ventricular end-diastolic pressure.
(Previously published. See “Credit Lines” section.)

rule in or exclude HF. Various algorithms have been proposed without obvious congestion. A key positive finding supporting
to diagnose HFpEF, but the central components are 1) clinical the diagnosis of HFpEF is elevated filling pressures, which can
symptoms compatible with HF (dyspnea, fatigue) with objec- be estimated based upon physical examination findings, radiog-
tive evidence of 2) cardiac dysfunction (eg, elevated cardiac raphy, echocardiography, or elevated natriuretic peptide levels
filling pressures or low cardiac output at rest or with exercise), (Table 92.4). Echocardiography plays a key role in diagnosis
and 3) normal EF (typically ≥50%). Component 1 is satisfied by and evaluation of patients with HFpEF as previously described,
observing 2 major Framingham criteria or 1 major and 2 minor and all suspected patients should undergo echocardiography
Framingham criteria for HF diagnosis (Figure 92.5). with assessment of diastolic function. Common findings include
As previously stated, a number of cardiovascular diseases concentric LV remodeling or hypertrophy, LA enlargement, and
can produce HF symptoms in the presence of a normal EF elevated Doppler-estimated pulmonary artery pressures (often
(Table 92.3), and each of these must be carefully considered related to high left heart pressures). The E/e′ ratio is the most
and ruled out prior to establishing the diagnosis of HFpEF. widely embraced noninvasive measure of diastolic dysfunction
Myocardial ischemia in particular causes acute diastolic dys- and/or elevated filling pressures.
function, and evaluation for coronary disease should be strongly Even though HFrEF is associated with more dramatic cham-
considered in all patients with normal EF and symptoms of HF, ber dilation, patients with HFpEF display the same degree of car-
especially if chest discomfort is also reported. Finally, many diomegaly on chest film. Pulmonary edema or venous congestion
patients with pulmonary disease may present with similar symp- is useful if present but not helpful to exclude HFpEF if absent.
toms, and pulmonary function testing may help to rule in or ECG may show LV hypertrophy, LA enlargement or atrial fibril-
exclude pulmonary disease. It should, however, be remembered lation, though none of these findings are specific for HFpEF.
that some abnormalities (eg, reduced diffusion capacity for car- Laboratories may show elevated BNP levels, but BNP levels
bon monoxide) can also be observed with HF. are lower in HFpEF compared with patients with reduced EF,
In contrast to HFrEF, where echocardiography establishes and normal BNP (or NT-proBNP) levels do not exclude HFpEF.
the cause of symptoms readily, the diagnosis of HFpEF is Similar to patients with reduced EF, patients with HFpEF may
cumbersome, especially in patients presenting in an outpatient display anemia, hyponatremia, and renal dysfunction. Troponin
clinic with exertional dyspnea and multiple comorbidities but elevation is uncommon in typical presentations of HFpEF, and its
856 IX Cardiomyopathy and Heart Failure

Table 92.3. Diseases Commonly Confused With Heart Table 92.4. Factors Increasing the Probability of HFpEF in
Failure With Preserved Left Ventricular Ejection Fraction Patients With Unexplained Dyspnea
Cardiovascular Typical demographics (age ≥60 years, female, hypertensive, obese,
Hypertrophic cardiomyopathy diabetic)
Infiltrative or restrictive cardiomyopathy (eg, amyloid, sarcoid, Fabry Examination, ECG, and CXRa findings (jugular distention, gallop sounds,
disease) peripheral edema, cardiomegaly, LV hypertrophy, atrial fibrillation)
Idiopathic pulmonary arterial hypertension Typical cardiac structural alterations (left atrial enlargement, concentric
Constrictive pericarditis LV hypertrophy, or remodeling)
High-output heart failure (eg, arteriovenous fistula, thyrotoxicosis, or Paget LV echocardiographic diastolic dysfunction (elevated E/e’ ratio, elevated
disease) pulmonary artery systolic pressure)
Valvular heart disease Elevated BNP or NT-proBNPa
Coronary artery disease Abnormal cardiopulmonary exercise test (decreased peak oxygen
Pulmonary embolism consumption, decreased ventilatory efficiency)
Right ventricular myopathies (eg, arrhythmogenic right ventricular Clinical improvement with diuretics or worsening with atrial fibrillation
dysplasia) Cardiac catheterization (increased LV filling pressures at rest or with
Noncardiovascular exercise, decreased cardiac output reserve)
Pulmonary disease
Abbreviations: BNP, brain natriuretic peptide; CXR, chest radiographic; ECG,
Anemia
electrocardiographic; HFpEF, heart failure with preserved left ventricular
Obesity ejection fraction; LV, left ventricle; NT-proBNP, N-terminal prohormone of
Deconditioning brain natriuretic peptide.
Renal artery stenosis a
These findings are often absent in early stage (euvolemic) HFpEF.
Thyroid disease
Neuromuscular disease
treatment in HFpEF. Standard HF medications such as ACEIs,
ARBs, and digoxin have not been found to be effective in
presence should trigger consideration of coronary artery disease, HFpEF, though they are still prescribed. Other treatments such
infiltrative cardiomyopathy (eg, amyloid) and myocarditis. as β-blockers, aldosterone antagonists, and statins require further
In general, if patients meet the Framingham criteria and show study. Devices (eg, pacemakers, defibrillators, and resynchroni-
several of the typical Doppler echocardiographic findings, fur- zation devices) have not been evaluated in HFpEF and have no
ther diagnostics are not required. When the results are equivo- indication in the absence of other specific indications (eg, heart
cal from examination and echocardiography, additional testing block or resuscitated sudden cardiac death).
is required, such as cardiopulmonary exercise testing, pulmo- The PEP-CHF study was the first major randomized-con-
nary function tests, and cardiac catheterization.
. Patients with trolled trial on the use of ACEIs in HFpEF patients, comparing
HFpEF typically display reduced VO2 with a plateau suggesting perindopril 4 mg daily to placebo in 850 elderly patients with
impaired cardiac output reserve. Exercise echocardiography may HF and EF>40%. There was no reduction in the primary end
show findings suggestive of high filling pressures, but these tests point of all-cause mortality or HF hospitalization at 3 years of
have not been fully validated. Invasive catheterization remains follow-up, though a trend toward lower event rates at 1 year was
the standard diagnostic test for HFpEF. Elevation in LV filling noted, and improvements in 6-minute walk distance and NYHA
pressures (wedge pressure or LV end-diastolic pressure) at rest or class were observed. The CHARM-Preserved trial randomized
during exercise in the absence of acute ischemia or other second- 3,023 patients with HF and EF>40% to candesartan or placebo.
ary cause provides stand-alone evidence of HFpEF. After a median follow-up 36 months, treatment with candesartan
was associated with a nonsignificant reduction in the composite
end point in mortality and cardiovascular hospitalizations. The
Treatment
largest study to date in HFpEF, the I-PRESERVE trial, assigned
In contrast to the wealth of treatments that unequivocally improve 4,128 patients with HF and an EF>45% to irbesartan or placebo.
morbidity and mortality in HFrEF, there is no established After 4 years of follow-up there was no difference in death or car-
diovascular hospitalizations, with consistent absence of benefit
across all subgroups. An ancillary trial of the DIG trial found no
Major Criteria Minor Criteria benefit to digoxin in HFpEF, though cardiac glycosides remain
a useful adjunct to control the ventricular response in HFpEF
- PND - Peripheral edema patients with atrial fibrillation.
- Orthopnea - Night cough By slowing heart rate, β-adrenergic antagonists may allow for
- Elevated JVP - DOE a longer diastolic filling period, and it has been proposed that this
- Rales - Hepatomegaly may be effective in HFpEF. Unfortunately, adequate prospec-
- S3 - Pleural effusion tive trial data regarding β-blockers in HFpEF are not currently
- CXR cardiomegaly - HR >120/min available. The SENIORS Trial, enrolled patients aged ≥70 years
- CXR pulm edema - Wt loss ≥4.5 kg in with both reduced and preserved EF and demonstrated the ben-
5 days with diuretic efit of nebivolol versus placebo in the primary outcome of mor-
tality and cardiovascular hospitalization. However, there were
Validated CHF if 2 major or 1 major and relatively few patients with truly “normal” EF (>50%) in this
2 minor are present concurrently
trial, and nebivolol is highly β-1 specific and has additional nitric
Figure 92.5. Framingham Criteria for Diagnosis of Congestive oxide-dependent vasodilating properties, so the SENIORS find-
Heart Failure. Notably, many patients with HFpEF may present in a non- ings may not apply to other commonly used β-blockers. In the
congested state and thus not fulfill these criteria at all points in time. OPTIMIZE-HF registry, discharge use of β-blockers and ACEIs
 92 Heart Failure With Preserved Ejection Fraction 857

were not associated with any reduction in 1-year mortality or fluid content. In these patients, use of diuretics can precipitate
hospitalization rates of HFpEF patients, though each signifi- hypotension and should be avoided. Nitrates may be useful as
cantly improved both end points in HFrEF patients. Collectively, an alternative, reducing cardiac filling pressures by venodilation
the evidence to date indicates that established treatments in without promoting hypovolemia. Atrial systole may play a more
HFrEF are not effective in HFpEF, though more data is needed important role to maintain adequate LV filling volume in patients
for β-blockers and aldosterone antagonists. with HFpEF, and atrial fibrillation is a common precipitator of
In the absence of trial-based data, recommendations are decompensation. The ACC/AHA guidelines recommend consid-
driven by expert consensus opinion. The 2009 ACC/AHA eration of cardioversion for patients with atrial fibrillation, and
guidelines recommend control of blood pressure, control of the adequate rate control is certainly essential.
ventricular response in atrial fibrillation, consideration of car- It must be remembered that clinical outcome in HFpEF is
dioversion for atrial fibrillation, and evaluation for ischemia in also driven importantly by comorbidities, which are increased
appropriate patients (Table 92.5). Note that each of these recom- in this aged population, and interventions of proven efficacy
mendations carries level of evidence “C,” indicating the lack of should be prescribed for all comorbid medical conditions in
clinical trial-supported benefit. Specific drug classes prescribed HFpEF, including diabetes, coronary disease, sleep-disor-
for blood pressure or heart rate control should be based upon dered breathing, and chronic obstructive pulmonary disease.
observed efficacy for in the individual patient and other estab- Occasionally, patients that present with recurrent hypertensive
lished disease-specific indications (eg, ACEIs for patients with pulmonary edema are found to have occult renal artery steno-
atherosclerosis or renal disease, β-blockers in patients after myo- sis, in which case their “heart failure” can be cured by renal
cardial infarction). Diuretics play a key role in management of revascularization. Finally, a recent randomized trial found that
fluid overload and carry a class I, level C recommendation. In 16 weeks of supervised exercise training was associated with
an ancillary analysis from the ALLHAT study, the thiazide diu- significant enhancement in functional capacity in older adults
retic chlorthalidone was found to be associated with lower rates with HFpEF, but cardiac rehabilitation is not consistently reim-
of incident HFpEF compared with lisinopril and amlodipine, bursed for the indication of heart failure, and the use of exercise
though this drug has not been tested in patients with prevalent training for HFpEF is not specifically dealt with in the most
(ie, already diagnosed) HFpEF. Many patients with HFpEF recent guidelines.
develop elevation in cardiac filling pressures exclusively dur-
ing exercise stress, in the absence of an increase in total body
Abbreviations
ACC American College of Cardiology
Table 92.5. ACC/AHA Recommendations for Treatment ACEI angiotensin-converting enzyme inhibitor
of Patients With Heart Failure and Normal Left Ventricular AHA American Heart Association
Ejection Fraction ARB angiotensin receptor blocker
BNP brain natriuretic peptide
Recommendation Class Level of Evidence DPVR diastolic pressure-volume relationship
ECG electrocardiographic
Physicians should control systolic and I A
EF ejection fraction
diastolic hypertension, in accordance
HF heart failure
with published guidelines
HFpEF heart failure with preserved left ventricular
Physicians should control ventricular rate I C
ejection fraction
in patients with atrial fibrillation
HFrEF heart failure with reduced ejection fraction
Physicians should use diuretics to control I C
LA left atrial, left atrium
pulmonary congestion and peripheral
LV left ventricular, left ventricle
edema
NT-proBNP N-terminal prohormone of brain natriuretic
Physicians might recommend coronary IIa C
peptide
revascularization in patients with
NYHA New York Heart Association
coronary artery disease in whom
TDE
. tissue Doppler echocardiography
symptomatic or demonstrable
VO2 peak oxygen consumption
myocardial ischemia is judged to be
having an adverse effect on cardiac
function Names of Clinical Trials
Restoration and maintenance of sinus IIb C
rhythm in patients with atrial ALLHAT Antihypertensive and Lipid-Lowering Treat-
fibrillation might be useful to improve ment to Prevent Heart Attack Trial
symptoms CHARM-Preserved Candesartan in Heart Failure—Assessment of
The use of β-adrenergic blocking agents, IIb C
Reduction in Mortality
angiotensin converting enzyme
DIG Digoxin Investigation Group
inhibitors, angiotensin receptor
I-PRESERVE Irbesartan in Heart Failure with Preserved
blockers, or calcium antagonists in
Systolic Function
patients with controlled hypertension
PEP-CHF Perindopril in Elderly People with Chronic
might be effective to minimize
Heart Failure
symptoms of heart failure
OPTIMIZE-HF Organized Program to Initiate Lifesaving
Treatment in Hospitalized Patients with Heart
The use of digitalis to minimize IIb C
Failure
symptoms of heart failure might be
SENIORS Study of the Effects of Nebivolol Intervention
considered
on Outcomes and Rehospitalization in Seniors
Previously published. See “Credit Line” section. with Heart Failure

Heart Failure: Diagnosis and Evaluation
RICHARD J. RODEHEFFER, MD, and MARGARET M. REDFIELD, MD
Heart Failure
Heart failure has been defined as the pathophysiologic state in
which an abnormality of cardiac function prevents the heart from
pumping blood at a rate commensurate with metabolic require-
ments, or allows it to do so only when ventricular filling pressures
are excessively increased. However, only some patients with this
circulatory pathophysiology manifest the signs and symptoms of
the clinical syndrome we call heart failure (Figure 93.1). In rec-
ognition of this distinction, heart failure is now divided into four
stages: stage A patients have risk factors for heart failure (eg,
hypertension, coronary artery disease, diabetes, and obesity) but
normal ventricular structure and function and no heart failure
signs or symptoms; stage B patients have the ventricular struc-
tural and functional abnormalities but no heart failure signs or
symptoms; stage C patients have structural and functional abnor-
malities and symptoms of inadequate cardiac output (exercise
intolerance) or fluid overload (congestion); and stage D patients
have severely deranged cardiac structure and function, advanced
symptoms, and severe disability. The stages are associated with
increasing morbidity and mortality (Table 93.1).
Clinical Presentation of Patients
With Heart Failure
An important advance in the assessment and treatment of heart
failure has been the development of evidence-based guidelines
by committees of independent experts. These guidelines have
been published and are regularly updated by leading cardiovas-
cular disease societies. Implementation of guidelines has been
shown to improve outcomes in large populations.
Abbreviations and acronyms are expanded at the end of this chapter.
858
93
Heart Failure with Reduced or Preserved EF
Systolic heart failure refers to the pathophysiologic state in which
the predominant form of ventricular dysfunction is impaired
systolic contraction, and its hallmarks are reduced EF and a
dilated left ventricle. In contrast, the predominant abnormality in
heart failure with preserved EF (sometimes called diastolic heart
failure) is impaired diastolic filling, and the left ventricle is not
markedly dilated. In either form of heart failure cardiac output is
impaired and ventricular filling pressures are elevated.
Epidemiology of Heart Failure
Community data for persons older than 45 years show that about
25% have stage A heart failure, 33% have stage B heart fail-
ure, 2% to 3% have stage C heart failure, and fewer than 1%
have stage D heart failure. On the basis of objectively measured
ventricular function, approximately 6% of persons in such a
community cohort have any degree of systolic dysfunction (EF
<50%) and 2% have moderate to severe systolic dysfunction
(EF <40%). Approximately 7% have moderate to severe dias-
tolic dysfunction. Among those with stage C symptomatic heart
failure, about half have reduced systolic function and half have
heart failure with preserved EF.
More than 500,000 new cases of heart failure are diagnosed
each year in the United States, and 2 to 3 million people have
stage C heart failure. The population with stage B, asymptom-
atic ventricular function, is estimated to be an additional 3 to
5 million people. The annual number of deaths attributed to heart
failure approaches 200,000, and heart failure is the leading cause
for hospital admission for patients over 65 years of age.
The most important risk factors for heart failure con-
tinue to be age, hypertension, coronary disease, and diabetes.
 93 Heart Failure: Diagnosis and Evaluation
Figure 93.1. Natural History of Congestive Heart Failure (CHF). AHA indicates American Heart Association; CAD, coronary artery disease;
DM, diabetes mellitus; HTN, hypertension.
Approximately half of new-onset heart failure cases occur after
age 80 years, making heart failure a major disease of the very
elderly. The aging of the population, combined with decreasing
systolic heart failure mortality, have contributed to an increase
in the prevalence of heart failure. Improvements in hypertension
control and myocardial infarction survival have resulted in more
patients surviving with chronic hypertensive heart disease and
coronary disease who are at increased risk for heart failure.
Natural History of Heart Failure
Accurate estimation of prognosis for individual patients with
heart failure is remains challenging because of the multiple
causes of the heart failure syndrome, the large number of factors
that influence prognosis, different causes of death (eg, ischemic
events, progressive heart failure, and sudden death), and evolv-
ing treatment strategies. Current data from population-based
community studies show a five-year survival rate of about 50%.
Longitudinal data show a small but measurable improvement in
survival after heart failure diagnosis over the last several decades
that can be accounted for by better survival in patients with sys-
tolic heart failure (Figure 93.2). All-cause mortality rates for
patients across symptom classes who participated in heart failure
Table 93.1. ACC/AHA Heart Failure Stages
Stage A: Heart failure risk factors (hypertension, coronary disease,
diabetes, obesity) but no heart failure signs or symptoms or
abnormalities of cardiac structure and function.
Stage B: Abnormal cardiac structure and function but no heart failure signs
or symptoms. Asymptomatic ventricular dysfunction.
Stage C: Abnormal cardiac structure and function with overt heart failure
signs and symptoms.
Stage D: Refractory heart failure symptoms with advanced abnormalities
of cardiac structure and function.
859
randomized trials are shown in Table 93.2 and Figure 93.3. Trial
participants, it should be noted, are younger, have fewer comor-
bidities, and are more closely managed than heart failure patients
in community population studies.
Factors with an impact on heart failure prognosis are listed in
Table 93.3. Advanced age is a potent risk factor for a poor prog-
nosis. The severity of symptoms (functional class on optimal
medical therapy) is reflected in measurements of exercise capac-
ity and remains an important predictor of outcome. A rough, but
widely used, estimate of symptom severity is NYHA functional
class (Table 93.4). Most studies have documented poorer sur-
vival for patients with ventricular dysfunction due to coronary
artery disease than for patients with nonischemic ventricular
dysfunction. Patients with nonischemic dilated cardiomyopathy
and a potentially reversible etiology have a better chance of expe-
riencing improvement in systolic function in response to medi-
cal therapy (tachycardia-mediated cardiomyopathy, myocarditis,
peripartum cardiomyopathy, or alcoholic cardiomyopathy) and,
thus, a better prognosis.
Although EF correlates poorly with symptoms, it is an inde-
pendent prognostic predictor in systolic heart failure. The devel-
opment of right ventricular dysfunction in association with left
ventricular dysfunction results in worsening symptoms and
portends a worse prognosis.
Hemodynamic measurements obtained after optimization of
therapy have value in predicting prognosis, especially pulmonary
capillary wedge pressure and stroke work index. Likewise, the
presence of severe (restrictive) diastolic dysfunction as determined
with Doppler echocardiography has prognostic implications.
Neurohormonal factors that reflect activation of the renin-
angiotensin-aldosterone and sympathetic nervous systems and
increased ventricular filling pressure are predictive of a poor
prognosis. The most extensively evaluated of these is BNP.
Chronic atrial fibrillation or ventricular arrhythmias portend a
poorer outcome.
860 IX Cardiomyopathy and Heart Failure
Overt CHF
Risk factors
Myocyte injury
Cardiac remodeling
Asymptomatic LV dysfunction
Figure 93.2. Progression to Heart Failure. Myocardial damage leads to ventricular dilatation and hypertrophy (cardiac remodeling), but in the
early stages a patient may be well compensated hemodynamically without fluid retention or symptoms or signs of congestive heart failure (CHF).
Ultimately, adverse neurohumoral activation and progressive ventricular dysfunction lead to excessive vasoconstriction, sodium retention, and clini-
cal CHF. LV indicates left ventricular.
Causes of Systolic Ventricular Dysfunction
Systolic dysfunction and heart failure are the common end points
of a range of cardiovascular disease processes (Table 93.5).
Hypertension is the most common precursor of the develop-
ment of heart failure, occurring in 75% of patients. Coronary
artery disease is also a major risk factor for systolic heart failure.
Population studies indicated that hypertension and coronary dis-
ease each account for about 20% of incident heart failure cases.
Even after thorough evaluation the cause of nonischemic
dilated cardiomyopathy is identifiable in a minority of cases.
Well-conducted studies have demonstrated that at least 25% of
idiopathic nonischemic dilated cardiomyopathy patients have a
genetic basis, and over thirty different mutations have been shown
to cause the dilated cardiomyopathy phenotype. It now is recog-
nized that patients with persistent and excessive tachycardia (eg,
in atrial fibrillation with poorly controlled ventricular rates), may
develop reversible dilated cardiomyopathy. Since these patients
are often unaware of their tachycardia and may have controlled
resting heart rates, exercise testing or Holter monitoring may be
needed to detect the poor rate control.
Although uncommon, inflammatory myocarditis may be
associated with viral infection or systemic inflammatory condi-
tions and can evolve into dilated cardiomyopathy. In addition to
the known relationship between enteroviral infection (eg, cox-
sackievirus) and myocarditis, HIV has also been associated with
cardiac inflammation.
Table 93.2. Approximate 2-Year Mortality
Among Patients With Left Ventricular Systolic
Dysfunction Enrolled in Clinical Trials
NYHA Class Mortality
I 10%
II 20%
III 30–40%
IV 40–50%
Ventricular remodeling
and dysfunction
Sodium retention
Vasoconstriction
Adverse neurohumoral
activation
Cancer chemotherapeutic agents such as the anthracyclines
and traztuzumab have become increasingly common causes of
nonischemic systolic dysfunction. Cocaine use is associated with
myocarditis with persistent ventricular dysfunction as well as
microvascular disease, coronary artery spasm, and myocardial
infarction. Reversible systolic ventricular dysfunction associated
with severe sleep apnea has been described.
Causes of Diastolic Ventricular Dysfunction
The pathophysiology of diastolic dysfunction is discussed in
more detail elsewhere in this book. The principal factors associ-
ated with heart failure and preserved EF are age, hypertension,
coronary artery disease, and female gender. Infiltrative cardio-
myopathies such as amyloidosis or idiopathic fibrosis account for
a relatively small number of diastolic dysfunction cases.
As noted previously, the largest groups of heart failure
patients are those with stage B disease, asymptomatic ventric-
ular dysfunction. Such patients may be discovered in the course
of evaluating and managing their heart failure risk factors such
as coronary disease, hypertension, or diabetes. Or they may be
identified by chance due to abnormality found on an ECG or chest
radiography. Stage C heart failure patients present, by definition,
with overt signs or symptoms of ventricular dysfunction. Heart
failure symptoms and signs are similar regardless of whether the
EF is decreased or preserved (Table 93.6). Exertional dyspnea
and fatigue are early symptoms, but they are nonspecific and
can be produced by pulmonary disease, obesity, deconditioning,
and advanced age. Rales, edema, jugular venous distention, and
ascites are also nonspecific. Paroxysmal nocturnal dyspnea and
orthopnea are more specific for heart failure, but are relatively
insensitive indicators.
Evaluation of Heart Failure
The goals of the evaluation of patients with chronic heart failure
are 1) to characterize the type of cardiac dysfunction (decreased
or preserved EF) and the extent of structural involvement of the
 93 Heart Failure: Diagnosis and Evaluation 861

Table 93.3. Factors Associated with a Poor Prognosis in Heart failure laboratory evaluation is outlined in Table 93.7,
Ventricular Dysfunction A comprehensive echocardiogram is an essential tool for evalua-
Functional Class
tion of chamber size and function, valve function, wall thickness,
New York Heart Association class III or IV and pericardial effusion. In selected patients MRI scanning can
Peak oxygen consumption <10 mL/kg/min provide important information about pericardial disease, myo-
6-minute walking distance <350 m cardial fibrosis, or infiltrative myocardial diseases. Basic studies
Advanced age include blood count, urinalysis, electrolytes, calcium, magne-
Ischemic etiology sium, creatinine, BNP, creatine kinase, fasting glucose, lipid
Duration of symptoms profile, liver function tests, and thyroid-stimulating hormone.
Ejection fraction An ECG and chest radiograph should be evaluated. In selected
Left ventricular <25% patients at risk, measurement of ferritin and determination of
Right ventricular <35%
HIV status are appropriate.
Hemodynamics
Low cardiac index, low stroke work index
Because many patients with systolic dysfunction have
High pulmonary capillary wedge pressure, pulmonary hypertension ischemic cardiomyopathy it is important to evaluate for coronary
Restrictive filling pattern on Doppler echocardiography artery disease in patients who are potential candidates for revas-
Neurohormonal factors at increased levels cularization. Coronary angiography is recommended in patients
Norepinephrine with angina, a positive stress test, or who are at high risk for
Plasma renin activity coronary disease.
Aldosterone Endomyocardial biopsy for the detection of lymphocytic
Angiotensin II myocarditis is not routinely recommended because significant
Brain natriuretic peptide sampling error leads to a high rate of false-negative results, and
Arginine vasopressin
role of immunosuppression remains unproven. Endomyocardial
Endothelin
Tumor necrosis factor
biopsy may be of value if a systemic disease (eg, amyloidosis,
Arrhythmias hemochromatosis, and sarcoidosis) is suggested by the clinical
Atrial fibrillation and ventricular tachycardia presentation and cannot be diagnosed by less invasive means. If
giant cell myocarditis is suspected (ie, acute heart failure onset
in a young patient with chest pain, ventricular arrhythmias,
and normal coronary arteries), a biopsy should be considered
ventricles, atria, valves, and pericardium, 2) to identify correcta- because of the poor prognosis associated with giant cell myocar-
ble etiologic factors, 3) to assess the degree of circulatory impair- ditis and the potential for improvement with immunosuppression
ment, 4) to estimate prognosis, and 5) to guide therapy. or transplantation.
The history should include information on alcohol, illicit It is important to determine functional class based on an assess-
drugs, alternative therapies, and cancer chemotherapy. The abil- ment of the patient’s daily activity and the limitations imposed
ity to perform activities of daily living must be assessed. Physical by heart failure symptoms. Although subjective and insensitive,
examination should include special attention to fluid volume sta- the NYHA classification has long been used to categorize heart
tus, orthostatic blood pressure measurements, and measurement failure symptom severity (Table 93.4). While NYHA classifi-
of body mass index. cation provides prognostic information when applied to large

SOLVD SAVE
30 30 28%
Mortality From All Causes, %

Mortality Rate, %

21%
20 20 18%

10%

10 10

0 0
0 1 2 3 4 0 1 2 3 4
Years Years
Figure 93.3. Mortality Among Untreated Patients With Asymptomatic Ventricular Dysfunction in the SAVE and SOLVD Trials. (See “Credit
Lines” section.)
862 IX Cardiomyopathy and Heart Failure

Table 93.4. New York Heart Association Classification for Table 93.7. Heart Failure Laboratory Evaluation
Congestive Heart Failure Indicated
Class Description Chest radiography
Electrocardiogram
I Cardiac disease without resulting limitations of physical activity. Complete blood cell count
Ordinary physical activity does not cause undue fatigue, Urinalysis
palpitation, dyspnea, or anginal pain. Sodium, phosphorus, magnesium, calcium, blood urea nitrogen, creatinine,
II Cardiac disease resulting in slight limitation of physical activity. glucose, serum albumin, TSH
Comfortable at rest. Ordinary physical activity results in Transthoracic echocardiography
fatigue, palpitation, dyspnea, or anginal pain. Noninvasive stress testing to detect ischemia in patients who are potential
III Cardiac disease resulting in marked limitation of physical candidates for revascularization:
activity. They are comfortable at rest. Less than ordinary Without angina but with a high probability of coronary artery disease
physical activity causes fatigue, palpitation, dyspnea, or Without angina but with previous myocardial infarction, to detect
anginal pain. viability and residual ischemia
IV Cardiac disease resulting in inability to carry on any physical Coronary angiography for patients with:
activity without discomfort. Symptoms of heart failure present Angina
even at rest. If any physical activity is undertaken, discomfort Significant area of infarction or ischemia on noninvasive stress testing
is increased. Screening for other causes
Previously published. See “Credit Lines” section.
Only as suggested by history and physical examination findings
Reasonable, May Be Considered
Diagnostic tests for systemic diseases in patients when there is a clinical
suspicion (hemochromatosis, HIV, pheochromocytoma, sleep apnea,
amyloidosis, rheumatic diseases)
BNP measurement to assist in differentiating dyspnea of cardiac or
pulmonary origin
Noninvasive stress testing
To detect ischemia in all patients with unexplained heart failure who are
Table 93.5. Etiologies of Systolic Heart Failure candidates for revascularization
Coronary angiography
Coronary artery disease
For all patients with unexplained heart failure who are candidates for
Hypertension
revascularization
Diabetes
Endomyocardial biopsy for patients with:
Familial cardiomyopathies
Recent onset of rapidly deteriorating cardiac function, especially if
Tachycardia-induced cardiomyopathy
characterized by chest pain, arrhythmias, and normal coronary
Infectious agents: bacterial, viral (including human immunodeficiency virus)
arteries (giant cell myocarditis)
Infiltrative disorders: amyloidosis, hemochromatosis, sarcoidosis
Systemic disease and possible cardiac involvement (hemochromatosis,
Toxic: heroin, cocaine, alcohol, amphetamines, doxorubicin,
sarcoidosis, amyloidosis, Loffler endocarditis, endomyocardial
cyclophosphamide, sulfonamides, lead, arsenic, cobalt, phosphorus,
fibroclastosis) if diagnosis cannot be made by less invasive means
ethylene glycol, some antiviral agents
Exercise testing
Nutritional deficiencies: protein, thiamine, selenium
To quantify functional limitation
Electrolyte disorders: hypocalcemia, hypophosphatemia, hyponatremia,
To address specific clinical questions (rate control in patients with atrial
hypokalemia
fibrillation, chronotropic competence, exercise-induced arrhythmias,
Collagen vascular disorders: lupus rheumatoid arthritis, systemic sclerosis
adequacy of blood pressure control)
polyarteritis nodosa, hypersensitivity vasculitis, Takayasu syndrome,
Exercise measurement of respiratory gas exchange to assist in identification
polymyositis, Reiter syndrome
of patients who may be candidates for cardiac transplantation or
Endocrine and metabolic diseases: diabetes mellitus, thyroid disease,
ventricular assist pump
hypoparathyroidism with hypocalcemia, pheochromocytoma, acromegaly
Generally not indicated
Miscellaneous: peripartum cardiomyopathy, sleep apnea syndrome
Screening for asymptomatic arrhythmias in the absence of symptoms
Idiopathic
suggestive of sustained arrhythmia or syncope
Previously published. See “Credit Lines” section. Serial echocardiographic or radionuclide studies in the absence of change
in clinical status
Coronary angiography for patients who are not candidates for
revascularization, valve surgery, or heart transplant

Table 93.6. Presenting Clinical Features of Heart Failure

None Table 93.8.
. Functional Classification for Heart Failure
Truly asymptomatic Based on VO2max
Asymptomatic because of sedentary lifestyle
. Maximal Cardiac Index,
Dyspnea on exertion
Symptom Severity V O2max, mL/kg/m2 L/min/m2
Decreased exercise tolerance
Paroxysmal nocturnal dyspnea, orthopnea, pulmonary rales None >20 >8
Edema, ascites, jugular venous distention Mild 16–20 6–8
Abdominal pain and distention Moderate 10–15 4–6
Palpitations, presyncope, syncope Severe 6–9 2–4
Embolic events Very severe <6 <2
 93 Heart Failure: Diagnosis and Evaluation 863

cohorts of patients (Table 93.2), it is less helpful when applied to Table 93.9. Precipitating Factors for Decompensation of
individual patients. Stable Chronic Heart Failure
Objective quantification of functional capacity can be obtained Myocardial ischemia or infarction
by exercise testing with respiratory gas analysis and calculation Hypertension
of oxygen consumption. This provides useful information or car- Noncompliance with diet, medication, volume restriction
diac limitations,
. pulmonary limitations, and general decondi- Arrhythmia (often atrial fibrillation)
tioning. VO2max has prognostic value even in patients with severe Infection
systolic dysfunction but must be interpreted in view of age, sex, Pulmonary embolism, COPD exacerbation
and conditioning status. A functional Comorbid condition (renal failure, gastrointestinal fluid or blood loss)
. classification scheme based Anemia
on respiratory gas exchange and VO2max is .presented in Table
93.8. Patients with advanced symptoms and VO2max less than 10 Toxins (alcohol, street drugs)
Inappropriate drug therapy (negative inotrope, salt retention)
mL/kg per m2 have a poor two year survival with medical ther-
High-output states (pregnancy, hyperthyroidism, arteriovenous shunt)
apy and may benefit from left ventricular assist device or cardiac
transplantation. In addition to measurement of gas exchange, the
test also provides information on heart rate and blood pressure
responses to exercise.
When patients present with increased dyspnea it may be unclear
whether the dyspnea is due to left ventricular failure or exacer-
Associated or Systemic Conditions
bation of chronic lung disease. In this clinical setting measure-
Screening for sleep-disordered breathing, hemochromatosis, ment of BNP concentration may be of value in clarifying this
HIV, amyloidosis, rheumatologic disorders, and pheochromocy- distinction.
toma should be undertaken if there is clinical suspicion of these Previously healthy persons or patients with chronic heart fail-
diseases. Holter monitoring should be considered if ventricular ure may present with increased fluid volume manifested by pul-
or atrial arrhythmias are suspected. monary and peripheral edema, or low cardiac output manifested
by hypotension and cool vasoconstricted extremities. In addition
Clinical Monitoring of Chronic Heart Failure to the list of common precipitants in Table 93.9 the clinical cir-
cumstances may lead to consideration of uncommon causes of
Heart failure patients need regular long-term monitoring. At fol- heart failure such as acute inflammatory myocarditis, pericardial
low-up visits they should be reassessed new symptoms, for abil- tamponade, or apical ballooning syndrome.
ity to perform activities of daily living, for fluid volume status,
and for changes in diet or medications. Electrolytes and creati-
nine should be followed routinely. Repeat measurement of EF Abbreviations
or BNP should be guided by changes in clinical status. Four to ACC American College of Cardiology
six months after the initiation of a comprehensive management AHA American Heart Association
program it is appropriate to reassess ventricular function. This BNP brain natriuretic peptide
helps in making recommendations for defibrillator implantation CABG coronary artery bypass grafting
and may, if ventricular function has markedly improved, affect ECG electrocardiographic
estimates of prognosis. EF ejection fraction
The evaluation of patients who experience decompensation of NYHA
. New York Heart Association
stable chronic heart failure includes reassessment for events that VO2max maximal oxygen consumption
could precipitate hemodynamic instability. Cardiac and noncar-
diac precipitants must be considered (Table 93.9). In the acutely Names of Clinical Trials
ill patient a fundamental judgment must be made as to whether
CONSENSUS The Effects of Enalapril on Mortality in Severe
hospitalization is required for management of the decompensa-
Congestive Heart Failure
tion episode. SAVE Survival and Ventricular Enlargement Trial
SOLVD The Effects of Enalapril on Survival in Patients with
Evaluation of Acute Heart Failure Reduced Left Ventricular Ejection Fractions and
Congestive Heart Failure
Hospitalization for acute heart failure accounts for a large por- V-HeFT II Efficacy and Outcome With Felodipine in Heart
tion of the overall cost of heart failure care in the community. Failure
 94
Right Ventricular Failure and Cor Pulmonale
ROBERT P. FRANTZ, MD, and JOSEPH G. MURPHY, MD
RVF is a clinical syndrome that is often challenging to diagnose
and treat. It should not be considered a freestanding diagnosis
but instead should lead to a search for an underlying disease.
RVF most often occurs secondary to pulmonary hypertension
or left heart failure when it is often a harbinger of a poor prog-
nosis. Causes of isolated RVF include congenital heart disease,
severe tricuspid incompetence, right ventricular infarction, and,
rarely, arrhythmogenic right ventricular cardiomyopathy. Causes
of RVF are listed in Box 94.1a, while signs and symptoms of
RVF are listed in Box 94.1b.
Cor pulmonale is disease of the right side of the heart which
initially presents as structural and morphological changes in the
right ventricle (right ventricular hypertrophy and or right ventric-
ular dilatation) which may progress further over time to clinical
RVF. The pathological mechanism leading to cor pulmonale is
increased resistance to pulmonary blood flow with a compen-
satory increase in pulmonary perfusion pressure which leads
to morphological changes in the right heart chambers. A large
number of pulmonary diseases may cause cor pulmonale, includ-
ing pulmonary vascular disease, restrictive or obstructive lung
disease, neuromuscular disorders that affect ventilation, chest
wall disorders, and upper airway obstruction such as obstructive
sleep apnea.
• In addition to the underlying pulmonary disorder, right ventricular
output may be further compromised by an increased right ventric-
ular afterload associated with mechanical ventilation with positive
end-expiratory pressure
• Congenital heart disease, tricuspid valve disease, and left ven-
tricular diastolic dysfunction can all cause right-sided heart
disease, but these disorders are excluded from the definition of
cor pulmonale.
Abbreviations and acronyms are expanded at the end of this chapter.
864
Right ventricular infarction, pulmonary hypertension, heart
failure with preserved left ventricular function, and tricus-
pid valve disease are addressed separately in other chapters in
this text.
Diagnostic Evaluation of Suspected RVF
If RVF is suspected, the clinician should consider the long list of
associated conditions and the differential diagnosis. Peripheral
edema has many causes and often does not indicate RVF but
rather venous stasis or insufficiency, venous thrombosis, lymph-
edema, lymphatic obstruction, increased intra-abdominal pres-
sure or obstruction that impedes lower extremity venous or
lymphatic return, intrinsic liver disease, constrictive pericardi-
tis, nephrotic syndrome, or the side effects of medications such
as dihydropyridine calcium channel antagonists and nonsteroi-
dal anti-inflammatory drugs. Important clues from the medical
history include questions targeting symptoms of obstructive
sleep apnea; prior exposure to diet drugs, tobacco, and alcohol;
illicit drug use; risk factors for pulmonary embolus, including
recent prolonged travel or bed rest and use of birth control pills
or estrogen replacement therapy; and risk factors for human
immunodeficiency virus. Obstructive sleep apnea is increas-
ingly prevalent amid the obesity epidemic, and associated
nocturnal hypoxemia can result in pulmonary vasoconstric-
tion that may aggravate RVF. During the physical examination,
particular attention should be paid to the neck veins, since the
presence or absence of abnormal jugular venous waveforms
and JVP can be extremely helpful in defining whether RVF is
present. Abdominal compression to elicit the abdominojugu-
lar reflex should be routinely performed while examining the
neck veins. Physical signs of RVF include an elevated JVP with
a prominent “A” wave, right ventricle-generated left paraster-
nal or subxiphoid lift, and a right-sided third or fourth heart
 94 Right Ventricular Failure and Cor Pulmonale 865

