Neurology and General Medicine E-Book
2533 pages

Vous pourrez modifier la taille du texte de cet ouvrage

Neurology and General Medicine E-Book


Obtenez un accès à la bibliothèque pour le consulter en ligne
En savoir plus
2533 pages

Vous pourrez modifier la taille du texte de cet ouvrage

Obtenez un accès à la bibliothèque pour le consulter en ligne
En savoir plus


Better understand your patients' complete medical profile and provide the best possible care! This one-of-a-kind reference provides a practical look at neurological disease and how it affects, and is affected by, other disease. It helps neurologists manage patients with co-existing medical conditions, and helps internists understand and treat the neurological manifestations of patients' primary diseases. A new emphasis on diagnosis and management—including advances in pharmacology, genetic-based therapies, and new imaging techniques—makes this 4th Edition more clinically valuable than ever!
  • Focused content highlights the vital links between neurology and other medical specialties, promoting a better understanding of all disciplines, as well as enhancing patient care.
  • Comprehensive coverage of advances in pharmacology, such as new antibiotics for infectious diseases, helps you successfully manage a full range of diseases and disorders.
  • An interdisciplinary team of authors provides insight into the neurological aspects of the conditions you see in daily practice.
  • Easy-to-read chapters apply equally well to neurologists and non-neurologists, providing essential knowledge that covers the full spectrum of medical care.
  • Expanded chapters emphasize key diagnostic and therapeutic information, including appropriate testing and treatments for neurological disease.
  • An emphasis on advances in pharmacology and new imaging techniques helps you better manage your patients and understand how new drugs or therapies will affect your patients and practice.
  • New chapters on auditory and vestibular disease, ocular disease, and cutaneous disease provide a well-rounded look at the specialty.
  • Updated illustrations make complex concepts easier to understand and apply.


Derecho de autor
Vértigo (desambiguación)
Cardiac dysrhythmia
Lepromatous leprosy
Parkinson's disease
Human T-lymphotropic virus
Spinal cord
Pertussis vaccine
Systemic lupus erythematosus
Atrial fibrillation
Myocardial infarction
Alzheimer's disease
Hematologic disease
Bone disease
Paraneoplastic syndrome
Tonic?clonic seizure
Endocrine disease
Gastrointestinal physiology
Systemic disease
Contrast medium
Partial seizure
Atopic dermatitis
Connective tissue disease
Hepatic encephalopathy
Digestive disease
Traumatic brain injury
Subdural hematoma
Subarachnoid hemorrhage
Interventional cardiology
Sex steroid
Peripheral neuropathy
Tuberous sclerosis
Low molecular weight heparin
Infective endocarditis
Diabetic neuropathy
Nutrition disorder
Pulmonary edema
Sexual dysfunction
Human T-lymphotropic virus 1
Multiple myeloma
Health care
Immunosuppressive drug
Internal medicine
General practitioner
Ventricular fibrillation
Urinary incontinence
Rapid eye movement sleep
Organ transplantation
Orthostatic hypotension
Cardiac arrest
Multiple sclerosis
Hearing impairment
Diabetes mellitus
Transient ischemic attack
Epileptic seizure
Nervous system
Magnetic resonance imaging
Erectile dysfunction
Major depressive disorder
Central nervous system
Hypertension artérielle
Divine Insanity
Headache (EP)
Delirium tremens
On Thorns I Lay
Hypotension artérielle
Maladie infectieuse


Publié par
Date de parution 06 décembre 2007
Nombre de lectures 0
EAN13 9780702036064
Langue English
Poids de l'ouvrage 10 Mo

Informations légales : prix de location à la page 0,1019€. Cette information est donnée uniquement à titre indicatif conformément à la législation en vigueur.


Neurology and General Medicine
Fourth Edition

Michael J. Aminoff, MD, DSc, FRCP
Professor, Department of Neurology, School of Medicine, University of California, San Francisco, California
Churchill Livingstone
1600 John F. Kennedy Blvd.
Suite 1800
Philadelphia, PA 19103-2899
ISBN: 978-0-443-06707-5
Copyright © 2008, 2001, 1995, 1989 by Churchill Livingstone, an imprint of Elsevier Inc.
All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permissions may be sought directly from Elsevier’s Rights Department: phone: (+1) 215 239 3804 (US) or (+44) 1865 843830 (UK); fax: (+44) 1865 853333; e-mail: . You may also complete your request on-line via the Elsevier website at <> .

Knowledge and best practice in this field are constantly changing. As new research and experience broaden our knowledge, changes in practice, treatment, and drug therapy may become necessary or appropriate. Readers are advised to check the most current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be administered, to verify the recommended dose or formula, the method and duration of administration, and contraindications. It is the responsibility of the practitioners, relying on their own experience and knowledge of the patients, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety precautions. To the fullest extent of the law, neither the Publisher nor the Editor assumes any liability for any injury and/or damage to persons or property arising out or related to any use of the material contained in this book.
The Publisher

Library of Congress Cataloging-in-Publication Data
Neurology and general medicine / edited by Michael J. Aminoff.—4th ed.
p. ; cm.
Includes bibliographical references and index.
ISBN 978-0-443-06707-5
1. Neurologic manifestations of general diseases. 2. Nervous system—Diseases—Complications.I. Aminoff, Michael J. (Michael Jeffrey) II. Title.
[DNLM: 1. Nervous System Diseases—complications. 2. Neurologic Manifestations. WL 340 N4932 2008]
RC347.N479 2008
Acquisitions Editor: Adrianne Brigido
Developmental Editor: Joan Ryan
Design Direction: Steven Stave
Printed in China
Last digit is the print number: 9 8 7 6 5 4 3 2 1
This book is dedicated to the memory of Abraham S. Aminoff, my father and friend.
It is also dedicated to my three children, Alexandra, Jonathan, and Anthony, as they go their own ways and face the many opportunities and challenges ahead of them.

Gary M. Abrams, MD , Associate Professor, Department of Neurology, School of Medicine, University of California, San Francisco, Rehabilitation Section Chief, San Francisco Veterans Affairs Medical Center, San Francisco, California, Other Endocrinopathies and the Nervous System

Gregory W. Albers, MD , Professor, Department of Neurology and Neurological Sciences, Stanford University School of Medicine, Stanford, Director, Stanford Stroke Center, Stanford University Medical Center, Palo Alto, California, Stroke as a Complication of General Medical Disorders

Bradley L. Allen, MD, PhD , Clinical Associate Professor, Department of Medicine, Division of Infectious Diseases, Indiana University School of Medicine, Chief, Medicine Service and Section of Infectious Diseases, Richard L. Roudebush Veterans Affairs Medical Center, Indianapolis, Indiana, Neurological Manifestations of Infective Endocarditis

Michael J. Aminoff, MD, DSc, FRCP , Professor, Department of Neurology, School of Medicine, University of California, San Francisco, California, Postural Hypotension; Neurological Dysfunction and Kidney Disease; Sexual Dysfunction in Patients With Neurological Disorders ; Pregnancy and Disorders of the Nervous System ; The Neurology of Aging ; Seizures and General Medical Disorders ; Movement Disorders Associated With General Medical Diseases ; Neuromuscular Complications of General Medical Disorders ; Care at the End of Life

Bruce O. Berg, MD , Professor Emeritus, Departments of Neurology and Pediatrics, School of Medicine, University of California, San Francisco, California, Neurocutaneous Syndromes

Joseph R. Berger, MD , Professor and Chair, Department of Neurology, University of Kentucky College of Medicine, Lexington, Kentucky, AIDS and the Nervous System

Timothy G. Berger, MD , Professor, Department of Dermatology, School of Medicine, University of California, San Francisco, California, Dermatological–Neurological Interactions

Adil E. Bharucha, MD , Associate Professor, Department of Medicine, Mayo Clinic College of Medicine, Rochester, Minnesota, Disturbances of Gastrointestinal Motility and the Nervous System

Charles F. Bolton, MD, FRCP(C) , Adjunct Professor, Department of Neurology, School of Medicine, Queen’s University, Kingston, Ontario, Canada, Neurological Complications in Critically Ill Patients

David L. Brown, MD , Edward Rotan Distinguished Professor and Chair, Department of Anesthesiology and Pain Medicine, University of Texas, M. D. Anderson Cancer Center, Houston, Texas, Neurological Complications of Anesthesia

Christine E. Burness, MSc, MB, ChB, MRCP , Research Fellow, Department of Neurology, University of Sheffield Medical School, Sheffield, England, Thyroid Disease and the Nervous System

Michael Camilleri, MD , Professor, Departments of Medicine and Physiology, Mayo Clinic College of Medicine, Rochester, Minnesota, Disturbances of Gastrointestinal Motility and the Nervous System

Vinay Chaudhry, MD, FRCP , Professor, Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, Maryland, Other Neurological Disorders Associated With Gastrointestinal, Liver, or Pancreatic Diseases

Chadwick W. Christine, MD , Assistant Professor, Department of Neurology, School of Medicine, University of California, San Francisco, California, Movement Disorders Associated With General Medical Diseases

Kimberly P. Cockerham, MD, FACS , Associate Clinical Professor, Department of Ophthalmology, Stanford University School of Medicine, Stanford, California, Orbital and Ocular Manifestations of Neurological Disease

Gary M. Cox, MD , Associate Professor, Department of Medicine, Duke University School of Medicine, Durham, North Carolina, Fungal Infections of the Central Nervous System

G.A.B. Davies-Jones, MD, FRCP , Lecturer in Medicine, University of Sheffield Medical School, Consultant Neurologist, Royal Hallamshire Hospital, Sheffield, England, Neurological Manifestations of Hematological Disorders

Larry E. Davis, MD , Professor, Department of Neurology, University of New Mexico School of Medicine, Chief, Neurology Service, New Mexico Veterans Affairs Health Care System, Albuquerque, New Mexico, Nervous System Complications of Systemic Viral Infections

Lisa M. Deangelis, MD , Professor, Department of Neurology and Neuroscience, Weill Medical College of Cornell University, Chair, Department of Neurology, Memorial Sloan-Kettering Cancer Center, New York, New York, Neurological Complications of Chemotherapy and Radiation Therapy

Philip R. Delio, MD , Director of Stroke Services, Department of Neurology, Santa Barbara Cottage Hospital, Santa Barbara, California, Stroke as a Complication of General Medical Disorders

William P. Dillon, MD , Professor, Departments of Radiology, Neurology, and Neurosurgery, School of Medicine, University of California, San Francisco, California, Neurological Complications of Imaging Procedures

Christopher F. Dowd, MD , Associate Professor, Departments of Radiology and Neurological Surgery, School of Medicine, University of California, San Francisco, California, Neurological Complications of Imaging Procedures

AdrÉ J. Du Plessis, MBChB, MPH , Associate Professor, Department of Neurology, Harvard Medical School, Associate in Neurology and Director, Fetal-Neonatal Neurology Program, Department of Neurology, Children’s Hospital, Boston, Massachusetts, Neurological Complications of Congenital Heart Disease and Cardiac Surgery in Children

David T. Durack, MB, DPhil, FRCP , Consulting Professor, Department of Medicine, Duke University School of Medicine, Durham, North Carolina, Senior Vice President, Corporate Medical Affairs, Becton, Dickinson and Company, Franklin Lakes, New Jersey, Fungal Infections of the Central Nervous System

Jacob S. Elkins, MD , Professor, Department of Neurology, School of Medicine, University of California, San Francisco, California, Neurological Complications of Hypertension

John W. Engstrom, MD , Professor, Department of Neurology, School of Medicine, University of California, San Francisco, California, HTLV-I Infection and the Nervous System

Randolph W. Evans, MD , Clinical Professor, Department of Neurology and Neuroscience, Weill Medical College of Cornell University, New York, New York, Clinical Associate Professor, Department of Neurology, Baylor College of Medicine, Houston, Texas, The Postconcussion Syndrome

Eva L. Feldman, MD, PhD , Russell N. DeJong Professor, Department of Neurology, University of Michigan Medical School, Ann Arbor, Michigan, Diabetes and the Nervous System

Bruce J. Fisch, MD , Professor, Department of Neurology, Louisiana State University School of Medicine, New Orleans, Louisiana, Neurological Aspects of Sleep

Joseph M. Furman, MD, PhD , Professor, Departments of Otolaryngology and Neurology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, Otoneurological Manifestations of Otological and Systemic Disease

Douglas J. Gelb, MD, PhD , Professor, Department of Neurology, University of Michigan Medical School, Ann Arbor, Michigan, Abnormalities of Thermal Regulation and the Nervous System

David J. Gladstone, MD, FRCP(C), PhD , Assistant Professor, Department of Neurology, University of Toronto Faculty of Medicine, Toronto, Ontario, Canada, Neurological Manifestations of Acquired Cardiac Disease, Arrhythmias, and Interventional Cardiology

Douglas S. Goodin, MD , Professor, Department of Neurology, School of Medicine, University of California, San Francisco, California, Neurological Complications of Aortic Disease and Surgery

Sean A. Grimm, MD , Fellow, Department of Neurology, Memorial Sloan-Kettering Cancer Center, New York, New York, Neurological Complications of Chemotherapy and Radiation Therapy

John J. Halperin, MD , Clinical Professor, Department of Neurology, New York University School of Medicine, New York, New York, Spirochetal Infections of the Nervous System

J. Claude Hemphill, III , MD , Associate Professor, Department of Neurology, School of Medicine, University of California, San Francisco, California, Disorders of Consciousness in Systemic Diseases

John R. Hotson, MD , Professor, Department of Neurology and Neurological Sciences, Stanford University School of Medicine, Stanford, California, Neurological Complications of Cardiac Surgery

Cheryl A. Jay, MD , Clinical Professor, Department of Neurology, School of Medicine, University of California, San Francisco, California, Other Endocrinopathies and the Nervous System

S. Claiborne Johnston, MD, PhD , Professor, Department of Neurology, School of Medicine, University of California, San Francisco, California, Neurological Complications of Hypertension

Charles H. King, MD , Professor, Center for Global Health and Diseases, Case Western Reserve University School of Medicine, Cleveland, Ohio, Parasitic Infections of the Central Nervous System

Nerissa U. Ko, MD , Assistant Professor, Department of Neurology, School of Medicine, University of California, San Francisco, California, Cardiac Manifestations of Acute Neurological Lesions

Allan Krumholz, MD , Professor, Department of Neurology, University of Maryland School of Medicine, Baltimore, Maryland, Sarcoidosis of the Nervous System

Colin D. Lambert, BM, FRCP, FRCP(C) , Associate Professor, Department of Medicine, University of Toronto Faculty of Medicine, Toronto, Ontario, Canada, Neurological Manifestations of Acquired Cardiac Disease, Arrhythmias, and Interventional Cardiology

J. William Langston, MD , Scientific Director, Parkinson’s Institute, Sunnyvale, California, Neuropsychiatric Complications of Substance Abuse

John M. Leonard, MD , Professor, Department of Medicine, Division of Infectious Diseases, Vanderbilt University School of Medicine, Nashville, Tennessee, Tuberculosis of the Central Nervous System

Catherine Limperopoulos, PhD , Assistant Professor and Canada Research Chair in Brain and Development, Departments of Neurology and Neurosurgery, McGill University Faculty of Medicine, Montreal, Quebec, Canada, Neurological Complications of Congenital Heart Disease and Cardiac Surgery in Children

Alan H. Lockwood, MD , Professor, Departments of Neurology and Nuclear Medicine, University at Buffalo School of Medicine and Biomedical Sciences, Buffalo, New York, Hepatic Encephalopathy

W.T. Longstreth, JR. , MD, MPH , Professor, Department of Neurology, University of Washington School of Medicine, Adjunct Professor, Department of Epidemiology, University of Washington School of Public Health and Community Medicine, Seattle, Washington, Neurological Complications of Cardiac Arrest

Elliott L. Mancall, MD , Professor, Department of Neurology, Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, Nutritional Disorders of the Nervous System

Frank L. Mastaglia, MD , Professor, Centre for Neuromuscular and Neurological Disorders, University of Western Australia School of Medicine, Nedlands, Western Australia, Australia, Drug-Induced Disorders of the Nervous System

Una D. Mccann, MD , Associate Professor, Department of Psychiatry and Behavioral Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland, Neuropsychiatric Complications of Substance Abuse

Robert O. Messing, MD , Professor, Department of Neurology, School of Medicine, University of California, San Francisco, Associate Director, Ernest Gallo Clinic and Research Center, Emeryville, California, Alcohol and the Nervous System

Andrea Olmos, BA , Research Assistant, Department of Ophthalmology, Stanford University School of Medicine, Stanford, California, Orbital and Ocular Manifestations of Neurological Disease

Richard K. Olney, MD , Professor, Department of Neurology, School of Medicine, University of California, San Francisco, California, The Neurology of Aging

Jack M. Parent, MD , Associate Professor, Department of Neurology, University of Michigan Medical School, Ann Arbor, Michigan, Seizures and General Medical Disorders

Gareth J. Parry, MD , Professor, Department of Neurology, University of Minnesota Medical School, Minneapolis, Minnesota, Neurological Complications of Toxin Exposure in the Workplace

Roy A. Patchell, MD , Professor, Department of Neurosurgery, University of Kentucky Medical School, Chief of Neuro-oncology, University of Kentucky Medical Center, Lexington, Kentucky, Neurological Complications of Organ Transplantation and Immunosuppressive Agents

John R. Perfect, MD , Professor, Department of Medicine, Division of Infectious Diseases, Duke University School of Medicine, Durham, North Carolina, Fungal Infections of the Central Nervous System

Ann Noelle Poncelet, MD , Professor, Department of Neurology, School of Medicine, University of California, San Francisco, California, Neurological Disorders Associated With Bone and Joint Disease

Rodica Pop-Busui, MD, PhD , Assistant Professor, Department of Internal Medicine, University of Michigan Medical School, Ann Arbor, Michigan, Diabetes and the Nervous System

Jerome B. Posner, MD , Professor, Department of Neurology and Neuroscience, Weill Medical College of Cornell University, Attending Neurologist, Memorial Sloan-Kettering Cancer Center, New York, New York, Paraneoplastic Syndromes Involving the Nervous System

Jeffrey W. Ralph, MD , Assistant Professor, Department of Neurology, School of Medicine, University of California, San Francisco, California, Neuromuscular Complications of General Medical Disorders

William J. Ravich, MD , Associate Professor, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland, Other Neurological Disorders Associated With Gastrointestinal, Liver, or Pancreatic Diseases

George A. Ricaurte, MD, PhD , Professor, Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, Maryland, Neuropsychiatric Complications of Substance Abuse

Jack E. Riggs, MD , Professor, Department of Neurology, West Virginia University School of Medicine, Morgantown, West Virginia, Neurological Manifestations of Electrolyte Disturbances

Karen L. Roos, MD , Professor, Departments of Neurology and Neurosurgery, Indiana University School of Medicine, Indianapolis, Indiana, Acute Bacterial Infections of the Central Nervous System

Andrew P. Rose-Innes, MBChB , Assistant Professor, Department of Neurology, University of Washington School of Medicine, Seattle, Washington, Neurological Disorders Associated With Bone and Joint Disease

Richard B. Rosenbaum, MD , Clinical Professor, Department of Neurology, School of Medicine, Oregon Health & Science University, Attending Neurologist, Oregon Clinic, Portland, Oregon, Connective Tissue Diseases, Vasculitis, and the Nervous System

Thomas D. Sabin, MD , Professor, Department of Neurology, Tufts University School of Medicine, Boston, Massachusetts, Neurological Complications of Leprosy

Robert A. Salata, MD , Professor, Department of Medicine, Case Western Reserve University School of Medicine, Cleveland, Ohio, Parasitic Infections of the Central Nervous System

Hyman M. Schipper, MD, PhD, FRCP(C) , Professor, Departments of Neurology and Neurosurgery and Department of Medicine (Geriatrics), McGill University Faculty of Medicine, Neurologist and Director, Center for Neurotranslational Research, Lady Davis Institute for Medical Research, Sir Mortimer B. Davis Jewish General Hospital, Montreal, Quebec, Canada, Sex Hormones and the Nervous System

Pamela J. Shaw, MD, FRCP , Professor, Department of Neurology, University of Sheffield Medical School, Sheffield, England, Thyroid Disease and the Nervous System

Roger P. Simon, MD , Director and Chair, Dow Neurobiology Laboratories, Portland, Oregon, Breathing and the Nervous System

Michele B. St. Martin, MD , Visiting Instructor, Department of Otolaryngology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, Otoneurological Manifestations of Otological and Systemic Disease

Barney J. Stern, MD , Professor, Department of Neurology, University of Maryland School of Medicine, Baltimore, Maryland, Sarcoidosis of the Nervous System

Kelli A. Sullivan, PhD , Assistant Research Professor, Department of Neurology, University of Michigan Medical School, Ann Arbor, Michigan, Diabetes and the Nervous System

Jon D. Sussman, MB, ChB, PhD, FRCP , Honorary Lecturer, Department of Neurology, University of Manchester Medical School, Consultant Neurologist, Greater Manchester Neuroscience Centre, Hope Hospital, Salford, Greater Manchester, England, Neurological Manifestations of Hematological Disorders

Michael Swash, MD, FRCP , Professor, Department of Neurology, Royal London Hospital, London, England, Sphincter Disorders and the Nervous System

Thomas R. Swift, MD, FAAN , Professor, Department of Neurology, Medical College of Georgia, Augusta, Georgia, Neurological Complications of Leprosy

Michael R. Trimble, MD, FRCP, FRCPsych , Emeritus Professor of Behavioural Neurology, Institute of Neurology, University College London, London, England, Psychiatry and Neurology

Angela T. Truong, MD , Assistant Professor, Department of Anesthesiology and Pain Medicine, University of Texas, M. D. Anderson Cancer Center, Houston, Texas, Neurological Complications of Anesthesia

Alex C. Tselis, MD, PhD , Associate Professor, Department of Neurology, Wayne State University School of Medicine, Detroit, Michigan, Neurological Complications of Vaccination

David B. VoduŠek, MD, DSc , Professor and Chair, Department of Neurology, School of Medicine, University of Ljubljana, Ljubljana, Slovenia, Sexual Dysfunction in Patients With Neurological Disorders

Kevin C. Wang, MD, PhD , Resident Physician, Department of Dermatology, School of Medicine, University of California, San Francisco, California, Dermatological–Neurological Interactions

Linda S. Williams, MD , Professor, Department of Neurology, Indiana University School of Medicine, Indianapolis, Indiana, Neurological Manifestations of Infective Endocarditis

Marc D. Winkelman, MD , Associate Professor, Department of Neurology, Case Western Reserve University School of Medicine, Cleveland, Ohio, Neurological Complications of Thermal and Electrical Burns

G. Bryan Young, MD, FRCP(C) , Professor, Department of Neurology, University of Western Ontario Faculty of Medicine, Consultant Neurologist, London Health Sciences Centre, London, Ontario, Canada, Neurological Complications in Critically Ill Patients

Jonathan G. Zaroff, MD , Assistant Professor, Department of Medicine, School of Medicine, University of California, San Francisco, California, Cardiac Manifestations of Acute Neurological Lesions
Preface to the Fourth Edition
Medical practice continues to evolve, but with changes that occasion both excitement and dismay. Technological developments have led to a gradual erosion of clinical skills; easy access to information technology has led to more informed but not necessarily better educated patients; an increasing bureaucracy has required physicians to spend time on paperwork rather than on patient care; translational research has lagged behind the spectacular advances occurring in the laboratory—the list seems endless. In addition, the mistaken belief is perpetuated that the excellence of a health care delivery system is reflected by its cost. In many countries of the developed world, costs have risen sharply but the quality of care has seemed to decline. A lack of clinical common sense, a failure to grasp the importance of the quality of life, and an inability to optimize the use of limited resources help to round out the picture of a field in some disarray. Even so, advances have been spectacular. For example, new imaging modalities and new therapeutic agents allow clinicians in the developed world to manage patients with greater precision and optimism, and advances in basic and applied immunology and an increased understanding of molecular biology at the level of genes, proteins, and ion channels have helped practitioners to clarify and solve previously impenetrable clinical problems.
The medical literature, which has grown at an alarming rate, is now all but unmanageable in size. With new journals proliferating, quarterly journals becoming monthlies, and monthly journals now appearing weekly, the reader is faced with a daunting prospect in endeavoring to keep abreast of advances in the field. Furthermore, to clinicians, much of the material published in the leading medical journals seems divorced from the realities of clinical medicine and of little relevance to its practice. Specialties are fragmenting into ever smaller subspecialties, many with their own certification process to give them legitimacy, and—in part because of the jargon that has evolved within each area—communication among specialists or even among subspecialists within the same overall specialty is becoming increasingly limited.
It is in this context that the new edition of this book has been developed. As with earlier editions, which received a generous acceptance, it is hoped that the book will appeal to neurologists, specialists in other clinical fields, hospitalists, and primary health-care providers by both defining the neurological aspects of general medical disorders and discussing the non-neurological (general medical and other) aspects of various neurological diseases. The book is intended not to provide a comprehensive account of clinical neurology but to serve as an interface between neurology and the other clinical specialties. Despite the trend toward specialization, it is essential that all physicians—and especially neurologists—gain or retain expertise in general medicine. Many patients with neurological disorders are middle-aged or elderly and have a wide variety of general medical disorders, and many general medical disorders have neurological complications or manifest as neurological diseases. This is exemplified by patients who present with strokes. Cardiac disease and hematological disorders account for a sizable number of strokes; infection (e.g., infective endocarditis and meningitis), inflammatory diseases (e.g., vasculitis and connective tissue disease), and neoplastic diseases are other well-recognized causes. Clearly, competent neurologists require knowledge of general medicine to manage such patients successfully. Further, as neurologists and other specialists work alongside and communicate with those in other medical areas, they require basic knowledge of specialties outside their own at a time when the advance of medicine is so rapid that it is hard to remain current. The lack of consensus among primary care physicians and specialists concerning the appropriate extent of specialist involvement in the care of patients with neurological conditions requires not only that patient care be coordinated among physicians but also that neurologists have some understanding of the general medical issues relating to their patients and that primary care physicians, in turn, have an appreciation of their neurological disorders.
This book is aimed at both general physicians and neurologists, regardless of their level of training or experience. I hope that it will help to guide more junior physicians in the diagnosis and management of patients and in the development of a fundamental base of clinical knowledge. Experience is necessary to become an outstanding physician, and it requires not only years of training to acquire the requisite knowledge and skills to be a neurologist but also lifelong learning to maintain these skills. As diagnostic strategies and treatment modalities advance, more senior practitioners can be left behind. Thus, it is my hope that this book will appeal to them also as a source of reference and guide to the care of patients with diseases that may already be familiar to them.
This edition includes chapters updated from the third edition (or replaced by new chapters on the same topic), as well as three new chapters on additional topics. The references have also been updated. Although many of the older references have been replaced (but can be found by interested readers in earlier editions), a number have been retained because they are classic papers or provide seminal descriptions of particular diseases, clinical phenomena, or treatment to which reference may still usefully be made. This edition also features the addition of a companion website with full-text search capability and links to PubMed. It is my hope that the online site will allow readers to access the text easily while out of the office.
I am grateful to the authors who contributed to this volume and gracefully tolerated my editorial suggestions and interventions. Their generosity in sharing their knowledge and skill with others is particularly appreciated at a time when there are so many competing and conflicting demands on their time.
During the production of the second edition of Neurology and General Medicine , my father died unexpectedly, and I dedicated the book to his memory. In the 13 years that have passed since that time, I have thought of him often. I still remember how he—an engineer by profession—read my first book, a small monograph on spinal angiomas published in the mid-1970s, from cover to cover when first it appeared, even though he must have found much of it difficult to understand, and how he showed it proudly to all his friends. I like to think that he would have had the same pride in the present volume, which is once again dedicated to his memory as a measure of my affection, esteem, and gratitude. This edition is also again dedicated to my three children, who have given me so much happiness over the years. My daughter, Alexandra, is now a final-year medical student; my son Jonathan is a final-year law student; and my younger son, Anthony, is in his final year as an undergraduate at the University of California at Berkeley, from which he also hopes to go on to law school. It is my hope that they will find their professional lives as fulfilling, rewarding, challenging, and enjoyable as I have my own.
I could not have undertaken the compilation of this book without the assistance of my wife, Jan, who gave me the support, encouragement, and time to complete this new revision, taking on many additional chores, without complaint, to ease my burden. Finally, I am grateful to my editors at Elsevier, and especially to Susan Pioli, Joan Ryan, and Adrianne Brigido for their unfailing assistance in the development of this book, and to Joan Sinclair and Joan Vidal for seeing it through the production process. They all were a joy to work with and went out of their way to help with any special requests and to satisfy my every concern.

Michael J. Aminoff, MD, DSc, FRCP
Preface to the First Edition
The increasing sophistication and complexity of modern medicine have led to greater specialization among practitioners and to more restricted communication between physicians in different disciplines. Perhaps, inevitably, this trend has created certain major problems. These difficulties are particularly well exemplified by the relationship between neurology and general medicine.
For non-neurologists, evaluation of patients with neurological symptoms and signs has always been difficult because of the compexity of the anatomy and physiology of the nervous system and frustrating because the therapeutic options have seemed somewhat limited. Nevertheless, a number of neurological diseases are exacerbated by, or occur as specific complications of, general medical disorders. Appropriate management of these neurological disturbances requires their early recognition and an appreciation of their prognosis. It is equally important to recognize the manner in which such neurological disorders may influence the management of the primary or coexisting medical condition, as well as the manner in which systemic complications of neurological disorders may require somewhat different management than when these complications occur in other settings.
For neurologists, who are being asked increasingly to evaluate neurological disturbances presenting in the context of other medical disorders, the difficulty is equally apparent. The general background of cases is frequently confusing, the relationship of the neurological to the other medical problems is commonly not appreciated, and the manner in which treatment needs to be “tailored” to the specific clinical context is often not clear. Furthermore, neurological disturbances may themselves be the presenting feature of general medical disorders or lead to general medical complications requiring speedy recognition and effective management.
I hope that the present volume will appeal to both neurologists and physicians in other specialties by providing a guide to the neurological aspects of general medical disorders and to some of the medical complications of certain neurological diseases. It is not intended to be a textbook of neurology, but rather a “bridge” between neurology and the other medical specialties.
It is a pleasure to acknowledge the help that I received from various people in developing this book. I am grateful to the various contributors, who devoted a great deal of time and energy to reviewing developments in their own fields of interest and showed considerable tolerance of the many demands that I made upon them. I am grateful also to Mr. Robert Hurley and Ms. Margot Otway at Churchill Livingstone for their help and advice during the preparation of this book. Finally, the support and encouragement of my wife, Jan, and of our children, Alexandra, Jonathan, and Anthony, did much to ease the burden involved in seeing this volume to its conclusion.

Michael J. Aminoff, MD, FRCP
Table of Contents
Preface to the Fourth Edition
Preface to the First Edition
Chapter 1: Breathing and the Nervous System
Chapter 2: Neurological Complications of Aortic Disease and Surgery
Chapter 3: Neurological Complications of Cardiac Surgery
Chapter 4: Neurological Complications of Congenital Heart Disease and Cardiac Surgery in Children
Chapter 5: Neurological Manifestations of Acquired Cardiac Disease, Arrhythmias, and Interventional Cardiology
Chapter 6: Neurological Manifestations of Infective Endocarditis
Chapter 7: Neurological Complications of Hypertension
Chapter 8: Postural Hypotension
Chapter 9: Neurological Complications of Cardiac Arrest
Chapter 10: Cardiac Manifestations of Acute Neurological Lesions
Chapter 11: Neurocutaneous Syndromes
Chapter 12: Dermatological–Neurological Interactions
Chapter 13: Neurological Manifestations of Hematological Disorders
Chapter 14: Hepatic Encephalopathy
Chapter 15: Other Neurological Disorders Associated With Gastrointestinal, Liver, or Pancreatic Diseases
Chapter 16: Disturbances of Gastrointestinal Motility and the Nervous System
Chapter 17: Nutritional Disorders of the Nervous System
Chapter 18: Neurological Dysfunction and Kidney Disease
Chapter 19: Neurological Manifestations of Electrolyte Disturbances
Chapter 20: Thyroid Disease and the Nervous System
Chapter 21: Diabetes and the Nervous System
Chapter 22: Sex Hormones and the Nervous System
Chapter 23: Other Endocrinopathies and the Nervous System
Chapter 24: Neurological Disorders Associated With Bone and Joint Disease
Chapter 25: Otoneurological Manifestations of Otological and Systemic Disease
Chapter 26: Orbital and Ocular Manifestations of Neurological Disease
Chapter 27: Paraneoplastic Syndromes Involving the Nervous System
Chapter 28: Neurological Complications of Chemotherapy and Radiation Therapy
Chapter 29: Connective Tissue Diseases, Vasculitis, and the Nervous System
Chapter 30: Psychiatry and Neurology
Chapter 31: The Postconcussion Syndrome
Chapter 32: Neurological Aspects of Sleep
Chapter 33: Sphincter Disorders and the Nervous System
Chapter 34: Sexual Dysfunction in Patients With Neurological Disorders
Chapter 35: Pregnancy and Disorders of the Nervous System
Chapter 36: Drug-Induced Disorders of the Nervous System
Chapter 37: Alcohol and the Nervous System
Chapter 38: Neuropsychiatric Complications of Substance Abuse
Chapter 39: Neurological Complications of Toxin Exposure in the Workplace
Chapter 40: Acute Bacterial Infections of the Central Nervous System
Chapter 41: Spirochetal Infections of the Nervous System
Chapter 42: Tuberculosis of the Central Nervous System
Chapter 43: Neurological Complications of Leprosy
Chapter 44: Nervous System Complications of Systemic Viral Infections
Chapter 45: AIDS and the Nervous System
Chapter 46: Neurological Complications of Organ Transplantation and Immunosuppressive Agents
Chapter 47: HTLV-I Infection and the Nervous System
Chapter 48: Fungal Infections of the Central Nervous System
Chapter 49: Parasitic Infections of the Central Nervous System
Chapter 50: Neurological Complications of Vaccination
Chapter 51: Sarcoidosis of the Nervous System
Chapter 52: Neurological Complications in Critically Ill Patients
Chapter 53: Neurological Complications of Imaging Procedures
Chapter 54: Neurological Complications of Anesthesia
Chapter 55: Neurological Complications of Thermal and Electrical Burns
Chapter 56: Abnormalities of Thermal Regulation and the Nervous System
Chapter 57: The Neurology of Aging
Chapter 58: Seizures and General Medical Disorders
Chapter 59: Movement Disorders Associated With General Medical Diseases
Chapter 60: Neuromuscular Complications of General Medical Disorders
Chapter 61: Stroke as a Complication of General Medical Disorders
Chapter 62: Disorders of Consciousness in Systemic Diseases
Chapter 63: Care at the End of Life
Chapter 1 Breathing and the Nervous System

Roger P. Simon

Alteration of Gas Exchange
Pulmonary Hydrostatic Pressure
Capillary Permeability
Central Effects on Ventilation
Autonomic Dysfunction
Extrapyramidal Disorders
Forebrain Influences on Ventilation
Apraxia of Ventilatory Movements
Posthyperventilation Apnea
Hindbrain Control of Ventilation
Other Ventilatory Patterns
Cheyne–Stokes Breathing
Central Hyperventilation
Alveolar Hypoventilation
Acute Hypoxia
Chronic Hypercapnia
Acute Hypercapnia
Acute Hypocapnia
Chronic Hypocapnia
The relationship between breathing and the nervous system can be considered from two perspectives, both of which are important to neurologists as well as to general physicians. First, neurological dysfunction can have effects on respiration that may be the most disturbing aspects of the underlying neurological disease. Second, primary respiratory dysfunction may affect the nervous system and lead to a request for neurological consultation. Both interactions are considered in this chapter. In revising this chapter for the current edition, many old but classic references were removed, but interested readers will find these cited in earlier editions, to which they are referred.


Alteration of Gas Exchange
One of the most dramatic and life-threatening effects of nervous system dysfunction on respiration is the impairment of alveolar gas exchange by a neurologically induced increase in pulmonary interstitial and alveolar fluid: the phenomenon of acute pulmonary edema. The fluid producing pulmonary edema originates in the pulmonary capillaries. Fluid movement from the pulmonary capillary bed into the alveolar air space is governed by the variables in the classic Starling equation. In its simplest form, the Starling equation expresses transcapillary fluid flux as a balance between intravascular pressures (tending to push fluid out of the vascular lumen) and plasma osmotic forces (which tend to retain fluid within the vascular lumen) ( Fig. 1-1 ). 1 Although the mechanisms by which neurogenically induced pulmonary edema occurs remain uncertain, 2 the major recognized factors are discussed in the following sections.

FIGURE 1-1 Relationships between microvascular hydrostatic pressure (P MV ); perimicrovascular hydrostatic pressure, that is, within the interstitial space (P PMV ); plasma colloid osmotic pressure (π MV ); and perimicrovascular pericolloid osmotic pressure (π Mv ). Under normal conditions, the sum of forces is slightly positive, producing a small vascular fluid flux into the pericapillary interstitium of the lung, which is drained as lymph.
(From Fein A, Grossman RF, Jones JG, et al: The value of edema fluid protein measurement in patients with pulmonary edema. Am J Med 67:32, 1979, with permission.)

Pulmonary Hydrostatic Pressure
The main variable under the control of the nervous system affecting pulmonary capillary fluid flux is pulmonary intravascular pressure. A marked increase in this pressure can force fluid from the vascular compartment, flood the interstitial space ( Fig. 1-1 ), produce pulmonary edema, and impair oxygenation.
An elevation in intracranial pressure can unbalance the Starling equation and result in neurogenic pulmonary edema. 2 This early experimental observation in animals has been confirmed in patients with traumatic head injury. 3 Experimental studies have demonstrated that the effect of increased intracranial pressure on pulmonary vascular pressure and transcapillary fluid flux occurs as intracranial pressure approaches systemic pressure. An increase in systemic pressure (the Cushing response) then occurs to protect cerebral perfusion. In most studies, an increase in intracranial pressure alone, in the absence of the Cushing response, has no effect on transcapillary fluid flux in the lung. During the Cushing response, pulmonary vascular pressure increases in concert with systemic pressure, with a resultant increase in pulmonary transcapillary fluid flux. 4 Only one experimental study has shown an increase in pulmonary transcapillary fluid flux in the absence of elevated pulmonary vascular pressure. 5 Other classic models of induction of neurogenic pulmonary edema also appear to be those of centrally induced pulmonary vascular hypertension. These models include “sympathetic activation” induced by intracisternal veratrine and intracisternally administered thrombin and fibrinogen with or without vagotomy in rabbits. 6 - 8
Focal central nervous system (CNS) lesions can cause both an elevation of systemic vascular pressure and pulmonary edema. Although hemodynamic data in humans are lacking, there are many reports that the brainstem, particularly the medulla, is the site of focal CNS injuries that result in pulmonary edema. 2, 9 - 11 In unanesthetized small animals, brainstem lesions in the region of the nucleus tractus solitarius produce marked systemic hypertension and fulminant pulmonary edema; pulmonary vascular pressure cannot be measured in these small animals. Following bilateral lesion placement in the ventral lateral nucleus tractus solitarius in sheep, however, pulmonary arterial pressures and transcapillary fluid flux in the lung can be measured and may increase significantly without a change in systemic or left atrial pressures. 12 This pattern of response to a CNS injury is similar to that reported for neurogenic pulmonary edema in humans. 2 Furthermore, a patient has been reported in whom a unilateral injury occurred to the tractus solitarius during a neurosurgical procedure and in whom the contralateral tractus solitarius was absent because of a congenital brainstem syrinx. The patient died of pulmonary edema and hypoxemia 34 hours postoperatively. 9 The localization by magnetic resonance imaging (MRI) of a lesion at the obex ( Fig. 1-2 ) in patients with recently diagnosed multiple sclerosis (MS) and acute pulmonary edema supports this anatomical site as that inducing neurogenic pulmonary edema. 13, 14 Recent corroborative data come from a subset of patients with EV71 encephalitis in whom brainstem encephalitis and a polio-like acute flaccid paresis picture occur associated with neurogenic pulmonary edema. Brain MRI performed within hours of onset of pulmonary edema showed restricted diffusion in the posterior medulla, anterior to the inferior aspect of the fourth ventricle 15 ( Fig. 1-3 ).

FIGURE 1-2 Rostral to caudal ( A to C ) schematic reconstruction of the medullary lesion in a patient with multiple sclerosis and pulmonary edema, based on magnetic resonance imaging, illustrating the major nuclear groups and tracts involved. AP, area postrema; 4th, fourth ventricle; LRN, lateral reticular nucleus; MLF, medial longitudinal fasciculus; MRN, medial reticular nucleus; NA, nucleus ambiguus; NTS, nucleus of the solitary tract; Ob, obex; ST, solitary tract; V, spinal trigeminal nucleus; X, dorsal motor nucleus of the vagus; XII, hypoglossal nucleus.
(From Simon RP: Respiratory manifestations of neurologic diseases. p. 496. In Goetz CG, Tanner CM, Aminoff MJ [eds]: Handbook of Clinical Neurology. Vol 63. Elsevier, Amsterdam, 1993, with permission.)

FIGURE 1-3 A, Diffusion magnetic resonance image at the level of the fourth ventricle, performed within hours of onset of neurogenic pulmonary edema, showing paired areas of restricted diffusion paracentrally in the region of the dorsal motor nucleus of the vagus, nucleus tractus solitarius, and medial reticular formation. Axial (B) and sagittal (C) T1-weighted images of same patient performed 4 weeks after onset of neurogenic pulmonary edema. Note the well-defined signal abnormality anterior to the inferior aspect of the fourth ventricle consistent with encephalomyelomalacia.
( A kindly provided by Dr. P. Ian Andrews; B and C modified from Nolan MA, Craig ME, Lahra MM, et al: Survival after pulmonary edema due to enterovirus 71 encephalitis. Neurology 60:1651, 2003.)
Generalized seizures produce an abrupt, marked increase in sympathetic outflow from the brain, and both systemic and pulmonary vascular pressures increase. 16 The degree of systemic pressure elevation cor relates with the number of seizures and is maximal during status epilepticus. The magnitude of the pressure elevation in the pulmonary vasculature is independent of the number of seizures, however, although the duration of the elevation is maximal with status epilepticus ( Fig. 1-4 ). The increase in transcapillary fluid flux resulting from this transient pulmonary vascular hypertension persists for hours after the pressure transient 17 and probably explains the phenomenon of pulmonary edema following seizures in humans.

FIGURE 1-4 Vascular pressure changes that occur during seizures in sheep. Mean values have been plotted at 10-second intervals. Spinal cord refers to animals with cervical spinal cord transection prior to seizures. Single, 5, and 20 shocks refer to the number of electroconvulsive seizures induced; bicuculline refers to bicuculline-induced status epilepticus. LA, left atrial; PA, pulmonary arterial.
(From Bayne LL, Simon RP: Systemic and pulmonary vascular pressures during generalized seizures in sheep. Ann Neurol 10:566, 1981, with permission.)
The development of postictal pulmonary edema requires an increase in pulmonary vascular pressures. If these pressure transients are aborted by a diversion of blood from the pulmonary artery and left atrium during experimental status epilepticus, pulmonary edema does not occur. As these peripheral vascular manipulations do not alter central sympathetic output during the seizure, the studies support a hydrodynamic mechanism for postictal pulmonary edema rather than the pulmonary edema being a manifestation of increased activity of the sympathetic nervous system. 18

Capillary Permeability
Fulminant neurogenic pulmonary edema occurs in the setting of an alteration in pulmonary capillary permeability, possibly independent of 5 or in association with 19 an imbalance of the forces in the Starling equation. The classic explanation for the pathogenesis of the altered permeability is that the rapid elevation of pulmonary vascular pressures and blood flow mechanically disrupts the pulmonary capillary bed, resulting in a pulmonary capillary leak phenomenon and noncardiogenic pulmonary edema. 20 Although this explanation is likely, some studies indicate the possibility that altered capillary permeability occurs in the absence of altered intravascular pressure. 5 Other studies in animals have demonstrated an inverse correlation between maximal pulmonary vascular pressures and altered capillary permeability, suggesting that a combination of “cardiogenic” and “noncardiogenic” factors may be the most common cause of neurogenic pulmonary edema. 19 A similar conclusion has been reached from the study of patients in whom the ratio of the protein concentration of edema fluid to plasma protein concentration has been used as an index of altered capillary permeability. 21

Central Effects on Ventilation

Autonomic Dysfunction
Neural pathways subserving volitional ventilation descend from cortex through the brainstem and spinal cord in the region of the corticospinal tract. The neuronal pools subserving rhythmic involuntary ventilation originate in the caudal medulla and give rise to descending pathways in the ventrolateral brainstem and spinal cord. Accordingly, appropriately placed focal lesions may interfere with voluntary or involuntary ventilation independently.
Impairment of autonomic but not volitional ventilation produces the phenomenon of sleep apnea, or “Ondine’s curse.” This term was taken from a 1956 play by Jean Giraudoux, who recreated a German mythical legend. The sea nymph Ondine cursed the unfaithful knight Hans with the necessity of voluntary control over all of his autonomic functions: “He died, they will say, because it was a nuisance to breathe.” In the brainstem, bilateral medullary infarctions ( Fig. 1-5A ) have resulted in sleep apnea, as has unilateral medullary infarction 22 ( Fig. 1-5B ). In the latter case, the lesion depicted in Figure 1-5B will have destroyed primary ventilatory nuclei in and about the nucleus retroambigualis and the nucleus tractus solitarius as well as fibers from these nuclear groups, which descend contralaterally. Transient vertebrobasilar ischemia has also resulted in transient episodes of Ondine’s curse. 23 Congenital disorders of central alveolar hypoventilation may represent a primary defect in neural crest cell migration and function, resulting in altered central chemoreceptors. Accordingly neuroblastoma formation and Hirschsprung’s disease sometimes occur in these patients. 24 Patients with myotonic dystrophy and alveolar hypoventilation have lost catecholaminergic neurons in the medullary reticular formation. 25 Incomplete and asymmetric involvement in the region of the dorsal and ventral ventilatory complex of the medulla at about the obex has been described in two patients with multiple sclerosis who died of sleep apnea. 26

FIGURE 1-5 A, Location of bilateral brainstem infarcts in a patient with automatic respiratory failure. B, Brainstem section showing a unilateral lesion that resulted in failure of autonomic respiration.
( A from Devereaux MW, Keane JR, Davis RL: Automatic respiratory failure associated with infarction of the medulla. Arch Neurol 29:46, 1973, with permission. B from Levin BE, Margolis G: Acute failure of automatic respirations secondary to a unilateral brain stem infarct. Ann Neurol 1:583, 1977, with permission.)
Primary involvement of autonomic ventilatory nuclei was a common consequence of bulbar poliomyelitis ( Fig. 1-6 ). As with lesions of the descending pathways from these nuclear groups, these lesions led to temporary or permanent sleep apnea. There are rare reports of hypoventilation in patients with systems degeneration, 27, 28 and pathological material from such cases suggests that the causal abnormalities are located in the region of the solitary tracts in the caudal medulla. Vertebral artery dissection involving the dorsal medulla and anterior spinal artery with resultant central ventilatory failure has been reported. 29

FIGURE 1-6 Medullary lesions found in 17 patients with bulbar poliomyelitis who died of respiratory failure.
(From Baker AB, Matzke HA, Brown JR: Poliomyelitis. III: Bulbar poliomyelitis: a study of medullary function. Arch Neurol Psychiatry 63:257, 1950, with permission.)
Iatrogenic sleep apnea occurs in some patients following bilateral cervical tractotomy performed for intractable pain (6 of 112 patients reported by Tranmer and associates 30 ). Figure 1-7 shows the most common site of the cordotomy lesion and the descending autonomic pathways in the reticulospinal tract. Descending pathways for voluntary ventilation are located in the corticospinal tract and thus are distant from the lesion site ( Fig. 1-7 ).

FIGURE 1-7 Cervical spinal cord at the C1–C2 level showing the area commonly damaged in cervical cordotomies and the site of the descending autonomic pathway subserving ventilation.
(From Tranmer BI, Tucker WS, Bilbao JM: Sleep apnea following percutaneous cervical cordotomy. Can J Neurol Sci 14:262, 1987, with permission.)
Sleep apnea also occurs on an obstructive or mixed basis. Such patients are usually obese, hypertensive men older than 40 years. Excessive daytime sleepiness and sleep attacks are associated symptoms. Nocturnal breath cessation is associated with prominent snoring, snorting, and gasping sounds. Obstructive sleep apnea has been associated with neurodegenerative diseases, such as syringobulbia and olivopontocerebellar degeneration, and miscellaneous unilateral lesions of the rostrolateral medulla, 31 which may produce oropharyngeal weakness. Nonobstructive ventilatory dysfunction may occur as well. 32 Treatment with continuous positive airway pressure (CPAP) during sleep is effective. 33 Further discussion of this syndrome can be found in Chapter 32 .
Impairment of voluntary ventilatory efforts with preservation of autonomic ventilation may also occur. Cases have been reported from a demyelinating lesion in the high cervical cord and a bilateral pyramidal tract lesion in the medulla resulting from syphilitic arteritis. In another case, an infarct of the basal pons produced quadriplegia; autonomic ventilation was modulated normally by laughing, crying, and anxiety, supporting a nonpyramidal location of descending pathways from limbic structures to medullary ventilatory nuclei. 27 The most common cause, however, is a midpontine lesion that produces the “locked-in” syndrome. Patients may have a regular ventilatory pattern and a preserved response to CO 2 stimulation, or a Cheyne–Stokes pattern that is volitionally unalterable.

Extrapyramidal Disorders
Symptomatic or asymptomatic ventilatory dysfunction is an infrequently recognized but relatively common manifestation of extrapyramidal syndromes of multiple causes. Respiratory dysrhythmias were common in postencephalitic parkinsonism. 34 Tachypnea, the most common abnormality, may be episodic or continuous during sleep or wakefulness; rates as high as 100 per minute are reported. Ventilatory dysrhythmias are less common and manifest as breath-holding spells, sighing, forced or noisy expiration, inversion of the inspiration/expiration ratio, or the Cheyne–Stokes pattern. Respiratory tics occur as well, manifesting as yawning, hiccupping, spasmodic coughing, and sniffing.
In a study by Kim, all nine patients with postencephalitic parkinsonism had an increase in respiratory rate, and the normal variation in respiratory amplitude did not occur. 35 The most striking abnormality in these patients was their inability to alter the respiratory rhythm voluntarily so that, for instance, they were unable to hold their breath.
Direct fiberoptic visualization of the upper airway in patients with extrapyramidal disease (essential tremor, parkinsonian tremor, rigid parkinsonism, or dyskinesia) 36 has disclosed rhythmic or irregular glottic and supraglottic involuntary movements. Symptomatic stridor and ventilatory failure that could be reversed by endotracheal intubation were described in a number of these patients and suggested upper airway obstruction. Abnormal flow-volume curves were commonly found. Such upper airway dysfunction may be a factor in the retention of secretions and respiratory infections that occur in many patients. Alternatively, a reduction in both maximal static inspiratory and expiratory pressures precluding the ability to rapidly increase peak expiratory flow for maximally effective coughing may be an important factor. 37
Respiratory distress and dyspnea are also described in patients with extrapyramidal dysfunction in whom no cardiopulmonary cause is found, but in whom respiratory rates are irregular owing to involuntary respiratory dyskinesias that are either levodopa induced or related to a tardive dyskinesia. 38 Respiratory dyskinesias, then, may be an accompaniment of choreiform movement disorders and may account for subjective complaints of dyspnea in Parkinson’s disease and dystonia. 39

Forebrain Influences on Ventilation
That the forebrain influences both ventilatory rate and rhythm is documented by the volitional acts of overbreathing and breath-holding as well as by the coordinated semivoluntary or involuntary rhythmic alterations in ventilatory pattern that occur as part of speaking, singing, laughing, and crying. Furthermore, during sleep, normal ventilatory patterns become more irregular, total ventilatory volume decreases, Paco 2 is elevated, and the CO 2 response curve shifts to the right. Cortical “readiness potentials” originating from supplementary motor and primary motor cortex can be recorded from humans prior to volitional but not automatic inspiration or expiration. 40 Using positron emission tomography of changes in regional cerebral blood-flow, areas of cortical activation during volitional inspiration and expiration have been identified. 41 Inspiration is associated with increased cerebral blood-flow in primary motor cortex bilaterally, the right supplementary motor cortex, and left ventrolateral thalamus. In expiration, the structures implicated are similar and overlapping but extend beyond those in inspiration and include the cerebellum. In the cortex, the identified regions activated during ventilation conform to the homuncular regions of thoracic and abdominal muscles. Diaphragmatic contraction induced from these cortical regions with magnetic stimulation does not, however, affect automatic breathing. 42
Hemispheric stroke results in attenuation of diaphragmatic excursion on the hemiplegic side but only during volitional breathing; thus, the diaphragm lacks bilateral cortical representation. 43 The intercostal muscles are similarly affected by hemispheric stroke. 44 Sleep-disordered breathing is common in acute supra- and infratentorial stroke but rarely has localizing value. 45
The cortical areas effective in inducing apnea in humans are similar to those in primates ( Fig. 1-8 ) and include the anterior portion of the hippocampal gyrus, the ventral and medial surfaces of the temporal lobe, the anterior portion of the insula, and the anterior portion of the limbic gyrus. An episode of partial seizures with ictal apnea following encephalitis in humans has been studied with ictal-interictal subtraction single-photon emission computed tomography (SPECT), showing an abnormality in the left posterior lateral temporal region consistent with the ictal electroencephalographic (EEG) findings. 46 Respiratory changes have also been associated with paroxysmal abnormalities on the electroencephalogram. Such episodes have been implicated in epileptic sudden death. 47

FIGURE 1-8 Points on the anterior lateral (top) and ventromedial (bottom) cerebral cortex of Macaca mulatta where electrical stimulation elicited inhibition of respiration. C, cingulate gyrus; CC, corpus callosum; CF, central fissure; HG, hippocampal gyrus; IN, insula; LO, lateral orbital gyrus; OLF, olfactory tract; OT, optic tract; PO, posterior orbital gyrus; R, gyrus rectus; ST, superior temporal gyrus.
(From Kaada BR: Somato-motor, autonomic, and electrocorticographic responses to electrical stimulation of “rhinencephalic” and other structures in primates, cat, and dog. Acta Physiol Scand 24:1, 1951, with permission.)

Apraxia of Ventilatory Movements
The inability to take or hold a deep breath despite normal motor and sensory function is termed respiratory apraxia. This phenomenon is noted most often in elderly patients with evidence of mild or moderate cerebrovascular disease. For example, in a patient with progressive supranuclear palsy, rhythmic breathing movements persisted during planned volitional inspiration or breath-holding. 48 As cortical magnetic stimulation of primary motor cortex produces diaphragmatic contraction but does not affect ongoing nonvolitional ventilation, cortical or subcortical regions other than primary motor cortex must be the site of respiratory apraxia in such patients.

Posthyperventilation Apnea
In 1867, Hering observed that brief periods of apnea followed hyperventilation in anesthetized animals, and in 1908, Haldane reported apnea after voluntary hyperventilation in humans. 49, 50 Modern reanalysis of posthyperventilation apnea in awake normal human subjects shows that both hyperpnea and apnea of 10 to 30 seconds may occur in an individual subject; apneic pauses occur about 1 minute after cessation of hyperventilation; the apnea’s length and occurrence, although variable among subjects, was reproducible in individual subjects; and the occurrence of apnea was unrelated to the Pco 2 during hyperventilation. 51 In patients with brain injury, apnea occurred for more than 10 seconds with equal frequency in patients with unilateral (67%) and bilateral (70%) damage. 52 No correlation was found between the decrease in end-tidal CO 2 and the occurrence of apnea. A depressed level of consciousness in normal subjects, as during drowsiness, sleep, or anesthesia, also leads to posthyperventilation apnea. Posthyperventilation apnea has also been described in normal patients engaged in an intellectual task. 53

Hindbrain Control of Ventilation
The concept that the hindbrain controls ventilatory function, rate, and rhythm has grown from the experiments of Lumsden ( Fig. 1-9 ). These studies in anesthetized cats localized the brainstem ventilatory centers to regions below the inferior colliculus because transection at this level did not alter the ventilatory pattern when the vagi were intact. Transection at the medullary-cervical junction produced the cessation of all ventilatory functions. Accordingly, the neuronal centers responsible for ventilation are located between these levels. Transection at the pontomedullary junction resulted in rhythmic breathing with a gasping quality unchanged by vagal transection, demonstrating that the most primitive respiratory oscillator is located within the medulla. The higher brainstem “centers” play a modulatory role. A modern example of such experiments in anesthetized cats is found in Figure 1-10 .

FIGURE 1-9 A, The original illustration from Lumsden (1923) showing the level of “crucial sections” producing ventilatory alteration in cats. Ventilatory effects produced with lesions: 1, no alteration; 2, apneusis; between 3 and 4, uncoordinated inspiratory spasms and gasping; 4, gasping; between 5 and 6, cessation of all respiratory movements. B, Respiratory tracings from Lumsden (1923). a, normal animal; b, after vagotomy; c, apneusis (transection 2); d, gasping (transection 4).
(From Lumsden T: Observation on the respiratory centres in the cat. J Physiol [Lond] 57:153, 1923, with permission.)

FIGURE 1-10 Effects of brainstem and vagal transection on the ventilatory pattern in an unanesthetized cat. APC, apneustic center; CP, cerebellar peduncle; DRG, dorsal ventilatory group; IC, inferior colliculus; PNC, pneumotaxic center; VRG, ventral ventilatory groups. Transections at different levels are indicated by roman numerals. Tracings on right represent the tidal volume with inspiration upward.
(From Berger AJ, Mitchell RA, Severinghaus JW: Regulation of respiration. N Engl J Med 297:139, 1977, with permission.)

Classic studies of the role of the cerebellum in ventilation focused on the inhibitory effects of the anterior lobe induced by stimulation. Modern studies have extended these observations to the posterior lobe, showing stimulation-induced ventilatory inhibition from the fastigial nucleus and uvula. Stimulation of large regions of the cerebellum, however, produced no ventilatory alteration. Stimulation of the fastigial nucleus produced early termination of bursting in both the inspiratory and the expiratory medullary neurons in the cat. 54 Functional magnetic resonance imaging and positron emission tomography studies have also shown activation of the cerebellum along with other brainstem and basal forebrain structures during volitional breathing in humans. In some studies, expiration particularly involved the cerebellum. 41, 55, 56 A congenital syndrome associated with hypoplastic posterior cerebellar vermis (Joubert’s syndrome) is characterized by prominent ventilatory abnormalities: episodic hyperpnea and apnea. 57

Pneumotaxic Center
Lumsden named the pneumotaxic center ( pneumotaxy: normal rhythmic ventilation) and localized it to the rostral pons in the parabrachial complex. 58 Transection at this level results in regular breathing, and the rate of this breathing, but not the rhythm, is slowed by vagotomy. Destruction of this region or transection below it produces the phenomenon of apneusis ( Fig. 1-9B ), which is discussed in the next section. Modern electrophysiological and cytoarchitectural studies have localized respiratory-related neuronal activity to multiple nuclei in the dorsal and ventral pons and its connections. 59 Electrical stimulation within this region produces premature switching of respiratory phases. This off-switching is modified at least in part by the classic Hering–Breuer (and possibly other) afferents carried within the vagus. 60 Glycinergic and GABAergic input is critical for off-switching. 61 Neuroanatomical and neurophysiological studies in animals support the belief that the pneumotaxic center functions as a relay station, finely tuning the ventilatory pattern generator. Stimulation by glutamate injection of the parabrachial complex Kolliker–Fuse nucleus to include the margins of the sensory and motor trigeminal nuclei have identified functionally distinct cell populations producing specific but sometimes opposing ventilatory responses, which include both respiratory facilitation and inhibition. 62

Apneustic Center
The phenomenon of apneusis consists of prominent, prolonged end-inspiratory pauses that can be pro- duced by pontine transection in vagotomized animals ( Fig. 1-9 ). 58, 63 Although the phenomenon of apneusis is well recognized, anatomical definition of a neuronal aggregate that can reasonably be called the apneustic center is still lacking. Apneusis is defined operationally as a failure of activation of normal inspiratory off-switching. The phenomenon of apneusis may result from one of a number of lesions ( Fig. 1-11 ) or pharmacological manipulations. Systemic, but not local, administration of antagonists of the N -methyl-d-aspartate (NMDA) subset of the glutamate receptor, but not non-NMDA antagonism 64 induces apneusis, thus defining the neurotransmitter system involved and the lack of a specific inducing site. 65 However, altered membrane potentials in neurons of the ventral respiratory group are produced by NMDA antagonists. 66

FIGURE 1-11 Areas of the brainstem infarction in two patients with apneustic breathing.
(From Plum F, Alvord EC: Apneustic breathing in man. Arch Neurol 10:101, 1964, with permission.)
Apneustic respiration is rare in humans. Children with brainstem damage from hypoxic-ischemic injury or other brainstem lesions may have apneustic breathing, with cyanosis during inspiratory pauses. Tandospirone or buspirone, serotonin-1A agonists, normalize breathing. 67, 68 Five patients with cervicomedullary compression from achondroplasia had apneustic breathing patterns that were “reduced in the majority” following decompressive surgery. The absence of a compressive effect at the level of the pneumotaxic center and the integrity of the vagus nerves are notable in this clinical description. 69

Medullary Center
Rhythmic ventilatory excursions persist with brainstem transection at the pontomedullary level, and all ventilatory movements are abolished by transection at the medullary-cervical junction. Accordingly, attention has been focused on the medulla as the generator of rhythmic ventilatory movements. Medullary centers responsible for inspiration and expiration were identified and were held to explain both ventilatory function and ventilatory rhythmicity. Two major neuronal pools are responsible for ventilation. Primary inspiratory cells located in the ventrolateral nucleus tractus solitarius constitute the dorsal respiratory group, which receives all primary pulmonary afferents from the vagus nerves. GABA B receptors are the major modulators. 70 Inspiratory and expiratory neurons are found in a separate grouping within the nucleus ambiguus and the nucleus retroambigualis, which together constitute the ventral respiratory group ( Fig. 1-12 ). Excitatory amino acid neurotransmitter function is necessary to modulate ventral respiratory group function. NMDA receptors are the major mediators of ventral respiratory group ventilatory drive, with modulation by non-NMDA glutamate systems. 71 Thus, ventilatory rhythmicity is mediated by the dorsal respiratory group, and projection to spinal respiratory motor neurons and vagally mediated auxiliary muscles of respiration occurs via the ventral respiratory group. Although rhythmic ventilatory responses occur from the medulla following ponto-medullary transection, this respiratory pattern has a rather gasping quality and is not normal rhythmic ventilation. A gasping center has been found just rostral and ventral to the dorsal respiratory group. 72 The primary ventilatory rhythm generator appears to reside in a limited region of the ventral medulla (the pre-Bötzinger complex) just rostral to the rostral ventilatory group ( Fig. 1-12 ). Rhythm generation is eliminated by removal of this region, and medullary slices containing this region generate respiratory-related oscillations. 73 The pre-Bötzinger complex responds to hypoxia, and this response is modified by glutamate receptors. 74 The network and intrinsic membrane properties of this region are an intensive area of current investigation. 75, 76

FIGURE 1-12 A, Dorsal view of brainstem and cervical spinal cord indicating regions involved in control of breathing and progression of labeling with a viral tracer injected into the phrenic nerve. The percentage of labeled third-order neurons (propiobulbar neurons) in the pre-Bötzinger complex and adjacent regions is plotted in the set at right. Note that the pre-Bötzinger complex contains almost entirely third-order neurons, whereas adjacent regions, rVRG and BötC, contain 0 to 20 percent. BötC, Bötzinger complex; cVRG, caudal ventral respiratory group; KF, Kölliker-Fuse nucleus; NTS, nucleus tractus solitarius; PB, parabrachial nuclei; PGi, paragigantocellular reticular nucleus; preBötC, pre-Bötzinger complex; RTN, retrotrapezoid nucleus; rVRG, rostral ventral respiratory group. B, Sagittal and transverse view of the location of the pre-Bötzinger complex. cNA, caudal nucleus ambiguus; LRN, lateral reticular nucleus; rNA, rostral nucleus ambiguus; VII, facial nucleus.
(From Rekling JC, Feldman JL: PreBötzinger complex and pacemaker neurons: hypothesized site and kernel for respiratory rhythm generation. Annu Rev Physiol 60:385, 1998, with permission.)

Other Ventilatory Patterns

Cheyne–Stokes Breathing
Periodic, or Cheyne–Stokes, breathing suggests left ventricular failure or nervous system dysfunction. 77 Its original description by Cheyne was in a patient who died of heart failure, but both CNS and cardiac dysfunction (or a combination of the two) can produce this ventilatory pattern. 78
The Cheyne–Stokes pattern is that of escalating hyperventilation followed by decremental hypoventilation and finally apnea, which recurs in cycles. Cycle lengths of 40 to 100 seconds have been reported in humans. 79 Arterial blood gas assays during Cheyne–Stokes breathing indicate a rising pH and a falling Paco 2 , which become maximal at the apnea point and never return to normal values 80 ( Fig. 1-13 ). Cheyne–Stokes patterns are seen in 30 to 40 percent of patients in congestive heart failure, and Cheyne–Stokes breathing is associated with an increased mortality. 81 - 83 This ventilatory pattern also occurs in normal premature infants, during normal sleep, in subjects at high altitude, and with equal frequency in association with supratentorial and infratentorial stroke. 80, 84 Associated changes in arousal, pupillary size, 85 cardiac rhythm, heart rate, blood pressure, 86 muscle tone, and consciousness may occur cyclically in patients with Cheyne–Stokes breathing. The alterations in Paco 2 also affect the cerebral vasculature, producing changes in the intracerebral volume of the vascular compartment with associated alterations in cerebral blood-flow and intracranial pressure. The periodicity of ventilation can be eliminated by intravenous theophylline or by oxygen inhalation. 80

FIGURE 1-13 Periodicity of arterial oxygen saturation (Sao 2 ; upper trace ), chest wall motion (middle trace), and CO 2 concentration in the expired air (lower trace) in a stroke patient with Cheyne–Stokes respiration. The phase shift between the upper and middle traces is due to the sampling time of the pulse oximeter of approximately 40 seconds. The drops in CO 2 concentration during hypopnea are due to dead space ventilation.
(From Nachtmann MD, Siebler M, Rose G, et al: Cheyne-Stokes respiration in ischemic stroke. Neurology 45:820, 1995, with permission.)
Based on studies in patients with heart failure, the ventilatory oscillations result from Pco 2 fluctuations about the apneic threshold. 87 The reciprocal fall in Po 2 results from attenuated ventilatory drive. Cheyne–Stokes breathing is abolished by inhalation of CO 2 (increasing the Pco 2 over the apneic point) but not by inhalation of oxygen. 88
A host of factors that might explain Cheyne–Stokes ventilatory oscillations has been addressed experimentally and clinically. The possibility that a prolonged circulation time may itself produce ventilatory oscillations by creating a feedback loop delay to central receptors was classically considered as the factor responsible for the Cheyne–Stokes ventilatory pattern. However, Hoffman and associates, studying patients with cardiogenic pulmonary edema, found no differences in left ventricular ejection fractions in patients with or without Cheyne–Stokes breathing. 89 Hall and colleagues found that circulatory delay did correlate with Cheyne–Stokes cycle length, but not with apnea length. 90 Lorenzi-Filho and co-workers showed that CO 2 inhalation blocked Cheyne–Stokes breathing in patients with heart failure and argued that reduction in Paco 2 sensed by peripheral chemoreceptors triggered central apneas. 88 The issue of an abnormal feedback to the CNS in the genesis of respiratory oscillations was studied in animals by Cherniack 91 who used the normal phrenic nerve stimulus to trigger a mechanical ventilator that had been modified so that the gain could be varied to amplify or retard the induced tidal volume triggered by the phrenic stimulus. This model produced periodic ventilations when the gain was increased. Supporting the concept of abnormal feedback loops generating Cheyne–Stokes breathing, ventilatory periodicity was eliminated by destruction of peripheral chemoreceptors but was unchanged by vagotomy. Furthermore, all animals had a persistent respiratory alkalosis. Duplicating observed clinical phenomena, the oscillations were enhanced by hypoxia and eliminated by increasing the oxygen or CO 2 content of inspired air. Hypoxemia (during sleep) also induces Cheyne–Stokes breathing in humans. 92
Originally described as a variant of Cheyne–Stokes breathing, Biot breathing is characterized by clusters of breaths having equal and regular inspiratory and expiratory phases, rather than the spindle characteristics of Cheyne–Stokes breathing. The similarity to Cheyne–Stokes breathing is in the separation of the ventilatory periods by apnea, which in Biot breathing occurs in end-expiration. Although first described in patients with meningitis, a ganglioglioma involving the cerebellum and pons was responsible in one patient, 93 and bihemispheric infarction in another. 94

Central Hyperventilation
Hyperventilation was thought, at one point, to be the respiratory pattern characteristic of midbrain dysfunction during transtentorial herniation. 95 The exhaustion resulting from such hyperventilation may be fatal; morphine or methadone will suppress the abnormal ventilatory drive. 96
A specific midbrain localization for lesions pro-ducing this ventilatory pattern cannot be supported any longer. Cases of isolated brainstem tumors and sustained tachypnea offered the possibility of an unambiguous anatomical localization of the source of this ventilatory pattern. In some instances the pons or medulla was involved. Extra-axial medullary compression has also caused central hyperventilation. 97 Central hyperventilation has been associated with CNS lymphoma, the infiltrating nature of which has been suggested as the common feature in such cases. 98, 99 Table 1-1 shows the incidence of various abnormal ventilatory patterns associated with lesions at different CNS sites.

TABLE 1-1 Incidence of Various Abnormal Ventilatory Patterns Associated With Lesions at Different Central Nervous System Sites
The possibility of central stimulation of medullary chemoreceptors due to local lactate production from tumors or stroke has been suggested to explain the lack of correlation between anatomical lesion site and ventilatory pattern. However, a markedly alkaline cerebrospinal fluid (CSF) pH was reported in a patient with central hyperventilation and a pontine tumor. 100
Pulmonary congestion of neurogenic cause (neurogenic pulmonary edema) might induce this respiratory pattern via stimulation of pulmonary receptors in the pulmonary interstitial space. At the San Francisco General Hospital, the author saw three patients with sustained tachypnea following stroke. Their in vivo lung water content was measured with a double indicator dilution technique (by Dr. Frank Lewis), and no elevation was found.

Alveolar Hypoventilation
Hypoventilation, hypoxia, and apnea are major risks in diseases of the anterior horn cells, peripheral nerves, myoneural junctions, and muscles. Motor neuron disease, polyneuropathy, myasthenia gravis, and the muscular dystrophies are, respectively, the most common examples of such diseases that cause ventilatory disturbances. 101 In part because of the decreased exercise demands induced by the disease processes, dyspnea is often absent and arterial blood gases may show little alteration immediately prior to fatal ventilatory compromise. Furthermore, the amount of muscle weakness in extremity and girdle muscles is often a poor predictor of ventilatory muscle function. Vincken and associates examined this point and documented that maximal inspiratory (diaphragmatic, intercostal, and accessory neck muscles) and expiratory (abdominal and intercostal muscles) pressure measurements were required to assess the risk of respiratory compromise in patients with chronic neuromuscular disease. 102 Unsuspected ventilatory dysfunction was found in one half of the 30 patients studied, and in one third of patients, it was severe. In no case was the ventilatory dysfunction clinically suspected. Traumatic myelopathies or myelopathies resulting from infiltrative tumors produce ventilatory insufficiency with lesions above the cervical roots innervating the phrenic nerve (C3, C4, C5). Such patients’ ventilatory dysfunction has been successfully managed without mechanical ventilation by electrical pacing of the diaphragm. In patients with lesions between C3 and C5, this treatment is feasible if the C5 root is preserved below the level of the lesion. Each of eight patients with traumatic tetraplegia reported by Vanderlinden and co-workers were successfully weaned from ventilator support using this technique. 103 Cervical and cortical magnetic stimulation have been used to assess diaphragm strength in patients proposed for phrenic pacing. 104
Although ventilatory compromise is often the terminal event in advanced motor neuron disease, isolated respiratory insufficiency may be the presenting feature of the disease. In patients with primary bulbar disease, sleep apnea or nocturnal hypoventilation occurs, manifesting itself by both obstructive and central apnea. Orthopnea may be the presenting symptom of motor neuron disease. Such patients have predominantly diaphragmatic weakness, and this may be unilateral or bilateral. Paradoxical chest wall and abdominal movements are seen during inspiration, and vital capacity is reduced, especially when the patient is tested in the supine position. In this group of patients with diaphragmatic weakness in the absence of bulbar impairment), symptomatic relief is obtained with ventilatory support (CPAP or nocturnal intermittent positive pressure ventilation) without unwarranted prolongation of life. 105 Continuous bimodal positive airway pressure (BiPAP) is now an important alternative to tracheostomy. 105, 106
Ventilatory failure requiring mechanical assistance has been reported in 10 to 80 percent of patients with Guillain–Barré syndrome. Intubation and ventilation were needed in 43 percent of the 111 patients from the French plasmapheresis study 104 and in 47 percent of the 123 patients in the American study. 107 The mean duration of the assisted ventilation was 31 days in the French study, and it was reduced to 18 days by plasmapheresis. Intubation is usually required when vital capacity falls below 18 ml/kg. Sunderrajan and Davenport analyzed the presenting and early stages of their patients’ illness and were unable to identify any characteristics or neurological features that would predict the need for assisted ventilation. 108 While the mean hospital day on which intubation was required was 4.4, the range was broad (0 to 21 days). The hospital day on which the patient was extubated had an equally wide range: hospital days 5 to 90. Two unusual cases required ventilatory support for more than a year. This experience suggests that extubation will be successful when vital capacities exceed 1 liter. A detailed study of diaphragmatic performance in patients with Guillain–Barré syndrome suggested that improvement in the maximal transdiaphragmatic pressure was the best predictor of recovery, and this measure was correlated with maximal inspiratory force, but not forced vital capacity. 109 The duration of mechanical ventilatory support required in patients with the Guillain–Barré syndrome was nearly halved by treatment with plasma exchange; treatment with intravenous gamma globulin is equivalent. 110
An acute, primary axonal degeneration of motor and sensory fibers occurs in the setting of prolonged sepsis (approximately 2 weeks) with multiple organ failure. 111 This syndrome has been termed critical illness neuropathy and is described in detail in Chapter 52 . The neuropathy is characterized clinically by distal weakness with reduced or absent tendon reflexes; when it is severe, there is paralysis with areflexia. The syndrome is frequently recognized only because of unexpected difficulty in weaning patients from assisted ventilation. Phrenic nerve conduction velocities have been abnormal, and autopsy studies have shown axonal degeneration of the phrenic nerve, with denervation atrophy in the intercostal muscles and diaphragm. 112 Complete recovery over a period of 6 months is the rule in mild and moderate cases, but patients with severe polyneuropathy may fail to improve and have a fatal outcome. A similar syndrome of critical care myopathy is recognized, with a similar prognosis. 113, 114 Neuromuscular blockade and corticosteroid treatment may be risk factors, especially in transplant patients. 113
Temporary ventilatory support may be required in myasthenia gravis. Indications include the post-thymectomy period and failure of outpatient pharmacological therapy. Of 22 such patients seen at the Mayo Clinic, the duration of ventilatory support required was 1 to 32 days, 115 with 1 to 41 days reported by O’Donohue and colleagues. 116 Both plasma exchange and intravenous immunoglobulin treatments may be useful in myasthenic crisis. 117
In patients with myopathy, ventilatory dysfunction may occur and may be disproportionate to the severity of the muscle weakness. Although the poor prognosis in the muscular dystrophies usually commits patients to ventilatory support for the remainder of their lives, two patients with Duchenne muscular dystrophy were weaned from continual positive pressure ventilation with intermittent negative pressure techniques. 39 Recurrent episodes of ventilatory failure independent of muscle weakness have been reported in patients with mitochondrial myopathies. 118 Patients have depressed respiratory responses to hypoxia and often to hypercapnia as well. Life-threatening hypoventilation often occurs in the setting of surgery, sedation, or infection. Reported cases include typical Kearns-Sayre syndrome, MERRF (myoclonic epilepsy and ragged-red fibers) syndrome, and familial mitochondrial myopathy. No specific biochemical defect has been found, although a defect in cytochrome- c oxidase has been suggested. 118 The cause of the hypoventilation may be central rather than muscular. 119



Acute Hypoxia
The terms hypoxic and anoxic encephalopathy are frequently used to describe neurological syndromes that occur following cardiac arrest. The encephalopathy, however, is due primarily to cerebral ischemia. Acute hypoxia results in transient alterations of cognitive function similar to those due to intoxication with alcohol. Hallucinations and alterations in judgment and behavior are well known in mountain climbers at high altitudes. Climbers to the Mt. Everest summit, at 8,854 meters (29,000 feet) have been studied to determine the potential acute and long-term neurological deficits from hypoxia at these altitudes. The results of simple tests of short-term memory (number recall) and simple motor tasks (finger tapping) are shown in Figure 1-14 . Significant reductions in performance in both tests were found immediately after the expedition, and significant impairments persisted 12 months later. 120

FIGURE 1-14 Results of finger-tapping and short-term memory tests performed before, immediately after, and 1 year following an expedition to Mt. Everest.
(From West JB: Do climbs to extreme altitude cause brain damage? Lancet 2:387, 1986, with permission.)
Neuropathological studies of the CNS in primates subjected to hypoxia have revealed lesions only in the watershed distribution between major arterial territories. Thus, the effects of acute hypoxia on the brain are those of cerebral hypoperfusion. 121 Structural abnormalities do not occur in the brain in the setting of hypoxia without ischemia, 122, 123 However, polymerase chain reaction (PCR) techniques to simultaneously amplify long random segments of DNA have shown that pure hypoxia for 30 minutes in vivo produces both nuclear and mitochondrial DNA damage, which have dissimilar repair kinetics. 124


Chronic Hypercapnia
A reversible syndrome of headache, papilledema, and impaired consciousness with “tremor of the extremities” has been described in patients with chronic pulmonary insufficiency. 125 Tremulousness is most prominent with the fingers outstretched and has the characteristics of an action tremor or the features of asterixis; in some patients, it resembles myoclonus. The headaches are attributed to the increased intracranial pressure. Arterial oxygen saturations in one study 125 ranged from 81 to 94 percent (but may be as low as 40%), and the Paco 2 levels ranged from 39 to 68 mmHg (but can be higher). The electroencephalogram shows slowing in the theta or delta range. The etiology of such CO 2 narcosis is probably multifactorial, including hypercapnia, hypoxia, and elevated intracranial pressure. The increased intracranial pressure may produce papilledema that can progress to blindness. 126 Ventilatory support and discontinuation of sedative drugs constitute effective treatment.

Acute Hypercapnia
Nervous system abnormalities from hypercapnia are related in significant measure to the rate of increase of Paco 2 . The rapid diffusibility of carbon dioxide across the blood–brain barrier produces a prompt fall in CSF pH in respiratory acidosis, a decrease that does not occur in metabolic acidosis. A potent inhibitory effect of H + upon the postsynaptic receptor for glutamate, the brain’s major excitatory neurotransmitter, has been described and may be responsible for the acute encephalopathy of hypercapnia. 127


Acute Hypocapnia
Acute hypocapnia occurs during hyperventilation. The symptom complex of dizziness, lightheadedness, faintness, paresthesias, and impaired consciousness can be reproduced in normal subjects during hyperventilation, supporting a cause-and-effect relationship between acute hypocapnia and the symptoms of the hyperventilation syndrome ( Table 1-2 ), although some have found hyperventilation as a consequence, rather than a cause, of the attacks. 128 Asthma was significantly associated in one series. 129 This syndrome has its maximal incidence during the third decade. Distal paresthesias are notable and may be asymmetric, prompting evaluation for a more sinister cause. Alteration or loss of consciousness is common (31% in the series of Perkin and Joseph; Table 1-2 ), leading to an inappropriate diagnosis of epilepsy. Symptoms can often be reproduced by voluntary hyperventilation, and the electroencephalographic findings while the patient is symptomatic can help to exclude a diagnosis of seizure disorder. The effects of hypocapnia include cerebral vasoconstriction, alteration in the ionic balance of calcium, and a shift in the oxyhemoglobin dissociation curve with reduced delivery of oxygen to peripheral tissues. A combination of these events is responsible for the clinical symptoms.
TABLE 1-2 Symptoms During Attacks in 78 Patients With Hyperventilation Syndrome * Symptoms Number Percent Neurological Giddiness 46 59 Paresthesias 28 36 Loss of consciousness 24 31 Visual disturbance 22 28 Headache 17 22 Ataxia 14 18 Tremor 8 10 Tinnitus 2 3 Cardiorespiratory Dyspnea 41 53 Palpitations 33 42 Chest discomfort 6 8 Gastrointestinal Nausea 15 16 Abdominal pain 1 1 Vomiting 1 1
* Most patients had more than one symptom.
From Perkin GD, Joseph R: Neurological manifestations of the hyperventilation syndrome. J R Soc Med 79:448, 1986, with permission.

Chronic Hypocapnia
Fixed respiratory alkalosis is a common or even diagnostic finding in a number of metabolic disorders, the most prominent being hepatic encephalopathy; sepsis and salicylate poisoning are additional examples. However, the role of the alkalosis itself in causing CNS dysfunction is uncertain. Potential mechanisms by which alkalosis may affect the brain include a shift in the oxyhemoglobin dissociation curve (which decreases oxygen availability to tissues), a decrease in cerebral blood-flow resulting from cerebral vasoconstriction, and alkalosis-induced hypophosphatemia. Posner and Plum also found that control of the alkalosis by mechanical ventilation did not alter the encephalopathic manifestations in patients with hepatic failure. 130 Accordingly, alkalosis per se appears to have a minimal effect on the CNS.

