You are on page 1of 410

1600 John F. Kennedy Blvd.

Ste 1800 Philadelphia, PA 19103-2899

THE EPILEPSIES 3 Copyright ! 2009 by Saunders, an imprint of Elsevier Inc.

ISBN: 978-1-4160-6171-7

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 Elseviers Rights Department: phone: (+1) 215 239 3804 (US) or (+44) 1865 843830 (UK); fax: (+44) 1865 853333; e-mail: healthpermissions@elsevier.com. You may also complete your request on-line via the Elsevier website at http://www.elsevier.com/permissions.

Notice 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 practitioner, relying on his or her own experience and knowledge of the patient, 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 authors assume any liability for any injury and/or damage to persons or property arising out of or related to any use of the material contained in this book. The Publisher

Acquisitions Editor: Adrianne Brigido Developmental Editor: Joan Ryan Senior Project Manager: David Saltzberg Design Direction: Steve Stave

Printed in China Last digit is the print number:

5 4

CONTRIBUTING AUTHORS

Gail D. Anderson, PhD


Professor of Pharmacology Pharmaceutics, and Neurological Surgery University of Washington Health Science Complex Seattle, Washington

Carlo Antozzi, MD
Professor of Clinical Neurosciences Neuroimmunology and Muscle Pathology Unit Carlo Besta Neurological Institute Foundation Milan, Italy

Tallie Z. Baram, MD, PhD


Professor of Pediatrics, Anatomy, and Neurobiology University of California Irvine Irvine, California

Dina Battino, MD
Professor of Neurophysiology Epilepsy Center Carlo Besta Neurological Institute Foundation Milan, Italy

Edward H. Bertram, MD
Professor of Neurology The F.E. Dreifuss Comprehensive Epilepsy Program University of Virginia Charlottesville, Virginia

Paul Boon, MD, PhD


Professor of Neurology Laboratory for Clinical and Experimental Neurophysiology Ghent University Hospital Gent, Belgium

xi

xii

CONTRIBUTING AUTHORS

Catherine Chiron, MD, PhD


Professor of Neurology UMR 663 University Paris Descartes, Paris 5 Chief, Research Unit on Epilepsy in Children UMR 663, INSERM Pediatric Neurologist Neurometabolic Department Necker-Enfants Malades Hospital Paris, France

Chantal Depondt, MD, PhD


Department of Neurology University of Brussels pital Erasme Ho Brussels, Belgium

line M. Dube , PhD Ce


Assistant Project Scientist Departments of Anatomy/Neurobiology and Pediatrics University of California Irvine Irvine, California

John S. Ebersole, MD
Professor of Neurology and Director Adult Epilepsy Center Department of Neurology The University of Chicago Chicago, Illinois

Howard P. Goodkin, MD, PhD


Assistant Professor of Neurology and Pediatrics University of Virginia Charlottesville, Virginia

Tiziana Granata, MD
Professor of Pediatric Neurosciences Carlo Besta Neurological Institute Foundation Milan, Italy

CONTRIBUTING AUTHORS

xiii

Renzo Guerrini, MD
Pediatric Neurology Unit and Laboratories Neuroscience Department Childrens Hospital A. Meyer-University of Florence Florence, Italy

Andres M. Kanner, MD
Professor of Neurological Sciences and Psychiatry Rush Medical College at Rush University Director Laboratory of Electroencephalography and Video-EEG-Telemetry Associate Director, Section of Epilepsy Rush University Medical Center Chicago, Illinois

Henrik Klitgaard, PhD


Vice President CNS Research UCB NewMedicines UCB Pharma S.A., Belgium

Michael Koutroumanidis, MD
Senior Lecturer in Neurology Department of Academic Neuroscience Kings College London Consultant Neurologist and Clinical Neurophysiologist Department of Clinical Neurophysiology and Epilepsies Guys, St. Thomas, and Evelina NHS Foundation Trust London, United Kingdom

David Krieger, MD, MS


Resident, Department of Neurosurgery Thomas Jefferson University Philadelphia, Pennsylvania

Louis Lemieux, PhD


Professor of Physics Applied to Medical Imaging Department of Clinical and Experimental Epilepsy University College London Institute of Neurology London, United Kingdom

xiv

CONTRIBUTING AUTHORS

Brian Litt, MD
Associate Professor of Neurology and Bioengineering University of Pennsylvania Director, Epilepsy Surgery Program Department of Neurology Hospital of the University of Pennsylvania Philadelphia, Pennsylvania

Francesco Mari, MD
Pediatric Neurology Unit and Laboratories Neuroscience Department Childrens Hospital A. Meyer-University of Florence Florence, Italy

Anthony G. Marson, MD, FRCP


Reader in Neurology, Division of Neurological Science The University of Liverpool Consultant Neurologist The Walton Centre for Neurology and Neurosurgery NHS Trust Liverpool, United Kingdom

Alain Matagne, MSc


Senior Director, Research and Development CNS Research UCB NewMedicines UCB Pharma S.A., Belgium

Anil Mendiratta, MD
Assistant Clinical Professor of Neurology Columbia University New York, New York

Nicholas Moran, MBChB, MRCP, MSc


Epilepsy Research Group Kings College Hospital London, United Kingdom Department of Neurology Kent and Canterbury Hospital Canterbury, United Kingdom

CONTRIBUTING AUTHORS

xv

Lina Nashef, MBChB, MD, FRCP


Professor of Neurology Department of Clinical Neuroscience Kings College London; Neurologist, Kings College Hospital London, United Kingdom

Soheyl Noachtar, MD
Professor of Neurology and Head of Epilepsy Center University of Munich Munich, Germany

Chrysostomos P. Panayiotopoulos, MD, PhD, FRCP


Consultant Emeritus Department of Clinical Neurophysiology and Epilepsies St. Thomas Hospital London, United Kingdom

Timothy A. Pedley, MD
Henry and Lucy Moses Professor of Neurology and Chairman Department of Neurology College of Physicians and Surgeons Columbia University; Neurologist-in-Chief Neurological Institute of New York New York-Presbyterian Hospital Columbia University Medical Center New York, New York

mi, MD Jan Re
Sleep and Epilepsy Fellow Department of Neurology Klinikum Grosshadern University of Munich Munich, Germany

xvi

CONTRIBUTING AUTHORS

Jong M. Rho, MD
Associate Professor of Clinical Neurology University of Arizona College of Medicine; Senior Staff Scientist, Neurology Research Barrow Neurological Institute; Associate Director of Child Neurology Childrens Health Center St. Josephs Hospital & Medical Center Phoenix, Arizona

Fergus J. Rugg-Gunn, MBBS, MRCP, PhD


Consultant Neurologist Department of Clinical and Experimental Epilepsy Institute of Neurology, University College London Department of Clinical and Experimental Epilepsy The National Hospital for Neurology and Neurosurgery London, United Kingdom

Simon Shorvon, MD
Professor in Clinical Neurology Institute of Neurology University College Consultant Neurologist National Hospital for Neurology and Neurosurgery London, United Kingdom

Rachel Thornton, MA, MBBS


Clinical Research Fellow Department of Clinical and Experimental Epilepsy University College London Institute of Neurology London, United Kingdom

rn Tomson, MD, PhD Torbjo


Professor of Clinical Neuroscience Karolinska Institute; Department of Neurology Karolinska University Hospital Stockholm, Sweden

Kristl Vonck, MD, PhD


Professor of Neurology and Laboratory for Clinical and Experimental Neurophysiology Ghent University Hospital Gent, Belgium

SERIES PREFACE
The Blue Books of Neurology have a long and distinguished lineage. Life began as the Modern Trends in Neurology series and continued with the monographs forming BIMR Neurology. The present series was first edited by David Marsden and Arthur Asbury, and saw the publication of 25 volumes over a period of 18 years. The guiding principle of each volume, the topic of which is selected by the Series Editors, was that each should cover an area where there had been significant advances in research and that such progress had been translated to new or improved patient management. This has been the guiding spirit behind each volume, and we expect it to continue. In effect, we emphasize basic, translational, and clinical research but principally to the extent that it changes our collective attitudes and practices in caring for those who are neurologically afflicted. Tony Schapira took over as joint editor in 1999 following Davids death, and together with Art oversaw the publication and preparation of a further 8 volumes. In 2005, Art Asbury ended his exceptional co-editorship after 25 years of distinguished contribution and Martin Samuels was asked to continue the co-editorship with Tony. The current volumes represent the beginning of the next stage in the development of the Blue Books. The editors intend to build upon the excellent reputation established by the Series with a new and attractive visual style incorporating the same level of high-quality review. The ethos of the Series remains the same: up-to-date reviews of topic areas in which there have been important and exciting advances of relevance to the diagnosis and treatment of patients with neurological diseases. The intended audience remains those neurologists in training and those practicing clinicians in search of a contemporary, valuable, and interesting source of information.
Anthony H.V. Schapira Martin A. Samuels Series Editors

xvii

S a m u e l H. B a r o n d e s
BORN:

Brooklyn, New York December 21, 1933


EDUCATION:

Columbia College, A.B. (1954) Columbia College of Physicians and Surgeons, M.D. (1958) Peter Bent Brigham Hospital (Medicine, 1958-1960) National Institutes of Health (Postdoctoral Research, 1960-1963) McLean Hospital (Psychiatry, 1963-1966)
APPOINTMENTS:

Albert Einstein College of Medicine (1966) University of California, San Diego (1969) University of California, San Francisco (1986) Chair, Psychiatry, UCSF and Director, Langley Porter Psychiatric Institute (1986-1993) Founding Director, Center for Neurobiology and Psychiatry, UCSF (1993) Jeanne and Sanford Robertson Professor, UCSF (1996)
HONORS AND AWARDS (SELECTED):

Fogarty International Scholar, NIH (1979) Herman Stillmark Lectin Centennial Medal, Estonia (1988) McKnight Endowment Fund for Neuroscience, President (1989-1998) Institute of Medicine (1990) J. Elliott Royer Award in Psychiatry, University of California (1990) J. Robert Oppenheimer Memorial Lecturer, Los Alamos (2000) National Institute of Mental Health, Board of Scientific Counselors, Chair (2000-2003)
Samuel Barondes played a major role in bringing a molecular and genetic approach to neuroscience and psychiatry. In early work he helped establish the requirement for brain protein synthesis in long-term memory and demonstrated the rapid transport of proteins in brain axons. Turning his attention to cellular interactions, he discovered discoidinsmslime mold relatives of the discoidin-domain proteins involved in synaptogenesis--as well as galectins, a family of glycoconjugate-binding proteins, some of which are found in neurons. A gifted writer, he has published three books about psychiatric genetics and molecular psychiatry for a general audience.

S a m u e l H. B a r o n d e s

Brighton Beach Childhood

riting this memoir has caused me to reflect on my good fortune. I have been very lucky to have had the privilege of participating in such an exciting period of discovery in neuroscience and psychiatry and of enjoying warm personal relationships with so many talented members of these rich scientific communities. I am particularly grateful to the mentors who shaped me, the colleagues and trainees who sustained me, and the continuing elaboration of our work by others. My parents were born in Eastern Europe in 1902, and each left for America in their late teens, in the aftermath of the First World War. My father, Solomon, was raised in Zbaraz, which was part of the Austro-Hungarian empire and is now in Ukraine. My mother, Yetta Kaplow, came from Kraisk in what is now Belarus. These were both Fiddler-on-the-Roof-type villages in which their families, each with eight children, eked out a living and were guided by the Jewish traditions of the time. Although my parents had almost no formal schooling they were literate in Yiddish, Russian, and Hebrew. As immigrants they went to night school and became fluent in English. Both my parents came to America following in the footsteps of an older brother. My father's voyage was arranged by his brother Nathan who had settled in Quincy, Massachusetts and had become a junk dealer. But neither Quincy nor the junk business appealed to my father who soon moved to New York City with the hope of becoming a professional singer. He settled in the lower east side of Manhattan, which then had a thriving Yiddish theater, worked as a waiter in restaurants in this theater district, then as a salesman in a clothing store. His singing ambitions were not fulfilled until many years later. My mother's immigration was made possible by her older brother, Joe, who arrived in New York City shortly before World War I and soon found himself in the US Army. Sent to France as an infantryman he was gassed in the trenches but survived without disability. Upon his discharge he was offered some schooling and became an accountant in New York and, over the years, a wealthy man. When my mother arrived she went to work in a clothing store near the one where my father was employed. When my parents married several years later they moved to Brighton Beach, a seaside community of six-story brick apartment buildings in Brooklyn. Helped by a gift from Joe they opened a small fabric store on a

Samuel H. Barondes

street lined with mom-and-pop businesses. Being in the store all day provided a great opportunity for reading, which my mother took full advantage of. The store and their apartment were just a few blocks from a wide beach and a splendid boardwalk, which both extended for several miles to Coney Island, Brooklyn's famous amusement park district. It was golden America, the promised land, even when the Depression came. By the time I was born in 1933 the Depression was in full force. My parents told me that we were very poor during my early childhood, but I, of course, remember none of it. I do know that the bad economy influenced the elementary school I attended, a Jewish parochial school for boys called the Yeshiva of Brighton Beach, which provided religious studies in the morning and secular studies in the afternoon. The long school day was particularly attractive to my parents because they both worked in the store and it was a convenient way for them to have me, their only child, occupied all afternoon in those days when there were no after-school programs. My parents were also pleased that I would be getting a Biblical and Talmudic education as well as a secular one, but a school day that went until 5 PM was clearly a major attraction. The Yeshiva of Brighton Beach was, for me, mainly a blessing. In the first grade it became clear that I was an enthusiastic student, the first in my class to read the Bible in Hebrew. By the time I was 10 I eagerly participated in discussions of ethical arguments in the Talmud, a series of volumes of commentaries by eminent rabbinical scholars, parts of which go back 2 millennia. These talents were greatly prized by my teachers who considered the study of the Bible and its Talmudic commentaries as an act of worship. I also did well in the usual elementary school curriculum, which got me a gold medal on graduation but was not valued as highly. The downside of my schooling was that it was largely in the hands of orthodox rabbis who expected not just the study of the Bible but also the relentless practice of the way of life that it prescribed--a practice that was not always in tune with that of the greater world around me. Some of their demands were not hard to fulfill. For example they required that we always wear a traditional head covering, but they were willing to accept a baseball cap in its place. Others were more difficult to satisfy, especially those that concerned the observance of the Sabbath, because it is forbidden to do work of any kind on this day of rest, and working on the Sabbath is a concept that orthodox rabbis have interpreted very broadly. In the modern world they have extended this prohibition to such simple activities as flipping a switch to turn on the lights or the radio. Even as a child this seemed unreasonable to me. But the rabbis at my school had zero tolerance for any behavior they believed to be forbidden. Despite my discomfort with the rules of orthodoxy I went to a high school that was organized along similar lines. I did this mainly to please my mother who had several eminent rabbinical relatives and who hoped

Samuel H. Barondes

that I too would become a rabbi. I was already strongly inclined against this but bowed to her wishes, knowing that I would eventually conduct my life as I pleased. My father also was in favor of this choice of high school but for a different reason. As the family store began to bring in a comfortable living he had the time and the means to indulge his passion for singing by studying with excellent teachers. With this training, his haunting lyric tenor voice, and a great talent improvising on traditional synagogue themes, he became a superb Cantor. In the process he befriended other Cantors, such as Jan Pierce and Richard Tucker, who studied operatic singing and became leading tenors at the Metropolitan Opera. My father, too, sang operatic arias, mainly by Puccini, Verdi, and Mozart, but his greatest talent was in interpreting Jewish liturgical music. Despite this he refused to become a full-time Cantor because he did not want to be dependent on the whims of a congregation. So he restricted his cantoring to special holidays and to making records, while guaranteeing his independence with the earnings from the store. Nevertheless, he believed that it was appropriate for the son of a Cantor to attend a religious high school, much to my mother's delight. The high school I attended, Talmudical Academy, introduced me to a world outside of Brighton Beach. Located in a central part of Brooklyn, it was a 20-minute subway ride away. One of its attractions was that it was just a few blocks from Ebbetts Field, home of the Brooklyn Dodgers. Another attraction was its excellent secular afternoon classes that followed a morning of Talmud. Taught by teachers who were also employed by the public high schools, the afternoon curriculum was particularly strong in science. The success of our education was assessed each year when we, like all high school students in New York, took standardized achievement tests. Students like me, who scored close to the top in these exams, were viewed with respect by our classmates. Instead of being ostracized as a hopeless nerd I was elected president of the student council. But it was to Brighton Beach that I returned every night and with happy anticipation. There was always the beach and the boardwalk where I liked to hang out. If I needed pocket money I would collect discarded soda bottles that were easy to find on the beach and brought 2 r apiece, quite a bonanza at a time when a movie ticket cost only a dime. There were also fascinating discussions going on at various spots on the boardwalk where adults would congregate to dream together about the creation of a harmonious socialist world in the aftermath of World War II. And just down the boardwalk were the amusements of Coney Island with its rides, shows, and penny arcades. My teenage years were also greatly enriched by 2 months each summer at camps in upstate New York. I started these adventures just before high school, and over the years I graduated to senior boys counselor. Going to

Samuel H. Barondes

camp was 2 months of continuous outdoor activities in the countryside. And, for the first time in my life, there were girls, dances, and coed theatrical productions that I participated in avidly, often writing original skits and the lyrics for musicals. This new world facilitated my transition from the cloistered existence of all-boy parochial schools to the liberation of college. There was, however, a final challenge: my mother was still interested in making me a rabbi and urged me to go to Yeshiva University. But I had decided to move into the greater world. Since I was only 16, and my parents insisted that I continue to live at home during college, I narrowed the choices down to Brooklyn College, which was only about 20 minutes away by subway, and Columbia, which would extend the ride by at least an hour more in each direction. Despite this inconvenience I liked the feel of the Columbia campus, and I was thrilled by the possibility of a daily escape to Manhattan. When Columbia admitted me I was ready to go, despite my mother's disappointment. My father was p r o u d ~ a n immigrant from Zbaraz with a son at Columbia. As I look back on my experiences in Jewish parochial schools I consider them a valuable preparation for my subsequent life and career. Although I was, from a young age, skeptical about the aspects of my education that were based on revealed truth, I could put that aside while incorporating its emphasis on living a virtuous life and a life of learning. My father, who had a strong philosophical bent, would also talk with me about wisdom and morality on our frequent evening walks on the boardwalk. He sometimes liked to base our conversations on phrases from the Talmud, many of which have stayed with me ever since. One of his favorites, roughly translated: "Who is a rich man? He who rejoices in his portion."

College at Columbia
When I started Columbia College in the Fall of 1950 I had no clear plans for the future. This was just fine with the College, which did not require the selection of a major course of study. Instead it prided itself on offering a liberal education shaped by its famous core curriculum. This emphasis on great ideas of western civilization was particularly valuable for me because I was less informed about these matters than many of my classmates, and I was delighted to be offered so many samples of the wisdom of the ages. My 3 hours a day on the subway were regularly devoted to reading, and the jostling ride went unnoticed as I immersed myself in Aristotle, Sophocles, Spinoza, and Freud. There was also a science course requirement that I began to satisfy in my sophomore year by taking the course in introductory psychology. Taught at Columbia as an experimental science, it was organized around Principles of Psychology: A Systematic Text in the Science of Behavior by Fred S. Keller and William N. Schoenfeld, the two professors in charge of the course.

Samuel H. Barondes

Having just been published in 1950 this was a very unusual introductory textbook. Instead of providing a broad survey of approaches to psychology it confined its attention to behaviorism and emphasized the ideas of the world's leading behaviorist, B.F. Skinner, whom the authors idolized. Discovering that psychology could be based on experimentation rather than introspection was very exciting to me. I was captivated by the simple experimental apparatus that was used to study bar-pressing by rats and the manipulation of their rates of response by schedules of reinforcement with food pellets. I was particularly impressed that the pattern of responses could be recorded graphically in the course of the experiment as a cumulative response curve, providing a quantitative picture of behavior on a piece of paper, and that the results were predictable and reliable. I was also swept up by the almost religious zeal of my teachers who insisted that this was the approach that would finally lead us to an objective understanding of the forces that control all h u m a n behavior. I wanted to learn more. At the start of my junior year I supplemented my scientific education with a course in chemistry, a subject that I already loved since high school, and also signed up for several more psychology courses. The one in abnormal psychology taught by Ralph Hefferline really grabbed me. As the year progressed I began thinking more and more about a career in academic psychology. Then Uncle Joe, my mother's older brother, intervened. He had, by then, made a small fortune in real estate and established himself as the senior member of our extended family. He had always taken a great interest in me since I was the first of the American-born generation, and he was eager to see me prosper. One afternoon, in the first semester of my junior year, he came to our house to talk to me about my future plans. I proudly informed him that I wanted to do research in psychology and hoped some day to be a professor. His response, which is permanently etched in my memory, was not at all what I expected. He began by telling me he approved of my ambition, even thought it was a great idea. "But first," he said, "you have to go to medical school. That will broaden your horizon and provide you with some security if your research doesn't work out. When you've finished medical school you'll be in a great position to start doing exactly what you want to do--research in psychology. And you'll also be in a great position to do many other interesting things, should you decide to change your mind." The amazing thing about my conversation with Joe--this down to earth man with no formal education~is that it immediately altered the course of my life. When I greeted him that day I was pretty clear that I was on my way to a Ph.D. in psychology. In a matter of minutes I was seriously~and for the first time entertaining the possibility of becoming a medical doctor. As the idea sunk in, I arranged to take the additional science courses that were required for medical school.

Samuel H. Barondes

Joe's intervention soon paid me a great intellectual dividend. In the second semester of my junior year I began a course in physics, a subject I knew nothing about and that, until then, I had no intention of studying. Fortunately Columbia was among the first to offer an introductory course that did not require much background in math, a forerunner of those now referred to as "physics for poets." For me it was transformative. Based on a new textbook--Introduction to Concepts and Theories in Physical Science by Gerald Holtonmit opened my eyes to the way science uses quantitative methods of observation and experimentation to explain the world. I had already had a taste of this in looking at the graphs that rats generated in Skinner boxes. But that now seemed like child's play in comparison with the work of Galileo, Newton, Faraday, and Einstein. And even though I did not understand all the nuances I got the big picture, a picture of a world made a bit more comprehensible by the cumulative discoveries of generations of scientists. A great benefit of my enthusiasm for so many classes was my election to Phi Beta Kappa in my junior year. This guaranteed my admission to most medical schools even though I had not taken the usual premedical program of studies. Urged by my parents to remain close to home, I decided to continue at Columbia by moving on to its medical campus further uptown. In anticipation of this move I filled my senior year with sciences. To round out my understanding of psychology I also took a class called "The Biology of Behavior" that was not a mainstream listing of the behavioristdominated psychology department but was offered, instead, by Columbia's School of General Studies. Its instructor, Murray Jarvik, opened my eyes to the value of brain research in the study of behavior. Murray will also figure later in my story and became a lifelong friend. Columbia Medical School Columbia's College of Physicians and Surgeons is on 168th street in Manhattan, about 3 miles north of the main campus. When I started there in the Fall of 1954 I signed up for lodgings at Bard Hall, the student residence. Freed at last from the long daily subway rides to Brighton Beach, I had a comfortable room overlooking the Hudson River and the continuous company of stimulating classmates. I approached all the medical school classes with high hopes. Some were taught by leading researchers such as Erwin Chargaff, whose discovery of the ratios of the four bases in DNA was crucial for the Watson-Crick model of the double helix. Elvin Kabat's lectures on immunology, which emphasized his own experimental work, were particularly inspiring. I even liked anatomy. But the big disappointment for me was psychiatry. Having started medical school with the belief that academic psychiatrists would be engaged in

Samuel H. Barondes

experimental studies of behavior, I was surprised to learn that my teachers cared only about psychoanalysis and were not interested in research. Even though I was intensely curious about psychopathology, and was fascinated by the concurrent brilliance and craziness of a manic-depressive patient I worked with, I was uncomfortable with the prospect of devoting myself to a field that seemed to rely so heavily on strongly held opinions rather than scientific evidence. As I progressed into the clinical years of medical school I was drawn to endocrinology, a field that had already developed quantitative laboratory tests to aid in diagnosis and treatment. Endocrinology also appealed to me because the pituitary, which controls other glands, is itself controlled by secretions from nerve cells in the hypothalamus, and these, in turn, are influenced by emotions. Viewed in this way endocrinology was not only grounded in science but also relevant to aspects of human behavior that I found interesting. I was also pleased to discover that I liked working in the clinics. This was particularly true in endocrinology, which offered excellent treatments for some prevalent disorders such as hypothyroidism, thus guaranteeing many satisfied patients. By the end of my third year in medical school I decided to get training in endocrinology and opted for a medical internship at the Peter Bent Brigham Hospital in Boston, just down the street from Harvard Medical School. A main attraction was its chief of medicine, George Thorn, a distinguished endocrinologist.

Becoming a Physician at the Brigham My first day at the Brigham opened another exciting chapter in my long education. Arriving on the ward in one of the starched white cotton suits the hospital provided, but with little understanding of what I was supposed to do, I was surprised to find my fellow intern, Donald Harrison, already busily at work with the patients. Donald had gone to medical school at the University of Alabama, which offered a much more practical education than I had received. He had been a whiz student and I was immediately in awe of the way he combined exuberant enthusiasm, Southern charm, high intelligence, and hands-on medical knowledge. Fortunately he was also an enormously generous person who was eager to teach me the rudiments of patient care in exchange for a few bits of the book-learning I had accumulated at Columbia. A few months with him and I had picked up the tricks of the trade. My 2 years at the Brigham, the first as an intern and the second as a medical resident, were filled with many such comradely experiences that come when a small group of young people keep working to exhaustion for a worthy cause. The only time off from our continuous duty was every other

Samuel H. Barondes

night and weekend. But the work was exciting and this intense process of initiation had the desired effect of minting skilled physicians. The Brigham was also an important site of medical innovation, which attracted some unusual patients. Being interested in endocrinology, I had the opportunity to work with people with adrenal diseases who were drawn there because of George Thorn. We were also encouraged to start thinking about doing research. Having decided that population control was the most important world problem, I submitted a pie-in-the-sky proposal to Dr. Somers Sturgis, a professor of gynecology, on male contraception which proposed to make antibodies to sperm cells and to use them for birth control. As I now reread the proposal I see how naive it was. But even though nothing came of it, I found that I enjoyed thinking about experiments. The Brigham's interest in making us into medical scientists was not restricted to encouraging such armchair speculation. In order to be invited to complete the medical residency, it was necessary to leave after 2 years of clinical training, for at least 2 years in the lab. In those days, in which all medical doctors were subject to the draft, the most desirable way to do this research was as a member of the United States Public Health Service (USPHS)--one of the uniformed services--and to be assigned to a research unit at the National Institutes of Health (NIH). Located in Bethesda, MD, a suburb of Washington, the NIH was, at the time, establishing itself as the world leader in biomedical research. And every year they accepted a handful of young doctors for research training while concurrently serving as commissioned officers in the USPHS. For trainees like me at the Brigham, getting one of these plum positions was the perfect way to learn to do research while fulfilling the requirement for 2 years of uniformed service as a medical doctor. Of the units at the NIH that offered these positions which interested me most was in the Clinical Endocrinology Branch of the National Institute of Arthritis and Metabolic Diseases. Competition for this position was intense, and applications and interviews had to be arranged more than a year advance, in the midst of my hectic internship. Of all the jobs I ever applied for this was the one that I was most eager to get, and most worried would elude me. I was very relieved to learn, in April 1959, that I would be appointed as a Senior Assistant Surgeon in the USPHS and was assigned to the Clinical Endocrinology Branch under the supervision of J. Edward Rall, beginning July 1, 1960.

Becoming a Scientist at NIH


The 3 years I spent at the NIH were, in my mind, the time of transition from being a student to being an adult. For the first time in my life I would be making a living, earning about $6,000 dollars a year, a princely sum 20 times as much as the $300 dollars per year I had earned as an intern.

10

Samuel H. Barondes

This meant I could have a real residence; first, a house I shared with two friends; and then, a little apartment of my own at 2700 Q Street in Georgetown, a 20-minute drive to NIH. And I had a serious girlfriend, Ellen Slater, whom I had known since medical school and would eventually marry. Yet, in many ways I was still a student. Even though I had become a competent physician, I would need a lot of guidance to get started in research. This need was particularly pressing because Ed Rall, who hired me, had just decided to turn his attention to administration and encouraged me to figure out my own path. Fortunately there was Ira Pastan. Ira had arrived as a trainee in the Clinical Endocrinology Branch a year before I did and had settled into the lab of Jim Fields, one of the labs that was open to me. Warm, insightful, and already a productive young scientist, Ira took me under his wing and taught me the essentials of biochemical experimentation. He was doing metabolic studies on slices of thyroid gland and, after some brief discussion, I decided that I would do similar experiments with slices of the pituitary gland, that master gland seated beneath the hypothalamus that had first attracted me in medical school. Having learned that the hypothalamus contained serotonin and norepinephrine, two neurotransmitters that might regulate pituitary functions, I decided to study their effects on the metabolism of pituitary slices using the same techniques that Ira was using in the thyroid. The results were dramatic. Both serotonin and norepinephrine increased glucose oxidation by pituitary slices by way of the hexose monophosphate pathway, and the increases were impressive--up to fivefold. Within months of arriving at NIH I submitted a paper to Endocrinology, a top journal, which soon accepted it. But the micromolar concentrations of the neurotransmitters needed to produce these effects suggested that they might not be acting as ligands for receptors but in some other way. I soon found evidence for an alternative mechanism by blocking the action of the amines with monoamine oxidase inhibitors, drugs already in use by psychiatrists to treat depression. The effect of the drugs indicated that metabolites of norepinephrine and serotonin were the active agents, rather than the neurotransmitters themselves, raising questions about the physiological significance of this in vitro effect. My paper about this was promptly accepted by the Journal of Biological Chemistry, then the top journal in the field. These early experiences at NIH influenced me greatly by showing m e - and o t h e r s - - t h a t I was a competent experimentalist. They also brought me into contact with several outstanding NIH scientists to whom I turned for help. Among them was Julie Axelrod who was then doing his Nobel Prizewinning work on norepinephrine metabolism and who gave me reagents, encouragement, and advice. Julie and I became friends and remained in touch for the rest of his life. But my career in the laboratory was almost aborted by an unexpected event. Shortly after I arrived at NIH as a commissioned officer

Samuel H. Barondes

11

in the USPHS, John F. Kennedy was elected President. As is now well known, but was then top secret, Kennedy had Addison's disease--adrenal insufficiency--with very low levels of corticosteroids that required daily replacement therapy. Because symptoms of Addison's disease can be exacerbated by stress, Kennedy's doctors wanted to have someone around who could immediately deal with any signs of deterioration in his condition. From what I have been able to piece together they decided to assign this duty to a young endocrinologist in the uniformed services who had worked with patients with Addison's disease. Because of my experience at the Brigham I was a reasonable candidate and, after being sworn to secrecy, was asked if I would like to be considered. I said yes, and assuming they would never pick me, thought no more of it. A few days later my mother called me in a panic. Government agents, she told me, were asking questions about me all over Brighton Beach, and she was worried that I was in some kind of trouble. She was greatly relieved by my explanation that they must be checking out my security clearance for an important assignment but still worried because one of the people they interviewed was an artist who had a studio in our basement. To understand why she continued to be worried I need to tell you more about the ideologies of the people of Brighton Beach. As I described my childhood you might have the picture of a community filled with orthodox rabbis on the lookout for violators of the laws governing the Sabbath. But the fact of the matter is that the most prominent belief system in the Brighton Beach of my youth was not Judaism but Socialism, and many residents were even members of the Communist party. My mother had reason to believe that this might include the man with the studio in our basement who was interviewed by the government agents. Whether or not her belief was correct, I was soon politely informed that my services at the White House would not be needed after all. The explanation I was given was that they had chosen someone from the Navy because this had been Kennedy's branch of the service. I never found out what really happened, and my FBI file, which I later obtained, makes no mention of this episode. Whatever the reasons, I was free to continue with my research at NIH.

Gordon Tomkins and Marshall Nirenberg: From Endocrinology to Molecular Biology


My research was about to take a major turn because of a conversation with Gordon Tomkins, a brilliant and charismatic scientist who would soon become my mentor (Fig. 1). About 7 years older than me, Gordon was a Californian who had gone to medical school at Harvard and also interned at the Brigham. Interested in hormones he had thought about doing some clinical work but decided he belonged in the lab and got a Ph.D. in biochemistry from UC Berkeley. When I met him he was settling into a position as

12

Samuel H. Barondes

Fig. 1. Gordon Mayer Tomkins (1926-1975) in 1974.

head of a newly formed Laboratory of Molecular Biology, which was located just around the corner from the Clinical Endocrinology Branch in NIH's massive Building 10. My first conversation with Gordon is as memorable to me as the one I described with my uncle Joe. When he asked me what I was interested in I answered "endocrinology," and he offered a startling response: "You know what endocrinology is? Endocrinology is just molecular biology." To which I replied, "What exactly is molecular biology?" To justify my ignorance you must understand that this conversation happened early in 1961, when molecular biology was not exactly a household word. But Gordon had already realized that hormones work by regulating gene expression, and he decided, on the spot, to tell me why. In the course of the next 2 hours he explained the central dogma of molecular biology: that regions of DNA act as templates for the synthesis of specific messenger RNAs that, in turn, encode the structures of specific proteins. In Gordon's view hormones work by changing the synthesis of certain messenger RNAs and the proteins they encode, thereby influencing biological functions. His explanations were so convincing and his personality so warm and inspiring that, by the end of the conversation, I asked him if I could join his lab. He told me he had a better idea. Instead of immediately working with h i m ~ w h i c h was especially problematic because he would soon be going to Paris for a sabbatical~I should first work with a young biochemist whom he had recently hired as a member of his unit. In Gordon's view this young man, whose tiny lab was just a few doors away, was a brilliant experimentalist who could teach me a lot. Furthermore he had only one post-doc in

Samuel H. Barondes

13

his lab, and Gordon thought he could also use another pair of hands. This was how I came to work with Marshall Nirenberg. Moving into Marshall's lab turned out to be a very lucky break, because the following month he and his post-doc, Heinrich Matthaei, made an extraordinary discovery. They had developed a system for studying in vitro protein synthesis by extracts of Escherichia coli, and they were adding various types of RNAs to the extracts to see if they could direct the incorporation of radioactive amino acids into proteins. Among those they tested was a synthetic RNA, polyuridylic acid (poly-U). This simple experiment had two amazing results: poly-U did, indeed, direct the synthesis of radioactive protein; and the protein product contained only a single amino acid, phenylalanine. Because there was already reason to believe that sequences of three nucleotides in RNA directed the incorporation of a particular amino acid into p r o t e i n ~ t h e so-called triplet code~these results raised the possibility that the code for phenylalanine was a sequence of three uridines, a possibility that was soon confirmed. Amazing as that result was in itself, it quickly became clear that Marshall and Heinrich had not only found the first component of the genetic code but also a way of finding the nucleotide triplets that encoded all 20 of the amino acids found in proteins. As they set out to follow this lead, I was given the assignment of finding out what happens to poly-U when it is added to the E. coli extract. Over the next year I discovered that poly-U associates with clusters of ribosomes that were just being implicated in the translation of messenger RNA into protein. This was not only interesting in itself but also was further evidence that the synthetic polynucleotide was, indeed, acting like a real messenger RNA. I also discovered that a single molecule of poly-U could direct the synthesis of multiple copies of the artificial protein called polyphenylalanine, providing direct evidence that messenger RNA could be used over and over again, and was not used up in the synthesis of a single protein molecule. These findings were considered to be of such great importance that when Marshall and I submitted them to Science they were promptly published as back-to-back papers. Even more exciting than these successful experiments was my immersion in a lab that was engaged in one of molecular biology's greatest adventures, the race to decipher the genetic code. Marshall had become the frontrunner with the initial finding with poly-U. But as soon as that became known, others began using the same approach. Most notable among them was Severo Ochoa, who had already won a Nobel prize for work with polynucleotide phosphorylase, the enzyme used to make poly-U. Over the next few years Marshall would earn his own Nobel prize. But despite my association with this groundbreaking research and the thrill of being pictured in the newspapers as a member of NIH's "code of life team" I decided to follow a plan I had made with Gordon to join him in Paris for the last

14

Samuel H. Barondes

3 months of his sabbatical. In July of 1962, with Marshall's blessing, I set sail for France. Those 3 months were a welcome respite from years of hard work. Although Gordon had a lab at the CNRS laboratory in Gif-sur-Yvette, in the outskirts of Paris, he had wound down his experiments because of the summer vacation season, and there was a lot of free time for conversations, visits with scientists at the Institut Pasteur, and exploration of Paris. It was also a great opportunity for me to travel through the French countryside with Gordon and his wife Millicent, a gifted artist and singer, who gave solo performances at churches along the way. In the course of these excursions Gordon and I frequently talked about our future plans. Mine were once again in flux. The exciting discoveries in Marshall's lab, the inspiring mentorship of Gordon, and my considerable personal success at the lab bench had, together, hooked me on a career in science. But instead of simply cruising in the wake of these two exceptional young men I wanted to find my own way, and this led me back to my college ambition to be an experimental psychologist and my short-lived interest in psychiatry. Having learned from Gordon that "endocrinology is molecular biology" it did not require much imagination to consider that the brain mechanisms that control behavior could also be thought of in terms of this exciting new field. And because I was also interested doing something that would be clinically relevant, it occurred to me that much of psychiatry is also molecular biology and that my new training might even qualify me for a career in psychiatric research. I also had an idea about bridging the gap between molecular biology and behavior which came from Mike Sporn and Wes Dingman, two contemporaries of mine at NIH. Mike and Wes were interested in messenger RNA, which had then been identified in bacteria but not yet in mammals, and we collaborated on a study showing that RNA isolated from rat liver nuclei had potent messenger activity in E. coli extracts, a simple experiment then considered to be so significant that it was published in Nature. But even more important to me than this bit of work was that Mike and Wes got me interested in the idea that messenger RNA synthesis was involved in the storage of memories in the brain, which suggested a way of using molecular biology to study a mental mechanism. Mike and Wes had already done a pioneering experiment that supported this idea by injecting 8-azaguanine, an analogue of a normal precursor of RNA, into the cerebrospinal fluid of rats just before training them in a swimming maze. They found that rats injected with this chemical did not show the same progressive improvement of performance as controls that were injected with saline. This raised the possibility that the drug-treated rats had difficulty learning because their brains were making dysfunctional 8-azaguanine-containing messenger RNA and that synthesis of functional brain messenger RNA was needed for normal learning. These findings,

Samuel H. Barondes

15

and the alternative interpretations they considered, were published in 1961 in the first issue of The Journal of Psychiatric Research, but got little attention because of their appearance in this obscure new publication. When I mentioned this line of research to Gordon he suggested that I discuss it with a psychologist friend whom he had met as an undergraduate at UCLA. The friend, who was studying memory in mice, turned out to be none other than Murray Jarvik, who taught me behavioral biology at Columbia. Murray happily agreed to come to NIH for some experiments, bringing with him a simple apparatus for measuring passive-shockavoidance learning. To test the role of messenger RNA in learning we used a new drug, actinomycin-D, which inhibits RNA synthesis, and measured the effects of injecting different doses into mouse brains by studying incorporation of a radioactive precursor into RNA. We were disappointed to find that mice whose brain RNA synthesis was substantially inhibited learned the simple passive avoidance task as well as controls and had normal memory 3 hours later; and experiments with larger doses of actinomycin were abandoned because of the toxicity of the drug. Nevertheless, the approach was so exciting that it got me several invitations to present symposium papers, including one at the American Psychological Association in 1964. These presentations, which combined the tutelage of Gordon Tomkins with speculations about the molecular processes that control synaptic connections, were subsequently summarized in Nature as "Relationship of Biological Regulatory Mechanisms to Learning and Memory." While the experiments with Murray Jarvik were ongoing I began getting tempting job offers. But I had already decided to satisfy my ambition to become a psychiatric researcher by signing up for a residency in psychiatry at McLean Hospital, a Harvard-affiliated mental hospital in a suburb of Boston. Gordon, who was skeptical, assured me that if I hated working with psychiatric patients NIH would take me back and give me a laboratory of my own. Comforted by his continuing friendship I put my three eventful years at NIH behind me and set out for three more at McLean.

Becoming a Psychiatrist and Neuroscientist at McLean My life was also about to change in another way. Right after the move from NIH, Ellen Slater and I got married. Fortunately I had been awarded a special fellowship, which paid a living wage rather than the meager resident stipend of the time. This allowed us to live in a small apartment in Cambridge at 60 Brattle Street and to pay tuition for Ellen at the nearby Harvard School of Education, which she soon began attending. We could also afford an occasional babysitter when our first daughter, Elizabeth, was born the following year.

16

Samuel H. Barondes

Both of us really needed that help. Ellen, who had majored in poetry and French literature at Vassar, was struggling to integrate graduate studies in a more practical subjectmpsychological counselingmwith the mothering of a demanding infant; and I was working overtime learning psychiatry, setting up a lab, and doing my share of baby care. But it was also the best of times, building a family while each of us learned a new field. McLean proved to be a wonderful place for me to learn psychiatry, providing stimulating teachers such as Alan Stone and Alfred Stanton and fascinating patients, many drawn from the vast faculty and student populations in the Boston area. I was also pleased with the laboratory facilities. Directed by Jordi Folch-Pi, a distinguished neurochemist, they were located in a brick building on the bucolic McLean campus, a short walk from the ward where I worked. The scientists, who all had appointments at Harvard, greeted me warmly as a young colleague. I was the first psychiatry resident to work in their midst and they were happy to forge this link with the clinical world that surrounded them. But the biggest break of all was the appearance of Harry (Hersh) Cohen, a graduate psychology student from Tufts who sought me out shortly after I arrived at McLean. Hersh had heard about my interest in memory and received permission to do the research for his Ph.D. thesis with me. This stroke of good luck made it possible for me to keep working on the molecular basis of memory s t o r a g e ~ a field that was heating u p ~ while holding down my job as a full time psychiatry resident. It was also wonderful to feel that I was no longer simply a trainee, because I now had my first graduate student.

Early Studies of Protein Synthesis and Memory


Hersh did not waste much time in getting started. Because we had no behavioral equipment at McLean he went to the hardware store and got the materials to build it. In a matter of weeks he made two mazes with electrified floors, which we used to train mice to escape or avoid shock, a more complex task than Murray Jarvik and I had employed. With these mazes we reexamined the effects of actinomycin D and again were stymied by its toxicity. This led us to shift our attention from RNA synthesis to protein synthesis using a new drug, puromycin, that had already been used by others for this purpose. Puromycin, whose general effects on mammalian protein synthesis were discovered around 1959, was first used in memory experiments by Josefa and Louis Flexner and colleagues at the University of Pennsylvania. In July 1963, just as I arrived at McLean, the Flexners published a paper in Science showing that injections of puromycin into mouse brains 1 day after maze-learning impaired memory tested 3 days later. In contrast, identical

Samuel H. Barondes

17

injections 6 days after training had no effect on memory. In the next 2 years Bernie Agranoff and colleagues at the University of Michigan, who were studying learning in goldfish, also began reporting effects of puromycin on memory. But Agranoff's group found that injections of puromycin after learning impaired memory 0nly if given within 30 minutes after completion of training, which was a much shorter interval than the Flexner group reported. When Hersh and I examined the effects of puromycin injections before trainingmso that protein synthesis was already inhibited when the mice learned to solve a m a z e m o u r results fit better with those of the Agranoff group. In our first study we found that injections of puromycin into the mouse brain before training did not interfere with learning but that memory deteriorated in the 3-hour period after training and was virtually absent thereafter. These and other controlled experiments were interpreted to mean that memory during training and for minutes after training ("short term memory") is not dependent on brain protein synthesis, whereas memory thereafter ("long term memory") requires brain protein synthesis. In the paper we published in Science we also raised the possibility that the Flexners' finding of an amnesic effect of puromycin injections 1 day after training could mean that there is a third phase of memory storage that operates over this longer time frame.

Axoplasmic Transport of Brain Proteins


In the course of these behavioral experiments, I turned my attention to a distinctive feature of neurons that might have bearing on the role of protein synthesis in memory storage. It was known, from experiments with peripheral nerves by Paul Weiss and others, that neuronal proteins are slowly transported from a site of synthesis in nerve cell bodies down the axon to nerve terminals. If protein synthesis was, indeed, required for memory, and if this protein works by facilitating synaptic functions in nerve terminals (in addition to or instead of on the postsynaptic side), it became important to know how quickly new proteins are transported to the nerve terminals along the short axons in mouse brains. To study this I took advantage of the recent finding that homogenization of brains under appropriate conditions shears off nerve endings, which can be isolated on sucrose gradients as particles called synaptosomes. The method I invented was to inject radioactive leucine into mouse brain and to compare the rate of radioactive protein appearance in whole brain homogenates and in synaptosomes--including soluble and particulate fractions of synaptosomes that could be separated by further disruption and gradient centrifugation. Early in my psychiatric residency I worked out this method, which I described in a paper in Science in November 1964. In subsequent work I showed that radioactive protein begins to appear at nerve endings within 15 minutes after its synthesis,

18

Samuel H. Barondes

during the same period that short term memory is being converted into protein-synthesis-dependent long term memory. My work with axoplasmic transport of proteins proved particularly auspicious because it brought me to the attention of Frank Schmitt. Frank, a distinguished biophysicist at MIT, had founded the Neuroscience Research Program (NRP) in 1962 to help develop the nascent field of neuroscience. This was largely accomplished by organizing small work sessions at NRP headquarters in the Brandegee Estate in Brookline. While I was still a resident at McLean, Frank asked me if I would participate in a work session on axoplasmic transport, which would be held the following year, in April 1967. The conference would be an important one because it would bring together a rich mix of people including the great Paul Weiss, the acknowledged leader in the field. But Frank was very concerned that Paul, a senior scientist with a strong personality, would be too domineering to serve as chair of this work session, and he and his colleague, Fred Samson, asked me to take on that task~jokingly explaining that my psychiatric training might be helpful in chairing a meeting that was likely to become very stormy. I was particularly pleased by this request because it helped me realize that I was becoming recognized as a player in neuroscience despite my lowly position as a psychiatry resident. Fortunately the Work Session on Axoplasmic Transport, which was published as an NRP Bulletin, proved to be a great success, and Paul Weiss and I became friends.

First Job at Einstein


In my final year of residency I was actively recruited by the faculties of several venerable East Coast schools. But I was most attracted to a New York City newcomer, Albert Einstein College of Medicine, which offered me an assistant professorship of psychiatry with a joint appointment in molecular biology. The Department of Molecular Biology, which had been founded by Bernie Horecker, a distinguished biochemist, was probably the first of its kind in a medical school, reflecting the innovativeness that was typical of the Einstein of that time. My startup package consisted of a brand new laboratory in the Department of Psychiatry built to my design by remodeling a large tile-walled space that was originally used as a l a v a t o r y ~ a n indication of the lack of research space in psychiatry departments of that period. My only duties as a psychiatrist were to meet with and supervise residents and to spend a few hours a week seeing patients. The rest of my time could be devoted to research, which was soon supported by a research grant and a Career Development Award from the National Institute of Mental Health (NIMH). Coming back to New York, our home town, was exhilarating for Ellen and me. It became even more so with the birth of our second daughter Jessica less than a month after we arrived. We rented an apartment with a

Samuel H. Barondes

19

view of the Hudson on Palisades Avenue in Riverdale and settled into our wonderful new life.

Protein Synthesis and Long Term Memory


Work in the laboratory had also gotten off to a fast start because Hersh Cohen, now a newly minted Ph.D., joined me at Einstein as a post-doc to continue our work on protein synthesis and memory. That work had been greatly advanced by the Agranoff group's experiments with acetoxycyclohemide, a protein synthesis inhibitor that worked by a different mechanism than puromycin. In 1966 they published a paper in the first volume of a new journal, Brain Research, which reported that acetoxycycloheximide injections into goldfish brain immediately after learning blocked memory measured 3 days later, confirming their results with puromycin. But the picture was clouded by the Flexners' 1966 report that, in striking contrast with their pioneering study with puromycin, acetoxycycloheximide injections into mouse brain 1 day after training did not interfere with memory~which raised the possibility that the amnesic effect of injections of puromycin so long after training was due to some other effect of the drug. Fortunately Hersh and I quickly discovered the reasons for this discrepancy. We found that puromycin produces abnormalities in brain electrical activity, including occult seizures, suggesting that this action~which is not shared by cycloheximide or acetoxycycloheximide--contributes to puromycin's amnesic effect. This interpretation was supported by the finding that diphenylhydantoin, an anticonvulsant, attenuated the amnesic effect of puromycin but not that of the other drugs. The upshot of these studies was that puromycin, which had played such an important role in sparking this line of research, was not really useful in studying the relationship of brain protein synthesis to memory because of the drug's powerful side effect on brain function, and that its amnesic effect when injected 1 day after training appeared to be due to occult seizures (like the retrograde amnesic effect of electroconvulsive shock) rather than to inhibition of brain protein synthesis. Having cleared up the confusion generated by puromycin's side effects, Hersh and I went on to a long series of experiments that showed that intracerebral or subcutaneous injections of cycloheximide or acetoxycyloheximide before training have no effects on initial learning but do indeed interfere with memory measured a few hours after training and thereafter, and that the critical protein synthesis is initiated within minutes after training under our experimental conditions. These results were consistently obtained in carefully controlled studies with mice that studied maze learning motivated by either shock avoidance or a water reward. When taken together they strongly supported our conclusion, and that of the Agranoff group, that learning and "short term" memory are not

20

Samuel H. Barondes

Fig. 2. Participants at Consolidation of the Memory Trace Work Session, Neuroscience Research Program, Brookline, MA, November 1967. Left to right. B a c k r o w : Hersh Cohen, David Quartermain, Joe Parks, Ted Melnechuk, Richard Roberts, Bruce McEwen, Adrian Rake, Jan Bures. M i d d l e r o w : Patricia Dimond, Catherine LeBlanc, George Adelman, Wardwell Holman, Everett Johnson, Tony Deutsch, Bernie Agranoff. F r o n t r o w : Roy John, Steve Chorover, Samuel Barondes, George Koelle, Seymour Kety, Gardner Quarton, Neal Miller (Work Session Chair), Frank Schmitt, Murray Jarvik.

dependent on brain protein synthesis, whereas "long term" memory, which is being established in the few hours after training, is, indeed, dependent on brain protein synthesis. The implication of these results was that the newly synthesized proteins play a role in the alteration of the functional synaptic connections that store the memory, which became the topic of a historic work session at NRP in 1967 (Fig. 2).
Brain Protein and Glycoprotein Metabolism

While this work on memory was going on I continued to study the rapid transport of newly synthesized brain proteins to nerve endings. Further proof came from autoradiographic studies of synaptosomes with the electron microscope, in collaboration with Bernard Droz. We found that labeled protein could be directly visualized in synaptosomes within 15 minutes after injections of radioactive amino acids, and the significance of the finding coupled with the novelty of the technique led to its publication

Samuel H. Barondes

21

in Science. I also became interested in the possibility that not all proteins at synapses originated in the neuronal cell bodies and that some could be made by mitochondria in nerve terminals. F u r t h e r work in this general area was done with Gary Dutton, a new post-doc and with Howard Feit, an M.D.-Ph.D. student. With them I began studying the metabolism and axoplasmic transport of microtubule protein (tubulin), a specific brain protein known to be a major component of axons. Tubulin was a particularly attractive subject because it was easy to purify from brain homogenates by precipitation with vinblastine followed by electrophoresis on polyacrylamide gels. A major result of these studies was that this structural protein, a major component of axons, turns over with a half life of several days. This finding helped change the picture of the brain, which had been viewed as structurally very stable. The true picture was that the neuron was constantly changing, a theme elaborated in Cellular Dynamics of the Neuron, the 1969 book I edited, which was based on a Paris meeting on the subject. Included in that book was a report of our findings in a new field, the modification of brain proteins by glycosylation, which I had begun studying with my post-doc, Gary Dutton, and with Marty Zatz, an M.D.-Ph.D. student. I had become interested in glycosylation for two reasons. First it seemed to me that this posttranslational modification might be an important way to modulate the function of brain proteins and could even play a role in short term memory in ways already envisioned for posttranslational phosphorylation. Second I had already become interested in the possibility that the glycoproteins on cell surfaces and in the extracellular matrix might play a role in cell adhesion and recognition, an interest that I would pursue for many years.

Life Changes
In the midst of this scientific excitement, disaster struck. Little more than a year after we arrived in New York Ellen found a lump in her breast. Although she was only 29, and the experts we consulted assured us that it was a benign fibroadenoma, it proved to be a cancer. Nevertheless, after radical surgery and extensive radiation, we were convinced that she was cured. As we gradually went back to our normal lives we began thinking about schools for the children and buying a house, which raised questions about where to settle down. I had, by then, come to the attention of other universities, which prompted Einstein to quickly promote me to tenure. I had also been appointed Director of Einstein's Interdepartmental Institute for Training in Research in the Behavioral and Neurological Sciences, a pioneering program organized by Saul Korey in 1957 with generous support from the NIH, and a program that fit with my personal goal of increasing the role of

22

Samuel H. Barondes

biological science in psychiatry. But despite my professional satisfaction at Einstein, Ellen and I were reluctant to put down roots in New York. Instead we were increasingly drawn to California, which I had visited several times because of a generous offer from Stanford. The turning point came when Salvador Luria invited me to give a lecture at the Salk Institute for their nonresident fellows. When I accepted this invitation I was contacted by Arnold Mandell, a young biological psychiatrist who had just been appointed as the founding Chair of Psychiatry at the brand new medical school at the University of California in San Diego. Arnie said he would come to my talk and would like to show me around. It was a thrilling visit. I was stunned by the majestic appearance and intellectual vitality of the Salk Institute, the warm reception I received from Jonas Salk and the assembled scientists, and the grand vision for UCSD that was just beginning to rise on a vast campus across the road. When Arnie explained his dreams of a research-based Department of Psychiatry, and offered me a full professorship on the spot, the opportunity to participate in this new adventure seemed irresistible. On a return visit with Ellen she was as enthusiastic as I was. Before making our final decision we had another treat in store for us. The whole family had been invited to spend three weeks in Boulder, Colorado at the Second Intensive Study Program of the Neurosciences Research P r o g r a m ~ a n o t h e r of the contributions of NRP to my personal scientific development and that of the emerging field of neuroscience. It was a great experience for me because it led to lasting friendships with other participants, such as Gunter Stent. It was also a wonderful holiday for the children, their first time in the mountains and on horseback, and helped to convince us that we were ready to move west. Four months later, in December 1969, we all got in the car to move to La Jolla.

Building Psychiatry and Neuroscience at UCSD


Our departure came at a convenient time for my coworkers. Hersh Cohen, who was being recruited for faculty positions, made a surprise move to Wall Street and now holds a major position with Citigroup. My two M.D.Ph.D. students got their degrees and moved on to clinical training, Marty Zatz in psychiatry and Howard Feit in neurology, and both went on to academic careers. Gary Dutton, a post-doc, decided to join me at UCSD with salary support from my NIH grant. So did Larry Squire, whom I had met at Einstein where he was a post-doc in Murray Jarvik's lab, and who welcomed the opportunity to check out California. Having Gary and Larry join me in this move allowed me to get a research program going while I attended to my other duties at UCSD. These were quite numerous because our Department of Psychiatry was then made up of just two faculty members" Arnie Mandell and me. I had tried very hard to entice Sol Snyder to sign up with us, and UCSD was willing to offer

Samuel H. Barondes

23

him a full professorship despite his young age, but Johns Hopkins had the good sense to make him the same offermand to eventually make him the founding chair of their new Department of Neuroscience. Which left it up to Arnie and me to do all the teaching, clinical work, and administration, as well as to build up our labs. It was a great relief when Lew Judd arrived the following summer to share the enormous load. As soon as I arrived at UCSD I also began working to help build an interdepartmental neuroscience program like that at Einstein. The nucleus of this program had already been established by Bob Livingston. Working with him and Jon Singer, a professor of biology, I submitted a proposal to the Alfred P. Sloan Foundation, which was, at the time, very interested in fostering neuroscience. They gave us a generous grant, which I administered, to support students, post-docs, and research. This grant provided funding for about 10 years and played an important role in the development of UCSD's neuroscience program, which is now a world leader. Ellen also had a lot to do to get us settled. After months of searching she found us an affordable Frank Lloyd Wright-style house with an ocean view at 1642 Kearsarge Road, on the lower part of Mount Soledad. It was within walking distance of downtown La Jolla and the La Jolla Elementary School, which my daughters would attend, and a 5-minute drive to UCSD. We could not believe our good fortune. But as soon as we moved in, disaster struck again. Ellen's cancer had spread to her liver; and this time we knew it was a death sentence. Seizing what time she had left we lived through grim treatments and exhilarating remissions for a year and a half. When it ended she was only 33, and I was alone with my two little girls, barely 5 and 7 years old. What saved me during my darkest time was Ellen's parents. In the midst of her illness they retired and moved to La Jolla, just a few miles from our home; and when Ellen died they were there to help me with the responsibilities of a single parent. Their presence was also a godsend for Elizabeth and Jessica who knew that they could always rely on their beloved Grandma Fanny.

Molecules and Memory


While I was reeling from this tragedy, the work in the lab went on. Larry Squire began using cycloheximide to more clearly define the sequential phases of memory in mice, an interest he would maintain in his later research on human memory. Much of his work used the Deutsch Carousel, an automated machine designed by Tony Deutsch, a professor of psychology at UCSD, to study learning and memory of a discrimination task. With this apparatus Larry found evidence that a protein-synthesis dependent component of mouse memory can already be detected during the course of prolonged training, which helped refine our view of the stages of memory. He went on to show that anisomycin, a protein synthesis inhibitor that is structurally different from cycloheximide, is equally effective in blocking

24

Samuel H. Barondes

long term memory, which greatly increased our confidence that the amnesic effect of these drugs is really due to inhibition of protein synthesis rather than to some unknown side effect. While completing these studies Larry established his own research program at UCSD where he is now Distinguished Professor of Psychiatry, Neurosciences, and Psychology (as well as editor of this series of volumes). As we became convinced that newly synthesized brain proteins are needed to store long term memories, and that reversible modifications of existing proteins are probably essential for short term memory, the next problem was to identify the relevant proteins. In the early 1970s this seemed to be an impossible task, because the proteins involved in a particular memory were likely to be confined to a limited number of neurons in the vast mouse brain. Encouraged by conversations with Eric Kandel, who had already discovered electrophysiological correlates of learning in single cells of Aplysia, I decided to try to follow his lead in the hope of ultimately identifying the proteins involved in plasticity at identified synapses. Working with Aplysia californica also was attractive because these animals live along the coast of La Jolla and could be readily harvested from tide pools that were within a few miles of our lab. And even though I had no experience in cellular neurophysiology, I had recruited Werner Schlapfer, a post-doc, and Paul Woodson and Jacques Tremblay, two graduate students, who were eager to give it a try. By 1974 we began describing various forms of synaptic plasticity measured in identified cells in the abdominal ganglion of Aplysia and the influence of exogenous neurotransmitters and drugs. To examine the molecular effects of these reagents we turned to Irwin Levitan, a postdoc with a background in biochemistry. He found that serotonin and octopamine increased levels of cyclic AMP in the abdominal ganglion as well as the phosphorylation of a prominent protein peak that could be resolved by electrophoresis on a polyacrylamide gel. The combined results of the electrophysiological and molecular studies was very encouraging because they raised the possibility that this phosphoprotein might be involved in a form of memory. But the amount of tissue in the abdominal ganglion was too small for detailed studies of such proteins with the molecular tools of the mid-1970s. To me this seemed like a decisive limitation of this line of research that made it less attractive than the other project I was concurrently engaged in, which was turning up abundant pure proteins with intriguing functional properties.