Box 94.1a. Causes of RVF

Left-sided heart failure from any cause
Pulmonary hypertension from any cause
Lung disease with cor pulmonale
Pulmonary or tricuspid valve stenosis or
regurgitation
Congenital heart disease with intracardiac
shunting
Pericardial constriction
Right ventricular infarction or cardiomyopathy
Sleep apnea and hypoventilation syndromes

sound which augments with inspiration, peripheral edema, and

ascites.
Chest radiography and electrocardiography may provide
clues of underlying left heart disease, which commonly precipi-
tates RVF. The chest radiograph may show signs of right heart
enlargement, including enlargement of the cardiac border to the
right of the spine on the frontal chest radiograph and loss of the
anterior mediastinal clear space on the lateral view due to right
ventricular enlargement. Other features that may accompany
RVF are enlargement of the central pulmonary arteries in PAH,
hyperinflation in chronic obstructive pulmonary disease, or
interstitial lung changes that may accompany various inflamma-
tory or fibrotic lung diseases. Chest radiographs from a patient
with PAH are shown in Figure 94.1. The electrocardiogram may

Box 94.1b. Manifestations of RVF

Symptoms
Diminished appetite

Abdominal distention/pain
Peripheral swelling

Nasal congestion

Dyspnea on exertion
Exertional chest discomfort
Weight gain
Figure 94.1. Chest Radiographs From a Patient With Pulmonary
Signs Arterial Hypertension. Upper, In this anteroposterior view, peripheral
hypovascularity suggests pruning of the vascular tree consistent with
Jugular venous distention pulmonary arterial hypertension (long arrow). The hilar pulmonary
Hepatomegaly arteries are prominent (arrowhead). The right ventricle is enlarged
(short arrow). Lower, In this lateral view, right ventricular enlargement
Ascites is apparent in the retrosternal clear space (arrow).
Peripheral edema
Right ventricular lift show signs of right bundle branch block, right atrial enlargement,
right axis deviation, or right ventricular hypertrophy. A typical
Right-sided third heart sound electrocardiogram from a patient with idiopathic PAH resulting
Accentuated pulmonic closure sound in right ventricular hypertrophy is shown in Figure 94.2.
Echocardiography is mandatory in all patients with sus-
Murmur of tricuspid insufficiency
pected RVF with particular reference to congenital heart dis-
Murmur of pulmonic insufficiency ease, valvular heart disease and diastolic dysfunction of the left
ventricle. Other echocardiographic findings include tricuspid
866 IX Cardiomyopathy and Heart Failure
I aVR
II aVL
III aVF
II
Figure 94.2. Pulmonary Arterial Hypertension. Electrocardiogram indicates right atrial enlargement, right axis deviation, and right ventricular
hypertrophy.
regurgitation, right atrial enlargement, and failure of the inferior
vena cava to collapse with inspiration. Noninvasive estimation of
right ventricular systolic pressure is performed by interrogation
of the tricuspid regurgitant signal and application of the modi-
fied Bernoulli equation. The modified Bernoulli equation is the
following:
Right ventricular pressure = [4× (Triscuspid regurgitant
velocity)2] + Estimated right atrial pressure
It is important to realize that echocardiographically derived
estimates of pulmonary artery pressure are subject to a variety
of sources of potential error and must be viewed in a clinical con-
text. Echocardiography will also show cardiac chamber size and
morphology. An echocardiogram that shows a D-shaped septum
with an underfilled left ventricle, with normal or small left atrial
size in the setting of right ventricular enlargement and dysfunc-
tion is most compatible with cor pulmonale.
• A normal left atrial size argues against left ventricular diastolic
dysfunction as a cause RVF.
• Left atrial enlargement in a patient with RVF is suggestive of a
left heart disease etiology but is not definitive as it may be a coin-
cidental finding.
• Definitive diagnosis and classification of PAH requires cardiac
catheterization.
• A reduction in the DLCO is a sensitive and early indicator of
abnormal gas exchange in pulmonary vascular disease.
• A normal DLCO argues against a diagnosis of significant pulmo-
nary hypertension.
Classification of Pulmonary Hypertension
Pulmonary hypertension often accompanies RVF. When present,
pulmonary hypertension should be classified according to the
Dana Point classification system (Box 94.2). This system divides
patients into those with PAH (Dana Point group 1) and those
with non-PAH. Non-PAH is further categorized into group 2 (left
heart disease), group 3 (pulmonary hypertension associated with
V1 V4
V2 V5
V3 V6
hypoxemia or ventilatory disorders), group 4 (thromboembolic
pulmonary hypertension) and group 5 (miscellaneous causes,
some of which may respond to PAH therapies in a fashion sim-
ilar to group 1). A special category (1′) designates pulmonary
veno-occlusive disease and pulmonary capillary hemangioma-
tosis. The diagnosis of PAH requires that the following criteria
be met:
1. Mean pulmonary artery pressure greater than 25 mm Hg
2. Pulmonary capillary wedge pressure or left ventricular end-diastolic
pressure within the reference range (<15 mm Hg)
3. Absence of left heart disease or other diseases categorized as
non-PAH
Generally, an additional requirement of PVR greater than 3
Wood units is also utilized to define PAH, since inclusion of only
mean pulmonary artery pressure and pulmonary capillary wedge
pressure could result in erroneous inclusion of patients with
Box 94.2. Summary of Updated Clinical Classifi-
cation of Pulmonary Hypertension (Dana Point, 2008)
Group 1: Pulmonary arterial hypertension (PAH)
Group 1′
Pulmonary veno-occlusive disease and/or pulmonary capillary
hemangiomatosis
Group 2: Pulmonary hypertension owing to left
heart disease
Group 3: Pulmonary hypertension owing to lung
diseases and/or hypoxia
Group 4: Chronic thromboembolic pulmonary
hypertension
Group 5: Pulmonary hypertension with unclear
multifactorial mechanisms
Previously published. See “Credit Lines“ section.
 94 Right Ventricular Failure and Cor Pulmonale
elevated cardiac output, such as patients with portopulmonary
pulmonary hypertension who may have pulmonary hypertension
but often have a normal PVR. Such patients technically fall into
the “other diseases categorized as non-PAH,” but inclusion of
a PVR requirement further prevents errors in classification that
could result in errors in treatment.
The distinction between PAH and non-PAH is useful because
patients with PAH may respond to selective pulmonary vaso-
dilators (Box 94.3) while patients with non-PAH who have an
elevated pulmonary capillary wedge pressure may worsen after
administration of selective pulmonary vasodilators and are at risk
of developing pulmonary edema owing to the increase in blood
flow into the lung that may further congest the pulmonary vas-
cular bed. Patients with non-PAH but normal pulmonary capil-
lary wedge pressure (eg, obstructive lung disease) have generally
been excluded from randomized trials of selective pulmonary
vasodilators, so no confident recommendations can be made on
the utility of selective pulmonary vasodilators in those patients.
Management of RVF Associated With PAH
RVF is the commonest cause of hospitalization in patients with PAH
and is a strong predictor of poor long-term survival in pulmonary
hypertension. Treatment of RVF associated with PAH includes use
of diuretics, warfarin to reduce the risk of associated venous throm-
bosis or in situ pulmonary vascular thrombosis, and supplemental
oxygen to maintain optimal systemic oxygen saturation. Digoxin
does not provide a comparable inotropic effect in RVF associated
with pulmonary hypertension, as it does in left ventricular failure
and is generally avoided in RVF, excepting coexisting atrial fibril-
lation. Digoxin has also been associated with pulmonary vasocon-
striction and worsening of pulmonary hypertension.
The patient should ideally be referred to a tertiary pulmonary
hypertension program for assistance with management of this
complex problem. Use of selective pulmonary vasodilators such
as endothelin receptor antagonists (eg, bosentan or ambrisentan),
phosphodiesterase-5 inhibitors (eg, sildenafil or tadalafil), and
prostanoids (eg, inhaled or intravenous iloprost, inhaled, subcu-
taneous or intravenous treprostinil or intravenous epoprostenol)
should be considered. Recommendations can be found in the
Box 94.3. Classification of Pulmonary Arterial
Hypertension
Idiopathic PAH
Familial PAH (documented bone morphogenetic
protein receptor II mutation)
PAH related to
Connective tissue disease
Human immunodeficiency virus
Portal hypertension
Anorexigens
Congenital heart disease
Persistent pulmonary hypertension of the newborn
PAH with venule/capillary involvement
Abbreviation: PAH, pulmonary arterial hypertension.
867
PAH practice guidelines (see the “Resources” list at the end of
this chapter and Chapter 79 “Pulmonary Hypertension.”) Lung
transplant may be necessary in patients with PAH that is refrac-
tory to medical therapy. Surgical thromboendarterectomy may
be indicated for selected patients with pulmonary thromboem-
bolic disease.
• Localized hypoxia dilates systemic circulations but vasoconstricts
the pulmonary vasculature leading to an increase in PVR.
• PVR acts as the afterload for the right ventricle which can be
reduced by administration of supplemental oxygen to hypoxic
patients, as oxygen will frequently reverse hypoxic pulmonary
vasoconstriction.
• Left ventricular dysfunction can exacerbate pulmonary hyperten-
sion due to its direct effect by increasing pulmonary venous pres-
sure and indirectly through interstitial pulmonary edema, which
causes hypoxia and further pulmonary arterial vasoconstriction.
Management of RVF Complicating Left
Ventricular Systolic Dysfunction
Management of RVF that is complicating left ventricular sys-
tolic dysfunction should focus on optimizing left ventricular per-
formance, lowering left ventricular end-diastolic pressure, and
lessening mitral regurgitation by effective afterload reduction.
This includes optimization of doses of angiotensin-converting
enzyme inhibitors or angiotensin II receptor antagonists (or
occasionally both). In some situations, tenuous renal func-
tion greatly complicates the management of biventricular fail-
ure. If the use of these drugs needs to be stopped or reduced,
use of nitrates and hydralazine should be considered in order
to maintain optimal preload and afterload reduction. Avoiding
hypotension may be important in preserving renal perfusion. A
systematic search for hypoxemia, including overnight oximetry,
is very important since sleep-disordered breathing is common
in congestive heart failure and correction of hypoxemia second-
ary to sleep-disordered breathing may improve heart failure
management.
Patients with biventricular failure may present with rapidly
progressive ascites and peripheral edema; this leads to edema
in the intestinal wall, impeding absorption of oral diuretics and
preventing effective diuresis with oral agents. Patients should
be questioned carefully about recent fluid or salt indiscretion
or use of nonsteroidal anti-inflammatory drugs since these are
common precipitants of progressive fluid retention. In this situ-
ation, intravenous loop diuretics (either as a bolus or as a drip)
can be extremely helpful in achieving diuresis. Use of oral meto-
lazone can also help to achieve diuresis if the patient has a poor
response to loop diuretics. By blocking reuptake in the distal
tubule, metolazone works synergistically with loop diuretics.
Metolazone has the potential for precipitation of electrolyte
disturbances (eg, hypokalemia and hyponatremia) and renal
failure. Patients should be weighed daily and have electrolytes
checked at least twice weekly during periods of augmented
diuresis. Use of metolazone once or twice a week 30 minutes
before administration of a loop diuretic is often effective but
is not a substitute for fluid and salt restriction. Aldosterone
antagonists (eg, spironolactone or eplerenone) can be used as a
potassium-sparing diuretic for patients already receiving a loop
diuretic; efficacy is particularly well established for patients
with congestive heart failure due to left ventricular systolic dys-
function, Supplemental potassium requirements may decrease
or vanish and electrolyte levels must be closely monitored for
hyperkalemia.
868 IX Cardiomyopathy and Heart Failure
Aldosterone antagonists should generally be avoided in
patients with significant renal insufficiency because of the risk
of precipitating hyperkalemia. Paracentesis can be useful if
massive ascites is complicating RVF. This may also improve
renal function by decompressing intra-abdominal pressure on
the renal veins. Ultrafiltration may also be useful in removing
large fluid volumes, but it may aggravate renal dysfunction.
Although inotropes such as dopamine, dobutamine, or milri-
none may be useful for treatment of the hospitalized patient
with refractory RVF, they increase the risk of arrhythmia,
and milrinone may cause hypotension and precipitate renal
failure.
Research is ongoing to understand whether phosphodi-
esterase-5 inhibitors and soluble guanylate cyclase activators
and/or stimulators may have roles in management of patients
with pulmonary hypertension associated with left ventricular
systolic or diastolic failure.
Patients being considered for placement of a left ventricu-
lar assist device should have optimization of volume status to
decrease right heart pressures before device placement. This
reduces the risk of intractable RVF after placement of a left ven-
tricular assist device.
Arrhythmogenic Right Ventricular
Cardiomyopathy or Dysplasia
ARVC also called ARVD is a rare genetically determined cardio-
myopathy frequently associated with mutations in cell adhesion
genes and desmosomal proteins that morphologically affect the
free wall of the right ventricle, which is typically dyskinetic on
echocardiography. Microscopically there is replacement of right
ventricular myocytes by noncontractile fibrous and fatty tissue.
The left ventricular myocardium can also be involved, but there
is usually relative sparing of the ventricular septum. While the
right ventricle is often dilated and the right ventricular free wall
is poorly contractile, overt RVF is uncommon, and the common-
est clinical presentation of ARVD is palpitations or dizziness
due to ventricular arrhythmias. Sudden cardiac death may also
occur which is often exercise-induced, particularly in athletes in
whom ARVD is one of a number of rare causes of sudden cardiac
death.
Suggested Reading
Haddad F, Doyle R, Murphy DJ, Hunt SA. Right ventricular function in
cardiovascular disease, part II: pathophysiology, clinical importance,
and management of right ventricular failure. Circulation. 2008 Apr
1;117(13):1717–31.
McLaughlin VV, Archer SL, Badesch DB, Barst RJ, Farber HW,
Lindner JR, et al; ACCF/AHA. ACCF/AHA 2009 expert consen-
sus document on pulmonary hypertension: a report of the American
College of Cardiology Foundation Task Force on Expert Consensus
Documents and the American Heart Association: developed in col-
laboration with the American College of Chest Physicians, American
Thoracic Society, Inc., and the Pulmonary Hypertension Association.
Circulation. 2009 Apr 28;119(16):2250–94. Epub 2009 Mar 30.
Erratum in: Circulation. 2009 Jul 14;120(2):e13.
Simonneau G, Robbins IM, Beghetti M, Channick RN, Delcroix M,
Denton CP, et al. Updated clinical classification of pulmonary hyper-
tension. J Am Coll Cardiol. 2009 Jun 30;54(1 Suppl):S43–54.
Abbreviations
ARVC arrhythmogenic right ventricular cardiomyopathy
ARVD arrhythmogenic right ventricular dysplasia
DLCO diffusing capacity for carbon monoxide
PAH pulmonary arterial hypertension
PVR pulmonary vascular resistance
RVF right ventricular failure
 95
Pharmacologic Therapy of Systolic Ventricular
Dysfunction and Heart Failure
RICHARD J. RODEHEFFER, MD, and MARGARET M. REDFIELD, MD
Principles of Treatment
All patients with systolic ventricular dysfunction should be
considered for aggressive treatment of the underlying coro-
nary artery disease, valvular heart disease, or hypertension.
Concurrent with the evaluation and management of the under-
lying cause of systolic ventricular dysfunction, heart failure-spe-
cific medical therapy should be started to reduce both morbidity
and mortality. Treatment strategies for all patients with systolic
dysfunction, regardless of the underlying myocardial disorder,
are outlined according to the stage of heart failure in Table 95.1.
Therapies that have been proved to decrease morbidity and mor-
tality (ie, ACE inhibitors and β-blockers should be used for all
patients with decreased systolic function. Use of therapies that
control symptoms (ie, digoxin and diuretics), but have not been
proved to reduce mortality, should be guided by symptoms.
Nonpharmacologic measures appropriate for all patients with
systolic dysfunction are also outlined in Table 95.1.
ACC/AHA practice guidelines for the evaluation and man-
agement of chronic heart failure in adults suggest a stepped
approach to the classification of heart failure that emphasizes the
progression of the disease. Accordingly, 4 heart failure stages
have been identified:
1. Stage A—heart failure risk factors without structural heart disease
2. Stage B—structural heart disease without symptoms
3. Stage C—structural heart disease associated with past or present
heart failure symptoms
4. Stage D—end-stage disease that may require specialized treatment
strategies
Abbreviations and acronyms are expanded at the end of this chapter.
ACE Inhibitors
The cleavage of angiotensinogen by renin gives rise to angio-
tensin I, which is modified by ACE to form angiotensin II, an
important regulator of vascular tone, myocardial hypertrophy
and fibrosis, aldosterone secretion, and glomerular filtration rate
(Figure 95.1). In addition to its systemic production, angiotensin
is produced at a local level by several tissues (including the heart
and vasculature), and local tissue renin-angiotensin systems
may be upregulated independently from systemic production.
The RAAS is activated in heart failure and in patients taking
diuretics. Antagonism of the deleterious actions of angiotensin II
is beneficial in all patients with systolic dysfunction, regardless
of the severity of symptoms. This approach is supported by the
results of large clinical trials demonstrating that ACE inhibitor
therapy reduces symptoms, retards the progression of heart fail-
ure (including the need for hospitalization), and reduces mortality
among patients with systolic dysfunction regardless of functional
class. Mortality reduction is most striking in the most sympto-
matic patients and in those with severe systolic dysfunction.
As combined arterial and venous vasodilators, ACE inhibi-
tors have favorable hemodynamic effects in patients with systolic
dysfunction. Reductions in afterload, preload, and wall stress are
observed without an increase in heart rate. ACE inhibitors aug-
ment renal blood flow and reduce production of aldosterone and
antidiuretic hormone. Thus, they promote excretion of sodium
and water.
ACE inhibitors have effects on cellular metabolism indepen-
dent of their hemodynamic effects. Antihypertrophic effects on
ventricular and vascular cells are at least partially independent
of blood pressure reduction and contribute to their role in the
prevention of ventricular remodeling. ACE inhibitors blunt the
progressive dilatation of the ventricle in patients with systolic
869
870 IX Cardiomyopathy and Heart Failure

Table 95.1. General Approach to Medical Therapy for
Patients With Systolic Dysfunction
Stage A (risk factors)
Treat atherosclerosis risk factors, ischemia, and chronic coronary
disease
Control hypertension and diabetes
Appropriate dietary management
Exercise program
Aggressive management of obesity
Stage B (asymptomatic left ventricular dysfunction)
All stage A measures
ACE inhibitor/ARB
β-blocker
Stage C (symptomatic left ventricular dysfunction)
All stage A and B measures
Salt restriction
Diuretics
Selected patients: aldosterone antagonist, ARB, digitalis, hydralazine
and nitrates, CRT, ICD, warfarin
Stage D (end-stage heart failure)
All appropriate stage A, B, and C measures
Palliative care, hospice
Selected patients: LVAD, cardiac transplant, chronic intravenous
inotropic support
dysfunction and are potent agents for reducing ventricular hyper-
trophy in patients with hypertension.
Specific ACE inhibitors differ in chemical structure, which
imparts differences in potency, half-life, bioavailability, route of
elimination, and affinity for tissue-bound ACE. Use of the opti-
mal dose of ACE inhibitors is clinically important. Uptitration to
target doses established to be efficacious in clinical trials should
be attempted in all patients (Table 95.2). Monitoring of serum
creatinine and potassium level is important during the titration
phase. Concomitant use of diuretics may need to be decreased if
systemic hypotension occurs. Mild renal impairment and a mild
increase in the serum level of creatinine during the uptitration of
ACE inhibitors are not contraindications to initiation of therapy,
although meticulous monitoring of electrolytes during uptitra-
tion is required. In the ELITE trial, more than 70% of the elderly
patients in the ACE inhibitor group tolerated full doses of the
drug. ACE inhibitors are contraindicated in the presence of sig-
nificant renal artery stenosis.
In addition to renal insufficiency, hyperkalemia, and hypoten-
sion, other class-related side effects of ACE inhibitors include
cough, angioedema, and dysgeusia (metallic taste). Cough occurs
in 5% to 10% of white persons but may be more common among
Asians. ACE inhibitors may be teratogenic and should not be
used by females of childbearing age who are not using effective
birth control.
Angiotensin II Receptor Blockers
ARBs were developed as antagonists of the RAAS that do not
cause potentiation of bradykinin effects, such as cough. The find-
ings of the ELITE and ELITE II trials suggest that angiotensin II
antagonists and ACE inhibitors have similar beneficial effects on
symptoms and survival. In the CHARM trial the addition of can-
desartan to ACE inhibitor in systolic heart failure resulted in a
modest but significant incremental benefit in mortality and heart
failure hospitalization, even in patients also taking β-blockers.
Like ACE inhibitors, ARBs also carry the risk of hyperkalemia
and increased creatinine.
Digitalis Glycosides in Systolic Heart Failure
In the RADIANCE trial, patients receiving digoxin treatment
who were in sinus rhythm and had an EF ≤35% and NYHA
class II or III symptoms were randomly assigned to continue tak-
ing digoxin or to substitute a placebo for digoxin. All patients
were also taking ACE inhibitors and diuretics. The probabil-
ity of clinical deterioration was 25% among those who stopped
receiving digoxin treatment and 5% among those who continued
receiving digoxin (Figure 95.2). The Digoxin Study prospectively
randomized 6,800 patients with symptomatic heart failure and

Local
RAS
Aogen ACE ACE
Renin AI All

Thirst Aldosterone NE VC VC Chronotropy

GFR Hypertrophy Inotropy?
Mesangial contraction Hyperplasia Lusitropy
Tubular Na reabsorption Hypertrophy
Fibrosis
Arrhythmogenesis
Figure 95.1. Circulating and Local Renin-Angiotensin System (RAS). Renin from the kidney converts angiotensinogen (Aogen) to angiotensin
I (AI), which is converted to angiotensin II (AII) by angiotensin-converting enzyme (ACE) in the lungs. AII, either from this pathway or from inde-
pendent production in tissues, exerts effects on the brain, adrenals, sympathetic nervous system, kidney, vasculature, and heart. NE indicates norep-
inephrine; GFR, glomerular filtration rate; Na, sodium; VC, arterial vasoconstriction.
 95 Pharmacologic Therapy of Systolic Ventricular Dysfunction and Heart Failure
Table 95.2. Recommended Dosages of
Commonly Used Angiotensin-Converting
Enzyme Inhibitors for Treatment of Systolic
Dysfunction
Agent Dose, mg Doses Per Day, No.
Captopril 50 3 increase in EF is positively correlated to the β-blocker dose and
Enalapril 10 2 is more evident in the absence of widespread myocardial scar
Lisinopril 20–40 1 and fibrosis.
sinus rhythm to digoxin or placebo. Digoxin treatment did not
alter mortality but did reduce risk of hospitalization for worsen-
ing heart failure. Digoxin, therefore, improves clinical stability
and reduces heart failure hospitalizations. The dose of digoxin
should be decreased in patients with renal insufficiency and in
those taking amiodarone or propafenone. Plasma levels should
not exceed 1.0 mg/mL.
Aldosterone Antagonists
The RALES trial evaluated spironolactone in 1,663 patients with
severe heart failure and an EF ≤35% who were being treated with
standard therapy. Spironolactone decreased mortality (46% in the
placebo group and 35% in the spironolactone group), as well as
heart failure hospitalization. This 24% reduction in mortality was
due to a lower risk of sudden death and death from progressive
heart failure. In addition, patients who received spironolactone
had improvement in NYHA symptom class. A similar controlled
trial using eplerenone likewise showed decreased mortality
among 6,632 patients with systolic heart failure. Therefore, the
addition of aldosterone blockade to standard therapy provides fur-
ther decreases in both morbidity and death among patients with
moderate to severe heart failure. Addition of aldosterone antago-
nists can cause serious hyperkalemia, and serum potassium levels
should be monitored carefully during initiation of treatment.
β -Blockers in Heart Failure
Left ventricular dysfunction is accompanied by increased sympa-
thetic nervous system activation, which intensifies as heart failure
becomes more severe. Experimental data suggest that long-term
augmentation of the sympathetic nervous system with increased
0.3
Probability of Worsening
Placebo
Digoxin
Heart Failure
0.2 P<.001
0.1
0.0
0 20 40 60 80
Days After Randomization
Figure 95.2. RADIANCE Study. Kaplan-Meier analysis of cumu-
lative probability of worsening heart failure in patients whose treat-
ment was switched to placebo (n-93). Patients in the placebo group had
a higher risk of worsening heart failure throughout the 12-week study
(relative risk, 5.9; 95% confidence interval, 2.1–17.2; P<.001).
871
plasma norepinephrine levels exacerbates myocyte dysfunction
and cell death, and down-regulates β-receptor responsiveness.
Although short-term administration of β-adrenergic blockers
decreases contractility and heart rate in systolic heart failure,
long-term administration improves contractility. This effect may
not become apparent until after 3 to 6 months of treatment. The
Controlled trials have demonstrated that β-blockers pro-
long survival and improve functional class, cardiac hemody-
namics, left ventricular EF, and quality of life in patients with
NYHA class II, III, or IV congestive heart failure (Table 95.3).
Carvedilol provides significant mortality benefit in patients with
NYHA class II to IV heart failure. Trials with cardioselective
β-blockers, CIBIS-II (bisoprolol) and MERIT-HF (metoprolol
succinate), also showed significant improvement in survival.
After initiation, β-blocker therapy should be continued indef-
initely, as discontinuation may result in deterioration of cardiac
function.
Hydralazine and Isosorbide
Dinitrate in Heart Failure
In the V-HeFT I trial, the combination of hydralazine and iso-
sorbide dinitrate reduced mortality and improved symptoms in
patients with heart failure. In the V-HeFT II trial, ACE inhibitors
reduced mortality more than hydralazine and isosorbide dini-
trate. In a trial focused specifically on African American patients
the addition of hydralazine and isosorbide dinitrate to ACE inhi-
bition and β-blocker therapy produced significant incremental
benefit.
Calcium Channel Blockers in Heart Failure
First-generation calcium channel blockers (eg, verapamil, dilti-
azem, and nifedipine) are to be avoided in systolic heart failure
because of their negative inotropic properties.
Newer-generation dihydropyridines have higher vascular-to-
myocardial specificity and fewer negative inotropic effects. The
PRAISE and PRAISE-II trials tested amlodipine in patients with
heart failure and showed that amlodipine had no beneficial effect
on survival or hospitalization, but could be given safely. The
V-HeFT III trial examined the use of felodipine in patients with
heart failure. Again, no beneficial effect on survival or symptoms
was demonstrated but safety was documented. Amlodipine and
felodipine may be an option for treating patients who are still
hypertensive while receiving maximal doses of standard heart
failure therapy.
Oral Inotropic Agents in Heart Failure
Phosphodiesterase inhibitors and other “calcium sensitizing
agents” developed as inotropic drugs have been used for the
treatment of heart failure. While safe and efficacious when given
intravenously to decompensated patients in a monitored hospital
setting, chronic oral administration of these drugs has resulted in
100 increased mortality. Hence, they are not used for long-term heart
failure treatment.
Intermittent and Continuous Intravenous
Inotropic Therapy
Small clinical studies suggested that the periodic use of short-term
intravenous inotropic support provided long-lasting symptomatic
872 IX Cardiomyopathy and Heart Failure
Table 95.3. Selected Clinical Trials of β-Blockers in Systolic Dysfunction
Patient Population
Study NYHA Class N Drugs
US Carvedilol II-IV 1,094 Carvedilol vs placebo
CIBIS-II II-IV 641 Bisoprolol vs placebo
MERIT-HF II-IV 3,991 Metoprolol (succinate) vs placebo
COPERNICUS IV 2,289 Carvedilol vs placebo
COMET II-IV 3,029 Carvedilol vs metoprolol tartrate
benefit. Larger, more definitive clinical trials have not been per-
formed, and the value of this form of therapy is unproven. It is
not approved for heart failure management.
Continuous low-dose intravenous inotropic therapy is occa-
sionally administered on an outpatient basis in highly selected
patients with end-stage disease as a palliative measure. However,
controlled clinical trials have not been performed to demonstrate
the safety and efficacy of this therapy. Long-term intravenous
inotropic support carries the risk of arrhythmia, as well as cen-
tral line infection and central line–associated venous thrombosis
complications.
Diuretics in Heart Failure
Data from the SHEP trial showed that diuretic-based therapy for
hypertension reduces new onset of heart failure, especially in
patients with a history of MI (80% decrease in initial episodes
of heart failure). Diuretics are necessary for patients with estab-
lished heart failure who remain symptomatic despite treatment
with ACE inhibitors, β-blockers, and digoxin. Patients who pre-
sent with pulmonary edema need diuretics immediately.
Only the minimum dose of diuretics needed to control con-
gestion should be used. Overdiuresis exacerbates the activation
of the RAAS and may result in hypotension, prerenal azotemia,
hyponatremia, hypokalemia, and hypomagnesemia. It may also
result in excessive thirst, consequent excessive fluid intake, and
apparent diuretic refractoriness. Education of patients about the
transient use of higher doses of diuretics in response to increases
in fluid retention can prevent episodes of decompensation.
In patients who seem to be developing diuretic resistance it
is important to determine whether the dose of diuretics is effec-
tive (ie, produces a diuretic response). If not, the dose should be
increased. In advanced heart failure, twice-daily dosing may be
required. If high doses of furosemide are ineffective, a loop diu-
retic with better absorption, such as torsemide or bumetanide,
should be tried. In the absence of renal failure, the need for furo-
semide doses greater than 120 mg twice should prompt consid-
eration of combination therapy with a thiazide diuretic which
potentiates the action of loop diuretics by blocking sodium reup-
take in the distal nephron. Metolazone is often used in this setting
and may be given daily, every other day, or as needed in response
to an increase in body weight. Such combination diuretic ther-
apy can result in profound hypokalemia and volume depletion, so
careful monitoring of electrolytes and blood pressure is required.
Patients with refractory hypokalemia may benefit from a small
dose of a potassium-sparing diuretic such as spironolactone.
Careful monitoring of potassium levels is required, especially in
patients taking ACE inhibitors or ARBs.
Anticoagulation in Heart Failure
Heart failure is associated with an increased risk of thrombo-
embolic events. Risk factors include atrial fibrillation and the
Decrease in All-Cause
Follow-Up, y Mortality, %
0.7 65
1.3 34
1 34
0.94 35
5.8 18
severity of cardiac dysfunction. The efficacy of warfarin in
reducing embolic events in atrial fibrillation patients has been
well documented. Similarly, patients with a recent anterior or
large MI have a markedly increased risk of cardioembolic events,
and warfarin anticoagulation is recommended for at least the first
3 to 6 months after infarction. Patients with systolic dysfunc-
tion who have had a previous cardioembolic event are also at
increased risk regardless of underlying rhythm, and warfarin
anticoagulation is recommended for them.
The value of long-term anticoagulation in patients with sinus
rhythm and chronic ischemic or nonischemic dilated cardio-
myopathy without a recent large MI or previous cardioembolic
event is a matter of controversy. The WASH trial failed to find
benefit for warfarin or aspirin compared to placebo in heart
failure patients with EF ≤35%. In similar patients the WATCH
trial showed no difference in stroke prevention (ischemic plus
hemorrhagic stroke) between warfarin, clopidogrel, or aspirin
(annual stroke rate 1.7%, 2.5%, 2.9% respectively).
At this time, it is reasonable to recommend warfarin anticoag-
ulation for heart failure in sinus rhythm if they have had previous
systemic or pulmonary emboli, or if there is visible thrombosis
on echocardiography.
Sudden Death and Sustained Ventricular
Tachycardia in Heart Failure
Sudden death is common among patients with heart failure and
accounts for approximately one-third of the deaths among these
patients. Although sudden death is widely assumed to be a con-
sequence of ventricular tachycardia or fibrillation, bradyarrhyth-
mias, electromechanical disassociation, acute MI, or embolic
events may account for a substantial proportion of these sudden
death events.
All patients should be monitored for electrolyte derange-
ments, treated with medications known to reduce sudden death
(β-blockers, aldosterone blockers), and assessed for ongo-
ing myocardial ischemia that may precipitate arrhythmias.
Antiarrhythmic drugs for suppression of premature ventricular
contractions or nonsustained ventricular tachycardia have not
been effective for improving survival. Patients who survive an
episode of cardiac arrest or who present with sustained ventricu-
lar tachycardia are candidates for ICD implantation. For patients
with ischemic or nonischemic heart failure, left ventricular EF
less than 35%, and an expected survival of at least one year, pro-
phylactic insertion of an ICD is also indicated on the basis of out-
comes of controlled trials. Further discussion of the indications
for defibrillator therapy in heart failure patients is provided in
Chapter 31 (“Implantable Cardioverter-Defibrillator Trials and
Prevention of Sudden Cardiac Death”).
Cardiac Resynchronization Therapy
Multiple trials have documented that biventricular pacing can
reduce heart failure symptoms and hospitalization and improve
 95 Pharmacologic Therapy of Systolic Ventricular Dysfunction and Heart Failure 873

exercise tolerance and survival among patients with significant pulmonary dyspnea. It is critical to exclude an evolving acute
dyssynchrony of left ventricular contraction. Biventricular pac- coronary syndrome with ECG and troponin measurements.
ing should be undertaken only after a comprehensive pharmaco- Initial evaluation should include should include measurement
logic treatment program for heart failure has been established. of oxygen saturation and administration of oxygen as needed.
Patients with NYHA class III symptoms, sinus rhythm, EF If reduced systemic perfusion and decreased urine output sug-
≤35%, and QRS duration of >120 msec are appropriate candi- gest the onset of shock, urgent fluid resuscitation is indicated. In
dates for CRT. It is worth noting that subanalyses in large CRT the presence of pulmonary edema and hypoxemia, intravenous
trials have shown that benefit is more likely to occur in patients morphine may be helpful to urgently relieve dyspnea. Pulmonary
with QRS much longer than 120 msec, ie, greater than 150 msec. and systemic congestion require prompt initiation of intrave-
Trials of similar patients in atrial fibrillation indicate that they nous loop diuretic therapy (by bolus or continuous infusion). If
may also derive benefit from CRT (See Chapter 29 for further there is evidence of poor systemic perfusion and cardiac filling
discussion of CRT). pressures cannot be determined by clinical evaluation, invasive
hemodynamic monitoring may be helpful in selected patients. If
fluid resuscitation does not promptly restore arterial pressure and
Exercise
urine output, intravenous inotropic support is important. After
Although exercise programs do not alter mortality among patients initial stabilization, standard chronic heart failure pharmaco-
with congestive heart failure, physical conditioning improves logic therapy and volume status should be optimized. This may
exercise capacity and the sense of well-being. It is recommended require upward titration of medication doses at a rate that does
that patients with chronic heart failure engage in regular exercise not cause hypotension and disturbance of end-organ perfusion.
to maintain skeletal fitness and muscle tone. For many patients, a Before discharge patient education is essential, and early outpa-
monitored exercise program is important adjunctive therapy. tient follow-up is critical to assure smooth transition from hospi-
tal management to chronic outpatient care.
Atrial Fibrillation in Heart Failure
Abbreviations
Approximately 20% of heart failure patients experience atrial
fibrillation, and in those in whom it occurs a worse prognosis ACC American College of Cardiology
is observed. Atrial fibrillation results in loss of the atrial aug- ACE angiotensin-converting enzyme
mentation of ventricular filling and excessive tachycardia, with AHA American Heart Association
ARB angiotensin II receptor blocker
consequent impairment of systolic and diastolic function and
CRT cardiac resynchronization therapy
reduced stroke volume. Its other major effect is an increased risk ECG electrocardiography
of atrial thrombosis and peripheral thromboemboli. Two broad EF ejection fraction
approaches to atrial fibrillation treatment have been restoration ICD implantable cardioverter-defibrillator
of sinus rhythm and management of chronic atrial fibrillation MI myocardial infarction
with rate control and anticoagulation. These two management NYHA New York Heart Association
strategies have been found to have equivalent clinical outcomes. RAAS renin-angiotensin-aldosterone system
Given the difficultly of achieving long-term maintenance of
sinus rhythm after cardioversion, the preferred approach for most Clinical Trials
patients with atrial fibrillation and systolic dysfunction is rate
control and anticoagulation. CHARM Candesartan in Heart Failure Assessment of
Assessment of the adequacy of rate control may require evalu- Reduction in Mortality and Obesity
CIBIS II The Cardiac Insufficiency Bisoprolol Study II
ation with Holter monitor and stress testing. β-Blockers are more
COMET Carvedilol or Metoprolol European Trial
effective than digoxin for achieving rate control, especially dur- COPERNICUS Effect of Carvedilol on Survival in Severe Chronic
ing exercise. If pharmacologic rate control cannot be achieved Heart Failure
atrioventricular node ablation may be necessary. Anticoagulation ELITE Evaluation of Losartan in the Elderly
should be maintained unless there are serious risks for hemor- ELITE II Losartan Heart Failure Survival Study
rhagic complication in individual patients. MERIT-HF Metoprolol Cr/Xl Randomized Intervention Trial
in Congestive Heart Failure
PRAISE I, II Prospective Randomized Amlodipine Survival
Acute Heart Failure Evaluation 1, 2
Clinical decompensation requiring hospitalization is a constant RADIANCE Withdrawal of Digoxin From Patients With
risk in chronic heart failure. Initial evaluation of acute heart fail- Chronic Heart Failure Treated With Angiotensin-
Converting Enzyme Inhibitors
ure should focus on determining adequacy of systemic perfu-
RALES Randomized Aldactone Evaluation Study
sion, assessment of volume status, identification of precipitating SHEP Systolic Hypertension in Elderly Program
factors, and assessment of ventricular function. If dyspnea is the V-HeFT I, II, III Vasodilator-Heart Failure Trial 1, 2, 3
principal symptom and its cause is unclear after physical exam, WASH Warfarin/Aspirin Study in Heart Failure
history, chest radiography, and ECG, then brain natriuretic pep- WATCH Warfarin and Antiplatelet Trial in Chronic Heart
tide measurement may be useful to help distinguish cardiac from Failure