Persistent or intractable hiccup is an abnormality resulting from many systemic, 131 pharmacological, 132 and CNS causes, including brainstem neoplasm, 133 multiple sclerosis, 134 and thoracic herpes zoster. 135 It may also occur with cortical pathology. 136, 137 Cases of intractable hiccup are much more common in men than women. 138 Hiccup results from CNS-induced synchronous contraction of the diaphragm and the external (inspiratory) intercostal muscles, followed rapidly by inhibition of expiratory intercostal muscles and glottal closure. 139 The glottal closure minimizes air exchange. However, with tracheostomy, the induced ventilatory movements of hiccup cause air exchange, and a respiratory alkalosis is produced. 139, 140
The frequency, but not amplitude, of hiccuping is modulated by arterial Paco 2 . Hiccup frequency is reduced with elevated Paco 2 levels and increased with a fall in arterial Paco 2 . 139 This observation is in keeping with the traditional cure for hiccups—breath-holding. Another common lay remedy for hiccup is swallowing or pharyngeal stimulation, maneuvers that may increase vagal tone. Thus, hiccups are most common at maximal inspiration because vagal afferents are inhibited by maximal lung inflation. High-frequency diaphragmatic flutter, responsive to carbamazepine, is responsible for hiccups in some patients. 137 Chlorpromazine remains the most popular pharmacological treatment, although baclofen and gabapentin are now also popular 141 - 143 ; a host of other approaches has been suggested. 144

The coordinated act of sneezing arises from a caudal brainstem center near the nucleus ambiguus. 145 A medullary mass lesion or lateral medullary syndrome can prevent sneezing despite the urge to do so. 146 - 148
Cortical input to sneezing has long been recognized. The central mediation of sneezing was noted by Penfield and Jasper 149 in a patient during temporal lobe stimulation when both sneezing and chewing movements were induced. A common reflex that induces sneezing is that which occurs on sudden exposure to bright light. 150 This reflex was found in 80 percent of the families of medical students in whom the phenomenon of light-induced sneezing was reported. It has been suggested that this reflex is inherited in an autosomal dominant manner. 151

Yawning is coordinated from brainstem sites near the paraventricular nucleus 152 via extrapyramidal pathways using a number of neurotransmitters and neuropeptides. 153 This reflex may occur in patients “locked in” from pontine transection who have nonvolitional mouth opening with spontaneous yawns. 154 Yawning in the setting of a pyramidal lesion (capsular infarction) may be associated with arm stretching in the paretic limb, supporting the involvement of extrapyramidal circuitry. 155 Cortical input also occurs, as reflected by the yawning related to boredom and somnolence. 156 Yawning has been seen to initiate a temporal lobe seizure. 149


1 Hux W. A new view of Starling’s hypothesis at the microstructural level. Microvasc Res . 1999;58:281.
2 Simon RP. Neurogenic pulmonary edema. Neurol Clin . 1993;11:309.
3 Rogers FB, Shackford SR, Trevisani GT, et al. Neurogenic pulmonary edema in fatal and nonfatal head injuries. J Trauma . 1995;39:860.
4 Simon RP, Bayne LL. Pulmonary lymphatic flow alterations during intracranial hypertension in sheep. Ann Neurol . 1984;15:188.
5 McClellan MD, Dauber IM, Weil JV. Elevated intracranial pressure increases pulmonary vascular permeability to protein. J Appl Physiol . 1989;67:1185.
6 Maron MB, Holcomb PH, Dawson CA, et al. Edema development and recovery in neurogenic pulmonary edema. J Appl Physiol . 1994;77:1155.
7 Bosso FJ, Lang SA, Maron MB. Role of hemodynamics and vagus nerves in development of fibrin-induced pulmonary edema. J Appl Physiol . 1990;69:2227.
8 Lane SM, Maender KC, Awender NE, et al. Adrenal epinephrine increases alveolar liquid clearance in a canine model of neurogenic pulmonary edema. Am J Respir Crit Care Med . 1998;158:760.
9 Brown RHJr, Beyerl BD, Iseke R, et al. Medulla oblongata edema associated with neurogenic pulmonary edema. Case report. J Neurosurg . 1986;64:494.
10 Keegan MT, Lanier WL. Pulmonary edema after resection of a fourth ventricle tumor: possible evidence for a medulla-mediated mechanism. Mayo Clin Proc . 1999;74:264.
11 Inobe JJ, Mori T, Ueyama H, et al. Neurogenic pulmonary edema induced by primary medullary hemorrhage: a case report. J Neurol Sci . 2000;172:73.
12 Darragh TM, Simon RP. Nucleus tractus solitarius lesions elevate pulmonary arterial pressure and lymph flow. Ann Neurol . 1985;17:565.
13 Simon RP, Gean-Marton AD, Sander JE. Medullary lesion inducing pulmonary edema: a magnetic resonance imaging study. Ann Neurol . 1991;30:727.
14 Crawley F, Saddeh I, Barker S, et al. Acute pulmonary oedema: presenting symptom of multiple sclerosis. Mult Scler . 2001;7:71.
15 Nolan MA, Craig ME, Lahra MM, et al. Survival after pulmonary edema due to enterovirus 71 encephalitis. Neurology . 2003;60:1651.
16 Bayne LL, Simon RP. Systemic and pulmonary vascular pressures during generalized seizures in sheep. Ann Neurol . 1981;10:566.
17 Simon RP, Bayne LL, Tranbaugh RF, et al. Elevated pulmonary lymph flow and protein content during status epilepticus in sheep. J Appl Physiol . 1982;52:91.
18 Johnston SC, Darragh TM, Simon RP. Postictal pulmonary edema requires pulmonary vascular pressure increases. Epilepsia . 1996;37:428.
19 Maron MB. Analysis of airway fluid protein concentration in neurogenic pulmonary edema. J Appl Physiol . 1987;62:470.
20 West JB, Mathieu-Costello O. Structure, strength, failure, and remodeling of the pulmonary blood-gas barrier. Annu Rev Physiol . 1999;61:543.
21 Smith WS, Matthay MA. Evidence for a hydrostatic mechanism in human neurogenic pulmonary edema. Chest . 1997;111:1326.
22 Bogousslavsky J, Khurana R, Deruaz JP, et al. Respiratory failure and unilateral caudal brainstem infarction. Ann Neurol . 1990;28:668.
23 Kraus J, Heckmann JG, Druschky A, et al. Ondine’s curse in association with diabetes insipidus following transient vertebrobasilar ischemia. Clin Neurol Neurosurg . 1999;101:196.
24 Strauser LM, Helikson MA, Tobias JD. Anesthetic care for the child with congenital central alveolar hypoventilation syndrome (Ondine’s curse). J Clin Anesth . 1999;11:431.
25 Ono S, Takahashi K, Jinnai K, et al. Loss of catecholaminergic neurons in the medullary reticular formation in myotonic dystrophy. Neurology . 1998;51:1121.
26 Auer RN, Rowlands CG, Perry SF, et al. Multiple sclerosis with medullary plaques and fatal sleep apnea (Ondine’s curse). Clin Neuropathol . 1996;15:101.
27 Munschauer FE, Loh L, Bannister R, et al. Abnormal respiration and sudden death during sleep in multiple system atrophy with autonomic failure. Neurology . 1990;40:677.
28 Glass GA, Josephs KA, Ahlskog JE. Respiratory insufficiency as the primary presenting symptom of multiple-system atrophy. Arch Neurol . 2006;63:978.
29 Lanczik O, Szabo K, Lecei O, et al. Central respiratory dysfunction following vertebral artery dissection. Neurology . 2006;66:944.
30 Tranmer BI, Tucker WS, Bilbao JM. Sleep apnea following percutaneous cervical cordotomy. Can J Neurol Sci . 1987;14:262.
31 Morrell MJ, Heywood P, Moosavi SH, et al. Unilateral focal lesions in the rostrolateral medulla influence chemosensitivity and breathing measured during wakefulness, sleep, and exercise. J Neurol Neurosurg Psychiatry . 1999;67:637.
32 Nogues M, Gene R, Benarroch E, et al. Respiratory disturbances during sleep in syringomyelia and syringobulbia. Neurology . 1999;52:1777.
33 Bradley TD, Shapiro CM. ABC of sleep disorders. Unexpected presentations of sleep apnoea: use of CPAP in treatment. BMJ . 1993;306:1260.
34 Turner WA, Critchley M. Respiratory disorders in epidemic encephalitis. Brain . 1925;48:72.
35 Kim R. The chronic residual respiratory disorder in post-encephalitis parkinsonism. J Neurol Neurosurg Psychiatry . 1968;31:393.
36 Vincken WG, Gauthier SG, Dollfuss RE, et al. Involvement of upper-airway muscles in extrapyramidal disorders. A cause of airflow limitation. N Engl J Med . 1984;311:438.
37 Boggard JM, Hovestadt A, Meervaldt J, et al. Maximum expiratory and inspiratory flow-volume curves in Parkinson’s disease. Am Rev Respir Dis . 1989;39:610.
38 Zupnick HM, Brown LK, Miller A, et al. Respiratory dysfunction due to L-dopa therapy for parkinsonism: diagnosis using serial pulmonary function tests and respiratory inductive plethysmography. Am J Med . 1990;89:109.
39 Braun N, Abd A, Baer J, et al. Dyspnea in dystonia. A functional evaluation. Chest . 1995;107:1309.
40 Macefield G, Gandevia SC. The cortical drive to human respiratory muscles in the awake state assessed by premotor cerebral potentials. J Physiol . 1991;439:545.
41 Ramsay SC, Adams L, Murphy K, et al. Regional cerebral blood flow during volitional expiration in man: a comparison with volitional inspiration. J Physiol . 1993;461:85.
42 Corfield DR, Murphy K, Guz A. Does the motor cortical control of the diaphragm ‘bypass’ the brain stem respiratory centres in man? Respir Physiol . 1998;114:109.
43 Cohen E, Mier A, Heywood P, et al. Diaphragmatic movement in hemiplegic patients measured by ultrasonography. Thorax . 1994;49:890.
44 Przedborski S, Brunko E, Hubert M, et al. The effect of acute hemiplegia on intercostal muscle activity. Neurology . 1998;38:1882.
45 Bassetti C, Aldrich MS, Quint D. Sleep-disordered breathing in patients with acute supra- and infratentorial strokes. A prospective study of 39 patients. Stroke . 1997;28:1765.
46 Lee HW, Hoang SB, Tae WA, et al. Partial seizures manifesting as apnea only in an adult. Epilepsia . 1991;40:1828.
47 Nashef L, Walker F, Allen P, et al. Apnoea and bradycardia during epileptic seizures: relation to sudden death in epilepsy. J Neurol Neurosurg Psychiatry . 1996;60:297.
48 Haouzi P, Chenuel B, Barroche G. Interactions between volitional and automatic breathing during respiratory apraxia. Respir Physiol Neurobiol . 2006;152:169.
49 Plum F, Brown HW, Snoep E. Neurologic significance of post-hyperventilation apnea. JAMA . 1962;181:1050.
50 Haldane JS, Poulton EP. The effects of want of oxygen on respiration. J Physiol . 1908;37:390.
51 Meah MS, Gardner WN. Post-hyperventilation apnoea in conscious humans. J Physiol . 1994;477:527.
52 Jennett S, Ashbridge K, North JB. Post-hyperventilation apnoea in patients with brain damage. J Neurol Neurosurg Psychiatry . 1974;37:288.
53 Chin K, Ohi M, Fukui M, et al. Inhibitory effect of an intellectual task on breathing after voluntary hyperventilation. J Appl Physiol . 1996;81:1379.
54 Xu F, Frazier DT. Medullary respiratory neuronal activity modulated by stimulation of the fastigial nucleus of the cerebellum. Brain Res . 1995;705:53.
55 Colebatch JG, Adams L, Murphy K, et al. Regional cerebral blood flow during volitional breathing in man. J Physiol (Lond) . 1991;443:91.
56 Gozal D, Omidvar O, Kirlew KA, et al. Functional magnetic resonance imaging reveals brain regions mediating the response to resistive expiratory loads in humans. J Clin Invest . 1996;97:47.
57 Maria BL, Hoang KB, Tusa RJ, et al. “Joubert syndrome” revisited: key ocular motor signs with magnetic resonance imaging correlation. J Child Neurol . 1997;12:423.
58 Lumsden T. Observation on the respiratory centres in the cat. J Physiol (Lond) . 1923;57:153.
59 Song G, Yu Y, Poon CS. Cytoarchitecture of pneumotaxic integration of respiratory and nonrespiratory information in the rat. J Neurosci . 2006;26:300.
60 St. John WM, Zhou D. Rostral pontile mechanisms regulate durations of expiratory phases. J Appl Physiol . 1991;71:2133.
61 Schmid K, Foutz AS, Denavit-Saubie M. Inhibitions mediated by glycine and GABAA receptors shape the discharge pattern of bulbar respiratory neurons. Brain Res . 1996;710:150.
62 Chamberlin NL, Saper CB. Topographic organization of respiratory responses to glutamate microstimulation of the parabrachial nucleus in the rat. J Neurosci . 1994;14:6500.
63 Wang W, Fung ML, St. John WM. Pontile regulation of ventilatory activity in the adult rat. J Appl Physiol . 1993;74:2801.
64 Fung ML, Wang W, St. John WM. Involvement of pontile NMDA receptors in inspiratory termination in rat. Respir Physiol . 1994;96:177.
65 Haji A, Okazaki M, Yamazaki H, et al. NMDA receptor-mediated inspiratory off-switching in pneumotaxic-disconnected cats. Neurosci Res . 1998;32:323.
66 Haji A, Pierrefiche O, Takeda R, et al. Membrane potentials of respiratory neurones during dizocilpine-induced apneusis in adult cats. J Physiol (Lond) . 1996;495:851.
67 Wilken B, Lalley P, Bischoff AM, et al. Treatment of apneustic respiratory disturbance with a serotonin-receptor agonist. J Pediatr . 1997;130:89.
68 Saito Y, Hashimoto T, Iwata H, et al. Apneustic breathing in children with brainstem damage due to hypoxic-ischemic encephalopathy. Dev Med Child Neurol . 1999;41:560.
69 Mador MJ, Tobin MJ. Apneustic breathing. A characteristic feature of brainstem compression in achondroplasia? Chest . 1990;97:877.
70 Wagner PG, Dekin MS. GABAb receptors are coupled to a barium-insensitive outward rectifying potassium conductance in premotor respiratory neurons. J Neurophysiol . 1993;69:286.
71 Abrahams TP, Hornby PJ, Walton DP, et al. An excitatory amino acid(s) in the ventrolateral medulla is (are) required for breathing to occur in the anesthetized cat. J Pharmacol Exp Ther . 1991;259:1388.
72 St. John WM. Neurogenesis, control, and functional significance of gasping. J Appl Physiol . 1990;68:1305.
73 Smith JC, Ellenberger HH, Ballanyi K, et al. Pre-Botzinger complex: a brainstem region that may generate respiratory rhythm in mammals. Science . 1991;254:726.
74 Solomon IC. Glutamate neurotransmission is not required for, but may modulate, hypoxic sensitivity of pre-Botzinger complex in vivo. J Neurophysiol . 2005;93:1278.
75 Ramirez JM, Richter DW. The neuronal mechanisms of respiratory rhythm generation. Curr Opin Neurobiol . 1996;6:817.
76 Koshiya N, Smith JC. Neuronal pacemaker for breathing visualized in vitro. Nature . 1999;400:360.
77 Naughton MT. Pathophysiology and treatment of Cheyne-Stokes respiration. Thorax . 1998;53:514.
78 Nopmaneejumruslers C, Kaneko Y, Hajek V, et al. Cheyne-Stokes respiration in stroke: relationship to hypocapnia and occult cardiac dysfunction. Am J Respir Crit Care Med . 2005;171:1048.
79 Lange RI, Hecht HH. The mechanism of Cheyne-Stokes respiration. J Clin Invest . 1962;41:42.
80 Nachtmann A, Siebler M, Rose G, et al. Cheyne-Stokes respiration in ischemic stroke. Neurology . 1995;45:820.
81 Ponikowski P, Anker SD, Chua TP, et al. Oscillatory breathing patterns during wakefulness in patients with chronic heart failure: clinical implications and role of augmented peripheral chemosensitivity. Circulation . 1999;100:2418.
82 Hanly PJ, Zuberi-Khokhar NS. Increased mortality associated with Cheyne-Stokes respiration in patients with congestive heart failure. Am J Respir Crit Care Med . 1996;153:272.
83 Leung RS, Floras JS, Bradley TD. Respiratory modulation of the autonomic nervous system during Cheyne-Stokes respiration. Can J Physiol Pharmacol . 2006;84:61.
84 Khoo MC, Anholm JD, Ko SW, et al. Dynamics of periodic breathing and arousal during sleep at extreme altitude. Respir Physiol . 1996;103:33.
85 Gonyea EF. The abnormal pupil in Cheyne-Stokes respiration. Case report. J Neurosurg . 1990;72:810.
86 Leung RS, Bradley TD. Respiratory modulation of heart rate and blood pressure during Cheyne-Stokes respiration. J Electrocardiol . 2003;36(suppl):213.
87 Hanly P, Zuberi N, Gray R. Pathogenesis of Cheyne-Stokes respiration in patients with congestive heart failure. Relationship to arterial PCO2. Chest . 1993;104:1079.
88 Lorenzi-Filho G, Rankin F, Bies I, et al. Effects of inhaled carbon dioxide and oxygen on Cheyne-Stokes respiration in patients with heart failure. Am J Respir Crit Care Med . 1999;159:1490.
89 Hoffman R, Agatston A, Krieger B. Cheyne-Stokes respiration in patients recovering from acute cardiogenic pulmonary edema. Chest . 1990;97:410.
90 Hall MJ, Xie A, Rutherford R, et al. Cycle length of periodic breathing in patients with and without heart failure. Am J Respir Crit Care Med . 1996;154:376.
91 Cherniack NS, Longobardo G, Evangelista CJ. Causes of Cheyne-Stokes respiration. Neurocrit Care . 2005;3:271.
92 Cherniack NS. Apnea and periodic breathing during sleep. N Engl J Med . 1999;341:985.
93 Kuna ST, Smickley JS, Murchison LC. Hypercarbic periodic breathing during sleep in a child with a central nervous system tumor. Am Rev Respir Dis . 1990;142:880.
94 Freeman WD, Sen S, Roy TK, et al. Cluster breathing associated with bihemispheric infarction and sparing of the brainstem. Arch Neurol . 2006;63:1487.
95 Plum F, Posner JB. The Diagnosis of Stupor and Coma, 3rd Ed. Philadelphia: FA Davis, 1980.
96 Jaeckle KA, Digre KB, Jones CR, et al. Central neurogenic hyperventilation: pharmacologic intervention with morphine sulfate and correlative analysis of respiratory, sleep, and ocular motor dysfunction. Neurology . 1990;40:1715.
97 Dubaybo BA, Afridi I, Hussain M. Central neurogenic hyperventilation in invasive laryngeal carcinoma. Chest . 1991;99:767.
98 Shibata Y, Meguro K, Narushima K, et al. Malignant lymphoma of the central nervous system presenting with central neurogenic hyperventilation. Case report. J Neurosurg . 1992;76:696.
99 Tarulli AW, Lim C, Bui JD, et al. Central neurogenic hyperventilation: a case report and discussion of pathophysiology. Arch Neurol . 2005;62:1632.
100 Siderowf AD, Balcer LJ, Kenyon LC, et al. Central neurogenic hyperventilation in an awake patient with a pontine glioma. Neurology . 1996;46:1160.
101 Sivak ED, Shefner JM, Sexton J. Neuromuscular disease and hypoventilation. Curr Opin Pulm Med . 1999;5:355.
102 Vincken W, Elleker MG, Cosio MG. Determinants of respiratory muscle weakness in stable chronic neuromuscular disorders. Am J Med . 1987;82:53.
103 Vanderlinden RG, Epstein SW, Hyland RH, et al. Management of chronic ventilatory insufficiency with electrical diaphragm pacing. Can J Neurol Sci . 1988;15:63.
104 Efficiency of plasma exchange in Guillain-Barré syndrome: role of replacement fluids. French Cooperative Group on Plasma Exchange in Guillain-Barré syndrome. Ann Neurol . 1987;22:753.
105 Lo Coco D, Marchese S, Pesco MC, et al. Noninvasive positive-pressure ventilation in ALS: predictors of tolerance and survival. Neurology . 2006;67:761.
106 Aboussouan LS, Khan SU, Meeker DP, et al. Effect of noninvasive positive-pressure ventilation on survival in amyotrophic lateral sclerosis. Ann Intern Med . 1997;127:450.
107 Plasmapheresis and acute Guillain-Barré syndrome. The Guillain-Barré Syndrome Study Group. Neurology . 1985;35:1096.
108 Sunderrajan EV, Davenport J. The Guillain-Barré syndrome: pulmonary-neurologic correlations. Medicine (Baltimore) . 1985;64:333.
109 Borel CO, Tilford C, Nichols DG, et al. Diaphragmatic performance during recovery from acute ventilatory failure in Guillain-Barré syndrome and myasthenia gravis. Chest . 1991;99:444.
110 Randomised trial of plasma exchange, intravenous immunoglobulin, and combined treatments in Guillain-Barré syndrome. Plasma Exchange/Sandoglobulin Guillain-Barré Syndrome Trial Group. Lancet . 1997;349:225.
111 Zochodne DW, Bolton CF, Wells GA, et al. Critical illness polyneuropathy. A complication of sepsis and multiple organ failure. Brain . 1987;110:819.
112 Witt NJ, Zochodne DW, Bolton CF, et al. Peripheral nerve function in sepsis and multiple organ failure. Chest . 1991;99:176.
113 Lacomis D, Giuliani MJ, Van Cott A, et al. Acute myopathy of intensive care: clinical, electromyographic, and pathological aspects. Ann Neurol . 1996;40:645.
114 Lacomis D, Petrella JT, Giuliani MJ. Causes of neuromuscular weakness in the intensive care unit: a study of ninety-two patients. Muscle Nerve . 1998;21:610.
115 Gracey DR, Divertie MB, Howard FMJr. Mechanical ventilation for respiratory failure in myasthenia gravis. Two-year experience with 22 patients. Mayo Clin Proc . 1983;58:597.
116 O’Donohue WJJr, Baker JP, Bell GM, et al. Respiratory failure in neuromuscular disease. Management in a respiratory intensive care unit. JAMA . 1976;235:733.
117 Qureshi AI, Choudhry MA, Akbar MS, et al. Plasma exchange versus intravenous immunoglobulin treatment in myasthenic crisis. Neurology . 1999;52:629.
118 Barohn RJ, Clanton T, Sahenk Z, et al. Recurrent respiratory insufficiency and depressed ventilatory drive complicating mitochondrial myopathies. Neurology . 1990;40:103.
119 Ono S, Takahashi K, Kanda F, et al. Decrease of neurons in the medullary arcuate nucleus in myotonic dystrophy. Acta Neuropathol (Berl) . 2001;102:89.
120 West JB. Do climbs to extreme altitude cause brain damage? Lancet . 1986;2:387.
121 Graham DI. Hypoxia and vascular disorders. In: Adams JH, Duchen LW, editors. Greenfield’s Neuropathology . 5th Ed. New York: Oxford University Press; 1992:153.
122 Pearigen P, Gwinn R, Simon RP. The effects in vivo of hypoxia on brain injury. Brain Res . 1996;725:184.
123 Miyamoto O, Auer RN. Hypoxia, hyperoxia, ischemia, and brain necrosis. Neurology . 2000;54:362.
124 Englander EW, Greeley GHJr, Wang G, et al. Hypoxia-induced mitochondrial and nuclear DNA damage in the rat brain. J Neurosci Res . 1999;58:262.
125 Austen FK, Carmichael MW, Adams RD. Neurologic manifestations of chronic pulmonary insufficiency. N Engl J Med . 1957;257:579.
126 O’Halloran HS, Berger JR, Baker RS, et al. Optic nerve edema as a consequence of respiratory disease. Neurology . 1999;53:2204.
127 Tang CM, Dichter M, Morad M. Modulation of the N-methyl-D-aspartate channel by extracellular H + . Proc Natl Acad Sci U S A . 1990;87:6445.
128 Hornsveld HK, Garssen B, Dop MJ, et al. Double-blind placebo-controlled study of the hyperventilation provocation test and the validity of the hyperventilation syndrome. Lancet . 1996;348:154.
129 Saisch SG, Wessely S, Gardner WN. Patients with acute hyperventilation presenting to an inner-city emergency department. Chest . 1996;110:952.
130 Posner JB, Plum F. Toxic effects of carbon dioxide and acetazolamide in hepatic encephalopathy. J Clin Invest . 1980;39:1246.
131 Launois S, Bizec JL, Whitelaw WA, et al. Hiccup in adults: an overview. Eur Respir J . 1993;6:563.
132 Lauterbach EC. Hiccup and apparent myoclonus after hydrocodone: review of the opiate-related hiccup and myoclonus literature. Clin Neuropharmacol . 1999;22:87.
133 Nagayama T, Kaji M, Hirano H, et al. Intractable hiccups as a presenting symptom of cerebellar hemangioblastoma. Case report. J Neurosurg . 2004;100:1107.
134 Funakawa I, Terao A. Intractable hiccups and syncope in multiple sclerosis. Acta Neurol Scand . 1998;98:136.
135 Berlin AL, Muhn CY, Billick RC. Hiccups, eructation, and other uncommon prodromal manifestations of herpes zoster. J Am Acad Dermatol . 2003;49:1121.
136 Jansen PHP, Joosten EMG, Vingerhoets HM. Persistent periodic hiccups following brain abscess: a case report. J Neurol Sci . 1991;103:144.
137 Vantrappen G, Decramer M, Harlet R. High-frequency diaphragmatic flutter: symptoms and treatment by carbamazepine. Lancet . 1992;339:265.
138 Liaw CC, Wang CH, Chang HK, et al. Gender discrepancy observed between chemotherapy-induced emesis and hiccups. Support Care Cancer . 2001;9:435.
139 Newsom-Davis J. An experimental study of hiccup. Brain . 1970;93:851.
140 Campbell LA, Schwartz SH. An unusual cause of respiratory alkalosis. Chest . 1991;100:1159.
141 Friedman NL. Hiccups: a treatment review. Pharmacotherapy . 1996;16:986.
142 Petroianu G, Hein G, Petroianu A, et al. Idio-pathic chronic hiccup: combination therapy with cisapride, omeprazole, and baclofen. Clin Ther . 1997;19:1031.
143 Hernandez JL, Pajaron M, Garcia-Regata O, et al. Gabapentin for intractable hiccup. Am J Med . 2004;117:79.
144 Kolodzik PW, Eilers MA. Hiccups (singultus): review and approach to management. Ann Emerg Med . 1991;20:565.
145 Nonaka S, Unno T, Ohta Y, et al. Sneeze-evoking region within the brainstem. Brain Res . 1990;511:265.
146 Martin RA, Handel SF, Aldama AE. Inability to sneeze as a manifestation of medullary neoplasm. Neurology . 1991;41:1675.
147 Hersch M. Loss of ability to sneeze in lateral medullary syndrome. Neurology . 2000;54:520.
148 Seijo-Martinez M, Varela-Freijanes A, Grandes J, et al. Sneeze related area in the medulla: localisation of the human sneezing centre? J Neurol Neurosurg Psychiatry . 2006;77:559.
149 Penfield W, Jasper H. Epilepsy and the Functional Anatomy of the Human Brain. Boston: Little, Brown, 1954.
150 Whitman BW, Packer RJ. The photic sneeze reflex: literature review and discussion. Neurology . 1993;43:868.
151 Peroutka SJ, Peroutka LA. Autosomal dominant transmission of the “photic sneeze reflex”. N Engl J Med . 1984;310:599.
152 Sato-Suzuki I, Kita I, Oguri M, et al. Stereotyped yawning responses induced by electrical and chemical stimulation of paraventricular nucleus of the rat. J Neurophysiol . 1998;80:2765.
153 Argiolas A, Melis MR. The neuropharmacology of yawning. Eur J Pharmacol . 1998;343:1.
154 Krasnianski M, Gaul C, Neudecker S, et al. Yawning despite trismus in a patient with locked-in syndrome caused by a thrombosed megadolichobasilar artery. Clin Neurol Neurosurg . 2003;106:44.
155 Wimalaratna HS, Capildeo R. Is yawning a brainstem phenomenon? Lancet . 1988;1:300.
156 Perriol MP, Monaca C. “One person yawning sets off everyone else”. J Neurol Neurosurg Psychiatry . 2006;77:3.
Chapter 2 Neurological Complications of Aortic Disease and Surgery

Douglas S. Goodin

Spinal Cord Ischemia
Ischemic Cord Syndromes
Cerebral Ischemia
Strokes and Transient Ischemic Attacks
Peripheral Neuropathy
Autonomic Neuropathies
Syphilitic Aortitis
Takayasu’s Arteritis
Giant Cell Arteritis
Aortic Aneurysms
Nondissecting Aneurysms
Dissecting Aortic Aneurysms
Traumatic Aortic Aneurysm
Coarctation of the Aorta
Surgery and Other Procedures
Aortic Surgery
Aortography and Other Procedures on the Aorta
Intraoperative Adjuncts to Avoid Spinal Cord Ischemia
The aorta is the main conduit through which the heart supplies blood to the body, including the brain, brainstem, and spinal cord. In addition, this vessel is situated close to important neural structures. In consequence, both disease of the aorta and operations on it may have profound but variable effects on nervous system function. Often the neurological syndrome produced by aortic disease or surgery depends more on the part of the aorta involved than on the nature of the pathological process itself. For example, either syphilis or atherosclerosis may produce symptoms of cerebral ischemia if the disease affects the aortic arch or of spinal cord ischemia if the pathological process is in the descending thoracic aorta. Even when the nature of the pathological process is important in determining the resultant neurological syndrome, several diseases may result in the same pathological process. Thus, atherosclerosis, infection, inflammation, and trauma may each result in the formation of aortic aneurysms; similarly, coarctation of the aorta may be congenital, a result of Takayasu’s arteritis, or a sequela of radiation exposure during childhood.
The initial focus of this chapter is on the three major areas of neurological dysfunction resulting from aortic disease and surgery: spinal cord ischemia, cerebral ischemia, and peripheral neuropathy. Specific conditions that merit special consideration are then discussed individually. The normal anatomical relationships are also considered in order to provide insight into the pathogenesis of the resulting neurological syndromes.

Aortic disease may produce a variety of neurological syndromes. The specific syndrome depends to a large extent on the site of involvement along the aorta.

Spinal Cord Ischemia


Embryological Development
During embryological development, primitive blood vessels arise along the spinal nerve roots bilaterally and at each segmental level. Each of these segmental vessels then divides into anterior and posterior branches, which ramify extensively on the surfaces of the developing spinal cord. As development proceeds, most of these vessels regress and a few enlarge, so that by birth, the blood supply to the spinal cord depends on a small but highly variable number of persisting segmental vessels 1 - 11 ( Fig. 2-1 ). In the thoracic region, where the aorta is situated to the left of the midline, the persisting vessels entering the spinal canal are those from the left in 70 to 80 percent of cases. 5, 6, 8

FIGURE 2-1 Extraspinal contributions to the anterior spinal arteries showing the three arterial territories. In the cervical region, an average of three arteries (derived from the vertebral arteries and the costocervical trunk) supply the anterior spinal artery. The anterior spinal artery is narrowest in the midthoracic region, often being difficult to distinguish from other small arteries on the anterior surface of the cord; occasionally it is discontinuous with the anterior spinal artery above and below. In addition, this region is often supplied by only a single small radiculomedullary vessel. The lumbosacral territory is supplied by a single large artery, the great anterior medullary artery of Adamkiewicz, which turns abruptly caudal after joining the anterior spinal artery. If it gives off an ascending branch, that branch is usually a much smaller vessel. This artery is usually the most caudal of the anterior radiculomedullary arteries, but when it follows a relatively high thoracic root, there is often a small lumbar radiculomedullary artery below. In this and subsequent illustrations, a indicates artery; m, muscle; n, nerve.

Anterior Spinal Artery
The anterior spinal artery is formed rostrally from paired branches of the intracranial vertebral arteries that descend from the level of the medulla ( Fig. 2-1 ). These two arteries fuse to form a single anterior spinal artery that overlies the anterior longitudinal fissure of the spinal cord. 1 This artery is joined at different levels by anterior radiculomedullary arteries, which are branches of certain segmental vessels ( Fig. 2-2 ). The number of these vessels is variable among individuals, ranging from 2 to 17, although 85 percent of individuals have between 4 and 7. 5, 6

FIGURE 2-2 Anatomy of the spinal cord circulation, showing the relationship of the segmental arteries and their branches to the spinal canal and cord. The left rib and the left pedicle of the vertebra have been cut away to show the underlying vascular and neural structures.
The anterior spinal artery in the region that includes the cervical enlargement (C1 to T3) is particularly well supplied, receiving contributions from an average of three segmental vessels. 6 One constant artery arises from the costocervical trunk and supplies the lower segments; the others arise from the extracranial vertebral arteries and supply the middle cervical segments. 6 In addition, branches of the vertebral arteries have rich anastomotic connections with other neck vessels, including the occipital artery, deep cervical artery, and ascending cervical artery. 6
The anterior spinal artery in the midthoracic portion of the cord (T4 to T8) often receives only a single contribution from a small artery located at about T7, most often on the left. 6, 8 The anterior spinal artery has its smallest diameter in this region, and it is sometimes discontinuous with the vessel in more rostral or caudal regions. 5, 6
The anterior spinal artery in the region of the lumbar enlargement (T9 to the conus medullaris) is, as at the cervical enlargement, richly supplied, deriving its blood supply predominantly from a single large (1.0 to 1.3 mm in diameter) artery, the great anterior medullary artery of Adamkiewicz. This artery almost always accompanies a nerve root between T9 and L2, usually on the left, although rarely it may accompany a root above or below these levels. 5, 6 Identification of the actual location of this great vessel has become an important part of the planning and execution of operations on the aorta such as repair of thoracoabdominal aortic aneurysms. Although digital subtraction angiography has been used for this purpose, the use of contrast-enhanced magnetic resonance angiography has recently been proposed to offer a noninvasive alternative. 11 Caudally, at the conus medullaris, the anterior spinal artery anastomoses with both posterior spinal arteries. 6

Posterior Spinal Arteries
The paired posterior spinal arteries are formed rostrally from the intracranial portion of the vertebral arteries. They are distinct paired vessels only at their origin, however, and thereafter become intermixed with an anastomotic posterior pial arterial plexus 5, 6 ( Fig. 2-3 ). This plexus is joined at different levels by a variable number (10 to 23) of posterior radiculomedullary vessels that accompany the posterior nerve roots. 5

FIGURE 2-3 Vascular anatomy of the spinal cord. The anterior spinal artery gives off both peripheral and sulcal branches. The sulcal branches pass posteriorly, penetrating the anterior longitudinal fissure. On reaching the anterior white commissure, they turn alternately to the right and to the left to supply the gray matter and deep white matter on each side. 5 Occasionally two adjacent vessels pass to the same side, and on other occasions, a common stem vessel bifurcates to supply both sides. Terminal branches of these vessels overlap those from vessels above and below on the same side of the cord. The peripheral branches of the anterior spinal artery pass radially and form an anastomotic network of vessels, the anterior pial arterial plexus, which supplies the anterior and lateral white matter tracts by penetrating branches. The posterior pial arterial plexus is formed as a rich anastomotic network from the paired posterior spinal arteries. Penetrating branches from this plexus supply the posterior horns and posterior funiculi.

Intrinsic Blood Supply of the Spinal Cord
In contrast to the extreme interindividual variability in the extraspinal arteries that supply the spinal cord, the intrinsic blood supply of the cord itself is more consistent. The anterior spinal artery gives off central (sulcal) arteries that pass posteriorly, penetrating the anterior longitudinal fissure and supplying most of the central gray matter and the deep portion of the anterior white matter ( Fig. 2-4 ). The number of these sulcal vessels is variable, with 5 to 8 vessels per centimeter in the cervical region, 2 to 6 vessels per centimeter in the thoracic region, and 5 to 12 vessels per centimeter in the lumbosacral region. 5, 6

FIGURE 2-4 Intrinsic blood supply of the spinal cord. The vascular territories are depicted on the right half of the cord. The hatched lines indicate the territory supplied by the posterior spinal arterial system. The remainder is supplied by the anterior circulation, with the dark stippling indicating the area supplied exclusively by the sulcal branches of the anterior spinal artery.
The anterior spinal artery also gives off peripheral arteries that pass radially on the anterior surface of the spinal cord to supply the white matter tracts anteriorly and laterally. These arteries form the anterior pial arterial plexus, which is often poorly anastomotic with its posterior counterpart. 5 The posterior horns and posterior funiculi are supplied by penetrating vessels from the posterior pial arterial plexus.

Ischemic Cord Syndromes
Ischemia of the spinal cord may be produced either by the interruption of blood flow through critical feeding vessels or by aortic hypotension. The resulting neurological syndrome depends on the location of ischemic lesions along and within the spinal cord, which depends, in turn, on the vascular anatomy discussed previously. A wide variety of pathological disturbances of the aorta result in spinal cord ischemia. They include both iatrogenic causes, such as surgery and aortography, 12 - 15 and intrinsic aortic diseases, such as dissecting and nondissecting aneurysms, 16, 17 inflammatory aortitis, 6, 18, 19 occlusive atherosclerotic disease, 20 infective and noninfective emboli, 21, 22 and congenital coarctation. 6, 23 Spinal cord ischemia is a rare complication of pregnancy, 24 possibly due to aortic compression, which can occur toward the end of gestation. 25
Some authors have suggested that the midthoracic region (T4 to T8) is particularly vulnerable to ischemia because of the sparseness of vessels feeding the anterior spinal artery in this region and its poor anastomotic connections. 6 Others have stressed the vulnerability of the watershed areas between the three anterior spinal arterial territories. Although the concept is theoretically appealing, documentation of the selective vulnerability of these regions is not completely convincing. For example, a review of 61 case reports 16, 26 - 29 with respect to the distribution of ischemic myelopathies resulting from surgery on the aorta does not especially suggest that either of these areas is more vulnerable than other cord segments ( Table 2-1 ). Even when the operation was performed on the thoracic aorta (and thus the proximal clamp was placed above the midthoracic cord feeder), the lumbosacral cord segments were the site of the ischemic damage more often than the supposedly more vulnerable midthoracic segments ( Table 2-1 ). Similarly, the watershed area between these two arterial territories (T8 to T9) does not seem particularly vulnerable. In fact, the most frequently affected cord segment within each vascular territory in these 61 cases was centrally placed—T6 in the midthoracic territory and T12 in the lumbosacral territory—rather than at the borders, as would be anticipated with watershed vulnerability ( Fig. 2-5 ).
TABLE 2-1 Influence of Location of Aortic Surgery on the Vascular Territory of Resulting Spinal Cord Ischemia   Location of Surgery Vascular Territory of Ischemia Abdominal Aorta Thoracic Aorta Cervical region (C1–T3) 0 0 Midthoracic region (T4–T8) 1 14 Lumbosacral region (T9–conus) 25 21
Based on 61 reported cases. 16, 26 - 29

FIGURE 2-5 Upper segmental level of spinal cord involvement in 61 cases of spinal cord ischemia after surgery on the aorta (based on previously published reports 16, 26 - 29 ).
Moreover, of the 25 cases of spinal cord infarction in an unselected autopsy series of 300 cases, two thirds were in cervical cord segments 6 ; the most commonly affected segment was C6. Such a distribution would be unexpected if either the midthoracic or the watershed area was particularly vulnerable. Perhaps relating to such observations, it was recently found that, contrary to earlier reports, the anterior spinal artery is continuous along its length without interruption in all 51 cadavers studied. 9 If this observation can be generalized, it may be the case that the poorly vascularized thoracic cord, which has much less gray matter than the cervical and lumbar enlargements, actually matches its sparse blood supply with its reduced metabolic requirements. 3, 6, 30
The site of aortic disease also plays an important role in the location of the lesion along the spinal cord. For example, distal aortic occlusion often presents with lumbosacral involvement, 6, 20 whereas dissecting aneurysm of the thoracic aorta commonly presents with infarction in the midthoracic region. 6, 17, 31, 32 Similarly, cord ischemia following surgery on the abdominal aorta is essentially confined to the lumbosacral territory, whereas surgery on the thoracic aorta not infrequently involves the midthoracic segments ( Table 2-1 ). Regardless of the pathological process affecting the aorta, however, it generally involves the suprarenal portion if there is cord ischemia 6, 33 because the important radiculomedullary arteries usually originate above the origin of the renal arteries.

Anterior Spinal Artery Syndrome
Ischemic injury of the spinal cord at a particular segmental level may present with a complete transverse myelopathy. 6 Within the spinal cord, however, there are certain vascular territories that can be affected selectively. In particular, the territory of the anterior spinal artery, especially its sulcal branch, is prone to ischemic injury. 3, 6 This increased vulnerability probably relates to two factors. First, the anterior circulation receives a much smaller number of feeding vessels than the posterior circulation. 1, 3, 4, 6 Second, the posterior circulation is a network of anastomotic channels 1, 3, 6 and therefore probably provides better collateral flow than the single anterior artery, which in some patients is discontinuous along its length. The relative constancy of the resulting syndrome 6, 34, 35 presumably reflects the relative constancy of the intrinsic vascular anatomy of the cord.
As mentioned earlier, the anterior spinal artery supplies blood to much of the spinal gray matter and to the tracts in the anterior and lateral white matter. Ischemia in this arterial territory therefore gives rise to a syndrome of diminished pain and temperature sensibility with preservation of vibratory and joint position sense. Weakness (either paraparesis or quadriparesis, depending on the segments involved) occurs below the level of the lesion and may be associated with other evidence of upper motor neuron involvement, such as Babinski signs, spasticity, and hyperreflexia. Bowel and bladder functions are affected, owing to interruption of suprasegmental pathways. Segmental gray matter involvement may also lead to lower motor neuron deficits and depressed tendon reflexes at the level of the lesion. Thus, a lesion in the cervical cord may produce flaccid areflexic paralysis with amyotrophy in the upper extremities, spastic paralysis in the lower extremities, and dissociated sensory loss in all limbs. In contrast, a lesion in the thoracic cord typically presents with only spastic paraplegia and dissociated sensory loss in the legs. The syndrome usually comes on abruptly, although occasionally it is more insidious and progressive. 36

Motor Neuron Disease
On occasion, diseases of the aorta (e.g., dissecting aneurysms or atherosclerosis) that interfere with the blood supply to the anterior spinal artery result in more restricted cord ischemia, perhaps because of better anastomotic connections between the anterior and the posterior pial arterial plexuses in some individuals or because of greater vulnerability of the anterior horn cells with their greater metabolic activity. 6, 14, 36 The ischemic injury in these circumstances is limited to the gray matter supplied by the sulcal branches ( Fig. 2-6 ). Clinical impairment is then confined to the motor system and is associated with amyotrophy. When the onset is abrupt, 6 the ischemic nature of the lesion usually is apparent, but when the onset is more gradual, 6, 36 and especially when pyramidal signs are also present, it may mimic other diseases, such as amyotrophic lateral sclerosis or spinal cord tumors.

FIGURE 2-6 Area of infarction within the spinal cord over four adjacent spinal segments in a patient reported by Herrick and Mills (Herrick MK, Mills PE: Infarction of spinal cord. Two cases of selective gray matter involvement secondary to asymptomatic aortic disease. Arch Neurol 24:228, 1971). The infarction was extensive but limited to the gray matter, particularly the anterior horns.

Posterior Spinal Artery Syndrome
In contrast to the anterior spinal artery syndrome, selective ischemia of the posterior circulation, characterized by prominent loss of posterior column function with relative sparing of other functions, is rarely recognized clinically 2, 14 and only occasionally reported pathologically. 10, 20, 37 For example, in a review of 27 cases of nonsurgical spinal cord ischemia, only 2 (7%) had posterior spinal artery patterns. 10 The relative infrequency of this syndrome presumably relates to the more abundant feeding vessels and better anastomotic connections in this arterial system compared to the anterior spinal artery.

Unilateral Cord Syndromes
In some cases, the area of ischemic damage can be confined to only a small portion of the spinal cord. For example, in the review cited previously, 10 eight (29%) of the patients with nonsurgical spinal cord ischemia had unilateral syndromes involving either the anterior or posterior aspects of the spinal cord.

Intermittent Claudication
Intermittent claudication (limping) refers to a condition in which a patient experiences difficulty in walking that is brought about by use of the lower extremities. Charcot initially described this syndrome in 1858 and related it to occlusive peripheral vascular disease in the lower extremities. 38 In 1906, Dejerine distinguished claudication caused by ischemia of the leg muscles from that caused by ischemia of the spinal cord. 39 In the latter condition, the arterial pulses in the legs tend to be preserved, pain tends to be dysesthetic or paresthetic in quality and may not occur, and neurological signs are frequently present, especially after exercise. In 1961, Blau and Logue identified another form of neurogenic claudication caused by ischemia or compression of the cauda equina and resulting from a narrowed lumbosacral canal (either congenital or due to degenerative disease). 40 This condition is similar to that produced by ischemia of the spinal cord; however, the sensory complaints tend to have a more radicular distribution, and signs of cord involvement (e.g., Babinski signs) are not present.
The clinical distinction between various types of claudication, particularly between the two neurogenic varieties, is sometimes difficult. The cauda equina variety, however, is far more common than the spinal cord form. 6, 41 Intermittent spinal cord ischemia, when it occurs, is often associated with intrinsic diseases of the aorta, such as coarctation or atherosclerotic occlusive disease. 6
Bony erosion through vertebral bodies from an abdominal aortic aneurysm with direct compression of the spinal nerve roots has also been reported to produce intermittent neurological symptoms. 42 The clinical details of the single reported case, however, are not sufficient to determine whether the symptoms resemble those of intermittent claudication.

Cerebral Ischemia

The aortic arch gives rise to all the major vessels that provide blood to the brain, brainstem, and cervical spinal cord ( Fig. 2-7 ). The first major branch is the innominate (brachiocephalic) artery, which subsequently divides into the right common carotid and right subclavian arteries. The latter artery subsequently gives rise to the right vertebral artery, which ascends through the foramina of the transverse processes of the upper six cervical vertebrae to join with its counterpart on the left and form the basilar artery. The basilar artery provides blood to the posterior fossa and posterior regions of the cerebral hemispheres. The second major branch of the aortic arch is the left common carotid artery, and the third is the left subclavian artery, which, in turn, gives rise to the left vertebral artery.

FIGURE 2-7 Vascular anatomy of the aortic arch and its branches.

Strokes and Transient Ischemic Attacks
Diseases of the aortic arch, such as atherosclerosis, aneurysms, and aortitis as well as surgery on this segment of the aorta, may give rise to symptoms of cerebrovascular insufficiency, such as strokes or transient ischemic attacks (TIAs). 6, 13, 18, 32, 43 - 47 A young woman has even been reported with a stroke secondary to an occlusion of the aorta that was associated with the use of birth control pills and recurrent venous thromboses. 48 Cerebral ischemia is produced either by occlusion of a major vessel or by embolization of atheromatous or other material to more distal arteries. The resulting neurological syndromes are not specific for any disease process but depend on the location and duration of the vascular occlusion.

Atherosclerosis of the aortic arch and its branches, compared with atherosclerosis at the origin of the internal carotid artery, is an infrequent cause of stroke or TIAs, probably for two reasons. First, atherosclerosis is much less common in this location than at the carotid bifurcation 49 ( Table 2-2 ). Second, the anastomotic connections between the major vessels in the neck are extensive, 6, 50 and an occlusion at their origin from the aortic arch is therefore less likely to be associated with symptoms of ischemia than a more peripheral obstruction.
TABLE 2-2 Distribution of Atherosclerosis in the Aorta and Its Branches Location Number of Lesions External carotid artery 9 Internal carotid artery 256 Common carotid artery 16 Innominate artery 16 Subclavian artery 29 Vertebral artery 55 Aortoiliac region 952 Femoropopliteal region 772
Based on data from Crawford ES, DeBakey ME, Cooley DA, et al: Surgical considerations of aneurysms and atherosclerotic occlusive lesions of the aorta and major arteries. Postgrad Med 29:151, 1961.

Transient Emboligenic Aortoarteritis
Transient emboligenic aortoarteritis has been reported by Wickremasinghe and colleagues to be a cause of stroke in young patients. 51 They described 10 patients (aged 16 to 36 years), all of whom had presented with pathologically verified thromboembolic strokes, and 3 of whom had a history of TIAs preceding the event by as much as 4 years. All these patients had both active and healed inflammatory lesions of the central elastic arteries, such as the aorta, innominate, common carotid, and proximal subclavian arteries. Active lesions were small (200 to 300 μm in diameter) and associated with a mural thrombus on the intimal surface. Healed lesions usually were associated with fibrous plaques but not with a mural thrombus. More peripheral arteries supplying the brain were normal. This condition seems to be distinct from segmental aortitis of the Takayasu type. Clinically it is an acute, intermittent disorder with an approximately equal sex incidence, whereas Takayasu’s disease is more chronic and has a strong female predominance. The systemic symptoms of Takayasu’s disease are absent, and occlusion of the central arteries does not occur in this condition.

Subclavian (Cerebral) Steal
Disease of the aortic arch may result in occlusion of either the innominate artery or the left subclavian artery proximal to the origin of the vertebral artery. This, in turn, may result in the reversal of the usual cephalad direction of blood flow in the ipsilateral vertebral artery ( Fig. 2-8 ), depending on individual variations in the collateral circulation and may result in ischemia in the posterior cerebral circulation. 52 - 56 In some patients, this is particularly evident when the metabolic demand (and therefore the blood flow) of the affected arm is increased during exercise. 52 If the innominate artery is blocked proximally, it may also cause a reversal of blood flow in the right common carotid artery, resulting in anterior circulation ischemia ( Fig. 2-8 ).

FIGURE 2-8 Mechanisms producing subclavian steal syndrome in diseases of the aortic arch and its branches. A, Obstruction of the left subclavian artery at its origin, resulting in reversal of blood flow in the left vertebral artery. B, Obstruction of the right subclavian artery distal to the takeoff of the right common carotid artery, resulting in reversal of blood flow in the right vertebral artery. C, Obstruction of the innominate artery at its origin, producing reversal of blood flow in the right common carotid artery.
Killen and colleagues reviewed the clinical features of a series of patients with demonstrated reversals of arterial blood-flow in a vertebral artery (i.e., with flow from the vertebral artery into the ipsilateral subclavian artery). 52 The left subclavian artery was affected more than twice as often as the right subclavian and innominate arteries combined, probably as a result of the more frequent involvement of this artery by atherosclerosis ( Table 2-2 ). Men were affected three times as often as women, probably reflecting the greater prevalence of atherosclerosis in men. Of the 87 patients in this series with symptoms that were adequately described, 75 (86%) had symptoms referable to the central nervous system (CNS). These symptoms were usually transient, lasting seconds to a few minutes, although the deficits were sometimes permanent. The neurological manifestations of steal were varied but most frequently included motor difficulties, vertigo, visual deficits, or syncope. Ischemic symptoms in the arms occurred in only a few patients, and precipitation of CNS symptoms by exercise of the arm ipsilateral to the occlusion was uncommon. Although reconstructive surgery relieved symptoms in most patients in this series, 52 it was the frequent failure of surgery to correct such nonspecific symptoms that led to a more recent reassessment of the importance of cerebral steal. 56
Thus, when noninvasive techniques such as Doppler ultrasonography have been used to define the direction of blood flow in the great vessels in a large spectrum of patients with vascular disease, the majority (50% to 75%) of patients with documented subclavian steal are found to be asymptomatic, even when the steal is bilateral. 53 - 55 When symptoms do occur, they are suggestive of transient vertebrobasilar insufficiency in only 7 to 37 percent of patients with steal; the occurrence of infarcts in this vascular territory is distinctly rare. 54, 55 For this reason, a recent review of this topic concluded that subclavian steal is a actually a marker of generalized atherosclerotic disease and that it is rarely a cause for symptoms of cerebral ischemia. 56

Peripheral Neuropathy
The peripheral nervous system is sometimes affected by aortic disease or surgery. The syndromes produced may be the presenting manifestations of aortic disease and may mimic less life-threatening conditions.


Left Recurrent Laryngeal Nerve
The left recurrent laryngeal nerve descends in the neck as part of the vagus nerve and wraps around the aortic arch just distal to the ligamentum arteriosum ( Fig. 2-7 ) before reascending in the neck to innervate all the laryngeal muscles on the left side except the cricothyroideus. It may be compressed by disease of the aortic arch, such as dissecting and nondissecting aneurysms or aneurysmal dilatation proximal to a coarctation of the aorta. 56 The resulting hoarse, low-pitched voice may be one of the earliest presenting symptoms of these conditions, although it is often overshadowed by other symptoms or signs, such as chest pain, shortness of breath, congestive heart failure, or hypertension. 57

Femoral Nerve
The femoral nerve arises from the nerve roots of L2, L3, and L4. It forms within the belly of the psoas muscle and then exits on its lateral aspect to innervate the quadriceps femoris, iliacus, pectineus, and sartorius muscles and the skin of the anterior thigh and medial aspect of the leg. The nerve is located considerably lateral to the aorta ( Fig. 2-9 ) and hence is rarely involved by direct compression. It may, however, be compressed by a hematoma from a ruptured aortic aneurysm into the psoas muscle and thereby signal a life-threatening condition that requires an urgent operation. 6, 58 - 60

FIGURE 2-9 Anatomy of the abdominal aorta showing its relationship to the femoral and obturator nerves, which form within the psoas muscle from branches of the L2, L3, and L4 segmental nerves.
The femoral nerve may also be injured as a consequence of aortic surgery. Boontje and Haaxma reported this complication in 3.4 percent of 1,006 abdominal aortic operations for atherosclerotic or aneurysmal disease, the left femoral nerve being involved unilaterally in two thirds of the cases and jointly with the right femoral nerve in another 6 percent. 61 The mechanism of injury in these cases was presumed to be ischemic and related to poor collateral blood supply to the intrapelvic portions of the femoral nerves, especially on the left.

Obturator Nerve
The obturator nerve also forms within the belly of the psoas muscle by the union of fibers from the L2, L3, and L4 segments, but, in contrast to the femoral nerve, exits medially from this muscle ( Fig. 2-9 ). It innervates the adductors of the leg and the skin on the medial aspect of the thigh. It too is lateral to the aorta and not usually involved by direct compression. Like the femoral nerve (and often together with it), the obturator nerve may be compressed by a hematoma in the psoas muscle. 6

Nerve roots, particularly L4, L5, S1, and S2, which lie almost directly underneath the terminal aorta and iliac arteries ( Fig. 2-10 ), may be directly compressed by an aortic aneurysm in this region. The syndromes produced are typical of radicular disease, with unilateral radiating pain and a radicular pattern to the sensory and motor findings. 6

FIGURE 2-10 Anatomy of the terminal branches of the aorta in relationship to the nerve roots that subsequently join to form the sciatic nerve. Aneurysmal dilatation of the abdominal aorta often includes dilatation of these branch vessels, which can compress the nerve roots, particularly the L4, L5, S1, and S2 nerve roots, which lie directly underneath.
Radiculopathies may also be produced by erosion of one or more vertebral bodies by an aortic aneurysm, with consequent compression of the nerve roots in the cauda equina or at the root exit zones. The syndrome produced is not necessarily associated with back pain; it may result in multisegmental involvement on one side or even in paraplegia. 62 - 64


Ischemic Monomelic Neuropathy
Ischemic monomelic neuropathy was described in detail by Wilbourn and co-workers, who reported 3 patients (and alluded to another 11) who had a distal “polyneuropathy” in one limb after sudden occlusion of a major vessel. 65 One of their patients had a saddle embolus to the distal aorta that occluded the right common iliac artery, another had a superficial femoral artery occlusion after placement of an intra-aortic balloon pump, and the third had upper-extremity involvement. The syndrome consists of a predominantly sensory neuropathy with a distal gradient. It affects all sensory modalities and is associated with a constant, deep, causalgia-like pain. The symptoms persist for months, even after revascularization or without evidence of ongoing ischemia. The results of nerve conduction studies and needle electromyography suggest an axonal neuropathy. There is no evidence of ischemic muscle injury, such as induration, muscle tenderness, or elevated serum creatine kinase levels. This condition is rare, although a similar syndrome has been reported in the setting of an acute aortic dissection, 66 and it may be that it is more prevalent than currently appreciated.

Autonomic Neuropathies

The autonomic nerves, particularly the lower thoracic and lumbar sympathetic fibers that lie close to the aorta and its branches, may be injured by disease of or surgery on the aorta. The preganglionic efferent sympathetic finerve fibers originate in the intermediolateral cell column in the spinal cord ( Fig. 2-4 ) and exit segmentally between T1 and L2 with the ventral roots. 67 The sympathetic fibers part company with the segmental nerves through the white rami communicantes ( Fig. 2-2 ), which enter the paravertebral sympathetic ganglia and trunks to form bilateral sympathetic chains; these chains are situated lateral to and parallel with the vertebral column ( Fig. 2-11 ). Some of these fibers synapse on postganglionic neurons in the ganglia of their segmental origin, whereas others ascend or descend in the trunk to different segmental levels before making such synapses. In the lumbosacral and cervical segments, where there are no white rami (i.e., below L2 or above T1), the segmental ganglia receive preganglionic contributions only from cord segments either above them (lumbosacral ganglia) or below them (cervical ganglia). 67 The postganglionic fibers rejoin the segmental nerves through the gray rami communicantes ( Fig. 2-2 ) to provide vasomotor, sudomotor, and pilomotor innervation throughout the body.

FIGURE 2-11 Anatomy of the terminal aorta and pelvis in the male in relationship to the sympathetic and parasympathetic nerves in the region.
Some of the preganglionic fibers, in contrast, do not synapse in the paravertebral ganglia but pass through them to form splanchnic nerves, which then unite in a series of prevertebral ganglia and plexuses (many of which overlie the thoracic and abdominal aorta). These structures, in turn, provide sympathetic innervation to the viscera. The plexus that overlies the aorta in the region of its bifurcation, the superior hypogastric plexus ( Fig. 2-11 ), is responsible (via the inferior hypogastric and other pelvic plexuses) for sympathetic innervation of the pelvic organs, including the prostate, prostatic urethra, bladder, epididymis, vas deferens, seminal vesicles, and penis in men ( Fig. 2-12 ) and the uterus, bladder, fallopian tubes, vagina, and clitoris in women. This plexus is formed by the union of the third and fourth lumbar splanchnic nerves with fibers from the more rostral inferior mesenteric plexus. Its segmental contribution usually derives from T11 to L2.

FIGURE 2-12 Distribution of sympathetic (left) and parasympathetic (right) nerves to the pelvic viscera and sexual organs in the male.
The visceral afferent fibers accompany the efferent autonomic fibers and pass uninterrupted back through the trunk, ganglia, and white rami to reach their nerve cells of origin in the dorsal root.

Postsympathectomy Neuralgia
Operations on the distal aorta to treat symptomatic aortic disease from atherosclerosis or other causes frequently include intentional sympathectomy as part of the effort to improve blood flow to the legs. This is usually done by dividing the sympathetic chain below the last white ramus at L2, thereby depriving the lower lumbar and sacral ganglia of their preganglionic innervation. Such an operation is often followed by a distinctive pain syndrome, 68, 69 which Raskin and associates termed postsympathectomy neuralgia. 68 In their experience with 96 such operations, this syndrome occurred in 35 percent of the patients. In each case, the sympathetic chain was interrupted at L3 by removal of the segmental ganglion. The pain was characterized as deep, boring, nonrhythmic, and nonradiating; it had an abrupt but delayed onset. The mean delay from sympathectomy to onset of pain was 12 days. The pain was located predominantly in the thigh, either medially or laterally, and was associated with tenderness in the area of pain. The course was always self-limited, with an average duration of 3 weeks.

Disorders of Sexual Function
Normal male sexual function has two distinct components. The first, erection, is a response mediated predominantly through the parasympathetic nervous system by the pelvic splanchnic nerves (nervi erigentes) arising from segments S2, S3, and S4 ( Fig. 2-12 ). Activation of these nerves causes vasodilatation and engorgement of the penile musculature and sinuses. 6, 70 The blood supply to the penis is provided by the internal pudendal artery via the internal iliac artery ( Fig. 2-10 ). The sympathetic nervous system, however, must have at least a modifying influence on erection because sympathectomy may disturb it. 6, 70 The second component, ejaculation, can be divided into two phases. The first phase, expulsion of seminal fluid into the prostatic urethra, is a response mediated predominantly by the sympathetic nervous system through the superior hypogastric plexus. 6, 70 The second phase, emission, is produced by the clonic contraction of penile musculature (bulbocavernosus and ischiocavernosus) innervated by somatic (pudendal) nerves ( Fig. 2-12 ).
Male sexual function may be disturbed by aortic disease or surgery. 6, 71 - 76 Female sexual function has not been as well studied in these circumstances, although it seems to be affected to a similar degree as in men. 77 Because the superior hypogastric plexus lies close to the aortic bifurcation ( Fig. 2-11 ), most preoperative and postoperative sexual disturbances occur with disease of this portion of the aorta, and most involve ejaculation ( Table 2-3 ). The pelvic splanchnic nerves are not situated near the aorta ( Fig. 2-11 ) and usually are not affected by aortic disease or surgery. Disturbances in erection, however, do occur, possibly because of sympathetic dysfunction, a reduction in blood flow to the internal pudendal artery and penis, or cavernovenous leakage. 71 - 76 To determine whether blood flow or sympathetic function was more important in this regard, Ohshiro and Kosaki examined the outcome of (1) terminal aortic operations either done traditionally or designed to spare the superior hypogastric plexus and (2) operations that did or did not preserve internal iliac blood-flow. 74 Their results indicated that preservation of the hypogastric plexus appeared to be more important for maintenance of normal erection and ejaculation than was preservation of internal iliac artery blood-flow ( Table 2-4 ). Other authors have also found that modification of operative technique to spare the superior hypogastric plexus considerably improves postoperative sexual function. 6, 71, 73

TABLE 2-3 Male Sexual Dysfunction in Patients With Disease of or Surgery on the Aorta

TABLE 2-4 Influence of Blood Flow and Sympathetic Function on Male Sexual Function After Abdominal Aortic Operations
Despite the importance of operative technique in preserving sexual function, preservation of blood flow is probably also important. Thus, Nevelsteen and colleagues reported a clear relationship between the occurrence of preoperative impotence and the adequacy of blood flow through the internal iliac arteries. 75 In this study, however, no special attempt was made to improve blood flow in the internal iliac artery during surgery, so that it is unclear whether a different operative approach might have been beneficial in restoring postoperative sexual function.

Certain conditions affecting the aorta merit special consideration because of the variety of nervous system syndromes that each can produce.

Injury to the aorta by a variety of infectious, toxic, allergic, or idiopathic causes may produce similar inflammatory pathological changes in the elastic media ( Table 2-5 ). 18 Such aortic damage may lead to neurological syndromes either primarily through direct involvement of important branch arteries by the pathological process or secondarily through the development of aneurysms, aortic stenosis, or atherosclerosis. The neurological syndromes produced either primarily or secondarily by aortitis depend on both the nature and the location of the resulting aortic lesion.
TABLE 2-5 Causes of Aortitis
Stenosing Aortitis
Takayasu’s arteritis
Postradiation during infancy
Nonstenosing Aortitis
Mycotic aneurysms
Rheumatic fever
Rheumatoid arthritis
Giant cell arteritis
Collagen vascular and other diseases *
Ankylosing spondylitis
Reiter’s syndrome
Relapsing polychondritis
* Systemic lupus erythematosus, scleroderma, psoriasis, Crohn’s disease, and ulcerative colitis.

Syphilitic Aortitis
During the prepenicillin era, syphilis was a common cause of aortitis, 18, 78 although by the 1950s, its occurrence had markedly declined. 78 A report in 1958 on the relative occurrence of atherosclerotic and syphilitic thoracic aortic aneurysms showed cases of syphilis outnumbering atherosclerosis by a ratio of 1.3:1.0. 79 A similar report published in 1982 gave this ratio as 0.13:1.0. 79 The pathological process in syphilitic aortitis is almost always in the thoracic aorta, 19, 78 in contrast to the distribution of atherosclerosis, which is more prevalent in the abdominal aorta ( Table 2-2 ). The aortitis is accompanied by aneurysmal dilatation of the aorta in approximately 40 percent of cases. 78 Rarely, it presents with multiple arterial occlusions and mimics Takayasu’s arteritis, 6 although patients are generally older than those with Takayasu’s arteritis and are usually men.

Takayasu’s Arteritis
Takayasu’s arteritis is an idiopathic inflammatory condition affecting the large arteries, particularly the aorta and its branches. 6, 45, 80, 81 The pathogenesis seems to involve cell-mediated autoimmunity, although the responsible antigen is unknown. The onset of disease is typically between the ages of 15 and 30 years, and the condition seems most prevalent in Asian and Hispanic populations. 6 More than 85 percent of affected individuals are women. In the early (prepulseless) phase, the disease may be characterized by systemic symptoms such as fever, night sweats, weight loss, myalgia, arthralgia, arthritis, and chest pain. In some patients, however, the systemic symptoms are either inconspicuous or absent. The later (pulseless) phase of the disease is characterized by occlusion of the major vessels of the aortic arch, producing symptoms such as Takayasu’s retinopathy, hypertension (secondary to renal artery stenosis, coarctation, or both), aortic regurgitation, and aortic aneurysms. Symptoms of cerebral ischemia can occur; however, they are typically reported in only a few patients. 6, 45, 81 Nevertheless, a report from South Africa on 272 patients who were diagnosed with Takayasu’s arteritis, based on the criteria of the American College of Rheumatology, found that 20 percent of the cohort had symptoms of cerebrovascular disease, including TIAs and stroke. 81 In addition, 32 percent of this cohort experienced intermittent claudication of either upper or lower limbs. 81 Seizures and headaches have also been reported but are uncommon. 6 Involvement along the aorta is typically diffuse, although some patients (perhaps as many as 20%) present with symptoms related to more restricted aortic involvement. 18, 81 The disorder is discussed further in Chapter 29 .

Giant Cell Arteritis
Giant cell arteritis (GCA) seems to be a particularly important cause of aortitis in the elderly 6, 18, 19, 82 ; although it typically affects medium-sized vessels, as many as one fourth or more of affected individuals have large-artery involvement. 83 For example, in one series of eight consecutive patients with aortitis, GCA seemed to be its basis in many. 18 Thus, four had definite GCA diagnosed based on their age at onset, the new onset of headaches, and an elevation in erythrocyte sedimentation rate. However, all these patients were older than 57 years, each of the eight had an elevation of some serum inflammatory marker (e.g., increase in Creactive protein levels, erythrocyte sedimentation rate, or fibrinogen levels), and three had symptoms of polymyalgia rheumatica. 6 In another series of 45 patients undergoing aortic resection and who had microscopic evidence of active noninfectious aortitis, the majority had either unclassifiable aortitis (47%) or GCA (31%), two entities that were histopathologically indistinguishable. 82 The presenting symptoms in patients with GCA or unclassified aortitis are generally nonspecific and include exhaustion, night sweats, weight loss, chest and back pain, headache, fevers of unknown origin, TIAs, and arm claudication. 19, 82 Typically, all segments of the aorta (ascending aorta, arch, and descending aorta) are involved in the inflammatory process, although involvement can be more restricted. 6, 18, 19 Between 10 and 20 percent of patients with unclassified aortitis or GCA will subsequently develop either dissecting or, more commonly, nondissecting aortic aneurysms. 82, 83

Aortic Aneurysms

Nondissecting Aneurysms
Nondissecting aortic aneurysms can be caused by any pathological process that weakens the arterial wall, such as inflammation, infection, or atherosclerosis. 6, 15, 18, 53, 60, 84, 85 In the past, syphilis was an important cause, 86 but at present, almost all these aneurysms are caused by atherosclerosis. As a result, the distribution of aortic aneurysms essentially parallels the distribution of atherosclerosis within the aorta, with most occurring in the abdominal aorta ( Tables 2-2 and 2-6 ). In a study from Sweden, it was found that the incidence of ruptured abdominal aortic aneurysms in men (but not women) had increased by more than 100 percent between 1971 to 1986 and 2000 to 2004. 87 The reason for this increased incidence is unclear, and it is unknown whether a similar increase has occurred in other parts of the world. The prognosis of untreated aneurysms is grave, with 80 percent of patients dying of rupture within 12 months of the onset of symptoms. 84 Disturbances of neurological function in aortic aneurysms are uncommon, but when they occur, they are variable and depend in part on the location and extent of the lesion. Abdominal aneurysms may result in sexual dysfunction, compressive neuropathies, 6, 58 - 62 , 71 - 77 or, rarely, spinal cord ischemic syndromes, including intermittent claudication, asymmetric paraparesis, and paraplegia 88 ; descending thoracic aneurysms may produce spinal cord ischemia, 16 and aortic arch aneurysms may result in cerebral ischemia or recurrent laryngeal nerve dysfunction. 6, 57 Most commonly, neurological symptoms are produced by either rupture or direct compression. Even when aneurysms result in paraplegia, the neurological deficit is often caused by bony erosion through the vertebral bodies and direct compression of the spinal cord or cauda equina rather than by ischemia. 64, 89
TABLE 2-6 Distribution and Nature of Aortic Aneurysms Site Number of Cases Nondissecting Aneurysms Aortic arch 56 Descending thoracic aorta 116 Thoracoabdominal aorta 25 Abdominal aorta 829 Dissecting Aneurysms Thoracic aorta 62
Based on data from Crawford ES, DeBakey ME, Cooley DA, et al: Surgical considerations of aneurysms and atherosclerotic occlusive lesions of the aorta and major arteries. Postgrad Med 29:151, 1961.

Dissecting Aortic Aneurysms
Dissecting aortic aneurysms, in contrast to nondissecting aortic aneurysms, predominantly involve the thoracic aorta, either at the beginning of the ascending segment (type A) or immediately distal to the left subclavian artery (type B). 6, 17, 32, 57, 66, 89, 90 Their etiology has not been established. Atherosclerosis is probably not a major factor because atherosclerosis is seldom found in the region of the intimal tear because the distribution of these aneurysms along the aorta is unlike that of atherosclerosis and because atherosclerosis is only infrequently present. 90 - 92 Nevertheless, hypertension probably is a factor as it is present in the large majority of patients with either type A or type B dissections. 91, 92 Moreover, dissecting aortic aneurysms have been associated with cystic medial necrosis, a degenerative condition focally affecting the arterial media, which may itself be related to hypertension. This condition is increased in patients with Marfan’s syndrome, as are dissecting aneurysms. Most aneurysms, however, do not occur in patients with Marfan’s syndrome or other identifiable collagen disorders, and the pathophysiology remains unknown. 89, 91, 92 Neurological involvement from dissecting aneurysms (due to the cutoff of important arteries by the dissection or by embolization) is well described but uncommon. It occurs more frequently with type A than type B dissections, 91, 92 and it usually involves either spinal or cerebral ischemia. Neurological involvement may also occur during surgery to repair the aneurysm. Thus, in one large series of 527 patients, 57 preoperative stroke occurred in 4 percent, and paraparesis occurred in another 2 percent. Patients with aortic dissection usually present with acute chest or back pain, which generally leads to the proper diagnosis. 57, 91, 92 On occasion, however, pain is absent, and the neurological syndrome is the presenting feature. 6, 66 Moreover, the neurological deficit produced by the dissecting aneurysm is sometimes only transient, lasting for several hours, and thereby mimicking other transient disturbances of neurological function. 6

Traumatic Aortic Aneurysm
Brutal deceleration injuries to the chest, especially from motor vehicle accidents, may result in traumatic rupture of the thoracic aorta, often just distal to the left subclavian artery. Many of these patients die immediately, but some present with an acute paraplegia. 93 - 96 Still others have a chronic aortic aneurysm that may present years later with acute spinal cord ischemia 93 or other neurological symptoms. 94

Coarctation of the Aorta
Coarctation of the aorta, a relatively common congenital condition, 97 typically results in constriction of the thoracic aorta just distal to the origin of the left subclavian artery. Less commonly, it occurs as part of Takayasu’s arteritis, and this condition should be suspected if the location of the coarctation is atypical. 6, 81 It may also follow radiation exposure during infancy 6, 98 ; in these cases, the pathological process is focal and limited to the segment of aorta that was in the field of irradiation. Coarctation can result in a variety of neurological symptoms ( Table 2-7 ). 6, 23, 97, 98 Cerebral infarcts probably result from embolization of thrombotic material in the dilated aorta proximal to the obstruction. 6
TABLE 2-7 Neurological Sequelae of Coarctation of the Aorta * Sequela Incidence (%) Ruptured intracerebral aneurysms 2.5 Ischemic stroke during childhood 1.0 Neurogenic intermittent claudication † 7.5 Headache 25.0 Episodic loss of consciousness 3.0 Intracerebral hemorrhage ‡ <1.0 Spinal cord compression ‡ <1.0
* Based on a review of 200 patients with coarctation of the aorta. 97
† Patients with exercise-induced motor or sensory disturbances in the lower extremities.
‡ These complications were not found in the series reported by Tyler and Clark 97 but have been reported by others, as described elsewhere. 6
Subarachnoid hemorrhage from ruptured saccular aneurysms can occur with coarctation. In the general population, aneurysmal hemorrhage has an annual incidence of approximately 8 per 100,000 6,99 and rarely occurs before the age of 20 years. 6, 99 Accordingly, the reported occurrence of ruptured aneurysms in 2.5 percent of patients with coarctation of the aorta 97 suggests an association of these two disorders, although the coincidental occurrence of the two conditions cannot be completely excluded. 6
Headache is a common accompaniment of coarctation, perhaps as a result of secondary hypertension, but, again, the incidental occurrence of two unrelated conditions cannot be excluded.
Episodic loss of consciousness may also occur in patients with coarctation of the aorta. It may result either from syncope due to associated cardiac abnormalities or from seizures. 6, 97 It is unclear, however, whether seizures unrelated to cerebrovascular disease are more prevalent in these patients than in the general population. 97
Neurogenic intermittent claudication can result from aortic coarctation. In patients with coarctation of the aorta, blood flow to the legs is often provided by collateral connections between the spinal arteries and the distal aorta. In these situations, the blood flow through the radiculomedullary and intercostal arteries distal to the obstruction is reversed, 6, 52 and the exercise-related spinal ischemia may be related to “steal” by the increased metabolic demands (and thus increased blood flow) of the legs 6 rather than aortic hypotension distal to the coarctation ( Fig. 2-13 ). These collaterals are sometimes so extensive that they cause spinal cord compression and mimic (clinically and radiologically) vascular malformations of the spinal cord. 6, 14, 98

FIGURE 2-13 Mechanism of steal in coarctation of the aorta. Obstruction of the aorta at the isthmus causes dilatation of the anterior spinal artery and reversal of blood flow in anterior radiculomedullary arteries distal to the obstruction. In this circumstance, the blood flow to the lower extremities is provided by these (and other) collaterals, and use of the lower extremities may cause shunting of blood from the spinal circulation to the legs, which, in turn, sometimes results in spinal cord ischemia.

Surgery and Other Procedures

Aortic Surgery
As with diseases of the aorta, the risks of aortic surgery depend in part on the site of operation. Thus, operations on the aortic arch may produce cerebral ischemia either by intraoperative occlusion of major vessels or by embolization of material such as calcified plaque loosened during surgery. 6, 13, 46, 47 Operations on the suprarenal aorta may result in spinal ischemia, 13 whereas operations on the distal aorta may result in sexual dysfunction or ischemia of the femoral nerve. 6, 58 - 62 , 71 - 77
The major complication of all aortic operations, however, is intraoperative spinal cord ischemia with resultant paraplegia or paraparesis. The occurrence of this complication varies with the location of the surgery and the nature of the pathological process affecting the aorta ( Table 2-8 ). Thus, operations on dissecting or nondissecting aortic aneurysms that are entirely abdominal are associated with a lower incidence of this complication than operations on aneurysms confined to the thoracic aorta. 6, 26, 100 - 104 Surgery on aneurysms involving the entire abdominal and thoracic aorta carries the highest risk of producing cord ischemia. 100 Operations on the distal aorta for occlusive disease only rarely result in spinal ischemia, especially when confined to the infrarenal portion. 26, 33 This variability presumably occurs because important feeding arteries to the spinal circulation are more likely to be ligated during surgery, included within the segment of the aorta that is cross-clamped, or subjected to distal hypotension when the aortic lesion is above the level of origin of the renal arteries.
TABLE 2-8 Spinal Cord Ischemia During Surgery and Procedures on the Aorta Diagnosis Number of Patients Percentage With Spinal Cord Damage Nondissecting aortic aneurysm     Abdominal * 1,724 0.46 Thoracic † 585 6.3 Thoracoabdominal ‡ 102 21.6 Dissecting aortic aneurysm § 102 30.4 Abdominal aortic occlusion ¶ 1,089 0 Coarctation of aorta ¶ 12,532 0.41 Aortography * 17,949 0.01
* Based on a report by Szilagyi and associates. 26
† Based on reports by DeBakey and associates, 79 Kahn and Sloan, 104 Livesay and colleagues, 101 Crawford and associates (group I), 100 Bloodwell and coworkers, 103 and Neville and associates. 102
‡ Based on the findings of Crawford and co-workers (group II). 100
§ Based on a report by Crawford and associates. 100 The relative risk of operation in these patients depended on the location along the aorta and essentially paralleled the experience in nondissecting aneurysms, although the numbers in each subcategory were too small to be included separately.
¶ Based on the findings of Brewer and associates (J Thorac Cardiovasc Surg 64:368, 1972).
Operations on the thoracic aorta for coarctation are much less frequently complicated by spinal ischemia than thoracic operations done for other reasons. 23 There are probably at least two reasons for this difference. First, the former patients are younger, and the extent of overall arterial disease is therefore less. Second, as mentioned earlier, the flow in the radiculomedullary vessels below the coarctation is frequently reversed, 6, 105 so obstruction of blood flow in them (either by ligation or cross-clamping the aorta above and below their origin) may actually result in an increased blood supply to the spinal cord.

Aortography and Other Procedures on the Aorta
Aortography may be associated with either spinal 16 or cerebral 6, 106 ischemia, depending on the portion of the aorta visualized. This complication, however, is distinctly rare ( Table 2-8 ). Paraplegia may also occur during intra-aortic balloon assistance after myocardial revascularization. 6, 107

Intraoperative Adjuncts to Avoid Spinal Cord Ischemia
Several adjuncts are commonly used during surgery in an attempt to avoid spinal cord injury. They include the use of deep hypothermia and circulatory arrest in addition to thiopental anesthesia and intraoperative corticosteroids, all of which are thought to reduce the metabolic requirements of the cord or otherwise enhance tolerance to ischemia. 57 In addition, many authors have reported that minimization of cross-clamp time results in a lower incidence of spinal ischemia.
Other adjunctive methods such as the reattachment of intercostal arteries, the use of shunts to maintain distal perfusion pressure, and the use of cerebrospinal fluid drainage have not proved consistently effective at preventing cord ischemia, 7, 46, 49, 108 - 114 although the more recent experience with such adjunctive techniques has been quite favorable. 7, 49, 113, 114 Part of the difficulty with these procedures may relate to the extreme variability of the blood supply to the spinal cord. For example, if a crucial spinal artery leaves the aorta within the cross-clamped section, the preservation of distal blood-flow is irrelevant. Furthermore, because the important intercostal arteries are few and unpredictably situated, the random reattachment of a few intercostal arteries may be fruitless.
There has been considerable interest in the use of somatosensory evoked potentials (SEPs) and motor evoked potentials (MEPs) for assessing spinal cord function during operations on the aorta. 94, 115 - 125 The combined use of SEPs and MEPs may ultimately prove better than either technique alone, 124 and, indeed, the most recent reports with both techniques have been encouraging. 7, 46, 123 - 125 An approach that seems particularly valuable is the use of intraoperative MEPs or SEPs to identify those vessels that perfuse the spinal cord and therefore need reattachment, should not be sacrificed, or should not be included within the aortic cross-clamp. 7, 121, 124 Another approach that has been reported to be useful is the use of intraoperative MEPs to monitor patients and to quickly increase both the distal aortic pressure and the mean arterial pressure in response to a drop in MEP amplitude. 123 Nevertheless, although these reports are quite encouraging, the best method of monitoring intraoperative spinal cord function and how best to use the information to alter the operative technique so that postoperative spinal cord function is maintained are still being determined.


1 Adamkiewicz A. I. Die Gefasse der Ruckenmarkersubstanz. Sitzungsb Akad Wissensch Wien Math-Naturw . 1881;84:469.
2 Adamkiewicz A. Die Blutgefasse des menschlichen Ruckenmarkes: II. Die Gefasse der Ruckenmarksoberflache. Sitzungsb Akad Wissensch Wien Math-Naturw . 1882;85:101.
3 Kadyi H. Über die Blutgefasse des menschlichen Ruckenmarker. Anat Anz . 1886;1:304.
4 Kadyi H. Über die Blutgefasse des menschlichen Ruckenmarkes: Nach einer im XV Bande der Denkschriften d Math-Naturw C1 d Akad d Wissensch in Krakau erschienenen Monographie aus dem Polnischen Ulsersatzt vom Verfasser. Lemberg: Grubrynowicz & Schmidt, 1889.
5 Turnbull IM. Blood supply of the spinal cord. Vinken PJ, Bruyn GW, editors. Handbook of Clinical Neurology . Amsterdam: North Holland; 1972;Vol 12:478.
6 Goodin DS. Neurologic sequelae of aortic disease and surgery. In: Aminoff MJ, editor. Neurology and General Medicine . 3rd Ed. New York: Churchill Livingstone; 2001:23.
7 Griepp RB, Ergin MA, Galla JD, et al. Looking for the artery of Adamkiewicz: a quest to minimize paraplegia after operations for aneurysms of the descending thoracic and thoracoabdominal aorta. J Thorac Cardiovasc Surg . 1996;112:1202.
8 Milen MT, Bloom DA, Culligan J, et al. Albert Adamkiewicz (1850–1921): his artery and its significance for the retroperitoneal surgeon. World J Urol . 1999;17:168.
9 Biglioli P, Roberto M, Cannata A, et al. Upper and lower spinal cord blood supply: the continuity of the anterior spinal artery and the relevance of the lumbar arteries. J Thorac Cardiovasc Surg . 2004;127:1188.
10 Novy A, Carruzzo A, Maeder P, et al. Spinal cord ischemia: clinical and imaging patterns, pathogenesis, and outcomes in 27 patients. Arch Neurol . 2006;63:1113.
11 Nijenhuis RJ, Jacobs MJ, van Engelshoven JMA, et al. MR angiograph of the Adamkiewicz artery and anterior radiculomedullary vein: postmortem validation. AJNR Am J Neuroradiol . 2006;17:573.
12 Skillman JJ. Neurological complications of cardiovascular surgery: I. Procedures involving the carotid arteries and abdominal aorta. Int Anesthesiol Clin . 1986;24:135.
13 Shaw PJ. Neurological complications of cardiovascular surgery: II. Procedures involving the heart and thoracic aorta. Int Anesthesiol Clin . 1986;24:159.
14 Aminoff MJ. Vascular disorders of the spinal cord. In: Davidoff RA, editor. Handbook of the Spinal Cord . New York: Marcel Dekker; 1987:259.
15 Ross RT. Spinal cord infarction in disease and surgery of the aorta. Can J Neurol Sci . 1985;12:289.
16 Kewalramani LS, Orth MS, Katta RSR. Atraumatic ischaemic myelopathy. Paraplegia . 1981;19:352.
17 Leramo OB, Char G, Coard K, et al. Spinal stroke as a presentation of dissecting aneurysm of the aorta. West Indian Med J . 1986;35:203.
18 Scheel AK, Meller J, Vosshenrich R, et al. Diagnosis and follow up of aortitis in the elderly. Am Rheum Dis . 2004;63:1507.
19 Tavora F, Burke A. Review of isolated ascending aortitis: differential diagnosis, including syphilitic, Takayasu’s and giant cell aortitis. Pathology . 2006;38:302.
20 Nedeltchev K, Loher TJ, Stepper F, et al. Long-term outcome of acute spinal cord ischemia syndrome. Stroke . 2004;35:560.
21 Dickson AP, Lum SK, Whyte AS. Paraplegia following saddle embolism. Br J Surg . 1984;71:321.
22 Syrjanen J, Iivanainen M, Kallio M, et al. Three different pathogenic mechanisms for paraparesis in association with bacterial infections. Ann Clin Res . 1986;18:191.
23 Rao PS. Coarctation of the aorta. Curr Cardiol Rep . 2005;7:425.
24 Barré JA, D’Andrade C. Paraplegie par ramollissement aigu unisegmentaire de la moelle survenne au cours de la grossesse: étude anatomo-clinique. Rev Neurol (Paris) . 1938;69:133.
25 Chen GY, Kuo CD, Yang MJ, et al. Return of autonomic nervous activity after delivery: role of aortocaval compression. Br J Anaesth . 1999;82:932.
26 Szilagyi DE, Hageman JH, Smith RF, et al. Spinal cord damage in surgery of the abdominal aorta. Surgery . 1978;83:38.
27 Lynch C, Weingarden SI. Paraplegia following aortic surgery. Paraplegia . 1982;20:196.
28 Costello TG, Fisher A. Neurological complications following aortic surgery. Anaesthesia . 1983;38:230.
29 Picone AL, Green RM, Ricotta JR, et al. Spinal cord ischemia following operations on the abdominal aorta. J Vasc Surg . 1986;3:94.
30 Duggal N, Lach B. Selective vulnerability of the lumbosacral spinal cord after cardiac arrest and hypotension. Stroke . 2002;33:116.
31 Jacobs MJ, Elenbaas TW, Schurink GW, et al. Assessment of spinal cord integrity during thoracoabdominal aortic aneurysm repair. Ann Thorac Surg . 2002;74:S1864.
32 Ohmi M, Shibuya T, Kawamoto S, et al. Spinal cord ischemia complicated with acute aortic dissection and intramural hematoma; report of two cases. Kyobu Geka . 2003;56:473.
33 Webb TH, Williams GM. Thoracoabdominal aneurysm repair. Cardiovasc Surg . 1999;7:573.
34 Hakimi KN, Massagli TL. Anterior spinal artery syndrome in two children with genetic thrombotic disorders. J Spinal Cord Med . 2005;28:69.
35 Ogun SA, Adefuye B, Kolapo KB, et al. Anterior spinal artery syndrome complicating aortic dissecting aneurysm: case report. East Afr Med J . 2004;81:549.
36 Skinhoj E. Arteriosclerosis of the spinal cord. Acta Psychiatr Neurol Scand . 1954;29:139.
37 Blanc R, Hosseini H, Le Guerinel C, et al. Posterior cervical spinal cord infarction complicating the treatment of an intracranial dural arteriovenous fistula embolization. Case report. J Neurosurg Spine . 2006;5:79.
38 Charcot JM. Sur la claudication intermittente observée dans un cas d’oblitération complète de l’une des artéres iliaques primitives. C R Mem Soc Biol . 1858;5:225.
39 Dejerine J. Sur la claudication intermittente de la moelle épinière. Rev Neurol (Paris) . 1906;14:341.
40 Blau JN, Logue V. Intermittent claudication of the cauda equina. Lancet . 1961;1:1081.
41 Truumees E. Spinal stenosis: pathophysiology, clinical and radiologic classification. Instr Course Lect . 2005;54:287.
42 Crawford ES, Snyder DM, Cho GC, et al. Progress in treatment of thoracoabdominal and abdominal aortic aneurysms involving celiac, superior mesenteric, and renal arteries. Ann Surg . 1978;188:404.
43 Jex RK, Schaff HV, Piehler JM, et al. Early and late results following repair of dissections of the descending thoracic aorta. J Vasc Surg . 1986;3:226.
44 Frist WH, Baldwin JC, Starnes VA, et al. A reconsideration of cerebral perfusion in aortic arch replacement. Ann Thorac Surg . 1986;42:273.
45 Mwipatayi BP, Jeffery PC, Beningfield SJ, et al. Takayasu arteritis: clinical features and management: report of 272 cases. Aust N Z J Surg . 2005;75:110.
46 Erlich M, Grabenwoger M, Luckner D, et al. Operative management of aortic arch aneurysm using profound hypothermia and circulatory arrest. J Cardiovasc Surg . 1996;37:63.
47 Okita T, Ando M, Minatoya K, et al. Predictive factors for mortality and cerebral complications in arteriosclerotic aneurysm of the aortic arch. Ann Thorac Surg . 1999;33:72.
48 Bogousslavsky J, Regli F. Ischemic stroke in adults younger than 30 years of age. Arch Neurol . 1987;44:479.
49 Crawford ES, DeBakey ME, Cooley DA, et al. Surgical considerations of aneurysms and atherosclerotic occlusive lesions of the aorta and major arteries. Postgrad Med . 1961;29:151.
50 Wylie EJ, Effeney DJ. Surgery of the aortic arch branches and vertebral arteries. Surg Clin North Am . 1979;59:669.
51 Wickremasinghe HR, Peiris JB, Thenabadu PN, et al. Transient emboligenic aortoarteritis. Arch Neurol . 1978;35:416.
52 Killen DA, Foster JH, Walter GGJr, et al. The subclavian steal syndrome. J Thorac Cardiovasc Surg . 1966;51:539.
53 Aseem WM, Makaroun MS. Bilateral subclavian steal syndrome through different paths and different sites. Angiology . 1999;50:149.
54 Ackermann H, Diener HC, Dichgans J. Stenosis and occlusion of the subclavian artery: ultrasonographic and clinical findings. J Neurol . 1987;234:396.
55 Hennerici M, Klemm C, Rautenberg W. The subclavian steal phenomenon: a common vascular disorder with rare neurologic deficits. Neurology . 1988;38:669.
56 Taylor SL, Selman WR, Ratcheson RA. Steal affecting the central nervous system. Neurosurgery . 2002;50:670.
57 DeBakey ME, McCollum CH, Crawford ES, et al. Dissection and dissecting aneurysms of the aorta: twenty-year follow-up of five hundred twenty-seven patients treated surgically. Surgery . 1982;92:1118.
58 Merchant RFJr, Cafferata HT, DePalma RG. Ruptured aortic aneurysm seen initially as acute femoral neuropathy. Arch Surg . 1982;17:811.
59 Owens ML. Psoas weakness and femoral neuropathy: neglected signs of retroperitoneal hemorrhage from ruptured aneurysm. Surgery . 1982;91:363.
60 Sterpetti AV, Blair EA, Schultz RD, et al. Sealed rupture of abdominal aortic aneurysms. J Vasc Surg . 1990;11:430.
61 Boontje AH, Haaxma R. Femoral neuropathy as a complication of aortic surgery. J Cardiovasc Surg . 1987;28:286.
62 Wilberger JE. Lumbosacral radiculopathy secondary to abdominal aortic aneurysms. J Neurosurg . 1983;58:965.
63 Rothschild BM, Cohn L, Aviza A, et al. Aortic aneurysm producing back pain, bone destruction, and paraplegia. Clin Orthop . 1982;164:123.
64 Higgins R, Peitzman AB, Reidy M, et al. Chronic contained rupture of an abdominal aortic aneurysm presenting as a lower extremity neuropathy. Ann Emerg Med . 1988;17:284.
65 Wilbourn AJ, Furlan AJ, Hulley W, et al. Ischemic monomelic neuropathy. Neurology . 1983;33:447.
66 Beach C, Manthey D. Painless acute aortic dissection presenting as left lower extremity numbness. Am J Emerg Med . 1998;16:49.
67 Afifi AK, Bergman RA. Functional Neuroanatomy, 2nd Ed. New York: McGraw-Hill, 1998.
68 Raskin NH, Levinson SA, Hoffman PM, et al. Post-sympathectomy neuralgia: amelioration with diphenylhydantoin and carbamazepine. Am J Surg . 1974;128:75.
69 Smead WL, Vaccaro PS. Infrarenal aortic aneurysmectomy. Surg Clin North Am . 1983;63:1269.
70 Quayle JB. Sexual function after bilateral lumbar sympathectomy and aortoiliac by-pass surgery. J Cardiovasc Surg . 1980;21:215.
71 DePalma RG. Impotence in vascular disease: relationship to vascular surgery. Br J Surg . 1982;69(suppl):S14.
72 Ohshiro T, Takahashi A, Kosaki G. Sexual function in patients with aortoiliac vascular disorders. Int Surg . 1982;67:49.
73 Flanigan DP, Schuler JJ, Keifer T, et al. Elimination of iatrogenic impotence and improvement of sexual function after aortoiliac revascularization. Arch Surg . 1982;117:544.
74 Ohshiro T, Kosaki G. Sexual function after aortoiliac vascular reconstruction. J Cardiovasc Surg . 1984;25:47.
75 Nevelsteen A, Beyens G, Duchateau J, et al. Aortofemoral reconstruction and sexual function: a prospective study. Eur J Vasc Surg . 1990;4:247.
76 Cormio L, Edgre J, Lepäntalo M, et al. Aortofemoral surgery and sexual function. Eur J Vasc Endovasc Surg . 1996;11:453.
77 Hultgren R, Sjögren B, Söderberg M, et al. Sexual function in women suffering from aortoiliac occlusive disease. Eur J Vasc Endovasc Surg . 1999;17:306.
78 Heggtveit HA. Syphilitic aortitis: a clinicopathologic autopsy study of 100 cases, 1950 to 1960. Circulation . 1964;29:346.
79 DeBakey ME, Cooley DA, Crawford ES, et al. Aneurysms of the thoracic aorta: analysis of 179 patients treated by resection. J Thorac Surg . 1958;36:393.
80 Bickerstaff LK, Pairolero PC, Hollier LH, et al. Thoracic aortic aneurysms: a population-based study. Surgery . 1982;92:1103.
81 Rizzi R, Bruno S, Stellacci C, et al. Takayasu’s arteritis: a cell-mediated large-vessel vasculitis. Int J Clin Lab Res . 1999;29:8.
82 Miller DV, Isotalo PA, Weyand CM, et al. Surgical pathology of non-infectious ascending aortitis: a study of 45 cases with an emphasis on an isolated variant. Am J Surg Pathol . 2006;30:1150.
83 Bongartz T, Matteson EL. Large vessel involvement in giant cell arteritis. Curr Opin Rheumatol . 2006;18:10.
84 Haimovici H. Abdominal aortic aneurysm. In: Haimovici H, editor. Vascular Surgery . East Norwalk, CT: Appleton-Century-Crofts; 1984:685.
85 Johansen K, Devin J. Mycotic aortic aneurysms: a reappraisal. Arch Surg . 1983;118:583.
86 Brindley P, Schwab EH. Aneurysms of the aorta, with a summary of pathological findings in 100 cases at autopsy. Tex State J Med . 1930;25:757.
87 Acosta S, Ögren M, Bengtsson H, et al. Increasing incidence of ruptured abdominal aortic aneurysms: a population-based study. J Vasc Surg . 2006;44:237.
88 Roquer J, Martí N, Cano A, et al. Spinal cord ischemia indicating aneurysm of the abdominal aorta: report of three cases. Neurologia . 1995;10:201.
89 Rosen SA. Painless aortic dissection presenting as spinal cord ischemia. Ann Emerg Med . 1988;17:840.
90 Dalen JE, Pape LA, Cohn LH, et al. Dissection of the aorta: pathogenesis, diagnosis, and treatment. Prog Cardiovasc Dis . 1980;23:237.
91 Xu SD, Huang FJ, Yang JF, et al. Endovascular repair of acute type B aortic dissection: Early and mid-term results. J Vasc Surg . 2006;43:1090.
92 Tsai TT, Evangelista A, Nienaber CA, et al. Long-term survival in patients presenting with type A acute aortic dissection: insights from the international registry of acute aortic dissection (IRAD). Circulation . 2006;114(suppl 1):350.
93 Conti VR, Calverley J, Safley WL, et al. Anterior spinal artery syndrome with chronic traumatic thoracic aortic aneurysm. Ann Thorac Surg . 1982;33:81.
94 Mitchell RL, Enright LP. The surgical management of acute and chronic injuries of the thoracic aorta. Surg Gynecol Obstet . 1983;157:1.
95 Schmidt CA, Jacobson JG. Thoracic aortic injury: a ten-year experience. Arch Surg . 1984;119:1244.
96 Naude GP, Back M, Perry MO, et al. Blunt disruption of the abdominal aorta: report of a case and review of the literature. J Vasc Surg . 1997;25:931.
97 Tyler HR, Clark DB. Neurologic complications in patients with coarctation of aorta. Neurology . 1958;8:712.
98 Lerberg DB, Hardesty RL, Siewers RD, et al. Coarctation of the aorta in infants and children: 25 years of experience. Ann Thorac Surg . 1982;33:159.
99 Brisman JL, Song JK, Newell DW. Cerebral aneurysms. N Engl J Med . 2006;355:928.
100 Crawford ES, Crawford JL, Safi HJ, et al. Thoracoabdominal aortic aneurysms: preoperative and intraoperative factors determining immediate and long-term results of operations in 605 patients. J Vasc Surg . 1986;3:389.
101 Livesay JJ, Cooley DA, Ventemiglia RA, et al. Surgical experience in descending thoracic aneurysmectomy with and without adjuncts to avoid ischemia. Ann Thorac Surg . 1985;39:37.
102 Neville WE, Cox WD, Leininger B, et al. Resection of the descending thoracic aorta with femoral vein to femoral artery oxygenation perfusion. J Thorac Cardiovasc Surg . 1968;56:39.
103 Bloodwell RD, Hallman GL, Cooley DA. Partial cardiopulmonary bypass for pericardiectomy and resection of descending thoracic aortic aneurysms. Ann Thorac Surg . 1968;6:46.
104 Kahn DR, Sloan H. Resection of descending thoracic aneurysms without left heart bypass. Arch Surg . 1968;97:336.
105 Krieger KH, Spencer FC. Is paraplegia after repair of coarctation of the aorta due principally to distal hypotension during aortic cross-clamping? Surgery . 1985;97:2.
106 Galbreath C, Salgado ED, Furlan AJ, et al. Central nervous system complications of percutaneous transluminal coronary angioplasty. Stroke . 1986;17:616.
107 Ussia GP, Marasini M, Pongiglione G. Paraplegia following percutaneous balloon angioplasty of aortic coarctation: a case report. Catheter Cardiovasc Interv . 2001;54:510.
108 Laschinger JC, Cunningham JN, Cooper MM, et al. Prevention of ischemic spinal cord injury following aortic cross-clamping: use of corticosteroids. Ann Thorac Surg . 1984;38:500.
109 Robertson CS, Foltz R, Grossman RG, et al. Protection against experimental ischemic spinal cord injury. J Neurosurg . 1986;64:633.
110 Kumagai H, Isaka M, Sugawara Y, et al. Intra-aortic injection of propofol prevents spinal cord injury during aortic surgery. Eur J Cardiothorac Surg . 2006;29:714.
111 Oldfield EH, Plunkett RJ, Nylander WAJr, et al. Barbiturate protection in acute experimental spinal cord ischemia. J Neurosurg . 1982;56:511.
112 Patel HJ, Shillingford MS, Mihalik S, et al. Resection of the descending thoracic aorta. Ann Thorac Surg . 2006;82:90.
113 Safi HJ, Mille CCIII. Spinal cord protection in descending thoracic and thoracoabdominal aortic repair. Ann Thorac Surg . 1999;67:1937.
114 Coselli JS, LeMaire SA. Left heart bypass reduces paraplegia rates after thoracoabdominal aortic aneurysm repair. Ann Thorac Surg . 1999;67:1931.
115 Laschinger JC, Cunningham JNJr, Nathan IM, et al. Experimental and clinical assessment of the adequacy of partial bypass in maintenance of spinal cord blood flow during operations on the thoracic aorta. Ann Thorac Surg . 1983;36:417.
116 Ogino H, Sasaki H, Minatoya K, et al. Combined use of Adamkiewicz artery demonstration and motor evoked potentials in descending thoracoabdominal repair. Ann Thorac Surg . 2006;82:592.
117 Kaplan BJ, Friedman WA, Alexander JA, et al. Somatosensory evoked potential monitoring of spinal cord ischemia during aortic operations. Neurosurgery . 1986;19:82.
118 Mizrahi EM, Crawford ES. Somatosensory evoked potentials during reversible spinal cord ischemia in man. Electroencephalogr Clin Neurophysiol . 1984;58:120.
119 MacDonald DB. Intraoperative motor evoked potential monitoring: overview and update. J Clin Monit Comput . 2006;20:347.
120 Coles JG, Wilson GJ, Sima AF, et al. Intraoperative detection of spinal cord ischemia using somatosensory cortical evoked potentials during thoracic aortic occlusion. Ann Thorac Surg . 1982;34:299.
121 Svensson LG, Patel V, Robinson MF, et al. Influence of preservation or perfusion of intraoperatively identified spinal cord blood supply on spinal motor evoked potentials and paraplegia after aortic surgery. J Vasc Surg . 1991;13:355.
122 Reuter DG, Tacker WA, Badylak SF, et al. Correlation of motor-evoked potential response to ischemic spinal cord damage. J Thorac Cardiovasc Surg . 1992;104:262.
123 Jacobs MJHM, Meylaerts SA, de Haan P, et al. Strategies to prevent neurologic deficit based on motor-evoked potentials in type I and II thoracoabdominal aortic aneurysm repair. J Vasc Surg . 1999;29:48.
124 Guerit JM, Witdoeckt C, Verhelst R, et al. Sensitivity, specificity, and surgical impact of somatosensory evoked potentials in descending aorta surgery. Ann Thorac Surg . 1999;67:1943.
125 Galla JD, Ergin MA, Lansman SL, et al. Use of somatosensory evoked potentials for thoracic and thoracoabdominal aortic resections. Ann Thorac Surg . 1999;67:1947.
Chapter 3 Neurological Complications of Cardiac Surgery

John R. Hotson

Consequences During “Normal” Convalescence
Brain Disorders
Peripheral Nerve Disorders
Neuro-ophthalmological Disorders
Neurological complications are a potential consequence of cardiac surgery that can nullify or limit any benefits of such surgery. 1 - 8 The probability of these complications increases as coronary artery bypass graft surgery is used for treating ischemic heart disease in older patients, in patients with multiple comorbid conditions, and as heart transplantation programs expand. 9 In spite of the increased use of catheter-based coronary revascularization, more than 400,000 people per year have coronary artery bypass graft surgery in the United States. 10 A substantial number of patients have a postoperative adverse cerebral outcome such as stroke or hypoxic-ischemic encephalopathy. 6, 11 Many more patients may develop loss of cognitive performance after heart surgery. 8, 12 Prevention of perioperative neurological complications remains an important medical problem.