Slime Molds, Discoidins, and Vertebrate Lectins


The competing project in my lab grew out of my interest in glycosylation of brain proteins and their potential role in the formation of synaptic connections. The development of synaptic connections was of particular interest

Samuel H. Barondes

25

to me because I believed that abnormalities in the molecules that control this process might be responsible for the variations in neuronal circuits that may lead to certain forms of mental illness; and my fascination with cell surface glycoconjugates had been kindled by Vic Ginsburg, another young member of Gordon's Laboratory of Molecular Biology, whose lab was across the hall from Marshall's. Vic was one of the early proponents of the now widely accepted idea that specific cellular associations may be controlled by the precise structures of the complex carbohydrates on and around cell surfaces. Just as Marshall was working on a nucleic acid code that determined the structure of proteins, Vic believed there is a sugar code that determines intercellular interactions. But unlike the nucleic acid code which is inscribed as a template made up of four nucleic acid building blocks, the sugar code was presumed to be based on progressive incorporation of six sugar building blocks (galactose, mannose, N-acetyl-glucosamine, N-acetyl-galactosamine, fucose, and sialic acid) into complex sugar chains under the direction of a particular combination of enzymes, the glycosyltransferases, that are expressed in particular cells. How these complex sugar structures on cell surfaces actually mediate cell-cell interactions was not something Vic worried about. But their potential role in this process was what excited him. It was Vic's idea, and its relevance to the formation of specific synaptic connections that stimulated my work with Marty Zatz and Gary Dutton, and that formed the basis of one of the p a p e r s ~ " B r a i n Glycomacromolecules and Interneuronal R e c o g n i t i o n " ~ t h a t I presented at the NRP meeting in Boulder in 1969. The direction of my thinking about this problem took a big turn in 1972 with the arrival of Steve Rosen, a new post-doc. Steve had begun his graduate work at Cornell with an interest in memory but went on to do his thesis on cell adhesion in Dictyostelium discoideum, a cellular slime mold. This organism exists in two forms: as a unicellular ameba that lives on soil bacteria; and as a member of a colony of thousands of cells that stream together, adhere to each other, and differentiate into a multicellular organism called a fruiting body. The transformation from unicellular nonadhesive cells to aggregating adhesive cells is induced by starvation and occurs over the course of about 8 hours. As a graduate student Steve made the serendipitous discovery that, in the course of this transformation, the aggregating cells make a substance that agglutinates erythrocytes. This raised the possibility that the agglutinin is responsible for the developmentally regulated adhesion, and we agreed that when he came to my lab we would try to find out. When Steve arrived he quickly confirmed his earlier observations. Working with David Simpson, another post doc, he set out to isolate the active component of the extract by a standard protein purification technique, gel filtration on a Sepharose column, which separates proteins on the basis of their molecular weight. To his great dismay none of the fractions

26

Samuel H. Barondes

that came through the column had any agglutination activity. But because of our preconceived notion that the agglutinin might bind to carbohydrates, and knowing that Sepharose is a cross-linked polymer of galactose, he washed the column with a solution of galactose. This released the pure galactose-binding-protein, which we named discoidin, a new member of a class of proteins called lectins that had previously been found in plant extracts. Studies of the effects of simple sugars as inhibitors of hemagglution by purified discoidin showed that N-acetyl-galactosamine binds discoidin better than galactose, and that other simple sugars do not bind at all. These encouraging results, which we began publishing in 1973, suggested that cell adhesion involves interactions between carbohydratebinding-proteins and their ligands on cell surfaces or in the extracellular matrix. In the next few years Steve and Dave, working with Bill Frazier, Chen-Min Chang, and Dick Reitherman, accumulated evidence in support of this idea. In the course of this work they found that there are actually two discoidins, discoidin I and II, which are synthesized at different stages in the development of the multicellular organism, and that other species of slime molds also have their own distinct lectins. Stimulated by this work Tom Nowak, a new graduate student, began looking for developmentally regulated lectins in embryonic chick tissues by making extracts and screening for substances that agglutinate erythrocytes. He found some agglutination activity, but none that was blocked by simple sugars. Then, in the midst of these discouraging results, a paper describing an animal lectin was published by Vivian Teichberg and colleagues in the April 1975 issue of PNAS. Using methods like those in our papers on discoidin, they had detected agglutination activity in extracts of the electric organ of an eel, purified the relevant protein on a Sepharose column by elution with lactose, a beta-galactoside, and named the pure protein electrolectin. The main difference between their method and ours is that they included dithiothreitol, a reducing agent, in all their solutions. If dithiothreitol was omitted the lectin was quickly inactivated by oxidation. When Tom repeated his experiments using dithiothreitol or another reducing agent, beta-mercaptoethanol, he too found lactose-binding agglutinins in various tissue extracts. Concentrating on an agglutinin from embryonic chick muscle he discovered that its synthesis, like that of discoidin, was under striking developmental control, with an increase of 10to 100-fold between 8 and 16 days of embryonic development and a decline thereafter. Tom went on to purify the lectin with the help of David Kobiler and Larry Roel and to show that it is present on the surface of differentiating muscle cells. We soon found that it is also expressed in other cell types of interest, including neurons. These discoveries, following on the heels of our work with discoidin, led me to rethink my research priorities. Although I had worked for more than a decade on the molecular basis of memory storage, and Irwin Levitan had

Samuel H. Barondes

27

some promising findings that might allow for identification of molecules involved in synaptic plasticity in Aplysia, the work on lectins seemed more attractive for several reasons. First, the developmentally regulated lectins we found in slime molds and chick tissues were abundant proteins that we could easily purify in milligram quantities, in striking contrast with the minute quantities of poorly characterized proteins that Irwin had identified as radioactive peaks on polyacrylamide gels. In the 1970s, before recombinant DNA techniques made possible the synthesis of limitless quantities of any protein by translation of its cDNA, having milligrams of pure protein was a very big deal. Furthermore, the lectins had intriguing properties: they bound specific sugar-containing molecules on and around cell surfaces, which made them reasonable candidates for roles in specific cellular interactions. If we were lucky, lectins might even influence the formation of specific synaptic connections, a hypothesis I put forth once again in Neuronal Recognition, a book I edited in 1976. These considerations led me to gradually phase out my work on learning and memory as well as the Ph.D. explorations of my other students--Steve Flanagan, Elaine Traynor, Susan Newlin, and Paula Shadle--and commit myself to studies of lectins.

The Impact of Genetic Technology


The focus on lectins led to several discoveries. The first was stimulated by work in the laboratory of Rick Firtel, a colleague at UCSD. In 1981, using newly developed recombinant DNA techniques, Rick and his colleagues discovered that there were actually three genes encoding discoidin I, and published the deduced amino acid sequences of the three proteins. When we inspected these sequences we found that all forms of discoidin I contain the sequence arg-gly-asp, a sequence also found in fibronectin, a h u m a n cell adhesion molecule. Because it was already known that synthetic peptides containing this sequence block attachment of fibroblasts to extracellular matrix, Wayne Springer, a post-doc, and Doug Cooper, a graduate student, tested the effects of similar peptides on the adhesion of slime mold cells to various coated surfaces, and on their streaming into aggregates. We found that the synthetic peptides blocked attachment and streaming, as did univalent antibodies to discoidin I. From these and other experiments we concluded that, as with fibronectin, this sequence of three amino acids in discoidin I is a critical element in its biological function and that this part of the protein, rather than its carbohydrate-binding site, is the one clearly involved in cell adhesion. So the role of discoidin I in adhesion was confirmed, but the biological significance of its interaction with s u g a r s ~ the reason we were interested in it in the first place~is still not completely clear. Exciting though these findings were, the path of their discovery, and the ambiguous role of the carbohydrate site of discoidin I made me reconsider

28

Samuel H. Barondes

my commitment to continued work with slime molds. The main attraction of this experimentally favorable model system was as a potential source of discoveries that would guide research on humans. Indeed Wayne Springer and Andy Feinberg were busily developing methods to examine cell sorting in slime molds, which we hoped to apply to human cell recognition. But in our analysis of the action of discoidin I the guidance was clearly the other way around: our progress in understanding slime molds was based on discoveries already made with human fibronectin. This conclusion was premature. Beginning in 1993, with extensive sequencing of human DNA, many investigators found human proteins that contain discoidin-domains, as well as evidence that these domains play important roles in cell signaling and adhesion. Among these human proteins are two that are now called discoidin-domain-receptor-1 and -2 (DDR1 and DDR2). These integral membrane proteins were first identified in a genetic screen for tyrosine kinases. Then, in comparing their sequences with those in the gene data base, the computer revealed the surprising finding that the tyrosine kinases each have an extracellular domain that resembles discoidin I. This resemblance raised the possibility that these proteins bind to extracellular matrix; and, because their ligands were not then known, they were named discoidin-domain proteins. The suspicion that they might bind to extracellular proteins soon led to the discovery that they bind specifically to collagen and that this interaction activates specific intracellular signaling pathways. Of particular interest to neuroscientists, DDR1 is abundant in the brain, and has been directly implicated in synapse formation. So too are other discoidin-domain-containing proteins such as neuropilins and neurexins, which participate in brain cell adhesion and synapse formation by interactions with semaphorins and with neuroligin. Recently RS1, a cell adhesion protein that interacts with neuronal cells in the retina, has been shown to be an octamer of eight subunits each largely composed of a discoidin domain; and mutations in this domain cause retinoschisis, a common X-linked form of hereditary macular degeneration that affects males early in life. So the discovery of discoidin did contribute to studies of neuronal cell adhesion and synapse formation after all. Although I did not then know how genetic technology would make our discovery of discoidin relevant to cellular interactions in the human brain, it was already clear in the mid 1980s that vertebrate genes and tissues had become experimentally accessible in ways that I did not anticipate when plunging into slime mold research. Meanwhile we kept turning up new galactose-binding lectins in a variety of animal tissues. First Eric Beyer, an M.D.-Ph.D. student, purified one from chicken intestine that had different properties than the one we had found in chick muscle. Then Howard Ceri, Robert Cerra, Hakon Leffler, and Carl Sparrow found several galactose-binding lectins in rat and human tissues, and Marie Roberson

Samuel H. Barondes

29

found another lectin in Xenopus. When our immunohistochemical studies with fluorescence and electron microscopes localized these lectins on cell surfaces and in extracellular matrix, I drew attention to this new class of extracellular proteins with a review in Science. In other immunohistochemical studies with Tom Jessell, Jane Dodd, and their student, L.J. Regan, we found evidence for two different lectins in subsets of neurons of dorsal root ganglion and spinal cord, a step toward addressing their potential role in neuronal interactions that first attracted me to this field. Further discovery of related lectins followed when Michael Gitt began supplementing our biochemical work with gene cloning in the mid-1980s. While screening human cDNA clones with an antibody raised against a purified mammalian lectin, Michael found a few that encoded a second related lectin. When he identified the human genes that encoded these two lectins in genomic DNA he named them LGALS1 and LGALS2 (which encode the lectins we now call galectin-1 and galectin-2). As I became increasingly familiar with the new genetic technology I also began paying attention to its application to studies of heritable human diseases. Having served briefly as a consultant for the Hereditary Disease Foundation, I had been informed by Nancy Wexler of the search for the Huntington's disease gene; and, when the gene was mapped in 1983, I decided to explore the application of this technology to psychiatry. With the help of David Housman, who had played a key role in the Huntington's disease project, I organized a small conference---"Looking for Genes Related to Mental Illness"--at the Neurosciences Institute, the successor to the Neuroscience Research Program, which had been moved to Rockefeller University under the direction of Gerry Edelman. The highlight of this conference, held in October 1984, was the report by Housman and his post doc, Daniella Gerhard, of genetic studies of manic-depressive patients from Amish families identified by Janice Egeland. Using the same approach that had located the Huntington's disease gene, they had some evidence that a gene on chromosome 11 might influence the risk of developing manic-depression (bipolar disorder), a finding they would publish in Nature a few years later. As I developed my interest in lectin genes, and in the gene variants that influence the risk of mental illness, my professional life was about to undergo another major change. In 1985 I received a letter from Zach Hall asking if I would be interested in moving to the University of California School of Medicine in San Francisco (UCSF) as Chair of the Department of Psychiatry. I was already a big fan of UCSF, which my mentor Gordon Tomkins had joined in 1968, and which I visited several times around 1974 when, shortly before his untimely death, Gordon tried to recruit me to help set up a new neuroscience program. Zach Hall was eventually hired to head that program which he developed into a world leader. Now Zach was asking if I would help build bridges between neuroscience and psychiatry.

30

Samuel H. Barondes

Zach's inquiry came at a time when I was ready to consider such a move. Both my daughters were in college at UCLA, which greatly diminished my parenting responsibilities. With their departure I was eager to begin a new life, even find a wife, and it seemed to me that this might be easier if I moved to San Francisco. The city itself was yet another attraction. And UCSF was rich in colleagues and close friends, including Steve Rosen who was studying a lymphocyte lectin (L-selectin) like those we had been looking for in slime molds. Although I was not eager to take on the administration of a vast psychiatry department, the resources it could provide seemed an ample reward for these new responsibilities. The hardest thing about leaving La Jolla was the many friends I had made. Most of them were the people I worked with every day, such as David Segal and Lew Judd. But I would also miss the members of the First Thursday Dinner Club, which I had joined at its inception in 1979 along with Gustav Arrhenius, Francis Crick, Sandy Lakoff, Richard Lerner, Walter Munk, Leslie Orgel, Roger Revelle, Ellie Schneour, and Charlie Thomas. Drawn from UCSD, the Salk Institute, and the Scripps Research Institute, we met every month in La Valencia Hotel for conversations that I always looked forward to. It was also hard to leave my home on the La Jolla hillside, where my children had grown up, and my mother-in-law Fanny. Nevertheless, I accepted Zach's challenge. In September 1986 1 drove up to Los Angeles to visit my daughters and continued north to San Francisco to begin yet another new life.

Building Psychiatric Science at UCSF


The challenge proved to be even greater than I had imagined. The psychiatry department I inherited was still steeped in the psychoanalytic traditions that I had found so limiting when I was a medical student, and many of its faculty members were not happy with Dean Rudi Schmidt's decision to bring me in to steer it in a more scientific direction. Furthermore, the Langley Porter Psychiatric Institute, in which the department is headquartered, had just a few tiny labs, so space and resources would have to be diverted to build facilities for research. Such changes were bound to meet with the resistance I continuously struggled with during my 7 years as Psychiatry Chair. Fortunately there were some young people at UCSF who helped me move the department in this new direction. The first one I hired was David Cox, a medical geneticist. Along with Victor Reus, a biological psychiatrist, we set out to build an NIMH-sponsored program to hunt for gene variants in patients with bipolar disorder, following the lead of Housman and others. To flesh out this program I hired Nelson Freimer, a recent Langley Porter graduate, and enlisted the participation of Rick Meyers, a young geneticist in the physiology department. They soon joined forces in a new

Samuel H. Barondes

31

Neurogenetics Laboratory we built on the site of an abandoned kitchen in Langley Porter. Stanford, which is just a short drive from UCSF, was another source of talent. Among its psychiatry residents was Rob Malenka, a cellular neurophysiologist, and John Rubenstein, a molecular biologist with training in child psychiatry, both of whom I recruited as assistant professors. They were soon joined by three more young psychiatrist-scientists: Larry Tecott, a behavioral geneticist from our own residency program; Mark Von Zastrow, a cellular neurobiologist from Stanford; and Allison Doupe, who had done a psychiatry residency at UCLA and a post doctoral fellowship in systems neuroscience at Caltech. This group formed the core of UCSF's Center for Neurobiology and Psychiatry, which I have led since concluding my term as Chair of Psychiatry at the start of 1994. While these recruitments were going on I continued my research on lectins with the help of Hakon Leffler, Doug Cooper, and Michael Gitt, in a new lab built on the site of a former suite for occupational t h e r a p y ~ a more elegant shell for remodeling than the lavatory I had converted when moving to Einstein 30 years earlier. I also continued a collaboration with Tom Jessell on lectin expression in the nervous system, along with two of his Columbia trainees, Mary Hynes and Linda Buck~before her Nobel Prize-winning discovery of olfactory receptors. But my main aim during that period was to find all the vertebrate lectins we could, instead of concentrating on their biological functions. With the help of Steve Massa, Yuko Oda, Phillipe Marschal, and Margaret Huflejt, we discovered and characterized a number of new human, rodent, and frog galactose-binding-lectins. Hakon Leffler was particularly interested in discovering the specificity of each lectin by examining its affinity for various galactose-containing saccharides, a project he now continues as Professor of Laboratory Medicine at Lund University in his native Sweden. We did, however, devote a lot of attention to an unusual feature of these lectins. Like discoidin, none of them has a signal peptide characteristic of secreted proteins. Yet we knew that they become concentrated on the surface of cells and in the extracellular matrix. Working with a cell line of mouse myoblasts, Doug Cooper found that its lectin is initially confined to the cytoplasm and then accumulates in membrane evaginations which pinch off, releasing it outside the cell. The details of this novel secretory mechanism have still not been worked out. As it became clear that all the lectins we found share a carbohydraterecognition domain of about 130 amino acids, and that members of this family were also being discovered in other laboratories, and given a variety of names, I contacted the main investigators to reach a consensus about nomenclature. We agreed on the name galectin, with each of the known mammalian galectins receiving a number, and new ones to be numbered in the order in which they were found. This name has caught on: PubMed

32

Samuel H. Barondes

citations of galectin continue to grow exponentially, and galectins were recently the subject of a special issue of Glycoconjugate Journal, which the editors kindly dedicated to me.

Other Activities
Throughout my career I have been an eager participant in communities outside of my home institution. In recent years this has included membership in the scientific boards of biotechnology companies such as Dupont-Merck Pharmaceuticals, Guilford Pharmaceuticals, and Renovis Inc. The repeated meetings of these groups have taught me a lot about the practical applications of biomedical research and have fostered friendships with other members such as Sidney Brenner, David Martin, Sol Snyder, and Corey Goodman. Sidney also invited me to serve with him on Singapore's International Advisory Committee for Biomedical Research, and it has been fascinating to see how this small nation has positioned itself as a significant player in this field. But my most enduring and gratifying extramural association has been with the McKnight Endowment Fund for Neuroscience. An offshoot of the Minneapolis-based McKnight Foundation, its support of neuroscience began with consultations in the mid-1970s with Fred Plum and Julius Axelrod. They convened a founding committee that I was asked to join along with Edward Evarts, Seymour Kety, and James McGaugh. At our initial meeting in July 1976, chaired by Julie, we created a Scholars Awards program to help young faculty establish an independent career. This remains an influential program that supports about six new Scholars each year, many of whom have become leaders in the field. I served on the initial selection committee for these awards and have been associated with the development of new McKnight awards ever since. In 1986 Russ Ewald, the Executive Director, persuaded the Board of Directors of the McKnight Foundation to provide long term support for this program by spinning it off as an independent nonprofit organization~ the McKnight Endowment Fund for Neuroscience. Fred Plum was the founding President, and I succeeded him 3 years later, serving for almost a decade. I was succeeded by Torsten Wiesel and then Corey Goodman (Fig. 3) but continue to participate in various McKnight activities. I am very gratified by my long association with the McKnight Foundation and believe that its sustained commitment to neurosciences has had an impact that rivals that of the Neuroscience Research Program, which helped me so much in my youth. My involvement with the McKnight Foundation also changed my life in another way, by bringing me together with my wife, Louann Brizendine. Already on the UCSF faculty, Louann was acquainted with Larry Ellison, the billionaire founder of Oracle, and had interested him in the possibility

Samuel H. Barondes

33

Fig. 3. Presidents of the McKnight Endowment Fund for Neuroscience at the McKnight Conference on Neuroscience, Aspen CO, May 2000. Left to right: Torsten Wiesel, Samuel Barondes, Fred Plum, Corey Goodman. (Photo kindly provided by Peter Mombaerts.)

of establishing a foundation to support biomedical research. Knowing little about such activities she was advised by Eugene Roberts, a mutual friend, to consult me because of my work with McKnight. Early in 1996 Louann came to see me about this, and we soon developed a personal relationship, marrying in 2002. The foundation she worked to establish, with the help of Joshua Lederberg became the Ellison Medical Foundation, which Josh continues to lead. Aside from the McKnight Foundation, my most important extramural affiliation had been with the National Institutes of Health, a beloved alma mater which I have served almost continuously since I left in 1963 and which I returned to for several brief sabbaticals as a Fogarty Scholar, beginning in 1979. One of my favorite assignments was as Chair of the NIMH Genetics Workgroup whose members were Aravinda Chakravarti, Mary Claire King, Eric Lander, Bob Nussbaum, Ted Reich, Joe Takahashi, and Steve Warren. In 1997 we prepared a report that continues to shape NIMH policy on psychiatric genetics and that established the principle of sharing clinical data and DNA samples. I most recently served on NIMH's Board of Scientific Counselors, which I chaired from 2001 to 2003. I have also developed a new career as a writer of books for a general audience. My interest in publishing goes back to Cellular Dynamics of the Neuron in 1969 and continued with my editorship of Current Topics in Neurobiology, a series that included my 1976 book Neuronal Recognition.

34

Samuel H. Barondes

Over the years I served as Chair of the Publications Committee of the Society for Neuroscience and was on the founding editorial boards of The Journal of Neurobiology, The Journal of Neuroscience, Glycobiology, and

Molecular Psychiatry.
But I really got the writing bug when I was commissioned to write

Molecules and Mental Illness, a volume on biological psychiatry in the


Scientific American Library, which was published in 1993. It was followed in 1998 by Mood Genes, about the hunt for the gene variants that predispose people to mania and depression, and, 5 years later, by Better Than Prozac, about psychiatric drugs. Another one is in the works. Looking Ahead In reviewing my life in science it is hard to avoid the temptation to speculate about the future. I do this with trepidation because my past record of prognostication is a mixed one. On the one hand the prediction that molecular biology would transform neuroscience, which came to me via Gordon Tomkins, has been fulfilled so convincingly that students find it hard to believe that it was not always self-evident. On the other hand the impact of molecular research in psychiatry has not yet been as great as I expected. This is not to say that psychiatry has remained the same. During my training the intellectual core of the field was mainly Freudian, psychiatric residents were taught that psychopathology was the result of the traumas of early childhood, and the standard treatment was a form of wide-ranging psychoanalytic psychotherapy designed to undo this damage. Now the field has incorporated a great deal of brain science; residents are well aware that genetic susceptibility to mental disorders has an etiological role on a par with life events, and the more focused psychotherapies that have largely replaced psychoanalysis are heavily supplemented with drugs that influence molecular targets in the brain. But, impressive though these changes have been, progress has been slower than I anticipated. Geneticists who have been searching for relevant gene variants in patients with mental disorders have been frustrated by the complexity of the problem, because so many genes appear to be involved that it is hard to implicate any one with certainty. And despite the identification of a series of intriguing molecular targets for new psychiatric drugs, creating them has proved to be extremely difficult. As scientists at pharmaceutical companies know only too well, many promising compounds have been abandoned because of undesirable side effects, whereas others simply do not work or are no better than those already available. Nevertheless, I remain optimistic. We keep finding more affordable ways to identify variants in individual genomes, so the massive screens this makes possible are bound to identify many of the genes that influence the risk of mental illness. We also keep learning so much about the proteins

S a m u e l H. Barondes

35

and small molecules in the brain, and about ways to manipulate them, t h a t new t r e a t m e n t s will certainly follow. So it seems to me t h a t it is just a m a t t e r of time until we successfully translate this growing knowledge and technology into new ways to alleviate mental suffering. In the end, what has been most rewarding to me about my activities in neuroscience and psychiatry is t h a t it has allowed me to combine my interests in elegant science with my enduring fascination with the moral and psychological issues t h a t I was introduced to in my childhood. This rich mixture was the allure t h a t drew me to these fields. The allure remains.

Selected Bibliography
Barondes SH. The Influence of neuroamines on the oxidation of glucose by the anterior pituitary. I. The role of monoamine oxidase. J Biol Chem 1962;237:204-207. Barondes SH. Delayed appearance of labeled protein in isolated nerve endings and axoplasmic flow. Science 1964;146:779-781. Barondes SH. The relationship of biological regulatory mechanisms to learning and memory. Nature 1965;205:18-21. Barondes SH. On the site of synthesis of the mitochondrial protein of nerve endings. J Neurochem 1966;13:721-727. Barondes SH. Synaptic plasticity and axoplasmic transport, an essay. In Barondes SH, ed. Axoplasmic transport. Neurosciences Research Program Bulletin. 1967;5:365-370. Barondes SH. Further studies of the transport of proteins to nerve endings. J Neurochem 1968;15:343-350. Barondes SH. Incorporation of radioactive glucosamine into protein at nerve endings. J Neurochem 1968;15:699-706. Barondes SH, ed. Cellular dynamics of the neuron. New York: Academic Press, 1969. Barondes SH. Brain glycomacromolecules and interneuronal recognition. In Schmitt, FO, ed. The neurosciences: A second study program. New York: Rockefeller University Press, 1970;747-760. Barondes SH. Multiple steps in the biology of memory. In Schmitt FO, ed. The neurosciences: A second study program. New York: Rockefeller University Press, 1970;272-278. Barondes SH. Cerebral protein synthesis inhibitors block long term memory. Int Rev Neurobiol 1970;12:177-205. Barondes SH. Synaptic macromolecules: Identification and metabolism. Annu Rev Biochem 1974;43:147-168. Barondes SH, ed. Neuronal recognition. New York: Plenum Press, 1976.

36

S a m u e l H. Barondes

Barondes SH. Lectins: Their multiple endogenous cellular functions. Annu Rev Biochem 1981;50:207-231. Barondes SH. Soluble lectins: A new class of extracellular proteins. Science 1984 ;223:1259-1264. Barondes SH. The biological approach to psychiatry: History and prospects. J Neurosci 1990;10:1707-1710. Barondes SH. Molecules and mental illness. New York: Scientific American Library, 1993 (revised paperback edition, 1999). Barondes SH. Thinking about Prozac. Science 1994;263:1102-1103. Barondes SH. Galectins: A personal overview. Trends Glycosci Glycotech 1997;9:1-7. Barondes SH. Mood genes: Hunting for origins of mania and depression. New York: W.H. Freeman & Co, 1998 (Penguin Books, 1999; Oxford University Press, 1999). Barondes SH. An agenda for psychiatric genetics. Arch Gen Psychiatry 1999;56: 549-556. Barondes SH. Report of the National Institute of Mental Health's Genetics Workgroup. Biol Psychiatry 1999;45:559-602. Barondes SH. Drugs, DNA, and the analyst's couch. In Brockman J, ed. The next fifty years. New York: Vintage, 2002. Barondes SH. The double helix at fifty: From the gene to the brain. Cerebrum 2002;4(4):17-26. Barondes SH. Better than Prozac: Creating the next generation of psychiatric drugs. New York: Oxford University Press, 2003. Barondes SH, Alberts BM, Andreasen NC, Bargmann C, Benes F, Goldman-Rakic P, Gottesman I, Heinemann SF, Jones EG, Kirschner M, Lewis D, Raft M, Roses A, Rubenstein J, Snyder S, Watson SJ, Weinberger DR, Yolken RH. Workshop on schizophrenia. Proc Natl Acad Sci U S A 1997;94:1612-1614. Barondes SH, Castronovo V, Cooper DNW, Cummings RD, Drickamer K, Feizi T, Gitt MA, Hirabayashi J, Hughes C, Ken-ichi K, Leffler H, Liu FT, Lotan R, Mercurio AM, Monsigny M, Pillai S, Poirer F, Raz A, Rigby PWJ, Rini JM, Wang JL. Galectins: A family of animal fl-galactoside-binding lectins. Cell 1994;76:597-598. Barondes SH, Cohen HD. Puromycin effect on successive phases of memory storage. Science 1966;151:595-597. Barondes SH, Cohen HD. Comparative effects of cycloheximide and puromycin on cerebral protein synthesis and consolidation of memory in mice. Brain Res 1967;4:44-51. Barondes SH, Cohen H. Delayed and sustained effect of acetoxycycloheximide on memory in mice. Proc Natl Acad Sci U S A 1967;58:157-164. Barondes SH, Cohen HD. Arousal and conversion of short term to long term memory. Proc Natl Acad Sci U S A 1968;61:923-929. Barondes SH, Cohen HD. Memory impairment after subcutaneous injection of acetoxycycloheximide. Science 1968;160:556-557. Barondes SH, Cooper DNW, Gitt MA, Leffler H. Galectins: Structure and function of a large family of animal lectins. J Biol Chem 1994;269:20807-20810.

S a m u e l H. B a r o n d e s

37

Barondes SH, Cooper DN, Haywood-Reid PL. Discoidin I and discoidin II are localized differently in developing Dictyostelium discoideum. J Cell Biol 1983;96:291-296. Barondes SH, Dingman CW, Sporn MB. In vitro stimulation of amino acid incorporation into protein by liver nuclear RNA. Nature 1962;136:145-147. Barondes SH, Haywood-Reid PL. Externalization of an endogenous chicken muscle lectin with in vivo development. J Cell Biol 1981;91:568-572. Barondes SH, Haywood-Reid PL, Cooper DN-~. Discoidin I, an endogenous lectin, is externalized from Dictyostelium discoideum in multilamellar bodies. J Cell Biol 1985;100:1825-1833. Barondes SH, Jarvik ME. The influence of actinomycin-D on brain RNA synthesis and on memory. J Neurochem 1964;11:187-195. Barondes SH, Johnson P, Field J. Stimulation of anterior pituitary and cerebral glucose oxidation by neurohumoral agents. Endocrinology 1961;69:808-818. Barondes SH, Nirenberg MW. Fate of a synthetic polynucleotide directing cell-free protein synthesis. I. Characteristics of degradation. Science 1962;138:810-813. Barondes SH, Nirenberg MW Fate of a synthetic polynucleotide directing cell free protein synthesis. II. Association with ribosomes. Science 1962;138:813-817. Beyer EC, Barondes SH. Quantitation of two endogenous lactose-inhibitable lectins in embryonic and adult chicken tissue. J Cell Biol 1982;92:23-27. Beyer EC, Barondes SH. Secretion of endogenous lectin by chicken intestinal goblet cells. J Cell Biol 1982;92:28-33. Beyer EC, Tokuyasu K, Barondes SH. Localization of an endogenous lectin in chicken liver, intestine and pancreas. J Cell Biol 1979;82:565-571. Beyer EC, Zweig SE, Barondes SH. Two lactose binding lectins from chicken: Purified lectin from intestine is different from those in liver and muscle. J Biol Chem 1980;255:4236-4239. Bols NC, Roberson MM, Haywood-Reid PL, Cerra RE Barondes SH. Secretion of a cytoplasmic lectin from Xenopus laevis skin. J Cell Biol 1986;102:492-499. Ceri H, Kobiler D, Barondes SH. Heparin-inhibitable lectin: Purification from chicken liver and embryonic chicken muscle. J Biol Chem 1981;256:390-394. Cerra RE Gitt MA, Barondes SH. Three soluble rat B-galactoside-binding lectins. J Biol Chem 1985;260:10474-10477. Cerra RE Haywood-Reid PL, Barondes SH. Endogenous mammalian lectin localized extracellularly in lung elastic fibers. J Cell Biol 1984;98:1580-1589. Chang CM, Reitherman RW, Rosen SD, Barondes SH. Cell surface location of discoidin, a developmentally regulated carbohydrate-binding protein from Dictyostelium discoideum. Exp Cell Res 1975;95:136-142. Cleves AE, Cooper DNW, Barondes SH, Kelly RB. A new pathway for protein export in Saccharomyces cerevisiae. J Cell Biol 1996;133:1017-1026. Cohen HD, Barondes SH. Further studies on learning and memory after intracerebral actinomycin-D. J Neurochem 1966;13:207-211. Cohen HD, Barondes SH. Puromycin effect on memory may be due to occult seizures. Science 1967;157:333-334.

38

S a m u e l H. B a r o n d e s

Cohen HD, Barondes SH. Acetoxycycloheximide effect on learning and memory of a light-dark discrimination. Nature 1968;218:271-273. Cohen HD, Barondes SH. Cycloheximide impairs memory of an appetitive task. Comm Behav Biol 1968;1:337-340. Cohen HD, Ervin F, Barondes SH. Puromycin and cycloheximide: Different effects of hippocampal electrical activity. Science 1966;154:1557-1558. Cooper DN, Barondes SH. Isolectins from Dictyostelium purpureum: Purification and characterization of seven functionally distinct forms. J Biol Chem 1981; 256:5046-5051. Cooper DNW, Barondes SH. Co-localization of discoidin-binding ligands with discoidin in developing' Dictyostelium discoideum. Dev Biol 1984;105:59-70. Cooper DNW, Barondes SH. Evidence for export of a muscle lectin from cytosol to extracellular matrix and for a novel secretary mechanism. J Cell Biol 1990;110:1681-1691. Cooper DNW, Barondes SH. God must love galectins because He's made so many of them. Glycobiology 1999;9:979-984. Cooper DNW, Haywood-Reid PL, Springer WR, Barondes SH. Bacterial glycoconjugates are natural ligands for the carbohydrate binding site of discoidin I and influence its cellular compartmentalization. Dev Biol 1986;114:416-425. Cooper DN, Lee S-C, Barondes SH. Discoidin-binding polysaccharide from Dictyostelium discoideum. J Biol Chem 1983;258:8745-8750. Cooper DNW, Massa SM, Barondes SH. Endogenous muscle lectin inhibits myoblast adhesion to laminin. J Cell Biol 1991;115:1437-1448. DeVries G, Barondes SH. Incorporation of (14C) N-acetyl neuraminic acid into brain glycoproteins and gangliosides in vivo. J Neurochem 1971;18:101-105. Droz B, Barondes SH. Nerve endings: Rapid appearance of labelled protein shown by electron microscope radioautography. Science 1969;165:1131-1133. Dutton G, Barondes SH. Microtubular protein: Synthesis and metabolism in developing brain. Science 1970;1969;1637-1638. Dutton G, Barondes SH. Glycoprotein metabolism in developing mouse brain. J Neurochem 1970;17, 913-920. Feinberg A, Springer WR, Barondes SH. Segregation of pre-stalk and pre-spore cells of Dictyostelium discoideum: Observations consistent with selective cell cohesion. Proc Natl Acad Sci U S A 1979;76:3977-3981. Feit H, Dutton G, Barondes SH, Shelanski M. Microtubule protein: Identification in and transport to nerve endings. J Cell Biol 1971;51:138-147. Flanagan SD, Barondes SH. Affinity partitioning: A method for purification of proteins using specific polymer-ligands in aqueous polymer two-phase systems. J Biol Chem 1975;250:1484-1489. Flanagan SD, Barondes SH, Taylor P. Affinity partitioning of membranes: Cholinergic receptor-containing membranes from Torpedo californica. J Biol Chem 1976;251:858-865. Flanagan SD, Taylor P, Barondes SH. Affinity partitioning of acetylcholine receptor enriched membranes and their purification. Nature 1975;254:441-443.

S a m u e l H. B a r o n d e s

39

Frazier WA, Rosen SD, Reitherman RW, Barondes SH. Purification and comparison of two developmentally regulated lectins from Dictyostelium discoideum: Discoidin I and II. J Biol Chem 1975;250:7714-7721. Freimer NB, Reus VI, Escamilla MA, McIness LA, Spesny M, Leon P, Service SK, Smith LB, Silva S, Rojas E, Gallegos A, Meza L, Fournier E, Baharloo S, Blankenship K, Tyler DJ, Batki S, Vinogradov S, Weissenbach J, Barondes SH, Sandkuijl LA. Genetic mapping using haplotype, association and linkage methods suggests a locus for severe bipolar disorder (BPI) at 18q22-q23. Nat Genet 1996;12:436-441. Gabius H-J, Springer WR, Barondes SH. Receptor for the cell binding site of discoidin I. Cell 1985;42:449-456. Geller A, Robustelli F, Barondes SH, Cohen H, Jarvik M. Impaired performance by post-trial injections of cycloheximide in a passive-avoidance task. Psychopharmacologia 1969; 14:371-376. Gitt MA, Barondes SH. Evidence that a human soluble B-galactoside-binding lectin is encoded by a family of genes. Proc Natl Acad Sci U S A 1986;83: 7603-7607. Gitt MA, Barondes SH. Genomic sequence and organization of two members of a human lectin gene family. Biochemistry 1991;30:82-89. Gitt MA, Colnot C, Poirier F, Nani KJ, Barondes SH, Leffler H. Galectin-4 and galectin-6 are two closely related lectins expressed in mouse gastrointestinal tract. J Biol Chem 1998;273:2954-2960. Gitt MA, Massa SM, Leffler H, Barondes SH. Isolation and expression of a gene encoding L-14-II, a new human soluble lactose-binding lectin. J Biol Chem 1992 ;267:10601-10606. Gitt MA, Wiser ME Leffler H, Herrmann J, Xia Y, Massa SM, Cooper DNW, Lusis AJ, Barondes SH. Galectin-5: Sequence and mapping of galectin-5, a beta-galactoside-binding lectin found in rat erythrocytes. J Biol Chem 1995; 270:5032-5038. Gitt MA, Xia Y, Atchison RE, Lusis AJ, Barondes SH, Leffler H. Sequence, structure, and chromosomal mapping of the mouse Lgals6 gene, encoding galectin-6. J Biol Chem 1998;273:2961-2970. Gremo F, Kobiler D, Barondes SH. Distribution of an endogenous lectin in the developing chick optic tectum. J Cell Biol 1978;79:491-499. Herrmann J, Turck CW, Atchison R, Huflejt M, Poulter L, Gitt MA, Burlingame AL, Barondes SH, Leffler H. Primary structure of the soluble lactose binding lectin L-29 from rat and dog and interaction of its non-collagenous Pro, Gly, Try-rich sequence with bacterial and tissue collagenase. J Biol Chem 1993;268: 26704-26711. Hinek A, Wrenn DS, Mecham RP, Barondes SH. The elastin receptor: A galactosidebinding protein. Science 1988;239:1539-1541. Huflejt ME, Jordan ET, Gitt MA, Barondes SH, Leffler H. Strikingly different localization of galectin-3 and galectin-4 in human colon adenocarcinoma T84 cells. J Biol Chem 1997;272:14294-14303.

40

S a m u e l H. Barondes

Huflejt ME, Turck CW, Lindstedt R, Barondes SH, Leffler H. L-29, a soluble lactose binding lectin, is phosphorylated on serine-6 and serine-12 in vivo and by casein kinase I. J Biol Chem 1993;268:26712-26718. Hynes MA, Buck LB, Gitt M, Barondes S, Dodd J, Jessell TM. Carbohydrate recognition in neuronal development: Structure and expression of surface oligosaccharides and B-galactoside-binding lectins. In Carbohydrate recognition in cellular function, CIBA Foundation Symposium 145. Chichester, UK: John Wiley & Sons, 1989;189-218. Hynes MA, Gitt MA, Barondes SH, Jessell TM, Buck LB. Selective Expression of an endogenous lactose-binding lectin gene in subsets of central and peripheral neurons. J Neurosci 1990;10:1004-1013. Kobiler D, Barondes SH Lectin Activity from embryonic chick brain, heart and liver: Changes with development. Dev Biol 1977;60:326-330. Kobiler D, Beyer EC, Barondes SH. Developmentally regulated lectins from chick muscle, brain, and liver have similar chemical and immunological properties. Dev Biol 1978;64:265-272. Leffler H, Barondes SH. Specificity of binding of three soluble rat lung lectins to substituted and unsubstituted mammalian B-galactosides. J Biol Chem 1986;261:10119-10126. Leffler H, Masiarz FR, Barondes SH. Soluble lactose-binding vertebrate lectins: A growing family. Biochemistry 1989;28:9222-9229. Levitan IB, Barondes SH. Octopamine- and serotonin-stimulated phosphorylation of specific protein in the abdominal ganglion of Aplysia california. Proc Natl Acad Sci U S A 1974;71:1145-1148. Levitan IB, Madsen CJ, Barondes SH. Cyclic AMP and amine effects on phosphorylation of specific protein in abdominal ganglion of Aplysia california: Localization and kinetic analysis. J Neurobiol 1974;5:511-525. Lindstedt R, Apodaca G, Barondes SH, Mostov KE, Leffler H. Apical secretion of a cytosolic protein by Madin-Darby canine kidney cells. J Biol Chem 1993;268:11750-11757. Lipsick JS, Beyer EC, Barondes SH, Kaplan NO. Lectins from chicken tissues are mitogenic for Thy-1 negative murine spleen cells. Biochem Biophys Res Comm 1980;97:56-61. Lobsanov YD, Gitt MA, Leffler H, Barondes SH, Rini JM. Crystallization and preliminary x-ray diffraction analysis of the human dimeric S-Lac lectin (L-14-II). J Mol Biol 1993;233:553-555. Lobsanov YD, Gitt MA, Leffler H, Barondes SH, Rini JM. X-ray crystal structure of the human dimeric S-Lac lectin, L-14-II, in complex with lactose at 2.9 .~kresolution. J Biol Chem 1993;268:27034-27038. Loomis WF, Wheeler SA, Springer WR, Barondes SH. Adhesion mutants of Dictyostelium discoideum lacking the saccharide determinant recognized by two adhesion-blocking monoclonal antibodies. Dev Biol 1985;109:111-117. Mahanthappa NK, Cooper DNW, Barondes SH, Schwarting GA. Rat olfactory neurons can utilize the endogenous lectin, L-14, in a novel adhesion mechanism. Development 1994;120:1373-1384.

S a m u e l H. B a r o n d e s

41

Marschal P, Cannon V, Barondes SH, Cooper DNW Xenopus laevis L-14 lectin is expressed in a typical pattern in the adult, but is absent from embryonic tissues. Glycobiology 1994;4:297-305. Marschal P, Herrmann J, Leffler H, Barondes SH, Cooper DN~. Sequence and specificity of a soluble lactose-binding lectin from Xenopus laevis skin. J Biol Chem 1992;267:12942-12949. Massa SM, Cooper DNW, Leffler H, Barondes SH. L-29, an endogenous lectin, binds to glycoconjugate ligands with positive cooperativity. Biochemistry 1993;32: 260-267. McDonough JP, Springer WR, Barondes SH. Species-specific cell cohesion in cellular slime molds: Demonstration by several quantitative assays and with multiple species. Exp Cell Res 1980;125:1-14. McInnes AL, Escamilla MA, Service SK, Reus VI, Leon P, Silva S, Rojas E, Spesny M, Baharloo S, Blankenship K, Peterson A, Tyler D, Shimayohsi N, Tobey C, Batki S, Vinogradov S, Meza L, Gallegos A, Fournier E, Smith LB, Barondes SH, Sandkuijl LA, Freimer NB. A complete genome screen for genes predisposing to severe bipolar disorder in two Costa Rican pedigrees. Proc Natl Acad Sci U S A 1996;93:13060-13065. Mehrabian M, Gitt MA, Sparkes RS, Leffler H, Barondes SH, Lusis AJ. Two members of the S-Lac lectin gene family, LGALS1 and LGALS2, reside in close proximity on human chromosome 22q12-q13. Genomics 1993;15: 418-420. Mir-Lechaire F, Barondes SH. Two distinct developmentally regulated lectins in chick embryo muscle. Nature 1978;272:256-258. Newlin SA, Schlapfer WT, Barondes SH. Heterosynaptic stimulation modulates the duration of post-tetanic potentiation at an Aplysia synapse without affecting other aspects of synaptic transmission. Brain Res 1980;181:107-125. Newlin SA, Schlapfer WT, Barondes SH. Separate serotonin and dopamine receptors modulate the duration of post-tetanic potentiation at an Aplysia synapse without affecting other aspects of synaptic transmission. Brain Res 1980;181: 89-106. Nirenberg MW, Matthaei JH, Jones OW, Martin RG, Barondes SH. Approximation of genetic code via cell-free protein synthesis directed by template RNA. Fed Proc 1963;22:55-61. Nowak TP, Haywood PL, Barondes SH. Developmentally regulated lectin in embryonic chick muscle and a myogenic cell line. Biochem Biophys Res Comm 1976;68:650-657. Nowak TP, Kobiler D, Roel L, Barondes SH. Developmentally regulated lectin from embryonic chick pectoral muscle: Purification by affinity chromatography. J Biol Chem 1977;252:6026-6030. Oda Y, Herrmann J, Gitt MA, Turck CW, Burlingame AL, Barondes SH, Leffler H. Soluble lactose-binding lectin from rat intestine with two different carbohydrate-binding domains in the same peptide chain. J Biol Chem 1993 ;268: 5929-5939.

42

S a m u e l H. B a r o n d e s

Oda Y, Leffler H, Sakakura Y, Kasai K-I, Barondes SH. Human breast carcinoma cDNA sequence encoding a galactoside-binding lectin homologous to mouse Mac-2 antigen. Gene 1991;99:279-283. Outenreath RL, Roberson MM, Barondes SH. Endogenous lectin secretion into the extracellular matrix of early embryos of Xenopus laevis. Dev Biol 1988;125: 187-194. Regan LJ, Dodd J, Barondes SH, Jessell TM. Selective expression of endogenous lactose-binding lectins and lactoseries glycoconjugates in subsets of rat sensory neurons. Proc Natl Acad Sci U S A 1986;83:2248-2252. Reitherman RW, Rosen SD, Barondes SH. Lectin purification using formalinized erythrocytes as a general affinity adsorbant. Nature 1974;248:599-600. Reitherman RW, Rosen SD, Frazier WA, Barondes SH. Cell surface species-specific high affinity receptors for discoidin: Developmental regulation in Dictyostelium discoideum. Proc Natl Acad Sci U S A 1975;72:3541-3545. Roberson MM, Barondes SH. Lectin from embryos and oocytes of Xenopus laevis: Purification and properties. J Biol Chem 1982;257:7520-7524. Roberson MM, Barondes SH. Xenopus laevis lectin is localized at several sites in Xenopus oocytes, eggs and embryos. J Cell Biol 1983;97:1875-1881. Roberson MM, Wolffe AP, Tata JR, Barondes SH. Galactoside-binding serum lectin of Xenopus laevis: Estrogen-dependent hepatocyte synthesis and relationship to oocyte lectin. J Biol Chem 1985;260:11027-11032. Rosen SD, Chang CM, Barondes SH. Intercellular adhesion in the cellular slime mold P. pallidum inhibited by interaction of asialofetuin or specific univalent antibody with endogenous cell surface lectin. Dev Biol 1977;61:202-213. Rosen SD, Haywood PL, Barondes SH. Inhibition of intercellular adhesion in a cellular slime mold by univalent antibody against a cell surface lectin. Nature 1976;263:425-427. Rosen SD, Kafka JA, Simpson DL, Barondes SH. Developmentally-regulated, carbohydrate-binding protein in Dictyostelium discoideum. Proc Natl Acad Sci U S A 1973;70:2554-2557. Rosen SD, Simpson DL, Rose JE, Barondes SH. Carbohydrate-binding protein from Polysphondylium pallidum implicated in intercellular adhesion. Nature 1974;252:128, 149-151. Schlapfer WR, Tremblay JP, Woodson PBJ, Barondes SH. Frequency facilitation and post-tetanic potentiation of a unitary synaptic potential in Aplysia california are limited by different processes. Brain Res 976;109:1-20. Schlapfer WT, Woodson PBJ, Smith GA, Tremblay JP, Barondes SH. Marked prolongation of post-tetanic potentiation at a transition temperature, and its adaptation. Nature 1975;258:623-625. Schlapfer WT, Woodson PBJ, Tremblay JP, Barondes SH. Depression and frequency facilitation at a synapse in Aplysia california: Evidence for regulation by availability of transmitter. Brain Res 1974;76:267-280. Seetharaman J, Kanigsberg A, Slaaby R, Leffler H, Barondes SH, Rini JM. X-Ray crystal structure of the human galectin-3 carbohydrate recognition domain at 2.1-A resolution. J Biol Chem 1998;273:13047-13052.

S a m u e l H. B a r o n d e s

43

Segal DS, Squire LR, Barondes SH. Cycloheximide: Its effects on activity are dissociable from its effects on memory. Science 1971;172:82-84. Shadle PJ, Barondes SH. Adhesion of human platelets to immobilized type I trimeric collagen. J Cell Biol 1982;95:361-365. Shadle PJ, Barondes SH. Platelet-collagen adhesion: Evidence for participation of antigenically distinct entities. J Cell Biol 1984;99:2048-2055. Shadle PJ, Ginsberg MH, Plow EF, Barondes SH. Platelet-collagen adhesion: Inhibition by a monoclonal antibody that binds glycoprotein IIb. J Cell Biol 1984;99:2056-2060. Simpson DL, Rosen SD, Barondes SH. Discoidin, a developmentally regulated carbohydrate-binding protein from Dictyostelium discoideum: Purification and characterization. Biochemistry 1974;13:3487-3493. Sparrow CP, Leffler H, Barondes SH. Multiple soluble B-galactoside-binding lectins from human lung. J Biol Chem 1987;262:7383-7390. Springer WR, Barondes SH. Direct measurement of species-specific cohesion in cellular slime molds. J Cell Biol 1978;78:937-942. Springer WR, Barondes SH. Cell adhesion molecules: Detection with univalent second antibody. J Cell Biol 1980;87:703-707. Springer WR, Barondes SH. Evidence for another cell adhesion molecule in Dictyostelium discoideum. Proc Natl Acad Sci U S A 1982;79:6561-6565. Springer WR, Barondes SH. Externalization of the endogenous intracellular lectin of a cellular slime mold. Exp Cell Res 1982;138:213-240. Springer WR, Barondes SH. Monoclonal antibodies block cell-cell adhesion in Dictyostelium discoideum. J Biol Chem 1983;258:4698-4701. Springer WR, Barondes SH. Protein-linked oligosaccharide implicated in cell-cell adhesion in two Dictyostelium species. Dev Biol 1985;109:102-110. Springer WR, Cooper DNW, Barondes SH. Discoidin I is implicated in cellsubstratum attachment and ordered cell migration of Dictyostelium discoideum and resembles fibronectin. Cell 1984;39:557-564. Springer WR, Haywood PL, Barondes SH. Endogenous cell surface lectin in Dictyostelium: Quantitation, elution by sugar, and elicitation by divalent immunoglobulin. J Cell Biol 1980;87:682-690. Squire LR, Barondes SH. Actinomycin-D: Effect on memory at different times after training. Nature 1970;225:649-650. Squire LR, Barondes SH. Inhibitors of cerebral protein or RNA synthesis and memory. In Gaito J, ed. Macromolecules and behavior, 2nd ed. AppletonCentury-Crofts, 1972;61-82. Squire LR, Barondes SH. Variable decay of memory and its recovery in cycloheximide treated mice. Proc Natl Acad Sci 1972;69:1416-1421. Squire LR, Barondes SH. Memory impairment during prolonged training in mice given inhibitors of cerebral protein synthesis. Brain Res 1973;56:215-225. Squire LR, Barondes SH. Anisomycin, like other inhibitors of cerebral protein synthesis, impairs long-term memory of a discrimination task. Brain Res 1974;66:301-308.

44

S a m u e l H. B a r o n d e s

Squire LR, Barondes SH. Amnesic effect of cycloheximide not due to depletion of a constitutive brain protein with short half-life. Brain Res 1976;103:184-189. Squire LR, Kuczenski R, Barondes SH. Tyrosine hydroxylase inhibition by cycloheximide and anisomycin is not responsible for their amnesic effects. Brain Res 1974;82:241-248. Squire LR, Smith GA, Barondes SH. Cycloheximide can affect memory within minutes after the onset of training. Nature 1973;242:201-202. Tecott LH, Barondes SH. Genes and aggressiveness. Curr Biol 1996;6:238-240. Traynor AE, Schlapfer WT, Barondes SH. Stimulation is necessary for development of tolerance to a neuronal effect of ethanol. J Neurobiol 1980;11:633-638. Traynor ME, Woodson PBJ, Schlapfer WT, Barondes SH. Sustained tolerance to a specific effect of ethanol on post-tetanic potentiation in Aplysia california. Science 1976;193:510-511. Tremblay JP, Schlapfer WR, Woodson PBJ, Barondes SH. Morphine and related compounds: Evidence that they decrease available neurotransmitter in Aplysia californica. Brain Res 1974;8:107-118. Tremblay JP, Woodson PBJ, Schlapfer WT, Barondes SH. Dopamine, serotonin and related compounds: Presynaptic effects on synaptic depression, frequency facilitation and post-tetanic potentiation at a synapse in Aplysia california. Brain Res 1976;109:61-81. Woodson PBJ, Schlapfer WT, Barondes SH. Amplitude and rate of decay of post-tetanic potentiation are controlled by different mechanisms. Brain Res 1978;157:33-46. Woodson PBJ, Schlapfer WR, Tremblay JP, Barondes SH. Cholinergic agents affect two receptors that modulate transmitter release at a central synapse in Aplysia california. Brain Res 1975;88:455-474. Woodson PBJ, Schlapfer WR, Tremblay JP, Barondes SH. Resting and stimulated values of model parameters governing transmitter release at a synapse in Aplysia california. Brain Res 1976;109:21-40. Woodson PBJ, Schlapfer WT, Tremblay JP, Barondes SH. Synaptic depression at a synapse in Aplysia california: Analysis in terms of a material flow model of neurotransmitter. Brain Res 1976;109:41-59. Woodson PBJ, Traynor ME, Schlapfer WT, Barondes SH. Increased membrane fluidity implicated in acceleration of decay of post-tetanic potentiation by alcohols. Nature 1976;260:797-799. Woodson PBJ, Tremblay JP, Schlapfer WT, Barondes SH. Heterosynaptic inhibition modifies the presynaptic plasticities of the transmission process at a synapse in Aplysia california. Brain Res 1976;109:83-95. Zatz M, Barondes SH. Fucose incorporation into glycoproteins of mouse brain. J Neurochem 1970;17:157-163. Zatz M, Barondes SH. Particulate and solubilized fucosyl transferases from mouse brain. J Neurochem 1971;18:1625-1637. Zatz M, Barondes SH. Rapid transport of fucosyl glycoproteins to nerve endings in mouse brain. J Neurochem 1971;18:1125-1133.

S a m u e l H. B a r o n d e s

45

Additional Publications
Agranoff BW, Davis RE, Brink JJ. Memory fixation in the goldfish. Proc Natl Acad Sci U S A 1965;54:788-793. Dingman W, Sporn MB. The incorporation of 8-azaguanine into rat brain RNA and its effect on maze-learning by the rat: An inquiry into the biochemical basis of memory. J Psychiatr Res 1961;1:1-11. Flexner JB, Flexner LB, Stellar E. Memory in mice as affected by intracerebral puromycin. Science 1963;141:57-59. Leffler H, ed. Special issue on galectins. Glycoconjugate J 2004;19:433-629. Teichberg VI, Silman I, Beitsch DD, Resheff G. A beta-D-galactoside binding protein from electric organ tissue of Electrophorus electricus. Proc Natl Acad Sci U S A 1975;72:1383-1387. Vogel W. Discoidin domain receptors: Structural relations and functional implications. FASEB J 1999;13(Suppl):S77-82.

THE EPILEPSIES 3

Seizure Prediction: Its Evolution and Therapeutic Potential


DAVID KRIEGER  BRIAN LITT

Introduction Historical Background Gathering Momentum High-Frequency Oscillations, Broad-Band EEG and Seizure Generation

Therapeutics, Devices, and Seizure Prediction Seizure Prediction Devices and Their Therapeutic Potential Future Directions

Introduction
Seizure prediction has a long history, starting in the 1970s1 with very small data sets looking only at preseizure (preictal) events minutes to seconds before seizures. It has progressed over the past almost 40 years up to current methods, which use mathematical to analyze continuous days of multiscale intracranial electroencephalogram (IEEG) recordings.2 Seizure prediction research, most important, has given hope for new warning and therapeutic devices to the 25% of epilepsy patients who cannot be successfully treated with drugs or surgery.3 One of the most insidious aspects of seizures is their unpredictability. In this light, in the absence of completely controlling a patients epilepsy, seizure prediction is an important aim of clinical management and treatment. From a broader view, seizure prediction research has also transformed the way we understand epilepsy and the basic mechanisms underlying seizure generation. Seizures were once viewed as isolated and abrupt events, but we now view them as processes that develop over time and space in epileptic networks. Thus, what started as a goal of predicting seizures for clinical applications has expanded into a field dedicated to understanding seizure generation. The study of seizure generation necessarily encompasses a large collaborative effort between mathematicians, engineers, physicists, clinicians, and neuroscientists. However, it also requires large volumes of clinical data, which has led to more specific collaborations between epilepsy centers. These partnerships have come about through The International Seizure Prediction Group (ISPG), which held its

THE EPILEPSIES 3

Third Collaborative Workshop on Seizure Prediction in Freiburg, Germany, in October 2007. This workshop, and its two predecessors, allowed various groups to share computational methods, data, and ideas, and to focus on basic research and its translation to clinical relevance. There is a large gulf between understanding how seizures are generated and the eventual goal of preventing seizure occurrence. Although studies in the literature have focused on prospectively testing seizure prediction methods,4,5 no study to date has yet confirmed the ability of a method to predict seizures better than random, or with accuracy sufficient for prospective clinical trials or eventual implementation in patients. Much of the blame for this performance failure resides in two important challenges, both of which have recently been solved. First, until recently, there was no consensus on the amount and quality of data required to conduct appropriate prospective prediction studies. Second, the statistical methods for designing experiments, and the metrics by which to judge successful algorithm performance, were not in place.2 Recent research has definitively advanced progress in these areas.68 Looking further ahead, for successful prediction devices to emerge, many technical questions will need to be resolved to design systems that not only warns the patient of a seizure but also intervenes to preempt it. For example, the intervention strategy (drug versus stimulation or other method), the clinical interface (sensors, classifiers, etc.), and the number and site of electrode placement are just a few of the problems under investigation that will need definitive solutions. With the advent of new brain sensors, stimulation technologies, and the availability of large data sets of continuous EEG recordings for collaborative research, our progress toward understanding seizure generation and preventing its occurrence is accelerating.

Historical Background
Epileptologists have long been aware that many patients with epilepsy know that their seizures are not abrupt in onset, and that they can often predict periods of time when seizures are more likely to occur. Many clinical findings support the idea that seizures are predictable. An increase in blood flow in the epileptic temporal lobe has been seen as much as 12 minutes before the onset of seizures.9,10 Clinical prodromes are noted in more than 50% of patients, according to Rajna et al.,11 with more refined measurements made recently by Haut et al.12,13 An increase in oxygen availability and blood oxygen level dependent signals on magnetic resonance imaging (MRI) have also been noted before seizures.14,15 Preictal changes in heart rate have been reported in several studies.1618 However, these clinical findings do not reveal how long before a seizure the first changes in seizure generation occur or by what method a seizure might be predicted. Seizure prediction research has evolved via diverse mathematical and engineering approaches, from its starting point in the 1970s. Viglione and Walsh not only implemented the first electronic classifier to solve this problem, in the form of an analog neural network, but they also created an actual device and tested it on patients with epilepsy19 (Figure 1-1). This device used scalp EEG electrodes to record signals for classification.1,20 Linear approaches looking at absence seizures using surface electrodes detected preictal changes up to 6 seconds before seizure onset.21 Another early study found differences between 1-minute preictal EEG epochs and

1  Seizure Prediction: Its Evolution and Therapeutic Potential

BIOTELEMETRY

Figure 11

A patient using one of the earliest seizure warning devices invented by Viglione et al.19 The biotelemetry unit was developed by Biocom, Inc., Culver City, CA.

control epochs.22 Using another approach, many studies have evaluated the rate of spikes in EEG before seizures with some23 finding predictive differences, but with the majority finding no predictive value.24,25 Novel mathematical approaches started with nonlinear systems using the largest Lyapunov exponent. They detected decreases in chaotic behavior in the minutes before temporal lobe epileptic seizures,26,27,28,29 with subsequent studies claiming prediction over hours5,30 (Figure 1-2). These and other early studies are important because they support the conceptual idea that seizures are not isolated events but rather develop over time. However, there were methodological problems with these early studies. First, they focused on the preictal period and did not include prolonged interictal recordings; thus there was doubt about the specificity of the findings with regard to time. Second, they also lacked specificity with regard to space because they used univariate measures and one-channel recordings. Third, most of these studies employed small numbers of patients and small selected data sets. Finally, and perhaps most important, these studies did not have specific, well-thought-out statistical criteria for success and did not demonstrate performance measured against a chance predictor. These criteria would later reveal hidden biases in experimental data sets, not apparent in initial studies. The identification of these difficulties was one of the main products of the First International Collaborative Workshop on Seizure Prediction.2

THE EPILEPSIES 3

Figure 12 Some examples of different methods for seizure prediction. A, A derivation of the principal Lyapunov exponents of two sites (represented by the blue and red lines) converge as a seizure threshold is reached (represented by the horizontal dashed line). The vertical dashed lines represent the start and end times for the seizure. B, An estimate of the correlation dimension (D*, top row) and the mean phase coherence (R) as a measure of phase synchronization (bottom row) discriminate between interictal and preictal data. C, Spatial and temporal changes in dynamical similarity in a patient with temporal lobe epilepsy. D, A cascade of neurophysiological events occurring in patterns associated with oncoming seizures (e.g., bursts of complex interictal epileptiform activity, bursts of increased signal energy or power, rhythmic seizurelike events (chirps), and energy accumulating at higher rates preseizure than during interictal periods). (Adapted from Litt B, Echauz J. Prediction of epileptic seizures. Lancet Neurol. May 2002; and Litt B, Lehnertz K. Seizure Prediction and the Preseizure Period. Curr Opin Neurol. 2002 Apr;15(2):173-7.)

As the field evolved, progressive attempts were made to conquer these challenges. The first problem was addressed in multiple studies in which a number of mathematical measures, starting with the correlation dimension,31 were shown to distinguish preictal and interictal periods using expanded data sets. The second problem was addressed with bivariate and multivariate measures, such as the largest Lyapunov exponent of two channels,3234 and other methods fusing information from multiple channels and measures over time,4,35,36 simulated neuronal cell bodies,37 and measures for phase synchronization and cross correlation.36,38,39 But none of these studies addressed the lack of large data sets, containing prolonged interictal periods and sufficient numbers of seizures for classifier training and testing. As a result, their findings came into question in later studies carried out on unselected and more extended EEG recordings. Though the correlation dimension31 reliably demonstrated preseizure changes, these changes were also seen at other times, when seizures did not occur, a problem of many early (and present) methods.

1  Seizure Prediction: Its Evolution and Therapeutic Potential

For this reason, although many methods were considered promising, they were found not to predict seizures, as verified by other investigators using alternative statistical methods.40,41 Statistical concerns also engendered criticism of studies involving prediction work invoking Lyapunov exponents.42 The predictive value of accumulated energy measures35 was also called into question by Harrison et al. and Maiwald et al.43,44 though these groups did not separate recordings into wakefulness and sleep, as was required in the original study. Therefore, the suitability of nonlinear mathematics,45 as well as linear methods in seizure prediction, was called into question. These questions spawned The First International Collaborative Workshop on Seizure Prediction, held in Bonn, Germany, in 2002. The goal of the workshop was to give a tutorial on state-of-the-art methods for predicting seizures to students and investigators, to share scientific ideas, to compare different mathematical methods on a large common data set (five continuous intracranial EEG recordings from patients undergoing presurgical evaluation for medically intractable epilepsy), and to set goals and standards for future work in the field. The meeting attendees published a collection of papers summarizing this meeting, including a summary paper outlining a consensus on future research in seizure prediction, under the name of The International Seizure Prediction Group.2 Suggestions for future research studies in the field included: (1) that some of the first thorough easiest studies to be performed, to guarantee sampling data from networks involved in generating seizures, are recordings demonstrating should examine data from very focal disorders, such as unilateral temporal lobe; (2) that seizure prediction should focus on EEG events rather than events having only overt clinical symptoms; (3) that a top priority for the field was to establish an international database of high-quality intracranial EEG recordings for collaborative research; (4) that it was key that prolonged recordings, with statistically sufficient amounts of interictal and ictal data, be analyzed; and (5) that there needed to be better development of statistical models for seizure prediction and a consensus on what constitutes success.2 These last two issues were seen as major impediments to success in the area of seizure prediction over the years. Unfortunately, there was, at that time, no consensus on what statistical performance measures should be used to present results. There was also much debate about the predictability of seizures from patients with types of epilepsy other than that originating from the temporal lobe. Results from the individual group presentations at the meeting were of interest. The conclusions of studies on this large test data set showed poor performance for univariate measures46,47 and better performance for bi- and multivariate measures.30,48,49 Nonlinear measures were not found to outperform linear ones.49 Of interest, the use of machine learning for automated feature exploration was introduced at this meeting, though this method did not yield results significantly better than other multivariate measures.4 Presentations at the meeting also included results on spatial contributions to seizure generation in the brain. The finding of preictal changes in channels remote from the seizure onset zone4,48,49 necessarily changed the conceptual framework for understanding seizure generation. Whereas earlier studies taught that seizures are not abrupt in onset, but more likely develop over time, these new studies demonstrated that even focal seizures likely develop across a network of distributed regions that are functionally connected. These findings challenged the more common accepted view of the epileptic focus.