Mechanical Ventricular Assist Devices
JOHN A. SCHIRGER, MD, SOON J. PARK, MD,
and SUDHIR S. KUSHWAHA, MD
CHF remains a significant burden in terms of both morbidity
and mortality in the United States and in the developed world.
While there have been remarkable advances in pharmacologic
therapy, leading to decreased hospitalization rates, improved
myocardial remodelling, and prolonged life, there remains a sig-
nificant portion of the heart failure population that progresses
to end-stage heart failure. CRT-D devices have been shown to
improve exercise capacity and quality of life in some of these
patients; however, there remains a significant number of patients
whose options include cardiac transplantation or LVAD therapy.
Cardiac transplantation is the gold standard in therapy for this
population, but it is available to a limited number of patients due
to organ shortages.
LVAD therapy has emerged as an option in several clinical
settings (Figures 96.1 and 96.2). It can be used to bridge patients
to cardiac transplantation, the so called BTT strategy. It can
also be utilized as DT in patients who are ineligible for cardiac
transplantation. A new category of patient is emerging, the BTD
patient. These individuals may not be transplant candidates at the
time of implantation of LVAD but may become eligible for trans-
plant sometime in the future; examples include reversal of sec-
ondary pulmonary hypertension, improved kidney function, and
optimized nutritional and metabolic state. The concept of bridge
to recovery remains controversial and it is a subject of ongoing
investigation.
While LVAD and mechanical support have been availa-
ble clinically for many years, they have often been utilized in
so-called “salvage” situations or primarily as a BTT. In 2001, the
results of the landmark REMATCH trial were published, dem-
onstrating that LVAD therapy was superior to medical therapy in
Abbreviations and acronyms are expanded at the end of this chapter.
874
96
advanced heart failure patients in terms of survival. Nevertheless,
despite these early encouraging results, after two years the sur-
vival curves with these two therapies were closely approximated
and dismal. The primary cause of morbidity and mortality with
LVAD therapy at that time was infection.
In the post-REMATCH era, improvements in surgical tech-
nique, technical advances in device design, and medical manage-
ment have resulted in improved survival. Major limitations with
first generation LVADs included the size of the device, and the
fact that the mechanically these pulsatile devices were usually
sustainable for less than two years.
Newer-generation continuous flow devices have been shown
to be successful in the BTT setting and superior to older-
generation devices in survival at 2 years. Continuous flow will
likely have increased longevity compared to the older genera-
tion pulsatile flow pumps, and initial studies suggest that con-
tinuous flow is not detrimental to end-organ function in terms
of renal and hepatic function. Additional studies are needed to
evaluate the effects of continuous flow on the neurovasculature,
gastrointestinal tract, and other end-organ systems. Infectious
complications have decreased, but gastrointestinal bleeding
remains a problem post implantation, with up to 40% of patients
suffering gastrointestinal bleeding in the first year after implan-
tation. This complication may be related to breakdown of von
Willebrand factor, though further studies are required to define
this phenomenon.
• The REMATCH trial showed survival benefit vs medical therapy
in advanced NYHA IIIb CHF in patients optimally medically
managed including inotropic therapy.
• Newer-generation continuous axial flow LVAD therapy is superior
to older-generation pulsatile flow LVAD therapy, and also improves
quality of life and functional capacity.
 96 Mechanical Ventricular Assist Devices
Figure 96.1. Example of a left ventricular assist device (LVAD)
implanted in the chest. (Previously published. See Credit Lines section.)
• Infectious complications have decreased but right heart failure
and gastrointestinal bleeding (possibly related to alterations in
von Willebrand factor) remain challenges with newer generation
LVAD therapy.
Selection for patients for LVAD consideration is often done
in tandem with the heart transplantation team. Patients who are
being considered for heart transplantation and LVAD therapy
and are thought to be ineligible for heart transplantation can be
considered for DT LVAD therapy. The INTERMACS criteria
are useful to consider in evaluating patients for LVAD therapy.
INTERMACS criteria for advanced heart failure are defined as
follows:
Level 1: critical cardiogenic shock
Level 2: progressive decline and heart failure despite inotropic therapy
Level 3: stable but inotropic-dependent patients
Level 4: recurrent, advanced heart failure
Level 5: exertion intolerance
Level 6: exertion-limited CHF
Level 7: advanced NYHA class III symptomatology
It is important to note that patients with critical cardiogenic
shock or INTERMACS level 1 should not typically be consid-
ered candidates LVAD placement; primarily INTERMACS lev-
els 2 through 4 at this time should be considered.
Any patient who is being considered for chronic inotropic
therapy may be considered for LVAD therapy. In addition to
INTERMACS criteria, decreased EF, 25% or less, an enlarged
875
Figure 96.2. Left ventricular assist device (LVAD) with controller and
external power source. (Previously published. See Credit lines section.)
.
left ventricle, a peak VO2 of 12 mL/kg/min or less and NYHA
class IV symptoms for greater than 30 days or IABP or inotrope
dependence are criteria for DT LVAD therapy. One of the major
complications after placement of LVAD includes right heart fail-
ure; this complication is best predicted by elevated creatinine and
bilirubin. Specialty consultation should always be undertaken at
experienced centers when the question of LVAD therapy arises.
If a patient is a candidate for heart transplant, then LVAD
can be utilized as BTT. In general, patients with enlarged left
ventricles and elevated filling pressures, with an ischemic sub-
strate, with a history of ventricular dysrhythmias, and with a pro-
jected long wait time (such as patients with blood type O) are
good candidates for LVAD placement. Increasing evidence sug-
gests that patients can undergo successful BTT despite increased
immunosensitization.
Contraindications to LVAD placement include INTERMACS
level of 1 and poor neurologic status. In general, nonsystolic heart
failure is considered a contraindication. Coexisting illness with
life expectancy of less than two years is a general contraindica-
tion as is active, severe bleeding or a bleeding diathesis. Severe
right ventricular dysfunction or multiorgan failure is considered
a contraindication as well.
• LVAD therapy is successful as BTT and should be considered in
patients with low EF, ischemic substrate, hypotension, ventricular
arrhythmias, severe exercise limitations, and long projected wait
times.
876 IX Cardiomyopathy and Heart Failure
• LVAD as DT can be considered . with NYHA IV symptoms, EF less
than or equal to 25%, Peak VO2 of 14 mL/kg/min or less, with opti-
mal tolerated medical therapy and not considered for transplant.
• LVAD should not be placed in critically ill INTERMACS level
1 patients as salvage therapy, in patients with unclear neurologic
status, or patients with noncardiac disease processes that decrease
life expectancy to less than 2 years.
Risk factors prior to left ventricular assist device placement
for poor outcomes including longer hospital stay, right heart
failure, and mortality are delineated by the Leitz-Miller crite-
ria. These criteria include nine preoperative factors that predict
90-day in-hospital mortality and are weighted by points:
Platelets, 148 × 103/μL or less, 7 points
Albumin, 3.3 g/dL or less, 5 points
International normalized ratio exceeding 1.1, 4 points
Vasodilator therapy, 4 points
Mean pulmonary artery pressure of 25 mm Hg or less, 3 points
Aspartate aminotransferase exceeding 45 U/L, 2 points
Hematocrit of 34% or less, 2 points
Blood urea nitrogen exceeding 51 mg/dL, 2 points
Lack of intravenous inotropic support, 2 points
Patients are stratified into low (≤16) and high (≥17) risk groups.
Patients are often admitted to the cardiac care unit to try to
optimize these hemodynamic and biochemical criteria prior to
LVAD placement. There is no prospective study to date that dem-
onstrates the benefit of this hemodynamic optimization, although
it is accepted clinical practice. Hemodynamic optimization occurs
often with IABP and with inotropic therapy combined. There
is now some suggestion that improvement in renal parameters
leads to improved renal outcomes three months postoperatively.
In general, goals to be attained include a right atrial pressure of
less than 12 to 14 with a cardiac index of 2.2 to 2.4 L/min/kg
and improvements in blood urea nitrogen, creatinine, and liver
function tests.
Complications postoperatively include right heart failure and
dialysis, as well as embolic stroke and bacteremia and sepsis,
although this is less common. Use of current devices is as suc-
cessful several months post-transplant as inotropic therapy is in
terms of bridging people to transplant. With DT, LVAD survival
rates at two years approach 60% in a population that otherwise
have a survival rate in the range of 10%. While aortic insuffi-
ciency had been considered a contraindication, now with uti-
lization of the Park stitch, LVAD can be placed in the setting
of aortic insufficiency, but this should be done by experienced
surgeons in experienced centers.
• Risk factors for poor outcomes include increased blood urea nitro-
gen, decreased hematocrit, increased aspartate aminotransferase,
increased right atrial pressure, decreased pulmonary artery pressure,
decreased albumin and increased international normalized ration.
• Patients are admitted to the cardiac care unit prior to LVAD implant
of hemodynamic and nutritional optimization.
• In experienced hands, the Park stitch allows LVAD implant in
some situations with aortic insufficiency, generally considered a
contraindication to LVAD implant.
In summary, LVAD offers opportunities for BTT in patients
with advanced heart failure and dilated, enlarged left ventricles
especially with ischemic substrate or arrhythmias and a long wait
time. A number of patients who are initially not BTT candidates
can be served with LVAD as a bridge to candidacy in, for exam-
ple, reversal of pulmonary hypertension with LVAD treatment.
Furthermore, LVAD is an option for patients with NYHA class IV
symptoms with an EF of less than 25% on maximally-tolerated
medical therapy and especially INTERMAC levels 2 through
4 without obvious contraindications. While LVAD therapy is
not helpful in treating isolated right ventricular failure, newer
RVADs and the total artificial heart provides treatment options
with right ventricular failure. LVAD therapy provides treatment
options to prolong duration and quality of life in patients with
end-stage CHF who previously had very limited options availa-
ble. The field provides ample opportunities for ongoing research
to continue to benefit this group of patients.
Abbreviations
BTD bridge to decision
BTT bridge to transplant
DT destination therapy
EF ejection fraction
CHF congestive heart failure
CRT-D cardiac resynchronization therapy defibrillator
IABP intra-aortic balloon pump
LVAD left ventricular assist device
NYHA New York Heart Association
RVAD
. right ventricular assist device
VO2 oxygen consumption
Names of Clinical Trials
INTERMACS Interagency Registry for Mechanically Assisted
Circulatory Support
REMATCH Randomized Evaluation of Mechanical Assistance
for the Treatment of Congestive Heart Failure
 97
Myocarditisa
LESLIE T. COOPER JR, MD, and LORI A. BLAUWET, MD
Introduction
Myocarditis is defined histologically as inflammation of the myo-
cardium associated with injury to cardiac myocytes (Figure 97.1).
Common viral infections are the most frequent cause of myo-
carditis, but other pathogens, hypersensitivity reactions, and
systemic and autoimmune diseases should be considered and
excluded during the initial evaluation (Table 97.1). Patients with
myocarditis present with a range of symptoms varying from mild
dyspnea and fatigue to cardiogenic shock or even sudden death.
The diagnosis of probable myocarditis can usually be made from
clinical findings and noninvasive imaging, but endomyocardial
biopsy may be considered in certain cases.
Occasionally, the physiologic manifestation of acute myocar-
ditis is more suggestive of hypertrophic or restrictive cardiomy-
opathy than of pure DCM. Important clinicopathologic variants
include the endocardial eosinophilic and fibrotic diseases and
pericarditis associated with focal or diffuse epicarditis (also
called myopericarditis). Typically, in acute lymphocytic myo-
carditis presenting as DCM, LV function improves over 1 to 6
months with standard heart failure treatment. However, in many
patients a cardiotropic virus is not cleared or persistent inflam-
mation develops and leads to chronic DCM. If progressive heart
failure develops despite guideline-based medical management,
mechanical cardiac assist devices or heart transplant may be
indicated and the risk of death is significantly increased.
a
Portions previously published in Blauwet LA, Cooper LT. Myocarditis.
Prog Cardiovasc Dis. 2010 Jan-Feb;52(4):274–88; Cooper LT Jr.
Myocarditis. N Engl J Med. 2009 Apr 9;360(15):1526–38; and Cooper
LT Jr. Acute heart failure due to fulminant and giant cell myocarditis.
Herz. 2006 Nov;31(8):767–70. Used with permission.
Abbreviations and acronyms are expanded at the end of this chapter.
Etiology of Acute Myocarditis
Since the 1960s, seroepidemiologic studies have linked entero-
viral infections, most commonly from CVB3, with acute cardi-
omyopathy. With the advent of molecular biologic techniques,
the ability to diagnose viral infection has improved and more
than 20 viruses have been associated with acute and chronic
DCM. Adenovirus, parvovirus B19, human herpesvirus 6, and
hepatitis C virus (in Japan) have been identified with increasing
frequency in pediatric and adult populations (Table 97.2). Acute
cardiomyopathy associated with enteroviral infection is often
associated with a lymphocytic infiltrate or altered expression
of immunologic markers on cardiac myocytes, suggesting acute
myocarditis.
• CVB3 (an enterovirus) is associated with lymphocytic myocarditis.
Although the possible causes of acute myocarditis are many,
viral cardiomyopathy is the most commonly identified form of
acute myocarditis in North American adults. The diagnosis of
viral cardiomyopathy may be suggested by an antecedent viral
prodrome or by acute changes in serologic titers of antiviral
immunoglobulins. However, the specificity of acute changes in
viral serologic titers is low, because similar serologic changes
have been observed in unaffected family members of patients
with myocarditis. Indeed, during outbreaks of CVB3 infection,
only 3.5% to 5% of patients have symptoms suggestive of myo-
carditis. The diagnosis may be confirmed by PCR from a biopsy
sample of fresh frozen myocardium, but this is not routinely per-
formed at most medical centers.
Although much clinical research in the 1980s and 1990s
examined the effects of the immune response to viral infection as
a cause of cardiomyopathy, later research has focused on the role
of direct viral injury. Viral infection may cause myocardial injury,
lead to cardiomyopathy, and induce myocyte death. Picornavirus
877
878 IX Cardiomyopathy and Heart Failure

Figure 97.1. A, Normal myocardium is shown in longitudinal section (left) and cross section (right). B, Lymphocytic myocarditis is shown with a
mixed inflammatory infiltrate and associated myocyte necrosis. (See “Credit Lines” section.)

protease 2A can inhibit host cell synthesis, and CVB3 protease
2A can cleave dystrophin, which may lead to cardiomyopathy
independently of any immune response.
Both innate and adaptive immune responses have key roles
in the development of myocarditis. Cytokines, such as tumor
necrosis factor and interleukins, mediate further cardiac injury.
Antibodies and autoantibodies to numerous cardiac antigens,
including cardiac myosin and troponin I, are associated with
this inflammatory reaction and can impair cardiac myocyte con-
tractility. Released cellular products, such as major basic protein
from eosinophils, can lead directly to myocyte cell death. The
viral pathogen is cleared and the immune system is downregu-
lated in most patients with viral myocarditis, but in a minority of
patients, the virus is not cleared, resulting in persistent myocyte
damage.
A proposed 3-phase model characterizes the progression of
acute viral infection to DCM. In the first phase, acute infection of
cardiac myocytes results in myocyte cell death and activation of
the innate immune response, including interferon-γ, natural killer
cells, and nitric oxide. Antigen-presenting cells phagocytose
released viral particles and cardiac proteins and migrate from
the heart to regional lymph nodes. The second phase consists of
an adaptive immune response in a subset of patients. Antibodies
to viral proteins and to some cardiac proteins (often including
cardiac myosin and receptors such as β1-adrenergic receptors)
are produced along with a proliferation of T-helper cells. In the
third phase, the immune response is downregulated, with fibrosis
replacing a cellular infiltrate in the myocardium. Under neurohu-
moral stimulation and hemodynamic stress, the ventricles dilate,
leading to chronic DCM.
In acute myocarditis, inflammation may have the beneficial
effect of complete viral clearance. This may explain the largely
negative results of treatment trials aimed at altering this acute
immune response. Most patients with acute myocarditis have
 97 Myocarditis 879

Table 97.1. Common Causes of Myocarditis

Associated Disorder or Agent Clinical Clues Diagnostic Method
Virus (eg, coxsackievirus B) Flulike prodrome Endomyocardial biopsy
Acute rheumatic fever Jones criteria Throat culture; antistreptolysin O titer
Lyme disease History of tick bite Serology
Anthracyclines Previous cancer treatment Clinical; endomyocardial biopsy
Chagas disease Travel to Central or South America Serology
Peripartum cardiomyopathya Last trimester or first 6 mo post partum Clinical
a
Reviewed in Chapter 80 (“Pregnancy and the Heart”).

a mild illness with partial or complete recovery of ventricular Physical Examination

function. Patients who have no response to usual heart failure
The physical examination findings in patients with acute myo-
care, or who have progressive heart block or ventricular arrhyth-
carditis are similar to those of acute decompensated heart fail-
mias with acute cardiomyopathy, may have a distinct pathologic
ure. Patients may show signs of fluid overload with rales, elevated
disorder such as granulomatous myocarditis or GCM, and heart
jugular venous pressure, hepatomegaly, and lower extremity
biopsy should be considered.
edema. The patients are usually tachycardic. The cardiac apex
In contrast, patients who have evidence of persistent inflamma-
may be laterally displaced, suggesting cardiomegaly. The first
tion in the myocardium or persistent viral infection with chronic,
heart sound may be soft and there may be a third or fourth heart
symptomatic congestive heart failure usually do not improve if
sound (or both).
they are already receiving optimal standard heart failure treat-
Certain physical examination findings suggest specific causes
ment. In these patients with chronic viral or autoimmune DCM,
of myocarditis:
therapy aimed at eliminating the virus or altering the immune
response may be beneficial. The management of chronic myo- 1. Enlarged lymph nodes—systemic sarcoidosis
carditis is an area of active clinical research. 2. A pruritic, maculopapular rash—hypersensitivity reaction, often to a
drug or toxin
3. Sustained ventricular tachycardia or new heart block with rapidly
Clinical Presentation and Diagnosis progressive congestive heart failure—GCM
The clinical presentation of patients with myocarditis is highly
variable, ranging from subclinical disease to fulminant heart Biomarkers of Cardiac Injury
failure. In most case series, there is a slight male predominance,
Cardiac enzyme elevations occur in a minority of patients with
which may be mediated by sex hormones. Estrogenic hormones
acute myocarditis and can help confirm the diagnosis. Standard
have been shown to protect against viremia and viral infectivity
markers of myocardial damage, including troponin T and
of cardiac myocytes and to decrease the inflammatory response,
CK-MB, have a high specificity but limited sensitivity in the
while testosterone inhibits the anti-inflammatory response. Of
diagnosis of myocarditis. Clinical and experimental data sug-
the 111 patients enrolled in the Myocarditis Treatment Trial,
gest that cardiac troponin T levels are increased more frequently
62% were male. In a cardiac MRI study of 65 patients with acute
than CK-MB levels in patients with myocarditis. Other serum
myocarditis, myocardial fibrosis was more frequent in men and
immunologic biomarkers, including cytokines, complement, and
in patients younger than 40 years. Myocarditis most commonly
cardiac autoantibodies, have not been prospectively validated
affects young persons. The reported mean age of patients with
for sensitivity and specificity in the diagnosis of myocarditis.
acute lymphocytic myocarditis ranges from 20 to 51 years; the
Elevated levels of serum soluble Fas and soluble Fas ligand have
median age for patients with GCM is 42 years.
been associated with a poor prognosis for patients with severe
Patients may report fatigue, dyspnea, decreased exercise toler-
acute myocarditis; elevated serum levels of interleukin 10 predict
ance, palpitations, precordial chest pain, or syncope. Chest pain
increased mortality among patients with fulminant myocarditis.
may mimic typical angina and be accompanied by ECG changes,
The erythrocyte sedimentation rate, C-reactive protein level,
including ST-segment elevation. Chest pain may also mimic
and leukocyte count may be elevated, and aspartate aminotrans-
pericarditis, suggesting pericardial involvement. Cardiac rhythm
ferase, alanine aminotransferase, and lactase dehydrogenase
disturbances may include new atrial or ventricular arrhythmias
levels may be increased in acute myocarditis, but these are non-
or complete heart block. Of the 3,044 adult patients with sus-
specific signs of inflammation.
pected myocarditis in the European Study of Epidemiology and
Treatment of Cardiac Inflammatory Diseases, 72% had dyspnea,
32% had chest pain, and 18% had arrhythmias. Patients with Electrocardiogram
myocarditis who present with syncope are at particularly high The ECG often shows sinus tachycardia with nonspecific
risk for death or need for a transplant. ST-segment and T-wave abnormalities in patients with acute
• Myocarditis should be suspected in a previously healthy young myocarditis. The presence of widened QRS and Q waves has
person who presents with new-onset heart failure symptoms been associated with greater risk of heart transplantation and
and DCM. death. Occasionally, the ECG changes are suggestive of acute
• Most patients report a flulike prodrome. myocardial infarction or pericarditis; intraventricular conduction
delay is common. A small proportion of patients have various
• The mean age of affected persons is 20 to 51 years.
degrees of AV block. In the Myocarditis Treatment Trial, pace-
• Young men are at greatest risk for severe myocardial injury.
makers were implanted in approximately 1% of subjects because
• Myocarditis is an important cause of chronic DCM. of high-grade AV block. Ventricular arrhythmias may also be
Table 97.2. Viral Myocarditis
Cause Clinical Features Pathologic Findings Comments
Coxsackievirus See text Lymphocytic infiltrate with myocyte necrosis Sensitivity of endomyocardial biopsy is about 35%
Influenza Tachycardia, ECG abnormalities, dyspnea, anginal chest Myocarditis present in one-third of fatal cases; Treat type A with oseltamivir or zanamivir
pain, congestive heart failure, complete heart block, biventricular dilatation, subendocardial and
and death subepicardial hemorrhage, and mononuclear
perivascular inflammation
Cytomegalovirus Symptomatic disease rare, pericardial effusion Focal lymphocytic infiltration and fibrosis
occasionally present
Poliomyelitis Often found in fatal cases; cardiovascular collapse, heart Immunization effective; maintain proper
failure, and pulmonary edema oxygenation and pulmonary function
Infectious mononucleosis ECG changes common, cardiac symptoms unusual, Myocardial infiltrates of atypical lymphocytes and Consider treatment with corticosteroids
congestive heart failure or death rare necrosis
Human immunodeficiency virus Pericarditis, arrhythmias, ECG abnormalities, and dilated Lymphocytic myocarditis, opportunistic myocardial Consider adverse drug effects, opportunistic
cardiomyopathy infections, and ventricular dilatation infections, and malignancy
Hepatitis C Usually transient; ECG abnormalities (bradycardia, Often nonspecific histological changes
ventricular premature complexes, and ST-segment and
T-wave abnormalities); congestive heart failure and
sudden death in severe cases
Mumps Clinical and cardiac involvement uncommon, ST-segment Cardiac dilatation and hypertrophy, mural thrombi,
and T-wave abnormalities more frequent interstitial fibrosis, infiltration of mononuclear cells,
and focal necrosis; possible relationship between
maternal mumps and fetal endocardial fibroelastosis
Rubeola Transient ECG abnormalities, congestive heart failure Perivascular lymphocytic infiltrate
rare
Varicella Rare bundle branch block, conduction defects, heart Intranuclear inclusion bodies within myocardial cells,
failure, and sudden death; nonsustained ventricular interstitial edema, cellular infiltrates, and myonecrosis
tachycardia and fibrillation
Variola and vaccinia Myocarditis rare but may be fatal; myocarditis may follow Mononuclear infiltrate, interstitial edema, and necrosis Variola now eradicated and immunization no
vaccination by 2 wk longer recommended; severe myocarditis after
vaccination has responded to a short course of
corticosteroids
Arbovirus (chikungunya, dengue) Myocarditis frequent; chest pain, dyspnea, palpitations, Embolization
murmurs, gallops, and cardiomegaly; ST-segment and
T-wave abnormalities, supraventricular and ventricular
arrhythmias, and sudden death; chronicity
Respiratory syncytial viruses Rare congestive heart failure, complete heart block, and
arrhythmias
Herpes simples virus Chronic interstitial inflammation and fibrosis Treat with adenine arabinoside or acyclovir, but
effectiveness not proved
Adenovirus Myocarditis rare Dilated right and left ventricles, paucicellular infiltrate
Yellow fever virus Hepatitis, gastrointestinal tract bleeding, cardiovascular Pericardial petechial hemorrhages and myocyte
collapse, and bradycardia inappropriate to the fever degeneration
Rabies Rare tachycardia, gallop rhythm, and hypotension Diffuse interstitial infiltrate, myocardial necrosis
Abbreviation: ECG, electrocardiographic.
Previously published. See “Credit Lines” section.
 97 Myocarditis
present, but they occur more commonly in cardiac sarcoidosis
and GCM. Although an ECG is widely used as a screening tool,
its sensitivity for myocarditis is only 46%.
Echocardiography
There are no specific echocardiographic features of myocardi-
tis, since patterns consistent with dilated, hypertrophic, restric-
tive, and ischemic cardiomyopathy have all been described
in histologically proven myocarditis. LV systolic dysfunction
is common. RV dysfunction is uncommon. Segmental wall
motion abnormalities are often present and may mimic myo-
cardial infarction. Diastolic relaxation abnormalities are com-
mon, and a restrictive pattern is frequently present. Pericardial
effusions are common. One study has shown that patients with
acute myocarditis typically have marked LV dilatation and nor-
mal wall thickness, while patients with fulminant myocarditis
tend to present with small cardiac chambers and ventricular
hypertrophy.
Certain echocardiographic variables may predict prognosis in
acute myocarditis. RV dysfunction may be an independent pre-
dictor of death or need for cardiac transplant. In a study of 23
patients with biopsy-confirmed myocarditis, the likelihood of
death or heart transplant was greater for patients with abnormal
RV function than for patients with normal RV function. After 2
years, 60% of patients with RV dysfunction had died or required
transplant, compared with none of the patients with normal RV
function (P = .03). In that study, multivariate analysis revealed
that RV dysfunction as quantitated by RV descent was the most
powerful predictor of death or cardiac transplant. In another
study, contractile reserve that was assessed by dobutamine ech-
ocardiography was used to predict LV functional recovery in
22 patients who had new-onset DCM, some of which may have
resulted from myocarditis. Baseline variables that were signifi-
cantly predictive of follow-up LV ejection fraction were deceler-
ation time (r = 0.69, P<.001), wall motion score index (r = −0.63,
P = .002), LV mass (r = 0.56, P = .008), and LV ejection fraction
after dobutamine (r = 0.84, P<.001).
• Echocardiographic findings are not specific for acute myocarditis
but are useful to exclude other known causes of heart failure.
• LV systolic and diastolic function are both common.
• RV dysfunction is a powerful predictor of death or need for cardiac
transplant.
Cardiac Magnetic Resonance
Cardiac MRI is becoming an increasingly important tool for the
diagnosis of myocarditis. Serial T1-weighted images with gad-
olinium have been used to visualize myocardial injury in myo-
carditis and to track its progression. The myocardial delayed
enhancement technique provides additional diagnostic value,
with improved sensitivity when combined with T2-weighted
sequences. Histopathologic evaluation of biopsy samples from
biopsies directed by cardiac MRI with delayed enhancement
showed active myocarditis in 19 of 21 patients. In contrast, only 1
in 7 patients showed active myocarditis in samples from biopsies
that were guided by cardiac MRI without myocardial delayed
enhancement.
Cardiac MRI can be used to evaluate markers of tissue injury,
including intracellular and interstitial edema, hyperemia and cap-
illary leakage, and necrosis and fibrosis. According to a recent
paper by the International Consensus Group on Cardiovascular
881
Magnetic Resonance in Myocarditis, cardiac MRI should be
performed for symptomatic patients who have clinical suspi-
cion of myocarditis and for whom MRI results will affect clini-
cal management. Imaging criteria (the “Lake Louise Criteria”)
have been proposed (Box 97.1). Myocardial inflammation can be
predicted with a diagnostic accuracy of 78% when at least 2 of
the 3 criteria are present, but the diagnostic accuracy decreases
to 68% if only delayed post–gadolinium-enhanced imaging is
performed.
Cardiac MRI may differentiate ischemic from nonischemic
cardiomyopathy. In 1 study, gadolinium-enhanced cardiac MRI
was performed in 90 patients with heart failure and LV sys-
tolic dysfunction, 63 of whom had idiopathic DCM. All the
patients with ischemic cardiomyopathy had either subendocar-
dial or transmural enhancement. In contrast, the DCM group
had 3 patterns of enhancement: no enhancement, myocardial
enhancement indistinguishable from that of patients with coro-
nary artery disease, and patchy or longitudinal striae of mid-wall
enhancement. This study and others suggest that diffuse or het-
erogeneous involvement in the lateral wall, subepicardium, or
mid-myocardium or combined enhancement is highly suggestive
of myocarditis.
Because of its availability and low risk, cardiac MRI is being
used with increasing frequency as a diagnostic tool in suspected
acute myocarditis. Current limitations of cardiac MRI include the
Box 97.1. Proposed Cardiac MRI Diagnostic Criteria
for Myocarditis (“Lake Louise Criteria”)
In the setting of clinically suspected myocarditis,
cardiac MRI findings are consistent with
myocardial inflammation if at least 2 of the
following criteria are present:
1. Regional or global myocardial signal intensity
increase in T2-weighted images
2. Increased global myocardial early enhancement
ratio between myocardium and skeletal muscle in
gadolinium-enhanced T1-weighted images
3. There is at least 1 focal lesion with nonischemic
regional distribution in inversion-recovery
prepared gadolinium-enhanced T1-weighted
images (delayed enhancement)
Cardiac MRI study is consistent with myocyte
injury or scar caused by myocardial inflammation
(or both) if criterion 3 is present
A second cardiac MRI study 1 to 2 weeks after the
initial cardiac MRI study is recommended if either
of the following is true:
None of the criteria are present, but the onset of
symptoms is very recent and there is strong clinical
evidence for myocardial inflammation
One of the criteria is present
The presence of left ventricular dysfunction
or pericardial effusion provides additional,
supportive evidence for myocarditis
Abbreviation: MRI, magnetic resonance imaging.
Previously published. See “Credit Lines” section.
882 IX Cardiomyopathy and Heart Failure

inability to differentiate between myocarditis that may require
specific therapy, such as granulomatous or GCM, and myocardi-
tis due to specific diseases, such as idiopathic hypereosinophilic
syndrome. The diagnostic features on MRI decrease over several
months after symptom onset, and therefore the diagnostic utility
of MRI depends on the chronicity of symptoms. Not all patients
are candidates for cardiac MRI, because they may have claus-
trophobia, implantable devices, tachyarrhythmias, significant
renal dysfunction, or hemodynamic instability. The presence of
delayed gadolinium enhancement predicted cardiovascular death
in one retrospective case series.
• Cardiac MRI is useful for assessing myocardial inflammation in
patients with suspected myocarditis.
• Cardiac MRI may differentiate ischemic from nonischemic
cardiomyopathy.
• Use of cardiac MRI to guide endomyocardial biopsy may improve
sensitivity.
Endomyocardial Biopsy
The criterion standard for the diagnosis of acute myocarditis is
transvenous RV EMB (Figure 97.2). Although the 79% specifi-
city of EMB for lymphocytic myocarditis is high, the sensitiv-
ity is only 35% when compared with the clinical standard of
improved LV function over time. An EMB sample positive for
myocarditis unequivocally establishes the diagnosis. However,
owing to widespread sampling error, the absence of histologic
evidence should not preclude the diagnosis of myocarditis in the
appropriate clinical settings.
The proportion of positive RV biopsy findings in specimens
from patients with suspected myocarditis averages 10%. In the
Myocarditis Treatment Trial, only 214 of 2,233 patients with
heart failure and suspected myocarditis had diagnostic biopsy
findings. The relatively low proportion of diagnostic findings on
EMB is likely a result of the small size and number of biopsy
specimens obtained, the insensitivity of routine histologic tech-
niques for detecting inflammation (Dallas criteria) (Box 97.2),
and considerable intraobserver variability in the identification of
inflammatory infiltrates.
Routine use of EMB is also limited by the complications of
this invasive technique. Historically, the risks are about 1 in
1,000 for risk of death and 1 in 250 for risk of perforation when
performed by an experienced operator. These risks are decreased
when smaller, more flexible bioptomes are used. Many clinicians
have questioned the role of routine EMB in the diagnosis of acute
lymphocytic myocarditis. EMB for uncomplicated acute DCM
is listed as a class IIb recommendation in the current American
College of Cardiology and American Heart Association scien-
tific statement (ie, it is not recommended for routine evalua-
tion of suspected myocarditis). This recommendation is based
in part on the evidence from the Myocarditis Treatment Trial,
which did not demonstrate a benefit for EMB-guided treatment
in acute myocarditis. In most cases of acute DCM, EMB findings
do not affect treatment, and therefore there is no need to obtain a
specific histologic diagnosis. However, in patients with a rapidly
deteriorating course, particularly if complicated by heart block
or ventricular arrhythmias, an EMB is the only way to diag-
nose GCM, a disorder that carries a poor prognosis but usually
responds to multidrug immunosuppression or heart transplant.

Figure 97.2. Right Ventricular Endomyocardial Biopsy. A, Diagram shows ease of access to right ventricular apex via right internal jugular
vein. B through D, Autopsy specimen was used to simulate biopsy sample procurement. B, Bioptome jaws are opened around a trabecula carnea. C,
Jaws are closed. D, Appearance of biopsy site (arrow). RA indicates right atrium; RV, right ventricle; TV, tricuspid valve. (Previously published. See
“Credit Lines” section.)
 97 Myocarditis
Box 97.2. Dallas Criteria for the Diagnosis of
Lymphocytic Myocarditisa
First biopsy
Active myocarditis (with or without fibrosis)
Borderline myocarditis (not diagnostic and requiring
further biopsy)
No evidence of myocarditis
Subsequent biopsies
Ongoing (persistent) myocarditis
Resolving (healing) myocarditis
Resolved (healed) myocarditis
a
The histologic diagnosis of active myocarditis requires an
inflammatory infiltrate with necrosis or degeneration (or both) of
adjacent myocytes without evidence of Chagas disease or features of
ischemic heart disease.
The role of EMB in the evaluation of patients with chronic
DCM is more controversial than its role in acute DCM. Expression
of major histocompatability antigens is a more sensitive marker
than the Dallas criteria for myocardial inflammation and can be
detected with immunoperoxidase-based stains for HLA-ABC
and HLA-DR antigens. PCR of viral RNA and DNA from EMB
specimens can be used to diagnose viral myocarditis and to
assess whether viral clearance occurs. In patients with biopsy-
proven viral myocarditis, viral persistence in the myocardium is
associated with progressive cardiac dysfunction, and spontane-
ous viral elimination is associated with significant improvement
in cardiac function. In virus-negative inflammatory cardiomyop-
athy, the use of cell-specific immunocytochemistry (CD45- or
CD3-positive T cells) can identify patients who have chronic
DCM and who may respond to immunosuppression. Because
of the potential to influence outcome with antiviral or immuno-
modulatory treatment in chronic DCM, there is renewed interest
in the use of EMB in combination with these novel methods of
tissue analysis to manage chronic DCM.
Treatment of Myocarditis
The Myocarditis Treatment Trial was a prospective, randomized,
double-blinded, placebo-controlled trial of prednisolone plus
cyclosporine or azathioprine for the treatment of biopsy-proven
lymphocytic myocarditis in patients with acute congestive heart
failure. There was no benefit to immunosuppression over placebo
in survival or improvement of LV function. The mean LV ejec-
tion fraction increased from 24% to 35% in both groups, and
approximately 56% of patients in both groups died or underwent
heart transplant within 4.5 years.
The IMAC trial evaluated the role of intravenous immuno-
globulin treatment and enrolled patients with acute DCM who
had an LV ejection fraction of less than 40% and heart failure
symptoms for less than 6 months. The trial showed an average
increase in LV function of 14% regardless of treatment. These
results were in agreement with the Myocarditis Treatment
Trial findings, which showed that immune modulation in most
cases of acute DCM does not provide additional benefit beyond
usual care.
In contrast, immunosuppression may benefit patients who
have chronic DCM that is unresponsive to standard heart failure
883
therapy. The TIMIC trial randomly assigned 85 patients who had
chronic stable DCM to receive either azathioprine and predni-
sone or placebo for 6 months. In 38 of the 43 patients treated with
immunosuppression, LV ejection fraction improved by more
than 10%; LV ejection fraction did not improve in any patient in
the placebo group. Patients treated with immunosuppression also
showed better clinical improvement, with significantly lower
average New York Heart Association classes at 6 months.
Most patients with acute cardiomyopathy improve with treat-
ment that includes the standard heart failure regimen of angio-
tensin-converting enzyme inhibitors or angiotensin receptor
blocking agents; β-blockers such as carvedilol, metoprolol suc-
cinate, or bisoprolol; and diuretics as needed. Heart block and
tachyarrhythmia may be treated with standard agents and, in
rare cases, pacemaker implantation. Treatment with nonsteroidal
anti-inflammatory drugs should be avoided in patients with LV
systolic dysfunction, since they may worsen the outcome. Patients
with profound circulatory shock may benefit from mechani-
cal circulatory support, including implantation of an LV assist
device as a bridge to transplant or recovery. Cardiac transplant is
reserved for patients with conditions that are refractory to opti-
mal medical therapy and mechanical circulatory support.
• Most patients with acute myocarditis respond well to standard
heart failure treatment.
• Routine treatment with immunosuppressive agents for acute viral
or lymphocytic myocarditis is not recommended.
• Immunosuppressive therapy may benefit patients who have chronic
DCM that is unresponsive to standard heart failure therapy.
Specific Forms of Myocarditis
Lymphocytic Myocarditis
The prognosis for myocarditis primarily depends on the cause of
the myocardial inflammation (Figure 97.3). A multivariate anal-
ysis demonstrated that predictors of death or need for heart trans-
plant after acute lymphocytic myocarditis include syncope, low
ejection fraction (<40%), left bundle branch block, and increased
pulmonary artery pressure.
Giant Cell Myocarditis
Idiopathic GCM is a rare and usually fulminant form of myo-
carditis. Diagnosis is based on the presence of multinucleated
1.0 Giant cell myocarditis
Proportion Surviving
Lymphocytic myocarditis
0.8
0.6
0.4
0.2
0.0
0 1 2 3 4 5
Years
Figure 97.3. Kaplan-Meier Survival Curves for Patients With Giant
Cell Myocarditis or Lymphocytic Myocarditis. (Previously published.
See “Credit Lines” section.)
884 IX Cardiomyopathy and Heart Failure

A B

Figure 97.4. Idiopathic Giant Cell Myocarditis. A, Diffuse endomyocardial inflammatory infiltrate has multinucleated giant cells in the absence
of granuloma original magnification, ×25). B, Widespread mixed inflammatory infiltrate has multinucleated giant cells and myocyte necrosis original
magnification, ×100).