Cardiopulmonary bypass was first used successfully in cardiac surgery in 1953 and was the pivotal development that led to modern cardiac surgery. 13 Its early use in humans resulted in frequent complications, which restricted its employment to seriously ill patients with progressive heart failure. Although the safety of extracorporeal circulation has increased, it remains a potential cause of neurological complications independent of other heart surgery procedures.

Open heart surgery with cardiopulmonary bypass begins with exposure of the heart, usually by a median sternotomy, followed by cannulation of the ascending aorta and vena cava or right atrium. 13 Insertion of the aortic cannula can dislodge atheromatous material in a severely diseased aorta, thereby leading to cerebral embolization. 14 In addition, this procedure can produce rotational torsion or compression of the brachial plexus, with subsequent injury. 15, 16
Extracorporeal circulation is used in association with systemic heparinization, hypothermia, and hemodilution. 13 Anticoagulation is used to prevent clot formation when blood is exposed to the nonendothelial surfaces of the bypass pump oxygenator and microaggregation filtration system. Core hypothermia is often used in combination with selective cooling of the heart, or cold cardioplegia, in order to protect the heart, brain, and other vital organs from ischemic damage. Infusion of ice slush solutions into the pericardium is one technique for inducing cold cardioplegia, but it occasionally produces focal phrenic nerve injuries. 17 - 19 Normothermic cardiopulmonary bypass may also be used in patients with few risk factors for stroke because it provides better hemodynamic function and decreases cardiopulmonary bypass time. 20, 21
Normovolemic hemodilution is used in part to conserve blood loss. It also compensates for the progressive hemoconcentration, decrease in plasma vol ume, and reduced blood flow that is associated with hypothermia. 13
During cardiac surgery with extracorporeal circulation, the ascending aorta is routinely cross-clamped; during valve-replacement surgery, congenital heart disease repair, or left ventricular aneurysm resection, the cardiac chambers are entered. These procedures may disrupt diseased tissue and produce emboli. Arterial systolic, diastolic, and mean pressure, pump pressure and flow rate, and central venous pressure are monitored during cardiopulmonary bypass. Mean arterial pressure is usually maintained above 40 to 50 mmHg by vasodilators, pressors, or volume expanders. 13

Consequences During “Normal” Convalescence
Extracorporeal circulation has predictable effects that result in a postperfusion syndrome and systemic inflammatory response during “normal” convalescence. 13, 22 This syndrome includes the following conditions.

Reduced Clotting Factors.
Exposure of blood to an abnormal environment during cardiopulmonary bypass leads to consumption of platelets and coagulation factors. Platelets adhere to the unphysiological surface of the oxygenators and filtration system of the bypass pump. This clumping of platelets not only predisposes to platelet emboli but also can reduce the number of circulating platelets. The exposure to foreign surfaces also causes release and depletion of granule-stored aggregating proteins in surviving platelets. Therefore, the remaining platelets have decreased adhesiveness. 13, 22
Coagulation factors are also consumed during cardiopulmonary bypass. A variety of carrier proteins and lipoproteins are denatured when blood passes through the bypass pump oxygenator. Even with adequate heparin levels, these damaged proteins can initiate a cascade in several coagulation and inflammatory systems. 22, 23 The clinical significance of these hematological changes is usually minor. They may contribute, however, to the intracranial hemorrhages that are occasionally observed after open heart surgery. 24

Cardiovascular Response.
During the early postoperative period, the degree of peripheral vasoconstriction provides a clinical estimate of cardiac output. 13 Transient atrial fibrillation, which carries a risk of cardiac emboli, is common during the convalescent period. 25, 26 A metabolic acidosis is also common during the 2 hours immediately after operation and reflects a washout of lactic acid from regions of poor perfusion during extracorporeal circulation. Persistence of a metabolic acidosis may indicate inadequate tissue perfusion secondary to low cardiac output. 13

Red Blood Cell Fragmentation.
Exposure of blood to nonendothelial surfaces during bypass surgery causes a breakdown of red blood cells, with subsequent anemia, hemoglobinemia, and hemoglobinuria. 13

Mild Mental Confusion.
Transient mild disturbances of orientation, memory, and level of alertness that resolve within the first few days after open heart surgery with cardiopulmonary bypass are frequent. 13 Whether the changes in mentation are totally reversible events that accompany most major operations or whether they indicate long-term sequelae is an area of sustained interest. 8, 11

Brain Swelling.
Brain swelling is present when magnetic resonance imaging (MRI) is obtained immediately following coronary artery bypass surgery. When brain imaging studies are repeated 2 to 3 weeks after the surgery, the brain swelling has remitted. There are no clinical findings associated with the brain swelling, and its cause is unknown. 27
These expected consequences of cardiopulmonary bypass are functionally reversible and compensated for during convalescence. Firm evidence that extracorporeal circulation itself permanently harms the brain is lacking. Cardiopulmonary bypass, however, does create numerous potential hazards that, if augmented by procedural mishap, may lead to permanent injury of the central nervous system (CNS). Cardiac operations using extracorporeal circulation carry the risks of embolus formation (from platelets, fibrin, tissue or surgical debris, air, or fat), cerebral hypoperfusion, and even hemorrhage. 13, 22 For these reasons, there is interest in performing coronary artery surgery without the use of cardiopulmonary bypass. 28 This off-pump technique can produce excellent cardiac outcomes. It is associated with fewer cerebral microemboli and less short-term neurocognitive decline when compared to on-pump coronary artery surgery. 29, 30 The short-term difference in cognitive performance, however, is limited, has not been consistently found across studies, and is not statistically significant at 1 year. 28, 30, 31 The off-pump technique does not lower the overall stroke rate, but may decrease stroke in a high-risk subgroup of patients with a severely atheromatous aorta. 32 Comparisons of performing coronary artery bypass grafting off-pump and with cardiopulmonary bypass have not proven the overall superiority of either method. 28

Diffuse or multifocal anoxic-ischemic damage, focal cerebral infarction, and brachial plexus injuries remain the main causes of permanent, disabling neurological complications after cardiac surgery. 1, 6, 15, 33 Therefore, the common, obvious postoperative symptoms include diffuse impairment of cognition and level of consciousness, focal deficits from stroke, and isolated peripheral weakness and sensory loss in one arm and hand.

Brain Disorders
Stroke, encephalopathy, coma, and seizures are the major brain disorders complicating cardiac surgery. 6, 34 Stroke is reported in prospective studies to occur in 1 percent to more than 5 percent of patients following coronary artery bypass surgery, and the incidence increases in association with preoperative stroke risk factors. 6, 21, 35 - 37 Stroke after cardiac surgery increases hospital mortality approximately five- to sixfold, prolongs intensive care, and typically requires inpatient rehabilitation or nursing home placement. 38, 39 The majority of stroke patients who survive to hospital discharge are substantially disabled. 39 Most strokes occur in the first 2 to 3 days after coronary artery bypass surgery, but they may continue with increased frequency during the first 2 postoperative weeks. 8, 39, 40 Stroke occurs more frequently when valvular heart surgery is combined with coronary artery bypass graft operations. 41, 42 This increase is due to the additional risk of cerebral macroemboli with operations that require opening a heart chamber and removal or repair of diseased mitral or aortic valves. Imaging and clinical studies, including cerebral arteriograms, suggest that the main cause of cerebral infarction with either coronary artery bypass surgery or valvular heart surgery is embolization and not hypoperfusion. 40, 43 Diffusion-weighted brain MRI is more sensitive that computed tomography (CT) for detecting acute stroke after heart surgery and, when combined with measurements of the apparent diffusion coefficient, distinguishes acute from chronic ischemic lesions. 44 Intra-arterial thrombolysis for stroke occurring 1 to 14 days after heart surgery has been reported as a potential treatment option that merits further study. 45 - 47
Intracranial hemorrhage is an infrequent cause of stroke, but its rapid identification is important so that surgical evacuation can be undertaken if necessary. 40, 48, 49 Hematomas may be located in the brain parenchyma or in subdural or epidural spaces. Intracranial hemorrhage may be related to reduced platelet adhesion and coagulation factors during cardiopulmonary bypass.
The global encephalopathy that can follow heart surgery varies from coma to confusion or delirium with impaired cognition. Stupor or coma after uncomplicated surgery is infrequent, occurring in less than 1 percent of patients. 34 It may be due to global anoxia-ischemia, massive stroke, or multiple brain infarctions. Postoperative hyponatremic encephalopathy is important to recognize and reverse because it can lead to brain damage and death, particularly in younger women. 50 Additional, rarely reported causes of encephalopathy or coma include hypoglycemia, 51 a hypernatremic hyperosmolar state, 52 and acute obstructive hydrocephalus. 53
When anoxia-ischemia is the cause of coma, myoclonus, at times accompanied by seizures, may be prominent. Recurrent postanoxic myoclonus and seizures are often poorly responsive to anticonvulsant therapy. The outcome in these comatose patients is usually extremely poor, with only a rare patient making a meaningful recovery. 54
Clinical assessment identifies confusion or delirium after cardiac surgery in greater than 8 percent of patients. 55, 56 Its prevalence is even greater in patients older than 65 to 70 years and in patients with known preexisting cerebrovascular disease. 57 Confusion and delirium after cardiac surgery increase postoperative morbidity and prolong postoperative hospitalization.
These encephalopathic patients may be slow to emerge from anesthesia, are often agitated, and have fluctuating moderate to severe impairment of cognitive function. Hallucinations may be present, and occasionally there are bilateral Babinski signs. Improvement often occurs during the first postoperative week. In comparison, patients who are matched for age and clinical condition but who have major surgery for peripheral vascular disease without cardiopulmonary bypass rarely develop such transient impairment of intellectual function. 3 Some confused patients have multiple, acute, small ischemic brain lesions detected with diffusion-weighted MRI, suggesting multiple emboli. 44 However, in many patients, the cause of confusion or delirium cannot be clinically defined. Coronary artery bypass grafting without the use of cardiopulmonary bypass results in less frequent postoperative delirium, whereas prolonged operating time increases its frequency. 55 Therefore, exposure to cardiopulmonary bypass appears to be a contributing factor to a transient encephalopathic state in otherwise uncomplicated cardiac surgery.
Acute psychosis after open heart surgery has been attributed to a situational psychiatric reaction if the level of alertness and memory remain intact. 58 When the latter processes are also impaired, the psychotic behavior has been called an organic delirium. In patients undergoing cardiac surgery, this differentiation may be incorrect. When the psychotic response has cleared and neuropsychological testing is performed, both groups have similar, multiple cognitive impairments compared to patients without neurobehavioral complications. 59 The diagnosis of an intensive care unit psychosis is usually restricted to reactions that begin 2 to 5 days after surgery, are associated with preserved memory and alertness, and rapidly resolve after treatment with neuroleptic agents or discharge from the intensive care unit. Psychotic reactions that occur during the first 48 postoperative hours in a previously stable patient probably represent a behavioral response to an anoxic-ischemic insult associated with cardiac surgery and cardiopulmonary bypass. 58, 60
Seizures may accompany coma, encephalopathy, or delirium, or they may occur independently after cardiac surgery. 6, 34 They occur in fewer than 1 percent of patients, usually early in the postoperative period and often within the first 24 hours. Tonic-clonic or partial motor seizures are clinically apparent, but partial complex seizures in an encephalopathic patient may be difficult to recognize clinically. Choreoathetosis after heart surgery, a complication that occurs mainly in children, sometimes raises the question of a seizure disorder. 61 Nonconvulsive status epilepticus may occur with stroke complicating cardiac surgery and will then contribute to a prolonged confusional state that is treatable with anticonvulsant drugs. 62 Therefore, the electroencephalographic evaluation of patients with a persistent encephalopathy may be valuable.

Peripheral Nerve Disorders
The brachial plexus and phrenic nerves are the most frequent peripheral nerves injured during cardiac surgery. A polyneuropathy may also occur under certain circumstances.
A persistent brachial plexopathy after median sternotomy has been reported to occur in more than 5 percent of patients. 1, 15, 16 Transient and minor brachial plexus injuries may be even more common. 63 Most frequently, the lower trunk of the brachial plexus is injured. Therefore, the intrinsic hand muscles are often most severely impaired, and the triceps reflex may be decreased in the affected arm. Sensory loss is sometimes present in the affected hand. Pain is prominent in some patients, and a minority have Horner’s syndrome. Injuries of the upper brachial plexus can also occur but are less frequent. Although not life-threatening and usually reversible in 1 to 3 months, a brachial plexus injury may produce permanent disability, particularly if it affects the dominant hand or produces intractable causalgia.
The plexus injuries may be due to torsional traction or compression during the open heart surgery. 16, 63 Intraoperative electrophysiological monitoring of upper extremity sensory nerve conduction reveals significant disturbances during sternal retraction in the majority of patients. This intraoperative monitoring technique can detect functional disorders of the brachial plexus during surgery, predict postoperative nerve injury, and identify intraoperative factors that predispose to brachial plexus injury. 64, 65 Brachial plexus injuries may be reduced by minimizing the opening of the sternal retractor, placing the retractor in the most caudal location, and avoiding asymmetric traction. 16
Unilateral phrenic nerve injuries with hemidiaphragmatic paralysis occur in at least 10 percent of patients during open heart surgery. 18, 66, 67 The location of the phrenic nerve adjacent to the pericardium makes it particularly vulnerable to injury from hypothermia associated with topical cold cardioplegia, as well as injury from manipulation and ischemia. Unilateral phrenic nerve injury causes atelectasis and inspiratory muscle weakness, predisposing to postoperative respiratory complications. Phrenic nerve injury in patients with preoperative chronic obstructive pulmonary disease adds particularly to postoperative morbidity. In most patients, however, morbidity is low. Some recovery is usually evident by about 6 months after injury, but there may be a more protracted course consistent with axonal injury and regeneration. Severe, bilateral phrenic nerve injury is a rare complication of heart surgery and leads to prolonged mechanical respiration. 68
Mononeuropathies resulting from compression or trauma during surgery may involve the accessory, facial, lateral femoral cutaneous, peroneal, radial, recurrent laryngeal, saphenous, long thoracic, and ulnar nerves. 15, 69 - 72 A recurrent laryngeal nerve injury with vocal cord paralysis and a persistent peroneal neuropathy can be disabling. 15 Ischemia to the cochlea-auditory nerve can result in severe hearing loss. 73 Most compressive mononeuropathies, however, are transient. This reversibility, usually within 4 to 8 weeks, may reflect the focal selective injury to myelin, with relative sparing of nerve axons, which occurs with compression neuropathies. 74 Awareness of possible intraoperative compression sites helps to prevent these complications.
Diffuse paralysis as a result of the Guillain–Barré syndrome may follow otherwise uncomplicated cardiac surgery as well as other surgical procedures. 75 Persistent paralysis also occurs after cardiac surgery in critically ill patients who have renal failure and require days of vecuronium to facilitate mechanical respiration. 76 If heart surgery is complicated by sepsis and multiorgan failure lasting for more than a week, a “critical illness” polyneuropathy and myopathy may develop, with difficulty in weaning from a respirator, distal weakness, and reduced tendon reflexes. 77

Neuro-ophthalmological Disorders
Visual disorders from cardiac surgery are frequent but usually asymptomatic and reversible. Retinal disorders include multifocal areas of retinal nonperfusion in almost all patients, cotton wool spots consistent with retinal infarctions in 10 to 25 percent of patients, and visualization of retinal emboli in fewer individuals. These retinal disorders are infrequently associated with reduced visual acuity. 5, 78
An anterior ischemic optic neuropathy is an uncommon, disabling complication of heart surgery. It produces infarction of the optic nerve head, optic disc swelling with a painless and usually permanent decrease in visual acuity. An anterior ischemic optic neuropathy may produce a monocular altitudinal, arcuate, or central scotoma. 78 A retrobulbar or posterior ischemic optic neuropathy due to infarct of the intraorbital nerve is even less common. 79 It produces acute blindness without optic disc swelling accompanied by impaired pupillary reactions. 80 Both the anterior and posterior ischemic optic neuropathies may produce unilateral or bilateral blindness. 79
Homonymous visual field defects occur with focal ischemic injury of the visual cortex or retrochiasmal visual pathways. An occasional patient is found to be cortically blind after heart surgery, usually from bilateral ischemia of the occipital cortex. Retinal and pupillary examination are both normal in patients with cortical blindness. Some of these patients deny any visual impairment. At least partial recovery from cortical blindness is possible. 5, 78
Horner’s syndrome occurs in association with injuries to the lower brachial plexus and may result from concomitant injury to the preganglionic sympathetic fibers that travel through the eighth cervical and first thoracic ventral roots. 5 It also develops in the postoperative period in hypertensive and diabetic patients, presumably due to ischemic injury to sympathetic fibers. 81
Gaze deviations, gaze paralysis, and dysconjugate gaze may occur in postoperative patients who have a brainstem or large hemispheric stroke involving eye movement systems. Intermittent gaze deviation with nystagmoid movements raises concerns about postoperative focal seizures. 82
Pituitary apoplexy resulting from acute hemorrhage or infarction of a pituitary adenoma is a rare complication of cardiopulmonary bypass. 83 The pituitary tumor is usually not recognized prior to surgery and is particularly susceptible to the ischemic and hemorrhagic risks associated with cardiopulmonary bypass. After heart surgery, patients awake with headache, ptosis, ophthalmoplegia, and visual impairment from compression of the adjacent cranial nerves and the anterior visual pathways. Transsphenoidal surgical decompression has been used safely in some patients. Infarction of a normal pituitary during coronary artery bypass surgery also occurs, may initially be silent, and leads to panhypopituitarism. 84
Visual hallucinations solely on eye closure have been reported following cardiovascular surgery. 85, 86 Patients are otherwise fully alert and lucid and can stop the hallucinations simply by opening their eyes. Atropine or lidocaine toxicity and complex partial seizures have been associated with such hallucinations.

Neuropsychological studies of cognitive function before and after cardiac surgery have identified both a transient early and a subsequent late decline in cognitive function occurring after heart surgery. 7, 87 - 91 The early postoperative changes in cognition have been shown by comparing repetitive neuropsychological test results in patients undergoing coronary artery surgery with extracorporeal circulation to nonsurgical control subjects. Performance declines on tests of attention, visuospatial ability, and memory 3 days after coronary artery bypass surgery compared to preoperative testing. A similar decline does not occur in age- and gender-matched nonsurgical control subjects. The impaired neurocognitive performance, however, typically returns to the preoperative level within weeks. 87, 89 - 92 Although numerous factors may contribute to this transient postoperative cognitive impairment, direct evidence of a specific etiology in individual patients is often lacking. 7
A late decline in cognitive function occurs in the interval from 1 to 5 years after coronary artery bypass surgery. 88, 90, 93 The cause of this late decline is unproven, in part because few of the longitudinal studies included control groups. 92 One postulated etiology is that diffuse brain microemboli associated with extracorporeal circulation injure a neuronal reserve that is needed to compensate for brain aging and to prevent dementia. 94, 95 Transcranial Doppler ultrasonography of the middle cerebral artery and carotid artery can detect microemboli during heart surgery. During cardiopulmonary bypass, there is a continuous dissemination of brain microemboli producing microvascular occlusions followed by reperfusion. 78, 94, 96 - 98 The washout of brain emboli and reperfusion may be impaired if there is concurrent systemic or localized hypoperfusion. 99
Patients with a high total microemboli count during heart surgery have a significantly higher frequency of neuropsychological test deficits than patients with low microemboli counts. Patients with long cardiopulmonary bypass durations also have a higher total microemboli count and higher frequency of neuropsychological decline. 100 - 102 If extracorporeal circulation does lead to a late decline in cognitive performance, then patients with off-pump coronary surgery on the beating heart should have less of a late decline. This comparative study is pending. 103
Evidence exists against the belief that disseminated brain microemboli from the extracorporeal circulation account for the late cognitive decline. Cognitive function 5 years after patients are randomly assigned to undergo either coronary surgery or angioplasty is similar. 104 Cognitive performance at 3 years is also similar in patients receiving on-pump coronary artery bypass surgery and a nonsurgical control group with coronary artery disease. 103 One study with a small number of patients in which individuals with preexisting neurological or psychiatric diseases or impaired cognition were excluded showed no late decline in cognitive performance after 3 to 5 years. These patients also had very good vascular risk factor control over the interval of neuropsychological testing. 105 A case-control study found a similar incidence of coronary artery bypass surgery in control subjects and patients with dementia, including a subgroup with a diagnosis of Alzheimer’s disease. Coronary artery bypass surgery does not appear to be a major risk factor for dementia. 106
A slow accumulation of microvascular brain ischemia due to vascular risk factors is an alternative explanation for the late decline in neurocognitive performance after cardiac surgery. Elderly subjects with asymptomatic ischemic lesions on brain imaging who have not had heart surgery have a greater decline in cognitive function over a period of 3 to 4 years than individuals without ischemic lesions. 107 - 109 Similarly, subjects with symptomatic cerebrovascular disease have increased progressive cognitive decline compared to control subjects. 110, 111 Patients undergoing coronary artery bypass grafts typically have vascular risk factors for asymptomatic and symptomatic brain lesions that are associated with cognitive decline. 112 It would be valuable to know whether very good control of vascular risk factors eliminates the late decline in cognitive performance that follows heart surgery. 7, 105

Several preoperative factors have been identified as placing a patient at a higher risk of a neurological complication ( Table 3-1 ). Increasing older age is associated with increasing frequency of neurological and cognitive disorders following coronary artery bypass surgery. 6, 113, 114 A multicenter, prospective study of 5,000 patients found that the occurrence of stroke was 1 percent in patients younger than 50 years of age, almost 2 percent in patients aged 50 to 59 years, approaching 4 percent in patients aged 60 to 69 years, and greater than 5 percent in patients older than 70 years. 115 With a growing elderly population, the number of patients older than 80 years who are evaluated for coronary artery bypass grafts may increase. 116
TABLE 3-1 Risk Factors for Cerebral Ischemia During Cardiac Surgery
Older age
Atheromatous aorta
Diabetes mellitus
Previous stroke
Carotid artery disease
Prolonged cardiopulmonary bypass
High count of cerebral microemboli
Combined coronary artery bypass and valvular heart surgeries
Large hemodynamic fluctuations
Atrial fibrillation
Dislodgment of prothrombotic atheroma during instrumentation of the aorta is a risk factor for stroke. 14, 32, 117 - 120 Atheromatous aortic disease can be identified with intraoperative ultrasonographic scanning and transesophageal echocardiography. Approximately 25 to 50 percent of patients receiving a coronary artery bypass graft have atherosclerotic plaques in their ascending aorta identified by these techniques. The frequency of such aortic disease increases with age and is particularly prominent in patients older than 70 years. Identification of a moderately to severely atheromatous aorta may alter surgical management. 118, 121
A preoperative history of hypertension, diabetes mellitus, stroke, severe stenosis of the carotid artery (>70%), and other markers of vascular disease are also risk factors for stroke following coronary artery bypass surgery. 6, 8, 38, 114, 115, 122 Cardiac surgery within 3 months of a stroke may carry a risk of worsening preoperative neurological deficits. 123 The greater the number of preoperative risk factors, the higher is the probability of a perioperative stroke. For example, a 65- to 75-year-old patient with a history of a stroke and hypertension has a risk of postoperative stroke that is three times greater than that of a patient of the same age without a history of stroke or hypertension. A patient older than 75 years with a history of stroke and hypertension has a probability for postoperative stroke that is 13 times greater than a patient younger than 65 years with no previous stroke or history of hypertension. Stroke risk indexes may identify patients prior to coronary artery bypass surgery who are at high risk of a perioperative stroke. 124, 125
Intraoperative factors also influence the frequency of stroke ( Table 3-1 ). As noted previously, individuals who require coronary artery bypass surgery combined with a left-sided intracardiac procedure have a relatively high rate of stroke. 38, 126 Patients who require cardiopulmonary bypass lasting more than 2 hours have a higher number of intraoperative cerebral microemboli detected by transcranial Doppler ultrasound monitoring and a higher frequency of stroke. 38, 115, 127, 128 A large fluctuation in hemodynamic parameters during surgery, such as mean arterial pressure, has been associated with postoperative stroke and encephalopathy. 129 The risk from sustained intraoperative hypotension with a mean arterial pressure below 40 to 50 mmHg during cardiopulmonary bypass remains unclear.
Atrial fibrillation occurs in approximately one third of continuously monitored patients following cardiac surgery and is a risk factor for stroke. Its initial occurrence is most common during the first 3 postoperative days, and 20 percent of patients have more than one episode. Advancing age is a risk factor for atrial fibrillation, and patients older than 70 years are at high risk of arrhythmia. Withdrawal from β-adrenergic receptor–blocking agents or angiotensin-converting enzyme inhibitors is also associated with recurrent atrial fibrillation. Use of β-blocking drugs or angiotensin-converting enzyme inhibitors preoperatively and postoperatively and β-blockers postoperatively is associated with a reduced risk of atrial fibrillation. 25, 115, 122, 130, 131

Attempts to prevent neurological sequelae after cardiac surgery have focused on improved surgical and cardiopulmonary bypass techniques and neuroprotective drugs. 8, 132, 133
Identification of surgical and technical factors that carry particular risks of neurological complications after cardiac surgery has led to the adoption of preventive measures. 8, 13, 132 An arterial line microfilter system has been incorporated into the extracorporeal circulation with the aim of reducing cerebral embolization. Improved surgical techniques reduce the bypass time and the total number of cerebral microemboli. Maintenance of the mean arterial blood pressure above 50 mmHg provides a safety margin against periods of relative hypoperfusion. The use of a membrane oxygenator decreases the magnitude of air emboli. Preoperative or early postoperative administration of β-blocking agents decreases the incidence of postoperative atrial fibrillation. Early postoperative use of aspirin decreases ischemic complications of multiple organs including the brain. 134 Perioperative monitoring and control of hyperglycemia may influence outcome. 132, 135 Delaying heart surgery for 4 or more weeks after a recent stroke has been recommended if the cardiac condition allows such a delay. 132
The benefit of combined carotid revascularization and coronary artery bypass surgery in patients with asymptomatic carotid stenosis awaits confirmation in a prospective clinical trial. 114, 136 - 139 Carotid endarterectomy for patients with severe (70% to 99%) internal carotid stenosis that has been neurologically symptomatic in the past 6 months is of established benefit independent of cardiac surgery. 140 Performing a carotid endarterectomy before or simultaneous with coronary artery bypass surgery in such patients is an accepted practice. 132, 141, 142 Carotid stenting has evolved as an alternative procedure for such patients. 8, 138, 139, 143, 144
An increased concern that an atheromatous aorta is a primary source of embolic stroke has led to intraoperative identification with transesophageal echocardiography and epiaortic scanning. 14 The presence of prominent aortic atheroma alters surgical techniques including the site of aortic cannulation, the aorta clamping technique, the use of intra-aortic filtration, and using off-pump coronary artery surgery to avoid manipulation of a severely diseased aorta. 28, 32, 118, 132, 145, 146
The magnitude of cerebral microemboli and the frequent neuropsychological and anoxic-ischemic findings associated with cardiac surgery suggest a need and opportunity to study brain protective agents. Proposed mechanisms of pharmacological neuroprotection include decreasing cerebral oxygen consumption; decreasing cerebral blood-flow and, with it, the total number of microemboli; interrupting the cascade of cerebral ischemic events that are mediated via excitotoxins such as glutamate; and decreasing the inflammatory response and coagulation cascade associated with cardiopulmonary bypass. 133, 147 - 151 Clinical trials, however, have yet to identify an effective pharmacological neuroprotective agent that has wide clinical application during coronary artery bypass surgery.

Cardiac transplantation is an established treatment for selected patients with progressive, preterminal heart failure. Cardiac transplantation centers now report survival rates at 1 year of greater than 80 to 85 percent, at 5 years of 60 to 80 percent, at 10 years of approximately 50 percent, and at 15 years of 30 to 40 percent. 152, 153 The annual number of heart transplantations worldwide is estimated to be greater than 4,000. 153 Neurological sequelae occurring either in the perioperative period or as a late complication may negate an otherwise successful heart transplantation. The early identification of treatable complications offers the best opportunity to prevent severe disability.
The perioperative neurological sequelae from cardiac transplantation are similar to the complications associated with valvular or bypass graft surgery, dis cussed previously, except that neurological complications occur more frequently in transplant recipients. 41 Postoperative encephalopathy, stroke, headaches, psychosis, seizures, and peripheral nerve disorders are the most common problems. 154 - 160
Vascular headaches accompanied by nausea and vomiting may occur in the first week after transplantation. 156 The headaches are associated with a rapid shift from low preoperative to high postoperative mean arterial pressures. Similar headaches may rarely precede an intraparenchymal hemorrhage. These vascular headaches respond to β-adrenergic receptor–blocking agents.
Seizures have been reported in as many as 15 percent of patients with cardiac transplants. 157 They commonly occur during the perioperative period. They also occur as a side effect of cyclosporine or as a late complication of a brain infection or tumor. Seizures in the perioperative period are usually due to stroke and may not require long-term anticonvulsant therapy. When anticonvulsant drugs are indicated, phenytoin, phenobarbital, and carbamazepine are often avoided because they induce the hepatic metabolism of cyclosporine, tacrolimus, and sirolimus. When these antiepileptic agents are used, immunosuppression may be reduced. Levatiracetam and gabapentin have negligible hepatic enzyme–inducing effects and few drug interactions and may be preferred anticonvulsants. 161 Pregabalin, a newer antiepileptic drug, has similar characteristics.
Psychotic behavior with hallucinations, delusional thought processes, and disorganized behavior can occur during the first 2 weeks after transplantation or as a late complication. When it occurs during the postoperative period, multiple causal factors may be present, but with time, the psychotic behavior usually resolves. When psychotic behavior occurs as a late complication, it is often a manifestation of an intracranial infection, most commonly viral. A thorough neurological evaluation is therefore indicated when a cardiac transplant recipient develops an acute late psychosis. 154, 155
Immunosuppression remains a cause of late neurological complications after cardiac transplantation. Opportunistic infections can occur as early as 2 weeks after surgery and immunosuppression, but usually there is an interval of at least 1 month. Focal meningoencephalitis or brain abscess, meningitis, and diffuse encephalitis are three common presentations of infections in cardiac transplant recipients. 154, 155, 162 Aspergillus fumigatus, Toxoplasma gondii, Cryptococcus neoformans, Listeria monocytogenes, and herpesvirus infections are the more common causes of CNS infections in heart transplant recipients. 162, 163
Aspergillosis is the most frequent fungal infection that produces a necrotizing meningoencephalitis and single or multiple brain abscesses. 162 Cerebral aspergillosis is almost always disseminated from a preceding pulmonary infection. The abscesses may have ring, irregular, or no contrast enhancement on MRI and CT scans. 164 Diffusion-weighted MRI may demonstrate restricted diffusion of water in the center of the abscess. 165 Aspergillosis also causes an invasive necrosis of intracranial vessels that may lead to hemorrhagic infarction. Therefore, focal hemorrhage on brain imaging is suggestive of Aspergillus infection. The diagnosis is confirmed by direct needle aspiration or biopsy; cerebrospinal fluid studies and cultures usually are not helpful. If the diagnosis is made late, the disease is fatal; early diagnosis and treatment in an immunosuppressed patient, however, improve survival. 166, 167
T. gondii is the second most common cause of focal or multifocal meningoencephalitis and abscess formation following cardiac transplantation. 162, 168 It can produce multiple ring-enhancing lesions, seen with contrast CT scans. MRI may demonstrate additional lesions not apparent on CT and may also show a rapid response to antibiotic therapy. Serological evidence of T. gondii is supportive evidence, particularly if there is seroconversion after transplantation or an increase in titer compared to the preoperative baseline serology. 168 While tissue diagnosis with material aspirated from a brain abscess is diagnostic, a presumptive diagnosis based on imaging and serological testing may lead to a therapeutic trial. Consideration of the diagnosis is mandatory because of the excellent therapeutic response to antitoxoplasmic antibiotics. 169 Toxoplasmosis may also cause an inflammatory myopathy in association with intracranial and multisystem infection. 170, 171
Other less frequent opportunistic infections that produce focal meningoencephalitis or brain abscess include the rhinocerebral phycomycotic organisms, Candida albicans in the setting of disseminated candidiasis, Nocardia infections, Klebsiella (abscess), and Rhodococcus equi. 155, 162, 172 - 175
Meningitis after cardiac transplantation is most commonly due to C. neoformans when the symptoms are subacute or chronic and the white blood count in the cerebrospinal fluid is mildly to moderately elevated with predominantly mononuclear cells. L. monocytogenes is the most common organism when the symptoms are acute and there is a prominent cerebrospinal fluid pleocytosis consisting of polymorphonuclear and mononuclear cells. Coccidioides immitis and Pseudoallescheria boydii, as well as previously mentioned fungi, are less frequent causes of meningitis. 154, 157, 176, 177
Cytomegalovirus, herpes simplex, and herpes zoster encephalitis also occur, in association with a disseminated viremia, in patients who have undergone cardiac transplantation. 157, 163, 178 Immunosuppression, however, transforms the acute necrotizing focal herpes simplex encephalitis into a more diffuse and slowly progressive process. Progressive multifocal leukoencephalopathy after heart transplantation is thought to be due to reactivation of a JC virus infection that is initially acquired during childhood. 179 Pathogens can also be transferred from donor to transplant recipients, typically causing neurological deterioration during the first post-transplantation month. West Nile virus has been transferred from a donor heart to a heart recipient causing an encephalitis shortly after transplantation. 180 Rabies virus, lymphocytic chorimeningitis virus, and Trypanosoma cruzi are also reported donor-derived infections. 181 - 183
Immunosuppression for cardiac transplantation combined with Epstein–Barr virus infection can cause a post-transplantation lymphoproliferative disorder that leads to systemic malignant lymphoproliferation including involvement of the brain. 184 Post-transplantation lymphoproliferative disorders may regress with reduction of immunosuppressive therapy or they may require radiotherapy. 185, 186 The CNS can be the only site of malignant lymphoproliferation in association with Epstein–Barr virus in brain tissue. The response of this post-transplantation primary CNS lymphoma to multimodal therapies is often poor. 184, 187 Glioma is another isolated brain tumor that can occur after heart transplantation, although the relationship to immunosuppression is unclear. 188
Immunosuppressive agents can also cause neurological side effects more directly. Prior to the use of cyclosporine, high-dose prednisone in combination with azathioprine was commonly used. The main side effect of the prednisone was weakness of the proximal lower extremities, osteoporosis with lower thoracic and lumbosacral compression fractures, or psychiatric complications. 176 With the use of calcineurin inhibitors, cyclosporine, and tacrolimus, the dose of prednisone has been lowered, thereby reducing its side effects. 189, 190 Cyclosporine and tacrolimus themselves, however, may have neurotoxic side effects, including prominent tremor, headache, a lowered seizure threshold, paresthesias, mental confusion, acute mania, weakness, ataxia, dysarthria, visual hallucinations, and cortical blindness. Brain imaging may reveal a posterior leukoencephalopathy. Diffusion-weighted MRI studies suggest that the onset of neurotoxicity is due to vasogenic brain edema. Vasogenic edema is reversible, which is consistent with the typical remission of adverse effects and MRI findings following reduction of the cyclosporine or tacrolimus dose. Prolonged drug exposure at neurotoxic levels, however, may cause residual neurological impairment. 191 - 193 The newer immunosuppressive agent sirolimus has few reported neurotoxic effects and may be used as an alternative to cyclosporine or tacrolimus when neurotoxicity occurs. 194
Cyclosporine also induces gout and produces chronic nephrotoxicity. 190 When gout is treated with colchicines, the impaired renal function may lead to colchicine toxicity and a peripheral neuromuscular disorder that improves when the colchicine is stopped. 195
The monoclonal anti-CD3 antibody (OKT3) is used to prevent and treat graft rejection following cardiac transplantation. Aseptic meningitis with fever, headache, seizures, and a variable encephalopathy is reported to occur in 5 percent of patients as a reaction to it. This aseptic meningitis may occur during the course of OKT3 therapy or in the weeks immediately subsequent to it. The aseptic meningitis and encephalopathy resolve within days of onset. 176, 196, 197
As noted previously, most of the neurological complications of cardiac transplantation with immunosuppression may present with a confusional state in which headache and focal neurological findings may be present or absent. It is not uncommon, however, for more than one complication of immunosuppression to cause symptoms in an individual cardiac transplant recipient. 155


1 Shaw PJ, Bates D, Cartlidge NEF, et al. Neurological complications of coronary artery bypass graft surgery. BMJ . 1986;293:165.
2 Shaw PJ, Bates D, Cartlidge NEF, et al. Early intellectual dysfunction following coronary bypass surgery. QJM . 1986;58:59.
3 Shaw PJ, Bates D, Cartlidge NEF, et al. Neurologic and neuropsychological morbidity following major surgery: comparison of coronary artery bypass and peripheral vascular surgery. Stroke . 1987;18:700.
4 Shaw PJ, Bates D, Cartlidge NEF, et al. Long-term intellectual dysfunction following coronary artery bypass graft surgery. QJM . 1987;62:259.
5 Shaw PJ, Bates D, Cartlidge NEF, et al. Neuro-ophthalmological complications of coronary artery bypass graft surgery. Acta Neurol Scand . 1987;76:1.
6 Roach GW, Kanchuger M, Mangano CM, et al. Adverse cerebral outcomes after coronary artery bypass surgery. N Engl J Med . 1996;335:1857.
7 Selnes OA, McKhann GM, Borowicz LMJr, et al. Cognitive and neurobehavioral dysfunction after cardiac bypass procedures. Neurol Clin . 2006;24:133.
8 McKhann GM, Grega MA, Borowicz LMJr, et al. Stroke and encephalopathy after cardiac surgery: an update. Stroke . 2006;37:562.
9 McKhann GM, Goldsborough MA, Borowicz LM, et al. Predictors of stroke risk in coronary artery bypass patients. Ann Thorac Surg . 1997;63:516.
10 Thom T, Haase N, Rosamond W, et al. Heart disease and stroke statistics—update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation . 2006;113:e85.
11 Seines O, Goldsborough M, Borowicz L, et al. Neurobehavioral sequelae of cardiopulmonary bypass. Lancet . 1999;353:1601.
12 Murkin JM, Baird DL, Martzke JS, et al. Long-term neurological and neuropsychological outcome 3 years after coronary artery bypass surgery. Anesth Analg . 1996;82:328S.
13 Kirklin J, Barratt-Boyes B. Cardiac Surgery: Morphology, Diagnostic Criteria, Natural History, Techniques, Results and Indications, 2nd Ed. New York: Churchill Livingstone, 1993.
14 Djaiani G, Fedorko L, Borger M, et al. Mild to moderate atheromatous disease of the thoracic aorta and new ischemic brain lesions after conventional coronary artery bypass graft surgery. Stroke . 2004;35:e356.
15 Lederman RJ, Breuer RC, Hanson MR, et al. Peripheral nervous system complications of coronary artery bypass graft surgery. Ann Neurol . 1982;12:297.
16 Vahl CF, Carl I, Muller-Vahl H, et al. Brachial plexus injury after cardiac surgery. J Thorac Cardiovasc Surg . 1991;102:724.
17 Katz MG, Katz R, Schachner A, et al. Phrenic nerve injury after coronary artery bypass grafting: will it go away? Ann Thorac Surg . 1998;65:32.
18 Dimopoulou I, Daganou M, Dafni U, et al. Phrenic nerve dysfunction after cardiac operations. Chest . 1998;113:8.
19 Cohen AJ, Katz MG, Katz R, et al. Phrenic nerve injury after coronary artery grafting: is it always benign? Ann Thorac Surg . 1997;64:148.
20 Mclean RF, Wong BI. Normothermic versus hypothermic cardiopulmonary bypass. J Cardiothorac Vasc Anesth . 1996;10:45.
21 Engleman R, Pleet A, Rousou J, et al. Influence of cardiopulmonary bypass perfusion temperature on neurologic and hematologic function after coronary artery bypass grafting. Ann Thorac Surg . 1999;67:1547.
22 Murphy GJ, Angelini GD. Side effects of cardiopulmonary bypass: what is the reality? J Card Surg . 2004;19:481.
23 Butler J, Rocker GM, Westaby S. Inflammatory response to cardiopulmonary bypass. Ann Thorac Surg . 1993;55:552.
24 Nakajima M, Tsuchiya K, Kanemaru K, et al. Subdural hemorrhagic injury after open heart surgery. Ann Thorac Surg . 2003;76:614.
25 Almassi GH, Schowalter T, Nicolosi AC, et al. Atrial fibrillation after cardiac surgery: a major morbid event? Ann Surg . 1997;226:501.
26 Mathew J, Parks R, Savino J, et al. Atrial fibrillation following coronary artery bypass graft surgery. JAMA . 1996;276:300.
27 Harris DNF, Baily SM, Smith PLC, et al. Brain swelling in the first hour after coronary artery bypass surgery. Lancet . 1993;342:586.
28 Sellke FW, DiMaio JM, Caplan LR, et al. Comparing on-pump and off-pump coronary artery bypass grafting: numerous studies but few conclusions: a scientific statement from the American Heart Association council on cardiovascular surgery and anesthesia in collaboration with the interdisciplinary working group on quality of care and outcomes research. Circulation . 2005;111:2858.
29 Lund C, Hol PK, Lundblad R, et al. Comparison of cerebral embolization during off-pump and on-pump coronary artery bypass surgery. Ann Thorac Surg . 2003;76:765.
30 van Dijk D, Moons KG, Keizer AM, et al. Association between early and three month cognitive outcome after off-pump and on-pump coronary bypass surgery. Heart . 2004;90:431.
31 Lund C, Sundet K, Tennoe B, et al. Cerebral ischemic injury and cognitive impairment after off-pump and on-pump coronary artery bypass grafting surgery. Ann Thorac Surg . 2005;80:2126.
32 Sharony R, Grossi EA, Saunders PC, et al. Propensity case-matched analysis of off-pump coronary artery bypass grafting in patients with atheromatous aortic disease. J Thorac Cardiovasc Surg . 2004;127:406.
33 Shaw PJ, Bates D, Carlidge NEF, et al. An analysis of factors predisposing to neurological injury in patients undergoing coronary bypass operations. QJM . 1989;72:633.
34 Coffey CE, Massey EW, Roberts KB, et al. Natural history of cerebral complications of coronary artery bypass graft surgery. Neurology . 1983;33:1416.
35 McKhann GM, Goldsborough MA, Borowicz LM, et al. Cognitive outcome after coronary artery bypass: a one-year prospective study. Ann Thorac Surg . 1997;63:510.
36 Redmond JM, Greene PS, Goldsborough MA, et al. Neurologic injury in cardiac surgery patients with a history of stroke. Ann Thorac Surg . 1996;61:42.
37 Tuman KJ, McCarthy RJ, Najfi H, et al. Differential effects of advanced age on neurologic and cardiac risks of coronary artery operations. J Thorac Cardiovasc Surg . 1992;104:1510.
38 Bucerius J, Gummert JF, Borger MA, et al. Stroke after cardiac surgery: a risk factor analysis of 16,184 consecutive adult patients. Ann Thorac Surg . 2003;75:472.
39 Salazar JD, Wityk RJ, Grega MA, et al. Stroke after cardiac surgery: short- and long-term outcomes. Ann Thorac Surg . 2001;72:1195.
40 Likosky DS, Marrin CA, Caplan LR, et al. Determina-tion of etiologic mechanisms of strokes secondary to coronary artery bypass graft surgery. Stroke . 2003;34:2830.
41 Inque K, Luth JU, Pottkamper KM, et al. Incidence and risk factors of perioperative cerebral operations: heart transplantation compared to coronary artery bypass grafting and valve surgery. J Cardiovasular Surg . 1998;39:201.
42 Goldsborough MA, Borowicz LM, Mckhann GM, et al. Variation in stroke occurrence by cardiac procedures. Perfusion . 1997;12:47.
43 Hise JH, Nipper ML, Schnitker JC. Stroke associated with coronary artery bypass surgery. AJNR Am J Neuroradiol . 1991;12:811.
44 Wityk RJ, Goldsborough MA, Hillis A, et al. Diffusion- and perfusion-weighted brain magnetic resonance imaging in patients with neurologic complications after cardiac surgery. Arch Neurol . 2001;58:571.
45 Katzan IL, Masaryk TJ, Furlan AJ, et al. Intra-arterial thrombolysis for perioperative stroke after open heart surgery. Neurology . 1999;52:1081.
46 Moazami N, Smedira NG, McCarthy PM, et al. Safety and efficacy of intraarterial thrombolysis for perioperative stroke after cardiac operation. Ann Thorac Surg . 2001;72:1933.
47 Fukuda I, Imazuru T, Osaka M, et al. Thrombolytic therapy for delayed, in-hospital stroke after cardiac surgery. Ann Thorac Surg . 2003;76:1293.
48 Yokote H, Itakura T, Funahashi K, et al. Chronic subdural hematoma after open heart surgery. Surg Neurol . 1985;24:520.
49 Humphreys RP, Hoffman HJ, Mustard WT, et al. Cerebral hemorrhage following heart surgery. J Neurosurg . 1975;43:671.
50 Ayus JC, Arieff AI. Brain damage and postoperative hyponatremia. Neurology . 1996;46:323.
51 Criado A, Dominguez E, Carmona J, et al. Hypoglycemic coma after cardiac surgery. Crit Care Med . 1984;12:409.
52 Hiramatsu Y, Sakakibara Y, Mitsui T, et al. Clinical features of hypernatremic hyperosmolar delirium following open heart surgery. Nippon Kyobu Geka Gakkai Zasshi . 1991;39:1945.
53 Gonzalez-Santos JM, Gonzalez-Santos ML, Vallejo JL. Acute obstructive hydrocephalus: an unusual complication after cardiopulmonary bypass. Thorac Cardiovasc Surg . 1986;77:586.
54 Krumholz A, Stern B, Weiss H. Outcome from coma after cardiopulmonary resuscitation: relation to seizures and myoclonus. Neurology . 1988;38:401.
55 Bucerius J, Gummert JF, Borger MA, et al. Predictors of delirium after cardiac surgery delirium: effect of beating-heart (off-pump) surgery. J Thorac Cardiovasc Surg . 2004;127:57.
56 Eriksson M, Samuelsson E, Gustafson Y, et al. Delirium after coronary bypass surgery evaluated by the organic brain syndrome protocol. Scand Cardiovasc J . 2002;36:250.
57 Rolfson DB, McElhaney JE, Rockwood K, et al. Incidence and risk factors for delirium and other adverse outcomes in older adults after coronary artery bypass graft surgery. Can J Cardiol . 1999;15:771.
58 Dubin WR, Field HL, Gastfriend DR. Postcardiotomy delirium: a critical review. J Thorac Cardiovasc Surg . 1979;34:201.
59 Juolasmaa A, Outakoski J, Hirvenoja R, et al. Effect of open heart surgery on intellectual performance. J Clin Neuropsychol . 1981;3:181.
60 Smith LW, Dimsdale JE. Postcardiotomy delirium: conclusions after 25 years. Am J Psychiatry . 1989;146:452.
61 DeLeon S, Ilbawi M, Arcilla R, et al. Choreoathetosis after deep hypothermia without circulatory arrest. Ann Thorac Surg . 1990;50:714.
62 Fagan KJ, Lee SI. Prolonged confusion following convulsions due to generalized nonconvulsive status epilepticus. Neurology . 1990;40:1689.
63 Canbaz S, Turgut N, Halici U, et al. Brachial plexus injury during open heart surgery—controlled prospective study. Thorac Cardiovasc Surg . 2005;53:295.
64 Hickey C, Gugino LD, Aglio LS, et al. Intraoperative somatosensory evoked potential monitoring predicts peripheral nerve injury during cardiac surgery. Anesthesiology . 1993;78:29.
65 Jellish WS, Blakeman B, Warf P, et al. Hands-up positioning during asymmetric sternal retraction for internal mammary artery harvest: a possible method to reduce brachial plexus injury. Anesth Analg . 1997;84:260.
66 Katz ES, Tunik PA, Rusinek H, et al. Protruding aortic atheromas predict stroke in elderly patients undergoing cardiopulmonary bypass: experience with intraoperative transesophageal echocardiography. J Am Coll Cardiol . 1992;20:70.
67 Canbaz S, Turgut N, Halici U, et al. Electrophysiological evaluation of phrenic nerve injury during cardiac surgery—a prospective, controlled, clinical study. BMC Surg . 2004;4:2.
68 Werner RA, Geiringer SR. Bilateral phrenic nerve palsy associated with open-heart surgery. Arch Phys Med Rehabil . 1990;71:1000.
69 Marini SG, Rook JL, Green RF, et al. Spinal accessory nerve palsy: an unusual complication of coronary artery bypass. Arch Phys Med Rehabil . 1991;72:247.
70 Parsonnet V, Karasakalides A, Gielchinsky I, et al. Meralgia paresthetica after coronary bypass surgery. J Thorac Cardiovasc Surg . 1991;101:219.
71 Lavee J, Schneiderman J, Yorav S, et al. Complications of saphenous vein harvesting following coronary artery bypass surgery. J Cardiovasc Surg . 1989;30:989.
72 Tewari P, Aggarwal SK. Combined left-sided recurrent laryngeal and phrenic nerve palsy after coronary artery operation. Ann Thorac Surg . 1996;61:1721.
73 Plasse HM, Mittleman M, Frost JO. Unilateral sudden hearing loss after open heart surgery: a detailed study of seven cases. Laryngoscope . 1981;91:101.
74 Burns T. Neuropathy caused by compression, entrapment or physical injury. In: Dyck PJ, Thomas PK, editors. Peripheral Neuropathy . 4th Ed. Philadelphia: Elsevier Saunders; 2005:1391.
75 Hogan JC, Briggs TP, Oldershaw PJ. Guillain-Barré syndrome following cardiopulmonary bypass. Int J Cardiol . 1992;35:427.
76 Segredo V, Caldwell JE, Matthay MA, et al. Persistent paralysis in critically ill patients after long-term administration of vecuronium. N Engl J Med . 1992;327:524.
77 Bolton CF. Neuromuscular manifestations of critical illness. Muscle Nerve . 2005;32:140.
78 Shahian DM, Speert PK. Symptomatic visual deficits after open heart operations. Ann Thorac Surg . 1989;48:275.
79 Nuttall GA, Garrity JA, Dearani JA, et al. Risk factors for ischemic optic neuropathy after cardiopulmonary bypass: a matched case/control study. Anesth Analg . 2001;93:1410.
80 Buono LM, Foroozan R. Perioperative posterior ischemic optic neuropathy: review of the literature. Surv Ophthalmol . 2005;50:15.
81 Barbut D, Gold JP, Heinemann MD, et al. Horner’s syndrome after coronary artery bypass surgery. Neurology . 1996;46:181.
82 Tusa RJ, Kaplan PW, Hain TC, et al. Ipsiversive eye deviation and epileptic nystagmus. Neurology . 1990;40:662.
83 Cooper DM, Bazaral MG, Furlan AJ, et al. Pituitary apoplexy: a complication of cardiac surgery. Ann Thorac Surg . 1986;41:547.
84 Davies JS, Scanlon MF. Hypopituitarism after coronary artery bypass grafting. BMJ . 1998;316:682.
85 Laloux P, Osseman M. Visual hallucinations on eye closure after cardiovascular surgery. J Clin Neuroophthalmol . 1992;12:242.
86 Eissa A, Baker RA, Knight JL. Closed-eye visual hallucinations after coronary artery bypass grafting. J Cardiothorac Vasc Anesth . 2005;19:217.
87 Mullges W, Berg D, Schmidtke A, et al. Early natural course of transient encephalopathy after coronary artery bypass grafting. Crit Care Med . 2000;28:1808.
88 Newman MF, Kirchner JL, Phillips-Bute B, et al. Longitudinal assessment of neurocognitive function after coronary-artery bypass surgery. N Engl J Med . 2001;344:395.
89 Selnes OA, Grega MA, Borowicz LMJr, et al. Cognitive changes with coronary artery disease: a prospective study of coronary artery bypass graft patients and nonsurgical controls. Ann Thorac Surg . 2003;75:1377.
90 Stygall J, Newman SP, Fitzgerald G, et al. Cognitive change 5 years after coronary artery bypass surgery. Health Psychol . 2003;22:579.
91 Keith JR, Puente AE, Malcolmson KL, et al. Assessing postoperative cognitive change after cardiopulmonary bypass surgery. Neuropsychology . 2002;16:411.
92 McKhann GM, Grega MA, Borowicz LMJr, et al. Is there cognitive decline 1 year after CABG? Comparison with surgical and nonsurgical controls. Neurology . 2005;65:991.
93 Selnes OA, Royall RM, Grega MA, et al. Cognitive changes 5 years after coronary artery bypass grafting: is there evidence of late decline? Arch Neurol . 2001;58:598.
94 Moody DM, Bell MA, Challa VR, et al. Brain microemboli during cardiac surgery or aortography. Ann Neurol . 1990;28:477.
95 Brown W, Moody D, Tytell M, et al. Microembolic brain injuries from cardiac surgery: are the seeds of future Alzheimer’s disease? Ann N Y Acad Sci . 1997;826:386.
96 Blauth CI, Arnold JV, Schulenberg WE, et al. Cerebral microembolism during cardiopulmonary bypass. J Thorac Cardiovasc Surg . 1988;95:668.
97 Challa V, Lovell M, Moody D, et al. Laser microprobe mass spectrometric study of aluminum and silicon in brain emboli related to cardiac surgery. J Neuropathol Exp Neurol . 1998;57:140.
98 van der Linden J, Casimir-Ahn H. When do cerebral emboli appear during open heart operations? Ann Thorac Surg . 1991;51:231.
99 Caplan LR, Hennerici M. Impaired clearance of emboli (washout) is an important link between hypoperfusion, embolism, and ischemic stroke. Arch Neurol . 1998;55:1475.
100 Pugsley W, Klinger L, Paschalis C, et al. The impact of microemboli during cardiopulmonary bypass on neuropsychological functioning. Stroke . 1994;25:1393.
101 Stump DA, Tegeler CH, Rogers AT, et al. Neuropsychological deficits are associated with the number of emboli detected during cardiac surgery. Stroke . 1993;24:509.
102 Sylivris S, Christopher L, Matalanis G, et al. Pattern and significance of cerebral microemboli during coronary artery bypass grafting. Ann Thorac Surg . 1998;66:1674.
103 Selnes OA, Grega MA, Borowicz LMJr, et al. Cognitive outcomes three years after coronary artery bypass surgery: a comparison of on-pump coronary artery bypass graft surgery and nonsurgical controls. Ann Thorac Surg . 2005;79:1201.
104 Hlatky MA, Bacon C, Boothroyd D, et al. Cognitive function 5 years after randomization to coronary angioplasty or coronary artery bypass graft surgery. Circulation . 1997;96(9 Suppl):11.
105 Mullges W, Babin-Ebell J, Reents W, et al. Cognitive performance after coronary artery bypass grafting: a follow-up study. Neurology . 2002;59:741.
106 Knopman DS, Petersen RC, Cha RH, et al. Coronary artery bypass grafting is not a risk factor for dementia or Alzheimer disease. Neurology . 2005;65:986.
107 Vermeer SE, Prins ND, den Heijer T, et al. Silent brain infarcts and the risk of dementia and cognitive decline. N Engl J Med . 2003;348:1215.
108 Prins ND, van Dijk EJ, den Heijer T, et al. Cerebral white matter lesions and the risk of dementia. Arch Neurol . 2004;61:1531.
109 Prins ND, van Dijk EJ, den Heijer T, et al. Cerebral small-vessel disease and decline in information processing speed, executive function and memory. Brain . 2005;128:2034.
110 Sachdev PS, Brodaty H, Valenzuela MJ, et al. Progression of cognitive impairment in stroke patients. Neurology . 2004;63:1618.
111 Altieri M, Di Piero V, Pasquini M, et al. Delayed poststroke dementia: a 4-year follow-up study. Neurology . 2004;62:2193.
112 Knopman D, Boland LL, Mosley T, et al. Cardiovascular risk factors and cognitive decline in middle-aged adults. Neurology . 2001;56:42.
113 Newman MF, Wolman R, Kanchuger M, et al. Multicenter preoperative stroke risk index for patients undergoing coronary artery bypass graft surgery. Circulation . 1996;94(suppl II):74.
114 Naylor AR, Mehta Z, Rothwell PM, et al. Carotid artery disease and stroke during coronary artery bypass: a critical review of the literature. Eur J Vasc Endovasc Surg . 2002;23:283.
115 Almassi G, Sommers T, Moritz T, et al. Stroke in cardiac surgical patients: determinants and outcome. Ann Thorac Surg . 1999;63:391.
116 Smith KM, Lamy A, Arthur HM, et al. Outcomes and costs of coronary artery bypass grafting: comparison between octogenarians and septuagenarians at a tertiary care centre. CMAJ . 2001;165:759.
117 Mackensen GB, Ti LK, Phillips-Bute BG, et al. Cerebral embolization during cardiac surgery: impact of aortic atheroma burden. Br J Anaesth . 2003;91:656.
118 Hangler HB, Nagele G, Danzmayr M, et al. Modification of surgical technique for ascending aortic atherosclerosis: impact on stroke reduction in coronary artery bypass grafting. J Thorac Cardiovasc Surg . 2003;126:391.
119 van der Linden J, Hadjinikolaou L, Bergman P, et al. Postoperative stroke in cardiac surgery is related to the location and extent of atherosclerotic disease in the ascending aorta. J Am Coll Cardiol . 2001;38:131.
120 Mathew JP, Rinder HM, Smith BR, et al. Transcerebral platelet activation after aortic cross-clamp release is linked to neurocognitive decline. Ann Thorac Surg . 2006;81:1644.
121 Gaudino M, Glieca F, Alessandrini F, et al. Individualized surgical strategy for the reduction of stroke risk in patients undergoing coronary artery bypass grafting. Ann Thorac Surg . 1999;67:1246.
122 Eagle K, Guyton R, Davidoff R, et al. ACC/AHA guidelines for coronary artery bypass graft surgery: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol . 1999;34:1262.
123 Rorick MB, Furlan AJ. Risk of cardiac surgery in patients with prior stroke. Neurology . 1990;40:835.
124 McKhann GM, Grega MA, Borowicz LMJr, et al. Encephalopathy and stroke after coronary artery bypass grafting: incidence, consequences, and prediction. Arch Neurol . 2002;59:1422.
125 Charlesworth DC, Likosky DS, Marrin CA, et al. Development and validation of a prediction model for strokes after coronary artery bypass grafting. Ann Thorac Surg . 2003;76:436.
126 Boeken U, Litmathe J, Feindt P, et al. Neurological complications after cardiac surgery: risk factors and correlation to the surgical procedure. Thorac Cardiovasc Surg . 2005;53:33.
127 Barbut D, Lo Y, Gold JP, et al. Impact of embolization during coronary artery bypass grafting on outcome and length of stay. Ann Thorac Surg . 1997;63:998.
128 Libman RB, Wirkowski E, Neystat M, et al. Stroke associated with cardiac surgery. Arch Neurol . 1997;54:83.
129 Ganushchak YM, Fransen EJ, Visser C, et al. Neurological complications after coronary artery bypass grafting related to the performance of cardiopulmonary bypass. Chest . 2004;125:2196.
130 Amar D. Postoperative atrial fibrillation. Heart Dis . 2002;4:117.
131 Mathew JP, Fontes ML, Tudor IC, et al. A multicenter risk index for atrial fibrillation after cardiac surgery. JAMA . 2004;291:1720.
132 Eagle KA, Guyton RA, Davidoff R, et al. ACC/AHA 2004 guideline update for coronary artery bypass graft surgery: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1999 Guidelines for Coronary Artery Bypass Graft Surgery). Circulation . 2004;110:e340.
133 Arrowsmith JE, Harrison MJG, Newman SP, et al. Neuroprotection of the brain during cardiopulmonary bypass. Stroke . 1998;29:2357.
134 Mangano DT. Aspirin and mortality from coronary bypass surgery. N Engl J Med . 2002;347:1309.
135 Guyton RA, Mellitt RJ, Weintraub WS. A critical assessment of neurological risk during warm heart surgery. J Card Surg . 1995;10:488.
136 Naylor R, Cuffe RL, Rothwell PM, et al. A systematic review of outcome following synchronous carotid endarterectomy and coronary artery bypass: influence of surgical and patient variables. Eur J Vasc Endovasc Surg . 2003;26:230.
137 Hill MD, Shrive FM, Kennedy J, et al. Simultaneous carotid endarterectomy and coronary artery bypass surgery in Canada. Neurology . 2005;64:1435.
138 Pullicino P, Halperin J. Combining carotid endarterectomy with coronary bypass surgery: is it worth the risk? Neurology . 2005;64:1332.
139 Randall MS, McKevitt FM, Cleveland TJ, et al. Is there any benefit from staged carotid and coronary revascularization using carotid stents? A single-center experience highlights the need for a randomized controlled trial. Stroke . 2006;37:435.
140 Chaturvedi S, Bruno A, Feasby T, et al. Carotid endarterectomy—an evidence-based review: report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Neurology . 2005;65:794.
141 Moore W, Barnett H, Beebe H, et al. Guidelines for carotid endarterectomy: A multidisciplinary consensus statement from the Ad Hoc Committee, American Heart Association. Stroke . 1995;26:188.
142 Durand DJ, Perler BA, Roseborough GS, et al. Mandatory versus selective preoperative carotid screening: a retrospective analysis. Ann Thorac Surg . 2004;78:159.
143 Yadav JS, Wholey MH, Kuntz RE, et al. Protected carotid-artery stenting versus endarterectomy in high-risk patients. N Engl J Med . 2004;351:1493.
144 Ziada KM, Yadav JS, Mukherjee D, et al. Comparison of results of carotid stenting followed by open heart surgery versus combined carotid endarterectomy and open heart surgery (coronary bypass with or without another procedure). Am J Cardiol . 2005;96:519.
145 Tsang JC, Morin JF, Tchervenkov CI, et al. Single aortic clamp versus partial occluding clamp technique for cerebral protection during coronary artery bypass: a randomized prospective trial. J Card Surg . 2003;18:158.
146 Hammon JW, Stump DA, Butterworth JF, et al. Single crossclamp improves 6-month cognitive outcome in high-risk coronary bypass patients: the effect of reduced aortic manipulation. J Thorac Cardiovasc Surg . 2006;131:114.
147 Nagels W, Demeyere R, Van Hemelrijck J, et al. Evaluation of the neuroprotective effects of S(+)-ketamine during open-heart surgery. Anesth Analg . 2004;98:1595.
148 Kanbak M, Saricaoglu F, Avci A, et al. Propofol offers no advantage over isoflurane anesthesia for cerebral protection during cardiopulmonary bypass: a preliminary study of S-100beta protein levels. Can J Anaesth . 2004;51:712.
149 Taggart DP, Browne SM, Wade DT, et al. Neuroprotection during cardiac surgery: a randomised trial of a platelet activating factor antagonist. Heart . 2003;89:897.
150 Luciani GB, Menon T, Vecchi B, et al. Modified ultrafiltration reduces morbidity after adult cardiac operations: a prospective, randomized clinical trial. Circulation . 2001;104(suppl I):I253.
151 Whitaker DC, Newman SP, Stygall J, et al. The effect of leucocyte-depleting arterial line filters on cerebral microemboli and neuropsychological outcome following coronary artery bypass surgery. Eur J Cardiothorac Surg . 2004;25:267.
152 Ozduran V, Yamani MH, Chuang HH, et al. Survival beyond 10 years following heart transplantation: the Cleveland Clinic Foundation experience. Transplant Proc . 2005;37:4509.
153 Taylor DO, Edwards LB, Boucek MM, et al. Registry of the International Society for Heart and Lung Transplantation: twenty-second official adult heart transplant report—2005. J Heart Lung Transplant . 2005;24:945.
154 Hotson J, Enzmann D. Neurologic complications of cardiac transplantation. Neurol Clin . 1988;6:349.
155 Hotson J, Pedley T. The neurological complications of cardiac transplantation. Brain . 1976;99:673.
156 Furlan A, Sila C, SC C. Neurologic complications related to cardiac surgery. Neurol Clin . 1992;10:145.
157 Sila C. Spectrum of neurologic events following cardiac transplantation. Neurology . 1989;20:1586.
158 Jarquin-Valdivia AA, Wijdicks EF, McGregor C. Neurologic complications following heart transplantation in the modern era: decreased incidence, but postoperative stroke remains prevalent. Transplant Proc . 1999;31:2161.
159 Perez-Miralles F, Sanchez-Manso JC, Almenar-Bonet L, et al. Incidence of and risk factors for neurologic complications after heart transplantation. Transplant Proc . 2005;37:4067.
160 Mayer TO, Biller J, O’Donnell J, et al. Contrasting the neurologic complications of cardiac transplantation in adults and children. J Child Neurol . 2002;17:195.
161 Glass GA, Stankiewicz J, Mithoefer A, et al. Levetiracetam for seizures after liver transplantation. Neurology . 2005;64:1084.
162 Britt R, Enzmann D, Remington J. Intracranial infections in cardiac transplant recipients. Ann Neurol . 1981;9:107.
163 Fishman JA, Rubin RH. Infection in organ-transplant recipients. N Engl J Med . 1998;338:1741.
164 DeLone DR, Goldstein RA, Petermann G, et al. Disseminated aspergillosis involving the brain: distribution and imaging characteristics. AJNR Am J Neuroradiol . 1999;20:1597.
165 Gaviani P, Schwartz RB, Hedley-Whyte ET, et al. Diffusion-weighted imaging of fungal cerebral infection. AJNR Am J Neuroradiol . 2005;26:1115.
166 Marr KA, Boeckh M, Carter RA, et al. Combination antifungal therapy for invasive aspergillosis. Clin Infect Dis . 2004;39:797.
167 Schwartz S, Ruhnke M, Ribaud P, et al. Improved outcome in central nervous system aspergillosis, using voriconazole treatment. Blood . 2005;106:2641.
168 Luft B, Naot Y, Araujo F, et al. Primary and reactivated toxoplasma infection in patients with cardiac transplants. Ann Intern Med . 1983;99:27.
169 Renold C, Sugar A, Chave J-P, et al. Toxoplasma encephalitis in patients with the acquired immunodeficiency syndrome. Medicine (Baltimore) . 1992;71:224.
170 Cuturic M, Hayat GR, Vogler CA, et al. Toxoplasmic polymyositis revisited: case report and review of literature. Neuromuscul Disord . 1997;7:390.
171 Plonquet A, Bassez G, Authier FJ, et al. Toxoplasmic myositis as a presenting manifestation of idiopathic CD4 lymphocytopenia. Muscle Nerve . 2003;27:761.
172 Montero C, Martinez A. Neuropathology of heart transplantation: 23 cases. Neurology . 1986;36:1149.
173 Lopez FA, Johnson F, Novosad DM, et al. Successful management of disseminated Nocardia transvalensis infection in a heart transplant recipient after development of sulfonamide resistance: case report and review. J Heart Lung Transplant . 2003;22:492.
174 Hall WA, Martinez AJ, Dummer JS, et al. Nocardial brain abscess: diagnostic and therapeutic use of stereotactic aspiration. Surg Neurol . 1987;28:114.
175 Kohl O, Tillmanns HH. Cerebral infection with Rhodococcus equi in a heart transplant recipient. J Heart Lung Transplant . 2002;21:1147.
176 Patchell RA. Neurological complications of organ transplantation. Ann Neurol . 1994;36:688.
177 Alsep SG, Cobbs CG. Pseudoallescheria boydii infection of the central nervous system in a cardiac transplant recipient. South Med J . 1986;79:383.
178 Pollard R, Arvin A, Gamberg P, et al. Specific cell-mediated immunity and infections with herpes viruses in cardiac transplant recipients. Am J Med . 1982;73:679.
179 Flomenbaum MA, Jarcho JA, Schoen FJ. Progressive multifocal leukoencephalopathy fifty-seven months after heart transplantation. J Heart Lung Transplant . 1991;10:888.
180 Iwamoto M, Jernigan DB, Guasch A, et al. Transmission of West Nile virus from an organ donor to four transplant recipients. N Engl J Med . 2003;348:2196.
181 Srinivasan A, Burton EC, Kuehnert MJ, et al. Transmission of rabies virus from an organ donor to four transplant recipients. N Engl J Med . 2005;352:1103.
182 Fischer SA, Graham MB, Kuehnert MJ, et al. Transmission of lymphocytic choriomeningitis virus by organ transplantation. N Engl J Med . 2006;354:2235.
183 Centers for Disease Control and Prevention (CDC). Chagas disease after organ transplantation—United States, 2001. MMWR Morb Mortal Rep Wkly . 2002;51:210.
184 Aull MJ, Buell JF, Trofe J, et al. Experience with 274 cardiac transplant recipients with posttransplant lymphoproliferative disorder: a report from the Israel Penn International Transplant Tumor Registry. Transplantation . 2004;78:1676.
185 Swinnen L. Treatment of organ transplant-related lymphoma. Hematol Oncol Clin North Am . 1997;11:963.
186 Kang SK, Kirkpatrick JP, Halperin EC. Low-dose radiation for posttransplant lymphoproliferative disorder. Am J Clin Oncol . 2003;26:210.
187 Phan T, O’Neill B, Habermann T. Posttransplant primary central nervous system lymphoma. Ann Neurol . 1999;46:463.
188 Schiff D, O’Neill B, Wijdicks E, et al. Gliomas arising in organ transplant recipients: an unrecognized complication of transplantation? Neurology . 2001;57:1486.
189 Mueller XM. Drug immunosuppression therapy for adult heart transplantation. Part 2: clinical applications and results. Ann Thorac Surg . 2004;77:363.
190 Mueller XM. Drug immunosuppression therapy for adult heart transplantation. Part 1: immune response to allograft and mechanism of action of immunosuppressants. Ann Thorac Surg . 2004;77:354.
191 Wijdicks EF. Neurotoxicity of immunosuppressive drugs. Liver Transpl . 2001;7:937.
192 Small SL. Pharmacotherapy of aphasia: a critical review. Stroke . 1994;25:1282.
193 Bechstein WO. Neurotoxicity of calcineurin inhibitors: impact and clinical management. Transplant Int . 2000;13:313.
194 Maramattom BV, Wijdicks EF. Sirolimus may not cause neurotoxicity in kidney and liver transplant recipients. Neurology . 2004;63:1958.
195 Rana SS, Giuliani MJ, Oddis CV, et al. Acute onset of colchicine myoneuropathy in cardiac transplant recipients: case studies of three patients. Clin Neurol Neurosurg . 1997;99:266.
196 Adair J, Woodley S, O’Connell J, et al. Aseptic meningitis following cardiac transplantation: clinical characteristics and relationship to immunosuppressive regimen. Neurology . 1991;41:249.
197 Pittock SJ, Rabinstein AA, Edwards BS, et al. OKT3 neurotoxicity presenting as akinetic mutism. Transplantation . 2003;75:1058.
Chapter 4 Neurological Complications of Congenital Heart Disease and Cardiac Surgery in Children

Catherine Limperopoulos, Adré J. Du Plessis

Chromosomal Disorders
Cerebral Dysgenesis
Acquired Preoperative Cerebrovascular Injury
Intraoperative Brain Injury
Focal or Multifocal Hypoxic-Ischemic Injury
Global Hypoxic-Ischemic Injury
Mechanisms of Postoperative Injury
Delayed Recovery of Consciousness
Postoperative Seizures
Periventricular White Matter Injury
Movement Disorders
Spinal Cord Injury
Brachial Plexus and Peripheral Nerve Injury
Pediatric neurologists have become increasingly challenged by diagnostic and management decisions in children with congenital or acquired heart disease experiencing neurological dysfunction. Of the 30,000 infants born with heart defects in the United States each year, approximately half require some form of surgical intervention within the first year of life. 1, 2 Since the late 1960s, there have been major changes in the clinical profile of neurological injury in children with congenital heart disease. In earlier years, the neurological complications of congenital heart disease were mediated for the most part by chronic hypoxia and polycythemia in cyanotic children, uncorrected right-to-left shunts, and the effects of repeated palliative heart operations. 3, 4 Advances in surgical technique and intensive care management have allowed the anatomical correction of many heart lesions in early infancy. These early-life corrective procedures have resulted in major decreases in the mortality rate of congenital heart disease. Consequently, neurological sequelae are now increasingly seen in adult survivors of congenital heart disease. Heart defects considered inoperable in the mid-1980s are now successfully repaired with a very low mortality rate. More infants with critical congenital heart disease and profound hemodynamic disturbances in the newborn period are now rescued, only to manifest later the neurological consequences of this early-life morbidity. Furthermore, the same surgical support techniques responsible for advancing survival have, paradoxically, been associated with an incidence of neurological complications that approaches 25 percent in some centers. 5 Consequently, mechanisms of brain injury during cardiac surgery have been the focus of intense investigation over the past two decades. Understanding of these intraoperative mechanisms has been advanced through animal experimental models 6, 7 and several large clinical trials, 7 - 12 as well as by intraoperative cerebral monitoring and perioperative magnetic resonance imaging (MRI).
More recently, there has been increased recognition that both acquired and developmental brain disturbances in infants with congenital heart disease may have their origin prior to surgical intervention, in many cases during the fetal period. 13 - 22 It is expected that these mechanisms will receive particular attention over the next few years as the role of fetal imaging expands and the potential for fetal interventions is explored. 23, 24

Recent studies have demonstrated a high prevalence of neurological abnormalities prior to infant cardiac surgery, in some studies exceeding 50 percent. These abnormalities include microcephaly, hypotonia, behavioral dysregulation, and feeding difficulties 8, 13, 15, 25 as well as abnormal electrophysiological studies. 14, 16 These preoperative neurological and electrophysiological abnormalities are increasingly recognized as significant predictors of longer term neurodevelopmental sequelae. 14, 16, 25 - 28 The presence of these preoperative abnormalities in the early neonatal period strongly suggests a fetal onset of neurological impairment.

Chromosomal Disorders
A number of chromosomal disorders have a phenotype that includes both cardiac and neurological malformations, including trisomies 11, 18, and 21. The most common neurological manifestation in children with trisomy 21, or Down syndrome, is cognitive dysfunction. Epilepsy develops in approximately 5 percent of children with trisomy 21. Congenital heart defects, most commonly endocardial cushion defects, are present in 40 percent of children with Down syndrome. Gross structural brain alterations in Down syndrome include a narrow superior temporal gyrus and a disproportionately small cerebellum and brainstem. 29 Trisomy 13 is associated with ventricular septal defects and patent ductus arteriosus; the associated cerebral dysgenesis in this syndrome is often severe, with holoprosencephaly and agenesis of the corpus callosum being the most common lesions. The most common cardiac lesions in infants with trisomy 18 are ventricular septal defects and patent ductus arteriosus, with neuronal migration defects the usual form of brain dysgenesis. 30
The phenotypic spectrum of specific chromosome 22 deletions, particularly in the 22q11 region, includes a variety of cardiac malformations and neurological features. 31 Recent population-based data suggest that at least 700 infants with chromosome 22 deletion syndromes are born annually in the United States. 32 The acronym CATCH 22 ( c ardiac defect, a bnormal facies, t hymic hypoplasia, c left palate, h ypocalcemia, chromosome 22 q11 deletions) has been used to designate this group of apparently related syndromes. The two most common syndromes, DiGeorge and velocardiofacial (or Shprintzen) syndromes, have neurological and cognitive manifestations in association with structural cardiac defects. 33 The fundamental problem in DiGeorge syndrome is a developmental defect of the third and fourth pharyngeal pouches, manifesting with thymic and parathyroid hypoplasia and conotruncal cardiac malformations, which include interrupted aortic arch (type B), truncus arteriosus, and tetralogy of Fallot.
A common neurological presentation in both DiGeorge and the velocardiofacial syndrome is hypocalcemic seizures due to hypoparathyroidism. In addition to the usual cardiac lesions (i.e., ventricular septal defect or tetralogy of Fallot), the velocardiofacial syndrome is associated with cleft palate or velopharyngeal insufficiency and a typical facial appearance, including a broad, prominent nose and retrognathia, and microcephaly in up to 40 percent. Neuroimaging and autopsy studies may show a small posterior fossa and vermis, small cystic lesions adjacent to the frontal horns of the lateral ventricles, dysgenesis of the corpus callosum, and abnormal cortical gyrification patterns. 34 - 42 Delayed opercular development and disproportionately enlarged sylvian fissures as well as white matter abnormalities might underlie some of the developmental problems in these children, 43 including nearly universal learning difficulties. 44 The mean intelligence quotient (IQ) in this syndrome is around 70, 33, 44 with mild to moderate mental retardation in up to 50 percent of patients.
In recent years, a high rate of autism spectrum disorders and attention deficit/hyperactivity disorder has been described in this group. 45, 46 Psychiatric disorders occur in up to 22 percent of patients with 22q11 deletion syndromes. 34, 36 A peculiar and inappropriately blunt affect may be evident during childhood, 47 often evolving to frank psychosis during adolescence and adulthood. 48 Altered prefrontal cortex-amygdala circuitry, reduced cerebellar and thalamic volumes, and increased basal ganglia and corpus callosal volumes, as shown by quantitative neuroimaging studies, may underlie the disrupted emotional processing and form the neurobiological substrate for the psychiatric disturbances in these children. 38, 39, 44, 49 - 51

Cerebral Dysgenesis
The prevalence of brain dysgenesis in children with congenital heart disease approaches 30 percent in some autopsy studies. 29, 52, 53 The risk of cerebral dysgenesis appears related to the underlying cardiac lesion. For example, infants with hypoplastic left heart syndrome may be at particular risk of associated developmental brain lesions, which range in severity from microdysgenesis to gross malformations, 52 including agenesis of the corpus callosum, holoprosencephaly, and immature cortical mantle. With advances in neuroimaging, the relationship between cardiac and brain dysgenesis is becoming more clearly defined. Clinically, these dysgenetic lesions may present in the newborn period with seizures, alterations in level of consciousness, and abnormalities in motor tone. Conversely, these lesions may remain clinically occult until later infancy and childhood, when they present with developmental delay, epilepsy, and cerebral palsy. For these reasons, it is important to consider cerebral dysgenesis in any child with congenital heart disease and neurological manifestations.

Acquired Preoperative Cerebrovascular Injury
Infants with congenital heart disease are at increased risk of acquired antenatal or perinatal brain injury. During fetal life, congenital heart lesions may be associated with changes in cerebrovascular blood-flow distribution and resistance. Fetuses with hypoplastic left heart syndrome, whose cerebral perfusion is supplied retrograde through the ductus arteriosus, may be at particular risk. 19, 20 Preoperative MRI studies have demonstrated that brain injury is common in infants with critical congenital heart disease and during invasive diagnostic procedures (e.g., balloon-atrial septostomy). 54, 55 Preoperative findings detected by MRI include intracranial hemorrhage, cerebral venous thromboses, thromboembolic infarctions, dilation of the ventricles and subarachnoid spaces (suggestive of cerebral atrophy), periventricular leukomalacia, and gray matter injury. 17, 18, 20, 56, 57 Elevated preoperative brain lactate levels have been found by magnetic resonance spectroscopy in over half of newborns. 17, 24, 56
Complex corrective operations are performed in ever smaller and less mature newborn infants. 58 Consequently, the vascular lesions associated with less mature infants are now seen. Intraventricular/periventricular hemorrhage (IVH-PVH) is a common neurological complication in the newborn. 59 The risk of IVH-PVH is related to the severity of the vascular insult and inversely to the infant’s gestational age. Prematurity predisposes to IVH-PVH because of the structural and physiological vulnerability of the immature periventricular germinal matrix. The hemodynamic instability commonly seen in more severe forms of congenital heart disease predisposes to the systemic hypotension or fluctuations in blood pressure that trigger IVH-PVH. 59 Compared with the incidence of 3.5 percent for IVH-PVH in term infants overall, the incidence in term infants with congenital heart disease is as high as 16 percent in some studies. 22 At autopsy, 25 percent of infants with hypoplastic left heart syndrome have intracranial hemorrhage. 60 Infants with coarctation of the aorta are also at increased risk of intracranial hemorrhage because of the intracranial vascular malformations and hypertension associated with this condition. 59, 61
The detection of intraventricular hemorrhage in infants with congenital heart disease in the preoperative period creates a major management dilemma because the risk of extending such hemorrhage is increased by cardiopulmonary bypass and cardiac surgery. Specifically, cardiopulmonary bypass requires anti-coagulation to prevent clot formation in the bypass circuit; in addition, it has been associated with enhanced systemic fibrinolytic activity. 62 The more complex operations require periods of decreased perfusion to approach the cardiac defect. The requirement for anticoagulation and the potentially severe intraoperative perfusion changes increase the risk of extending any preoperative intracranial hemorrhage. The dilemma is further complicated by the fact that intracranial hemorrhage occurs most commonly in infants with the most critical cardiac lesions, that is, those in greatest need of early surgical repair.
There are no prospectively tested protocols for managing the dilemma of preoperative intracranial hemorrhage in infants requiring cardiac surgery. At our center, we use the following guidelines. We perform preoperative cranial ultrasonography to exclude IVH-PVH in all premature infants with a birth weight less than 1,500 g and newborn infants with preoperative neurological dysfunction, coagulation disturbances, or hemodynamic instability causing significant metabolic acidosis. In those infants with IVH-PVH, surgical planning is based on the severity of the cardiac illness (which may directly affect the risk of hemorrhage extension), the likely complexity of surgery, and the severity of preoperative IVH-PVH. Minor subependymal hemorrhage carries a low risk of extension 63, 64 and should not delay cardiac surgery. In infants with hemorrhage into the ventricles or the parenchyma, we delay cardiopulmonary bypass for at least 7 days if the cardiac condition permits.

Neurological dysfunction presenting during the early postoperative period is likely due in most cases to intraoperative hypoxic-ischemic/reperfusion injury. However, the risk of cerebrovascular injury extends into the postoperative period, when cardiorespiratory instability, together with cerebral autoregulatory dysfunction, predisposes to cerebral hypoxic-ischemic injury. Despite the remarkable advances facilitated by deep hypothermia and pharmacological agents, the persistent neurological morbidity in the postoperative period attests to the incomplete neuroprotection offered by these strategies. 65, 66
The precise onset and evolution of hypoxic-ischemic/reperfusion injury may be difficult to establish. First, the mechanisms of both parenchymal and vascular hypoxic-ischemic/reperfusion injury are known to evolve over time. Second, during the early posthypoxic-ischemic period, cells that have sustained an insult may be at particular risk of irreversible injury from subsequent disturbances in oxygen supply. Consequently, it is difficult to ascribe with any certainty hypoxic-ischemic/re-perfusion injury to one of the preoperative, intraoperative, or postoperative periods. Rather, it is likely that in many cases the injury is multifactorial and cumulative.