THE EPILEPSIES 3

The conference ended on a humorous but poignant note summing up the results of different prediction methods presented. This was a quote from what would later be a paper by the Bonn group: The null hypothesis of the non-existence of a preseizure state cannot be disproved.49,50 Papers detailing the research and presentations of each group at this workshop were later published in a special edition of the journal Clinical Neurophysiology (vol. 116, no. 3, 2005).

Gathering Momentum
There was a 5-year hiatus between the first Seizure Prediction Workshop and the second, which took place in Bethesda, Maryland, in February 2006 and was sponsored by the National Institute of Neurological Disease and Stroke (NINDS). During this period, investigators in the field continued to critically examine their work and developed new insights into why prior methods had failed to produce reliable algorithms that could predict seizures better than random chance. This introspection was released in a flurry of talks and papers leading up to and after the meeting. This body of work focused primarily on statistical concerns about current methods for tracking seizure generation over time,68,34,41,49,51 but also on clinical reports from patients to bolster seizure prediction work, reinforcing the concept that seizures are generated over time.12,52 The second workshop provided an explosive outlet for these ideas, with harsh criticism of much of the early work in this field. Rather than interpreting this as a failure of the field of seizure prediction to make progress, most investigators came away from the meeting with a renewed sense of focus and a feeling that tremendous progress had been made. One of these leaps forward was the acceptance of seizure generation as likely being a stochastic process. In this scheme there are periods of increased probability of seizure onset, which may or may not lead to seizures, depending on unknown internal and external factors. This focused investigators interested in seizure generation on forecasting probabilities, rather than specific events. At the Second International Workshop two other points of interest were made, emphasized in a lecture by Leonard Smith, PhD, from the London School of Economics. These were (1) that identifying seizure-free periods may be as or more useful to patients than seizure warning, depending on the relative accuracy of forecasting algorithms and the frequency of seizure events; and (2) that statistical validation of a particular forecasting model might be better judged by its effect on clinical outcome, that is the overall effect on the patient and quality of life, rather than on a rigorous statistical model and performance of a seizure prediction paradigm. Although there was no consensus on this issue at the meeting: how much to weigh clinical impact versus statistical performance, there was general agreement that both patient impact and statistical performance should be used to assess seizure prediction methods. During the time leading up to the Third Collaborative Workshop on Seizure Prediction, held in Freiburg, Germany, at the end of September 2007, it became clear that the priority items for progress in the field, identified at the first workshop in 2001, had been aggressively pursued. There was now a consensus on statistical methods required to prove success at seizure prediction (see).6,7 Standards for data to be used in prediction experiments were also put forward, and a model of seizure generation as a stochastic process in distributed neuronal networks was growing in

1  Seizure Prediction: Its Evolution and Therapeutic Potential

popularity. At the same time, there was increasing interest in the role of neurophysiologic markers for epileptic brain and in the spatial and temporal evolution leading up to seizures. The idea behind this work is that there are particular neurophysiological markers of an epileptic brain that appear interictally and that may evolve in their temporal and spatial distribution as the probability of seizure increases.35 In particular, attention began to focus on high-frequency oscillations (HFOs) during this period.

High-Frequency Oscillations, Broad-Band EEG, and Seizure Generation


The idea of what constitutes a pathological HFO in human electrophysiology is a concept that is evolving over time. In clinical EEG studies from patients implanted with intracranial macroelectrodes, the frequency spectrum of these oscillations is limited by signal filtering on clinical systems. This is due to hard-wired antialiasing filters set at 100 Hz in many systems, despite sample rates that can go much higher than 250 Hz.53,54 Other studies using custom recording systems with macro- and microelectrodes in humans and animal models of epilepsy have observed much higher frequency oscillations in the 100 to 200 Hz range. Investigators have called these oscillations ripples, which are felt to be seen in both normal and abnormal processes. Presumably pathological oscillations, called fast ripples (FR), were also identified at 500 Hz or higher5557 (Figure 1-3). Some investigators are actively involved in mapping oscillations in and below the ripple frequencies in normal cognitive processes, such as spatial navigation, memory storage, and recall.5862 The potential role of ripples in network dysfunction in epilepsy is not yet well defined, and there appears to be considerable overlap in the distribution of these two populations of events (ripples and fast ripples), without a clear frequency cutoff to distinguish them.63 Various studies show an increase in HFO and broad-band fast activity preictally and at seizure onset.53,6466 Very high frequency (VHF) oscillations or fast ripples, between 250 and 500 Hz, are postulated to be pathological because they are seen primarily in epileptogenic areas. They are proposed to represent hypersynchronous interconnected neurons capable of generating seizures.67 VHF oscillations appear to be present in patients with mesial temporal lobe epilepsy before the onset of seizures.68 In temporal lobe and neocortical epilepsy, HFO activity is postulated to play an important role in seizure generation.53,63,69,70 The use of HFO activity to predict seizures is an area of important future research.71 Automated methods for detecting HFO activity in localizing epileptic networks is an active area of investigation.67,72,73 It is important to note that there is extensive literature exploring these oscillations in animal models of epilepsy spanning back into the 1960s and earlier, up to the present.7478 A discussion of this rich literature is beyond the scope of this brief review. It is important, however, to note that interest in HFOs and epilepsy extends well beyond the temporal lobe to the neocortex and deeper structures, such as the thalamus. The role of the thalamus in generating these oscillations and their relationship to epilepsy is a particular area of intense interest.79,80 When pathological conditions occur, such as in absence epilepsy, these circuits can be co-opted to create hypersynchronous activity.81 However, the exact relationship of this research to spontaneous partial seizures, and our ability to predict these events, is still undetermined.

THE EPILEPSIES 3

D
Figure 13

AD, The spectrum of frequencies identified in human brain, as transduced by different recording techniques. Antialiasing filters limit normal recordings to below 100 Hz, but custom electrodes can identify ripples and pathologic fast ripples. EI, Intracranial micro- and macroelectrodes used in IEEG. E, Standard intracranial subdural grid and strip electrodes. F, Hybrid subdural grid electrodes with 4x4 group of microwire electrodes in the center of four typical subdural disks. G, Drawing of a Utah microelectrode array for recording single-unit activity (note, the size of this entire array is the same as the size of one of the single circular disk electrodes in the picture above it). H, Four typical intracranial depth electrodes, each with six contacts (white arrow) to be embedded in brain. I, Drawing of a single Banke-Freid microelectrode for recording single units and multiunit activity through a set of eight microwires (black arrow). This electrode is threaded inside each of the depth electrodes pictured in H. (Tracings AD from Bragin A, Wilson CL, Staba RJ, Reddick M, Fried I, Engel J Jr., Interictal high-frequency oscillations (80-500 Hz) in the human epileptic brain: entorhinal cortex. Ann Neurol. 2002 Oct; 52(4): 407-15, with permission. Part F courtesy of Gregory Worrell, M.D., Ph.D., Mayo Clinic, Rochester.)

Therapeutics, Devices, and Seizure Prediction


One of the strongest motivations for seizure prediction research is its potential for driving a therapeutic device. Focal brain stimulation for the treatment of epilepsy has been explored for over 20 years.8284 Early trials were often uncontrolled and thus inconclusive, and it was only later, when the Food and Drug Administration (FDA) instituted stricter standards and regulation of medical devices, that device safety, evaluation, and testing have considerably improved.85 The Vagus Nerve Stimulator (VNS, Cyberonics, Inc.), approved by the FDA in 1997 for pharmaco-resistant partial epilepsy, is the first antiepileptic device to be approved by this body and be widely used. Its operation is an example of a chronic stimulation open-loop protocol. The device functions by periodically stimulating the

1  Seizure Prediction: Its Evolution and Therapeutic Potential

vagus nerve with no direct feedback to modulate operation. In early trials, 30 to 40% of patients with medically intractable seizures experienced seizure reduction of approximately 50% with this therapy, though less than one patient in ten became seizure-free.8688 Driven by proof of principle that devices such as the VNS can reduce or stop seizures, and by the great success of analogous devices in cardiology, there is currently an explosion of research and development for neuro-implantables to treat epilepsy. The SANTE trial (Medtronic, Inc.) employs the same Deep Brain Stimulator (DBS) used for treating Parkinsons disease (Figure 1-4). It activates in an open-loop protocol to deliver intermittent electrical stimulation to the anterior nucleus of the thalamus (ANT) for treating partial onset epilepsy.89,90 Small trials have shown improvement in seizure severity.91,92 A multicenter pivotal clinical trial is currently underway to evaluate the devices efficacy. Early attempts to make this therapy responsive, or closed loop, have been inconclusive.93 Continuous hippocampal electrical stimulation for mesial temporal lobe epilepsy has also been attempted, with a reported reduction in seizure frequency of up to 50%, though this remains an active area of investigation.9497 Over the past 5 years there has been increasing interest in developing true closed-loop devices for treating epilepsy. Although there is some interest in making existing technologies, such as the VNS and ANTs stimulators, closed loop, more attention has been paid to technologies triggering direct cortical or hippocampal stimulation in response to electrocorticographic recordings from subdural strip and hippocampal depth electrodes. In such systems, a seizure detection algorithm is developed, validated, and implemented in a closed-loop stimulation protocol to prevent seizures.98 In these protocols, focal stimulation is only triggered when needed to abort a seizure. An intracranial Responsive Neurostimulator (RNS, NeuroPace, Inc., Mountain View, CA) is currently being investigated in a multicenter pivotal clinical trial for treatment of medically refractory partial onset seizures. This closed-loop protocol triggers electrical stimulation when the input algorithm detects that a seizure has started or is imminent on the intracranial EEG.98 There is some suggestion that triggering responsive stimulation earlier in the onset of seizure activity on the EEG may improve system efficacy (personal communication), though this has not been validated. An early study found a 45% decrease in seizures in seven out of eight patients99 and in an initial safety study not designed to test efficacy.100 Newer strategies for responsive stimulation err on the side of early detection, with relatively low specificity, in the hope of stimulating and potentially arresting seizures earlier. This is done at the expense of significant numbers of false-positive stimulations. The approach is based on the premise that there is no evidence that pulse stimulations that do not cause after-discharges have adverse effects on brain. These conclusions are based on extrapolations from the animal literature, as there are no human studies available to draw on.101 Some investigators propose that stimulation based on seizure prediction, or algorithms that identify periods of increased probability of seizure onset, may be an even more effective strategy. Such paradigms are an area of intense investigation. It is important to note that there is considerable research and commercial activity in the area of other forms of responsive stimulation and therapy for refractory seizures. Among modalities being investigated are open-loop and responsive trigeminal nerve stimulation,102,103 focal cooling,104 electrical field stimulation,105,106 and a variety of other therapeutic modalities.

10

THE EPILEPSIES 3

C
Figure 14

Two examples of implantable therapeutic antiepileptic devices, currently in clinical trials. AB, Two MRI scans from a single patient implanted with a Medtronic DBS (deep brain stimulation) device for electrically stimulating the anterior nucleus of the thalamus. A shows one electrode with four contacts from sagittal view (electrodes appear larger due to artifact), and B, same electrode in transverse brain image (Courtesy of Dr. Baltuch and the Penn Epilepsy Center). CD, A patient with an intracranial Responsive Neurostimulator (RNS) developed by Neuropace Inc., Mountain View, CA, picture on a skull x-ray from the lateral (C) and anteroposterior views (D). (CD, Reprinted with permission from Wood DA. Career management. Nursing Spectrum. 2004.)

Seizure Prediction Devices and Their Therapeutic Potential


Seizure prediction and antiepileptic device research are on trajectories that are already intersecting. Now that there is agreement on statistical methodologies to properly assess prediction algorithms, it is likely that many previously discarded methods will be found to predict seizures better than randomly. The question then becomes, what kind of performance is required for clinical utility? Such a

1  Seizure Prediction: Its Evolution and Therapeutic Potential

11

discussion is complex and depends on the application. For example, a device solely dedicated to warning of impending seizures will need to perform with a high degree of accuracy. Too many false-positive alarms will cause the device to be ignored. Rare false-negative warnings have the potential to be more dangerous, if patients engage in risky behavior reinforced by overconfidence in less-than-perfect device performance still, given that the unpredictability of seizures is perhaps the most common complaints in epilepsy patients, there is enormous potential to improve quality of life in those affected through warning systems alone. This remains an area of intense research. Systems that do not inform patients, but trigger therapeutic intervention that is not apparent to their hosts, may be easier to implement. This is because these devices can theoretically embody less-accurate performance, with regard to false-positive rates, if there is no penalty for false detections. Falsepositive detections may trigger unneeded intervention, but this is acceptable, if no short- or long-term side effects result. Again, the rationale for implementing seizure prediction in such devices is the hypothesis that earlier intervention in the process of seizure generation will be more effective in preventing clinical events than later interventions. Prediction devices, despite their increased complexity, also have theoretical advantages over open-loop systems, or those based on detection, in that they could potentially require less power, saving stimulation or other interventions for when they are needed, rather than continuously. Other complex issues, such as device cost, training, and maintenance, are beyond the scope of this discussion. Similarly, advances in implantable device technologies and electronics are making more complex systems possible by allowing increased computational power at lower energy costs than were possible only a few years ago. Rechargeable systems are also becoming more practical, though their implementation remains relatively rare in the medical device market. A variety of platforms are then possible for constructing seizure prediction devices. The question then becomes, what will they monitor, how will algorithms be implemented, and what kind of warning or intervention will they trigger? The earliest seizure warning device, crafted by Viglione in the 1970s, monitored standard scalp EEG and used an analog implementation of a neural network classifier for seizure prediction.1 Though rigorous statistical validation of device performance is not documented, the inventor was encouraged by his results. Research and now standard clinical practice demonstrate that intracranial EEG detects seizure onset 10 seconds or more in advance of scalp recordings, and that implanted electrodes are more stable and their recordings more artifact free than their scalp counterparts.107 For this reason, research in seizure prediction and antiepileptic devices has focused primarily on IEEG. IEEG electrodes are now frequently chronically implanted and used for sensing and stimulation in therapeutic neurological devices. Early results assessing long-term function and tissue compatibility are encouraging. These results suggest that early seizure prediction devices are likely to process IEEG and may potentially use the same electrodes for sensing signals and delivering therapy, if electrical stimulation remains the intervention of choice, due to its relative simplicity. Other sensors capable of recording intracranial-quality signals without requiring entry into the subdural space are also an intense area of investigation. The next question becomes what information to sense, where to place sensors, and where, when, and how to deliver therapy. These remain open questions and very active areas of research. There is encouraging evidence that pathological HFOs may help identify regions important to seizure generation in epileptic networks,55,63

12

THE EPILEPSIES 3

though thorough studies of continuous human broad-band EEG for this purpose are just getting under way. Other methods for localizing epileptic networks based on functional imaging are also encouraging, but remain experimental as well. Early work measuring very high frequency IEEG, particularly activity from individual neurons transduced by arrays of microelectrodes in regions felt to be important to seizure generation, is under way, and there is early evidence to suggest that HFOs and unit activity will allow seizures to be predicted or detected earlier than field potential recordings. This work is, as of yet, too preliminary to draw conclusions. Finally, it is not yet clear how intervention will be controlled by seizure prediction algorithms, as algorithm performance will be the primary determinant of this implementation. Several strategies have been suggested. In one, therapeutic intervention will escalate as the probability of seizure onset increases over time. In this scheme, more benign therapy, such as low current, very confined electrical stimulation, might be applied with modest increases in the probability of seizure onset and then increase in its spatial delivery and intensity if this probability continues to rise over time. Should seizure onset be detected in this scheme, maximal therapy, perhaps coupled with local drug infusion, might be triggered as a last resort, albeit with more potential side effects, to stop imminent clinical symptoms.35 Another strategy that has been suggested models developing seizures as a critical system, similar to avalanches and volcanoes. In this scheme, increasing probability of seizure onset is likened to snow gathering on avalanche-prone mountains. When this probability rises beyond a certain threshold, electrical stimulation is applied to provoke small, subclinical seizure-like events to release energy in the system. In this way a large, potentially convulsive event might be channeled into a series of small, subclinical electrical discharges that are not associated with any symptoms.108 These are only two potential methods of implementing an implantable antiepileptic device based on seizure prediction. The exact way in which a practical seizureprediction device might be employed will depend on algorithm performance and the results of ongoing research into delivering responsive therapy.

Future Directions
The actual embodiment of antiepileptic devices based on seizure prediction will be determined by the results of ongoing research focused on determining what signals to monitor, how, when, and where to deliver preemptive therapy to abort seizures, and how to optimize seizure prediction performance. This assumes that prospective seizure prediction can be demonstrated, which the authors think is likely, despite the large amount of time devoted to this research over the past 20 years. This optimism is the result of recent breakthroughs in the major areas of direction for the field, identified first at the ISPGs Bonn Workshop in 2001. It is our opinion, however, that the greatest hope for new seizure prediction-based devices will come in the form of collaborations. The International Collaborative Workshops on Seizure Prediction and the International Seizure Prediction Group provide a framework for this research and its clinical translation, which continues to drive it forward. Its next project is a large international database of human and animal model intracranial data on which prospective algorithms can be tested. This joint project will continue to accelerate the groups efforts and its progress toward making antiepileptic devices based on the neurophysiology of seizure generation a clinical reality.

1  Seizure Prediction: Its Evolution and Therapeutic Potential

13

REFERENCES
1. Viglione S, Ordon V, Martin W, Kesler C, inventors; US government, assignee. Epileptic seizure warning system. US patent 3 863 625. November 2, 1973. 2. Lehnertz K, Litt B. The First International Collaborative Workshop on Seizure Prediction: summary and data description. Clin Neurophysiol. 2005;116(3):493-505. 3. Annegers JF. The Epidemiology in Epilepsy. Baltimore: Williams and Wilkins; 1996. 4. DAlessandro M, Vachtsevanos G, Esteller R, et al. A multi-feature and multi-channel univariate selection process for seizure prediction. Clin Neurophysiol. 2005;116(3):506-516. 5. Iasemidis LD, Shiau DS, Chaovalitwongse W, et al. Adaptive epileptic seizure prediction system. IEEE Trans Biomed Eng. 2003;50(5):616-627. 6. Mormann F, Andrzejak RG, Elger CE, Lehnertz K. Seizure prediction: the long and winding road. Brain. 2007;130(Pt 2):314-333. 7. Wong S, Gardner AB, Krieger AM, Litt B. A stochastic framework for evaluating seizure prediction algorithms using hidden Markov models. J Neurophysiol. 2007;97(3):2525-2532. 8. Winterhalder M, Schelter B, Schulze-Bonhage A, Timmer J. Comment on: Performance of a seizure warning algorithm based on the dynamics of intracranial EEG.. Epilepsy Res. 2006;72(1):80-81; discussion 82-84. 9. Baumgartner C, Serles W, Leutmezer F, et al. Preictal SPECT in temporal lobe epilepsy: regional cerebral blood flow is increased prior to electroencephalography-seizure onset. J Nucl Med. 1998;39(6):978-982. 10. Weinand ME, Carter LP, el-Saadany WF, Sioutos PJ, Labiner DM, Oommen KJ. Cerebral blood flow and temporal lobe epileptogenicity. J Neurosurg. 1997;86(2):226-232. 11. Rajna P, Clemens B, Csibri E, et al. Hungarian multicentre epidemiologic study of the warning and initial symptoms (prodrome, aura) of epileptic seizures. Seizure. 1997;6(5):361-368. 12. Haut SR, Hall CB, LeValley AJ, Lipton RB. Can patients with epilepsy predict their seizures? Neurology. 2007;68(4):262-266. 13. Haut SR, Hall CB, Masur J, Lipton RB. Seizure occurrence: precipitants and prediction. Neurology. 2007;69(20):1905-1910. 14. Adelson PD, Nemoto E, Scheuer M, Painter M, Morgan J, Yonas H. Noninvasive continuous monitoring of cerebral oxygenation periictally using near-infrared spectroscopy: a preliminary report. Epilepsia. 1999;40(11):1484-1489. 15. Federico P, Abbott DF, Briellmann RS, Harvey AS, Jackson GD. Functional MRI of the pre-ictal state. Brain. 2005;128(Pt 8):1811-1817. 16. Delamont RS, Julu PO, Jamal GA. Changes in a measure of cardiac vagal activity before and after epileptic seizures. Epilepsy Res. 1999;35(2):87-94. 17. Kerem DH, Geva AB. Forecasting epilepsy from the heart rate signal. Med Biol Eng Comput. 2005;43(2):230-239. 18. Novak VV, Reeves LA, Novak P, Low AP, Sharbrough WF. Time-frequency mapping of R-R interval during complex partial seizures of temporal lobe origin. J Auton Nerv Syst. 1999;77(2-3):195-202. 19. Viglione SS, Ordon VA, et al. Detection and Prediction of Epileptic Seizures: McDonnell Douglas Astronautics Company; November 1972. Contract No. SRS-ORDT-71-50. 20. Viglione SS, Walsh GO. Proceedings: Epileptic seizure prediction. Electroencephalogr Clin Neurophysiol. 1975;39(4):435-436. 21. Rogowski Z, Gath I, Bental E. On the prediction of epileptic seizures. Biol Cybern. 1981;42(1):9-15. 22. Siegel A, Grady CL, Mirsky AF. Prediction of spike-wave bursts in absence epilepsy by EEG powerspectrum signals. Epilepsia. 1982;23(1):47-60. 23. Lange HH, Lieb JP, Engel J, Jr., Crandall PH. Temporo-spatial patterns of pre-ictal spike activity in human temporal lobe epilepsy. Electroencephalogr Clin Neurophysiol. 1983;56(6):543-555. 24. Gotman J, Marciani MG. Electroencephalographic spiking activity, drug levels, and seizure occurrence in epileptic patients. Ann Neurol. 1985;17(6):597-603. 25. Katz A, Marks DA, McCarthy G, Spencer SS. Does interictal spiking change prior to seizures? Electroencephalogr Clin Neurophysiol. 1991;79(2):153-156. 26. Iasemidis LD, Sackellares JC, Zaveri HP, Williams WJ. Phase space topography and the Lyapunov exponent of electrocorticograms in partial seizures. Brain Topogr. Spring. 1990;2(3):187-201. 27. Le Van Quyen M, Martinerie J, Navarro V, et al. Anticipation of epileptic seizures from standard EEG recordings. Lancet. 2001;357(9251):183-188. 28. Litt B, Lehnertz K. Seizure prediction and the preseizure period. Curr Opin Neurol. 2002;15(2): 173-177. 29. Litt B, Echauz J. Prediction of epileptic seizures. Lancet Neurol. 2002;1(1):22-30.

14

THE EPILEPSIES 3 30. Iasemidis LD, Shiau DS, Pardalos PM, et al. Long-term prospective on-line real-time seizure prediction. Clin Neurophysiol. 2005;116(3):532-544. 31. Elger CE, Lehnertz K. Seizure prediction by non-linear time series analysis of brain electrical activity. Eur J Neurosci. 1998;10(2):786-789. 32. Iasemidis LD, Pardalos PM, Sackellares JC, Shiau DS. Quadratic binary programming and dynamical system approach to determine the predictability of epileptic seizures. J Comb Optim. 2001;5:9-26. 33. Jerger KK, Weinstein SL, Sauer T, Schiff SJ. Multivariate linear discrimination of seizures. Clin Neurophysiol. 2005;116(3):545-551. 34. Schelter B, Winterhalder M, Maiwald T, et al. Testing statistical significance of multivariate time series analysis techniques for epileptic seizure prediction. Chaos. 2006;16(1):013108. 35. Litt B, Esteller R, Echauz J, et al. Epileptic seizures may begin hours in advance of clinical onset: a report of five patients. Neuron. 2001;30(1):51-64. 36. Mormann F, Lehnertz K, David P, Elger CE. Mean phase coherence as a measure for phase synchronization and its applications to the EEG of epilepsy patients. Physica D. 2000;144:358-369. 37. Schindler K, Wiest R, Kollar M, Donati F. EEG analysis with simulated neuronal cell models helps to detect pre-seizure changes. Clin Neurophysiol. 2002;113(4):604-614. 38. Mormann F, Andrzejak RG, Kreuz T, et al. Automated detection of a preseizure state based on a decrease in synchronization in intracranial electroencephalogram recordings from epilepsy patients. Phys Rev E Stat Nonlin Soft Matter Phys. 2003;67(2 Pt 1):021912. 39. Mormann F, Kreuz T, Andrzejak RG, David P, Lehnertz K, Elger CE. Epileptic seizures are preceded by a decrease in synchronization. Epilepsy Res. 2003;53(3):173-185. 40. Aschenbrenner-Scheibe R, Maiwald T, Winterhalder M, Voss HU, Timmer J, Schulze-Bonhage A. How well can epileptic seizures be predicted? An evaluation of a nonlinear method. Brain. 2003;126(Pt 12):2616-2626. 41. Harrison MA, Osorio I, Frei MG, Asuri S, Lai YC. Correlation dimension and integral do not predict epileptic seizures. Chaos. 2005;15(3):33106. 42. Lai YC, Harrison MA, Frei MG, Osorio I. Inability of Lyapunov exponents to predict epileptic seizures. Phys Rev Lett. 2003;91(6):068102. 43. Harrison MA, Frei MG, Osorio I. Accumulated energy revisited. Clin Neurophysiol. 2005;116(3): 527-531. 44. Maiwald T, Winterhalder M, Aschenbrenner-Scheibe R, Voss HU, Schulze-Bonhage A, Timmer J. Comparison of three nonlinear seizure prediction methods by means of the seizure prediction characteristic. Physica D. 2004;194:357-368. 45. McSharry PE, Smith LA, Tarassenko L. Prediction of epileptic seizures: are nonlinear methods relevant? Nat Med. 2003;9(3):241-242; author reply 242. 46. Esteller R, Echauz J, DAlessandro M, et al. Continuous energy variation during the seizure cycle: towards an on-line accumulated energy. Clin Neurophysiol. 2005;116(3):517-526. 47. Jouny CC, Franaszczuk PJ, Bergey GK. Signal complexity and synchrony of epileptic seizures: is there an identifiable preictal period?. Clin Neurophysiol. 2005;116(3):552-558. 48. Le Van Quyen M, Soss J, Navarro V, et al. Preictal state identification by synchronization changes in long-term intracranial EEG recordings. Clin Neurophysiol. 2005;116(3):559-568. 49. Mormann F, Kreuz T, Rieke C, et al. On the predictability of epileptic seizures. Clin Neurophysiol. 2005;116(3):569-587. 50. Andrzejak RG, Mormann F, Kreuz T, et al. Testing the null hypothesis of the nonexistence of a preseizure state. Phys Rev E Stat Nonlin Soft Matter Phys. 2003;67(1 Pt 1):010901. 51. Schelter B, Winterhalder M, Feldwisch genannt Drentrup H, et al. Seizure prediction: the impact of long prediction horizons. Epilepsy Res. 2007;73(2):213-217. 52. Schulze-Bonhage A, Kurth C, Carius A, Steinhoff BJ, Mayer T. Seizure anticipation by patients with focal and generalized epilepsy: a multicentre assessment of premonitory symptoms. Epilepsy Res. 2006;70(1):83-88. 53. Worrell GA, Parish L, Cranstoun SD, Jonas R, Baltuch G, Litt B. High-frequency oscillations and seizure generation in neocortical epilepsy. Brain. 2004;127(Pt 7):1496-1506. 54. Niederhauser JJ, Esteller R, Echauz J, Vachtsevanos G, Litt B. Detection of seizure precursors from depth-EEG using a sign periodogram transform. IEEE Trans Biomed Eng. 2003;50(4):449-458. 55. Bragin A, Engel J, Jr., Wilson CL, Fried I, Buzsaki G. High-frequency oscillations in human brain. Hippocampus. 1999;9(2):137-142. 56. Staba RJ, Frighetto L, Behnke EJ, et al. Increased fast ripple to ripple ratios correlate with reduced hippocampal volumes and neuron loss in temporal lobe epilepsy patients. Epilepsia. 2007.

1  Seizure Prediction: Its Evolution and Therapeutic Potential 57. Staba RJ, Wilson CL, Bragin A, Jhung D, Fried I, Engel J, Jr. High-frequency oscillations recorded in human medial temporal lobe during sleep. Ann Neurol. 2004;56(1):108-115. 58. Ekstrom A, Viskontas I, Kahana M, et al. Contrasting roles of neural firing rate and local field potentials in human memory. Hippocampus. 2007;17(8):606-617. 59. Ekstrom AD, Caplan JB, Ho E, Shattuck K, Fried I, Kahana MJ. Human hippocampal theta activity during virtual navigation. Hippocampus. 2005;15(7):881-889. 60. Ekstrom AD, Kahana MJ, Caplan JB, et al. Cellular networks underlying human spatial navigation. Nature. 2003;425(6954):184-188. 61. Kahana MJ. The cognitive correlates of human brain oscillations. J Neurosci. 2006;26(6): 1669-1672. 62. Jacobs J, Kahana MJ, Ekstrom AD, Fried I. Brain oscillations control timing of single-neuron activity in humans. J Neurosci. 2007;27(14):3839-3844. 63. Worrell G, Gardner A, Stead M, et al. High-frequency oscillations in human temporal lobe: simultaneous microwire and clinical electrode recordings. Brain. 2008;131:928-937. 64. Allen PJ, Fish DR, Smith SJ. Very high-frequency rhythmic activity during SEEG suppression in frontal lobe epilepsy. Electroencephalogr Clin Neurophysiol. 1992;82(2):155-159. 65. Fisher RS, Webber WR, Lesser RP, Arroyo S, Uematsu S. High-frequency EEG activity at the start of seizures. J Clin Neurophysiol. 1992;9(3):441-448. 66. Huang CM, White LE, Jr. High-frequency components in epileptiform EEG. J Neurosci Methods. 1989;30(3):197-201. 67. Staba RJ, Wilson CL, Bragin A, Fried I, Engel J, Jr. Quantitative analysis of high-frequency oscillations (80-500 Hz) recorded in human epileptic hippocampus and entorhinal cortex. J Neurophysiol. 2002;88(4):1743-1752. 68. Jirsch JD, Urrestarazu E, LeVan P, Olivier A, Dubeau F, Gotman J. High-frequency oscillations during human focal seizures. Brain. 2006;129(Pt 6):1593-1608. 69. Bragin A, Mody I, Wilson CL, Engel J, Jr. Local generation of fast ripples in epileptic brain. J Neurosci. 2002;22(5):2012-2021. 70. Schiff SJ, Colella D, Jacyna GM, et al. Brain chirps: spectrographic signatures of epileptic seizures. Clin Neurophysiol. 2000;111(6):953-958. 71. Rampp S, Stefan H. Fast activity as a surrogate marker of epileptic network function? Clin Neurophysiol. 2006;117(10):2111-2117. 72. Firpi H, Smart O, Worrell G, Marsh E, Dlugos D, Litt B. High-frequency oscillations detected in epileptic networks using swarmed neural-network features. Ann Biomed Eng. 2007;35(9): 1573-1584. 73. Gardner AB, Worrell GA, Marsh E, Dlugos D, Litt B. Human and automated detection of high-frequency oscillations in clinical intracranial EEG recordings. Clin Neurophysiol. 2007;118(5): 1134-1143. 74. Dichter M, Spencer WA. Hippocampal penicillin spike discharge: epileptic neuron or epileptic aggregate? Neurology. 1968;18(3):282-283. 75. Dichter M, Spencer WA. Penicillin-induced interictal discharges from the cat hippocampus. II. Mechanisms underlying origin and restriction. J Neurophysiol. 1969;32(5):663-687. 76. Dichter M, Spencer WA. Penicillin-induced interictal discharges from the cat hippocampus. I. Characteristics and topographical features. J Neurophysiol. 1969;32(5):649-662. 77. Dzhala VI, Staley KJ. Transition from interictal to ictal activity in limbic networks in vitro. J Neurosci. 2003;23(21):7873-7880. 78. Dzhala VI, Staley KJ. Mechanisms of fast ripples in the hippocampus. J Neurosci. 2004;24(40): 8896-9906. 79. McCormick DA, Contreras D. On the cellular and network bases of epileptic seizures. Annu Rev Physiol. 2001;63:815-846. 80. Traub RD, Contreras D, Cunningham MO, et al. Single-column thalamocortical network model exhibiting gamma oscillations, sleep spindles, and epileptogenic bursts. J Neurophysiol. 2005;93(4):2194-2232. 81. Huguenard JR, McCormick DA. Thalamic synchrony and dynamic regulation of global forebrain oscillations. Trends Neurosci. 2007;30(7):350-356. 82. Durand D. Electrical stimulation can inhibit synchronized neuronal activity. Brain Res. 1986;382(1):139-144. 83. Litt B, Baltuch G. Brain Stimulation for Epilepsy. Epilepsy Behav. 2001;(2):S61-S67. 84. Theodore WH, Fisher RS. Brain stimulation for epilepsy. Lancet Neurol. 2004;3(2):111-118. 85. Litt B. Evaluating devices for treating epilepsy. Epilepsia. 2003;44(Suppl 7):30-37.

15

16

THE EPILEPSIES 3 86. Fisher RS, Handforth A. Reassessment: vagus nerve stimulation for epilepsy: a report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Neurology. 1999;53(4):666-669. 87. Fisher RS, Krauss GL, Ramsay E, Laxer K, Gates J. Assessment of vagus nerve stimulation for epilepsy: report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Neurology. 1997;49(1):293-297. 88. Morrell M. Brain stimulation for epilepsy: can scheduled or responsive neurostimulation stop seizures? Current Opinion in Neurology. 2006;19(2):164-168. 89. Graves NM, Fisher RS. Neurostimulation for epilepsy, including a pilot study of anterior nucleus stimulation. Clin Neurosurg. 2005;52:127-134. 90. Mirski MA, Rossell LA, Terry JB, Fisher RS. Anticonvulsant effect of anterior thalamic high frequency electrical stimulation in the rat. Epilepsy Res. 1997;28(2):89-100. 91. Hodaie M, Wennberg RA, Dostrovsky JO, Lozano AM. Chronic anterior thalamus stimulation for intractable epilepsy. Epilepsia. 2002;43(6):603-608. 92. Kerrigan JF, Litt B, Fisher RS, et al. Electrical stimulation of the anterior nucleus of the thalamus for the treatment of intractable epilepsy. Epilepsia. 2004;45(4):346-354. 93. Osorio I, Frei MG, Sunderam S, et al. Automated seizure abatement in humans using electrical stimulation. Ann Neurol. 2005;57(2):258-268. 94. Tellez-Zenteno JF, McLachlan RS, Parrent A, Kubu CS, Wiebe S. Hippocampal electrical stimulation in mesial temporal lobe epilepsy. Neurology. 2006;66(10):1490-1494. 95. Velasco AL, Velasco F, Velasco M, Trejo D, Castro G, Carrillo-Ruiz JD. Electrical stimulation of the hippocampal epileptic foci for seizure control: a double-blind, long-term follow-up study. Epilepsia. 2007;48(10):1895-1903. 96. Vonck K, Boon P, Achten E, De Reuck J, Caemaert J. Long-term amygdalohippocampal stimulation for refractory temporal lobe epilepsy. Ann Neurol. 2002;52(5):556-565. 97. Vonck K, Boon P, Claeys P, Dedeurwaerdere S, Achten R, Van Roost D. Long-term deep brain stimulation for refractory temporal lobe epilepsy. Epilepsia. 2005;46 (Suppl 5):98-99. 98. Kossoff EH, Ritzl EK, Politsky JM, et al. Effect of an external responsive neurostimulator on seizures and electrographic discharges during subdural electrode monitoring. Epilepsia. 2004;45(12): 1560-1567. 99. Fountas KN, Smith JR, Murro AM, Politsky J, Park YD, Jenkins PD. Implantation of a closed-loop stimulation in the management of medically refractory focal epilepsy: a technical note. Stereotact Funct Neurosurg. 2005;83(4):153-158. 100. Worrell G, Wharen R, Goodman R, et al. Safety and evidence for efficacy of an implantable responsive neurostimulator (RNS [R]) for the treatment of medically intractable partial onset epilepsy in adults. Paper presented at: Annual Meeting of the American Epilepsy Society; December, 2005. 101. Wada JA, Tsuchimochi H. Cingulate kindling in Senegalese baboons, Papio papio. Epilepsia. 1995;36(11):1142-1151. 102. DeGiorgio CM, Shewmon DA, Whitehurst T. Trigeminal nerve stimulation for epilepsy. Neurology. 2003;61(3):421-422. 103. Fanselow EE, Reid AP, Nicolelis MA. Reduction of pentylenetetrazole-induced seizure activity in awake rats by seizure-triggered trigeminal nerve stimulation. J Neurosci. 2000;20(21):8160-8168. 104. Yang XF, Duffy DW, Morley RE, Rothman SM. Neocortical seizure termination by focal cooling: temperature dependence and automated seizure detection. Epilepsia. 2002;43(3):240-245. 105. Bikson M, Lian J, Hahn PJ, Stacey WC, Sciortino C, Durand DM. Suppression of epileptiform activity by high frequency sinusoidal fields in rat hippocampal slices. J Physiol. 2001;531(Pt 1):181-191. 106. Gluckman BJ, Nguyen H, Weinstein SL, Schiff SJ. Adaptive electric field control of epileptic seizures. J Neurosci. 2001;21(2):590-600. 107. Engel J, ed. Surgical Treatment of the Epilepsies. 1st ed. New York: Raven Press; 1987; No. 1. 108. Worrell GA, Cranstoun SD, Echauz J, Litt B. Evidence for self-organized criticality in human epileptic hippocampus. NeuroReport. 2002;13(16):2017-2021. 109. Wood DA. Career management. Nursing Spectrum. 2004.

THE EPILEPSIES 3

Febrile Seizures
LINE M. DUBE  TALLIE Z. BARAM CE

What Are Febrile Seizures? Frequency and Pathophysiology of Febrile Seizures Types of Febrile Seizures

Outcome of Febrile Seizures Evaluation and Management of Febrile Seizures

What are Febrile Seizures?


Several definitions of febrile seizures exist. The National Institutes of Health (NIH) Consensus Conference1,2 defined febrile seizures as events in infancy or childhood, usually occurring between 3 months and 5 years of age, associated with fever but without evidence of intracranial infection or defined cause. Seizures with fever in children who have suffered a previous nonfebrile seizure are excluded. More recently the International League Against Epilepsy (ILAE)3 defined febrile seizures as seizures occurring in childhood after age one month, associated with a febrile illness not caused by an infection of the central nervous system, without previous neonatal seizures or a previous unprovoked seizure, and not meeting criteria for other acute symptomatic seizures. Note that the NIH and ILAE definitions differ in both the minimal and maximal age permitted for febrile seizures, and this may cause confusion among parents and physicians. In addition, although both the NIH and ILAE definitions exclude children with prior afebrile seizures and those with seizures due to an intracranial infection or other specific cause, they do not exclude children with preexisting neurological deficits. This may confound studies looking at the cognitive outcome of these seizures. Note also that fever is not defined. In general, a temperature higher than 388C4,5 or 38.48C (1018F) has been defined as constituting fever. The majority of febrile seizures occur early in the febrile episode and often are the presenting sign of the fever.68 A typical scenario is a child who presents with a seizure, and the presence of fever is discovered later, once the child arrives in an emergency room. Whether the rate of temperature rise is an important determinant of the occurrence of a febrile seizure, rather than the actual temperature achieved, is not well substantiated.9,10

17

18

THE EPILEPSIES 3

Frequency and Pathophysiology of Febrile Seizures


Several large population-based studies examined the incidence (the number of new cases occurring in a defined population over a specified period of time) and the prevalence (proportion of individuals in a population that has ever had the disorder) of febrile seizures. They found similar rates (2 to 5%) in the Western world and higher rates in Japan and Guam (Figure 2-1). For examples, in Rochester, Minnesota, the annual incidence of febrile seizures was 4/1000 children younger than age 5, and 2% of children experienced a febrile seizure by the age of 5.11,12 The large Oakland, California,13 and British studies14 also found that by the age of 5 years 2 to 2.3% of children had had a febrile seizure. The National Collaborative Perinatal Project (NCPP) reported a somewhat higher rate (3.5%15,16), that was similar to results from Sweden (cumulative prevalence of 4.1% in children under age 517) and Holland (a rate of 3.9%18). In contrast to their frequency in the Western world, febrile seizures occur in 8 to 10% of Japanese children4 and of those in Guam.19,20 This increased incidence is likely a result of both environmental factors (e.g., parents and children sleep in close proximity, so that febrile seizures are more likely to be observed), as well as genetic elements. Why will a child have a febrile seizure? Current evidence supports the notion that in a given child, environmental and genetic factors may interact to determine whether that child will have a febrile seizure. Furthermore, the relative contribution of each varies significantly: febrile seizures occur sporadically as well as run in families, where the contribution of genetic background to their onset is likely higher.21 Looking at risk factors for febrile seizures, Bethune et al.22 found that both attending day care (increasing the chance of fever and infection) and a history of a family member with febrile seizure tripled the risk of having such seizures. Contribution of presumed genetic background (presence of febrile seizures in a relative), as well as environmental factors (e.g., maternal smoking during pregnancy), was reported also by Berg et al.23

6 5 Incidence (%) 4 3 2 1 0 0 6 12 18 24 30 36 42 48 54 60 66 72 78 Japan Rochester, Minnesota Oakland, California

Age (months)

Figure 21

Age-specific incidence of febrile seizures in the United States and in Japan. (Reprinted with permission from Hauser [1994], Blackwell Publishing.)

2  Febrile Seizures

19

How does fever lead to a febrile seizure? Temperature influences neuronal electrophysiological activity.24 For example, the function of ion channels such as the TRPV4 (transient receptor potential vanilloid 4) are markedly dependent on temperature in the physiological and fever ranges, 368C to 428C.25 Children can also develop seizures on simple hyperthermia and not fever, such as hyperthermia related to anticholinergic overdose26 or hot showers,27 suggesting that increased temperature in itself can be a convulsant in children. Recently, hyperthermia in an animal model has been shown to induce hyperventilation resulting in alkalosis that in turn provoked seizures.28 This happened 25 to 30 minutes after the onset of the hyperthermia and was not observed in a second febrile seizure model in an immature rat.29 In the latter case, seizures commenced within A: 2.9 8 0.1 minutes of the onset of the hyperthermia procedure (n = 52), and respiratory rates when seizures started B: 167 8 4.9, n = 12 did not differ significantly from those rates prior to C: 161.8 8 2.6 or during the first minute of D: 173 8 3.7 the hyperthermia. Whether alkalosis plays a role in febrile seizures in children remains unknown. Genetic susceptibility may increase the ability of fever to provoke seizures. Recently, mutations of several ion channels that predispose to febrile seizures have been described,3033 including mutations of sodium30,31 and chloride (GABAA receptor32,33) channels. In many families, these febrile seizures are associated with epilepsy of diverse phenotypes, a syndrome termed GEFS+ (generalized epilepsy with febrile seizures plus).34 Other single-gene mutations, such as in the interleukin gene promoter (see later discussion) might also render individuals more susceptible to febrile seizures,35,36 and multigene interactions might contribute to the occurrence of these seizures in a more complex and subtle way. Fever may promote seizures via the function of pyrogenic cytokines including interleukin-1.37,38 This notion is supported by reports of mutations in the interleukin-1 gene promoter that result in increased production of the cytokine in individuals with febrile seizures,36 although the significance of this finding has been debated.39 In summary, the contribution of gene-environment interaction to the generation of febrile seizures, and to their potential causal relationship to epilepsy,40 is a topic of active investigation; increasingly sophisticated experimental tools should rapidly advance our understanding of this question. Specific infectious etiology of the fever might contribute to the generation of human febrile seizures. A disproportionate number of febrile seizures has been associated with infection with the human herpes virus 6 (HHV6) in some,41 but not all, studies.42 This suggests that mechanisms specific to this virus, including perhaps a unique profile of cytokine induction, might selectively augment neuronal excitability and thus preferentially provoke seizures.

Types of Febrile Seizures


Febrile seizures are divided into: (a) simple (70 to 80% of the total5,15,18,4345), (b) complex, and (c) febrile status epilepticus (FSE). Simple febrile seizures are short generalized seizures, typically but not necessarily motor, that last less than 108,46 or 1516,45 minutes. Specifically, these seizures do not have focal features and do not recur within the same febrile episode (or within 24 hours from the first seizure). Complex febrile seizures are either longer, lasting

20

THE EPILEPSIES 3

>10 minutes8,46 or >15 minutes,16,45 have focal features, or recur within 24 hours of the first episode15,46 or within the same febrile illness.47 Febrile status epilepticus (FSE) is defined as febrile seizures longer than 30 minutes. Note that Scott et al.48,49 define seizures lasting more than 30 minutes in normal children without intracranial infection as prolonged febrile seizures rather than status epilepticus. Complex febrile seizures, and particularly febrile status epilepticus, are associated with a higher risk of later epilepsy (see later discussion), so recognizing them is of clinical importance. Recent research has found that the duration of a seizure is regulated in each child independent of the probability of having a seizure. Thus, Shinnar et al.50 have found that once a child has had a prolonged seizure, a subsequent seizure is likely to be prolonged as well. This is important for targeting children for abortive therapy to prevent long febrile seizures (see later discussion).

Outcome of Febrile Seizures


The outcome of febrile seizures is measured in terms of cognitive function (e.g., school performance, cognitive tests) and the risk for developing epilepsy. Several large studies have suggested that simple febrile seizures, nonfocal and with a duration of less than 10 to 15 minutes, do not lead to long-term sequelae16,43,46,51,52 in either of these realms. However, a smaller study from Taiwan cautioned that infants less than a year of age at the time of their seizure might function less well in certain learning tasks.53,54 By and large, however, the overwhelming evidence regarding simple febrile seizures is that they are benign. The outcome of complex febrile seizures and of febrile status epilepticus is more controversial.16,45,46,55 In terms of cognitive function, both the American National Collaborative Perinatal Project16,56 and the British cohort51 used general broad measures such as school performance as outcome and evaluated children at age 6 or 10, not later. They concluded that complex febrile seizures do not alter cognition, but their design, as mentioned earlier, limits the confidence of these conclusions. Evaluations of cognitive outcome of prolonged febrile seizures or febrile status are complicated by the fact that many epidemiological studies do not exclude children who are prior neurologically compromised (see definition of febrile seizures in the section What Are Febrile Seizures?). Inclusion of children with neurological deficits before the seizures occurred complicates attempts to look at the impact of these seizures on intellect. Notably, British investigators have recently begun to better define the populations they study to exclude children with neurological deficits. The results of these studies should be helpful to sort out the consequences of these seizures on the normal developing brain.57 Whether prolonged febrile seizures and febrile status epilepticus lead to epilepsy has been a subject of debate. Whereas prospective epidemiological evaluations have provided little evidence for epileptogenesis,16,45,46 retrospective analyses have strongly linked a history of complex, and particularly of prolonged, febrile seizures to temporal lobe epilepsy,5861 suggesting a potential contribution of febrile seizures to epileptogenesis. It is not easy to reconcile these conflicting data, yet the reader should bear in mind that each type of study has inherent limitations. Prospective epidemiological studies benefit from large numbers and from the fact that they are not skewed by clinical populations in referral centers. In contrast, they usually do not have long follow-up and may miss important associations. Retrospective studies

2  Febrile Seizures

21

involve people who are identified after the fact (or with a known outcome, such as epilepsy) and evaluated retroactively. These studies can identify associations in relatively small populations, but are open to ascertainment and recall biases and may assess selected (e.g., more severely affected) populations that do not reflect a general principle. Animal models of long febrile seizures have been used to address the question of the potential epileptogenic consequences of these seizures. These models involved induction of experimental febrile seizures in a normal rodent brain28,29,37,6271 or in animals that have been subjected to prior insults.7273 Recently, the use of an experimental model of prolonged febrile seizures in immature rats found that these seizures may promote limbic, temporal lobe epilepsy (TLE) in a minority of animals. The mechanisms by which these seizures contribute to temporal lobe epilepsy in this model are not fully resolved. However, the evidence suggests that the seizures do not lead to brain damage (neuronal death).65,74,75 Specifically, transient neuronal injury was provoked by these seizures, but the involved neurons did not die.65 Neuronal counts in several brain regions with established vulnerability to seizures failed to reveal cell loss.65,74,75 Studies for apoptosis did not show increased cell death at any time point after the seizures,65 even when the seizures were carried out for 60 minutes. In addition, neurogenesis that may be induced in other developmental seizures74 did not follow experimental febrile seizures. Other expected structural changes, including mossy fiber sprouting, were also minimal after these seizures74,76,77 and did not explain the conversion of the hippocampal circuit into a hyperexcitable one. The likely mechanisms for seizure-evoked hippocampal hyperexcitability29,78,79 involve profound and persistent molecular changes in hippocampal neurons that lead to a change of their intrinsic excitability and their responses to input from other neurons. Whereas the repertoire and sequence of molecular changes evoked by experimental prolonged febrile seizures in this model have not yet been fully determined, persistent changes in the expression of specific ion channels likely play a role, as do alterations in endocannabinoid signaling.67 Following experimental prolonged febrile seizures, the properties of Ih, a hyperpolarization-triggered cationic current that contributes to the maintenance of neuronal membrane potential, subthreshold oscillations, and dendritic integration,8082 in hippocampal neurons were altered.28,79 This led to increased probability of frequency-dependent rebound depolarization in response to hyperpolarizing input.78,79 At the molecular level, these changes were a result of longlasting altered expression of hyperpolarization-activated cyclic-nucleotidegated (HCN) channels that conduct this current, particularly a reduction of HCN1 isoforms, and perhaps also of increased formation of HCN1/HCN2 heteromeric channels.83,84 The relevance of the alterations in HCN channels and Ih to human epileptogenesis remains to be fully defined. Altered HCN1 channel expression was found in a subset of resected hippocampi from patients with temporal lobe epilepsy and mesial temporal sclerosis, often with a history of early life seizures.85 This suggests that HCN channels are altered in human epilepsy. However, whether mutations in HCN channel genes that alter Ih contribute to human epilepsy remains unknown. Altered endocannabinoid signaling also followed prolonged experimental febrile seizures, contributing to hyperexcitability. This resulted from an increase in the number of presynaptic cannabinoid type 1 receptors, which increased retrograde

22

THE EPILEPSIES 3

inhibition of GABA release, promoting hyperexcitability.67 It is very likely that many other enduring changes in the expression of key genes that govern neuronal and network excitability may take place after experimental febrile seizures and perhaps human prolonged febrile seizures. These are expected to coordinately alter the probability of the occurrence of spontaneous seizures (i.e., epilepsy).

Evaluation and Management of Febrile Seizures


The evaluation and management of the different types of febrile seizures depend on the age and clinical setting of the child that presents with the seizure, as well as the type of the seizures (simple, complex, or febrile status epilepticus). The first consideration is whether the febrile seizure might be a herald of meningitis or meningoencephalitis, especially in young children (less than 12 months), where clinical signs of CNS infection might be few.86 Because seizures may be the first sign of meningitis in 13 to 16% of children with meningitis, and meningeal sign may be absent or subtle in younger children, the American Academy of Pediatrics recommends a lumbar puncture for all individuals younger than 12 months who present with a febrile seizure and strong consideration of the procedure for older individuals, up to 18 months of age.87 Beyond this consideration, several guidelines recommend minimal evaluation and no treatment of a normal child with a simple febrile seizure.1,8790 Thus, EEG, imaging, and preventive or abortive therapy are not indicated.87,91,92 For prolonged febrile seizures (and less clearly for recurrent or focal types of complex febrile seizures) the approach of clinicians is in flux. In the proper clinical setting (e.g., with a strong family history) evaluation for GEFS+ (e.g., DNA for SCN1A mutations) might be indicated. With a prolonged seizure, particularly if focal, magnetic resonance imaging (MRI) changes have been reported,48,49,93 and imaging may be indicated, together with follow-up. The role of electroencephalography (EEG) is still not clear. An early, large study94 described EEG abnormalities in about a third of children with febrile seizures, mostly consisting of global slowing. However, the presence of an abnormal EEG did not correlate with seizure recurrence,94,95 or with the subsequent development of epilepsy95 and is not a required element of the evaluation of febrile seizures.87 Novel data from the multicenter study of febrile status epilepticus (FEBSTAT) is now emerging, suggesting that focal slowing after these seizures that last more than an hour might predict MRI abnormalities (S. Shinnar, personal communication). Treatment of febrile seizures is rarely indicated. Prophylactic treatment of a child with febrile seizures using barbiturates or valproate may reduce the number of further seizures, but there is no evidence that this approach alters the probability of developing epilepsy,52 and the side effects of daily medication are not justified.21,46,52,96 Attempts at initiating prophylactic diazepam at the onset of fever failed9799 because children became drowsy and irritable, thus reducing compliance. In addition, most febrile seizures occur at the onset of the febrile illnessbefore diazepam can be initiated.100 An emerging consensus is the notion that prolonged febrile seizures should be prevented in anyone who has had a previous prolonged febrile seizure. This can be achieved via abortive therapy using a benzodiazepine. The administration of rectal

2  Febrile Seizures

23

diazepam99,101,102 or buccal or nasal midazolam103,104 is rapidly effective, preventing a short febrile seizure from transforming to a prolonged one or to febrile status epilepticus and often preventing stressful emergency room visits.

Summary
Febrile seizures are common, and simple febrile seizures are benign. Much remains to be learned about the pathophysiology of febrile seizures, the biological differences of simple and prolonged seizures, as well as the consequences of the latter. The emerging availability of biomarkers (e.g., imaging) for the consequences of long febrile seizures might allow clinicians to determine whether a child is at risk of cognitive or pro-epileptogenic sequelae from these prolonged febrile seizures and treat these children specifically. For the majority of febrile seizures, treatment is not indicated. Acknowledgments The authors thank Ms. Joy Calara for excellent editorial help. This chapter is supported by NIH grant R37 NS35439. REFERENCES
1. National Institutes of Health. Febrile seizures: consensus development conference summary. Bethesda, National Institutes of Health 3:2, 1980. 2. Nelson KB, Ellenberg JH. Febrile Seizures. New York, Raven Press; 1981. 3. Commission on Epidemiology and Prognosis, International League Against Epilepsy. Guidelines for epidemiologic studies on epilepsy. Epilepsia. 1993;34:592-596. 4. Tsuboi, T. Epidemiology of febrile and afebrile convulsions in children in Japan. Neurology. 1984;34:175-181. 5. Al-Eissa Y A. Febrile seizures: rate and risk factors of recurrence. J. Child Neurol. 1995;0:315-319. 6. Wolf SM, Carr A, Davis D. The value of phenobarbital in the child who has had a single febrile seizure: a controlled prospective study. Pediatrics. 1977;59:378-385. 7. Berg AT, Shinnar S, Hauser WA. Predictors of recurrent febrile seizures: a prospective study of the circumstances surrounding the initial febrile seizure. N Engl J Med. 1992;327:1122-1127. 8. Berg AT, Shinnar S, Doufsky AS. Predictors of recurrent febrile seizures: a prospective cohort study. Arch Pediatr Adolesc Med. 151:371-378. 9. Michon PE, Wallace SJ. Febrile convulsions: electroencephalographic changes related to rectal temperature. Arch Dis Child. 1984;59:371-373. 10. Berg AT. Are febrile seizures provoked by a rapid rise in temperature? Am J Dis Child. 1993;147:1101-1103. 11. Hauser W, Hersdorffer D. Epilepsy: Frequency, Causes and Consequences. New York: Demos, 1990. 12. Hauser WA, Kurland LT. The epidemiology of epilepsy in Rochester, Minnesota, 1935 through 1967. Epilepsia. 1975;16:1-166. 13. van den Berg BJ, Yerushalmy J. Studies on convulsive disorders in young children. I. Incidence of febrile and nonfebrile convulsions by age and other factors. Pediatr Res. 1969;3:298-304. 14. Ross EM, Peckham CS, West PB, Butler NR. Epilepsy in childhood: findings from the National Child Development Study. Brit Med J. 1980;280:207-210. 15. Nelson KB, Ellenberg JH. Predictors of epilepsy in children who have experienced febrile seizures. N Engl J Med. 1976;295:1029-1033. 16. Nelson KB, Ellenberg JH. Prognosis in children with febrile seizures. Pediatrics. 1978;61:720-727. 17. Forsgren L, Sidenvall R, Blomquist HK, Heijbel J. A prospective incidence study of febrile convulsions. Acta Paediatr Scand. 1990;79:550-557. 18. Offringa M et al. Prevalence of febrile seizures in Dutch schoolchildren. Paediatr Perinat Epidemiol. 1991;5:181-188.

24

THE EPILEPSIES 3 19. Stanhope JM et al. Convulsions among the Chamorro people of Guam, Mariana Islands. Am J Epidemiol. 1972;95:299-304. 20. Pal DK. Methodologic issues in assessing risk factors for epilepsy in an epidemiologic study in India. Neurology. 1999;53:2058-2063. 21. Berg AT et al. Childhood-onset epilepsy with and without preceding febrile seizures. Neurology. 1999;53:1742-1748. 22. Bethune P, et al. Which child will have a febrile seizure? Am J Dis Child. 1993;147:35-39. 23. Berg AT et al. Risk factors for a first febrile seizure: a matched case-control study. Epilepsia. 1995;36:334-341. 24. Hodgkin AL, Katz B. The effect of temperature on the electrical activity of the giant axon of the squid. J Physiol. 1949;109:240-249. 25. Shibasaki K et al. Effects of body temperature on neural activity in the hippocampus: regulation of resting membrane potentials by transient receptor potential vanilloid 4. J Neurosci. 2007;27:1566-1575. 26. Olson KR et al. Seizures associated with poisoning and drug overdose. Am J Emerg Med. 1994;12:392-395. 27. Fukuda M et al. Clinical study of epilepsy with severe febrile seizures and seizures induced by hot water bath. Brain Dev. 1997;19:212-216. 28. Schuchmann S et al. Experimental febrile seizures are precipitated by a hyperthermia-induced respiratory alkalosis. Nat Med. 2006;12:817-823. C et al. Prolonged febrile seizures in the immature rat model enhance hippocampal excitability 29. Dube long-term. Ann Neurol. 2000;47:336-344. 30. Wallace RH et al. Febrile seizures and generalized epilepsy associated with a mutation in the Na+-channel beta1 subunit gene SCN1B. Nat Genet. 1998;19:366-370. 31. Escayg A et al. Mutations of SCN1A, encoding a neuronal sodium channel, in two families with GEFS+2. Nat Genet. 2000;24:343-345. 32. Harkin LA et al. Truncation of the GABA(A)-receptor gamma2 subunit in a family with generalized epilepsy with febrile seizures plus. Am J Hum Genet. 2002;70:530-536. 33. Kang JQ et al. Why does fever trigger febrile seizures? GABAA receptor gamma2 subunit mutations associated with idiopathic generalized epilepsies have temperature-dependent trafficking deficiencies. J Neurosci. 2006;26:2590-2597. 34. Berkovic S, Scheffer I. Febrile seizures: genetics and relationship to other epilepsy syndromes. Curr Opin Neurol. 1998;11:129-134. 35. Kanemoto K et al. Interleukin (IL)1beta, IL-1alpha, and IL-1 receptor antagonist gene polymorphisms in patients with temporal lobe epilepsy. Ann Neurol. 2000;47:571-574. 36. Virta, M et al. Increased frequency of interleukin-1beta (-511) allele 2 in febrile seizures. Pediatr Neurol. 2002;26:192-195. , C et al. Interleukin-1 beta contributes to the generation of experimental febrile seizures. 37. Dube Ann Neurol. 2005;57:152-155. 38. Vezzani A, Baram TZ. New roles for interleukin-1 beta in the mechanisms of epilepsy. Epilepsy Curr. 2007;7:45-50. 39. Tan NC et al. Genetic association studies in epilepsy: the truth is out there. Epilepsia. 2004;45:1429-1442. 40. Duncan JS et al. Adult epilepsy. Lancet. 2006;367:1087-1100. 41. Barone SR et al. Human herpesvirus-6 infection in children with first febrile seizures. J Pediatr. 1995;127:95-97. 42. Zerr DM et al. A population-based study of primary human herpesvirus 6 infection. N Engl J Med. 2005;352:768-776. 43. Verity CM et al. Febrile convulsions in a national cohort followed up from birth. I. Prevalence and recurrence in the first five years of life. Brit Med J. 1985;290:1307-1310. 44. Annegers JF et al. Recurrence risk of febrile convulsions in a population-based cohort. Epilepsy Res. 1990;5:209-216. 45. Berg AT, Shinnar S. Complex febrile seizures. Epilepsia. 1996;37:126-133. 46. Annegers JF et al. Factors prognostic of unprovoked seizures after febrile convulsions. N Engl J Med. 1987;316:493-498. 47. Camfield P, Camfield C. Febrile seizures. In Shinnar S, Amir N, Branski D, eds. Childhood Seizures. Basel, Karger,1995, 32-38. 48. Scott RC et al. Magnetic resonance imaging findings within 5 days of status epilepticus in childhood. Brain. 2002;125:1951-1959.