giant cells associated with eosinophils and myocyte destruction
in the absence of granulomas (Figure 97.4). Unlike lympho-
cytic myocarditis, GCM generally causes progressive LV failure
complicated by arrhythmias. Of the 63 patients enrolled in the
multicenter Giant Cell Myocarditis Treatment Trial, 75% pre-
sented with congestive heart failure and 14% with ventricular
arrhythmia.
GCM is also associated with a worse prognosis than lympho-
cytic myocarditis; the median survival from the onset of symp-
toms is only 5.5 months, and the rate of death or transplant is
89% (Figure 97.3). With this poor prognosis, a biopsy-proven
diagnosis of GCM prompts consideration of immunosuppression
and early evaluation for cardiac transplant. Timely institution of
combination immunosuppressive therapy, including cyclosporine
and prednisone, prolongs time to death or transplant. Despite a
25% incidence of posttransplant recurrence of GCM detected by
biopsy, the 5-year survival after transplant is about 71%, which is
comparable to survival after transplant for cardiomyopathy.
Cardiac Sarcoidosis
Historically, cardiac sarcoidosis was thought to occur in about
5% of patients with clinical sarcoidosis, although autopsy
series have reported higher rates of about 25%. One study of
81 patients with biopsy-proven extracardiac sarcoidosis reported
that cardiac MRI demonstrated evidence of cardiac involvement
in 21 (26%) of these patients. Patients with isolated cardiac sar-
coidosis can present with ventricular tachycardia, heart block, or
congestive heart failure. Definitive diagnosis is by EMB, which
shows characteristic noncaseating granulomas (Figure 97.5).
However, the diagnosis can be inferred from a tissue diagnosis
of sarcoidosis from an extracardiac source and cardiomyopathy
of unknown origin. Transplant-free survival of persons with car-
diac sarcoidosis is similar to that of persons with lymphocytic
myocarditis, but the rates of syncope, pacemaker placement,
and automatic implantable cardioverter-defibrillator placement
are higher.
Acute Rheumatic Fever
Patients with acute rheumatic fever may present with pancardi-
tis that affects the endocardium, myocardium, and pericar-
dium. Although the incidence of acute rheumatic fever is low
in the United States, at about 2 cases per 100,000 persons, it
is much higher in Asia, Africa, and South America, with 100
cases per 100,000 persons. Acute rheumatic fever is one of the

A B

Figure 97.5. Cardiac Sarcoidosis. A, Well-formed granuloma with giant cells may be seen without myocyte necrosis original magnification,
×125). B, Follicular granuloma is present in cardiac sarcoidosis original magnification, ×400).
 97 Myocarditis 885

most important cardiovascular diseases in developing countries. Box 97.3. Possible Effects of Various Drugs on the
It is thought to result from an immunologic response to a pha- Heart
ryngeal infection with certain strains of group A streptococci Hypersensitivity myocarditis
and occurs in about 3% of persons with untreated streptococcal Acetazolamide
pharyngitis. p-Aminosalicylic acid
The risk of rheumatic fever in genetically susceptible individ- Amitriptyline
uals is related to the presence of certain virulent types of strep- Amphotericin B
tococci that induce a strong immune response to M-type mucoid Carbamazepine
Cephalosporins
proteins that encapsulate the organism. Patients with HLA-DR Chloramphenicol
1, 2, 3, and 4 haplotypes may be at increased risk of rheumatic Clozapine
fever. A more detailed discussion of rheumatic fever is presented Diphenylhydantoin
in CHAPTER 38 (“RHEUMATIC HEART DISEASE”). Diphtheria toxin
Horse serum
Hydrochlorothiazide
Hypersensitivity and Eosinophilic Myocarditis Indomethacin
Isoniazid
Drug-induced hypersensitivity reactions sometimes affect the Methyldopa
myocardium. Numerous medications, including antidepres- Penicillins
sants (tricyclics), antibiotics (penicillins, cephalosporins, and Phenindione
sulfonamides), and antipsychotics (clozapine) have been impli- Phenylbutazone
cated in hypersensitivity myocarditis (Box 97.3 and Table 97.3). Smallpox vaccine
Clinically, the presentation differs from that of lymphocytic Spironolactone
Streptomycin
myocarditis in that patients with hypersensitivity myocarditis
Sulfonamides
are generally older (mean age, 58 years) and are often taking Sulfonylureas
several medications. Patients commonly present with acute rash, Tetanus toxoid
fever, and liver function test abnormalities. ECG changes are Tetracycline
similar to those of lymphocytic myocarditis, with sinus tach- Toxic myocarditis
ycardia, nonspecific T-wave abnormalities, and ST-segment Amphetamines
elevation. Antihypertensives
Eosinophilic myocarditis may occur in association with sys- Antimony
temic diseases such as hypereosinophilic syndrome, Churg- Arsenicals
Strauss syndrome, and Löffler endomyocardial fibrosis (Box Barbiturates
Caffeine
97.4). These syndromes are commonly characterized by periph-
Catecholamines
eral eosinophilia and an infiltration of mature eosinophils in Cocaine
many organs, including the heart. Clinical manifestations of Cyclophosphamide
eosinophilic myocarditis include endocardial fibrosis, fibrosis of Emetine
the cardiac valves leading to regurgitation, biventricular conges- 5-Fluorouracil
tive heart failure, and formation of thrombi on the endocardial Immunosuppressives
surface. Eosinophilic myocarditis may also occur in association Lithium
Paraquat
with parasitic helminthic or protozoal infections, such as Chagas Phenothiazines
disease, toxoplasmosis, schistosomiasis, trichinosis, hydatid Plasmocid
cysts, and visceral larval migrans. Acute necrotizing eosino- Quinidine
philic myocarditis is an aggressive form of eosinophilic myocar- Rapeseed oil
ditis with rapid onset and a high mortality rate. Theophylline
In general, the treatment of eosinophilic myocarditis depends Dilated cardiomyopathy
on the underlying cause. Therapy with high-dose corticosteroids Amphetamines
may be beneficial for systemic disease, whereas surgical treat- Anthracyclines
ment may aid in endomyocardial fibrosis. Chloroquine
Cobalt
Cocaine
Lyme Myocarditis Ephedrine
Ethanol
Lyme disease, an infection by the spirochete Borrelia burgdor- Lithium
feri, may result in both acute and chronic myocarditis. One study
Endocardial fibrosis
of 207 children with early disseminated Lyme disease found Anthracyclines
that mild to fulminant myocarditis had occurred in 16% of the Busulfan
children, 14% of whom had high-grade AV block. Lyme myo- Ergotamine
carditis should be suspected in patients who have a history of Mercury
travel to regions where Lyme disease is endemic or a history of Methysergide
a tick bite. Clinical presentation may include transient or perma- Serotonin
nent heart block or cardiac arrhythmia. Although most patients Myocardial fibrosis
fully recover, Lyme carditis sometimes persists and may lead to Cyclosporinea
chronic heart failure.
a
The diagnosis of Lyme disease is confirmed by serologic test- In transplanted hearts only.
ing; however, this does not establish the diagnosis of myocardi- Previously published. See “Credit Lines” section.
tis. EMB may show a lymphocytic myocarditis with a prominent
886 IX Cardiomyopathy and Heart Failure

Table 97.3. Hypersensitivity Myocarditis

Drug Symptoms Mode of Death Pathology
Methyldopa Shortness of breath, malaise, headache, Sudden Myocarditis, vasculitis, hepatitis
fever, cerebrovascular accident
Sulfonamides (sulfadiazine, sulfisoxazole, Fever, shortness of breath Sudden Myocarditis, petechial hemorrhages, vasculitis,
sulfamethoxypyridazine, carbutamide) granulomas
Penicillin, ampicillin Rash, congestive heart failure Sudden Myocarditis, granulomas, pericarditis,
myocardial infarction
Phenylbutazone Shortness of breath, fever, rash, Sudden Myocarditis, hepatitis, myocardial giant
chest pain, hypotension cells, perivascular granulomas, fibrinoid
degeneration, pericarditis
Oxyphenbutazone Congestive heart failure Sudden Myocarditis
Chlortetracycline Fever, tachycardia Interstitial and perivascular infiltrates
Chloramphenicol Sudden Myocarditis, hepatitis
Streptomycin Chest pain, fever, rash Sudden Myocarditis, petechial hemorrhage, pericardial
effusion
p-Aminosalicylic acid Heart failure, hypotension, ventricular
irritability
Phenytoin Epistaxis Sudden Myocarditis
Carbamazepine Jaundice, fever, rash Sudden Myocarditis
Indomethacin Cardiac arrest, fever Brain damage Myocarditis
Spironolactone with hydrochlorothiazide Low back pain, fever Sudden Myocarditis
Acetazolamide Fever, rash Uremia Myocarditis, hepatitis
Amitriptyline Sudden Myocarditis
Phenindione Myocarditis
Interleukin 2 Congestive heart failure; may occur Myocarditis
within days after therapy initiated
Previously published. See “Credit Lines” section.

plasmacytic component. The organism may be visualized in the
section with special stains. In a patient with suspected Lyme
disease after a tick bite, the possibility of coinfection with
Ehrlichia (ehrlichiosis) and Babesia (babesiosis) should be con-
sidered since both can also cause myocarditis. Serologic tests are
available for both organisms.
Chagas Cardiomyopathy
Chagas cardiomyopathy is an acute myocarditis that may result
from Trypanosoma cruzi infection or from chronic cardiomyopa-
thy. It is a major cause of cardiomyopathy worldwide and may be
seen in immigrants from rural Central or South America, where
it is endemic. In those areas, Chagas cardiomyopathy is a leading
cause of DCM and congestive heart failure.
Approximately 30% of patients with Chagas cardiomyopathy
present with symptomatic heart block or ventricular arrhyth-
mias (Table 97.4). These may result from progressive damage
to the myocardium, extracellular matrix, autonomic innerva-
tion, and coronary microvessels. An additional 20% to 30% of
affected persons have asymptomatic cardiac involvement. The
disease is clinically suspected if the patient has a strong history
of environmental exposure, and the diagnosis is confirmed with
serologic testing or PCR. The ECG may show evidence of con-
duction system disease, including right bundle branch block.
Echocardiography or contrast ventriculography may show an LV
apical aneurysm, regional wall motion abnormalities, or diffuse
cardiomyopathy.
There is no specific treatment for Chagas cardiomyopa-
thy; rather, institution of preventive measures is recommended.
Improvements in housing and the use of pesticides may eradicate
the reduviid bugs that transmit the disease, thus decreasing infec-
tion and disease rates. Antiparasitic treatments directed against
T cruzi may eradicate the disease in acute or subacute infections.
Congestive heart failure can be managed symptomatically with
angiotensin-converting enzyme inhibitors, diuretics, and dig-
oxin. Ventricular arrhythmias may respond to electrophysiolog-
ically guided treatments. Although heart transplant for Chagas
cardiomyopathy has been successfully performed, reactivation
of T cruzi is common.
HIV Myocarditis
HIV has long been associated with myocarditis and DCM.
HIV-associated myocarditis is often characterized by a focal
nonspecific myocardial infiltrate with LV dysfunction. The
pathogenesis of myocarditis and LV dysfunction in patients
with HIV is unclear. Direct damage by the HIV virus is infre-
quent, so it is speculated that myocardial inflammation results
from 1 or more of the following: coinfections, antiretroviral
treatment, or inhibition of cardiac contractility by HIV type 1
glycoprotein 120.
In studies performed before the HAART era, myocarditis
was identified in as many as 70% of HIV-infected patients.
One prospective study of these patients showed that the mean
annual incidence of progression to DCM was 15.9 cases per
1,000 patients. Since the introduction of HAART, the inci-
dence of myocarditis in HIV-infected patients in developed
countries has significantly decreased. In developing countries,
however, where the availability of HAART is sometimes lim-
ited, the incidence of HIV-associated myocarditis continues to
increase.
 97 Myocarditis 887

Table 97.4. Clinical Presentation of North American

Box 97.4. Causes of Various Histopathologic Forms Patients With Chagas Cardiomyopathy
of Myocarditis in Biopsy Tissues
Patients
Lymphocytic myocarditis
Idiopathic Feature No. %
Viral syndrome Atrioventricular block 9 21
Polymyositis Congestive heart failure 8 19
Sarcoidosis Chest pain 6 14
Conduction abnormality on electrocardiography 8 19
Lyme disease
Aborted sudden death 3 7
Mucocutaneous lymph node syndrome (Kawasaki Sustained ventricular tachycardia 3 7
disease) Embolic event 3 7
Acquired immunodeficiency syndrome Other 2 5
Total 42
Mycoplasma pneumoniae
Drug toxicity Previously published. See “Credit Lines” section.

Neutrophilic or mixed myocarditis

Bacterial infection
The prognosis for patients with HIV-associated myocarditis
Acute drug toxicity
is considerably poorer than that for patients with acute lympho-
Infarction cytic myocarditis. In a large cardiomyopathy cohort, HIV-related
Eosinophilic myocarditis myocarditis was the strongest predictor of death.
Idiopathic
Hypereosinophilia Abbreviations
Restrictive cardiomyopathy AV atrioventricular
CK-MB creatinine kinase MB
Churg-Strauss syndrome
CVB3 coxsackievirus B
Parasitic infestations DCM dilated cardiomyopathy
Drug hypersensitivity ECG electrocardiographic
EMB endomyocardial biopsy
Giant cell or granulomatous myocarditis
GCM giant cell myocarditis
Idiopathic HAART highly active antiretroviral therapy
Sarcoidosis HIV human immunodeficiency virus
LV left ventricular
Infective
MRI magnetic resonance imaging
Rheumatoid PCR polymerase chain reaction
Rheumatic RV right ventricular
Drug hypersensitivity
Foreign body reaction Names of Clinical Trials
IMAC Intervention in Myocarditis and Acute Cardiomyopathy
Previously published. See “Credit Lines” section. TIMIC Tailored Immunosuppression in Inflammatory Cardio-
myopathy
 98
Dilated Cardiomyopathya
HORNG H. CHEN, MD, BCH and SANJAY DANDAMUDI, MD
DCM is a myocardial disease characterized by dilation and
impaired contraction of one or both ventricles. It is an impor-
tant topic to review for cardiovascular examinations, and special
attention should be paid to the general features, etiologies and
evaluation of this disease process.
Definitions and Classifications
DCM is classified as a separate disease entity from RCM or HCM,
which are reviewed in detail in other chapters of this book.
Over the last several decades, multiple major scientific societ-
ies have proposed definitions and classifications for cardiomyo-
pathies in accordance with an evolving understanding of this
group of diseases. Originally, the World Health Organization and
the International Federation and Society of Cardiology defined
cardiomyopathies as diseases of heart muscle associated with
cardiac dysfunction. The primary classifications included DCM,
HCM, RCM, arrhythmogenic right ventricular cardiomyopathy
and unclassified cardiomyopathies based on ventricular mor-
phology and hemodynamic pathophysiology. The World Health
Organization’s list of specific cardiomyopathies is reproduced at
the end of the chapter.
a
The “Specific Cardiomyopathies” section has been published
previously in Richardson P, McKenna W, Bristow M, Maisch B, Mautner
B, O’Connell J, et al. Report of the 1995 World Health Organization/
International Society and Federation of Cardiology Task Force on the
Definition and Classification of Cardiomyopathies. Circulation. 1996
Mar 1;93(5):841–2. Used with permission. Portions of the “Pathological
Features” and “Prognosis and Survival” sections published previously
in Dec WG, Fuster V. Idiopathic dilated cardiomyopathy. N Engl J Med.
1994 Dec 8;331(23):1564–75. Used with permission.
Abbreviations are expanded at the end of this chapter.
888
In 2006, the AHA refined this definition to describe car-
diomyopathies as “a heterogeneous group of diseases of the
myocardium associated with mechanical and/or electrical dys-
function that usually (but not invariably) exhibit inappropriate
ventricular hypertrophy or dilatation and are due to a variety
of causes that frequently are genetic. Cardiomyopathies either
are confined to the heart or are part of generalized systemic
disorders, often leading to cardiovascular death or progressive
heart failure–related disability.” The AHA and, more recently,
the ESC, incorporated genetic or inherited components as a
part of their respective classification schemes (Figure 98.1 and
Figure 98.2).
Characteristic Features of Cardiomyopathies
The characteristic features of each of the main types of cardio-
myopathy can readily be identified with two-dimensional and
Doppler echocardiography on the basis of chamber size, wall
thickness, and systolic and diastolic function of the left ventricle
(Table 98.1).
DCM consists of an enlarged left ventricular cavity with
depressed systolic function. HCM is characterized by a small-
to-normal size left ventricular cavity, massive hypertrophy of the
myocardium, and hyperdynamic systolic function. The major
abnormality in RCM is diastolic dysfunction of the myocardium.
It is important to remember that there is some overlap among
the types of cardiomyopathies. End-stage HCM may exhibit ven-
tricular dilatation and have features of both HCM and DCM.
Some cases of HCM in which the ventricular walls are only
mildly thickened may mimic the restrictive hemodynamic pro-
file of RCM. RCM may also exhibit some degree of ventricular
dilatation, which is referred to as “minimally dilated restrictive
cardiomyopathy.” Although each type of cardiomyopathy has a
 98 Dilated Cardiomyopathy 889

Cardiomyopathies Genetics
Analysis of asymptomatic relatives of patients with dilated cardi-
HCM DCM ARVC RCM Unclassified omyopathy indicates that familial disease accounts for one-third
to one-half of cases, although with incomplete and age-depen-
dent penetrance. More than 40 disease genes have been identi-
Familial Genetic Nonfamilial/Nongenetic fied, and the most common mode of inheritance is autosomal
dominant transmission, with autosomal recessive, X-linked, and
mitochondrial forms occurring less frequently.
Unidentified Disease subtype* Idiopathic Disease subtype* DCM is often the end phenotype of diverse mutations in
gene defect heterogeneous pathways that encode cytoskeletal, sarcomeric
and calcium-handling proteins leading to cellular compromise
Figure 98.1. AHA Cardiomyopathy Classification. (Previously pub- and fibrotic repair, and ultimately gross cardiac dilatation and
lished. See “Credit line” section.) reduced function.
Family members of patients with DCM should undergo
echocardiographic screening as asymptomatic or mildly symp-
pure form, some degree of clinical overlap can exist among these tomatic relatives may be identified. In 2010, the ESC published a
entities. position statement describing the aims and goals genetic counsel-
ing and testing in cardiomyopathies in addition to initial screen-
Dilated Cardiomyopathy ing guidelines.
DCM is the most common type of cardiomyopathy with an esti- Clinical Presentation
mated prevalence in the general population of 1:2500. HCM is
Most patients with DCM present between the ages of 20 and 50
about one-fifth as common, and RCM and arrhythmogenic right
years, although the disease may affect children and older adults.
ventricular cardiomyopathy are rarer. DCM is the third most
The most common presenting symptoms are of those associ-
common cause of heart failure and the most frequent indication
ated with heart failure, such as progressive exertional dyspnea,
for cardiac transplantation.
impaired exercise tolerance, orthopnea, paroxysmal nocturnal
DCM occurs more frequently in males than females (about
dyspnea and peripheral edema. Other presentations include inci-
3:1) and is more common in African Americans than in whites
dental detection of asymptomatic cardiomegaly and coexisting
(about 2.5:1).
arrhythmias, conduction disturbances, thromboembolic disease
and sudden cardiac death.
Etiologies
DCM may be idiopathic, familial and/or genetic, viral and/or Pathological Features
immune, alcoholic and/or toxic, or associated with recognized The primary morphologic feature of DCM is ventricular dilata-
cardiovascular disease in which the degree of myocardial dys- tion. Mural thrombi are frequently present in the left ventricle
function is not explained by the abnormal loading conditions or and not infrequently in the atria, which are also usually dilated.
the extent of ischemic damage. Table 98.2 is a list of commonly The heart is increased in weight, indicating hypertrophy, but the
recognized causes of DCM. maximal thicknesses of the ventricular free wall and septum

Primary cardiomyopathies
(Predominantly involving the heart)

Genetic Mixed Acquired

Hypertrophic Dilated Inflammatory (myocarditis)

ARVC/D Restrictive Stress provoked

(nonhypertrophied (Takotsubo)
LVNC and nondilated) Peripartum
PRKAG2 Glycogen
storage Tachycardia-induced
Danon
Conduction defects Infants of insulin-dependent
diabetic mothers
Mitochondrial myopathies

Ion channel disorders

LQTS Brugada SQTS CVPT Asian

SUNDS

Figure 98.2. ESC Cardiomyopathy Classification. (Previously published. See “Credit line” section.)
890 IX Cardiomyopathy and Heart Failure

Table 98.1. Differentiation of Cardiomyopathies by Echocardiographic Features

Feature DCM HCM RCM
Cavity Size Enlarged Small Normal
Wall thickness Normal Marked Normal or increased
Systolic dysfunction Severely depressed Hyperdynamic Normal or reduced
Diastolic dysfunction Abnormal Abnormal Abnormal
Other Outflow tract obstruction

are typically normal because of abnormally dilated chambers.
Epicardial coronary vessels are generally normal.
In clinical practice, patients may present in the early stage of
the disease with only left ventricular dilatation, followed later by
left atrial dilatation, and finally with dilatation of all four cardiac
chambers. Right ventricular dilatation in DCM carries a poor
prognosis.
Microscopic features of DCM include substantial hypertro-
phy and degeneration of myocytes, varying degrees of interstitial
fibrosis, and occasional small clusters of lymphocytes.
Diagnostic Characteristics
Echocardiography is the most useful initial diagnostic procedure
in evaluating DCM. Echocardiographic findings are virtually
similar in all patients with DCM and include left ventricular
dilatation, normal or reduced septal and free-wall thickness, and
global hypokinesis. Regional wall motion abnormalities can be
superimposed on DCM even if no flow-limiting coronary disease
is present. It is important to recognize that regional wall motion
abnormalities do not exclude a diagnosis of idiopathic DCM.
Over the last decade, increased use of CMR has provided
an additional imaging modality in the diagnostic assessment of
DCM. With its high spatial resolution, CMR enables accurate
assessment of ventricular volumes, ventricular and systolic func-
tion, and myocardial mass and wall thickness. The additional use
of late-gadolinium enhancement allows clinicians to noninva-
sively distinguish ischemic from nonischemic cardiomyopathy.
This technique has also led to the recognition of a characteris-
tic mid-wall ventricular enhancement or fibrosis that is seen in
nearly 20% of patients with DCM, and is independently associ-
ated with an increased incidence of adverse cardiac events.
CMR is also useful in diagnosing specific forms of car-
diomyopathy, such as hypertrophic cardiomyopathy, cardiac
Table 98.2. Major Causes of Dilated Cardiomyopathy
Infection (viral, bacterial, mycobacterial, fungal, parasitic)
Infiltration (hemochromatosis, amyloid)
Medications (anthracyclines, cyclophosphamide, trastuzumab,
antiretrovirals, phenothiazines)
Toxins (alcohol, cocaine, amphetamine, cobalt, lead, mercury, carbon
monoxide)
Rheumatologic (systemic lupus erythematosus, scleroderma, giant cell
arteritis)
Endocrinologic (hypothyroidism, pheochromocytoma, diabetes mellitus,
Cushing’s disease)
Nutritional deficiencies (thiamine, carnitine, selenium)
Electrolyte abnormalities (hypocalcemia, hypophosphatemia)
Sarcoidosis
Familial
Peripartum
Tachycardia-induced
Sleep apnea
sarcoidosis, amyloidosis, hemochromatosis, myocarditis, and
Chagas disease.
Prognosis and Survival
Although clinical, hemodynamic, and imaging features are often
helpful in determining risk in the larger population of patients
with DCM, the assessment of prognosis for an individual patient
is quite variable and dependent on the underlying etiology.
Table 98.3 lists prognostic factors that are predictive of poor sur-
vival in DCM.
Since cardiac dilatation and dysfunction may result from a
variety of acquired or hereditary disorders, the differentiation
of idiopathic from secondary, and potentially reversible, forms
of myocardial disease has important prognostic and therapeutic
implications.
Tachycardia-induced cardiomyopathy, stress-induced car-
diomyopathy and peripartum cardiomyopathy are examples
of reversible forms of cardiomyopathy with more favorable
prognoses.
Treatments
Treatment of the underlying cause is the mainstay of therapy for
DCM when an etiology is identified. When the cause is unknown
Table 98.3. Prognostic Factors in Dilated Cardiomyopathy
That Predict Poor Survival
Abnormal ventricular function
Decreased ejection fraction is the most power prognostic indicator
in DCM
Increased left ventricular size
Right ventricular dilatation
Functional class
New York Heart Association class III or IV
Maximal oxygen uptake <12 mL/kg per minute on cardiopulmonary
exercise testing
Electrocardiography
Left bundle branch block
Asymptomatic nonsustained ventricular tachycardia
Clinical features
Clinical left or right heart failure
Syncope
Endocrine activation and electrolyte levels
Hyponatremia
Increased plasma concentrations of norepinephrine, atrial natriuretic
factor, and renin
Impaired hemodynamics
Elevated left ventricular end-diastolic pressure or pulmonary capillary
wedge pressure (>18–20 mm Hg)
Low cardiac index (<2.5 L/min per m 2)
Pulmonary hypertension (pulmonary artery systolic pressure >35 mm Hg)
Cardiac biopsy
Loss of intracellular cardiac myofilaments
Persistence of enteroviral RNA
 98 Dilated Cardiomyopathy
or specific treatment unavailable, the general principles of treat-
ing DCM focus on risk factor modification, symptom manage-
ment, and prevention of disease progression and associated
complications.
Pharmacotherapy in dilated cardiomyopathy follows the same
strategies used in treating heart failure with reduced systolic
function, as discussed in detail in previous chapters.
The use of device therapy, such as implantable cardioverter-
defibrillator, cardiac resynchronization therapy, and ventricular
assist devices as bridge or destination therapy, are becoming
more commonplace in the treatment of advanced DCM. Cardiac
transplantation remains an option for appropriately selected can-
didates with end-stage DCM. The details of cardiac transplanta-
tion are discussed in a subsequent chapter of this textbook.
Specific Cardiomyopathies
Ischemic cardiomyopathy presents as dilated cardiomyopathy
with impaired contractile performance not explained by the
extent of coronary artery disease or ischemic damage.
Valvular cardiomyopathy presents with ventricular dysfunc-
tion that is out of proportion to the abnormal loading conditions.
Hypertensive cardiomyopathy often presents with left ven-
tricular hypertrophy in association with features of dilated or
restrictive cardiomyopathy with cardiac failure.
Inflammatory cardiomyopathy is defined by myocardi-
tis in association with cardiac dysfunction. Myocarditis is an
inflammatory disease of the myocardium and is diagnosed by
established histologic, immunologic, and immunohistochemi-
cal criteria. Idiopathic, autoimmune, and infectious forms of
inflammatory cardiomyopathy are recognized. Inflammatory
myocardial disease is involved in the pathogenesis of dilated car-
diomyopathy and other cardiomyopathies (eg, Chagas disease,
human immunodeficiency virus, enterovirus, adenovirus, and
cytomegalovirus).
Metabolic cardiomyopathy includes the following categories:
endocrine (eg, thyrotoxicosis, hypothyroidism, adrenal cortical
insufficiency, pheochromocytoma, acromegaly, and diabetes
mellitus), familial storage disease and infiltrations (eg, hemo-
chromatosis, glycogen storage disease, Hurler syndrome, Refsum
syndrome, Niemann-Pick disease, Hand-Schüller-Christian dis-
ease, Fabry-Anderson disease, and Morquio-Ullrich disease),
891
deficiency (eg, disturbances of potassium metabolism, magne-
sium deficiency, and nutritional disorders such as kwashiorkor,
anemia, beri-beri, and selenium deficiency), amyloid (eg, pri-
mary, secondary, familial, and hereditary cardiac amyloidoses,
familial Mediterranean fever, and senile amyloidosis).
General system diseases include connective tissue disorders
(eg, systemic lupus erythematosus, polyarteritis nodosa, rheuma-
toid arthritis, scleroderma, and dermatomyositis). Infiltrations
and granulomas include sarcoidosis and leukemia.
Muscular dystrophies include Duchenne, Becker-type, and
myotonic dystrophies.
Neuromuscular disorders include Friedreich ataxia, Noonan
syndrome, and lentiginosis.
Sensitivity and toxic reactions include reactions to alcohol,
catecholamines, anthracyclines, irradiation, and miscellaneous.
Alcoholic cardiomyopathy is associated with a heavy alcohol
intake. Currently, we cannot define a causal versus a condition-
ing role of alcohol or apply precise diagnostic criteria.
Peripartal cardiomyopathy may first manifest in the peripar-
tum period. This is likely a heterogeneous group.
Arrhythmogenic right ventricular dysplasia is a cardiomyo-
pathy characterized by progressive fibrofatty replacement of the
right ventricle, ventricular arrhythmias, and sudden death at a
relatively young age. Autosomal dominance with incomplete
penetrance is common.
Stress-induced (“tako-tsubo”) cardiomyopathy is a clinical
entity, first described in Japan, characterized by acute but rapidly
reversible left ventricular systolic dysfunction in the absence of
atherosclerotic coronary artery disease. This distinctive form of
ventricular stunning is typically brought on by profound psycho-
logical stress, predominantly affecting older women, and pref-
erentially involves the apical portion of the left ventricle (apical
ballooning).
Abbreviations
AHA American Heart Association
CMR cardiac magnetic resonance
DCM dilated cardiomyopathy
ESC European Society of Cardiology
HCM hypertrophic cardiomyopathy
RCM restrictive cardiomyopathy

Restrictive Cardiomyopathya
SUDHIR S. KUSHWAHA, MD
Definition and Etiology
RCM is defined as myocardial disease that results in impaired
ventricular filling with 1) normal or reduced diastolic volume of
either or both ventricles and 2) normal or near-normal systolic
function and wall thickness. The condition usually results from
increased stiffness of the myocardium, which causes pressure
within the ventricles to increase precipitously with only small
increases in volume. Affecting either or both ventricles, RCM
may cause symptoms or signs of right or left ventricular failure.
Often, right-sided findings predominate, with elevated jugular
venous pressure, peripheral edema, and ascites. When the left
ventricle is affected, breathlessness and evidence of pulmonary
edema are present, usually with normal cardiac dimensions. The
diagnosis should therefore be considered when a patient has heart
failure but no evidence of cardiomegaly or systolic dysfunction.
The importance of an accurate diagnosis lies in distinguishing
RCM from constrictive pericarditis, which can also occur with
restrictive physiology but which can often be cured surgically.
RCM is the least common of the cardiomyopathies. Outside
the tropics, cardiac amyloidosis is the most thoroughly stud-
ied cause of RCM. Endomyocardial fibrosis, another cause, is
endemic in parts of Africa, Central America, and Asia and occurs
sporadically throughout the world. RCM may be caused by vari-
ous other local and systemic disorders (Box 99.1), many of which
may not be encountered in clinical practice. Patients may have
more common conditions, such as amyloidosis, and present with
a Amyloidosis
Portions previously published in Kushwaha SS, Fallon JT, Fuster V.
Restrictive cardiomyopathy. N Engl J Med. 1997 Jan 23;336(4):267–76.
Used with permission.
Abbreviations and acronyms are expanded at the end of this chapter.
892
99
congestive heart failure. In idiopathic RCM, the hemodynamic
abnormalities occur without specific histopathologic changes.
Pathogenesis and Natural History
Idiopathic Restrictive Cardiomyopathy
Idiopathic RCM may be familial and associated with a distal
skeletal myopathy and atrioventricular block. There may be a
genetic predisposition, but some cases are sporadic and result
from spontaneous mutation. Although patients with RCM can
present at any age, most are older than 60 years (72% in a Mayo
Clinic retrospective analysis). The most common symptoms at
presentation are dyspnea and edema. In childhood, idiopathic
RCM may be more common in girls and carry a worse prognosis,
with a median survival of only 1 year. Mutations in sarcomere
protein genes are an important cause of idiopathic RCM in child-
hood. The clinical course is more variable in adults, and most
small series suggest a protracted clinical course. The condition is
characterized by a mild to moderate increase in cardiac weight.
Biatrial enlargement is common, with thrombi often present in
the atrial appendages. Systolic function tends to be preserved
or mildly reduced, and ventricular size tends to be normal.
Depending on the degree of pulmonary hypertension, the right
ventricle may be enlarged. On microscopy, the pericardium is
normal and patchy intersitial fibrosis may be present. If fibrosis
affects the conducting system, complete heart block may occur
and require permanent pacing.
Cardiac involvement is relatively common in primary amyloido-
sis, which is caused by the production of immunoglobulin light
chains by plasma cells and is often due to multiple myeloma. It
 99 Restrictive Cardiomyopathy
Box 99.1. Causes of Restrictive Cardiomyopathy
Primary (idiopathic) restrictive cardiomyopathy
Eosinophilic endomyocardial disease and
endomyocardial fibrosis
Infiltrative cardiomyopathies
Amyloid heart disease
Hemochromatosis
Glycogen storage disease
Fabry disease
Mucopolysaccharidoses
Sarcoidosis
Scleroderma
Heart transplant
Mediastinal irradiation
Pseudoxanthoma elasticum
Doxorubicin and daunorubicin chemotherapy
Carcinoid heart disease
Malignant disease with encasement of the heart
from pericardial metastases
Eosinophilia-myalgia syndrome due to
toxic contamination of L-tryptophan dietary
supplements
can also occur with familial amyloidosis. RCM results from tissue
injury due to replacement of normal myocardium with infiltrative
interstitial deposits. Amyloid deposition begins in the subendo-
cardium with focal deposits and extends within the myocardium
between the muscle fibers. This leads to an overall increase in
myocardial thickness (Figure 99.1). Abnormalities of diastolic
filling can occur even without clinical evidence of RCM. Patients
with cardiac amyloid may also present with angina. The presence
of RCM is strongly predictive of a poor prognosis, with 55% of
patients dying of arrhythmia or cardiac failure. Amyloid infiltra-
tion of the heart is common in the elderly and in patients with
chronic heart disease, such as rheumatic or congenital disease.
Cardiac amyloid can be characterized by analyzing endomyocar-
dial biopsy tissue, and immunohistochemical staining may help
in distinguishing the various types.
Figure 99.1. Amyloid Deposits in Thickened Myocardium. The
appearance resembles that in hypertrophic cardiomyopathy.
893
The myocardium of patients with cardiac amyloidosis is firm
and noncompliant, with small or dilated ventricular cavities and
thrombi in the atrial appendages. Typically there is interstitial
deposition of insoluble amyloid fibrils in all 4 cardiac chambers,
which can result in increased wall thickness without cavity dila-
tation. The pericardium, cardiac valves, and coronary arteries
may also be involved. Left ventricular wall thickness is a prog-
nostic variable. In patients with normal wall thickness, survival
may be up to 2.4 years; in patients with markedly increased wall
thickness, only a few months. Increased wall thickness is also
correlated with the characteristic granular sparkling appear-
ance seen on 2-dimensional echocardiography (Figure 99.2).
Abnormalities of diastolic filling can also occur and may be
predictive of decreased survival.
Depending on the stage of cardiac amyloidosis, the patient
can present with a combination of asymmetric septal thickening,
angina, heart failure, abnormal diastolic function, and reduced
ejection fraction. Amyloid deposits in the conducting system can
cause various cardiac arrhythmias, including complex ventricu-
lar arrhythmias. The severity of the arrhythmias tends to be cor-
related with the severity of the heart failure and the abnormalities
on echocardiography. Frequently a combination of atrial fibrilla-
tion with low voltage on electrocardiography is seen with a slow
ventricular response. A classic scenario for cardiac amyloidosis
is a patient with atrial fibrillation and low voltage on electrocar-
diography undergoing pacemaker placement; it is difficult to find
a spot in the right ventricle with adequate pacing thresholds.
• Cardiac involvement is relatively common in primary amyloidosis.
It can also occur with familial amyloidosis.
• Atrial fibrillation with low voltage and a slow ventricular response
on electrocardiography is characteristic of amyloidosis.
Other Infiltrative and Storage Diseases
Various infiltrative conditions can result in RCM. Gaucher dis-
ease and related syndromes can result in the accumulation of
cerebroside in several organs, including the heart (Figure 99.3).
Hurler syndrome causes RCM through the deposition of muco-
polysaccharide in the myocardium and in the cardiac valves and
coronary arteries. Patients with Fabry disease can also present
with RCM.
Figure 99.2. Echocardiogram of Amyloid Heart Disease With
Abnormal Myocardial Texture (“Scintillating Granular Myocardium”)
and Wall Thickening. AV indicates aortic valve; LA, left atrium; LV, left
ventricle; PE, pericardial effusion; RV, right ventricle.
894 IX Cardiomyopathy and Heart Failure
Figure 99.3. Gaucher Disease. Right ventricular specimen is shown.
Patients with cardiac sarcoidosis may initially present with
impaired diastolic function and normal systolic function.
Subsequent injury and fibrosis can result in systolic dysfunc-
tion and diffuse hypokinesis and in regional wall motion abnor-
malities. In patients with systemic sarcoidosis the myocardium
may be involved, resulting in subclinical cardiac dysfunction.
Patients with cardiac sarcoidosis may present with sudden death
or high-degree atrioventricular block due to involvement of the
conducting system. Endomyocardial biopsy may be useful in the
diagnosis of cardiac sarcoidosis (Figure 99.4), although negative
biopsy findings do not rule out the diagnosis.
• Patients with cardiac sarcoidosis may present with sudden death or
high-degree atrioventricular block due to involvement of the con-
ducting system.
Endomyocardial Fibrosis and Eosinophilic
Endomyocardial Disease
Endomyocardial fibrosis and Löffler endocarditis (eosinophilic
cardiomyopathy) are different manifestations of restrictive oblit-
erative cardiomyopathy; both are associated with eosinophilia
(Figure 99.5). Restrictive obliterative cardiomyopathy results
from severe, prolonged eosinophilia from any cause (aller-
gic, autoimmune, parasitic, leukemic, or idiopathic) that leads
to eosinophilic infiltration of the myocardium, which can lead
to decreased myocardial compliance. In Löffler endocarditis,
Figure 99.4. Cardiac Sarcoidosis. Right ventricular endomyocardial
biopsy specimen shows solitary granuloma (arrow).
Figure 99.5. Eosinophilic Restrictive Cardiomyopathy. Right ven-
tricular endomyocardial biopsy specimen is shown (16% eosinophils).
eosinophilic degranulation results in myocardial damage through
the action of major basic and cationic proteins. In its late stages
eosinophilic infiltration of the myocardium is associated with
dense endomyocardial fibrosis, intraventricular thrombus forma-
tion, and obliteration of the ventricular cavity. Occasionally the
fibrosis is localized and produces valvular regurgitation or steno-
sis, which can be treated with valve replacement. The eosino-
philia-myalgia syndrome associated with the use of tryptophan
containing a contaminant has resulted in RCM, suggesting that
eosinophilia may have been responsible for the RCM. The end-
stage cardiac structure in eosinophilia-myalgia syndrome is
similar to that of tropical endomyocardial fibrosis.
Tropical endomyocardial fibrosis is a common form of RCM
that occurs almost exclusively in equatorial Africa, Asia, and
South America. Because most patients with this condition also
have heavy parasite loads, it was thought for many years that
endomyocardial fibrosis represented an end stage of eosinophilic
endomyocardial disease in response to an allergic or immune
reaction to parasitosis. This is now considered incorrect because
patients with the tropical form of endomyocardial fibrosis do not
have eosinophilia. The prognosis for patients with endomyocar-
dial fibrosis is poor, but surgical removal of fibrotic tissue and
valve repair or replacement may be helpful. The disease may
have an insidious onset with development of right- or left-sided
cardiac failure. Sudden death and syncopal episodes are not com-
mon in endomyocardial fibrosis compared with other causes of
RCM. Atrial fibrillation does occur. Other electrocardiographic
abnormalities include low QRS voltage, first degree atrioven-
tricular block, and left atrial enlargement. On echocardiography,
endocardial deposition of thrombus with apical obliteration and
mitral regurgitation may be observed. Systemic or pulmonary
emboli are uncommon.
Clinical Presentation
The underlying cause of RCM may not be obvious on presen-
tation. Symptoms include dyspnea, paroxysmal dyspnea, ortho-
pnea, peripheral edema, ascites, general fatigue, and weakness.
Angina does not occur except in amyloidosis, in which it may
be the presenting symptom. In advanced cases, all the signs
of heart failure may be present except cardiomegaly. The find-
ings may resemble those of severe constrictive pericarditis. Up
to one-third of patients with idiopathic RCM may present with
 99 Restrictive Cardiomyopathy
thromboembolic complications. In amyloidosis and sarcoido-
sis, cardiac conduction disturbances are particularly common.
Atrial fibrillation is common in idiopathic RCM and cardiac
amyloidosis. In the elderly, RCM is a diagnosis of exclusion, but
it should be differentiated from age-related changes in diastolic
compliance.
• In advanced RCM, all the signs of heart failure may be present
except cardiomegaly.
Diagnostic Evaluation
The initial diagnostic approach should attempt to rule out con-
strictive pericarditis, which results in clinical signs and symp-
toms similar to those of RCM, as described below.
The physical examination is remarkable for increased venous
pressure with rapid x and y descents if sinus rhythm is pres-
ent, but the most prominent wave is the y descent. The jugular
venous pulse does not decrease during inspiration and may actu-
ally increase (Kussmaul sign). Peripheral edema and ascites are
present in advanced cases, and the liver may be enlarged and
pulsatile. The carotid upstroke may be low volume and sinus
tachycardia may be present, as in a low-output state. The left ven-
tricular systolic impulse is usually normal. The first and second
heart sounds are usually normal with normal splitting of the sec-
ond heart sound. The pulmonary component of the second heart
sound is not accentuated. There may be a third heart sound, orig-
inating in the right or left ventricle, and less commonly a fourth
heart sound. The precordium is usually normal. Regurgitation
murmurs of the mitral or tricuspid valve (or of both valves) may
be present.
On chest radiography the heart size is usually normal. Atrial
enlargement may be present, particularly if there is atrioventric-
ular valvular regurgitation. Pulmonary congestion, interstitial
edema with Kerley B lines, and pleural effusion may be appar-
ent. The electrocardiogram shows nonspecific ST-segment and
T-wave abnormalities. Depolarization abnormalities may also be
present with bundle branch block, atrioventricular block, or find-
ings of ventricular hypertrophy.
One should consider serum protein electrophoresis, iron
studies, angiotensin-converting enzyme levels, and appropri-
ate parasitic screening if indicated by the history and clinical
findings.
Echocardiography
The imaging modality of choice for diagnosing RCM is echocar-
diography. In most instances, a 2-dimensional echocardiogram
shows normal left ventricular size and function and marked dila-
tation of both atria (Figure 99.6). The wall thickness is usually
normal in idiopathic RCM, but it may be thicker in infiltrative
disease causing RCM (eg, amyloidosis). Doppler echocardiog-
raphy shows features of a restrictive filling pattern, indicating a
marked decrease in chamber compliance. There is an increased
early diastolic filling velocity (≥1.0 m/s), a decreased atrial fill-
ing velocity (≤0.5 m/s), an increased ratio of early diastolic fill-
ing to atrial filling (≥2), a decreased deceleration time (≤150
ms), and a decreased isovolumic relaxation time (≤70 ms). The
ratio of early to late diastolic mitral inflow velocity is high, with
a short deceleration time on the mitral inflow velocities, indicat-
ing an abrupt cessation of ventricular filling, and a low systolic
to diastolic flow ratio of the pulmonary venous flow veloci-
ties. Evidence of moderate pulmonary hypertension is usually
present.
895
Figure 99.6. Doppler Echocardiogram From a Patient With
Restrictive Cardiomyopathy. The characteristic restrictive inflow profile
demonstrates increased peak early diastolic mitral inflow velocity with
rapid deceleration time and diminutive peak late diastolic mitral inflow
velocity.
Cardiac Catheterization
The characteristic hemodynamic feature in RCM is a deep
and rapid early decrease in ventricular pressure at the onset
of diastole, with a rapid increase to a plateau in early diastole
(Figure 99.7). This is the dip-and-plateau square root sign, mani-
fested in the atrial pressure tracing as a prominent y descent fol-
lowed by a rapid rise to a plateau. The right atrial pressure is
elevated, as in constrictive pericarditis, and the y descent may
become deeper during inspiration. Although the right ventricular
pressure may be elevated, diastolic hypertension is usually more
prominent with mean right atrial pressures of 15 to 20 mm Hg.
The left ventricular diastolic pressure has the same wave form as
the right ventricular diastolic pressure; the left may be 5 mm Hg
higher than the right ventricular pressure, or the 2 values may be
the same. When the left is higher, the difference may be accentu-
ated by exercise.
Cardiac Magnetic Resonance Imaging
When echocardiographic features suggest amyloidosis or other
causes of RCM, cardiovascular magnetic resonance imaging may
be useful diagnostically. It is a noninvasive means of evaluating for
concentric ventricular myocardial thickening, impaired systolic
function with reduced ejection fraction, restriction of diastolic
filling, and disproportionate atrial enlargement (Figure 99.8).
Cardiovascular magnetic resonance imaging can be used to help
distinguish between symmetrical hypertrophic cardiomyopathy
and amyloidosis by showing low signal intensity in amyloid-infil-
trated myocardium on T1 and T2 sequences. Patients who have
amyloidosis tend to have diffuse, heterogeneous, early myocar-
dial enhancement. Histologically, these patients have transmural
amyloid distribution with corresponding amounts of extracellu-
lar amyloid, which may be useful in determining a prognosis and
in assessing the response to therapy.
Distinction Between Restrictive Cardiomyopathy
and Constrictive Pericarditis
Constrictive pericarditis is a more likely diagnosis than RCM
if the patient has a clinical history of pericarditis. A history of
896 IX Cardiomyopathy and Heart Failure