Intraoperative Brain Injury
There are multiple interrelated mechanisms by which brain injury may occur during cardiac surgery. However, hypoxic-ischemic/reperfusion injury is likely the principal mechanism, a notion supported by the topography of injury seen at autopsy, 60, 67 that is, laminar cortical necrosis and periventricular white matter injury. 68, 69 Animal models of deep hypothermic circulatory arrest have also demonstrated selective neuronal necrosis in a distribution that corresponds closely to that seen after normothermic hypoxic-ischemic/reperfusion injury. 70 Neuropathological studies of infants after deep hypothermic cardiac surgery suggest that cerebral white matter lesions tend to be the most prevalent and severe, followed by a spectrum of gray matter lesions. 67
The changes in cerebral perfusion and metabolism during cardiac surgery are complex, interrelated, and often extreme. When these changes exceed the brain’s ability to maintain a balance between cerebral oxygen/substrate supply and utilization, a hypoxic-ischemic/reperfusion insult is triggered. The multiple factors determining intraoperative cerebral oxygen availability may be categorized as extrinsic, that is, related to the extracorporeal circulation (e.g., loss of pulsatility, low or no pump flow, hypothermia, emboli) or intrinsic (e.g., disturbances in cerebral blood-flow autoregulation). During deep hypothermic cardiac surgery, cerebral oxygen delivery may be impaired by focal or multifocal vaso-occlusive phenomena generated by the bypass circuit or global hypoperfusion due to excessive attenuation of bypass flow rate. 65, 66

Focal or Multifocal Hypoxic-Ischemic Injury
The relatively small intravascular volume of the young infant compared with the large blood volume required to “prime” the cardiopulmonary bypass circuit results in increased exposure to insults related to the bypass. 65, 66 These may be either embolic 71 or inflammatory due to the extensive interface between blood and artificial surfaces. 72 The replacement of bubble oxygenators with membrane devices has decreased but not eradicated the embolic “load” of bypass circuits. Both gaseous and particulate emboli may enter the circulation directly from the surgical field. Because the bypass circuit delivers oxygenated blood directly to the aorta, circulating emboli circumvent the normal pulmonary filtration bed and enter the systemic (and cerebral) arterial circulation directly. Earlier autopsy studies described cerebral embolic brain injury after cardiac surgery, and subsequent studies following cardiopulmonary bypass have described a significant prevalence of cerebral capillary-bed aneurysmal dilatations. 71
Cardiopulmonary bypass activates a host of inflammatory cascades that cause diffuse vascular injury, resulting in a postperfusion syndrome that in severe cases is associated with multiple organ failure. 73 Pathways triggered include the eicosanoid, complement, and kallikrein pathways. These pathways activate free radical generation, 74 cause antioxidant depletion, 75 and upregulate adhesion molecules on neutrophils and endothelial cells. These activated neutrophils appear to be potent mediators of reperfusion injury in the brain. Hypothermia delays and modifies the effect of these processes but does not completely prevent them. 76

Global Hypoxic-Ischemic Injury
Several techniques used during neonatal cardiac surgery jeopardize cerebral oxygen delivery by altering cerebral perfusion, arterial oxygen content, and tissue oxygen delivery. Under deep hypothermic conditions, cerebral oxygen availability may be limited by cold-induced increases in cerebral vascular resistance, 77, 78 impairment of cerebral pressure-flow autoregulation, 79, 80 and increased oxygen-hemoglobin affinity. 81 Normally, during periods of decreased perfusion pressure, cerebral oxygen delivery is maintained by an initial vasodilatory response followed by an increase in oxygen extraction. 82 However, both these compensatory responses are compromised at deep hypothermia. 83
To approach the often tiny cardiac defects, the bypass flow rate is decreased and often arrested for periods depending on the complexity of the lesion. Although there are general guidelines for “safe periods” of deep hypothermic circulatory arrest at different temperatures, these remain controversial and unpredictable in the individual infant. In addition, the safety of low-flow bypass compared with hypothermic circulatory arrest has been debated. Low-flow bypass prolongs exposure to bypass-related phenomena, as well as increasing the risk of incomplete ischemia. Conversely, deep hypothermic circulatory arrest (DHCA) allows more rapid completion of the intracardiac phases of the repair and reduces the exposure to bypass perfusion; however, it exposes the infant to periods of complete ischemia. 65, 66 Over the past 15 years, a number of studies have focused on the relationship between DHCA and neurological outcome; most studies have reported a deleterious effect on outcome. 84 - 87 In the first major clinical trial randomizing infants to a strategy of predominant hypothermic circulatory arrest or low-flow bypass, infants exposed to the former were at significantly greater risk of perioperative 8 and 1-year neurological sequelae. 9 At age 4 years, the DHCA group had significantly worse behavior, speech, and language function, 10 but no difference in mean intelligence score. Furthermore, at 8-year follow-up, those assigned to DHCA scored worse on motor and speech domains, whereas those assigned to low-flow bypass had worse scores for impulsivity and behavior. 11 Therefore, the long-term follow-up of this large cohort has provided important insights into the evolution of neurodevelopmental outcome in this complex population over time. 88 Although it is now generally accepted that prolonged periods of uninterrupted DHCA may have adverse neurological effects, certain studies have shown that shorter durations of DHCA are not consistently associated with adverse outcomes. 89 - 91 In fact, available data suggest that the relationship between DHCA duration and neurodevelopmental sequelae is not linear and that the risk of brain injury increases significantly after about 40 minutes of DHCA. 84, 92 These studies have again emphasized that the neurological dysfunction in this population is undoubtedly mediated by numerous interrelated preoperative and postoperative risk factors in addition to DHCA.
In addition to the bypass flow rate, a number of other factors associated with cardiopulmonary bypass may affect cerebral perfusion and predispose to hypoxic-ischemic/reperfusion injury. Most centers in the United States use nonpulsatile bypass devices as well as hemodilution to reduce the magnitude of blood cell trauma. Deep hypothermia is widely used to suppress oxygen consumption during infant cardiac surgery. In addition to their intended beneficial effects, these techniques all have potential adverse effects on cerebral oxygen delivery. The nonpulsatile perfusion of cardiopulmonary bypass, particularly at low-flow rates, may fail to maintain perfusion in distal capillary beds. 93 Furthermore, nonpulsatile blood-flow may disrupt pressure-flow and metabolism-flow autoregulation. 77, 80, 94 Hemodilution is used during bypass to reduce rheologic injury to circulating red cells during deep hypothermia. However, because the concentration of oxygenated hemoglobin is the major determinant of oxygen-carrying capacity, hemodilution may limit cerebral oxygen delivery. In animal studies, extreme hemodilution (to hematocrit levels less than 10%) was associated with neurological injury, whereas hematocrit levels above 30 percent improved cerebral recovery after DHCA. 6 These experimental results were confirmed by a randomized clinical trial in which infants randomized to a hematocrit of 20 percent during cardiac surgery had significantly worse developmental scores at 1 year than those randomized to a hematocrit of around 30 percent. 7
Another important intraoperative factor is the management of acid-base status during cardiopulmonary bypass. In a randomized, single-center trial, infants undergoing cardiac operations were assigned to the alpha-stat versus pH-stat strategy during deep hypothermic cardiopulmonary bypass. 12 The use of pH-stat management was associated with lower overall early postoperative morbidity. 12 Treatment assignment had no effect on neurodevelopmental outcome at 1, 2, and 4 years of age. 95 Despite these equivocal findings, many centers are currently using pH-stat management during core cooling.
After repair of the cardiac defect, bypass flow rates are increased using rewarmed and highly oxygenated blood. Rewarming aims to reactivate cellular enzyme function and oxygen utilization. During this period of reperfusion, a number of factors may predispose to free radical injury. 65, 66 Several studies have suggested a delay in recovery of mitochondrial function, 96 possibly by nitric oxide, which is generated in abundance during the bypass. 97 The combination of a highly oxygenated reperfusion and persistent mitochondrial dysfunction may be a major source of injurious oxygen free radicals. 98 Excessively rapid rewarming after deep hypothermia may be deleterious. 99 Hyperthermia is a trigger for glutamate release, 100 predisposing to excitotoxicity as well as further stressing the recovering cerebral metabolism.

Mechanisms of Postoperative Injury
During the postoperative period of intensive care, a number of factors may predispose to brain injury. Cerebral perfusion pressure may be compromised by a combination of decreased cardiac output and elevated central venous pressure resulting from postoperative cardiac dysfunction. In addition to these systemic circulatory factors, there may be intrinsic cerebrovascular disturbances in the postoperative period. Specifically, elevated cerebral vascular resistance, decreased cerebral blood-flow, 3, 66, 79, 101, 102 and impaired vasoregulation have been described after deep hypothermic circulatory arrest. Together, these factors may render the brain vulnerable to injury in the postoperative period.

Recent studies suggest a decrease in acute neurological morbidity following surgery. 103 However, intraoperative and postoperative insults may injure the neuraxis at any level. Because a detailed discussion of the entire spectrum of neurological manifestations is not possible in the current context, this review focuses on the more common clinical issues confronting the child neurologist.

Delayed Recovery of Consciousness
Prolonged impairment of mental status after cardiac surgery, anesthesia, and postoperative sedation is a common reason for neurological consultation. The evaluation should follow the well-established clinical guidelines for assessing impaired consciousness. 104 Certain specific etiologies should be excluded, including postoperative hepatic or renal impairment, which may impair the metabolism or excretion of sedating drugs. Prolonged use of neuromuscular blocking agents in the preoperative or postoperative period may delay the recovery of motor function (discussed later) 105 and, if severe, may suggest impaired consciousness. This condition should be excluded at the bedside with a peripheral nerve stimulator or formal nerve conduction studies. Postoperative seizures are a common complication of cardiac surgery (as discussed in the next section), and not infrequently these seizures are clinically silent. 8 Such “occult” seizures or a prolonged postictal state should be considered in the evaluation of a depressed postoperative mental state. In spite of this approach, the precise cause of an impaired postoperative mental status is not established in most cases. Many of these children ultimately demonstrate features suggestive of hypoxic-ischemic/reperfusion injury.

Postoperative Seizures
Seizures early in the postoperative period are among the most common neurological complications after open heart surgery. Postoperative clinical seizures have been reported to occur in up to 19 percent of survivors of neonatal cardiac surgery, 106 and in certain subgroups this risk may reach 50 percent. Clinical seizures are reported less frequently than those detected by continuous electroencephalographic monitoring 8, 107 since postoperative seizures may occur without typical motor correlates. 8
Postoperative seizures may be divided into two broad groups. First are those seizures with a readily identifiable cause, such as hypoglycemia, hypocalcemia, and cerebral dysgenesis. Postoperative seizures may also result from hypoxic-ischemic/reperfusion injury due to either generalized cerebral hypoperfusion (e.g., cardiac arrest) or focal vaso-occlusive insults. The second and more common category of postoperative seizures is that in which the etiology remains unknown. Although these cryptogenic seizures, commonly referred to as postpump seizures, are often assumed to relate to hypoxic-ischemic/reperfusion injury, their etiology is likely multifactorial with risk factors that include the use and duration of deep hypothermic circulatory arrest, 8, 108 younger age at surgery, the type of heart defect (e.g., aortic arch obstruction), and genetic conditions. 106 Furthermore, postpump seizures differ in several respects from other forms of posthypoxic seizures. First, these seizures typically develop later than, for instance, those occurring after perinatal asphyxia. Second, although less benign than previously believed, 9 the prognosis of postpump seizures is significantly better than that of asphyxial seizures, in which up to 50 percent of survivors are neurologically disabled. 109 Both the delayed onset and more favorable outcome of postpump seizures may be due to the partial protective effect of hypothermia at the time of intraoperative hypoxic-ischemic/reperfusion insult. 110
The clinical course of postpump seizures is fairly typical. These seizures appear confined to a relatively narrow time-window, with onset between 24 and 48 hours after surgery. This is followed by several days during which serial seizures occur, often evolving to status epilepticus; thereafter, the tendency toward further seizures wanes rapidly. The clinical manifestations of these electrographic seizures are often subtle even in the absence of sedating and paralyzing drugs and may be confined to paroxysmal autonomic changes. When evident, convulsive activity is usually focal or multifocal.
The therapeutic approach to postpump seizures should be based on their typical clinical course. After excluding reversible etiologies such as hypoglycemia, hypomagnesemia, and hypocalcemia, 111, 112 the tendency toward repeated seizures and status epilepticus should be countered by rapid achievement of therapeutic anticonvulsant levels by an intravenous route. Most postpump seizures are controlled by lorazepam, followed by phenobarbital or phenytoin. Potential cardiotoxicity due to these agents in children recovering from cardiac surgery should be monitored carefully, particularly during the initiation of treatment. The apparently circumscribed window of susceptibility to postpump seizures often allows early withdrawal of anticonvulsants.
Prospective studies have demonstrated a significant correlation between postoperative seizures and risk of perioperative 8 and 1-year neurological sequelae, 9 as well as abnormal MRI studies. 9, 106, 113 The longer term impact of postpump seizures may be less than previously thought. 11, 88, 107 The development of subsequent epilepsy is rare; however, West syndrome (infantile spasms, mental retardation, and epilepsy) 114 has been described after more intractable postpump seizures.
When postoperative seizures have an identified cause, the long-term outcome is related to etiology. For instance, cerebral dysgenesis, which is increased in congenital heart disease, may present with seizures in the early postoperative period, and here the long-term outcome is usually poor, with epilepsy a common sequela. 106 Infants with seizures due to postoperative stroke have a 20 to 30 percent risk of subsequent epilepsy. 115

Periventricular White Matter Injury
Brain MRI of neonates following cardiac surgery has shown a prevalence of periventricular leukomalacia in excess of 50 percent in some studies 116 ; this is a rare finding in older infants. The precise onset of these lesions remains unclear, but the MRI features appear to be transient in many cases. 56 Reported risk factors for these MRI lesions include prolonged exposure to cardiopulmonary bypass (with or without DHCA), possibly related to inflammatory mechanisms activated by cardiopulmonary bypass. In addition, early postoperative hypotension (especially diastolic) and hypoxemia appear to increase the risk of periventricular leukomalacia in these MRI studies. 116, 117 Magnetic resonance spectroscopy studies are beginning to provide insights into disturbed brain metabolism in the postoperative period. 17, 118, 119 Although significant decreases in brain N -acetyl-aspartate, a neuronal-axonal marker, have been described, 118, 119 more recent data have shown an apparently improved cerebral oxidative metabolism postoperatively as evidenced by improved lactate/choline ratios. 17 The long-term significance of these acute structural and metabolic disturbances in children who survive cardiac surgery remains to be determined.

The incidence of cerebrovascular accidents (strokes) in children ranges from 2.5 to 8 per 100,000. 120 Congenital heart disease is the leading known association of childhood stroke and is present in 25 to 30 percent of cases. 120 - 122 In earlier autopsy studies, almost 20 percent of children with congenital heart disease demonstrated features of cerebrovascular injury.
Stroke associated with heart disease (cardiogenic stroke) may be classified on the basis of the likely embolic or thrombotic source as (1) cardioembolic (i.e., a probable intracardiac embolic source); (2) paradoxical (i.e., a cardiac anatomy that permits an embolus of systemic venous origin access to the cerebral circulation); or (3) venous (i.e., cerebral vein thrombosis due to central venous hypertension and venous stasis).
Risk factors for cardiogenic stroke include the elements of Virchow’s triad—altered vascular surface, stasis, and hypercoagulability—as well as the presence of “paradoxical” embolic pathways. Risk factors for cardiogenic stroke have changed over the years. In earlier studies, the risk of stroke was related to the effects of long-standing heart defects, such as chronic hypoxia and polycythemia, and uncorrected paradoxical pathways (e.g., right-to-left shunts). The trend in recent decades toward earlier corrective surgery 1 has reduced exposure to such stroke risk factors, shifting the focus to intraoperative and postoperative mechanisms for stroke.
A number of intraoperative mechanisms related to cardiopulmonary bypass may predispose to cerebral vascular occlusion. Embolic material (particulate/gaseous) 71 generated during bypass avoids filtration by the pulmonary bed, gaining direct entry to the systemic arterial circulation. Earlier autopsy data demonstrated a substantial incidence of cerebral embolic infarction after surgery for congenital heart disease. Advances in bypass technique, including refinements in membrane oxygenators, in-line arterial filters, and anticoagulation, have reduced the incidence of macroembolization and large-vessel occlusion. 123 The impact of these advances on the incidence of microembolization and small-vessel disease is difficult to evaluate.
The extensive interface between circulating blood and the artificial surface of the bypass circuit may trigger an inflammatory response, which in turn activates complex physiological cascades, including endothelium–leukocyte interactions. 72 This process further enhances the risk of ischemic injury during the intra-operative and postoperative periods.
In the postoperative period, factors predisposing to stroke include stasis (intracardiac and extracardiac), altered vascular surfaces (native or prosthetic), and, in some situations, a potential procoagulant shift in the humoral clotting systems. 124 Intracardiac stasis may result from localized areas of low flow 125 or global ventricular dysfunction. Transient or sustained elevations of right heart and, hence, central venous pressure in the early postoperative period predispose to local thrombosis in the right atrium and central veins. 126 Prosthetic material in such areas of disturbed flow increases the likelihood of thrombus formation, and the presence of a right-to-left shunt (native or iatrogenic) compounds the risk of paradoxical embolization. Elevated right atrial pressure transmitted to the cerebral venous circulation predisposes to venous thrombosis, particularly in the dural venous sinuses. Elevated systemic venous pressure may cause a protein-losing enteropathy, 127 liver impairment, and pleural effusions, factors that may disturb the humoral coagulant systems. 124 A number of the aforementioned stroke risk factors may be present after the Fontan operation, as highlighted in several reports. 125, 128 In one study, a 2.6 percent incidence of stroke was found in a retrospective review of 645 patients after the Fontan operation; the risk extended over 3 years after the procedure. 125 Rosenthal and co-workers found a 20 percent incidence of thromboembolic complications after the Fontan procedure. 128
Strokes originating during or immediately after cardiac surgery may escape clinical recognition for several days because of the effects of postoperative sedating and paralyzing agents. In the young infant, stroke often presents with focal seizures 129 or changes in mental status; focal motor deficits may be subtle. In older infancy and childhood, stroke usually presents with acute focal motor deficits, language disturbance, or visual dysfunction.
The therapeutic approach to stroke in the child with heart disease includes (1) preventive and (2) “rescue” strategies. Experience with rescue therapies remains confined to adult and experimental stroke. These rescue therapies aim to salvage potentially viable brain using techniques designed to revascularize ischemic brain regions (thrombolytic therapy) or to curtail injurious biochemical cascades. 130 This discussion focuses on the principles of stroke prophylaxis using antithrombotic agents. Preventive stroke therapy may be categorized as primary or secondary. 131 Primary stroke prevention aims to identify and treat high-risk patients before a stroke, whereas secondary prevention aims at minimizing the risk of stroke recurrence. Consistent and universally accepted guidelines for both primary and secondary stroke prophylaxis in children are lacking. Current guidelines are largely empirical, anecdotal, and derived from experience in adults. Established indications for primary stroke prophylaxis in children include prosthetic heart valves, dilated cardiomyopathy, or intracardiac thrombus on echocardiogram.
The decision regarding whether and when to initiate secondary stroke prophylaxis with antithrombotic agents should aim to balance the risk of (1) recurrent cerebral embolization and (2) potentiating secondary hemorrhage into an area of cerebral infarction. Embolus recurrence risks after cardioembolic stroke are unknown in children. In adults (after myocardial infarction), this risk is highest in the early poststroke period, at approximately 1 percent per day (10% to 20% over the first 2 weeks). 132 Cardioembolic strokes are particularly prone to hemorrhagic transformation, especially in the early poststroke period. 133 Hemorrhagic transformation occurs (often silently) in 20 to 40 percent of adult cardioembolic strokes. 133 The risk of significant clinical deterioration after hemorrhagic transformation is greater in the anticoagulated patient.
Although it is difficult to predict which infarcts will undergo hemorrhagic transformation, certain guidelines have been formulated. Among infarcts destined to undergo hemorrhagic transformation, 75 percent do so within 48 hours after stroke onset. 83 Large infarcts, particularly those larger than 30 percent, or one lobe, of a cerebral hemisphere, are at greater risk of hemorrhagic transformation. 134 Uncontrolled systemic hypertension and stroke due to septic emboli and cerebral venous thrombosis are additional risk factors for hemorrhagic infarction. The details of antithrombotic management are discussed elsewhere.
The cerebrovascular disease associated with infective endocarditis warrants brief mention. The protean neurological manifestations of infective endocarditis include meningitis, brain abscess, and seizures. However, cerebrovascular injury, specifically septic embolism and hemorrhage, is the most common complication. Even with advanced antibiotic agents, neurological complications occur in one third of infective endocarditis cases in children; in one half of such cases, the complications are embolic in origin. 135 Cerebrovascular complications carry the highest mortality rate (up to 80% to 90%), primarily due to intracranial hemorrhage. The risk of and mortality rate of cerebral hemorrhage in this population contraindicates anticoagulant therapy. In all cases of cardiogenic stroke, the possibility of septic embolism should be considered before initiating anticoagulant therapy.

Movement Disorders
Reports of serious postoperative movement disorders 136 go back to the early 1960s and the early days of deep hypothermic cardiac surgery. 137, 138 Choreoathetosis is the most frequent form of dyskinesia complicating cardiac surgery; other rarer postoperative movement disorders include oculogyric crises and parkinsonism. The reported incidence of postoperative choreoathetosis reached 19 percent 88 in earlier years 139 ; fortunately, this complication has become rare in recent years. Despite their relative rarity, these movement disorders are often dramatic, frequently intractable, and, in severe cases, associated with a substantial mortality rate.
Postoperative movement disorders have a fairly typical clinical course. The involuntary movements are preceded in most cases by a 2- to 7-day latent period during which postoperative neurological recovery appears to be uncomplicated. Thereafter, a subacute delirium (marked irritability, insomnia, confusion, and disorientation) develops, followed closely by the emergence of involuntary movements. Typically, these movements start in the distal extremities and orofacial muscles, progressing proximally to involve the girdle muscles and trunk. In severe cases, violent ballismic thrashing may develop. The abnormal movements are present during wakefulness, peak with distress, and resolve during brief periods of sleep. Oculomotor and oromotor apraxia are common, with loss of voluntary gaze as well as feeding and expressive language skills. The involuntary movements often intensify over a 1-week period, followed by a 1- to 2-week period during which movements remain relatively constant. The period of recovery is highly variable in duration. The long-term outcome of postoperative movement disorders depends largely on their initial severity. Mild cases tend to resolve within weeks to months, whereas severe cases have a mortality rate approaching 40 percent and a high incidence of associated, significant, long-term neurodevelopmental deficits, including diffuse hypotonia, persistent dyskinesia (47%), and pervasive deficits in memory, attention, language, and motor abilities. 137, 139
The diagnosis of postoperative hyperkinetic syndromes is essentially clinical. Currently available neurodiagnostic studies are useful only for excluding other disorders. Cerebral changes by computed tomography (CT), MRI, and single-photon emission CT are nonspecific, seldom focal, and most commonly consist of diffuse cerebral atrophy 137 and a high incidence of both cortical and subcortical perfusion defects. The electroencephalogram is usually normal or diffusely slow, with no ictal changes associated with the involuntary movements. Descriptions of the neuropathological findings at autopsy are limited 140 and inconsistent. Findings have ranged from normal to extensive neuronal loss and gliosis, particularly in the external globus pallidus. 140 Typical features of infarction are characteristically absent.
Certain risk factors have been suggested, including (1) cyanotic congenital heart disease, particularly with systemic-to-pulmonary collaterals from the head and neck; (2) age at surgery older than 9 months; (3) excessively short cooling periods before attenuation of intraoperative blood-flow; (4) alpha-stat pH management strategy; (5) deep hypothermia and extracorporeal circulation; and (6) preexisting developmental delay. 141 - 143 Postoperative dyskinesias, usually mild and transient, have been reported after prolonged use of fentanyl and midazolam. 144, 145
Once manifest, the involuntary movements are very refractory to treatment and generally respond poorly to a wide variety of conventional antidyskinetic medications, including dopamine receptor blockers (phenothiazines and butyrophenones), dopamine-depleting agents (reserpine, tetrabenazine), dopamine agonists (levodopa), GABAergic agents (benzodiazepines, barbiturates, baclofen), and other agents such as valproic acid, carbamazepine, phenytoin, diphenhydramine, and chloral hydrate. In general, successful movement control has been achieved only at the expense of excessive sedation.
Given these limitations, the management of postoperative movement disorders should focus on the often severe agitation and insomnia. General measures, such as a decrease in the level of external (e.g., noise, light) and internal (e.g., pain) stimuli, are useful in decreasing the intensity of involuntary movements. Judicious use of sedation should aim to restore the fragmented sleep-wakefulness cycle. Oromotor dyskinesia is often severe enough to impair feeding and predispose to aspiration. Nasogastric or even gastrostomy tube feedings may be necessary to meet the high caloric demands of the constant involuntary movements.

Spinal Cord Injury
Spinal cord injury is a relatively rare complication of pediatric cardiac surgery 146, 147 and usually occurs after aortic coarctation repair, in which 0.4 to 1.5 percent of cases may be affected. Intraoperative spinal cord injury is mediated by hypoxic-ischemic/reperfusion injury to watershed territories in the cord, most commonly in the lower thoracic cord, where transverse infarction results in postoperative paraplegia. An additional watershed zone runs between the supply territories of the anterior and posterior spinal arteries; ischemia in this region results in predominant or selective anterior horn cell loss.

Brachial Plexus and Peripheral Nerve Injury
Prolonged immobility during and after cardiac catheterization and surgery predisposes peripheral nerves to pressure and traction injury. Pressure palsies may occur at any dependent site, but most commonly involve the peroneal and ulnar nerves. Brachial plexus injury is not uncommon after cardiac catheterization. 148, 149 Injury to the lower plexus results from prolonged traction during the extreme and sustained arm abduction required in some procedures. This neuropraxic lesion resolves gradually but usually completely. During cardiac catheterization, the insertion of indwelling central venous catheters through the internal jugular vein may injure the upper brachial plexus by direct physical trauma or extravasation of blood into the plexus.
Phrenic nerve injury results from hypothermic injury by ice packed around the heart or from direct intraoperative transection. 150 Postoperative phrenic nerve injury has also been described after malposition of chest tubes. 151, 152 Intraoperative phrenic nerve injury presents with diaphragmatic palsy and prolonged postoperative ventilator dependence. The lesion may be confirmed at the bedside by nerve conduction studies and electromyography. 153 Most phrenic nerve injuries resolve spontaneously, but occasionally diaphragmatic plication or, in rare instances, diaphragmatic pacing is required. 154 Younger infants are more likely than older children to require diaphragm plication. 151, 152
Postoperative ventilation is commonly facilitated by the use of neuromuscular blocking agents. Prolonged use of nondepolarizing agents, especially vecuronium and pancuronium, has been associated with neuromuscular dysfunction. 105, 155 - 157 The concomitant use of steroids may increase the risk. 157 The neuropathological spectrum in these conditions is highly variable, ranging from necrotizing myopathy to axonal motor neuropathy with variable sensory involvement. 156 These conditions may be difficult to distinguish from “critical illness polyneuropathy.”

Cardiac transplantation has become a rescue treatment for children with either primary (myocarditis/cardiomyopathy) or secondary (to associated congenital heart disease) end-stage myocardial failure. 158 Reported 10-year survival rates from various pediatric institutions range from 42 to 73 percent. 159
More effective immunosuppression has advanced the survival of transplant recipients; however, long-term immunosuppression remains a major challenge and has well-recognized neurological complications. The passage to heart transplantation is itself fraught with risk of neurological injury, particularly hypoxic-ischemic, as is the transplantation procedure, which may be complex and involve long periods of bypass support. Adult autopsy studies have described brain injury in more than 80 percent of transplant recipients, consisting of vascular (up to 60%), infectious (20%), and lymphoproliferative disorders (13%). In a recent pediatric autopsy study, brain injury was described in 87 percent of transplant recipients. 160
In the first 2 weeks after transplantation, the most common complications are stroke, drug neurotoxicity, hypoxic-ischemic encephalopathy, and acute psychosis. In a recent report, seizures occurred in 21 percent of children post-transplantation. 161 Later, the complications of chronic immunosuppression, such as opportunistic infections, lymphoma, drug neurotoxicity, and metabolic encephalopathy, are more common. 162 Further discussion of these complications is provided in Chapters 3 and 46 .


1 Benson D. Changing profile of congenital heart disease. Pediatrics . 1989;83:790.
2 Rudolph AM. Congenital Diseases of the Heart: Clinical-Physiologic Considerations in Diagnosis and Management. Chicago: Year Book Medical Publishers, 2001.
3 du Plessis AJ. Neurologic complications of cardiac disease in the newborn. Clin Perinatol . 1997;24:807.
4 du Plessis AJ. Mechanisms of brain injury during infant cardiac surgery. Semin Pediatr Neurol . 1999;6:32.
5 Ferry PC. Neurologic sequelae of open-heart surgery in children. An ‘irritating question.’. Am J Dis Child . 1990;144:369.
6 Shin’oka T, Shum-Tim D, Jonas RA, et al. Higher hematocrit improves cerebral outcome after deep hypothermic circulatory arrest. J Thorac Cardiovasc Surg . 1996;112:1610.
7 Jonas RA, Wypij D, Roth SJ, et al. The influence of hemodilution on outcome after hypothermic cardiopulmonary bypass: results of a randomized trial in infants. J Thorac Cardiovasc Surg . 2003;126:1765.
8 Newburger JW, Jonas RA, Wernovsky G, et al. A comparison of the perioperative neurologic effects of hypothermic circulatory arrest versus low-flow cardiopulmonary bypass in infant heart surgery. N Engl J Med . 1993;329:1057.
9 Bellinger DC, Jonas RA, Rappaport LA, et al. Developmental and neurologic status of children after heart surgery with hypothermic circulatory arrest or low-flow cardiopulmonary bypass. N Engl J Med . 1995;332:549.
10 Bellinger DC, Wypij D, Kuban KC, et al. Developmental and neurological status of children at 4 years of age after heart surgery with hypothermic circulatory arrest or low-flow cardiopulmonary bypass. Circulation . 1999;100:526.
11 Bellinger DC, Wypij D, duDuplessis AJ, et al. Neurodevelopmental status at eight years in children with dextro-transposition of the great arteries: the Boston Circulatory Arrest Trial. J Thorac Cardiovasc Surg . 2003;126:1385.
12 du Plessis AJ, Jonas RA, Wypij D, et al. Perioperative effects of alpha-stat versus pH-stat strategies for deep hypothermic cardiopulmonary bypass in infants. J Thorac Cardiovasc Surg . 1997;114:991.
13 Manzar S, Nair AK, Pai MG, et al. Head size at birth in neonates with transposition of great arteries and hypoplastic left heart syndrome. Saudi Med J . 2005;26:453.
14 Limperopoulos C, Majnemer A, Rosenblatt B, et al. Multimodality evoked potential findings in infants with congenital heart defects. J Child Neurol . 1999;14:702.
15 Limperopoulos C, Majnemer A, Shevell MI, et al. Neurologic status of newborns with congenital heart defects before open heart surgery. Pediatrics . 1999;103:402.
16 Limperopoulos C, Majnemer A, Shevell MI, et al. Neurodevelopmental status of newborns and infants with congenital heart defects before and after open heart surgery. J Pediatr . 2000;137:638.
17 Miller SP, McQuillen PS, Vigneron DB, et al. Preoperative brain injury in newborns with transposition of the great arteries. Ann Thorac Surg . 2004;77:1698.
18 Tavani F, Zimmerman RA, Clancy RR, et al. Incidental intracranial hemorrhage after uncomplicated birth: MRI before and after neonatal heart surgery. Neuroradiology . 2003;45:253.
19 Kaltman JR, Di H, Tian Z, et al. Impact of congenital heart disease on cerebrovascular blood flow dynamics in the fetus. Ultrasound Obstet Gynecol . 2005;25:32.
20 Licht DJ, Wang J, Silvestre DW, et al. Preoperative cerebral blood flow is diminished in neonates with severe congenital heart defects. J Thorac Cardiovasc Surg . 2004;128:841.
21 Te Pas AB, van Wezel-Meijler G, Bokenkamp-Gramann R, et al. Preoperative cranial ultrasound findings in infants with major congenital heart disease. Acta Paediatr . 2005;94:1597.
22 van Houten J, Rothman A, Bejar R. High incidence of cranial ultrasound abnormalities in full-term infants with congenital heart disease. Am J Perinatol . 1996;13:47.
23 Tworetzky W, Marshall AC. Fetal interventions for cardiac defects. Pediatr Clin North Am . 2004;51:1503.
24 Tworetzky W, Wilkins-Haug L, Jennings RW, et al. Balloon dilation of severe aortic stenosis in the fetus: potential for prevention of hypoplastic left heart syndrome: candidate selection, technique, and results of successful intervention. Circulation . 2004;110:2125.
25 Robertson CM, Joffe AR, Sauve RS, et al. Outcomes from an interprovincial program of newborn open heart surgery. J Pediatr . 2004;144:86.
26 Limperopoulos C, Majnemer A, Shevell MI, et al. Predictors of developmental disabilities after open heart surgery in young children with congenital heart defects. J Pediatr . 2002;141:51.
27 Limperopoulos C, Majnemer A, Shevell MI, et al. Functional limitations in young children with congenital heart defects after cardiac surgery. Pediatrics . 2001;108:1325.
28 Majnemer A, Limperopoulos C, Shevell M, et al. Long-term neuromotor outcome at school entry of infants with congenital heart defects requiring open-heart surgery. J Pediatr . 2006;148:72.
29 Miller G, Vogel H. Structural evidence of injury or malformation in the brains of children with congenital heart disease. Semin Pediatr Neurol . 1999;6:20.
30 Eskedal L, Hagemo P, Eskild A, et al. A population-based study of extra-cardiac anomalies in children with congenital cardiac malformations. Cardiol Young . 2004;14:600.
31 Derbent M, Bikmaz YE, Yilmaz Z, et al. Variable phenotype and associations in chromosome 22q11.2 microdeletion. Am J Med Genet A . 2006;140:659.
32 Botto LD, May K, Fernhoff PM, et al. A population-based study of the 22q11.2 deletion: phenotype, incidence, and contribution to major birth defects in the population. Pediatrics . 2003;112:101.
33 Moss E, Wang P, McDonald-McGinn D, et al. Characteristic cognitive profile in patients with a 22q11.2 deletion: verbal IQ exceeds nonverbal IQ. Am J Hum Genet A . 1995;57:20.
34 Bingham PM, Lynch D, McDonald-McGinn D, et al. Polymicrogyria in chromosome 22 deletion syndrome. Neurology . 1998;51:1500.
35 MacKenzie-Stepner K, Witzel MA, Stringer DA, et al. Abnormal carotid arteries in the velocardiofacial syndrome: a report of three cases. Plast Reconstr Surg . 1987;80:347.
36 Mitnick R, Bello J, Shprintzen R. Brain anomalies in velo-cardiofacial syndrome. Am J Med Genet . 1994;54:100.
37 Antshel KM, Conchelos J, Lanzetta G, et al. Behavior and corpus callosum morphology relationships in velocardiofacial syndrome (22q11.2 deletion syndrome). Psychiatry Res . 2005;138:235.
38 Bish JP, Pendyal A, Ding L, et al. Specific cerebellar reductions in children with chromosome 22q11.2 deletion syndrome. Neurosci Lett . 2006;399:245.
39 Campbell LE, Daly E, Toal F, et al. Brain and behaviour in children with 22q11.2 deletion syndrome: a volumetric and voxel-based morphometry MRI study. Brain . 2006;129:1218.
40 Natowicz M, Chatten J, Clancy R, et al. Genetic disorders and major extracardiac anomalies associated with the hypoplastic left heart syndrome. Pediatrics . 1988;82:698.
41 Schaer M, Schmitt JE, Glaser B, et al. Abnormal patterns of cortical gyrification in velo-cardio-facial syndrome (deletion 22q11.2): an MRI study. Psychiatry Res . 2006;146:1.
42 Sugama S, Bingham PM, Wang PP, et al. Morphometry of the head of the caudate nucleus in patients with velocardiofacial syndrome (del 22q11.2). Acta Paediatr . 2000;89:546.
43 Barnea-Goraly N, Menon V, Krasnow B, et al. Investigation of white matter structure in velocardiofacial syndrome: a diffusion tensor imaging study. Am J Psychiatry . 2003;160:1863.
44 Zinkstok J, van Amelsvoort T. Neuropsychological profile and neuroimaging in patients with 22Q11.2 deletion syndrome: a review. Child Neuropsychol . 2005;11:21.
45 Lajiness-O’Neill R, Beaulieu I, Asamoah A, et al. The neuropsychological phenotype of velocardiofacial syndrome (VCFS): relationship to psychopathology. Arch Clin Neuropsychol . 2006;21:175.
46 Vorstman JA, Morcus ME, Duijff SN, et al. The 22q11.2 deletion in children: high rate of autistic disorders and early onset of psychotic symptoms. J Am Acad Child Adolesc Psychiatry . 2006;45:1104.
47 Golding-Kushner K, Weller G, Shpintzen R. Velo-cardio-facial syndrome: language and psychological profiles. J Craniofac Genet . 1985;5:259.
48 Shprintzen R, Goldberg R, Golding-Kushner K, et al. Late-onset psychosis in the velo-cardio-facial syndrome. Am J Med Genet . 1992;42:141.
49 Bish JP, Nguyen V, Ding L, et al. Thalamic reductions in children with chromosome 22q11.2 deletion syndrome. Neuroreport . 2004;15:1413.
50 Eliez S, Barnea-Goraly N, Schmitt JE, et al. Increased basal ganglia volumes in velo-cardio-facial syndrome (deletion 22q11.2). Biol Psychiatry . 2002;52:68.
51 Kates WR, Miller AM, Abdulsabur N, et al. Temporal lobe anatomy and psychiatric symptoms in velocardiofacial syndrome (22q11.2 deletion syndrome). J Am Acad Child Adolesc Psychiatry . 2006;45:587.
52 Glauser T, Rorke L, Weinberg P, et al. Congenital brain anomalies associated with the hypoplastic left heart syndrome. Pediatrics . 1990;85:984.
53 Jones M. Anomalies of the brain and congenital heart disease: a study of 52 necropsy cases. Pediatr Pathol . 1991;11:721.
54 Cheng TO. That balloon atrial septostomy is associated with preoperative stroke in neonates with transposition of the great arteries is another powerful argument in favor of therapeutic closure of every patent foramen ovale. Am J Cardiol . 2006;98:277.
55 McQuillen PS, Hamrick SE, Perez MJ, et al. Balloon atrial septostomy is associated with preoperative stroke in neonates with transposition of the great arteries. Circulation . 2006;113:280.
56 Mahle WT, Tavani F, Zimmerman RA, et al. An MRI study of neurological injury before and after congenital heart surgery. Circulation . 2002;106:I109.
57 McConnell JR, Fleming WH, Chu WK, et al. Magnetic resonance imaging of the brain in infants and children before and after cardiac surgery. A prospective study. Am J Dis Child . 1990;144:374.
58 Reddy VM, Hanley FL. Cardiac surgery in infants with very low birth weight. Semin Pediatr Surg . 2000;9:91.
59 Volpe JJ. Intracranial hemorrhage. In Neurology of the Newborn , 4th Ed., Philadelphia: WB Saunders; 2001:397.
60 Glauser T, Rorke L, Weinberg P, et al. Acquired neuropathologic lesions associated with the hypoplastic left heart syndrome. Pediatrics . 1990;85:991.
61 Young R, Liberthson R, Zalneraitis E. Cerebral hemorrhage in neonates with coarctation of the aorta. Stroke . 1982;13:491.
62 Giuliani R, Szwarcer E, Aquino E, et al. Fibrin-dependent fibrinolytic activity during extracorporeal circulation. Thromb Res . 1991;61:369.
63 Rudack D, Baumgart S, Gross G. Subependymal (grade 1) intracranial hemorrhage in neonates on extracorporeal membrane oxygenation. Clin Pediatr . 1994;33:583.
64 von Allmen D, Babcock D, Matsumoto J, et al. The predictive value of head ultrasound in the ECMO candidate. J Pediatr Surg . 1992;27:36.
65 du Plessis AJ. Cerebral hemodynamics and metabolism during infant cardiac surgery. Mechanisms of injury and strategies for protection. J Child Neurol . 1997;12:285.
66 du Plessis AJ, Johnston MV. The pursuit of effective neuroprotection during infant cardiac surgery. Semin Pediatr Neurol . 1999;6:55.
67 Kinney HC, Panigrahy A, Newburger JW, et al. Hypoxic-ischemic brain injury in infants with congenital heart disease dying after cardiac surgery. Acta Neuropathol (Berl) . 2005;110:563.
68 Volpe JJ. Hypoxic-ischemic encephalopathy: biochemical and physiological aspects. In In Neurology of the Newborn , 4th Ed., Philadelphia: WB Saunders; 2001:217.
69 Volpe JJ. Hypoxic-ischemic encephalopathy: neuropathology and pathogenesis. In Neurology of the Newborn , 4th Ed., Philadelphia: WB Saunders; 2001:296.
70 Redmond J, Gillinov A, Zehr K, et al. Glutamate excitotoxicity: a mechanism of neurologic injury associated with hypothermic circulatory arrest. J Thorac Cardiovasc Surg . 1994;107:776.
71 Moody D, Bell M, Challa V, et al. Brain microemboli during cardiac surgery or aortography. Ann Neurol . 1990;28:477.
72 Casey L. Role of cytokines in the pathogenesis of cardiopulmonary-induced multisystem organ failure. Ann Thorac Surg . 1993;56:S92.
73 Westaby S. Organ dysfunction after cardiopulmonary bypass: a systemic inflammatory reaction by the extracorporeal circuit. Intensive Care Med . 1987;13:89.
74 Pesonen EJ, Korpela R, Peltola K, et al. Regional generation of free oxygen radicals during cardiopulmonary bypass in children. J Thorac Cardiovasc Surg . 1995;110:768.
75 Pyles LA, Fortney JE, Kudlak JJ, et al. Plasma antioxidant depletion after cardiopulmonary bypass in operations for congenital heart disease. J Thorac Cardiovasc Surg . 1995;110:165.
76 Ledeist F, Menasche P, Kucharski C, et al. Hypothermia during cardiopulmonary bypass delays but does not prevent neutrophil-endothelial cell adhesion. A clinical study. Circulation . 1995;92:II.
77 Andersen K, Waaben J, Husum B, et al. Nonpulsatile cardiopulmonary bypass disrupts the flow-metabolism couple in the brain. J Thorac Cardiovasc Surg . 1985;90:570.
78 Speziali G, Russo P, Davis D, et al. Hypothermia enhances contractility in cerebral arteries of newborn lambs. J Surg Res . 1994;57:80.
79 Greeley W, Ungerleider R, Smith L, et al. The effects of deep hypothermic cardiopulmonary bypass and total circulatory arrest on cerebral blood flow in infants and children. J Thorac Cardiovasc Surg . 1989;97:737.
80 Murkin JM, Farrar JK, Tweed WA, et al. Cerebral autoregulation and flow/metabolism coupling during cardiopulmonary bypass: the influence of PaCO 2 . Anesth Analg . 1987;66:825.
81 Coetzee A, Swanepoel C. The oxyhemoglobin dissociation curve before, during and after cardiac surgery. Scand J Clin Lab Invest Suppl . 1990;203:149.
82 Powers W. Hemodynamics and metabolism in ischemic cerebrovascular disease. Neurol Clin . 1992;10:31.
83 Dexter F, Hindman B. Theoretical analysis of cerebral venous blood hemoglobin oxygen saturation as an index of cerebral oxygenation during hypothermic cardiopulmonary bypass. Anesthesiology . 1995;83:405.
84 Forbess JM, Visconti KJ, Hancock-Friesen C, et al. Neurodevelopmental outcome after congenital heart surgery: results from an institutional registry. Circulation . 2002;106:I95.
85 Kern JH, Hinton VJ, Nereo NE, et al. Early developmental outcome after the Norwood procedure for hypoplastic left heart syndrome. Pediatrics . 1998;102:1148.
86 Uzark K, Lincoln A, Lamberti JJ, et al. Neurodevelopmental outcomes in children with Fontan repair of functional single ventricle. Pediatrics . 1998;101:630.
87 Wernovsky G, Stiles KM, Gauvreau K, et al. Cognitive development after the Fontan operation. Circulation . 2000;102:883.
88 McGrath E, Wypij D, Rappaport LA, et al. Prediction of IQ and achievement at age 8 years from neurodevelopmental status at age 1 year in children with D-transposition of the great arteries. Pediatrics . 2004;114:e572.
89 Kirshbom PM, Flynn TB, Clancy RR, et al. Late neurodevelopmental outcome after repair of total anomalous pulmonary venous connection. J Thorac Cardiovasc Surg . 2005;129:1091.
90 Mahle WT, Clancy RR, Moss EM, et al. Neurodevelopmental outcome and lifestyle assessment in school-aged and adolescent children with hypoplastic left heart syndrome. Pediatrics . 2000;105:1082.
91 Wernovsky G, Wypij D, Jonas RA, et al. Postoperative course and hemodynamic profile after the arterial switch operation in neonates and infants: a comparison of low-flow cardiopulmonary bypass and circulatory arrest. Circulation . 1995;92:2226.
92 Wypij D, Newburger JW, Rappaport LA, et al. The effect of duration of deep hypothermic circulatory arrest in infant heart surgery on late neurodevelopment: the Boston Circulatory Arrest Trial. J Thorac Cardiovasc Surg . 2003;126:1397.
93 Sorensen H, Husum B, Waaben J, et al. Brain microvascular function during cardiopulmonary bypass. J Thorac Cardiovasc Surg . 1987;94:727.
94 Lundar T, Lindegaard K, Froysaker T, et al. Dissociation between cerebral autoregulation and carbon dioxide reactivity during nonpulsatile cardiopulmonary bypass. Ann Thorac Surg . 1985;40:582.
95 Bellinger DC, Wypij D, du Plessis AJ, et al. Developmental and neurologic effects of alpha-stat versus pH-stat strategies for deep hypothermic cardiopulmonary bypass in infants. J Thorac Cardiovasc Surg . 2001;121:374.
96 du Plessis A, Newburger J, Jonas R, et al. Cerebral oxygen supply and utilization during infant cardiac surgery. Ann Neurol . 1995;37:488.
97 Ruvolo G, Greco E, Speziale G, et al. Nitric oxide formation during cardiopulmonary bypass. Ann Thorac Surg . 1994;57:1055.
98 Dugan LL, Sensi SL, Canzoniero LM, et al. Mitochondrial production of reactive oxygen species in cortical neurons following exposure to N-methyl-D-aspartate. J Neurosci . 1995;15:6377.
99 Nathan HJ, Polls T. The management of temperature during hypothermic cardiopulmonary bypass: 2. Effect of prolonged hypothermia. Can J Anaesth . 1995;42:672.
100 Ginsberg M, Sternau L, Globus M-T, et al. Therapeutic modulation of brain temperature: relevance to ischemic brain injury. Cerebrovasc Brain Metab Rev . 1992;4:189.
101 Astudillo R, van der Linden J, Ekroth R, et al. Absent diastolic cerebral blood flow velocity after circulatory arrest but not after low flow in infants. Ann Thorac Surg . 1993;56:515.
102 Jonassen A, Quaegebeur J, Young W. Cerebral blood flow velocity in pediatric patients is reduced after cardiopulmonary bypass with profound hypothermia. J Thorac Cardiovasc Surg . 1995;110:934.
103 Menache CC, du Plessis AJ, Wessel DL, et al. Current incidence of acute neurologic complications after open-heart operations in children. Ann Thorac Surg . 2002;73:1752.
104 Plum F, Posner J. Multifocal, diffuse, and metabolic brain diseases causing stupor or coma. In The diagnosis of stupor and coma , 3rd Ed., Philadelphia: FA Davis; 1985:177.
105 Waitling S, Dasta J. Prolonged paralysis in intensive care unit patients after use of neuromuscular blocking agents: a review of the literature. Crit Care Med . 1994;22:884.
106 Clancy RR, McGaurn SA, Wernovsky G, et al. Risk of seizures in survivors of newborn heart surgery using deep hypothermic circulatory arrest. Pediatrics . 2003;111:592.
107 Clancy RR, Sharif U, Ichord R, et al. Electrographic neonatal seizures after infant heart surgery. Epilepsia . 2005;46:84.
108 Gaynor JW, Jarvik GP, Bernbaum J, et al. The relationship of postoperative electrographic seizures to neurodevelopmental outcome at 1 year of age after neonatal and infant cardiac surgery. J Thorac Cardiovasc Surg . 2006;131:181.
109 Mizrahi EM, Clancy RR. Neonatal seizures: early-onset seizure syndromes and their consequences for development. Ment Retard Dev Disabil Res Rev . 2000;6:229.
110 Dietrich W, Busto R, Alonso O, et al. Intraischemic but not postischemic brain hypothermia protects chronically following global forebrain ischemia in rats. J Cereb Blood Flow Metab . 1993;13:541.
111 Lynch B, Rust R. Natural history and outcome of neonatal hypocalcemic and hypomagnesemic seizures. Pediatr Neurol . 1994;11:23.
112 Satur C, Jennings A, Walker D. Hypomagnesemia and fits complicating pediatric cardiac surgery. Ann Clin Biochem . 1993;30:315.
113 Rappaport LA, Wypij D, Bellinger DC, et al. Relation of seizures after cardiac surgery in early infancy to neurodevelopmental outcome. Boston Circulatory Arrest Study Group. Circulation . 1998;97:773.
114 du Plessis A, Kramer U, Jonas R, et al. West syndrome following deep hypothermic cardiac surgery. Pediatr Neurol . 1994;11:246.
115 Yang J, Park Y, Hartlage P. Seizures associated with stroke in childhood. Pediatr Neurol . 1995;12:136.
116 Gaynor JW. Periventricular leukomalacia following neonatal and infant cardiac surgery. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu . 2004;7:133.
117 Galli KK, Zimmerman RA, Jarvik GP, et al. Periventricular leukomalacia is common after neonatal cardiac surgery. J Thorac Cardiovasc Surg . 2004;127:692.
118 Ashwal S, Holshouser B, Schell R, et al. Proton magnetic resonance spectroscopy in the evaluation of children with congenital heart disease and acute central nervous system injury. J Thorac Cardiovasc Surg . 1996;112:403.
119 Ashwal S, Holshouser BA, del Rio MJ, et al. Serial proton magnetic resonance spectroscopy of the brain in children undergoing cardiac surgery. Pediatr Neurol . 2003;29:99.
120 deVeber G. Arterial ischemic strokes in infants and children: an overview of current approaches. Semin Thromb Hemost . 2003;29:567.
121 Hutchison JS, Ichord R, Guerguerian AM, et al. Cerebrovascular disorders. Semin Pediatr Neurol . 2004;11:139.
122 Barker PC, Nowak C, King K, et al. Risk factors for cerebrovascular events following Fontan palliation in patients with a functional single ventricle. Am J Cardiol . 2005;96:587.
123 Nussmeier N, McDermott J. Macroembolization: prevention and outcome modification. In: Hilberman M, editor. Brain Injury and Protection During Cardiac Surgery . Boston: Martinus Nijhoff; 1988:85.
124 Cromme-Dijkhuis A, Henkens C, Bijleveld C, et al. Coagulation factor abnormalities as possible thrombotic risk factors after Fontan operations. Lancet . 1990;336:1087.
125 du Plessis A, Chang A, Wessel D, et al. Cerebrovascular accidents following the Fontan procedure. Pediatr Neurol . 1995;12:230.
126 Dobell A, Trusler G, Smallhorn J, et al. Atrial thrombi after the Fontan operation. Ann Thorac Surg . 1986;42:664.
127 Hess J, Kruizinga A, Bijleveld C, et al. Protein-losing enteropathy after Fontan operation. J Thorac Cardiovasc Surg . 1984;88:606.
128 Rosenthal DN, Friedman AH, Kleinman CS, et al. Thromboembolic complications after Fontan operations. Circulation . 1995;92(suppl II):287.
129 Levy SR, Abroms IF, Marshall PC, et al. Seizures and cerebral infarction in the full-term newborn. Ann Neurol . 1985;17:366.
130 Ginsberg MD. Emerging strategies for the treatment of ischemic brain injury. In: Waxman SG, editor. Molecular and Cellular Approaches to the Treatment of Neurological Disease . New York: Raven Press; 1993:207.
131 Anderson D. Cardioembolic stroke: primary and secondary prevention. Postgrad Med . 1991;90:67.
132 Cerebral Embolism Study Group. Cardioembolic stroke, immediate anticoagulation, and brain hemorrhage. Arch Intern Med . 1987;147:636.
133 Hart R, Easton J. Hemorrhagic infarcts. Stroke . 1986;17:586.
134 Yatsu F, Hart R, Mohr J, et al. Anticoagulation of embolic strokes of cardiac origin: an update. Neurology . 1988;38:314.
135 Saiman L, Prince A, Gersony W. Pediatric infective endocarditis in the modern era. J Pediatr . 1993;122:847.
136 Bjork V, Hultquist G. Contraindications to profound hypothermia in open-heart surgery. J Thorac Cardiovasc Surg . 1962;44:1.
137 Wong PC, Barlow CF, Hickey PR, et al. Factors associated with choreoathetosis after cardiopulmonary bypass in children with congenital heart disease. Circulation . 1992;86:118.
138 Curless RG, Katz DA, Perryman RA, et al. Choreoathetosis after surgery for congenital heart disease. J Pediatr . 1994;124:737.
139 du Plessis AJ, Bellinger DC, Gauvreau K, et al. Neurologic outcome of choreoathetoid encephalopathy after cardiac surgery. Pediatr Neurol . 2002;27:9.
140 Kupsky WJ, Drozd MA, Barlow CF. Selective injury of the globus pallidus in children with post-cardiac surgery choreic syndrome. Dev Med Child Neurol . 1995;37:135.
141 Deleon GA, Radkowski MA, Crawford SE, et al. Persistent respiratory failure due to low cervical cord infarction in newborn babies. J Child Neurol . 1995;10:200.
142 Levin DA, Seay AR, Fullerton DA, et al. Profound hypothermia with alpha-stat pH management during open-heart surgery is associated with choreoathetosis. Pediatr Cardiol . 2005;26:34.
143 Hamrick SE, Gremmels DB, Keet CA, et al. Neurodevelopmental outcome of infants supported with extracorporeal membrane oxygenation after cardiac surgery. Pediatrics . 2003;111:e671.
144 Bergman I, Steeves M, Burckart G, et al. Reversible neurologic abnormalities associated with prolonged intravenous midazolam and fentanyl administration. J Pediatr . 1991;119:644.
145 Petzinger G, Mayer SA, Przedborski S. Fentanyl-induced dyskinesias. Mov Disord . 1995;10:679.
146 Puntis J, Green S. Ischemic spinal cord injury after cardiac surgery. Arch Dis Child . 1985;60:517.
147 Christenson JT, Sierra J, Didier D, et al. Repair of aortic coarctation using temporary ascending to descending aortic bypass in children with poor collateral circulation. Cardiol Young . 2004;14:39.
148 Souza Neto EP, Durand PG, Sassolas F, et al. Brachial plexus injury during cardiac catheterisation in children. Report of two cases. Acta Anaesthesiol Scand . 1998;42:876.
149 Liu XY, Wong V, Leung M. Neurologic complications due to catheterization. Pediatr Neurol . 2001;24:270.
150 Mok Q, Ross-Russell R, Mulvey D, et al. Phrenic nerve injury in infants and children undergoing cardiac surgery. Br Heart J . 1991;65:287.
151 Hwang MS, Chu JJ, Su WJ. Diaphragmatic paralysis caused by malposition of chest tube placement after pediatric cardiac surgery. Int J Cardiol . 2005;99:129.
152 Joho-Arreola AL, Bauersfeld U, Stauffer UG, et al. Incidence and treatment of diaphragmatic paralysis after cardiac surgery in children. Eur J Cardiothorac Surg . 2005;27:53.
153 Bolton C. Clinical neurophysiology of the respiratory system. Muscle Nerve . 1993;16:809.
154 Weese-Mayer DE, Hunt CE, Brouillette RT, et al. Diaphragm pacing in infants and children. J Pediatr . 1992;120:1.
155 Segredo V, Caldwell JE, Matthay MA, et al. Persistent paralysis in critically ill patients after long-term administration of vecuronium. N Engl J Med . 1992;327:524.
156 Danon M, Carpenter S. Myopathy with thick filament (myosin) loss following prolonged paralysis with vecuronium during steroid treatment. Muscle Nerve . 1991;14:1131.
157 Benzing G, Iannacone S, Bove K, et al. Prolonged myasthenic syndrome after one week of muscle relaxants. Pediatr Neurol . 1990;6:190.
158 Johnston JK, Chinnock RE, Zuppan CW, et al. Limitations to survival for infants with hypoplastic left heart syndrome before and after transplant: the Loma Linda experience. J Transpl Coord . 1997;7:180.
159 Bauer J, Thul J, Kramer U, et al. Heart transplantation in children and infants: short-term outcome and long-term follow-up. Pediatr Transplant . 2001;5:457.
160 McClure CD, Johnston JK, Fitts JA, et al. Post-mortem intracranial neuropathology in children following cardiac transplantation. Pediatr Neurol . 2006;35:107.
161 Raja R, Johnston JK, Fitts JA, et al. Post-transplant seizures in infants with hypoplastic left heart syndrome. Pediatr Neurol . 2003;28:370.
162 Hotson J, Enzmann D. Neurologic complications of cardiac transplantation. Neurol Clin . 1988;6:349.
Chapter 5 Neurological Manifestations of Acquired Cardiac Disease, Arrhythmias, and Interventional Cardiology

Colin D. Lambert, David J. Gladstone

Clinical Features
Atrial Fibrillation and Flutter
Cardioversion in Atrial Fibrillation or Flutter
Chronic Sinoatrial Disorder
Myocardial Infarction and Left Ventricular Dysfunction
Rheumatic Heart Disease
Atrial Myxoma
Marantic (Nonbacterial Thrombotic) Endocarditis
Other Echocardiographic Abnormalities Linked to Stroke
Acute Medical Treatment of Cardiogenic Embolism
Coronary Catheterization
Percutaneous Transluminal Coronary Angioplasty and Stenting
Thrombolytic Therapy for Acute Myocardial Infarction
The neurological manifestations of acquired cardiac disease fall into several categories:
1. The sudden onset of a focal neurological deficit due to occlusion of a cerebral or retinal artery by an embolus that has developed within the heart (cardiogenic embolism)
2. Transient, self-limited episodes of generalized cerebral ischemia that occur as a consequence of brief failures of cardiac output, due to either rhythm disturbances or outflow obstruction, resulting in presyncope or syncope
3. The complications of invasive techniques for the investigation or management of cardiac disease
The major exceptions to these generalizations occur with atrial fibrillation (AF), which is associated with embolus formation rather than syncope, and with chronic sinoatrial disorder, which predisposes to both syncopal and embolic disturbances.
Topics that are the focus of other chapters are not considered here. In this chapter, the term stroke is used to mean the sudden onset of a focal neurological deficit of ischemic origin. Cerebral embolus is used where the deficit is thought to be of embolic origin. The term cardiogenic embolism is reserved for events in which the embolic occlusion is considered to be the result of a cardiac source of emboli.
This chapter addresses three major situations: (1) cardiogenic embolism, (2) arrhythmias and their manifestations (syncope), and (3) interventional procedures.


Clinical Features
Ischemic stroke or transient ischemic attack (TIA) may be classified into six major etiological categories, which have implications for treatment and prognosis. 1 This is the TOAST (Trial of ORG 10172 in Acute Stroke Treatment) classification, the standard now for clinical studies. These categories are cardioembolism, large-artery atherosclerosis, small-artery (lacunar) occlusion, stroke of other determined etiology, stroke of undetermined etiology, and events of multiple possible etiologies. The first four categories are also subdivided into probable or possible. Strokes in the undetermined group are classed as either completely or incompletely evaluated. The last category accommodates those in whom more than one established cause is present.
Cardiogenic brain embolism accounts for about 20 percent of acute ischemic strokes overall. Coexistent pathology (i.e., arterial and heart disease in the same patient) may be present in up to one third of patients with a potential cardiac source of embolism. 2 The most common cardiac cause of ischemic stroke is AF, which accounts for about one sixth of all strokes. 3 Other cardiac causes of stroke are listed in Table 5-1 .
TABLE 5-1 Established and Putative Cardiac Causes of Stroke
Atrial fibrillation
Atrial flutter
Sick-sinus syndrome
Valvular heart disease
Mitral valve prolapse
Calcific aortic stenosis
Aortic sclerosis
Mitral annular calcification
Myocardial infarction (acute and chronic)
Left ventricular dysfunction
Congestive heart failure
Other echocardiographic abnormalities
Patient foramen ovale ± atrial septal aneurysm
Left atrial thrombus
Spontaneous left atrial echo contrast
Cardiac tumors
Marantic nonbacterial thrombotic
Iatrogenic causes
Cardiac surgery
Cardiac catheterization
Percutaneous coronary interventions
Thrombolytic therapy for acute myocardial infarction
Cardioversion for atrial fibrillation/flutter
In the young stroke population (generally regarded as patients who have their first stroke around the age of 15 to 45 years), 60 or so causes had to be considered in one study. 4 In that study of 329 patients, cardioembolism was thought to be responsible in 64 (just under 20 percent). There were 13 diagnoses in these 64 patients, with the top three being paradoxical embolism and prosthetic or rheumatic valve disease. No patients had AF, a feature also noted in a Swedish study. 5 Strokes attributable to a cardiac source show striking differences in various studies. In a Persian study of 124 patients, 54 percent were thought to be of cardiac origin. 6 Rheumatic heart disease was the major culprit. In contrast, a French study of 296 patients attributed less than 9 percent to a cardiac cause. 7 In Italy, the figure was 34 percent. 8 This was a hospital-based study of 394 consecutive young adults with ischemic stroke submitted to a comprehensive diagnostic protocol. Of the 133 considered to be of cardiac origin, these were subdivided into two groups according to TOAST criteria. The smaller group (23) had a probable cause including recent myocardial infarction, AF, valvulopathy, patent foramen ovale (PFO) with deep vein thrombosis (DVT) and atrial myxoma. The much larger group (110 patients) had various possible causes: PFO with right to left shunt (60), atrial septal aneurysm (ASA) (22), and PFO plus ASA (16). Looked at another way, 23 of 394 patients (6%) had an established cardiac cause. Attribution was less certain in 28 percent. In Korea and Taiwan, around 18 percent of cases were attributed to a cardiac cause. 9, 10 Comparison of etiological factors in the occurrence of TIAs in younger, as opposed to older, patients disclosed that only two cardiac sources were encountered more frequently in the younger age group: valvular heart disease and mitral valve prolapse. 11
Features suggesting cardioembolism are usually derived from analysis of the clinical presentation and neuroimaging features of acute ischemic strokes that occur in patients with cardiac abnormalities thought to predispose to thrombus formation ( Table 5-2 ). 12 - 19 Emboli may lodge in either the anterior (carotid) or the posterior (vertebrobasilar) circulation. The anterior circulation is affected four times more commonly than the posterior. Least likely to be affected are the entire internal carotid artery, deep branches of the middle cerebral artery, and brainstem. 20 Although the posterior circulation is less commonly affected, studies of the mechanism of infarction in specific territories (e.g., those of the posterior inferior cerebellar artery and superior cerebellar artery) implicate cardiogenic embolism in 50 percent of cases. 21 A cardioembolic mechanism occurred in 67 percent of cases with isolated cerebellar infarcts (i.e., without concomitant brainstem or occipital infarcts). 22 Embolism is also a common mechanism of infarction within the territory of the posterior cerebral artery. 23
TABLE 5-2 Clinical Features Suggesting Cardioembolic Rather Than Non-cardioembolic Stroke
Cortical signs (e.g., aphasia, neglect, visual field defect)
Isolated global aphasia or Wernicke’s aphasia (without hemiparesis)
Impaired consciousness at stroke onset
Sudden onset, reaching maximal deficit within 5 minutes of onset
Rapid dramatic neurological recovery (“spectacular shrinking deficit”)
Simultaneous or sequential strokes in different vascular territories
Evidence of systemic embolism
Atrial fibrillation, valvular heart disease
Stroke recurrence rate and prognosis have been estimated in several studies. A meta-analysis showed that the 3-month risk of recurrent stroke was 12 percent if the etiology was cardioembolism, compared to 19 percent for large-vessel atherosclerosis, 3 percent for small-vessel disease, and 9 percent for unknown cause. 24 In a population-based study of first stroke in Bavaria, patients with cardioembolic stroke had the lowest 2-year survival rate (55%) and were three times more likely to be dead at 2 years compared to those with small-artery occlusion. 25

The first neurological investigation for suspected stroke is usually a computed tomography (CT) scan of the head to exclude intracranial hemorrhage or other nonischemic pathological processes and to identify signs of acute infarction or vessel occlusion. In patients at high-risk of cardioembolism, cranial CT disclosed infarcts that were more likely to involve one half of a lobe or more, or the infarcts involved both superficial and deep structures. 12 Deep small infarcts were underrepresented and were considered to have a predictive value of 90 percent for the absence of a major cardiac source. 12 Similar conclusions were drawn in an earlier study, namely, that the mechanism underlying lacunes is infrequently embolic and that infarctions in the pial (superficial) artery territory are usually indicative of an embolic mechanism. 26
The potential for embolic infarcts to develop hemorrhagic transformation remains a concern, especially when anticoagulant therapy has to be considered. A hemorrhagic infarct was seen on the initial CT scan of 6 percent of patients in a series of 244 cases, none of whom were receiving anticoagulants. 12 In a series of scans performed within 48 hours of onset, the figure rose to 24 percent 27 ; on prospective follow-up scanning, a total of 40 percent was found at 1 month. 28 With the more sophisticated technology of magnetic resonance imaging (MRI), the figure rose to nearly 70 percent at 3 weeks. Both of the latter studies showed that larger infarcts were more liable to demonstrate hemorrhagic transformation, with a figure of 90 percent for infarcts with a volume greater than 10 cm 3 . 29 Thus, the key factors that determine whether hemorrhagic transformation occurs appear to be the time of the study, size of the infarct, and technology applied. The age of the patient may also be a factor in that patients older than 70 years may be more liable to hemorrhagic transformation. 28
Because of concerns for the complications of acute stroke treatment by thrombolysis or anticoagulation, early pointers to hemorrhagic transformation have been sought. The only independent predictor identified in a study of 150 consecutive patients was focal hypodensity found by CT scanning within the first 5 hours after stroke onset. Mortality was twice as high in the hemorrhagic-transformation group owing to the larger size of infarcts in that group. Evolution of the transformation process was similar in anticoagulated and nonanticoagulated patients. 29
MRI is the most sensitive test for detecting early infarction. Diffusion-weighted images are superior to T2-weighted images and to CT. 30 The pattern of diffusion-weighted imaging abnormalities can help to determine the most likely etiological diagnosis. For example, a pattern of multiple acute lesions in more than one vascular territory (bilateral lesions or lesions in the anterior and posterior circulations) suggests a shower of cardiogenic emboli. Single cortical-subcortical lesions are also associated with a cardiac source of emboli.
Conventional catheter angiography remains the definitive method for assessing structural abnormalities of the extra- and intracranial circulation. Use of this invasive procedure requires recognition of the associated risks. A review of 15 studies (8 prospective) concluded that the mortality rate was very low (less than 0.1%) but that the risk of a neurological complication (TIA or stroke) was approximately 4 percent and that of a permanent neurological deficit was 1 percent. 31 The characteristic angiographic appearance of an embolic occlusion is of a proximal, meniscus-like filling defect in an artery that is otherwise normal and lacks evidence of atherosclerotic change. Emboli tend to fragment. In a study of 142 patients who underwent angiography, the initial procedure, performed at a median of 1.5 days after the precipitating event, revealed an occlusion in 82 percent. Follow-up angiography, at a median of 20 days, showed reopening of the vessels in 95 percent. 28 Distal branch occlusions are often also considered to be embolic manifestations. Conventional catheter angiography has now been largely replaced by noninvasive contrast-enhanced CT angiography or magnetic resonance angiography in many countries because of increased availability and lower complication rates.
Echocardiography has come to occupy a preeminent place in the structural evaluation of the heart. Transthoracic echocardiography (TTE) is noninvasive but has limitations that can be overcome by using the transesophageal (transesophageal echocardiography [TEE]) route. For the latter procedure, the patient is usually mildly sedated and topical anesthetic is applied to the posterior pharynx. In experienced hands, the procedure was successfully accomplished in 98 percent of instances. The complication rate was less than 1 percent. 32 The technique employed (TEE or TTE) depends on the area to be visualized. The two procedures can be considered complementary; TTE images the left ventricle well, but TEE is required for adequate assessment of the left atrium and its appendage. TEE is also better for visualizing a PFO. TEE is the most sensitive and specific test for detecting a cardiac source of embolism. For patients with AF, TEE may assist in risk stratification and guide cardioversion. 3
A review of papers published between 1966 and 1998 evaluated the yield of TTE or TEE, or both, in various subgroups of patients with stroke. The figures reached were, for TTE, an overall yield of less than 1 percent in patients without clinical evidence of cardiac disease, rising to 13 percent in those with cardiac disease. The corresponding figures for TEE were less than 2 percent and 19 percent. 33 The recommendations reached highlight some uncertainties. It was concluded that there was fair evidence to recommend echocardiography in patients with stroke and clinical evidence of heart disease (grade B recommendation). Because the yield from TEE is higher than that for TTE, controversy arises as to whether this should be the first intervention. Some have preferred a sequential approach with TTE followed by TEE, if indicated, 34 but others have suggested that it is more cost-effective to proceed directly to TEE. 35 Clearly, the area to be visualized is a major consideration.
Cardiac MRI is emerging as a new technology for noninvasive structural imaging of the heart. MRI is more sensitive than TTE and comparable to TEE for the detection of cardiac thrombi.
Transcranial Doppler ultrasonography is a noninvasive tool that can be of value in the acute stroke setting for detecting acute intracranial vascular obstruction (e.g., due to an occlusive embolus in the middle cerebral artery) and can monitor recanalization following treatment with thrombolysis. It can also be used to detect right-to-left cardiac shunts due to PFO. By identifying microbubbles reaching the middle cerebral arteries, especially following the Valsalva maneuver, contrast-enhanced transcranial Doppler ultrasonography has shown near-perfect correlation with contrast-enhanced TEE for the detection and quantification of such shunts. 36, 37
It remains necessary for the clinician to balance extensive investigation against its impact on patient management, usually the justification for lifelong anticoagulant therapy and its consequent risks. In several situations, there are no established guidelines for management. The onus remains on the clinician to determine the significance of potential sources of emboli and their implications for management.


Atrial Fibrillation and Flutter
Atrial fibrillation, the most common arrhythmia in medical practice, is a major risk factor for stroke and death. This arrhythmia accounts for nearly half of all cardiac causes of stroke and about one quarter of strokes in the elderly. 3 Strokes associated with AF are generally severe, and 1-year mortality is 50 percent. 38 AF is also a risk factor for silent strokes and vascular dementia. 39, 40
The prevalence of AF is strongly age dependent, ranging from 0.1 percent among adults older than 55 years to 9 percent in those 80 years or older. 38 Over 2 million individuals have AF in the United States, and prevalence is rising. 38 AF typically occurs in patients with underlying cardiac disease (i.e., valvular heart disease, heart failure, coronary disease, hypertension, cardiomyopathy, mitral valve prolapse, mitral annular calcification, and cardiac tumors), but may also occur as “lone AF” in young patients who have no cardiac disease. It may be paroxysmal (self-terminating episode, lasting less than 7 days), recurrent (2 or more episodes), persistent (more than 7 days), or permanent (cardioversion failed or not attempted). Reversible or temporary causes include alcohol, surgery, hyperthyroidism, acute myocardial infarction, pulmonary embolism, and pericarditis, among others. 3
The average annual risk of stroke in individuals with AF is 5 percent and is heavily dependent on age and the presence of additional risk factors ( Table 5-3 ). In the Framingham Study, stroke risk was 1.5 percent in the age group 50 to 59 years and 23.5 percent in those 80 to 89 years. 41

TABLE 5-3 Two-Year Stroke Risk for Patients With Atrial Fibrillation Stratified by Additional Risk Factors
It is well established that the risk of stroke in AF is related to the presence or absence of associated structural cardiac disease and other risk factors. For example, in the absence of rheumatic heart disease, there is a fivefold increase in stroke incidence, but this increases to 17-fold when associated with rheumatic mitral valve disease. 42 Only in lone AF (i.e., fibrillation in the absence of overt cardiovascular disease or precipitating illness) developing in middle age is the prognosis relatively benign. Follow-up at 15 years disclosed a rate of thromboembolic events of 0.55 per 100 person-years. 43 This was equivalent to 1.3 percent of the patients experiencing a stroke on a cumulative actuarial basis.
The most important predictor of stroke risk in patients with AF is a history of thromboembolism (i.e., previous TIA, stroke, or systemic arterial embolism). Other independent risk factors for stroke in AF are hypertension, heart failure, increasing age, and diabetes mellitus. Other factors that have been associated with increased stroke risk in some studies include female sex, systolic hypertension, and left ventricular dysfunction. 3
Echocardiographic features that have been used for risk stratification in patients with AF include left ventricular systolic dysfunction, atrial thrombus, dense spontaneous echo contrast or reduced blood flow velocity within the left atrium or left atrial appendage on TEE, and aortic atheroma. Left atrial size does not appear to predict risk of thromboembolism. 44 TEE is the method of choice for evaluating the left atrial appendage, the site at which most thrombi form, and the left atrium. In a prospective study of patients with AF considered on clinical grounds to be at high risk of stroke, risk was 18 percent per year in those with dense spontaneous echo contrast who were treated with low-dose warfarin (international normalized ratio [INR] 1.2 to 1.5) plus aspirin compared to 4.5 percent for those on dose-adjusted warfarin. Prevalence of thrombus in the left atrial appendage was similar initially in the two treatment groups (10% to 12%) when TEE was performed more than 2 weeks after study entry, but atrial thrombus was present in 6 percent of those on warfarin compared to 18 percent of those on combination therapy, and stroke rate was 13 percent per year in the latter group. Absence of thrombus predicted a low rate of ischemic events (2.3% per year); the presence of thrombus predicted a high rate (18% per year). 45
That the risk of stroke in AF can be significantly reduced by anticoagulation was clearly established by four independent studies. 46 - 49 A fifth, Canadian, study was terminated prior to completion because the other studies had shown clear evidence of benefit. 50 A meta-analysis published in 1999 evaluated 16 trials. 51 Six were of dose-adjusted warfarin versus placebo. The conclusions drawn from the original four studies were upheld. Warfarin reduced stroke risk by 62 percent overall. Absolute risk reductions were higher for secondary prevention (8.4% per year) than primary prevention (2.7%). These percentages translate into the numbers needed to treat (NNT) of 12 and 37, respectively. Although more intracranial hemorrhages (ICHs) occurred in the warfarin group (0.3% per year) compared to those on placebo (0.1%), this was not statistically significant. Major extracranial hemorrhage occurred in 0.6 percent per year of patients on placebo, with a relative risk of those on warfarin of 2.4 (absolute risk increase, 0.3% per year). The total number of patients in the six trials was 2,900, with an average follow-up of 1.7 years. The aforementioned risk reduction with warfarin was based on intention-to-treat analysis; the on-treatment analysis reveals more than 80 percent relative risk reduction in stroke.
This meta-analysis also evaluated adjusted-dose warfarin compared to aspirin. There were five trials, all unblinded, totaling 2,837 individuals. Excluding one study because the range of the INR was wide (2.0 to 4.5), the relative risk reduction for warfarin compared to aspirin was 46 percent.
The issue of aspirin as an alternative to warfarin has also been addressed in several trials. 51 Aspirin dose ranged from 25 to 1,200 mg daily. More than 3,000 patients were studied, with an average follow-up of 1.5 years. In patients receiving placebo, the stroke rate was 5.2 percent per year for primary prevention and 12.9 percent for secondary prevention. Aspirin reduced stroke risk by 22 percent, resulting in numbers needed to treat of 67 and 40, respectively. The trials showed only a trend toward reduced stroke in aspirin-treated patients. All-cause mortality was not reduced. The authors suggested that the benefit of aspirin is to prevent nondisabling stroke that is not of cardioembolic origin. Therefore, published guidelines strongly recommend warfarin rather than aspirin for stroke prevention in individuals with AF who are at high risk. 52
In practice, despite the clear benefit of warfarin in stroke prevention in patients with AF, this therapeutic intervention is frequently underused. Many studies from different countries have demonstrated suboptimal rates of appropriate antithrombotic therapy for patients with AF. 53 - 55 Several potential reasons account for underuse of warfarin, including physician factors, patient factors, and geographic practice variations. Warfarin is a difficult medication for patients because of the inconvenience of INR monitoring, drug and food interactions, and bleeding risks. However, physicians frequently overestimate the bleeding risks but underestimate the benefits of warfarin and overestimate the benefits of aspirin. 56 Although major adverse bleeding events associated with warfarin occur with a relatively low incidence, 57 they may profoundly bias physician prescribing behavior. 58 There is often a bias against prescribing warfarin to patients of advanced age, especially elderly women, despite the fact that safety in patients 80 years and older has been established. 59
Individual patient preferences, knowledge, and attitudes affect compliance with long-term anticoagulation therapy. Among AF patients taking warfarin in one study, about one half did not know that AF was a risk factor for stroke and could not state why they were taking warfarin; ethnic differences in knowledge about their diagnosis and treatment were also identified. 60 Methods to encourage compliance with appropriate antithrombotic prophylaxis include use of a patient decision aid. One such tool is available for download at and is highly recommended for use by primary care physicians and specialists who are counseling AF patients about the benefits and risks of warfarin compared to those of aspirin for stroke prevention. A home INR finger-stick device for self-monitoring may increase the duration patients spend in the therapeutic INR range. 61
Bleeding is the major concern with anticoagulant therapy. The average risk of major bleeding in the clinical trials was 1.3 percent per year with warfarin compared to 1 percent with aspirin or placebo. 62 The Stroke Prevention in Atrial Fibrillation study had a higher rate of major bleeding at 2.3 percent on warfarin and 1.1 percent per year on aspirin. 62 Rates of ICH were 0.9 percent per year and 0.3 percent per year, respectively. Age older than 75 years increased the risk of major hemorrhage to 4.2 percent per year (relative risk = 2.6) compared to 1.7 percent per year in the younger population. Of patients on warfarin, 16 were in the therapeutic range, 4 were below, and 13 were above at the time of their bleed. All had had therapeutic levels on their last routine prothrombin time ratios. Intensity of anticoagulation was a risk factor for bleeding only in those older than 75 years. The other identified risk factor was the use of more than three prescription drugs. 62 Interestingly, in another study, patients with cerebral ischemia of presumed arterial origin had a substantially higher risk of ICH than those anticoagulated for AF. Leukoariosis is a newly identified risk factor. 63
Analysis of a cohort of patients attending five anticoagulation clinics documented the cumulative risk of bleeding over an 8-year period. Serious bleeds occurred at a rate of 7.5 events per 100 patient-years. Points that emerged were that the incidence of bleeding and thromboembolic complications remained approximately constant, with a prothrombin time ratio of 1.3 to 2.0, but it increased sharply above or below those limits (i.e., thromboembolism was much more likely with a prothrombin time ratio of less than 1.3). No increase in bleeding complication was found related to any specific indication for therapy, including cerebrovascular disease. Older patients did not have a greater risk of bleeding. The highest risk of bleeding was seen during the first 3 months of therapy, and then it tended to plateau somewhat. Of particular note was the high risk of recurrence (32%) in patients who experienced one serious bleed. It was also noted that patients who had more than four dose adjustments per year bled 25 percent more often than those who had fewer adjustments. 64
With the exception of some patients with lone AF, all patients with AF (regardless of whether this is paroxysmal, persistent, or permanent) require some form of antithrombotic therapy unless contraindicated. It remains necessary to individualize management strategies for specific patients, taking into account compliance, risk of bleeding complications, and other medical conditions. Risk stratification is essential to determine the optimal treatment, i.e., warfarin or aspirin. Many different schemes have been devised for identifying patients with AF unassociated with valvular heart disease that are at high, moderate, or low risk of stroke. According to the 2006 American Heart Association guidelines, high risk factors are previous stroke, TIA, or systemic embolism, mitral stenosis, and prosthetic heart valves. 3 Moderate risk factors include age 75 years or older; hypertension; heart failure; left ventricular ejection fraction 35 percent or lower; and diabetes. Warfarin is recommended for patients with any high risk factor or more than one moderate risk factor. This means that all patients with a previous ischemic stroke or TIA are considered at high risk and require warfarin anticoagulation for secondary stroke prevention, unless contraindicated. Warfarin or aspirin (81 to 325 mg) is recommended for those with only one moderate risk factor. Aspirin alone (81 to 325 mg) is considered sufficient for patients without any of these risk factors.
For patients receiving warfarin, the target INR should be 2.5 (range 2.0 to 3.0). The INR should be monitored closely: usually weekly initially and then monthly once stable. A minimum INR of 2.0 is recommended for stroke prevention; stroke risk increases exponentially as the intensity of anticoagulation declines. 65
In addition to protecting against stroke, antithrombotics attenuate stroke severity: patients taking warfarin at the time of stroke have less-disabling strokes compared to individuals taking aspirin or no antithrom- botic therapy, and stroke severity is negatively correlated with INR at stroke onset. 66, 67 Table 5-4 gives a summary of the indications for warfarin in secondary stroke prevention for patients with selected cardiac conditions.
TABLE 5-4 Summary of Indications for Warfarin in Secondary Stroke Prevention for Patients With Selected Cardiac Conditions 52
Strong or Moderate Indication for Warfarin
Mechanical heart valve
Atrial fibrillation
Atrial flutter
Cardioversion in atrial fibrillation or flutter
Bioprosthetic heart valve
Rheumatic mitral valve disease
Acute myocardial infarction and left ventricular thrombus
Possible/Uncertain Indication for Warfarin
Dilated cardiomyopathy
Left ventricular dysfunction
Patent foramen ovale associated with atrial septal aneurysm
Mitral annular calcification associated with mitral regurgitation
Warfarin Usually Not Indicated
Isolated patent foramen ovale
Isolated mitral valve prolapse
Isolated mitral annular calcification
Isolated aortic valve disease
The only class I evidence in support of warfarin for stroke prevention exists for atrial fibrillation and mechanical heart valves. Treatment recommendations are expected to change over time as new evidence emerges; the reader is advised to consult published guidelines for more detailed information.
For patients with a mechanical heart valve, the INR should be maintained above 2.5, and for secondary stroke prevention, the target INR should be 3.0 (range 2.5 to 3.5) 52
Dual antiplatelet therapy (aspirin plus clopidogrel) was investigated in a randomized trial and found to be inferior to warfarin for stroke prevention in AF and associated with a higher rate of adverse bleeding events than warfarin. 68
If warfarin therapy needs to be interrupted for surgical procedures, temporary discontinuation for up to 1 week is usually considered reasonable for patients without mechanical heart valves. However, this practice can be associated with increased stroke risk. Heparin may be substituted in high-risk patients.
In addition to medical therapy for stroke prevention in AF, interventional techniques are being investigated. These include percutaneously implanted left atrial appendage occlusive devices and surgical resection of the left atrial appendage, given that 91 percent of thrombi are localized at that site. 69 Carotid artery endovascular devices to filter emboli are also under investigation.
Cardioversion of AF to sinus rhythm (either pharmacological or electrical) does not reduce the risk of stroke and therefore does not obviate the need for continued anticoagulation therapy for stroke prevention. 70, 71
AF occurring in the postoperative setting following cardiac surgery is fairly common and usually self-limited. Anticoagulation is reasonable if AF persists for more than 48 hours, but it may not need to be continued long-term if sinus rhythm is restored. Similarly, other conditions associated with transient AF (e.g., alcohol, thyrotoxicosis) usually do not need long-term antithrombotic prophylaxis. 3
In patients with atrial flutter, the risk of thromboembolism is thought to be less than that for AF but higher than for patients in sinus rhythm. These patients frequently go on to develop AF. For practical purposes, the antithrombotic treatment recommendations are similar to those for AF. 3

Cardioversion in Atrial Fibrillation or Flutter
Cardioversion (electrical or pharmacological) undertaken to convert AF back to sinus rhythm is associated with an increased risk of thromboembolism. Review of 22 series published over a 30-year period showed an overall risk of embolism of 1.5 percent. 72 Figures have changed little in recent years, with an incidence of 1.3 percent. 73 It appears that up to 3 weeks may be required for atrial mechanical activity to recover. 74 It is therefore recommended that warfarin (INR 2.0 to 3.0) be given for at least 3 weeks before elective cardioversion of patients who have been in AF for 2 days or more or when the duration of AF is unknown and that it be continued until normal sinus rhythm has been maintained for 4 weeks. 3
For patients requiring immediate cardioversion, intravenous heparin is recommended concurrently followed by warfarin for at least 4 weeks. 3 Alternatively, TEE prior to cardioversion can be performed; if no thrombus is detected, then cardioversion can occur as soon as the patient is anticoagulated and continue for at least 4 weeks. If a thrombus is detected on TEE, warfarin is recommended for at least 3 weeks before and may need to be continued for a longer duration afterward.
The recommendations for cardioversion in atrial flutter are the same as for AF. 3 Atrial flutter has been studied less extensively than AF, but embolism can occur in relation to cardioversion or during subsequent months. The total incidence of acute and chronic events was found to be 7 percent over a period of 26 ± 18 months in a consecutive series of 191 unselected patients undergoing cardioversion. 75 The same percentage was found in a smaller study of 86 patients who were followed for a longer period (mean, 4.5 years). Annual risk was estimated at 1.6 percent, one third of the rate for those with AF. 76 Prior transesophageal echocardiography is not an adequate predictor of those at risk. A total of 3 of 41 patients who had no left atrial clot developed ischemic neurological syndromes within 48 hours of elective cardioversion. 77 In another study, spontaneous echo contrast was a more common finding than atrial thrombosis (34% versus 11%). 78

Chronic Sinoatrial Disorder
As with atrioventricular block, chronic sinoatrial disease (sick sinus syndrome) presents usually with syncope and dizziness but differs in predisposing to systemic embolism. In a study comparing age- and sex-matched control subjects with atrioventricular heart block to those having chronic sinoatrial disorder, prevalence of systemic embolism was found in 16 percent of those with sick-sinus syndrome compared to 1.3 percent of those with atrioventricular block. 79 Other studies have disclosed similar figures; patients with the “brady-tachy” form of the disorder appear to be particularly at risk. 80, 81 Insertion of a pacemaker does not protect against embolic phenomena. In one series, 6 of 10 strokes developed after pacemaker insertion. Only one of these patients was anticoagulated at the time. 82
Concern was raised that, although ventricular pacing provides symptomatic relief, this modality may worsen the underlying disease process by increasing the rate at which AF, congestive heart failure, and thromboembolism occur. 83 Many studies relating to various pacemaker types have followed. A Cochrane review noted poor quality of reporting but concluded that physiological (primarily dual-chamber) pacing had a statistically significant benefit in preventing the development of AF compared to ventricular pacing. 84 A nonsignificant preference for stroke prevention was found. A large subsequent study, also comparing ventricular with dual-chamber pacing, concluded that clinical features were the key predictors of stroke. 85 In the same year, a Danish study showed single-chamber atrial pacing to be superior to dual-chamber pacing in the prevention of AF and thromboembolism. 86 Patients in the brady-tachy group were noted to be more at risk of developing AF and stroke. It was concluded that warfarin treatment should be considered for these patients.