2  Febrile Seizures 49. Scott RC et al. Hippocampal abnormalities after prolonged febrile convulsion: a longitudinal MRI study. Brain. 2003;126:2551-2557. 50. Shinnar S et al. How long do new-onset seizures in children last? Ann Neurol. 2001;49: 659-664. 51. Verity CM et al. Long-term intellectual and behavioral outcomes of children with febrile convulsions. N Engl J Med. 1998;338:1723-1728. 52. Berg AT, Shinnar S. Do seizures beget seizures? An assessment of the clinical evidence in humans. J Clin Neurophysiol. 1997;14:102-110. 53. Chang YC et al. Working memory of school-aged children with a history of febrile convulsions: a population study. Neurology. 2001;57:37-42. 54. Baram TZ, Shinnar S. Do febrile seizures improve memory? Neurology. 2001;57:7-8. 55. Raol YS et al. Epilepsy after early-life seizures can be independent of hippocampal injury. Ann Neurol. 2003;53:503-511. 56. Hirtz DG et al. Febrile convulsions. In: Engel Jr. J, Pedley TA, eds. Epilepsy: A Comprehensive Textbook. Vol. 3. Philadelphia: Lippincott-Raven, 1997, 2483-2488. 57. Raspall-Chaure M et al. Outcome of paediatric convulsive status epilepticus: a systematic review. Lancet Neurol. 2006;5:769-779. 58. Cendes F et al. Atrophy of mesial temporal lobe structures in patients with temporal lobe epilepsy: cause or consequence of repeated seizures? Ann Neurol. 1993;34:795-801. 59. French JA et al. Characteristics of medial temporal lobe epilepsy: I. Results of history and physical examination. Ann Neurol. 1993;34:774-780. 60. Hamati-Haddad A, Abou-Khalil B. Epilepsy diagnosis and localization in patients with antecedent childhood febrile convulsions. Neurology. 1998;50:917-922. 61. Theodore WH et al. Hippocampal atrophy, epilepsy duration, and febrile seizures in patients with partial seizures. Neurology. 1999;52:132-136. 62. Holtzman D et al. Hyperthermia-induced seizures in the rat pup: a model for febrile convulsions in children. Science. 1981;213:1034-1036. 63. Morimoto, T et al. Electroencephalographic study of rat hyperthermic seizures. Epilepsia. 1991;32:289-293. 64. Baram TZ et al. Febrile seizures: an appropriate-aged model suitable for long-term studies. Dev Brain Res. 1997;98:265-270. 65. Toth Z et al. Seizure-induced neuronal injury: vulnerability to febrile seizures in an immature rat model. J Neurosci. 1998;18:4285-4294. 66. Chang YC et al. Febrile seizures impair memory and cAMP-response element binding protein activation. Ann Neurol. 2003;4:706-718. 67. Chen K et al. Long-term plasticity of endocannabinoid signaling induced by developmental febrile seizures. Neuron. 2003;39:599-611. 68. Heida JG et al. Lipopolysaccharide-induced febrile convulsions in the rat: short-term sequelae. Epilepsia. 2004;45:1317-1329. 69. Heida JG, Pittman QJ. Causal links between brain cytokines and experimental febrile convulsions in the rat. Epilepsia. 2005;46:1906-1913. 70. Lemmens EM et al. Gender differences in febrile seizure-induced proliferation and survival in the rat dentate gyrus. Epilepsia. 2005;46:1603-1612. 71. Kamal A et al. Persistent changes in action potential broadening and the slow afterhyperpolarization in rat CA1 pyramidal cells after febrile seizures. Eur J Neurosci. 2006;23:2230-2234. 72. Germano IM et al. Neuronal migration disorders increase susceptibility to hyperthermia-induced seizures in developing rats. Epilepsia. 1996;37:902-910. 73. Scantlebury MH et al. Febrile seizures in the predisposed brain: a new model of temporal lobe epilepsy. Ann Neurol. 2005;58:41-49. 74. Bender RA et al. Mossy fiber plasticity and enhanced hippocampal excitability, without hippocampal cell loss or altered neurogenesis, in an animal model of prolonged febrile seizures. Hippocampus. 2003;13:357-370. C et al. Temporal lobe epilepsy after experimental prolonged febrile seizures: prospective 75. Dube analysis. Brain. 2006;129:911-922. 76. Jiang W et al. The neuropathology of hyperthermic seizures in the rat. Epilepsia. 1999;40:5-19. 77. Baram TZ et al. Is neuronal death required for seizure-induced epileptogenesis in the immature brain? Prog Brain Res. 2002;135:365-375. 78. Chen K et al. Febrile seizures in the developing brain result in persistent modification of neuronal excitability in limbic circuits. Nat Med. 1999;5:888-894.

25

26

THE EPILEPSIES 3 79. Chen K et al. Persistently modified h-channels after complex febrile seizures convert the seizureinduced enhancement of inhibition to hyperexcitability. Nat Med. 2001;7:331-337. 80. Magee JC. Dendritic lh normalizes temporal summation in hippocampal CA1 neurons. Nat Neurosci. 1999;2:508-514. 81. Robinson RB, Siegelbaum SA. Hyperpolarization-activated cation currents: from molecules to physiological function. Annu Rev Physiol. 2003;65:453-480. 82. Santoro B, Baram TZ. The multiple personalities of h-channels. Trends Neurosci. 2003;26:550-554. 83. Brewster AL et al. Developmental febrile seizures modulate hippocampal gene expression of hyperpolarization-activated channels in an isoform- and cell-specific manner. J Neurosci. 2002;22:4591-4599. 84. Brewster AL et al. Formation of heteromeric hyperpolarization-activated cyclic nucleotide-gated (HCN) channels in the hippocampus is regulated by developmental seizures. Neurobiol Dis. 2005;19:200-207. 85. Bender RA et al. Enhanced expression of a specific hyperpolarization-activated cyclic nucleotidegated cation channel (HCN) in surviving dentate gyrus granule cells of human and experimental epileptic hippocampus. J Neurosci. 2003;23:6826-6836. 86. Rosman PN. Evaluation of the child with febrile seizures. In Baram TZ, Shinnar S, eds. Febrile Seizures. Academic Press, 2002, 266-271. 87. American Academy of Pediatrics Provisional Committee on Quality Improvement, Subcommittee on Febrile Seizures. Practice parameter: the neurodiagnostic evaluation of the child with a first simple febrile seizure. Pediatrics. 1996;97:769-775. 88. Knudsen FU. Febrile seizurestreatment and outcome. Brain Dev. 1996;18:438-449. 89. Knudsen FU. Febrile seizures: treatment and prognosis. Epilepsia. 2000;41:2-9. 90. Baumann RJ, Duffner PK. Treatment of children with simple febrile seizures: the AAP practice parameter. American Academy of Pediatrics. Pediatr Neurol. 2000;23:11-17. 91. Berg AT et al. An EEG should not be obtained routinely after first unprovoked seizure in childhood. Neurology. 2000;55:898-899. 92. Hirtz D, Berg A, Bettis D, et al, for the Quality Standards Subcommittee of the American Academy of Neurology; Practice Committee of the Child Neurology Society. Practice parameter: treatment of the child with a first unprovoked seizure. Report of the Quality Standards Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society. Neurology. 2003;60:166-175. 93. VanLandingham KE et al. Magnetic resonance imaging evidence of hippocampal injury after prolonged focal febrile convulsions. Ann Neurol. 1998;43:413-426. 94. Frantzen E et al. Longitudinal EEG and clinical study of children with febrile convulsions. Electroencephalogr Clin Neurophysiol. 1968;24:197-212. 95. Stores G. When does an EEG contribute to the management of febrile seizures? Arch Dis Child. 1991;66:554-557. 96. American Academy of Pediatrics: Committee on Quality Improvement, Subcommittee on Febrile Seizures. Practice parameter. Long-term treatment of the child with simple febrile seizures. Pediatrics. 1999;103:1307-1309. 97. Autret E et al. Double-blind, randomized trial of diazepam versus placebo for prevention of recurrence of febrile seizures. J Pediatr. 1990;117:490-494. 98. Rosman NP et al. A controlled trial of diazepam administered during febrile illnesses to prevent recurrence of febrile seizures. N Engl J Med. 1993;329:79-84. 99. Uhari M et al. Effect of acetaminophen and of low intermittent doses of diazepam on prevention of recurrences of febrile seizures. J Pediatr. 1995;126:991-995. 100. Berg AT. Diazepam to prevent febrile seizures. N Engl J Med. 1993;329:2033-2034. 101. Rossi LN et al. Short-term prophylaxis of febrile convulsions by oral diazepam. Acta Paediatr. 1993;82:99. 102. ODell C et al. Rectal diazepam gel in the home management of seizures in children. Pediatr Neurol. 2005;33:166-172. 103. Lahat E et al. Comparison of intranasal midazolam with intravenous diazepam for treating febrile seizures in children: prospective randomised study. BMJ. 2000;321:83-86. 104. Scott RC et al. Intranasal midazolam for treating febrile seizures in children. Buccal midazolam should be preferred to nasal midazolam. BMJ. 2001;322:107.

THE EPILEPSIES 3

Mechanisms of Action of Levetiracetam and Newer SV2A Ligands


HENRIK KLITGAARD  ALAIN MATAGNE

Introduction Levetiracetam Epilepsy pharmacology Mechanism of action Brivaracetam

Epilepsy pharmacology Mechanism of action Seletracetam Epilepsy pharmacology Mechanism of action

Introduction
Levetiracetam (LEV; ucb L059; (S)-a-ethyl-2-oxo-pyrrolidine acetamide; Figure 3-1) is the (S)-enantiomer of the ethyl analog of the nootropic drug piracetam. Random screening in audiogenic susceptible mice initially showed that both systemic and central administration of LEV protect against seizure activity in these animals, whereas the R-enantiomer (ucb L060) and main metabolite (ucb L057) of LEV were inactive.1 This suggests that the parent compound mediates an antiepileptic effect by a central action. However, the therapeutic utility of this observation was challenged by other findings, which showed LEV to be inactive in the two conventional screening models in rodents for antiepileptic drugs (AEDs), the maximal electroshock and pentylenetetrazol seizure tests.2 Despite these conflicting results, UCB Pharma initiated a number of small, open-label studies in patients with treatment refractory seizures. The outcome suggested LEV to be an effective and well-tolerated add-on therapy devoid of clinically relevant pharmacokinetic interactions. This triggered a clinical development program resulting in approval of LEV as add-on treatment of adult patients with partial onset seizures by the FDA in 1999 and the EMEA in 2000.3 Both agencies later extended this approval to include children, and more recent approvals include adjunctive treatment of both myoclonic seizures in patients with juvenile myoclonic epilepsy and primary generalized tonic clonic seizures in patients with idiopathic generalized epilepsy and monotherapy treatment of

27

28

THE EPILEPSIES 3

LEVETIRACETAM BRIVARACETAM

SELETRACETAM F F

N O

O NH2

N O

O NH2

N O

O NH2

Figure 31 Chemical structures of levetiracetam, brivaracetam, and seletracetam.

patients with new onset epilepsy (by EMEA only).3 Taken together, these studies demonstrate LEVs efficacy as a broad-spectrum add-on therapy as well as a monotherapy treatment for epilepsy. More than 1 million patients had been treated with LEV by 2006, and by 2007 the drug had become the most prescribed new AED for the treatment of epilepsy.3 Together with its novel pharmacological profile and unique mechanisms of action, this triggered significant drug discovery activities at UCB Pharma targeting the identification of LEV analogs with improved mechanistic and antiepileptic properties. The first successful outcome of these efforts was the discovery of brivaracetam (BRV) and seletracetam (SEL) (Figure 3-1). These two clinical AED candidates currently undergo phase III and II studies, respectively, as add-on treatment of drug-refractory adult patients with partial onset seizures. This chapter will briefly describe the epilepsy pharmacology and review the mechanisms of action of levetiracetam, brivaracetam, and seletracetam.

Levetiracetam
EPILEPSY PHARMACOLOGY The lack of activity of LEV in the maximal electroshock and pentylenetetrazol seizure test appears to reflect a general absence of anticonvulsant activity in rodents in acute seizure tests employing either maximal electroshocks or administration of CD97 doses of chemoconvulsants.2 This contrasts a significant ability of LEV to suppress seizures in animals with an acquired, chronic epilepsy, as revealed by LEVs potent seizure protection in a number of different kindling models, or in various genetic animal models of epilepsy.4 These results reveal a selective, broad-spectrum action of LEV in animal models of epilepsy that mimics both partial and generalized seizures in mana profile that it does not share with any other AED. Several studies have shown LEVs remarkable ability to counteract kindling acquisition, induced by either PTZ administration to mice1 or electrical stimulation of amygdala in rats.5 Two independent experiments in the latter model showed that LEV permanently abolishes the kindling-induced increase in afterdischarge duration, even after cessation of treatment.5,6 This persistent effect against kindling acquisition after prolonged treatment distinguishes LEV from other AEDs7 and suggests that it does not simply mask the expression of kindled seizures through seizure suppression, but potentially possesses antiepileptogenic properties. Systemic administration of high doses of LEV to rodents only induces minor sedative and ataxic effects.2 Psychomimetic behaviors are absent2 and cognitive

3  Mechanisms of Action of Levetiracetam and Newer SV2A Ligands

29

performance unaltered.8 Combined with the potent seizure suppression by LEV in animal models of epilepsy, this results in a very high separation between doses inducing seizure protection and CNS-related adverse effects.2 The findings summarized earlier indicate that the pharmacological properties of LEV in animal models of seizures and epilepsy are unique and distinguish it from all other AEDs.4 LEV is the only AED to reveal an absence of anticonvulsant activity in conventional screening tests. This contrasts broad-spectrum seizure suppression and kindling inhibition with a wide safety margin in animal models of acquired and genetic epilepsy. This novel preclinical profile has nourished the desire to determine LEVs mechanism of action. MECHANISM OF ACTION Electrophysiological Properties A vast number of both in vitro and in vivo electrophysiological studies in rodents has consistently shown an absence of effect of LEV on normal neuronal responses and neuronal characteristics.9 This contrasts several other studies reporting that LEV exerts a preferential action against hypersynchronization of epileptiform activity. It has been observed that LEV differs from other AEDs by its ability to reduce increases in the amplitude of evoked population spikes, reflecting hypersynchronization, in rat hippocampal slices expressing epileptiform activity due to perfusion with high K+/low Ca2+.10 Further exploration with simultaneous dual extra- and intracellular recordings in the same model showed that LEV was the only AED to decrease the number of population spikes per extracellular response, without altering the number of action potentials per intracellular burst.11 These results suggest that LEV may counteract the transition from interictal to ictal activity. This could explain both its absence of activity against acute seizures induced by immediate, ictal activity in normal animals as well as its selective seizure protection, and low induction of adverse effects in animals with acquired and genetic epilepsy.2 Conventional AED Mechanisms The antiepileptic action of AEDs traditionally relates to a primary action on one or more of three main mechanisms. These consist of facilitation of GABAA/BZ receptors; inhibition of voltage-gated Na+ channels, and inhibition of low voltage-gated (T-type) Ca2+ channels. Numerous studies have failed to show a direct interaction of LEV with any of these three mechanisms.9 This confirms that LEVs unique profile in epilepsy pharmacology must reflect a novel mechanism of action. Other Mechanisms It has repeatedly been observed that LEV possesses an ability to inhibit AMPA-gated currents in rat hippocampal and cortical neurons.12,13 This effect only becomes significant at concentrations above therapeutic relevance and can therefore not be expected to contribute to LEVs antiepileptic action. However, other studies conducted at therapeutically relevant concentrations have discovered two novel cellular effects that probably contribute to LEVs unique mechanism of action.

30

THE EPILEPSIES 3

TABLE 31

Levetiracetam, Brivaracetam, and Seletracetam: Mechanisms of Action


Levetiracetam Brivaracetam Seletracetam

SV2A (pKi) Reversal of inhibition by Zn2+ on GABA-gated currents (20 mM) in hippocampal neurons Reversal of inhibition by Zn2+ on glycine-gated currents (100 mM) in spinal cord neurons HVA Ca2+ channel current (IC50 value) Na+ channel current (IC50 value)

6.1 Reversal (301000 mM) Reversal (0.1100 mM) 13.9 mM Inactive up to 1 mM

7.1 Reversal (1100 mM) Reversal (3-100 mM)

7.1 No effect (up to 100 mM) Total reversal (from 10 pM)

Inactive up to 1 mM 0.27 mM 7 mM Inactive up to 1 mM

This table was compiled by data reported for LEV9,14,17,18,22; BRV29,30,31,32; SEL36,37,38; and UCB data on file. pKi: -log of the unlabeled drug equilibrium constant; IC50: inhibitory half-maximal concentration.

LEV was shown to differ from other AEDs by an ability to reverse the inhibition of Zn2+ and b-carbolines of both GABA-gated currents in rat hippocampal and dentate granule neurons and glycine-gated currents in spinal cord neurons (Table 3-1).14 This distinguishes LEV from other AEDs and associates it with a potentially novel mechanism, in particular when viewed in the context of the sprouted mossy fiber/ Zn2+-sensitive GABAA receptor hypothesis.15 Indeed, it has been proposed that epileptogenesis involves alterations in both the subunit composition of the GABAA receptor as well as mossy fiber sprouting from dentate granule cells with Zn2+ containing terminals. This is believed to induce a vicious circle of disinhibition in hippocampus resulting in epileptic dischargesa condition that LEV may counteract. LEV was also been shown to reduce high-voltage gated Ca2+ currents.16,17 This effect appears to relate to a selective modulation of N-type Ca2+ channels.18 Furthermore, several studies have reported on LEVs ability to reduce both ryanodine and IP3-receptor mediated Ca2+ release from the endoplasmic reticulum.19,20 Taken together, these results suggest a novel effect of LEV on Ca2+ transients by a dual action on both N-type Ca2+ channels, incorporated in the plasma membrane, as well as on intraneuronal Ca2+ stores. LEVs ability to oppose Zn2+ inhibition of GABA- and glycine-gated currents and on 2+ Ca transients are only of a modest magnitude. Thus, although novel, these mechanisms cannot constitute the primary antiepileptic mechanism of action of LEV. This appears, instead, to relate to LEVs binding to the synaptic vesicle protein 2A. Synaptic Vesicle Protein 2A LEV (10 mM) has not been found to displace radioligands specific for a variety of receptors, ion channel proteins, reuptake sites, and second messenger systems.21 This contrasts to the observation of a specific binding site for 3H-LEV in rat brain

3  Mechanisms of Action of Levetiracetam and Newer SV2A Ligands

31

membranes, which has become known as the Levetiracetam Binding Site (LBS).21 LEV was found to bind saturably, reversibly, and stereospecifically to LBS. A strong correlation also existed between the affinity of a series of LEV analogs to the LBS and their seizure protection in epilepsy models.21 This suggests an important functional role for LBS in the antiepileptic mechanism of LEV and provided a strong rationale to identify its molecular nature. Approximately 10 years after the discovery of LBS, it was finally documented that synaptic vesicle protein 2A (SV2A) is the molecular correlate to LBS.22 SV2 is a 12-transmembrane protein incorporated in the membrane of synaptic vesicles, present in the presynaptic terminal. It exists in three isoforms, SV2A, SV2B, and SV2C, of which SV2A is the most widely distributed in the brain and also present on many neuroendocrine cells.23 The role of SV2 in the modulation of synaptic events remains obscure, but it is presumed to exert a modulatory effect on maturation and/or fusion of vesicles with the plasma membrane of the presynaptic terminal.23,24 Further evidence that the SV2A isoform has an impact on neurotransmission is derived from studies on animals lacking SV2A. These have shown that SV2A homozygous knockout (KO) mice express a lethal seizure phenotype,23 and that SV2A heterozygous KO mice reveal accelerated kindling acquisition.25 This supports the theory that SV2A has an important role in the control of vesicle exocytosis and may be involved in the pathophysiology of epilepsy. LEV has been documented to bind to SV2A expressed in fibroblasts (Table 3-1) but reveals no significant binding to SV2B or SV2C.22 Binding of 3H-LEV to brain membranes and purified synaptic vesicles from SV2A KO mice was absent, further supporting that SV2A is the molecular correlate of LBS (Figure 3-2). A strong correlation for LEV analogs was confirmed between their affinity for SV2A and LBS as well as between their affinity for SV2A and seizure protection in animal models of epilepsy (Figure 3-3). No other AED tested, up to 100 mM, revealed any significant affinity for SV2A. These results clearly prove that SV2A is the binding site of LEV and represents the novel and primary mechanism of action of this AED. However, LEV only possesses a moderate affinity for SV2A (Table 3-1). Together with the correlation between SV2A affinity and seizure protection, this provided a strong rationale to pursue a drug discovery program aimed at identifying high-affinity SV2A ligands with antiepileptic properties potentially superior to LEV. In this quest, approximately 12,000 compounds were screened in vitro for SV2A binding affinity, 900 compounds were examined for seizure protection in audiogenic susceptible mice, and 30 compounds were characterized more widely in animal models of seizures and epilepsy. As a first outcome of these efforts, brivaracetam (BRV; ucb 34714; (2S)-2-[(4R)-2-oxo-4-propylpyrrolidin-1-yl] butanamide; Figure 3-1) and seletracetam (SEL; ucb 44212; (2S)-2-[(4S)-4-(2,2-difluorovinyl)-2-oxo-pyrrolidin-1-yl]-butyramide; Figure 3-1) were discovered.

Brivaracetam
EPILEPSY PHARMACOLOGY BRV differs from both LEV and SEL by inducing seizure protection, albeit at relatively high doses, against maximal electroshock and pentylenetetrazol seizures in

32

THE EPILEPSIES 3

SV2-Cross Reactive MAb WT AKO BKO A/BKO

SV2A-specific Polyclonal WT AKO BKO A/BKO

A
[3H]ucb 30889 bound (DPM/Assay) 1000

500

0 WT

SV2A -/-

SV2B-/-

SV2A/B -/SV2A KO P2 LP1

H SV2A

P2

WT LP1

LP2

LP2

C synaptophysin
[3H]ucb 30889 bound (DPM/Assay) 5000 4000 3000 2000 1000 0
H P1 P2 LP 2 LP 1 W T KO W KO W LP 2 H KO P2 KO

D
Figure 32

Binding of the tritiated levetiracetam analog [3H]ucb 30889 to WT and KO brain membranes and synaptic vesicles. A, Western blot of brain membranes from WT and homozygous KO mice probed with an antiSV2 monoclonal antibody (cross-reactive to all isoforms) or with an anti-SV2A-specific polyclonal antibody. Lanes 1, WT; lanes 2, SV2A-/- KO; lanes 3, SV2B-/- KO; lanes 4, SV2A-/-/B-/- double KO. B, Binding of [3H]ucb 30889 to brain membranes from WT, SV2A-/-, SV2B-/-, and SV2A-/-/SV2B-/KO mice. Binding is observed only to membranes from animals expressing SV2A. , [3H]ucb 30889 alone; , [3H]ucb 30889 plus 1 mM levetiracetam. Error bars are the SD of experiments performed with five WT brains and four KO brains, with three replicates within each experiment. C, Purification of synaptic vesicles enriched for the synaptic vesicle proteins and levetiracetam binding. Shown are blots of mouse brain homogenate (H), crude synaptosomes (P2), plasma and heavy membranes (LP1), and synaptic vesicles (LP2) (2 mg of each fraction) that were probed for the synaptic vesicle proteins SV2A (Upper) and synaptophysin (Lower). The synaptic vesicle fraction from WT animals displays enrichment of both synaptic vesicle proteins and LEV-binding proteins, whereas material from SV2A KOs shows enrichment of synaptophysin only. D, Binding to the different fractions using [3H]ucb 30889 shows significant binding only to the WT LP2 fraction, containing SV2A-rich synaptic vesicles. Shown are [3H]ucb 30889 alone (open bars) and [3H]ucb 30889 plus 1 mM levetiracetam (filled bars). Shown are representative examples of two experiments. Error bars are the SD of two replicates. (Reprinted with permission from Lynch, B et al. The synaptic vesicle protein SV2A is the binding site for the antiepileptic drug levetiracetam. PNAS. 2004;101(26):9861-9866, ! 2004 National Academy of Sciences, U.S.A.)

3  Mechanisms of Action of Levetiracetam and Newer SV2A Ligands

33

r2=0.84

Correlation between binding affinity and antiseizure potency of levetiracetam analogs. Correlation of binding of a series of levetiracetam-related compounds to human SV2A assayed in transiently transfected COS-7 cells (pIC50s measured by using the tritiated levetiracetam analog [3H]ucb 30889) and of antiseizure potencies shown as the -log ED50s (pED50s) in the mouse audiogenic seizure model. (Reprinted with permission from Lynch, B et al. The synaptic vesicle protein SV2A is the binding site for the antiepileptic drug levetiracetam. PNAS. 2004;101(26):98619866, ! 2004 National Academy of Sciences, U.S.A.)

Figure 33

plC50hSV2A

4 2 3 4 5 6 7 pED50 Audiogenic seizures in mice

mice (Table 3-2).26 In kindling models, BRV shows a more potent seizure protection from LEV and a more complete seizure suppression than both LEV and SEL (Table 3-2).26 The latter is reflected by a superior ability to reduce both motor seizure severity as well as afterdischarge duration in fully amygdala-kindled rats (Table 3-2). BRV is also associated with a more potent seizure protection than LEV in genetic animal models of epilepsy, like audiogenic susceptible mice, and reveal a more complete suppression of cortical spike-and-wave discharges than LEV in an experimental model of absence epilepsy, the Genetic Absence Epilepsy Rat from Strasbourg (GAERS) (Table 3-2).26 Intravenous bolus administration of BRV has also been assessed for its ability to interfere with the process of self-sustaining status epilepticus (SSSE). An SSSE was induced in rats by perforant path stimulation. The cumulative duration of seizure activity was reduced dose-dependently to 11% and 8% of controls at 20 and 300 mg/kg, respectively.27 This compares favorably to the 35% and 15% obtained with 200 mg/kg of LEV and 10 mg/kg of diazepam, respectively.27 Kindling acquisition, induced by corneal stimulation in mice, was counteracted to the same extent by pretreatment with LEV (1.754 mg/kg i.p.) and BRV at 10-fold lower doses (0.216.8 mg/kg i.p.).26 However, continuation of corneal stimulation following cessation of treatment showed a significantly more persistent inhibition of the kindling process by BRV than by LEV. BRV is devoid of any psychomimetic reactions in amygdala-kindled rats and demonstrates a higher separation than LEV between doses inducing adverse effects and seizure protection in that model. BRV also possesses a high margin between these two parameters in corneally kindled mice, despite a separation inferior to LEV.26 Taken together, the profile of BRV in animal models of seizures and epilepsy suggests that it may represent a well-tolerated broad-spectrum agent for the symptomatic treatment of epilepsy in humans. BRV may provide a more potent and efficacious seizure protection than LEV and SEL against partial seizures and status epilepticus and possess a promising potential in animal models to also affect the course of the epileptic disease process.

34

THE EPILEPSIES 3

TABLE 32

Levetiracetam, Brivaracetam, and Seletracetam: Seizure Protection in Animal Models of Seizures and Epilepsy
Levetiracetam Brivaracetam Seletracetam

Acute Seizure Tests in Normal Animals Maximal electroshock seizures (mice ED50, mg/kg i.p.) Pentylenetetrazol seizures (mice, ED50, mg/kg i.p.) Fully Kindled Animals Corneally kindled mice (ED50, mg/kg, i.p.) Hippocampal kindled rats (MAD, mg/kg, p.o.) Amygdala kindled rats. (Motor seizure severity, MAD, mg/kg i.p.) (% reduction of afterdischarge duration at highest tested dose) Genetic Models of Chronic Epilepsy Audiogenic susceptible mice (ED50, mg/kg, i.p.) Genetic absence epilepsy rats from Strasbourg (MAD, mg/kg, i.p.) (Seizure suppression at highest tested dose)

>540 >540 7.3 54 170 13% 30 5.4 Partial

113 30 1.2 0.2 21.2 69% 2.4 6.8 Complete

>232 >232 0.7 0.02 74.3 5% 0.2 0.2 Complete

This table was compiled by data reported for LEV1,2,4; BRV26; SEL33; and UCB data on file. ED50: an effective dose protecting 50% of the animals against the convulsive endpoint; MAD: the minimal active dose providing significant protection against the convulsive endpoint; i.p.: intraperitoneal; p.o.: per os (oral).

MECHANISM OF ACTION BRV has been reported to possess a more potent and efficacious action than LEV against epileptiform responses in rat hippocampal slices expressing epileptiform activity due to perfusion with high K+/low Ca2+.28 BRV (3.2 mM) was more potent and active against repetitive firing than LEV (32 mM) and also inhibited spontaneous field bursting, an epileptiform marker resistant to the action of LEV. BRV (10 mM) was not observed to induce any significant displacement of radioligands specific for various receptors, ion channel proteins, reuptake sites, and second messenger systems. Instead, BRV possesses a high and selective affinity for SV2A (Table 3-1) and does not reveal any significant affinity for SV2B or SV2C.29 Patch clamp studies in cultured hippocampal neurons have shown that BRV is devoid of any direct interaction with excitatory and inhibitory neurotransmission, with the exception of a weak and minor inhibition of the NMDA receptor current.30

3  Mechanisms of Action of Levetiracetam and Newer SV2A Ligands

35

BRV is also devoid of effect on low- (T-type) and high-voltage-gated Ca2+ currents, the latter being different from the action of both LEV and SEL (Table 3-1).31 Another study has shown BRV to be without impact on voltage-gated K+ currents in cultured mice hippocampal neurons (UCB data on file). In contrast, BRV has a potent ability, like LEV, to reverse the inhibitory action of Zn2+ on both GABA- and glycine-gated current in hippocampal and spinal cord neurons, respectively (Table 3-1).30 Interestingly, BRV was also shown to produce a concentration-dependent inhibition of voltage-gated Na+ currents, recorded in rat cortical neurons in culture, with an IC50 value of 7 mM and a maximal inhibition of approximately 65% appearing from a concentration of 30 mM (Table 3-1).31 The experimental data accumulated regarding BRVs mechanism of action suggest that its primary antiepileptic mechanism relates to selective, high-affinity binding with SV2A, superior to LEV. Its ability to reverse Zn2+ inhibition of GABA- and glycine-gated currents may also contribute. However, a more important contribution probably derives from its ability to inhibit voltage-gated Na+ currents, an effect that markedly distinguishes BRV from both LEV and SEL.

Seletracetam
EPILEPSY PHARMACOLOGY SEL resembles LEV, but differs from BRV, by its absence of activity against maximal electroshock and pentylenetetrazol seizures in mice (Table 3-2).33 In kindling models, mimicking partial epilepsy in man, SEL shows a very potent protection against secondarily generalized seizure activity, superior to both LEV and BRV, but only reveals modest activity against the focal/partial seizure activity (Table 3-2). The latter is illustrated by its inability to reduce epileptiform afterdischarges in supramaximally stimulated amygdala-kindled rats (Table 3-2). In genetic animal models, mimicking primary generalized epilepsy in man, SEL also reveals a very potent suppression of seizure activity in audiogenic susceptible mice and of cortical spike-and-wave discharges in GAERS ratsan action superior to both LEV and BRV (Table 3-2).33 In the latter model, SEL is also superior to LEV in its ability to induce complete suppression of the spike-and-wave discharges. SEL is devoid of any psychomimetic reactions in amygdala-kindled rats and reveals an exceptionally high CNS tolerability.33 Previous studies with LEV revealed that it possesses a very high separation between doses inducing significant CNS-related adverse effects, measured by impairment of performance in the rotarod test, and seizure protection when compared with reference AEDs. The ratio between these two parameters for LEV in corneally kindled mice and GAERS rats was 148 and 235, respectively.2 For relevant reference AEDs, they were between 2-21 and 2-5,2 respectively. The similar numbers for SEL are 1048 and 3075, respectively.33 Taken together, the data generated with SEL in animal models of seizures and epilepsy suggest that it may possess an outstanding tolerability and a broad spectrum potential for the symptomatic treatment of epilepsy in humans. Compared to both LEV and BRV, SEL appears to provide a particular benefit by a very potent suppression of both primary- and secondary-generalized seizures.

36

THE EPILEPSIES 3

MECHANISM OF ACTION SEL (3.2 mM) has been reported to induce a more potent suppression than LEV (32 mM) of epileptiform responses in rat hippocampal slices expressing epileptiform activity due to perfusion with high K+/low Ca2+.34 In that model, LEV has been reported to differ from other AEDs by its significant ability to reduce increases in the amplitude of population spikes. Interestingly, SEL has been observed to possess a more potent and complete effect than both LEV and BRV on this measure of hypersynchronization.35 SEL (10 mM) was not observed to induce any significant displacement of radioligands specific for various receptors, ion channel proteins, reuptake sites, and second messenger systems. Instead, SEL possesses a high and selective affinity for SV2A (Table 3-1) and does not reveal any significant affinity for SV2B and SV2C (UCB data on file). Patch clamp studies in cultured hippocampal neurons have shown that SEL is devoid of any direct interaction with excitatory and inhibitory neurotransmission.36 SEL is also without effect on voltage-gated K+ currents in cultured mice hippocampal neurons (UCB data on file) and differs from BRV by an absence of effect on voltage-gated Na+ currents in cultured rat cortical neurons (Table 3-1).37 Like LEV and BRV, SEL is also devoid of effect on low-voltage-gated (T-type) Ca2+ currents (UCB data on file) but is distinct from BRV by a more potent ability than LEV to inhibit high-voltage-gated Ca2+ currents (Table 3-1).38 The ability of LEV and BRV to oppose the inhibitory action of Zn2+ on both GABA- and glycine-gated currents potentially contributes to their mechanism of action. Interestingly, SEL reveals a different profile by a very potent and selective effect only against the inhibition by Zn2+ on glycine-gated currents (Table 3-1).36 The experimental data accumulated regarding SELs mechanism of action support that it shares with BRV a primary antiepileptic mechanism related to a selective, high-affinity binding with SV2A, superior to LEV. Its selective ability to reverse Zn2+ inhibition of glycine-gated currents may also contribute. However, a more important contribution probably derives from its ability to inhibit high-voltage-gated Ca2+ currents, an effect that markedly distinguishes SEL from BRV.

Conclusion
The novel profile of LEV in epilepsy pharmacology together with its affinity for a unique brain binding site provided a strong rationale to search for the molecular nature of its primary mechanism of action. This finally led to the revelation that SV2A is the molecular correlate to LBS, a discovery that enriched the current understanding of the molecular mechanisms of the epilepsy pathophysiology and identified a new validated target class for AED drug discovery. Brivaracetam and seletracetam represent the first successful outcome of efforts focused on generating optimized, high-affinity SV2A ligands. They are both superior to LEV with respect to their affinity for SV2A and their ability to provide seizure protection in vivo. But they differ significantly, in particular by their effect on other relevant antiepileptic mechanisms, a finding further supported by their distinct profile in in vitro and in vivo models of epilepsy. Taken together, this highlights

3  Mechanisms of Action of Levetiracetam and Newer SV2A Ligands

37

the promise that both may represent an important addition to the existing armamentarium of AEDs available for the treatment of epilepsy. REFERENCES
1. Gower A, Noyer M, Verloes R, Gobert J, Wu lfert E. ucb L059, a novel anti-convulsant drug: pharmacological profile in animals. Eur J Pharmacol. 1992;222:193-203. 2. Klitgaard H, Matagne A, Gobert J, Wu lfert E. Evidence for a unique profile of levetiracetam in rodent models of seizures and epilepsy. Eur J Pharmacol. 1998;353:191-206. 3. Klitgaard H, Verdru P. Levetiracetam: the first SV2A ligand for the treatment of epilepsy. Expert Opin on Drug Discov. 2007;2:1537-1545. 4. Klitgaard H. Levetiracetam: the preclinical profile of a new class of antiepileptic drugs? Epilepsia. 2001;42(suppl 4):13-18. 5. Lo scher W, Ho nack D, Rundfeldt C. Antiepileptogenic effects of the novel anticonvulsant levetiracetam (ucb L059) in the kindling model of temporal lobe epilepsy. J Pharmacol Exp Ther. 1998;284:474-479. 6. Stratton S, Large C, Cox B, Davies G, Hagan R. Effects of lamotrigine and levetiracetam on seizure development in a rat amygdala kindling model. Epilepsy Res. 2003;53:95-106. 7. Silver JM, Shin C, McNamara JO. Antiepileptogenic effects of conventional anticonvulsants in the kindling model of epilepsy. Ann Neurol. 1991;29:356-363. 8. Lamberty Y, Margineanu D, Klitgaard H. Absence of negative impact of levetiracetam on cognitive function and memory in normal and amygdala-kindled rats. Epilepsy Behav. 2000;333-342. 9. Margineanu DG, Klitgaard H. Levetiracetam mechanisms of action. In Levy RH, Mattson RH, Meldrum BS, Perucca E, eds. Antiepileptic Drugs. 5th ed. Philadelphia: Lippincott Williams & Wilkins; 2002;419-427. 10. Margineanu DG, Klitgaard H. Inhibition of neuronal hypersynchrony in vitro differentiates levetiracetam from classical antiepileptic drugs. Pharmacol Res. 2000;42(4):281-285. 11. Niespodziany I, Klitgaard H, Margineanu DG. Desynchronizing effect of levetiracetam on epileptiform responses in rat hippocampal slices. Neuroreport. 2003;14(9):1273-1276. 12. Hans G, Nguyen L, Rocher V, Belachew S, Moonen, Matagne A, Klitgaard H. Levetiracetam: no relevant effect on ionotropic excitatory glutamate receptors. Epilepsia. 2000;41(Suppl 7):35. 13. Carunchio I, Pieri M, Ciotti MT, Albo F, Zona C. Modulation of AMPA receptors in cultured cortical neurons induced by the antiepileptic drug levetiracetam. Epilepsia. 2007;48(4):654-662. 14. Rigo JM, Hans G, Nguyen L, et al. The anti-epileptic drug levetiracetam reverses the inhibition by negative allosteric modulators of neuronal GABA- and glycine-gated currents. Br J Pharmacol. 2002;136(5):659-672. 15. Coulter DA. Mossy fiber zinc and temporal lobe epilepsy: pathological association with altered epileptic gamma-aminobutyric acid A receptors in dentate granule cells. Epilepsia. 2000;41 (Suppl 6):S96-S99. 16. Niespodziany I, Klitgaard H, Margineanu DG. Levetiracetam inhibits the high-voltage-activated Ca(2+) current in pyramidal neurons of rat hippocampal slices. Neurosci Lett. 2001;306(1-2):5-8. 17. Pisani A, Bonsi P, Martella G, et al. Intracellular calcium increase in epileptiform activity: modulation by levetiracetam and lamotrigine. Epilepsia. 2004;45(7):719-728. 18. Lukyanetz EA, Shkryl VM, Kostyuk PG. Selective blockade of N-type calcium channels by levetiracetam. Epilepsia. 2002;43(1):9-18. ngehagen M, Margineanu DG, Ben-Menachem E, Ronnback L, Hansson E, Klitgaard H. 19. A Levetiracetam reduces caffeine-induced Ca2+ transients and epileptiform potentials in hippocampal neurons. Neuroreport. 2003;14(3):471-475. 20. Cataldi M, Lariccia B, Secondo A, Di Renzo G, Annunziato L. The antiepileptic drug levetiracetam decreases the inositol 1,4,5-trisphosphate-dependent [Ca2+] increase induced by ATP and bradykinin in PC12 cells. J Pharmacol Exp Ther. 2005;313(2):720-730. 21. Noyer M, Gillard M, Matagne A, Henichart JP, Wu lfert E. The novel antiepileptic drug levetiracetam (ucb L059) appears to act via a specific binding site in CNS membranes. Eur J Pharmacol. 1995;286(2):137-146. 22. Lynch BA, Lambeng N, Nocka K, et al. The synaptic vesicle protein SV2A is the binding site for the antiepileptic drug levetiracetam. Proc Natl Acad Sci U S A. 2004;101(26):9861-9866. 23. Crowder KM, Gunther JM, Jones TA, et al. Abnormal neurotransmission in mice lacking synaptic vesicle protein 2A (SV2A). Proc Natl Acad Sci U S A. 1999;96(26):15268-15273.

38

THE EPILEPSIES 3 24. Janz R, Goda Y, Geppert M, Missler M, Sudhof TC. SV2A and SV2B function as redundant Ca2+ regulators in neurotransmitter release. Neuron. 1999;24(4):1003-1016. 25. Leclercq K, Dassesse D, Lambeng N, Klitgaard H, Matagne A. Unaltered seizure susceptibility contrasts accelerated kindling acquisition in heterozygous SV2A KO mice. Program No. 82.7 2006, Abstract viewer/itinerary planner. Washington DC, Society for Neurosciences 2006. On-line. 26. Matagne A, Kenda B, Michel PH, Klitgaard H. ucb 34714, a new pyrrolidone derivative: comparison with levetiracetam in animal models of chronic epilepsy in vivo. Epilepsia. 2003;44(suppl 9):261. 27. Wasterlain C, Suchomelova L, Matagne A, Klitgaard H, Mazarati A, Shinmei S, Baldwin R. Brivaracetam is a potent anticonvulsant in experimental status epilepticus. Epilepsia. 2005;46 (suppl 8):219. 28. Margineanu D, Kenda B, Michel Ph, Matagne A, Klitgaard H. ucb 34714, a new pyrrolidone derivative: comparison with levetiracetam in hippocampal slice epilepsy models in vitro. Epilepsia. 2003;44(suppl 9):261. 29. Gillard M, Fuks B, Lambeng N, Chatelain P, Matagne A. Binding characteristics of brivaracetam, a novel antiepileptic drug candidate. Program No.3.186. 2007 Abstract Viewer. Philadelphia, PA: American Epilepsy Society. 30. Rigo JM, Nguyen L, Hans G, et al. ucb 34714: effect on inhibitory and excitatory neurotransmission. Epilepsia. 2004;45(suppl 3):56. 31. Kostyuk PG, Lukyanetz EA, Klitgaard H, Margineanu DG. ucb 34714, a new pyrrolidone derivative, without impact on high- and low-voltage activated calcium currents in rat isolated neurons. Epilepsia. 2004;45(suppl 7):141-142. 32. Zona C, Pieri M, Klitgaard H, Margineanu DG. ucb 34714, a new pyrrolidone derivative, inhibits Na+ currents in rat cortical neurons in culture. Epilepsia. 2004;45(suppl 7):146. 33. Matagne A, Margineanu DG, Michel Ph, Kenda B, Klitgaard H. Seletracetam (ucb 44212), a new pyrrolidone derivative, reveals potent activity in in vitro and in vivo models of epilepsy. J Neurol Sciences. 2005;238(suppl 1):S133. 34. Margineanu DG, Michel Ph, Kenda B, Matagne A, Klitgaard H. Seletracetam (ucb 44212), a new pyrrolidone derivative, inhibits epileptiform responses in hippocampal slices in vitro. Epilepsia. 2005;46(suppl 6):121. 35. Margineanu D, Klitgaard H. The novel SV2A ligands brivaracetam and seletracetam manifest different effects against the epileptiform markers of field potentials in a high K+-low Ca2+ rat hippocampal slice model. Epilepsia. 2006;47(suppl 3):74-75. 36. Rigo JM, Nguyen L, Hans G, et al. Seletracetam (ucb 44212): effect on inhibitory and excitatory neurotransmission. Epilepsia. 2005;46(suppl 8):110. 37. Zona C, Niespodziany I, Pieri M, Klitgaard H, Margineanu DG. Seletracetam (ucb 44212), a new pyrrolidone derivative, lacks effect on Na+ currents in rat brain neurons in vitro. Epilepsia. 2005;46(suppl 8):116. 38. Pisani A, Bonsi P, Martella G, Cuomo D, Klitgaard H, Margineanu DG. Seletracetam (ucb 44212), a new pyrrolidone derivative, inhibits high-voltage-activated Ca2+ currents and intracellular [Ca2+] increase in rat cortical neurons in vitro. Epilepsia. 2005;46(suppl 8):119.

THE EPILEPSIES 3

Long-Term Effects of Seizures on Brain Structure and Function


HOWARD P. GOODKIN  EDWARD H. BERTRAM

Neonatal Seizures The Childhood Epilepsies: Good versus Bad Benign Rolandic Epilepsy West Syndrome Febrile Seizures and Mesial Temporal Lobe Epilepsy

Mesial Temporal Lobe Epilepsy Laboratory Insights Other Effects of Seizures Seizures Beget Seizures?

Perhaps one of the most common questions asked by patients with epilepsy or by parents of children with epilepsy is, What are the seizures doing to the brain? Implicit in the question is a concern that the seizures are having a negative effect. The number of questions that derive from this one are almost endless: Are the seizures causing brain damage? Will my childs cognitive development be impaired by having seizures? Will the seizures cause more seizures? The answers to these and other related questions are as varied as the causes of epilepsy, the types of seizure, and the ages of the patient when the seizures occur. In short, there is no single, simple answer. The physiology associated with seizures varies from a short run of rhythmic spike and wave activity to prolonged tonic discharges. Some types of epilepsy are seemingly benign and self-limited, whereas others are inexorably progressive. Seizures can occur in a neonatal brain with years of development to come or in aging brains that have suffered the ravages of time and are hanging by their fingertips over the abyss of oblivion. In approaching these questions, we therefore have to consider, among other things, which kind of seizure, what cause, and what type of brain. Many of the answers are often colored by the physicians interpretation of imperfect data, an interpretation that can be further clouded by attempts to extrapolate data across seizure types, developmental state, and species. In this chapter, we will provide an overview of some of the relevant data. However, we cannot offer clarity where none exists. Thus, readers will find answers that contain enough caveats to cast doubt on any attempt at certainty. Our goal is to provide a

39

40

THE EPILEPSIES 3

perspective on the complexities involved in answering your patients and their families legitimate questions. Before we enter the discussion regarding the effect of seizures on the brain, it is essential that we differentiate among the different types of epilepsy and seizures because it is clear that the pathophysiology of the many syndromes and situations are quite varied in both cause and consequence. These are important distinctions to make, as one of the problems that we encounter in determining the effect of seizures on the brain is that many of the causes involve preexisting abnormalities or destructive injuries. Thus, claiming any change found as a potential consequence of seizure activity cannot be supported. The primary distinction to be made between seizure types is between status epilepticus (SE) and intermittent spontaneous seizures. Even though SE is often caused by events such as stroke, infection, or trauma, there is ample evidence, especially from the laboratory, that after a duration of as short as 30 minutes, the seizure activity itself can induce neuronal loss.1,2 However, it is likely that even in the general area of SE, there are types that cause injury and types that do not. Convulsive SE or SE associated with high-frequency discharges on electroencephalogram (EEG) clearly can cause damage, even when the high-intensity physical convulsions are controlled.3,4 Absence SE or spontaneous episodes of prolonged partial SE associated with lower frequency discharges (3 c/s spike and wave or rhythmic delta/theta activity) may not be, as there appear to be no long-term consequences to these episodes. The other major grouping of seizures are intermittent spontaneous, which is the type that will be encountered most often in clinical practice. These seizures range from benign neonatal convulsions to infantile spasms (West syndrome) to absence seizures and partial seizures associated with well-defined focal pathology with many types and etiologies. Each seizure type has a distinctive physiology on EEG, a variety of underlying pathologies, and a defined natural history ranging from those that stop on their own with no identified poor outcomes, to intractable epilepsy associated with severe neurological and cognitive problems. The extent to which these different outcomes are the consequence of continued seizures and the contribution is from the underlying neuropathology, which is at times quite severe, is unclear. In our attempt to clarify the problem, we will draw on a variety of animal studies as well as a number of clinical observations. In each instance, keep in mind that no finding can be universally applicable to all seizure types and all forms of epilepsy. To answer the question for any given human syndrome, it will ultimately be necessary to extrapolate, often imperfectly, from what is known about the syndromes natural history and pathophysiology, as well as relevant experimental data, recognizing that not all laboratory observations are applicable to a particular (or in some cases any) human syndrome. Answering the questions for any one patient will require careful consideration and judgment. We will begin the discussion of these issues with some clinical examples, which will illustrate the complexity of differentiating cause from consequence as well as the inappropriateness of applying a universal rule to all epilepsies, which are a wondrously varied collection of syndromes.

Neonatal Seizures
The developing brain is predisposed to seizures,5,6 and seizures are common in the neonate with an estimated incidence of 1 to 6 per 1000 live births.711 The term

4  Long-Term Effects of Seizures on Brain Structure and Function

41

neonatal seizures encompasses seizures of diverse etiologies ranging from acute symptomatic causes (e.g., hypoxia-ischemia, intracranial hemorrhage, infections, and metabolic derangements such as hypocalcemia) to remote symptomatic causes (e.g., cerebral dysgenesis and inborn errors of metabolism) to idiopathic, genetically determined epilepsies (e.g., early myoclonic encephalopathy of infancy and familial neonatal convulsions). As a group, the outcome for children with neonatal seizures, although improving, is poor.12 The mortality rate for neonates with seizures is high, with values ranging from 15 to 30%. In survivors, approximately 30% will develop neurological sequelae that include epilepsy, mental retardation, and cerebral palsy. Those most at risk for mortality and neurological morbidity are those neonates with persistent, difficult-totreat seizures.1317 It has been proposed that these seizures produce neurological injury as the result of energy depletion and cerebral metabolism dysfunction.18 However, because these prolonged, persistent seizures tend to be observed in those neonates with the most severe CNS involvement and injury (e.g., severe hypoxic ischemic encephalopathy), it is difficult to distinguish between the secondary damage, if any, imposed by the seizures and the primary injury imposed on the brain by the underlying etiology. There are good examples in which neonatal seizures tend to be associated with a more favorable outcome. One example is the genetically determined benign familial neonatal convulsions, an autosomal dominantly inherited epilepsy syndrome resulting from mutations in voltage-gated potassium channel. These convulsions are brief events commencing as tonic posturing with apnea and other autonomic features, which can evolve to clonic movements in otherwise healthy neonates. These convulsions typically commence during the first week of life, can be observed multiple times per day, and may be difficult to treat. In most cases, the seizures will remit by 16 months of age. Although there is a increased risk of febrile and afebrile seizures later in life, psychomotor development tends to be normal.19 For those with psychomotor delay, it is likely that this outcome is the result of genetic factors and not the seizures.20 In summary, caveats abound when attempting to answer the question of whether seizures damage the neonatal brain. Although all recognize that the answer to this question is important, clinical data are limited.21 It is assumed that few would argue against the concept that in some circumstances seizures in the neonatal brain impose additional injury, but not all would agree on the circumstances or the extent of the injury, especially as some neonatal seizures have a benign outcome.

The Childhood Epilepsies: Good versus Bad


For many children with epilepsy, the outcome is good. The seizures remit and may not recur, even after medication is discontinued. Although cognitive and behavioral disorders may be observed transiently during the time period when the epilepsy was most active and required treatment with an antiepileptic medication, these children have no apparent evidence of a significant, long-term disability resulting from their epilepsy. In contrast, a group of children with epilepsy will have an associated psychomotor slowing or regression that persists even in the absence of seizures. In some of these children, data suggest that the seizures or the interictal

42

THE EPILEPSIES 3

TABLE 41
Benign Epilepsies

Epilepsy Syndromes with Good or Bad Outcomes

Benign familial neonatal convulsions Benign myoclonic epilepsy of infancy Childhood absence epilepsy Juvenile absence epilepsy Benign rolandic epilepsy Panayiotopoulos syndrome Juvenile myoclonic epilepsy
Malignant Epilepsies

Ohtahara syndrome Early myoclonic encephalopathy West syndrome Dravet syndrome Lennox-Gastaut syndrome Landau-Kleffner syndrome

EEG abnormalities contributed to the regression. Most childhood epilepsies fall somewhere between these two extremes (Table 4-1). The epilepsies in which the outcome tends to be good, the so-called benign epilepsies, include benign rolandic epilepsy, childhood absence epilepsy, and juvenile absence epilepsy. The epilepsies in which a potentially long-lasting psychomotor slowing or regression is observed include Ohtahara syndrome (early infantile epileptic encephalopathy with suppression burst), West syndrome (infantile spasms), Dravet syndrome (severe myoclonic epilepsy of infancy), Lennox-Gastaut syndrome, and Landau-Kleffner syndrome. For purposes of illustration, we will review two of these disorders, the evidence supporting the absence of a long-term effect of the epilepsy in one (benign rolandic epilepsy), and evidence supporting the long-term effect of the epilepsy in the other (West syndrome). BENIGN ROLANDIC EPILEPSY Benign rolandic epilepsy (BRE; also known as benign childhood epilepsy with centrotemporal spikes or BECT) is a common childhood epilepsy accounting for 15 to 25% of all childhood epilepsies. This syndrome is named for the sleep-activated interictal centrotemporal spikes with a tangential dipole observed on EEG and its characteristic partial seizures of the face and tongue that, especially during sleep, may secondarily generalize. It is most commonly observed in boys between the ages of 3 and 13 years with a peak at 9 to 10 years. Given that the seizure frequency associated with this syndrome is low,22 and that the seizures most commonly occur during sleep, the majority of children with this syndrome do not require antiepileptic therapy.23 In addition, these children outgrow the epilepsy, and the interictal EEG abnormality resolves during their teenage years.24 That the seizures remit and the rate of epilepsy in these children in

4  Long-Term Effects of Seizures on Brain Structure and Function

43

adulthood is no greater than that in the general population24 is in contrast to the adage that seizures beget seizures. Although BRE is considered a benign epilepsy, cognitive and behavioral difficulties may be observed.2528 However, for the most part, these difficulties are transient and appear to be related to the severity of the interictal discharges and not the result of a permanent injury to the brain.29 Indeed, as a group, the developmental outcome for these children, including those with cognitive and behavioral difficulties, whether treated or not, is good. A potential alternative explanation is that an underlying genetic predisposition underlies both the epilepsy and cognitive dysfunction. WEST SYNDROME West syndrome is a cryptogenic or symptomatic generalized epilepsy syndrome composed of infantile spasms, a hypsarrhythmic EEG, and mental retardation. The spasms, which tend to occur in clusters, are characterized by a brief, massive flexion, extension, or mixed contraction of the axial musculature coincident with an electrodecrement on the EEG. The incidence for this syndrome ranges from 3 to 4.5 per 10,000 live births, and the majority of children present during the first year of life.3032 Causes of West syndrome are multiple and include the neurocutaneous syndromes (e.g., tuberous sclerosis complex), brain malformation and tumors, inborn disorders of metabolism, and genetic disorders, as well as acquired injuries from fetal infections, trauma, and hypoxic-ischemic injury. As a group, the long-term prognosis is poor and includes medically refractory epilepsy, cognitive impairments ranging from learning disorders to mental retardation, and autism. To a large extent, like neonatal seizures, the prognosis is dependent on the etiology.33 However, medical and surgical reports demonstrating improved developmental outcome in those children with the shortest epilepsy duration and prompt response to therapy suggest that the seizure activity itself may be detrimental. But once again, caution in interpretation is required, as those with a short treatment lag or who responded quickly to therapy or potentially those most amenable to surgical resection may be those who would have had a better developmental outcome, even in the absence of effective treatment.34

Febrile Seizures and Mesial Temporal Lobe Epilepsy


One of the areas of great uncertainty regarding the long-term consequences of seizures is the potential relationship between febrile seizures as a young child and the development of mesial temporal lobe epilepsy (MTLE) with hippocampal sclerosis/atrophy as an adult. Patients with MTLE are much more likely to have a history of febrile seizures as a child, and this association has raised the issue of whether febrile seizures may in some way cause the limbic pathology that eventually leads to MTLE. On the other hand, a number of epidemiological studies indicate that the risk for developing epilepsy (or any other negative consequence) following febrile seizures is not greater than for children who never experienced them. However, the studies have also revealed that febrile seizures are not a single entity. Some are classified as simple and last less than 5 minutes, and these do not carry any additional risk

44

THE EPILEPSIES 3

for a poor long-term outcome. Complex febrile seizures (focal ictal behavior or seizures lasting more than 30 minutes), especially those that are prolonged, have a much higher risk for poor long-term outcome and the development of chronic epilepsy. This latter association between complex febrile seizures (a minority of all febrile seizures) and long-term problems has raised the question of whether the poor outcomes are the result of the febrile seizures or whether the febrile seizures are a symptom of a preexisting abnormality that was likely to cause problems as the child grew up. There is no definitive answer at the moment, but some interim data have emerged from a multicenter FEBSTAT study headed by Dr. Shlomo Shinnar that suggest that both may be true in individual patients. Thus, some patients show evidence for a preexisting abnormality (e.g., hippocampal malrotation), and in others the history and follow-up studies suggest that the prolonged seizures result in focal temporal edema. This study is not yet complete, and, until the long-term follow-up data are available, all that can be said is that atypical febrile seizures are associated with poorer long-term outcomes.

Mesial Temporal Lobe Epilepsy


This syndrome has been the source of the greatest debates regarding the potential negative impact that seizures have on the brain, as it goes straight to the heart of the issue of whether uncontrolled (often single) seizures cause neuronal loss. This is the single area in which the thoughts of the epilepsy community are most divided. The opinions range from the conclusion that each single seizure causes a measurable neuronal loss to the position that no evidence exists for progressive neuronal loss as the consequence of any number of recurrent, spaced seizures, even thousands. The mesial temporal lobe epilepsy (MTLE) or chronic limbic (CLE) syndrome is associated with specific patterns of neuronal loss in the hippocampus and adjacent mesial temporal regions known as hippocampal or mesial temporal sclerosis.35 This particular syndrome has a well-established pattern of neuronal loss that can be quantified in vivo through magnetic resonance imaging (MRI) volume measurements36 or spectroscopy for N-acetyl-aspartase (NAA), a compound that is predominantly neuron specific.37 Anatomical studies have been variably interpreted to show evidence of significant progressive neuronal loss or to show minimal to no progressive loss over prolonged periods of time.38,39 In view of the strong opinions on both sides of the issue (progressive vs. no loss), what do the data really show? A number of clinical correlational studies use surgical or postmortem specimens to correlate the duration of epilepsy or, less commonly, total numbers of seizures (always difficult to estimate) in comparison to the severity of neuronal loss. Although there is not an absolute consensus, the most recent studies have suggested that there is not tremendous neuronal loss as a cumulative consequence of recurrent seizures.38,40 One study that examined tissue obtained at the time of surgery found a slight reduction of neuronal density in patients with a 30-year epilepsy history. However all patients had significant neuronal loss, even patients with 5-year histories and relatively few seizures. This finding suggested that the overwhelming majority of neuronal loss preceded the onset of seizures.38

4  Long-Term Effects of Seizures on Brain Structure and Function

45

Laboratory Insights
Animal studies may avoid the issues of cause versus consequence, as one is able to control the inducing lesion and evaluate the tissue at specified times after the induction of the causative lesion and after specified numbers of seizures. Although there is a lack of unanimity,41 the data tend to support the conclusion that much of the neuronal loss precedes the onset of epilepsy and is not a consequence of recurrent seizures.4245 In animals in which seizures were induced, the method used was direct electrical stimulation in the limbic system (the kindling model).46 As in the human studies there is some disagreement about the interpretation of the data, but overall the studies suggest that recurrent seizures do not cause progressive neuronal loss, even after 1000 induced seizures over a number of months.42,43 This observation does not mean that the seizures have no effect on the brain. They clearly activate glial cells and cause them to hypertrophy,45,47 It is just that, at least in this one model of intermittent seizures, neuronal loss is not a consequence of seizures. Is there a human parallel of induced seizures in animals with normal brains so that we might extrapolate from the models to the clinic regarding the potential effect of seizures on the brain? There is in one circumstance: electroconvulsive therapy (ECT). Concern has been raised that this therapy causes brain damage, but in recent follow-up studies using MRI volumetrics, no evidence was found for detectable brain injury as a consequence of ECT. Although most courses of ECT involve a small number of induced seizures and most patients undergo the therapy once, there is a report of a patient who had approximately 1200 stimulated seizures through ECT over the course of his life. On autopsy his brain showed no evidence of damage.48 Whether seizures are harmful to the developing brain has also been addressed in the laboratory in which seizures are induced in young animals through the use of chemoconvulsants such as the excitotoxin kainate acid, the cholinergic agonist pilocarpine, and the general anesthetic flurothyl. Some studies have demonstrated that repetitive seizures over several days starting soon after4953 or even a single episode of repetitive seizures over an approximately 3-hour period in rat pups can result in long-term changes in seizure susceptibility and cognitive function. Yet, in another study, a single prolonged seizure induced in young rats had no effect on later seizure susceptibility or performance on tests of visuospatial learning, memory, responses to novel environment, and emotionality.54 One of the primary lessons learned from laboratory studies is that the developing hippocampus, unlike the adult hippocampus, is resistant to gross neuronal injury secondary to seizures.5557 However, recurrent seizures in rat pups can induce synaptic rearrangement,58 a reduction in neurogenesis, and altered expression of neurotransmitter receptors. In addition, neuronal loss can be observed in other parts of the brain. Following severe prolonged seizures induced with a combination of lithium and pilocarpine, neuronal loss was observed in the central and lateral segments of the mediodorsal nucleus,59 and repetitive prolonged seizures may result in a change in the distribution of interneurons in the cortex. Although these studies are confounded by the use of excitotoxins, and the nature of the seizures varies by method of induction and the age of the animal, it is reasonable to conclude that the developing brain is less susceptible to seizure-induced injury, but not immune to it.