LV

100
RV

50

0 Inspiration

Figure 99.7. Hemodynamic Tracing From a Patient With Restrictive Cardiomyopathy. A indicates atrial pressure; LV, left ventricular pressure;
RV, right ventricular pressure.

tuberculosis may suggest constrictive pericarditis, particularly
in nonindustrialized nations. Constrictive pericarditis may also
develop after trauma, cardiac surgical procedures, and radia-
tion therapy, or it may develop years after acute pericarditis.
Although rare, some causes of RCM, including sarcoidosis and
amyloidosis, may also lead to constrictive pericarditis. Table 99.1
summarizes the important differences between the 2 conditions.
With cardiac magnetic resonance imaging, both conditions show
features of impaired ventricular filling, with a small or normal-
sized ventricular cavity, and dilatation of the atria, superior
and inferior venae cavae, and hepatic veins. However, when the
pericardium is thickened (typically >4 mm), a hypointense line
on spin echo T1 sequences makes the diagnosis of constrictive
pericarditis more likely. No technique, though, is completely
reliable—sometimes the only way to make the diagnosis is to
perform pericardiectomy.
Treatment
Symptomatic Therapy for Restrictive
Cardiomyopathy
Diuretics are used to treat pulmonary and systemic venous con-
gestion, but they must be used with caution because excessive
use may decrease ventricular filling pressures and consequently
decrease cardiac output with symptoms and signs of hypotension

A B

Figure 99.8. Cardiac Magnetic Resonance Images. A, Concentric left ventricular wall thickening, biatrial enlargement, a small amount of peri-
cardial effusion, and abnormally diffuse retention of myocardial contrast material are all consistent with cardiac amyloidosis. B, Short-axis perfusion
image shows hypoenhancing rind of subendocardial tissue.
 99 Restrictive Cardiomyopathy 897

Table 99.1. Differential Diagnosis of Restrictive Cardiomyopathy and Constrictive Pericarditis

Type of Evaluation Restrictive Cardiomyopathy
Physical examination Kussmaul sign may be present
Apical impulse may be prominent
S3 (rarely S4) may be present
Regurgitant murmurs common
Electrocardiography Low voltage (especially in amyloidosis),
pseudoinfarction, left axis deviation, atrial
fibrillation, conduction disturbances common
Echocardiography Increased wall thickness (especially thickened
interatrial septum in amyloidosis)
Thickened cardiac valves (amyloidosis)
Granular sparkling texture (amyloid)
Doppler studies Decreased RV and LV velocities with inspiration
Inspiratory augmentation of hepatic vein diastolic
flow reversal
Mitral and tricuspid regurgitation common
Cardiac catheterization LVEDP often >5 mm Hg greater than RVEDP,
but may be identical
Endomyocardial biopsy May reveal specific cause of restrictive
cardiomyopathy
CT or MRI Pericardium usually normal
Abbreviations: CT, computed tomography; LV, left ventricular; LVEDP, left ventricular end-diastolic pressure; MRI, magnetic resonance
imaging; RV, right ventricular; RVEDP, right ventricular end-diastolic pressure; S3, third heart sound; S4, fourth heart sound.
Previously published. See “Credit Lines” section.
Constrictive Pericarditis
Kussmaul sign usually present
Apical impulse usually not palpable
Pericardial knock may be present
Regurgitant murmurs uncommon
Low voltage (<50%)
Normal wall thickness
Pericardial thickening may be seen
Prominent early diastolic filling with abrupt
displacement of interventricular septum
Increased RV systolic velocity and decreased LV
systolic velocity with inspiration
Expiratory augmentation of hepatic vein diastolic flow
reversal
RVEDP and LVEDP usually equal
RV systolic pressure <50 mm Hg
RVEDP greater than one-third of RV systolic pressure
May be normal or show nonspecific myocyte
hypertrophy or myocardial fibrosis
Pericardium may be thickened

and hypoperfusion. Digoxin is usually not recommended because
it is potentially arrhythmogenic, particularly in patients with
amyloidosis. An important goal is to maintain sinus rhythm. The
development of atrial fibrillation, with loss of the atrial contribu-
tion to ventricular filling, may worsen existing diastolic dysfunc-
tion, and this may be further compromised by a rapid ventricular
response. Advanced conduction system disease may need to be
treated with implantation of a pacemaker. Because stroke volume
tends to be fixed in RCM, the development of a bradyarrhythmia
can precipitate cardiac failure, so the heart rate needs to be sup-
ported. For this reason, calcium channel blockers and β-blockers
are not helpful. Frequently, patients with cardiac sarcoidosis pres-
ent with malignant ventricular arrhythmias, which may need to
be treated with an implantable cardioverter-defibrillator system.
Anticoagulation with warfarin is recommended for most patients
because of the propensity for thrombus formation in the atrial
appendage and the resulting risk of embolic complications.
Specific Therapy
Cardiac Amyloidosis
Overall, the prognosis for patients with cardiac amyloidosis
tends to be poor, although chemotherapy for cardiac and sys-
temic manifestations may benefit specific patients. In special-
ized centers, cardiac transplant has been performed with success.
Without subsequent stem cell transplant in primary amyloidosis
or liver transplant in familial amyloidosis, however, the condition
will recur in the transplanted heart.
Endomyocardial Fibrosis and Eosinophilic
Cardiomyopathy
Treatment with corticosteroids and cytotoxic drugs may improve
symptoms and survival in the early stages of Löffler endocarditis.
Surgical excision of the fibrotic endocardium and replacement of
the mitral and tricuspid valves may also provide symptomatic
relief in the fibrotic stage of the disease.
Other Conditions
The degree of iron overload determines the prognosis and com-
plications in hemochromatosis. Early treatment with phlebotomy
or iron-chelation therapy may reverse many of the hemodynamic
abnormalities associated with heart failure in this condition.
Combined heart and liver transplant has been successful.
In idiopathic or familial RCM, cardiac transplant has been
successful and should be considered. Although transplant may
be an option in cardiac sarcoidosis, the condition tends to recur
in the transplanted heart.
Abbreviation
RCM restrictive cardiomyopathy
 100
Hypertrophic Cardiomyopathya
STEVE R. OMMEN, MD
Definition
The clinical definition of HCM is left ventricular hypertrophy
in the absence of other causes of left ventricular hypertrophy
(Figure 100.1). The size and volume of the left ventricle are
usually normal or smaller. Most cases are familial disease with
autosomal dominant inheritance. Histologically, loose connec-
tive tissue is surrounded by myocyte hypertrophy and disarray.
Arrhythmias and systolic LVOT gradients are common, and
premature sudden death occurs in approximately 1% of HCM
patients annually.
Genetic Etiology
HCM is present in approximately 1 in 500 persons, a rate that
appears constant across many ethnic backgrounds. Mutations in
the gene encoding for cardiac β-myosin heavy chain were the
first to be associated definitively with HCM. To date, hundreds
of mutations encoding for defects in at least 21 genes have been
recognized as responsible for HCM. Most of these mutations
involve the proteins of the cardiac sarcomere: β-myosin heavy
chain, myosin-binding protein C, cardiac troponin T, cardiac
troponin I, α-tropomyosin, essential myosin light chain, regula-
tory myosin light chain, cardiac α-actin, and titin (Table 100.1).
These mutations are inherited in an autosomal dominant pattern.
Mutations in other cellular structures (Z disk, mitochondria, etc)
have also been described.
Evidence suggests that gene mutations may alter sarcomeric
function and secondarily lead to hypertrophy and fibrosis.
a
Portions previously published in Ommen SR, Nishimura RA.
Hypertrophic cardiomyopathy. Curr Probl Cardiol. 2004 May;29(5):
239–91. Used with permission.
Abbreviations and acronyms are expanded at the end of this chapter.
898
Putative abnormalities include alteration of the protein structure,
change in sensitivity to regulators such as calcium or adenosine
triphosphate, impaired energy metabolism, and a decrease in the
force or velocity of myocyte contraction. Hypertrophy may thus
be a compensatory mechanism for sarcomeric dysfunction. Other
determinants or disease-modifying genetic polymorphisms may
be involved in the extent and distribution of hypertrophy.
Pathologic and Morphologic Features
The histologic hallmark of HCM is myocardial disarray:
Short runs of severely hypertrophied fibers are interrupted by
connective tissue (Figure 100.2). Disarray is observed not
only in the ventricular septum but also in all myocardial seg-
ments (Figure 100.3). Additional findings include large and
bizarre myocyte nuclei, fibrosis, and degenerating muscle fibers
(Figure 100.4); intramyocardial fibrosis, which is an integral
determinant of arrhythmogenesis and diastolic dysfunction; and
thrombosis and obliteration of the small vessels within the myo-
cardium, which can predispose to subendocardial ischemia.
Asymmetric hypertrophy of the septum from the basal to the
midventricular levels is common, but any pattern of left ventricu-
lar hypertrophy can occur in HCM (Figures 100.5–100.7). The
severity of hypertrophy is also highly variable (Figure 100.8).
The maximal left ventricular wall thickness is frequently 20 to
30 mm but can be more than 35 to 40 mm. Endocardial plaque
or a thickened subaortic portion of the ventricular septum is fre-
quently seen in the obstructive variant of the disease where the
anterior mitral leaflet contacts the ventricular septum. In addition
to subaortic obstruction, there may be diffuse hypertrophy of the
septum and papillary muscles at the midventricular level, result-
ing in combined midventricular obstruction. Abnormalities of
the mitral valve and its support structures are common in HCM:
Elongation of the mitral leaflets and anterior displacement of the
 100 Hypertrophic Cardiomyopathy 899

A B

Figure 100.1. Hypertrophic Cardiomyopathy. Autopsy specimen of

heart shows marked septal hypertrophy (arrow), a small left ventricular Figure 100.2. A, Myocardial disarray in hypertrophic cardiomyo-
cavity, a moderately dilated left atrium, and a subendocardial myocar- pathy. B, Normal myocardium.
dial infarct (arrowhead).

in higher or lower risks of sudden cardiac death. However, sub-

papillary muscles are common, and very short chordae tendineae sequent investigations of unrelated patients with HCM suggest
or direct insertion of the papillary muscle onto the mitral leaflets that neither the phenotypic expression nor the natural history of
has been observed. HCM can be distinguished on the basis of a specific mutation.
Notably, mutations related to myosin heavy chain, cardiac actin,
troponin T, α-tropomyosin, and titin can result in either hyper-
Genotype-Phenotype Correlations trophic or dilated cardiomyopathy. Several reports have shown
The question of whether the clinical features of HCM are deter- that delayed penetrance (late-onset hypertrophy in the fifth or
mined by individual mutations or gene defects has led to intense sixth decade) can occur, particularly if the underlying mutation
investigation. Studies of large families affected by HCM have involves myosin-binding protein C. Contrary to the previous
suggested that mutations related to β-myosin heavy chain are belief that phenotypic expression is complete in early adult-
associated with severe hypertrophy and mutations related to car- hood, this finding has led to an important change in the recom-
diac troponin are associated with only mild hypertrophy. Some mendations for family screening: It is now recommended that
family-based studies have suggested that certain mutations result adult family members of HCM patients undergo surveillance
echocardiography every 5 years and that adolescents undergo
screening every 12 to 18 months (or more often if they want to
Table 100.1. Frequency of Identified Chromosome and participate in competitive athletics). Among patients with clear
Protein Abnormalities in Hypertrophic Cardiomyopathy clinical HCM, the yield of currently available genetic testing is
only about 50%, limiting its practicality in clinical management
Protein (Gene Name) Chromosome Frequency, % of HCM.
Cardiac myosin-binding protein C 11 15–25
(MYBPC3) • Among patients with clear clinical HCM, the yield of currently
β-Myosin heavy chain (MYH7) 14 15–25
available genetic testing is only about 50%.
α-Myosin heavy chain (MYH6) 14 <1
Regulatory myosin light chain (MYL2) 12 <2
Essential myosin light chain (MYL3) 3 <1
Cardiac troponin T (TNNT2) 1 <5
Cardiac troponin I (TNNI3) 19 <5
α-Tropomyosin (TPM1) 15 <5
Cardiac α-actin (ACTC1) 15 <1
Titin (TTN) 2 <1
Muscle LIM protein (CSRP3) 11 <1
Telethonin (TCAP) 17 <1
Vinculin or metavinculin (VCL) 10 <1
Z band alternatively spliced PDZ 10 <5
protein (ZASP/LBD3)
α-Actinin (ACTN2) 1 <1
Cardiac ryanodine receptor (RyR2) 1 <1
Junctophilin-2 (JPH2) 20 <1
Adenosine monophosphate–activated 7 <1
protein kinase (PRKAG2)
Lysosome-associated membrane X <1
protein 2 (LAMP2)
Frataxin (FRDA) 9 <1 Figure 100.3. Myofiber Disarray in Familial Hypertrophic Cardio-
myopathy. Specimen is from right ventricular biopsy.
900 IX Cardiomyopathy and Heart Failure
Figure 100.4. Myofiber Disarray in Hypertrophic Cardiomyopathy.
Pathophysiology of HCM
The complex pathophysiology of HCM results from a com-
bination of left ventricular diastolic dysfunction, myocardial
ischemia, arrhythmias, LVOT obstruction, mitral regurgitation,
and autonomic dysfunction.
Diastolic Dysfunction
Diastolic dysfunction is thought to be a major pathophysiologic
mechanism in all patients with HCM, frequently leading to dia-
stolic heart failure. Marked abnormalities of both ventricular
relaxation and chamber stiffness may be present. Multiple abnor-
mal loads are imposed on the left ventricle, including contrac-
tion loads (from outflow obstruction) and relaxation loads as well
as nonuniformity of relaxation. The thick, hypertrophied ven-
tricular muscle causes abnormalities of passive filling because
of the increased stiffness of the left ventricle. Abnormalities of
relaxation have been observed with tissue Doppler echocardiog-
raphy before the development of hypertrophy in animals and in
humans.
Figure 100.5. Autopsy specimen of heart with hypertrophic cardio-
myopathy shows predominantly apical involvement of the left ventricle
and massive left atrial enlargement (arrowhead). The papillary muscles
are malpositioned and misoriented (arrow).
Figure 100.6. Autopsy specimen of heart with hypertrophic car-
diomyopathy shows hypertrophy involving the septum and lateral and
inferior walls of the left ventricle. Hypertrophy of the inferior and free
walls of the right ventricle is also present.
LVOT Obstruction and Mitral Regurgitation
LVOT obstruction is present under resting conditions or with
physiologic provocation in up to 70% of patients. It can be
severe in the resting state, labile (mild in the resting state but
significant with provocation), or latent (absent in the resting
state but occurring with provocation). Outflow obstruction is
exacerbated in the presence of a decreased ventricular pre-
load, decreased ventricular afterload, or increased ventricular
contractility. The mechanism of outflow obstruction involves
basal septal hypertrophy and flow-mediated displacement of
the mitral valve anteriorly into the LVOT such that the mitral
leaflet can contact the ventricular septum (Figures 100.9 and
100.10). The temporal sequence has been described as eject-
obstruct-leak. The accelerated flow around the hypertrophied
basal septum pushes the mitral leaflet, while suction forces may
Subaortic Midventricular
Apical Diffuse
Figure 100.7. Diagram of Hypertrophic Cardiomyopathy Variants.
 100 Hypertrophic Cardiomyopathy
Figure 100.8. Hypertrophic Cardiomyopathy (Short-Axis View).
contribute to the systolic anterior motion of the mitral valve.
Other anatomical features that may contribute to the obstruc-
tion include anterior displacement of papillary muscles, hyper-
trophied papillary muscles, and direct insertion of the papillary
muscles onto the mitral valve.
The LVOT obstruction can produce symptoms by several
mechanisms. The obstruction itself can limit cardiac output and
result in effort-related symptoms such as dyspnea or presyncope.
Obstruction also increases left ventricular pressures, which can
induce ischemia through increased myocardial oxygen demand
and decreased perfusion pressure. The high contraction load on
the left ventricle impairs diastolic filling of the left ventricle.
Mitral regurgitation associated with the obstruction can further
elevate the left atrial pressure.
Arrhythmias
Arrhythmias contribute to the pathophysiology of HCM. Atrial
arrhythmias are common (occurring in up to 25% of patients in
some series) and may cause severe hemodynamic deterioration
Figure 100.9. Hypertrophic Cardiomyopathy With Simulated Systolic Anterior Motion of the Anterior Leaflet of the Mitral Valve. Left, Mitral
valve closed in systole. Right, Mitral valve open in diastole, with systolic anterior motion of the anterior leaflet.
901
from loss of atrial contraction as well as from rapid heart rates.
Patients with HCM are very dependent on normal atrial func-
tion for filling of the stiff hypertrophied left ventricle. Cardiac
output may decrease as much as 40% if atrial fibrillation
occurs, leading to rapid deterioration of the patient’s symp-
toms. Ventricular ectopy is a common finding on Holter moni-
toring. Sustained ventricular tachycardia and fibrillation are
the most likely mechanisms of syncope and sudden death in
these patients. For a more comprehensive review of the arrhyth-
mias in HCM, SEE CHAPTER 20 (“HERITABLE CARDIOMYOPATHIES
AND CHANNELOPATHIES: CLINICAL P RESENTATIONS, GENETICS, AND
IMPLICATIONS OF GENETIC TESTING”).
Autonomic Dysfunction
Approximately 25% of HCM patients have an abnormal blood
pressure response to exercise, as defined by either a failure of
systolic blood pressure to increase more than 20 mm Hg or a
decrease in systolic blood pressure. Even with an appropriate
increase in cardiac output, inappropriate systemic vasodilatation
during exercise leads to an inability to augment and sustain sys-
tolic blood pressure. There is speculation that patients with HCM
have a high degree of abnormal autonomic tone, which is associ-
ated with a poorer prognosis.
• Outflow obstruction is exacerbated in the presence of a decreased
ventricular preload, decreased ventricular afterload, or increased
ventricular contractility.
• Cardiac output may decrease as much as 40% if atrial fibrillation
occurs.
Clinical Management of HCM
Successful treatment of HCM requires appreciation of the
pathophysiology as it applies to each patient.
Clinical Presentation and Diagnosis
HCM may be newly diagnosed in patients of any age, from early
childhood to advanced age. The clinical presentation varies
902 IX Cardiomyopathy and Heart Failure
Figure 100.10. Outflow Tract Obstruction in Hypertrophic Cardio-
myopathy. Subaortic septal patch (mitral contact lesion) is shown (arrow).
widely. Patients may be asymptomatic, with the diagnosis made
on the basis of a heart murmur or abnormal electrocardiographic
findings (Figure 100.11) or during screening before participation
in competitive athletics. Even patients with massive hypertrophy
of the heart can be asymptomatic until they present with sudden
cardiac death.
The typical triad of symptoms in HCM is dyspnea on exer-
tion, angina, and presyncope or syncope. Patients with atrial
fibrillation may also present with systemic embolism. Dyspnea
in HCM is caused by increased left atrial pressure, which can
result from abnormal left ventricular diastolic function, LVOT
obstruction, or significant mitral regurgitation (Figure 100.12).
Angina pectoris is common even without epicardial coronary
artery disease and is related to an abnormal myocardial oxygen
supply-demand mismatch due to hypertrophied left ventricular
I aVR V1 V4
II aVL V2 V5
III aVF V3 V6
Figure 100.11. Electrocardiogram From Patient With Apical Variant
of Hypertrophic Cardiomyopathy. Deeply inverted T waves predomi-
nantly occur in chest leads V2 through V6.
Obstruction Diastolic
SAM/MR sequelae
Ischemia
MR
LAP
CO CO
Figure 100.12. Pathophysiologic Mechanisms in Hypertrophic
Cardiomyopathy. CO indicates cardiac output; LAP, left atrial pressure;
MR, mitral regurgitation; SAM, systolic anterior motion of the mitral
valve.
walls, increased arteriolar compressive wall tension caused by
diastolic relaxation abnormalities, and endothelial dysfunction.
Syncope may result from arrhythmias or a sudden increase in
LVOT obstruction. Patients with HCM frequently have abnormal
autonomic function, and vasodepressor syncope may be part of
the mechanism of syncope.
HCM is a clinical disease that is defined by the finding of left
ventricular hypertrophy without an identifiable provocative cause.
The primary diagnostic test is 2-dimensional echocardiography.
Cardiac magnetic resonance imaging and computed tomography
can also be used to confirm the presence, extent, and distribu-
tion of left ventricular hypertrophy. It is crucial that other stimuli
for hypertrophy (hypertension and valvular or subvalvular aortic
stenosis) or increased wall thickness (cardiac amyloidosis, Fabry
cardiomyopathy, and Friedreich ataxia) are considered before the
diagnosis of HCM is given to a patient.
HCM may be distinguished from athletic heart by careful
evaluation that includes echocardiography and potentially other
imaging such as cardiac magnetic resonance imaging. In HCM,
left ventricular septal wall thickness is typically unusual or
asymmetric and more than 15 mm, the left atrium is dilated (>4
cm), and the left ventricular end-diastolic diameter is less than
45 mm. In athletic heart, the distribution of hypertrophy is more
commonly concentric, the left ventricular septal wall thickness
is less than 15 mm, the left atrium is less than 4 cm, and, impor-
tantly, the left ventricular cavity is often dilated (left ventricu-
lar end-diastolic diameter >45 mm). In athletic heart, the left
ventricular hypertrophy regresses after training stops, usually
within 3 months. Diastolic function is normal in athletic heart
but abnormal in HCM.
Physical Examination
Physical examination findings are always abnormal when there
is obstruction of the LVOT, but in nonobstructive HCM the
abnormalities may be less obvious. The hallmark in HCM is the
finding of severe myocardial hypertrophy. This is detected by
palpation of the left ventricular apical impulse, which is localized
but sustained. Frequently there is a palpable presystolic impulse
of the augmented atrial contraction (palpable S4 or bisferiens api-
cal impulse). In the presence of outflow obstruction, there is a
“triple ripple,” although this classic finding is rarely observed.
A bisferiens apical impulse or double apical impulse is more
common. The first impulse is the large atrial kick (atrial boost or
 100 Hypertrophic Cardiomyopathy 903

presystolic boost), and the next impulse is a sustained left ven-
tricular apical impulse. The atrial kick is from a forceful atrial
systole caused by atrial hypertrophy in response to the chroni-
cally elevated left ventricular diastolic pressure and mitral regur-
gitation. The jugular venous pressure may be slightly increased,
with a prominent a wave indicating abnormal diastolic function
of the right side of the heart.
The carotid pulse has a rapid upstroke due to the hyperdy-
namic systolic function and rapid ventricular emptying. In the
presence of LVOT obstruction, the carotid upstroke has a dis-
tinctive, jerky bisferiens quality (“spike-and-dome” pulse). The
spike indicates the initially rapid ventricular emptying phase; the
dome corresponds to the onset of ventricular obstruction, fol-
lowed by the more gradual increase in ventricular pressure to
overcome the gradient.
A harsh systolic ejection murmur is heard across the entire
precordium and radiates to the apex and base of the heart but not
to the neck. In many instances, a separate mitral regurgitation
murmur may be auscultated. Both murmurs respond in a similar
manner to examination maneuvers that change the loading con-
ditions of the left ventricle.
The murmur of HCM is increased by maneuvers that decrease
left ventricular end-diastolic volume (decreased venous return,
decreased afterload, increased contractility, pure vasodilators,
inotropes, dehydration, and the Valsalva maneuver). The murmur
decreases with squatting, passive leg raising, negative inotropes
(eg, β-blockers, verapamil, disopyramide), and any maneuver
that increases left ventricular end-diastolic volume. After the
patient walks briskly or climbs stairs, the murmur is intensified.
In addition to an S4 gallop, there may be an S3 gallop.
Previously, it was thought that a dilated chamber was necessary
for an S3 gallop, but both S3 and S4 gallops are frequent in HCM,
even in the absence of left ventricular dilatation.
• The typical triad of symptoms in HCM is dyspnea on exertion,
angina, and presyncope or syncope.
• Although rare, a “triple ripple” at the apex is a classic physical
finding.
• A “spike-and-dome” carotid contour is another classic physical
finding with LVOT obstruction; the spike indicates the initially
rapid ventricular emptying phase and the dome corresponds to the
onset of ventricular obstruction.
Echocardiography in HCM
Two-Dimensional Echocardiography
Two-dimensional and Doppler echocardiography are the stan-
dard tests for the diagnosis and evaluation of HCM. With
2-dimensional echocardiography, the site and extent of the
myocardial hypertrophy can be visualized and characterized.
Although in the majority of cases the hypertrophy is classically
asymmetric septal hypertrophy with anterolateral extensions, it
can also be diffuse concentric hypertrophy or be localized to
specific areas, such as the apex or free wall of the left ventricle.

1 m/s

Rest Valsalva Amyl nitrite

Figure 100.13. Doppler Dagger-Shaped Late-Peaking Signal of the Intracavitary Gradient in Hypertrophic Cardiomyopathy. The signal is accen-
tuated by the Valsalva response and by inhaled amyl nitrite. At rest, the velocity is 3.0 m/s (gradient, 36 mm Hg). The velocity increases to 3.5 m/s
(gradient, 50 mm Hg) during the Valsalva maneuver and to 4.7 m/s (gradient, 88 mm Hg) after inhalation of amyl nitrite.
904 IX Cardiomyopathy and Heart Failure

Echocardiographic contrast enhancement, cardiac magnetic

resonance imaging, or cardiac computed tomography can aid in
detecting some variants such as apical HCM. Systolic anterior
motion of the mitral valve is frequently present when there is LV
LVOT obstruction.
150 mm Hg

Doppler Echocardiography
Doppler echocardiography is used to study the pathophysiology Ao
of HCM. Dynamic outflow obstruction can be diagnosed and
accurately quantified by a continuous wave Doppler examination
across the outflow tract (Figure 100.13). The modified Bernoulli
equation is used to obtain the peak systolic gradient from the
peak velocity. Coexistent mitral regurgitation can be diagnosed
and semiquantified with color flow Doppler imaging. Diastolic
filling of the left ventricle can also be characterized with the
LA
mitral valve inflow and tissue Doppler assessment of the mitral
annular velocity.
It is important to establish whether the HCM variant is 0 mm Hg
obstructive or nonobstructive. Nonobstructive HCM is a diagno- VPC
sis of exclusion: If no resting obstruction (defined by an LVOT
gradient >30 mm Hg) is noted, provocative maneuvers must be
used. The Valsalva maneuver and inhalation of amyl nitrite are
often used to detect the presence of latent obstruction. If these Figure 100.14. The Brockenbrough sign shows an increase in left
methods do not detect a significant gradient and if the patient ventricular (LV) systolic pressure, a decrease in ascending aortic (Ao)
has effort-related symptoms that may be attributable to obstruc- systolic pressure, and an increase in the pressure gradient between the
tion, the next step is exercise echocardiography or, occasionally, LV and the Ao. Note that there is also a decrease in the pulse pressure
isoproterenol infusion. When obstruction is present, the site and in the Ao (systolic blood pressure minus diastolic blood pressure). LA
severity should be quantified and defined, the associated mitral indicates left atrium; VPC, ventricular premature contraction.
regurgitant severity should be defined, and the diastolic ventricu-
lar dysfunction should be assessed. children and family members participating in competitive athletics
Doppler echocardiography is used to quantify and localize should undergo screening every 12 to 18 months.
the level and severity of obstruction. It is important to distin- 3. HCM patients should avoid participating in competitive athletics or
guish between true outflow tract signals and signals arising from other strenuous activity, but they may participate in low-level aerobic
mitral regurgitation or intracavitary obliteration. Typically, the exercise to promote general cardiovascular health.
HCM Doppler signal appears as a late-peaking, dagger-shaped 4. Patients should avoid dehydration.
signal (Figure 100.13). The aortic valve closure signal is used to 5. Holter monitoring should be performed for 24 to 48 hours to detect
ventricular arrhythmias and to determine risk stratification.
distinguish HCM from mitral regurgitation. The mitral regurgi-
6. Administration of pure vasodilators, high-dose diuretics, and posi-
tant signal in HCM may be a late-peaking signal, but it continues tive inotropes should be avoided since they may exacerbate LVOT
until mitral valve forward flow begins in diastole, whereas the obstruction.
LVOT obstruction signal ends with aortic valve closure. Doppler-
derived gradients in patients with HCM correlate well with cath-
eter-derived gradients during simultaneous examinations. β-Blockers, Calcium Channel Blockers,
Cardiac catheterization is not necessary to diagnose HCM or and Disopyramide
to evaluate the severity of LVOT obstruction or mitral regurgita- For patients who have obstructive cardiomyopathy and symptoms,
tion; however, it may be useful in select patients who have chal- first-line pharmacologic therapy should be negative inotropic
lenging results from echocardiographic studies. agents. β-Blockade is often the first-line therapy. β-Blockers
relieve symptoms in about 50% of patients by 1) slowing the heart
• Typically, the HCM Doppler signal appears as a late-peaking,
dagger-shaped signal.
rate, resulting in a longer diastolic filling time and decreased
• The Brockenbrough sign is a classic hemodynamic function in
HCM (Figure 100.14). Table 100.2. Comparison of Surgical Septal Myectomy and
Percutaneous Septal Ablation
Management of Obstructive HCM Feature Myectomy Ablation
General Principles Procedural mortality, % ≤1 1–2
There are several general guidelines for management of HCM: Residual gradient (immediate), mm Hg <10 <25
Success rate, % ≥95 ≥85
1. All first-degree relatives of patients with HCM should undergo Permanent pacemaker rate, % 1–2 ≥5–10
screening with either genetic testing or cardiac imaging, and younger Subsequent sudden death or ICD Very low Uncertain
affected family members should receive genetic counseling if they discharge risk
plan to have a family. Intramyocardial scarring None Present
2. With the use of an imaging-based screening algorithm, surveil-
lance echocardiography should be repeated every 5 years for adults; Abbreviation: ICD, implantable cardioverter-defibrillator.
 100 Hypertrophic Cardiomyopathy 905

Table 100.3. Selection Criteria for Invasive Therapy in Obstructive Hypertrophic Cardiomyopathy
Therapy Good Candidates Poor Candidates
Surgical septal myectomy LVOTO with NYHA class III or IV symptoms Severe comorbidity that markedly increases risks
Any age
Coexistent mitral valve or other structural
cardiac abnormalities
Percutaneous septal ablation LVOTO with NYHA class III or IV symptoms More diffuse hypertrophy
Isolated basal septal hypertrophy Extreme gradients (>80 mm Hg)
Moderate gradient (40–60 mm Hg) Left bundle branch block
Older age Abnormal mitral valve or mitral support structures
Preexisting PPM Preexisting risk factors for SCD in the absence of ICD
Contraindication to surgical septal myectomy
Dual-chamber pacing LVOTO with NYHA class III or IV symptoms Rapid native atrioventricular conduction
Coexistent chronotropic incompetence Abnormal mitral valve or mitral support structures
Contraindication to surgical septal myectomy
and percutaneous septal ablation
Abbreviations: ICD, implantable cardioverter-defibrillator; LVOTO, left ventricular outflow tract obstruction; NYHA, New York Heart Association; PPM, permanent
pacemaker; SCD, sudden cardiac death.

myocardial oxygen consumption, thus reducing myocardial
ischemia, and 2) diminishing the exercise-induced augmentation
of LVOT obstruction through a direct negative inotropic effect.
If β-blockade does not adequately decrease the intraven-
tricular gradient and control symptoms, a calcium channel
blocker—either verapamil or diltiazem—may be administered.
Care must be taken when prescribing calcium channel blockers
for patients with large outflow tract obstruction, advanced heart
failure symptoms, or signs of pulmonary congestion because
peripheral vasodilatation may cause acute hemodynamic deteri-
oration. Dihydropyridine calcium channel blockers are contrain-
dicated in obstructive HCM since they are pure vasodilators.
Disopyramide, a class I antiarrhythmic agent with strong nega-
tive inotropic properties, may also be used to treat HCM in con-
junction with either a β-blocker or a calcium channel blocker;
however, anticholinergic effects can cause urinary retention (in
men) and dry mouth.
Surgery
For patients who have obstructive HCM and who continue to have
severe, disabling, effort-related symptoms despite appropriate
Narrow
Pace RV apex LVOT Pace RV apex
Abnormal Venturi Remodeling
septal motion effect
Acute SAM Chronic
medication adjustment, surgical septal myectomy is usually the
primary treatment (Tables 100.2 and 100.3). Septal myectomy
is a well-proven therapy that can abolish the LVOT gradient and
provide excellent long-term relief. The resected muscle does not
regrow. Procedural complications are infrequent (<5%), and
mortality associated with the operation is rare (<1%) across all
age ranges in reports from high-volume centers. With surgery,
the LVOT is effectively widened, thereby abolishing the gradi-
ent and eliminating mitral regurgitation mediated by systolic
anterior motion. In nonrandomized comparisons, patients who
undergo septal myectomy have better survival than patients who
receive nonsurgical management.
Alcohol Septal Ablation
Alcohol septal ablation has emerged as a potential alternative to
surgical myectomy for selected patients with obstructive HCM
and drug-refractory symptoms. In this percutaneous, catheter-
based procedure, pure ethanol is injected into the septal perfora-
tor coronary artery that supplies the hypertrophied myocardium
adjacent to the point of outflow obstruction (the systolic anterior
motion septal contact point). This portion of the septum becomes
akinetic, eventually leading to scarring (thinning) in this area,
which dramatically reduces the outflow obstruction. A relatively
high rate of complete heart block necessitates permanent pac-
ing (5%–10% in experienced centers). The procedural mortality
rate of about 2% in pooled data is similar to that of myectomy
as reported from meta-analyses with short-term follow-up. This
procedure induces a nonreperfused septal infarction, which has
raised concerns about proarrhythmia. Septal ablation cannot
offer relief from coexistent structural valve abnormalities, and
the ability to treat the precise site depends on appropriate coro-
nary anatomy. Data suggest that patients younger than 65 years
have better survival free from symptoms with surgical myectomy.
With these limitations and complication profile, patient eligibil-
ity is limited and septal ablation should be performed only after
careful discussion with the patient.