This continues to be a rapidly changing field. A new definition and classification were proposed in 2006. 87 Cardiomyopathies are defined as “a heterogeneous group of diseases of the myocardium associated with mechanical and/or electrical dysfunction that usually ‘but not invariably’ exhibit inappropriate ventricular hypertrophy or dilatation and are due to a variety of causes that frequently are genetic.” 87 Specifically excluded are those diseases of the myocardium secondary to congenital or valvular heart disease, systemic hypertension, or atherosclerotic coronary disease. The cardiomyopathies are then divided into two major groups based on predominant organ involvement. The primary cardiomyopathies are those solely or predominantly confined to heart muscle. Genetic, mixed, and acquired forms are recognized. Both hypertrophic and dilated cardiomyopathies are considered primary diseases. Also now included are the ion channel disorders, in which there is a primary electrical disturbance without structural cardiac pathology. These are further considered in the section devoted to syncope. The list of secondary cardiomyopathies is extensive.
Neuromuscular or neurological causes listed are Friedreich’s ataxia, Duchenne or Becker muscular dystrophy, Emery–Dreifuss muscular dystrophy, neurofibromatosis, and tuberous sclerosis. Surprisingly, the mitochondrial cytopathies, quintessentially multisystem disorders, are listed as primary cardiomyopathies. The secondary cardiomyopathy table classification does not include infective processes, such as Chagas’ disease or infection with human immunodeficiency virus, although these are briefly mentioned in the text.
In North America, the most common cardiomyopathy is hypertrophic cardiomyopathy, which is an autosomal-dominant disease affecting 1:500 of the general population. 88 The disorder is notorious as a major cause of sudden cardiac death in athletes but is compatible with survival until old age. 89 Mortality rates overall have been estimated at 1.0 to 1.5 percent for ages 16 to 65, 3.9 percent over the next decade, and 4.7 percent for ages older than 75 years. Risk was generally similar in Western and Asian populations. 90 Stroke risk in hypertrophic cardiomyopathy has been studied in a group of 900 patients. 91 Stroke occurred in 44 patients over a period of 7 ± 7 years. A small number (7) of other vascular events were noted. Age at first event ranged from 29 to 86 years, with a mean of 61 ± 14 years. Stroke was particularly associated with advanced age, congestive symptoms, and AF. The cumulative incidence of events was significantly higher in nonanticoagulated patients with AF compared to those receiving warfarin. Other studies confirm increased risk of stroke when AF develops in hypertrophic cardiomyopathy, but surprisingly also identified a subgroup in which the course was largely benign. 92 Outflow tract obstruction also increased the risk of stroke. 92, 93 The odds ratio for stroke in patients with AF was 17.7. 92
There are considerable geographic variations in the causes of cardiomyopathy. In Latin America, American trypanosomiasis (Chagas’ disease) is a major cause. Stroke has been increasingly well documented as a complication. A study of 94 consecutive stroke patients with the cardiomyopathy of Chagas’ disease compared these with 150 consecutive stroke patients without Chagas’ disease. 94 A cardioembolic basis for stroke was considered present in 56 percent of the former compared to 9 percent of the controls. Most strokes in the group with Chagas’ disease were in the anterior circulation (85%); the posterior circulation was rarely affected (5%) and less than 10 percent of the patients presented with lacunar syndromes.
In Chagasic cardiomyopathy, the apical region of the left ventricle is the typical site for formation of thrombosis or aneurysm. Echocardiography in this study revealed an apical aneurysm in 37 percent and mural thrombosis in 12 percent, but the most common finding was left ventricular diastolic dysfunction (49%). The ECG was abnormal in 67 percent. The most common abnormality was a right bundle branch block pattern (35%), followed by left His fascicular block (17%), and AF (15%). A pacemaker had been inserted in 10. Oral anticoagulation has been recommended for all individuals with Chagasic stroke who demonstrated risk factors for cardioembolism. 94
In Africa, the major cardiomyopathy is the dilated type, but peripartum cardiomyopathy is ubiquitous with an incidence ranging from 1:100 to 1:1,000. 95 There are regional variations: endomyocardial fibrosis is restricted to the tropical regions of East, Central, and West Africa. 95 The incidence of human immunodeficiency virus (HIV)–associated cardiac disease, including cardiomyopathy, is increasing in contrast to developing countries where the availability of highly active antiretroviral therapy has significantly reduced the incidence of myocarditis. 96
In Japan, hypertrophic cardiomyopathy is the most common cause of cardiomyopathy, followed closely by dilated cardiomyopathy (DCM). 97 Cardiomyopathy associated with the prolonged QT interval syndrome came in a distant third, followed by mitochondrial disease, arrhythmogenic right ventricular dysplasia, and Fabry’s disease of the heart. 97
In young adults arrhythmogenic right ventricular dysplasia/cardiomyopathy (ARVD/C) is another rare hereditary disorder causing sudden death. 98 In a natural history study of 130 patients, 100 male, age at onset of symptoms was 32 ± 14 years. The annual mortality rate was 2.3 percent; all patients who died had a history of ventricular tachycardia. 99 Diagnosis requires a high index of suspicion. 87
In the dilated cardiomyopathies, a necropsy study showed a high incidence of embolic events (systemic or pulmonary) at 60 percent of 152 cases. In the living, once a TIA or stroke has occurred, either warfarin or antiplatelet therapy should be considered. 52 There is insufficient evidence to recommend warfarin or antiplatelet therapy for primary prevention, in the absence of other indications. 100

Myocardial Infarction and Left Ventricular Dysfunction
Patients with a history of coronary artery disease have a threefold increase in stroke risk. 101 This risk is particularly high within the first month after myocardial infarction (MI). 102 Mechanisms include embolism from left ventricular mural thrombosis and the development of AF (which occurs in up to 20% of patients after MI). 103
A community-based study of 2,160 patients hospitalized between 1979 and 1998 found stroke risk during the 30 days after a first MI to be increased 44-fold, and it remained two to three times higher than expected during the subsequent 3 years. 102 Of note, the 20-year duration of the study enabled the conclusion to be drawn that acute MI treatment by thrombolysis did not reduce stroke risk. 102 Overall, stroke risk following MI is approximately 1 percent during the first month and about 2 percent at 1 year. 104, 105 For a non-ST elevation acute coronary syndrome, the early stroke risk was found to be 0.7 percent at 3 months. 106 In large randomized trials of aspirin versus the combination of aspirin and clopidogrel in patients with MI or acute coronary syndrome, the stroke rate ranged between 0.9 and 1.7 percent. 103
In a meta-analysis, predictors of stroke following MI included advanced age, diabetes, hypertension, previous stroke or MI, anterior MI, AF, heart failure, and nonwhite race. 104 Anterior wall MI has been a predictor of stroke in some, but not all, studies. 107 Left ventricular thrombus develops in about one third of individuals in the first 2 weeks following an anterior MI. 108 A meta-analysis of 11 studies concluded that mural thrombus formation after an MI poses a significantly increased risk of embolization, which is reduced by anticoagulation. 109
The current recommendation, in the absence of thrombolytic therapy, is that, after acute MI, heparin should be initiated and followed by warfarin for 3 months in patients considered to be at increased risk of embolism, either pulmonary or systemic. High-risk patients are those with severe left ventricular dysfunction, congestive heart failure, a history of pulmonary or systemic embolism, echocardiographic evidence of mural thrombosis, or the presence of AF. Because of the increased frequency of mural thrombosis in anterior as opposed to inferior myocardial infarcts, it is also recommended that patients with an anterior Q-wave infarction receive heparin followed by warfarin. 110
In patients with TIA/ischemic stroke related to an acute MI in which LV mural thrombus is identified, oral anticoagulation is recommended for at least 3 months and up to 1 year (INR 2 to 3) in addition to aspirin for coronary artery disease (up to 162 mg/day). 52
Stroke risk is inversely proportional to left ventricular ejection fraction (LVEF). In a study of 2,231 patients with LV dysfunction after an acute MI, those with LVEF less than 29 percent had a stroke risk nearly double that of patients with LVEF exceeding 35 percent: the annual stroke rate was 1.5 percent overall. Thus, reduced LVEF is an independent risk factor for subsequent stroke. 111 Another study found a 58 percent increase in risk of embolic events for every 10 percent decrease in LVEF in women, but not men. 112
Congestive heart failure carries a two- to threefold increase in the relative risk of stroke. Among patients enrolled into heart failure trials, the overall annual stroke risk has ranged between 1.3 to 3.5 percent; most patients were taking aspirin or warfarin. 113 In the absence of clinically overt heart failure or MI, the presence of asymptomatic left ventricular systolic dysfunction, even of mild degree, is an independent risk factor for stroke. 114
The optimal antithrombotic prophylaxis for patients with poor LV function remains uncertain; the efficacy of warfarin versus aspirin is the subject of ongoing trials. 115

Rheumatic Heart Disease
Extensive experience has accumulated over several decades concerning the association of systemic embolism with rheumatic heart disease. A 1973 review concisely summarized relevant features. 116 A minimum of 20 percent of patients with rheumatic heart disease experience a thromboembolic complication at some time, and 40 percent of these arterial emboli involve the brain. Embolic events are the cause of death in 16 to 35 percent of adults dying of rheumatic heart disease, and subgroups of patients having a much greater frequency of embolic complication can be identified.
The risk of embolism is substantially increased when atrial thrombus is present (risk increases from 16% to 41%) or AF develops (risk increases from 7% to 30%). The proportion of patients developing left atrial thrombus increases from 9 to 41 percent when AF is present; conversely, 80 percent of patients with atrial thrombus are in AF. Embolism is most likely to occur when the dominant valvular lesion is that of mitral stenosis, either alone or in combination with aortic valve disease or mitral insufficiency. Isolated aortic valve disease is rarely associated with embolic events. Older patients more frequently have AF, atrial thrombus, and embolic events.
Studies of atrial thrombosis initially involved TTE, an insensitive method. Of 293 patients in one study who were to undergo open heart surgery, TTE disclosed thrombi in the left atrium in 33. At surgery, this was confirmed in 30 of the cases, but the study had missed 21 additional patients, including all 11 in whom thrombus was located in the left atrial appendage. 117
Once embolization has occurred, recurrence rate is high, approaching 60 percent. 118 Current recommendations are therefore strongly in favor of the use of long-term warfarin (to prolong the INR to 2.0 to 3.0) in patients with rheumatic mitral valve disease who have a history of systemic embolism or who develop AF, either chronic or paroxysmal. It is also recommended that the same treatment be given to patients in normal sinus rhythm if the left atrial diameter is in excess of 5.5 cm. Furthermore, if recurrent systemic embolism occurs despite adequate warfarin therapy, addition of aspirin should be considered. 119 The beneficial effect of adding aspirin, 100 mg daily, to warfarin has been demonstrated in the context of prosthetic heart valves. 120

Atrial Myxoma
Atrial myxomas have long been recognized as a cause of cerebral embolism. They are uncommon. A French hospital reviewed experience with 112 cases collected over 40 years. 121 Women outnumbered men 72 to 40; ages ranged from 5 to 84 years. The presenting symptoms were cardiac, constitutional, and embolic in 67, 34, and 29 percent, respectively. Younger and male patients were more liable to have embolic events. Neurological manifestations, in 113 patients, were evaluated in a literature review. 122 Ischemic stroke was the most common at 83 percent, often at multiple sites. Syncope (28%), psychiatric presentations (23%), headache (15%), and seizures (12%) were all encountered. In a Spanish study of 28 patients, it was noted that in the 9 with stroke, TIA had preceded the stroke in 7. 123 Treatment is surgical. A rare delayed complication is that of distal multiple cerebral aneurysm formation. 124 Transient ischemic attacks led to this diagnosis 5 years after successful surgery in one person. Symptoms were controlled with clopidogrel.

Marantic (Nonbacterial Thrombotic) Endocarditis
Although there are several causes of nonbacterial thrombotic endocarditis, a review of 14 series, predating the era of echocardiography, found an underlying malignancy in half. The most common tumor was lung cancer. Cancers of gastrointestinal origin accounted for a similar number of cases. Breast cancer appeared underrepresented. The mitral valve was most commonly affected (43%), followed by the aortic valve (36%). Overall, embolism occurred in 42 percent of patients. 125 An autopsy series (171 cases) found cancers of the ovaries, biliary system, pancreas, stomach, and lung to be most common. 126
The widespread availability of echocardiography has facilitated recognition of vegetations. A prospective study of 200 unselected ambulatory patients with solid tumors found vegetations in 19 percent compared to 2 percent in controls. Vegetations were seen in 50 percent of pancreatic cancers, 28 percent of lung cancers, and 19 percent of lymphomas. Only two patients had cerebral events. 127 On MRI, numerous lesions of various sizes may be found in multiple arterial territories.
At one cancer center, 96 stroke patients were assessed. Echocardiography (TTE) was performed in 61; none had TEE. An embolic mechanism was thought to be causative in 52. The heart was implicated in 14, but nonbacterial thrombotic endocarditis in only 3. Stroke of embolic origin carried a dismal prognosis. Life expectancy was just over 2 months, and treatment had no apparent influence. 128

Other Echocardiographic Abnormalities Linked to Stroke

Patent Foramen Ovale and Atrial Septal Aneurysm
A PFO is present in about one quarter of adults and represents a potential mechanism for cardiogenic embolism. 129 Case-control studies of young adults (younger than 55 years) with cryptogenic stroke found a fivefold increase in prevalence of PFO compared to control subjects without stroke. 130
In a French prospective study of individuals with stroke and an isolated PFO, the 4-year stroke recurrence risk was 2.3 percent. For those with both PFO and ASA, the rate was 15.2 percent compared to 0 percent for those with ASA alone. In the “control group” with neither PFO nor ASA, the rate was 4.2 percent. All patients in this study were taking aspirin. 131 In another study, the presence of a PFO (with or without ASA) did not confer a significant increase in stroke recurrence rate over a 2-year follow-up; furthermore, recurrence rate did not differ between patients on aspirin or warfarin or in those with large or small PFO. 132
ASA was found in 2 percent of persons in a population-based study. 133 In elderly patients undergoing cardiac surgery, the incidence was nearly 5 percent. No patient had a cerebrovascular event over a follow-up period of 70 months; most were receiving aspirin. It was concluded that the risk of embolic stroke was low. 134
The optimal management of patients with PFO is not currently known. Treatment options include (1) antiplatelet therapy, (2) anticoagulation, (3) percutaneous device closure, and (4) surgical closure. Opinions differ between specialists: neurologists are more likely to recommend medical management, whereas cardiologists are more likely to suggest device closure. 135 Randomized trials are currently under way to compare the efficacy and safety of medical therapy with percutaneous closure. For patients with cryptogenic stroke and isolated PFO, antiplatelet therapy is usually recommended. 136 For patients with PFO and ASA, anticoagulation or device closure may be considered, although evidence to support these treatments is lacking.

Left Atrial Spontaneous Echo Contrast
Left atrial spontaneous echo contrast (smoke) may be detected by TEE and is thought to represent stasis of blood within the atrium. The finding may thus indicate a predisposition to thrombus formation. It is most commonly encountered in patients with either AF or mitral stenosis and has been found to be highly associated with previous stroke or peripheral embolism in this context. 137

Mitral Annular Calcification
Mitral annular calcification (MAC) has been suggested as a potential source of calcific or thrombotic emboli to the cerebral and retinal circulations, but the evidence has been conflicting on whether it is an independent risk factor for stroke. A Framingham study documented a doubled stroke risk in those with MAC compared to those without, but it was unclear whether this relationship is causal or a marker for other risk factors; for example, in MAC, the risk of developing AF is increased 12-fold. 138, 139 Although one study found MAC to be an independent predictor of stroke, 140 two others did not. 141, 142 One of these involved nearly 6,000 patients followed over 6 to 7 years. 142 MAC appears to be a marker of generalized atherosclerotic disease including carotid stenosis, calcified aortic plaque, and coronary disease. 143, 144

Mitral Valve Prolapse
Mitral valve prolapse (MVP) is the most frequent valve disease in adults, with a prevalence of about 2 percent. 145 Initially postulated as a cause of stroke/TIA in the young, 146 this has not been confirmed in more recent studies. 147, 148 Stroke risk is increased with older age and the development of cardiac conditions: AF, mitral valve thickening, left atrial enlargement, and mitral regurgitation. 149 In those with an auscultatory diagnosis of MVP alone, confirmed by echocardiography, no increase in risk was found. 150 In the Framingham cohort, no significant difference was found in the prevalence of stroke/TIA in those with or without MVP. 151
Treatment guidelines are therefore (1) no antithrombotic therapy for primary prevention in individuals with MVP who have not experienced embolic events 151 and (2) long-term antiplatelet therapy for secondary prevention in MVP patients who have had ischemic stroke or TIA. 52 If other cardiac abnormalities develop, these are treated according to their own merits.

Aortic Valve Sclerosis and Stenosis
Systemic embolism in patients with aortic valve disease is uncommon in the absence of AF or other risk factors. Aortic sclerosis (valve thickening without outflow obstruction) is a common finding in the elderly and is associated with generalized atherosclerotic vascular disease and increased cardiovascular mortality. 152
A prospective study of patients with echocardiographically documented aortic valve calcification showed no statistically significant difference in stroke risk in patients with calcification without stenosis (8%) compared to those with stenosis (5%) or control subjects (5%). 153 Additionally, aortic valve disease was not associated with the presence of silent brain infarcts in this study. A larger study compared stroke risk in those with stenosis to those with sclerosis. Over a mean follow-up of 5 years, stroke risk was 12 percent in those with stenosis and 8 percent in those with sclerosis compared to 6 percent in those with a normal aortic valve. After adjusting for other variables, there was no statistically significant increase in stroke risk in those with aortic sclerosis. 152 A similar conclusion was found in another cohort study. 153 With regard to aortic stenosis, only if severe was it an independent predictor of stroke in addition to age and AF. 154

Acute Medical Treatment of Cardiogenic Embolism
The landmark study comparing thrombolysis of acute ischemic stroke with intravenous tissue plasminogen activator (t-PA) against placebo showed improved clinical outcome at 3 months for all stroke subtypes. 155 Cardioembolism accounted for 28 percent of the patients. Therefore, this acute intervention should be considered for stroke of cardioembolic origin. The two fundamental eligibility criteria are (1) treatment initiated within 3 hours of stroke onset (therefore the time of stroke onset must be clearly defined) and (2) absence of hemorrhage on CT brain scan. Prompt assessment and treatment are required because the odds of a favorable outcome with t-PA decline rapidly the longer the interval to t-PA injection time.
A series of inclusion and exclusion criteria exist. 156 The purpose is to minimize the risk of intracerebral hemorrhage, the major complication of intravenous t-PA, and to avoid treating minor or rapidly resolving processes such as TIAs, or nonischemic events. The dose of t-PA for stroke thrombolysis (0.9 mg/kg) is lower than that for acute MI. The risk of intracerebral hemorrhage in the treated group (6.4%) was 10 times higher than that of the placebo group in one report. 155 This risk appears increased if the treatment window is extended beyond 3 hours. Patients treated with t-PA cannot receive heparin, warfarin, or aspirin for the first 24 hours after infusion. Subsequently, long-term anticoagulation for secondary stroke prevention must be considered.
Other interventional approaches to achieve recanalization include direct intraclot lysis via a microcatheter and mechanical clot disruption, but availability of such procedures is limited. Mechanical clot removal devices may especially have a role in the acute treatment of patients with severe stroke in whom thrombolysis is contraindicated (e.g., recent cardiac surgery). 157
Some embolic infarcts undergo secondary hemorrhagic transformation, which may lead to clinical deterioration. Factors found to increase this possibility in one study were large infarct size and initiation of early anticoagulation (less than 12 hours from presentation). 158
The optimal timing of initiation of anticoagulation after cardioembolic stroke is not known. One recommendation is that nonhypertensive patients without evidence of hemorrhage on CT scan performed 24 to 48 hours after stroke can start anticoagulation. Anticoagulation is usually delayed for about 7 days in those with large infarcts. The American Stroke Association states that initiation of warfarin is generally recommended within 2 weeks after a stroke, but longer delays may be appropriate in patients with large infarcts or uncontrolled hypertension. 52 Decisions must be individualized.

Transient self-limited interruptions of cardiac output result in generalized cerebral ischemia, a condition that is termed syncope when it results in a loss of consciousness. 159 Syncope is discussed in Chapter 8 but is considered further here with particular regard to its occurrence in patients with acquired cardiac disease and arrhythmias.
A study of syncope induced in 14 patients with pacemakers noted that consciousness was lost 9 or more seconds after induction of a ventricular arrhythmia (fibrillation or tachycardia). 160 Patients felt distant, dazed, or as if they were “fading out” before loss of consciousness. Motor activity was noted in 10 of 15 episodes, with generalized tonic contraction of axial muscles followed or accompanied by irregular jerking of the extremities, generalized rigidity without clonic activity, or irregular facial movement or eyelid flutter without tonic activity. None of the patients bit their tongue or was incontinent. During the recovery phase, tonic flexion of the trunk was seen in three patients. Patients remained dazed or confused for up to 30 seconds or more after restoration of the circulation. This study confirmed that motor phenomena occur in association with syncope without corresponding electroencephalographic (EEG) evidence of epileptic discharges. The authors noted variability in EEG findings and poor correlation of these changes with the clinical ones. 160
Videometric analysis of syncope lasting on average 12 seconds induced in 42 healthy volunteers showed that myoclonic activity occurred in 90 percent. Head turns, oral automatisms, and writhing movements were common. Upward eye deviation was also common, and eyes remained open in three quarters of the subjects. Visual hallucinations occurred in 60 percent and were associated with auditory hallucinations in 36 percent, although never with intelligible speech. 161
Focal neurological symptoms are rare with cardiac arrhythmia. Evaluation of 290 patients who required pacemaker insertion disclosed that only 4 had focal neurological symptoms or signs; among these, only 2 had focal symptoms that could be related to a specific episode of cardiac dysfunction. 162 Rarely, features suggestive of complex partial seizures are seen.
The clinical spectrum of abnormalities that occur with generalized cerebral hypoperfusion is thus an extended one, ranging from nonspecific “dizziness” through a variety of sensory disturbances, including paresthesias and alterations of vision to loss of consciousness, sometimes with convulsive features. This has long been recognized in the context of blood donation, where 12 percent of syncopal reactions were shown to have some convulsive features. 163 Confusion may occur upon recovery. These observations highlight potential difficulties in distinguishing syncope from seizure.
A collaborative study between cardiologists and neurologists identified historical criteria to identify seizure patients among those presenting with presumed syncope: waking with a lacerated tongue, loss of consciousness with emotional stress, head turning to one side during loss of consciousness, and postictal confusion or abnormal behavior. 164 Of note, syncopal events indistinguishable from seizures have been observed in the context of cardiac arrhythmias. Additionally, seizures may cause arrhythmias.
Syncope is common, especially in the elderly, who show a high recurrence rate. 165 Of the many causes, it is important to identify those of cardiac origin because mortality is significantly increased in this group of patients. A cardiac basis, in different studies, ranged from 1 to 8 percent for organic heart disease and 4 to 38 percent for arrhythmias. 159 In addition to common structural causes, aortic tract outflow stenosis or intermittent obstruction to outflow may occur, for example, by a mobile thrombus or tumor in the left atrium. Echocardiography is the test of choice. In the case of arrhythmias, the prime objective is to document a relevant abnormality during an episode.
In the past, no cause for syncope was found in about one third of patients, but diagnostic yields as high as 76 percent have been achieved, for example, in a Swiss study of 788 patients presenting to an emergency department. 166 Evaluations were completed in 650 of those patients. History and clinical examination led to a diagnosis in 38 percent. In 10 percent, a possible cause for syncope was identified, and in about 3 percent this was refuted. In 21 percent, the cause of syncope was not initially determined, and the majority of this group underwent an extensive work-up. A probable cause of syncope was found in only 30 of the 122 patients in this group. Among the 650 patients, 69 (11%) were considered to have a cardiac cause, and arrhythmias were most prominent (44 patients). A sinus bradycardia or pause was seen in 15, as was atrial ventricular block, whereas 4 showed a supraventricular tachycardia and 1 had a pacemaker malfunction. Acute coronary syndromes were found in 9, aortic stenosis in 8, and pulmonary embolism in 8. The 18-month mortality in the cardiac group, noncardiac group, and group with unidentified cause was 26, 6, and 7 percent, respectively.
A relatively common disorder predisposing to paroxysmal supraventricular tachycardia is the Wolff–Parkinson–White syndrome, usually a sporadic disorder, with a prevalence of up to 1 in 1,000 persons. 167 AF may develop. Dizziness, syncope, and, rarely, sudden death may occur. The characteristic electrocardiographic (ECG) hallmarks are a short PR interval and a slowly rising prolonged QRS complex.
It is those patients with an apparently normal heart that present a special challenge and raise the possibility of disorders of the conducting tissues. 168 The long QT syndrome is seen throughout the world. A recessive form is associated with deafness, whereas the more common form, without deafness, has autosomal-dominant inheritance. Acquired forms, often drug related, are more common. Exertion or emotion may trigger events. The characteristic feature, as the name implies, is a prolonged QT interval (corrected for heart rate) on a standard ECG. The disorder predisposes to polymorphic ventricular tachycardia, which in turn predisposes to syncope and sudden death. 87
Recently, a short QT interval syndrome has been identified. 169 Especially affected are the young, including infants. It is rare and predisposes to paroxysmal AF and episodes of ventricular fibrillation, which may lead to syncope and sudden death. It has been suggested that the disorder may be responsible for some cases of sudden infant death. 170
Sudden death in males from Southeastern Asia attributable to ventricular fibrillation has been recognized for more than 20 years. Episodes indistinguishable from generalized seizures may occur in sleep, and the sudden death is presumed due to ventricular fibrillation. 171 It is now known as SUNDS (sudden unexplained nocturnal death syndrome) and has been linked to the Brugada syndrome, which is said to be phenotypically, genetically, and functionally the same. 172 However, the Brugada syndrome has been described in Europe and in females. The Brugada syndrome is also characterized by sudden death due to malignant arrhythmias. The baseline ECG may be abnormal in showing ST elevation in leads V1, 2, and 3, together with the presence of a right bundle branch block pattern. However, this pattern may be concealed and require unmasking by the use of sodium channel blockers.
Another disorder, but one in which the resting ECG is unremarkable, although it may show a sinus bradycardia and prominent U waves in some patients, is catecholamine-induced polymorphic ventricular tachycardia. 173 This is a disorder of childhood with an average age at symptom onset of 8 years. Syncope or events indistinguishable from seizures are triggered by exercise or emotional stress.
Evaluation of syncope therefore requires attention to family history, age at onset, unexplained sudden deaths, note of apparent epileptic disorders, relation of events to exertion and distress, and effects of postural change. In the presence of an apparently normal heart, evaluation of the standard ECG may suggest a cause, as indicated previously. In the context of a normal ECG or with intermittent events, prolonged recordings may be required in order to capture an episode. With daily events, a Holter monitor may suffice. More prolonged recording techniques are available, as are sophisticated electrophysiological studies. Details are beyond the scope of this chapter; a useful modern overview of an approach to the investigation of syncope is available. 174
Cardiac arrhythmias may also result from epileptic events. Tachycardia is the most commonly observed rhythm disturbance. 175 Sinus bradycardia, complete atrioventricular block, and cardiac arrest all have been documented as epileptic effects. 176 - 178 Asystole secondary to an epileptic event, often a partial seizure, has been documented to last for up to 60 seconds. 179 It should be noted that carbamazepine can cause heart block, especially in elderly women. 180


Coronary Catheterization
Coronary angiography carries a small (0.2%) risk of central nervous system (CNS) complications. An unexplained observation is the preponderance of embolic events within the posterior circulation, regardless of the route of catheterization. 181 The corresponding clinical features are visual disturbances that may be migrainous, transient, or persistent; confusion may also occur. 182, 183 For patients experiencing an iatrogenic ischemic stroke in the context of coronary catheterization, thrombolytic therapy is a potential treatment option that should be considered. 184

Percutaneous Transluminal Coronary Angioplasty and Stenting
Percutaneous transluminal coronary angioplasty had an overall mortality of 0.1 percent in a large series of more than 12,000 patients. Of the 121 who died, low-output failure was the most common cause (66% of deaths); stroke was responsible for 4 percent. 185 Another study showed that in the presence of peripheral vascular disease the risk of any major complication (stroke included) was higher: 12 versus 8 percent. 186
Angioplasty has also been compared to coronary stenting. Stroke rate (0.2%) was equal in the two groups. 187 To prevent stent thrombosis, an antithrombotic regimen is required. The addition of clopidogrel to aspirin reduces stroke incidence both before and after percutaneous coronary intervention. 188 In a study of more than 18,000 patients with a non–ST-segment elevation acute coronary syndrome, the 6-month stroke risk was 1.3 percent (1.1% for those not undergoing coronary artery bypass graft surgery) and the 6-month mortality in these patients was 27 percent. 189 Independent predictors of stroke risk were coronary bypass surgery (especially when performed early), previous stroke, diabetes, and older age, among others. Percutaneous coronary intervention was not associated with an increased risk of stroke in this group.
In a study of 12,407 percutaneous coronary interventions (1990 to 1999), the periprocedural risk of stroke and TIA was 0.38 and 0.12 percent, respectively. 190 More than 90 percent of patients in this study underwent balloon angioplasty, and nearly half also underwent coronary stenting. Independent predictors of stroke were advanced age, use of an intra-aortic balloon pump, and need for saphenous vein graft intervention. 190
Primary angioplasty, compared to thrombolysis, has been noted to decrease significantly the risk of stroke, 191 but when used as a rescue therapy, stroke risk was marginally increased. 192 In a review of 23 randomized trials involving more than 7,000 patients with acute MI and ST-segment elevation who were randomly assigned to either primary percutaneous transluminal coronary angioplasty or thrombolysis, the overall stroke rate was 1 percent with angioplasty compared to 2 percent with thrombolysis (a statistically significant reduction in favor of angioplasty). 193

Thrombolytic Therapy for Acute Myocardial Infarction
Concern that the introduction of thrombolytic therapy for MI would result in an increase in stroke was not substantiated by the results of the initial large Italian trial of nearly 12,000 patients. Stroke rate was 0.77 percent in the streptokinase group compared to 0.92 percent in the control group. An excess of stroke was evident only during the first day after randomization to streptokinase. After this time, patients in the control group had more stroke or TIA events. The study did not include CT scan results. 194 Extended experience from this group, specifically stressing stroke risk, found that stroke occurred in 236 (1.14%) of 20,768 patients. Autopsy or CT scanning enabled the cause of stroke to be identified in 74 percent. Perhaps surprisingly, infarction was more common (42%) than hemorrhage (31%). Patients receiving recombinant t-PA showed a small but significant excess of stroke. 195 Comparison of four thrombolytic strategies confirmed a slight excess of hemorrhagic stroke in those receiving t-PA and in those receiving combined thrombolytic agents. This excess risk was on the order of 2 to 3 per 1,000 treated. 196 In the four groups, stroke risk rate ranged from a low of 1.22 percent in those treated with streptokinase and subcutaneous heparin to 1.64 percent in those treated with intravenous heparin and both t-PA and streptokinase. These percentages are equal to or less than those documented in recent large prethrombolytic studies of acute MI.
The risk of ICH following thrombolytic therapy has been linked to the intensity of heparin anticoagulation and timing of partial thromboplastin time (PTT) monitoring. Recent trials that have used reduced-dose heparin regimens and 3-hour PTT monitoring have reduced ICH rates. 197
When ICH does occur, it is likely to be large in size, supratentorial in site, and more often lobar than deep. Mass effect is common, and blood may extend into the ventricles or subarachnoid space. Of the 244 cases in the study referred to earlier, symptoms emerged within 8 hours of treatment in 55, after 30 hours in 58, and between these times in the remainder. 194 A small percentage (3%) of hemorrhages were subdural. Syncope within 48 hours of treatment, or facial or head trauma within 2 weeks of treatment were disproportionately noted, but numbers were small (7). 198
Review of risk factors in 150 patients with documented ICH identified four factors as independent predictors: age older than 65 years, body weight less than 70 kg, hypertension on hospital admission, and administration of alteplase. The same risk factors for ICH were identified in the GUSTO-I trial; additional predictors included a history of cerebrovascular disease or hypertension, and elevated systolic and diastolic blood pressure on admission. 199
If ICH is suspected, immediate brain CT scan and discontinuation or reversal of thrombolytic or antithrombotic therapy are recommended. Neurosurgical consultation should be considered. 200


1 Goldstein LB, Jones MR, Matchar DB, et al. Improving the reliability of stroke subgroup classification using the Trial of ORG 10172 in Acute Stroke Treatment (TOAST) criteria. Stroke . 2001;32:1091.
2 Cardiogenic brain embolism. The second report of the Cerebral Embolism Task Force. Arch Neurol . 1989;46:727.
3 Fuster V, Ryden LE, Cannom DS, et al. ACC/AHA/ESC 2006 Guidelines for the Management of Patients with Atrial Fibrillation: 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 Guidelines (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation): developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Circulation . 2006;114:e257.
4 Adams HPJr, Kappelle LJ, Biller J, et al. Ischemic stroke in young adults. Experience in 329 patients enrolled in the Iowa Registry of stroke in young adults. Arch Neurol . 1995;52:491.
5 Kristensen B, Malm J, Carlberg B, et al. Epidemiology and etiology of ischemic stroke in young adults aged 18 to 44 years in northern Sweden. Stroke . 1997;28:1702.
6 Ghandehari K, Moud ZI. Incidence and etiology of ischemic stroke in Persian young adults. Acta Neurol Scand . 2006;113:121.
7 Ducrocq X, Lacour JC, Debouverie M, et al. Cerebral ischemic accidents in young subjects. A prospective study of 296 patients aged 16 to 45 years. Rev Neurol (Paris) . 1999;155:575.
8 Rasura M, Spalloni A, Ferrari M, et al. A case series of young stroke in Rome. Eur J Neurol . 2006;13:146.
9 Lee TH, Hsu WC, Chen CJ, et al. Etiologic study of young ischemic stroke in Taiwan. Stroke . 2002;33:1950.
10 Kwon SU, Kim JS, Lee JH, et al. Ischemic stroke in Korean young adults. Acta Neurol Scand . 2000;101:19.
11 Giovannoni G, Fritz VU. Transient ischemic attacks in younger and older patients. A comparative study of 798 patients in South Africa. Stroke . 1993;24:947.
12 Kittner SJ, Sharkness CM, Sloan MA, et al. Features on initial computed tomography scan of infarcts with a cardiac source of embolism in the NINDS Stroke Data Bank. Stroke . 1992;23:1748.
13 Kittner SJ, Sharkness CM, Sloan MA, et al. Infarcts with a cardiac source of embolism in the NINDS Stroke Data Bank: neurologic examination. Neurology . 1992;42:299.
14 Knepper LE, Biller J, Tranel D, et al. Etiology of stroke in patients with Wernicke’s aphasia. Stroke . 1989;20:1730.
15 Hanlon RE, Lux WE, Dromerick AW. Global aphasia without hemiparesis: language profiles and lesion distribution. J Neurol Neurosurg Psychiatry . 1999;66:365.
16 Timsit SG, Sacco RL, Mohr JP, et al. Early clinical differentiation of cerebral infarction from severe atherosclerotic stenosis and cardioembolism. Stroke . 1992;23:486.
17 Arboix A, Oliveres M, Massons J, et al. Early differentiation of cardioembolic from atherothrombotic cerebral infarction: a multivariate analysis. Eur J Neurol . 1999;6:677.
18 Arboix A, Garcia-Eroles L, Massons JB, et al. Atrial fibrillation and stroke: clinical presentation of cardioembolic versus atherothrombotic infarction. Int J Cardiol . 2000;73:33.
19 Minematsu K, Yamaguchi T, Omae T. ‘Spectacular shrinking deficit’: rapid recovery from a major hemispheric syndrome by migration of an embolus. Neurology . 1992;42:157.
20 Bogousslavsky J, Van MG, Regli F. The Lausanne Stroke Registry: analysis of 1,000 consecutive patients with first stroke. Stroke . 1988;19:1083.
21 Kase CS, Norrving B, Levine SR, et al. Cerebellar infarction. Clinical and anatomic observations in 66 cases. Stroke . 1993;24:76.
22 Bogousslavsky J, Regli F, Maeder P, et al. The etiology of posterior circulation infarcts: a prospective study using magnetic resonance imaging and magnetic resonance angiography. Neurology . 1993;43:1528.
23 Yamamoto Y, Georgiadis AL, Chang HM, et al. Posterior cerebral artery territory infarcts in the New England Medical Center Posterior Circulation Registry. Arch Neurol . 1999;56:824.
24 Lovett JK, Coull AJ, Rothwell PM. Early risk of recurrence by subtype of ischemic stroke in population-based incidence studies. Neurology . 2004;62:569.
25 Kolominsky-Rabas PL, Weber M, Gefeller O, et al. Epidemiology of ischemic stroke subtypes according to TOAST criteria: incidence, recurrence, and long-term survival in ischemic stroke subtypes: a population-based study. Stroke . 2001;32:2735.
26 Ringelstein EB, Koschorke S, Holling A, et al. Computed tomographic patterns of proven embolic brain infarctions. Ann Neurol . 1989;26:759.
27 Weisberg LA. Computerized tomographic findings in cardiogenic cerebral embolism. Comput Radiol . 1985;9:189.
28 Okada Y, Yamaguchi T, Minematsu K, et al. Hemorrhagic transformation in cerebral embolism. Stroke . 1989;20:598.
29 Toni D, Fiorelli M, Bastianello S, et al. Hemorrhagic transformation of brain infarct: predictability in the first 5 hours from stroke onset and influence on clinical outcome. Neurology . 1996;46:341.
30 Gonzalez RG, Schaefer PW, Buonanno FS, et al. Diffusion-weighted MR imaging: diagnostic accuracy in patients imaged within 6 hours of stroke symptom onset. Radiology . 1999;210:155.
31 Hankey GJ, Warlow CP, Sellar RJ. Cerebral angiographic risk in mild cerebrovascular disease. Stroke . 1990;21:209.
32 Daniel WG, Erbel R, Kasper W, et al. Safety of transesophageal echocardiography. A multicenter survey of 10,419 examinations. Circulation . 1991;83:817.
33 Kapral MK, Silver FL. Preventive health care, 1999 update: 2. Echocardiography for the detection of a cardiac source of embolus in patients with stroke. Canadian Task Force on Preventive Health Care. CMAJ . 1999;161:989.
34 Thompson CR. Echocardiography in stroke: which probe when? CMAJ . 1999;161:981.
35 McNamara RL, Lima JA, Whelton PK, et al. Echocardiographic identification of cardiovascular sources of emboli to guide clinical management of stroke: a cost-effectiveness analysis. Ann Intern Med . 1997;127:775.
36 Belvis R, Leta RG, Marti-Fabregas J, et al. Almost perfect concordance between simultaneous transcranial Doppler and transesophageal echocardiography in the quantification of right-to-left shunts. J Neuroimaging . 2006;16:133.
37 Sloan MA, Alexandrov AV, Tegeler CH, et al. Assessment: transcranial Doppler ultrasonography: report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Neurology . 2004;62:1468.
38 Go AS, Hylek EM, Phillips KA, et al. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and stroke prevention: the AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA . 2001;285:2370.
39 Feinberg WM, Seeger JF, Carmody RF, et al. Epidemiologic features of asymptomatic cerebral infarction in patients with nonvalvular atrial fibrillation. Arch Intern Med . 1990;150:2340.
40 Ott A, Breteler MM, de Bruyne MC, et al. Atrial fibrillation and dementia in a population-based study. The Rotterdam Study. Stroke . 1997;28:316.
41 Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke . 1991;22:983.
42 Wolf PA, Dawber TR, Thomas HEJr, et al. Epidemiologic assessment of chronic atrial fibrillation and risk of stroke: the Framingham study. Neurology . 1978;28:973.
43 Kopecky SL, Gersh BJ, McGoon MD, et al. The natural history of lone atrial fibrillation. A population-based study over three decades. N Engl J Med . 1987;317:669.
44 Echocardiographic predictors of stroke in patients with atrial fibrillation: a prospective study of 1066 patients from 3 clinical trials. Arch Intern Med . 1998;158:1316.
45 Transesophageal echocardiographic correlates of thromboembolism in high-risk patients with nonvalvular atrial fibrillation. The Stroke Prevention in Atrial Fibrillation Investigators Committee on Echocardiography. Ann Intern Med . 1998;128:639.
46 Petersen P, Boysen G, Godtfredsen J, et al. Placebo-controlled, randomised trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation. The Copenhagen AFASAK study. Lancet . 1989;1:175.
47 Stroke Prevention in Atrial Fibrillation Study. Final results. Circulation . 1991;84:527.
48 The effect of low-dose warfarin on the risk of stroke in patients with nonrheumatic atrial fibrillation. The Boston Area Anticoagulation Trial for Atrial Fibrillation Investigators. N Engl J Med . 1990;323:1505.
49 Ezekowitz MD, Bridgers SL, James KE, et al. Warfarin in the prevention of stroke associated with nonrheumatic atrial fibrillation. Veterans Affairs Stroke Prevention in Nonrheumatic Atrial Fibrillation Investigators. N Engl J Med . 1992;327:1406.
50 Connolly SJ, Laupacis A, Gent M, et al. Canadian Atrial Fibrillation Anticoagulation (CAFA) Study. J Am Coll Cardiol . 1991;18:349.
51 Hart RG, Benavente O, McBride R, et al. Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: a meta-analysis. Ann Intern Med . 1999;131:492.
52 Sacco RL, Adams R, Albers G, et al. Guidelines for prevention of stroke in patients with ischemic stroke or transient ischemic attack: a statement for healthcare professionals from the American Heart Association/American Stroke Association Council on Stroke: co-sponsored by the Council on Cardiovascular Radiology and Intervention: the American Academy of Neurology affirms the value of this guideline. Stroke . 2006;37:577.
53 Paciaroni M, Agnelli G, Caso V, et al. Atrial fibrillation in patients with first-ever stroke: frequency, antithrombotic treatment before the event and effect on clinical outcome. J Thromb Haemost . 2005;3:1218.
54 Indredavik B, Rohweder G, Lydersen S. Frequency and effect of optimal anticoagulation before onset of ischaemic stroke in patients with known atrial fibrillation. J Intern Med . 2005;258:133.
55 Waldo AL, Becker RC, Tapson VF, et al. Hospitalized patients with atrial fibrillation and a high risk of stroke are not being provided with adequate anticoagulation. J Am Coll Cardiol . 2005;46:1729.
56 Bungard TJ, Ghali WA, McAlister FA, et al. Physicians’ perceptions of the benefits and risks of warfarin for patients with nonvalvular atrial fibrillation. CMAJ . 2001;165:301.
57 Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation. Analysis of pooled data from five randomized controlled trials. Arch Intern Med . 1994;154:1449.
58 Choudhry NK, Anderson GM, Laupacis A, et al. Impact of adverse events on prescribing warfarin in patients with atrial fibrillation: matched pair analysis. BMJ . 2006;332:141.
59 Kagansky N, Knobler H, Rimon E, et al. Safety of anticoagulation therapy in well-informed older patients. Arch Intern Med . 2004;164:2044.
60 Lip GY, Kamath S, Jafri M, et al. Ethnic differences in patient perceptions of atrial fibrillation and anticoagulation therapy: the West Birmingham Atrial Fibrillation Project. Stroke . 2002;33:238.
61 Matchar DB, Jacobson AK, Edson RG, et al. The impact of patient self-testing of prothrombin time for managing anticoagulation: rationale and design of VA Cooperative Study #481—the Home INR Study (THINRS). J Thromb Thrombolysis . 2005;19:163.
62 Bleeding during antithrombotic therapy in patients with atrial fibrillation. The Stroke Prevention in Atrial Fibrillation Investigators. Arch Intern Med . 1996;156:409.
63 Gorter JW. Major bleeding during anticoagulation after cerebral ischemia: patterns and risk factors. Stroke Prevention In Reversible Ischemia Trial (SPIRIT). European Atrial Fibrillation Trial (EAFT) study groups. Neurology . 1999;53:1319.
64 Fihn SD, McDonell M, Martin D, et al. Risk factors for complications of chronic anticoagulation. A multicenter study. Warfarin Optimized Outpatient Follow-up Study Group. Ann Intern Med . 1993;118:511.
65 Hylek EM, Skates SJ, Sheehan MA, et al. An analysis of the lowest effective intensity of prophylactic anticoagulation for patients with nonrheumatic atrial fibrillation. N Engl J Med . 1996;335:540.
66 O’Donnell M, Oczkowski W, Fang J, et al. Preadmission antithrombotic treatment and stroke severity in patients with atrial fibrillation and acute ischaemic stroke: an observational study. Lancet Neurol . 2006;5:749.
67 Hylek EM, Go AS, Chang Y, et al. Effect of intensity of oral anticoagulation on stroke severity and mortality in atrial fibrillation. N Engl J Med . 2003;349:1019.
68 Connolly S, Pogue J, Hart R, et al. Clopidogrel plus aspirin versus oral anticoagulation for atrial fibrillation in the Atrial fibrillation Clopidogrel Trial with Irbesartan for prevention of Vascular Events (ACTIVE W): a randomised controlled trial. Lancet . 2006;367:1903.
69 Nageh T, Meier B. Intracardiac devices for stroke prevention. Prev Cardiol . 2006;9:42.
70 Cordina J, Mead G. Pharmacological cardioversion for atrial fibrillation and flutter. Cochrane Database Syst Rev . 2005. CD003713
71 Mead GE, Elder AT, Flapan AD, et al. Electrical cardioversion for atrial fibrillation and flutter. Cochrane Database Syst Rev . 2005. CD002903
72 Stein B, Halperin JL, Fuster V. Should patients with atrial fibrillation be anticoagulated prior to and chronically following cardioversion? Cardiovasc Clin . 1990;21:231.
73 Arnold AZ, Mick MJ, Mazurek RP, et al. Role of prophylactic anticoagulation for direct current cardioversion in patients with atrial fibrillation or atrial flutter. J Am Coll Cardiol . 1992;19:851.
74 Manning WJ, Leeman DE, Gotch PJ, et al. Pulsed Doppler evaluation of atrial mechanical function after electrical cardioversion of atrial fibrillation. J Am Coll Cardiol . 1989;13:617.
75 Seidl K, Hauer B, Schwick NG, et al. Risk of thromboembolic events in patients with atrial flutter. Am J Cardiol . 1998;82:580.
76 Wood KA, Eisenberg SJ, Kalman JM, et al. Risk of thromboembolism in chronic atrial flutter. Am J Cardiol . 1997;79:1043.
77 Mehta D, Baruch L. Thromboembolism following cardioversion of “common” atrial flutter. Risk factors and limitations of transesophageal echocardiography. Chest . 1996;110:1001.
78 Irani WN, Grayburn PA, Afridi I. Prevalence of thrombus, spontaneous echo contrast, and atrial stunning in patients undergoing cardioversion of atrial flutter. A prospective study using transesophageal echocardiography. Circulation . 1997;95:962.
79 Fairfax AJ, Lambert CD, Leatham A. Systemic embolism in chronic sinoatrial disorder. N Engl J Med . 1976;295:190.
80 Bathen J, Sparr S, Rokseth R. Embolism in sinoatrial disease. Acta Med Scand . 1978;203:7.
81 Simonsen E, Nielsen JS, Nielsen BL. Sinus node dysfunction in 128 patients. A retrospective study with follow-up. Acta Med Scand . 1980;208:343.
82 Fisher M, Kase CS, Stelle B, et al. Ischemic stroke after cardiac pacemaker implantation in sick sinus syndrome. Stroke . 1988;19:712.
83 Camm AJ, Katritsis D. Ventricular pacing for sick sinus syndrome—a risky business? Pacing Clin Electrophysiol . 1990;13:695.
84 Dretzke J, Toff WD, Lip GY, et al. Dual chamber versus single chamber ventricular pacemakers for sick sinus syndrome and atrioventricular block. Cochrane Database Syst Rev . 2004. CD003710
85 Greenspon AJ, Hart RG, Dawson D, et al. Predictors of stroke in patients paced for sick sinus syndrome. J Am Coll Cardiol . 2004;43:1617.
86 Kristensen L, Nielsen JC, Mortensen PT, et al. Incidence of atrial fibrillation and thromboembolism in a randomised trial of atrial versus dual chamber pacing in 177 patients with sick sinus syndrome. Heart . 2004;90:661.
87 Maron BJ, Towbin JA, Thiene G, et al. Contemporary definitions and classification of the cardiomyopathies: an American Heart Association Scientific Statement from the Council on Clinical Cardiology, Heart Failure and Transplantation Committee; Quality of Care and Outcomes Research and Functional Genomics and Translational Biology Interdisciplinary Working Groups; and Council on Epidemiology and Prevention. Circulation . 2006;113:1807.
88 Maron BJ. Hypertrophic cardiomyopathy: a systematic review. JAMA . 2002;287:1308.
89 Maron BJ. Sudden death in young athletes. N Engl J Med . 2003;349:1064.
90 Pokorski RJ. Hypertrophic cardiomyopathy: risk factors for life and living benefits insurance. J Insur Med . 2002;34:43.
91 Maron BJ, Olivotto I, Bellone P, et al. Clinical profile of stroke in 900 patients with hypertrophic cardiomyopathy. J Am Coll Cardiol . 2002;39:301.
92 Olivotto I, Cecchi F, Casey SA, et al. Impact of atrial fibrillation on the clinical course of hypertrophic cardiomyopathy. Circulation . 2001;104:2517.
93 Maron MS, Olivotto I, Betocchi S, et al. Effect of left ventricular outflow tract obstruction on clinical outcome in hypertrophic cardiomyopathy. N Engl J Med . 2003;348:295.
94 Carod-Artal FJ, Vargas AP, Horan TA, et al. Chagasic cardiomyopathy is independently associated with ischemic stroke in Chagas disease. Stroke . 2005;36:965.
95 Sliwa K, Damasceno A, Mayosi BM. Epidemiology and etiology of cardiomyopathy in Africa. Circulation . 2005;112:3577.
96 Barbaro G. HIV-associated cardiomyopathy etiopathogenesis and clinical aspects. Herz . 2005;30:486.
97 Matsumori A, Furukawa Y, Hasegawa K, et al. Epidemiologic and clinical characteristics of cardiomyopathies in Japan: results from nationwide surveys. Circ J . 2002;66:323.
98 Corrado D, Basso C, Thiene G, et al. Spectrum of clinicopathologic manifestations of arrhythmogenic right ventricular cardiomyopathy/dysplasia: a multicenter study. J Am Coll Cardiol . 1997;30:1512.
99 Hulot JS, Jouven X, Empana JP, et al. Natural history and risk stratification of arrhythmogenic right ventricular dysplasia/cardiomyopathy. Circulation . 2004;110:1879.
100 Sirajuddin RA, Miller AB, Geraci SA. Anticoagulation in patients with dilated cardiomyopathy and sinus rhythm: a critical literature review. J Card Fail . 2002;8:48.
101 Kannel WB, Wolf PA, Verter J. Manifestations of coronary disease predisposing to stroke. The Framingham study. JAMA . 1983;250:2942.
102 Witt BJ, Brown RDJr, Jacobsen SJ, et al. A community-based study of stroke incidence after myocardial infarction. Ann Intern Med . 2005;143:785.
103 Dutta M, Hanna E, Das P, et al. Incidence and prevention of ischemic stroke following myocardial infarction: review of current literature. Cerebrovasc Dis . 2006;22:331.
104 Witt BJ, Ballman KV, Brown RDJr, et al. The incidence of stroke after myocardial infarction: a meta-analysis. Am J Med . 2006;119:354.
105 Fibrinolytic Therapy Trialists’ (FTT) Collaborative Group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomised trials of more than 1000 patients. Lancet . 1994;343:31.
106 Kassem-Moussa H, Mahaffey KW, Graffagnino C, et al. Incidence and characteristics of stroke during 90-day follow-up in patients stabilized after an acute coronary syndrome. Am Heart J . 2004;148:439.
107 Bodenheimer MM, Sauer D, Shareef B, et al. Relation between myocardial infarct location and stroke. J Am Coll Cardiol . 1994;24:61.
108 Fuster V, Halperin JL. Left ventricular thrombi and cerebral embolism. N Engl J Med . 1989;320:392.
109 Vaitkus PT, Barnathan ES. Embolic potential, prevention and management of mural thrombus complicating anterior myocardial infarction: a meta-analysis. J Am Coll Cardiol . 1993;22:1004.
110 Cairns JA, Theroux P, Lewis HDJr, et al. Antithrombotic agents in coronary artery disease. Chest . 1998;114:611S.
111 Loh E, Sutton MS, Wun CC, et al. Ventricular dysfunction and the risk of stroke after myocardial infarction. N Engl J Med . 1997;336:251.
112 Dries DL, Rosenberg YD, Waclawiw MA, et al. Ejection fraction and risk of thromboembolic events in patients with systolic dysfunction and sinus rhythm: evidence for gender differences in the studies of left ventricular dysfunction trials. J Am Coll Cardiol . 1997;29:1074.
113 Pullicino PM, Halperin JL, Thompson JL. Stroke in patients with heart failure and reduced left ventricular ejection fraction. Neurology . 2000;54:288.
114 Hays AG, Sacco RL, Rundek T, et al. Left ventricular systolic dysfunction and the risk of ischemic stroke in a multiethnic population. Stroke . 2006;37:1715.
115 Pullicino P, Thompson JL, Barton B, et al. Warfarin versus aspirin in patients with reduced cardiac ejection fraction (WARCEF): rationale, objectives, and design. J Card Fail . 2006;12:39.
116 Abernathy WS, Willis PWIII. Thromboembolic complications of rheumatic heart disease. Cardiovasc Clin . 1973;5:131.
117 Shrestha NK, Moreno FL, Narciso F, et al. Two-dimensional echocardiographic diagnosis of left-atrial thrombus in rheumatic heart disease. A clinicopathologic study. Circulation . 1983;67:341.
118 Carter AB. Prognosis of cerebral embolism. Lancet . 1965;2:514.
119 Levine HJ, Pauker SG, Salzman EW, et al. Antithrombotic therapy in valvular heart disease. Chest . 1992;102:434S.
120 Turpie AG, Gent M, Laupacis A, et al. A comparison of aspirin with placebo in patients treated with warfarin after heart-valve replacement. N Engl J Med . 1993;329:524.
121 Pinede L, Duhaut P, Loire R. Clinical presentation of left atrial cardiac myxoma. A series of 112 consecutive cases. Medicine (Baltimore) . 2001;80:159.
122 Ekinci EI, Donnan GA. Neurological manifestations of cardiac myxoma: a review of the literature and report of cases. Intern Med J . 2004;34:243.
123 Alvarez-Sabin J, Lozano M, Sastre-Garriga J, et al. Transient ischaemic attack: a common initial manifestation of cardiac myxomas. Eur Neurol . 2001;45:165.
124 Jean WC, Walski-Easton SM, Nussbaum ES. Multiple intracranial aneurysms as delayed complications of an atrial myxoma: case report. Neurosurgery . 2001;49:200.
125 Lopez JA, Ross RS, Fishbein MC, et al. Nonbacterial thrombotic endocarditis: a review. Am Heart J . 1987;133:773.
126 Steiner I. Nonbacterial thrombotic endocarditis—a study of 171 case reports. Cesk Patol . 1993;29:58.
127 Edoute Y, Haim N, Rinkevich D, et al. Cardiac valvular vegetations in cancer patients: a prospective echocardiographic study of 200 patients. Am J Med . 1997;102:252.
128 Cestari DM, Weine DM, Panageas KS, et al. Stroke in patients with cancer: incidence and etiology. Neurology . 2004;62:2025.
129 Kizer JR, Devereux RB. Clinical practice. Patent foramen ovale in young adults with unexplained stroke. N Engl J Med . 2005;353:2361.
130 Overell JR, Bone I, Lees KR. Interatrial septal abnormalities and stroke: a meta-analysis of case-control studies. Neurology . 2000;55:1172.
131 Mas JL, Arquizan C, Lamy C, et al. Recurrent cerebrovascular events associated with patent foramen ovale, atrial septal aneurysm, or both. N Engl J Med . 2001;345:1740.
132 Homma S, Sacco RL, Di Tullio MR, et al. Effect of medical treatment in stroke patients with patent foramen ovale: Patent Foramen Ovale in Cryptogenic Stroke Study. Circulation . 2002;105:2625.
133 Meissner I, Whisnant JP, Khandheria BK, et al. Prevalence of potential risk factors for stroke assessed by transesophageal echocardiography and carotid ultrasonography: the SPARC study. Stroke Prevention: Assessment of Risk in a Community. Mayo Clin Proc . 1999;74:862.
134 Burger AJ, Sherman HB, Charlamb MJ. Low incidence of embolic strokes with atrial septal aneurysms: a prospective, long-term study. Am Heart J . 2000;139:149.
135 Messe SR, Cucchiara B, Luciano J, et al. PFO management: neurologists vs cardiologists. Neurology . 2005;65:172.
136 Messe SR, Silverman IE, Kizer JR, et al. Practice parameter: recurrent stroke with patent foramen ovale and atrial septal aneurysm: report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology . 2004;62:1042.
137 Chimowitz MI, DeGeorgia MA, Poole RM, et al. Left atrial spontaneous echo contrast is highly associated with previous stroke in patients with atrial fibrillation or mitral stenosis. Stroke . 1993;24:1015.
138 Benjamin EJ, Plehn JF, D’Agostino RB, et al. Mitral annular calcification and the risk of stroke in an elderly cohort. N Engl J Med . 1992;327:374.
139 Savage DD, Garrison RJ, Castelli WP, et al. Prevalence of submitral (anular) calcium and its correlates in a general population-based sample (the Framingham Study). Am J Cardiol . 1983;51:1375.
140 Kizer JR, Wiebers DO, Whisnant JP, et al. Mitral annular calcification, aortic valve sclerosis, and incident stroke in adults free of clinical cardiovascular disease: the Strong Heart Study. Stroke . 2005;36:2533.
141 Boon A, Lodder J, Cheriex E, et al. Mitral annulus calcification is not an independent risk factor for stroke: a cohort study of 657 patients. J Neurol . 1997;244:535.
142 Gardin JM, McClelland R, Kitzman D, et al. M-mode echocardiographic predictors of six- to seven-year incidence of coronary heart disease, stroke, congestive heart failure, and mortality in an elderly cohort (the Cardiovascular Health Study). Am J Cardiol . 2001;87:1051.
143 Allison MA, Cheung P, Criqui MH, et al. Mitral and aortic annular calcification are highly associated with systemic calcified atherosclerosis. Circulation . 2006;113:861.
144 Tunca A, Karanfil A, Koktener A, et al. Association between mitral annular calcification and stroke. Jpn Heart J . 2004;45:999.
145 Freed LA, Levy D, Levine RA, et al. Prevalence and clinical outcome of mitral-valve prolapse. N Engl J Med . 1999;341:1.
146 Barnett HJ, Boughner DR, Taylor DW, et al. Further evidence relating mitral-valve prolapse to cerebral ischemic events. N Engl J Med . 1980;302:139.
147 Gilon D, Buonanno FS, Joffe MM, et al. Lack of evidence of an association between mitral-valve prolapse and stroke in young patients. N Engl J Med . 1999;341:8.
148 Petty GW, Orencia AJ, Khandheria BK, et al. A population-based study of stroke in the setting of mitral valve prolapse: risk factors and infarct subtype classification. Mayo Clin Proc . 1994;69:632.
149 Avierinos JF, Brown RD, Foley DA, et al. Cerebral ischemic events after diagnosis of mitral valve prolapse: a community-based study of incidence and predictive factors. Stroke . 2003;34:1339.
150 Orencia AJ, Petty GW, Khandheria BK, et al. Risk of stroke with mitral valve prolapse in population-based cohort study. Stroke . 1995;26:7.
151 Salem DN, Stein PD, Al-Ahmad A, et al. Antithrombotic therapy in valvular heart disease–native and prosthetic: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest . 2004;126:457S.
152 Otto CM, Lind BK, Kitzman DW, et al. Association of aortic-valve sclerosis with cardiovascular mortality and morbidity in the elderly. N Engl J Med . 1999;341:142.
153 Boon A, Lodder J, Cheriex E, et al. Risk of stroke in a cohort of 815 patients with calcification of the aortic valve with or without stenosis. Stroke . 1996;27:847.
154 Petty GW, Khandheria BK, Whisnant JP, et al. Predictors of cerebrovascular events and death among patients with valvular heart disease: a population-based study. Stroke . 2000;31:2628.
155 The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med . 1995;333:1581.
156 Adams HP, del Zoppo G, Alberts MJ, et al. Guidelines for the early management of adults with ischemic stroke. A guideline from the American Heart Association/American Stroke Association Stroke Council, Clinical Cardiology Council, Cardiovascular Radiology and Intervention Council, and the Atherosclerotic Peripheral Vascular Disease and Quality of Care Outcomes in Research Interdisciplinary Working Groups. Stroke . 2007;38:1655.
157 Smith WS, Sung G, Starkman S, et al. Safety and efficacy of mechanical embolectomy in acute ischemic stroke: results of the MERCI trial. Stroke . 2005;36:1432.
158 Cardioembolic stroke, early anticoagulation, and brain hemorrhage. Cerebral Embolism Study Group. Arch Intern Med . 1987;147:636.
159 Kapoor WN. Syncope. N Engl J Med . 2000;343:1856.
160 Aminoff MJ, Scheinman MM, Griffin JC, et al. Electrocerebral accompaniments of syncope associated with malignant ventricular arrhythmias. Ann Intern Med . 1988;108:791.
161 Lempert T, Bauer M, Schmidt D. Syncope: a videometric analysis of 56 episodes of transient cerebral hypoxia. Ann Neurol . 1994;36:233.
162 Reed RL, Siekert RG, Merideth J. Rarity of transient focal cerebral ischemia in cardiac dysrhythmia. JAMA . 1973;223:893.
163 Lin JT, Ziegler DK, Lai CW, et al. Convulsive syncope in blood donors. Ann Neurol . 1982;11:525.
164 Sheldon R, Rose S, Ritchie D, et al. Historical criteria that distinguish syncope from seizures. J Am Coll Cardiol . 2002;40:142.
165 Savage DD, Corwin L, McGee DL, et al. Epidemiologic features of isolated syncope: the Framingham Study. Stroke . 1985;16:626.
166 Sarasin FP, Louis-Simonet M, Carballo D, et al. Prospective evaluation of patients with syncope: a population-based study. Am J Med . 2001;111:177.
167 Al-Khatib SM, Pritchett EL. Clinical features of Wolff-Parkinson-White syndrome. Am Heart J . 1999;138:403.
168 Goldschlager N, Epstein AE, Grubb BP, et al. Etiologic considerations in the patient with syncope and an apparently normal heart. Arch Intern Med . 2003;163:151.
169 Brugada R, Hong K, Cordeiro JM, et al. Short QT syndrome. CMAJ . 2005;173:1349.
170 Wolpert C, Schimpf R, Veltmann C, et al. Clinical characteristics and treatment of short QT syndrome. Expert Rev Cardiovasc Ther . 2005;3:611.
171 Otto CM, Tauxe RV, Cobb LA, et al. Ventricular fibrillation causes sudden death in Southeast Asian immigrants. Ann Intern Med . 1984;101:45.
172 Hong K, Berruezo-Sanchez A, Poungvarin N, et al. Phenotypic characterization of a large European family with Brugada syndrome displaying a sudden unexpected death syndrome mutation in SCN5A. J Cardiovasc Electrophysiol . 2004;15:64.
173 Leenhardt A, Lucet V, Denjoy I, et al. Catecholaminergic polymorphic ventricular tachycardia in children. A 7-year follow-up of 21 patients. Circulation . 1995;91:1512.
174 Strickberger SA, Benson DW, Biaggioni I, et al. AHA/ACCF scientific statement on the evaluation of syncope: from the American Heart Association Councils on Clinical Cardiology, Cardiovascular 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 collaboration with the Heart Rhythm Society: endorsed by the American Autonomic Society. Circulation . 2006;113:316.
175 Blumhardt LD, Smith PE, Owen L. Electrocardiographic accompaniments of temporal lobe epileptic seizures. Lancet . 1986;1:1051.
176 Constantin L, Martins JB, Fincham RW, et al. Bradycardia and syncope as manifestations of partial epilepsy. J Am Coll Cardiol . 1990;15:900.
177 Wilder-Smith E. Complete atrioventricular conduction block during complex partial seizure. J Neurol Neurosurg Psychiatry . 1992;55:734.
178 Liedholm LJ, Gudjonsson O. Cardiac arrest due to partial epileptic seizures. Neurology . 1992;42:824.
179 Rocamora R, Kurthen M, Lickfett L, et al. Cardiac asystole in epilepsy: clinical and neurophysiologic features. Epilepsia . 2003;44:179.
180 Takayanagi K, Hisauchi I, Watanabe J, et al. Carbamazepine-induced sinus node dysfunction and atrioventricular block in elderly women. Jpn Heart J . 1998;39:469.
181 Keilson GR, Schwartz WJ, Recht LD. The preponderance of posterior circulatory events is independent of the route of cardiac catheterization. Stroke . 1992;23:1358.
182 Vik-Mo H, Todnem K, Folling M, et al. Transient visual disturbance during cardiac catheterization with angiography. Cathet Cardiovasc Diagn . 1986;12:1.
183 Kosmorsky G, Hanson MR, Tomsak RL. Neuro-ophthalmologic complications of cardiac catheterization. Neurology . 1988;38:483.
184 Serry R, Tsimikas S, Imbesi SG, et al. Treatment of ischemic stroke complicating cardiac catheterization with systemic thrombolytic therapy. Catheter Cardiovasc Interv . 2005;66:364.
185 Malenka DJ, O’Rourke D, Miller MA, et al. Cause of in-hospital death in 12,232 consecutive patients undergoing percutaneous transluminal coronary angioplasty. The Northern New England Cardiovascular Disease Study Group. Am Heart J . 1999;137:632.
186 Weaver WD, Simes RJ, Betriu A, et al. Comparison of primary coronary angioplasty and intravenous thrombolytic therapy for acute myocardial infarction: a quantitative review. JAMA . 1997;278:2093.
187 Grines CL, Cox DA, Stone GW, et al. Coronary angioplasty with or without stent implantation for acute myocardial infarction. Stent Primary Angioplasty in Myocardial Infarction Study Group. N Engl J Med . 1999;341:1949.
188 Sabatine MS, Cannon CP, Gibson CM, et al. Effect of clopidogrel pretreatment before percutaneous coronary intervention in patients with ST-elevation myocardial infarction treated with fibrinolytics: the PCI-CLARITY study. JAMA . 2005;294:1224.
189 Cronin L, Mehta SR, Zhao F, et al. Stroke in relation to cardiac procedures in patients with non-ST-elevation acute coronary syndrome: a study involving >18000 patients. Circulation . 2001;104:269.
190 Fuchs S, Stabile E, Kinnaird TD, et al. Stroke complicating percutaneous coronary interventions: incidence, predictors, and prognostic implications. Circulation . 2002;106:86.
191 Cucherat M, Bonnefoy E, Tremeau G. Primary angioplasty versus intravenous thrombolysis for acute myocardial infarction. Cochrane Database Syst Rev . 2003. CD001560
192 Patel TN, Bavry AA, Kumbhani DJ, et al. A meta-analysis of randomized trials of rescue percutaneous coronary intervention after failed fibrinolysis. Am J Cardiol . 2006;97:1685.
193 Keeley EC, Boura JA, Grines CL. Primary angioplasty versus intravenous thrombolytic therapy for acute myocardial infarction: a quantitative review of 23 randomised trials. Lancet . 2003;361:13.
194 Maggioni AP, Franzosi MG, Farina ML, et al. Cerebrovascular events after myocardial infarction: analysis of the GISSI trial. Gruppo Italiano per lo Studio della Streptochinasi nell’Infarto Miocardico (GISSI). BMJ . 1991;302:1428.
195 Maggioni AP, Franzosi MG, Santoro E, et al. The risk of stroke in patients with acute myocardial infarction after thrombolytic and antithrombotic treatment. Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto Miocardico II (GISSI-2), and The International Study Group. N Engl J Med . 1992;327:1.
196 An international randomized trial comparing four thrombolytic strategies for acute myocardial infarction. The GUSTO investigators. N Engl J Med . 1993;329:673.
197 Giugliano RP, McCabe CH, Antman EM, et al. Lower-dose heparin with fibrinolysis is associated with lower rates of intracranial hemorrhage. Am Heart J . 2001;141:742.
198 Gebel JM, Sila CA, Sloan MA, et al. Thrombolysis-related intracranial hemorrhage: a radiographic analysis of 244 cases from the GUSTO-1 trial with clinical correlation. Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries. Stroke . 1998;29:563.
199 Gore JM, Granger CB, Simoons ML, et al. Stroke after thrombolysis. Mortality and functional outcomes in the GUSTO-I trial. Global Use of Strategies to Open Occluded Coronary Arteries. Circulation . 1995;92:2811.
200 Mahaffey KW, Granger CB, Sloan MA, et al. Neurosurgical evacuation of intracranial hemorrhage after thrombolytic therapy for acute myocardial infarction: experience from the GUSTO-I trial. Global Utilization of Streptokinase and tissue-plasminogen activator (tPA) for Occluded Coronary Arteries. Am Heart J . 1999;138:493.
Chapter 6 Neurological Manifestations of Infective Endocarditis

Linda S. Williams, Bradley L. Allen

Site of Infection
Infecting Organism
Acuteness of Infection
Valvular Vegetations
Hematological Risk Factors
Clinical Presentation
Evaluation of Patients
Treatment of Ischemic Stroke
Anticoagulation in Native Valve Endocarditis
Anticoagulation in Prosthetic Valve Endocarditis
Surgical Treatment
Clinical Presentation
Treatment of Hemorrhagic Stroke
Intraparenchymal Hemorrhage
Mycotic Aneurysms
Clinical Presentation
Treatment of Cerebral Infection

The relationship between infection of the heart valves and arterial embolization was first recognized by Rudolf Virchow in the mid-1800s 1 and the classic clinical triad of fever, heart murmur, and hemiplegia was described 30 years later by Osler in his Gulstonian Lectures of 1885. 2 The understanding of infective endocarditis has evolved since these early descriptions to a concept of the disease having different predisposing conditions, different propensity for sites of valve infection, different infecting organisms, and different treatments, but the proportion of patients with neurological manifestations has remained relatively constant. It is important to recognize any neurological complications not only because they are frequent but also because they may require alterations in treatment and are often associated with increased morbidity and mortality in infective endocarditis. Although the key to treating neurological complications is appropriate antibiotic therapy, the presence of neurological manifestations often alters concomitant medical or surgical treatment of infective endocarditis. This chapter reviews the most common neurological manifestations of infective endocarditis, detailing their epidemiology and clinical presentations, suggesting appropriate diagnostic evaluations, and discussing treatment options.

Neurological events have long been recognized as frequent and severe complications of infective endocarditis. In series of patients from the 1950s onward, the overall frequency of neurological complications has remained relatively constant at approximately 20 to 40 percent ( Table 6-1 ). 3 - 15 One reason for the similarity of these reports is that cerebral emboli are almost always symptomatic; the only study to date that has systematically performed cerebral and abdominal imaging in patients with infective endocarditis regardless of symptoms showed that the overall proportion of cases with cerebral embolization was 34 percent and that more than 90 percent of these cerebral emboli were symptomatic. 9 Nevertheless, because of the high overall incidence of stroke in the general population, infective endocarditis is an unusual cause of stroke. Neurological complications of infective endocarditis can be divided into three major types: ischemic stroke, hemorrhagic stroke, and cerebral infection ( Table 6-1 ). Ischemic stroke is by far the most common, occurring in 20 to 30 percent of patients and accounting for 50 to 75 percent of all neurological complications. Primary cerebral hemorrhage, usually intraparenchymal or subarachnoid, is less common, reported in 2 to 17 percent of patients. Secondary hemorrhagic transformation of an ischemic stroke, however, is not uncommon and is estimated to occur in 20 to 40 percent of ischemic strokes. Cerebral infections may manifest without previous clinical evidence of ischemic or hemorrhagic stroke in less than 10 percent of cases; typical infections include cerebritis, meningitis, and microabscesses or macroabscesses. Other neurological symptoms, including seizures, headache, mental status changes, and neuropsychological abnormalities, sometimes occur but are usually secondary to one of the three major complications. Rarely, endocarditis has been associated with spinal cord infarction or abscess, discitis, retinal ischemia, and ischemic cranial and peripheral neuropathies.

TABLE 6-1 Common Neurological Complications in Patients With Infective Endocarditis

Almost all the neurological complications of infective endocarditis have embolization as their primary cause. Although cerebral emboli are probably not more common than extracerebral emboli, 9 they are more often symptomatic and thus more frequently reported, and they are associated with an increased morbidity and mortality compared to other systemic emboli. Cerebral emboli most often affect the middle cerebral artery (MCA) territory and may be septic or nonseptic; either type can cause ischemic stroke. Septic emboli may also lead to hemorrhagic stroke through the development of arteritis or mycotic aneurysm, to cerebral microabscess or macroabscess, usually by seeding of ischemic tissue, and to cerebritis or meningitis by seeding of the meninges ( Fig. 6-1 ). Although the term bacterial intracranial aneurysm has been suggested as more appropriate, 16 the term mycotic aneurysm continues to be widely used and is therefore used here.

FIGURE 6-1 Embolization to various cerebral structures is responsible for most of the neurological complications of infective endocarditis. Emboli that lodge in the lumen of cerebral vessels may lead to ischemic stroke and can lead to arteritis or mycotic aneurysm formation with resultant vessel rupture and cerebral hemorrhage. Emboli to the meninges may produce meningitis, and emboli to the brain parenchyma, especially when associated with cerebral ischemia, may result in meningoencephalitis or abscess.
(Reprinted with permission from Solenski NJ, Haley EC Jr: Neurological complications of infective endocarditis. p. 331. In Roos KL [ed]: Central Nervous System Infectious Diseases and Therapy. Marcel Dekker, New York, 1997.)
Most primary intracerebral hemorrhages in infective endocarditis result from septic embolism, followed by septic necrosis and rupture of the vessel wall; less commonly, they result from rupture of mycotic aneurysms. 5, 17 - 19 Masuda and colleagues found that 10 of 16 patients with infective endocarditis and intracerebral bleeding had pyogenic arteritis, in 5 of whom rupture occurred without evidence of concomitant mycotic aneurysm; 13 of the 16 had either septic emboli or arteritis, or both. 18 Intracerebral hemorrhage may also occur owing to a secondary hemorrhage into an ischemic infarct. In one histopathological series of 17 patients, it was due to secondary transformation of ischemic infarction in 24 percent of cases, necrotic arteritis in 24 percent, mycotic aneurysm in 12 percent, and other causes in 11 percent; in 29 percent it was of unknown etiology. 17
Mycotic aneurysm formation has been related to (1) septic embolization to the arterial lumen, producing intraluminal wall necrosis and outward extension of infection, 20 and (2) septic embolization to the adventitial layer of the artery, resulting in destruction of the adventitia and muscularis layers and subsequent aneurysmal dilation. 21, 22 Mycotic aneurysms are usually small, located at distal arterial bifurcations, rather than on the circle of Willis, and can be single or multiple. Branches of the MCA are the most common location for mycotic aneurysms; in one series, almost 40 percent of all mycotic aneurysms involved distal MCA vessels. 23 Rarely, mycotic aneurysms involve extracranial vessels, including the internal carotid artery ( Fig. 6-2 ).

FIGURE 6-2 This patient with fungal endocarditis developed headache, confusion, and decreased level of consciousness without focal deficits. A, Head computed tomography (CT) showed subarachnoid hemorrhage (increased density) in the perimesencephalic cistern, left more than right, and dilatation of the temporal horns of the lateral ventricles. B, Digital subtraction angiography showed a large aneurysm of the cavernous portion of the left internal carotid artery. The aneurysm was treated with endovascular coils to occlude the carotid artery.
Brain macroabscesses account for less than 1 percent of all neurological complications of infective endocarditis and may occur secondary to ischemic infarction from a septic embolus or to extension of infection from adjacent arteritis or mycotic aneurysm. Brain microabscesses are more common than macroabscesses and usually occur in cases with multiple ischemic infarctions as a result of distal migration of septic embolic fragments. Microabscesses have been associated most commonly with Staphylococcus aureus infections. Meningoencephalitis is usually a result of embolization to meningeal vessels, with subsequent parenchymal or cerebrospinal fluid (CSF) invasion of the infecting organism. Aseptic meningitis may also occur with subarachnoid hemorrhage due to a necrotic arteritis or ruptured mycotic aneurysm.

A variety of clinical and laboratory variables have been associated with an increased risk of neurological complications ( Table 6-2 ), including site and type of valve infection, virulence of the infecting organism, acuteness of infection, presence of valvular vegetations, increased size and mobility of vegetations, and certain hematological factors.
TABLE 6-2 Suggested Risk Factors for Embolization in Infective Endocarditis Risk Factor Proposed Mechanism Mitral valve infection Increased valve mobility and left-sided position predispose to systemic embolization “Virulent” organism More rapid endothelial invasion leads to more friable, unstable valve surface Acuteness of infection More rapid endothelial invasion leads to more friable, unstable valve surface Valvular vegetations Increasing vegetation size and vegetation mobility may predispose to embolism Hematological factors Increased endothelial cell activity, platelet aggregability, and antiphospholipid antibodies may be associated with increased risk of embolization

Site of Infection
Neurological complications are more common with left-sided than with right-sided valve involvement, 11, 24, 25 although some series have found increased embolism in patients with right-sided infective endocarditis. 26, 27 Cerebral embolization in right-sided endocarditis may occur via embolization through a patent foramen ovale or a pulmonary arteriovenous fistula. 28, 29 Mitral valve infection has been associated most commonly with neurological complications; in one series, mitral valve infection was found in 76 percent of cases with neurological complications compared to 37 percent of other cases ( P < 0.005), 30 and this association has also been reported by others. 3, 10, 31 - 33 However, Wong and colleagues reported associations between aortic valve infection and stroke, with 44 percent of those with stroke having large aortic valve vegetations compared with a 9 percent prevalence in those without stroke. 34 Some authors have found no relationship between the site of infection and the occurrence of neurological complications. 4, 13, 26, 35 Although disagreement exists, 36, 37 most reports comparing native valve and prosthetic valve endocarditis indicate no significant difference in the proportion of patients with neurological complications. Among patients with prosthetic valve endocarditis, however, mechanical valves may be associated with complications more often than bioprosthetic valves. 38

Infecting Organism
Several important changes in the type and characteristics of the infecting organism in infective endocarditis have become evident in the past few years. Although streptococci, staphylococci, and enterococci remain the three most prevalent infecting organisms, some recent studies report that staphylococcal is now more common than viridans group streptococcal infective endocardi tis. 12, 39 More problematic than a shift in type of infecting organism is the growing prevalence of antibiotic resistance among these organisms, especially resistant viridans group streptococci and methacillin- and van-comycin-resistant S. aureus. 40, 41 This changing resistance pattern is reflected in updated guidelines from the American Heart Association on diagnosis, antimicrobial treatment, and management of complications in patients with infective endocarditis. 42
It is unclear whether antibiotic susceptibility changes have an impact on the risk of embolic complications, although an infection with a resistant organism that takes longer to control might well be associated with an increased risk of embolization. Previous studies have linked an increased risk of cerebral embolization to endocarditis due to S. aureus, 3, 4, 30, 33, 35, 43 enterococci, 3 Escherichia coli, 3 Streptococcus bovis, 44 various fungi, enterobacteriaceae, 3 and anaerobic bacteria. 3, 45 Several studies have shown that, even after adjusting for other factors, S. aureus 14, 15 and S. bovis 14 were independently associated with embolism. In prosthetic valve endocarditis, specifically, Staphylococcus epidermidis has been associated with more neurological complications than S. aureus. 46 Endocarditis due to Streptococcus pneumoniae has been associated with an increased risk of meningitis (50% to 90% of cases), 47, 48 and S. aureus endocarditis has been associated with brain abscess. 49 The current summation of these varied reports is that the virulence of the organism, the availability of effective antimicrobial therapy, and the potential development of large, friable vegetations all contribute to the propensity for embolization.

Acuteness of Infection
There is a higher risk of neurological complications with acute endocarditis than with subacute endocarditis. This probably relates to the typical etiological agents noted in acute disease ( S. aureus and beta-hemolytic streptococci), the potential for large vegetations or valve damage, and the subsequent increased risk of cerebral embolization. Many authors have observed that the risk of cerebral embolization is highest in the first 1 to 2 weeks of infection, with most patients either presenting with a neurological complication or experiencing an acute event in the first 48 hours after diagnosis. 3, 5, 10, 13, 14, 33, 50 Similarly, the risk of embolization decreases as the duration of effective antibiotic treatment increases, with most events occurring in the first 2 weeks of therapy. 14, 33, 50

Valvular Vegetations
Valvular vegetations are detected by two-dimensional echocardiography in 50 to 80 percent of patients with infective endocarditis and by transesophageal echocardiography (TEE) in more than 90 percent of cases. 32, 36, 51 - 53 Because of its increased sensitivity and ability to evaluate the more posteriorly located aortic valve, transesophageal echocardiography appears to be cost-effective as the initial study if clinical suspicion of infective endocarditis is high. 54, 55 Although some older clinical series revealed no significant difference in the development of neurological complications between patients with and without vegetations, 4, 8, 50, 56 - 58 most recent studies have linked either the presence of vegetations, increased vegetation size, or vegetation mobility to an increased risk of embolization. 13, 26, 27, 32, 59 - 63 The emergence of this relationship may be related to greater access and improved technical capabilities of echocardiography in the more recent series. A prospective study of 384 patients with infective endocarditis, all of whom had transesophageal echocardiography, found that vegetation length greater than 10 mm and vegetation mobility increased the risk of embolism and that vegetation length greater than 15 mm independently increased 1-year mortality. 14 The significance of changes in vegetations on serial echocardiography remains unclear: some investigators report that morphological changes in vegetation size or consistency are not associated with complications, 64 whereas others find that an increase in vegetation size during antibiotic treatment is associated with increased complications. 33, 65, 66 A final echocardiographic variable that may be related to complications is the presence of spontaneous echo contrast imaging. In a multivariate analysis, Rohmann and colleagues found that spontaneous echo contrast on transesophageal echocardiography was an independent predictor for embolization and hypothesized that this finding signified increased spontaneous platelet aggregation. 67 Current recommendations suggest that repeat echocardiography may be useful if clinical changes that suggest treatment failure occur during antibiotic therapy and that it should be performed urgently for unexplained progression of heart failure, new heart murmurs, or the development of atrioventricular block. 42

Hematological Risk Factors
In addition to spontaneous echo contrast, some reports also present evidence of an association between coagulation system activation and embolic events. 13, 68 - 71 In a series of 91 patients with infective endocarditis, antiphospholipid antibodies were present in 62 percent of patients with embolic events compared to 23 percent of those without such events ( P = 0.008) and were also positively correlated with other markers of endothelial cell activation, thrombin generation, and impaired fibrinolysis. 69 Antiphospholipid antibodies have also been reported to decrease after successful treatment of infective endocarditis. 70 Whether antiphospholipid antibodies independently increase the risk of embolism or this risk results from the association of these antibodies with increased numbers and size of vegetations remains to be determined. Similarly, soluble adhesion molecules have also been reported to independently increase the risk of embolism. 13, 71 At present, however, these hematological studies do not clearly aid in risk prediction for patients with infective endocarditis.

Ischemic stroke secondary to embolization of friable valvular material is the most common neurological complication of infective endocarditis. Most cerebral emboli are symptomatic, but they can be asymptomatic in as many as 35 percent of patients. 9 Ischemic stroke is the presenting symptom of infective endocarditis in up to 20 percent of cases 3, 30 and is most common in the acute stage of the infection, that is, before antibiotic treatment is begun or during the first several days of treatment (median time, 4 to 10 days). 4, 5, 10 Because of this clustering of symptoms in the acute phase, transient focal neurological symptoms in a febrile patient, especially in the presence of a regurgitant murmur, should always raise suspicion for infective endocarditis.

Clinical Presentation
In accordance with their embolic etiology, the majority of ischemic strokes involve the cortex, rather than being confined to subcortical brain tissue. One series found that 62 percent of strokes affected the cerebral or cerebellar cortex (with or without additional subcortical involvement), and only 16 percent were exclusively subcortical. 5 Brainstem strokes account for 10 percent or less of all strokes in infective endocarditis. 3
Because of their cortical involvement, ischemic strokes often present with aphasia, if the dominant hemisphere is involved, or visual or spatial neglect, if the nondominant hemisphere is affected. If the embolus lodges in the posterior cerebral artery, homonymous hemianopia can result. In addition to the more typical focal cerebral hemispheric or brainstem syndromes, multiple microemboli are clinically manifest in as many as 11 percent of cases 3 and in more than 50 percent of cases systematically evaluated with neuroradiological studies. 72 Patients with microemboli can present with nonlocalizing symptoms, including diminished level of consciousness, encephalopathy, or psychosis.
Clinical worsening of ischemic stroke may result from a variety of mechanisms, including development of cerebral edema, recurrent embolization and stroke, secondary hemorrhage into the ischemic area, and development of cerebral abscess. Cerebral edema may occur regardless of ischemic stroke mechanism, is more likely to be symptomatic in larger strokes and younger patients, and is typically maximal between 72 to 96 hours after stroke. Recurrent embolization should be suspected if new focal deficits develop; this complication is most likely to occur early in the course of treatment or if infection is uncontrolled. Hemorrhagic transformation of an ischemic stroke occurs in 18 to 42 percent of all patients with ischemic stroke and has been reported to be more common in cardioembolic strokes. 73 An autopsy series of patients with neurological complications of infective endocarditis found hemorrhagic transformation of an ischemic infarct in 9 of 16 patients. 18 Hemorrhagic transformation of an ischemic stroke is often asymptomatic, although development of intrainfarct hematoma is more likely to be symptomatic than is the development of petechial hemorrhage. 73 The term septic infarction has been used when, several days to weeks after an ischemic stroke, a cerebral abscess develops within the infarcted tissue 72 ( Fig. 6-3 ). The frequency with which this occurs is not known.

FIGURE 6-3 This patient presented with left hemiparesis and mitral valve endocarditis. A, Noncontrast head CT showed a focal low-density lesion in the right internal capsule and lentiform nucleus with a central area of hemorrhage (increased density) and cortical hemorrhage in the insula. B, With contrast, large confluent areas of enhancement representing leaky blood–brain barrier can be seen in the right caudate and lentiform nuclei, the insula, and the temporal cortex. C, Fluid-attenuated inversion recovery (FLAIR) magnetic resonance imaging (MRI). MRI 2 days after the head CT showed diffuse increased signal in the regions of CT enhancement and the right thalamus. D, After gadolinium, ring-like enhancement in the area of a previous infarct can be seen, representing possible secondary infection. This pattern is sometimes referred to as a “septic infarction.” This enhancement pattern resolved with antibiotic treatment and without development of a macroabscess.

Although seizures can occur in patients with infective endocarditis as a result of toxic or metabolic disturbances (e.g., hypoxia, antibiotic toxicity), most often seizures are secondary to ischemic or hemorrhagic stroke. The proportion of patients with seizure as the presenting symptom of infective endocarditis was 2 percent in one large series; 11 percent of patients had seizures during the course of their illness. 3 Seizures that are secondary to focal brain injury are usually focal in nature, with or without secondary generalization, whereas seizures due to metabolic or toxic factors are more often primarily generalized. The development of seizures during antibiotic treatment often signifies clinical worsening from either recurrent stroke, hemorrhagic transformation, or abscess formation. Thus, new onset of seizures should always prompt an urgent neuroimaging study. Rarely, seizures are secondary to antibiotic therapy, with imipenem the antibiotic having the greatest seizure proclivity.