46

THE EPILEPSIES 3

The basis for the difference is unclear, but it likely includes differences in intrinsic cell properties as well as receptors, channels, and type of connectivity. Even though it is clear that injury can have a long-term effect on seizure susceptibility and cognitive function in these animal models, given the differences in the development of neural networks in the animal and human brain, extrapolation of the results from animal studies to the preterm and term infant must be done with caution. Therefore, although the clinical and animal studies suggest that seizures in the neonate may be harmful, in view of the severe injuries or abnormalities that cause these seizures, a definitive conclusion cannot be reached at this time.

Other Effects of Seizures


If seizures dont cause neuronal loss, what effects do they have? There is no question that seizures activate astrocytes.45,47 The activated glia gradually subside if seizures do not occur over several months.45 In addition to the activation of glia, clear changes can be observed in the physiology of neurons following seizures, but the nature of the change varies from region to region. Although there are clear changes in a multitude of channels and receptors in the neurons from epileptic animals and people, the limited available data suggest that many of these changes precede the onset of epilepsy, although one cannot exclude the possibility that some of the changes are a consequence of recurrent seizures. Many of the ion channels have not been studied to the same degree in kindled animals as they have in the poststatus epilepticus spontaneous epilepsy models. However, where they have been studied, very little overlap is seen in terms of changes in ion channel expression in epileptic and kindled animals, suggesting that the changes in epilepsy are part of the disorder as opposed to a consequence of the seizures. There are reports about A-type potassium channels and T-type calcium channels from kindled animals,60,61 as well as other calcium currents that do exhibit short- and long-term alterations.6265 Changes in the GABA receptor after just a few induced seizures have also been reported.66 A number of reports have demonstrated that the changes associated with kindling are quite different than the changes found in epilepsy. Although there are still many areas for study, the data to date suggest that the consequences of intermittent seizures are quite different and likely of less consequence than the changes that cause the epilepsy (Table 4-2). Sprouting of mossy fibers in the dentate gyrus has been observed in MTLE as well as in the kindling models. These aberrant axons, which originate in the granule cells of the dentate gyrus and normally project to the pyramidal cells of CA3 in the

TABLE 42

Changes Associated with Induced Seizures

Activation of astrocytes Altered expression of voltage-gated ion channels (sodium, potassium, calcium) Altered expression of receptors (GABA, NMDA) Synaptic rearrangement (minimal mossy fiber spouting in dentate gyrus)

4  Long-Term Effects of Seizures on Brain Structure and Function

47

Figure 41 Comparison of hippocampal mossy fiber staining in kindled and epileptic rats. A and C are from kindled rats with relatively few (A, 100) and many (C, 870) stimulated seizures. Neither animal was observed to have spontaneous seizures. B and D are from rats that became epileptic after an episode of status epilepticus. B was from a rat with 124 documented spontaneous seizures and D from a rat with over 1500 documented spontaneous seizures. Arrows point to mossy fiber staining in the hilus of the dentate gyrus, arrowheads to aberrant mossy fiber sprouting in the inner molecular layer of the dentate gyrus. In A there is no appreciable sprouting, and in C there is a minimal amount. In comparison, the epileptic rats (B and D) have very dense mossy fiber sprouting, and the rat with the greatest number of seizures (D) has much greater sprouting than the rat with relatively few (B). In addition, the hilus in the kindled animals is much larger than the atrophied hilus in the epileptic rats. This latter difference has been shown in a number of studies. This figure demonstrates that seizures have a much less significant effect on the structure of the hippocampus than the status epilepticus that induced the epilepsy in the other rats.

hippocampus, turn back and move into the inner molecular layer in MTLE, where they form excitatory synapses.67,68 Although they have been described in kindling, the degree of mossy fiber sprouting in kindling is far less than in models of MTLE44 (Figure 4-1). This observation suggests that although seizures may cause some synaptic reorganization, it is a relatively minor change compared to the changes that occur in MTLE. Changes in the physiology of the dentate gyrus have been described in kindling, so that these changes, along with others, are likely having an effect on the local system physiology. The net effect of these changes with regard to changes in the larger limbic system excitability is still unknown.

48

THE EPILEPSIES 3

Seizures Beget Seizures?


This phrase has been around for over a century.69 It implies that each seizure induces changes that increase the probability that another seizure will occur. If true, Seizures beget seizures also implies that there is some urgency in suppressing seizures to improve long-term outcomes. The observation that seizures can induce changes in channels and receptors could support that hypothesis, but what is the clinical experience regarding the natural history of the epilepsies? As noted earlier, there is a significant variability in the ultimate outcomes of the different disorders. Some, such as the Lennox-Gastaut syndrome, are difficult to control and are ultimately associated with a long history of intractable seizures. Others, such as rolandic or benign epilepsy of centrotemporal spikes, enter into spontaneous remission regardless of whether the seizures were controlled or not. Absence epilepsy has a high rate of spontaneous remission (although not complete). These observations would suggest that early treatment may not have an effect on outcome, as it appears that the outcome regarding ultimate remission is independent of treatment. So how did the concept arise? Likely from the observation that some patients who had more than a few seizures were likely to have more, and if they had more seizures, they might be more difficult to control. Although the data have never really been analyzed in this way, it is very possible that the patients who are having more seizures before they start treatment may be having them in a shorter period of time and that their frequent seizures are a sign of a less-benign syndrome. Animal studies do not completely clarify the issue, unfortunately. Kindling, during which seizures are induced repeatedly over days and weeks (and sometimes months) in rats by direct brain stimulation, results in longer and more behaviorally severe seizures (that ultimately plateau), but do not consistently result in spontaneous seizures. Although some investigators do report the development of spontaneous seizures after many (150 to 300 stimulated seizures),70,71 it is not a universal finding across labs. Primate kindling shows some evidence that genetics may play a role, as there are some primate species that neither kindle well nor ever develop spontaneous seizures.72,73 As a consequence, we are left with the conclusion that although in some situations after many seizures the seizure activity itself may cause changes that result in spontaneous seizures, little evidence supports the notion of seizures begetting seizures. It is far more likely that the natural history of the particular epilepsy syndrome reflects the natural history of the underlying cause. The issue of kindling to spontaneous seizures is relevant for the theoretical question of secondary epileptogenesis in people. It is a hypothetical process in which seizure activity that initiates in one region of the brain spreads to other regions. Over time the repeated seizures could potentially cause changes in other regions that eventually result in another seizure focus that could act independently. One of the primary reasons for developing this hypothesis was the gradual appearance of independent spikes away from the presumed original (or primary) spike focus. It was also used to explain the presence of a second focus and seizure type in patients who had a single seizure type for many years before the second became active. The first focus was often associated with an obvious abnormality, whereas the second was not. This phenomenon has been called into question because of several developments, not the least of which was improved neuroimaging. This allowed the visualization of subtle cortical changes that indicate the existence of multiple

4  Long-Term Effects of Seizures on Brain Structure and Function

49

abnormalities that could be independent foci for seizures. In addition, unless there is a well-documented second seizure type that is independent of the primary seizure, the existence of multifocal spikes does not prove the existence of multiple seizure foci, as spikes are not always good indicators of a seizure focus. Rather, they are a sign of a potential to have seizures. Thus, in humans, although we cannot completely exclude the eventual kindling of a secondary focus, at this time there is no convincing evidence that the original primary seizures kindle a secondary and eventually independent seizure focus.

Summary
A recurring theme in this chapter has been the difficulty in assessing what kind of effect seizures are having on the brain. Other than status epilepticus associated with prolonged high-frequency discharges such as in convulsive or limbic status epilepticus, there is no strong evidence that seizures cause significant brain damage, with the potential exception (and yet to be demonstrated definitively) of several of the severe childhood epileptic encephalopathies such as West syndrome. Clinical experience suggests that the evolution of a particular epilepsy (self-limited, static, or progressive) is more related to the natural history of the underlying neurological disorder rather than the cumulative consequence of recurrent seizures. There are limited data regarding the subtler effects of seizures on neurons and glial cells in humans. Animal data clearly demonstrate that seizures can have more subtle effects such as glial activation and hypertrophy as well as on the expression of subsets of receptors and channels. The functional effect of these changes, when placed in the context of the overall system, is unknown, as some of the changes may be attempts by the system to reestablish equilibrium, whereas others may support seizure generation or spread such as occurs in the process of kindling. What is clear from animal studies is that the effects of seizures on the brain are quite different from the underlying changes that are associated with epilepsy and are physiologically much less significant. As we consider the questions that we asked at the beginning of the chapter, how can we answer our patients questions? It is clear that there is no simple answer, but the current data, as imperfect as they are, generally suggest that individual seizures themselves are not severely harmful, although the underlying cause of the seizure might be. For this reason, more emphasis should be placed on determining the cause. In advising patients, for the moment we will continue to rely on the likely natural history of their particular epilepsy as it relates to the causative syndrome. In addition, one should continue to encourage patients to seek prompt medical assistance should seizures last too long and threaten to become a continuous status epilepticus, as there is no controversy that status epilepticus can cause significant damage. REFERENCES
1. Lothman E. The biochemical basis and pathophysiology of status epilepticus. Neurology. 1990;40:13. 2. Fountain NB, Lothman EW. Pathophysiology of status epilepticus. J Clin Neurophysiol. 1995;12:326. 3. Blennow G, Brierley JB, Meldrum BS, et al. Epileptic brain damage: the role of systemic factors that modify cerebral energy metabolism. Brain. 1978;101:687. 4. Nevander G. Ingvar M. Auer R, et al. Status epilepticus in well-oxygenated rats causes neuronal necrosis. Ann Neurol. 1985;18:281-290.

50

THE EPILEPSIES 3 5. Wong M. Advances in the pathophysiology of developmental epilepsies. Semin Pediatr Neurol. 2005;12:72. 6. Ben-Ari Y, Holmes GL. Effects of seizures on developmental processes in the immature brain. Lancet Neurol. 2006;5:1055. 7. Hauser WA, Kurland LT. The epidemiology of epilepsy in Rochester, Minnesota, 1935 through 1967. Epilepsia. 1975;16:1. 8. Ellenberg JH, Hirtz DG, Nelson KB. Age at onset of seizures in young children. Ann Neurol. 1984;15:127. 9. Lanska MJ, Lanska DJ, Baumann RJ, Kryscio RJ. A population-based study of neonatal seizures in Fayette County, Kentucky. Neurology. 1995;45:724. 10. Ronen GM, Penney S, Andrews W. The epidemiology of clinical neonatal seizures in Newfoundland: a population-based study. J Pediatr. 1999;134:71. 11. Cowan LD. The epidemiology of the epilepsies in children. Mental Retard Dev Disabil Res Rev. 2002;8:171. 12. Ronen GM, Buckley D, Penney S, et al. Long-term prognosis in children with neonatal seizures: a population-based study. Neurology. 2007;69:1816. 13. Mellits ED, Holden KR, Freeman JM. Neonatal seizures. II. A multivariate analysis of factors associated with outcome. Pediatrics. 1982;70:177. 14. Legido A, Clancy RR, Berman PH. Neurologic outcome after electroencephalographically proven neonatal seizures. Pediatrics. 1991;88:583. 15. Ortibus EL, Sum JM, Hahn JS. Predictive value of EEG for outcome and epilepsy following neonatal seizures. Electroencephalogr Clin Neurophysiol. 1996;98:175. 16. Bye AM, Cunningham CA, Chee KY, Flanagan D. Outcome of neonates with electrographically identified seizures, or at risk of seizures. Pediatr Neurol. 1997;16:225. 17. Pisani F, Cerminara C, Fusco C, Sisti L. Neonatal status epilepticus vs recurrent neonatal seizures: clinical findings and outcome. Neurology. 2007;69:2177. 18. Miller SP, Weiss J, Barnwell A, et al. Seizure-associated brain injury in term newborns with perinatal asphyxia. Neurology. 2002;58:542. 19. Zonana J, Silvey K, Strimling B. Familial neonatal and infantile seizures: an autosomal-dominant disorder. Am J Med Genet. 1984;18:455. 20. Steinlein OK, Conrad C, Weidner B. Benign familial neonatal convulsions: always benign? Epilepsy Res. 2007;73:245. 21. Clancy RR. Summary proceedings from the neurology group on neonatal seizures. Pediatrics. 2006;117:S23. 22. Kramer U, Zelnik N, Lerman-Sagie T, Shahar E. Benign childhood epilepsy with centrotemporal spikes: clinical characteristics and identification of patients at risk for multiple seizures. J Child Neurol. 2002;17:17. 23. Peters JM, Camfield CS, Camfield PR. Population study of benign rolandic epilepsy: is treatment needed? Neurology. 2001;57:537. 24. Bouma PA, Bovenkerk AC, Westendorp RG, Brouwer OF. The course of benign partial epilepsy of childhood with centrotemporal spikes: a meta-analysis. Neurology. 1997;48:430. 25. Staden U, Isaacs E, Boyd SG, Brandl U, Neville BG. Language dysfunction in children with rolandic epilepsy. Neuropediatrics. 1998;29:242. 26. Ong HT, Wyllie E. Benign childhood epilepsy with centrotemporal spikes: is it always benign? Neurology. 2000;54:1182. 27. Baglietto MG, Battaglia FM, Nobili L, et al. Neuropsychological disorders related to interictal epileptic discharges during sleep in benign epilepsy of childhood with centrotemporal or rolandic spikes. Dev Med Child Neurol. 2001;43:407. 28. Massa R, de Saint-Martin A, Carcangiu R, et al. EEG criteria predictive of complicated evolution in idiopathic rolandic epilepsy. Neurology. 2001;57:1071. 29. Boxerman JL, Hawash K, Bali B, et al. Is rolandic epilepsy associated with abnormal findings on cranial MRI? Epilepsy Res. 2007;75:180. 30. Brna PM, Gordon KE, Dooley JM, et al. The epidemiology of infantile spasms. Can J Neurol Sci. 2002;28:309. 31. Luthvigsson P, Olafsson E, Sigurthardotir S, et al. Epidemiologic features of infantile spasms in Iceland. Epilepsia. 1994;35:802. 32. Sidenvall R, Eeg-Olofssn O. Epidemiology of infantile spasms in Sweden. Epilepsia. 1995;36:572. 33. Koo B, Hwang PA, Logan WJ. Infantile spasms: outcome and prognostic factors of cryptogenic and symptomatic groups. Neurology. 1993;43:2322.

4  Long-Term Effects of Seizures on Brain Structure and Function 34. Hrachovy RA, Glaze DG, Frost JD, Jr. A retrospective study of spontaneous remission and long-term outcome in patients with infantile spasms. Epilepsia. 1991;32:212. 35. Margerison JH, Corsellis JA. Epilepsy and the temporal lobes. A clinical, electroencephalographic and neuropathological study of the brain in epilepsy, with particular reference to the temporal lobes. Brain. 1966;89:499. 36. Quigg MS, Bertram EH, Jackson T, et al. Evidence for bilateral atrophy in unilateral mesial temporal lobe epilepsy. Epilepsia. 1997;38:588. 37. Suhy J, Laxer KD, Capizzano AA, et al. 1H MRSI predicts surgical outcome in MRI-negative temporal lobe epilepsy. Neurology. 2002;58:8213. 38. Mathern GW, Adelson PD, Cahan LD, et al. Hippocampal neuron damage in human epilepsy: Meyers hypothesis revisited. Prog in Brain Res. 2002;135:237. 39. Cendes F, Andermann F, Gloor P, et al. Atrophy of mesial structures in patients with temporal lobe epilepsy: cause or consequence of repeated seizures? Ann of Neurol. 1993;34:795. 40. Thom M, Zhou J, Martinian L, et al. Quantitative post-mortem study of the hippocampus in chronic epilepsy: seizures do not inevitably cause neuronal loss. Brain. 2005;128:1344. 41. Cavazos JE, Das I, Sutula TP. Neuronal loss induced in limbic pathways by kindling: evidence for induction of hippocampal sclerosis by repeated brief seizures. J Neurosci. 14:3106. 42. Bertram EH, Lothman EW, Lenn NJ. The hippocampus in experimental chronic epilepsy: a morphometric analysis. Ann Neurol. 1990;27:43. 43. Bertram EH, Lothman EW. Morphometric effects of intermittent kindled seizures and limbic status epilepticus in the dentate gyrus of the rat. Brain Res. 1993;603:25. 44. Mathern GW, Bertram III EH, Babb TL, et al. In contrast to kindled seizures, the frequency of spontaneous epilepsy in the limbic status model correlates with greater aberrant fascia dentata excitatory and inhibitory axon sprouting, and increased staining for N-methyl-D-aspartate, AMPA and GABAA receptors. Neurosci. 1997;77:1003-1019. 45. Adams B, Von Ling E, Vaccarella L, et al. Time course for kindling-induced changes in the hilar area of the dentate gyrus: reactive gliosis as a potential mechanism. Brain Res. 1998;804:331. 46. Goddard GV, McIntyre DC, Leech CK. A permanent change in brain function resulting from daily electrical stimulation. Experimental Neurol. 1969;25:295. 47. Torre ER, Lothman E, Steward O. Glial response to neuronal activity: GFAP-mRNA and protein levels are transiently increased in the hippocampus after seizures. Brain Res. 1993;631:256. 48. Lippman S, Manshadi M, Wehry M, et al. 1250 electroconvulsive treatments without evidence of brain injury. Brit Jour of Psychiat. 1985;147:203. 49. Holmes GL, Gairsa JL, Chevassus-Au-Louis N, et al. Consequences of neonatal seizures in the rat: morphological and behavioral effects. Ann Neurol. 1989;44:845-857. 50. Huang L, Cilio MR, Silveira DC, et al. Long-term effects of neonatal seizures: a behavioral, electrophysiological, and histological study. Brain Res Dev Brain Res. 1999;118:99. 51. de Rogalski Landrot I, Minokoshi M, Silveira DC, et al. Recurrent neonatal seizures: relationship of pathology to the electroencephalogram and cognition. Brain Res Dev Brain Res. 2001;129:27. 52. Sogawa Y, Monokoshi M, Silveira DC, et al. Timing of cognitive deficits following neonatal seizures: relationship to histological changes in the hippocampus. Brain Res Dev Brain Res. 2001;131:73. 53. Sayin U, Sutula TP, Stafstrom CE. Seizures in the developing brain cause adverse long-term effects on spatial learning and anxiety. Epilepsia. 2004;45:1539. 54. Stafstrom CE, Chronopoulos A, Thurber S, et al. Age-dependent cognitive and behavioral deficits after kainic acid seizures. Epilepsia. 1993;34:420. 55. Nitecka L, Tremblay E, Charton G, et al. Maturation of kainic acid seizure-brain damage syndrome in the rat. II. Histopathological sequelae. Neuroscience. 1984;13:1073. 56. Cavalheiro EA, Silva DF, Turski WA, et al. The susceptibility of rats to pilocarpine-induced seizures is age-dependent. Brain Res. 1987;465:43. 57. Sperber EF, Haas KZ, Romero MT, et al. Flurothyl status epilepticus in developing rats: behavioral, electrographic histological and electrophysiological studies. Brain Res Dev Brain Res. 1999;116:59. 58. Holmes GL, Sarkisian M, Ben-Ari Y, et al. Mossy fiber sprouting after recurrent seizures during early development in rats. J Comp Neurol. 1999;404:537. 59. Kubova H, Druga R, Lukasiuk K, et al. Status epilepticus causes necrotic damage in the mediodorsal nucleus of the thalamus in immature rats. J Neurosci. 2001;21:3593. 60. Vreugdenhil M, Wadman WJ. Potassium currents in isolated CA1 neurons of the rat after kindling epileptogenesis. Neuroscience. 1995;66:805.

51

52

THE EPILEPSIES 3 61. Hendriksen H, Kamphius W, Lopes da Silva FH. Changes in voltage-dependent calcium channel alpha1-subunit mRNA levels in the kindling model of epileptogenesis. Brain Res Mol Brain Res. 1997;50:257. 62. Faas GC, Vreugdenhil M, Wadman WJ. Calcium currents in pyramidal CA1 neurons in vitro after kindling epileptogenesis in the hippocampus of the rat. Neuroscience. 1996;75:57. 63. Blalock EM, Chen K-C, Vanaman TC, et al. Epilepsy-induced decrease of L-type Ca2+ channel activity and coordinate regulation of subunit mRNA in single neurons of rat hippocampal zipper slices. Epilepsy Res. 2001;43:211. 64. Bernstein GM, Medonca A, Wadia J, et al. Kindling induces an asymmetric enhancement of N-type Ca2+ channel density in the dendritic fields of the rat hippocampus. Neurosci Lett. 1999;268:155. 65. Bernstein GM, Medonca A, Wadia J, et al. Kindling induces a long-term enhancement in the density of N-type calcium channels in the rat hippocampus. Neuroscience. 1999;94:1083. 66. Evans MS, Cady CJ, Disney KE, et al. Three brief epileptic seizures reduce inhibitory synaptic currents, GABAA currents, and GABAA-receptor subunits. Epilepsia. 2006;47:1655. 67. Tauck DL, Nadler JV. Evidence of functional mossy fiber sprouting in hippocampal formation of kainic acid-treated rats. J Neurosci. 1985;5:1016. 68. Sutula T, Cascino G, Cavazos J, et al. Mossy fiber synaptic reorganization in the epileptic human temporal lobe. Ann Neurol. 1989;26:321. 69. Gowers WR. Epilepsy and Other Chronic Convulsive Disorders: Their Causes, Symptoms and Treatment. London: J&A Churchill; 1881. 70. Pinel JPJ, Rovner LI. Experimental epileptogenesis: kindling-induced epilepsy in rats. Experimental Neurol. 1978;58:190. 71. Sayin U, Osting S, Hagen, et al. Spontaneous seizures and loss of axo-axonic and axo-somatic inhibition induced by repeated brief seizures in kindled rats. J Neurosci. 2003;23:2759. 72. Wada JA, Mizoguchi T, Osawa T. Secondarily generalized convulsive seizures induced by daily amygdaloid stimulation in rhesus monkeys. Neurology. 1978;28:1026. 73. Wada JA, Osawa T. Spontaneous recurrent seizure state induced by daily electrical amygdaloid stimulation in Senegalese baboons (Papio papio). Neurology. 1976;26:273.

THE EPILEPSIES 3

Dipole Source Modeling in Epilepsy: Contribution to Clinical Management


JOHN S. EBERSOLE

Introduction EEG Fields from Cortical Sources The Dipole as a Model of Cortical Sources Interpretation of Dipole Models

Clinical Studies Using Dipole Models Usefulness of EEG Dipole Modeling in Clinical Practice

Introduction
Of the principal clinical uses for electroencephalography (EEG) in the management of epilepsy, localization of the epileptogenic focus is perhaps second in importance only to diagnosis. This is particularly true for those patients with medically uncontrolled partial seizures who may be surgical candidates. Source localization by EEG analysis has a long history. Unfortunately, many of the traditional methods are based on principles that are at best simplistic and at worst scientifically unsound. With the advent of digital EEG and advanced computer techniques, localization of epileptogenic foci is easily transformed from an art form to a science. Dipole source modeling is an advanced method of analyzing and interpreting EEG data. It is based, however, on simple principles that need to be understood before the technique can be fully appreciated. In reality, the EEG is a time series of continually changing voltage fields of differing polarity and magnitude over the surface of the head. The EEG, as most know it, namely traces of voltage potential difference over time between electrodes, is simply a creation of measurement and display techniques. It is important to think of EEG as voltage fields and not simply as lines on a display. This is because the contours of these fields, the location and amplitude of negative and positive field maxima, convey all the information that is necessary for proper source localization. Traditional localization of spike and seizure foci by visual inspection of EEG is dependent on identifying certain features of the tracing. It was recognized early in

53

54

THE EPILEPSIES 3

the history of EEG that sharp negative potentials are the interictal hallmark of an epileptogenic process. Similarly, seizure potentials, particularly those of focal origin, tend to produce rhythmic negative potentials that also have a local maximum. Penwriting EEG machines were engineered to emphasize this negative field maximum by producing the so-called phase reversal deflections toward one another when using a linked bipolar montage. EEG machines could have just as easily been made to emphasize positive field maxima. However, this feature of epileptiform fields has been neglected until recently. Somewhat similarly, referential EEG montages are commonly analyzed by searching for the channel displaying the largest negative potential. Thus, in traditional EEG, the act of localizing epileptic foci is largely dependent on identifying the electrode recording the maximal negative potential. The basis for doing this rests on the assumption that the epileptic source lies under this electrode. With renewed appreciation for the biophysics of EEG generation, we now realize that this assumption is true only in limited cases in which the spike or seizure voltage field is purely radial in its orientation.1 Such fields are typically generated by the crowns of convexity cortex. However, many brain regions that are highly epileptogenic are not found on the cortical convexity. This includes the entire base of the brain, in particular orbitofrontal and basal temporal regions. These cortices produce EEG fields that are tangential to the head surface rather than radial. Similarly, epileptogenic foci in major fissures, such as Sylvian or interhemispheric, produce tangentially oriented EEG fields. Such sources are important to understand because little or no EEG potential is recorded directly above them; rather, the negative and positive EEG fields on the head are displaced on either side of their true location. In such a situation, considering the negative field maximum as the source location will result in false localization and even a false lateralization in certain situations (Figure 5-1).

EEG Fields from Cortical Sources


In the generation of any cortical EEG potential, be it epileptiform or normal, current sinks and sources are created on opposite ends of the palisaded pyramidal cells. By electromagnetic necessity, the extracellular space at one end of this cell layer is relatively negative, and the other end is relatively positive. This separation of charge creates a dipole. EEG potentials are characteristically dipolar in nature because of this generating mechanism. In epileptiform activity, the superficial laminae are often depolarized initially, leading to the prominence of negative potentials for epileptic spikes and seizure potentials. Of note, this is the opposite of normal sensory-evoked potentials, where depolarization of layer 4 results in passive hyperpolarization of the superficial laminae and a resultant surface-positive potential. Recurrent circuits within the cortex produce reverberating depolarization/hyperpolarization sequences. Thus many cortical potentials, including epileptic activity, exhibit recurrent cycling of negative and positive waveforms, such as in spikewave complexes. All cortical EEG potentials have a dipolar field configuration; however, given limited spatial sampling of scalp electrodes, we may not record both sides of this dipole field. This is particularly true for radial sources near the vertex that have a positive field maximum at the bottom of the head. Conversely, for basal brain sources only the positive field maximum may be recordable from

5  Dipole Source Modeling in Epilepsy

55

top

back 2

top

back

top

back

Figure 51

Three cortical sources are noted by color overlay on the MRIinterhemispheric (blue), lateral convexity (green), and temporal base (red). Scalp voltage fields produced by depolarization of these sources (1, 2, and 3, respectively) are depicted surrounding the MRI. Isopotential lines denote field amplitude; speckled field is negative in polarity. An equivalent current dipole for each voltage field is shown adjacent to the cortical source. Note that the orientation of the dipole is orthogonal to the net orientation of the source cortex, and the position of the dipole is deep to the cortical surface.

standard 10 to 20 electrode positions. However, for sources in the lateral aspect of the brain, both maxima are typically evident in traditional montages (see Figure 5-1). A three-dimensional line drawn between the negative and positive field maximum of a spike potential on the head has the same orientation as that of the pyramidal cells generating the potential. The actual cortical source may be geometrically complex, but this line represents the net orientation of those cells. Because pyramidal cells are orthogonal to the cortical surface, the orientation of an EEG field is also orthogonal to the source cortex. Theoretically, the center of the cortical source must lie along this 3-D line. Its location is proportionally nearer the field maximum with greater amplitude. Thus, in most cases sources are located nearer the higher amplitude negative field maximum. However, when the voltage field is tangential, the source location is equidistant between field maxima. Simply by inspecting the spike/seizure voltage fields over the head, one can gain considerable information about the likely source of these fields. Such visual analysis is essentially a form of

56

THE EPILEPSIES 3

source modeling. This same spatial information regarding voltage fields is the basic data used by all source modeling mathematical algorithms.

The Dipole as a Model of Cortical Sources


The most common source modeling technique used for clinical applications is the single equivalent current dipole.2 As noted earlier, cortical sources of EEG spikes or seizure rhythms produce voltage fields that are dipolar in nature. It makes sense, therefore, to use a model that is dipole based. However, the dipole used in modeling is a theoretical point source with a separation of charge, whereas the actual source of an EEG potential is a relatively large region of cortex that forms a dipole layer. Although the geometry of the actual dipole layer may be complex and comprised of several gyri and sulci, the voltage fields produced by this convoluted source will add or cancel one another in a linear way to produce a resultant simple dipolar field at the scalp. In fact, recent studies of simultaneous intracranial and scalp EEG have shown that a region of gyral cortex 10 or more square centimeters in area is necessary to produce an EEG potential that is distinguishable from the ongoing background rhythm.3 In fact, large and easily identified epileptiform potentials are often generated by cortical sources of 20 to 30 cm2 in size, thus encompassing a substantial sublobar region. Even though more complex multipolar geometry such as quadrapoles and octopoles may be produced by these large convoluted sources at the cortex, the dipolar field component dominates at the scalp. Because a point source dipole model attempts to explain scalp EEG fields produced by a large cortical patch, these dipole solutions are usually deep to the actual generating cortex in order to project a field of equivalent area (Figure 5-2). Physical laws of electromagnetic field theory state that the voltage fields on the surface of a spherical volume conductor can be accurately predicted if the location, orientation, and strength of a dipole source within the conductor are known.4,5 This is called the forward solution, and it is unique. However, what we attempt to do in clinical dipole modeling is the reverse. We measure the surface voltage field on the head and attempt to identify the source within this volume conductor. This is the inverse solution, and it is not unique because a number of different dipolar source configurations within the brain can produce the same surface field.68 Therefore, we have to use certain assumptions to minimize the possibilities, and the principal one is that the source is a single current dipole. Various algorithms exist for identifying an equivalent dipole source.811 Most of these algorithms use an iterativeminimizing approach, whereby an estimate of source location is made. The forward solution is performed, and the difference between the forward solution and the actual measured field is characterized. Subsequent random movements of the dipole model attempt to minimize this difference. When the smallest difference is obtained, the putative dipolar source has been identified. Dipole models display in three-dimensional terms the same information that a person can perceive by visually inspecting the voltage fields as explained earlier. In addition, these equivalent dipoles can be coregistered either with a head schematic or with an actual three-dimensional magnetic resonance imaging (MRI) to identify the putative source within the brain. With a head or brain schematic, it is possible to identify the most likely lobe or perhaps even sublobar area containing the source.

5  Dipole Source Modeling in Epilepsy

57

Figure 52 Major cortical sources and their equivalent dipoles for temporal lobe spikes/seizures are illustratedlateral cortex (top left), tip cortex (bottom left), and base cortex (top right). To the right of each source is the scalp voltage field produced by the source. Note that a three-dimensional line connecting the negative (blue) and positive (red) field maximum defines the orientation of the field and that of the equivalent dipole. Lateral temporal cortex dipoles are horizontal and radial; temporal tip dipoles are horizontal and tangential; temporal base dipoles are vertical and tangential. Note, too, that dipoles are deep to the cortical sources.

It is very tempting to coregister these data with three-dimensional MRIs, as is commonly done with other functional imaging techniques. The problem with doing so is that sources or colored blobs that are placed on a real brain image take on a seductive pseudorealism that in reality must be verified. In the case of EEG dipole models, sources in the cortical convexity are accurately localized. This is because even simple spherical head models used by source-modeling algorithms work well in this most spherical part of the head and brain. However, when sources were near the base of the brain, such as in the temporal lobe, systematic dipole location errors occurred for known sources.12,13 This error was typically in the vertical direction toward the vertex, and it could be as great as 2 or more centimeters in magnitude. A spherical head model was the reason for this error.1315 Such a model cannot be used to localize sources near nonspherical parts of the head, such as the base of the skull (Figure 5-3). Those surfaces that are important for the degradation of scalp EEG signals, namely the inner and outer layers of the skull and outer layer of the scalp, can be segmented with modern volumetric imaging. Using the shape of these surfaces, as determined from three-dimensional MRI and computed tomography (CT), a realistic head model can be achieved. In clinical practice, boundary element models (BEM) are most commonly used, particularly when localizing the source of potentials in nonspherical parts of the brain. The accuracy of correctly identifying temporal lobe sources, for example, is improved considerably with such models (Figure 5-3).1618 BEM head models can now be calculated rapidly, even on a regular PC, in a matter of minutes. Even without a patients individual MRI, the benefits of a realistic head model can be obtained by using one derived from standardized head images.19

58

THE EPILEPSIES 3

Figure 53

Fifty milliseconds of the same left temporal spike are depicted in moving dipole models at the top and bottom of the figure. A single dipole has been calculated for each of 11 sequential voltage field measurements at 5 msec intervals (red to green). Dipoles illustrated on the upper brain were calculated using a three-concentric-sphere head model. Dipoles on the lower brain were derived using a realistic head model. Note that both initial and late dipoles from the spherical head model are erroneously located outside the temporal lobe, whereas dipoles from the realistic head model remain in the temporal lobe.

5  Dipole Source Modeling in Epilepsy

59

Finite element head models are more complex and take into consideration differences in regional resistivity and anisotropy within the head and brain.2023 Accordingly, finite element models are much more difficult to compute. Regardless, it is important to use realistic head models whenever dipole models are coregistered in 3-D anatomy. Spatial resolution of the EEG sensors influences the accuracy of source localization afforded by dipole modeling. The typical international 10-20 electrode configuration coarsely samples only the top half of the head. Recording from the southern hemisphere is critical for characterizing sources at the base of the brain. Twenty-six EEG channels, including a subtemporal row of three electrodes on each side of the head, is probably a minimal configuration for clinical EEG dipole modeling.2 Electrode arrays that approximate the 10-10 system are better. In a study in which 128 channels of scalp EEG were progressively downsampled to 64-, 32-, and 21-channel recordings, Lantz et al.24 documented an improvement in source localization accuracy up to 64 channels. More electrodes provided little additional accuracy because the spatial frequency of dipolar scalp fields from spontaneous EEG is relatively low, and the sources of these fields are relatively large. INTERPRETATION OF DIPOLE MODELS It is clear from studies of simultaneous scalp and intracranial EEGs that the cortical sources of scalp spikes and seizure rhythms are quite large, namely 10 to 30 cm2.3 A point source dipole is thus a simple model of this extended cortical generator, and for it to be useful, it must be interpreted properly. EEG dipoles model major cortical surfaces, not individual gyri or sulci. Dipole location identifies only the center of the source region generating the greatest potential and certainly not its extent or the specific cortex involved. Because this pointlike model tries to explain the scalp field resulting from an extended source, the model dipole is typically located deep to actual source cortex. Accordingly, one should not be concerned about dipole models in white matter. It is dipole orientation that identifies the likely source cortex within the region.2527 Dipole orientation conveys the net orientation of the pyramidal cell generating the field, and this is orthogonal to the net orientation of the generator cortex. If there is no cortex in the region of a dipole model that has an appropriate orientation, the validity of the source solution should be questioned (Figure 5-2). Only initial epileptiform EEG activity may represent the spike source or seizure origin. Single dipole models may be misleading at the spike peak or after the early seconds of a seizure discharge. Source propagation is common, thus the temporal evolution of a dipole model is important. A dipole solution for each time point, the so-called moving dipole model, can portray propagation reasonably well if the propagation is unidirectional and until repolarization of the original source confounds the resultant voltage field (Figure 5-3). Several investigators have concluded that modeling the rising phase of the spike is more likely to represent the initial spike source2831 than modeling the spike peak. Obviously there is a trade-off because the earlier time points typically have lesser signal to noise (S/N), which may make modeling less accurate. Averaging closely similar spikes or sequential ictal waveforms can improve the S/N, which will provide a more confident solution.32

60

THE EPILEPSIES 3

Clinical Studies Using Dipole Models


Early investigations of major clinical import were that of Ebersole and Wade.33,34 They characterized the voltage topography of spikes from a series of patients with temporal lobe epilepsy who were surgical candidates. They noted two distinctive patterns of voltage fields. One had an inferior temporal negative maximum and a vertex positive maximum, whereas the other had a lateral temporal negative maximum and a contralateral temporal positive maximum. They termed the former a Type 1 spike field and the latter a Type 2 field. Constructing a three-dimensional line between these field maxima demonstrates that the Type 2 spike has a radial field orientation, and the Type 1 has more of a tangential orientation. A variety of clinical correlations, as well as simultaneous scalp and intracranial EEG investigations, suggested that the Type 2 field topography and its corresponding horizontal radial dipole model originated from a lateral temporal cortical source, whereas Type 1 topography and its corresponding more vertical tangential dipole model was associated with basomesial temporal sources. However, the Type 1 spike field was not thought to reflect hippocampal or amygdalar activity directly. It was shown that spikes confined to these structures did not generate scalp-recordable voltage fields due to the small source area and curved source shape, which favor voltage cancellation.35 Rather, it was the common and preferred propagation of this epileptiform activity from mesial structures into the entorhinal, fusiform, and temporal basal cortex that resulted in a generator of sufficient area to produce scalp EEG potentials.28,35,36 Subsequent investigation demonstrated that a rigid categorization of temporal spike topography was overly simplistic. Spike voltage fields commonly evolve over tens of milliseconds, which usually signifies a propagating spike source. A Type 1 pattern may evolve into a Type 2 pattern and vice versa. Spikes that originate in basal temporal cortex frequently propagate into temporal tip cortex by the time sufficient cortex is activated to produce a scalp EEG field. Temporal tip cortex has a net anterior-facing orientation. Accordingly, spike sources in this cortex produce a scalp voltage field with a frontotemporal to frontopolar negative maximum and a posterior positive maximum 27,29,30 (Figure 5-2). Other investigators confirmed and validated these results in additional patient populations. Most came to the same conclusionthat dipole orientation was essential for proper interpretation of source solutions.37-40 In a series of studies, Boon et al.4143 demonstrated that dipole orientation could differentiate patients with the basomesial temporal lobe epilepsy from those with lateral neocortical temporal lobe epilepsy. This distinction was based on dipole orientation and could be made by analyzing either spike discharges or seizure rhythms. Dipoles with a radial and horizontal orientation reflected spike or seizure activity from the lateral temporal cortices, whereas dipole models with a more vertical orientation reflected activity in the base of the temporal lobe that originated from more mesial structures. Using these criteria, Boon et al. were able to predict surgical outcomes and intracranial EEG findings. A later study by this same group showed that dipole analysis of ictal rhythms identified incongruence between a structural abnormality and the source of epileptiform EEG activity. This information could be useful in deciding not to proceed with invasive studies or surgical resection.44 Numerous other publications appeared in the late 1990s that demonstrated

5  Dipole Source Modeling in Epilepsy

61

that the localization provided by spike dipole modeling correlated well with other localization tools such as intracranial EEG,45,46 positron emission tomography,47 single photon emission tomography,47,48 and MRI lesions in both adults and pediatric patients.49 Assaf and Ebersole50 examined ictal recordings from the scalp EEGs of 40 patients with temporal lobe epilepsy who required invasive monitoring as part of a presurgical evaluation. The earliest recognizable seizure rhythms for each patient were averaged and analyzed using a source montage based on a fixed set of dipoles modeling defined cortical areas of the temporal lobe. Results were compared to ictal localization from intracranial recordings. A high positive predictive value was seen between vertical tangential dipoles and seizure onset in the hippocampus, between horizontal tangential dipoles and seizure onset in the temporal tip or entorhinal cortex, and finally between horizontal radial dipoles and seizure onset lateral temporal neocortex. These results from ictal data are thus comparable to findings from previous interictal studies. In a later publication, these authors51 demonstrated that the same type of dipole analysis could be used to predict surgical success. They found that those patients with a temporal ictal rhythm having a dominant and leading basal source were more likely to maintain seizure freedom following surgery than those with a lateral source. Michel et al.52 recently performed the largest prospective study of interictal source modeling to date. In a group of 40 candidates for epilepsy surgery, they analyzed epileptiform EEG spikes using a distributed, rather than dipole, model. In a subgroup of 24 patients (17 with temporal and 7 with extratemporal foci) who underwent the surgery, source models of their spikes fell within the borders of the nearest surgical resection in 18 patients. Sixteen of these 18 had a class 1 Engel outcome. They also noted that the localization by source models agreed with the intracranial EEG in the seven patients who had invasive studies. The same group recently examined and demonstrated the usefulness of EEG source modeling in a presurgical cohort of pediatric epilepsy patients.53

Usefulness of EEG Dipole Modeling in Clinical Practice


Given the numerous clinical studies cited earlier, it is clear that EEG dipole modeling is useful in a variety of clinical settings and for a variety of clinical questions. Localization of cortical sources for epileptiform EEG is, however, clearly the predominant goal. In an effort to define the irritative cortical zone noninvasively, EEG spikes have been the principal waveforms that are analyzed. The advantage afforded by dipoles over traditional visual inspection is that this source localization is accurate at a defined sublobar level, as opposed to a scalp electrode region. This utility exists for all cortical areas, not only in the temporal lobe, and in particular wherever there is a source producing other than a simple radial field. In these latter cases visual inspection of EEG traces commonly leads to false localization or lateralization. EEG dipole modeling is not limited to interictal spikes, however. Ictal rhythms can be modeled in a similar fashion, and the same benefits accrue with this type of analysis as compared to simple visual inspection of traces. The present status of EEG source localization in the evaluation of focal epilepsy has recently been reviewed by Plummer et al.54 Dipole modeling and its basis, voltage topography, can take full advantage of the high sampling rate of modern digital EEG recorders. The presence and direction of

62

THE EPILEPSIES 3

spike propagation can be characterized on a millisecond time scale. Spikes propagate between lobar regions or even between homotopic bilateral cortices in tens of milliseconds. It is clinically essential to differentiate the spike origin from later recruited cortex. This temporal resolution is almost never appreciated by inspection of traces. More commonly than one would think, EEG dipole models of spikes or seizures can lead to the identification of subtle cortical abnormalities that were previously missed. These can be structural, including malformations and dysplasias, or functional, such as small regions of hypometabolism on PET. Knowing where to look for an imaging abnormality increases the likelihood of finding it. In such instances, dipole modeling results in the successful acquisition of several data types. When dipole models of a patients EEG spikes and ictal rhythms converge on the same cortical location, it increases the likelihood that the epileptogenic focus has been identified. A further coherence with structural and other clinical data may allow patients to proceed directly to surgery in instances where there was previously insufficient or inconclusive functional data. Similarly, localizing putative spike and seizure sources at a sublobar level can provide needed information in planning electrode placement for an intracranial EEG study. This is particularly important in cases of multiple lesions, such as tubers, or large lesions where one needs to choose which lesion or which aspect of the lesion to investigate. Fishing expeditions with scattered intracranial electrodes in an attempt to find the focus are no longer acceptable. Finally, identifying very divergent localizations for EEG sources of epileptiform activity by itself or in relation to other clinical data may provide the insight needed to decide not to proceed with epilepsy surgery.

Conclusions
Dipole modeling is simply an extension of EEG field analysis. All the information used by this mathematical construct is contained in the contours of the voltage fields measured at the scalp. An appreciation for the significance of these field features, such as the location of both negative and positive maxima, the gradients between them, and how they evolve over time, leads to an understanding and easy acceptance of dipole source modeling. Unfortunately, traditional EEG interpretation has been fixated on pattern recognition of EEG traces, rather than on understanding the underlying voltage fields. For the most part, clinical EEG localization has failed to advance beyond identifying the largest negative potential and assuming that the source of that potential must underlie the electrode recording it, even though that assumption is biophysically incorrect in most situations. Modern digital EEG machines and software can make the display of voltage fields and calculation of an equivalent dipole as simple as a mouse click. That these data and models are not used more often in clinical work appears to be mostly a matter of inertia and the unfounded concern that significant additional time and effort will be required. With a more widespread appreciation for the strengths and limitations of dipole modeling and an understanding of how to interpret these models clinically, this EEG analysis tool is ready to take its place among the routine practices for evaluating patients with epilepsy.

5  Dipole Source Modeling in Epilepsy

63

REFERENCES
1. Ebersole JS. Cortical generators of EEG voltage fields. In: Ebersole JS, Pedley T, eds. Current Practice of Clinical Electroencephalography. 3rd ed. Philadelphia: Lippincott Williams & Wilkins; 2003:12-31. 2. Ebersole JS, Hawes-Ebersole S. Clinical application of dipole models in the localization of epileptiform activity. J Clin Neurophysiol. 2007;24:120-129. 3. Tao J, Ray A, Hawes-Ebersole S, Ebersole JS. Intracranial substrates of scalp EEG interictal spikes. Epilepsia. 2005;46(5):669-676. 4. Helmholtz, H. Uber einige Gesetze der Vertheilung elektrischer Strome in korperlichen Leitern, mit Anwendung auf die thierischelekrischen Versuche. Ann Phys Chem. 1853;29:211-233,353-377. 5. Wilson FN, Bayley RH. The electric field of an eccentric dipole in a homogeneous spherical conducting medium. Circulation. 1950;1:84-92. 6. Shaw JC, Roth M. Potential distribution analysis. II. A theoretical consideration of its significance in terms of electric field theory. Electroencephalogr Clin Neurophysiol. 1955;7:285-292. 7. Rush S, Driscoll DA. Current distribution in the brain from surface electrodes. Anaesthesia Analgesia Current Res. 1968;47:717-723. 8. Brody DA, Terry FH, Ideker RE. Eccentric dipole in spherical medium: generalized expression for surface potentials. IEEE Trans Biomed Eng. 1973;20:141-143. 9. Kavanagh RN, Darcey TM, Lehmann D, Fender DH. Evaluation of methods for three-dimensional localization of electrical sources in the human brain. IEEE Trans Biomed Eng. 1978;25:421-429. 10. Sidman RD, Giambalvo V, Allison T, Bergey P. A method for localization of sources of human cerebral potentials evoked by sensory stimuli. Sensory Proc. 1978;2:116-129. 11. Darcey TM, Ary JP, Fender DH. Methods for the localization of electrical sources in the human brain. Prog Brain Res. 1980;54:128-134. 12. Cuffin BN. Effects of head shape on EEGs and MEGs. IEEE Trans Biomed Eng. 1980;37:44-52. 13. Cuffin BN. EEG localization accuracy improvements using realistically shaped head models. IEEE Trans Biomed Eng. 1996;43:299-303. 14. Roth BJ, Balish M, Gorbach A. How well does a three-sphere model predict positions of dipoles in a realistically shaped head? Electroencephalogr Clin Neurophysiol. 1993;87:175-184. 15. Yvert B, Bertrand O, Thevenet M, Echallier JF, Pernier J. A systematic evaluation of the spherical model accuracy in EEG dipole localization. Electroencephalogr Clin Neurophysiol. 1996;102:452-459. 16. Roth BJ, Ko D, von Albertini-Carletti IR, Scaffidi D, Sato, S. Dipole localization in patients with epilepsy using the realistically shaped head model. Electroencephalogr Clin Neurophysiol. 1997;102:159-166. 17. Fuchs M, Drenckhahn R, Wischmann HA, Wagner M. An improved boundary element method for realistic volume-conductor modeling. IEEE Trans Biomed Eng. 1998;45:980-997. 18. Fuchs M, Wagner M, Kohler T, Wischmann HA. Linear and nonlinear current density reconstructions. J Clin Neurophysiol. 1999;16:267-295. 19. Fuchs M, Kastner J, Wagner M, Hawes S, Ebersole JS. A standardized boundary element method volume conductor model. Clin Neurophysiol. 2002;113:702-712. 20. Fuchs M, Wagner M, Kastner J. Boundary element method volume conductor models for EEG source reconstruction. Clin Neurophysiol. 2001;112:1400-1407. 21. Herrendorf G, Steinhoff BJ, Kolle R, et al. Dipole-source analysis in a realistic head model in patients with focal epilepsy. Epilepsia. 2000;41:71-80. 22. Nunez, P, Srinivasan, R. Electric fields and currents in biological tissue. In: Nunez P, Srinivasan R, eds. Electric Fields of the Brain: The Neurophysics of EEG. 2nd ed. New York: Oxford University Press; 2006:147-202. 23. Fuchs M, Wagner M, Kastner J. Development of volume conductor and source models to localize epileptic foci. J Clin Neurophysiol. 2007;24:101-119. 24. Lantz G, Grave de Peralta R, Spinelli L, Seeck M, Michel CM. Epileptic source localization with high density EEG: how many electrodes are needed? Clin Neurophysiol. 2003;114:63-69. 25. Ebersole JS. EEG dipole modeling in complex partial epilepsy. Brain Topogr. 1991;4:113-123. 26. Ebersole, JS. Equivalent dipole modeling: a new EEG method for epileptogenic focus localization. In: Pedley TA, Meldrum BS, eds. Recent Advances in Epilepsy 5, Edinburgh: Churchill Livingstone; 1991:51-72. 27. Ebersole JS. Non-invasive localization of the epileptogenic focus by EEG dipole modeling. Acta Neurol Scand Suppl. 1994;152:20-28. 28. Ebersole JS. Defining epileptogenic foci: past, present, future. J Clin Neurophysiol. 1997;14:470-483. 29. Ebersole JS. Noninvasive localization of epileptogenic foci by EEG source modeling. Epilepsia. 2000;41:S24-S33.

64

THE EPILEPSIES 3 30. Ebersole, JS. Sublobar localization of temporal neocortical epileptogenic foci by EEG source modeling. In: Williamson P, Siegel A, Roberts D, Thadani V, Gazzaniga M, eds. Advances in Neurology, Neocortical Epilepsies. Vol. 84. Philadelphia: Lippincott Williams & Wilkins; 2000:353-363. 31. Lantz G, Spinelli L, Seeck M, de Peralta Menendez RG, Sottas CC, Michel CM. Propagation of interictal epileptiform activity can lead to erroneous source localizations: a 128-channel EEG mapping study. J Clin Neurophysiol. 2003;20:311-319. 32. Bast T, Boppel T, Rupp A, et al. Noninvasive source localization of interictal EEG spikes: effects of signal-to-noise ration and averaging. J Clin Neurophysiol. 2006;23:487-497. 33. Ebersole JS, Wade PB. Spike voltage topography and equivalent dipole localization in complex partial epilepsy. Brain Topogr. 1990;3:21-34. 34. Ebersole JS, Wade PB. Spike voltage topography identifies two types of frontotemporal epileptic foci. Neurology. 1991;41:1425-1433. 35. Ebersole JS, Hawes S, Scherg M. Intracranial EEG validation of spike propagation predicted by dipole models. Electroencephalogr Clin Neurophysiol. 1995;95:18. 36. Pacia SV, Ebersole JS. Intracranial EEG substrates of scalp ictal patterns from temporal lobe foci. Epilepsia. 1997;38:642-654. 37. Lantz G, Ryding E, Rosen I. Three-dimensional localization of interictal epileptiform activity with dipole analysis: comparison with intracranial recordings and SPECT findings. J Epilepsy. 1994;7: 117-129. 38. Baumgartner C, Lindinger G, Ebner A, et al. Propagation of interictal epileptic activity in temporal lobe epilepsy. Neurology. 1995;45:118-122. 39. Lantz G, Holub M, Ryding E, Rosen I. Simultaneous intracranial and extracranial recording of interictal epileptiform activity in patients with drug resistant partial epilepsy: patterns of conduction and results from dipole reconstructions. Electroencephalogr Clin Neurophysiol. 1996;99:69-78. 40. Lantz G, Ryding E, Rosen I. Dipole reconstruction as a method for identifying patients with mesolimbic epilepsy. Seizure. 1997;6:303-310. 41. Boon P, DHave M. Interictal and ictal dipole modelling in patients with refractory partial epilepsy. Acta Neurol Scand. 1995;92:7-18. 42. Boon P, DHave M, Adam C, et al. Dipole modeling in epilepsy surgery candidates. Epilepsia. 1997;38:208-218. 43. Boon P, DHave M, Van Hoey G, et al. Interictal and ictal source localization in neocortical versus medial temporal lobe epilepsy. Adv Neurol. 2000;84:365-375. 44. Boon P, DHave M, Vanrumste B, et al. Ictal source localization in presurgical patients with refractory epilepsy. J Clin Neurophysiol. 2002;19:461-468. 45. Mine S, Yamaura A, Iwasa H, Nakajima Y, Shibata T, Itoh T. Dipole source localization of ictal epileptiform activity. Neuroreport. 1998;9:4007-4013. 46. Homma I, Masaoka Y, Hirasawa K, Yamane F, Hori T, Okamoto Y. Comparison of source localization of interictal epileptic spike potentials in patients estimated by the dipole tracing method with the focus directly recorded by the depth electrodes. Neurosci Lett. 2001;304:1-4. 47. Shindo K, Ikeda A, Musha T, et al. Clinical usefulness of the dipole tracing method for localizing interictal spikes in partial epilepsy. Epilepsia. 1998;39, 371-379. 48. Rojo P, Caicoya AG, Martin-Loeches M, Sola RG, Pozo MA. [Localization of the epileptogenic zone by analysis of electroencephalographic dipole]. Rev Neurol.;32:315-20. 49. Krings T, Chiappa KH, Cuffin BN, Buchbinder BR, Cosgrove GR. Accuracy of electroencephalographic dipole localization of epileptiform activities associated with focal brain lesions. Ann Neurol. 1998;44:76-86. 50. Assaf BA, Ebersole JS. Continuous source imaging of scalp ictal rhythms in temporal lobe epilepsy. Epilepsia. 1997;38:1114-1123. 51. Assaf BA, Ebersole JS. Visual and quantitative ictal EEG predictors of outcome after temporal lobectomy. Epilepsia. 1999;40:52-61. 52. Michel CM, Lantz G, Spinelli L, De Peralta RG, Landis T, Seeck M. 128-channel EEG source imaging in epilepsy: clinical yield and localization precision. J Clin Neurophysiol. 2004;21:71-83. 53. Sperli F, Spinelli L, Seeck M, Kurian M, Michel CM, Lantz G. EEG source imaging in pediatric epilepsy surgery: a new perspective in presurgical workup. Epilepsia. 2006;47:981-990. 54. Plummer C, Harvey AS, Cook M. EEG source localization in focal epilepsy: where are we now? Epilepsia. 2008;49:201-218.

THE EPILEPSIES 3

EEG-Correlated fMRI in Epilepsy: Current State of the Art


RACHEL THORNTON  LOUIS LEMIEUX

Introduction Interictal Activity in Focal Epilepsy Localization of BOLD Changes in Focal Epilepsy BOLD Changes Associated with Specific Pathologies in Focal Epilepsy EEG-fMRI and the Pediatric Epilepsies Ictal Studies in Focal Epilepsy EEG-fMRI as a Tool to Study the Neurobiology of Epilepsy

Evolving Analysis Methodology Generalized Epilepsies Technical Aspects Principles of EEG-MR System Interactions Safety Data Quality Strategies for the Reduction and Correction of Artifacts on the EEG

Introduction
EEG-correlated fMRI, commonly referred to as EEG-fMRI, is a magnetic resonance imaging (MRI) technique that uses the simultaneously recorded electroencephalogram (EEG) as a marker of brain activity. EEG is a measure of a neuronal activity, whereas fMRI reflects hemodynamic changes, in particular blood oxygen level dependent (BOLD) signal, which is the main contrast mechanisms exploited by fMRI. Although it is clear that BOLD signal change is indirectly linked to neuronal activity, it is not a clear relationship, and the nature of this neurovascular coupling is not fully understood. The study of the hemodynamic correlates of pathological EEG patterns observed in epilepsy by use of EEG-fMRI has developed since the mid1990s and indeed provided the original impetus for this development. While the initial motivation for combining the two modalities was to overcome some of the deficiencies in each technique, in particular the problem of EEG source localization (the inverse solution) and the low temporal resolution of fMRI, EEG-fMRI uniquely allows the study of the haemodynamic correlates of spontaneous brain activity and in particular individual interictal discharges and seizures.

65

66

THE EPILEPSIES 3

Despite the methodological challenges encountered in implementing the technique, it has been shown to provide a new form of localizing information and even shed new light on the physiology of interictal and ictal epileptic phenomena. This chapter is organized as follows: we begin by reviewing the results of the application of EEG-fMRI in patients with both focal and generalized epilepsy. Central to the technique is the modeling of the BOLD signal based on the observed EEG, which forms a large part of the discussion. This has important implications for not only the techniques sensitivity, and therefore its potential clinical usefulness, but also our understanding of the relationship between EEG activity and brain hemodynamics. The main technical and methodological issues of EEG-fMRI are reviewed at the end of the chapter.

Interictal Activity in Focal Epilepsy


The primary aim of developing EEG-fMRI in the first instance was to infer the location of irritative and seizure onset zones with the hope of providing useful clinical information, particularly in patients undergoing presurgical evaluation. Using an interictal epileptic discharge (IED) triggered technique, whereby images were only acquired following the observation of an EEG event of interest and compared voxel by voxel to images acquired following periods of background EEG, initial studies in selected patients with focal epilepsy revealed significant BOLD increases in a large proportion of cases that were reproducible,1 although the statistical tests applied varied widely28 (see Table 6-1). It should be noted that the early studies focused on the identification of BOLD signal increases, neglecting the possibility of BOLD decreases, which were subsequently shown to be of considerable interest (see discussion later in this chapter). The continuous EEG-fMRI acquisition approach was made possible by the use of EEG processing methods to correct scanning-related artifact9 (see the section on technical aspects at the end of this chapter for further discussion). The analysis of this type of data requires the creation of a model of the BOLD signal over the entire experiment by identifying images that coincide with EEG events of interest, such as IED.10 This model is then used to find voxels with BOLD signal time courses using correlation. A map of the IED-correlated BOLD signal change is then created across the brain (for examples, see Figures 6-1 and 6-2). The identification of the events of interest is usually made by human observers and suffers from the well-documented limitations of this approach.11 In addition to the limits of an observer-derived EEG model per se, not all patients will have IEDs in the scanner, thus presenting a further difficulty. In the two largest case series of continuously recorded EEG-fMRI in focal epilepsy,12,13 around 50% of patients had IEDs during scanning. Approximately 60% of these had a BOLD signal change recorded in the vicinity of the electroclinical localization of seizure onset (known as concordant BOLD signal change; see Figure 6-1 for illustration). This figure represents the yield of EEG-fMRI in a given study. It is improved by modeling runs of IEDs rather than single events in addition to improvement in the EEG model by adequate spike detection and classification by the observer.11,13,14 The patterns of IED-related BOLD signal changes can be complex, with clusters commonly observed in close proximity to or overlapping the presumed seizure onset zone and remote from the epileptogenic zone and irritative zone regions of BOLD decrease most often observed at remote sites.

TABLE 61

Significant Case Series and Milestones in EEG-fMRI (Interictal Studies)


Subjects: Number and Type Study Type Findings

Study Case Reports

Warach et al., 19962 Seeck et al., 19983

1 Frequent IEDs 1 Frequent IEDs

EEG-triggered fMRI EEG-triggered fMRI

Symms et al., 19991 Krakow et al., 19995 Lemieux et al., 200110 Baudewig et al., 200165

1 Frequent IEDs 1 Frequent IEDs 1 Frequent, stereotyped IEDs 1 Frequent GSW

EEG-triggered fMRI EEG-triggered fMRI First case using continuous simultaneous EEG-fMRI First case recording findings in GSW (interleaved EEG-fMRI) EEG-triggered fMRI: Comparison of analysis methods First major series of focal epilepsy: EEG-triggered fMRI EEG-triggered fMRI EEG-triggered fMRI

Bilateral activation where EEG suggested left temporal localization and anterior cingulate activation in relation to generalized epileptiform activity Multiple areas of signal enhancement on fMRI. Confirmed on 3D-EEG source localization with evidence of a focal onset. Focus later confirmed on subdural recordings. Results of EEG-fMRI in a patient with focal epilepsy are reproducible across sessions In a case with stereotyped frequent IEDs, BOLD signal change concordant with the seizure onset zone was recorded Unilateral insular activation shown in relation to generalized epileptiform discharges

6  EEG-Correlated fMRI in Epilepsy

Case Series: Focal Epilepsy

Patel et al., 19996 Krakow et al., 19995 Lazeyras et al., 20004 Krakow et al., 20018

20 focal epilepsy, 10 frequent IEDs 10, focal epilepsy Frequent IEDs 11 Frequent IEDs 24 Frequent IEDs

Individual spike analysis versus resting state resulted in high yield of BOLD activation 60% IED correlated BOLD activation concordant with seizure focus 8/11 concordant with seizure focus. 5/11 confirmed on icEEG 12 patients BOLD response concordant with seizure onset, 2 nonconcordant. No activation in 10 Table continued on following page

67

68
THE EPILEPSIES 3

TABLE 61

Significant Case Series and Milestones in EEG-fMRI (Interictal Studies) (Continued)


Subjects: Number and Type Study Type Findings

Study

Ja ger et al., 20027 nar et al., 200216 Be Al Asmi et al., 200313 Benar et al., 200628 Aghakhani et al., 200670 Salek-Hadaddi et al., 200612 Zijlmans et al., 200733

10 Frequent IEDs, focal epilepsy 4 Frequent IEDs 48/31 Frequent IEDs 5/5 sEEG candidates 64/40 Focal epilepsy with frequent unilateral or bilateral IEDs 63 Focal epilepsy frequent IEDs; IEDs recorded in 50% 29 Focal epilepsy, declined for surgery 15 with IEDs

EEG-triggered fMRI Simultaneous EEG/fMRI EEG-triggered fMRI in 16, continuous acquisition in remainder Simultaneous EEG/fMRI Simultaneous EEG/fMRI

Simultaneous EEG/fMRI Simultaneous EEG/fMRI, first systematic assessment of role of EEG fMRI in presurgical evaluation

5/10 patients had IED correlated BOLD response, concordant in 4/5 The average HRF presented a wider positive lobe in three patients and a longer undershoot in two. BOLD activation in 39% of studies. Concordant with seizure focus in almost all. 4 patients had concordant intracranial recording (by lobe). Intracranial electrodes <40 mm from BOLD are usually active. A positive thalamic response was seen in 12.5% of studies with unilateral and 55% with bilateral spikes. Cortical activation was more concordant with focus than deactivation. Significant hemodynamic correlates were detectable in over 68% of patients and were highly, but not entirely, concordant with site of presumed seizure onset. 8/15 subjects: IED correlated BOLD response at site of focus. Multifocal in 4, unifocal in 4. Concordant with IC data in 2.