Figure 100.15. Possible Mechanisms for the Beneficial Effects of
Dual-Chamber Pacing in Hypertrophic Cardiomyopathy. LVOT indi-
cates left ventricular outflow tract; RV, right ventricular; SAM, systolic
anterior motion of the mitral valve.
Dual-Chamber Pacing
Dual-chamber pacing has been advocated as another means of
relieving outflow obstruction. The proposed mechanisms for the
beneficial effects of pacing include optimizing atrioventricular
906 IX Cardiomyopathy and Heart Failure

synchrony and altering the ventricular activation sequence (ie,
apex to base), with potentially long-term remodeling and wid-
ening of the LVOT (Figure 100.15). Although initial observa-
tional series were encouraging, randomized controlled trials
did not demonstrate significant, objective improvements. It now
appears that only about 30% of patients treated with pacemak-
ers to relieve obstruction have lasting benefit. Unfortunately, no
preprocedural variables can identify which patients will respond
favorably to pacing therapy. On the basis of these findings, pace-
maker therapy to relieve symptomatic outflow obstruction is
reserved for patients who have contraindications to myectomy
and septal ablation.
Management of Nonobstructive HCM
The majority of patients with HCM do not have outflow obstruc-
tion. If these patients become symptomatic, the symptoms are
usually related to heart failure due to inherent diastolic dysfunc-
tion or atrial arrhythmias (or both). The treatment options for
these patients are limited. Diuretics are often useful in the early
stages of management. Emerging data suggest that inhibition of
the renin-angiotensin-aldosterone cascade may favorably alter
the diastolic properties of the left ventricle. If symptoms continue
to progress, cardiac transplant becomes an option.
Risk Assessment for Sudden
Cardiac Death in HCM
The natural history of HCM is highly variable, but population-
based studies suggest that most HCM patients have near-normal
longevity. However, the annual rate of sudden cardiac death is
still approximately 1%. HCM is the leading cause of sudden
death in competitive and school-aged athletes. The challenge for
the clinician is to identify patients who are at increased risk, so
that preventive strategies (eg, implantable defibrillators) may be
appropriately used. A personal history of resuscitated cardiac
arrest or sustained ventricular tachycardia is a clear indication
for implanting a defibrillator. Several other clinical features
also portend a poorer prognosis, including a family history of
sudden death; a history of recent, unexplained syncope in the
young; repetitive nonsustained ventricular tachycardia; mas-
sive (>30 mm) left ventricular hypertrophy; and failure to aug-
ment systolic blood pressure by 20 mm Hg with exercise. The
risk appears to increase with the number of risk factors present.

Figure 100.16. ICD Indications in Hypertrophic Cardiomyopathy. ABPR indicates ambulatory blood pressure; FHSCD, family history of sud-
den cardiac death; HCM, hypertrophic cardiomyopathy; ICD, implantable cardioverter-defibrillator; LVOT, left ventricular outflow tract; MLVWT,
maximum left ventricular wall thickness; NSVT, nonsustained ventricular tachycardia; SCD, sudden cardiac death; VF, ventricular fibrillation; VT,
ventricular tachycardia.
 100 Hypertrophic Cardiomyopathy
Risk should be a part of the discussion involving decisions about
implanting a defibrillator (Figure 100.16). If only 1 of these fac-
tors is present, individualized discussion with the patient about
all the risks and benefits is required.
• The annual rate of sudden cardiac death is approximately 1%.
• HCM is the leading cause of sudden death in competitive and
school-aged athletes.
• Clinical features that portend a poorer prognosis include a family
history of sudden death; a history of recent, unexplained syncope
in the young; repetitive nonsustained ventricular tachycardia; mas-
sive (>30 mm) left ventricular hypertrophy; and failure to augment
systolic blood pressure by 20 mm Hg with exercise.
Summary
HCM is a disease marked by heterogeneity. It is generally
thought to be an autosomal dominant disease with mutations
907
often localized to the genes for the proteins comprising the car-
diac sarcomere. Although sudden cardiac death is more com-
mon in patients with HCM than in the general population, it is
still rare in HCM. Risk stratification is a challenging art. LVOT
obstruction is a common feature that predisposes patients to
effort-related dyspnea, angina, and presyncope. Pharmacologic
treatment is usually successful in controlling symptoms. Surgical
septal myectomy is the primary treatment strategy for sympto-
matic patients who have no response to medications, and alcohol
septal ablation can be offered to select patients (ie, those with
considerable comorbidity) as a nonsurgical alternative.
Abbreviations
HCM hypertrophic cardiomyopathy
LVOT left ventricular outflow tract
S3 third heart sound
S4 fourth heart sound
 101
Cardiac Transplantationa
BARRY A. BOILSON, MD, RICHARD C. DALY, MD,
and SUDHIR S. KUSHWAHA, MD
Background
In the United States, an estimated 5.7 million people carry a
diagnosis of heart failure and almost 300,000 people die of heart
failure each year. The prevalence of heart failure is increasing
as our society ages. Advances in medical therapy have resulted
in improved survival of patients with moderate or severe heart
failure, but the prognosis for patients with end-stage heart failure
remains poor.
Orthotopic cardiac transplant is a highly successful procedure
for the treatment of end-stage heart disease. The modern era of
human cardiac transplantation started with research at Stanford
University in the 1950s and 1960s and ultimately led to the first
human cardiac transplant, which was performed by Christiaan
Barnard in Cape Town, South Africa, in 1967.
Despite major advances since that first transplant, the num-
ber of transplants performed worldwide has plateaued, primarily
because of the limited supply of donor organs. Approximately
2,000 cardiac transplants are performed yearly in the United
States, and cardiac transplant is the gold standard of car-
diac replacement therapy. However, cardiac transplant is not
appropriate for many patients because of their age and comor-
bidities. This has led to the development of new therapeutic
avenues, including ventricular assist devices and the total arti-
ficial heart.
a
Portions previously published in Boilson BA, Raichlin E, Park SJ,
Kushwaha SS. Device therapy and cardiac transplantation for end-
stage heart failure. Curr Probl Cardiol. 2010 Jan;35(1):8–64. Used with
permission.
Abbreviations and acronyms are expanded at the end of this chapter.
908
Indications for Cardiac Transplant
The current AHA/ACC guidelines on selecting adult patients for
cardiac transplant are reported in the AHA/ACC practice guide-
lines for managing chronic heart failure in adults (Box 101.1).
The primary focus is on patients with severe functional impair-
ment of cardiac origin or dependence on inotropic support (or
both). Rare indications include recurrent malignant arrhythmias
refractory to medical therapy and debilitating refractory angina
pectoris resulting from severe nonrevascularizable coronary
artery disease. Contraindications for cardiac transplant are rela-
tive and dependent on whether they can be modified before sur-
gery; they include pulmonary disease, pulmonary hypertension,
diabetes mellitus with complications, systemic disease (including
malignancy), and peripheral vascular disease (Box 101.2). Age is
included, but data on outcomes among older recipients have been
conflicting.
Potential transplant recipients must be able to demonstrate
that they understand the heart transplant procedure and can take
medications as directed and adhere to a complex long-term med-
ical regimen essential for survival after transplant. They need
to demonstrate emotional stability and a realistic attitude in
response to past and current illness. The patient must receive a
suitable plan for caring for the transplanted organ.
Donor and Recipient Matching
A potential donor must meet several criteria:
1. The donor must meet national or regional criteria for brain death.
2. Electrocardiographic and echocardiographic findings should be
normal.
3. A donor older than 45 years (often considered for an older recip-
ient) should undergo coronary angiography to exclude significant
 101 Cardiac Transplantation 909

Box 101.1. Indications for Cardiac Transplant Box 101.2. Contraindications for Cardiac Transplant
Absolute indications Severe pulmonary hypertension
Hemodynamic compromise due to heart failure PVR >3 Wood Units or PA systolic pressure >70 mm
Hg and unresponsive to vasodilators or mechanical
a. Refractory cardiogenic shock
circulatory support
b. Dependence on intravenous inotropic support for
adequate organ perfusion PVR >6 Wood Units
.
c. Peak VO2 <10 mL/kg/min Transpulmonary gradient >20
Severely limiting nonrevascularizable ischemic heart Pulmonary disease
disease affecting activities of daily living Significant primary lung disease with FEV1 <1 L or
Recurrent symptomatic ventricular arrhythmias <40% of the predicted result
refractory to therapy Recent (<4 wk) pulmonary infarct
Relative indications Renal disease
.
Peak VO2 11–14 mL/kg/min with major limitations Primary renal disease that would shorten life
affecting activities of daily living expectancy

Recurrent unstable angina refractory to current Irreversible renal dysfunction not explained by heart
therapy failure

Recurrently labile fluid balance or renal function in Creatinine clearance <35 mL/min with optimized
chronic heart failure despite full patient adherence hemodynamics
to therapy Hepatic disease
Insufficient indications Primary hepatic disease that would shorten life
expectancy
Presence of the following without other indications
for transplant: Irreversible hepatic dysfunction not explained by
heart failure
a. Impaired left ventricular systolic function
Gastrointestinal disease
b. Previous history of NYHA class III or IV heart failure
. Conditions potentially affecting absorption of
c. Peak VO2 >15 mL/kg/min
medications
.
Abbreviations: NYHA, New York Heart Association; VO2, oxygen Uncontrolled gastrointestinal tract bleeding
consumption per unit time. Severe uncorrected peripheral or cerebrovascular
Previously published. See “Credit Lines” section. disease
Diabetes mellitus with evidence of significant
end-organ dysfunction
coronary artery disease. Otherwise, the risk factor profile for coro- Malignancy thought not to be in a curative state
nary artery disease should be low and there should be no evidence of or with a high likelihood of recurrence
untreated acute infection or systemic malignancy.
Uncontrolled infection or sepsis
4. Results from human immunodeficiency virus infection and hepatitis
screens should be negative. HIV infection
5. Potential donors with cardiac trauma are usually excluded. Poor psychosocial status
Donor-recipient matching depends on a few key issues: History of a behavior pattern or psychiatric illness
1. Blood type: ABO matching is mandatory. Matching of rhesus factor likely to interfere with adherence to a disciplined
status is not required since cardiac myocytes do not express the rhe- medical regimen
sus antigen. Inadequate social support symptom
2. Body size: Generally, the donor’s body weight should be at least 80% Current or recent tobacco use
of the recipient’s.
3. Pulmonary hypertension: If the recipient has an elevated pulmo- Unresolved drug or alcohol dependency
nary vascular resistance or pulmonary artery systolic pressure Obesity
(Box 101.2), a larger donor heart is usually selected to ensure ade-
quate right ventricular functional reserve. BMI >35
4. Geographic location of donor: Ensure the shortest possible cold Weight >140% predicted for height and sex
ischemia time for the heart after it has been explanted from the
Malnutrition
donor. If the time exceeds 4 hours, outcomes may be compromised,
especially if the donor organ has significant hypertrophy. BMI <18
5. Anti-HLA antibody titer: For cardiac transplant, unlike renal trans- Age >65 y
plant, HLA crossmatching is performed only if the recipient has a
significant titer of preformed antibodies. The recipient’s titer of pre- Abbreviations: BMI, body mass index (calculated as weight in
formed antibodies is measured in the routine pretransplant assess- kilograms divided by height in meters squared); FEV1, forced
ment. The titer reflects the degree of sensitization of the patient to expiratory volume in the first second of expiration; HIV, human
foreign antigens of HLA classes HLA-A, HLA-B, and HLA-DR. immunodeficiency virus; PA, pulmonary artery; PVR, pulmonary
Significant elevation of the anti-HLA antibody titer is more likely vascular resistance.
in patients who have received multiple blood transfusions and in
910 IX Cardiomyopathy and Heart Failure
multiparous female patients. Elevations have been associated with
increased risk of hyperacute rejection, antibody- and cell-mediated
rejection, and cardiac allograft vasculopathy. Currently, Luminex
anti-HLA antibody screening is used to quantitatively measure the
levels of antibody to different HLA antigens by immunofluores-
cence. This is much more sensitive and accurate than prior methods
and is now incorporated by many centers for donor selection. With
this technique, a specific antigen can be avoided by not accepting
a donor who carries that antigen if the recipient has an excessively
high antibody level to it (ie, a virtual crossmatch).
Operative Details
The original operative technique described by Shumway involved
removal of the native heart and anastomosis of the recipient heart
at the midatrial level bilaterally (Figures 101.1A and 101.1B).
This biatrial anastomosis technique has been shown to increase
the incidence of atrial arrhythmia, right atrial thrombus, and
tricuspid valve dysfunction.
In the late 1980s, a new approach preserved the entire donor
heart but involved 8 anastomoses: the 4 pulmonary veins, the
superior and inferior venae cavae, the pulmonary artery, and the
aorta. The disadvantage of this technique was the increased time
required for anastomosis of the pulmonary veins and significantly
increased cold ischemia time of the donor heart. Therefore, in
1991, the procedure was simplified by preserving the recipi-
ent pulmonary veins and anastomosing a small cuff of recipi-
ent left atrial tissue to the donor left atrium while retaining the
Figure 101.1. Cardiac Transplant Operative Techniques. A and B, Standard Shumway (biatrial) anastomosis. C and D, Bicaval anastomosis. Ao
indicates aorta; IVC, inferior vena cava; LA, left atrium; PA, pulmonary artery; RA, right atrium; SVC, superior vena cava.
anastomosis of the venae cavae and thus sparing the integrity of
the donor right atrium. This bicaval anastomosis technique has
been reported to preserve atrial contractility, sinus node func-
tion, and tricuspid valve competence, but it also increases the
operative time, including the cold ischemia time (Figures 101.1C
and 101.1D). Recent evidence has again shown short-term clin-
ical benefits of the bicaval technique, when compared with
the biatrial technique, but without a significant difference in
long-term survival.
Immunosuppression
Immunosuppression can be considered as initial induction ther-
apy or long-term maintenance immunosuppression. Induction
therapy traditionally involved the use of the anti-CD3 cell recep-
tor monoclonal antibody muromonab-CD3. However, this agent
is now rarely used (<1% of patients) since evidence has linked
it to an elevated risk of PTLD. Most institutions that use induc-
tion therapy currently use interleukin 2 antagonists (eg, basil-
iximab, daclizumab) or polyclonal antilymphocytic antibodies
(eg, rabbit or equine antithymocyte globulin, as used at Mayo
Clinic). Newer agents under evaluation for this purpose include
the anti-CD52 antibody alemtuzumab (Campath). The ration-
ale for induction therapy is to decrease the risk of early acute
rejection through enhanced immunosuppression and to decrease
the risk of postoperative renal dysfunction by delaying the start
of calcineurin inhibitor therapy. However, the use of induction
 101 Cardiac Transplantation
Ao
D
SVC
PA
LA
IVC
Figure 101.1. (Continued)
therapy varies among centers worldwide; at almost half, its use
is avoided completely.
Maintenance immunosuppression varies among centers. The
traditional model includes use of a calcineurin inhibitor (cyclo-
sporine or tacrolimus), an antiproliferative agent (azathioprine or
mycophenolate), and prednisone, but preference for tacrolimus
over cyclosporine has increased in recent years. Tacrolimus is
now the most frequently used calcineurin inhibitor worldwide.
The largest randomized clinical trial that compared tacrolimus
with cyclosporine in over 300 patients showed a significantly
lower incidence of severe rejection in the tacrolimus group at
6 months but no difference in patient or graft survival at 18
months. Studies to date comparing cyclosporine and tacrolimus
have shown a lower incidence of hypertension and dyslipidemia
in tacrolimus-treated patients compared with those treated with
cyclosporine. However, the incidence of new-onset diabetes
mellitus with tacrolimus appears to be higher. Cyclosporine is
classically associated with gingival hyperplasia, hirsutism, neph-
rotoxicity, and hypertension.
Mycophenolate has replaced azathioprine as the antiprolifera-
tive agent of choice and is currently used in over 80% of trans-
plant recipients at 1 year. The advantages of mycophenolate over
azathioprine include a decreased incidence of acute rejection and
911
mortality, and possibly a decreased incidence of cardiac allo-
graft vasculopathy. The underlying mechanisms for this benefit
may include preferential anti–B-cell activity compared with aza-
thioprine and decreased production of anti-HLA antibodies after
transplant. Mycophenolate causes less myelosuppression than aza-
thioprine, but it does cause dose-related leukopenia and it is associ-
ated with gastrointestinal tract side effects (eg, nausea, diarrhea).
Newer agents such as the mTOR inhibitor rapamycin (siroli-
mus and its derivative everolimus, also known as PSIs) are
becoming more common. They have a potent immunosuppres-
sive capacity coupled with cytostatic effects outside the immune
system. Studies have shown that PSIs used in combination with
calcineurin inhibitors may reduce the incidence of CAV, and
subsequent data have demonstrated a significant renal-sparing
effect of sirolimus monotherapy compared with cyclosporine. In
heart transplant recipients, PSIs may have other beneficial ther-
apeutic effects, including regression of cardiac hypertrophy and,
possibly, antineoplastic effects. For these reasons, over 20% of
patients worldwide have been transitioned to sirolimus at 5-year
follow-up; however, sirolimus impairs wound healing, so its
use must be delayed for 3 to 6 months after cardiac transplant.
Sirolimus has also been associated with pulmonary toxicity,
hyperlipidemia, and peripheral edema.
912 IX Cardiomyopathy and Heart Failure
Everolimus is shorter acting than sirolimus and remains under
evaluation in clinical trials. Studies have shown that, compared
with azathioprine-based therapy, everolimus in combination with
full-dose cyclosporine in de novo cardiac transplant recipients
reduced the risk of acute cellular rejection and the development
of allograft vasculopathy at 24 months.
Complications After Cardiac Transplant
Early Complications
Early complications after cardiac transplant include primary
graft failure, acute and hyperacute rejection, arrhythmia, bleed-
ing, and infection.
Primary Graft Failure
Primary graft failure includes ischemic or reperfusion injury and
right heart failure due to pulmonary hypertension. It accounts
for up to 40% of deaths within the first 30 days after cardiac
transplant. Extended cold ischemia time of the donor heart and
elevated pulmonary vascular resistance in the recipient before
transplant are significant risk factors. Treatment usually requires
inotropic support, use of vasodilators to reduce pulmonary
and systemic afterload, and, occasionally, mechanical support.
Rarely, emergent retransplant is required.
Hyperacute Rejection
Hyperacute rejection is a rare form of early rejection that occurs
when the donor heart is initially perfused with blood from the
recipient. It is caused by preformed donor-specific antibodies of
the recipient circulating within the coronary circulation of the
donor heart, severe microvascular injury, and thrombosis, fre-
quently resulting in loss of the graft. It is associated with a high
titer of preformed anti-HLA antibodies in the recipient. The titer
is usually detected on pretransplant screening. Patients in this
category usually have prolonged waiting times for a suitable
organ because they require crossmatching (virtual or prospec-
tive) to a potential donor heart to minimize the risk of this rare
but devastating complication. With crossmatching, hyperacute
rejection is encountered very rarely.
Acute Rejection
Acute rejection is common and usually T-cell mediated (cel-
lular immune response), but sometimes it is caused by recipi-
ent antibodies to donor antigens (humoral immune response).
Interestingly, in an ISHLT analysis, patients treated with all
forms of induction therapy had more acute rejection episodes in
the first year after transplant than patients not treated with induc-
tion therapy. These findings may reflect a selection bias, in which
patients at higher risk of acute rejection (female recipients and
younger patients) are more likely to receive induction therapy. In
addition, patients treated with tacrolimus in preference to cyclo-
sporine soon after transplant have fewer rejection episodes in the
first year, especially if tacrolimus was given in combination with
mycophenolate.
Acute rejection does not usually cause symptoms unless it is
fulminant and severe. Detection of acute rejection is important
because frequent episodes, especially in the first year, are associ-
ated with reduced graft survival and possibly with an increased
incidence of cardiac allograft vasculopathy. Therefore, screening
is required with frequent endomyocardial biopsy. Biopsies are
usually performed weekly for the first month, then every 2
weeks until 3 months after transplant, and then less frequently.
Alternative approaches are under investigation to decrease the
number of biopsies since the endomyocardial biopsy procedure
itself has been associated with an increased risk of tricuspid valve
injury. Alternative approaches include the use of microarrays for
identification of key (candidate) genes upregulated and down-
regulated in early rejection. The CARGO study demonstrated
the utility of gene expression profiling to rule out significant car-
diac rejection, which may be important in reducing the burden
of endomyocardial biopsy on cardiac transplant recipients. The
IMAGE study, published in 2010, demonstrated noninferiority of
this approach compared with a traditional biopsy-based approach
according to the incidence of the primary outcome of rejection
with hemodynamic compromise, graft dysfunction due to other
causes, death, or retransplant. However, to date only a few pro-
grams have adopted gene expression profiling into their standard
practice.
The ISHLT grading system for acute cellular rejection
was changed in 2004. Currently, the following system is used
(Figure 101.2):
• Grade 0R: no rejection. Normal.
• Grade 1R, mild: interstitial or perivascular infiltrate (or both) with
≤1 focus of myocyte damage.
• Grade 2R, moderate: ≥2 foci of infiltrate with associated myocyte
damage.
• Grade 3R, severe: diffuse infiltrate with multifocal myocyte dam-
age, with or without edema, hemorrhage, or vasculitis.
Humoral rejection or AMR (Figure 101.3) is less well recog-
nized in cardiac transplantation medicine and is probably under-
diagnosed. However, it is associated with an increased incidence
of cardiac allograft vasculopathy and mortality. Histologically, it
is characterized by endothelial swelling and the presence of mac-
rophages and neutrophils in the capillaries with fibrin deposition.
The immunofluorescence markers used for diagnosis of AMR
have changed repeatedly over the years. Previously, they failed to
correlate with the clinical severity of the condition. The current
ISHLT working formulation suggests that AMR diagnosis should
be based on a combination of histologic and immunopathologic
biopsy findings (eg, C3d and C4d staining), presence of donor-
specific antibodies, and allograft dysfunction. However, the
methods used in the diagnosis of AMR vary among institutions.
One series showed that AMR most commonly occurs early after
cardiac transplant, and in those cases, association with elevated
anti-donor HLA antibodies is frequent. However, this is less
common in patients in whom AMR develops later after trans-
plant. In those patients, AMR is associated with malignancy
or recent infection, suggesting activation of antibody-mediated
immunity by a new antigen present on an invading pathogen or
expressed on tumor cells.
Arrhythmia
Arrhythmia occurs frequently after cardiac transplant. Sinus
tachycardia results from vagal denervation of the donor heart.
Sinus node injury is also common early after transplant, par-
ticularly when the biatrial anastomosis technique is used. Sinus
node function generally improves with time, but occasionally
(4%–12% of patients) permanent pacemaker implantation is
required before the patient is dismissed from the hospital, and 1
 101 Cardiac Transplantation
0R
2R
Figure 101.2. Cell-Mediated Cardiac Rejection. Current (2004) International Society for Heart and Lung Transplantation grading system. Grade
0R indicates no rejection; 1R, mild rejection; 2R, moderate rejection; 3R, severe rejection. (hematoxylin-eosin, original magnification ×40 [OR],
×100 [1R], ×200 [2R and 3R]).
study has suggested that dual-chamber atrioventricular sequen-
tial pacing would be preferable. Later, sinus node dysfunction
may occur from CAV and involvement of the sinus node artery.
Surgical atrioventricular nodal injury is less frequent. As with
sinoatrial dysfunction, late development of atrioventricular block
is a poor prognostic sign and is often associated with subsequent
CAV. Right bundle branch block also occurs and is often a result
of prolonged cold ischemia time or injury to the right bundle dur-
ing endomyocardial biopsy.
Atrial fibrillation may occur early after transplant during
healing of the atrial anastomoses (resulting from the hetero-
geneity of atrial tissue refractoriness) or later in association
with rejection. Atrial flutter is often of the typical cavotricus-
pid isthmus–dependent type and may be responsive to radiof-
requency ablation between the tricuspid annulus and the right
atrial suture line for donor and recipient.
Atrial and ventricular premature beats can occur early after
transplant and are not significantly associated with rejection.
Nonsustained ventricular tachycardia is common initially after
transplant and then decreases in incidence with time. Sustained
ventricular tachycardia is associated with severe CAV or
rejection. Ventricular fibrillation with sudden death is a hallmark
of severe CAV.
913
1R
3R
Late Complications
Cardiac Allograft Vasculopathy
CAV is the leading cause of late morbidity and death among
heart transplant recipients. Angiographic studies indicate that
CAV occurs in 42% of all heart transplant patients; IVUS, a more
sensitive technique, detects CAV in 75% of patients by 3 years
after transplant. The ISHLT registry indicates that 5 years after
cardiac transplant, CAV and late graft failure (likely from CAV)
together account for 30% of deaths, and over 50% of adult
recipients will have angiographic evidence of CAV at 10 years
(Figure 101.4).
CAV is a diffuse process that affects large epicardial vessels
and the microcirculation, and it may involve the coronary veins
of the transplanted heart. Histologically CAV is characterized
by concentric fibrous intimal hyperplasia, smooth-muscle prolif-
eration and inflammation, and fibrous replacement of the media
(Figure 101.4). These factors contribute to progressive luminal
narrowing, reduction of myocardial blood flow, and endothelial
dysfunction.
• CAV is the leading cause of late morbidity and death among heart
transplant recipients.
914 IX Cardiomyopathy and Heart Failure

0 1+

2+ 3+

Figure 101.3. Antibody-Mediated Cardiac Rejection. Intensity of C4d staining is graded 0 to 3+ according to Mayo Clinic grading (diaminoben-
zidine, original magnification ×200). Grade 0 indicates no rejection, and grades 1+ to 3+ indicate increasing intensity of C4d staining.

The diagnosis of CAV has traditionally relied on coronary
angiography, which has a high specificity (97.8%) but only a
moderate sensitivity (79.3%) in long-term follow-up. The inti-
mal changes in CAV are best detected with IVUS, which is the
gold standard for the early diagnosis of CAV. IVUS, however, is
limited to imaging only the larger epicardial arteries and is not
useful for smaller vessels and branches. A growing body of evi-
dence has demonstrated that virtual histology IVUS is a reliable
tool and offers an in vivo opportunity to characterize the dif-
ferent types of plaque morphology (eg, fibrous, fibrofatty, dense
calcium, and necrotic core). Simultaneous assessment of virtual
histology with IVUS has been suggested as a useful tool for
studying the mechanism and predicting the progression of CAV.
Endothelial dysfunction and plaque formation may lead to rup-
ture and acute coronary syndromes, as in classic atherosclerotic
coronary artery disease. Although there is evidence of some rein-
nervation of cardiac allografts, most transplant recipients do not
experience anginal pain with myocardial ischemia or infarction.
Therefore, patients with CAV most commonly present with silent
myocardial infarction, congestive cardiac failure, or arrhythmia,
which may manifest as sudden cardiac death. CAV, particularly
involving the small vessels, is also a cause of chronic allograft
dysfunction. After CAV is established, it is generally irrevers-
ible, and because of its diffuse (rather than localized) nature,
angioplasty or aortocoronary bypass provide limited long-term
benefit, although either one may have a role in the treatment of
localized lesions. Modification of traditional risk factors may
attenuate disease progression and improve outcome. There is evi-
dence for the role of statins in the prevention of CAV in animal
models and in humans. In severe CAV, however, the prognosis is
grave and the only treatment option is retransplant.
• Endothelial dysfunction and plaque formation may lead to rupture
and acute coronary syndromes, as in classic atherosclerotic coro-
nary artery disease.
The pathophysiology of CAV is thought to be multifactorial,
involving several immunologic and nonimmunologic factors.
In addition to well-known risk factors such as hypertension,
younger recipient age, male sex, and preexisting donor coronary
artery disease, markers of metabolic syndrome have been shown
to be associated with an increased incidence of CAV and a worse
prognosis after heart transplant. Although CAV may develop at
any stage after transplant, events during the first year appear to
be important in its pathogenesis, and the donor’s risk factors may
also have a role.
Experimental evidence has shown that immune activation
may lead to an inflammatory process in the vascular endothe-
lium and result in tissue destruction and potentiation of CAV.
Therefore, in many ways, CAV is thought to be a manifestation of
chronic rejection. Studies have shown that increased frequency of
 101 Cardiac Transplantation 915

Figure 101.4. Coronary Angiography Showing Evidence of Severe Cardiac Allograft Vasculopathy (CAV). Evidence of occlusion is apparent in
the distal left anterior descending coronary artery (LAD) (upper left, arrow), the first diagonal vessels and distal circumflex coronary artery (upper
right), and the severe distal tapering of the right coronary artery (arrowhead) with occlusion of the posterior descending artery and posterolateral
branches (lower left). Photomicrograph shows transmural inflammation and necrosis of the distal LAD at autopsy (lower right) (hematoxylin-eosin;
original magnification ×200).

rejection episodes, anti-HLA antibody titer level, and the degree
of HLA mismatch between donor and recipient are also CAV
risk factors. The importance of systemic inflammation in CAV
has been shown in several studies. CMV infection has also been
associated with the development of CAV.
There is growing evidence that the use of mTOR inhibitors,
such as sirolimus (rapamycin) and everolimus, may attenuate the
progression of CAV. Sirolimus was shown to be less deleterious
to the vasculature than cyclosporine by maintaining nitric oxide
homeostasis and reducing plasma endothelin levels. Sirolimus-
based immunosuppression, compared with cyclosporine-based
immunosuppression, results in less pronounced coronary epi-
cardial endothelial dysfunction. Moreover, the lower systemic
blood pressure in the sirolimus group suggested that the ben-
eficial vascular effects of sirolimus may extend beyond the
coronary circulation. IVUS has shown that everolimus results
in less progression of intimal thickening compared with a
cyclosporine-azathioprine–based regimen, and similar findings
have been reported for comparisons of mycophenolate with aza-
thioprine. Furthermore, C-reactive protein levels also appear to
be significantly decreased with a mycophenolate-based regimen
compared with azathioprine.
Chronic Kidney Disease
Chronic kidney disease after cardiac transplant is an important
cause of morbidity and death, contributing to up to 10% of deaths
by 10 years. However, ISHLT analysis of data from 2001 to 2008
showed that the incidence of renal insufficiency in cardiac trans-
plant recipients has been decreasing. At 5 years after cardiac
transplant, the prevalence of severe renal insufficiency (defined
as serum creatinine >2.5 mg/dL, dialysis, or renal transplant)
decreased from 27% in 2000 to 16% in 2008. This decrease may
be from the increased use of mTOR inhibitors in preference to
calcineurin inhibitors for long-term immunosuppression.
916 IX Cardiomyopathy and Heart Failure
Risk factors for the development of renal dysfunction after
heart transplant include long-term use of calcineurin inhibitors,
renal dysfunction before transplant, older recipient and donor
ages, and the presence of diabetes mellitus and hypertension
before transplant. BK polyomavirus infection has also been
linked with renal dysfunction after heart transplant.
Infection
Infection with community-acquired or opportunistic pathogens is
increased in patients receiving chronic immunosuppression ther-
apy. The risk of infection for specific cardiac transplant recipient
depends on the epidemiologic exposures of that patient and on
the degree of immunosuppression. Exposure is carefully ascer-
tained during the evaluation of the future recipient to ensure that
appropriate precautions are recommended and that antibiotic
prophylaxis is provided if necessary. If the patient presents with
infection after transplant, consideration of exposure is also crit-
ically important to ensure that a timely and accurate diagnosis
is made; untreated infection in immunocompromised patients
frequently progresses rapidly.
The immunosuppression requirements of the recipient change
over time—they are highest immediately after transplant, partic-
ularly when induction therapy is used. This period also coincides
with the highest incidence of posttransplant infection, which
remains high for the first 6 months after transplant. With increas-
ing time after transplant, in the absence of significant rejection,
corticosteroid therapy is tapered and often discontinued, and use
of other maintenance immunosuppressive agents is decreased.
These steps usually result in a reduced risk of infection. However,
if significant or recurrent rejection occurs and immunosuppres-
sion is increased, the infection risk increases.
Common pathogens include community-acquired pathogens
such as common respiratory viruses (eg, influenza, parainflu-
enza, respiratory syncytial virus, adenovirus), and common bac-
teria such as streptococci, Mycoplasma, Legionella, Listeria, and
Salmonella. Vaccinations for influenza virus and pneumococci
are recommended but may have reduced efficacy in this popula-
tion. Endemic organisms, such as Histoplasma and Coccidioides
in the United States may also be seen. In general, the most com-
mon infections reflect the most common organisms prevalent in
the recipient’s environment. The fundamental difference between
the transplant recipient and the general population is the rapidity
of onset of symptoms and signs, the relative severity of infection,
and the likelihood of coinfection with more than 1 pathogen.
Latent infection is of particular concern for transplant recip-
ients—both reactivation of latent infection and the possibility of
acquired latent infection from the donor. In common practice,
this concern extends mainly to reactivation of CMV, varicella-
zoster virus, and herpes simplex virus. However, reactivation of
tuberculosis, toxoplasmosis, and, in endemic areas, histoplasmo-
sis and blastomycosis may occasionally occur.
In the pretransplant assessment of potential recipients, exten-
sive serologic evaluation is performed to assess the patient’s
immune status for hepatitis viruses A, B, and C; CMV; EBV;
varicella-zoster virus; herpes simplex virus; human immunode-
ficiency virus; Toxoplasma; Treponema pallidum; and the mea-
sles virus. A tuberculin skin test is also performed. Vaccinations
are then updated as necessary with vaccines for hepatitis A,
hepatitis B, measles, and varicella-zoster virus. Tetanus toxoid,
pneumococcal, and influenza virus vaccines are also updated as
necessary.
At the time of transplant, administration of perioperative anti-
biotics is routine, but the practice may vary among institutions.
Administration of intranasal mupirocin calcium ointment pre-
operatively is also usual and is continued for several days post-
operatively to eliminate nasal carriage of methicillin-resistant
Staphylococcus aureus.
Postoperatively, prophylaxis against CMV infection is rou-
tine, usually for 3 months. As discussed in detail above, the ben-
efits of CMV prophylaxis include reduced incidence of acute and
chronic rejection and possibly of posttransplant CAV infection
and PTLD. Ganciclovir is the antiviral agent with efficacy against
CMV that has been most extensively studied, and the efficacy of
ganciclovir in preventing CMV-related infection in solid organ
transplant recipients has been demonstrated in smaller studies
and in 1 meta-analysis. The main side effect of ganciclovir is
myelosuppression. Ganciclovir has limited oral bioavailability,
but oral ganciclovir is currently licensed for long-term CMV pro-
phylaxis, with evidence of its efficacy as maintenance therapy
for solid organ transplant recipients when it has been preceded
by intravenous induction therapy. Valganciclovir is a valine ester
of ganciclovir that has enhanced oral bioavailability and proven
efficacy as CMV preemptive therapy in cardiac transplant recipi-
ents. CMV resistance to ganciclovir (and valganciclovir) has
been reported. In these cases, foscarnet is used for CMV ther-
apy, but its main dose-limiting side effect is renal impairment.
Acyclovir has some efficacy against CMV, as suggested in a
meta-analysis of 12 randomized trials. Like ganciclovir, its oral
bioavailability is limited but enhanced when delivered as its valyl
ester, valacyclovir.
The transmission of Toxoplasma gondii is a concern for car-
diac transplant recipients who are seronegative for Toxoplasma
antibody and who receive an organ from a seropositive donor.
The Toxoplasma trophozoites or cysts reside in the skeletal and
cardiac muscle of previously infected persons. Patients in the
immunosuppressed state after transplant have an increased risk
of local or disseminated toxoplasmosis. Therefore, it has been
routine to administer prophylactic therapy for toxoplasmosis to
this group for at least 3 months, with pyrimethamine, sulfadia-
zine, or folinic acid. The efficacy of co-trimoxazole (trimethop-
rim-sulfamethoxazole) against Toxoplasma has also been shown
and is adequate prophylaxis against both Pneumocystis carinii
and Toxoplasma. The practice in many institutions, including
Mayo Clinic, is to administer high-dose oral co-trimoxazole for
3 months and then to continue maintenance low-dose therapy
for life.
For cardiac transplant patients (unlike heart-lung and lung
transplant recipients) antifungal prophylaxis is not routine and is
not supported by data from clinical trials.
Malignancy
Malignancy is a common complication of long-term immunosup-
pression and an increasing contributor to death after transplant.
The incidence of all cancers is increased in solid organ trans-
plant recipients compared with the general population. The most
common form of malignancy is skin cancer, and its incidence is
dramatically increased among solid organ transplant recipients
compared with that of the general population. By 10 years after
cardiac transplant, the incidence of skin cancer is approximately
20% according to current ISHLT data. The incidence of lym-
phoproliferative malignancies, including PTLD (Figure 101.5),
is also markedly increased compared with that for the general
 101 Cardiac Transplantation
Figure 101.5. Posttransplant Lymphoproliferative Disease Involving the Small Bowel Mesentery, Successfully Resected. The patient responded
well to reduction in immunosuppression.
population (4% incidence at 10 years after transplant). The sum
total incidence of all other cancers including lung, colon, breast,
prostate, bladder cancer, and Kaposi sarcoma, is approximately
15% at 10 years.
Much of the increased incidence of cancer after transplant is
from the facilitation of chronic opportunistic infection by onco-
genic viruses such as EBV in PTLD, human herpesvirus 8 in
Kaposi sarcoma, and human papillomavirus in skin cancers. All
immunosuppressive agents have been implicated to some degree,
with the possible exception of sirolimus, for which evidence is
mounting that the risk of malignancy is significantly reduced.
Most PTLD is of B-lymphocyte origin (Figure 101.5). The
etiology of PTLD is multifactorial, but EBV infection has been
strongly associated with its development. Other risk factors
include younger recipient and donor ages (<18 years), more than
5 acute rejection episodes after transplant, and CMV infection.
The evidence for a protective effect of antiviral prophylaxis on
the development of PTLD is mixed and inconclusive.
T cells are thought to have a role in preventing malignant trans-
formation of EBV-infected B cells. Therefore, there is a strong
association between the intensity of immunosuppression and
the development of PTLD. Induction therapy, particularly with
muromonab-CD3 (now rarely used), has been associated with an
increased risk. Alemtuzumab does not appear to carry as high a
risk of future PTLD as other agents, such as polyclonal antilym-
phocytic antibodies and interleukin 2 receptor antagonists.
Over 50% of PTLD manifests extranodally, in sites such as
the gastrointestinal tract, lungs, skin, liver, and central nervous
system. PTLD has also been reported in allografts (2%–6%).
More unusual locations for PTLD have been described, such as
gingival tissue where PTLD mimicks cyclosporine-induced gin-
gival hyperplasia. The potential for PTLD to manifest in these
unusual locations underlines the importance of obtaining a tissue
sample for analysis if possible when the diagnosis is in doubt.
917
The treatment of PTLD starts with decreasing immunosup-
pressive therapy as much as possible. The next step is usually
to begin anti–B-cell monoclonal antibody therapy, most fre-
quently rituximab, an anti-CD20 monoclonal antibody that has
demonstrated efficacy in small studies. Anti–interleukin 6 anti-
body therapy has been used with success in a small number of
patients. At some institutions, monoclonal antibody therapy is
used only if decreasing immunosuppression does not help; at
others, it is used routinely, especially if PTLD occurs early after
transplant and immunotherapy cannot be decreased. If treat-
ment with rituximab fails (which is often the case if the tumor
does not express CD20), salvage chemotherapy is required; this
strategy has been described in a small study. Other therapies
with reported positive outcomes are adoptive immunotherapy,
involving the administration of banked HLA-matched or autolo-
gously cloned EBV-specific cytotoxic T cells, and intravenous
immunoglobulin.
Skin cancer occurs up to 100 times more often in the heart
transplant recipient than in the general population. It is often
recurrent and more aggressive. Cardiac transplant recipients are
probably at higher risk of skin malignancy compared with renal
transplant recipients owing to the higher threshold of immuno-
suppression required. The risk of skin malignancy may vary
with different immunosuppressive regimens; the use of siroli-
mus as an alternative agent may be protective and may even
induce remission of skin cancers in transplant recipients. The
appropriate response of transplant physicians confronted with
recurrent skin malignancy is to decrease the patient’s immu-
nosuppression if possible. The importance of patient education
and specialist advice on appropriate risk-reducing measures
cannot be underestimated. The role of retinoids in preventing
skin malignancy in this population remains under review. Initial
results have been encouraging, but the limiting factor is patient
tolerability.
918 IX Cardiomyopathy and Heart Failure