Evaluation of Patients
All patients with acute focal neurological deficits should have a noncontrast head computed tomography (CT) scan or brain magnetic resonance imaging (MRI). Noncontrast CT allows for the most accurate distinction between hemorrhagic and ischemic events and can be done more quickly than MRI in most settings. If infective endocarditis is known or suspected, head CT with and without contrast may be useful; areas of increased contrast enhancement, representing possible cerebral abscess may then be distinguished from areas of ischemia ( Fig. 6-3 ). Although published radiological series are few, brain MRI appears to be more sensitive than CT in detecting the multiplicity of neurological lesions seen in infective endocarditis. In one series, multiple lesions were found in 10 of the 12 patients studied, with embolic branch infarction (8), multiple emboli and microabscesses (7), and hemorrhagic stroke (4) being the most common findings. 72 MRI findings have been categorized into four patterns: (1) embolic infarction, (2) multiple patchy infarctions (nonenhancing), (3) small nodular or ring-enhancing white matter lesions (probably microabscesses), and (4) hemorrhagic infarctions (intraparenchymal or subarachnoid). 74 Microabscesses usually develop several days after the ischemic stroke and can be asymptomatic or associated with clinical deterioration. 75 Multiple microabscesses are often responsible for nonfocal encephalopathy. MRI is superior to CT for symptoms referable to the brainstem or cerebellar regions.
Once cerebral embolism has occurred, serial neuroimaging studies or subsequent angiography can be performed to assess the presence of secondary complications such as microabscess or macroabscess formation, hemorrhagic transformation of ischemic stroke, or development of a mycotic aneurysm. Most authors agree that patients without neurological symptoms do not require cerebral angiography and that those with intracerebral hemorrhage do require angiography, 76 but whether to perform cerebral angiography after ischemic stroke in patients with infective endocarditis is especially controversial. The 2005 AHA statement on diagnosis and treatment of infective endocarditis suggests that diagnostic pursuit of mycotic aneurysms should be considered in patients with severe headache, erythrocytes or xanthochromia in CSF, or focal neurological signs. 42
Based on the evidence that subarachnoid hemorrhage can occur without previous symptoms in more than 50 percent of patients with mycotic aneurysm, some authors recommend that all patients with cerebral embolism have arterial imaging performed at some time beyond 48 hours after the initial event. 77 The basis for the timing of this recommendation is that mycotic aneurysm formation after septic embolization takes at least 48 hours to develop, and angiography immediately after embolization may therefore be negative. Although some studies suggest a more rapid angiographic evaluation based on early mycotic aneurysmal rupture within 24 hours of the onset of neurological symptoms, 78 others argue that a mycotic aneurysm develops in so few patients that angiographic complications present a greater risk. Using the published literature, van der Meulen and colleagues estimated the probability of 12-week survival in patients with infective endocarditis and ischemic stroke and found no added survival benefit for patients who had angiography, largely owing to the low prevalence of mycotic aneurysms and the low risk of their rupture in patients with adequate antibiotic therapy. 76 Since so few patients with infective endocarditis harbor mycotic aneurysms, the need to perform initial or serial angiography depends on the clinical presentation and proposed treatment. Patients with hemorrhagic stroke or hemorrhagic transformation of an ischemic stroke should have angiography to delineate mycotic aneurysm from arteritis because this distinction often influences subsequent evaluation and treatment. Patients with ischemic or hemorrhagic stroke who require long-term anticoagulation for mechanical valves or treatment of systemic thromboembolism, for example, may also benefit from angiography to exclude a mycotic aneurysm. Patients with ischemic stroke without hemorrhagic transformation or any indication for long-term anticoagulation probably do not benefit from repeated neuroimaging studies or conventional angiography.
The diagnosis of infective endocarditis depends on the documentation of an infecting organism on serial blood cultures and, in part, on the presence of valve abnormalities on echocardiography. 42 Echocardiography is also important in assessing valve function and excluding conditions such as valve thrombosis or abscess formation that would change clinical management. Transesophageal echocardiography is more sensitive to mitral and aortic valve pathology and has been reported to change patient management in as many as one third of cases. 79 Whether serial echocardiography provides data that reliably predict risk of subsequent thromboembolism or otherwise influence management is not known.
CSF examination is regarded by some authors as part of the standard evaluation of patients with infective endocarditis and neurological symptoms. The manner in which the CSF results will influence therapy, however, is not clear. The interpretation of CSF findings as a diagnostic tool for infective endocarditis in patients with acute stroke is complicated by the tendency for patients with cerebral embolism unrelated to endocarditis also to have mild to moderate increases in either white blood cells, red blood cells, or protein concentration in the CSF shortly after stroke. 80, 81 In one large series, CSF was abnormal in 48 of 69 patients with infective endocarditis in whom it was examined. 3 Of these, 28 percent had a purulent profile, 25 percent were aseptic, 13 percent were hemorrhagic, and 30 percent were normal. With the exception of purulent CSF in patients with meningismus, the type of neurological event in these patients did not correlate with the CSF pattern. For these reasons, CSF examination does not usually aid in the diagnosis or management of patients with neurological symptoms and infective endocarditis.

Treatment of Ischemic Stroke
The cornerstone of treatment of infective endocarditis is appropriate antibiotic therapy directed at the infecting organism. Numerous studies have shown that the risk of either initial or recurrent thromboembolism decreases sharply after the first few days of adequate antibiotic therapy. 4 - 7 ,33 ,51 Although this association may result in part from an ascertainment bias, it is critical to ensure that antibiotics are begun empirically, immediately after drawing initial blood for cultures (preferably three sets from separate sites) in febrile patients with stroke in whom infective endocarditis is among the differential diagnoses. Since effective long-term antimicrobial therapy will be required to treat infective endocarditis, the isolation and susceptibility testing of the pathogen are of critical importance. Involvement of an infectious diseases consultant is recommended. Thorough discussion of a current approach to diagnosis and antimicrobial treatment in various clinical scenarios can be found in the 2005 AHA guideline statement. 42
Recent studies have addressed the question of whether acute antiplatelet therapy is beneficial in reducing the risk of thromboembolism in infective endocarditis. In animal models of the disease, aspirin or aspirin plus ticlopidine has been found to reduce vegetation weight, echocardiographic evidence of vegetation growth, bacterial titer of vegetations, or systemic emboli. 82 - 84 Although one pilot study confirmed this finding, 85 a larger randomized controlled trial found no reduction of embolic events in patients treated with 325 mg aspirin compared to those given placebo, and there was a nonsignificant trend toward increased bleeding in the aspirin-treated group. 86 Based on this study, routine use of antiplatelet therapy for the purpose of decreasing embolic risk in patients with acute infective endocarditis is not recommended. 42
Anticoagulation in patients with infective endocarditis remains a controversial and complicated topic. Hemorrhagic complications are clearly more common in anticoagulated patients, with one retrospective study finding that 50 percent of the hemorrhages occurred in the 13 percent of subjects receiving anticoagulation. 87 However, patients with mechanical prosthetic valves may be receiving long-term anticoagulation, and the decision as to whether and for how long to withhold anticoagulants in this setting is especially difficult. Given the divergent management strategies required, it is useful to consider anticoagulation in native and prosthetic valve endocarditis separately.

Anticoagulation in Native Valve Endocarditis
Many authors have documented an increased risk of hemorrhagic complications in anticoagulated patients with native valve endocarditis and ischemic stroke, and the risk of recurrent embolism is low in patients receiving appropriate antibiotic therapy. Accordingly, there appears to be little benefit to anticoagulating patients with native valve endocarditis. Whether lower-level anticoagulation (e.g., for prevention of deep venous thrombosis) is safe in patients with stroke and infective endocarditis is unknown. Because other strategies, such as using sequential compression devices, have been shown to be equally efficacious, a conservative approach is to use these nonpharmacological methods of prevention of venous thrombosis.

Anticoagulation in Prosthetic Valve Endocarditis
Patients with bioprosthetic valves are typically not on long-term anticoagulation and have a lower risk of stroke in infective endocarditis than patients with mechanical valves 38, 46 ; thus, the same rationale applies to them as for patients with native valve endocarditis. Patients with mechanical prostheses who have endocarditis and stroke, however, present especially difficult management dilemmas. Most studies indicate that the proportions of patients with native and prosthetic valves having endocarditis and cerebral embolism are similar 4, 5 ; initiating anticoagulation in a previously nonanticoagulated patient with infective endocarditis and a mechanical valve thus appears unwarranted.
If a patient with a mechanical valve is receiving long-term anticoagulation and develops a cerebral embolus as a complication of infective endocarditis, the decision as to whether to continue anticoagulation or temporarily withhold it depends on several factors, including the size of the stroke and type of mechanical valve. Some authors have suggested that anticoagulation decreases the risk of cerebral embolism and should be instituted in all patients with newly diagnosed prosthetic valve endocarditis. 87, 88 Because larger strokes, especially those secondary to emboli, may be more likely to develop secondary hemorrhagic complications, 76 other authors favor withholding anticoagulation for several days in patients with acute cerebral embolism and mechanical valve endocarditis, especially when S. aureus is the infecting organism. 89, 90
Regardless of the timing of anticoagulation, it is safer to convert the patient from oral anticoagulation to the more controlled intravenous route of therapy during the acute phase of infective endocarditis. Some authors have not found a decrease in cerebral emboli in patients with acute prosthetic valve endocarditis anticoagulated with warfarin 46 or have documented a rate of hemorrhagic complications as high as 36 percent in this subgroup of patients, 88 thus leading to the position that anticoagulants should not be initiated and perhaps should be temporarily discontinued in previously anticoagulated patients with prosthetic valve endocarditis. 5, 91 If temporary discontinuation of anticoagulation is considered, determination of the patient’s type of mechanical valve and consultation with a cardiologist or a cardiothoracic surgeon concerning the risk of valve thrombosis with that valve type will help guide the decision about how long the patient can safely remain off anticoagulation. Although the use of anticoagulants remains controversial, converting to the most controllable (i.e., intravenous) form of therapy and frequent monitoring of anticoagulation parameters (activated partial thromboplastin time or international normalized ratio [INR]) are recommended. Solenski and Haley recommend that large cerebral infarctions, hemorrhage on CT scan, presence of mycotic aneurysm, uncontrolled infection or infection with S. aureus, history of bleeding diathesis, and possibly advanced patient age are factors arguing against the use of anticoagulation in patients with neurological complications of mechanical valve endocarditis. 92

Surgical Treatment
Valve replacement is not usually recommended as a therapy for preventing initial or recurrent stroke, although multiple emboli, infection with a “virulent” organism, and the presence of large vegetations may be relative indications for surgery. 26, 61 Typically, surgery is reserved for patients with acute or refractory congestive heart failure, perivalvular abscess, unstable valve prosthesis, continued embolism, infection with a pathogen resistant to effective antimicrobial agents, or inability to clear the infection. If surgery is required, the timing of the procedure in a patient with ischemic or hemorrhagic stroke is controversial. If surgery is contemplated to prevent embolization, early surgery is associated with greatest benefit since the risk of embolization is greatest in the first 2 weeks of the infection. If stroke has occurred, the first 72 to 120 hours after stroke are the period of maximal risk of cerebral edema and disruption of cerebral autoregulation; thus, most authors recommend delaying cardiac surgery for at least 1 week after stroke if possible. One retrospective assessment of 247 patients operated on for left-sided native valve endocarditis found that operation at approximately 3 weeks after the neurological deficit appeared was as safe for patients with previous neurological complications as for those without neurological manifestations of endocarditis. 93

Intracerebral hemorrhage in infective endocarditis may be primary or secondary to ischemic stroke or other pharmacological or hematological conditions ( Table 6-3 ; Fig. 6-4 ). Of the primary hemorrhages, intraparenchymal and subarachnoid hemorrhage are most common. Secondary transformation of an ischemic stroke is the most common form of intracerebral hemorrhage in infectious endocarditis, accounting for 24 to 56 percent of all hemorrhages in this condition. 17, 18 Intracerebral hemorrhage is a much less common complication than ischemic stroke, accounting for 2 to 17 percent of all neurological complications. In one recent series, only 8 cases of subarachnoid hemorrhage occurred among 489 patients with infective endocarditis; in 6 of these, no cause for the hemorrhage was identified by autopsy or angiography. 94 The prevalence of asymptomatic mycotic aneurysms in patients with infective endocarditis is not known, but seems to be small. 17
TABLE 6-3 Causes of Intracerebral Hemorrhage in Infective Endocarditis
Primary Intracerebral Hemorrhage
Arterial rupture secondary to arteritis
Rupture of a mycotic aneurysm
Secondary Intracerebral Hemorrhage
Hemorrhagic conversion of ischemic stroke
Hematological disorder
Disseminated intravascular coagulopathy
Vitamin K deficiency
Preexisting central nervous system lesion (e.g., aneurysm, arteriovenous malformation)

FIGURE 6-4 This patient had tricuspid valve endocarditis secondary to intravenous drug abuse. Initially, the patient had no neurological symptoms but left the hospital against medical advice after completing 6 days of antibiotic therapy. He returned 2 days later with a decreased level of consciousness and a right gaze preference. A toxicology screen was positive for cocaine. Noncontrast axial head CT at that time showed an approximately 3 × 4-cm hemorrhage in the right frontal lobe with intraventricular extension and subfalcial herniation. Cerebral angiography did not show a mycotic aneurysm. Echocardiography showed a large patent foramen ovale with right-to-left shunting and vegetations on the tricuspid valve. This case underscores several clinical points: (1) neurological complications of endocarditis are more common during uncontrolled infection; (2) neurologically asymptomatic patients may have silent cerebral emboli, particularly in the nondominant hemisphere; and (3) patients with right-sided endocarditis may develop cerebral embolization via a right-to-left shunt.
As described previously, in at least 40 percent of patients, septic embolization is the first event leading to intracerebral hemorrhage. 3, 17, 77 Depending on the location of the embolus, arteritis with secondary vessel rupture or development of a mycotic aneurysm may occur. Several series have documented that hemorrhagic complications are more common in anticoagulated patients, with one third of patients with endocarditis and subsequent intracerebral hemorrhage either anticoagulated or having an underlying bleeding diathesis. 17 In one series, 23 percent of all intracerebral hemorrhages occurred in the 3 percent of anticoagulated patients 3 ; in another, 50 percent of all such bleeds occurred in the 13 percent of patients who were anticoagulated. 87 These observations have led to the consensus to avoid anticoagulation in native valve endocarditis and to a judicious approach to its use in prosthetic valve endocarditis. Other conditions that sometimes accompany infective endocarditis may also predispose to bleeding complications, including disseminated intravascular coagulation, thrombocytopenia, and vitamin K deficiency.
Although mycotic aneurysms are most commonly found in the intracranial vessels, rarely these aneurysms may involve the extracranial carotid ( Fig. 6-1 ), thoracic, or abdominal vessels. 95 - 97 Management in these cases should be individualized but may include surgical or endovascular interventions or vessel ligation.

Clinical Presentation
Intracerebral hemorrhage usually presents with focal neurological symptoms as in ischemic stroke, but nonlocalizing symptoms, such as headache and decreased level of consciousness, may also predominate. Seizures may occur at the onset of the hemorrhage or later in its course. If subarachnoid hemorrhage occurs, either from rupture of an arteritic vessel or from a mycotic aneurysm, meningismus may be a prominent feature. Headaches may be more diffuse and subacute than is typical with saccular aneurysm rupture. 94 A transient ischemic attack (TIA) may precede intracerebral hemorrhage in as many as 25 percent of patients or may be the presenting symptom. 98

As in ischemic stroke, noncontrast head CT is the best initial neuroimaging procedure. The hematoma appears as an increased-density signal on CT ( Fig. 6-4 ) and can be localized to the intraparenchymal, subarachnoid, subdural, or intraventricular space. Hemorrhagic transformation of an ischemic infarct is most often patchy and may follow the contour of the gyri ( Fig. 6-3A ), but may appear as a homogeneous hematoma within an infarct. MRI is also useful and can better delineate stroke in the posterior fossa, although the signal change of blood products over time may make MRI more difficult to interpret in hemorrhagic stroke. A clue to the presence of an underlying mycotic aneurysm may be a focal area of cortical enhancement adjacent to an area of hemorrhage. 99
All patients with intracerebral hemorrhage complicating infective endocarditis should have imaging of the cerebral vasculature to visualize any underlying mycotic aneurysm. Since mycotic aneurysms tend to be small and to occur distally, rather than at the more proximal arterial branch-points as do saccular aneurysms, conventional cerebral angiography is preferred over magnetic resonance angiography (MRA) or CT angiography (CTA) for aneurysm detection. Although the resolution of these techniques continues to improve, at present they are adequate for screening in patients with infective endocarditis and ischemic stroke but should not be the primary diagnostic tool for evaluating patients with infective endocarditis and hemorrhagic stroke. They may be useful, however, for serial monitoring of aneurysm size following conventional angiography. One study reported the utility of monitoring mycotic aneurysms with serial thin-slice CT or MRI and found that all of six aneurysms identified with conventional angiography could be successfully followed for 6 to 8 weeks. 100 Repeat angiography at the end of antibiotic treatment confirmed the resolution (in 2) or persistence (3 enlarged, 1 unchanged) of the aneurysms.

Treatment of Hemorrhagic Stroke

Intraparenchymal Hemorrhage
The mainstay of treatment for either primary or secondary intracerebral hemorrhage in patients with infective endocarditis is the same as that for cerebral emboli: effective treatment of the underlying infectious organism. This is especially true for patients with pyogenic arteritis but is also critical for the treatment of mycotic aneurysms. Some patients with intracerebral hemorrhage and progressive neurological deterioration, either from expanding hematoma or edema, may benefit from surgical evacuation of the clot, but no firm guidelines exist for assisting with management in these cases. Similarly, although recombinant factor VIIa has been used successfully to reduce hematoma growth and improve outcomes in patients with intracerebral hemorrhage, 101 no data are available for its use in patients with infective endocarditis and cerebral hemorrhage. The increased risk of thrombosis and stroke associated with its use would be of concern in this population. As discussed previously, patients with mechanical valves and receiving anticoagulation therapy may have their anticoagulant discontinued temporarily or converted to an intravenous form. All patients should have close neurological monitoring in an intensive care setting because deterioration from recurrent hemorrhage or edema is not uncommon.

Mycotic Aneurysms
The natural history of mycotic aneurysms is that approximately one third resolve with 6 to 8 weeks of antibiotic treatment, one third remain unchanged in size, and the remaining one third are equally divided among those that increase and those that decrease in size. 17, 78, 100, 102, 103 Because of their propensity to resolve with antibiotic therapy, the evaluation and treatment of mycotic aneurysms are controversial. Aspects of care that remain unclear are whether serial angiography is needed in patients with mycotic aneurysms and the indications for surgical therapy.
Because more than one third of mycotic aneurysms either are unchanged in size or enlarge during antibiotic therapy, some authors recommend serial angiography every 2 weeks during antibiotic treatment. 77, 104, 105 If an aneurysm enlarges, surgical treatment to prevent rupture may be advocated. 104 - 106 Late hemorrhage from a ruptured mycotic aneurysm in patients who have completed adequate antibiotic therapy is rare, occurring in none of 122 patients with a mean 40-month follow-up, 77 but it has been reported. 17, 107 As discussed previously, the need for ongoing or subsequent long-term anticoagulation is another factor that may favor angiographic surveillance and surgical treatment, especially in patients with known cerebral embolization.
Once an aneurysm is discovered, controversy also exists about its treatment. Asymptomatic aneurysms are often treated medically, with surgical intervention reserved for those that either enlarge or do not resolve after antibiotic therapy is completed. 105, 106 Although symptomatic aneurysms may also resolve with antibiotic treatment and the risk of rebleeding is low, some authors favor surgical treatment of symptomatic mycotic aneurysms in addition to antibiotic therapy. 105, 106 This recommendation is usually based on the fear of recurrent bleeding, the associated increased morbidity and mortality, and the potential development of new aneurysms. Aneurysm accessibility and number are other features that influence the decision for surgical treatment; single aneurysms in a peripheral location are more likely to be treated surgically. Inaccessible aneurysms may be successfully treated endovascularly, although the management and outcomes in these cases are highly individualized. 108 - 111 Whether to undertake surgery at presentation or to wait until the completion of antibiotic therapy is debatable. For unruptured mycotic aneurysms, some authors have suggested serial angiography every 2 weeks during antibiotic therapy, 77, 104, 105 although the proportion of aneurysms that enlarge and thus may require urgent surgery is small. Since at least half of mycotic aneurysms persist after adequate antibiotic treatment and since new aneurysms can appear, it seems reasonable to repeat angiography, either conventional or MRA, at the conclusion of antibiotic therapy (usually 4 to 6 weeks) or to undertake serial imaging with a noninvasive procedure, such as MRA.
Accessible aneurysms that persist after adequate antibiotic therapy or that enlarge during therapy are usually treated surgically. Because mycotic aneurysms often lack a defined neck amenable to clipping, other surgical techniques, including wrapping, excision, or endovascular obliteration, may be necessary. Because mycotic aneurysms are often difficult to locate at the time of surgery, new techniques, including stereoscopic brain-surface imaging with MRA 112 and stereotactic angiographic localization, 113 are sometimes used to aid in aneurysm localization.

Cerebral infection, most commonly abscess or meningitis, has been reported as a primary complication in 6 to 31 percent of cases, although these cases typically represent less than 10 percent of the entire population of patients with endocarditis and neurological complications ( Table 6-1 ). These infections most typically occur after cerebral embolism; infection arising without clinical evidence of prior cerebral embolization is unusual. Encephalitis has also been reported, although the usual pathology in these cases is multiple emboli with microabscess formation.
Meningitis accounts for 4 to 7 percent of all neurological manifestations of infective endocarditis and is reported to be more common with either S. aureus or S. pneumoniae infections. 3, 114, 115 When meningitis is associated with involvement of the cerebral cortex, evidenced by gyral enhancement on MRI, the terms cerebritis and meningoencephalitis are used. Rarely, cerebritis can lead to the development of a parameningeal abscess in the cortex. Meningitis typically results from septic emboli to the meningeal vessels with subsequent CSF colonization. Less commonly, meningitis is nonseptic, resulting from sterile inflammation of the meninges due to blood products or circulating immune complexes in the CSF.
Cerebral abscesses are rare, accounting for approximately 2 percent of all neurological complications in infective endocarditis. 116 Small “microabscesses,” often defined as abscesses smaller than 1 cm 3 , are more common than “macroabscesses” but still account for less than 4 percent of all neurological complications. Cerebral abscess usually develops as the result of septic embolus and is not necessarily preceded by clinical symptoms. Radiographically, infarction-related abscesses are usually small and multiple and demonstrate areas of nodular or ringlike enhancement in an area of prior ischemic stroke 72, 74 ( Fig. 6-3D ). Abscess has also been reported as a consequence of mycotic aneurysm or septic arteritis. 117

Clinical Presentation
Although the clinical diagnosis of meningitis is infrequent in infective endocarditis, symptoms of meningitis are not. In one series, meningeal symptoms or signs occurred in more than 40 percent of 84 patients with endocarditis. 31 In addition to meningismus, headache, encephalopathy, cranial neuropathies, seizures, and increased intracranial pressure may occur. These symptoms may be subtle, especially in the elderly, and, when associated with fever, elevated white blood cell count, and regurgitant murmur should prompt an urgent evaluation for infective endocarditis.

All patients with known or suspected infective endocarditis and neurological symptoms, whether focal or nonfocal, should have imaging with noncontrast head CT prior to lumbar puncture. This is critical because multiple embolic strokes, intracerebral hemorrhage, and abscess may all present with nonfocal symptoms and can also cause significant compartmental increases in intracranial pressure, thus increasing the risk of cerebral herniation. Lumbar puncture should not be done in any patient with a focal lesion and evidence of mass effect on neuroimaging studies. Because patients with endocarditis have a propensity toward hematological abnormalities, coagulation tests, including a platelet count and INR, are especially important before one performs a spinal tap.

Treatment of Cerebral Infection
As for any type of meningitis, the goal of treatment is adequate antibiotic therapy to which the infecting organism is sensitive and that has good CSF penetration and activity in brain abscesses. Both microabscesses and macroabscesses usually respond to antibiotic treatment, although macroabscesses may occasionally produce significant mass effect and thus require stereotactic aspiration or surgical drainage.

Other extracerebral neurological complications may rarely occur. Although cerebral and systemic emboli appear to occur with similar frequency, 9 cerebral neurological complications predominate over extracerebral complications, probably because the brain receives more blood flow than peripheral neurological tissues and because cerebral complications are more likely to be symptomatic.
Mononeuropathy simplex or multiplex has been reported in as many as 1 percent of patients with infective endocarditis. 118 Both peripheral and cranial nerves may be involved, and viridans streptococci appear to be an especially prominent infectious organism in these cases. Discitis, occasionally with associated abscess or osteomyelitis, has also been reported and is more common with S. aureus infection. Other rare sites of embolization include the spinal cord and the retina.

The management of neurological complications of infective endocarditis is not standardized and substantial variations in care may be necessary based on the individual patient’s characteristics. Nonetheless, it is helpful to consider a treatment algorithm that includes pathways for the major neurological manifestations of the disease ( Fig. 6-5 ). This algorithm differs from some proposed previously in that cerebral angiography is not suggested for all patients with ischemic stroke, lumbar puncture is not recommended for all patients with neurological complications, and serial angiography every 2 weeks is optional. 92, 105 As many other authors have suggested, the two keys to managing patients, regardless of any neurological complications, are (1) a high level of suspicion for the possibility of infective endocarditis and (2) prompt initiation of appropriate antibiotic therapy after obtaining multiple sets of blood cultures.

FIGURE 6-5 Suggested management algorithm for patients with focal neurological deficits and known or suspected infective endocarditis. Factors favoring either surgical or medical treatment of mycotic aneurysms are presented; management of these cases is highly individualized. Repeat angiography at the conclusion of medical therapy is suggested for all patients with known mycotic aneurysms and may be considered either for patients with intracerebral hemorrhage and a negative initial angiogram or for patients with ischemic stroke who require long-term anticoagulation. LP, lumbar puncture, MRA, magnetic resonance angiography.

Among patients with infective endocarditis, mortality is increased in those with neurological complications compared to those without. 3, 4, 9, 10 Estimates of in-hospital mortality in various clinical series range from 16 to 58 percent compared with 14 to 20 percent for patients with and without neurological complications, respectively, although a population-based study from France reported 16 percent in-hospital mortality in 1999. 12 Mortality is higher in infections with more virulent organisms, with several large cohort studies showing an association between S. aureus and mortality. 14, 15, 39, 119 Intracerebral hemorrhage appears to confer added risk, as mortality in these patients is reported to be 40 to 90 percent. 17, 30, 119 Although rare, mycotic aneurysm rupture is associated with even higher mortality. 106 Mortality in patients with unruptured mycotic aneurysms appears no different from the aggregate mortality rate in all patients with neurological manifestations of endocarditis. 16 A multicenter, prospective study of 384 patients with infective endocarditis found that increasing age, female gender, serum creatinine greater than 2.0 mg/L, moderate to severe congestive heart failure, infection with S. aureus, increased medical comorbidity, and vegetation length greater than 15 mm were all independently associated with 1-year mortality. 14 Another study of factors related to in-hospital death also found an association with S. aureus infection and comorbidity as well as embolic events and diabetes. 119
The risk of recurrent neurological events, either embolic or hemorrhagic, is quite low. 3, 5, 6, 77 Recurrent ischemia has been documented in less than 0.5 percent of cases per day 5 and recurrent hemorrhage in less than 1 percent of all cases. 77 Elimination of recurrent events appears to depend more on effective antibiotic treatment than on other specific therapy. 33, 46, 50

Although infective endocarditis has evolved over the past several decades with regard to frequency of involvement of different valves and prevalence and susceptibility of infecting organisms, the proportion of patients with neurological manifestations and the type of neurological complications remain remarkably consistent. Most neurological complications are caused by embolization of friable valvular material resulting in either ischemic or hemorrhagic stroke. A high index of suspicion for infective endocarditis as the cause of stroke is critical because common treatments for acute stroke, such as thrombolysis or anticoagulation, are contraindicated in patients with native valve endocarditis and ischemic stroke. Management of patients with endocarditis and mechanical prosthetic valves is complicated, and decisions about continued anticoagulation in these patients must be individualized. Similarly, decisions about medical or medical plus surgical treatment of mycotic aneurysms must also be individualized because a number of clinical factors may influence treatment. Although many clinical decisions in patients with neurological manifestations of infective endocarditis must be individualized, it is clear that the cornerstone of prevention and treatment of all neurological complications is rapid delivery of appropriate antibiotic therapy.

Dr. Williams is supported by grants from the Department of Veterans Affairs and the National Institutes of Health.


1 Virchow R. Uber die akute Entzündung der Arterien. Virchows Arch . 1847;1:272.
2 Osler W. Malignant endocarditis (Gulstonian Lecture). Lancet . 1885;1:415.
3 Pruitt AA, Rubin R, Karchmer A. Neurologic complications of bacterial endocarditis. Medicine . 1978;57:329.
4 Salgado AV, Furlan AJ, Keys TF, et al. Neurologic complications of endocarditis: a 12-year experience. Neurology . 1989;39:173.
5 Hart RG, Foster JW, Luther MF, et al. Stroke in infective endocarditis. Stroke . 1990;21:695.
6 Paschalis C, Pugsley W, John R, et al. Rate of cerebral embolic events in relation to antibiotic and anticoagulant therapy in patients with bacterial endocarditis. Eur Neurol . 1990;30:87.
7 Matsushita K, Kuriyama Y, Sawada T, et al. Hemorrhagic and ischemic cerebrovascular complications of active infective endocarditis of native valve. Eur Neurol . 1993;33:267.
8 Weng MC, Chang FY, Young TG, et al. Analysis of 109 cases of infective endocarditis in a tertiary care hospital. Chung-Hua I Hsueh Tsa Chih (Taipei) . 1996;58:18.
9 Millaire A, Leroy O, Gaday V, et al. Incidence and prognosis of embolic events and metastatic infections in infective endocarditis. Eur Heart J . 1997;18:677.
10 Roder BL, Wandall DA, Espersen F, et al. Neurologic manifestations in Staphylococcus aureus endocarditis: a review of 260 bacteremic cases in nondrug addicts. Am J Med . 1997;102:379.
11 Heiro M, Nikoskelainen J, Engblom E, et al. Neurologic manifestations of infective endocarditis: a 17-year experience in a teaching hospital in Finland. Arch Intern Med . 2000;160:2781.
12 Hoen B, Alla F, Selton-Suty C, et al. Changing profile of infective endocarditis: results of a 1-year survey in France. JAMA . 2002;288:75.
13 Durante ME, Adinolfi LE, Tripodi MF, et al. Risk factors for “major” embolic events in hospitalized patients with infective endocarditis. Am Heart J . 2003;146:311.
14 Thuny F, Disalvo G, Belliard O, et al. Risk of embolism and death in infective endocarditis: prognostic value of echocardiography. A prospective multicenter study. Circulation . 2005;112:69.
15 Miro JM, Anguera I, Cabell CH, et al. Staphylococcus aureus native valve infective endocarditis: report of 566 episodes from the International Collaboration on Endocarditis Merged Database. Clin Infect Dis . 2005;41:507.
16 Bohmfalk GL, Story JL, Wissinger JP, et al. Bacterial intracranial aneurysms. J Neurosurg . 1978;48:369.
17 Hart RG, Kagan-Hallet K, Joerns SE. Mechanisms of intracranial hemorrhage in infective endocarditis. Stroke . 1987;18:1048.
18 Masuda J, Yutani C, Waki R, et al. Histopathological analysis of the mechanisms of intracranial hemorrhage complicating infective endocarditis. Stroke . 1992;23:843.
19 Cerebral Embolism Task Force. Cardiogenic brain embolism—the second report of the cerebral embolism task force. Arch Neurol . 1989;46:727.
20 Katz RI, Goldberg HI, Selzer ME. Mycotic aneurysm: case report with novel sequential angiographic findings. Arch Intern Med . 1974;134:939.
21 Molinari GF, Smith I, Goldstein MN, et al. Pathogenesis of cerebral mycotic aneurysm. Neurology . 1973;23:325.
22 Nakata T, Shinoya S, Kamiya K. The pathogenesis of mycotic aneurysm. Angiology . 1968;19:93.
23 Weir B. Special aneurysms (nonsaccular and saccular). In: Brown C, Eckhart C, editors. Aneurysms Affecting the Nervous System . Baltimore: Williams & Wilkins; 1987:134.
24 Hubbel G, Cheitlin MD, Rapaport E. Presentation, management, and follow-up evaluation of infective endocarditis in drug addicts. Am Heart J . 1981;102:85.
25 Levine DP, Crane LR, Zervos MJ. Bacteremia in narcotic addicts at the Detroit Medical Center. II. Infectious endocarditis: a prospective comparative study. Rev Infect Dis . 1986;8:374.
26 Di Salvo G, Habib G, Pergola V, et al. Echocardiography predicts embolic events in infective endocarditis. J Am Coll Cardiol . 2001;37:1069.
27 Macarie C, Iliuta L, Savulescu C, et al. Echocardiographic predictors of embolic events in infective endocarditis. Kardiol Pol . 2004;60:535.
28 Jones HRJr, Caplan LR, Come PC, et al. Cerebral emboli of paradoxical origin. Ann Neurol . 1983;13:314.
29 Stagaman DJ, Presti C, Rees C, et al. Septic pulmonary arteriovenous fistula: an unusual conduit for systemic embolization in right-sided valvular endocarditis. Chest . 1990;97:1484.
30 Le Cam B, Guivarch G, Boles JM, et al. Neurologic complications in a group of 86 bacterial endocarditis. Eur Heart J . 1984;5(suppl C):97.
31 Kanter MC, Hart RG. Neurologic complications of infective endocarditis. Neurology . 1991;41:1015.
32 Rohmann S, Erbel R, Gorge G, et al. Clinical relevance of vegetation localization by transoesophageal echocardiography in infective endocarditis. Eur Heart J . 1992;12:446.
33 Vilacosta I, Graupner C, San Roman JA, et al. Risk of embolization after institution of antibiotic therapy for infective endocarditis. J Am Coll Cardiol . 2002;39:1489.
34 Wong D, Chandraratna AN, Wishnow RM, et al. Clinical implications of large vegetations in infectious endocarditis. Arch Intern Med . 1983;143:1874.
35 Garvey GJ, Neu HC. Infective endocarditis—an evolving disease. Medicine (Baltimore) . 1978;57:105.
36 Schulz R, Werner GS, Fuchs JB, et al. Clinical outcome and echocardiographic findings of native and prosthetic valve endocarditis in the 1990’s. Eur Heart J . 1996;17:281.
37 Carpenter JL, McAllister CK. Anticoagulation in prosthetic valve endocarditis. South Med J . 1983;76:1372.
38 Keyser DL, Biller J, Coffman T, et al. Neurologic complications of late prosthetic valve endocarditis. Stroke . 1990;21:472.
39 Cabell CH, Jollis JG, Peterson GE, et al. Changing patient characteristics and the effect on mortality in endocarditis. Arch Intern Med . 2002;162:90.
40 Gordon KA, Beach ML, Biedenbach DJ, et al. Antimicrobial susceptibility patterns of beta-hemolytic and viridans group streptococci: report from the SENTRY Antimicrobial Surveillance Program (1997–2000). Diagn Microbiol Infect Dis . 2002;43:157.
41 Salgado CD, Farr BM, Calfe DP. Community-acquired methicillin-resistant Staphylococcus aureus: a meta-analysis of prevalence and risk factors. Clin Infect Dis . 2003;36:131.
42 Baddour LM, Wilson WR, Bayer AS, et al. Infective endocarditis: diagnosis, antimicrobial therapy, and management of complications: a statement for healthcare professionals from the Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease, Council on Stroke, and Cardiovascular Surgery and Anesthesia, American Heart Association: Endorsed by the Infectious Diseases Society of America. Circulation . 2005;111:294.
43 Gransden WR, Ekyn SJ, Leach RM. Neurological presentations of native valve endocarditis. QJM . 1989;73:1135.
44 Kupferwasser I, Darius H, Muller AM, et al. Clinical and morphological characteristics in Streptococcus bovis endocarditis: a comparison with other causative microorganisms in 177 cases. Heart . 1998;80:276.
45 Felner JM, Dowell VRJr. Anaerobic bacterial endocarditis. N Engl J Med . 1970;283:1188.
46 Davenport J, Hart RG. Prosthetic valve endocarditis 1976–1987: antibiotics, anticoagulation, and stroke. Stroke . 1990;21:993.
47 Powderly WG, Stanley SL, Medoff G. Pneumococcal endocarditis: report of a series and review of the literature. Rev Infect Dis . 1986;8:786.
48 Straus AL, Hamburger M. Pneumococcal endocarditis in the antibiotic era. Arch Intern Med . 1966;118:190.
49 Watanakunakorn C, Tan JS, Phair JP. Some salient features of Staphylococcus aureus endocarditis. Am J Med . 1973;54:473.
50 Steckelberg JM, Murphy JG, Ballard D, et al. Emboli in infective endocarditis: the prognostic value of echocardiography. Ann Intern Med . 1991;114:635.
51 Mügge A, Daniel WG, Frank G, et al. Echocardiography in infective endocarditis: reassessment of prognostic implications of vegetation size determined by the transthoracic and the transesophageal approach. J Am Coll Cardiol . 1989;14:631.
52 Daniel WG, Schroder E, Mügge A, et al. Transesophageal echocardiography in infective endocarditis. Am J Card Imaging . 1988;2:78.
53 Erbel R, Rohmann S, Drexler M. Improved diagnostic value of echocardiography in patients with infective endocarditis by transesophageal approach: a prospective study. Eur Heart J . 1988;1:43.
54 Reynolds HR, Jagen MA, Tunick PA, et al. Sensitivity of transthoracic versus transesophageal echocardiography for the detection of native valve vegetations in the modern era. J Am Soc Echocardiogr . 2003;16:67.
55 Heidenreich PA, Masoudi FA, Maini B, et al. Echocardiography in patients with suspected endocarditis: a cost-effectiveness analysis. Am J Med . 1999;107:198.
56 Jung HO, Seung KB, Kang DH, et al. A clinical consideration of systemic embolism complicated to infective endocarditis in Korea. Korean J Intern Med . 1994;9:80.
57 Heinle S, Wilderman N, Harrison JK, et al. Value of transthoracic echocardiography in predicting embolic events in active infective endocarditis. Duke Endocarditis Service. Am J Cardiol . 1994;74:799.
58 De Castro S, Magni G, Beni S, et al. Role of transthoracic and transesophageal echocardiography in predicting embolic events in patients with active infective endocarditis involving native cardiac valves. Am J Cardiol . 1997;80:1030.
59 Wann LS, Hallam CC, Dillon JC, et al. Comparison of m-mode and cross-sectional echocardiography in infective endocarditis. Circulation . 1979;60:728.
60 Buda AJ, Zotz RJ, LeMire MS, et al. Prognostic significance of vegetations detected by two-dimensional echocardiography in infective endocarditis. Am Heart J . 1986;112:1291.
61 Sanfilippo AJ, Picard MH, Newell JB, et al. Echocardiographic assessment of patients with infectious endocarditis: prediction of risk for complications. J Am Coll Cardiol . 1991;18:1191.
62 Lancellotti P, Galiuto L, Alert A, et al. Relative value of clinical and transesophageal echocardiographic variables for risk stratification in patients with infective endocarditis. Clin Cardiol . 1998;21:572.
63 Deprele C, Berthelot P, Lemetayer F, et al. Risk factors for systemic emboli in infective endocarditis. Clin Microbiol Infect . 2004;10:46.
64 Vuille C, Nidorf M, Weyman AE, et al. Natural history of vegetations during successful medical treatment of endocarditis. Am Heart J . 1994;128:1200.
65 Rohmann S, Erbel R, Darius H, et al. Prediction of rapid versus prolonged healing of infective endocarditis by monitoring vegetation size. J Am Soc Echocardiogr . 1991;4:465.
66 Rohmann S, Erbel R, Darius H, et al. Effect of antibiotic treatment on vegetation size and complication rate in infective endocarditis. Clin Cardiol . 1997;20:132.
67 Rohmann S, Erbel R, Darius H, et al. Spontaneous echo contrast imaging in infective endocarditis: a predictor of complications? Int J Cardiac Imaging . 1992;8:197.
68 Kupferwasser LI, Taha TH, Durrant S, et al. Hemostatic studies in patients with infective endocarditis: a report on nine consecutive cases with evidence of coagulopathy. Heart Vessels . 1991;6:102.
69 Kupferwasser LI, Harner G, Mohr-Kahaly S, et al. The presence of infection-related antiphospholipid antibodies in infective endocarditis determines a major risk factor for embolic events. J Am Coll Cardiol . 1999;33:1365.
70 Lydakis C, Apostolakis S, Lydataki N, et al. Stroke-complicated endocarditis with positive lupus anticoagulant—a case report. Angiology . 2005;56:503.
71 Korkmaz S, Ileri M, Hisar I, et al. Increased levels of soluble adhesion molecules, E-selectin and P-selectin, in patients with infective endocarditis and embolic events. Eur Heart J . 2001;22:874.
72 Bakshi R, Wright PD, Kinkel PR, et al. Cranial magnetic resonance imaging finding in bacterial endocarditis: the neuroimaging spectrum of septic brain embolization demonstrated in twelve patients. J Neuroimaging . 1999;9:78.
73 Moulin T, Crepin-Leblond T, Chopard J-L, Bogousslavsky J. Hemorrhagic infarcts. Eur Neurol . 1993;34:64.
74 Kim SJ, Lee JY, Kim TH, et al. Imaging of the neurological complications of infective endocarditis. Neuroradiology . 1998;40:109.
75 Bertorini TE, Laster RE, Thompson BF, et al. Magnetic resonance imaging of the brain in bacterial endocarditis. Arch Intern Med . 1989;149:815.
76 van der Meulen JHP, Weststrate W, van Gijn J, et al. Is cerebral angiography indicated in infective endocarditis? Stroke . 1992;23:1662.
77 Salgado AV, Furlan AJ, Keys TF. Mycotic aneurysm, subarachnoid hemorrhage, and indications for cerebral angiography in infective endocarditis. Stroke . 1987;18:1057.
78 Brust JC, Taylor Dickinson PC, Hughes JEO, et al. The diagnosis and treatment of cerebral mycotic aneurysms. Ann Neurol . 1990;27:238.
79 Ellis CJ, Waite ST, Coverdale HA, et al. Transoesophageal echocardiography in patients with prosthetic heart valves and systemic emboli: is it a useful investigation? N Z Med J . 1995;108:376.
80 Merritt HH, Fremont-Smith F. The Cerebrospinal Fluid. Philadelphia: WB Saunders, 1938.
81 Sornas R, Ostlund H, Muller R. Cerebrospinal fluid cytology after stroke. Arch Neurol . 1972;26:489.
82 Nicolau DP, Freeman CD, Nightingale CH, et al. Reduction of bacterial titers by low dose aspirin in experimental aortic valve endocarditis. Infect Immun . 1993;61:1593.
83 Nicolau DP, Tessier PR, Nightingale CH. Beneficial effect of combination antiplatelet therapy on the development of experimental Staphylococcus aureus endocarditis. Int J Antimicrob Agents . 1999;11:159.
84 Kupferwasser LI, Yeaman MR, Shapiro SM, et al. Acetylsalicylic acid reduces vegetation bacterial density, hematogenous bacterial dissemination, and frequency of embolic events in experimental Staphylococcus aureus endocarditis through antiplatelet and antibacterial effects. Circulation . 1999;99:2791.
85 Taha TH, Durrant SS, Mazeika PK, et al. Aspirin to prevent growth of vegetations and cerebral emboli in infective endocarditis. J Intern Med . 1992;231:543.
86 Chan KL, Dumesnil JG, Cujec B, et al. A randomized trial of aspirin on the risk of embolic events in patients with infective endocarditis. J Am Coll Cardiol . 2003;42:775.
87 Leport C, Vilde JL, Bricaire F, et al. Fifty cases of late prosthetic valve endocarditis: improvement in prognosis over a 15-year period. Br Heart J . 1987;58:66.
88 Wilson WR, Geraci JE, Danielson GK, et al. Anticoagulant treatment and central nervous system complications in patients with prosthetic valve endocarditis. Circulation . 1978;57:1004.
89 Delahaye JP, Poncet P, Malquarti V, et al. Cerebrovascular accidents in infective endocarditis: role of anticoagulation. Eur Heart J . 1990;11:1074.
90 Tornos P, Almirante B, Mirabet S, et al. Infective endocarditis due to Staphylococcus aureus: deleterious effect of anticoagulant therapy. Arch Intern Med . 1999;159:473.
91 Salgado AV. Central nervous system complications of infective endocarditis. Curr Concepts Cerebrovasc Dis Stroke . 1991;26:19.
92 Solenski NJ, Haley ECJr. Neurological complications of infective endocarditis. In: Roos KL, editor. Central Nervous System Infectious Diseases and Therapy . New York: Marcel Dekker; 1997:331.
93 Gillinov AM, Shah RV, Curtis WE, et al. Valve replacement in patients with endocarditis and acute neurologic deficit. Ann Thorac Surg . 1996;61:1125.
94 Chukwudelunzu FE, Brown RDJr, Wijdicks EF, et al. Clinical features and etiology of subarachnoid hemorrhage associated with infectious endocarditis. Neurology . 1999;52(suppl 2):A503.
95 Jebara VA, Acar C, Dervanian P, et al. Mycotic aneurysms of the carotid arteries—case report and review of the literature. J Vasc Surg . 1991;14:215.
96 Mansur AJ, Grinberg M, Leao PP, et al. Extracranial mycotic aneurysms in infective endocarditis. Clin Cardiol . 1986;9:65.
97 Heyd J, Yinnon AM. Mycotic aneurysm of the external carotid artery. J Cardiovasc Surg (Torino) . 1994;35:329.
98 Sikert RG, Jones HRJr. Transient cerebral attacks associated with subacute bacterial endocarditis. Stroke . 1970;1:178.
99 Simmons KC, Sage MR, Reilly PL. CT of intracerebral haemorrhage due to mycotic aneurysms: case report. Neuroradiology . 1980;19:215.
100 Ahmadi J, Tung H, Giannotta SL, et al. Monitoring of infectious intracranial aneurysms by sequential computed tomographic/magnetic resonance imaging studies. Neurosurgery . 1993;32:45.
101 Mayer SA, Brun NC, Begtrup K, et al. Recombinant activated factor VII for acute intracerebral hemorrhage. N Engl J Med . 2005;352:777.
102 Corr P, Wright M, Handler LC. Endocarditis-related cerebral aneurysms: radiologic changes with treatment. AJNR Am J Neuroradiol . 1995;16:745.
103 Morawetz RB, Karp RB. Evolution and resolution of intracranial bacterial (mycotic) aneurysms. Neurosurgery . 1984;15:43.
104 Bingham WF. Treatment of mycotic intracranial aneurysms. J Neurosurg . 1977;46:428.
105 Barrow DL, Prats AR. Infectious intracranial aneurysms: comparison of groups with and without endocarditis. Neurosurgery . 1990;27:562.
106 Monsuez JJ, Vittecoq D, Rosenbaum A, et al. Prognosis of ruptured intracranial mycotic aneurysms: a review of 12 cases. Eur Heart J . 1989;10:821.
107 Schold C, Earnest MP. Cerebral hemorrhage from a mycotic aneurysm developing during appropriate antibiotic therapy. Stroke . 1978;9:267.
108 Scotti G, Li MH, Righi C, et al. Endovascular treatment of bacterial intracranial aneurysms. Neuroradiology . 1996;38:186.
109 Frizzell RT, Vitek JJ, Hill DL, et al. Treatment of a bacterial (mycotic) intracranial aneurysm using an endovascular approach. Neurosurgery . 1993;32:852.
110 Chapot R, Houdart E, Saint-Maurice JP, et al. Endovascular treatment of cerebral mycotic aneurysms. Radiology . 2002;222:389.
111 Sugg RM, Weir R, Vollmer DG, Cacayorin ED. Cerebral mycotic aneurysms treated with a neuroform stent: technical case report. Neurosurgery . 2006;58:E381.
112 Kato Y, Yamaguchi S, Sano H, et al. Stereoscopic synthesized brain-surface imaging with MR angiography for localization of a peripheral mycotic aneurysm: case report. Minim Invasive Neurosurg . 1996;39:113.
113 Cunha e Sa M, Sisti M, Solomon R. Stereotactic angiographic localization as an adjunct to surgery of cerebral mycotic aneurysms: case report and review of the literature. Acta Neurochir (Wien) . 1997;139:625.
114 Grandsden WR, Eykyn SJ, Leach RM. Neurological presentations of native valve endocarditis. QJM . 1989;73:1135.
115 Wolff M, Regnier B, Witchitz S, et al. Pneumoccocal endocarditis. Eur Heart J . 1984;5:C7780.
116 Tunkel AR, Kaye D. Neurologic complications of infective endocarditis. Neurol Clin . 1993;11:419.
117 Pozzati E, Tognetti F, Padovani R, et al. Association of cerebral mycotic aneurysm and brain abscess. Neurochirurgia (Stuttg) . 1983;26:18.
118 Jones HRJr, Siekert RG. Embolic mononeuropathy and bacterial endocarditis. Arch Neurol . 1968;19:535.
119 Chu VH, Cabell CH, Benjamin DK, et al. Early predictors of in-hospital death in infective endocarditis. Circulation . 2004;109:1745.
Chapter 7 Neurological Complications of Hypertension

S. Claiborne Johnston, Jacob S. Elkins

Unruptured Cerebral Aneurysms
Subarachnoid Hemorrhage
Blood pressure was first measured in 1707 by an English divinity student, Stephan Hales, using a glass tube attached directly into the arteries of animals. 1 Methods of measurement improved slowly over the next 200 years, with Nikolai Korotkoff describing the modern cuff-and-stethoscope technique in 1905. Hypertension was recognized as an indicator of poor prognosis by Theodore Janeway, who published a case series of 7,872 hypertensive patients gathered from 1903 to 1912, in which hypertension was defined as a systolic blood pressure greater than 160 mmHg. He found a mean survival of 4 to 5 years after the development of symptoms of hypertension, with stroke being an important cause of death.
Hypertension was initially considered a compensatory phenomenon rather than a disease in itself. Even into the 1940s, physicians were concerned that lowering blood pressure would exacerbate end-organ damage, particularly in the kidneys. Treatment options were not available until 1925, when surgical sympathectomy was shown to reduce blood pressure without impairment in renal function. The first antihypertensive medication, tetraethylammonium, was used in a patient in 1946, but the agent was poorly tolerated because of severe anticholinergic side effects. A tolerable oral agent was not available until 1957, when chlorothiazide was shown to reduce blood pressure in patients with essential hypertension and rapidly became the most commonly prescribed medication.
Both acute hypertension and chronic hypertension produce neurological disease. Acute hypertension is associated with hypertensive encephalopathy, an uncommon presentation since the widespread identification and treatment of hypertension. Chronic hypertension is associated with stroke, which is its most important neurological complication. All stroke subtypes are linked to hypertension, including ischemic infarction, intraparenchymal hemorrhage, and aneurysmal subarachnoid hemorrhage. Chronic hypertension is also associated with dementia and with peripheral neuropathy in those with diabetes.

Both systolic and diastolic blood pressures are distributed approximately normally in the population. For convenience, physicians have defined pathological states such as hypertension based on specific blood pressure thresholds, typically a systolic blood pressure of 140 mmHg or greater or a diastolic blood pressure of 90 mmHg or greater, or both. Thus defined, hypertension is common, affecting approximately 50 million individuals in the United States and as many as 1 billion worldwide. 2 In the Framingham study, individuals who were normotensive at age 55 had an approximately 90 percent lifetime risk of developing hypertension. 3 Despite the frequent division of blood pressure into diagnostic categories such as hypertension and normotension, there is no obvious threshold at which higher blood pressure begins affecting the risk of complications, and even patients with diastolic blood pressures of 80 to 90 mmHg are at increased risk of stroke compared with those with blood pressures of 70 to 80 mmHg ( Fig. 7-1 ). 4 Reflecting a growing awareness of the continuous risk associated with blood pressure, blood pressures in the range of 120–140/80–90 mmHg, once considered to be “normal,” are now labeled as “prehypertensive.” 2 Throughout much of the twentieth century, blood pressure risk was assessed according to the diastolic blood pressure, and it was not until 1993 that systolic blood pressure was formally incorporated into the definition of hypertension in U.S. guidelines. 5 Since that time, however, it has been increasingly recognized that systolic blood pressure is somewhat more informative than diastolic blood pressure at predicting future cardiovascular events. 6

FIGURE 7-1 Relative risks of stroke. Estimates of the usual diastolic blood pressure (DBP) in each baseline DBP category are taken from mean DBP values 4 years after baseline in the Framingham study. Solid squares represent disease risks in each category relative to risk in the whole study population; sizes of squares are proportional to the number of events in each DBP category; and 95 percent confidence intervals for estimates of relative risk are denoted by vertical lines.
(From MacMahon S, Peto R, Cutler J, et al: Blood pressure, stroke, and coronary heart disease. Lancet 335:764, 1990, with permission.)

In the brain, the primary pathophysiologic process of hypertension is related to increases in vasomotor tone and peripheral arterial resistance. Acute elevation in blood pressure results in constriction of small arteries in the brain in a compensatory response termed autoregulation. Brain blood-flow is maintained at a relatively constant level over a range of pressures. At high pressures, vasoconstriction is thought to be protective by reducing pressure at smaller, more distal vessels. Acute severe hypertension overwhelms normal autoregulation at a mean arterial pressure of approximately 150 mmHg, with increased cerebral blood-flow occurring above this pressure threshold. Vasoconstriction in acute hypertension is patchy, and some small vessels are exposed to high pressures, which may lead to endothelial injury and focal breakdown of the blood–brain barrier. 7 Acute hypertensive encephalopathy is a fulminant presentation of this process. Fibrinoid necrosis of small vessels may also occur, lowering the threshold for future ischemic and hemorrhagic events.
Chronic hypertension results in cerebral vascular remodeling. The media hypertrophies, and the lumen becomes narrowed. 8 These changes are protective, with reduction in wall tension and shifting of the autoregulation curve to allow compensation at higher blood pressures. 9 However, vascular remodeling is accompanied by endothelial dysfunction, with impaired relaxation and poor compensation for hypoperfusion. The result is greater susceptibility to ischemic injury due to reduced collateral flow. 7
Hypertension also predisposes to atherosclerosis. Hypertension is proinflammatory and is accompanied by increased plasma oxygen free radicals. 10 Free radicals induce vascular smooth muscle cell proliferation and may oxidize low-density lipoproteins, which in turn promotes macrophage activation and monocyte extravasation. Angiotensin II is elevated in many hypertensives and may play a direct role in atherogenesis independent of its effects on blood pressure. 11 It directly stimulates smooth muscle cell growth, hypertrophy, and lipoxygenase activity, with resultant inflammation and low-density lipoprotein oxidation, 10 thus accelerating atherosclerosis. Angiotensin II also stimulates the production of transforming growth factor β (TGF-β), a cytokine that is linked to fibrosis in a number of disease states. In animal models, transforming growth factor β appears to play a causal role in the development of hypertension and pathological vessel remodeling. 12

The gold standard of blood pressure measurement is auscultation using a mercury sphygmomanometer. Newer devices can provide accurate readings but require calibration. Blood pressure should be measured in the seated position after a 5-minute rest with the patient’s feet resting on the floor and the arm supported at heart level during the measurement. Accurate readings depend on the use of an appropriate-sized cuff with the bladder covering at least 80 percent of the arm. The classification of blood pressure into specific diagnostic categories is based on the average of two or more readings on each of two or more office visits. 13 A complete history and physical examination with basic laboratory measurements are essential to evaluate for identifiable causes of hypertension and assess risk. Several patient characteristics may suggest an identifiable cause of hypertension including young age, severe hypertension, hypertension that is refractory to multiple interventions, and physical or laboratory findings suggestive of endocrinological disorders, such as truncal obesity or hypokalemia. Abdominal bruits or decreased femoral pulses may also be an indicator of renovascular disease or coarctation of the aorta. 14
Lifestyle modification is recommended as an initial therapy for patients with blood pressure of 120/80 mmHg or higher. 2 Effective lifestyle interventions include weight loss, limited alcohol intake, aerobic physical activity, adequate potassium intake (approximately 90 mmol/day), reduction in sodium intake, and dietary regimens such as the Dietary Approaches to Stop Hypertension (DASH) eating plan. 15 Antihypertensive medications are recommended in addition to lifestyle measures for patients with blood pressure of 140/90 mmHg or higher, and when the blood pressure is 130/80 mmHg or higher in those with diabetes and chronic kidney disease.
For subjects without a history of cardiovascular disease or other compelling indication, initiating therapy with a thiazide diuretic such as chlorthalidone, is generally recommended. In the Antihypertensive and Lipid Lowering to Prevent Heart Attack Trial (ALLHAT), involving more than 33,000 participants, therapy with chlorthalidone was either equivalent or superior to lisinopril and amlodipine for the primary prevention of cardiovascular endpoints, with a particular benefit for African American subjects both in terms of safety and efficacy. 16 When the blood pressure is 160/100 mmHg or higher, initiating therapy with two-drug combinations is generally recommended. 2
There are many benefits to treating hypertension, especially reductions in myocardial infarctions, congestive heart failure, retinopathy, renal failure, and overall mortality. The focus of the remainder of this chapter is on specific neurological complications of hypertension and the unique aspects of treatment that they necessitate.

Of all the identified modifiable risk factors for stroke, hypertension appears to be the most important, owing to its high prevalence and its associated three- to fivefold increase in stroke risk. 17 Based on epidemiological data, approximately 50 percent of strokes could be prevented if hypertension were eliminated ( Table 7-1 ). 18 Even small reductions in blood pressure result in large reductions in stroke risk. For example, in a meta-analysis of 37,000 hypertensive subjects from 14 studies, a reduction of 5 to 6 mmHg in diastolic blood pressure with active treatment was associated with a 42 percent reduction in stroke risk. 19 The benefits of blood pressure reduction on stroke risk extend similarly to the elderly with isolated elevations in systolic blood pressure. In the Systolic Hypertension in the Elderly Program (SHEP) trial of 4,736 subjects 60 years and older, a 36 percent reduction in stroke was seen with a 12-mmHg decline in systolic pressure, a finding confirmed in other large randomized trials. 20, 21 Although there is still some uncertainty surrounding the treatment of blood pressure in the oldest old (>85 years), the best available data suggest that benefits will be comparable with those seen in younger individuals. 22 Stroke rates have generally declined worldwide, especially throughout the 1970s and 1980s, although more recently they appear to have plateaued ( Figs. 7-2 and 7-3 ). 23 Although these historic trends are not entirely explained by better control of blood pressure, the rates of decline have roughly paralleled increased use of antihypertensive medications, suggesting that benefits of blood pressure therapy observed in randomized trials have been at least partially realized in community practice. 24

TABLE 7-1 Estimated Impact of Modifiable Risk Factors on Stroke in the United States *

FIGURE 7-2 Percent change in stroke mortality, men aged 35 to 74, 1972 to 1982.
(From Thom TJ: Stroke mortality trends: an international perspective. Ann Epidemiol 3:509, 1993. Copyright 1993, with permission from Elsevier Science.)

FIGURE 7-3 Age-adjusted death rates for stroke among men and women in the United States, 1900 through 1990.
(From Higgins M, Thom T: Trends in stroke risk factors in the United States. Ann Epidemiol 3:550, 1993. Copyright 1993, with permission from Elsevier Science.)
Hypertension contributes to each of the major intermediate causes of both ischemic and hemorrhagic stroke including carotid stenosis, intracranial atherosclerosis, small-vessel arteriosclerosis, and both macroscopic and microscopic aneurysms. Each of these conditions is considered separately in this chapter. In part because of the heterogeneity of its manifestations in the brain, there continues to be some uncertainty about the optimal management of blood pressure in both the acute and chronic phases after stroke.
In the acute phase of cerebral ischemia, hypertension may play a compensatory role in maintaining cerebral perfusion to viable but threatened areas of the brain. 25 Loss of normal cerebral autoregulation has been demonstrated in areas of ischemic brain. When autoregulation is lost, blood flow to the brain becomes directly proportional to mean arterial pressure, and therefore, in theory, pharmacological increases in blood pressure could have salutatory effects in preserving hypoperfused regions of the brain. 26 In some small studies, rapid pharmacological reductions in blood pressure have predicted worse outcomes, and there are numerous anecdotal reports of the recrudescence of stroke symptoms after a decrease in blood pressure. 27, 28 Therefore, most stroke guidelines recommend withholding pharmacological treatments of blood pressure in acute stroke in the absence of acute end-organ injury or administration of thrombolytics, unless the blood pressure exceeds 220/120 mmHg. 29 It is also possible, however, to make physiological arguments that would be supportive of acute blood pressure reduction, such as stabilization of an intra-arterial thrombus or to reduce edema formation. 30 Ongoing trials in this area will provide key data to help resolve this debate.
Although historically there has been concern about lowering blood pressure even in the chronic phases after stroke, there is now overwhelming evidence to support the use of pharmacological interventions to lower blood pressure for secondary stroke prevention. In 6,105 subjects with a history of stroke, the Perindopril Protection Against Recurrent Stroke Study (PROGRESS) demonstrated a 43 percent relative risk reduction for secondary stroke prevention when subjects were randomized to the combination of the angiotensin-converting enzyme (ACE) inhibitor perindopril and the thiazide diuretic indapamide. 31 Combination therapy with the ACE inhibitor and thiazide, which resulted in a mean blood pressure reduction of 12.3/5 mmHg, demonstrated a substantially more robust benefit for stroke prevention than monotherapy with ramipril (relative risk reduction 5%), which produced only a 4.9/2.8-mmHg average reduction in blood pressure ( P for heterogeneity between treatments <0.001). Combination therapy with an ACE inhibitor and a thiazide is now commonly recommended for secondary stroke prevention, with benefits appearing to be similar regardless of whether measured blood pressure is above or below the traditional cut points for hypertension. 32 Although other studies have supported the finding that therapy with renin-angiotensin system antagonists and diuretics provides especially strong benefits for stroke prevention, particularly when compared with β-blockers, 33 the degree of hypertension control that is achieved is usually the best predictor of protection against recurrent stroke. Therefore, response to therapy and other comorbidities, such as heart failure, diabetes, asthma, and arrhythmia, should be considered when deciding on an appropriate antihypertensive drug regimen. 2 There is still debate about how soon after stroke to initiate therapy, as the majority of patients in trials of blood pressure therapy after stroke have been randomized months after their qualifying event. Early initiation of therapy is increasingly practiced to improve patient compliance, and current guidelines recommend consideration of treatment once the “hyperacute” period has ended. 32, 34

Cerebral aneurysms are focal dilations of blood vessels. Subarachnoid hemorrhage, an important form of hemorrhagic stroke, occurs when a cerebral aneurysm ruptures ( Fig. 7-4 ). Hypertension is associated with cerebral aneurysm formation and with subarachnoid hemorrhage. In a large sample of Medicare patients, hypertension was listed as a secondary diagnosis in 43 percent of patients admitted with unruptured aneurysms and in 34 percent of hospitalized control subjects. 35 In a meta-analysis, the risk of subarachnoid hemorrhage was 2.8 times greater in those with a history of hypertension. 36

FIGURE 7-4 A ruptured anterior communicating artery aneurysm producing acute subarachnoid hemorrhage. A, Head computed tomography (CT) shows a large amount of blood at the base of the brain and a small amount of intraventricular blood. B, Angiogram reveals a complex saccular aneurysm.
The cause of the development and rupture of cerebral aneurysms is probably multifactorial. Epidemiological studies have found several environmental risk factors for subarachnoid hemorrhage other than hypertension. Cigarette smoking increases the risk of subarachnoid hemorrhage by 100 percent or more, 36 perhaps by increasing the release of proteolytic enzymes that affect blood vessel integrity. 37 Heavy alcohol consumption increases subarachnoid hemorrhage risk with a pooled odds ratio of 1.5 in case control studies and relative risk of 4.7 in cohort studies. 36 Alcohol-induced hypertension, relative anticoagulation, or increased cerebral blood-flow may be responsible. 37 Oral contraceptives are associated with a small but significant excess risk of subarachnoid hemorrhage, with a relative risk of 1.4 in current and past users. 38 The source of the association is unknown.
Genetic factors are also important to aneurysm formation and subarachnoid hemorrhage. The risk of subarachnoid hemorrhage is three to seven times greater in patients with an affected first-degree relative, 39 and the prevalence of unruptured aneurysms is probably at least twice as high as without a family history. 40 Females are twice as likely to have an aneurysm or present with subarachnoid hemorrhage. African Americans have twice the rate of subarachnoid hemorrhage as whites. 41 Polycystic kidney disease, Ehlers–Danlos syndrome type 4, and α 1 -antitrypsin deficiency are also associated with increased risk. Marfan’s syndrome has been considered a risk factor for aneurysm formation, but this association has recently been questioned. 42
The pathology of aneurysms reveals little about the underlying etiology. Two major morphological subtypes are recognized: fusiform and saccular or berry aneurysms. Fusiform aneurysms more commonly occur in children and in the elderly. The childhood form is thought to represent a genetic or early developmental abnormality in vessel wall structure. 43 Fusiform aneurysms in the elderly are often associated with intracranial atherosclerosis. 44 Saccular aneurysms commonly occur at vessel branch points at the base of the brain, with the middle cerebral artery bifurcation and origins of the anterior communicating artery and posterior communicating artery representing the most frequent locations. 45 Turbulent flow may be responsible for this tropism.

Unruptured Cerebral Aneurysms
Estimates of the prevalence of unruptured aneurysms vary widely. A recent meta-analysis of prospective studies in adults reported 3.6 percent prevalence in four autopsy series and 6.0 percent in nine angiographic studies. 46 Prevalence was 2.3 percent in those without a known risk factor. Approximately 90 percent of these aneurysms were less than 10 mm in diameter, and 70 percent were less than 6 mm. Based on the prevalence from angiographic studies, an estimated 11 million American adults have an unruptured aneurysm, and these are being detected more frequently with advances in imaging studies. Annual cost for unruptured aneurysms in the United States was estimated at $522 million in the 1980s 47 and is probably significantly greater now.
Unruptured aneurysms are often asymptomatic, discovered incidentally in a work-up for an unrelated problem. Some aneurysms produce symptoms by compressing neighboring structures. Presentation with a new cranial neuropathy is considered a worrisome sign for imminent rupture and often prompts urgent treatment. New headaches are also a presenting sign of unruptured aneurysm. Although migraine may simply represent an unrelated occurrence prompting head imaging, some headaches may be due to the aneurysm itself. A sudden, severe “thunderclap” headache may herald rapid aneurysm growth or a small leak without evidence of subarachnoid hemorrhage. 48
Catheter angiography is the gold standard for detection of aneurysms. Magnetic resonance (MR) angiography is approximately 85 percent sensitive for detecting aneurysms larger than 3 mm, with 85 percent specificity. 49 Head computed tomography (CT) does not reliably detect unruptured aneurysms.
Prognosis of unruptured aneurysms, as reflected in the rate of rupture, is a subject of controversy. In the largest prospective cohort study, the International Study of Unruptured Intracranial Aneurysms, 1,692 subjects with unruptured aneurysms who did not undergo surgery or endovascular treatment, were followed prospectively for an average of 4.1 years. The size of the aneurysm (≥7 mm in maximal diameter) and location at the basilar tip or posterior communicating artery were independent predictors of hemorrhage. Among 1,077 subjects with no history of subarachnoid hemorrhage, the annual risk of hemorrhage for an aneurysm less than 7 mm in diameter in the anterior circulation was 0 percent; it was 0.5 percent when the aneurysm was located in the posterior circulation. 50
The standard of care for treatment of aneurysms has historically been surgical clipping, in which a metal clip is placed over the neck of the aneurysm, isolating it from the circulation. Coil embolization is an alternative therapy and involves packing platinum coils into an aneurysm through a microcatheter in an angiographic endovascular procedure. The relative merits of the two procedures have been argued. Coil embolization appears to provide a safer approach 51 but may not reduce subsequent rupture rates as effectively as surgical clipping.
Whether a given aneurysm requires treatment depends on the anticipated rupture rate. For asymptomatic aneurysms smaller than 7 mm with no history of subarachnoid hemorrhage, treatment may not be justified, particularly when in the anterior circulation, given the risks of surgery and endovascular therapy. 52 Treatment of unruptured aneurysms appears to be cost-effective when they are larger or symptomatic or when there is a history of subarachnoid hemorrhage from a different aneurysm.
Controlling or eliminating risk factors, such as hypertension, smoking, and alcohol abuse, may reduce rupture rates, but this has not been systematically studied.

Subarachnoid Hemorrhage
Subarachnoid hemorrhage accounts for approximately 5 percent of all strokes, but it tends to occur at a younger age than other stroke subtypes, with median age at death being 59 years for subarachnoid hemorrhage, 73 years for intracerebral hemorrhage, and 81 years for ischemic stroke. 53 Subarachnoid hemorrhage accounts for nearly one third of the years of potential life lost before age 65 due to stroke. Case fatality rates approach 50 percent, and another 10 to 20 percent remain disabled and dependent at follow-up. 54 Approximately 25,000 Americans present with subarachnoid hemorrhage each year, with total costs estimated at $5.6 billion. 55
Presentation with subarachnoid hemorrhage generally involves sudden onset of severe headache, sometimes accompanied by neck pain. Alteration of consciousness occurs in a minority of patients, but it may be severe enough to produce coma or sudden death outside the hospital. Head CT often shows blood surrounding the base of the brain. Intraventricular and intraparenchymal hemorrhage may be present and can provide clues as to the location of the ruptured aneurysm. Lumbar puncture may rarely show signs of hemorrhage when there is no evidence of it on head CT. Blood in the spinal fluid that does not clear is suggestive of subarachnoid hemorrhage. Xanthochromia is present in nearly all cases and may persist for more than 3 weeks, but its appearance is delayed by more than 12 hours in 10 percent of cases. Angiography is required for the characterization of the aneurysm and to plan treatment.
Prognosis depends on the ability to treat the underlying aneurysm and on the patient’s condition at presentation. Recurrent hemorrhage occurs in more than 4 percent of untreated patients during the first day and then in 1 to 2 percent per day for the next 2 weeks and is associated with even greater fatality and morbidity than primary rupture. 45 Regardless of treatment and recurrent hemorrhage, the level of consciousness is the major predictor of mortality ( Table 7-2 ). 56 The World Federation of Neurological Surgeons developed a Universal Subarachnoid Hemorrhage Grading Scale, similar to the older Hunt and Hess scale, which has been widely adopted but offers little advantage over determinations of level of consciousness alone. 57

TABLE 7-2 Overall Outcome After Subarachnoid Hemorrhage by Consciousness Level on Admission *
To reduce the risk of recurrent hemorrhage, aneurysms are generally rapidly identified and repaired with surgical clipping or endovascular coil embolization. Controversy continues about optimal timing of treatment in high-risk patients. Early surgery may be technically more challenging when a large amount of subarachnoid clot is present, but early surgical treatment reduces the risk of recurrent rupture more quickly than does later surgery. Most neurosurgeons therefore recommend early treatment. 45
Hydrocephalus from obstruction of the cerebral aqueduct or the meninges by blood clot may require external drainage. Vasospasm is a common complication that produces cerebral ischemia due to blood vessel constriction in areas with aneurysmal subarachnoid clot. It becomes symptomatic in one third of cases, usually 3 to 14 days after hemorrhage, and results in cerebral infarction or death in 15 to 20 percent. 58 Transcranial Doppler ultrasonography can detect vasospasm before it becomes symptomatic and is helpful in monitoring patients. 59 Oral nimodipine, a calcium-channel antagonist, reduces poor outcomes from vasospasm and should be given as soon as possible after the initial bleed in all cases. Hypertension induced with pressors and hydration with intravenous fluids may reduce the risk of infarction, but these measures have never been studied in trials. They should not be used in patients with untreated ruptured aneurysms because of the risk of precipitating further episodes of bleeding. Vasodilatation through angioplasty or intra-arterial verapamil (or other vasodilators) immediately reverses angiographic vasospasm in many cases, but it requires further study before clinical benefits are proven.

Bleeding directly into the substance of the brain is termed intraparenchymal or intracerebral hemorrhage ( Fig. 7-5 ). It may occur as a complication of ischemic stroke, termed hemorrhagic conversion, or as the primary injury without preceding ischemia. Hypertension is the most important identified risk factor for intracerebral hemorrhage. More than 70 percent of patients with intracerebral hemorrhage have a history of hypertension, and the risk of hemorrhagic stroke is elevated 9.5-fold in the highest compared with the lowest quintile of systolic blood pressure. 60

FIGURE 7-5 Head CT of an acute basal ganglia intracerebral hemorrhage with mass effect compressing the ventricles.
Intracranial hemorrhage is responsible for 10 to 15 percent of all stroke deaths but for more than one third of the years of life lost before age 65 due to the younger age distribution of intracerebral hemorrhage. 53 Case fatality rates are high, with 35 to 50 percent dead at 1 month and only 20 percent returning to independence at 6 months. 61 In the United States, an estimated 37,000 cases of intracerebral hemorrhage occur each year, with the total estimated cost of care exceeding $6 billion annually. 55
Other risk factors for intracerebral hemorrhage include age, race, substance abuse, anticoagulation, platelet dysfunction, and vascular and structural anomalies. Rates of intracerebral hemorrhage increase with age. African Americans have 40 percent higher rates than those of whites, with larger differences at younger ages. 41 Cocaine and amphetamines are associated with increased risk, particularly acutely, possibly because of transient severe hypertension. 60 Abnormalities in clotting may account for an increased incidence of intracerebral hemorrhage with heavy alcohol use. Excessive warfarin anticoagulation and antiplatelet therapy also increase the risk of intracerebral hemorrhage. 62, 63 Thrombolytic agents used for ischemic stroke and myocardial infarction cause intracerebral hemorrhage in some cases. It may also occur with severe thrombocytopenia and platelet dysfunction.
Intracerebral hemorrhage may result from and occur in brain tumors, such as glioblastoma multiforme and metastatic melanoma, choriocarcinoma, renal cell carcinoma, and bronchogenic carcinoma. Congophilic amyloid angiopathy, a vasculopathy common in the elderly, is associated with lobar hemorrhages, often centered at the gray-white junction. Other punctate hemorrhages may be apparent on gradient-echo MR images ( Fig. 7-6 ), supporting the diagnosis. Arteriovenous malformations, abnormal complexes of arteries and veins in brain parenchyma, account for 5 percent of intracerebral hemorrhage. 60 Cavernous malformations are dense collections of thin-walled vascular channels and appear to be the cause of intracerebral hemorrhage in 5 percent of autopsies 63 ; they are not apparent on angiography but have a “popcorn” appearance in T2-weighted MR images, with a hyperintense core surrounded by hypointense hemosiderin from previous small hemorrhages ( Fig. 7-7 ). Aneurysms may produce intracerebral hemorrhages when blood is directed into the brain, and these rarely are mistaken for primary hypertensive hemorrhages.

FIGURE 7-6 Imaging findings of amyloid angiopathy, with no evidence of hemorrhage on CT (A) but multiple punctate hypointensities at the gray-white junction on T2-weighted multiplanar gradient-recalled (MPGR) magnetic resonance imaging (MRI) (B), suggesting old hemorrhage (two lesions are apparent on this image in the parieto-occipital region).

FIGURE 7-7 A cavernous malformation with a small amount of acute, intracerebral hemorrhage surrounding it on CT (A). T2-weighted MRI (B) shows a lesion with a focal area of high signal intensity surrounded by a thick rim of hypointense siderin. T1-weighted MRI (C) showing the typical “popcorn” appearance. The high signal intensity represents methemoglobin.
Primary hypertensive intracerebral hemorrhage was thought to be caused by chronic vascular injury, resulting in formation of microscopic aneurysms, first characterized by Charcot and Bouchard in 1868. Advances in pathological tissue preparation have raised doubts about the frequency and importance of microscopic aneurysms, attributing the appearance to complex vascular coils. 64 More recently, fibrinoid necrosis of small arteries has been proposed as the initial step in intracerebral hemorrhage. 65 When acute hypertension or clotting abnormality precipitates rupture, blood dissects into the brain parenchyma, sometimes producing a hematoma. Brain injury occurs because of compression of surrounding tissue and from the direct toxic effects of blood. Mass effect from the hematoma may lead to uncal, subfalcine, tonsillar, or transtentorial herniation, depending on location, and death may ensue.
Clinical presentation depends on the location and size of the hemorrhage ( Table 7-3 ). Nearly all intracerebral hemorrhage is characterized by sudden onset of neurological deficits, progressing over minutes and accompanied by headache, often with alteration of consciousness. Deterioration due to surrounding edema, hydrocephalus, or continuing or recurrent hemorrhage often occurs within the first 24 hours but may be delayed by days.

TABLE 7-3 Clinical Presentation of Intracerebral Hemorrhage
Prognosis is multifactorial. Hemorrhage volume, most easily measured by halving the product of the length, width, and depth on axial head CT images, is a powerful predictor of mortality, with 80 percent 30-day mortality in those with volumes greater than 60 ml and 22 percent mortality in hemorrhages less than 30 ml. 66 Mortality is much greater in those with intraventricular extension of blood. 67 Hydrocephalus due to intraventricular extension or cerebrospinal fluid (CSF) outflow obstruction predicts in-hospital mortality: 51 percent of those with and 2 percent of those without hydrocephalus died in one series. 68 Lower Glasgow Coma Scale scores, greater age, location, and blood pressure or pulse pressure are other independent predictors of mortality. Simple multivariable prediction models have been developed and validated. 69, 70
Urgent head CT is required in patients with suspected intracerebral hemorrhage. MR imaging (MRI) is probably as sensitive as CT for detecting hemorrhage and is more sensitive for detecting an underlying structural etiology, but the rapidity, availability, and ease of interpretation of CT favor its initial use. Contrast-enhanced head CT scan may show evidence of persistent hemorrhage at the time of presentation, a sign associated with poor prognosis. 71 Urgent catheter angiography is required whenever aneurysmal subarachnoid hemorrhage is possible, such as in cases with a large amount of subarachnoid blood, and should be considered for all patients without a clear etiology who would be surgical candidates. Early MRI may be indicated if a structural etiology is suspected, but findings are often obscured by the hemorrhage, and a scan delayed by 4 to 8 weeks may provide more useful information if urgent diagnosis is unnecessary. MRI is useful in diagnosing cavernous malformations and may suggest congophilic amyloid angiopathy.
Treatment is generally supportive, although surgical intervention is indicated in some cases. Severe hypertension is common after intracerebral hemorrhage, in part because it is a response to elevated intracranial pressure and brain injury. No clinical studies are available for determining optimal blood pressure control after intracerebral hemorrhage. Theoretically, persistent hypertension could increase the risk of ongoing hemorrhage, but antihypertensive treatment may reduce blood flow to ischemic brain surrounding a hematoma or reduce cerebral perfusion pressure. 72 Consensus guidelines have recommended antihypertensive medications for systolic blood pressure greater than 180 mmHg or diastolic blood pressure greater than 105 mmHg, and fluids or pressors for systolic blood pressure less than 90 mmHg, but these thresholds for treatment are frequently debated. 61 Increased intracranial pressure may lead to coma and is treated with extraventricular drainage, osmotherapy, or hyperventilation.
Surgical evacuation of primary intracerebral hemorrhages is commonly performed when there is posterior fossa hemorrhage with a risk of brainstem compression or when there is evolving neurological deterioration in patients with lobar hemorrhages and other prognostic signs are favorable. A large, international trial randomized 1,033 subjects with supratentorial hemorrhage to receive early surgical evacuation of the hematoma or initial conservative treatment followed by surgical evacuation only if it was necessitated by neurological deterioration. There was a favorable outcome at 6 months in 26 percent of those allocated to early surgery as compared with 24 percent in those allocated to initial conservative treatment ( P = 0.89). In subgroup analysis, it appeared that early surgery was more effective than conservative therapy when the hematoma was 1 cm or less from the cortical surface. Additional trials will be needed to resolve the issue of early surgical benefit for superficial hematomas. 73
After the acute period, aggressive treatment of hypertension is indicated. In addition to reducing cardiovascular disease and ischemic stroke, one study has shown that treating hypertension reduces the risk of intracerebral hemorrhage by more than 50 percent. 74

The term lacune was first introduced in 1843 by M. Durand-Fardel to describe small, subcortical areas lacking gray and white matter. These lesions were attributed to infarct and associated with particular clinical presentations by P. Marie and J. Ferrand more than 50 years later. In the 1950s C. Miller Fisher reintroduced the term into modern neurology. 75 In a rapid succession of articles, he described the clinical and pathological presentation, recognized the importance of hypertension as an etiology, and developed a theory of pathogenesis that survives today.
Less than 2 cm in diameter, lacunes are small infarcts that result from occlusion of small penetrating branches arising from large arteries ( Fig. 7-8 ). There is general agreement about the definition of lacune, but much argument about the interrelationship between lacunar infarcts, lacunar strokes (symptomatic lacunes), lacunar syndromes (symptom complexes often associated with lacunar strokes), and lacunar disease (lacunes due to intrinsic small-vessel changes). Arguments arise from imperfect correlations between these entities. First, not all lacunes produce lacunar strokes because some are silent. Second, lacunar syndromes are sometimes associated with large-vessel strokes. Third, lacunes are produced by intrinsic small-vessel disease and by other etiologies. These issues are discussed in greater detail later.

FIGURE 7-8 An acute right thalamocapsular lacunar stroke producing left sensorimotor syndrome. The lesion was hypodense in noncontrast head CT (A). With MRI, it was hyperintense on T2-weighted images (B), inconspicuous on T1-weighted images (C), and hyperintense on diffusion-weighted images (D).
More than 50 percent of lacunes are located in the basal ganglia and thalamus, with the remainder in the internal capsule, pons, cerebellum, and subcortical white matter. Approximately 20 to 30 percent of ischemic strokes are due to lacunes. In a recent study, 23 percent of randomly selected subjects 65 years or older had a lacune on MR scanning, and 89 percent of those with a lacune denied a history of stroke or transient ischemic attack. 76 These “silent” lacunes were associated with impairment in cognitive and functional tasks, suggesting that the overall clinical burden of lacunes may be greater than previously suspected.
Hypertension is an important risk factor for development of lacunes, ranking as the most important identified risk factor in multivariable models. 76, 77 However, the strength of the association may be no greater for lacunes than for other forms of ischemic stroke, 78 and hypertension is not always present. 79 Nonetheless, since other risk factors are less important for lacunes, eliminating hypertension would be expected to have a greater impact on the occurrence of lacunar infarction than other forms of ischemic stroke. Elevation in the level of serum creatinine is independently associated with lacunar infarction, perhaps because it is a marker for chronic end-organ damage from hypertension. 76
Diabetes mellitus is a risk factor for symptomatic lacunes, approximately doubling the risk. However, the influence of diabetes on lacunar stroke does not appear to differ from its effect on other ischemic stroke subtypes. 78 This is also true for cigarette smoking, which doubles the risk of all ischemic strokes, including lacunes. Carotid artery stenosis is associated with an increased risk of lacunar stroke, with more than twice the risk of a symptomatic lacune above a 50 percent or greater stenosis. 76 Cardiac disease is less common in patients with lacunes (20%) than in those with other ischemic stroke types (47%). 78
The etiology of lacunes has been argued bitterly. Some have suggested that the vast majority of lacunes is due to changes within small penetrating vessels, primarily because of chronic hypertension, 80 but others have argued that emboli to small vessels and intracranial atherosclerosis are responsible for a significant number of lesions. 79 Fisher produced much of the data supporting intrinsic small-vessel disease. He found degenerative changes in small vessels he termed lipohyalinosis and fibrinoid degeneration, characterized by layers of connective tissue within the vascular media, obstructing the lumen. 81 These changes were proximal to infarcts in some cases. Atherosclerosis at the origin appeared responsible for other infarcts. Fisher recognized that emboli may be responsible for some lacunes. 81 Animal models have shown that particles may embolize to small penetrating arteries, producing lacunes. 82
Risk factor profiles have been used to argue against hypertension and intrinsic small-vessel disease as the sole etiology of lacunes. 79 As discussed previously, the risks imparted by hypertension, diabetes, and cigarette smoking are similar for lacunes and for other forms of ischemic stroke, and carotid and cardiac disease appear to be independent risk factors for lacunes. However, carotid and cardiac disease are much more commonly associated with large-vessel infarctions than with lacunes. 83 The etiology of lacunes is almost certainly multifactorial. Intrinsic small-vessel disease may predominate, but emboli and intracranial atherosclerosis almost certainly account for a significant minority of cases.
Several classic presentations of lacunar strokes have been described, termed the lacunar syndromes. Pure motor hemiparesis is the most common, accounting for 45 percent of cases. Motor functions involving face, arm, and leg are impaired, but other neurological functions are spared. The appearance is different from that with cortical strokes, in which deficits in sensation or cognition often accompany motor changes. Pure motor hemiparesis is not always due to a lacune, with 10 to 20 percent of cases attributed to a cortical stroke. When a lacune is responsible, it is most often located in the posterior limb of the internal capsule or in the basis pontis, but any other site along the path of corticospinal fibers can produce the syndrome.
Sensorimotor syndrome is the second most common lacunar syndrome, accounting for 20 percent of cases. 78 Weakness and numbness are present in varying degrees, usually involving the face, arm, and leg. Other neurological deficits are absent. The syndrome is most commonly produced by a lacune involving the lateral thalamus and internal capsule, but 13 percent of cases are not due to lacunes. 78
Ataxic hemiparesis accounts for 10 to 18 percent of lacunar syndromes. 78, 84 In the affected limbs, pyramidal weakness is combined with elements normally attributed to cerebellar ataxia. Intention tremor, exaggerated rebound, and irregular rapid alternating movements are superimposed on ipsilateral weakness. The findings are highly suggestive of a lacunar stroke, with 95 percent attributable to lacunes. 78 Infarct locations are identical to those that cause pure motor hemiparesis.
Among patients with lacunar syndromes, 7 percent have a pure sensory stroke, 78, 84 characterized by impaired sensation without other accompanying neurological deficits. When the face, arm, and leg are involved, the lesion is nearly always a lacune in the contralateral thalamus. A lesion in the medial lemnicus in the midbrain or rostral pons may occasionally produce an identical syndrome. Pain and dysesthesia in the affected region may accompany the lesion acutely or may be delayed by weeks to months.
Many other lacunar syndromes have been described, including clumsy-hand dysarthria, hemiballism, and pure motor hemiparesis combined with various eye movement abnormalities. 81 Although lacunes occur more commonly in certain regions of the brain, they can occur anywhere, producing an infinite number of potential syndromes. 79 Even signs generally attributed to cortical lesions may be produced by lacunes, including aphasia, abulia, confusion, and neglect.
Prognosis for recovery after a lacunar stroke is generally more favorable than for ischemic strokes due to occlusion of major vessels. 85 Recurrent stroke and mortality rates are also lower than for other stroke subtypes. 86
Lacunar syndromes may be produced by cortical strokes or even by small hemorrhages. Also, some lacunes may be associated with carotid disease or intracranial atherosclerosis, and knowledge of this could alter treatment decisions. Therefore, diagnostic imaging has been recommended for all those presenting with lacunar syndromes. An immediate head CT scan will rule out hemorrhage as an etiology but may not distinguish lacunes from large-vessel infarctions. The absence of changes on a head CT scan delayed by more than 6 hours has been used as confirmatory evidence of lacunar infarction. MRI provides more definitive confirmation, and MR angiography may suggest intracranial atherosclerosis. For lacunar strokes in the internal carotid distribution, duplex ultrasonography or neck MR angiography should be performed because a stenosis proximal to the lacune would generally be considered symptomatic.
Acute therapy for lacunes has not been well studied. Most studies of ischemic stroke have not examined this subtype separately, except in post hoc analyses with numbers too small to find significant treatment effects. There has been much debate about whether lacunar strokes should be treated with tissue plasminogen activator (t-PA). Some have argued that the predominant etiology of lacunar syndromes is intrinsic small-vessel disease, which would not be expected to respond to thrombolytics, so that patients would be exposed to an increased risk of hemorrhage without a good chance for benefit. In the National Institute of Neurological Disorders and Stroke t-PA trial, t-PA was effective in the subgroup of patients judged to have small-vessel occlusive strokes prior to randomization. 85 In fact, absolute differences in favorable outcomes were greater for small-vessel strokes than for large-vessel occlusive and cardioembolic strokes, and differences in two indices reached statistical significance despite small numbers. Some have argued that the final stroke occurs suddenly, implying that acute thrombosis in a diseased small vessel may account for a lacunar infarction and that this explains the efficacy of thrombolysis. Others have argued that the correlation between lacunar stroke and lacunar syndrome is so poor that a diagnosis of nonthrombotic small-vessel occlusion cannot be made with accuracy and that all patients with stroke should receive t-PA because of its overall benefit in ischemic stroke. The American Heart Association guidelines do not recommend avoiding t-PA in lacunar syndromes. 87 Although treatment is not recommended for patients with mild neurological deficits, including isolated hemisensory deficits, a patient with a pure motor hemiparesis is considered an appropriate candidate for thrombolysis. Other acute treatment for lacunar strokes is supportive.
Aspirin appears to reduce the risk of subsequent ischemic strokes, regardless of etiology. Although differences in effectiveness in stroke subgroups have not been carefully studied, the Chinese Acute Stroke Trial found that aspirin’s effect on stroke recurrence was similar in patients with lacunar strokes and those with other ischemic stroke subtypes. 88 Clopidogrel and the combination of dipyridamole/aspirin are alternatives for secondary prevention in those who cannot tolerate aspirin or who have recurrent strokes despite aspirin. Control of hypertension reduces subsequent ischemic stroke risk, and risk reduction may be even greater for lacunes. Treatment of isolated systolic hypertension in elderly patients halved the risk of lacunar stroke, a more dramatic effect than that seen for other ischemic strokes. 74
A lacune was considered evidence that a carotid stenosis was symptomatic in previous studies. 89 Whether the risk–benefit ratio for endarterectomy in patients with lacunes is different from that in those with large-vessel strokes is not known.