Generalized Epilepsy

Archer et al., 200366 Hamandi et al., 200562 Aghakhani et al., 200467


Pediatric Series

5 GSW (IGE only) 46 GSW in IGE or SGE IEDs recorded in 30 15 GSW. IEDs recorded in 14 6 Focal epilepsy (lesional and nonlesional) pediatric 9 Mixed focal epilepsy, sedated pediatric 13 Symptomatic (lesional epilepsy), sedated pediatric

First group study in Generalized Epilepsy: Spike triggered EEG/fMRI Simultaneous EEG/fMRI Simultaneous EEG/fMRI

4/5 patients had deactivation (negative response) in posterior cingulated cortex. Thalamic activation and cortical deactivation observed at group level. Deactivation in the default brain areas. Bilateral thalamic activation in 80% of BOLD response. Cortical deactivation in 93%. Concordant activation with presumed focus in 4 cases. IC recording corroborative in 1. All had BOLD activation or deactivation concordant with seizure focus. Deactivation appears more common than in adults. Activation corresponding with the lesion was seen in 20% and deactivation in 52% of the studies.
6  EEG-Correlated fMRI in Epilepsy

De Tiege et al., 200742 Jacobs et al., 200744 Jacobs et al., 200743

Simultaneous EEG/fMRI Simultaneous EEG/fMRI Simultaneous EEG/fMRI

Abbreviations: IEDs, interictal epileptiform discharges; GSW, generalized spike and wave; sEEG, stereoencepholgraphic.

69

70

THE EPILEPSIES 3

Figure 61 Positive BOLD signal change correlated with frequent right midtemporal sharp waves in a patient with right temporal lobe epilepsy (maximum T8), overlaid on normalized T1-weighted structural MRI in MNI space (coronal plane). Intracranial recording confirmed an extensive irritative zone in the right midtemporal lobe extending posteriorly (SPM-T p < 0.05 familywise error correction for multiple comparisons, t = 6.42 at voxel of maximum intensity).

The interpretation of these remote changes remains one of the active areas in EEGfMRI research. As a general rule, the results of EEG-fMRI studies in focal epilepsy suggest that the time course of the BOLD increases (known as the hemodynamic response function, or HRF12,15) related to focal spikes matches the so called canonical shape (i.e., peaks at 5 to 6 seconds after the event, returning to baseline roughly 15 seconds later), of responses observed in relation to events (e.g. stimuli) in healthy brains. Deviations from the normal time course, however, have been observed in epilepsy. Although the BOLD response, both positive and negative (to external stimuli), has been studied extensively and is well documented in reference to neuronal activity in healthy subjects, some have suggested that this relationship may be altered in the case of interictal epileptiform discharges.1618 A formal study investigating this possibility found that the inclusion of noncanonical time courses does not lead to an important increase in yield.12 A more recent study suggests noncanonical HRFs associated with IEDs are often remote from the presumed seizure onset zone,19 but whether this represents artifact, propagation of epileptiform activity (timelocked activity) or another phenomenon is not clear. Others have, however,

6  EEG-Correlated fMRI in Epilepsy

71

Figure 62 SPM{F} from a patient with frequent GSW discharges. Top left: BOLD response correlated with GSW in glass brain display (the red arrow marks the global maxima). Top right: design matrix: first column shows the effect of interest (GSW) convolved with canonical HRF; subsequent columns represent motion and cardiac regressors. At the bottom: BOLD response overlaid onto normalized T1-weighted MRI in MNI space (p < 0.05 corrected). Note bilateral thalamic and right cingulate gyrus activation, right precuneus (global maxima showed by cross hair), bilateral middle frontal gyrus, right occipital lobe (lingual gyrus) deactivation (localization confirmed using Talairach coordinates in Talairach Daemon, http://ric.uthscsa.edu/project/talairachdaemon.html). (Image courtesy of Dr. A. E. Vaudano.)

72

THE EPILEPSIES 3

emphasized the importance of intersubject variability in the HRF and routinely use individualized HRFs to analyze IED-correlated fMRI,16,20 and the issue is still under debate. Early BOLD signal change (time locked to IEDs) has been reported in focal epilepsy,21 and BOLD activation has also been reported in the thalamus prior to the onset of generalized spike and wave discharges.22 It is conceivable that these reflect neuronal activity time-locked but preceding an event observed on the scalp EEG and for example may represent a phenomenon similar to observed preictal BOLD signal change in humans25 and preceding the occurrence of epileptic spikes in an animal model.26 Another possibility is that this is a reflection of abnormal neurovascular coupling. However, the specificity of this type of BOLD time course to epilepsy remains to be properly assessed. Furthermore the relationship between blood perfusion and BOLD changes linked to epileptiform discharges in focal (one case) and generalized epilepsies are consistent with preservation of neurovascular coupling in epilepsy.23,24 LOCALIZATION OF BOLD CHANGES IN FOCAL EPILEPSY As has been discussed earlier, IED-correlated BOLD signal change may be colocalized with the irritative or epileptogenic zones.27 We will now discuss the efforts to validate EEG-fMRI localization by comparing other noninvasive localization methods and the gold standard of intracranial recording. Some reports, predominantly within larger series, of concordance of IEDcorrelated BOLD response with seizure onset were identified by intracranial recording,3,4,12,13 but the only more systematic study was limited to a five-case series. Here it was found that where EEG-fMRI activations were identified, at least one active electrode was on intracranial recording in the same location. The same series made a comparison with source analysis from standard scalp EEG.28 Comparison of noninvasive EEG source localization and EEG-fMRI activations have been made using both spike-triggered fMRI and continuous acquisition. An initial study observed that dipolar sources were often remote (2 to 6 cm) from EEGfMRI localization.29 This was corroborated by a separate study that suggested distributed source analysis might be a more effective way of comparing EEG source and BOLD activation.30 A recent systematic study using calculated measurements over the cortex revealed IED-associated BOLD clusters that were highly concordant with distributed sources in most patients recordings, but also that other EEG-fMRI sources were present that were not concordant with the distributed sources,31 particularly negative BOLD responses. An initial study observed that multi-dipolar sources were generally in anatomical (lobar level) agreement with EEG-fMRI localization based on BOLD increases, and lesser agreement with BOLD decreases.32 Despite comparative studies made between modalities as described earlier, the role of EEG-fMRI in presurgical evaluation remains undefined. The only attempt to assess its practical benefit suggests a limited role in evaluating the group of patients where surgery was not able to be offered.33 Eight patients from this group had significant IED-correlated BOLD signal increase concordant with presumed seizure focus when studied with EEG-fMRI. Of these, four had a unifocal activation, which narrowed the location of the seizure onset zone and was concordant with intracranial recordings in two. In four patients, multifocal BOLD activation concordant with electroclinical evaluation was observed. Larger studies comparing multimodal

6  EEG-Correlated fMRI in Epilepsy

73

invasive and noninvasive investigation in addition to postsurgical outcome are required to complete the picture. BOLD CHANGES ASSOCIATED WITH SPECIFIC PATHOLOGIES IN FOCAL EPILEPSY It is known in lesional epilepsy that the irritative zone and epileptogenic zone may extend beyond the anatomical boundary of pathological abnormality, and in addition, both in vitro and animal models suggest abnormal subpopulations of neurons within dysplastic areas.34 EEG-fMRI has therefore been used as a tool to evaluate the hemodynamic response in areas of cortical malformation and other pathologies.3539 In general, EEG-fMRI studies of malformations of cortical development have shown variability in BOLD response within pathologically abnormal regions with broadly concordant activations reported in both gray matter heterotopia35 and focal cortical dysplasia.38 In both these studies, BOLD decreases were observed remote to the area of pathological abnormality. Other studies of malformations of cortical development (MCD) have supported these findings.37 The frequent IED-related BOLD decreases, particularly in MCD, have been attributed to loss of neuronal inhibition (in the presence of normal neurovascular coupling) in the regions surrounding the abnormality or abnormalities in neurovascular coupling itself. The significance of these deactivations will be discussed in more detail later. A recent study of IED-related BOLD signal change in cavernomas36 suggested BOLD signal change both within the area of anatomical abnormality and within areas remote to it. Caution is required in interpreting BOLD signal change in these very vascular lesions as T2* sequences used in EEG-fMRI are very sensitive to hemosiderin within the lesions. EEG-fMRI results in tuberous sclerosis reflect a similar pattern. In a recent study exclusively using pediatric patients, it was found that BOLD activation was variable between tubers and extended beyond the border of tubers as identified on structural images once again supporting the view that EEG-fMRI may be a useful technique in assessing the epileptic network beyond areas of structural abnormality.40 EEG-fMRI AND THE PEDIATRIC EPILEPSIES The pediatric population has been generally less well studied with EEG-fMRI than adults (although some of the studies mentioned earlier have included several children). The experiments are lengthy and can be difficult to tolerate, particularly because they require subjects to remain still for the duration of the procedure. However, intracranial recording is also difficult in children, and invasive procedures such as this may be less acceptable to the pediatric population, meaning that the need for development of noninvasive techniques for localization of the seizure onset zone is even more pressing.41 Two series have used EEG-fMRI in groups of children with focal epilepsies of mixed etiology illustrating that the experiments are tolerated, and results may also show concordance with the seizure onset zone.42,43 In the most recent of these studies, it was found that deactivations are more common and more widespread than in adults, although a common pattern was not identified in this group.44 Questions remain regarding whether this is a function of age and the developing brain, of the particular epilepsy syndrome under study, or whether it in fact is

74

THE EPILEPSIES 3

related to pharmacological sedation, the use of which is necessary in children undergoing EEG-fMRI. Previous studies have suggested that benzodiazepine sedation may affect BOLD response to IEDs.45 Specific syndromes are now being studied in children. Two groups of children with benign epilepsy of childhood with centrotemporal spikes (BECCTS) have undergone EEG-fMRI,32,46 and both studies reported BOLD signal change concordant with the electroclinical localization in addition to the tuberous sclerosis study mentioned earlier. A recent study of children with generalized spike-and-wave activity is discussed later.22 ICTAL STUDIES IN FOCAL EPILEPSY Ictal EEG-fMRI studies present specific technical difficulties due to increased movement artifact and specific safety concerns. Therefore, seizure-related activity is likely to be recorded by chance rather than by design. Clinical seizure onset has been used as an event marker in fMRI studies without the use of EEG,25,47,48 which has revealed concordance of seizure-related fMRI activation with electroclinical localization in some cases. In the few ictal EEG-fMRI studies in focal epilepsy, activations have been more widespread than those observed interictally, usually extending beyond the seizure onset zone. A large increase in BOLD signal was observed, time-locked to a focal subclinical seizure in electroclinically concordant gray matter structures.49 In a patient with multiple simple partial seizures, BOLD activation was revealed, concordant with the electroclinically determined seizure focus. In addition, widespread deactivation was observed on the contralateral side and in other areas of the brain, and increases in the area of activation with cumulative numbers of seizures were analyzed.50 One benefit of recording seizures has been the opportunity to observe preictal changes in BOLD signal. PET and SPECT data suggest preictal changes in cerebral metabolism, and studies of cerebral blood flow support this.51 An fMRI study without simultaneous EEG in three patients suggested BOLD signal change may occur up to minutes prior to the electroclinical onset of seizures,25 a phenomenon that requires further evaluation. These results suggest that analysis of EEG-fMRI may allow detection of early changes in brain state prior to changes observed on the scalp EEG, but it is difficult to expand these studies for the reasons discussed earlier. EEG-fMRI AS A TOOL TO STUDY THE NEUROBIOLOGY OF EPILEPSY Aside from the observations regarding localization of the seizure onset zone in focal epilepsy, the observation of activations extending beyond the lesional zone in addition to deactivations (negative hemodynamic response) has led to attempts at exploring the epileptic network using EEG-fMRI in the resting state. In focal epilepsy, initial investigations have focused on temporal lobe epilepsy, the most uniform and best-studied syndrome. An initial series of 19 patients with temporal lobe epilepsy showed the majority had BOLD signal change in the affected temporal lobe, although the activations were usually neocortical, and deactivations often occurred in extratemporal regions.52 Given the observation that the remote deactivations occur in temporal lobe epilepsy, a further study in a group of patients with temporal lobe epilepsy

6  EEG-Correlated fMRI in Epilepsy

75

showed that deactivation of the precuneus is common and associated with activation in the ipsilateral hippocampus in relation to interictal temporal lobe spikes.53 This effect was not observed in a similarly selected extratemporal lobe epilepsy group and was thought to reflect a suspension of the precuneus (a region commonly activated in an awake resting state in contrast to reduced conscious or task states54 (and correlated with alpha activity on the EEG55), specifically in response to temporal lobe spikes. The authors pointed out that this may reflect a subclinical suspension of the awake resting state analogous to the cortical deactivation/thalamic activation, which occurs in response to generalized spike-and-wave (GSW) discharge. See discussion later in the chapter. Methods to assess the functional and structural connectivity between activated regions, such as dynamic causal modeling (which assesses causality in functional connectivity or effective connectivity)56 and diffusion tensor imaging have recently been combined to investigate putative propagation of epileptiform activity,24 and the continued evaluation of connectivity in the epileptic network is an exciting area of evolving research. Beyond the BOLD changes related to epilepsy-specific stereotypical discharges, such as focal interictal spikes, there had been limited study of other electrophysiological abnormalities. Inspired by methods employed in the study of the hemodynamic correlates of normal brain rhythms using EEG-fMRI,55,57,58 EEG frequency-band-based approaches have been used to study patients with epilepsy, but these studies are restricted to case reports at present.37,59 In the first of these, delta power associated EEG-fMRI was correlated with the seizure onset zone, confirmed by intracranial recording. EVOLVING ANALYSIS METHODOLOGY As we have seen, significant BOLD changes are not revealed in an important proportion of cases in which pathological EEG activity is observed during scanning. Methods designed to improve the sensitivity and specificity of EEG-fMRI include the use of automated approaches to spike detection and classification14 and most recently a more symmetric fusion of EEG and fMRI.60 A major limitation of EEG-fMRI is the failure of the technique to register EEG abnormalities in a large proportion of cases studied, depending on the selection criteria. In such cases, the correlation approach cannot be used. This has led to the exploration of data-driven (i.e., without reference to any other data) fMRI analysis in an attempt to identify patterns of BOLD signal specifically linked to the IZ and EZ, such as temporal cluster analysis (TCA)61,62 and independent component analysis (ICA).63

Generalized Epilepsies
In generalized epilepsy syndromes, the aim of EEG-fMRI studies has been mainly an improved understanding of the neurobiology. One of the earliest observations was of four absence seizures, which were found to be associated with a striking pattern of widespread cortical BOLD decrease and thalamic BOLD increase using continuous EEG-fMRI. This was consistent with a reduction of cortical activity during GSW and a key role for the thalamus,64 although there were case reports of

76

THE EPILEPSIES 3

EEG-fMRI in generalized epilepsy before this.65 Archer et al. revealed a pattern of BOLD decrease in the posterior cingulate and bilateral precentral BOLD increase in relation to brief epochs of GSW in five patients using triggered EEG-fMRI.66 Continuous EEG-fMRI in larger series of patients revealed a varied pattern of GSW-related cortical BOLD changes dominated by decreases and changes in the thalamus dominated by increases6769 (Figure 6-2 illustrates this finding). By grouping the results for cases with IGE on one hand and secondarily generalized epilepsy on the other, Hamandi et al. were able to demonstrate that GSW is commonly associated with BOLD decreases in the precuneus/posterior cingulate and bilateral BOLD increases in the thalamus, and that there is considerable overlap between the patterns in the two groups. Interestingly, similar thalamic activations were observed in 55% of a group of patients with focal epilepsy with bilateral spikes, compared to just 12.5% in a group with unilateral spikes.70 It was proposed that involvement of the precuneus reflects a subclinical manifestation of the suspension of consciousness observed in association with generalized seizures, in particular absence seizures, due to that regions presumed role in consciousness54,71 It is important to note that the type of analysis employed in these studies, based on correlation, do not allow to infer causal links7274 Recently, it was shown that thalamic hemodynamic changes can systematically precede the onset of GSW by as much as 6 seconds in some children, which may reflect the key role of the thalamus in the generation of generalized discharges.22

Technical Aspects
Despite the fact that technique has been successfully applied in many imaging centers around the world, the recording of EEG inside the MR scanner still presents safety, image data quality, and EEG data quality challenges. This is due to the use of very strong and rapidly varying electromagnetic fields in the MR image acquisition process. Historically, the issue of EEG data quality has been the determining factor in the techniques evolution, from interleaved (recordings with gaps) to simultaneous, continuous acquisition of the EEG and fMRI. This reflects mainly the gradual degradation in EEG quality linked to cardiac activity and head movement that can be readily observed as subjects are moved inside the MR scanner, immediately precluding reliable EEG interpretation, even before issues of safety or other artifacts, such as those caused by the scanning process, or issues of image quality degradation are considered. Despite the development of a dynamic commercial market for MR-compatible EEG recording equipment, EEG-fMRI experiments remain a challenge, and the hardware and software technology necessary for the acquisition of good-quality EEG-fMRI data for application in epilepsy and increasingly in other areas of neuroscience research remains an area of active development. For this reason, the main technological issues related to the acquisition and postprocessing of EEGfMRI data for subsequent analysis and interpretation are discussed in the following section.

6  EEG-Correlated fMRI in Epilepsy

77

PRINCIPLES OF EEG-MR SYSTEM INTERACTIONS The following features of the MR image acquisition process are susceptible to interactions with the EEG system: strong static magnetic field (1 to 3 Tesla [T]), switching magnetic gradient fields (100 Tm1s1), and radio frequency (RF) pulses (10 mT and 100 MHz). Four main mechanisms are at the origin of EEG-MR instrumentation interactions: magnetic induction, which may be hazardous to health and cause EEG quality degradation; magnetic susceptibility differences, which can cause image degradation; radio frequency (RF) radiation emanating from the EEG recording system, which can cause image degradation; and magnetic force, which can lead to safety concerns. The implications of these effects on the EEG-fMRI acquisition process are discussed in the following sections. SAFETY The main safety concern in carrying out simultaneous EEG-fMRI experiments is the risk of heating of the EEG components and of induced currents due to the effect of RF and to a lesser extent gradient switching. To minimize these risks, methods such as the inclusion of current limiting resistors (e.g., 10 kV inserted at each electrode for a 1.5 T scanner in one study) and the twisting together of EEG leads to minimize large loops being formed within the scanner have been recommended.75 Commercial MR-compatible EEG systems are now available that incorporate these features into their design. A crucial consideration when placing wires in contact with the body is the type of RF transmit coil used and the length of wire exposed to the electrical component of the field. The important point is that EEG-fMRI should be limited to head-only RF transmit coils, given the current state of technology and in the absence of further studies addressing this problem.76 DATA QUALITY If EEG is acquired by standard methods in the MRI scanner, in the majority of cases the signal becomes uninterpretable during image acquisition due to the presence of repetitive artifact waveforms superimposed on the physiological signal due to the switching of gradients during EPI sequence acquisition2,9 (Figure 6-3). The first attempts at recording EEG inside MR scanners revealed the presence of significant pulse artifacts (often referred to as the BCG [ballistocardiogram] artifact).77 This effect has been shown to be common across subjects.78 The pulse artifact amplitude can reach 50 mV (at 1.5T) and may resemble epileptic spikes introducing an obvious complication in the study of epilepsy. The precise mechanism through which the circulatory system exposed to a strong magnetic field gives rise to these artifacts remains uncertain, but it is thought to represent a combination of the motion of the electrodes and leads (induction) and the Hall effect (voltage induced by flow of conducting blood in proximity of electrodes).79 This effect is proportional to the scanners main field strength. In addition to artifacts on EEG, interaction between EEG and MRI systems results in artifacts caused by electrodes and leads on the images acquired,77 and this has affected the choice of EEG component materials.5,80,81 Radio-frequency fields

78

THE EPILEPSIES 3

FP2-F4 F4-C4 C4-P4 P4-O2 FP1-F3 F3-C3 C3-P3 P3-O1 FP2-F8 F8-T8 T8-P8 P8-O2 FP1-F7 F7-T7 T7-P7 P7-O1

A
FP2-F4 F4-C4 C4-P4 P4-O2 FP1-F3 F3-C3 C3-P3 P3-O1 FP2-F8 F8-T8 T8-P8 P8-O2 FP1-F7 F7-T7 T7-P7 P7-O1

B
Figure 63

Scan Start

Scan Start

Scan Start

32-channel EEG recorded inside a 3 Tesla MRI Scanner (Symphony, General Electric, Milwaukee) using MRI-compatible EEG amplifier (Brain Products, Munich). A, As acquired during echo-planar imaging before postprocessing. B, Following removal of gradient artifact using algorithm based on AAS method. (Reprinted with permission from Allen PJ, Josephs O, Turner R. A method for removing imaging artifact from continuous EEG recorded during functional MRI. Neuroimage. 2000;12:230-239.)

6  EEG-Correlated fMRI in Epilepsy

79

radiating from the EEG recording equipment placed in the vicinity of the scanner can cause severe image degradation and may therefore require shielding. Various EEG-fMRI data acquisition strategies have been employed to minimize the impact of EEG artifacts. 1. Interleaved EEG-fMRI.80,82 This method requires a gap in the acquisition of fMRI where EEG features can be reliably observed and is most useful for studying evoked responses or slow variations in brain activity. 2. EEG-triggered fMRI. This involves the identification of EEG events online to trigger a burst of fMRI scanning and is of particular relevance to epilepsy research.2,5,8,13 3. Continuous EEG-fMRI acquisition, which requires specially designed amplifiers (with adequate dynamic range, bandwidth, and sampling rate), enables image acquisition artifact correction on- or off-line.9,83 STRATEGIES FOR THE REDUCTION AND CORRECTION OF ARTIFACTS ON THE EEG Reduction at Source Careful placement and immobilization of the EEG leads, twisting of the leads, bipolar electrode chain arrangement, and the reduction of head movement using vacuum can reduce artifacts. Low-pass filtering may be used to reduce the artifact significantly, although not sufficiently to result in adequate EEG quality. Modification of MR sequences and synchronization with EEG digitization can result in significant reductions of the measured artifact amplitude. Reduction and Correction of the Image Acquisition Artifact Filtering,84,85 template subtraction, and/or ICA/PCA86 methods have all been used to remove the image artifact following acquisition of the EEG. Simple filtering led to some improvement in image quality, but template subtraction has been shown to be much more effective,87 and this method is used most widely at present (including its incorporation into commercially available MRI-compatible EEG systems). Postprocessing methods to reduce image acquisition artifacts can be categorized as filtering, template subtraction methods, or PCA/ICA. The most commonly used image acquisition EEG artifact reduction method is based on average template artifact subtraction (AAS) method.9 It enables the artifact to be separated from physiological signal by averaging the EEG over repeated epochs (based on the fMRI scanning rate) and therefore relies on accurate knowledge of the timing of the scanner signal (for instance, by synchronizing scanner and EEG clocks). Subsequent subtraction from the ongoing EEG depends critically on the sampling rate, the number of averaging epochs, and the precision of their timing. Numerous refinements of the AAS method have been proposed, notably to deal with the residual artifact when averaging of signal is suboptimal due to subject movement changing the artifact.88,89 Possibly the most important practical development has been the demonstration that synchronized MR acquisition and EEG digitization led to significantly improved EEG quality, and in particular over a wider frequency range, when combined with an AAS-like method.90

80

THE EPILEPSIES 3

Reduction and Correction of the Pulse Artifact Correction of this artifact can be more problematic than the image acquisition artifact due to beat-to-beat variability. The standard artifact reduction algorithm is based on subtraction of a running average estimate of the artifact based on QRS complex detection, in principle similar to the method used to reduce imaging artifact.78 Spatial EEG filtering methods have been proposed as an alternative means of correction based on temporal principal components analysis (PCA) or independent components analysis (ICA).86,87,9195

Summary
The simultaneous recording of EEG and fMRI is a useful noninvasive research method uniquely capable of revealing hemodynamic changes related to epileptic events, often suggestive of a network of pathological activity in both focal and generalized epilepsies. The techniques evaluation as a clinical tool will require further validation studies in patients undergoing presurgical evaluation, and it may also have a key role in furthering the understanding of epileptic networks. REFERENCES
1. Symms MR, Allen PJ, Woermann FG, et al. Reproducible localization of interictal epileptiform discharges using EEG-triggered fMRI. Phys Med Biol. 1999;44:N161-N168. 2. Warach S, Ives JR, Schlaug G, et al. EEG-triggered echo-planar functional MRI in epilepsy. Neurology. 1999;47:89-93. 3. Seeck M, Lazeyras F, Michel CM, et al. Non-invasive epileptic focus localization using EEG-triggered functional MRI and electromagnetic tomography. Electroencephalography and Clin Neurophysiol. 1998;106:508-512. 4. Lazeyras F, Blanke O, Perrig S, et al. EEG-triggered functional MRI in patients with pharmacoresistant epilepsy. J Magn Reson Imaging. 2000;12:177-185. 5. Krakow K, Woermann FG, Symms MR, et al. EEG-triggered functional MRI of interictal epileptiform activity in patients with partial seizures. Brain. 1999;122(Pt 9):1679-1688. 6. Patel MR, Blum A, Pearlman JD, et al. Echo-planar functional MR imaging of epilepsy with concurrent EEG monitoring. Am J Neuroradiol. 1999;20:1916-1919. 7. Ja ger L, Werhahn KJ, Hoffmann A, et al. Focal epileptiform activity in the brain: detection with spike-related functional MR imagingpreliminary results. Radiology. 2002;223:860-869. 8. Krakow K, Lemieux L, Messina D, et al. Spatio-temporal imaging of focal interictal epileptiform activity using EEG-triggered functional MRI. Epileptic Disord. 2001;3:67-74. 9. Allen PJ, Josephs O, Turner R. A method for removing imaging artifact from continuous EEG recorded during functional MRI. Neuroimage. 2000;12:230-239. 10. Lemieux L, Salek-Haddadi A, Josephs O, et al. Event-related fMRI with simultaneous and continuous EEG: description of the method and initial case report. Neuroimage. 2001;14:780-787. 11. Salek-Haddadi A, Lemieux L, Merschhemke M, Diehl B, Allen PJ, Fish DR. EEG quality during simultaneous functional MRI of interictal epileptiform discharges. Magn Reson Imaging. 2003;21:1159-1166. 12. Salek-Haddadi A, Diehl B, Hamandi K, et al. Hemodynamic correlates of epileptiform discharges: an EEG-fMRI study of 63 patients with focal epilepsy. Brain Res. 2006;1088:148-166. nar CG, Gross DW, et al. fMRI activation in continuous and spike-triggered EEG-fMRI 13. Al Asmi A, Be studies of epileptic spikes. Epilepsia. 2003;44:1328-1339. 14. Liston AD, De Munck JC, Hamandi K, et al. Analysis of EEG-fMRI data in focal epilepsy based on automated spike classification and signal space projection. Neuroimage. 2006;31:1015-1024.

6  EEG-Correlated fMRI in Epilepsy 15. Friston KJ, Frith CD, Frackowiak RSJ, Turner R. Characterizing dynamic brain responses with fMRI: a multivariate approach. Neuroimage. 1995;2:166-172. nar CG, Gross DW, Wang Y, et al. The BOLD response to interictal epileptiform discharges. 16. Be Neuroimage. 2002;17:1182-1192. 17. Salek-Haddadi A, Friston KJ, Lemieux L, Fish DR. Studying spontaneous EEG activity with fMRI. Brain Res Brain Res Rev. 2003;43:110-133. 18. Gotman J, Kobayashi E, Bagshaw AP, Benar CG, Dubeau F. Combining EEG and fMRI: a multimodal tool for epilepsy research. J Magn Reson Imaging. 2006;23:906-920. 19. Lemieux L, Laufs H, Carmichael D, Paul JS, Walker MC, Duncan JS. Noncanonical spike-related BOLD responses in focal epilepsy. Hum Brain Mapp. 2008;29(3):329-345. 20. Kang JK, Benar CG, Al Asmi A, et al. Using patient-specific hemodynamic response functions in combined EEG-fMRI studies in epilepsy. Neuroimage. 2003;20:1162-1170. 21. Hawco CS, Bagshaw AP, Lu Y, Dubeau F, Gotman J. BOLD changes occur prior to epileptic spikes seen on scalp EEG. Neuroimage. 2007;35:1450-1458. 22. Moeller F, Siebner HR, Wolff S, et al. Changes in activity of striato-thalamo-cortical network precede generalized spike wave discharges. Neuroimage. 2008;39(4):1839-1849. 23. Stefanovic B, Warnking JM, Kobayashi E, et al. Hemodynamic and metabolic responses to activation, deactivation and epileptic discharges. Neuroimage. 2005;28:205-215. 24. Hamandi K, Laufs H, Noth U, Carmichael DW, Duncan JS, Lemieux L. BOLD and perfusion changes during epileptic generalised spike wave activity. Neuroimage. 2008;39:608-618. 25. Federico P, Abbott DF, Briellmann RS, Harvey AS, Jackson GD. Functional MRI of the pre-ictal state. Brain. 2005;128:1811-1817. 26. Ma kiranta M, Ruohonen J, Suominen K, et al. BOLD signal increase precedes EEG spike activitya dynamic penicillin induced focal epilepsy in deep anesthesia. Neuroimage. 2005;27:715-724. 27. Rosenow F, Lu ders H. Presurgical evaluation of epilepsy. Brain. 2001;124:1683-1700. 28. Benar CG, Grova C, Kobayashi E, et al. EEG-fMRI of epileptic spikes: concordance with EEG source localization and intracranial EEG. Neuroimage. 2006;30:1161-1170. 29. Lemieux L, Krakow K, Fish DR. Comparison of spike-triggered functional MRI BOLD activation and EEG dipole model localization. Neuroimage. 2001;14:1097-1104. 30. Bagshaw AP, Kobayashi E, Dubeau F, Pike GB, Gotman J. Correspondence between EEG-fMRI and EEG dipole localisation of interictal discharges in focal epilepsy. Neuroimage. 2006;30:417-425. 31. Grova C, Daunizeau J, Kobayashi E, et al. Concordance between distributed EEG source localization and simultaneous EEG-fMRI studies of epileptic spikes. Neuroimage. 2008;39:755-774. 32. Boor R, Jacobs J, Hinzmann A, et al. Combined spike-related functional MRI and multiple source analysis in the non-invasive spike localization of benign rolandic epilepsy. Clin Neurophysiol. 2007;118:901-909. 33. Zijlmans M, Huiskamp G, Hersevoort M, Seppenwoolde JH, van Huffelen AC, Leijten FS. EEG-fMRI in the preoperative work-up for epilepsy surgery. Brain. 2007;130(Pt 9):2343-2353. 34. Najm IM, Tilelli CQ, Oghlakian R. Pathophysiological mechanisms of focal cortical dysplasia: a critical review of human tissue studies and animal models. Epilepsia. 2007;48(Suppl 2):21-32. 35. Kobayashi E, Bagshaw AP, Grova C, Gotman J, Dubeau F. Grey matter heterotopia: what EEG-fMRI can tell us about epileptogenicity of neuronal migration disorders. Brain. 2006;129:366-374. 36. Kobayashi E, Bagshaw AP, Gotman J, Dubeau F. Metabolic correlates of epileptic spikes in cerebral cavernous angiomas. Epilepsy Res. 2007;73:98-103. 37. Diehl B, Salek-Haddadi A, Fish DR, Lemieux L. Mapping of spikes, slow waves, and motor tasks in a patient with malformation of cortical development using simultaneous EEG and fMRI. Magn Reson Imaging. 2003;21:1167-1173. 38. Federico P, Archer JS, Abbott DF, Jackson GD. Cortical/subcortical BOLD changes associated with epileptic discharges: an EEG-fMRI study at 3 T. Neurology. 2005;64:1125-1130. 39. Salek-Haddadi A, Lemieux L, Fish DR. Role of functional magnetic resonance imaging in the evaluation of patients with malformations caused by cortical development. Neurosurg Clin N Am. 2002;13:63-69. 40. Jacobs J, Rohr A, Moeller F, et al. Evaluation of epileptogenic networks in children with tuberous sclerosis complex using EEG-fMRI. Epilepsia. 2008;49(5):816-825. 41. Cross JH. Epilepsy surgery in childhood. Epilepsia. 2002;43(Suppl 3):65-70. 42. De Tiege X, Laufs H, Boyd SG, et al. EEG-fMRI in children with pharmacoresistant focal epilepsy. Epilepsia. 2007;48:385-389. 43. Jacobs J, Jacobs J, Boor R, et al. Localization of epileptic foci in children with focal epilepsies using 3-tesla simultaneous EEG-fMRI recordings. Clin Neurophysiol. 2007;118:e50-e51.

81

82

THE EPILEPSIES 3 44. Jacobs J, Kobayashi E, Boor R, et al. Hemodynamic responses to interictal epileptiform discharges in children with symptomatic epilepsy. Epilepsia. 2007;48:2068-2078. 45. Ricci GB, De Carli D, Colonnese C, et al. Hemodynamic response (BOLD/fMRI) in focal epilepsy with reference to benzodiazepine effect. Magn Reson Imaging. 2004;22:1487-1492. 46. Lengler U, Kafadar I, Neubauer BA, Krakow K. FMRI correlates of interictal epileptic activity in patients with idiopathic benign focal epilepsy of childhood: a simultaneous EEG-functional MRI study. Clin Neurophysiol. 2007;118:e70. 47. Jackson GD, Opdam HI. Ictal fMRI: methods and models. Adv Neurol. 2000;83:203-211. 48. Detre JA, Sirven JI, Alsop DC, OConnor MJ, French JA. Localization of subclinical ictal activity by functional magnetic resonance imaging: correlation with invasive monitoring. Ann Neurol. 1995;38:618-624. 49. Salek-Haddadi A, Merschhemke M, Lemieux L, Fish DR. Simultaneous EEG-correlated ictal fMRI. Neuroimage. 2002;16:32-40. 50. Kobayashi E, Hawco CS, Grova C, Dubeau F, Gotman J. Widespread and intense BOLD changes during brief focal electrographic seizures. Neurology. 2006;66:1049-1055. 51. Baumgartner C, Serles W, Leutmezer F, et al. Preictal SPECT in temporal lobe epilepsy: regional cerebral blood flow is increased prior to electroencephalography-seizure onset. J Nucl Med. 1998;39:978-982. 52. Kobayashi E, Bagshaw AP, Benar CG, et al. Temporal and extratemporal BOLD responses to temporal lobe interictal spikes. Epilepsia. 2006;47:343-354. 53. Laufs H, Hamandi K, Salek-Haddadi A, Kleinschmidt AK, Duncan JS, Lemieux L. Temporal lobe interictal epileptic discharges affect cerebral activity in default mode brain regions. Hum Brain Mapp. 2007;28(10):1023-1032. 54. Raichle ME, MacLeod AM, Snyder AZ, Powers WJ, Gusnard DA, Shulman GL. A default mode of brain function. PNAS. 2001;98:676-682. 55. Laufs H, Kleinschmidt A, Beyerle A, et al. EEG-correlated fMRI of human alpha activity. Neuroimage. 2003;19:1463-1476. 56. Friston KJ, Harrison L, Penny W. Dynamic causal modelling. Neuroimage. 2003;19:1273-1302. 57. Laufs H, Krakow K, Sterzer P, et al. Electroencephalographic signatures of attentional and cognitive default modes in spontaneous brain activity fluctuations at rest. Proc Natl Acad Sci U.S.A. 2003;100:11053-11058. 58. Mantini D, Perrucci MG, Del Gratta C, Romani GL, Corbetta M. Electrophysiological signatures of resting state networks in the human brain. PNAS. 2007;104:13170-13175. 59. Laufs H, Hamandi K, Walker MC, et al. EEG-fMRI mapping of asymmetrical delta activity in a patient with refractory epilepsy is concordant with the epileptogenic region determined by intracranial EEG. Magn Reson Imaging. 2006;24:367-371. 60. Daunizeau J, Grova C, Marrelec G, et al. Symmetrical event-related EEG/fMRI information fusion in a variational Bayesian framework. Neuroimage. 2007;36:69-87. 61. Morgan VL, Gore JC, Abou-Khalil B. Cluster analysis detection of functional MRI activity in temporal lobe epilepsy. Epilepsy Res. 2007;76:22-33. 62. Hamandi K, Salek Haddadi A, Liston A, Laufs H, Fish DR, Lemieux L. fMRI temporal clustering analysis in patients with frequent interictal epileptiform discharges: comparison with EEG-driven analysis. Neuroimage. 2005;26:309-316. 63. Rodionov R, De Martino F, Laufs H, et al. Independent component analysis of interictal fMRI in focal epilepsy: comparison with general linear model-based EEG-correlated fMRI. Neuroimage. 2007;38:488-500. 64. Salek-Haddadi A, Lemieux L, Merschhemke M, Friston KJ, Duncan JS, Fish DR. Functional magnetic resonance imaging of human absence seizures. Ann Neurol. 2003;53:663-667. 65. Baudewig J, Bittermann HJ, Paulus W, Frahm J. Simultaneous EEG and functional MRI of epileptic activity: a case report. Clin Neurophysiol. 2001;112:1196-1200. 66. Archer JS, Abbott DF, Waites AB, Jackson GD. fMRI deactivation of the posterior cingulate during generalized spike and wave. Neuroimage. 2003;20:1915-1922. 67. Aghakhani Y, Bagshaw AP, Benar CG, et al. fMRI activation during spike and wave discharges in idiopathic generalized epilepsy. Brain. 2004;127:1127-1144. 68. Gotman J, Grova C, Bagshaw A, Kobayashi E, Aghakhani Y, Dubeau F. Generalized epileptic discharges show thalamocortical activation and suspension of the default state of the brain. PNAS. 2005;102:15236-15240. 69. Hamandi K, Salek-Haddadi A, Laufs H, et al. EEG-fMRI of idiopathic and secondarily generalized epilepsies. Neuroimage. 2006;31:1700-1710.

6  EEG-Correlated fMRI in Epilepsy 70. Aghakhani Y, Kobayashi E, Bagshaw AP, et al. Cortical and thalamic fMRI responses in partial epilepsy with focal and bilateral synchronous spikes. Clin Neurophysiol. 2006;117:177-191. 71. Cavanna AE, Trimble MR. The precuneus: a review of its functional anatomy and behavioural correlates. Brain. 2006;129:564-583. 72. Meeren HKM, Pijn JP, Van Luijtelaar ELJM, Coenen AML, Lopes da Silva FH. Cortical focus drives widespread corticothalamic networks during spontaneous absence seizures in rats. J Neurosci. 2002;22:1480-1495. 73. Steriade M, Dossi RC, Nunez A. Network modulation of a slow intrinsic oscillation of cat thalamocortical neurons implicated in sleep delta waves: cortically induced synchronization and brainstem cholinergic suppression. J Neurosci. 1991;11:3200-3217. 74. Avoli M, Gloor P. Interaction of cortex and thalamus in spike and wave discharges of feline generalized penicillin epilepsy. Exp Neurol. 1982;76:196-217. 75. Lemieux L, Allen PJ, Franconi F, Symms MR, Fish DR. Recording of EEG during fMRI experiments: patient safety. Magn Reson Med. 1997;38:943-952. 76. Konings MK, Bartels LW, Smits HF, Bakker CJ. Heating around intravascular guidewires by resonating RF waves. J Magn Reson Imaging. 2000;12:79-85. 77. Ives JR, Warach S, Schmitt F, Edelman RR, Schomer DL. Monitoring the patients EEG during echo planar MRI. Electroencephalogr Clin Neurophysiol. 1993;87:417-420. 78. Allen PJ, Polizzi G, Krakow K, Fish DR, Lemieux L. Identification of EEG events in the MR scanner: the problem of pulse artifact and a method for its subtraction. Neuroimage. 1998;8:229-239. 79. Wendt RE III, Rokey R, Vick GW III, Johnston DL. Electrocardiographic gating and monitoring in NMR imaging. Magn Reson Imaging. 1988;6:89-95. 80. Bonmassar G, Schwartz DP, Liu AK, Kwong KK, Dale AM, Belliveau JW. Spatiotemporal brain imaging of visual-evoked activity using interleaved EEG and fMRI recordings. Neuroimage. 2001;13:1035-1043. 81. Bonmassar G, Purdon PL, Ja askela inen IP, et al. Motion and ballistocardiogram artifact removal for interleaved recording of EEG and EPs during MRI. Neuroimage. 2002;16:1127-1141. 82. Goldman RI, Stern JM, Engel J Jr., Cohen MS. Acquiring simultaneous EEG and functional MRI. Clin Neurophysiol. 2000;111:1974-1980. 83. Anami K, Mori T, Tanaka F, et al. Stepping stone sampling for retrieving artifact-free electroencephalogram during functional magnetic resonance imaging. Neuroimage. 2003;19:281-295. 84. Hoffmann A, Ja ger L, Werhahn KJ, Jaschke M, Noachtar S, Reiser M. Electroencephalography during functional echo-planar imaging: detection of epileptic spikes using post-processing methods. Magn Reson Med. 2000;44:791-798. 85. Sijbers J, Michiels I, Verhoye M, Van Audekerke J, Van der Linden A, Van Dyck D. Restoration of MR-induced artifacts in simultaneously recorded MR/EEG data. Magn Reson Imaging. 1999;17: 1383-1391. 86. Mantini D, Perrucci MG, Cugini S, Ferretti A, Romani GL, Del Gratta C. Complete artifact removal for EEG recorded during continuous fMRI using independent component analysis. Neuroimage. 2007;34:598-607. nar CG, Aghakhani Y, Wang Y, et al. Quality of EEG in simultaneous EEG-fMRI for epilepsy. 87. Be Clin Neurophysiol. 2003;114:569-580. 88. Negishi M, Abildgaard M, Nixon T, Constable RT. Removal of time-varying gradient artifacts from EEG data acquired during continuous fMRI. Clin Neurophysiol. 2004;115:2181-2192. 89. Niazy RK, Beckmann CF, Iannetti GD, Brady JM, Smith SM. Removal of fMRI environment artifacts from EEG data using optimal basis sets. Neuroimage. 2005;28:720-737. 90. Mandelkow H, Halder P, Boesiger P, Brandeis D. Synchronization facilitates removal of MRI artifacts from concurrent EEG recordings and increases usable bandwidth. Neuroimage. 2006;32:1120-1126. 91. Scarff CJ, Reynolds A, Goodyear BG, Ponton CW, Dort JC, Eggermont JJ. Simultaneous 3-T fMRI and high-density recording of human auditory evoked potentials. Neuroimage. 2004;23:1129-1142. 92. Otzenberger H, Gounot D, Foucher JR. Optimisation of a post-processing method to remove the pulse artifact from EEG data recorded during fMRI: an application to P300 recordings during e-fMRI. Neurosci Res. 2007;57:230-239. 93. Otzenberger H, Gounot D, Foucher JR. P300 recordings during event-related fMRI: a feasibility study. Brain Res Cogn Brain Res. 2005;23:306-315. 94. Briselli E, Garreffa G, Bianchi L, et al. An independent component analysis-based approach on ballistocardiogram artifact removing. Magn Reson Imaging. 2006;24:393-400. 95. Srivastava G, Crottaz-Herbette S, Lau KM, Glover GH, Menon V. ICA-based procedures for removing ballistocardiogram artifacts from EEG data acquired in the MRI scanner. Neuroimage. 2005;24:50-60.

83

THE EPILEPSIES 3

Epilepsy and Sleep


MI SOHEYL NOACHTAR  JAN RE

Introduction Physiology and Diagnostic Studies of Sleep Effects of Epilepsy on Sleep Sleep in Patients with Generalized Epilepsies Sleep in Patients with Focal Epilepsies Effects of Sleep and Sleep Deprivation on Epilepsy

Relation of Epileptiform Discharges to the Sleep-Wake Cycle and Sleep Stages in Generalized Epilepsy Relations of Epileptic Seizures and Sleep in Generalized Epilepsies Focal Epilepsy in Sleep Sleep Deprivation Effects of Epilepsy Treatment on Sleep Sleep Disorders and Epilepsy

Introduction
Epilepsy and sleep are related in several aspects, from physiology and pathophysiology to clinical appearance and neurophysiological diagnosis to treatment. It has been known since the days of Aristotle that epileptic seizures may occur exclusively during sleep.1 Later, several authors observed that certain epileptic seizures may show a tendency to occur during sleep or waking periods and recognized the relation to the sleepwake cycle.28 Also, epileptiform discharges occur more frequently during non-REM sleep than in REM sleep and waking periods,911 and arousals and transitions periods between sleep stages are considered to facilitate epileptiform discharges.1215 Antiepileptic drugs have effects on sleep, and it has even been speculated that some of their antiepileptic effect is due to their effect on sleep.16 In this chapter, we will discuss the relationships between epilepsy and sleep, the use and ambiguity of sleep in the diagnosis of epilepsy, and the effects of treatment on epilepsy and sleep.

Physiology and Diagnostic Studies of Sleep


Sleep is a complex behavioral trait, in which consciousness and responsiveness to environmental stimuli are reduced. There is an ongoing discussion as to why we

84

7  Epilepsy and Sleep

85

sleep, but some physiologic functions, such as memory consolidation,17 have been shown to be dependent on sleep. The pressure on the organism to initiate sleep depends largely on circadian and homeostatic influences.18 Several brain regions are actively involved in the sleep process. Commonly, the cortical activity, and with it the sleep electroencephalography (EEG), is described as synchronous in a periodical fashion. The synchronicity refers in particular to the cortical neurons.19 Deeper brain structures such as the midbrain and brainstem involved in this active process do not necessarily reflect the same level of synchronicity as the cortex. There are two general different states of sleep, rapid eye movement sleep (REM sleep) and nonrapid eye movement sleep (NREM sleep), with the latter being divided further into four stages.20 Over the course of a night, NREM and REM sleep alternate with an approximately 90-minute rhythmicity, creating four to six sleep cycles. The NREM stages differ in their amount of slow-wave sleep and special defining patterns like K-complexes and sleep spindles. Electrophysiologically, the dynamic of NREM sleep can be understood as the neurobehavioral attempt to reach a high level of synchrony, expressed as slow-wave sleep and a deafferentiation of sensory input.21 REM sleep is defined by fast eye movements, loss of muscle tone, and an EEG desynchrony. These sleep oscillations are mainly generated by the corticothalamic system, the reticular thalamic and thalamocortical circuit, and thalamically projecting brainstem structures.22 This difference in cellular and systemic neuronal activity is important for the difference in occurrence of seizures and interictal epileptiform discharges during sleep. During sleep, external (for example, environmental noises) and internal stimuli (for example, sleep apnea episodes) may evoke arousal reactions. In the arousals, an adequate reaction to the stimulus and the attempt to stay asleep are competing impulses. A night with many arousals will be not as restful as a night with few arousals. In the classic sleep evaluation according to the criteria of Rechtschaffen and Kales20 an arousal is an event of fast, desynchronized EEG discharges. Terzano and colleagues23 described a 20- to 40-second rhythmicity of changes in the EEG as the cyclic alternating pattern (CAP), defined by an alternation of phasic EEG events and background activity. Now CAP is considered an expression of different arousal mechanisms that are expressed differentially in the various phases of sleep and in sleep stage transitions.24 During arousals and during stage shifts, interictal epileptic discharges and seizures are more likely to develop, especially in generalized epilepsies.13,25,26 In evaluating and scoring sleep, polysomnography is the specific electrophysiological diagnostic procedure. The EEG, an electrooculogram (EOG), and a chin electromyograph (EMG) are needed for sleep-stage scoring. For more sleep specific diagnostics, respiratory parameters, ECG, and extremity EMG are derived. For polysomnography, an EEG derivation with the full 10 to 20 placed electrodes is usually only performed for research purposes or in the differential diagnosis of epilepsy and sleep disorders. The conventional EEG scoring requires only one EEG channel (electrodes C3A2 or C4A1). Despite not incorporating all the derivations needed for polysomnography, routine EEG or EEG-video monitoring will already enable the diagnosis of all sleep stages, although with some ambiguity. The prefrontal leads (Fp1 and Fp2) and the lateral temporal leads (FT9 and FT10 or T7 and T8) can be used to interpret eye movements. An epilepsy-focused EEG will allow the evaluation of sleepiness (eye rolling in the frontal leads, temporal theta of drowsiness), falling asleep (change of background, vertex waves, and posterior sharp transients of sleep [POSTS]), and

86

THE EPILEPSIES 3

sleep stages 2, 3, and 4 (K-complexes, sleep spindles in stage 2, increase of delta activity in stages 3 and 4). REM sleep can be detected by fast eye movements and the low-amplitude beta and gamma activity in the EEG, but its exact beginning may be obscured without an EMG recording of the chin muscles. The formal scoring of conventional sleep stages requires 30-second epochs starting with the lights-off signal.20 Excessive daytime sleepiness can be assessed by the multiple sleep latency test (MLST), which provides more objective data than questionnaires.27 The MLST is usually performed on a day following a polysomnography to provide accurate documentation of the preceding nights sleep.

Effects of Epilepsy on Sleep


Epilepsy patients typically report poorer sleep than healthy controls,28 and it seems that sleep disturbance affects their quality of life.29,30 The prevalence is higher in patients with comorbid anxiety and depression.30 SLEEP IN PATIENTS WITH GENERALIZED EPILEPSIES Janz16 described that patients with awakening epilepsy and those with sleep epilepsy had distinct sleep patterns and sleep habits. According to that study, patients with generalized epilepsy like to stay up late in the evening and often have difficulty falling asleep. Their sleep appeared to be disrupted. In the morning they feel drowsy and unrefreshed, and they prefer to get up late if they can.6,16 Early polygraphic studies seemed to support this concept,31,32 but the sleep EEG recordings were discontinuous in one study,32 which does not allow interpretation of the sleep structure. Other polysomnographic sleep studies in patients with idiopathic generalized epilepsy differentiated more and found normal sleep patterns unless seizures occurred during the night.33-35 These studies were performed either on patients with chronic antiepileptic medication or on patients in whom the drugs (usually phenobarbital and phenytoin) were discontinued a few days prior to the sleep investigations. Thus, chronic drug effects or rebound effects after discontinuation, which may influence the results, cannot be excluded. Only one study36 investigated the night sleep of unmedicated epilepsy patients. Adaptation nights to the sleep lab were also included in this study36 to avoid first-night effects.37 Photosensitive patients with generalized epilepsy had significantly more deep sleep (sleep stages 3 and 4) and less light sleep (sleep stages 1 and 2) than the other patients with generalized epilepsy.36 However, this effect was no more significant if the patients were age matched, as the photosensitive patients were younger, and younger patients tend to have more deep sleep. Thus, the systematic polysomnographic evaluation of unmedicated patients with generalized epilepsy36 did not reveal distinct sleep patterns in patients with awakening and sleep epilepsy, as hypothesized by Janz6 based on unstructured interviews. SLEEP IN PATIENTS WITH FOCAL EPILEPSIES Patients with focal epilepsy have poorer sleep as compared to controls in many categories: epilepsy patients sleep less, have a poorer quality of sleep, and

7  Epilepsy and Sleep

87

report a disturbed sleep.30 This was found not only for patients with catastrophic epilepsy, but also already for patients who are relatively well controlled on 1 or 2 AED.29 Deep sleep is reduced and there are more arousals and more stage shifts,38 especially in nights with seizures, but also in seizure-free nights. REM sleep is decreased if seizures occur during the night as well as during the preceding day. The earlier the seizures appear during the night, the shorter the total REM time,39 but the amount of REM sleep increases with better seizure control.40 Interictal epileptiform discharges (IEDs) alone seem to have little effect on the sleep architecture.41

Effects of Sleep and Sleep Deprivation on Epilepsy


Langdon-Down and Brain5 were the first to subdivide epilepsy patients according to the occurrence of their seizures. They described (1) a diurnal type, whose seizure occurred predominantly during the day with a maximum following morning awakening and two smaller peaks in the afternoon; (2) a nocturnal type, in whom seizures occurred during the night with maxima shortly after falling asleep and in the early morning hours; and (3) a group without any discernible pattern (diffuse type).5 Relations of epileptic seizures to the sleepwake cycle were evident from the observation that the times at which the seizures occurred changed when the sleep regimen was altered.3,42 The circadian sleepwake cycle also significantly influences the occurrence of epileptiform discharges.11 During non-REM sleep, generalized epileptiform discharges are more frequent than during waking.9,10,43 For this reason, sleep EEGs are performed on patients in whom the diagnosis of epilepsy is not established or the epilepsy syndrome is unclear and the waking EEG is unrevealing. RELATION OF EPILEPTIFORM DISCHARGES TO THE SLEEP-WAKE CYCLE AND SLEEP STAGES IN GENERALIZED EPILEPSY Generalized epileptiform discharges gradually increase with deepening of non-REM sleep34,44 In patients with absence epilepsy, the lowest discharge rates were found during REM sleep.34 Because deep sleep stages are most pronounced during the first sleep cycle, it is not surprising that the highest rate of interictal epileptiform discharges were found during the first sleep cycle.34 The morphology of generalized spike-wave complexes is more irregular during non-REM sleep and is similar to waking and REM sleep.34,35,44 The early hypothesis8 that transitional states between wake and sleep and vice versa may be epileptogenic in selected patients was supported by polygraphic studies. Several authors demonstrated the facilitating effects of transitional states of sleep such as sleep onset, changes between sleep stages, and arousals on the occurrence of epileptiform discharges in patients with absence epilepsy.12,33,35 Niedermeyer45 emphasized that an abnormal paroxysmal response to arousal and the influx of upward traveling stimuli seem to be the most important epileptogenic mechanism in primary generalized epilepsy. The concept of CAP, which is based on these observations, provides a new approach and supports the idea that transitional cyclic sleep patterns activate epileptiform discharges.13

88

THE EPILEPSIES 3

RELATIONS OF EPILEPTIC SEIZURES AND SLEEP IN GENERALIZED EPILEPSIES Generalized tonic-clonic seizures show a clearer relation to the sleepwake cycle than minor seizures.46 However, myoclonic seizures tend to occur predominantly in the early morning hours shortly after awakening from night sleep in juvenile myoclonic epilepsy.14,47 Based on the patients recollection of seizures and the patients history, Janz46 classified epilepsies according to the time of occurrence of the grand mal seizures. He coined the term awakening epilepsy for patients whose generalized tonic-clonic seizures predominantly occur in the first 2 hours after awakening, with a second peak in the afternoon.46 Janz6 reported earlier that in 45% of his 2110 patients with grand mal epilepsy, seizures occurred predominantly during sleep, in 34% during the first 2 hours after awakening from night sleep, and in 21% no relation to the sleepwake cycle was found. During the course of the epilepsy, the pattern may change as less patients have an awakening predominance (31%) and more patients show a diffuse pattern (26%).16 The awakening type seemed to be associated with idiopathic generalized epilepsy.46 Half of Janzs patients with hereditary idiopathic epilepsy belonged to the awakening group, whereas most patients with focal epilepsy secondary to brain tumors (53%) had a diffuse pattern of seizure occurrence, and only 8% showed an awakening pattern.46 The minor seizures in patients with epilepsy on awaking comprised predominantly absence seizures (94%) and myoclonic seizures (96%) and less frequently psychomotor seizures (16%) or Jacksonian seizures (9%).46 A major bias of the study of is the observation that the recollection of epileptic seizures by the patients as well as caregivers is unreliable.48 Also, the etiological diagnoses are subject to uncertainties in the study of Janz46 because at that time no modern imaging studies were available. Few recent studies dealt with the relation of the occurrence of generalized tonic-clonic seizures to the sleepwake cycle.49,50 The distribution of generalized tonic-clonic seizures was as follows in 77 of 141 patients with generalized epilepsy, in whom generalized tonic-clonic were observed:49 (1) 16.8% had seizures after awakening, (2) 36.3% had seizures during waking hours of the day, (3) 28.5% had seizures during night sleep, and (4) 18.1% had seizures both during night sleep and during waking periods of the day. Morning and nocturnal awakenings accounted for 36 of 51 seizures in 33 patients with juvenile myoclonic epilepsy, whereas the evening relaxation period (n = 6), sleep onset (n = 3), and sleep (n = 6) were less frequently associated with seizures.50 RELATIONS OF FOCAL EPILEPSY AND SLEEP In focal epilepsies, the rate of IED is increased in sleep, with the highest increase in deep NREM sleep (sleep stages 3 and 4) and a decrease in REM sleep.51-53 The localization of IED is more widespread in NREM sleep and has the best localizing value if occurring in REM sleep.52 An important mechanism in activating IED is the activity changes associated with arousals.15 In frontal lobe epilepsy, seizures have a tendency to occur more commonly during night sleep than in other focal epilepsies such as temporal lobe epilepsy54 (Figure 7-1). During sleep, seizures arising from frontal lobes evolve less often into generalized tonic-clonic seizures (22% of the seizures) than do seizures arising from

7  Epilepsy and Sleep

89

Figure 71

Ictal EEG recording in a 25-year-old patient with medically refractory focal epilepsy who underwent continuous EEG video monitoring for presurgical evaluation. The 36-channel montage referenced to the left mastoid (TP9) shows a right frontal seizure pattern that starts right after a sleep spindle.

the temporal lobes (45%), whereas during wakefulness, the percentage of generalization was not statistically different (20% vs. 19%).55 Also, the seizures mostly arose from NREM stages 1 and 2, rarely from deep sleep, and almost never from stage REM. Figure 7-1 shows the onset of a right frontal EEG seizure pattern following a sleep spindle. The patients MRI (Figure 7-2) depicts a right frontal focal cortical dysplasia. Subclinical seizures were especially prevalent during the night. In sleep, a higher level of synchronization is reached, which explains the smaller number of seizures during deep sleep but is also compatible with the increased portion of secondary generalization.55 An important point for differential diagnosis is that psychogenic nonepileptic seizures do not arise directly from sleep.55 A period of wakefulness after awakening always precedes the psychogenic nonepileptic seizures if they occur during sleep.55 SLEEP DEPRIVATION Sleep deprivation is one of the most potent precipitators of epileptic seizures and epileptiform discharges, especially in patients with generalized epilepsy. Janz6 reported that sleep deprivation and/or excessive alcohol consumption precipitated the first seizure in 28 of 47 patients with juvenile myoclonic epilepsy. Based on these observations, most neurologists counsel epilepsy patients to avoid sleep deprivation. However, there was no seizure-precipitating effect observed in a study of temporal lobe epilepsy patients who underwent sleep deprivation during EEG-video

90

THE EPILEPSIES 3

Figure 72

MRI (T1 inversion recovery technique) of the patient, whose ictal EEG is shown in Figure 7-1. The continuity of the right superior frontal gyrus is interrupted by an aberrant sulcus, including a focal cortical dysplasia (FCD). The FCD collocates well with the EEG seizure pattern.

monitoring for epilepsy surgery evaluation.56 Thus, it is likely that generalized epilepsies and focal epilepsies are differently sensitive to sleep deprivation. Sleep deprivation also facilitates epileptiform discharges in patients with generalized epilepsy.6,57,58 However, there is some debate on whether sleep deprivation has a genuine activating effect on epileptiform discharges or whether it acts through sleep induction.59 It has been shown that the IED yield in patients with clinically highly probable epilepsy but negative previous routine EEGs (that contained both wake and sleep) could be increased by 52% after sleep deprivation.60 IED were recorded in some of these patients in wake or sleep stage only.60 Thus, it is important that the EEG after sleep deprivation includes both sleep and wake phases. The effect of sleep deprivation on the occurrence of IED in the EEG was also shown in another study that directly compared drug-induced sleep EEG and EEG after sleep deprivation in patients with normal or borderline waking EEG. This study revealed epileptic discharge activation in 44% after sleep deprivation versus 14% during druginduced sleep.61 Another study compared the rate of epileptiform discharges of waking EEGs after sleep deprivation with sleep EEG after sleep deprivation and concluded that sleep deprivation had an independent activating effect because after sleep deprivation, the waking EEGs showed more epileptiform discharges than the sleep EEGs.62 An increased cortical excitability following sleep deprivation, which could be a reason for increased epileptic discharges, has also been demonstrated by transcranial magnetic stimulation in patients with focal and generalized epilepsies.63

7  Epilepsy and Sleep

91

Effects of Epilepsy Treatment on Sleep


The mainstay of treatment in epilepsy is by antiepileptic drugs (AEDs). Many AEDs exert some effects on sleep, most obviously those that have hypnotic properties.64 One has to be careful, though, when trying to interpolate AED effects on sleep from the effect on vigilance. Side effects like drowsiness do not translate into quicker and better sleep. Furthermore, a normal macrostructure with 4 to 6 macrocycles of non-REM and REM sleep does not necessarily imply a good quality of sleep, as sleep quality also depends on arousals and the microstructure of sleep.65 It was hypothesized that some of the antiepileptic effects may even be mediated through an influence on sleep.16 Most studies evaluated short-term effects of the drugs (often with healthy volunteers) on patients (mostly with psychiatric disorders) on chronic treatment, who were then compared with what was considered normal sleep.36 The older, established AEDs have been studied the most and also have more longterm data available. Phenobarbital (PHB) is still a widely used AED, especially in developing countries. Its potency and low cost are its great advantages, but its use is limited in more affluent societies due to enzyme induction and mostly to increased tiredness, its main subjective side effect. Phenobarbital changes the sleep macrostructure by reducing sleep onset latency and accentuating the distribution of deep sleep to the first half of the night and REM sleep to the second part of the night. It also reduces the number of periods of waking and movement time after sleep onset.66 PHB reduces the number of REM interruptions only in patients with generalized epilepsy (including patients with awakening epilepsy), which is an interesting finding with respect to the observation that transitions between sleep stages seem to provoke seizure discharges in these patients.12,33,49 The effect of PHB differs in patients with generalized and focal epilepsies, whereas in patients with generalized epilepsy, sleep stages 1 to 3 were decreased, and stages 2 to 4 in patients with focal epilepsy were increased.66 Phenytoin (PHT) is another classical AED. After starting PHT treatment, patients fell asleep more rapidly, deep sleep was increased, and light sleep was decreased when compared to the unmedicated baseline. These effects were more pronounced in the second part of the night. In contrast to the effects of PHB, the amount and the interruptions of REM-sleep remained unchanged with PHT.66 The comparison of patients with generalized and focal epilepsy revealed only minor insignificant differences. A long-term study (at least 6 months) on PHT showed that the aforementioned increase of deep sleep and decrease of light sleep returned to baseline values, and only the reduced sleep latency was lasting.36 The use of ethosuximide brought increased light sleep (stage 1), decreased deep sleep (stages 3 and 4), and prolonged REM sleep in the first sleep cycle as compared with unmedicated baseline in patients with generalized epilepsy.67 Valproic acid use led to an increase of light sleep stage 1, but no deep sleep changes occurred. As with ethosuximide, the first sleep cycle was prolonged.67 The effects of valproic acid and ethosuximide on the first sleep cycle are maybe related to the fact that these drugs were evaluated in patients with generalized epilepsy, and these patients showed similar influences of PHT and PHB on the first sleep cycle.67 Carbamazepine is one of the most widely used AEDs, and its effects on sleep are well documented. The effects differ in acute and long-term treatment. In the first

92

THE EPILEPSIES 3

days of treatment, deep sleep and sleep efficiency are increased, but REM sleep is decreased, an effect that is most pronounced after the first doses. In long-term therapy, the sleep macrostructure of patients treated with carbamazepine was closer to the normal macrostructure than before treatment.68,69 For the newer AEDs, there are several studies on sleep, but in most, long-term data is missing. A single dose of levetiracetam increased sleep stage 2 but decreased stage 4 NREM sleep in epileptic patients.70 Gabapentin increases deep sleep and REM and decreases sleep stage 1, but did not change the subjective daytime sleepiness in healthy volunteers.71,72 For some of the newer AEDs, among them tiagabine, vigabatrin, and topiramate, a GABAergic mechanism of action is hypothesized, and so these drugs can be expected to have effects on sleep. For topiramate, though, no effect on sleep could be demonstrated.73 Tiagabine, on the other hand, improved sleep efficiency and increased deep sleep.74 In summary, AEDs have several effects on the sleep of patients with epilepsy. However, most drug effects on sleep are nonspecific or temporary. Based on our present knowledge on the sleep of patients with epilepsy, it appears that the antiepileptic effect of the drugs is only little, if at all, related to drug effects on sleep.