Current Demographics and Outcomes appear to have a higher risk of death after transplant. In the past,
female sex, recipient history of malignancy, donor-recipient
The ISHLT has reported annual outcome data on transplant
CMV mismatch, and elevated panel reactive antibody level were
recipients for the past 25 years. The most common indications
risk factors for mortality in the first year, but they are not now
for transplant in the period from 2005 to 2009 were nonischemic
because of improvements in practice.
cardiomyopathy (53%) and ischemic cardiomyopathy (38%).
Graft failure remains the most common cause of death dur-
Congenital heart disease, valvular heart disease, and second
ing the first 30 days, and it includes ischemic-reperfusion injury,
transplant each accounted for approximately 3% of all heart
right heart failure, and acute rejection. From 30 days through the
transplants performed worldwide.
first year, infection is the most common cause of death (account-
There has been a trend toward performing transplants in older
ing for 29% of deaths within that period). Between 1 and 3
patients who have more medical comorbidities (>10% of patients
years after transplant, the rate of death from CAV progressively
are older than 65 years). More than 20% of patients have diabetes
increases, up to 15% by 10 years. However, the contribution of
mellitus before transplant, and 42% have had previous cardiac
CAV to death is likely to be underrecognized, since there is a
surgery. For end-stage heart failure, 20% of transplant patients
steady contribution from graft failure as the cause of death over
are supported with a left ventricular assist device, and 3% are
that time, much of which is likely from undiagnosed CAV. The
supported with a right ventricular assist device. In addition,
contribution of malignancy as a cause of death is approximately
the proportion of patients with preformed HLA antibodies has
20% at 3 years and the rate remains the same thereafter.
increased, with 9% of transplant patients having a panel reactive
antibody level greater than 10%.
Despite these trends, survival after cardiac transplant has Conclusions
continued to improve. The current overall median survival for
Cardiac replacement therapy in end-stage heart failure is at a
all patients (adult and pediatric) undergoing heart transplant is
crossroads. Although posttransplant outcomes have continued to
10 years; for those who survive the first year the median sur-
improve since the first transplant in 1967, cardiac transplant is
vival is 13 years (Figure 101.6). Much of this improvement has
a therapeutic option available to only the minority of end-stage
resulted from improved survival in the first year. The most recent
heart failure patients. Advances in destination mechanical circu-
data from the United Network for Organ Sharing demonstrate a
latory support are beginning to provide a long-term solution for
1-year patient survival of 89% after cardiac transplant, with a
many patients, but a fully implantable device is not yet available.
5-year survival of 74%. At Mayo Clinic, 1-year survival is more
Research continues in the fields of xenotransplant and cell ther-
than 90%, and 5-year survival approximately 80%.
apy. In the meantime, cardiac transplant remains the gold stand-
Currently, the most important risk factors for death after car-
ard for cardiac replacement therapy to which all future cardiac
diac transplant in the first year are the requirement for short-term
replacement therapies will be compared.
extracorporeal mechanical support after transplant, and congen-
ital heart disease as an indication for transplant. Patients bridged
to transplant with mechanical circulatory support devices also Abbreviations
AHA/ACC American Heart Association and American College of
Cardiology
100 Median survival =10.0 y AMR antibody-mediated rejection
Conditional median survival =13.0 y CAV cardiac allograft vasculopathy
80 CMV cytomegalovirus
Survival, %

EBV Epstein-Barr virus

60
ISHLT International Society for Heart and Lung Transplan-
tation
40
IVUS intravascular ultrasonography
mTOR mammalian target of rapamycin
20 N=80,038
No. at risk at 23 y=124
PSI proliferation signal inhibitor
PTLD posttransplant lymphoproliferative disease
0
0 5 10 15 20 25
Years Names of Clinical Trials
Figure 101.6. Kaplan-Meier Survival for All Cardiac Transplant CARGO Cardiac Allograft Rejection Gene Expression Observa-
Patients, January 1982 to June 2008. Conditional median survival is the tional
time to 50% survival for recipients surviving the first year after trans- IMAGE Invasive Monitoring Attenuation Through Gene
plant. (Previously published. See “Credit Lines” section.) Expression
 Section X

Invasive/Interventional Cardiology
This page intentionally left blank
 102

Diagnostic Cardiac Angiographya

ANDRÉ C. LAPEYRE III, MD

Introduction outweighed by the likely benefits of accurate diagnosis and the

This chapter focuses on the basics of cardiac angiography,
especially those aspects most pertinent to cardiology board
examinations. It is important to remember that angiograms are
2-dimensional projections of 3-dimensional structures. Therefore,
structures need to be visualized with at least 2 orthogonal views
to avoid missing important information because of overlapping
structures, eccentric or ribbonlike lesions, foreshortening, curva-
ture, or fracture. Additionally, the heart, coronary arteries, and
descending aorta lie obliquely in the chest with much variabil-
ity among patients. The “best” view or angle to see any given
structure must be determined for each patient. Therefore, gen-
eralizations about the best view are always approximations and
not absolutes.
• Orthogonal views are needed to avoid missing issues because of
overlap, foreshortening, or eccentric lesions.
• The exact angles for “best” visualization must be determined for
each patient.
Coronary Angiography
Indications for Coronary Angiography
Coronary angiography is indicated for the diagnosis and treat-
ment of coronary artery disease when the procedural risks are
a
The author thanks Peter Berger, MD, and Joseph G. Murphy, MD, who
wrote previous versions of this chapter.
An atlas showing an assortment of coronary angiograms,
ventriculograms, and aortograms appears in the Appendix at the end
of this chapter. Although the images in the atlas are typical of ones that
may be shown on cardiology examinations, the atlas is not inclusive for
all types of images.
Abbreviations and acronyms are expanded at the end of this chapter.
patient is willing to consider a therapeutic procedure if a sig-
nificant problem is found. Although epicardial coronary athero-
sclerosis is the most common concern, increasingly common
indications for angiography include other diseases of the coro-
nary arteries such as the following: congenital anomalies of the
origin of coronary arteries, coronary fistulas, coronary spasm,
coronary endothelial dysfunction, coronary emboli, coronary
arteritis, small vessel disease, and myocardial bridging. Coronary
angiography may also be considered in special situations when
coronary artery disease, although unlikely, needs to be excluded
(eg, in airline pilots with mildly abnormal stress tests, in cardio-
myopathies, in probable apical ballooning syndrome, in patients
preoperatively before heart valve surgery or high-risk noncardiac
surgery).
Contraindications to Coronary Angiography
Refusal of a patient to consent to the procedure is an absolute
contraindication to coronary angiography. Relative contraindica-
tions that should delay angiography until mitigated include the
presence of correctable electrolyte abnormalities or drug toxicity
(eg, hyperkalemia, digitalis toxicity), febrile illness, acute renal
failure, decompensated heart failure, severe allergy to radio-
graphic contrast agents, an anticoagulated state or a severe bleed-
ing diathesis, severe uncontrolled hypertension, and pregnancy.
• An absolute contraindication to coronary angiography is the refusal
of a patient to consent to the procedure.
Complications of Coronary Angiography
Major Complications
The most severe complications of coronary angiography
are death, myocardial infarction, and stroke. The combined
921
922 X Invasive/Interventional Cardiology
Table 102.1. Complications of Coronary Angiography
Complication Frequency, %
Death 0.10
Stroke 0.07
Myocardial infarction 0.05
Hemodynamically significant arrhythmia 0.30
Contrast agent reaction 0.30
Vascular complication (needing transfusion or surgical 0.30
or interventional repair)
Renal cholesterol emboli 0.15
Contrast-induced nephropathy (creatinine increase 5–50
≥0.5 mg/dL)
frequency of these adverse events is generally about 0.22%, but
it depends on the case mix. Other major complications include
dysrhythmias, contrast-associated reaction and nephropathy,
vascular damage, and emboli. (Table 102.1) The main risk fac-
tors for adverse events with coronary angiography are left main
coronary artery disease, aortic stenosis, and preexisting renal
insufficiency. Other important risk factors include advanced age,
angina at rest, left ventricular dysfunction, previous stroke, and
severe noncardiac disease (cerebrovascular and peripheral vas-
cular disease, diabetes mellitus, and pulmonary insufficiency).
Cardiogenic shock increases the risk of coronary angiography by
6-fold and the risk of acute myocardial infarction by 4-fold.
• The main risk factors for major adverse coronary artery events
with coronary angiography are left main coronary artery disease,
aortic stenosis, and preexisting renal insufficiency.
Contrast-Induced Nephropathy and Renal Failure
There are many definitions for CIN in the medical literature. One
of the most common is a serum creatinine level that increases 0.5
mg/dL or more. This increase is usually transient but may be per-
manent. Depending on the study, the incidence of CIN has been
reported to range from 5% to 50% among patients undergoing
coronary angiography. The frequency of renal failure—transient
or permanent—that requires dialysis is much less.
The frequency of CIN and renal failure from nephrotoxic con-
trast agents can be decreased by delaying angiography in patients
with acute renal failure, avoiding the concomitant administration
of other nephrotoxins, providing hydration, and, most impor-
tantly, decreasing the volume of contrast agent administered.
The volume of contrast agent can be decreased by minimiz-
ing the number of angiographic views taken and using biplane
angiography in patients with preexisting renal disease, who are
at increased risk of acute renal failure. N-acetylcysteine, given
before the contrast agent is administered, can ameliorate CIN;
dopamine and fenoldopam cannot. If ventricular function is nor-
mal, the nephrotoxic effects of contrast media may be mitigated
by volume expansion with isotonic saline or isotonic bicarbonate
solution.
Minor Complications
Minor complications include local complications at the vascu-
lar access site; the type of complication depends on the vascular
approach. These complications include hemorrhage, hematoma,
distal embolization, false aneurysm, and local injury to the
artery, vein, or nerve in the site-associated neurovascular bundle.
Hemodynamically significant vascular complications occur in
approximately 1% of patients undergoing diagnostic angiogra-
phy and are least likely to occur when the radial artery is used.
However, a radial artery approach can be used only when there
is a patent palmar arch; that is, when the ulnar artery and radial
artery communicate though the palmar arch, so that the ulnar
artery can supply blood to the hand should the radial artery
occlude. Adequacy of the palmar arch can be evaluated clini-
cally with the Allen test.
• Hemodynamically significant vascular complications occur in
approximately 1% of patients undergoing diagnostic angiography.
Arrhythmias During Angiography
Tachyarrhythmia, bradyarrhythmia, or vasovagal complications
occur in approximately 1% of patients. These complications are
usually self-limited. If need be, they generally can be treated
with electrical cardioversion, defibrillation, the administra-
tion of atropine, or intravenous volume expansion. Pulmonary
edema may develop during angiography, most commonly from
increased intravascular volume due to the contrast agent, a car-
diac complication (ie, an acute myocardial infarction), or the
recumbent position.
Rare Complications Occurring in Less Than
1% of Patients
Coronary Artery Injury. Coronary artery dissection gener-
ally is preventable by paying meticulous attention to pressure
waveforms and avoiding overly vigorous injection of the contrast
agent.
Coronary artery spasm from a reaction to the catheter tip gen-
erally responds to removal of the catheter. Nitroglycerin (sublin-
gual or intracoronary) may be required.
Contrast Agent Reactions. Serious reactions to contrast
agents can mimic an anaphylactic reaction but are immunologi-
cally distinct. These anaphylactoid reactions should be treated
with immediate intravenous administration of antihistamines
and corticosteroids. Anaphylactoid reactions are more likely to
occur in patients with a history of allergy to contrast agents and
can be minimized by prophylactically administering antihista-
mines and corticosteroids and using low-ionic agents.
Protamine Reactions. Severe protamine reactions may
result in shortness of breath, hypotension, flushing, and flank
pain. Protamine is given to reverse the effect of intravenous hep-
arin, which is administered in many catheterization laboratories
(in doses ranging from 2,500 to 5,000 units) to reduce the risk of
thromboembolic complications of the procedure. Such reactions
are more likely to occur in patients who have received NPH insu-
lin, because it contains protamine. Thus, the use of protamine
should be avoided in those patients. Patients with an allergy to
fish are also at increased risk.
Pseudocomplications
Up to one-half of the complications that occur within 24 hours
after coronary angiography are believed to be “pseudocomplica-
tions.” That is, these “complications” may have occurred during
the same period had angiography not been performed.
• Half of the complications that occur within 24 hours after coro-
nary angiography are probably “pseudocomplications” (ie, the
“complications” may have occurred even if angiography had not
been performed).
 102 Diagnostic Cardiac Angiography
Coronary Artery Anatomy
Dominance
Coronary artery dominance is defined by the artery that gives
rise to the posterior descending coronary artery. In most
patients (70%–86%), the right coronary artery is dominant
(Figure 102.1). In 7% to 20% of patients, the circumflex coronary
artery is dominant, and 7% to 10% of patients have codominant
arteries, with both the right coronary artery and the circumflex
coronary artery supplying the posterior descending coronary
artery. There is no particular clinical significance to whether a
patient is right dominant, left dominant, or codominant.
Coronary Arteries
The left main coronary artery is usually 5 to 10 mm in diameter
and generally less than 4 cm long. It bifurcates into the left ante-
rior descending coronary artery and circumflex coronary artery.
It may also trifurcate into those branches and a ramus interme-
dius artery.
The left anterior descending coronary artery lies in the anterior
interventricular groove. It usually wraps around the apex of the left
ventricle; its terminal branches reach those of the right posterior
descending coronary artery. Diagonal branches supply the anterolat-
eral wall, and septal branches supply the interventricular septum.
The circumflex coronary artery lies in the left atrioventricular
groove. Its terminal branches reach those of the right posterolat-
eral coronary artery. Obtuse marginal branches supply the lateral
wall of the left ventricle. The circumflex artery may also supply
left posterolateral branches supplying the posterior wall.
The right coronary artery lies in the right atrioventricu-
lar groove. Proximally, it generally gives off the conus artery
(in 50% of patients), the sinoatrial nodal artery (in 55%), and
acute marginal branches to the right ventricle. Frequently, these
branches arise from their own coronary ostium in the right
coronary sinus. Distally, it most commonly bifurcates into the
CX
LAD
RCA
PDA
Right Dominant
Figure 102.1. Coronary Artery Dominance. Left, In right coronary dominance, the posterior descending branch (PDA) arises from the right
coronary artery (RCA). Right, In left coronary dominance, the RCA is a diminutive vessel and the posterior descending coronary artery arises as
a continuation of the left atrioventricular groove artery, a branch of the circumflex (CX) system. LAD indicates left anterior descending coronary
artery. (Previously published. See “Credit Lines” section.)
923
posterior descending coronary artery and right posterolateral
coronary artery (or arteries). Other important branches to be
familiar with include the following:
1. Conus artery (usually from the proximal right coronary artery or
from the aorta very near the right coronary ostium)
2. Sinoatrial nodal artery supplying the sinoatrial node (slightly <60%
from the right coronary artery and slightly <40% from the circum-
flex coronary artery; the others have a dual supply)
3. Atrioventricular nodal artery (usually from the dominant coronary
artery near the point at which it gives rise to the posterior descending
branch)
Coronary Artery Anomalies
Coronary artery anomalies are found on 1.0% to 1.5% of coro-
nary angiograms (Table 102.2). Of these, 90% are abnormali-
ties in the origin or distribution of a coronary artery and 10%
are abnormal fistulas. Coronary anomalies are often classified as
benign or clinically significant; most are benign.
Benign Coronary Artery Anomalies. Benign anomalies
must be recognized by the angiographer. The most common
benign coronary artery anomalies are separate ostia of the left
anterior descending and circumflex coronary arteries. These
occur in 0.4% to 1% of patients and may be associated with a
bicuspid aortic valve.
The circumflex coronary artery may arise from the right cor-
onary sinus or as an early branch of the right coronary artery.
When present, the circumflex coronary artery virtually always
travels behind the aorta to lie in the left atrioventricular groove
(Figure 102.2).
Rarely, there may be no circumflex coronary artery; in that
case, a superdominant right coronary artery supplies the entire
left atrioventricular groove and left posterolateral wall.
Clinically Significant Anomalies. The most common clini-
cally significant anomaly is a coronary artery that originates
CX
LAD
RCA
PDA
Left Dominant
924 X Invasive/Interventional Cardiology
Table 102.2. Coronary Artery Anomalies in 126,595 Angiograms
Type of Anomaly
Benign (80%)
Separate, adjacent LAD and LCX ostia
LCX
LCX origin from RSV or RCA
Anomalous origin from PSV
Anomalous origin from aorta
LMCA
RCA
Absent LCX
Small fistulae
Clinically significant (20%)
Origin of coronary artery from opposite aortic sinus
LMCA from RSV
LAD from RSV
RCA from LSV
Anomalous origin from pulmonary artery
LMCA
LAD or RCA
Single coronary artery
Multiple or large coronary fistulae
Abbreviations: LAD, left anterior descending coronary artery; LCX, left circumflex coronary
artery; LMCA, left main coronary artery; LSV, left sinus of Valsalva; PSV, posterior sinus of
Valsalva; RCA, right coronary artery; RSV, right sinus of Valsalva.
Previously published. See “Credit Lines” section.
from the contralateral aortic sinus (ie, the left main or left ante-
rior descending coronary artery from the right sinus of Valsalva
or the right coronary artery from the left sinus of Valsalva). It is
important to identify the course as it relates to the great vessels.
If the anomalous vessel courses between the aorta and the pul-
monary artery, symptoms (including sudden death) may occur. In
tetralogy of Fallot, the left anterior descending coronary artery
arises from the right coronary artery in 4% of patients.
Figure 102.2. Anomalous Origin of the Circumflex Coronary
Artery (CX) From the Right Coronary Artery (RCA). This anomaly
occurs in 0.4% of patients. The CX branch arises from or near the RCA
ostium and courses posteriorly to supply the obtuse margin of the heart.
(Previously published. See “Credit Lines” section.)
Incidence, % Anomalies, %
0.40 30.0
0.40 30.0
<0.01 0.3
0.01 1.0
0.15 10.0
0.003 0.2
0.10 10.0
0.02 1.0
0.03 2.0
0.10 10.0
<0.01 <1.0
<0.01 <1.0
0.05 3.0
0.05 3.0
A coronary artery may arise from the pulmonary artery.
The incidence of these anomalies is very low (Table 102.2).
In relation to each other, the coronary artery that most often
has an anomalous origin from the pulmonary artery is the left
main coronary artery; the next most common is the left anterior
descending coronary artery, and the least common is the right
coronary artery. Nearly 90% of patients with these anomalies
die during infancy. If the patient survives, the anomalous artery
fills retrogradely through collaterals and drains into the pul-
monary artery (left-to-right shunt). This condition may cause
angina, infarction, and heart failure. It warrants surgical repair
(ligation and grafting or reanastomosis to the aorta or subclavian
artery).
Coronary Artery Fistula. A coronary artery fistula is an
abnormal connection between a coronary artery and another
structure, most commonly a venous structure or chamber on the
right side of the heart. The right coronary artery is the site of the
fistula in 55% of patients. The majority of coronary artery fistu-
las empty into the right ventricle, right atrium, or coronary sinus.
Less commonly, fistulas empty into the pulmonary artery, left
atrium, or left ventricle. If a branch from a native coronary artery
leaves the surface of the heart or can be seen to directly con-
nect to a venous structure, a fistula is the most likely diagnosis.
Fistulas must be differentiated from the very rare occurrence of a
tumor receiving its arterial supply from a coronary artery.
Generally, the shunt caused by a fistula is small and the myo-
cardial blood flow to the terminal branches of the involved coro-
nary artery is not compromised. Patients who have a fistula are
usually asymptomatic. However, if the shunt is large, the follow-
ing may occur: pulmonary hypertension, congestive heart fail-
ure, bacterial endocarditis, rupture, or myocardial ischemia in
the terminal portion of the involved coronary artery.
• Coronary artery dominance is defined by the artery that gives rise
to the posterior descending artery.
 102 Diagnostic Cardiac Angiography 925

• Anomalous coronary arteries originating from the pulmonary

artery or which course between the aorta and pulmonary artery
are serious and clinically significant. LM
• In tetralogy of Fallot, the left anterior descending coronary artery
arises from the right coronary artery in 4% of patients. LAD
• Arterial vessels that leave the heart are usually either coronary CX
artery bypass grafts or coronary fistulas.

Angiographic Views
Angiography should visualize all segments of the coronary arter-
ies and their branches (and bypass grafts, if present) in at least 2
orthogonal views. All segments of the coronary arteries must be
seen without foreshortening and without being obscured by over- OM
lapping branches, so that stenoses can be appropriately evaluated.
To do this, multiple projections must be used, combining cranial
and caudal angulation with right and left angulation. The average
“best” projections to see various coronary artery segments are
Figure 102.3. Right Anterior Oblique Caudal Projection of the Left
listed in Table 102.3. These may need to be adjusted to account for Coronary System. The circumflex coronary artery (CX) and its branches
differences in the heart’s position in the chest among patients. are seen clearly. LAD indicates left anterior descending coronary artery;
The 4 views of the left coronary arteries and 2 views of the LM, left main coronary artery; OM, obtuse marginal branches.
right coronary artery in Figures 102.3 through 102.8 are the most
commonly encountered views and the views from which one is
expected to identify the major coronary arteries and their major For example, the lumen of an artery reduced to 20% of normal
branches. is expressed as an 80% stenosis. Because stenoses are often
eccentric, orthogonal views must be obtained. When the degree
Coronary Artery Lesions of stenosis appears to differ significantly in orthogonal views,
the most severe stenosis is commonly reported. However, some
Stenoses
cardiologists report the average stenosis in the 2 views. Although
The grading of coronary artery stenoses is usually expressed as computer-assisted methods of quantifying coronary artery
a percentage of the nearest normal segment of the same artery. stenoses have been advocated because of their greater accuracy
and reproducibility, they have not become part of routine prac-
tice because of the time and expense associated with them.
Table 102.3. Angiographic Views for Coronary Artery Coronary stenoses are most commonly atherosclerotic.
Segmentsa However, lesions may also be related to thrombus, dissection,
vasculitis, spasm, and bridging. Figure 102.9 shows coronary
Coronary Artery
Segment “Best” View Orthogonal View
artery bridging.

LMCA—ostium Straight AP to LAO 60°,

LAO 0°–5° caudal 30°
LMCA—distal RAO 0°–20° LAO 60°, LM
caudal 30°
LAD—proximal RAO 30°, LAO 60°,
LAD
caudal 20° cranial 30°
LAD—mid RAO 30°, LAO 60°,
cranial 30° cranial 20°
LAD—distal RAO 30°, LAO 60°,
CX
cranial 0°–15° cranial 0°–10° OM
Septal perforators RAO 15°–30°, Difficult, variable
cranial 30°–45°
Diagonals RAO 30°, LAO 60°,
cranial 30° cranial 20°–30°
Circumflex RAO 30°, LAO 60°,
caudal 30°–40° caudal 30°–40°
Obtuse marginals RAO 30°, Difficult, variable
caudal 30°–40°
Right LAO 45°, RAO 30°,
cranial 20°–30° cranial 20°–30°
Abbreviations: AP, anteroposterior; LAD, left anterior descending coronary
Figure 102.4. Left Anterior Oblique Caudal View of the Left
artery; LAO, left anterior oblique; LMCA, left main coronary artery; RAO,
right anterior oblique.
Coronary Arteries. This “spider view” (which somewhat resembles a
a
The 4 most common left coronary artery injections are shown in Figures 102.3 spider) shows the distal left main coronary artery (LM), the proximal
through 102.6 (RAO caudal, LAO caudal, LAO cranial, and RAO cranial). The left anterior descending coronary artery (LAD), and the circumflex cor-
2 most common right coronary artery injections are shown in Figures 102.7 and onary artery (CX) and its branches (especially the proximal portion of
102.8 (LAO and RAO). its branches). OM indicates obtuse marginal branches.
926 X Invasive/Interventional Cardiology
LM
Septal
CX
LAD
Dx
Dx
Figure 102.5. Left Anterior Oblique Cranial View of the Left
Coronary Arteries. The left anterior descending coronary artery (LAD)
and diagonal branches (Dx) are shown. CX indicates circumflex coro-
nary artery; LM, left main coronary artery.
Aneurysms
Small aneurysms of the coronary arteries are frequently seen and
are most often related to atherosclerosis. Mild dilatation of the
artery just distal to a stenosis (ie, a poststenotic dilatation) is
also common. Coronary aneurysms may also be related to vas-
culitis (Kawasaki disease or polyarteritis), trauma (including iat-
rogenic/percutaneous coronary intervention, mycotic infection,
congenital conditions, and drug abuse (usually cocaine). Very
large aneurysms may be at risk of rupture, thrombosis, or distal
embolization.
Limitations of Coronary Angiography
A coronary angiogram is only a “luminogram.” Angiography
cannot be used to assess changes in wall thickness, a cardinal
LM
Septal Dx
LAD
Figure 102.6. Right Anterior Oblique Cranial View of the Left
Coronary Arteries. The left anterior descending coronary artery (LAD)
and its branches are shown clearly. Dx indicates diagonal branches; LM,
left main coronary artery.
RCA
RPL
PDA
Figure 102.7. Left Anterior Oblique Projection of the Right
Coronary Artery (RCA). The RCA (a dominant RCA) is clearly shown
with its 2 main branches, the right posterolateral artery (RPL) and the
posterior descending coronary artery (PDA).
feature of atherosclerosis. An angiographic image alone cannot
be used to assess the cause of stenotic or aneurysmal lesions.
Angiography relies on the presence of a normal segment,
which may not exist, for comparison with the diseased coro-
nary artery segment. Without a normal segment, the severity
and extent of lesions, especially atherosclerosis, will be under-
estimated. Furthermore, measurements of luminal diameter do
not account for any intraluminal obstruction that may be present
because of thrombus or tissue protrusion due to a ruptured plaque;
these can be suggested only indirectly by the presence of certain
characteristic angiographic features such as haziness, a mobile
object in the vessel lumen, and the appearance of contrast agent
on 3 sides of an unopacified filling defect. Angioscopy, intravas-
cular ultrasonography, endothelial function/spasm testing, and
flow reserve testing can provide important information about
RCA
RPL
PDA
Figure 102.8. Right Anterior Oblique Projection of the Right
Coronary Artery (RCA). The main portion of the RCA and the posterior
descending coronary artery (PDA) are shown, but individualized views
with cranial or caudal angulation are often required to show the right
posterolateral (RPL) branch well.
 102 Diagnostic Cardiac Angiography 927

A B

Figure 102.9. Coronary Artery Bridging. A, Right anterior oblique view of the left coronary artery in diastole shows only minimal narrowing
(arrow) of the left anterior descending coronary artery. B, Systolic frame shows obliteration of the middle left anterior descending coronary artery
(arrow) because of muscular bridging. (Previously published. See “Credit Lines” section.)

coronary arteries when the limitations of coronary angiography
become relevant in the management of a condition.
Ventriculography
Left Ventriculography
Indications for Left Ventriculography
Indications for left ventriculography include the need to assess
left ventricular function or the presence and severity of mitral
regurgitation.
Technique
A catheter is advanced to the aortic root, and the aortic valve is
crossed retrogradely. The catheter is placed in a stable position in
the left ventricle, and left ventricular pressure is measured. The
catheter then is connected to a power injector, and contrast agent
is injected through the catheter into the left ventricle. Biplane
images of the ventriculogram are preferred (generally, a 30° right
anterior oblique and a 60° left anterior oblique) because they are
more comprehensive and provide a much more accurate assess-
ment of the posterior left ventricle than the right anterior oblique
view alone.
Contraindications to Left Ventriculography
Alternative ways of determining left ventricular function should
be used for patients considered high risk for ventriculography.
High-risk patients include those with 1) severe or symptomatic
aortic stenosis, 2) severe congestive heart failure, 3) known left

Figure 102.10. Analysis of Left Ventricular Wall Motion. Left, The outline of the left ventricular cavity in the right anterior oblique view shows
anterobasal, anterolateral, apical diaphragmatic, and posterobasal segments. Right, The left anterior oblique view shows lateral, posterolateral, apical
septal, and basal septal segments. (Previously published. See “Credit Lines” section.)
928 X Invasive/Interventional Cardiology

A B

Figure 102.11. Right Anterior Oblique Ventriculograms of Hypertrophic Cardiomyopathy. A, Diastole. B, Systole with midcavity obliteration (arrow).

ventricular thrombus, 4) endocarditis involving the aortic or distortion from right-sided processes such as severe pulmonary
mitral valve, or 5) renal failure. Patients with mechanical aortic hypertension.
valve prostheses should not have catheters passed retrogradely
through the prostheses. Moderate congestive heart failure and Wall Motion. Wall motion (ie, the motion of each myocar-
rest angina are significant relative contraindications. dial region) is classified as normal, mildly hypokinetic, moder-
ately hypokinetic, severely hypokinetic, akinetic, or dyskinetic
Interpretation of Left Ventriculography (Figure 102.10).
Ejection Fraction. Determination of left ventricular ejection
Mitral Regurgitation. An assessment of the degree of mitral
fraction depends on defining the endocardial contours of the ven-
regurgitation is an integral part of left ventriculography. It is
tricle at end-systole and end-diastole, accurately calibrating, and
making assumptions about the shape of the left ventricle. The left
ventricle is assumed to be an ellipsoid with minor axes that are
equal. Calculation of the ejection fraction is therefore less accu-
rate in misshapen ventricles such as those with aneurysms, large
diverticula, severe regional or asymmetric hypertrophy, or septal

Figure 102.13. Right Ventriculogram Showing Anatomy of

Tetralogy of Fallot. The pulmonary arteries (arrow) are hypoplastic.
The right ventricle is enlarged. The aorta fills with contrast material
Figure 102.12. Pseudoaneurysm of Left Ventricle. The pseudoan- through the interventricular communication. (Previously published. See
eurysm (small arrow) has a well-defined neck (large arrow). “Credit Lines” section.)
 102 Diagnostic Cardiac Angiography 929

Figure 102.14. Left Ventriculograms. A, Right anterior oblique projection shows a large anteroapical left ventricular aneurysm. A thin rim of cal-
cification (arrowhead) is present along the anterolateral wall. The posterobasal and anterobasal segments showed systolic inward motion, but the other
segments were dyskinetic. B, Anteroapical aneurysm with extensive apical mural thrombus. (Previously published. See “Credit Lines” section.)

a visual estimation that is graded as 0 (none), 1+ (trivial), 2+ Aortic Root Angiography

(mild), 3+ (moderate), or 4+ (severe).
Right Ventriculography
Right ventriculography is used to assess right ventricular function,
tricuspid regurgitation, and many congenital cardiac conditions.
Since the right ventricle does not have an easily approximated
geometric shape, ejection fraction cannot be calculated. However,
visual estimation of global function is useful. Visual estimation
of tricuspid regurgitation is based on the rapidity and degree of
right atrial opacification and on the presence or absence of con-
trast agent reflux into the superior and inferior vena cava.
Uncommon Abnormalities
Figures 102.11 through 102.16 show examples of uncommon
abnormalities demonstrated by ventriculography.
Aortic angiography is discussed in CHAPTER 46 (“THE AORTA”).
Aortic root angiography, however, is discussed in the present chap-
ter since it is often performed as part of cardiac angiography.
Aortic root angiography is indicated for the following:
1. Assessment of aortic valve regurgitation
2. Assessment of aortic root and sinus of Valsalva dilatation
3. Localization of anomalous coronary artery ostia or coronary bypass
graft ostia that could not be easily engaged during selective coronary
or graft angiography
Although aortic dissection can be visualized with aortic root
angiography, computed tomography or magnetic resonance
imaging is usually preferred.
Ideally, root aortography uses a biplane coronary system so that
both projections are obtained with only one 40- to 60-mL injec-
tion of contrast agent. Alternatively, the procedure is performed

A B C
Figure 102.15. Gooseneck Appearances. A, Left ventriculogram shows the typical appearance of complete atrioventricular canal. B, Canada
goose. C, Pathologic specimen shows left ventricle. (Previously published. See “Credit Lines” section.)
930 X Invasive/Interventional Cardiology
Figure 102.16. Right Ventriculograms From a 21-Year-Old Woman With Ebstein Anomaly. A, Anteroposterior view shows large sail-like anterior
leaflet (arrow) displaced well to left of spine. Severe tricuspid regurgitation is evident. B, Lateral projection shows pronounced anterior displacement
of abnormal anterior tricuspid leaflet (arrow). (Previously published. See “Credit Lines” section.)
in a cardiac catheterization room equipped with specialized vas-
cular imaging and software packages that allow digital subtrac-
tion. Generally, the steep left anterior oblique view maximally
opens the aortic curvature. The use of just enough of a right ante-
rior oblique view to keep the ascending aorta from overlapping
the descending thoracic aorta gives a near-orthogonal view of
these structures. The largest field of view possible should be used
to include as much of the vasculature as possible. The field of
view should be centered to answer the most important clinical
question (eg, if the question relates to lesions of the arch and ostia
of the great vessels, include the clavicles; if the question relates to
the severity of aortic regurgitation, include the left ventricle).
Structure Sizing
The apparent size of a visualized structure during cardiac angiog-
raphy depends on several variables, including the following:
1. Size of the field of view
2. Distance of the structure from the x-ray tube
3. Distance of the structure from the imaging surface (flat panel, image
intensifier, etc)
4. Distortion due to overlap or foreshortening
True sizing of structures requires a known reference in approx-
imately the same imaging plane. For left ventriculography, true
sizing is often accomplished after the procedure by taking an
image of a metal ball of known size placed in the plane of the left
ventricle used for measuring during ventriculography. However,
vessel sizing during interventional procedures is a critical issue.
For noncoronary angiography, metal rulers or graduated metal
beads are often used. These are inadequate for coronary artery
sizing because of the highly variable angles used for imaging
and the motion of the coronary arteries. Since the most distal
portion of the angiographic catheter is in the plane of the vessel
being imaged and moves with the vessel, it can serve as a refer-
ence item. The size of the contrast-filled (visualized) lumen var-
ies among catheter types (diagnostic, guide, and high-flow) and
manufacturers. Angiographers must be familiar with the equip-
ment used in their laboratories (Table 102.4).
Radiation Safety
Cardiac catheterization and angiography cannot be performed
without the use of x-rays. Radiation and x-rays are invisible,
Table 102.4. Conversion Table for French Size
(Circumference) and Approximate External Diameter
of Catheter
French Size Catheter External Diameter, mma
3 1.0
4 1.3
5 1.7
6 2.0
7 2.3
8 2.7
9 3.0
10 3.3
11 3.7
12 4.0
a
Sizes are approximate and are specific for manufacturers and
catheters. The specific information for any catheter should be
checked before use.
 102 Diagnostic Cardiac Angiography
odorless, and noiseless, and they cannot be felt. The currently
accepted concepts of radiation exposure are as follows:
1. No level of radiation exposure is known to be absolutely safe
2. The effects of radiation exposure are cumulative over a person’s life-
time (for both the patient and the staff)
Everyone in the laboratory, especially the supervising physi-
cian or cardiologist, has a responsibility to ensure the lowest pos-
sible exposure to the patient and to the staff. The most important
component of this responsibility is to avoid doing any procedure
that is not clearly indicated. The procedure should be performed
only when it is indicated and the benefit to the patient outweighs
the risk to the patient and staff. Other components of radiation
safety include the following:
1. Limit the radiation
a. X-ray equipment should be checked regularly and calibrated to
the lowest doses for both fluoroscopy and “cine” modes that are
compatible with diagnostic imaging
b. Fluoroscopy uses less radiation than cine mode and so should be
used whenever possible
c. Use as little angulation as necessary to get diagnostic images
2. Limit the exposure
a. Use the shortest possible burst of x-rays at the lowest possible
exposure mode to get the required information
b. All personnel must wear full shielding equipment, which should
be checked regularly for integrity
931
c. All protective devices, such as movable side shields and cur-
tain shields, should be used to their maximum effect whenever
possible
d. All personnel should be as far away and behind as much shield-
ing as is compatible with performing the procedure safely for the
patient (the dose decreases as the square of the distance from the
radiation source)
e. All personnel should have their exposure monitored and reported
to them regularly
f. New equipment allowing measurement of the patient’s radia-
tion should be used and that information incorporated into the
patient’s medical record for future reference
g. To prevent accidental activation of the equipment and unneces-
sary radiation exposure, confirm that the equipment provides
visual and audible signals (and that these signals are working)
when x-rays are being generated
3. Promote education and teamwork
a. All staff should be educated and receive continuing education
about radiation hazards and radiation safety techniques
b. A spirit of teamwork and shared responsibility should be fos-
tered so that all persons involved feel that they can immedi-
ately speak up if they are aware of even a potentially avoidable
hazard
Abbreviation
CIN contrast-induced nephropathy
932 X Invasive/Interventional Cardiology

Appendix

Aorta

RV
LV
LV

Figure 102.A1. Right Anterior Oblique View of Left Ventriculogram.