With improvements in head imaging, changes in the white matter surrounding the lateral ventricles were recognized frequently in the elderly, prompting Vladimir Hachinski to coin the term leukoaraiosis for the finding. Head CT shows a periventricular mantle of hypodensity, often most profound at the frontal and occipital horns, which is hyperintense on T2-weighted MRI ( Fig. 7-9 ). Age is the most important risk factor, with 96 percent of those older than 65 years showing at least some evidence of such change. 90 Clinically, the changes are most frequently associated with insidious declines in cognitive and motor performance, particularly on tests that depend on reaction time and speed. 90, 91

FIGURE 7-9 Imaging findings of periventricular white matter disease, with hypodensity on head CT (A) and T2-weighted hyperintensities on MRI (B).
The etiology of white matter lesions is believed to be related to several distinct pathological processes, including hypoperfusion injury, cerebral amyloid angiopathy, dilated perivascular spaces, axonal loss, astrocytic gliosis, and loss of ependymal integrity with resulting cerebrospinal fluid extravasation. Lesions contiguous with the ventricles show fewer histological and molecular markers of ischemia than lesions in the deep subcortical areas, where they resemble areas of “incomplete” infarction on pathological examination. 92 Loss of vasomotor reactivity and autoregulation due to small-vessel vasculopathy is hypothesized to be a frequent cause of the ischemic changes. 93 Leukoaraiosis may be an important clinical indicator of end-organ injury from hypertension, integrating information about cumulative exposure to high blood pressure as well as susceptibility to injury. Individuals with white matter lesions in the brain are at high risk of incident stroke and other clinical cardiovascular events. 94 White matter burden is also one of strongest predictors of incident brain infarction defined by serial cranial MRI. 95

In 1894, Otto Binswanger described a form of early dementia distinguishable from syphilitic general paresis and large-vessel vascular dementia by its subcortical involvement. 96 Hypertension is an important risk factor for the disease, existing in 94 percent of cases in one series. 97 Diabetes also appears to be a risk factor. Multiple lacunar infarctions are combined with diffuse and focal myelin loss, particularly in periventricular areas, producing a pathological picture consistent with a combination of multiple lacunes and periventricular white matter disease. Patients present with a stepwise and gradual progression of motor, cognitive, and behavioral deficits, generally over 5 to 10 years. Periods of symptom stabilization or improvement may occur. Imaging studies show patchy periventricular white matter hypodensity or hyperintensity on T2-weighted MRI, with superimposed focal, subcortical lesions consistent with lacunes. Aspirin and control of hypertension are expected to slow progression, but this has not been shown systematically.

Cerebral autosomal-dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) is a recently recognized dementing illness caused by mutations in the NOTCH3 gene, which encodes a transmembrane receptor protein of unclear function. 98 Clinical presentation is similar to that of Binswanger’s disease, with stepwise decline in cognitive and motor functions. However, onset is earlier, beginning at 30 to 50 years of age, and it is often preceded by migraines with aura. Hypertension and diabetes are not associated. Head imaging shows multiple lacunes superimposed on periventricular white matter disease. 99 Degeneration of vascular smooth muscle cells and granular deposits characterize vessels in the brain and in other regions. Involvement of the dermis allows confirmation by skin biopsy. No treatment is available.

The first comprehensive description of carotid occlusion and stroke is attributed to J. R. Hunt, who in 1914 described a patient with decreased carotid pulsation contralateral to a hemiparesis. 100 Autopsy confirmed a hemispheric infarct and showed patent intracranial vessels. With the advent of angiography and surgical exploration, internal carotid artery occlusion with recent thrombus was confirmed in the 1940s.
The importance of internal carotid artery stenosis as a cause of stroke is unclear because it is difficult to definitively attribute a stroke to the stenosis. The Stroke Data Bank of the National Institute of Neurological Disorders and Stroke classified 69 of 1,273 ischemic stroke cases as atherothrombotic and 41 as embolic due to severe carotid stenosis or occlusion. 101, 102 Based on these numbers, approximately 9 percent of ischemic strokes are due to internal carotid stenosis or occlusion. In the Atherosclerosis Risk in Communities Study, 34 percent of randomly selected subjects aged 45 to 64 years had evidence of carotid plaque on ultrasonography. 103 An asymptomatic carotid stenosis of more than 60 percent is found in approximately 5 percent of 65-year-olds and increases with age. 104
Hypertension is an important risk factor for carotid stenosis. In the Framingham study, systolic hypertension was a powerful predictor of subsequent carotid stenosis, with twice the odds for each 20-mmHg increase in systolic blood pressure. 105 Systolic pressure is also a predictor of progression in patients with asymptomatic stenoses. 106 Cigarette smoking, high serum cholesterol level, and increased homocysteine are other risk factors for carotid stenosis. 105
Internal carotid artery stenosis is produced by atherosclerosis just distal to the common carotid bifurcation. The pathophysiology of carotid artery stenosis is complex. Hypertension induces vascular remodeling, resulting in medial thickening, luminal narrowing, and impaired smooth muscle relaxation. 8 These changes are concentrated in areas of nonlaminar flow, such as the common carotid bifurcation. Atherosclerotic plaques are thought to develop in these areas as a response to injury produced by hypertension, blood-flow abnormalities, lipids, and possibly infection. 10 This initiating injury induces endothelial cell expression of cell adhesion molecules that mediate local extravasation of mononuclear cells. Vessel wall inflammation results, with foamy, lipid-laden macrophages and T lymphocytes. Chronic injury leads to intimal hyperplasia and formation of complex plaques that may include a lipid core. When a plaque ruptures into the vessel lumen, thrombosis is induced, which may produce local occlusion, distal embolus, or, after organization, progressive luminal stenosis. Shear forces associated with a severe stenosis may induce platelet activation and thrombus formation without plaque rupture.
Clinically, symptomatic patients present with large-vessel ischemic strokes or transient ischemic attacks in the distribution of the ophthalmic, middle, or anterior cerebral artery. Transient monocular blindness (amaurosis fugax), weakness, numbness, aphasia, or neglect may occur, depending on the affected region of the anterior circulation. Border-zone ischemia due to distal hypoperfusion in the anterior and middle cerebral artery territories presents with proximal upper and lower extremity weakness and numbness (“man-in-the-barrel” syndrome) and may indicate a critical stenosis or occlusion with inadequate collateral blood-flow. Artery-to-artery emboli classically appear as cortical wedge-shaped infarcts, indistinguishable from emboli from other sources. Lacunar infarcts, often attributed to intrinsic small-vessel disease, probably represent embolic events from carotid artery stenoses in some instances because endarterectomy appears to reduce the risk of ipsilateral lacune. 107
A cervical bruit may be a sign of carotid stenosis, but it is absent in 25 to 40 percent of cases later confirmed to have a greater than 70 percent stenosis and is present in 25 to 40 percent of those without a severe stenosis. 108 Therefore, carotid imaging studies are generally indicated for patients with anterior circulation ischemic strokes or transient ischemic attacks.
Carotid Doppler ultrasonography, neck MR angiography, or catheter angiography is required for the determination of whether an internal carotid stenosis is present. All three tests are good predictors of the degree of stenosis determined at surgery. 109 Catheter angiography is considered the gold standard but is limited to producing two-dimensional projections and carries a 1 percent risk of stroke. Therefore, it is generally preferable to perform carotid Doppler ultrasonography or neck MR angiography first. Carotid Doppler ultrasonography is more widely available and usually less expensive. MR angiography provides three-dimensional views and can incorporate the intracranial vasculature. When the findings on either Doppler ultrasonography or MR angiography are positive, performing the other study provides confirmation of degree of stenosis and may obviate the need for catheter angiography.
From the perspective of society, it does not appear cost-effective to screen patients for asymptomatic carotid stenoses. 104 Because of the limited benefit of surgery and the costs of carotid ultrasonography, the pretest probability of finding a stenosis must be greater than 40 percent before it is cost-effective to screen those without symptoms. Carotid ultrasonography probably is not cost-effective even in asymptomatic elderly patients with bruits because the pretest probability of finding a high-grade stenosis is only approximately 15 percent, given the 5 percent prevalence of a high-grade stenosis in those older than 65 years and a threefold increased likelihood of stenosis in those with a bruit. 108
Aspirin has been shown to reduce risk of stroke and myocardial infarction in patients with ischemic stroke or transient ischemic attacks, with a risk reduction of about 20 percent. 110 Some clinicians use anticoagulation with heparin or warfarin to treat symptomatic carotid stenosis, but there are no reliable data supporting this approach. Based on results in coronary artery disease, a process with similar pathophysiology, and on overall risk reduction of ischemic stroke, treatment with cholesterol-lowering agents may be of benefit even in those without hypercholesterolemia.
Surgical removal of the obstructing plaque by endarterectomy is the established standard of therapy for symptomatic patients with carotid artery stenosis of at least 70 percent ( Table 7-4 ). 89, 111 Endarterectomy also reduces recurrent stroke rates in patients with symptomatic carotid stenoses of 50 to 69 percent, but the number needed to treat (NNT) to prevent one recurrent stroke is considerably higher in this group when compared with those with stenoses exceeding 70 percent (∼15.4 to prevent one stroke over 5 years in those with 50% to 69% stenosis versus ∼5.8 to prevent one stroke over 2 years in those with 70% to 99% stenosis). 107 Endarterectomy is not beneficial in patients with stenosis less than 50 percent and is generally impractical in those with carotid artery occlusion. The risk of stroke with medical therapy is greater in those with cerebral events compared with ocular events, with plaque surface irregularity consistent with ulceration, with a symptomatic event within 2 weeks of presentation, and with greater degrees of stenosis. 112 The risk of surgery is greater in females, in those with severe hypertension, and in those with peripheral vascular disease. These prognostic factors may be useful in fine-tuning patient selection for endarterectomy.

TABLE 7-4 Yearly Ipsilateral Stroke Rates With Carotid Artery Stenosis Based on 5-Year Follow-up
For patients with asymptomatic carotid artery stenosis, endarterectomy also prevents stroke when there is stenosis of at least 60 percent as assessed by carotid ultrasonography, but again the number needed to treat to prevent one stroke over 5 years remains large (∼20), and, therefore, current guidelines recommend consideration of endarterectomy for asymptomatic stenosis for patients with a surgical risk less than 3 percent and life expectancy of at least 5 years. 32, 113
Endovascular angioplasty and stenting are an evolving approach to treatment of carotid stenosis, and stenting has been shown to be not inferior to endarterectomy in patients with both symptomatic and asymptomatic stenoses who have comorbidities associated with high surgical risk during endarterectomy. Large-scale trials comparing endarterectomy and stenting in more representative patient populations are ongoing.
High-grade stenoses of the carotid arteries could impair distal cerebral blood-flow in some patients without adequate collateral supply. When endarterectomy is planned for the near future, physicians will routinely allow higher than normal blood pressures to reduce the risk of symptomatic hypoperfusion. Whether such a practice is truly justified when compared to the risk of plaque rupture associated with hypertension has not been addressed in clinical trials.

Atherosclerosis involving the large intracranial vessels causes about 8 percent of ischemic strokes. 114 African Americans, Hispanics, and Asians have a higher prevalence of intracranial atherosclerosis, and relatively low prevalence of extracranial carotid artery stenosis compared with whites. 114, 115 Extracranial carotid atherosclerosis is associated with a higher prevalence of peripheral vascular and coronary artery disease, but intracranial atherosclerosis is not. 114 Given racial and risk factor distribution differences, it seems appropriate to consider intracranial atherosclerosis an entity distinct from carotid artery disease rather than as an additional manifestation of widespread atherosclerotic changes.
Hypertension is an important risk factor for intracranial atherosclerosis, with a two- to threefold higher risk of disease in those with a history of hypertension. 116 Smoking may be the most important risk factor, with a 50 percent increase in odds of disease for every 10 years of smoking. Diabetics have about three times the risk of developing intracranial atherosclerosis. Hypercholesterolemia also increases risk, but probably to a lesser degree. The relative contribution of these factors to intracranial atherosclerosis as opposed to other stroke subtypes is unclear. Distribution of known risk factors probably accounts for some of the racial differences. 114
There are intriguing differences in the pathophysiology of intracranial atherosclerosis and other forms of vascular disease. Intracranial arteries are less susceptible to hypercholesterolemia than are extracranial arteries, 117 and atherosclerotic plaque rupture appears to be less common. 118 Release of endothelial adhesion molecules is greater with intracranial atherosclerosis than in other ischemic stroke subtypes, suggesting that inflammation is particularly important in its pathogenesis. 119 There is no accepted unifying theory on the etiology of intracranial atherosclerosis.
Clinical presentation is characterized by large-vessel or penetrating artery ischemia. The middle cerebral artery is most commonly involved, followed in order by the basilar, intracranial internal carotid, anterior cerebral, and posterior cerebral arteries. 114 Thrombosis at the site of the stenosis may lead to hypoperfusion in the entire distal territory or artery-to-artery embolus indistinguishable from events caused by extracranial carotid artery stenosis or cardiac embolus. Basilar thrombosis may result from underlying atherosclerosis in the basilar or vertebral arteries or after cardiac embolus. It is a life-threatening, often delayed diagnosis characterized by coma, quadriplegia, and cranial nerve findings. Involvement of the origin of penetrating small vessels may produce lacunar infarctions. Presentation with transient ischemic attack prior to infarction is more common with intracranial atherosclerosis than with other stroke subtypes.
Intracranial MR angiography may reveal narrowing or occlusion of large vessels. However, artifacts may suggest a stenosis where none is present, and sensitivity is low for medium-sized and smaller vessels. Transcranial Doppler ultrasonography shows increased blood flow velocities in large stenotic vessels. Its sensitivity and specificity are also low, so it may be most useful as an adjunct to MR angiography. Catheter angiography is the gold standard for establishing the diagnosis, but it is associated with a 1 percent stroke risk that is probably even higher in the population being evaluated for intracranial atherosclerosis. Given the risk of angiography, it is only justified if results will alter treatment decisions.
Prognosis in symptomatic patients is poor. Stenosis generally becomes more severe with time, but regression in some segments may occur. In the largest randomized trial of treatment for symptomatic intracranial atherosclerosis, the Warfarin-Aspirin Symptomatic Intracranial Disease (WASID) trial, 569 subjects were randomized to aspirin (1,300 mg/day) and warfarin (target international normalized ratio, or INR, of 2.0 to 3.0). 120 The trial was stopped early because of safety concerns among subjects randomized to warfarin, and there was no significant difference in the primary endpoint of stroke, brain hemorrhage, or vascular death ( P = 0.83). The 2-year rates of ischemic stroke were 19.7 percent in the aspirin group and 17.2 percent in the warfarin group ( P = 0.29), indicating that intracranial atherosclerosis is a high-risk condition for recurrent stroke. For comparison, the 2-year risk of ischemic stroke was 26 percent among individuals with symptomatic carotid stenosis between 70 and 99 percent assigned to medical therapy in the North American Symptomatic Carotid Endarterectomy Trial and approximately 11 percent among those assigned to aspirin in the Warfarin-Asprin Recurrent Stroke Study (WARSS), in which more than 75 percent of subjects had a lacunar or cryptogenic etiology for stroke. 121 In part because of the high recurrent stroke risk, intracranial angioplasty and stenting are sometimes considered for patients who have failed medical therapy because of recurrent ischemic symptoms. Given the current lack of efficacy data, such an approach remains investigational, but historical case-series data are encouraging, and as techniques and devices improve, the indications for endovascular treatment may broaden. 122
Patients with intracranial atherosclerosis also have a theoretical risk of hypoperfusion distal to the stenosis when blood pressure is lowered. Since these lesions are less commonly corrected compared with those in the carotid artery, some physicians may be less aggressive about treating hypertension in these patients. Newer imaging modalities provide a way to assess whether there is sufficient vascular dilatory capacity distal to a stenosis to preserve blood flow in response to systemic changes in arterial resistance, but the integration of this information into clinical practice has not been fully resolved. 123 There is currently no evidence to justify higher long-term blood pressure thresholds in patients with intracranial atherosclerosis or to support the belief that lower blood pressures could in fact increase the risk of infarction distal to a stenosis. 124 It would be prudent, however, to consider lower initial doses, slower titration schedules, and more frequent monitoring for orthostasis and other adverse effects of antihypertensive therapy in such patients.

The aortic arch has only recently come to be appreciated as a source of emboli to the brain. 125 An autopsy study found a 28 percent prevalence of ulcerated plaque in the aortic arch of patients with ischemic stroke compared with a 5 percent prevalence in a control population. 125 Transesophageal echocardiograms show evidence of a thickened aortic arch in 14 percent of stroke patients and 2 percent of control subjects. 126 In those with no other identified etiology, a thickened aortic arch is present in 28 percent. Aortic arch atherosclerosis is more strongly associated with peripheral vascular disease than with carotid stenosis. 127 Epidemiological studies have been small, and only cigarette smoking has been identified as an important risk factor. Hypertension, diabetes, and hypercholesterolemia may be risk factors, but this has not been confirmed.
Strokes and transient ischemic attacks produced by aortic atherosclerosis are identical to those produced by cardiac sources of emboli. Large-vessel territories are generally affected, producing weakness and numbness in similar distributions or cortical signs, such as aphasia and neglect.
Stroke patients with atherosclerotic plaque 4 mm thick or larger in the aortic arch have a fourfold increase in the risk of recurrence after correcting for other risk factors. 127 Optimal treatment is not known but should include an antiplatelet agent or anticoagulation. Aortic endarterectomy has been performed in some patients who have failed medical therapy, but it has not been systematically studied.

Hypertension increases the risk of myocardial infarction and atrial fibrillation. These diseases are associated with increased stroke risk from cardiac embolus, as discussed in Chapter 5 .

There is evidence from multiple cohort studies that cardiovascular risk factors, including hypertension, are risk factors for the development of dementia and cognitive impairment. 128 - 130 The biological basis for these associations remains unresolved. Although there are some data to support a direct association between hypertension and Alzheimer pathology, 131 there is increasing recognition that most individuals with dementia have a mixture of neurodegenerative and vascular pathology. 132 The association between hypertension and dementia is likely to be mediated in part by the accumulation of subclinical vascular injury in the brain, including infarct and leukoaraiosis, that results in the interruption of cognitive networks. Whether this process is simply additive to the cognitive effects of Alzheimer’s pathology or whether there is synergism between the two processes remains unresolved.
There is currently no convincing evidence that treatment of hypertension will make a large impact on the occurrence of dementia outside of its established benefits for stroke prevention. Although one large primary prevention trial using the calcium blocker nitrendipine found that the risk of dementia was reduced by 50 percent in those receiving active therapy, the finding has not been confirmed in other trials. 133 - 135 In the PROGRESS trial, which included subjects exclusively with a history of stroke, active therapy with perindopril and indapamide reduced the risk of dementia overall, but the benefit was not statistically significant when the analysis was restricted to those without recurrent stroke. 136
Although it hardly seems necessary to define additional benefits of treating hypertension, the recognition that cognitive decline may be an important manifestation of end-organ injury from hypertension has important implications for testing new treatments for cerebrovascular disease, assessing risk of stroke, and encouraging adherence to treatment. 137 If hypertension therapy is proven to prevent dementia among those without stroke, the cost–benefit ratio for more aggressive screening and therapy could also be substantially improved, an important issue, given that the elderly, who are at highest risk of dementia, are also the least likely to have their hypertension adequately treated. Some have suggested that aggressive blood pressure reduction could worsen cognition, particularly among those with loss of cerebral autoregulation due to small-vessel arteriopathy. 138 Therefore, the benefits of blood pressure therapy for cognition will need better definition in order to optimize treatment regimens.

In diabetics, hypertension appears to be a risk factor for the development of peripheral neuropathy. One recent prospective study found a fourfold increased risk of distal symmetrical sensorimotor polyneuropathy with hypertension among insulin-dependent diabetics. 139 Hypertension probably increases the risk of peripheral neuropathy in non–insulin-dependent diabetics as well, but not all studies have found an effect. Hypertension is thought to accelerate small-vessel changes in diabetics, producing microvascular ischemic injury to peripheral nerves. In the absence of diabetes, hypertension has not been demonstrated as a risk factor for peripheral neuropathy.

Hypertensive encephalopathy is one of several forms of reversible posterior leukoencephalopathy, a syndrome also encompassing other etiologies, including renal failure, immunosuppressive therapy, and eclampsia. 140
Hypertensive encephalopathy is an uncommon disease. There are no recent measurements of its incidence, but it is thought to have declined with greater use of antihypertensives. It tends to occur with a sudden elevation in blood pressure rather than with chronic hypertension. A number of medical conditions are known precipitants. Hyperadrenergic states may be responsible, including pheochromocytoma, tyramine ingestion with monoamine oxidase inhibitors, abrupt antihypertensive discontinuation, lower gastrointestinal irritation in paraplegic patients, and stimulant medications. 141 Structural precipitants include aortic coarctation and renal artery stenosis. Acute or chronic renal failure is another cause, probably through volume overload in addition to hypertension, and human recombinant erythropoietin may be a precipitant. 142 In patients in the postoperative period after endarterectomy, changes ipsilateral to the surgery may be identical to those seen with hypertensive encephalopathy, even in the absence of blood pressure elevation, probably because vessels compensate for chronic hypoperfusion above a severe stenosis and sudden return of blood flow produces relative hypertension.
Hypertensive encephalopathy is associated with vasogenic cerebral edema, particularly severe in the posterior regions of the cerebral hemispheres, which is sometimes sufficient to result in herniation. The pathophysiology linking hypertension and cerebral edema has been argued. At mean arterial pressures greater than 120 to 170 mmHg, cerebral blood-flow increases linearly with blood pressure, and some have argued that this is the threshold for hypertensive encephalopathy, when a “breakthrough of autoregulation” occurs. 140, 143 Angiotensin II may contribute to the formation of edema by increasing cerebrovascular permeability through oxygen free radicals. 144 A predilection toward involvement of the posterior hemispheres may be due to differential vascular innervation by the sympathetic nervous system. 143
Hypertensive encephalopathy is a neurological emergency that can lead to death if untreated. Diagnosis may be delayed when the connection between acute neurological dysfunction and hypertension is not obvious. High blood pressure may be attributed to an underlying neurological condition or agitation rather than identified as the causative agent. Headache is a common early complaint, sometimes accompanied by nausea and vomiting. Confusion with either agitation or lethargy may proceed to obtundation and coma if the process is untreated. Visual disturbance is frequent due to involvement of the retina and occipital lobes, with papilledema and subjective blurred vision, hemianopia, or cortical blindness. 140 Other cortical deficits may occur, including neglect, aphasia, and weakness. Focal or generalized seizures may complicate the course.
Head imaging should be performed to exclude hemorrhage or a structural etiology for both the encephalopathy and the hypertension. Since increased intracranial pressure can result in severe hypertension, which may be required to maintain cerebral perfusion, an urgent study is necessary. Head CT may show hypodensity in subcortical white matter, often most obvious in the occipital lobes ( Fig. 7-10 ). MRI can be dramatic, with multifocal T2-weighted hyperintensities particularly apparent in fluid-attenuated inversion recovery images. These changes are distinguished from infarcts by sparing of the cortex and absence of reduced diffusion, as expected with vasogenic edema. 145 Because cerebral edema may be severe, lumbar puncture should be avoided.

FIGURE 7-10 A patient with eclampsia. Typical findings in hypertensive encephalopathy are identical and include normal or subtle hypodensity on CT (A), subcortical hyperintensities on MRI fluid-attenuated inversion recovery (FLAIR) (B), enhancement with gadolinium on T1-weighted images (C), and no abnormality on diffusion-weighted MRI (D).
Once a structural etiology has been excluded, treatment of hypertension must be initiated. Target blood pressures are tailored to individual patients, with the goal of returning patients to their recent baseline. For patients without a history of hypertension, normal blood pressure parameters are appropriate, but for those with chronic hypertension, an abrupt return to 140/90 mmHg may result in hypoperfusion owing to chronic vascular compensatory changes. Close observation and intravenous antihypertensives are generally indicated. Nitroprusside is a good choice because of its effectiveness, rapid onset, and ease of adjustment. It produces some cerebral vasodilation, but there is no clinical evidence that this elevates the risk of herniation. Intravenous ACE inhibitors are also effective and easy to titrate and may have less-profound effects on cerebral vessels. Further, animal studies suggest that these agents may have more dramatic effects on cerebral edema by blocking increased vessel permeability from angiotensin II. 144 Anticonvulsant therapy may also be indicated. The underlying cause of the hypertensive episode should be sought.
Prognosis in treated patients is generally excellent. Neurological deficits usually recover completely within 2 weeks.

Eclampsia is a form of reversible posterior leukoencephalopathy. Occurring during the second half of pregnancy or the puerperium, eclampsia presents with proteinuria and clinical and imaging manifestations identical to hypertensive encephalopathy. Hypertension may not be severe, so additional effects on brain endothelial cell permeability are probably important. There is evidence of generalized endothelial cell dysfunction with abnormal vascular reactivity. 146 An underlying inflammatory response may be causative, but other potential etiologies have also been hypothesized.
Cerebral venous sinus thrombosis is another complication of pregnancy and delivery and can present with findings similar to those seen with eclampsia. MRI and venography are usually adequate to distinguish the two diseases, showing obstructed venous sinuses or ischemia with cytotoxic edema on diffusion-weighted images in cerebral venous sinus thrombosis.
Treatment includes delivery of the fetus and intravenous magnesium. 147 Other antihypertensive medications and anticonvulsants can also be used. Prognosis is good if treatment is initiated quickly.

Several immunosuppressive agents produce a reversible posterior leukoencephalopathy identical to hypertensive encephalopathy. Cyclosporine is the most commonly identified, and it may produce neurological symptoms at therapeutic levels and without evident hypertension. Tacrolimus, FK-506, interferon-alpha, cytarabine, and fludarabine have also been associated. 140 An alteration in the permeability of cerebral endothelial cells has been postulated. Lowering blood pressure and discontinuing immunosuppression generally reverses the process.


1 Comroe JH. Exploring the Heart: Discoveries in Heart Disease and High Blood Pressure. New York: Norton, 1983.
2 Chobanian AV, Bakris GL, Black HR, et al. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA . 2003;289:2560.
3 Vasan RS, Beiser A, Seshadri S, et al. Residual lifetime risk for developing hypertension in middle-aged women and men: the Framingham Heart Study. JAMA . 2002;287:1003.
4 MacMahon S, Peto R, Cutler J, et al. Blood pressure, stroke, and coronary heart disease. Part 1, Prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet . 1990;335:765.
5 The fifth report of the Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure (JNC V). Arch Intern Med . 1993;153:154.
6 Lewington S, Clarke R, Qizilbash N, et al. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet . 2002;360:1903.
7 Johansson BB. Hypertension mechanisms causing stroke. Clin Exp Pharmacol Physiol . 1999;26:563.
8 Gibbons GH, Dzau VJ. The emerging concept of vascular remodeling. N Engl J Med . 1994;330:1431.
9 Heistad DD, Baumbach GL. Cerebral vascular changes during chronic hypertension: good guys and bad guys. J Hypertens Suppl . 1992;10:S71.
10 Ross R. Atherosclerosis-an inflammatory disease. N Engl J Med . 1999;340:115.
11 Rossi G, Rossi A, Sacchetto A, et al. Hypertensive cerebrovascular disease and the renin-angiotensin system. Stroke . 1995;26:1700.
12 Zacchigna L, Vecchione C, Notte A, et al. Emilin1 links TGF-beta maturation to blood pressure homeostasis. Cell . 2006;124:929.
13 Pickering TG, Hall JE, Appel LJ, et al. Recommendations for blood pressure measurement in humans and experimental animals: part 1: blood pressure measurement in humans: a statement for professionals from the Subcommittee of Professional and Public Education of the American Heart Association Council on High Blood Pressure Research. Circulation . 2005;111:697.
14 Akpunonu BE, Mulrow PJ, Hoffman EA. Secondary hypertension: evaluation and treatment. Dis Mon . 1996;42:609.
15 Appel LJ, Moore TJ, Obarzanek E, et al. A clinical trial of the effects of dietary patterns on blood pressure. DASH Collaborative Research Group. N Engl J Med . 1997;336:1117.
16 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 hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA . 2002;288:2981.
17 Sacco RL. Risk factors and outcomes for ischemic stroke. Neurology . 1995;45:S10.
18 Gorelick PB. Stroke prevention. An opportunity for efficient utilization of health care resources during the coming decade. Stroke . 1994;25:220.
19 Collins R, Peto R, MacMahon S, et al. Blood pressure, stroke, and coronary heart disease. Part 2, Short-term reductions in blood pressure: overview of randomised drug trials in their epidemiological context. Lancet . 1990;335:827.
20 Shep Cooperative Research Group. Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension. Final results of the Systolic Hypertension in the Elderly Program (SHEP). JAMA . 1991;265:3255.
21 Staessen JA, Fagard R, Thijs L, et al. Randomised double-blind comparison of placebo and active treatment for older patients with isolated systolic hypertension. The Systolic Hypertension in Europe (Syst-Eur) Trial Investigators. Lancet . 1997;350:757.
22 Gueyffier F, Bulpitt C, Boissel JP, et al. Antihypertensive drugs in very old people: a subgroup meta-analysis of randomised controlled trials. INDANA Group. Lancet . 1999;353:793.
23 Thom TJ. Stroke mortality trends. An international perspective. Ann Epidemiol . 1993;3:509.
24 Higgins M, Thom T. Trends in stroke risk factors in the United States. Ann Epidemiol . 1993;3:550.
25 Meyer JS, Shimazu K, Fukuuchi Y, et al. Impaired neurogenic cerebrovascular control and dysautoregulation after stroke. Stroke . 1973;4:169.
26 Schwarz S, Georgiadis D, Aschoff A, et al. Effects of induced hypertension on intracranial pressure and flow velocities of the middle cerebral arteries in patients with large hemispheric stroke. Stroke . 2002;33:998.
27 Ahmed N, Nasman P, Wahlgren NG. Effect of intravenous nimodipine on blood pressure and outcome after acute stroke. Stroke . 2000;31:1250.
28 Britton M, de Faire U, Helmers C. Hazards of therapy for excessive hypertension in acute stroke. Acta Med Scand . 1980;207:253.
29 Adams HPJr, Adams RJ, Brott T, et al. Guidelines for the early management of patients with ischemic stroke: a scientific statement from the Stroke Council of the American Stroke Association. Stroke . 2003;34:1056.
30 Chamorro A, Vila N, Ascaso C, et al. Blood pressure and functional recovery in acute ischemic stroke. Stroke . 1998;29:1850.
31 PROGRESS Collaborative Group. Randomised trial of a perindopril-based blood-pressure-lowering regimen among 6,105 individuals with previous stroke or transient ischaemic attack. Lancet . 2001;358:1033.
32 Sacco RL, Adams R, Albers G, et al. Guidelines for prevention of stroke in patients with ischemic stroke or transient ischemic attack: a statement for healthcare professionals from the American Heart Association/American Stroke Association Council on Stroke: co-sponsored by the Council on Cardiovascular Radiology and Intervention: the American Academy of Neurology affirms the value of this guideline. Stroke . 2006;37:577.
33 Dahlof B, Devereux RB, Kjeldsen SE, et al. Cardiovascular morbidity and mortality in the Losartan Intervention For Endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol. Lancet . 2002;359:995.
34 Ovbiagele B, Saver JL, Fredieu A, et al. PROTECT: a coordinated stroke treatment program to prevent recurrent thromboembolic events. Neurology . 2004;63:1217.
35 Taylor CL, Yuan Z, Selman WR, et al. Cerebral arterial aneurysm formation and rupture in 20,767 elderly patients: hypertension and other risk factors. J Neurosurg . 1995;83:812.
36 Teunissen LL, Rinkel GJ, Algra A, et al. Risk factors for subarachnoid hemorrhage: a systematic review. Stroke . 1996;27:544.
37 Longstreth WTJr, Nelson LM, Koepsell TD, et al. Cigarette smoking, alcohol use, and subarachnoid hemorrhage. Stroke . 1992;23:1242.
38 Johnston SC, Colford JM, Gress DR. Oral contraceptives and the risk of subarachnoid hemorrhage: a meta-analysis. Neurology . 1998;51:411.
39 Schievink WI. Genetics and aneurysm formation. Neurosurg Clin North Am . 1998;9:485.
40 Magnetic Resonance Angiography in Relatives of Patients with Subarachnoid Hemorrhage Study Group. Risks and benefits of screening for intracranial aneurysms in first-degree relatives of patients with sporadic subarachnoid hemorrhage. N Engl J Med . 1999;341:1344.
41 Broderick JP, Brott T, Tomsick T, et al. The risk of subarachnoid and intracerebral hemorrhages in blacks as compared with whites. N Engl J Med . 1992;326:733.
42 Conway JE, Hutchins GM, Tamargo RJ. Marfan syndrome is not associated with intracranial aneurysms. Stroke . 1999;30:1632.
43 Johnston SC, Halbach VV, Smith WS, et al. Rapid development of giant fusiform cerebral aneurysms in angiographically normal vessels. Neurology . 1998;50:1163.
44 Drake CG, Peerless SJ. Giant fusiform intracranial aneurysms: review of 120 patients treated surgically from 1965 to 1992. J Neurosurg . 1997;87:141.
45 Schievink WI. Intracranial aneurysms. N Engl J Med . 1997;336:28.
46 Rinkel GJ, Djibuti M, van Gijn J. Prevalence and risk of rupture of intracranial aneurysms: a systematic review. Stroke . 1998;29:251.
47 Wiebers DO, Torner JC, Meissner I. Impact of unruptured intracranial aneurysms on public health in the United States. Stroke . 1992;23:1416.
48 Day JW, Raskin NH. Thunderclap headache: symptom of unruptured cerebral aneurysm. Lancet . 1986;2:1247.
49 Atlas SW, Sheppard L, Goldberg HI, et al. Intracranial aneurysms: detection and characterization with MR angiography with use of an advanced postprocessing technique in a blinded-reader study. Radiology . 1997;203:807.
50 Wiebers DO, Whisnant JP, Huston J3rd, et al. Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet . 2003;362:103.
51 Johnston SC, Wilson CB, Halbach VV, et al. Endovascular and surgical treatment of unruptured cerebral aneurysms: comparison of risks. Ann Neurol . 2000;48:11.
52 Johnston SC, Gress DR, Kahn JG. Which unruptured cerebral aneurysms should be treated? A cost-utility analysis. Neurology . 1999;52:1806.
53 Johnston SC, Selvin S, Gress DR. The burden, trends, and demographics of mortality from subarachnoid hemorrhage. Neurology . 1998;50:1413.
54 Hop JW, Rinkel GJ, Algra A, et al. Case-fatality rates and functional outcome after subarachnoid hemorrhage: a systematic review. Stroke . 1997;28:660.
55 Taylor TN, Davis PH, Torner JC, et al. Lifetime cost of stroke in the United States. Stroke . 1996;27:1459.
56 Kassell NF, Torner JC, Haley ECJr, et al. The International Cooperative Study on the Timing of Aneurysm Surgery. Part 1: Overall management results. J Neurosurg . 1990;73:18.
57 Germanson TP, Lanzino G, Kongable GL, et al. Risk classification after aneurysmal subarachnoid hemorrhage. Surg Neurol . 1998;49:155.
58 Mayberg MR, Batjer HH, Dacey R, et al. Guidelines for the management of aneurysmal subarachnoid hemorrhage. A statement for healthcare professionals from a special writing group of the Stroke Council, American Heart Association. Stroke . 1994;25:2315.
59 Grosse DG, Straiton J, McDonald I, et al. Use of transcranial Doppler sonography to predict development of a delayed ischemic deficit after subarachnoid hemorrhage. J Neurosurg . 1993;78:183.
60 Thrift AG, Donnan GA, McNeil JJ. Epidemiology of intracerebral hemorrhage. Epidemiol Rev . 1995;17:361.
61 Broderick JP, Adams HPJr, Barsan W, et al. Guidelines for the management of spontaneous intracerebral hemorrhage: a statement for healthcare professionals from a special writing group of the Stroke Council, American Heart Association. Stroke . 1999;30:905.
62 He J, Whelton PK, Vu B, et al. Aspirin and risk of hemorrhagic stroke: a meta-analysis of randomized controlled trials. JAMA . 1998;280:1930.
63 Kase CS, Mohr JP, Caplan LR. Intracerebral hemorrhage. In: Barnett HJM, Mohr JP, Stein BM, Yatsu FM, editors. Stroke Pathophysiology, Diagnosis, and Management . 3rd Ed. New York: Churchill Livingstone; 1998:649.
64 MacKenzie JM. Intracerebral haemorrhage. J Clin Pathol . 1996;49:360.
65 Rosenblum WI. The importance of fibrinoid necrosis as the cause of cerebral hemorrhage in hypertension. Commentary. J Neuropathol Exp Neurol . 1993;52:11.
66 Broderick JP, Brott TG, Duldner JE, et al. Volume of intracerebral hemorrhage. A powerful and easy-to-use predictor of 30-day mortality. Stroke . 1993;24:987.
67 Tuhrim S, Horowitz DR, Sacher M, et al. Volume of ventricular blood is an important determinant of outcome in supratentorial intracerebral hemorrhage. Crit Care Med . 1999;27:617.
68 Diringer MN, Edwards DF, Zazulia AR. Hydrocephalus: a previously unrecognized predictor of poor outcome from supratentorial intracerebral hemorrhage. Stroke . 1998;29:1352.
69 Tuhrim S, Horowitz DR, Sacher M, et al. Validation and comparison of models predicting survival following intracerebral hemorrhage. Crit Care Med . 1995;23:950.
70 Hemphill JC3rd, Bonovich DC, Besmertis L, et al. The ICH score: a simple, reliable grading scale for intracerebral hemorrhage. Stroke . 2001;32:891.
71 Becker KJ, Baxter AB, Bybee HM, et al. Extravasation of radiographic contrast is an independent predictor of death in primary intracerebral hemorrhage. Stroke . 1999;30:2025.
72 Adams RE, Powers WJ. Management of hypertension in acute intracerebral hemorrhage. Crit Care Clin . 1997;13:131.
73 Mendelow AD, Gregson BA, Fernandes HM, et al. Early surgery versus initial conservative treatment in patients with spontaneous supratentorial intracerebral haematomas in the International Surgical Trial in Intracerebral Haemorrhage (STICH): a randomised trial. Lancet . 2005;365:387.
74 Davis BR, Vogt T, Frost PH, et al. Risk factors for stroke and type of stroke in persons with isolated systolic hypertension. Systolic Hypertension in the Elderly Program Cooperative Research Group. Stroke . 1998;29:1333.
75 Fisher CM. Lacunes: small, deep cerebral infarcts. Neurology . 1965;15:774.
76 Longstreth WTJr, Bernick C, Manolio TA, et al. Lacunar infarcts defined by magnetic resonance imaging of 3660 elderly people: the Cardiovascular Health Study. Arch Neurol . 1998;55:1217.
77 You R, McNeil JJ, O’Malley HM, et al. Risk factors for lacunar infarction syndromes. Neurology . 1995;45:1483.
78 Gan R, Sacco RL, Kargman DE, et al. Testing the validity of the lacunar hypothesis: the Northern Manhattan Stroke Study experience. Neurology . 1997;48:1204.
79 Millikan C, Futrell N. The fallacy of the lacune hypothesis. Stroke . 1990;21:1251.
80 Caplan LR. Stroke: A Clinical Approach. Newton: Butterworth-Heineman, 1993.
81 Fisher SM. Lacunar stroke and infarcts: a review. Neurology . 1982;32:871.
82 Macdonald RL, Kowalczuk A, Johns L. Emboli enter penetrating arteries of monkey brain in relation to their size. Stroke . 1995;26:1247.
83 Mead GE, Wardlaw JM, Lewis SC, et al. Can simple clinical features be used to identify patients with severe carotid stenosis on Doppler ultrasound? J Neurol Neurosurg Psychiatry . 1999;66:16.
84 Chamorro A, Sacco RL, Mohr JP, et al. Clinical-computed tomographic correlations of lacunar infarction in the Stroke Data Bank. Stroke . 1991;22:175.
85 The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med . 1995;333:1581.
86 Salgado AV, Ferro JM, Gouveia-Oliveira A. Long-term prognosis of first-ever lacunar strokes. A hospital-based study. Stroke . 1996;27:661.
87 Adams HPJr, Brott TG, Furlan AJ, et al. Guidelines for thrombolytic therapy for acute stroke: a supplement to the guidelines for the management of patients with acute ischemic stroke. A statement for healthcare professionals from a Special Writing Group of the Stroke Council, American Heart Association. Circulation . 1996;94:1167.
88 CAST Collaborative Group. CAST: randomised placebo-controlled trial of early aspirin use in 20,000 patients with acute ischaemic stroke. Lancet . 1997;349:1641.
89 North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis. N Engl J Med . 1991;325:445.
90 Longstreth WTJr, Manolio TA, Arnold A, et al. Clinical correlates of white matter findings on cranial magnetic resonance imaging of 3301 elderly people. The Cardiovascular Health Study. Stroke . 1996;27:1274.
91 Longstreth WTJr, Arnold AM, Beauchamp NJJr, et al. Incidence, manifestations, and predictors of worsening white matter on serial cranial magnetic resonance imaging in the elderly: the Cardiovascular Health Study. Stroke . 2005;36:56.
92 Fernando MS, Simpson JE, Matthews F, et al. White matter lesions in an unselected cohort of the elderly: molecular pathology suggests origin from chronic hypoperfusion injury. Stroke . 2006;37:1391.
93 Isaka Y, Okamoto M, Ashida K, et al. Decreased cerebrovascular dilatory capacity in subjects with asymptomatic periventricular hyperintensities. Stroke . 1994;25:375.
94 Vermeer SE, Hollander M, van Dijk EJ, et al. Silent brain infarcts and white matter lesions increase stroke risk in the general population: the Rotterdam Scan Study. Stroke . 2003;34:1126.
95 Longstreth WTJr, Dulberg C, Manolio TA, et al. Incidence, manifestations, and predictors of brain infarcts defined by serial cranial magnetic resonance imaging in the elderly: the Cardiovascular Health Study. Stroke . 2002;33:2376.
96 Pantoni L, Garcia JH. The significance of cerebral white matter abnormalities 100 years after Binswanger’s report. A review. Stroke . 1995;26:1293.
97 Caplan LR. Binswanger’s disease—revisited. Neurology . 1995;45:626.
98 Kalimo H, Viitanen M, Amberla K, et al. CADASIL: hereditary disease of arteries causing brain infarcts and dementia. Neuropathol Appl Neurobiol . 1999;25:257.
99 Chabriat H, Vahedi K, Iba-Zizen MT, et al. Clinical spectrum of CADASIL: a study of 7 families. Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy. Lancet . 1995;346:934.
100 Hunt JR. The role of the carotid arteries in the causation of vascular lesions of the brain. Am J Med Sci . 147, 1914.
101 Timsit SG, Sacco RL, Mohr JP, et al. Early clinical differentiation of cerebral infarction from severe atherosclerotic stenosis and cardioembolism. Stroke . 1992;23:486.
102 Timsit SG, Sacco RL, Mohr JP, et al. Brain infarction severity differs according to cardiac or arterial embolic source. Neurology . 1993;43:728.
103 Li R, Duncan BB, Metcalf PA, et al. B-mode-detected carotid artery plaque in a general population. Atherosclerosis Risk in Communities (ARIC) Study Investigators. Stroke . 1994;25:2377.
104 Lee TT, Solomon NA, Heidenreich PA, et al. Cost-effectiveness of screening for carotid stenosis in asymptomatic persons. Ann Intern Med . 1997;126:337.
105 Wilson PW, Hoeg JM, D’Agostino RB, et al. Cumulative effects of high cholesterol levels, high blood pressure, and cigarette smoking on carotid stenosis. N Engl J Med . 1997;337:516.
106 Muluk SC, Muluk VS, Sugimoto H, et al. Progression of asymptomatic carotid stenosis: a natural history study in 1004 patients. J Vasc Surg . 1999;29:208.
107 Barnett HJ, Taylor DW, Eliasziw M, et al. Benefit of carotid endarterectomy in patients with symptomatic moderate or severe stenosis. North American Symptomatic Carotid Endarterectomy Trial Collaborators. N Engl J Med . 1998;339:1415.
108 Sauvé JS, Laupacis A, Ostbye T, et al. Does this patient have a clinically important carotid bruit? JAMA . 1993;270:2843.
109 Pan XM, Saloner D, Reilly LM, et al. Assessment of carotid artery stenosis by ultrasonography, conventional angiography, and magnetic resonance angiography: correlation with ex vivo measurement of plaque stenosis. J Vasc Surg . 1995;21:82.
110 Barnett HJ, Eliasziw M, Meldrum HE. Drugs and surgery in the prevention of ischemic stroke. N Engl J Med . 1995;332:238.
111 European Carotid Surgery Trialists’ Collaborative Group. MRC European Carotid Surgery Trial: interim results for symptomatic patients with severe (70–99%) or with mild (0–29%) carotid stenosis. Lancet . 1991;337:1235.
112 Rothwell PM, Warlow CP. Prediction of benefit from carotid endarterectomy in individual patients: a risk-modelling study. European Carotid Surgery Trialists’ Collaborative Group. Lancet . 1999;353:2105.
113 Halliday A, Mansfield A, Marro J, et al. Prevention of disabling and fatal strokes by successful carotid endarterectomy in patients without recent neurological symptoms: randomised controlled trial. Lancet . 2004;363:1491.
114 Sacco RL, Kargman DE, Gu Q, et al. Race-ethnicity and determinants of intracranial atherosclerotic cerebral infarction. The Northern Manhattan Stroke Study. Stroke . 1995;26:14.
115 Wityk RJ, Lehman D, Klag M, et al. Race and sex differences in the distribution of cerebral atherosclerosis. Stroke . 1996;27:1974.
116 Ingall TJ, Homer D, Baker HLJr, et al. Predictors of intracranial carotid artery atherosclerosis. Duration of cigarette smoking and hypertension are more powerful than serum lipid levels. Arch Neurol . 1991;48:687.
117 Postiglione A, Napoli C. Hyperlipidaemia and atherosclerotic cerebrovascular disease. Curr Opin Lipidol . 1995;6:236.
118 Lammie GA, Sandercock PA, Dennis MS. Recently occluded intracranial and extracranial carotid arteries. Relevance of the unstable atherosclerotic plaque. Stroke . 1999;30:1319.
119 Fassbender K, Bertsch T, Mielke O, et al. Adhesion molecules in cerebrovascular diseases. Evidence for an inflammatory endothelial activation in cerebral large- and small-vessel disease. Stroke . 1999;30:1647.
120 Chimowitz MI, Lynn MJ, Howlett-Smith H, et al. Comparison of warfarin and aspirin for symptomatic intracranial arterial stenosis. N Engl J Med . 2005;352:1305.
121 Mohr JP, Thompson JL, Lazar RM, et al. A comparison of warfarin and aspirin for the prevention of recurrent ischemic stroke. N Engl J Med . 2001;345:1444.
122 Lylyk P, Cohen JE, Ceratto R, et al. Angioplasty and stent placement in intracranial atherosclerotic stenoses and dissections. AJNR Am J Neuroradiol . 2002;23:430.
123 Ozgur HT, Kent Walsh T, Masaryk A, et al. Correlation of cerebrovascular reserve as measured by acetazolamide-challenged SPECT with angiographic flow patterns and intra- or extracranial arterial stenosis. AJNR Am J Neuroradiol . 2001;22:928.
124 Turan TN, Lynn M, Cotsonis G, et al: Target blood pressure and risk of recurrent stroke in patients with intracranial arterial stenosis. Paper presented at: 30th International Stroke Conference, Kissimmee, FL, 2005
125 Amarenco P, Duyckaerts C, Tzourio C, et al. The prevalence of ulcerated plaques in the aortic arch in patients with stroke. N Engl J Med . 1992;326:221.
126 Amarenco P, Cohen A, Tzourio C, et al. Atherosclerotic disease of the aortic arch and the risk of ischemic stroke. N Engl J Med . 1994;331:1474.
127 The French Study of Aortic Plaques in Stroke Group. Atherosclerotic disease of the aortic arch as a risk factor for recurrent ischemic stroke. N Engl J Med . 1996;334:1216.
128 Launer LJ, Masaki K, Petrovitch H, et al. The association between midlife blood pressure levels and late-life cognitive function. The Honolulu-Asia Aging Study. JAMA . 1995;274:1846.
129 Kivipelto M, Helkala EL, Laakso MP, et al. Apolipoprotein E epsilon4 allele, elevated midlife total cholesterol level, and high midlife systolic blood pressure are independent risk factors for late-life Alzheimer disease. Ann Intern Med . 2002;137:149.
130 Skoog I, Lernfelt B, Landahl S, et al. 15-year longitudinal study of blood pressure and dementia. Lancet . 1996;347:1141.
131 Petrovitch H, White LR, Izmirilian G, et al. Midlife blood pressure and neuritic plaques, neurofibrillary tangles, and brain weight at death: the HAAS. Honolulu-Asia aging Study. Neurobiol Aging . 2000;21:57.
132 Langa KM, Foster NL, Larson EB. Mixed dementia: emerging concepts and therapeutic implications. JAMA . 2004;292:2901.
133 Forette F, Seux ML, Staessen JA, et al. Prevention of dementia in randomised double-blind placebo-controlled Systolic Hypertension in Europe (Syst-Eur) trial. Lancet . 1998;352:1347.
134 Di Bari M, Pahor M, Franse LV, et al. Dementia and disability outcomes in large hypertension trials: lessons learned from the systolic hypertension in the elderly program (SHEP) trial. Am J Epidemiol . 2001;153:72.
135 Prince MJ, Bird AS, Blizard RA, et al. Is the cognitive function of older patients affected by antihypertensive treatment? Results from 54 months of the Medical Research Council’s trial of hypertension in older adults. BMJ . 1996;312:801.
136 Tzourio C, Anderson C, Chapman N, et al. Effects of blood pressure lowering with perindopril and indapamide therapy on dementia and cognitive decline in patients with cerebrovascular disease. Arch Intern Med . 2003;163:1069.
137 Elkins JS, Knopman D, Yaffe K, et al. Cognitive function predicts first-time stroke and heart disease. Neurology . 2005;64:1750.
138 Birns J, Markus H, Kalra L. Blood pressure reduction for vascular risk: is there a price to be paid? Stroke . 2005;36:1308.
139 Forrest KY, Maser RE, Pambianco G, et al. Hypertension as a risk factor for diabetic neuropathy: a prospective study. Diabetes . 1997;46:665.
140 Hinchey J, Chaves C, Appignani B, et al. A reversible posterior leukoencephalopathy syndrome. N Engl J Med . 1996;334:494.
141 Pentel P. Toxicity of over-the-counter stimulants. JAMA . 1984;252:1898.
142 Delanty N, Vaughan C, Frucht S, et al. Erythropoietin-associated hypertensive posterior leukoencephalopathy. Neurology . 1997;49:686.
143 Sheth RD, Riggs JE, Bodenstenier JB, et al. Parietal occipital edema in hypertensive encephalopathy: a pathogenic mechanism. Eur Neurol . 1996;36:25.
144 Blezer EL, Nicolay K, Bär D, et al. Enalapril prevents imminent and reduces manifest cerebral edema in stroke-prone hypertensive rats. Stroke . 1998;29:1671.
145 Schwartz RB, Mulkern RV, Gudbjartsson H, et al. Diffusion-weighted MR imaging in hypertensive encephalopathy: clues to pathogenesis. AJNR Am J Neuroradiol . 1998;19:859.
146 Redman CW, Sacks GP, Sargent IL. Preeclampsia: an excessive maternal inflammatory response to pregnancy. Am J Obstet Gynecol . 1999;180:499.
147 Mabie WC. Management of acute severe hypertension and encephalopathy. Clin Obstet Gynecol . 1999;42:519.
Chapter 8 Postural Hypotension

Michael J. Aminoff

Cardiovascular Disorders
Alterations of Effective Blood Volume
Endocrine and Metabolic Disorders
Inadequate Postural Adjustments
Central Lesions and Spinal Injury
Root and Peripheral Nerve Lesions
Toxic Exposure
Primary Degeneration of the Autonomic Nervous System
Miscellaneous Disorders
Postural Hypotension
Vasovagal Syncope
Deglutition Syncope
Micturition Syncope
Carotid Sinus Syncope
Syncope With Valsalva Maneuver
Cardiac Syncope
Postural Change in Blood Pressure
Postural Change in Heart Rate
Valsalva Maneuver
Other Cardiovascular Responses
Digital Blood-Flow
Cold Pressor Test
Norepinephrine Infusion
Response to Tyramine
Norepinephrine Response to Edrophonium
Sweat Tests
Other Studies
Pupillary Responses
Radiological Studies
General Precautions in Management of Dysautonomic Patients
When a normal person stands up after being recumbent, approximately 500 ml of blood (or more) pools in the vessels of the legs and abdomen, causing a reduction in filling pressure of the right atrium and thus a decrease in cardiac output and systemic blood pressure. This leads to changes in baroreceptor activity and thus to changes in impulse traffic in the ninth and tenth cranial nerves. These changes affect the activity of the brainstem vasomotor center, which, in turn, influences the autonomic neurons in the intermediolateral cell columns of the thoracolumbar spinal cord, producing reflex peripheral vasoconstriction and an increase in force and rate of myocardial contraction ( Figs. 8-1 and 8-2 ). Cardiopulmonary reflexes, subserved by vagal afferent fibers from mechanoreceptors in the heart and stretch receptors in the lungs, contribute to maintenance of the blood pressure, acting synergistically with the baroreceptor reflexes. The venoarteriolar axonal reflexes may also be important in limiting blood flow to the skin, muscle, and adipose tissues. Standing up also leads to release of norepinephrine. Venous return is aided during maintenance of the upright posture by mechanical factors, such as the tone in the leg muscles and the pumping action of these muscles during walking, and by maneuvers that increase intra-abdominal pressure. This has important therapeutic implications (see p. 155 ). 1 In addition, there is secretion of antidiuretic hormone (arginine vasopressin) and activation of the renin-angiotensin-aldosterone system, so that salt and water are conserved and blood volume increases. These, however, are typically longer term rather than immediate control mechanisms.

FIGURE 8-1 Anatomy of the autonomic pathways involved in maintaining the blood pressure on standing.

FIGURE 8-2 Sequence of events that ensure maintenance of the blood pressure after adoption of the upright posture. Only the immediate cardiovascular changes are shown. As indicated in the text, a variety of other humoral mechanisms are also activated.
Postural hypotension is defined as a decrease of at least 20 mmHg in systolic pressure or 10 mmHg in diastolic pressure on standing. It occurs when there is a failure of the autoregulatory mechanisms that maintain the blood pressure on standing. It may therefore occur with any neurological disorder that impairs baroreceptor function, disturbs the afferent input from these receptors, directly involves the brainstem vasomotor center or its central connections, or interrupts the sympathetic outflow pathway either centrally or peripherally. It may also occur with a number of non-neurological disorders, and it is important to consider these disorders if patients are to be managed correctly.


Cardiovascular Disorders
A variety of cardiac disorders may lead to postural hypotension or even syncope. Pathological processes such as mitral valve prolapse, aortic stenosis, or hypertrophic cardiomyopathy may limit cardiac output. Cardiac outflow may also be blocked in rare instances by a thrombus or myxoma when the patient is in the upright position. Certain paroxysmal cardiac dysrhythmias (bradycardias or tachycardias) may occur with activity or on standing and produce episodic hypotension or syncope; however, disturbances of cardiac rhythm are common in asymptomatic elderly persons, and their presence must be interpreted with caution. In patients with congestive heart failure, the heart rate and level of sympathetic tone may be such that compensatory adjustments cannot be made when the patient stands, and postural hypotension therefore results.

Alterations of Effective Blood Volume
Postural hypotension can occur because of loss of effective blood volume. Normal adults can withstand the loss of 500 ml of blood or bodily fluids with few if any symptoms, but greater volume depletion may occur acutely for a variety of reasons (e.g., hemorrhage or burns) and cause a postural drop in blood pressure. Hyponatremia and Addison’s disease may also lead to an absolute reduction in blood volume. Postural hypotension may occur owing to venous pooling in patients with severe varicose veins or congenital absence of venous valves or because of poor peripheral resistance and reduced muscle tone in patients with paralyzed limbs. Similarly, it may occur during the late stages of pregnancy owing to obstructed venous return by the gravid uterus. Marked vasodilatation, such as occurs in the heat or with the use of certain drugs or alcohol, sometimes causes postural hypotension.

Numerous drugs may produce postural hypotension, including those given to treat neurological disorders (e.g., dopamine agonists and levodopa) and psychiatric disturbances (e.g., tranquilizing, sedative, hypnotic, and antidepressant agents). 2 Antihypertensive drugs, diuretics, and vasodilators commonly lead to postural hypotension as a side effect. Insulin may cause nonhypoglycemic postural hypotension in diabetic patients with autonomic neuropathy, possibly because of vasodilatation and reduced venous return in the absence of functioning compensatory mechanisms or because of impaired baroreceptor responses to changes in arterial pressure. Iatrogenic and toxic autonomic neuropathy is considered later.

Endocrine and Metabolic Disorders
Autonomic neuropathy, with consequent postural hypotension, is a major and common complication of diabetes. Postural hypotension may be a feature of Addison’s disease, hypopituitarism, myxedema, thyrotoxicosis, pheochromocytoma, carcinoid syndrome, and hypokalemia. It may also occur with anorexia nervosa. Anemia may exacerbate or cause postural hypotension. A patient who had marked postural hypotension and extreme lability of the blood pressure in association with severe hypophosphatemia has been described. 3

Inadequate Postural Adjustments
Prolonged bed rest may result in postural hypotension when patients first begin standing again, but this problem is self-limited. Its cause is poorly understood, but it may be multifactorial. Carotid baroreceptor function is impaired, cardiac vagal activity is reduced, blood pooling is increased in the legs because of greater venous compliance, the total circulating blood volume and central venous pressure are reduced, and the red cell mass may decline. 4 - 6 Prolonged bed rest also leads to an increased incidence of cardiac dysrhythmias. In otherwise healthy subjects, vigorous exercise to the point of exhaustion may also cause a postural decline in blood pressure, possibly because of marked peripheral vasodilatation and venous pooling.

Many patients older than 70 years have a decline in systolic pressure of 20 mmHg or more on standing. Several causes for reduced orthostatic tolerance with advancing age have been identified. 7 Baroreflex sensitivity declines with age and certain adrenoreceptors exhibit reduced sensitivity. Loss of preganglionic neurons also occurs with age and becomes symptomatic when approximately 50 percent of the cells are lost. Diuretics (which are commonly taken by the elderly) reduce blood volume and may lead to postural hypotension. Finally, structural, mechanical, and functional changes in the vascular system, 8 such as loss of vascular elasticity and the occurrence of varicose veins, may be contributory, as may a reduction in the skeletal muscle mass. Prolonged bed rest, intercurrent illness, and adverse reactions to medication may also be important. Postural hypotension appears not to have an impact on mortality, at least among patients discharged from an acute geriatric ward. 9
Syncope is a common problem in the elderly. Often no precise explanation for it can be found, but postural hypotension is probably responsible in many instances. Nevertheless, it is best not to ascribe patients’ symptoms to postural hypotension unless they can be reproduced by a demonstrable fall in blood pressure on standing. Many of the homeostatic mechanisms that maintain intravascular volume and blood pressure may be impaired with advancing age, as discussed earlier, so that syncope is more likely to occur. Indeed, in many elderly patients a number of factors can be found to account for syncope, and it is then difficult to determine which of these factors is responsible in any individual instance.

The central nervous system (CNS) is important in regulating cardiovascular function. Various lower brainstem centers receive inputs from both the periphery and other central structures such as the cerebral cortex, temporal lobe, amygdala, hypothalamus, cerebellum, periaqueductal gray matter, and pontine nuclei. 10, 11 The nucleus tractus solitarius is the site of termination of baroreceptor, chemoreceptor, and cardiopulmonary afferent fibers; it connects with the nucleus ambiguus and dorsal nucleus of the vagus and with neurons in the lateral reticular formation that project to the cord in the bulbospinal pathway, thereby influencing the cardiovascular system. 12, 13
The vagus nerve has a major role in regulating the heart rate responses to various maneuvers. The sympathetic nervous system is important in influencing vasomotor tone and peripheral vascular resistance, but the sympathetic outflow to different regions and structures is regulated separately. The sympathetic nervous system causes a vasoconstriction in response to the release of norepinephrine. The occurrence of vasodilatation in the limbs probably depends on reduced sympathetic activity, and, to a lesser extent, on axon reflexes and antidromic conduction, but some of the vessels in limb muscles are probably also supplied by sympathetic vasodilator cholinergic fibers.
Microneurographic studies in humans have shown that bursts of impulses occur rhythmically in sympathetic efferent vasomotor fibers to the skin and muscles and are time-locked to the pulse. This rhythmic activity depends on supraspinal mechanisms and is not seen below the level of a complete cord transection. Such sympathetic impulse traffic to vessels in the limb muscles is markedly affected by baroreceptor activity, but not by brief mental stress, 14 whereas the traffic in human cutaneous nerves is markedly increased by mental stress. High-pressure arterial baroreceptors are located primarily in the carotid sinus and aortic arch, from which afferent fibers pass to the brainstem in the glossopharyngeal and vagus nerves, respectively.
Sympathetic efferent activity is inhibited by an increase in the pressure in the carotid sinus and aortic arch, whereas a reduced pressure causes increased sympathetic activity and a peripheral vasoconstriction. The heart rate is also influenced by the baroreceptors and cardiopulmonary stretch receptors, so that a bradycardia occurs when the pressure is increased and a tachycardia when the blood pressure declines.
Change from recumbency to an erect posture causes blood to pool in the legs and lower abdomen. There is a slight fall in systolic blood pressure; this leads to baroreceptor activation, a peripheral vasoconstriction, and an increase in heart rate and contractile force. Compensatory changes in the splanchnic vasculature, constriction of venous beds, and activation of the renin-angiotensin system also occur.
The carotid baroreceptor reflexes seem to be more important in responding to the immediate changes in blood pressure that occur on standing, whereas the aortic baroreceptors assume a greater role with maintenance of the upright posture. The cardiopulmonary stretch receptors act synergistically with the baroreceptor reflexes. The venoarteriolar axon reflex, which is activated by venous distention in the legs and an associated increase in transmural venous pressure, is also important in ensuring an increase in limb vascular resistance with change to an erect posture. During activity, the baroreceptors are reset by an uncertain, probably neural, mechanism to allow the blood pressure to increase with exercise. Unmyelinated chemoreceptor afferent fibers from skeletal muscles are also activated, 15 thereby increasing blood pressure and correcting any deficiency in muscle perfusion pressure during moderate to heavy exercise. In addition, activation of mechanically sensitive muscle receptors (muscle mechanoreflex) occurs, and these exercise pressor reflexes (peripheral neural reflexes originating in skeletal muscle) contribute significantly to cardiovascular regulation during exercise. 16 At the initiation of exercise, “central command” from higher brain centers leads to an immediate increase in heart rate and output as well as in blood pressure and respiration. 17, 18


Central Lesions and Spinal Injury
The autonomic consequences of spinal cord injuries depend on the level and severity of the lesion. In quadriplegic patients, the period of spinal shock that follows injury is associated with a dysautonomia in which the resting blood pressure and heart rate are typically low and postural hypotension is marked. This mandates that the patient be kept flat, without elevation of the head of the bed, and that any loss of blood volume be avoided or treated vigorously.
A few weeks after transection of the cervical cord, activity returns to the isolated spinal segment, but the brain is no longer able to control the sympathetic nervous system. Loss of regulation during postural change leads to orthostatic hypotension, whereas overactivity occurs if spinal sympathetic reflexes are activated and leads to the syndrome of autonomic hyperreflexia, which occurs in patients with cervical or high thoracic lesions. It is characterized by episodic hypertension, bradycardia, headache, and hyperhidrosis above the level of the lesion, with pallor and piloerection distal to it. Anxiety, confusion, nasal congestion, and facial flushing may also occur. Treatment of this syndrome thus requires avoidance of stimuli that activate spinal sympathetic reflexes (e.g., a distended bladder), elevation of the head of the bed, and, if necessary, use of short-acting antihypertensive agents such as calcium-channel blockers. In general, spinal cord transection produces postural hypotension if the lesion is above about the T6 level. Intramedullary and extramedullary tumors, transverse myelitis, and syringomyelia involving the cord above T6 may also produce dysautonomia.
A variety of brainstem lesions can impair autonomic function and affect control of the blood pressure, including syringobulbia 19, 20 and posterior fossa tumors. 21 Chiari malformation with tonsillar herniation may lead to syncopal episodes. 22 Impairment in Wernicke’s encephalopathy may relate to central or peripheral involvement. The extent to which autonomic reflex function, and particularly cardiovascular regulation, is impaired in Parkinson’s disease is disputed. Several recent studies have indicated that many patients with Parkinson’s disease have postural hypotension from cardiac sympathetic denervation. 23 In such patients, responses to the Valsalva maneuver are also abnormal. 24 Other dysautonomic symptoms, such as disturbances of bladder or gastrointestinal function, and excessive salivation, are relatively common.
The findings in certain other disorders with parkinsonian features (e.g., Shy–Drager syndrome, olivopontocerebellar atrophy, and striatonigral degeneration) are discussed on p. 147 . Mild postural hypotension occurs occasionally in progressive supranuclear palsy, but cardiovascular reflexes are preserved or show only minor abnormalities of dubious significance. 24 - 26 A variety of dysautonomic symptoms may occur in Huntington’s disease, 27 but any abnormalities of blood pressure regulation are usually mild and subclinical, 28 except when related to neuroleptic medication taken for chorea or behavioral disturbances. Postural hypotension or other disturbances of cardiovascular autonomic function occur occasionally in patients with multiple sclerosis, 29, 30 but disturbances of bladder and bowel function are much more common dysautonomic features of that disorder. Wallenberg’s syndrome or bilateral brainstem strokes may lead to bradycardia and hypotension that may exacerbate the underlying neurological problem. 31

Root and Peripheral Nerve Lesions
Postural hypotension may occur in patients with tabes dorsalis because of interruption of circulatory reflexes. In patients with polyneuropathies, autonomic involvement is not uncommon. It is particularly frequent in diabetic neuropathy, although usually relatively mild in severity 32 ; indeed, diabetes is the most common cause of autonomic neuropathy in the more developed countries. 33 Postural hypotension occurs in approximately 25 percent of patients with diabetic neuropathy. In addition to postural lightheadedness, the dysautonomia of diabetes may be manifest by impotence, postprandial bloating, early satiety, gastrointestinal motility disturbances, abnormalities of bladder control, and alterations of sweating. Cardiac vagal control is usually impaired early, before the development of postural hypotension; the quantitative sudomotor axon reflex test is commonly abnormal and indicates involvement of distal postganglionic sympathetic fibers. 34
Other polyneuropathies associated with postural hypotension include those of alcoholism, Guillain–Barré syndrome, malignant disease, idiopathic small-fiber neuropathies, and acute porphyria. In many patients with chronic alcoholism, however, there is no excessive decline in blood pressure on standing, although the cardiovascular responses to various maneuvers are often abnormal and indicate chronic vagal damage. 35
Primary amyloidosis and familial amyloid polyneuropathy of Portuguese type (FAP type 1) are often accompanied by dysautonomia consequent to the loss of predominantly unmyelinated and small myelinated peripheral fibers and of cells in the intermediolateral columns of the cord. Postural hypotension and impotence are early manifestations; episodic constipation and diarrhea, distal anhidrosis, impotence, urinary retention, and cardiac arrhythmias may also be conjoined. Tests of sympathetic and parasympathetic function are typically abnormal. 36
Autonomic dysfunction with abnormal cardiovascular responses occurs in some patients with chronic renal failure on intermittent hemodialysis, but the site of autonomic involvement is unclear. Vitamin B 12 deficiency may lead to autonomic neuropathy and postural hypotension that improves or resolves completely after vitamin supplementation. 37, 38 Autonomic involvement, with impairment of sweating and cardiovascular responses, may occur in leprosy, sometimes without conspicuous features of peripheral nerve involvement. 39, 40 Symptoms of autonomic impairment, including postural hypotension, may be a presenting feature of systemic autoimmune disorders. 41
In Fabry’s disease, disturbed sweating, reduced saliva and tear production, impaired pupillary responses, and gastrointestinal symptoms are common, but postural hypotension does not usually occur, and postural cardiovascular reflexes are normal. 42
Autonomic involvement in Guillain–Barré syndrome is usually mild, but paroxysmal cardiac arrhythmias or asystole or episodic hypertension may lead to a fatal outcome. Postural hypotension is common. 43 It has a number of possible causes including inactivity and bed rest, baroreceptor deafferentation, efferent sympathetic denervation, hypovolemia, cardiac abnormalities, or some combination of these and other factors. The severity of autonomic involvement in Guillain–Barré syndrome is not related to the degree of sensory or motor disturbance, and a wide variety of autonomic abnormalities is found if patients are studied in detail. The hypertensive episodes may relate to catecholamine supersensitivity or denervation of baroreceptors. Treatment of Guillain–Barré syndrome is by supportive measures or with plasmapheresis or intravenous immunoglobulin therapy depending on disease severity. Patients with autonomic instability require close observation and management in an intensive care unit. Further aspects of treatment are given on page 155 . Curiously, postural hypotension is uncommon in chronic inflammatory demyelinating polyneuropathy, although mild impairment may be found in many patients on tests of autonomic function. 44, 45
Autonomic neuropathy of acute or subacute onset, possibly on an autoimmune basis, sometimes occurs as a monophasic disorder in isolation or with associated sensory or motor involvement. It has occurred in the context of antecedent viral infections, malignancy, Hodgkin’s disease, infectious mononucleosis, certain connective tissue diseases, 46, 47 and ulcerative colitis. 48 In approximately 50 percent of patients, high titers of ganglionic acetylcholine receptor (AChR) antibody are found. In patients with a paraneoplastic etiology, other autoantibodies may be present, including antineuronal nuclear antibody 1 (ANNA-1 or anti-Hu) or 2 (ANNA-2), Purkinje cell antibody 2 (PCA-2), and collapsin response-mediator protein 5 antibody (CRMP-5). 49 The presence of such antibodies may suggest the likely site of an underlying primary tumor. Both sympathetic and parasympathetic fibers are usually involved, leading to marked postural hypotension accompanied by a fixed heart rate, anhidrosis or hypohidrosis, heat intolerance, sphincter disturbances, gastroparesis, ileus, and dryness of the eyes and mouth, but occasionally abnormalities are confined to postganglionic cholinergic neurons (acute cholinergic neuropathy), in which case postural hypotension does not occur. Pure adrenergic neuropathy has also been described. Autonomic function tests reflect the clinical findings. Nerve conduction study results are typically normal, but sensory abnormalities are sometimes found. Treatment is supportive; immunomodulating therapy may have a role in those with severe disease. Paraneoplastic dysautonomia may remit if the underlying malignancy is treated. The prognosis of patients with acute or subacute autonomic neuropathies is guarded: approximately one third of patients do well, but the remainder either fail to improve or are left with a major residual deficit, including marked postural hypotension.
Autonomic involvement may occur in a variety of hereditary polyneuropathies. In familial dysautonomia, or Riley–Day syndrome, many parts of the nervous system are affected. Presentation during infancy may be with inability to suck, but episodic vomiting, recurrent pulmonary infections, hypertension, tachycardia, and diaphoresis occur, especially after 3 years of age. There may also be emotional outbursts, difficulty in swallowing, hypothermia or hyperthermia, poor flow of tears, postural hypotension, and syncope. Sensory abnormalities include impaired pain and temperature appreciation, and the tendon reflexes are depressed. The tongue is smooth and lacks fungiform papillae. Cardiac arrest may occur on tracheal intubation. Treatment is essentially supportive. 50, 51 Postural hypotension usually is not a feature of the other hereditary sensory and autonomic neuropathies, whereas sudomotor function is often markedly impaired.
Autonomic symptoms or signs may occur in patients with hereditary motor and sensory neuropathy type 1, and abnormal vascular reflex responses may be present. 52 Postural hypotension is usually not a conspicuous feature of the disorder.
A familial disorder characterized by progressive distal weakness and muscle atrophy, distal sensory loss, pyramidal and visual pathway lesions, and dysautonomia has been reported. 53 Patients had a sensorimotor polyneuropathy. The dysautonomia consisted of postural hypotension, lack of normal sinus arrhythmia, an abnormal response to the sweat test, abnormal pupillary responses with denervation supersensitivity, and low serum norepinephrine levels in both the recumbent and upright positions.
Iatrogenic postural hypotension is common in the elderly 54 and relates most often to the use of antihypertensive agents or diuretics. Iatrogenic polyneuropathies may be responsible in other instances, as reviewed elsewhere 55 ; postural hypotension may be conspicuous in patients with the neuropathy caused by perhexiline maleate, cisplatin, paclitaxel (Taxol), vinca alkaloids, or amiodarone. 55

Toxic Exposure
Autonomic dysfunction may result from occupational or other exposure to certain neurotoxins but does not usually lead to postural hypotension. One study suggested that long-term occupational exposure to a mixture of organic solvents causes subtle disturbances of peripheral parasympathetic nerves as well as sensorimotor peripheral neuropathies, as reflected by cardiovascular reflex studies, 56 but other reports of autonomic involvement in this context are few. Intentional inhalation of n -hexane or methyl- n -butyl-ketone for recreational purposes may lead to a rapidly progressive neuropathy with associated postural hypotension. 55 Acrylamide neuropathy is usually accompanied by hyperhidrosis and cold, cyanotic extremities; in intoxicated rats, but not in humans, there is evidence that acrylamide alters common measures of cardiovascular function. 57 A variety of autonomic symptoms (in-cluding tachycardia, hypertension, and disturbances of sweating) may occur with thallium, arsenic, or mercury poisoning, but postural hypotension is not usually a feature.
McFarlane and associates described a patient with severe postural hypotension and neurological signs that were attributed to styrene intoxication. 58 The rodenticide N -3-pyridylmethyl- N’ - p -nitrophenyl urea (Vacor) has caused severe dysautonomia with disabling postural hypotension, 59 as well as sensorimotor peripheral neuropathy and encephalopathic states. Iatrogenic postural hypotension was considered earlier.

Primary Degeneration of the Autonomic Nervous System
Postural hypotension resulting from primary degeneration of the autonomic nervous system is well described. The postural hypotension is often exacerbated postprandially, and the normal circadian rhythm is reversed so that the blood pressure is highest at night and lowest in the morning. In addition, blood pressure typically declines with activity rather than increasing as in normal subjects. Other symptoms of dysautonomia in these patients include impotence, disturbances of bladder and bowel function, impaired thermoregulatory sweating, and xerostomia. Two distinct groups of patients are now recognized. In one, primary or pure autonomic failure leads to idiopathic orthostatic hypotension and other evidence of dysautonomia without peripheral neuropathy or CNS involvement. In the other, autonomic failure is associated with more widespread neurological degeneration (i.e., with evidence of multisystem atrophy ) such that there may be clinical features of parkinsonism (or striatonigral degeneration), and often of pyramidal, cerebellar, and lower motor neuron lesions as well (Shy–Drager syndrome). A disorder similar to olivopontocerebellar atrophy may also occur. The autonomic deficit may precede the somatic neurological one, or vice versa, but within a short period there is clinical evidence of both. 60 Occasionally there is a family history of dysautonomia.
The time course and pattern of the dysautonomia reportedly differ between these two disorders. In pure autonomic failure, syncope and sudomotor dysfunction may precede the onset of constipation, bladder dysfunction, or respiratory disturbances whereas in multisystem atrophy, urinary complaints occur early and are then followed by abnormalities of sweating or by postural hypotension. 61 Patients with pure autonomic failure had a slower functional deterioration and a better prognosis. 61 The ingestion of water temporarily increases the seated blood pressure by uncertain mechanisms in patients with chronic autonomic failure. 62 This occurs earlier in pure autonomic failure than multisystem atrophy, perhaps reflecting differing lesion sites in these two disorders.
In patients of both groups, plasma renin activity is usually subnormal. There are, however, a number of reported pharmacological differences between them. Patients with pure autonomic failure have low plasma norepinephrine levels when lying down, and these levels fail to increase appropriately on standing; they also have a lower threshold for the pressor response to infused norepinephrine. 63 The increase in plasma norepinephrine level in response to tyramine (see p. 154 ) is significantly less than in normal subjects or patients with multisystem atrophy. 64 Extensive cell loss occurs in the intermediolateral cell columns of the thoracic cord, and the autonomic dysfunction has been attributed primarily to loss of these preganglionic sympathetic neurons. However, the pharmacological studies described previously indicate that loss of postganglionic noradrenergic neurons also occurs, and norepinephrine may be depleted from sympathetic nerve endings.
By contrast, in multisystem atrophy, in which lesions are situated at multiple sites in the CNS, circulating norepinephrine levels are normal, suggesting that peripheral sympathetic neurons are intact, but plasma norepinephrine fails to increase appropriately with standing, implying that these neurons have not been activated. 63, 64 There is also an exaggerated pressor response to infused norepinephrine, but only patients with idiopathic orthostatic hypotension show a shift to the left in their dose–response curve, reflecting true adrenergic receptor supersensitivity. 64
Some investigators have not found any clear pattern in the resting level of plasma norepinephrine that distinguishes between these two types of autonomic failure. They have emphasized that low levels probably indicate the severity of sympathetic dysfunction rather than the site of lesions. 65
Endogenous arginine vasopressin is a powerful vasoconstrictor; it also acts on the kidney to control urinary concentrating mechanisms. The cardiovascular responses usually associated with arginine vasopressin are reduced cardiac output, heart rate, and plasma renin activity and increased vascular resistance and blood pressure. 66, 67 Arginine vasopressin helps maintain arterial pressure in certain hypotensive situations such as hemorrhage or volume depletion, but increased levels of arginine vasopressin do not normally affect the blood pressure significantly because the acute vasoconstrictor effects are buffered by the baroreceptor reflex. The chronic effects of vasopressin on renal function do not produce sustained retention of sodium and water, and so produce only minimal changes in mean arterial pressure.
Vasopressin release is influenced by the plasma’s osmotic pressure and by the activity of vascular stretch receptors. In normal people, plasma arginine vasopressin increases in response to standing, presumably because a decrease in venous return influences afferent activity from these stretch receptors. In patients with progressive autonomic failure or with multisystem atrophy, plasma levels similar to control values are found in the horizontal position, but the postural rise is only 10 percent of that in normal people. 68 If hypertonic saline is infused intravenously into such patients, plasma arginine vasopressin increases in a manner comparable with that in normal control subjects. 69 This suggests normal function of the efferent connections from the osmoreceptors within the hypothalamus and implies that in this clinical context, the loss of vasopressin response with head-up tilt is due to lesions in ascending pathways from cardiovascular receptors. 69 A vulnerability has been suggested of the noradrenergic neurons of the caudal ventrolateral medulla that are involved in activating hypothalamic neurons involved in vasopressin secretion. 70 However, in multisystem atrophy, loss of vasopressin neurons occurs in the posterior region of the hypothalamic paraventricular nucleus and may contribute to sympathetic failure, whereas loss of catecholaminergic input from the brainstem to the magnocellular vasopressin neurons may contribute to impaired vasopressin secretion following orthostatic stress. 71

Miscellaneous Disorders
Patients with Holmes–Adie syndrome may present with or develop postural hypotension or abnormalities of thermoregulatory sweating. Postural hypotension may occur in botulism; however, blurred vision, dry mouth, and constipation are much more common autonomic manifestations. In rare instances, it relates to excessive amounts of endogenous bradykinin (a vasodilator) or a congenital defect of norepinephrine release. In patients with dopamine β-hydroxylase deficiency, norepinephrine and epinephrine cannot be synthesized, and dopamine is released from central and peripheral adrenergic nerve terminals. 72, 73 Severe postural hypotension is accompanied by other autonomic disturbances in association with absent norepinephrine and excessive dopamine levels in the plasma.
Orthostatic symptoms may develop in association with a significant tachycardia (an increase of 30 beats per minute or more), but in the absence of consistent postural hypotension or an autonomic neuropathy. 74, 75 The designation postural tachycardia syndrome (POTS) is applied to this disorder, which is more common in women than men and tends to occur in patients between 20 and 60 years of age. Symptoms on standing include tremulousness, lightheadedness, palpitations, visual disturbances, weakness, fatigue, anxiety, hyperventilation, nausea, postprandial bloating, and sweating, and may occur cyclically. Thus, orthostatic symptoms are accompanied by symptoms of sympathetic activation. A preceding viral infection can sometimes be recognized, and the syndrome may also occur in association with mitral valve prolapse or a more specific dysautonomia. Its etiology remains uncertain. A reduced or redistributed blood volume of uncertain cause has been proposed. Partial sympathetic denervation in the legs may be responsible, as suggested by studies of norepinephrine spill-over into the venous circulation of the limbs in response to various stimuli. 76 There is impaired peripheral vasoconstriction on the Valsalva maneuver, but cardiovagal responses are normal. 77 The optimal therapy for the disorder is not clear, but treatment may include volume repletion, fludrocortisone, β-blocking agents, midodrine, and phenobarbital.

Postural hypotension is usually the most disabling feature of autonomic failure. Symptoms induced by postural hypotension reflect cerebral hypoperfusion and include faintness, lightheadedness, blurred vision, and syncope. They may be particularly troublesome after exercise or a heavy meal or in the morning when the blood pressure tends to be at its lowest (in contrast to healthy subjects). However, in some patients, marked postural hypotension may be clinically asymptomatic or may be accompanied by symptoms not usually regarded as suggestive of postural hypotension, such as nausea, breathlessness, heaviness or weakness of the limbs, episodic confusion, falling, staggering, and generalized weakness. Constipation may precipitate syncopal attacks during straining. Symptoms may also worsen in the heat because of vasodilatation and volume loss due to sweating. The symptoms of idiopathic postural tachycardia syndromes (discussed earlier) may be mistakenly attributed to postural hypotension, but occur without a significant decrease in blood pressure.
Impotence is a common initial symptom of autonomic dysfunction in men, often preceding other symptoms by several months or years. Bladder involvement may manifest by urinary frequency, urgency, incontinence, retention, and increased residual urine; urinary infections and renal calculi may occur in some patients with urinary stasis. Bowel dysfunction may lead to constipation, fecal incontinence, and diarrhea. Thermoregulatory sweating may be impaired. Pupillary abnormalities include Horner’s syndrome and anisocoria. Lacrimal dysfunction may lead to inadequate or excessive production of tears. Other symptoms of dysautonomia include night blindness, nasal congestion, and, sometimes, supine hypertension. Vocal abnormalities and respiratory disturbances (especially involuntary inspiratory gasps, cluster breathing, airway obstruction, and sleep apnea) sometimes occur, especially in patients with multisystem atrophy.

Syncope refers to a sudden, transient loss of consciousness due to diffuse cerebral ischemia or hypoxia. It is usually associated with flaccidity, but a generalized increase in muscle tone sometimes occurs with continuing cerebral ischemia/hypoxia, and there may be arrhythmic transient motor activity as well. Postictal confusion is usually brief (less than 30 seconds) when it occurs at all, unlike the marked postictal confusion that often follows a convulsion. Several causes of syncope have been recognized.

Postural Hypotension
In patients with orthostatic hypotension due to autonomic dysfunction, there is a fall in blood pressure on standing, without adequate compensatory change in total peripheral resistance or heart rate, and syncope may result. When postural hypotension occurs because of one of the non-neurological causes discussed earlier, it may also lead to syncope if autonomically mediated compensatory mechanisms fail to limit the decline in blood pressure.

Vasovagal Syncope
During vasovagal syncope, there is an initial increase in heart rate, blood pressure, total peripheral resistance, and cardiac output, followed by peripheral vasodilatation, increased blood flow to the muscles, decreased heart rate, and a decrease in venous return to the heart. Blood pressure falls owing to failure to increase the heart rate and cardiac output sufficiently, a decrease in systemic vascular resistance, or both. The vasodilatation and decline in systemic vascular resistance have been related to depressor Bezold–Jarisch reflexes arising from cardiac sensory receptors and subserved by vagal unmyelinated afferent fibers, perhaps in consequence of a decrease of central blood volume and decreased ventricular filling. 78, 79 Recordings from nerve fibers reveal that impulse traffic ceases in the sympathetic outflow to skeletal muscle during syncope and gradually builds up again over the following 5 minutes or so. 80
Syncope of this sort may be precipitated by pain, fear, emotional reactions, injury, and surgical manipulation. It may occur in association with missed meals, heat, or crowds; it usually occurs while subjects are standing. Warning symptoms include weakness, sweating, pallor, nausea, yawning, sighing, hyperventilation, blurred vision, impaired external awareness, and dilatation of pupils. Lying down or squatting at this time may abort actual loss of consciousness.

Deglutition Syncope
Swallowing precipitates syncope in some subjects (deglutition syncope). In such instances, there may be associated esophageal disorders. 81 The syncope has usually been attributed to atrioventricular heart block or cardiac arrhythmia. It is presumed that the prime factor is clinical or subclinical disease of the conducting system of the heart and that disturbances of cardiac rhythm are then triggered by reflexes originating in the esophagus. A pacemaker may prevent further episodes. 81, 82

Micturition Syncope
Micturition syncope occurs after urination, particularly when the patient has arisen from bed at night. It may relate to sudden release of the reflex vasoconstriction elicited by a full bladder. Assumption of the upright posture, the peripheral vasodilatation resulting from the warmth of the bed, and, particularly in elderly men, straining to micturate may also contribute to the drop in blood pressure. Occasionally, syncope occurs in response to cardiac dysrhythmia induced by a full bladder before micturition.

Carotid Sinus Syncope
Carotid sinus syncope may be provoked by neck-turning or a tight collar in susceptible subjects. Certain drugs have also been shown to predispose toward it, 83 especially propranolol, digitalis, and α-methyldopa. A hypersensitive carotid sinus reflex is defined by a slowing in heart rate of more than 50 percent or a decline in systolic pressure by more than 40 mmHg during carotid sinus massage. 83 However, less than 50 percent of patients with carotid hypersensitivity have syncope as a result. Conversely, in many patients with syncope of unidentifiable cause, the carotid sinus syndrome may have been overlooked.

Syncope With Valsalva Maneuver
Vigorous coughing or straining at stool may lead to syncope because of the reduced cardiac output and the peripheral vasodilatation caused by a high intrathoracic pressure. Cerebral perfusion may also be reduced by an increase in intracranial pressure.

Cardiac Syncope
As discussed earlier, postural hypotension may have a cardiac basis. In addition, atrioventricular conduction block may lead to sudden loss of consciousness, regardless of the position of the body (Adams–Stokes attacks). Further discussion of this topic is provided in Chapter 5 . Exertional syncope suggests obstructive valvular disease or a right-to-left shunt. Coronary artery disease may lead to arrhythmias that cause syncope.

Hyperventilation, with consequent hypocapnia and reduced cerebral perfusion, is a common cause of presyncopal symptoms, but actual loss of consciousness is uncommon.

After neurological, cardiological, and metabolic causes of syncope have been excluded, a number of patients remain in whom the diagnosis is unclear. The utility of autonomic studies in these circumstances was examined by Mathias and colleagues. 84 They found that screening autonomic function tests revealed postural hypotension and confirmed chronic autonomic failure in 5 percent, and neurally mediated syncope was diagnosed in 43.5 percent based on clinical features and autonomic studies. Thus, in recurrent syncope or presyncope, autonomic studies are worthwhile as they may clarify the diagnosis. In patients with unexplained syncope, the implantable loop recorder is an important diagnostic tool that may clarify the underlying pathophysiology. 85

Postural Change in Blood Pressure
In investigating patients with suspected autonomic dysfunction or postural hypotension, the blood pressure should be measured with the patient supine for at least 10 (preferably 20) minutes. The patient then stands up, and the blood pressure is measured again after 5 to 10 seconds, and again after 1, 2, and 5 minutes. There is normally an increase in pulse rate on standing, but the pulse rate may not change if there is already a high resting pulse or in patients with dysautonomia; furthermore, the change in heart rate may be blunted in the elderly. As for the blood pressure, there is normally a slight decline in systolic pressure, whereas diastolic pressure increases slightly. The response of the blood pressure is regarded as abnormal if systolic pressure decreases by at least 20 mmHg or diastolic pressure by 10 mmHg on standing. In some instances, postural hypotension develops only after exercise; it is therefore worthwhile to record the postactivity blood pressure if clinically feasible. It may be necessary to record the blood pressure on a number of occasions before the diagnosis of postural hypotension can be confirmed. In other instances, prolonged tilt (for up to 60 minutes) may be required to detect abnormalities.
The effect of postural change on blood pressure can be evaluated more accurately if the blood pressure is measured using an intra-arterial cannula or a noninvasive plethysmographic device with the patient resting quietly on a tilt-table; measurements are made continuously while the patient is supine and then at a 60-degree head-up tilt. The response to head-up tilt may differ from that obtained by standing because less enhancement occurs of the venous return to the heart by contraction of leg and abdominal muscles, and thus there is greater peripheral pooling of blood.
Mann and co-workers monitored ambulant intra-arterial blood pressure in six patients with autonomic failure and postural hypotension. 86 They found a consistent circadian trend in blood pressure that was the inverse of the pattern in normal subjects, with the highest pressures found at night and the lowest in the morning. Such temporal variation in blood pressure implies that physiological testing should be carried out at a standard time of day, especially if comparative studies are to be performed, and potentially harmful hypertension in response to treatment should be looked for especially during the early part of the night.