Sleep Disorders and Epilepsy


When patients report nightly seizures or seizurelike attacks, a wide field of possible differential diagnoses is opened. Epileptic seizures may occur either exclusively at night or predominantly at night. Several sleep disorders then have to be taken into consideration as differential diagnoses. Sleep myoclonusor hypnic jerksoccurs in the transition phase between wake and sleep and can be so strong that the sleeper and/or the bed partner awake.75 Sleep starts can more rarely appear as visual, auditory, or other sensory experiences and are then closely related to hypnagogic hallucinations, which are in turn probably only abnormal if occurring in connection with other sleep disorders, such as narcolepsy.76,77 Nightmares are frightening dreams that lead to awakening and are usually remembered by the dreamer. They have to be carefully weighted against epileptic seizures with fearful associations.78 Sleepwalking and pavor nocturnus are NREM parasomnias that lead to complex behavior and are often not remembered by the sleeper. Pavor nocturnus, which is a disorder of children, is also called night terrors because the children experience strong fear during an attack. The differential diagnosis between epileptic seizures associated with fear and sleep terrors may be difficult.79 Both disorders are fairly common.80 REM sleep behavior disorder (RBD) is a REM sleep parasomnia that is characterized by the loss of the REM muscle atonia81 and may be confused with frontal lobe seizures presenting with complex motor seizures. Patients act out their dreams in complex behavior; they often experience violent dreams and are mostly amnesic for these episodes. RBD precedes neurodegenerative disorders, especially multiple system atrophy and Parkinsons disease.82 Patients with narcolepsy have excessive daytime sleepiness and abnormal REM sleep, which is symptomatic by cataplexy (sudden loss of muscle tone, often after emotional triggers), sleep paralysis (inability to move for a short period of time on waking), or sleep-associated hallucinations.83

7  Epilepsy and Sleep

93

Primary sleep disorders are not only important differential diagnoses, but can also be more common in epilepsy patients. Obstructive sleep apnea (OSA) syndromes, for example, are more common in epilepsy patients.84 OSA in these patients was possibly related to medical treatment or AED-associated weight gain.85,86 OSA treatment was able to improve seizure control.87

Conclusion
An important goal of epilepsy treatment is the quality of life of patients. Disturbed sleep, insomnia, and daytime sleepiness rank highly among unfavorable effects of the disease and side effects of the treatment. Therefore, sleep-related issues should be considered when treating patients with epilepsy. It remains to be shown whether disturbed sleep in epileptic patients is caused by a disruption of physiological sleep mechanisms through epileptic mechanisms. However, AEDs should be favored that do not disturb sleep. REFERENCES
1. Passouant P. Historical aspects of sleep and epilepsy. Epilepsy Res Suppl. 1991;2:19-22. 2. Hopkins H. The time of appearance of epileptic seizures in relation to age, duration and type of the syndrome. J Nerv Ment Dis. 1993;77:153-162. 3. Gowers WR. Epilepsy and Other Chronic Convulsive Diseases. New York: William Wood; 1885. 4. Griffiths GN, Fox IT. Rhythm in epilepsy. Lancet. 1938;234:409-416. 5. Langdon-Down M, Brain WR. Time of day in relation to convulsions in epilepsy. Lancet. 1929;II:1029-1032. 6. Janz D. Aufwach-Epilepsien. Arch Psychiat Nervenkr. 1953;191:53-98. 7. Magnussen G. 18 cases of epilepsy with fits in relation to sleep. Acta Psychiat Scand. 1936;11:28.-3219. 8. Patry FL. The relation of time of day, sleep and other factors to the incidence of epileptic seizures. Am J Psychiat. 1931;10:789-813. 9. Gibbs EL, Gibbs FA. Diagnostic and localizing value of electroencephalic studies in sleep. Res Publ Ass Nerv Ment Dis. 1947;26:366-376. 10. Gloor P, Tsai C, Haddad F. An assessment of the value of sleep-electroencephalography for the diagnosis of temporal lobe epilepsy. Electroencephalogr Clin Neurophysiol. 1958;10:633-648. 11. Shouse MN, Martins da Silva A, Sammaritano M. Circadian rhythm, sleep, and epilepsy. J Clin Neurophysiol. 1996;13:32-50. 12. Halasz P. Sleep, arousal and electroclinical manifestations of generalized epilepsy with spike and wave pattern. Epilepsy Res Suppl. 1991;2:43-48. 13. Terzano MG, Parrino L, Anelli S, Halasz P. Modulation of generalized spike-and-wave discharges during sleep by cyclic alternating pattern. Epilepsia. 1989;30:772-781. 14. Gigli GL, Calia E, Marciani MG, et al. Sleep microstructure and EEG epileptiform activity in patients with juvenile myoclonic epilepsy. Epilepsia. 1992;33:799-804. 15. Parrino L, Halasz P, Tassinari CA, Terzano MG. CAP, epilepsy and motor events during sleep: the unifying role of arousal. Sleep Med Rev. 2006;10:267-285. 16. Janz D. Epilepsy and the sleeping-waking cycle. In: Vincken PJ, Bruyn GW, eds. Handbook of Clinical Neurology. Amsterdam: North Holland; 1974:457-490. 17. Stickgold R. Sleep-dependent memory consolidation. Nature. 2005;437:1272-1278. 18. Borbely AA. A two process model of sleep regulation. Hum Neurobiol. 1982;1:195-204. 19. Steriade M, McCormick DA, Sejnowski TJ. Thalamocortical oscillations in the sleeping and aroused brain. Science. 1993;262:679-685. 20. Rechtschaffen A, Kales A. A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects. Los Angeles: Brain Information Service/Brain Research Institute; 1968. 21. Steriade M, Contreras D, Curro Dossi R, Nunez A. The slow (< 1 Hz) oscillation in reticular thalamic and thalamocortical neurons: scenario of sleep rhythm generation in interacting thalamic and neocortical networks. J Neurosci. 1993;13:3284-3299.

94

THE EPILEPSIES 3 22. Steriade M. Grouping of brain rhythms in corticothalamic systems. Neuroscience. 2006;137: 1087-1106. 23. Terzano MG, Mancia D, Salati MR, Costani G, Decembrino A, Parrino L. The cyclic alternating pattern as a physiologic component of normal NREM sleep. Sleep. 1985;8:137-145. 24. Ferri R, Bruni O, Miano S, et al. The time structure of the cyclic alternating pattern during sleep. Sleep. 2006;29:693-699. 25. Shouse MN, Langer J, King A, et al. Paroxysmal microarousals in amygdala-kindled kittens: could they be subclinical seizures? Epilepsia. 1995;36:290-300. 26. Eisensehr I, Parrino L, Noachtar S, Smerieri A, Terzano MG. Sleep in Lennox-Gastaut syndrome: the role of the cyclic alternating pattern (CAP) in the gate control of clinical seizures and generalized polyspikes. Epilepsy Res. 2001;46:241-250. 27. Carskadon MA, Dement WC, Mitler MM, Roth T, Westbrook PR, Keenan S. Guidelines for the multiple sleep latency test (MSLT): a standard measure of sleepiness. Sleep. 1986;9:519-524. 28. Alanis-Guevara I, Pena E, Corona T, Lopez-Ayala T, Lopez-Meza E, Lopez-Gomez M. Sleep disturbances, socioeconomic status, and seizure control as main predictors of quality of life in epilepsy. Epilepsy Behav. 2005;7:481-485. 29. de Weerd A, de Haas S, Otte A, et al. Subjective sleep disturbance in patients with partial epilepsy: a questionnaire-based study on prevalence and impact on quality of life. Epilepsia. 2004;45: 1397-1404. 30. Xu X, Brandenburg NA, McDermott AM, Bazil CW. Sleep disturbances reported by refractory partial-onset epilepsy patients receiving polytherapy. Epilepsia. 2006;47:1176-1183. 31. Jovanovic UJ. Das Schlafverhalten der Epileptiker. I. Schlafdauer, Schlaftiefe und Besonderheiten der Schlafperiodik. Dtsch Z Nervenheilk. 1967;190:159-198. 32. Christian W. Schlaf-Wach-Periodik bei Schlaf- und Aufwachepilepsien. Nervenarzt. 1961;32: 266-275. 33. Passouant P, Besset A, Carriere A, Billiard M. Night sleep and generalized epilepsy. In: Koella WP, Levin P, eds. Sleep 1974. Basel: Karger; 1975:185-196. 34. Sato S, Dreifuss FE, Penry JK. The effect of sleep on spike-wave discharges in absence seizures. Neurology. 1973;23:1335-1345. 35. Tassinari CA, Bureau-Paillas M, Dalla Bernadina B, et al. Generalized epilepsies and sleep. A polygraphic study. In: Van Praag HM, Meinardi H, eds. Brain and Sleep. Amsterdam: De Erven Bohn; 1974:154-166. 36. Ro der-Wanner UU, Wolf P, Danninger T. Are sleep patterns in epileptic patients correlated with their type of epilepsy? In: Martins da Silva A, Binnie C, Meinradi H, eds. Biorhythms and Epilepsy. New York: Raven Press; 1985:109-121. 37. Agnew HW, Webb WB, Williams RL. The first night effect: an EEG study of sleep. Psychophysiology. 1966;2:263-266. 38. Touchon J, Baldy-Moulinier M, Billiard M. Sleep organization and epilepsy. In: Degen R, Rodin EA, eds. Epilepsy, Sleep and Sleep Deprivation. 2nd ed. Amsterdam: Elsevier; 1991:73-81. 39. Bazil CW, Kothari M, Luciano D, et al. Provocation of nonepileptic seizures by suggestion in a general seizure population. Epilepsia. 1994;35:768-770. 40. Hrachovy RA, Frost JD, Jr., Kellaway P. Sleep characteristics in infantile spasms. Neurology. 1981;31:688-793. 41. Malow BA, Lin X, Kushwaha R, Aldrich MS. Interictal spiking increases with sleep depth in temporal lobe epilepsy. Epilepsia. 1998;39:1309-1316. phale. 1931;26, 42. Marchand L. Des influences cosmiques sur les accidents epileptiques. LEnce Suppl. 10:237. 43. Billiard M, Besset A, Zachariev Z, Touchon J, Baldy-Moulinier M, Cadilhac J. Relation of seizures and seizure discharges to sleep stages. In: Wolf P, Dam M, Janz D, Dreifuss FE, eds. Advances in Epileptology. New York: Raven Press; 1987:665-670. 44. Ross JJ, Johnson LC, Walter RD. Spike and wave discharges during stages of sleep. Arch Neurol. 1966;14:399-407. 45. Niedermeyer E. Petit mal, primary generalized epilepsy and sleep. In: Sterman MB, Passouant P, Shouse MN, eds. Sleep and Epilepsy. New York: Academic Press; 1982:191-207. 46. Janz D. The grand mal epilepsies and the sleeping-waking cycle. Epilepsia. 1962;3:69-109. 47. Janz D, Christian W. Impulsiv-Petit mal. Dtsch Z Nervenheilk. 1957;176:346-356. 48. Blum DE, Eskola J, Bortz JJ, Fisher RS. Patient awareness of seizures. Neurology. 1996;47:260-264. 49. Billiard M. Epilepsies and the sleep wake cycle. In: Sterman MB, Passouant P, Shouse MN, eds. Sleep and Epilepsy. New York: Academic Press; 1982:269-286.

7  Epilepsy and Sleep 50. Touchon J, Baldy-Moulinier M, Billiard M, Besset A, Cadilhac J. Effect of awakening on epileptic activity in primary generalized myoclonic epilepsy. In: Sterman MB, Shouse MN, Passouant P, eds. Sleep and Epilepsy. New York: Academic Press; 1982:239-248. 51. Malow BA, Selwa LM, Ross D, Aldrich MS. Lateralizing value of interictal spikes on overnight sleep-EEG studies in temporal lobe epilepsy. Epilepsia. 1999;40:1587-1592. 52. Sammaritano M, Gigli GL, Gotman J. Interictal spiking during wakefulness and sleep and the localization of foci in temporal lobe epilepsy. Neurology. 1991;41:290-297. 53. Rossi GF, Colicchio G, Pola P. Interictal epileptic activity during sleep: a stereo-EEG study in patients with partial epilepsy. Electroencephalogr Clin Neurophysiol. 1984;58:97-106. 54. Crespel A, Baldy-Moulinier M, Coubes P. The relationship between sleep and epilepsy in frontal and temporal lobe epilepsies: practical and physiopathologic considerations. Epilepsia. 1998;39:150-157. 55. Bazil CW, Walczak TS. Effects of sleep and sleep stage on epileptic and nonepileptic seizures. Epilepsia. 1997;38:56-62. 56. Malow BA, Passaro E, Milling C, Minecan DN, Levy K. Sleep deprivation does not affect seizure frequency during inpatient video-EEG monitoring. Neurology. 2002;59:1371-1374. 57. Bechinger D, Kriebel J, Schlager M. Das Schlafentzugs-EEG, ein wichtiges diagnostisches Hilfsmittel bei cerebralen Anfa llen. Z Neurol. 1973;205:193-206. 58. Pratt KL, Mattson RH, Weikers NJ, Williams R. EEG activation of epileptics following sleep deprivation: a prospective study of 114 cases. Electroencephalogr Clin Neurophysiol. 1968;24:11-15. 59. Degen R, Degen HE. The diagnostic value of the sleep EEG with and without sleep deprivation in patients with atypical absences. Epilepsia. 1983;24:557-566. 60. Fountain NB, Kim JS, Lee SI. Sleep deprivation activates epileptiform discharges independent of the activating effects of sleep. J Clin Neurophysiol. 1998;15:69-75. 61. Rowan AJ, Veldhuisen RJ, Nagelkerke NJ. Comparative evaluation of sleep deprivation and sedated sleep EEGs as diagnostic aids in epilepsy. Electroencephalogr Clin Neurophysiol. 1982;54:357-364. 62. Klingler D, Tra gner H, Deisenhammer E. The nature of the influence of sleep deprivation on the EEG. In: Degen R, Rodin EA, eds. Epilepsy, Sleep and Sleep Deprivation. Amsterdam: Elsevier; 1991:231-234. 63. Badawy RA, Curatolo JM, Newton M, Berkovic SF, Macdonell RA. Changes in cortical excitability differentiate generalized and focal epilepsy. Ann Neurol. 2007;61:324-331. 64. Sammaritano M, Sherwin A. Effect of anticonvulsants on sleep. Neurology. 2000;54:S16-S24. 65. Terzano MG, Parrino L, Spaggiari MC, Palomba V, Rossi M, Smerieri A. CAP variables and arousals as sleep electroencephalogram markers for primary insomnia. Clin Neurophysiol. 2003;114:1715-1723. 66. Wolf P, Ro der-Wanner UU, Brede M. Influence of therapeutic phenobarbital and phenytoin medication on the polygraphic sleep of patients with epilepsy. Epilepsia. 1984;25:467-475. 67. Wolf P, Ro der-Wanner UU, Brede M, Noachtar S, Sengoku A. Influences of antiepileptic drugs on sleep. In: Martins da Silva A, Binnie C, Meinardi H, eds. Biorhythms and Epilepsy. New York: Raven Press; 1985:137-153. 68. Yang JD, Elphick M, Sharpley AL, Cowen PJ. Effects of carbamazepine on sleep in healthy volunteers. Biol Psychiatry. 1989;26:324-328. 69. Gigli GL, Placidi F, Diomedi M, et al. Nocturnal sleep and daytime somnolence in untreated patients with temporal lobe epilepsy: changes after treatment with controlled-release carbamazepine. Epilepsia. 1997;38:696-701. 70. Bell C, Vanderlinden H, Hiersemenzel R, Otoul C, Nutt D, Wilson S. The effects of levetiracetam on objective and subjective sleep parameters in healthy volunteers and patients with partial epilepsy. J Sleep Res. 2002;11:255-263. 71. Foldvary-Schaefer N, De Leon Sanchez I, Karafa M, Mascha E, Dinner D, Morris HH. Gabapentin increases slow-wave sleep in normal adults. Epilepsia. 2002;43:1493-1497. 72. Placidi F, Mattia D, Romigi A, Bassetti MA, Spanedda F, Marciani MG. Gabapentin-induced modulation of interictal epileptiform activity related to different vigilance levels. Clin Neurophysiol. 2000;111:1637-1642. 73. Bonanni E, Galli R, Maestri M, et al. Daytime sleepiness in epilepsy patients receiving topiramate monotherapy. Epilepsia. 2004;45:333-337. 74. Mathias S, Wetter TC, Steiger A, Lancel M. The GABA uptake inhibitor tiagabine promotes slow wave sleep in normal elderly subjects. Neurobiol Aging. 2001;22:247-253. 75. Oswald I. Sudden bodily jerks on falling asleep. Brain. 1959;82:92-103. 76. Dagnino N, Loeb C, Massazza G, Sacco G. Hypnic physiological myoclonias in man: an EEG-EMG study in normals and neurological patients. Eur Neurol. 1969;2:47-58. 77. Ohayon MM, Priest RG, Caulet M, Guilleminault C. Hypnagogic and hypnopompic hallucinations: pathological phenomena? Br J Psychiatry. 1996;169:459-467.

95

96

THE EPILEPSIES 3 78. Epstein AW. Recurrent dreams; their relationship to temporal lobe seizures. Arch Gen Psychiatry. 1964;10:25-30. 79. Lombroso CT. Pavor nocturnus of proven epileptic origin. Epilepsia. 2000;41:1221-1226. 80. Hublin C, Kaprio J, Partinen M, Heikkila K, Koskenvuo M. Prevalence and genetics of sleepwalking: a population-based twin study. Neurology. 1997;48:177-181. 81. Schenck CH, Bundlie SR, Ettinger MG, Mahowald MW. Chronic behavioral disorders of human REM sleep: a new category of parasomnia. Sleep. 1986;9:293-308. 82. Schenck CH, Bundlie SR, Mahowald MW. Delayed emergence of a parkinsonian disorder in 38% of 29 older men initially diagnosed with idiopathic rapid eye movement sleep behaviour disorder. Neurology. 1996;46:388-393. 83. Overeem S, Mignot E, van Dijk JG, Lammers GJ. Narcolepsy: clinical features, new pathophysiologic insights, and future perspectives. J Clin Neurophysiol. 2001;18:78-105. 84. Malow BA, Levy K, Maturen K, Bowes R. Obstructive sleep apnea is common in medically refractory epilepsy patients. Neurology. 2000;55:1002-1007. 85. Takhar J, Bishop J. Influence of chronic barbiturate administration on sleep apnea after hypersomnia presentation: case study. J Psychiatry Neurosci. 2000;25:321-324. 86. Peppard PE, Young T, Palta M, Dempsey J, Skatrud J. Longitudinal study of moderate weight change and sleep-disordered breathing. JAMA. 2000;284:3015-3021. 87. Malow BA, Weatherwax KJ, Chervin RD, et al. Identification and treatment of obstructive sleep apnea in adults and children with epilepsy: a prospective pilot study. Sleep Med. 2003;4:509-515.

THE EPILEPSIES 3

Cortical Myoclonus and Epilepsy: Overlap and Differences


RENZO GUERRINI  FRANCESCO MARI

Introduction Cortical Myoclonus (CM) Epileptic Syndromes and Neurological Disorders with CM Epileptic Syndromes and Neurological Disorders with Subcortical-Cortical (Thalamocortical) Myoclonus Idiopathic (Primary) Generalized Epileptic Myoclonus Epileptic Syndromes with Myoclonus of Unclear Neurophysiologic Characterization

Early Myoclonic Encephalopathy Myoclonic Status in Fixed Encephalopathies Epilepsy with Myoclonic Absences Eyelid Myoclonia with and without Absences Myoclonic Seizures Induced by Photic Stimuli Reticular Reflex Myoclonus Antiepileptic Drug-Induced Myoclonus Conclusion

Introduction
The term myoclonus is used to describe a brief and jerky involuntary movement involving antagonist muscles and originating from brief active contractions of muscles (positive myoclonus) or, more rarely, from brief interruptions of ongoing electromyographic activity (negative myoclonus).1 Clinically, myoclonus may be classified as focal if it involves a restricted, usually distal, group of muscles; multifocal when asynchronous focal jerks involve different body areas; or generalized when jerks involve most body segments in an apparently synchronous manner. Myoclonus is spontaneous if occurring irrespective from external triggering situations or can be induced by movement (action myoclonus) or by sensory or visual stimuli (reflex myoclonus). Finally, regarding periodicity, myoclonus may be either rhythmic or arrhythmic.

97

98

THE EPILEPSIES 3

Many conditions in which myoclonus is a prominent manifestation are known, allowing an etiological classification1,2 in which four major clinical syndrome categories are identified: (1) physiologic myoclonus (sleep-related, hiccup, myoclonus induced by anxiety or exercise); (2) essential myoclonus (individuals without other neurological signs); (3) epileptic myoclonus (conditions in which the predominant element is epilepsy); and (4) symptomatic myoclonus (conditions in which the predominant element is encephalopathy). Neurophysiological characteristics can be used to divide myoclonus into six main physiological categories, including cortical, cortical-subcortical, subcortical-supraspinal, spinal, and peripheral.3 The term epileptic myoclonus is still confusing. Some authors define as epileptic myoclonus what occurs within the setting of epilepsy.4 Others define as epileptic myoclonus those forms in which a paroxysmal depolarization shift is thought to be the underlying neurophysiological substrate, irrespective of which population of neurons (cortical or subcortical) is primarily involved.5 In our opinion, epileptic myoclonus can be comprehensively defined as an elementary electroclinical manifestation of epilepsy involving descending neurons, whose spatial (spread) or temporal (self-sustained repetition) amplification can trigger overt epileptic activity.6 Frequently, the electroencephalogram (EEG) correlate of epileptic myoclonus can be detected only by using jerk-locked (EEG or magnetoencephalogram) averaging. Yet, many patients with cortical myoclonus have rhythmic electromyogram (EMG) bursts at relatively high frequency (especially those with minipolymyoclonus, cortical tremor, Angelman syndrome or autosomal dominant cortical myoclonus, and epilepsy), which make it difficult to identify a cortical correlate. Recent works have demonstrated in these cases the role of EEG-EMG coherence by means of frequency analysis in demonstrating common cortical drives.7,8,9 Myoclonus can be only one component of a seizure (myoclonic jerks heralding a generalized tonic-clonic seizure in juvenile myoclonic epilepsy or in progressive myoclonus epilepsies), the only seizure manifestations (myoclonic jerks of benign myoclonic epilepsy), one of multiple seizure types (as observed in myoclonic-astatic epilepsy), or the basis of a movement-related disorder (action myoclonus in progressive myoclonus epilepsies). However, the relationships between myoclonus and epilepsy have been elucidated only in part, and the neurophysiological bases, nosology, and electroclinical characteristics of myoclonus in the setting of specific epilepsy syndromes are only partially understood. According to available evidence, epileptic myoclonus can be classified neurophysiologically as cortical (positive and negative), secondarily generalized, thalamocortical and reticular reflex myoclonus.10 Cortical epileptic myoclonus constitutes a fragment of partial or symptomatic generalized epilepsy; thalamocortical epileptic myoclonus is a fragment of idiopathic generalized epilepsy.5 Reflex reticular myoclonus, which does not have a time-locked EEG correlate, represents the clinical counterpart of hypersynchronous activity of neurons in the brainstem reticular formation. In the following sections, major epilepsy or neurological syndromes featuring different forms of epileptic myoclonus are described.

Cortical Myoclonus (CM)


CM originates from abnormal neuronal discharges in the sensorimotor cortex. Abnormally firing motoneurons may be primarily hyperexcitable or may be driven

8  Cortical Myoclonus and Epilepsy: Overlap and Differences

99

by abnormal inputs originating from hyperexcitable parietal11 or occipital12 neurons. Each jerk originates from the discharge of a small group of cortical motoneurons, somatotopically connected to a group of contiguous muscles. A cortical potential, time locked to the myoclonic potential and localized on the contralateral sensorimotor region, can be demonstrated by EEG, magnetoencephalogram, or jerk-locked averaging.13-16 Facilitation of interhemispheric and intrahemispheric spread of CM activity through transcallosal or corticocortical pathways seems to play a major role in producing generalized or bilateral myoclonus.17 In patients with cortical reflex myoclonus (CRM), appropriate stimuli administered to a resting somatic segment produce a reflex muscle response (jerk) with a latency of around 50 msec (C-reflex).18 A similar response is only observed in normal subjects during voluntary contraction. Somatosensory evoked potentials (SEPs) of giant amplitude are typically seen in association with CRM.19-22 The striking resemblance in latency and morphology of the giant SEPs to the myoclonus-related cortical spikes suggests that both originate from common cortical mechanisms.15 In the typical forms of CRM, the reflex jerk in the hand has a latency of 50 msec, and the CRT has a mean duration of about 7 msec.23 Typical CRM can be observed in patients with focal cortical lesions,18 spinocerebellar degeneration,14,24,25 multiple system atrophy,25,26,27 cerebral anoxia,14,28 childhood metabolic degenerations such as neuronal ceroid lipofuscinosis and sialidosis,11,19 Alzheimers disease,29,30 Down syndrome, and mitochondrial disorders.31,32,33 Epileptic negative myoclonus (ENM) is characterized by brief (50 to 400 msec) muscle inhibitions with focal, multifocal, or bilateral distribution and time locked to sharp-wave or spike-wave discharges on the contralateral central areas.34,35 ENM has a wide etiological spectrum ranging from idiopathic to symptomatic forms due to cortical dysplastic lesions.36 It may occasionally be precipitated by an adverse reaction to antiepileptic drugs.37,38,39 Previous studies35,40,41 hypothesized a cortical origin of ENM. Epileptic activity associated with ENM was described in the premotor42,43,44 and postcentral somatosensory cortex.40,45 Through cortical electrical stimulation studies it was suggested that negative motor areas might be present in the lateral and mesial portion of frontal lobe, encompassed in the supplementary sensorimotor area (SMA).46 However, no subsequent confirmatory studies are available. A possible role of the SMA in ENM was also proposed in a recent study47 in which electrical stimulation of the SMA constantly evoked ENM, with no preceding positive myoclonus, as it was instead observed following stimulation of the premotor, primary motor, and sensorimotor cortex. EPILEPTIC SYNDROMES AND NEUROLOGICAL DISORDERS WITH CM Cortical Action/Reflex Myoclonus Progressive Myoclonus Epilepsies (PMEs) Progressive myoclonus epilepsies represent a clinically and etiologically heterogeneous group of diseases with a progressive course, characterized by myoclonus, generalized, tonic-clonic seizures, and neurological deterioration.48 Onset is most frequently in late childhood or adolescence.49 Different forms are known, including Unverricht-Lundborg disease, Lafora disease, neuronal ceroid-lipofuscinosis, type III Gaucher disease, infantile and juvenile GM2-gangliosidosis, some mitochondrial

100

THE EPILEPSIES 3

L. Delt L. Ext L. Flex

R. Delt R. Ext R. Flex Fz-Cz Cz-Pz L. S. 27yrs 27.3.95 1 sec 100V

Figure 81 Twenty-seven-year-old woman with sialidosis (cherry red spot myoclonus) and movement-activated seizures. When the patient raises her right arm to seize an object, there is a typical cascade of myoclonic potentials that is initially arrhythmic and involves all sampled muscles on the right (right deltoid, wrist extensors, and flexors), then become more rhythmic and spread to the contralateral muscles. This pattern of spread of myoclonic activity is typical of action myoclonus.

encephalopathies, sialidosis, dentatorubropallidoluysian atrophy, and action myoclonus-renal failure (AMRF).49,50 Causative genes have been identified for most PMEs.51,52 Onset features comprise myoclonus and rare generalized tonic-clonic seizures, as in idiopathic myoclonic epilepsies.53,54 The tonic-clonic seizures can occur without any warning or after a long buildup of myoclonic jerks. The EEG shows generalized polyspike and spike-and-wave discharges, frequently precipitated by photic stimulation. Background EEG activity becomes progressively slower.53 Cortical reflex myoclonus is common to all PME syndromes, in which it is manifested with the classic combination of action myoclonus (Figure 8-1), spontaneous jerks, giant SEPs, C-reflex at rest, and the premyoclonus spike. According to Cantello et al.,55 focal subcortical reflex myoclonus can also be demonstrated in these patients. Initially mild, myoclonus becomes increasingly disabling during the course. Severe action myoclonus has a devastating impact on the patients level of autonomy. Rett Syndrome Rett syndrome is an X-linked dominant disorder, with an estimated prevalence of 1 in 10,000 to 15,000 females, making it one of the most common causes of severe mental

8  Cortical Myoclonus and Epilepsy: Overlap and Differences

101

retardation in females. Mutations in the exons 1 to 4 of the methyl CpG binding protein 2 gene (MECP2)56 have been identified in roughly 75 to 80% of girls with classical Rett syndrome;57 in MECP2-negative patients additional screening using (multiplex ligation probe amplification) MLPA enables detection of large deletions in nearly half of the remaining patients with the syndrome.58 The clinical phenotype includes progressive cognitive deterioration leading to dementia, autistic features, truncal ataxia/apraxia, loss of purposeful hand movements, breathing abnormalities, stereotypies, extrapyramidal signs, and epilepsy. A form of CRM characterized by prolonged C-reflex (65 5 msec) latency has been described in affected girls.59 Myoclonus is multifocal and arrhythmic, and major myoclonic seizures are not seen in these patients. A positive potential, localized on the contralateral centroparietal area, precedes myoclonus with a latency of 34 7 msec for the forearm muscle compatible with corticomotoneuronal conduction. The N20-P30 and P30-N35 components of the SEPs have significantly increased amplitude. In addition, the latency of the N20 component is delayed, and the N20-P30-N35 interval is significantly increased and has expanded morphology. It is probable that in Rett syndrome the following sequence of events occurs: slight delay in central conduction of the impulse afferent to the sensorimotor cortex (N20), slowing of the processing of the afferent impulse (interval N20-P30; mean = 11 msec), delay in corticocortical transmission to the precentral neurons subserving movement of the stimulated body segment (latency increase P30 - C reflex; mean = 32 msec), and rapid descending volley to the spinal motoneurons. Intracortical conduction time could be particularly prolonged because of the synaptic abnormalities that have been observed.60 Huntingtons Disease In Huntingtons disease, action myoclonus is a rare manifestation, but a few patients have been described in whom CRM was the presenting symptom.33,61 Seizures are an infrequent complication and are mainly seen with juvenile onset, rarely presenting with a typical PME syndrome.62 Postanoxic Encephalopathy Postanoxic encephalopathy is characterized by dysarthria, ataxia, pyramidal signs, rigidity, epilepsy, and myoclonus, which is usually spontaneous and action-induced, multifocal and generalized, and extremely disabling. Electromyographic silent periods following the jerks contribute to producing postural lapses.63 Postanoxic myoclonus may be cortical in origin, involving the sensorimotor cortex and rapidly conducting pyramidal pathways.14,28 More rarely it may have brainstem origin, either as an exaggerated startle reflex or as reticular reflex myoclonus.64,65 Forty percent of patients with postanoxic myoclonus suffer from generalized epileptic seizures. Focal Cortical Repetitive Myoclonus Epilepsia Partialis Continua Epilepsia partialis continua, or Kojewnikows syndrome,66 is characterized by almost continuous, focal, rhythmic (around 1 to 2 Hz) muscle jerks, which are observed both while awake and asleep, for periods ranging from hours to days, or rarely years.48 Unilateral somatomotor seizures are constantly associated. Two types of epilepsia partialis continua have been identified.67 The first type is due to fixed epileptogenic lesions involving the motor cortex. Causative factors include

102

THE EPILEPSIES 3

ischemia, posttraumatic head injury, cortical dysplasia, tumors, and vascular malformations.67-70 A stable motor deficit, predating seizure onset, and nonprogressive evolution are usual features. A second type of epilepsia partialis continua is observed in Rasmussens syndrome. Onset occurs during childhood, with continuous focal jerking and intractable homolateral motor or generalized seizures. Progressive hemiparesis, hemianopia, and eventually, cognitive deterioration follow. Magnetic resonance imaging (MRI) shows progressive atrophy of the affected hemisphere. Pathological studies reveal inflammation with perivascular infiltrates and microglia nodules.71 A viral etiology was originally hypothesized. More recently, the role of antibody-mediated mechanisms and more recently cell-mediated immunity have been hypothesized72,73,74 with inconclusive results. An analogous form of progressive epilepsia partialis continua has been observed in some children with mitochondrial myopathy, encephalopathy, lactic acidosis, and stroke (MELAS).75 Rhythmic High-Frequency Cortical Myoclonus (Cortical Tremor) Cortical tremor is a form of rhythmic myoclonus, presenting as postural or action tremor in some patients with progressive myoclonus epilepsy (PME),76,77 in Angelman syndrome, and in different forms of autosomal dominant epilepsy.7883 The nosologic boundaries between epilepsia partialis continua and this peculiar form of repetitive myoclonus are unclear. Angelman Syndrome Angelman syndrome is a neurogenetic disorder deriving from a defect in maternal chromosome 15q11 to q13. Seventy percent of patients present a cytogenetic or molecular deletion encompassing three subunits of receptor a for gammaaminobutyric acid (GABRB3, GABRA5, and GABRG3) and the gene UBE3A. Uniparental paternal disomy for chromosome 15, or mutations in the imprinting center or in the UBE3A gene, are more rarely found. Patients have microbrachycephaly, severe to moderate mental retardation, absence of speech, inappropriate paroxysmal laughter, epilepsy, ataxic gait, tremor, and jerky movements. Neurophysiologic investigations reveal a spectrum of manifestations of myoclonus.84 All patients present with rapid distal jerking of fluctuating amplitude, which causes a sort of coarse distal tremor combined with dystonic limb posturing. Jerks occur at rest in prolonged runs. In addition, the majority of patients have myoclonic and absence seizures, as well as episodes of myoclonic status. Bilateral jerks of myoclonic absences show rhythmic repetition at  2.5 Hz and are time locked with a cortical spike. Interside latency of both spikes and jerks is consistent with transcallosal spread, and spike-to-jerk latency indicates propagation through rapid conduction corticospinal pathway. A contralateral, central premyoclonic potential is uncovered by jerk-locked averaging. SEPs are normal, and C-reflex is absent. Familial Adult Myoclonic Epilepsy and Autosomal Dominant Cortical Reflex Myoclonus and Epilepsy A form of autosomal dominant epilepsy with cortical myoclonus manifested as cortical tremor has been described in several families, mostly of Japanese origin,79,85 and given the acronym BFAME (benign familial adult myoclonic epilepsy) or FAME (familial adult myoclonic epilepsy). Affected patients present homogeneous characteristics, including (a) autosomal dominant inheritance; (b) adult

8  Cortical Myoclonus and Epilepsy: Overlap and Differences

103

N35 C4-Fpz P30 C3-Fpz

| 12.5V

L.O.Oris

|100V

L. Delt

L.W.Ext

L.APB C L.TA

R.Delt

R.APB CIC

10 ms IV-4, 23yrs
Young adult lady with autosomal dominant cortical myoclonus and epilepsy. Cortical SEP to electrical stimulation of the left median nerve at the wrist. An enlarged N20P30-N35 complex is followed by C-reflex in the left deltoid, wrist extensors, and abductor pollicis brevis. The ipsilateral muscular response precedes a contralateral response by  8 ms. This delay is consistent with inter-hemispheric transcallosal spread. L = left; O. Oris = orbicularis oris; Delt = deltoid; W. Ext = wrist extensors; APB = abductor pollicis brevis; C = cortical reflex; TA = tibialis anterior; R = right; ClC = contralateral cortical reflex.

Figure 82

onset (mean age 38 years, range: 19 to 73); (c) nonprogressive course; (d) distal, rhythmic myoclonus enhanced during posture maintenance (cortical tremor); (e) rare, apparently generalized, seizures often preceded by worsening of myoclonus; (f) absence of other neurological signs; (g) generalized interictal spike-and-wave discharges; (h) photoparoxysmal response; (i) giant SEPs and hyperexcitability of the C-reflex (Figure 8-2); and (j) cortical EEG potential time locked to the jerks. The original Japanese families linked to chromosome 8q23.3-q24.86 However, European families with a similar phenotype did not link to the same locus.85,87,88

104

THE EPILEPSIES 3

Autosomal dominant cortical reflex myoclonus and epilepsy (ADCME)81 has been described in patients with a homogeneous syndromic core, including an association of nonprogressive cortical reflex myoclonus, manifested as semicontinuous rhythmic distal jerking (cortical tremor), generalized tonic-clonic convulsions (GTCs) preceded in some patients by generalized myoclonic jerks, and generalized EEG abnormalities. Age at onset of cortical tremor and of GTCs overlapped in a given individual but varied between individuals, ranging from 12 to 50 years. This clinical picture shares some features with FAME86; however, all ADCME patients had in addition focal frontotemporal EEG abnormalities, and some individuals also had focal seizures, of variable severity, starting around the same age as the other manifestations. The pattern of inheritance is autosomal dominant with high penetrance. Linkage analysis identified a critical region in chromosome 2p11.1-q12.2.81 Three Italian families with familial adult myoclonic epilepsy have been described, possibly linked to 2p11.1-q12.2, suggesting a possible allelism with ADCME.88,89 Recently two novel families with adult myoclonic epilepsy have been reported,82,83 neither linking to known loci. In the first report, clinical photosensitivity was a prominent feature; in the second report photosensitivity was not a prominent feature, but cerebellar ataxia, dementia, and progression of symptoms were observed, raising doubts about the nosology of this disorder. Epileptic Syndromes with Secondarily Generalized Epileptic Myoclonus Severe Myoclonic Epilepsy in Infancy (SMEI), or Dravet Syndrome SMEI, or Dravet syndrome, is observed in 6 to 7% of children with seizure onset in the first year of life90 and is a severe form of epilepsy, characterized by multiple seizure types and unfavorable prognostic outlook. Its classification as a form of myoclonic epilepsy is controversial because myoclonus, although present in most children, can be a transient phenomenon and often does not represent a hallmark of the syndrome. A subgroup of children does not exhibit myoclonic seizures at all.90,91 SMEI represents the prototype of an epileptic encephalopathy in which onset of severe, prolonged seizures precedes deterioration of cerebral functions.92 Mutations of the SCN1A gene are observed in about 80% of cases.93 Onset of epilepsy occurs during the first year of life with prolonged generalized or unilateral, clonic or tonicclonic seizures during fever, often evolving to status. They rapidly become associated with similar nonfebrile attacks. By the third to fourth year of life, resistant myoclonic, partial seizures and atypical absences also appear. EEG, normal at the beginning, subsequently shows multifocal and generalized abnormalities. Early photosensitivity is seen in some children. Neurological development appears delayed from the second year of life onward. Two main types of myoclonus have been described. Almost all children show arrhythmic, distal jerks, manifested as twitching of fingers, whereas some also have generalized jerks. Demonstrating a premyoclonic potential for multifocal jerks may be difficult, even using jerklocked averaging. Generalized jerks have an obvious EEG correlate, which appear to originate from spread of CM activity, when small time differences are measured (Figure 8-3).10,54 Lennox-Gastaut Syndrome Lennox-Gastaut syndrome (LGS) has a prevalence of 2 to 3% in children with epilepsy and is often observed in the brain damaged.94 Typical seizures start at 3 to

8  Cortical Myoclonus and Epilepsy: Overlap and Differences

105

F4 C4 P4 L. Mas L. Orb oris L. Delt L. Ext L. APB L. Quad F3 C3 P3 O1 R. Mas R. Delt

P. M.

14y

NS41

21.3.96

100V 1 sec

Figure 83 Fourteen-year-old boy with severe myoclonic epilepsy. Polygraphic recording. A discharge of bilateral and synchronous quasi-rhythmic multiple spike and waves and spikeand-wave complexes is accompanied by a series of myoclonic potentials, most of which are only visible on the left masseter. Muscle jerks are time locked with the spikes; one jerk is generalized. Such focal jerks, which were not reported by the patient, would have gone unrecognized should multiple muscles not have been sampled.

5 years of age as tonic, atonic, or atypical absences.94 A previous form of epilepsy, especially West syndrome, is frequently observed. Associated seizure types are myoclonic, generalized tonic-clonic, and rarely, focal. Epilepsy is drug resistant, and episodes of status are frequent. Interictal EEG shows abnormal background activity, slow (1.5 to 2.5 Hz) generalized spike waves, and often multifocal abnormalities. During sleep, all patients show typical rhythmic discharges around 10 Hz, accompanying tonic seizures or without apparent clinical correlate. Myoclonus is not a prominent feature of Lennox-Gastaut syndrome,95 but some patients exhibit generalized myoclonic jerks that seem to be produced by a secondary generalization of focal CM.54,96 Minipolymyoclonus, a term used to describe distal, small focal jerks, frequently leading to individual tiny finger movements, is observed in some patients

106

THE EPILEPSIES 3

with LGS8,97 in which back-averaged EEG shows a bilateral frontal negative slow wave, with 20 to 500 msec latency. In other patients, a sharper bilateral frontal negativity is demonstrable, leading the jerks by 40 to 70 msec.97 Minipolymyoclonus is strongly similar to the pattern of distal myoclonus observed in Angelman syndrome.8,84

Epileptic Syndromes and Neurological Disorders with Subcortical-Cortical (Thalamocortical) Myoclonus


IDIOPATHIC (PRIMARY) GENERALIZED EPILEPTIC MYOCLONUS Generalized epileptic myoclonus is spontaneous, is predominantly arrhythmic, and has an inconstant axial predominance. Patients may present with simple head nodding or raised shoulders or may stagger or fall. The generalized jerks appear to originate from afferent volleys from subcortical structures that act synchronously on a hyperexcitable cortex.5,98 As a consequence, muscles from both sides are activated synchronously, as in reticular myoclonus, and muscles innervated by the cranial nerves are involved through a rostrocaudal pattern of activation, as in cortical myoclonus. The EEG correlate is a generalized spike wave. The negative peak of the spike precedes the generalized jerks by 20 to 75 msec. Idiopathic Generalized Epilepsies Benign Myoclonic Epilepsy of Infancy (BME) BME affects 0.4 to 2% of all children with seizure onset by age 3 years.99,100,101 Age at onset ranges between 4 months and 5 years in children with normal development. Some patients, however, can have mild cognitive impairment.102,103 Seizures consist of generalized myoclonic jerks, which are brief, isolated, or repeated in small series. If the child is standing or sitting, the jerks often cause nodding with upward gaze deviation and eyelid myoclonus, accompanied by slight arm abduction or elbow bending. Staggering may occur, especially up to the second year of life, when walking is still unstable. If falls occur, the child collapses on the buttocks and then gets up immediately. In most cases, the jerks occur many times per day. A few patients may have generalized tonic-clonic seizures in adolescence.103 Treatment had been withdrawn in most patients aged more than 6 years at follow-up.102 However, the use of the term benign is questionable according to the most recent ILAE (International League Against Epilepsy) definitions in that outcome is often judged only in retrospect, and children with the same clinical presentation at onset might have cognitive or behavioral sequelae.104,105 About 10% of children with BME have photic-induced jerks.102 Some have both spontaneous and reflex myoclonus, the latter triggered by tactile or sudden acoustic stimuli.106 Neurophysiology of myoclonus reveals symmetric, rostrocaudal muscle activation and a premyoclonus negative spike preceding jerks by 30 2 msec.6 Duration of the myoclonic jerk is roughly 100 msec. Juvenile Myoclonic Epilepsy (JME) JME has a prevalence of between 3.4 and 11.9% and represents the most common form of idiopathic generalized epilepsy (23.3%).107 The syndrome is genetically

8  Cortical Myoclonus and Epilepsy: Overlap and Differences

107

heterogeneous and in most cases is presumed to be polygenic. However, mutations of three different genes have been identified in rare families having dominant (CLCN2 and GABRA1)108,109,110 or recessive (EFHC1) forms of the syndrome.111 Onset occurs at around age 14, with generalized myoclonus and generalized tonicclonic seizures. Myoclonic jerks constitute the initial symptom in 54% of patients. They are characteristically concentrated in the minutes following awakening and are bilateral, single or repetitive, arrhythmic, and more pronounced in the upper limbs. If intense, they may result in falls, but are too brief to be accompanied by loss of consciousness. Facial or lingual and perioral jerks, usually isolated, may be precipitated by talking in some patients,112 a phenomenon analogous to the jerking observed in patients with primary reading epilepsy. In 5% of patients, generalized jerks are also triggered by intermittent photic stimulation. Severe increase in frequency of jerks may herald episodes of myoclonic status epilepticus, which have become rarer113,114 with improved treatment. However, drug withdrawal or inappropriate drug choice are among the main factors that may precipitate symptoms.115 Generalized tonic-clonic seizures are present in 84% of patients and represent the initial symptom in 35% of cases. They are often preceded by a buildup of generalized myoclonic jerks. In 27% of patients, absences are also present, occurring infrequently (less than several times per week). Treatment with valproic acid in monotherapy or in association with clonazepam leads to total control of seizures in 80% of patients.107 Discontinuation of drug therapy is followed by a high rate of relapse (90%).107 Neurophysiologic analysis of myoclonus in JME indicates that muscles from both sides are activated synchronously, and those innervated by the cranial nerves are involved through a rostrocaudal pattern of activation. The EEG correlate is a generalized spike- or polyspikes-wave at 3 to 5 Hz, in which the negative peak of the spike precedes the generalized jerk by 10 to 30 msec.6 Duration of the EEG transient is 100 msec, and that of the myoclonic potential is less than 100 msec. In a recent work,116 a lateralized onset of the EEG transient has been suggested on the basis of an interside latency (9.5 1.7 msec) that was thought to be compatible with transcallosal spread. However, it remains to be explained why this supposed focal trigger constantly spreads to constantly generate a generalized phenomenon, without producing any focal jerking (as usually seen, for example in patients with PME, who constantly exhibit both). Myoclonic-Astatic Epilepsy (MAE) MAE has its onset between 2 and 6 years of age. Seizure types include massive myoclonus and atonic falls, atypical absences, generalized clonic or tonic-clonic seizures, and episodes of status epilepticus with erratic myoclonus and clouding of consciousness.117 Interictal EEG, often normal at onset, can become very disorganized.104 Outcome is unpredictable. Remission within a few months or years with normal cognition is possible, even after a severe early course.118,119 About 30% of children experience long-lasting intractability and cognitive impairment.120 A few children with MAE have been shown to have inherited SCN1A and GABRG2 gene mutations from parents with generalized epilepsy with febrile seizures plus.121 However, the genetics of MAE is complex. Myoclonus in MAE manifests as bilateral, synchronous whole body jerks, consistent with the hypothesis of a thalamocortical origin.54 The jerks, lasting around 100 msec, are preceded by a negative EEG potential by around 30 msec.96 Myoclonic status in MAE has neurophysiological

108

THE EPILEPSIES 3

characters of erratic CM with multifocal jerking, increase in muscle tone, and clouding of consciousness. Nonconvulsive status may be precipitated by carbamazepine (Figures 8-4 and 8-5).6,37 Treatment is primarily with valproate and ethosuximide, often in combination. Lamotrigine, topiramate, and benzodiazepines might be useful in some patients.

Epileptic Syndromes with Myoclonus of Unclear Neurophysiologic Characterization


EARLY MYOCLONIC ENCEPHALOPATHY Early myoclonic encephalopathy is a rare syndrome classified among the generalized symptomatic epilepsies with nonspecific etiology.48 Its causes are multiple and include some inborn errors of metabolism, such as methylmalonic acidemia and nonketotic hyperglycinemia. Onset is in the neonatal period or during the first month of life with severe myoclonus, followed by partial seizures and tonic spasms. Myoclonus has a multifocal distribution, leading to the description of erratic. Neurological development is severely delayed, with hypotonia, impaired alertness, and, often, vegetative state.122 The EEG is characterized by suppression bursts. Erratic myoclonus generally does not have an EEG correlate.122 MYOCLONIC STATUS IN FIXED ENCEPHALOPATHIES This condition is seen exclusively in severe encephalopathies with profound cognitive impairment and hypotonia and is characterized by recurrent, prolonged, and drug-resistant episodes of myoclonic status.123 Partial motor seizures, myoclonic absences, generalized myoclonus, and, rarely, unilateral or generalized clonic seizures can be associated. Myoclonic status is characterized by almost continuous absences accompanied by erratic, distal, multifocal, frequent myoclonic jerks, at times more rhythmic and diffuse. It is extremely important to recognize this condition and to differentiate it from a progressive encephalopathy. EPILEPSY WITH MYOCLONIC ABSENCES Onset is at about age 7 years with absences recurring many times a day, accompanied by bilateral rhythmic jerks, involving the shoulders, arms, or legs and, eventually, by a mild axial tonic contraction. Consciousness is cloudy but not completely interrupted.124 Ictal EEG shows bilateral, synchronous, and symmetric spike-wave discharges at 3 Hz and myoclonic jerks at the same frequency. Absences are often resistant to treatment. Evolution is variable featuring cognitive impairment in some patients, transition to a different type of epilepsy, or at times full recovery without sequelae. The physiology of myoclonus in myoclonic absences is difficult to study as jerks appear against a background of increased muscle tone.6 Tassinari and coworkers125 found a constant relationship between the spike-and-wave complex and the jerk, with the positive spike of the spike-and-wave complex being followed by a myoclonic jerk by a latency of 15 to 40 msec (proximal muscles).

8  Cortical Myoclonus and Epilepsy: Overlap and Differences

109

F4-AV C4-AV P4-AV O2-AV T3-AV L. Mass. L. O. Oris L. Delt. L. W. Ext. L. W. Flex. F3-AV C3-AV P3-AV O1-AV T3-AV L. Delt. L. W. Ext. L. W. Flex. Fz-AV Cz-AV Pz-AV

4y 6m

NS41 1s

100V

Figure 84 Myoclonic status epilepticus precipitated by carbamazepine in a child with myoclonic-astatic epilepsy. Clinically the child is unresponsive and exhibits multifocal and generalized jerks. The EEG shows diffuse slow waves, interspersed with diffuse or multifocal irregular slow spike-and-wave complexes. Simultaneous surface EMG shows arrhythmic, focal, and generalized myoclonic potentials on a background of mild tonic contraction.

110

THE EPILEPSIES 3

F2-F4 F4-C4 C4-P4 P4-O2 F8-T4 T4-T6 L.MAS. L.O.O. L.Delt. L.W.Ext. F1-F3 F3-C3 C3-P3 P3-O1 F7-T3 T3-T5 R.Delt. R.W.Ext. Fz-Cz Cz-Pz Pz-Oz B. A. 4y 6m T826 1 sec 100 V

Figure 85 Same patient as in Figure 8-4. Carbamazepine was briskly suspended after the recording of status myoclonicus shown in Figure 8-4. Status quickly resolved in the following days to disappear 4 days later, as shown in this recording, EEG shows rhythmic background, no paroxysmal discharges, and no myoclonic potentials are recorded on EMG channels.

8  Cortical Myoclonus and Epilepsy: Overlap and Differences

111

EYELID MYOCLONIA WITH AND WITHOUT ABSENCES Eyelid myoclonia with absences are characterized by prominent jerking of the eyelids with upward deviation of the eyes. Some authors126,127 emphasized the severity of eyelid jerking in these patients as compared with the slight flicker of the eyelids seen in typical absences. The phenomenon may be so short (1 to 2 seconds) that it may be impossible to find out whether there is concomitant lapse of consciousness. The intensity of the jerking justifies the inclusion of this condition within the group of myoclonic epilepsies, more so as the myoclonic phenomena are difficult to control and persist into adulthood, whereas the absences are relatively easily controlled. A marked photosensitivity and self-stimulation are features that eyelid myoclonia, with and without absences, share with other myoclonic epilepsies of infancy and childhood. MYOCLONIC SEIZURES INDUCED BY PHOTIC STIMULI Myoclonic attacks can be induced by photic stimuli. Jeavons and Harding126 found that only 1.5% of pure photosensitive epilepsies (i.e., epilepsies induced exclusively by exposure to visual stimuli without any spontaneous attacks) were myoclonic. Visually induced generalized myoclonic jerks are usually symmetrical and predominate in the upper limbs. In most cases, they are mild, only producing head nodding and slight arm abduction. More generalized jerks, involving the face, trunk, and legs, may occasionally cause the patient to fall. The relationship of myoclonic jerks to the stimulus is complex. Sometimes there is no definite time relationship. On other occasions, the jerks may be repeated rhythmically with the same frequency as the stimulus or at one of its subharmonics.128 The jerks are associated in the EEG recording with the photoparoxysmal response, consisting of a bilateral polyspike or polyspike-and-wave discharge.128,129 Spontaneous seizures are said to occur mainly, but not exclusively, when the polyspike-wave discharge persists after discontinuation of the stimulation (prolonged photoconvulsive response).130 Myoclonic attacks can be provoked by television watching, especially when the patients are close to the screen and while playing video games. Some patients, especially, but not exclusively, mentally retarded, induce the myoclonic attacks by waving a hand between their eyes and a source of light, flickering their eyelids in front of a light source, or staring at patterned surfaces or by similar maneuvers.126,131,132 There is no clear-cut nosologic distinction between eyelid myoclonia and photic-induced myoclonus.

Reticular Reflex Myoclonus


Reticular myoclonus presents most of the clinical and neurophysiological characteristics of epileptic myoclonus, although it lacks a time-locked EEG correlate.5 Clinically, myoclonic jerks are generalized, mostly involving proximal and flexor muscles, spontaneous, or induced by somatosensory, auditory, and visual stimuli, or by movement.5,64 Reticular myoclonus seems to originate from the brainstem reticular formation, as involvement of muscles innervated by XIth cranial nerve (trapezius and sternocleidomastoid muscles) precedes that of orbicularis oris (VIIth cranial nerve) and masseter (Vth cranial nerve).64 EEG discharges have a wide distribution

112

THE EPILEPSIES 3

and greater amplitude at the vertex and can follow the onset of the jerks, suggesting that they are projected and not directly responsible for the myoclonic jerks.5 SEPs have normal amplitude. In reflex reticular myoclonus, both slow and fast conducting pathways have been observed.20 Reticular myoclonus has been described in postanoxic encephalopathy and can appear alongside cortical myoclonus in some patients.14,24,65 In patients with progressive myoclonus epilepsy, Cantello described a form of focal subcortical reflex myoclonus whose latencies might be consistent with origin in the reticular formation.55 Clinically, there are both multifocal and generalized jerks. Neurophysiological study of reticular reflex myoclonus is, however, difficult, especially because of the coexistence of cortical myoclonus in most patients. As a consequence, its neurophysiological correlates and relationships with epilepsy are poorly understood. Electrophysiological recordings in the urea-induced myoclonus in the rat, which is considered to be a model of reticular reflex myoclonus, have demonstrated neuronal activity resembling paroxysmal depolarization shift in the nucleus reticularis gigantocellularis.133

Antiepileptic Drug-Induced Myoclonus


Antiepileptic drugs can aggravate or induce myoclonus or myoclonic seizures, either because of paradoxical reaction or inappropriate choice. Carbamazepine and vigabatrin have been reported to worsen or precipitate myoclonic seizures.134,135 De novo appearance of myoclonic jerks was described in children or young adults with cryptogenic or symptomatic partial epilepsy treated with add-on vigabatrin.136,137 Carbamazepine should be avoided in MAE, because it can trigger episodes of myoclonic status.6 Adolescents with juvenile absence epilepsy may experience myoclonic status if treated with carbamazepine when their absence seizures are misdiagnosed as complex partial seizures.138 Exacerbation of epileptic negative myoclonus has been reported in children with benign rolandic epilepsy after carbamazepine treatment.37,38 Lamotrigine may be useful in some children with myoclonic astatic epilepsy,139 but it has been reported to worsen SMEI,140 and occasionally, to precipitate seizure aggravation and de novo myoclonic status epilepticus if administered at high doses in other conditions.141145 LTG has also been reported to aggravate seizures in patients with JME.143 In Angelman syndrome worsening of myoclonic and absence seizures may be produced by carbamazepine or oxcarbazepine,146,147,148 phenytoin,146 or vigabatrin.149

Conclusion
Epileptic myoclonus can be defined as an elementary electroclinical manifestations of epilepsy involving descending neurons, whose spatial (spread) or temporal (selfsustained repetition) amplification can trigger overt epileptic activity6 and can be classified as cortical (positive and negative), secondarily generalized, thalamocortical, and reticular.10 Cortical epileptic myoclonus represents a fragment of partial or symptomatic generalized epilepsy; thalamocortical epileptic myoclonus is a fragment of idiopathic generalized epilepsy.5 Reflex reticular myoclonus represents the clinical counterpart of fragments of hypersynchronous epileptic activity of neurons in the

8  Cortical Myoclonus and Epilepsy: Overlap and Differences

113

brainstem reticular formation. EM, in the setting of an epilepsy syndrome, can be only one component of a seizure (i.e., myoclonic buildup in the clonic-tonic-clonic seizures of juvenile myoclonic epilepsy), the only seizure manifestations (myoclonic seizures of benign myoclonic epilepsy), one of the multiple seizure types (myoclonic astatic seizures in childhood epileptic encephalopathies), or a more stable condition that is manifested in a nonparoxysmal fashion and mimics a movement disorder (i.e., the continuous jerking of cortical tremor or of epilepsia partialis continua or the movement-activated jerks of progressive myoclonus epilepsy that can translate into a myoclonic cascade and a full-blown generalized tonic-clonic seizure). This complex correlation is more obvious in patients with epilepsia partialis continua in which cortical myoclonus (which underlies recurring focal jerks) and overt focal motor seizures usually start in the same somatic (and cortical) region. In patients with cortical tremor, this correlation is less obvious and requires neurophysiological studies to be demonstrated. REFERENCES
1. Marsden CD, Hallett M, Fahn S. The nosology and pathophysiology of myoclonus. In: Marsden CD, Fahn S, eds. Movement Disorders. London: Butterworths Scientific; 1982:196-249. 2. Fahn S, Marsden CD, Van Woert MH. Definition and classification of myoclonus. Adv Neurol. 1986;43:1-5. 3. Caviness JN, Brown P. Myoclonus: current concepts and recent advances. Lancet Neurol. 2006;3:598-607. 4. Patel VM, Jankovic J. Myoclonus. In: Appel SH, ed. Current Neurology, vol 8. Chicago: Year Book Medical Publishers; 1988:109-156. 5. Hallett M. Myoclonus: relation to epilepsy. Epilepsia. 1985;26(Suppl 1):S67-77. 6. Guerrini R, Bonanni P, Rothwell J, Hallet M. Myoclonus and epilepsy. In: Guerrini R, Aicardi J, Andermann F, Hallett M, eds. Epilepsy and Movement Disorders. Cambridge: Cambridge University Press; 2002:165-210. 7. Brown P, Farmer SF, Halliday DM, Marsden J, Rosenberg JR. Coherent cortical and muscle discharge in cortical myoclonus. Brain. 1999;122:461-472. 8. Grosse P, Guerrini R, Parmeggiani L, Bonanni P, Pogosyan A, Brown P. Abnormal corticomuscular and intermuscular coupling in high-frequency rhythmic myoclonus. Brain. 2003;126:326-342. 9. Van Rootselaar AF, Maurits NM, Koelman JHTM, et al. Coherence analysis differentiates between cortical myoclonic tremor and essential tremor. Mov Disord. 2006;21:215-222. 10. Guerrini R, Bonanni P, Parmeggiani L, Hallet M, Oguni H. Pathophysiology of myoclonic epilepsies. Adv Neurol. 2005;95:23-46. 11. Deuschl G, Ebner A, Hammers R, Lucking CH. Differences of cortical activation in spontaneous and reflex myoclonias. Electroencephalogr Clin Neurophysiol. 1991;80:326-328. 12. Kanouchi T, Yakota T, Kamata T, Ishii K, Senda M. Central pathway of photic reflex myoclonus. J Neurol Neurosurg Psychiatry. 1997;62:414-417. 13. Shibasaki H, Kuroiwa Y. Electroencephalographic correlates of myoclonus. Electroencephalogr Clin Neurophysiol. 1975;39:455-463. 14. Hallett M, Chadwick D, Marsden CD. Cortical reflex myoclonus. Neurology. 1979;29:1107-1125. 15. Shibasaki H, Kakigi R, Ikeda A. Scalp topography of giant SEP and pre-myoclonus spike in cortical reflex myoclonus. Electroencephalogr Clin Neurophysiol. 1991;81:31-37. 16. Mima T, Nagamine T, Ikeda A, Yazawa S, Kimura J, Shibasaki H. Pathogenesis of cortical myoclonus studied by magnetoencephalography. Ann Neurol. 1998;43:598-607. 17. Brown P, Day BL, Rothwell JC, Thompson PD, Marsden CD. Intrahemispheric and interhemispheric spread of cerebral cortical myoclonic activity and its relevance to epilepsy. Brain. 1991;114: 2333-2355. 18. Sutton GG, Mayer RF. Focal reflex myoclonus. J Neurol Neurosurg Psychiatry. 1974;7:207-217. 19. Shibasaki H, Yamashita Y, Neshige R, Tobimatsu S, Fukui R. Pathogenesis of giant somatosensory evoked potentials in progressive myoclonic epilepsy. Brain. 1985;108:225-240. 20. Rothwell JC, Obeso JA, Marsden CD. Electrophysiology of somatosensory reflex myoclonus. Adv Neurol. 1986;43:385-398.