The catheter passed from the right side of the heart through an atrial
septal defect and the mitral valve into the left ventricle (LV). This image Figure 102.A3. Left Anterior Oblique View of Left Ventriculogram.
shows a primum atrial septal defect with a typical gooseneck deformity This image shows a large muscular ventricular septal defect (arrow). LV
of the LV outflow tract (arrow). indicates left ventricle; RV, right ventricle.

Aorta

RV

LV LV

Figure 102.A2. Left Anterior Oblique View of Left Ventriculogram. Figure 102.A4. Left Anterior Oblique View of Left Ventriculogram.
This image shows a membranous ventricular septal defect (arrow). LV This image shows a subvalvular aortic stenosis (arrow). LV indicates
indicates left ventricle; RV, right ventricle. left ventricle.
 102 Diagnostic Cardiac Angiography 933

Aorta

PA

LAD

Figure 102.A5. Aortogram. This aortogram shows a patent ductus

arteriosus (arrow) with a left-to-right shunt. PA indicates pulmonary
artery.

RCA

Figure 102.A7. Abnormal Takeoff of the Left Anterior Descending

Coronary Artery (LAD) From the Right Coronary Sinus in a Patient
With Tetralogy of Fallot. RCA indicates right coronary artery.

Figure 102.A6. Left Anterior Oblique View of Left Ventriculogram.

This image shows a supravalvular aortic stenosis (arrow) associated
with Williams syndrome. Other features (not shown) are pulmonary
valve stenosis and infundibular stenosis.
934 X Invasive/Interventional Cardiology

PA
PA

RV

RV

Figure 102.A8. Pulmonary Angiogram. These images show severe pulmonary valve stenosis (arrows). Note the unilateral pulmonary artery (PA)
dilatation. RV indicates right ventricle.

A B

Figure 102.A9. Aortic Injection. These images show mild dilatation of the ascending aorta and moderate aortic coarctation (arrows). A, Left
anterior oblique view. B, Anteroposterior view.

Figure 102.A10. Right Anterior Oblique View of Left Ventriculogram. These images show hypertrophic cardiomyopathy with midcavitary
obstruction (arrowhead) and an apical secondary cavity (arrow).
 102 Diagnostic Cardiac Angiography 935

A B

Figure 102.A11. Right Anterior Oblique View of Left Ventriculogram. These images show moderate left ventricular dilatation and dysfunction
and severe hypokinesis of the diaphragmatic, posterobasal, and posterolateral segments (arrow). A, Diastole. B, Systole.

A B

Figure 102.A12. Left Anterior Oblique View of Left Ventriculogram. These magnified images show normal aortic valve leaflets (arrowheads).
A, Diastole. B, Systole.

A B

Figure 102.A13. Dextrocardia (Situs Solitus). A, Catheter progresses from the inferior vena cava to the right atrium, coronary sinus, and left
superior vena cava. B, Catheter progresses from the inferior vena cava to the right atrium and right superior vena cava.
936 X Invasive/Interventional Cardiology

RV
RA
RV
RA

A B

Figure 102.A14. Right Ventriculogram With Balloon-Tipped Catheter. A, Right anterior oblique view. B, Left anterior oblique view, which
superimposes right atrium (RA) and right ventricle (RV). These images show moderate RV dilatation and hypokinesis, severe tricuspid regurgitation,
and moderate RA enlargement. Sternal wires are present.

A B

Figure 102.A15. Right Anterior Oblique View of Left Ventriculogram. A, Diastole. B, Systole. The apical cavitary obliteration (arrow) is from
apical left ventricular hypertrophy or apical variant hypertrophic cardiomyopathy.

A B

Figure 102.A16. Right Anterior Oblique View of Left Ventriculogram. A, Mitral valve prolapse (arrow) without regurgitation is shown. B, Apical
filling defect suggests a mural thrombus (arrowhead).
 102 Diagnostic Cardiac Angiography 937

Ao      Ao     

LV     
LV     

A B

Figure 102.A17. Aortic Injection. A, Right anterior oblique view. B, Left anterior oblique view. The images show moderately severe dilatation
of the aortic root (Ao) and probable bicuspid aortic valve with severe aortic regurgitation. Note the indirect filling of the saphenous vein graft to the
right coronary artery. Sternal wires are present. LV indicates left ventricle.

A B

Figure 102.A18. Left Anterior Oblique View of Left Ventricle. A, Diastole. Note the prominent washout of contrast material from the brisk
mitral inflow (arrowhead); this probably indicates high left atrial pressure (ie, restrictive filling pattern). B, Systole with hyperdynamic function. Note
hypertrophic cardiomyopathy with midcavitary obstruction and apical cavity obliteration (arrow).

A B

Figure 102.A19. Right Coronary Artery Injection. A large fistula is shown from the right coronary artery to the right atrium. Jet flow is apparent
(arrowheads). A, Left anterior oblique view. B, Anteroposterior view.
938 X Invasive/Interventional Cardiology

RCA RCA

PL
PL

PDA
PDA
A B

Figure 102.A20. Constrictive Pericarditis. Left anterior oblique view shows fixation of mid-right (RCA), distal posterolateral (PL), and posterior
descending (PDA) coronary arteries. Note moderate ectasia or aneurysm (arrows) of mid-RCA. A, Systole. B, Diastole.

CB

RCA

Figure 102.A21. Anteroposterior View of Saphenous Vein Graft to Figure 102.A22. Left Anterior Oblique View of Right Coronary
First Diagonal Artery. A large myocardial “blush” (arrow) is present after Artery (RCA). This image shows a balanced dominant RCA with severe
percutaneous transluminal coronary angioplasty of the graft. This prob- stenoses of the ostium and conus branch (CB) (arrow). Contrast agent
ably occurred because of microembolization of the microcirculation. has filled the right sinus of Valsalva.
 102 Diagnostic Cardiac Angiography 939

A B

Figure 102.A23. Left Anterior Oblique Cranial Image of Occluded Distal Right Coronary Artery. A, Hazy tapered occlusion suggests thrombus.
B, During percutaneous transluminal coronary angioplasty (PTCA), there was moderate distal embolization and slow-reflow phenomenon. Note
placement of temporary pacemaker in right ventricular apex for post-PTCA bradycardia and hypotension (Bezold-Jarisch reflex).

RCA
RCA
PL
PL
AVN

A PDA B
PDA

Figure 102.A24. Normal Vessels. Normal right coronary artery (RCA) (right dominant), right posterolateral coronary artery (PL), posterior
descending coronary artery branch (PDA), and atrioventricular nodal artery (AVN) (determining dominance). A, Right anterior oblique view. B, Left
anterior oblique view.

A B

Figure 102.A25. Right Coronary Artery. Images show diffuse severe spasm (arrow) except in stented segment (midsection of the artery). A, Left
anterior oblique view. B, Right anterior oblique view.
940 X Invasive/Interventional Cardiology

A B

Figure 102.A26. Normal Nondominant Right Coronary Artery. Note the diminutive size, predisposing to catheter damping. The artery does not
supply the diaphragmatic myocardium. Also note the “shepherd’s crook” bend in the proximal right coronary artery. A, Left anterior oblique view.
B, Right anterior oblique view.

A B

Figure 102.A27. Left Anterior Oblique View of Right Coronary Artery. A, A large intracoronary “ball” thrombus (arrow) is adherent to the
guidewire after percutaneous transluminal coronary angioplasty (PTCA) for acute myocardial infarction. The thrombus was entwined in a second
wire and removed successfully without complication. B, Note the large post-PTCA dissection (arrowhead).

LAD sp sp
RCA LAD
sp RCA
PL

PDA PDA
A B

Figure 102.A28. Right Coronary Artery (RCA) Injection. Note the mild stenoses of the RCA. The left anterior descending coronary artery
(LAD) fills retrogradely through collateral vessels to the distal LAD and septal perforators (sp). The LAD is occluded proximally, and moderate
disease is scattered throughout it. A, Left anterior oblique view. B, Right anterior oblique view. PDA indicates posterior descending coronary artery;
PL, posterolateral coronary artery.
 102 Diagnostic Cardiac Angiography 941

RCA

PL

A PDA B

Figure 102.A29. Right Coronary Artery (RCA). Note the severe stenoses of the proximal RCA, mid-RCA, and posterolateral branch (PL) and
the atherosclerotic ulcer of the proximal RCA (arrow). There is either mild vessel ectasia or poststenotic dilatation. A, Left anterior oblique view. B,
Right anterior oblique view. PDA indicates posterior descending coronary artery.

A B

Figure 102.A30. Tilting Disk Mechanical Prosthetic Valve. A, Open. B, Closed. Pigtail catheter is in the ascending aorta. Arrow points to leaflet.

A B

Figure 102.A31. Normally Functioning Bileaflet Mechanical Prosthetic Aortic Valve. A, Systole. B, Diastole. Arrow points to leaflet.
942 X Invasive/Interventional Cardiology

LV

Figure 102.A32. Left Anterior Oblique View of Left Ventriculogram. Image shows a myocardial infarction–associated ventriculoseptal defect.
The intraventricular septum is seen as a filling defect. A small inferior wall rupture is apparent as a small jet immediately below the catheter (arrow).
LV indicates left ventricle.

A B

Figure 102.A33. Right Anterior Oblique View of Left Ventriculogram. The images show apical mural calcification (arrow) compatible with
an old mural thrombus, a calcified myocardial infarction scar, or a calcified ventricular aneurysm. A, Before injection of contrast agent. B, During
ventriculogram.

Ao
Ao

LA LA
LV
LV

A B

Figure 102.A34. Right Anterior Oblique View of Left Ventriculogram. The images show moderate left ventricular (LV) dilatation with normal
function and severe mitral regurgitation. A, Diastole. B, Systole. Ao indicates aorta; LA, left atrium.
 102 Diagnostic Cardiac Angiography 943

Ao
Ao
LA LA

LV

LV

A B

Figure 102.A35. Left Anterior Oblique View of Left Ventriculogram. The images show moderate left ventricular (LV) dilatation with normal
function and severe mitral regurgitation. A, Diastole. B, Systole. Ao indicates aorta; LA, left atrium.

A B

Figure 102.A36. Left Anterior Oblique View of Left Ventriculogram. The images show akinesis of the anteroapical wall segment (arrows). A,
Diastole. B, Systole.

A B

Figure 102.A37. Right Anterior Oblique View of Left Ventriculogram. The left ventricle has a normal size but moderately reduced function. Note
akinesis of the anterolateral and apical wall segments (arrows). A, Diastole. B, Systole.
944 X Invasive/Interventional Cardiology

A B

Figure 102.A38. Left Anterior Oblique View of Left Ventriculogram. Note the normal size of the left ventricle and akinesis of the posterobasal
and posterolateral wall segments. A temporary pacemaker has been placed in the right ventricular apex via the inferior vena cava. A, Diastole.
B, Systole.

Figure 102.A39. Saphenous Vein Graft. Image shows large aneurysm (arrow).

LM
LAD
LCX
LAD LCX RI

LPDA

LPDA
A B

Figure 102.A40. Normal Vessels. Images show normal left-dominant left main (LM), left anterior descending (LAD), left circumflex (LCX),
and ramus intermedius (RI) coronary arteries. The left posterior descending coronary artery (LPDA) wraps around from the LCX to the inferoseptal
wall, meeting the wraparound LAD. A, Left anterior oblique cranial view. B, Right anterior oblique caudal view.
 102 Diagnostic Cardiac Angiography 945

LCX
LM
LCX
LM

LAD
LAD

A B

Figure 102.A41. Left Main Coronary Artery (LM) Injection. Images show normal left anterior descending (LAD) and nondominant left circum-
flex (LCX) coronary arteries and age-related tortuosity. A, Left anterior oblique cranial view. B, Right anterior oblique cranial view.

LAD

LCX

Figure 102.A42. Right Anterior Oblique Caudal View. Image Figure 102.A43. Left Anterior Oblique View of Saphenous Vein
shows moderate ectasia or aneurysmal disease of the left circumflex Graft to Distal Right Coronary Artery. An extensive intraluminal
(LCX) and left anterior descending (LAD) coronary arteries. thrombus appears as multiple hazy filling defects (arrowhead).
946 X Invasive/Interventional Cardiology

SVG

PL
PL RCA

A B

Figure 102.A44. Saphenous Vein Graft (SVG) to Distal Right Coronary Artery (RCA). Note moderate stenoses in the graft body (arrows). PL
indicates posterolateral coronary artery.

LAD
LAD

A B

Figure 102.A45. Mid–Left Anterior Descending Coronary Artery (LAD). Images show moderate bridging (arrowheads).

RI
RI

AVG

OM OM

A B

Figure 102.A46. Normal Sequential Saphenous Vein Graft to Ramus Intermedius (RI) and Obtuse Marginal (OM) Coronary Arteries.
Note the total occlusion of the distal left circumflex coronary artery (no retrograde flow beyond the origin of the OM and atrioventricular groove
branch [AVG]).
 102 Diagnostic Cardiac Angiography 947

LM
LCX LM

LAD

LAD

A B

Figure 102.A47. Proximal Left Anterior Descending Coronary Artery (LAD). Images show severe stenosis (arrows). A, Left anterior oblique
cranial view. B, Right anterior oblique cranial view. LCX indicates left circumflex coronary artery; LM, left main coronary artery.

LM
LAD
LAD

LCX LM
LCX

A B

Figure 102.A48. Left Anterior Descending Coronary Artery (LAD). Images show severe stenosis (arrows) of the origin of the LAD. Note the
moderate stenoses of the left circumflex coronary artery (LCX). A, Right anterior oblique caudal view. B, Left anterior oblique caudal view. LM
indicates left main coronary artery.
948 X Invasive/Interventional Cardiology

Figure 102.A49. Right Anterior Oblique Cranial View. Image

shows heavy calcification (arrow) of the proximal left anterior descend-
ing coronary artery.

Ao
Figure 102.A51. Anteroposterior View of Aortic Injection. Image
shows large perforating atherosclerotic ulcer (arrow).

LV

Figure 102.A50. Right Anterior Oblique View of Aortic Injection.

Image shows severe aortic regurgitation. Note the equal density of the aorta
(Ao) and the left ventricle (LV). Note the relatively normal size of the LV.
 102 Diagnostic Cardiac Angiography 949

Figure 102.A52. Upper, Left anterior oblique (LAO) view of dominant right coronary artery in diastole (left) and systole (right). Lower, LAO
cranial view of left main coronary artery injection in diastole (left) and systole (right). Fixation of the right coronary artery and branches and the distal
left circumflex coronary artery is from constrictive pericarditis. Neovascularization of the pericarditis is apparent (arrows).

A B

Figure 102.A53. Anomalous Origin of the Left Circumflex Coronary Artery (LCX) From the Right Coronary Artery (RCA). A, Right anterior
oblique view of the LCX (arrow) and the RCA. B, Left anterior oblique view of the nondominant left anterior descending coronary artery.
950 X Invasive/Interventional Cardiology

A B

Figure 102.A54. Calcific Pericarditis. A, The dense rim of calcium is clearly seen outlining the border of the heart (arrows). B, Injection of the
right coronary artery (RCA) shows that the calcium is external to the heart and encases the heart and RCA.

1
1
2

Figure 102.A56. Dissection of Right Coronary Artery and Aorta.

Figure 102.A55. Pulmonary Hemorrhage. Selective pulmonary
artery injections demonstrate a large intraparenchymal pulmonary hem-
orrhage (arrows) that was associated with severe hemoptysis.
Image shows a catheter tip dissection with intramural injection of con-
trast agent. Arrows labeled 1 indicate the retrograde intramural dissec-
tion of contrast agent into the aortic wall of the right sinus of Valsalva.
Arrow 2 indicates the hard leading edge of the intramural injection in
the proximal right coronary artery; arrow 3, intraluminal flap of the dis-
section progressing down the right coronary artery.
 102 Diagnostic Cardiac Angiography 951

A B

Figure 102.A57. Apical Ballooning Syndrome. Right anterior oblique projection shows severe apical ballooning syndrome with associated severe
mitral regurgitation. Note the dyskinetic outward movement of the left ventricular apex in systole with the relatively preserved contraction only at
the base of the heart. A, Diastole. B, Systole.

Figure 102.A58. Aortic Dissection. Aortogram shows diffuse atheromatous disease, mild ascending dilatation, and a small spontaneous dissec-
tion (arrows) of the ascending aorta.
952 X Invasive/Interventional Cardiology

1 1
1 2

1
1
3

4
2

A B

Figure 102.A59. Anomalous Left Coronary Artery. Left anterior oblique projection of 2 types of origin of the left coronary artery (LCA) from
the right sinus of Valsalva next to the ostium of the right coronary artery (RCA). A, The LCA (arrows labeled 1) takes an anterior course across the
right ventricular outflow track. Arrow 2 indicates circumflex coronary artery; arrow 3, left anterior descending coronary artery (LAD); arrow 4,
RCA. B, The LCA (arrow 1) courses between the aorta and main pulmonary artery. Arrow 2 indicates LAD; arrow 3, circumflex coronary artery.

1 2
3
2

A B

Figure 102.A60. Coronary Artery Fistula. These images show a large fistula from the right coronary artery (RCA) to the pulmonary artery.
Arrow 1 indicates RCA; arrow 2, fistula; arrow 3, contrast agent entering the pulmonary artery. A, Left anterior oblique projection. B, Right anterior
oblique projection.
 103
Coronary Artery Physiology and
Intracoronary Imaginga
ABHIRAM PRASAD, MD
Normal Coronary Physiology
Myocardial Oxygen Consumption
The principal function of the coronary arteries is to. provide oxy-
gen and nutrients to the myocardium. Myocardial VO2 is equal to
the product of coronary blood flow and the arteriovenous oxygen
gradient across the coronary vascular bed—that is, arterial oxy-
gen content minus coronary sinus oxygen content. In the rest-
ing state, myocardial oxygen extraction is near the maximum
and coronary sinus oxygen saturation is typically 30% or less
(or PO2 <20 mm Hg). Because myocardial . oxygen extraction is
already near the maximum, myocardial VO2 can increase only by
an
. increase in coronary blood flow, and changes in myocardial
VO2 closely parallel changes in. coronary blood flow. Important
determinants of myocardial VO2 are heart rate, inotropic state .
(contractility), and intramyocardial wall stress. Myocardial VO2
can be approximated clinically by the product of systolic blood
pressure and heart rate (called the rate-pressure product). The
rate-pressure product is frequently used during exercise testing
as an estimate of cardiac workload.
.
• Myocardial Vo2 can be approximated clinically by the product of
systolic blood pressure and heart rate.
• In the resting state, myocardial oxygen extraction is near the
maximum.
.
• Important determinants of myocardial Vo2 are heart rate, inotropic
state (contractility), and intramyocardial wall stress.
a
Portions previously published in Higano ST, Lerman A, Garratt KN,
Nishimura RA, Holmes DR Jr. Assessing coronary flow physiology
with intracoronary Doppler following coronary interventions. J Interv
Cardiol. 1996;9(2):163–73. Used with permission.
Abbreviations and acronyms are expanded at the end of this chapter.
Coronary Blood Flow Regulation
During rest, normal coronary blood flow is approximately 60 to
90 mL/min per 100 g of myocardium. Its regulation is complex
and involves metabolic, autonomic, and mechanical factors. The
most important metabolic factors include adenosine, prostaglan-
dins, and endothelial-derived factors (eg, the vasodilator nitric
oxide and the vasoconstrictor endothelin). Myocardial oxygen
and carbon dioxide tensions and ATP-sensitive potassium chan-
nels also are important in regulating coronary blood flow.
Of the agents released from myocardial cells, adenosine is
probably the most important. Adenosine is produced from the
breakdown of high-energy phosphates (ATP), which cannot be
regenerated during ischemia owing to the low oxygen tension.
The breakdown product, adenosine monophosphate, accumu-
lates and is converted to adenosine.
The contribution of the autonomic nervous system to the
control of coronary blood flow is probably modest. Changes
in coronary blood flow with either sympathetic or parasympa-
thetic stimulation result predominantly from the accompanying
changes in loading conditions and contractility.
Mechanical factors have a major effect on coronary blood
flow. During myocardial contraction, intramyocardial pressure
increases, causing compression of small vessels and a reduction,
or throttling, of coronary blood flow. The result is a predomi-
nantly diastolic blood flow pattern (Figure 103.1). Approximately
60% of coronary blood flow occurs during diastole in the left
coronary artery. In contrast, the proximal right coronary artery
undergoes much less vessel compression during low-pressure
right ventricular contraction. As a result, there is a much smaller
reduction in blood flow during systole, and blood flow in the
proximal right coronary artery during systole is nearly equal to
that during diastole. However, in the distal right coronary artery
(beyond the right ventricular marginal branches), coronary blood
953
954 X Invasive/Interventional Cardiology

A B

S D S S D S

Figure 103.1. Intracoronary Doppler Velocities From the Left Anterior Descending Coronary Artery Showing Predominantly Diastolic Flow. A,
Flow during basal conditions. B, Flow after microvessel vasodilation with intracoronary adenosine. Coronary flow reserve (CFR) is 2.6. CFR is the
ratio of maximal diastolic flow to basal diastolic flow in the coronary vessel. Heart rate and aortic pressure are shown. D indicates onset of diastole;
S, onset of systole. (Previously published. See “Credit Lines” section.)

flow predominantly perfuses the inferior left ventricle, and dia-
stolic flow predominates.
The myocardial compressive effects are greater in the sub-
endocardial layer than in the subepicardial layer; thus, the sub-
endocardium is at increased risk of ischemia. During maximal
vasodilatation, myocardial perfusion is regulated primarily by
coronary perfusion pressure and myocardial compressive effects.
When coronary blood flow is reduced, as from an epicardial
coronary artery stenosis, the subendocardial layer is the first
region of the myocardium to become ischemic. Subendocardial
ischemia can be detected clinically with ST-segment depression
on an electrocardiogram. Although flow may be adequate at rest,
subendocardial ischemia may occur with exercise or stress. This
effect can be particularly pronounced in hypertrophied left ven-
tricles, even with normal coronary arteries.
Diastolic Pressure-Time Index
Coronary blood flow is closely correlated with the diastolic pres-
sure-time index, which is the product of the average difference
between aortic and left ventricular cavity pressure and the dura-
tion of diastole (ie, it is the area between diastolic aortic pressure
and left ventricular pressure curves). The diastolic pressure-time
index can be altered by changes in aortic diastolic pressure, left
ventricular diastolic pressure, and duration of diastole. Coronary
blood flow is decreased by systemic hypotension (by decreasing
aortic diastolic pressure), increased left ventricular end-diastolic
pressure (by increasing left ventricular diastolic pressure), and
tachycardia (by shortening diastole). Coronary blood flow can
be augmented by increased systemic pressure, decreased left
ventricular end-diastolic pressure, and slowing of the heart rate.
Intra-aortic balloon pumping can augment coronary blood flow
by increasing aortic diastolic pressure.
Myocardial Sinusoids
The coronary circulation drains primarily through the coronary
sinus and cardiac veins (Figure 103.2). A small portion of the
venous return drains into the thebesian veins and myocardial
sinusoids, which empty directly into the chambers of the left side
of the heart. A small right-to-left shunt occurs at this level, even
in normal hearts.
• Approximately 60% of coronary blood flow occurs during diastole
in the left coronary artery.
• Blood flow in the proximal right coronary artery during systole is
nearly equal to that during diastole.
• Coronary blood flow is decreased by systemic hypotension,
increased left ventricular end-diastolic pressure, and tachycardia.
Autoregulation of Coronary Blood Flow
During resting conditions, coronary blood flow is maintained
at a fairly constant level over a range of aortic pressures by
the process of autoregulation (Figure 103.3). As aortic pres-
sure decreases, coronary blood flow is maintained by dilation
of the resistance vessels. (The resistance vessels, or arterioles,
are small vessels proximal to the capillaries and are below the
resolution of coronary angiography). Conversely, as aortic pres-
sure increases, the resistance vessels constrict. Therefore, dur-
ing normal resting conditions, coronary blood flow is pressure
independent.
When pressure is extremely low or high, however, autoregu-
lation is overcome and coronary blood flow becomes pressure
dependent. At very high perfusion pressures, vasoconstriction is
maximal and an additional increase in pressure results in a linear
increase in blood flow. At low perfusion pressures, the resistance
vessels are dilated maximally and any additional decrease in
 103 Coronary Artery Physiology and Intracoronary Imaging 955

Extracardiac
branches Arterioluminal vessels Heart chambers

Veins Coronary Right

sinus atrium

Aorta Coronary Arterioles Capillaries

arteries

Thebesian Heart
veins chambers

Arteriosinusoidal Myocardial Heart chambers

vessels sinusoids
Figure 103.2. Diagram of the Coronary Circulation. (Previously published. See “Credit Lines” section.)

pressure results in a linear decrease in blood flow. At pressures
less than 70 mm Hg, the pressure-flow relationship becomes
linear, with blood flow decreasing in direct proportion to the
decrease in perfusion pressure. At extremely low perfusion pres-
sures (approximately 20 mm Hg), blood flow ceases altogether.
This effect is called the vascular waterfall phenomenon, which
is caused by the compressive effects of extravascular intramyo-
cardial pressure. The pressure at which flow ceases is called the
critical closure pressure, or the critical flow pressure.
• During normal resting conditions, coronary blood flow is pressure
independent.
• At pressures less than 70 mm Hg, blood flow decreases in direct
proportion to the decrease in perfusion pressure.
Coronary Flow Reserve
With physical or mental stress, the metabolic demands of the
myocardium increase and coronary blood flow and myocardial
.
VO2 increase. Coronary blood flow increases through dilation of
resistance vessels. When the resistance vessels are dilated maxi-
mally, coronary blood flow cannot be increased further without
an increase in aortic pressure. The vessels proximal to the resis-
tance vessels (ie, the epicardial and prearteriolar vessels) offer
only minimal resistance to coronary blood flow. The ratio of
Coronary Blood Flow, mL/min
maximal blood flow to resting (or basal) blood flow is termed the
CFR (Figure 103.4 and Table 103.1):
Maximal Coronary Blood Flow
CFR =
Resting Coronary Blood Flow
CFR, also called absolute flow reserve, is a measure of the abil-
ity to augment blood flow with stress. It can be measured easily
with intracoronary Doppler techniques. Maximal vasodilatation
is produced with a vasodilator such as adenosine. Adenosine has
been the easiest vasodilator to use because of its short half-life,
ability to promote maximal vasodilatation, and safety profile.
Bradycardia and complete heart block can occur, particularly
with injections into the right coronary artery, but they are rare at
the recommended doses (intracoronary bolus of 24–60 mcg for
the left coronary arteries and 12–36 mcg for the right coronary
arteries). Papaverine has a longer half-life and can prolong the
QT interval, rarely resulting in life-threatening arrhythmias.
• Coronary blood flow increases through dilation of resistance vessels.
• Maximal vasodilatation is produced with a vasodilator such as
adenosine.
Endothelial Function
The endothelium comprises the single layer of cells between
the vascular smooth muscle and the blood and circulating com-
ponents. It is the largest organ in the body, with approximately

200 1 trillion cells, a total surface area equivalent to 6 tennis courts,

and a total weight greater than that of the liver. Although the
160 endothelium functions as a semipermeable membrane, its role
in coronary artery physiology is more complex (Box 103.1). A
120 normally functioning endothelium is essential for maintaining
normal coronary blood flow. The increase in coronary blood
80 flow with both physical and mental stress is modulated largely by
endothelial-dependent changes in vasomotor tone. Endothelial
40 dysfunction is believed to be one of the earliest stages in the
pathogenesis of atherosclerosis.
0 The endothelium continuously produces substances to mod-
0 40 80 120 160 200 240 ulate vascular tone, including nitric oxide, prostacyclin, and
endothelial-derived contracting factors (eg, endothelin). The
Perfusion Pressure, mm Hg
relaxing factor produced by the endothelium and originally
Figure 103.3. Autoregulation Demonstrated by Changes in called endothelium-derived relaxing factor was identified sub-
Perfusion Pressure. (Previously published. See “Credit Lines” section.) sequently as nitric oxide. The endothelium produces nitric oxide
956 X Invasive/Interventional Cardiology

vasculitis or emboli) are caused by altered coronary artery phys-

iology and the resulting myocardial ischemia. Coronary artery
5 Maximal stenoses deform the column of blood flowing within the coronary
vasodilation artery. Although minor luminal irregularities have little effect
4 on blood flow, more severe stenoses may cause significant trans-
Coronary Flow

stenotic pressure gradients (Figure 103.5). The result is a reduc-

3
2 blood flow is maintained by vasodilation of the resistance vessels
1 coronary blood flow, the maximal coronary blood flow will be
0 the stenosis increases the pressure gradient.
Coronary Pressure
Figure 103.4. Coronary Flow Reserve. Coronary flow reserve is
the ratio of maximal blood flow to resting (or basal) blood flow. At rest
(circle), arteriolar vascular tone is relatively high. If arterioles are maxi-
mally dilated (triangle), coronary flow can increase 5-fold. (Previously
published. See “Credit Lines” section.)
by the action of the enzyme nitric oxide synthase, which metabo-
lizes L-arginine to citrulline. Endothelial cells are stimulated to
produce nitric oxide by several factors, including acetylcholine,
histamine, bradykinin, substance P, and platelet-derived sub-
stances. Fluid shear stress also stimulates the release of nitric
oxide, which contributes to flow-mediated dilation and the arte-
rial remodeling that occurs with changes in coronary blood
flow states. Nitric oxide acts by increasing intracellular levels of
cyclic guanosine monophosphate, which mediates smooth mus-
cle relaxation and platelet inhibition.
Therapeutically, L-arginine has been used to drive nitric oxide
synthase and promote nitric oxide production. The endothelial
cells also produce endothelin, a vasoactive peptide that directly
stimulates receptors on smooth muscle cells. Endothelin, one of
the most potent vasoconstrictors known, is produced in response
to endothelial cell stimulation by thrombin, transforming growth
factor β, interleukin 1, epinephrine, antidiuretic hormone, and
angiotensin II. The endothelium can convert angiotensin to angi-
otensin II through tissue-bound angiotensin-converting enzyme,
which also causes vasoconstriction.
• A normally functioning endothelium is essential for maintaining
normal coronary blood flow.
• Endothelin is one of the most potent vasoconstrictors known.
Altered Coronary Physiology
Obstructive Coronary Artery Disease
Essentially all the clinical manifestations of obstructive coro-
nary artery disease (due to coronary atherosclerosis and rarely to
tion in the perfusion pressures distal to the stenoses. Resting
(ie, autoregulation). Although this mechanism maintains resting
compromised. Furthermore, any attempt to increase flow across
Autoregulation fails when the severity of the stenosis severely
decreases the distal pressure beyond the lower limits of auto-
regulation. At this point, resting coronary blood flow decreases,
resulting in resting myocardial ischemia and rest angina.
Myocardial viability can be maintained with coronary blood
flow as low as 10 to 20 mL/min per 100 g if the mechanical
activity and metabolic needs of the heart are reduced (as during
hypothermic cardioplegia).
• Myocardial viability can be maintained with coronary blood flow
as low as 10–20 mL/min per 100 g.
Nonobstructive Coronary Artery Disease
Endothelial Dysfunction
Endothelial dysfunction precedes clinical atherosclerosis and
has been detected in patients with normal findings on angiog-
raphy and intravascular ultrasonography, although microscopic
abnormalities probably are present in the endothelial cells. The
relationship of endothelial dysfunction to cardiovascular disease
is complex. Endothelial dysfunction can be considered the initial
pathophysiologic step in the development of atherosclerosis, and
its presence is a predictor of adverse cardiovascular events.
Several factors cause endothelial dysfunction: hyperlipidemia,
cigarette smoking, uncontrolled hypertension, diabetes mellitus,
and decreased estrogen levels in postmenopausal women. Recent
evidence suggests that infectious pathogens and environmental
pollution may also lead to endothelial injury. Endothelial dys-
function leads to the expression of adhesion molecules on the
cell surface and to increased oxidative stress in the vessels,
which facilitates leukocyte recruitment, entry, and oxidation of
low-density lipoprotein into the intima, paving the way for the
formation of foam cells and atherosclerotic plaques. Endothelial
dysfunction also leads to reduced bioavailability of nitric oxide
through multiple deleterious mechanisms and stimulates the for-
mation of endothelin, which impairs vasomotion.
Acetylcholine can be selectively infused into the coronary
arteries to test for the presence of endothelial dysfunction (see

Table 103.1. Comparison of 3 Types of Coronary Flow Reserve

Feature
Definition
Independent of driving pressure
Easily applicable in humans
Applicable to 3-vessel disease
Unequivocal reference value
Abnormal value
Previously published. See “Credit Lines” section.
Absolute Flow Reserve Relative Flow Reserve Fractional Flow Reserve
Ratio of hyperemic flow Ratio of hyperemic flow in the Ratio of hyperemic flow in the stenotic
to resting flow stenotic region to hyperemic region to hyperemic flow in that
flow in a normal region same region if no lesion is present
− + +
+ ± +
+ − +
− + +
<2.0 ≤0.65 ≤0.80
 103 Coronary Artery Physiology and Intracoronary Imaging
Box 103.1. Functions of the Endothelium
Regulate vasomotor tone
Regulate thrombosis and fibrinolysis
Regulate vascular cell growth by producing
growth factors and inhibitors
Regulate leukocyte and platelet adhesion
Modulate lipid oxidation
Act as a selectively permeable barrier
Modulate thrombogenic response
below). Normally, acetylcholine stimulates the production and
release of nitric oxide from the endothelium, which results in
vasodilatation and increased coronary blood flow. In patients
with endothelial dysfunction, acetylcholine is ineffective in
releasing nitric oxide, and its direct action on smooth muscle
cells causes vasoconstriction (called paradoxical vasoconstric-
tion). Noninvasive methods for testing endothelial function have
been developed, but their role in clinical practice remains to be
established.
Endothelial cell dysfunction may be treated with lifestyle
modifications, including aerobic exercise, weight loss in obese
patients, low-fat or Mediterranean diets, and smoking cessation.
Aggressive treatment of risk factors such as diabetes mellitus,
metabolic syndrome, and hypertension would be expected to
improve vascular function. Pharmacologic therapy with statins
and angiotensin-converting enzyme inhibitors has been shown
to reverse endothelial dysfunction. The role of experimental
nitric oxide donors, fish oils, and antioxidants, has not been
established.
• Endothelial dysfunction can be considered the initial pathophysi-
ologic step in the development of atherosclerosis.
• Patients with endothelial dysfunction are at increased risk of
adverse cardiovascular outcomes.
• With normally functioning endothelium, acetylcholine stimulates
the production and release of nitric oxide, which results in vasodi-
latation and increased coronary blood flow.
Resting Coronary
85% 70%
Blood Flow
90% 80% 50%
Coronary Pressure
Figure 103.5. Effect of Stenosis Severity on Resting Coronary
957
Overload States
Any
. increase in the workload of the heart increases myocardial
VO2, and
. coronary blood flow increases in parallel with .myocar-
dial VO2. Acute pressure overload increases myocardial VO2 more
than acute volume overload, although most volume overloads
include some element of pressure overload.
Chronic overload results in myocardial hypertrophy, with
pressure overload causing concentric hypertrophy and volume
overload causing eccentric or dilated hypertrophy. The degree
of hypertrophy
. is related to the workload of the heart or to the
myocardial VO2 (and, thus, coronary blood flow). Therefore, cor-
onary blood flow per unit of myocardium (milliliters per minute
per 100 g) is frequently normal in hypertrophic states. The
increased resting flow often results in a decreased CFR. When
the heart is pressure overloaded, maldistributions in coronary
blood flow occur because of increased compressive forces in
the subendocardial layer. Even in patients with normal coronary
angiographic findings, decreased subendocardial flow can result
in angina pectoris and exercise-induced ST-segment depression,
a finding often seen in aortic stenosis, hypertrophic cardiomy-
opathy, and hypertensive left ventricular hypertrophy.
Intravascular Ultrasonography
Although coronary angiography is considered the reference
standard for coronary artery imaging, it has several inherent
limitations:
1. Angiography is a silhouette technique that detects only arterial dis-
ease that indents the luminal column of contrast medium; it reveals
little else about the atherosclerotic plaque.
2. When a normal proximal reference segment does not exist, as in dif-
fuse atherosclerotic disease, atherosclerosis is difficult to detect and
quantify with angiography.
3. Angiography underestimates the amount of atherosclerosis when
compensatory arterial enlargement occurs—that is, when the artery
at the site of an atherosclerotic plaque expands (positive remodel-
ing). Positive remodeling is associated with a larger plaque burden
and a large lipid-rich soft core, and it is thought to occur at sites of
active disease with “vulnerable” plaques. Negative remodeling can
occur when there is contraction of the vessel. The remodeling index
is defined as the ratio of the vessel external elastic membrane area at
the lesion site to the external elastic membrane area at the reference
site (either proximal or distal location, depending on the segment
that has the least plaque). A remodeling index greater than 1.05 indi-
cates positive remodeling (Figure 103.6), and a remodeling index
less than 0.95 indicates negative remodeling (Figure 103.7).
4. Frequently, angiography cannot detect intimal lesions, including
angioplasty-induced microfractures, intimal dissections, and mural
thrombi.
• Positive remodeling is associated with a larger plaque burden
and a large lipid-rich soft core, and it is thought to occur at sites
of “vulnerable” plaques.
• The remodeling index is defined as the ratio of the vessel exter-
nal elastic membrane area at the lesion site to the external elas-
tic membrane area at the reference site (either proximal or distal
location, depending on the segment that has the least plaque).
• A remodeling index greater than 1.05 indicates positive remod-
eling, and a remodeling index less than 0.95 indicates negative
remodeling.
Ca