Postural Change in Heart Rate
A simple, noninvasive test of autonomic function consists of evaluating the response in heart rate to change from a recumbent to a standing position. There is typically a rapid increase in heart rate that is maximal at approximately the fifteenth beat after standing, with a subsequent slowing from the initial tachycardia (i.e., a relative bradycardia) that is maximal at approximately the thirtieth beat ( Fig. 8-3 ). This response is mediated by the vagus nerve. For testing purposes, the R-R interval at beats 15 and 30 after standing can be measured to give the 30/15 ratio. 87 Values greater than 1.03 occur in normal subjects, whereas in diabetic patients with autonomic neuropathy (who typically show only a gradual increase in heart rate), values are 1.00 or less. Some prefer to measure the ratio of the absolute maximum to absolute minimum heart rate after standing, which may not coincide with the heart rates at beats 15 and 30. 88 This test does not depend on the resting heart rate and correlates well with the Valsalva ratio and the beat-to-beat variation in heart rate, described later. Studies suggest that the value for the 30/15 ratio declines with age in normal subjects. 88, 89

FIGURE 8-3 Heart rate responses to standing in a normal subject. Immediately on standing, there is a rapid increase in heart rate that is maximal at approximately the fifteenth beat after standing.
In some patients, an excessive and sustained tachycardia develops in response to standing or head-up tilt, without significant drop in blood pressure. A prolonged tilt (for up to 60 minutes) may be necessary to elicit the abnormality. The mechanisms underlying this postural tachycardia have not been clearly established.

Valsalva Maneuver
The Valsalva maneuver consists of a forced expiration maintained for at least 10 seconds (preferably 15 seconds) against a closed glottis after a full inspiration. Intrathoracic pressure should be increased by 30 to 40 mmHg. Clinically, this can be ensured by requiring the patient to blow into a mouthpiece connected to a manometer. The response can be recorded with an intra-arterial needle ( Fig. 8-4 ), a noninvasive photoplethysmographic recording device (Finapres), or an electrocardiograph (ECG) ( Fig. 8-5 ).

FIGURE 8-4 Cardiovascular responses to the Valsalva maneuver, as recorded with an intra-arterial needle. A, Normal response. B, Abnormal response in a patient with Shy–Drager syndrome.
(From Aminoff MJ: Electromyography in Clinical Practice. 3rd Ed. Churchill Livingstone, New York, 1998, p. 206, with permission.)

FIGURE 8-5 Valsalva maneuver as recorded using an electrocardiograph (ECG) or heart rate monitor in a normal subject. The tachycardia that occurs during the forced expiratory maneuver is clearly evident, as is the compensatory bradycardia that occurs when the maneuver is released.
The cardiovascular response is usually divided into four stages. Stage 1 is characterized by a transient increase in blood pressure at the onset of the forced expiration, reflecting the increased intrathoracic pressure. In stage 2, there is normally a gradual decrease in systolic and diastolic pressures, pulse pressure, and stroke volume for several seconds because of a reduction in venous return to the heart, with an associated reflex tachycardia. Reflex vasoconstriction arrests the decline in blood pressure after about 5 to 7 seconds. Stage 3 occurs when the patient releases the expiratory maneuver and is characterized by a transient fall in the blood pressure because of pooling of blood and expansion of the pulmonary vascular bed with the abrupt decline in intrathoracic pressure. In stage 4, there is an overshoot of the blood pressure above baseline value as a result of the peripheral vasoconstriction, with a compensatory bradycardia.
The Valsalva maneuver is an accurate indicator of baroreceptor reflex sensitivity. Abnormalities are found in patients with dysautonomia ( Fig. 8-4 ) and may consist of loss of the overshoot in systolic blood pressure and compensatory bradycardia in stage 4, a fall in mean blood pressure in stage 2 to less than 50 percent of the previous resting mean pressure, and loss of the tachycardia in stage 2 or a lower heart rate in stage 2 than stage 4. However, abnormalities may also be found in patients with severe congestive heart failure and in those with cardiac lesions other than primary myocardial dysfunction.
If the response is recorded noninvasively using an electrocardiograph, the ratio of the shortest R-R interval (the tachycardia) during the maneuver to the longest R-R interval (bradycardia) after it is determined and expressed as the Valsalva ratio. A value of 1.1 or less was arbitrarily defined by Ewing and associates as an abnormal response, 1.21 or greater as a normal response, and 1.11 to 1.20 as borderline. 90 Using such criteria, these authors found that the Valsalva maneuver was abnormal in 62 percent of diabetic patients with symptoms and signs suggestive of autonomic neuropathy. When more generous criteria for abnormality were used, with a lower limit for normal of 1.50, the value was abnormal in 86 percent of these patients, and such an abnormality correlated well with the presence of a significant postural drop in blood pressure. Subsequent studies have shown that age- and gender-based normal values should be used. 91, 92

Other Cardiovascular Responses
Other cardiovascular responses can also be measured noninvasively (e.g., the beat-to-beat variation in heart rate and the heart rate responses to deep breathing and sustained hand grip). Such tests of parasympathetic function appear to give abnormal results more often and earlier than tests of sympathetic function, at least with the dysautonomia that occurs in diabetes. 87
A particularly useful test is to measure the heart rate variation during deep breathing ( Fig. 8-6 ). In normal subjects, there is considerable heart rate variation, which is accentuated during deep breathing. This variation is reduced or absent in diabetic patients with autonomic neuropathy. The optimal breathing rate for this test is six breaths per minute (i.e., inspiration = expiration = 5 seconds). Heart rate variation scores can be calculated by measuring the difference between the maximal and minimal heart rates in inspiration and expiration, taking the average from 10 breaths in and 10 breaths out. Normal subjects usually have a score greater than 9, and autonomic neuropathy is probably absent if scores greater than 12 are obtained 93 ; the normal range, however, is age dependent. 89, 91, 94, 95 Thus, Freeman indicated that heart rate variability may decline by 3 to 5 beats per minute per decade in normal subjects. 92 The use of a single normal value regardless of age may therefore limit the utility of the test. Physical fitness, body weight and position, time of testing, and concomitant medication may affect the test results.

FIGURE 8-6 Normal variation in heart rate that occurs in response to deep breathing.
An increase in heart rate and blood pressure should also occur in response to startle, such as occurs with a sudden loud noise, and to mental stress, as is produced when the patient attempts to subtract 7 serially from 100 while constantly being distracted.

Digital Blood-Flow
Blood flow to a finger can be measured by conventional plethysmography or photoplethysmography. A sudden inspiratory gasp causes reflex digital vasoconstriction as a spinal reflex, and this is easily measured plethysmographically ( Fig. 8-7 ). The response is impaired or absent in patients with a lesion of the cord or sympathetic efferent pathway, as in peripheral neuropathy. In entrapment neuropathy, such as carpal tunnel syndrome, the vasoconstrictor response may be abolished in fingers supplied by the affected nerve but not in those supplied by other nerves. 96

FIGURE 8-7 Variation in blood volume after a deep inspiration in a normal subject, measured photoplethysmographically by means of an infrared emitter and detector placed on the pad of the index finger. The bottom trace represents the sensor output after it has been amplified by the photoplethysmographic module of a computerized autonomic testing system; it is a function of the absolute blood volume in the finger. Each peak represents a heartbeat, and the amplitude of each wave reflects blood volume in the area about the sensor. The apparent shift of the direct-current signal component is due to the long time constant that is necessary so that signal information is not lost. The relative voltage, representing the amplitude of each pulse, is shown in the upper trace. It is evident in both traces that after the deep inspiration there is a reduction in digital blood flow (i.e., reduced amplitude of the waveforms in the lower trace and a corresponding decline in the upper trace).

Cold Pressor Test
In the cold pressor test, one hand is immersed in ice water at 4°C, and this normally produces an increase in systolic pressure of 15 mmHg or more within 1 minute. The afferent pathway involves the spinothalamic tract, and if this tract is intact, the lack of a pressor response suggests a lesion centrally or in the sympathetic efferent pathway. A normal response in a patient with an abnormal Valsalva response and intact pain and temperature sensation suggests an afferent baroreceptor lesion.

Norepinephrine Infusion
The plasma norepinephrine level can be used as an index of sympathetic activity. Perhaps of greater value, the blood pressure can be measured in response to intravenous infusion of norepinephrine at several dose rates up to 20 μg/min. 64 In this way, a dose–response curve can be constructed. In normal subjects, it is usually necessary to administer 15 to 20 μg/min to increase systolic blood pressure to 40 mmHg above baseline. A similar increase in blood pressure results from doses of 5 to 10 μg/min in Shy–Drager syndrome and less than 2.5 μg/min in patients with idiopathic orthostatic hypotension. 64

Response to Tyramine
Tyramine, an indirectly acting sympathomimetic drug, can be used to test neuronal uptake and release of norepinephrine. Bolus injections ranging from 250 to 6,000 μg are administered and blood pressure measured at 1-minute intervals. The amount of norepinephrine released into plasma by tyramine can be quantified by obtaining a blood sample shortly after the rise in blood pressure. 64

Norepinephrine Response to Edrophonium
Intravenously administered edrophonium (10 mg) normally leads within 8 minutes to the postganglionic release of norepinephrine and thus to an increase in plasma norepinephrine levels. In patients with central dysautonomias (and intact postganglionic function), a normal response is obtained, whereas patients with more peripheral disease have absent or attenuated responses. 97 The sensitivity and specificity of the test remain to be established.

Sweat Tests
Cutaneous blood vessels and sweat glands are supplied by sympathetic fibers intermingled in the same fascicles but of different size and conduction velocity. 14 Commonly used tests of sweating are messy and require application of heat, which is time-consuming. A heat cradle placed over the trunk is used to produce an increase of 1°C (from a resting level of 36.5° to 37.0°C) in the oral temperature over the course of 30 to 60 minutes, and the presence of sweat over selected regions of the trunk and limbs is detected by the change in color of quinizarine powder or a starch-iodide mixture that it produces. Changes in hand blood flow can also be measured with a plethysmograph. The pattern of any impairment of sweating may be helpful in suggesting the underlying cause. For example, impairment is usually distal in the limbs in patients with polyneuropathies.
The volume of sweat produced by axon-reflex stimulation either electrically (faradic sweat response) or with parasympathomimetic drugs under specified conditions indicates the state of sudomotor innervation in the tested limb. After a short latent period, sweating occurs in an area that is approximately 4 to 5 cm in diameter about the site of stimulation. The reflex is subserved by sympathetic postganglionic fibers; impulses pass centripetally along these fibers until they reach a branch point and then pass distally again. The receptor involved in the reflex has not been defined. 98
Rather than detecting a sweat response by a color change, as described previously, it may be necessary to quantify the volume of sweat. Recordings can then be made of the humidity change of an air stream of defined flow. 98 Using such an approach, the group at the Mayo Clinic have quantified the sudomotor response to axon reflex stimulation using electrophoresed acetylcholine to stimulate the receptors involved in the axon reflex.
Another simple technique for evaluating sudomotor function is to measure changes in skin resistance. With sweating, there is a reduction in skin resistance. This is the so-called galvanic skin response, which can be elicited by painful or emotional stimuli or by deep inspiration.
Sudomotor function has also been evaluated in patients with polyneuropathies by recording the change in voltage measured from the skin surface after deep inspiration or startle or after electrical stimuli applied to the skin of the wrist or ankle (sympathetic skin response). Responses are recorded from a pair of electrodes placed on the palm and dorsum of the hand or the sole and dorsum of the foot. 99 The sympathetic skin response is simple to record, but responses tend to habituate and are affected by the recording technique and a number of other factors. The absence of a response, and not the absolute values of latency or amplitude, is regarded as significant for determining abnormality. The normal latency in the upper limb is in the order of 1.5 seconds and in the lower limb is about 2 seconds, reflecting the slow conduction velocity of postganglionic C fibers (approximately 1 m/sec) Abnormalities of the sympathetic skin response reportedly correlate well with the quantitative sudomotor axon reflex test. 100

Other Studies

Pupillary Responses
Pupillary constriction with 2.5 percent methacholine applied locally indicates denervation supersensitivity due to interruption of postganglionic parasympathetic fibers, as in the Holmes–Adie syndrome. Local instillation of 1:1,000 epinephrine hydrochloride (one or two drops) produces little or no response unless there is postganglionic sympathetic denervation, in which case marked pupillary dilatation occurs. A 4 percent solution of cocaine hydrochloride applied to the conjunctival sac dilates the normal pupil, but fails to do so if sympathetic innervation has been interrupted outside the CNS.

Radiological Studies
Radiological studies may be helpful in characterizing gastrointestinal and bladder function but are beyond the scope of this chapter.

The initial investigative approach to patients presenting with syncope or other symptoms suggestive of postural hypotension or autonomic dysfunction is to exclude reversible causes such as hypovolemia or certain medications (see pp. 142 to 143 ). The history must include a detailed account of illnesses and drug intake. Simple laboratory investigations should include a full blood count and erythrocyte sedimentation rate as well as determination of plasma urea, electrolytes, glucose, morning and evening cortisol levels, and lying and standing catecholamine concentrations. Urinary screen for porphyrins; serum protein electrophoresis and immunophoresis; hepatic, renal, and thyroid function tests; chest radiograph; and electrocardiogram (to exclude recent cardiac infarction or cardiac ischemia, heart block, or persisting cardiac dysrhythmia) are also performed, as are serological tests for syphilis and nerve conduction studies. Neuroimaging studies may be helpful if a structural intracranial lesion is suspected. An echocardiogram may help when evaluating patients with suspected structural lesions of the heart predisposing to syncope. Prolonged tilt-table evaluations and invasive cardiac electrophysiological studies may be necessary when an arrhythmia is likely. In patients with symptoms of uncertain etiology in whom general medical causes have been excluded, more detailed evaluation of autonomic function in the manner suggested earlier may be helpful.

If a specific reversible cause, such as a metabolic or endocrinological disturbance, can be recognized, it must be treated appropriately. The need for continuing with drugs likely to be responsible should be reviewed and, if feasible, treatment discontinued. Patients should be advised against using alcohol. Treatment with antiarrhythmic agents, cardiac pacemaker, or surgery may be indicated in patients with a cardiac cause of syncope or postural hypotension. Pacemaker therapy may also help patients with syncope due to carotid sinus hypersensitivity. 83
If no specific cause can be identified, treatment should be directed to the minimization of symptoms ( Table 8-1 ). The actual extent to which the blood pressure falls on standing, for example, is of less significance than the occurrence of symptoms. Patients with dysautonomia should avoid extreme heat, alcohol, large meals, rapid postural changes, prolonged periods of recumbency, and excessive straining (e.g., during micturition or defecation), each of which may exacerbate symptoms. Diuretics should be stopped, if possible, and salt intake liberalized.
TABLE 8-1 Management of Postural Hypotension
Treatment of Specific Underlying Cause or Aggravating Factors
Discontinue drugs that may be responsible, if feasible
Correct electrolyte/metabolic/hormonal disorders
Avoid alcohol
Eliminate conditions that favor pooling of blood or that impede venous return
Prescribe antiarrhythmic drugs, pacemaker, or surgery for selected cardiac disorders
Consider a cardiac pacemaker for carotid sinus hypersensitivity
Symptomatic Treatment
Nonpharmacological management
Stand up gradually
Eat small meals and avoid postprandial activity
Wear waist-high elastic stockings
Elevate the head of the bed
Eat a liberal salt diet
Pharmacological and other treatment
Indomethacin; ibuprofen
Dihydroergotamine; Cafergot
Sympathomimetic drugs (phenylephrine, ephedrine, amphetamines)
β-Blocker drugs (propranolol, pindolol)
Cardiac pacing in selected circumstances
Other approaches
Norepinephrine (by infusion pump)
Deep brain stimulation
When standing, patients often find it helpful to work the leg muscles because this aids the venous return to the heart. Symptoms may also be reduced if patients stand up gradually (e.g., by first adopting the seated position and, after a short pause, getting up from this position). Other physical maneuvers that may be helpful include standing with legs crossed, bending forward, squatting, or placing a foot on a chair, thereby slightly increasing the mean arterial pressure so that cerebral blood flow remains adequate. The underlying common mechanism is held to be an increase of thoracic blood volume by transfer from below the diaphragm to the chest. 1 Resistance exercise may increase orthostatic tolerance, plasma volume, and baroreflex gain.
Waist-high elastic stockings may be helpful in alleviating postural symptoms but are often difficult to put on (especially for elderly patients) and may be uncomfortable in hot weather. To be effective, the stockings must extend at least as high as the waist. Antigravity suits have been used in the past but are awkward, restrictive, impractical, and not generally available.
Many dysautonomic patients have a disturbance in the regulation of body fluids. In particular, there is defective sodium conservation, especially during recumbency at night, associated with, but not entirely due to, low aldosterone levels 101 ; there are also abnormal posture-dependent changes in urine volume ( Fig. 8-8 ) accompanied by an alteration in the secretion of antidiuretic hormone. 102 This leads to relative hypovolemia and postural hypotension that are worse in the morning and improve during the day. The disturbed regulation of body fluids could be due, at least in part, to diminished adrenergic activity in the renal nerves, which affects tubular reabsorption and renin release (and thus angiotensin formation). The effect of recumbency can be minimized by elevating the head of the bed by 5 to 20 degrees, which leads to reduced renal artery pressure, thereby stimulating the renin-angiotensin system and promoting sodium retention. Head-up tilt at night reduces nocturnal shifts of interstitial fluid from the legs into the circulation; furthermore, such interstitial fluid may exert hydrostatic force, opposing the tendency of blood to pool in the legs on standing. 103 Head-up tilt at night also reduces supine hypertension.

FIGURE 8-8 Renal responses of five dysautonomic patients and four parkinsonian (control) subjects to fluid deprivation for 36 hours, commencing at 6 p.m. Average results are presented for urine osmolality and volume as well as potassium and sodium excretion for each successive 4-hour period. Subjects were lying down (white bars) during the night (10 p.m. to 10 a.m.) and were up and about (black bars) during the day (10 a.m. to 10 p.m.).
(From Wilcox CS, Aminoff MJ, Penn W: Basis of nocturnal polyuria in patients with autonomic failure. J Neurol Neurosurg Psychiatry 37:677, 1974, with permission.)
If these measures are unsuccessful, the mineralocorticoid fludrocortisone can be tried. This agent seems to exert its effect in part by temporarily increasing plasma volume and also by increasing vascular sensitivity to norepinephrine and improving the vasoconstrictor response to sympathetic stimulation. Treatment is usually commenced with a daily dose of 0.1 mg, which can then be increased by 0.1 mg every 2 weeks or so until benefit occurs or there are intolerable side effects. Some patients may require as much as 1.0 or 2.0 mg daily, but usually a dose of 0.5 mg or less is sufficient. During treatment with fludrocortisone, a positive sodium balance should be ensured, with a sodium intake of at least 150 mEq/day. Side effects include pedal edema, weight gain, recumbent hypertension, cardiomegaly, hypokalemia, and retinopathy; coexisting diabetes mellitus may also be exacerbated.
Prostaglandin synthetase inhibitors should expand plasma volume and inhibit vasodilator prostaglandin synthesis. Indomethacin (25 to 75 mg three times daily with meals) increases peripheral vascular resistance, promotes fluid retention, and may increase the sensitivity of the peripheral vasculature to norepinephrine and angiotensin II. 104 It is said to be helpful in some patients with postural hypotension, 105 especially if they are also on fludrocortisone, but, in general, the results with it have been disappointing despite the theoretical advantages of its use. Ibuprofen (200 to 600 mg four times daily before meals) can also be tried and is sometimes helpful. Side effects include gastric irritation, nausea, constipation, and skin rashes.
Dihydroergotamine is a relatively selective constrictor of peripheral veins. Its action may be mediated partially through α-adrenoreceptors, and enhanced synthesis of a vasoconstrictor prostaglandin may also be important. It is sometimes helpful for treating postural hypotension, but may cause recumbent hypertension. Although it is effective when administered intravenously, its efficacy when taken orally (5 to 10 mg three times daily) is more limited and variable. Inhaled preparations may be effective. 106 Cafergot (caffeine and ergotamine) suppositories have sometimes been effective. 107
Sympathomimetic drugs that either act directly to constrict blood vessels (e.g., phenylephrine) or that have an indirect action, preventing the destruction of norepinephrine at sympathetic nerve terminals (e.g., ephedrine and amphetamines), have been used to treat postural hypotension. These drugs can sometimes be helpful, but any benefit is often mild and temporary, and they may cause severe recumbent hypertension. Other side effects include nervousness, anxiety, restlessness, tachycardia, and tachyphylaxis. Midodrine is a direct α-adrenergic agonist that causes constriction of arterioles and venous vessels. It is started in a low daily dose (2.5 mg three times daily) that is built up gradually to 10 mg three times daily) depending on response and tolerance. Side effects include supine hypertension, piloerection, and pruritus. 108, 109 Pyridostigmine bromide may reduce postural hypotension by enhancing ganglionic transmission without aggravating or precipitating supine hypertension. A recent clinical trial showed that its greatest effect was on diastolic blood pressure, suggesting that improvement is due to increased total peripheral resistance. 110 Its role merits further study.
Propranolol may help the postural hypotension associated with postural tachycardia syndrome and also reduces sodium excretion, leading to an increase in blood volume. Furthermore, it has been said to help the hypotension of both idiopathic orthostatic hypotension and the Shy–Drager syndrome; this action has been attributed to the β-blockade correcting an imbalance of α- and β-adrenoreceptor activity in the peripheral nervous system. 111 It can be initiated at a dose of 10 mg four times daily and built up to 40 mg four times daily. Pindolol, a β-blocker with intrinsic sympathomimetic activity, has also been helpful in some instances, but more often it produces no benefit and may lead to cardiac failure. 112
Clinical benefit and an increase in blood pressure on standing in patients with sympathetic efferent failure may occur with cardiac pacing. 113 - 115 However, in a patient with more severe sympathetic involvement, Goldberg and colleagues failed to show any benefit with atrial tachypacing. 115 Bannister and associates reported a patient with selective impairment of sympathetic cardiovascular reflexes but relative sparing of cardiac parasympathetic function. 48 Because of accentuation of bradycardia and the risk of dysrhythmia with any elevation in blood pressure, it was thought that the use of pressor agents would be especially hazardous. A demand pacemaker was therefore introduced, and, although it did not improve postural hypotension, it enabled blood pressure to be readily and safely controlled by drugs. Atrial tachypacing may thus prove helpful in patients with selective sympathetic autonomic neuropathy by protecting against vagal overactivity.
Vasopressin responses to upright posture are often defective in autonomic failure, and patients are hypersensitive to exogenous vasopressin. The long-term therapeutic utility of vasopressin is unclear. Kochar studied the effect of lysine vasopressin nasal spray in 10 patients with chronic orthostatic hypotension and found no significant alteration in the effect of tilt on heart rate, stroke volume, or cardiac output; there was, however, an increase in blood pressure and total peripheral resistance. 116 In five patients with chronic autonomic failure who were not receiving treatment with drugs, Mathias and associates found that the vasopressin analogue desmopressin (given intramuscularly) reduced nocturnal polyuria, raised supine blood pressure, and reduced postural drops in blood pressure. 117 Intranasal desmopressin administered once at night in patients with multisystem atrophy and nocturnal polyuria has led to an improvement in nocturia without serious adverse effects. 118 If this approach is to be used, serum sodium should be monitored, especially during the first 4 to 5 days of treatment and then at monthly intervals.
Recombinant erythropoietin helps the mild anemia that is common in dysautonomic patients and also reduces postural hypotension. It is expensive, may require concomitant iron supplementation, and sometimes leads to supine hypertension, but can be tried in cases that are otherwise difficult to manage. It is given subcutaneously, with the dose and frequency of administration individualized.
There have been a variety of other experimental therapeutic approaches. Yohimbine, a centrally acting α 2 -antagonist, increases mean arterial pressure and plasma norepinephrine levels in normal subjects 119 and dysautonomic patients, 120 but its clinical utility is not clear. The use of a dopamine receptor antagonist, metoclopramide, was suggested because dopaminergic drugs (e.g., bromocriptine) may depress the blood pressure. 121 It is probably most helpful in patients with a diabetic dysautonomia, particularly if gastroparesis is also conspicuous. It is given in a dose of 5 to 10 mg three or four times daily, before meals, but long-term use may lead to tardive dyskinesia. In a few patients with pure autonomic failure, the selective partial α-agonist clonidine may be helpful, 122 but it has caused profound hypotension in a patient with baroreceptor dysfunction due to irradiation of the neck; it may reduce supine hypertension and nocturnal natriuresis. 123 It is probably most likely to help in patients with an efferent sympathetic lesion. It is started in a dose of 0.1 mg taken in the morning, and the dose is then built up gradually to 0.2 to 1 mg twice daily. Side effects include xerostomia, constipation, drowsiness, and supine hypertension; hypotension may be exacerbated in patients with non-neurological causes of postural hypotension. Administration of caffeine with meals may markedly reduce postprandial hypotension and is worthy of trial when symptoms are particularly troubling after meals. 115 A possible role for deep brain stimulation of the periventricular/periaqueductal gray region has been suggested. 124
Patients with vasovagal syncope require reassurance coupled with advice about ensuring adequate fluid and salt intake and about sympathetic activation techniques (such as isometric hand exercises) to increase the blood pressure 125 ; sitting or lying down or sitting with head between the knees may help to abort attacks.

General Precautions in Management of Dysautonomic Patients
Patients may show postprandial falls in blood pressure because blood is diverted to the hepatic and splanchnic beds. Vasoactive substances may also contribute to the hypotensive response. To avoid or minimize this postprandial hypotension, it is helpful to eat smaller meals and to avoid excessive activity during the immediate postprandial period.
Dysautonomic patients often have low circulating catecholamine levels and denervation supersensitivity to sympathomimetic amines. Medications containing such substances should therefore be avoided, even though they are often available without prescription in over-the-counter preparations.
Patients with dysautonomia pose special problems during anesthesia. They are unable to tolerate hemodynamic stresses normally because of impaired cardiovascular reflexes. Maintenance of fluid balance is more difficult because of the abnormal manner in which they handle salt and water, and their enhanced sensitivity to volume changes influences blood pressure control.


1 Wieling W, van Lieshout JJ, van Leeuwen AM. Physical manoeuvres that reduce postural hypotension in autonomic failure. Clin Auton Res . 1993;3:57.
2 Schoenberger JA. Drug-induced orthostatic hypotension. Drug Saf . 1991;6:402.
3 Nanji AA, Freeman HJ. Postural hypotension and labile blood pressure associated with severe hypophosphatemia. Int J Cardiol . 1984;5:379.
4 Alfrey C. Regulation of red cell volume at microgravity. Med Sci Sports Exerc . 1996;28(suppl):542.
5 Convertino VA. Conditions of reduced gravity. In: Low PA, editor. Clinical Autonomic Disorders . 2nd Ed. Philadelphia: Lippincott-Raven; 1997:429.
6 Craven CG, Engelke KA, Pawelczyk JA, et al. Power spectral and time-based analysis of heart rate variability following 15 days of head-down bed rest. Aviat Space Environ Med . 1994;65:1105.
7 Low PA. The effect of aging on the autonomic nervous system. In: Low PA, editor. Clinical Autonomic Disorders . 2nd Ed. Philadelphia: Lippincott-Raven; 1997:161.
8 Mattace-Raso FU, van der Cammen TJ, Knetsch AM, et al. Arterial stiffness as the candidate underlying mechanism for postural blood pressure changes and orthostatic hypotension in older adults: the Rotterdam Study. J Hypertens . 2006;24:339.
9 Weiss A, Beloosesky Y, Kornowski R, et al. Influence of orthostatic hypotension on mortality among patients discharged from an acute geriatric ward. J Gen Intern Med . 2006;21:602.
10 Spyer KM. Physiology of the autonomic nervous system: CNS control of the cardiovascular system. Curr Opin Neurol Neurosurg . 1991;4:528.
11 Loewy AD. Anatomy of the autonomic nervous system: an overview. In: Loewy AD, Spyer KM, editors. Central Regulation of Autonomic Functions . New York: Oxford University Press; 1990:3.
12 Andersen MC, Kunze DL. Nucleus tractus solitarius: gateway to neural circulatory control. Annu Rev Physiol . 1994;56:93.
13 Dampney RA. Functional organization of central pathways regulating the cardiovascular system. Physiol Rev . 1994;74:323.
14 Wallin BG, Elam M. Microneurography and autonomic dysfunction. In: Low PA, editor. Clinical Autonomic Disorders . 2nd Ed. Philadelphia: Lippincott-Raven; 1997:233.
15 Sheriff DD, O’Leary DS, Scher AM, et al. Baroreflex attenuates pressor response to graded muscle ischemia in exercising dogs. Am J Physiol . 1990;258:H305.
16 Smith SA, Mitchell JH, Garry MG. The mammalian exercise pressor reflex in health and disease. Exp Physiol . 2006;91:89.
17 Joyner MJ, Shepherd JT. Autonomic regulation of circulation. In: Low PA, editor. Clinical Autonomic Disorders . 2nd Ed. Philadelphia: Lippincott-Raven; 1997:61.
18 Eldridge FL, Millhorn DE, Kiley JP, et al. Stimulation by central command of locomotion, respiration and circulation during exercise. Respir Physiol . 1985;59:313.
19 Nogues MA, Newman PK, Male VJ, et al. Cardiovascular reflexes in syringomyelia. Brain . 1982;105:835.
20 Heidel KM, Benarroch EE, Gene R, et al. Cardiovascular and respiratory consequences of bilateral involvement of the medullary intermediate reticular formation in syringobulbia. Clin Auton Res . 2002;12:450.
21 Hsu CY, Hogan EL, Wingfield W, et al. Orthostatic hypotension with brainstem tumors. Neurology . 1984;34:1137.
22 Weig SG, Buckthal PE, Choi SK, et al. Recurrent syncope as the presenting symptom of Arnold-Chiari malformation. Neurology . 1991;41:1673.
23 Goldstein DS, Holmes CS, Dendi R, et al. Ortho-static hypotension from sympathetic denervation in Parkinson’s disease. Neurology . 2002;58:1247.
24 Holmberg B, Kallio M, Johnels B, et al. Cardiovascular reflex testing contributes to clinical evaluation and differential diagnosis of parkinsonian syndromes. Mov Disord . 2001;16:217.
25 Sandroni P, Ahlskog JE, Fealey RD, et al. Autonomic involvement in extrapyramidal and cerebellar disorders. Clin Auton Res . 1991;1:147.
26 Gutrecht JA. Autonomic cardiovascular reflexes in progressive supranuclear palsy. J Auton Nerv Syst . 1992;39:29.
27 Aminoff MJ. Other extrapyramidal disorders. In: Low PA, editor. Clinical Autonomic Disorders . 2nd Ed. Philadelphia: Lippincott-Raven; 1997:577.
28 Andrich J, Schmitz T, Saft C, et al. Autonomic nervous system function in Huntington’s disease. J Neurol Neurosurg Psychiatry . 2002;72:726.
29 Thomaides TN, Zoukos Y, Chaudhuri KR, et al. Physiological assessment of aspects of autonomic function in patients with secondary progressive multiple sclerosis. J Neurol . 1993;240:139.
30 Haensch CA, Jorg J. Autonomic dysfunction in multiple sclerosis. J Neurol . 2006;253(suppl 1):I3.
31 Norrving B, Cronqvist S. Lateral medullary infarction: prognosis in an unselected series. Neurology . 1991;41:244.
32 Low PA, Benrud-Larson LM, Sletten DM, et al. Autonomic symptoms and diabetic neuropathy: a population-based study. Diabetes Care . 2004;27:2942.
33 Freeman R. Autonomic peripheral neuropathy. Lancet . 2005;365:1259.
34 Low VA, Sandroni P, Fealey RD, et al. Detection of small-fiber neuropathy by sudomotor testing. Muscle Nerve . 2006;34:57.
35 Duncan G, Johnson RH, Lambie DG, et al. Evidence of vagal neuropathy in chronic alcoholics. Lancet . 1980;2:1053.
36 Bernardi L, Passino C, Porta C, et al. Widespread cardiovascular autonomic dysfunction in primary amyloidosis: does spontaneous hyperventilation have a compensatory role against postural hypotension? Heart . 2002;88:615.
37 McCombe PA, McLeod JG. The peripheral neuropathy of vitamin B 12 deficiency. J Neurol Sci . 1984;66:117.
38 Eisenhofer G, Lambie DG, Johnson RH, et al. Deficient catecholamine release as the basis of orthostatic hypotension in pernicious anemia. J Neurol Neurosurg Psychiatry . 1982;45:1053.
39 Wilder-Smith EP, Wilder-Smith AJ, Nirkko AC. Skin and muscle vasomotor reflexes in detecting autonomic dysfunction in leprosy. Muscle Nerve . 2000;23:1105.
40 Abbot NC, Beck JS, Mostofi S, et al. Sympathetic vasomotor dysfunction in leprosy patients: comparison with electrophysiological measurement and qualitative sensation testing. Neurosci Lett . 1996;206:57.
41 Gudesblatt M, Goodman AD, Rubenstein AE, et al. Autonomic neuropathy associated with autoimmune disease. Neurology . 1985;35:261.
42 Cable WJL, Kolodny EH, Adams RD. Fabry disease: impaired autonomic function. Neurology . 1982;32:498.
43 Singh NK, Jaiswal AK, Misra S, et al. Assessment of autonomic dysfunction in Guillain-Barré syndrome and its prognostic implications. Acta Neurol Scand . 1987;75:101.
44 Lyu RK, Tang LM, Wu YR, et al. Cardiovascular autonomic function and sympathetic skin response in chronic inflammatory demyelinating polyradiculoneuropathy. Muscle Nerve . 2002;26:669.
45 Stamboulis E, Katsaros N, Koutsis G, et al. Clinical and subclinical autonomic dysfunction in chronic inflammatory demyelinating polyradiculoneuropathy. Muscle Nerve . 2006;33:78.
46 Mori K, Iijima M, Koike H, et al. The wide spectrum of clinical manifestations in Sjogren’s syndrome-associated neuropathy. Brain . 2005;128:2518.
47 Sandhu V, Allen SC. The effects of age, seropositivity and disease duration on autonomic cardiovascular reflexes in patients with rheumatoid arthritis. Int J Clin Pract . 2004;58:740.
48 Bannister R, Da Costa DF, Hendry WG, et al. Atrial demand pacing to protect against vagal overactivity in sympathetic autonomic neuropathy. Brain . 1986;109:345.
49 Etienne M, Weimer LH. Immune-mediated autonomic neuropathies. Curr Neurol Neurosci Rep . 2006;6:57.
50 Axelrod FB. Familial dysautonomia. Muscle Nerve . 2004;29:352.
51 Axelrod FB. Familial dysautonomia: a review of the current pharmacological treatments. Expert Opin Pharmacother . 2005;6:561.
52 Stojkovic T, de Seze J, Dubourg O, et al. Autonomic and respiratory dysfunction in Charcot-Marie-Tooth disease due to Thr124Met mutation in the myelin protein zero gene. Clin Neurophysiol . 2003;114:1609.
53 Rechthand E, Reife R, Kaplan JG. Hereditary neuropathy with upper motor-neuron, visual pathway, and autonomic disorders. Neurology . 1983;33:1495.
54 Poon IO, Braun U. High prevalence of orthostatic hypotension and its correlation with potentially causative medications among elderly veterans. J Clin Pharm Ther . 2005;30:173.
55 Low PA, Vernino S, Suarez G. Autonomic dysfunction in peripheral nerve disease. Muscle Nerve . 2003;27:646.
56 Matikainen E, Juntunen J. Autonomic nervous system dysfunction in workers exposed to organic solvents. J Neurol Neurosurg Psychiatry . 1985;48:1021.
57 Sterman AB, Panasci DJ, Sheppard RC. Autonomic-cardiovascular dysfunction accompanies sensory-motor impairment during acrylamide intoxication. Neurotoxicology . 1983;4:45.
58 MacFarlane IA, Wilkinson R, Harrington JM. Severe postural hypotension following home canoe construction from polyester resins. Postgrad Med J . 1984;60:497.
59 Benowitz NL, Byrd R, Schambelan M, et al. Dihydroergotamine treatment for orthostatic hypotension from Vacor rodenticide. Ann Intern Med . 1980;92:387.
60 Consensus Committee of the American Autonomic Society and the American Academy of Neurology: Consensus statement on the definition of orthostatic hypotension, pure autonomic failure, and multiple system atrophy. Neurology . 1996;46:1470.
61 Mabuchi N, Hirayama M, Koike Y, et al. Progression and prognosis in pure autonomic failure (PAF): comparison with multiple system atrophy. J Neurol Neurosurg Psychiatry . 2005;76:947.
62 Young TM, Mathias CJ. The effects of water ingestion on orthostatic hypotension in two groups of chronic autonomic failure: multiple system atrophy and pure autonomic failure. J Neurol Neurosurg Psychiatry . 2004;75:1737.
63 Polinsky RJ, Kopin IJ, Ebert MH, et al. Pharmacologic distinction of different orthostatic hypotension syndromes. Neurology . 1981;31:1.
64 Polinsky RJ. Clinical autonomic neuropharmacology. Neurol Clin . 1990;8:77.
65 Sever PS. Plasma noradrenalin in autonomic failure. In: Bannister R, editor. Autonomic Failure . Oxford: Oxford University Press; 1983:155.
66 Liard JF. Vasopressin in cardiovascular control: role of circulating vasopressin. Clin Sci . 1984;67:473.
67 Robertson GL. Antidiuretic hormone. Normal and disordered function. Endocrinol Metab Clin North Am . 2001;30:671.
68 Puritz R, Lightman SL, Wilcox CS, et al. Blood pressure and vasopressin in progressive autonomic failure. Brain . 1983;106:503.
69 Williams TDM, Lightman SL, Bannister R. Vasopressin secretion in progressive autonomic failure: evidence for defective afferent cardiovascular pathways. J Neurol Neurosurg Psychiatry . 1985;48:225.
70 Deguchi K, Sasaki I, Touge T, et al. Abnormal baroreceptor-mediated vasopressin release as possible marker in early diagnosis of multiple system atrophy. J Neurol Neurosurg Psychiatry . 2004;75:110.
71 Benarroch EE, Schmeichel AM, Sandroni P, et al. Differential involvement of hypothalamic vasopressin neurons in multiple system atrophy. Brain . 2006;129(pt 10):2688.
72 Timmers HJ, Deinum J, Wevers RA, et al. Congenital dopamine-beta-hydroxylase deficiency in humans. Ann N Y Acad Sci . 2004;1018:520.
73 Biaggioni I, Goldstein DS, Atkinson T, et al. Dopamine-β-hydroxylase deficiency in humans. Neurology . 1990;40:370.
74 Schondorf R, Low PA. Idiopathic postural orthostatic tachycardia syndrome: an attenuated form of acute pandysautonomia? Neurology . 1993;43:132.
75 Jacob G, Biaggioni I. Idiopathic orthostatic intolerance and postural tachycardia syndromes. Am J Med Sci . 1999;317:88.
76 Jacob G, Costa F, Shannon JR, et al. The neuropathic postural tachycardia syndrome. N Engl J Med . 2000;343:1008.
77 Sandroni P, Novak V, Opfer-Gehrking TL, et al. Mechanism of blood pressure alterations in response to the Valsalva maneuver in postural tachycardia syndrome. Clin Auton Res . 2000;10:1.
78 Almquist A, Goldenberg IF, Milstein S, et al. Provocation of bradycardia and hypotension by isoproterenol and upright posture in patients with unexplained syncope. N Engl J Med . 1989;320:346.
79 Kaufmann H. Neurally mediated syncope: pathogenesis, diagnosis, and treatment. Neurology . 1995;45(suppl 5):S12.
80 Wallin BG, Sundlof G. Sympathetic outflow to muscles during vasovagal syncope. J Auton Nerv Syst . 1982;6:287.
81 Gawrieh S, Carroll T, Hogan WJ, et al. Swallow syncope in association with Schatzki ring and hypertensive esophageal peristalsis: report of three cases and review of the literature. Dysphagia . 2005;20:273.
82 Omi W, Murata Y, Yaegashi T, et al. Swallow syncope, a case report and review of the literature. Cardiology . 2006;105:75.
83 Lipsitz LA. Syncope in the elderly. Ann Intern Med . 1983;99:92.
84 Mathias CJ, Deguchi K, Schatz I. Observations on recurrent syncope and presyncope in 641 patients. Lancet . 2001;357:348.
85 Kusumoto F, Goldschlager N. Implantable cardiac arrhythmia devices—Part II: implantable cardioverter defibrillators and implantable loop recorders. Clin Cardiol . 2006;29:237.
86 Mann S, Altman DG, Raftery EB, et al. Circadian variation of blood pressure in autonomic failure. Circulation . 1983;68:477.
87 Ewing DJ, Campbell IW, Clarke BF. Assessment of cardiovascular effects in diabetic autonomic neuropathy and prognostic implications. Ann Intern Med . 1980;92(part 2):308.
88 Ziegler D, Laux G, Dannehl K, et al. Assessment of cardiovascular autonomic function: age-related normal ranges and reproducibility of spectral analysis, vector analysis, and standard tests of heart rate variation and blood pressure responses. Diabet Med . 1992;9:166.
89 Vita G, Princi P, Calabro R, et al. Cardiovascular reflex tests: assessment of age-adjusted normal range. J Neurol Sci . 1986;75:263.
90 Ewing DJ, Campbell IW, Burt AA, et al. Vascular reflexes in diabetic autonomic neuropathy. Lancet . 1973;2:1354.
91 Low PA, Denq JC, Opfer-Gehrking TL, et al. Effect of age and gender on sudomotor and cardiovagal function and blood pressure response to tilt in normal subjects. Muscle Nerve . 1997;20:1561.
92 Freeman R. Assessment of cardiovascular autonomic function. Clin Neurophysiol . 2006;117:716.
93 Watkins PJ, MacKay JD. Cardiac denervation in diabetic neuropathy. Ann Intern Med . 1980;92(part 2):304.
94 Smith SA. Reduced sinus arrhythmia in diabetic autonomic neuropathy: diagnostic value of an age-related normal range. BMJ . 1982;285:1599.
95 Kaijser L, Sachs C. Autonomic cardiovascular responses in old age. Clin Physiol . 1985;5:347.
96 Aminoff MJ. Involvement of peripheral vasomotor fibres in carpal tunnel syndrome. J Neurol Neurosurg Psychiatry . 1979;42:649.
97 Gemmill JD, Venables GS, Ewing DJ. Noradrenaline response to edrophonium in primary autonomic failure: distinction between central and peripheral damage. Lancet . 1988;1:1018.
98 Low PA, Caskey PE, Tuck RR, et al. Quantitative sudomotor axon reflex test in normal and neuropathic subjects. Ann Neurol . 1983;14:573.
99 Shahani BT, Halperin JJ, Boulu P, et al. Sympathetic skin response—a method of assessing unmyelinated axon dysfunction in peripheral neuropathies. J Neurol Neurosurg Psychiatry . 1984;47:536.
100 Maselli RA, Jaspan JB, Soliven BC, et al. Comparison of sympathetic skin response with quantitative sudomotor axon reflex test in diabetic neuropathy. Muscle Nerve . 1989;12:420.
101 Wilcox CS, Aminoff MJ, Slater JDH. Sodium homeostasis in patients with autonomic failure. Clin Sci Mol Med . 1977;53:321.
102 Wilcox CS, Aminoff MJ, Penn W. Basis of nocturnal polyuria in patients with autonomic failure. J Neurol Neurosurg Psychiatry . 1974;37:677.
103 Onrot J, Goldberg MR, Hollister AS, et al. Management of chronic orthostatic hypotension. Am J Med . 1986;80:454.
104 Davies IB, Bannister R, Hensby C, et al. The pressor actions of noradrenaline and angiotensin II in chronic autonomic failure treated with indomethacin. Br J Clin Pharmacol . 1980;10:223.
105 Kochar MS, Itskovitz HD. Treatment of idiopathic orthostatic hypotension (Shy-Drager syndrome) with indomethacin. Lancet . 1978;1:1011.
106 Biaggioni I, Zygmunt D, Haile V, et al. Pressor effect of inhaled ergotamine in orthostatic hypotension. Am J Cardiol . 1990;65:89.
107 Toh V, Duncan E, Lewis N, et al. Ergotamine use in severe diabetic autonomic neuropathy. Diabet Med . 2006;23:574.
108 Jankovic J, Gilden JL, Hiner BC, et al. Neurogenic orthostatic hypotension: a double-blind, placebo-controlled study with midodrine. Am J Med . 1993;95:38.
109 Low PA, Gilden JL, Freeman R, et al. Efficacy of midodrine vs placebo in neurogenic orthostatic hypotension. JAMA . 1997;277:1046.
110 Singer W, Sandroni P, Opfer-Gehrking TL, et al. Pyridostigmine treatment trial in neurogenic orthostatic hypotension. Arch Neurol . 2006;63:513.
111 Brevetti G, Chiariello M, Giudice P, et al. Effective treatment of orthostatic hypotension by propranolol in the Shy-Drager syndrome. Am Heart J . 1981;102:938.
112 Davies B, Bannister R, Mathias C, et al. Pindolol in postural hypotension: the case for caution. Lancet . 1981;2:982.
113 Moss AJ, Glaser W, Topol E. Atrial tachypacing in the treatment of a patient with primary orthostatic hypotension. N Engl J Med . 1980;302:1456.
114 Kristinsson A. Programmed atrial pacing for orthostatic hypotension. Acta Med Scand . 1983;214:79.
115 Goldberg MR, Robertson RM, Robertson D. Atrial tachypacing for primary orthostatic hypotension. N Engl J Med . 1980;303:885.
116 Kochar MS. Hemodynamic effects of lysine-vasopressin in orthostatic hypotension. Am J Kidney Dis . 1985;6:49.
117 Mathias CJ, Fosbraey P, Da Costa DF, et al. The effect of desmopressin on nocturnal polyuria, overnight weight loss, and morning postural hypotension in patients with autonomic failure. BMJ . 1986;293:353.
118 Sakakibara R, Matsuda S, Uchiyama T, et al. The effect of intranasal desmopressin on nocturnal waking in urination in multiple system atrophy patients with nocturnal polyuria. Clin Auton Res . 2003;13:106.
119 Goldberg MR, Hollister AS, Robertson D. Influence of yohimbine on blood pressure, autonomic reflexes, and plasma catecholamines in humans. Hypertension . 1983;5:772.
120 Onrot J, Goldberg MR, Biaggioni I, et al. Oral yohimbine in human autonomic failure. Neurology . 1987;37:215.
121 Kuchel O, Buu NT, Gutkowska J, et al. Treatment of severe orthostatic hypotension by metoclopramide. Ann Intern Med . 1980;93:841.
122 Robertson D, Goldberg MR, Hollister AS, et al. Clonidine raises blood pressure in severe idiopathic orthostatic hypotension. Am J Med . 1983;74:193.
123 Shibao C, Gamboa A, Abraham R, et al. Clonidine for the treatment of supine hypertension and pressure natriuresis in autonomic failure. Hypertension . 2006;47:522.
124 Green AL, Wang S, Owen SL, et al. Controlling the heart via the brain: a potential new therapy for orthostatic hypotension. Neurosurgery . 2006;58:1176.
125 Mathias CJ. Autonomic diseases: management. J Neurol Neurosurg Psychiatry . 2003;74(suppl 3):42.
Chapter 9 Neurological Complications of Cardiac Arrest

W.T. Longstreth, JR

General Issues and Physical Examination
Ancillary Tests
Application of Prognostic Information
Minimizing Insult
Brain Resuscitation

Not long ago, cardiac arrest was an irreversible event that meant certain death. Mouth-to-mouth resuscitation, described since biblical times, 1 or other types of artificial ventilation alone may save a victim of respiratory arrest but not cardiac arrest. The understanding of the causes and treatment of cardiac arrest is intriguingly intertwined with the study of electricity. By the middle of the nineteenth century, investigators knew that electrical currents caused the ventricles of animals to be thrown into a state of Herzdelirium or fibrillation. A series of experiments by Prevost and Battelli in dogs in 1899 established that electrical current could not only induce ventricular fibrillation but also reverse it, 2 leading Jex-Blake to conclude in the 1913 Goulstonian lectures: “It is more than probable that the same treatment—a hair of the dog that bit them—could be applied with success to human beings apparently killed by electric currents.” 3
Progress was interrupted by World War I, and interest was not rekindled until the 1920s in the United States. 4 Concern about accidental death among linemen led Consolidated Edison Company of New York City to request help from the Rockefeller Institute. Several investigations were initiated, including in 1928 work at Johns Hopkins University by Kouwenhoven, an electrical engineer, and Langworthy, a neurologist. 5 They confirmed earlier works indicating that lower voltage shocks induced ventricular fibrillation and higher voltage shocks caused respiratory arrest. However, they were unable to reverse the ventricular fibrillation with various treatments. In 1930, a colleague pointed out to them the 1899 works by Provost and Battelli, which they subsequently confirmed. The Johns Hopkins investigators returned to Edison Electric Institute to have them create a portable defibrillator.
One of the defibrillators subsequently developed had paddles that had to be pushed against the chest before the current could be discharged. Investigators noted that forceful application of the paddles to the chest wall of dogs led to a rise in blood pressure while the heart was still fibrillating. This serendipitous observation led to closed chest compressions. Now the stage was set to treat effectively a common cause of sudden cardiac death—ventricular fibrillation. Closed chest compression and mouth-to-mouth respiration would keep the patient viable long enough for the portable defibrillator to be brought to the patient and applied. Progress was again interrupted by World War II, and it was not until 1956 that Zoll and colleagues reported the first successful external defibrillation in humans. 6 In 1960, Kouwenhoven and co-workers reported a series of successful resuscitations in hospital with closed chest cardiac massage and external defibrillation. 7 Several epidemiological aspects of sudden cardiac death led to the application of these same techniques outside the hospital. 8 First, the importance of heart disease as a cause of death was increasingly recognized. Second, sudden cardiac death was the presenting feature in up to one third of people with heart disease. Finally, many people dying of heart disease, especially in the setting of myocardial infarction, were doing so outside the hospital.
One of the first prehospital emergency medical systems was developed in Belfast, Ireland. 9 - 11 In their 1967 report, Pantridge and Geddes describe the out-of-hospital resuscitation of a 55-year-old man who had collapsed while dancing. A bystander performed chest compressions, and personnel from the mobile unit applied cardioversion to restore spontaneous circulation. 10 Unfortunately, the patient died a week later with brain damage. A similar emergency medical system began in Seattle, Washington, in 1970, 12, 13 and systems rapidly proliferated throughout the United States and elsewhere. Meanwhile, selective application of cardiopulmonary resuscitation in hospital had become common as patients, families, and clinicians tried to decide who was likely to benefit from this technique. Over these decades, the realization has grown that the outcome among those in whom spontaneous circulation can be restored is often dominated by brain damage suffered during the arrest. 14 - 16
Between 400,000 and 450,000 people are estimated to experience sudden cardiac death out of hospital or in the emergency department each year in the United States. 17 The numbers treated by emergency medical systems is lower and estimated to be 155,000 people for all rhythms based on an incidence of 55 per 100,000 person-years and 60,000 for ventricular fibrillation based on an incidence of 21 per 100,000 person-years. 18 In this review including studies from 1980 to 2004, survival was 8.4 percent for all rhythms and 17.7 percent for ventricular fibrillation. In many regions of the United States, cardiac arrest has become a leading cause of coma, along with head trauma and drug overdose.
The goal of this chapter is to review the neurological sequelae of cardiac arrest (outcomes) and address linked issues of predicting outcomes (prognostication) and improving outcomes (treatment). The focus is on adults, and extreme caution should be used in extrapolating to children. 19


To understand the outcomes of cardiac arrest, the insults must be understood. Interruption of cardiac function, as evidenced by an absent pulse, causes insufficient blood flow to the brain leading to global brain ischemia and global brain dysfunction, as evidenced by an absence of consciousness. The dysfunction can be transient or permanent, depending in large part on the duration and severity of the ischemia. At the extreme, the ischemia may be complete where the initial electrocardiogram (ECG) reveals ventricular fibrillation or asystole. With other pulseless conditions, such as rapid ventricular tachycardia or pulseless electrical activity, the ischemia may be incomplete or unknown. During the resuscitation of individuals with cardiac arrest, the inability of medical personnel to detect a pulse or blood pressure does not preclude the possibility of severe hypotension, which could be detected if more sensitive means of measuring blood pressure were available. In most cardiac arrests, both incomplete (low flow) and complete global brain ischemia (no flow) contribute to the brain dysfunction and possible injury. For example, a patient with myocardial ischemia may have hypotension (low flow) preceding deterioration to ventricular fibrillation (no flow). During cardiopulmonary resuscitation and basic life support, severe hypotension can exist (low flow). Finally, with restoration of spontaneous circulation, hypotension and cardiogenic shock (low flow) can complicate the hospital course. The duration and severity of these components of global brain ischemia probably determine the degree of brain damage and outcome. Unfortunately, duration and severity are usually unknown.
Cardiac arrest most commonly results from a primary problem with the heart. Although the list of potential cardiac problems is long, coronary artery disease and myocardial ischemia are the most likely culprits. 20 Cardiac arrest can also occur secondary to other toxic and metabolic problems. With the possible exception of hypothermia and overdoses of sedative drugs, these other problems can augment the insult to the brain, especially when a respiratory arrest precedes a cardiac arrest. The respiratory arrest may occur in many settings and is often not a simple event in itself, as with drowning, decompensation of chronic lung disease, drug overdoses, hanging, strangulation, trauma, asphyxia, and carbon monoxide poisoning. In the absence of a cytotoxic agent, such as carbon monoxide, the hypoxia of respiratory insufficiency can be surprisingly well tolerated by the brain. 21 However, if a cardiac arrest ensues, the resulting brain damage can be devastating.
Primary neurological problems, such as focal brain ischemia, hemorrhage, and trauma, can lead to a cardiac arrest, which most commonly results from an initial respiratory arrest. Rarely will the first rhythm be ventricular fibrillation. Some neurological events can result in cardiac arrest without intervening respiratory arrest. Brain–heart interactions are detailed in Chapter 10 . Their importance is supported by experiments in pigs. 22 When pigs were placed under psychological stress, occlusion of the left anterior descending coronary artery resulted in ventricular fibrillation. Functional blockage with cryoprobes of pathways from the frontal cortex through the posterior hypothalamus to the brainstem prevented or delayed the development of ventricular fibrillation. Such pathways may play a role in the sudden death experienced by some patients with epilepsy in whom a fatal arrhythmia may be the terminal event. 23
Thomas mused on the possibility that mechanisms have evolved to ensure sudden death when the end of an animal’s life is near. 24 If such mechanisms exist and are triggered inappropriately, premature sudden death could result. Psychological factors may be important in triggering these responses. Richter presented an experimental model in rats of the influence of psychological factors on sudden death and proposed that deaths in humans due to hexing could represent the same phenomena. 25 All such theories assume a strong interaction among cardiac, psychological, and neurological factors.

The consequences of global brain ischemia have intrigued investigators since well before resuscitation from cardiac arrest was a reality. In 1836, Cooper, an English physician, carried out experiments on rabbits. 26 After surgically occluding both common carotid arteries, he noted that pinching of the vertebral arteries led to sudden unconsciousness. If the blood flow to the brain was not interrupted for too long, the rabbit regained consciousness rapidly upon release of the vertebral arteries. These seminal observations were extended by Brown-Séquard, who introduced the concept of a rostral-caudal deterioration of brain function with global ischemia. 27 Simply put, cerebral cortical function was lost before brainstem function. In turn, upper brainstem function, such as the pupillary light reflex, was lost before lower brainstem function, such as spontaneous respiratory efforts. The clinical stimulus for these early investigations is uncertain, given that the only examples of global brain ischemia recognized in humans at the time were conditions like beheading and hanging.
Confirmation that cerebral cortical function was lost before brainstem function with global ischemia came from additional studies in the 1940s. These studies entailed rapid inflation of a cuff about the neck of human volunteers, with interruption of blood flow to the brain. Subjects were rendered rapidly unconscious. 28, 29 In a 1957 study in which they recorded electroencephalograms (EEGs) on 100 patients with syncope, Gastaut and Fischer-Williams attempted to induce transient cardiac arrest with vagal stimulation induced by ocular compression. 30 “When cardiac arrest lasted more than about 14 sec., one or two generalized clonic jerks appeared without affecting the E.E.G., followed by generalized tonic contraction resembling decerebrate rigidity and accompanied by complete ‘flattening’ of the E.E.G.” More recent examples include observations made during induction of ventricular arrhythmias to test automatic implantable defibrillators 31, 32 and head-up tilt testing for syncope. 33 Although the electrocerebral accompaniments of global ischemia may be more complex than suggested in the earlier studies, all studies agree that motor activity early in the course of global brain ischemia is common and not related to cerebral cortical hyperactivity. Careful video observations have also been made in patients rendered unconscious following hyperventilation, orthostasis, and the Valsalva maneuver to induce global brain ischemia. 34, 35 In addition to the motor activity, various eye movement abnormalities, including downbeat nystagmus, were documented. The cerebellum, in addition to the cerebral cortex, was held to be sensitive to brief global ischemia.
Given the observations that the cerebral cortex is more sensitive than the brainstem to global ischemia, the major clinical outcomes can be anticipated. 36, 37 Based on experiments of global ischemia in primates, Nemoto and colleagues proposed S-shaped curves relating the duration of ischemia with the degree of brain damage and thus with neurological deficits. 38 Assuming that similarly shaped curves exist for the human brain and that for any duration of ischemia, except at extremes, damage to the cerebral cortex is greater than damage to the brainstem, Figure 9-1 can be constructed. The outcomes predicted are similar to those defined in the Glasgow Outcome Scale. 39 At extreme durations of cardiac arrest, both the cerebrum and brainstem will be totally destroyed and brain death will result. Such an outcome after a primary cardiac arrest, as with ventricular fibrillation or asystole, is uncommon and probably reflects the inability to resuscitate individuals whose arrest has such a long duration. In a study of 459 consecutive patients admitted to one hospital in Seattle after out-of-hospital cardiac arrest and resuscitation, only about 1 percent had an outcome consistent with brain death. 16 With shorter durations of arrest, the cerebral cortex may still be entirely destroyed, but without as complete a devastation of the brainstem. Such individuals will not be brain dead in that they retain some brainstem function, but they may have evidence of forebrain failure, such as electrocerebral inactivity on the EEG. 40 Various terms have been applied to this type of brain damage, including cortex death, neocortical death, partial brain death, death of the forebrain, and acute failure of forebrain with sparing of brainstem function. Although this condition is well described after cardiac arrest, it also is unusual.

FIGURE 9-1 Hypothetical relationship between duration of ischemia and brain damage, assuming that cerebral cortex is more vulnerable to global ischemia than brainstem. 36 - 38 For example, in a study of 459 patients admitted to a single hospital in Seattle after out-of-hospital cardiac arrest, 279 (61%) awakened, of whom about two thirds regained independence. 16
More commonly, the cerebral cortex is severely damaged but not completely destroyed. The damage is extensive enough to preclude the patient ever regaining consciousness even though evidence of cerebral cortical activity may exist on the EEG. Evaluation of brainstem function may reveal some deficits or no abnormality. Spontaneous eye opening is common, and such patients are said to be in a vegetative state. 41, 42 In one survey of four nursing homes in Milwaukee, Wisconsin, about 3 percent of patients were in a vegetative state. 43 In about 20 percent, the vegetative state resulted from a respiratory or cardiac arrest. A vegetative state that has lasted at least 1 month can be termed persistent. 44 The time at which it can be considered permanent is a topic of debate, as detailed in a subsequent section. Some aspects of conscious behavior may be difficult to judge, but at a minimum, vegetative patients should not follow commands or have comprehensible speech.
With a shorter duration of ischemia, the damage to the cerebral cortex is less and recovery of consciousness or awakening can occur, as evidenced by the patient following commands or having comprehensible speech and entering a minimally conscious state. 45 Typically, brainstem function is normal. These patients can be severely disabled because of their neurological impairments, which can include spastic quadriparesis, cortical blindness, severe ataxia, bowel and bladder incontinence, and severe memory deficits. For these survivors, independence in activities of daily living is not possible. In the series of 459 consecutive patients with out-of-hospital cardiac arrest in Seattle, 18 (4%) had such an outcome, including cortical blindness in 4 patients. 16 With still shorter duration of ischemia, less cerebral cortex is damaged and impairments are thought to relate to the parts of the brain most sensitive to ischemia. Nonetheless, in one study of 17 survivors of out-of-hospital cardiac arrest, memory impairments were related more to global cerebral atrophy on magnetic resonance imaging (MRI) than to selective hippocampal injury. 46 Memory problems can be seen with subtle motor findings 47 or can occur in isolation. 48 - 50 The memory problems can range from severe dementia to mild forgetfulness, and some of these patients may be independent in many of their activities of daily living.
Finally, if the duration of ischemia is brief and the resuscitation swift and effective, the brain may escape serious injury. Upon awakening, such patients commonly have severe memory impairments, which clear over days. Those patients who have a complete cardiac arrest remain amnestic for the event and typically do not report near-death experiences. 51 In a prospective study of 509 successful resuscitations, 12 percent of patients reported at least some recollection of the event and 8 percent described core experiences, including most commonly positive emotions, awareness of being dead, meeting with deceased persons, and moving through a tunnel. 52 Depending on the cardiac status and other comorbidities, patients may be able to return to their previous level of neurological function and are typically independent in their activities of daily living. Neuropsychological testing was performed on 68 long-term survivors of cardiac arrest as part of a clinical trial and showed no deficits or only mild ones in approximately one half at 1 year. 53 The most common deficit was in delayed memory. Improvement in scores on the Mini-Mental State Examination, 54 a simple cognitive measure familiar to most clinicians, occurs mostly in the first 3 months after the arrest, with results at 1 year being comparable to those at 3 months ( Fig. 9-2 ). Quality of life and the level of function have also been assessed and, for many, are comparable to those of similar patients who have suffered a myocardial infarction without a cardiac arrest. 55

FIGURE 9-2 Scores on the Mini-Mental State Examination 54 at certain times after out-of-hospital cardiac arrest in patients enrolled in a randomized trial of nimodipine versus placebo. Because results did not differ by treatment status, combined results are presented. Patients whose scores were zero are included.
(Data from Roine RO, Kajaste S, Kaste M: Neuropsychological sequelae of cardiac arrest. JAMA 269:237, 1993.)
After cardiac arrest, any individual patient may pass through all of the stages illustrated in Figure 9-1 . At the initiation of the resuscitation, these patients may lack all evidence of brain function and, with time, move toward complete recovery. As with other forms of ischemic brain injury, most of the recovery occurs in the first 3 to 6 months, for example, as shown in Figure 9-2 for the Mini-Mental State Examination score. In the study from Seattle of 459 patients, 61 percent awakened after out-of-hospital cardiac arrest and about two thirds of those awakening were left without gross neurological deficits. 16 Unfortunately, recovery can become stalled at any stage short of complete recovery, with the best recovery for many being a vegetative state or minimally conscious state. 42
Assessment of outcomes after cardiac arrest has most commonly involved the Glasgow Outcome Scale 39 or a variation on it such as the Cerebral Performance Category (CPC) developed by investigators from Pittsburgh. 56 Certain dichotomous outcomes, as in the top part of Figure 9-1 , are important but vary in the ease with which they can be determined and timed. Death is clear-cut. Awakening is defined by the ability to produce comprehensible speech or follow commands, or both. The reliability of its components has been studied in detail as part of the Glasgow Coma Scale. 57 In addition, the timing of its occurrence can be determined with relative ease, like that of death. Many legal and judicial guidelines concerning limitation of medical support address the probability of awakening. Regaining independence is of great importance to patients, families, and clinicians but lacks a uniform definition. It takes longer to achieve, it is more susceptible to the effects of comorbidity than awakening, and the timing of its occurrence can be problematic.
The pathology of global brain ischemia has been well described. 21 It confirms what is seen clinically, namely, that some cell types and some regions of the brain are more vulnerable than others. The explanation for selective vulnerability remains incompletely understood. Rather than destroying multiple cell types, as may occur with focal brain ischemia or infarct, damage after cardiac arrest may be limited to the neurons. Ischemic neurons are seen most prominently in certain layers of the cerebral cortex, parts of the hippocampus, and the Purkinje cells of the cerebellar cortex. In the neocortex, layer 3 is the most sensitive, followed by layers 5 and 6. Layers 2 and 4 are the most resistant. In the allocortex, the parts of the hippocampus most sensitive are the Sommer sector (CA1) and the end folium (CA3 to CA4). The region between these two areas (CA2) is the most resistant. At least in rats, the initial loss of hippocampal CA1 neurons after global ischemia can be followed by their reappearance and by recovery of learning and memory. 58 Whether this capacity to form new neurons occurs in humans and explains delayed recovery of memory after cardiac arrest remains to be established.
The gross appearance of the brain of a person autopsied following a brief comatose survival after cardiopulmonary resuscitation can look surprisingly unremarkable. Microscopic examination reveals ischemic changes as described, but quantification of these changes is not readily available. The number of neurons that need to survive to result in the clinical outcomes described Figure 9-1 , except at the extremes of all or none, remains poorly defined.

Seizures and myoclonus complicate the clinical course in 30 percent or more of patients following cardiac arrest and resuscitation. 59 - 61 They tend to occur in the immediate postresuscitation period, and long-term epilepsy is rare. Myoclonus is common and can be diffuse and multifocal, as with myoclonus status epilepticus. Alternatively, the patient may have little clinical manifestation of seizure activity beyond some subtle abnormalities of eye movements, yet the EEG can show evidence of ongoing electrical seizure activity. 62, 63 Regardless of the manifestations, seizures and myoclonus can be difficult to control in this setting, and treatment does not seem to affect the outcome.
Brain swelling is common, as evidenced by the results of imaging and lumbar puncture, but it rarely leads to herniation and deterioration. Exceptions are patients in whom a respiratory arrest precedes the cardiac arrest. These patients sometimes develop brain swelling severe enough to result in herniation, neurological deterioration, and brain death. In such circumstances, the swelling probably reflects a severely damaged brain, and treatments aimed at controlling increased pressures do not seem to affect outcomes. After cardiac arrest, patients are also at risk of numerous cardiac and general medical complications, which are beyond the scope of this discussion. All these, especially recurrent cardiac arrest and hypotension, tend to worsen the initial injury and thus the outcome.
A number of other rare complications can occur in the setting of a mixture of brain insults, sometimes including cardiac arrest. The greater the proportion of incomplete compared with complete global ischemia, the greater is the risk of preferential damage to border-zone regions of the brain and spinal cord. In the brain, this can result in injury between the anterior and middle cerebral artery distributions and a clinical picture of a brachial diplegia. With the upper extremities being more involved than the lower extremities, this clinical picture has been referred to as the “man-in-the-barrel” syndrome. 64 A similar border-zone phenomenon in the spinal cord has been proposed as an explanation for the paraplegia that can complicate cardiac arrest, although a retrospective pathological study of patients dying after global ischemia found the lumbosacral spinal cord more commonly affected than the thoracic cord. 65 Emboli may arise from the heart and less commonly from bone marrow secondary to the trauma of chest compressions. 66 Interestingly, patients with focal ischemia of the cerebral cortex may have symmetric neurological examinations upon presentation after cardiac arrest. Their examination reflects a level of function below their focal insult. If the level rises, focal abnormalities may become evident. Given the masking capability of global brain ischemia and the elevated risk of cardiac emboli in many of these patients, cardioembolic infarcts may be missed in this setting.
Two well-described but rare complications seem to be more a function of respiratory failure mixed with cardiovascular insufficiency than simply of cardiac arrest: delayed postanoxic encephalopathy 67 and posthypoxic action myoclonus. 68 Many of the cases of delayed encephalopathy have occurred in the setting of carbon monoxide exposure. A period of improvement and sometimes return to seemingly normal function is followed by marked deterioration. 69 Diffuse demyelination is the apparent cause of the deterioration in those who have died. An association with a partial deficiency of arylsulfatase A has been described in some cases. 70
Those with posthypoxic action myoclonus awaken soon after their respiratory arrest, often with normal cognitive function but with severe action myoclonus. 71 Attempts at movement are interrupted by myoclonic jerks. These can also be triggered by loud noises and other sensory stimuli. The pathological substrate for the condition is uncertain. Treatment with valproic acid or benzodiazepines is sometimes successful. Clinicians not familiar with posthypoxic action myoclonus may confuse it with myoclonus status epilepticus. Early after the insult, the two conditions can look similar but carry very different prognoses. The confusion usually arises when soon after the inciting insult the action myoclonus is treated aggressively with benzodiazepines. If the benzodiazepines are reduced, the patient lightens but the myoclonus returns. Consequently, the benzodiazepines are reinstituted. What may not be appreciated is that the unresponsive state is more a result of the medications than the original insult. Table 9-1 contrasts the two conditions with the goal of minimizing the risk of a clinician confusing the two.
TABLE 9-1 Clinical Features of Postcardiac Arrest Myoclonus Status and Posthypoxic Action Myoclonus Postcardiac Arrest Myoclonus Status Posthypoxic Action Myoclonus Symmetric jerking spontaneously and with stimulation Symmetric jerking spontaneously and with stimulation Onset during coma Onset during coma Can occur with convulsions Can occur with convulsions Effect of acute treatment unknown Effect of acute treatment unknown After cardiac arrest After respiratory arrest Resolves within a few days Persistent Evidence of severe brain damage on physical examination and ancillary tests No evidence of severe brain damage on physical examination and ancillary tests EEG often has burst-suppression pattern EEG often has generalized epileptiform discharges Fatal outcome Good cognitive recovery
Other movement disorders, including akinetic, dystonic, and hyperkinetic syndromes, have also been described, but not always clearly following sole cardiac arrest with complete global brain ischemia. 71 Imaging and neuropathological examination can reveal injury in the basal ganglia. This topic is discussed further in Chapter 59 .


General Issues and Physical Examination
Key to making clinical decisions is information on prognosis. In patients with a poor prognosis, the decisions may range from initiation of potentially dangerous treatments aimed at brain resuscitation to limitations in the intensity of medical support. The outcomes to be predicted include death, awakening, and independence. For the reasons discussed earlier, independence can be difficult to determine and time. Death is easily determined and timed but may relate to cardiac problems and need not reflect the degree of neurological recovery. Consequently, awakening is often used as a simple, easily determined, and timed outcome for identifying important predictors. The accurate delineation of prognosis after cardiac arrest presents several challenges 72, 73 and has been the topic of many reports and comprehensive reviews. 74, 75 Some of the major findings of these studies are summarized in this section.
The severity and duration of global brain ischemia are probably the most important predictors of outcome but are usually unknown or, at best, crudely estimated. The cardiac rhythm identified at the start of the resuscitation is a predictor of outcome. The probability of restoring a spontaneous rhythm is higher when the initial rhythm is ventricular fibrillation rather than asystole or pulseless electrical activity. 76, 77 A similar finding holds for survival and awakening. Demographic factors such as age and gender are not important predictors of restoration of spontaneous circulation or survival to hospital discharge. 78, 79 African Americans are more likely to have poor outcomes. 80, 81 Comorbidity may be an important factor with respect to outcome and probably explains why resuscitation rates for out-of-hospital cardiac arrest in some regions of the country are better than those for in-hospital cardiac arrest. 82, 83
The amount of brain dysfunction as indicated by the physical examination reflects the severity of the insult. In general, the greater the function at any particular time after the arrest, the better is the eventual outcome. Nonetheless, complete loss of neurological function at the time that the resuscitation is initiated is still compatible with awakening and good recovery. The duration of arrest that is incompatible with recovery is uncertain and, as a practical matter, is often difficult to estimate. The human brain can probably not fully recover from more than 5 to 10 minutes of complete global brain ischemia, as with ventricular fibrillation or asystole. 84
Patients with some degree of neurological function at the start of the resuscitation, such as the presence of pupillary light reflex or spontaneous respiratory efforts, have a better outcome than those without that capacity. 85 - 87 As discussed earlier, neurological function is lost in a rostral-caudal manner when blood ceases flowing to the brain, with spontaneous respiratory efforts being the last to go. Thus, the presence of neurological function at the start of the resuscitation indicates that the cardiac arrest is likely less than a few minutes old. The physical examination performed once the patient has been stabilized with respect to cardiovascular status gives the clinician the next chance to assess prognosis. Evaluation includes the Glasgow Coma Scale 39 but should be augmented by further evaluation of brainstem function 56 to include pupillary light reflex, corneal reflex, cough, gag, spontaneous respiratory efforts, and eye movements elicited by vestibulo-ocular and cervico-ocular reflexes 88 ( Table 9-2 ).
TABLE 9-2 Aspects of the Neurological Examination Used for Prognostication After Cardiac Arrest Glasgow Coma Scale * Other brainstem reflexes Eye opening (E) Pupillary light responses 4 = spontaneous Corneal reflex 3 = to speech Cough 2 = to pain Gag 1 = nil Spontaneous respiratory efforts Best motor response (M) Eye movements 6 = obeys Spontaneous 5 = localizes Cervico-ocular reflex 4 = withdraws Vestibulo-ocular reflex 3 = abnormal flexion   2 = extensor response   1 = nil   Verbal response (V)   5 = oriented   4 = confused conversation   3 = inappropriate words   2 = incomprehensible sounds   1 = nil  
* Total score = (E + M + V), range 3 to 15.
Glasgow Coma Scale adapted from Teasdale G, Jennett B: Assessment of coma and impaired consciousness: a practical scale. Lancet 2:81, 1974, with permission.
Agreement across clinical studies is remarkable, given the diversity of populations studied and the statistical methods applied. For instance, two studies examined prognostication after cardiac arrest. One study included 389 consecutive unconscious patients admitted to a single hospital in Seattle after out-of-hospital cardiac arrest with ventricular fibrillation or asystole. 89 The other was a multicenter study of 210 patients with a mixture of hypoxic-ischemic insults occurring both in and out of the hospital. 90 The studies used different multivariable techniques to identify independent early predictors of outcome, but both found that pupillary light reflex, motor response, and eye movements were the most important predictors. The other independent predictor of outcome in the study from Seattle was the admission blood glucose, which is discussed later. Figure 9-3 shows a simplification of the multivariable rule from the Seattle study and its performance in predicting outcome based on information collected at the time of admission. By using this rule, the clinician can rapidly form an initial prognostic impression.

FIGURE 9-3 Performance of a simple rule applied at admission to predict awakening after out-of-hospital cardiac arrest due to ventricular fibrillation or asystole.
(Data from Longstreth WT Jr, Diehr P, Inui TS: Prediction of awakening after out-of-hospital cardiac arrest. N Engl J Med 308:1378, 1983.)
As would be surmised from Figure 9-1 , the finding of persistent brainstem dysfunction suggests a poor prognosis, given that the cerebral cortex, being more sensitive, has suffered an even worse injury. In a detailed review, the clinical signs most strongly related to a poor outcome, which could include patients who regained consciousness but not independence, were absent corneal response at 24 hours, absent pupillary response at 24 hours, absent withdrawal response to pain at 24 hours, no motor response at 24 hours, and no motor response at 72 hours. 91 Unfortunately, preservation of brainstem function does not ensure a good outcome in that the cerebral cortex may still have been damaged to a degree that precludes ever awakening. The pupillary light reflex and the motor function, as assessed in the Glasgow Coma Scale 57 shown in Table 9-2 , have emerged as important prognostic factors in a number of studies.
In a systematic review of factors that predict a poor outcome after cardiac arrest, which was defined as death or survival in a vegetative state (i.e., never awakening), the absence of a pupillary light reflex on admission had a specificity that ranged from 69 to 100 percent. 74 Specificity is the percentage of all patients with a good outcome who lack the finding. Only when patients were tested on day 3 after the arrest did absence of a pupillary light reflex have 100 percent specificity in all studies. A specificity of 100 percent means that the positive predictive value must be 100 percent and that the false-positive rate must be 0 percent, namely, that all those with the finding (absence of a pupillary light reflex on day 3) have a poor outcome and never awaken. Unfortunately, the sensitivity of this finding was low, ranging from 22 to 55 percent. Sensitivity is the percentage of all patients with a poor outcome who demonstrate the finding.
The Glasgow Coma Scale has been used by itself as a predictor of outcome but is dominated by the motor findings. Most of these patients are intubated as part of their resuscitation, so that the verbal response cannot be assessed. Eye opening is an uncertain predictor, with some patients having spontaneous eye opening soon after resuscitation and never regaining consciousness. In a systematic review, the worse the motor response, the worse was the outcome, with the accuracy of predictions improving over time. 74 Thus, a score of 1 on the motor scale (absent motor response to pain) on day 3 after the arrest had a specificity of 100 percent for a poor outcome, but the sensitivity ranged from 11 to 58 percent. In a subsequent large prospective study of 407 patients, specificity was 95 percent and the false-positive rate was 5 percent for no motor response at 72 hours. 92
As Jørgensen and Malchow-Møller first demonstrated 93 and others have confirmed, the longer the delay in the return of neurological function, the worse is the prognosis. For example, recovery of pupillary light reflex within 12 minutes was compatible with a good outcome, whereas its absence for more than 28 minutes predicted a poor outcome. 93, 94 Ultimately, duration of unconsciousness itself becomes a predictor of the likelihood of consciousness ever returning. For example, in the study from Seattle, although overall 61 percent of the 459 patients awakened, the probability of ever awakening fell precipitously during the first 3 days after admission from out-of-hospital cardiac arrest. 16 More than 90 percent of those destined to ever awaken did so by 3 days. All those awakening after 4 days had some persistent neurological deficits, and all those awakening after 14 days had persistent severe neurological deficits. Interestingly, in that study as well as a subsequent one, 95 about one in five of those still unconscious 1 week after the arrest eventually awakened.
Although these studies are replete with potential problems, 72 a basic, unavoidable one is the reliability of the information on which prognosis is based. 96 Even in research settings where the question of agreement between examiners has been evaluated systematically, 91 it is not 100 percent and is likely lower in a practice setting. In addition, many of the medications administered to these patients may alter the findings on the neurological examination and adversely affect the accuracy of predictions based on them. Consequently, a certain degree of random error or noise will always exist in predictive models based on the neurological examination. In the large prospective study of 407 patients, physical examination findings were inferior to other tests to be discussed. 92 Perhaps the best that can be expected from the neurological examination is to provide an early indication of prognosis, such as with the awakening score provided in Figure 9-3 , and to monitor neurological recovery.

Ancillary Tests
Laboratory tests typically have greater reliability or precision than that of findings from the physical examination, and several tests have been evaluated for their potential to predict outcome after cardiac arrest. To be clinically useful, such a test must improve on the good, but not perfect, prediction of outcome already available from the physical examination. Several ancillary tests have been used for prognostication after cardiac arrest.
Results of most routine laboratory tests are not related to outcome after cardiac arrest, the possible exceptions being thyroid function tests and admission blood glucose levels. For example, admission values for arterial blood gas or hematocrit are not related to outcome. 89 By contrast, in one study, lower total thyroxine, total triiodothyronine, and thyrotropin from blood collected at the end of resuscitation were associated with a poor outcome. 97 In addition, as predicted from animal experiments, 98 elevated admission blood glucose levels after cardiac arrest have been shown to be associated with poor neurological outcome. 89, 99, 100 In one study that attempted to explain this association, analyses were done of blood glucose levels determined during cardiopulmonary resuscitation in patients with out-of-hospital cardiac arrest and suggested that the admission blood glucose level simply reflected the duration of resuscitation. 101 Because a small detrimental effect of glucose could not be excluded in these observational studies and because of data from experimental studies in animals, a randomized trial in humans was performed. 102 During resuscitation of patients with out-of-hospital cardiopulmonary arrests, emergency medical services personnel randomized patients to receive either the fluid that had been routinely used, 5 percent dextrose in water, or a glucose-free alternative, 0.5 normal saline. Results indicated that awakening was not related to the glucose content of the fluids that were given ( Fig. 9-4 ), although the amount of fluid administered by emergency medical services personnel was small. An unexpected and not easily explained finding in the clinical trial was that the relationship between blood glucose level and awakening was reversed in patients whose cardiac arrest was not due to ventricular fibrillation or asystole on a presumed cardiac basis. For these other patients, most of whom had pulseless electrical activity, a high blood glucose level on admission was associated with a favorable outcome. Overall, these studies suggest that the admission blood glucose level after cardiac arrest depends on the duration of certain aspects of the arrest and resuscitation. Although it is an independent predictor of outcome ( Fig. 9-3 ), the blood glucose level is not as powerful as certain aspects of the physical examination already discussed. 89

FIGURE 9-4 Presentation of results on survival and awakening after cardiac arrest from clinical trial of brain-cardiopulmonary resuscitation. This randomized trial examined whether the type of fluids (5% dextrose in water versus 0.5 normal saline) administered in the field by emergency medical services personnel affected outcome. Shown in the graph is the proportion of 291 admitted patients, all of whom were initially comatose after out-of-hospital cardiopulmonary arrest, who went on to awaken or to die without awakening. Solid lines represent the experience among those who received 5 percent dextrose in water ( n = 141); dashed lines represent those who received 0.5 normal saline ( n = 150). A vertical line from the horizontal axis at any particular time indicates the proportion who had awakened (from the 0.0 horizontal line to the lower pair of curves), who were still comatose (from lower to upper pair of curves), and who had died without awakening (from the upper pair of curves to the 1.0 horizontal line ). For example, a vertical line at 3 days indicates for the entire cohort that about 34 percent had awakened regardless of treatment group (34% minus 0%), that about 37 percent were still comatose (71% minus 34%), and that the remaining 29 percent (100% minus 71%) had died without awakening. Of the 117 patients who awakened, 15 subsequently died during hospitalization and are classified in this figure as having awakened.
(From Longstreth WT Jr, Copass MK, Dennis LK, et al: Intravenous glucose after out-of-hospital cardiopulmonary arrest: a community-based randomized trial. Neurology 43:2534, 1993, with permission.)
Another study that is typically available in these patients is the EEG. In addition to electrocerebral inactivity and a burst-suppression pattern, other specific patterns have been described, such as spindle coma, theta-pattern coma, and alpha-pattern coma, although no universally accepted EEG classification scheme exists. 75 Both electrocerebral inactivity and a burst-suppression pattern are associated with a poor prognosis. 74, 75 In a systematic review, electrocerebral inactivity and a burst-suppression pattern had a specificity for a poor outcome of never awakening of 100 percent in all but one study. 74 Interestingly, the only study not to show a specificity of 100 percent was also the only study in which blinding was employed. 103 Of note, alpha-pattern coma was not necessarily associated with a poor prognosis. Part of the difficulty in assessing the prognostic utility of the EEG relates to its typically being obtained early in the hospital course on patients who are having uncontrolled seizures or myoclonus or later in those who remain unconscious. Thus, confounding by indication may be present, with the seizures and prolonged unconsciousness being markers of a poor prognosis. Whether the EEG provides additional independent prognostic information is uncertain.
The prognosis with myoclonus status epilepticus in the first 24 hours following the arrest is especially poor, leading some to equate myoclonus status epilepticus with an agonal brain phenomenon indicating severe cerebral cortical damage. 61, 75, 92, 104 In one review, myoclonus status was defined as spontaneous, repetitive, unrelenting, generalized multifocal myoclonus involving the face, limbs, and axial musculature in comatose patients after primary circulatory arrest, 75 and it can be associated with abnormal but variable EEG findings. 104 A meta-analysis yielded a specificity of 100 percent and a false-positive rate of 0 percent with 95 percent confidence interval of 0 to 8.8 percent, suggesting that the true percentage is 95 percent likely to fall somewhere within the interval. 75 Sensitivity has varied across studies. In the largest study that included 407 patients, only 4 percent of those who never awakened had myoclonus status. 92 Extreme caution is needed in distinguishing myoclonus status from action myoclonus ( Table 9-1 ), given the marked differences in prognosis. 105
Evoked potential studies have been investigated thoroughly, and bilateral median nerve somatosensory evoked potentials have become the dominant prognostic test after cardiac arrest. 74, 75 Bilateral absence of a short-latency cortical response (N20) is associated with a poor prognosis. 74, 75, 92, 106 In a systematic review, bilateral absence of the short-latency cortical response was 100 percent specific, with none of the 187 patients with this finding after cardiac arrest achieving a good outcome. Of note, exceptions exist if the somatosensory evoked potentials are performed within 24 hours of arrest. 107 In a subsequent prospective cohort study, serial somatosensory evoked potentials were performed at 24, 48, and 72 hours after resuscitation. Among the 301 patients still alive and unconscious at 72 hours, all 136 (45%) with bilaterally absent N20 at any of these times died or remained unconscious after 1 month. 92 Interestingly, in nine patients, N20s were present on an initial test but lost on follow-up, and in five patients, the reverse was seen. All these patients had a poor outcome. Although treating physicians were blinded to results of the testing at 24 and 48 hours, they were informed of the results of the tests at 72 hours and advised that chances for survival or recovery of consciousness were virtually nil with bilaterally absent N20. An exception has been reported, 108 which may reflect that although the reliability of somatosensory evoked potentials may be better than that for the physical examination, it is not perfect. 109 Recommendations to improve the reliability of the test have been offered ( Table 9-3 ). A meta-analysis based on these and other studies yielded a specificity of 99.3 percent and a false-positive rate of 0.7 percent with 95 percent confidence interval of 0 to 3.7 percent. 75 Unfortunately, the presence of N20 does not ensure a good outcome. Preliminary studies suggest that cognitive evoked potentials hold promise in predicting good outcomes. 110
TABLE 9-3 Recommendations for Recording and Interpreting Somatosensory Evoked Potentials
A standard recording technique should be used. Deviations from the standard recording techniques should be stated clearly.
It is imperative to obtain a good signal-to-noise ratio, which, in view of possibly low-amplitude signals, requires a very low remaining noise level. For both cortical N20 leads and the N13 lead, the noise level concerns the frequency range of 20 to 500 Hz and the peak-to-peak amplitude of noise after averaging should be lower than μ0.25 V.
Good signal to noise ratio can be ensured using the following steps:
Use up to 1,000 or even more stimuli to obtain an average.
Stimulus intensity may be set at a higher level than is used in conscious patients; however, in some patients this may result in more muscle activity.
Do not hesitate to use muscle relaxants.
Turn off nonessential electrical equipment in the patient’s surroundings.
Potentials should be recorded at an amplification of at least 1 μV/cm.
An N20 peak on one side can only be considered present if it fulfills all the following criteria:
It should have an appropriate latency (i.e., at least 4.5 msec longer than the corresponding N13 peak in normal-stature adults).
It should be present on the side contralateral to stimulation, and there should be a clear difference with the recording from the side ipsilateral to the stimulus.
An N13 peak must be present. This requires a recording below C2, preferentially at C5 to C7.
Any potentials should be reproducible with a second set of stimuli.
A failure to meet any one of these criteria for either the right or the left side means that the N20 peak cannot be considered bilaterally absent.
Modified from Zandbergen EG, Hijdra A, de Haan RJ, et al: Interobserver variation in the interpretation of SSEPs in anoxic-ischaemic coma. Clin Neurophysiol 117:1529, 2006.
Imaging of the head with computed tomography (CT) and MRI may show loss of gray-white matter differentiation, evidence of swelling, and other changes suggesting diffuse cortical injury. 75 In one study, agreement about the presence or absence of gray-white matter differentiation on CT was problematic. 111 The abnormalities, especially on MRI, may be dramatic ( Fig. 9-5 ), but more work is needed to clarify the prognostic utility of such tests. 75 Interestingly, evidence of swelling is more common among those who experienced respiratory failure prior to their cardiac arrest. 112 Later in the course, these studies may show atrophy in patients who remain unconscious. Other types of brain imaging have been studied, but only in a small number of patients, precluding any firm conclusions about their prognostic value.

FIGURE 9-5 Magnetic resonance imaging (MRI) after cardiac arrest. The patient was a 49-year-old woman with respiratory arrest followed by cardiac arrest. Paramedics found her cyanotic and asystolic. Initial neurological examinations showed her to be comatose with extensor posturing but intact brainstem function. MRI was performed about 32 hours after cardiac arrest. Cerebral cortical abnormalities are less obvious on fluid-attenuated inversion recovery (FLAIR) sequences ( A, bright ) than on diffusion-weighted imaging ( B, bright ) and apparent diffusion coefficient (ADC) maps ( C, dark ).
Several serum and cerebrospinal fluid (CSF) enzymes have been evaluated. 75, 113 Caution must be exercised in the interpretation of enzyme results when a mixture of brain insults has occurred because elevations are not specific to cardiac arrest, having been described with other types of injury. The highest quality evidence of predictions of outcomes after cardiac arrest exists for serum neuron-specific enolase (NSE). In the prospective cohort study mentioned earlier, serial NSE was assayed at 24, 48, and 72 hours after resuscitation. 92 Among the 231 patients