114

THE EPILEPSIES 3 21. Ikeda A, Shibasaki H, Nagamine T, et al. Peri-rolandic and fronto-parietal components of scalprecorded giant SEPs in cortical myoclonus. Electroencephalogr Clin Neurophysiol. 1995;96:300-309. 22. Hitomi T, Ikeda A, Matsumoto R, et al. Generators and temporal succession of giant somatosensory evoked potentials in cortical reflex myoclonus: epicortical recording from sensorimotor cortex. Clin Neurophysiol. 2006;117:1481-1486. 23. Thompson PD, Day BL, Rothwell JC, Brown P, Britton TC, Marsden CD. The myoclonus in corticobasal degeneration. Evidence for two forms of cortical reflex myoclonus. Brain. 1994;117: 1197-1207. 24. Chadwick D, Hallett M, Harris R, Jenner P, Reynolds EH, Marsden CD. Clinical, biochemical, and physiologic features distinguishing myoclonus responsive to 5-hydroxy-tryptophan, tryptophan with a monoamine oxidase inhibitor, and clonazepam. Brain. 1977;100:455-487. 25. Obeso J, Rothwell JC, Marsden CD. The spectrum of cortical myoclonus: from focal reflex jerks to spontaneous motor epilepsy. Brain. 1985;108:193-224. 26. Chen R, Ashby P, Lang AE. Stimulus-sensitive myoclonus in akinetic-rigid syndromes. Brain. 1992;115:1875-1888. 27. Rodriguez ME, Artieda AJ, Zubieta JL, Obeso JA. Reflex myoclonus in olivopontocerebellar atrophy. J Neurol Neurosurg Psychiatry. 1994;57:316-319. 28. Young RR, Shahani BT. Clinical neurophysiological aspects of post-hypoxic intention myoclonus. Adv Neurol. 1979;26:85-105. 29. Wilkins DE, Hallett M, Berardelli A, Walshe T, Alvarez N. Physiologic analysis of the myoclonus of Alzheimer disease. Neurology. 1984;34:898-903. 30. Ugawa Y, Kkohara N, Hirasawa H, Kuzuhara S, Iwata M, Mannen T. Myoclonus in Alzheimer disease. J Neurol. 1987;235:90-94. 31. Rosing HS, Hopkins LC, Wallace DC, Epstein CM, Weidenheim K. Maternally inherited mitochondrial myopathy and myoclonic epilepsy. Ann Neurol. 1985;17:228-237. 32. So N, Berkovic S, Andermann F, Kuzniecky R, Gendron D, Quesney LF. Myoclonus epilepsy and ragged-red fibres (MERRF). Electrophysiological studies and comparison with other progressive myoclonus epilepsies. Brain. 1989;112:1261-1272. 33. Thompson PD, Bathia KP, Brown P, et al. Cortical myoclonus in Huntingtons disease. Mov Disord. 1994;9:633-641. 34. Guerrini R, Dravet C, Genton P, et al. Epileptic negative myoclonus. Neurology. 1993;43:1078-1083. 35. Shibasaki H. Pathophysiology of negative myoclonus and asterixis. Adv Neurol. 1995;67:199-210. 36. Guerrini R, Parmeggiani L, Shewmon A, Rubboli G, Tassinari CA. Motor dysfunction resulting from epileptic activity involving the sensori-motor cortex. In: Guerrini R, Aicardi J, Andermann F, Hallett M, eds. Epilepsy and Movement Disorder. Cambridge: Cambridge University Press; 2002:77-96. 37. Guerrini R, Belmonte A, Genton P. Antiepileptic drug-induced worsening of seizures in children. Epilepsia. 1998;39(Suppl. 3):S2-S10. 38. Parmeggiani L, Seri S, Bonanni P, Guerrini R. Electrophysiological characterization of spontaneous and carbamazepine-induced epileptic negative myoclonus in benign childhood epilepsy with centro-temporal spikes. Clin Neurophysiol. 2004;115:50-58. 39. Cerminara C, Montanaro ML, Curatolo P, Seri S. Lamotrigine-induced seizure aggravation and negative myoclonus in idiopathic rolandic epilepsy. Neurology. 2004;63:373-375. 40. Noachtar S, Holthausen H, Lu ders HO. Epileptic negative myoclonus. Subdural EEG recording indicates a postcentral generator. Neurology. 1997;49:1534-1537. 41. Ikeda A, Ohara S, Matsumoto R, et al. Role of primary sensorimotor cortices in generating inhibitory motor response in humans. Brain. 2000;123:1710-1721. 42. Rubboli G, Parmeggiani L, Tassinari CA. Frontal inhibitory spike component associated with epileptic negative myoclonus. Electroencephalogr Clin Neurophysiol. 1995;95:201-205. 43. Baumgartner C, Podreka I, Olbrich A, et al. Epileptic negative myoclonus: an EEG-single-photon emission CT study indicating involvement of premotor cortex. Neurology. 1996;46:753-758. 44. Meletti S, Tinuper P, Bisulli F, Santucci M. Epileptic negative myoclonus and brief asymmetric tonic seizures. A supplementary sensorimotor area involvement for both negative and positive motor phenomena. Epileptic Disord. 2000;2:163-168. 45. Tassinari CA, Rubboli G, Parmeggiani L, et al. Epileptic negative myoclonus. Adv Neurol. 1995;67:181-197. 46. Lu ders HO, Dinner DS, Morris HH, Wyllie E, Comair YG. Cortical electrical stimulation in humans. The negative motor areas. Adv Neurol. 1995;67:115-129. 47. Rubboli G, Mai R, Meletti S, et al. Negative myoclonus induced by cortical electrical stimulation in epileptic patients. Brain. 2006;129:65-81.

8  Cortical Myoclonus and Epilepsy: Overlap and Differences 48. Commission on classification and terminology of the international league against epilepsy: proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia. 1989;30:389-399. 49. Genton P, Malafosse A, Moulard B, et al. The progressive myoclonus epilepsies. In: Roger J, Bureau M, Dravet C, Genton P, Tassinari CA, Wolf P, eds. Epileptic Syndromes in Infancy and Adolescence. 4th ed. Montrouge: John Libbey Eurotext; 2005:424-441. 50. Badhwar A, Berkovic SF, Dowling JP, et al. Action myoclonus-renal failure syndrome: characterization of a unique cerebro-renal disorder. Brain. 2004;127:2173-2182. 51. Delgado-Escueta, AV, Ganesh S, Yamakawa K. Advances in the genetics of progressive myoclonus epilepsy. Am J Med Genet. 2001;106:129-138. 52. Guerrini R, Bonanni P, Marini C, Parmeggiani L. The myoclonic epilepsies. In: Willie E, Gupta A, Lachhwani DK, eds. The Treatment of Epilepsy. 4th ed. Philadelphia: Lippincott Williams and Wilkins; 2006:407-427. 53. Roger J, Genton P, Bureau M, Dravet C. Progressive myoclonus epilepsies in childhood and adolescence. In: Roger J, Bureau M, Dravet C, Dreifuss FE, Perret A, Wolf P, eds. Epileptic Syndromes in Infancy, Childhood and Adolescence. 2nd ed. London and Paris: John Libbey Eurotext Ltd.; 1992:381-400. 54. Guerrini R, Parmeggiani L, Volzone A. Cortical myoclonus in early childhood epilepsy. In: Majkowski J, Owczarek K, Zwolinski P, eds. Third European Congress of Epileptology. Bologna: Monduzzi Editore; 1998:99-105. 55. Cantello R, Gianelli M, Civardi C, Mutani R. Focal subcortical reflex myoclonus. A clinical and neurophysiological study. Arch Neurol. 1997;54:187-196. 56. Amir RE, Van den Veyver IB, Wan M, Tran CQ, Francke U, Zoghbi HY. Rett syndrome is caused by mutations in X-linked MECP2, encoding methyl-CpG-binding protein 2. Nat Genet. 1999;23: 185-188. 57. Auranen M, Vanhala R, Vosman M, et al. MECP2 gene analysis in classical Rett syndrome and in patients with Rett-like features. Neurology. 2001;56:611-617. 58. Scala E, Longo I, Ottimo F, et al. MECP2 deletions and genotype-phenotype correlation in Rett syndrome. Am J Med Genet A. 2007;143:2775-2784. 59. Guerrini R, Bonanni P, Parmeggiani L, Santucci M, Parmeggiani A, Sartucci F. Cortical reflex myoclonus in Rett syndrome. Ann Neurol. 1998;43:1-8. 60. Armstrong DD. Neuropathology of Rett syndrome. Ment Retard Dev Disabil Res Rev. 2002;8:72-76. 61. Carella F, Scaioli V, Ciano C, Binelli S, Oliva D, Girotti F. Adult onset myoclonic Huntingtons disease. Mov Disord. 1993;8:210-215. 62. Gambardella A, Muglia M, Labate A, etal: Juvenile Huntington disease presenting as progressive myoclonic epilepsy. Neurology. 2001;57:708-711. 63. Lance JW, Adams RD. The syndrome of intention or action myoclonus as a sequel to hypoxic encephalopathy. Brain. 1963;86:111-136. 64. Hallett M, Chadwick D, Adam J, Marsden CD. Reticular reflex myoclonus: a physiologic type of human post-hypoxic myoclonus. J Neurol Neurosurg Psychiatry. 1977;40:253-264. 65. Brown P, Thompson PD, Rothwell JC, Day BL, Marsden CD. A case of postanoxic encephalopathy with cortical action and brainstem reticular reflex myoclonus. Mov Disord. 1991;6:139-144. 66. Kojewnikow AY. Eine besondere Form von corticaler epilepsie. Neurologisches Zentralblatt. 1895;14:47-48. 67. Bancaud J. Kojewnikows syndrome (epilepsia partialis continua) in children. In: Roger J, Bureau M, Dravet C, Dreifuss FE, Perret A, Wolf P, eds. Epileptic Syndromes in Infancy, Childhood and Adolescence. 2nd ed. London and Paris: John Libbey Eurotext Ltd.; 1992:363-369. 68. Thomas IE, Raegan TJ, Klass DW. Epilepsia partialis continua. A review of 32 cases. Arch Neurol. 1977;34:266-275. 69. Fusco L, Bertini E, Vigevano F. Epilepsia partialis continua and neuronal migration anomalies. Brain Dev. 1992;14:323-328. 70. Kuzniecky R, Powers R. Epilepsia partialis continua due to cortical dysplasia. J Child Neurol. 1993;8:386-388. 71. Andrews PI, McNamara JO, Lewis DV. Clinical and electroencephalographic correlates in Rasmussens encephalitis. Epilepsia. 1997;38:189-194. 72. Rogers SW, Andrews PI, Gahring LC, et al. Autoantibodies to glutamate receptor GluR3 in Rasmussens encephalitis. Science. 1994;265:648-651. 73. Hart Y. Rasmussens encephalitis. Epileptic Disord. 2004;6:133-144. 74. Watson R, Jiang Y, Bermudez I, et al. Absence of antibodies to glutamate receptor type 3 (GluR3) in Rasmussen encephalitis. Neurology. 2004;63:43-50.

115

116

THE EPILEPSIES 3 75. Veggiotti P, Colamaria V, Dalla Bernardina B, Martelli A, Mangione D, Lanzi G. Epilepsia partialis continua in a case of MELAS: clinical and neurophysiological study. Neurophysiol Clin. 1995;25:158-166. 76. Ikeda A, Kakigi R, Funai N. Cortical tremor: a variant of cortical reflex myoclonus. Neurology. 1990;40:1561-1565. 77. Toro C, Pascual-Leone A, Deuschl G, Tate E, Pranzatelli MR, Hallett M. Cortical tremor: a common manifestation of cortical myoclonus. Neurology. 1993;43:2346-2353. 78. Terada K, Ikeda A, Mima T, et al. Familial cortical tremor as a unique form of cortical reflex myoclonus. Mov Disord. 1997;12:370-377. 79. Okuma Y, Shimo Y, Shimura H, et al. Familial cortical tremor with epilepsy: an under-recognized familial tremor. Clin Neurol Neurosurg. 1998;100:75-78. 80. Elia M, Musumeci SA, Ferri R, et al. Familial cortical tremor, epilepsy, and mental retardation: a distinct clinical entity? Arch Neurol. 1998;55:1569-1573. 81. Guerrini R, Bonanni P, Patrignani A, et al. Autosomal dominant cortical myoclonus and epilepsy (ADCME) with complex partial and generalized seizures: a newly recognized epilepsy syndrome with linkage to chromosome 2p11.1-q12.2. Brain. 2001;124:2459-2475. 82. Gardella E, Tinuper P, Marini C, et al. Autosomal dominant early-onset cortical myoclonus, photic-induced myoclonus, and epilepsy in a large pedigree. Epilepsia. 2006;47:1643-1649. 83. Carr JA, van der Walt PE, Nakayama J, et al. FAME 3: a novel form of progressive myoclonus and epilepsy. Neurology. 2007;68:1382-1389. 84. Guerrini R, De Lorey TM, Bonanni P, et al. Cortical myoclonus in Angelman syndrome. Ann Neurol. 1996;40:39-48. 85. Labauge P, Amer LO, Simonetta-Moreau M, et al. Absence of linkage to 8q24 in a European family with familial adult myoclonic epilepsy (FAME). Neurology. 2002;58:941-944. 86. Mikami M, Yasuda T, Terao A, et al. Localization of a gene for benign adult familial myoclonic epilepsy to chromosome 8q23.3-q24.1. Am J Hum Genet. 1999;65:745-751. 87. van Rootselaar AF, Callenbach PM, Hottenga JJ, et al. A Dutch family with familial cortical tremor with epilepsy. Clinical characteristics and exclusion of linkage to chromosome 8q23.3-q24.1. J Neurol. 2002;249:829-834. 88. Striano P, Chifari R, Striano S, et al. A new benign adult familial myoclonic epilepsy (BAFME) pedigree suggesting linkage to chromosome 2p11.1-q12.2. Epilepsia. 2004;45:190-192. 89. De Falco FA, Striano P, de Falco A, et al. Benign adult familial myoclonic epilepsy: Genetic heterogeneity and allelism with ADCME. Neurology. 2003;60:1381-1385. 90. Dravet C, Bureau M, Oguni H, et al. Severe myoclonic epilepsy in infancy (Dravet syndrome). In: Roger J, Bureau M, Dravet C, Genton P, Tassinari CA, Wolf P, eds. Epileptic Syndromes in Infancy, Childhood and Adolescence. 3rd ed. London and Paris: John Libbey; 2002:81-103. 91. Oguni H, Hayashi K, Awaya Y, Fukuyama Y, Osawa M. Severe myoclonic epilepsy in infantsa review based on the Tokyo Womens Medical University series of 84 cases. Brain Dev. 2001;23:736-748. 92. Chiron C. Prognosis of severe myoclonic epilepsy in infancy (Dravet syndrome). In: Jallon P, Berg A, Dulac O, Hauser A, eds. Prognosis of Epilepsies. Paris: John Libbey Eurotext; 2003:239-248. 93. Marini C, Mei D, Temudo T, et al. Idiopathic epilepsies with seizures precipitated by fever and SCN1A abnormalities. Epilepsia. 2007;48:1678-1685. 94. Beaumanoir A, Dravet C. The Lennox-Gastaut syndrome. In: Roger J, Bureau M, Dravet C, Dreifuss FE, Perret A, Wolf P, eds. Epileptic Syndromes in Infancy, Childhood and Adolescence. 2nd ed. London and Paris: John Libbey Eurotext Ltd; 1992:115-132. 95. Gastaut H. The Lennox-Gastaut syndrome: comments on the syndromes terminology and nosological position amongst the secondary generalized epilepsies of childhood. Electroencephalogr Clin Neurophysiol. 1982;35(Suppl):S71-S84. 96. Bonanni P, Parmeggiani L, Guerrini R. Different neurophysiologic patterns of myoclonus characterize Lennox-Gastaut syndrome and myoclonic astatic epilepsy. Epilepsia. 2002;43:609-615. 97. Wilkins DE, Hallett M, Erba G. Primary generalized epileptic myoclonus: a frequent manifestation of minipolymyoclonus of central origin. J Neurol Neurosurg Psychiatry. 1985;48:506-516. 98. Gloor P. Generalized epilepsy with spike-and-wave discharge: a reinterpretation of its electrographic and clinical manifestations. Epilepsia. 1979;20:571-588. 99. Dalla Bernardina B, Colamaria V, Capovilla G. Nosological classification of epilepsies in the first ` G, Di Perri R, Meinardi H, eds. Epilepsy: an Update on Research and three years of life. In: Nistico Therapy. New York: Alan Liss; 1983:165-183. 100. Dravet C, Bureau M, Roger J. Benign myoclonic epilepsy in infants. In: Roger J, Bureau M, Dravet C, Dreifuss FE, Perret A, Wolf P, eds. Epileptic Syndromes in Infancy, Childhood and Adolescence. 2nd ed. London and Paris: John Libbey Eurotext Ltd.; 1992:67-74.

8  Cortical Myoclonus and Epilepsy: Overlap and Differences 101. Guerrini R, Dravet C, Gobbi G, et al. Idiopathic generalized epilepsies with myoclonus in infancy and childhood. In: Malafosse A, Genton P, Hirsch E, Marescaux C, Broglin D, Bernasconi R, eds. Idiopathic Generalized Epilepsies: Clinical, Experimental and Genetics Aspects. London: John Libbey and Company; 1994:267-280. nignes du nourrisson. Epilepsies. 1990;2:95-101. 102. Dravet C. Les epilepsies myocloniques be 103. Dravet C, Bureau M. Benign myoclonic epilepsy in infancy. Adv Neurol. 2005;95:127-137. 104. Guerrini R, Aicardi J. Epileptic encephalopathies with myoclonic seizures in infants and children (severe myoclonic epilepsy and myoclonic-astatic epilepsy). Journal of Clin Neurophysiol. 2003;20:449-461. 105. Engel J, Jr. A proposed diagnostic scheme for people with epileptic seizures and with epilepsy: report of the ILAE Task Force on Classification and Terminology. Epilepsia. 2001;42: 796-803. 106. Ricci S, Cusmai R, Fusco L, et al. Reflex myoclonic epilepsy in infancy: a new age-dependent idiopathic epileptic syndrome related to startle reaction. Epilepsia. 1995;36:342-348. 107. Genton P, Salas-Puig X, Tunon A, Lahoz C, Del Soccorro M. Juvenile myoclonic epilepsy and related syndromes: clinical and neurophysiological aspects. In: Malafosse A, Genton P, Hirsch E, Marescaux C, Broglin D, Bernasconi R, eds. Idiopathic Generalized Epilepsies: Clinical, Experimental and Genetics Aspects. London: John Libbey and Company; 1994:253-266. 108. Cossette P, Liu L, Brisebois K, et al. Mutation of GABRA1 in an autosomal dominant form of juvenile myoclonic epilepsy. Nat Genet. 2002;31:184-189. 109. Haug K, Warnstedt M, Alekov AK, et al. Mutations in CLCN2 encoding a voltage-gated chloride channel are associated with idiopathic generalized epilepsies. Nat Genet. 2003;33:527-532. 110. Annesi F, Gambardella A, Michelucci R, et al. Mutational analysis of EFHC1 gene in Italian families with juvenile myoclonic epilepsy. Epilepsia. 2007;48:1686-1690. 111. Suzuki T, Delgado-Escueta AV, Aguan K, et al. Mutations in EFHC1 cause juvenile myoclonic epilepsy. Nat Genet. 2004;36:842-849. 112. Wolf P, Mayer, T. Juvenile myoclonic epilepsy: a syndrome challenging syndromic concepts? In: Schmitz B, Sander T, eds. Juvenile Myoclonic Epilepsy: The Janz Syndrome. Petersfield: Wrightson Biomedical Publishing; 2000:33-39. 113. Asconape J, Penry JK. Some clinical and EEG aspects of benign juvenile myoclonic epilepsy. Epilepsia. 1984;25:108-114. pilepsie myoclonique juvee nile dans la population 114. Salas-Puig X, Camara da, Silva AM, Dravet C. Le du Centre Saint Paul. Epilepsies. 1990;2:108-113. 115. Thomas P, Genton P, Gelisse P, Wolf P. Juvenile myoclonic epilepsy. In: Roger J, Bureau M, Dravet C, Genton P, Tassinari CA, Wolf P, eds. Epileptic Syndromes in Infancy, Childhood and Adolescence. 3rd ed. London: John Libbey & Co Ltd; 2002:335-356. 116. Panzica F, Rubboli G, Franceschetti S, et al. Cortical myoclonus in Janz syndrome. Clin Neurophysiol. 2001;112:1803-1809. 117. Doose H. Myoclonic astatic epilepsy of early childhood. In: Roger J, Bureau M, Dravet C, Dreifuss FE, Perret A, Wolf P, eds. Epileptic Syndromes in Infancy, Childhood and Adolescence. 2nd ed. London and Paris: John Libbey Eurotext Ltd.; 1992:103-114. 118. Kaminska A, Ickowicz A, Plouin P, Bru MF, Dellatolas G, Dulac O. Delineation of cryptogenic Lennox-Gastaut syndrome and myoclonic astatic epilepsy using multiple correspondence analysis. Epilepsy Res. 1999;36:15-29. 119. Oguni H, Tanaka T, Hayashi K, et al. Treatment and long-term prognosis of myoclonic-astatic epilepsy of early childhood. Neuropediatrics. 2002;33:122-132. 120. Guerrini R, Parmeggiani L, Bonnanni P, et al. Myoclonic astatic epilepsy. In: Roger J, Bureau M, Dravet C, Genton P, Tassinari CA, Wolf P, eds. Epileptic Syndromes in Infancy, Childhood and Adolescence. 4th ed. London and Paris: John Libbey & Co Ltd.; 2005:115-124. 121. Meisler MH, Kearney J, Ottman R, Escayg A. Identification of epilepsy genes in human and mouse. Annu Rev Genet. 2001;35:567-588. 122. Aicardi, J. Early myoclonic encephalopathy (neonatal myoclonic encephalopathy). In: Roger J, Bureau M, Dravet C, Dreifuss FE, Perret A, Wolf P. eds. Epileptic Syndromes in Infancy, Childhood and Adolescence. 2nd ed. London and Paris: John Libbey Eurotext Ltd.; 1992:13-23. 123. Dalla Bernardina B, Fontana E, Darra F. Myoclonic status in nonprogressive encephalopathies. Adv Neurol. 2005;95:59-70. 124. Tassinari CA, Bureau M, Thomas P. Epilepsy with myoclonic absences. In: Roger J, Bureau M, Dravet C, Dreifuss FE, Perret A, Wolf P, eds. Epileptic Syndromes in Infancy, Childhood and Adolescence. 2nd ed. London and Paris: John Libbey Eurotext Ltd.; 1992:151-160.

117

118

THE EPILEPSIES 3 125. Tassinari CA, Michelucci R, Rubboli G. Myoclonic absence epilepsy. In: Duncan JS, Panayiotopoulos CP, eds. Typical Absences and Related Epileptic Syndromes. London: Churchill Communications Europe; 1995:187-195. 126. Jeavons PM, Harding GFA. Photosensitive Epilepsy. London: Heinemann; 1975. 127. Jeavons P. Myoclonic epilepsies: therapy and prognosis. In: Akimoto H, Kazamatsuri H, Seino M, Ward AA, eds. Advances in Epileptology: XIIIth Epilepsy International Symposium. New York: Raven Press; 1982:141-144. 128. Kasteleijn-Nolst Trenite DG, Guerrini R, Binnie CD, et al. Visual sensitivity and epilepsy: a proposed terminology and classification for clinical and EEG phenomenology. Epilepsia. 2001;42:692-701. 129. Gastaut H, Broughton R. Epileptic Seizures. Springfield, IL: Charles C. Thomas; 1972. 130. Reilly EL, Peters JF. Relationship of some varieties of electroencephalographic photosensitivity to clinical convulsive disorders. Neurology. 1973;23:1050-1057. 131. Binnie CD, Darby CE, De Korte RA, et al. Self-induction of epileptic seizures by eye closure: incidence and recognition. J Neurol Neurosurg Psychiatry. 1980;43:386-389. 132. Tassinari CA, Rubboli G, Michelucci R. Reflex epilepsy. In: Dam M, Gram L, eds. Comprehensive Epileptology. New York: Raven Press; 1990:233-243. 133. Zuckermann EG, Glaser GH. Urea induced myoclonic seizures. Arch Neurol. 1972;27:14-28. 134. Talwar D, Arora MS, Sher PK. EEG changes and seizure exacerbation in young children treated with carbamazepine. Epilepsia. 1994;35:1154-1159. 135. Viani F, Romeo A, Viri M. Seizure and EEG patterns in Angelmans syndrome. J Child Neurol. 1995;10:467-471. 136. Lortie A, Chiron C, Mumford J. The potential for increasing seizure frequency, relapse, and appearance of new seizure types with vigabatrin. Neurology. 1993;43:24-27. 137. Marciani MG, Gigli GL, Maschio M. Vigabatrin-induced myoclonus in four cases of partial epilepsy. Epilepsia. 1995;36:107. 138. Marini C, Parmeggiani L, Masi G, DArcangelo G, Guerrini R. Nonconvulsive status epilepticus precipitated by carbamazepine presenting as dissociative and affective disorders in adolescents. J Child Neurol. 2005;20:693-696. 139. Dulac O, Kaminska A. Use of lamotrigine in Lennox-Gastaut and related epilepsy syndromes. J Child Neurol. 1997;12(Suppl 1):S23-S28. 140. Guerrini R, Dravet C, Genton P, Belmonte A, Kaminska A, Dulac O. Lamotrigine and seizure aggravation in severe myoclonic epilepsy. Epilepsia. 1998;39:508-512. 141. Briassoulis G, Kalabalikis P, Tamiolaki M, Hatzis T. Lamotrigine childhood overdose. Pediatr Neurol. 1998;19:239-242. 142. Guerrini R, Belmonte A, Parmeggiani L, Perucca E. Myoclonic status epilepticus following high dosage lamotrigine therapy. Brain Dev. 1999;21:420-424. 143. Biraben A, Allain H, Scarabin JM, Schu ck S, Edan G. Exacerbation of juvenile myoclonic epilepsy with lamotrigine. Neurology. 2000;55:1758. sonyi G, Hala sz P, et al. Disabling erratic myoclonus during lamotrigine therapy with 144. Janszky J, Ra high serum level: report of two cases. Clin Neuropharmacol. 2000;23:86-89. 145. Carrazzana EJ, Wheeler SD. Exacerbation of juvenile myoclonic epilepsy with lamotrigine. Neurology. 2001;56:1424-1425. 146. Minassian BA, DeLorey TM, Olsen RW, et al. Angelman syndrome: correlations between epilepsy phenotypes and genotypes. Ann Neurol. 1998;43:485-493. 147. Laan LA, Renier WO, Arts WF, et al. Evolution of epilepsy and EEG findings in Angelman syndrome. Epilepsia. 1997;38:195-199. 148. Vendrame M, Khurana DS, Cruz M, Melvin J, Valencia I, Legido A, Kothare SV. Aggravation of seizures and/or EEG features in children treated with oxcarbazepine monotherapy. Epilepsia. 2007;48:2116-2120. 149. Kuenzle C, Steinlin M, Wohlrab G, Boltshauser E, Schmitt B. Adverse effects of vigabatrin in Angelman syndrome. Epilepsia. 1998;39:1213-1215.

THE EPILEPSIES 3

The Life-Threatening Epilepsies of Childhood and Their Treatment


CATHERINE CHIRON

Introduction Lennox-Gastaut Syndrome Diagnosis and Etiology Treatment Particular Issues of LGS Regarding AED Trials Infantile Spasms Diagnosis and Etiology Medical Treatment Particular Issues of IS Regarding AED Trials

Dravet Syndrome Diagnosis and Etiology Treatment Particular Issues of Dravet Syndrome Regarding AED Trials Continuous Spike Waves During Sleep Diagnosis and Etiology Treatment Particular Issues of CSWS Regarding AED Trials

Introduction
In this chapter, we will define what the epileptology community usually calls catastrophic epilepsies as life-threatening epilepsies. To define them more precisely, we should rather consider the notion of epileptic encephalopathy. Epileptic encephalopathies (EE) are conditions in which neurologic deterioration results from the epileptic phenomenon by itself. It is usually considered that EE result from subcontinuous paroxysmal interictal activity (such as continuous spikes and waves during sleep), but most authors also include in the spectrum of EE the severe conditions due to seizures themselves (such as Dravet syndrome). To extensively cover the different problems posed by these severe epilepsies at pediatric age, we shall extend childhood to early age, from 1 month to 2 years, usually distinguished as infancy. EE represent about half of the pharmacoresistent epilepsies in childhood, the other half being partial epilepsy. EE disclose several characteristics when compared to adult epilepsies. Most of these characteristics depend on the fact that epilepsy occurs in a developing brain and take into account the following as far as therapeutic trials are concerned:  EE are specific for childhood; they do not exist in adults (at least in the form they have been revealed as life-threatening in pediatrics).

119

120

THE EPILEPSIES 3

 EE are age related. They appear at a given age range, and seizures, as well as epileptic activity, tend to disappear or at least to modify themselves at another given age.  As a result, a same child may often switch from one EE syndrome to another with age.  EE carry a risk of cognitive deterioration, due to high seizure frequency and/or major interictal paroxysmal anomalies.  EE most often associate both partial and generalized features, so that this distinction, considered as a key point in adults, is nonapplicable to children with life-threatening epilepsies.  EE are highly pharmacoresistent and may even be aggravated by some drugs commonly prescribed in other epilepsy syndromes.  Because most EE are diseases, very few trials are dedicated to them, so that children with EE may be considered therapeutic orphans.  Because there is growing evidence that appropriate treatment may improve brain function provided it is given early enough, developing new drugs is an urgent need for children with life-threatening epilepsies.  Performing drug trials in such populations face various difficulties of recruitment, ethics, endpoints, and duration, which can be solved using some innovative methodological approaches.1 Rather than an exhaustive review of the life-threatening epilepsies of childhood, we will extensively present four syndromes that differently illustrate the various specificities of EE detailed earlier: infantile spasms (IS), Dravet syndrome (DS), Lennox-Gastaut syndrome (LGS), and continuous spike waves during sleep (CSWS). They all are highly pharmacoresistent syndromes, but they represent completely different conditions regarding developing drugs. LGS benefited from a significant number of randomized controlled trials with four new drugs approved within the last 15 years. IS has two alternative therapeutic options approved, but DS has only one and CSWS none. The historical approach of the various drug trials within these four syndromes provides a complete view of the problems and the possible solutions for developing new compounds in the life-threatening epilepsies of childhood. Therefore, other life-threatening epilepsies such as myoclonic-astatic epilepsy (MAE), Rasmussen encephalitis, hemimegalencephaly, or Sturge-Weber disease are not included in this chapter.

Lennox-Gastaut Syndrome
DIAGNOSIS AND ETIOLOGY2 LGS develops between 2 and 8 years of age. It is a combination of various seizure types, mostly tonic and/or atonic seizures (usually called drop-attacks for their risk of falls and recurrent injury) and atypical absences, sometimes appearing as status epilepticus. On electroencephalogram (EEG), bilateral slow spike waves (< 2.5 Hz), generalized with bifrontal predominance, and bursts of rapid (10 Hz) rhythms during slow sleep (often corresponding to subclinical seizures) are most important for diagnosis. Psychomotor delay is present in 90% of cases, with slow behavior and frontal disorders as predominant symptoms. LGS can be cryptogenic in children with a previously normal development or symptomatic of congenital or

9  The Life-Threatening Epilepsies of Childhood and Their Treatment

121

acquired brain anomalies. Other types of epilepsy, especially IS, may precede LGS. Even if development was delayed prior to the onset of LGS, further mental deterioration is the rule due to persistently high rates of seizures together with interictal abnormal activity. Treating epilepsy has therefore ever been a key point in LGS. TREATMENT The open antiepileptic drugs (AEDs), the most used in LGS (valproate, benzodiazepines, and phenytoin), as well as ACTH and steroids, or the nondrug treatments, including ketogenic diet, corpus callosotomy, or vagal nerve stimulator, all remain poorly beneficial.2 Felbamate was the first of the new drugs shown to be effective in LGS by using the methodology of double-blind placebo-controlled adjunctive therapy in a randomized controlled trial (RCT).3 The risk of medullar aplasia and hepatotoxicity let the drug be proposed as the third line of treatment in severe cases. Then lamotrigine showed 33% of the lamotrigine group and 16% of the placebo group experienced a more than a 50% reduction in the frequency of all major seizures, including drop attacks. Global evaluations of patients functioning in terms of speech, language, and attention were significantly improved in the lamotrigine group.4 The efficacy of topiramate was also demonstrated for tonic seizures, and drop attacks at 3 months5 and maintained at long term in more than 50% of the patients.6 More recently, patients on rufinamide experienced a significant reduction in total seizure frequency and in drop attacks compared to patients on placebo, with 43% of responders for drop attacks compared to 17% on placebo.7 Despite this relatively high number of randomized controlled trials (the highest for an EE in children), the current management of LGS remains somewhat disappointing. The new drugs finally disclosed moderate efficacy (one-quarter to one-third of responders, less than 4% seizure-free patients), and comparative data are not available. Three of the drugs are associated with potentially severe adverse reactions (aplasia and hepatitis with felbamate, skin rash with lamotrigine, cognitive disorders with topiramate). A need remains for new well-tolerated drugs for LGS. PARTICULAR ISSUES OF LGS REGARDING AED TRIALS The prevalence of LGS is considered to be a maximum of 1 per 10,000 that meets the prevalence proposed for an orphan indication. The last new drug, rufinamide, was developed as an orphan drug for LGS in Europe, as clobazam currently is in the United States (this benzodiazepine is commonly prescribed in Europe and Canada as adjunctive therapy in LGS). Several types of seizures coexist in LGS; each of them may be present in types of epilepsy that are not LGS. Clinical trials in LGS should therefore include EEG features in diagnosis criteria and several seizure types as efficacy endpoints to select the drug appropriate for the epileptic syndrome, as opposed to the current approach, which is based on the seizure type. LGS provided the first attempt to perform a trial dedicated to a specific syndrome, using felbamate in 1993.3 LGS is a recognized distinct medical condition, but one of the more challenging aspects of LGS is the distinction from other childhood epilepsies that might mimic either the EEG or clinical pattern. For example, trying to differentiate LGS from MAE may be highly difficult based on age of onset, seizure type (especially in LGS forms

122

THE EPILEPSIES 3

with myoclonic jerks and MAE forms with tonic seizures), and slow spike-and-wave patterns on EEG.8 Only one feature is pathognomonic of LGS, the bursts of rapid (10 Hz) rhythms during slow sleep. However, it has not been required as diagnostic criteria for trials until now, mainly because it needed sleep EEG, which could not be systematically performed. As a result, among the patients included in the LGS trials, some likely matched more closely with MAE than with LGS. Further trials should ideally include this EEG feature, at least for stratifying the analysis on a subgroup of pure LGS patients. Another issue is the lack of specific data for the pediatric group in LGS. The trials, which were conducted with felbamate, lamotrigine, and topiramate included both adults and children, but neither the study design nor statistical analysis specifically took into account the pediatric subgroups. Because the syndrome looks different and carries different prognosis in children and in adults, further trials should present the results separately in the two populations, as the last rufinamide trial did.

Infantile Spasms
DIAGNOSIS AND ETIOLOGY9 IS usually occurs between 3 and 8 months of age, with a peak at 4 months. More and more cases have been reported with onset later than 1 year, which seem to be considered separately. Diagnosis lies on the classical triad, which associates a specific type of seizures (epileptic spasms), a specific EEG pattern (hypsarrhythmia), and psychomotor deterioration. In fact, one of the three components may be lacking or be atypical. Epileptic spasms consist of axial contraction in flexion or extension, in clusters that correspond to a large slow wave or to flattening with rapid rhythms on EEG. Spasms may be clinically subtle or even subclinical, particularly at onset of the disease or when they are incompletely controlled, thus making EEG a necessary tool to prove their complete disappearance. The features of spasms may be either symmetric or asymmetric. Video recording is often required for detailed analysis because asymmetry of spasms or a focal discharge combined with them are arguments for a focal cortical lesion to be at the origin of IS. Psychomotor deterioration is usually rapid from epilepsy onset, affecting head control, reaching for objects, defect in visual and auditive attention, and eyehand coordination.10,11 It is important to assess the status of psychomotor development (normal or abnormal) before onset because it carries prognostic implications, with a better outcome for patients without regression of eye tracking. Although about 30% of IS present with normal MRI, the so-called cryptogenic/ idiopathic IS, a large panel of cerebral lesions can be the cause in the remaining cases, including cortical malformations (such as agyria-pachygyria, hemimegalencephaly or focal cortical dysplasia), sequelae of pre-, per-, or postnatal anoxic-ischemia (such as periventricular leukomalacia of premature infants, porencephaly, or sequelae of subdural hematoma), infection of the central nervous system (such as meningitis or encephalitis), neurocutaneous syndromes (such as tuberous sclerosis or neurofibromatosis), chromosome disorders (such as Down syndrome or mutations in ARX, STK9 or Kir6.2 genes),12 or inherited metabolic disorders (such as pyridoxine dependency, Menkes disease, or mitochondrial disease due to NARP mutation).

9  The Life-Threatening Epilepsies of Childhood and Their Treatment

123

As a result, specific etiologic treatment (pyridoxine, surgery) is rarely possible, although it must be actively evaluated. MEDICAL TREATMENT9 Most AEDs are usually inefficient in IS, and most of them were tested through open trials (valproate, benzodiazepines, piridoxine, lamotrigine, topiramate, felbamate, zonisamide, ketogenic diet, and thyrotropin-releasing hormone). The two major therapeutic approaches consist of hormonal treatment (ACTH and steroids) and vigabatrin. Considering an evidence-based approach, vigabatrin is less effective as a hormonal treatment at short term (2 weeks), but is as effective at 1-year followup.13,14 They both have side effects, although they are different in potential severity; vigabatrin may induce bilateral restriction of the peripheral visual field in around 20% of cases, whereas hormonal therapy carries a mortality rate up to 5%. However, a benefit with epilepsy seems to be associated with a mental benefit over the long term in IS: developmental and socialization outcome is favorably influenced by the initial and rapid control of spasms with vigabatrin in tuberous sclerosis15 and with steroids in cryptogenic and even symptomatic cases.16 ACTH usually controlled seizures initially in about 75% of the patients at a dose of 40 IU. A lower dose (20 IU) is less efficient; a higher dose (150 IU) is more efficient, but carries a higher relapse rate. The incidence of adverse events (infections, increased arterial blood pressure, gastritis, and hyperexcitability) reaches almost 100% if one considers Cushing effect. Tetracosactin (synthetic corticotropin) seems to be even less well tolerated than ACTH, whereas oral steroids (hydrocortisone, prednisone) induce side effects in less than 20% of cases. In a prospective, randomized, blind approach, the efficacy of prednisone (at 2 mg/kg/d) was equal to that of corticotropin. Unfortunately, no controlled study has been conducted comparing hormonal treatment to placebo in IS. Vigabatrin demonstrated its efficacy on IS as a first-line monotherapy at doses superior to 100 mg/kg/day compared to placebo or to low doses.17,18 Overall, more than one-third of children can be expected to have complete resolution of spasms, but the response rate mainly depends on etiology; it reached 90% in infants with tuberous sclerosis19 and 54% in patients with a variety of conditions other than tuberous sclerosis.13 In cryptogenic cases, the success rate may reach 100% when adding ACTH to patients not responding to vigabatrin monotherapy.20 The most preoccupying side effect of vigabatrin is its retinal toxicity, which induces visual field constriction. Although asymptomatic in most pediatric cases, rather less frequent in children (around 20%) than in adults (around 30%),21 and never reported in children who received less than 15 months of vigabatrin exposure,22 there is still no validated means to detect such a visual field defect before the age of 6 to 8 years. PARTICULAR ISSUES OF IS REGARDING AED TRIALS The risk of rapid cognitive decline allows a reduction in the duration of the evaluation period in double-blind conditions. Given the high rate of seizures, it was successfully limited to a maximum of 2 weeks in the two last RCTs in IS.13,17 Both epilepsy and its etiology contribute to the prognosis of IS. Epilepsy adds its own severity to this status, not only through the frequent seizures, but also through

124

THE EPILEPSIES 3

the interictal paroxysmal. The treatment of IS should therefore have two goals: to control seizures and to control hypsarrhythmia. Complete cessation of spasms is mandatory; the decrease of seizure rate by over 50%, which is how responders are usually defined in trials, is not adequate in IS trials. One should also consider the interictal EEG activity as a surrogate marker. Treatment strategy in IS depends on both the local use and the drug availability in various countries. Regarding vigabatrin, the question that remains is how long treatment should be continued; in Down syndrome, 6-month control of spasms was followed by no relapse after vigabatrin discontinuation, whereas a devastating relapse was possible until the age of 5 years in tuberous sclerosis.23 Further trials are still needed in IS for studying efficacy and safety of new compounds at short and long term, as well as the mechanisms of the retinal toxicity of vigabatrin in children.

Dravet Syndrome
DIAGNOSIS AND ETIOLOGY The first seizures occur between 2 and 9 months of age.24 They are tonic-clonic or clonic seizures, either generalized or affecting alternatively one side of the body, often prolonged, resulting in recurrent status epilepticus and often provoked by fever. Children typically have a normal perinatal history and initially present with normal psychomotor development, normal neurological examination, and normal EEG between seizures. The pattern changes from the second year on: tonic-clonic or clonic seizures persist with the same characteristics, but additional myoclonia, atypical absences, and partial seizures occur; generalized spike and waves are observed during sleep; patients develop ataxia, hyperactivity, and mental retardation; and EEG shows spontaneous generalized spike waves and polyspike waves, as well as a slowing down of the background activity. Borderline forms were identified by Japanese authors with potential later onset, no later myoclonia, better cognitive outcome, and less pharmacoresistance.25 The relationships with the typical form previously described are still to be established. Nonsense mutations of SCN1a gene, which codes for the a1 subunit of voltagedependant sodium channel, have been identified in up to 70% of patients, including some microdeletions and duplications, which require more sophisticated procedure.26,27,28 TREATMENT Most authors agree that valproate and benzodiazepines may decrease the frequency and duration of afebrile convulsive seizures, but the effect is only moderate. Some investigators associate phenobarbital, bromide, or phenytoin, depending on the country, usually with unsatisfactory results. Paradoxically, lamotrigine and carbamazepine can aggravate seizures and should be avoided.29 By contrast, two new drugs are helpful for treating patients with DS, stiripentol and topiramate. Stiripentol efficacy was shown in one open and then two randomized placebo-controlled trials, independently conducted in France and Italy in

9  The Life-Threatening Epilepsies of Childhood and Their Treatment

125

children with DS and receiving concomitant therapy with clobazam (CLB) and valproic acid (VPA).30,31 Despite a relatively small sample size in both trials (41 and 23 patients), 71% and 67% of patients, respectively, were responders on STP against 5% and 9%, respectively, on placebo. Tolerability was acceptable provided the dose of medication was diminished, because STP inhibits the cytochrome p450 (CYP) system in the liver, resulting in an increased plasma concentration of concomitant AED, particularly clobazam, mainly through CYP 2C19.31,32 In the long term, the frequency and duration of seizures remained significantly reduced, as was the number of episodes of convulsive status epilepticus.33 Topiramate has not been as extensively studied in DS, and only data from three open-labeled trials are available, with 55% of responders in two of them.34,35 Side effects are mainly related to rapid dosage titration and, to some extent, to the association with valproate (such as apathy and elevated blood ammonia levels). The association of topiramate to stiripentol does not need any particular adaptation of dosages and may be helpful and well tolerated in patients unsatisfactorily controlled with stiripentol.36 Preliminary open reports are also emerging using levetiracetam as adjunctive therapy with encouraging results.37 PARTICULAR ISSUES OF DRAVET SYNDROME REGARDING AED TRIALS The severity of Dravet syndrome is due to its unfavorable prognosis for both mental and vital status. The risk of sudden unexpected death in epilepsy (SUDEP) is among the highest within all epilepsy syndromes (about 15% compared to 5%). The psychomotor development of almost all affected children is poor, evolving from a normal status before the beginning of the disease to a severe mental retardation around the age of 4 to 5 years. Epileptic seizures, and especially the number of status epilepticus in the first years of life, are likely to carry some responsibility in the mental retardation,38 so that any drug that could decrease their frequency, as STP does, may be beneficial as soon as the diagnosis is confirmed. Dravet syndrome is a relatively easily identifiable disease, and diagnosis can be performed based on electroclinical criteria in a significant proportion of cases as early as 6 months of age. A score of early diagnosis is currently under validation in Japan.39 Thus, there is no reason to restrict the further trials to infants over the age of 1 year. Although identifying a SCN1A mutation provides a strong argument for diagnosis in atypical forms, diagnosis still lies in electroclinical arguments because any mutation is lacking in a significant proportion of patients with confirmed Dravet syndrome, whereas patients with other types of epilepsy may exhibit SCN1A mutation. Dravet syndrome is a rare disease affecting 1/30,000 to 1/40,000 of children. As for other rare epilepsy syndromes, only multicenter trials can be successfully conducted. A rational approach of therapeutics could be possible in the future for Dravet syndrome in light of the defect described on a sodium channel gene in these patients: AEDs working mainly by blocking sodium channels (such as carbamazepine, phenytoin, and lamotrigine) are likely not to be effective and even to worsen seizures, whereas broad-spectrum AEDs (like valproate, benzodiazepines, and topiramate) are likely to be effective.40

126

THE EPILEPSIES 3

Continuous Spike Waves During Sleep


DIAGNOSIS AND ETIOLOGY41 Onset is during childhood around 2 to 10 years of age. Seizures are rare, simple, or complex partial and atypical absences and drop attacks. The crucial feature is the regression of intellectual abilities, close to a CSWS pattern. All cognitive functions may be involved; some patients with verbal agnosia behave like deaf children, thus resulting in Landau-Kleffner syndrome, whereas others exhibit an acquired frontal syndrome42 with behavioral changes that may mimic psychosis or dementia. Negative myoclonus43 and oro-buccofacial apraxia have also been reported. Deterioration may be missed if the child presented with previous psychomotor delay, and CSWS diagnosis was therefore overlooked. Sleep EEG constantly shows bilateral continuous high-frequency spike wave activity (1.5 to 5 Hz). The proportion of EEG tracing with spikes is over 85% during slow sleep (but it may be around 50% in Landau-Kleffner syndrome).44 Focal activity is usually detected during awake EEG, in accordance with the particular functions involved. Etiology comprises both cryptogenic and symptomatic cases, the later mainly resulting from pre- or perinatal anoxic-ischemia or unilateral polymicrogyria.45 TREATMENT41 Benzodiazepines (BZ) in monotherapy and ethosuximide may be effective in some cryptogenic cases, whereas carbamazepine may worsen the condition, and barbiturates, lamotrigine, and even valproate may be inefficient. Steroids have a favorable and lasting effect and are justified when BZ are not efficient (most often in symptomatic patients), provided corticotherapy will be administered for at least 1 year to avoid possible relapse. Preliminary encouraging reports using open topiramate and levetiracetam46 need to be confirmed. PARTICULAR ISSUES OF CSWS REGARDING AED TRIALS Regarding the few therapeutic options available in CSWS and the complete absence of controlled data, randomized controlled trials with new drugs are dramatically needed in this epilepsy syndrome. Many studies indicate that the potential for early reversibility of clinical and EEG abnormalities will improve the outcome of children with CSWS. The longer the duration of CSWS, the more damage it causes. Trials would therefore ideally be needed early in the disease. Methodology should be adapted to the characteristics of the syndrome; sleep EEG pattern is preferred to seizure count as an efficacy primary endpoint, and neuropsychological course should be assessed using standardized age-related tests.

Conclusion
Contrary to what has been shown in adults with epilepsy, the early choice of an adapted treatment is a key point in infants and children with EE.47 However, they

9  The Life-Threatening Epilepsies of Childhood and Their Treatment

127

remain therapeutic orphans, although they represent the most frequent and deleterious disorders in the field of epilepsy.48 EE do not exist in adults and therefore require specific trials. Rather than apply the guidelines for AED trials in adults, the design and the methodology of trials in EE need to be adapted to the particulars of these epilepsy syndromes (i.e., rare and age-related diseases with rapid and severe cognitive deterioration). REFERENCES
1. Chiron C, Dulac O, Pons G. Antiepileptic drug development in children: considerations for a revisited strategy. Drugs. 2008;68:17-25. 2. Genton P, Dravet C. Lennox-Gastaut syndrome. In: Engel J Jr, Pedley TA, eds. Epilepsy: A Comprehensive Textbook. 2nd ed. Philadelphia: Lippincott Williams & Wilkins; 2008:2417-2428. 3. Efficacy of felbamate in childhood epileptic encephalopathy (Lennox-Gastaut syndrome). The Felbamate Study Group in Lennox-Gastaut Syndrome. N Engl J Med. 1993;328:29-33. 4. Motte J, Trevathan E, Arvidsson JF, Barrera MN, Mullens EL, Manasco P. Lamotrigine for generalized seizures associated with the Lennox-Gastaut syndrome. Lamictal Lennox-Gastaut Study Group. N Engl J Med. 1997;337:1807-1812. 5. Sachdeo RC, Glauser TA, Ritter F, Reife R, Lim P, Pledger G. A double-blind, randomized trial of topiramate in Lennox-Gastaut syndrome. Topiramate YL Study Group. Neurology. 1999;52: 1882-1887. 6. Glauser TA, Levisohn PM, Ritter F, Sachdeo RC. Topiramate in Lennox-Gastaut syndrome: openlabel treatment of patients completing a randomized controlled trial. Topiramate YL Study Group. Epilepsia. 2000;41(Suppl 1):S86-S90. 7. Arroyo S. Rufinamide. Neurotherapeutics. 2007;4:155-162. 8. Kaminska A, Ickowicz A, Plouin P, Bru MF, Dellatolas G, Dulac O. Delineation of cryptogenic Lennox-Gastaut syndrome and myoclonic astatic epilepsy using multiple correspondence analysis. Epilepsy Res. 1999;36:15-29. 9. Dulac O, Dalla BB, Chiron C. West syndrome. In: Engel J Jr, Pedley TA, eds. Epilepsy: A Comprehensive Textbook. 2nd ed. Philadelphia: Lippincott Williams & Wilkins; 2008:2329-2336. 10. Guzzetta F, Frisone MF, Ricci D, Rando T, Guzzetta A. Development of visual attention in West syndrome. Epilepsia. 2002;43:757-763. 11. Rando T, Baranello G, Ricci D, et al. Cognitive competence at the onset of West syndrome: correlation with EEG patterns and visual function. Dev Med Child Neurol. 2005;47:760-765. 12. Bahi-Buisson N, Eisermann M, Nivot S, et al. Infantile spasms as an epileptic feature of DEND syndrome associated with an activating mutation in the potassium adenosine triphosphate (ATP) channel, Kir6.2. J Child Neurol. 2007;22:1147-1150. 13. Lux AL, Edwards SW, Hancock E, et al. The United Kingdom Infantile Spasms Study comparing vigabatrin with prednisolone or tetracosactide at 14 days: a multicentre, randomised controlled trial. Lancet. 2004;364:1773-1778. 14. Lux AL, Edwards SW, Hancock E, et al. The United Kingdom Infantile Spasms Study (UKISS) comparing hormone treatment with vigabatrin on developmental and epilepsy outcomes to age 14 months: a multicentre randomised trial. Lancet Neurol. 2005;4:712-717. 15. Jambaque I, Chiron C, Dumas C, Mumford J, Dulac O. Mental and behavioural outcome of infantile epilepsy treated by vigabatrin in tuberous sclerosis patients. Epilepsy Res. 2000;38:151-160. 16. Riikonen R. Long-term outcome of patients with West syndrome. Brain Dev. 2001;23:683-687. 17. Appleton RE, Peters AC, Mumford JP, Shaw DE. Randomised, placebo-controlled study of vigabatrin as first-line treatment of infantile spasms. Epilepsia. 1999;40:1627-1633. 18. Elterman RD, Shields WD, Mansfield KA, Nakagawa J. Randomized trial of vigabatrin in patients with infantile spasms. Neurology. 2001;57:1416-1421. 19. Chiron C, Dumas C, Jambaque I, Mumford J, Dulac O. Randomized trial comparing vigabatrin and hydrocortisone in infantile spasms due to tuberous sclerosis. Epilepsy Res. 1997;26:389-395. 20. Granstrom ML, Gaily E, Liukkonen E. Treatment of infantile spasms: results of a population-based study with vigabatrin as the first drug for spasms. Epilepsia. 1999;40:950-957. 21. Wild JM, Ahn HS, Baulac M, et al. Vigabatrin and epilepsy: lessons learned. Epilepsia. 2007;48:1318-1327. 22. Vanhatalo S, Nousiainen I, Eriksson K, et al. Visual field constriction in 91 Finnish children treated with vigabatrin. Epilepsia. 2002;43:748-756.

128

THE EPILEPSIES 3 23. Kroll-Seger J, Kaminska A, Moutard ML, et al. Severe relapse of epilepsy after vigabatrin withdrawal: for how long should we treat symptomatic infantile spasms? Epilepsia. 2007;48:612-613. 24. Dravet C, Bureau M. Severe myoclonic epilepsy in infancy (Dravet syndrome). In: Engel J Jr, Pedley TA, eds. Epilepsy: A Comprehensive Textbook. 2nd ed. Philadelphia: Lippincott Williams & Wilkins; 2008:2337-2342. 25. Oguni H, Hayashi K, Oguni M, et al. Treatment of severe myoclonic epilepsy in infants with bromide and its borderline variant. Epilepsia. 1994;35:1140-1145. 26. Claes L, Del Favero J, Ceulemans B, Lagae L, Van Broeckhoven C, De Jonghe P. De novo mutations in the sodium-channel gene SCN1A cause severe myoclonic epilepsy of infancy. Am J Hum Genet. 2001;68:1327-1332. 27. Sugawara T, Mazaki-Miyazaki E, Fukushima K, et al. Frequent mutations of SCN1A in severe myoclonic epilepsy in infancy. Neurology. 2002;58:1122-1124. 28. Marini C, Temudo T, Ferrari AR, et al. Idiopathic epilepsies with seizures precipitated by fever and SCN1A abnormalities. Epilepsia. 2007;48:1678-1685. 29. Guerrini R, Dravet C, Genton P, Belmonte A, Kaminska A, Dulac O. Lamotrigine and seizure aggravation in severe myoclonic epilepsy. Epilepsia. 1998;39:508-512. 30. Perez J, Chiron C, Musial C, et al. Stiripentol: efficacy and tolerability in children with epilepsy. Epilepsia. 1999;40:1618-1626. 31. Chiron C, Marchand MC, Tran A, et al. Stiripentol in severe myoclonic epilepsy in infancy: a randomised placebo-controlled syndrome-dedicated trial. STICLO study group. Lancet. 2000;356:1638-1642. 32. Giraud C, Treluyer JM, Rey E, et al. In vitro and in vivo inhibitory effect of stiripentol on clobazam metabolism. Drug Metab Dispos. 2006;34:608-611. et tole rance ` 33. Thanh TN, Chiron C, Dellatolas G, et al. Efficacite a long terme du stiripentol dans le pilepsie myoclonique se ve `re du nourrisson (syndrome de Dravet). Arch Pediatr. traitement de le 2002;9:1120-1127. 34. Nieto-Barrera M, Candau R, Nieto-Jimenez M, Correa A, del Portal LR. Topiramate in the treatment of severe myoclonic epilepsy in infancy. Seizure. 2000;9:590-594. 35. Coppola G, Capovilla G, Montagnini A, et al. Topiramate as add-on drug in severe myoclonic epilepsy in infancy: an Italian multicenter open trial. Epilepsy Res. 2002;49:45-48. 36. Kroll-Seger J, Portilla P, Dulac O, Chiron C. Topiramate in the treatment of highly refractory patients with Dravet syndrome. Neuropediatrics. 2006;37:325-329. 37. Striano P, Coppola G, Pezella M, et al. An open-label trial of levetiracetam in severe myoclonic epilepsy of infancy. Neurology. 2007;69:250-254. 38. Nabbout R, Gennaro E, Dalla BB, et al. Spectrum of SCN1A mutations in severe myoclonic epilepsy of infancy. Neurology. 2003;60:1961-1967. 39. Hattori H, Ouchida M, Ono J, et al. A screening test for the prediction of Dravet syndrome before one year of age. Epilepsia. 2007;11:1-8. 40. Ceulemans B, Boel M, Claes L, et al. Severe myoclonic epilepsy in infancy: toward an optimal treatment. J Child Neurol. 2004;19:516-521. 41. Smith MC. Landau-Kleffner Syndrome and CSWS. In: Engel J Jr, Pedley TA, eds. Epilepsy: A Comprehensive Textbook. 2nd ed. Philadelphia: Lippincott Williams & Wilkins; 2008:2429-2436. 42. Roulet E, Davidoff V, Despland PA, Deonna T. Mental and behavioural deterioration of children with epilepsy and CSWS: acquired epileptic frontal syndrome. Dev Med Child Neurol. 1993;35:661-674. 43. Guerrini R, Dravet C, Genton P, et al. Epileptic negative myoclonus. Neurology. 1993;43:1078-1083. 44. Veggiotti P, Beccaria F, Guerrini R, Capovilla G, Lanzi G. Continuous spike-and-wave activity during slow-wave sleep: syndrome or EEG pattern? Epilepsia. 1999;40:1593-1601. 45. Guerrini R, Genton P, Bureau M, et al. Multilobar polymicrogyria, intractable drop attack seizures, and sleep-related electrical status epilepticus. Neurology. 1998;51:504-512. 46. Aeby A, Poznanski N, Verheulpen D, Wetzburger C, Van BP. Levetiracetam efficacy in epileptic syndromes with continuous spikes and waves during slow sleep: experience in 12 cases. Epilepsia. 2005;46:1937-1942. 47. Arroyo S, Brodie MJ, Avanzini G, et al. Is refractory epilepsy preventable? Epilepsia. 2002;43: 437-444. 48. Trevathan E. Antiepileptic drug development for therapeutic orphans. Epilepsia. 2003;44(Suppl 7):19-25.

THE EPILEPSIES 3

10

The Spectrum of Epilepsies Associated with Generalized Spike-and-Wave Patterns


MICHAEL KOUTROUMANIDIS  CHRYSOSTOMOS P. PANAYIOTOPOULOS

Introduction The International Classification of Epileptic Seizures and Syndromes: Basic Concepts and Definitions Conventions and Limitations of the ILAE Classification System and Syndromic Diagnosis in Clinical Practice IGE IGE Syndromes Recognized by the International League Against Epilepsy (ILAE) Benign Myoclonic Epilepsy in Infancy (BMEI) Epilepsy with Myoclonic-Astatic Seizures (MAE) Childhood Absence Epilepsy (CAE) Epilepsy with Myoclonic Absences (E-MA) Juvenile Absence Epilepsy (JAE) Juvenile Myoclonic Epilepsy (JME) IGE with GTCS Only (IGE/GTCS) IGE Syndromes that Have Not Been Recognized by the ILAE

Eyelid Myoclonia with Absences (ELMA or Jeavons Syndrome) IGE with Phantom Absences (PA) Perioral Myoclonia with Absences (PMA) The Syndrome of De Novo Absence-like Status of Late Onset Other Epileptic Syndromes and Conditions Associated with GSW Activity Cryptogenic/Symptomatic Generalized Epileptic Syndromes and Epileptic Encephalopathies Epileptic Encephalopathy with Spike-Wave during Sleep (CSWS) including Landau-Kleffner Syndrome (LKS) Idiopathic Focal Epilepsies of Childhood Cryptogenic/Symptomatic Focal Epilepsies

129

130

THE EPILEPSIES 3

Introduction
The generic term epilepsy embraces a particular group of paroxysmal brain disorders, but by itself is of limited clinical value. Intense clinical and genetic research over the last few decades has identified a large number of well-defined epilepsy syndromes with different clinical, EEG, neuropsychological, and neuroimaging profiles, natural history and prognosis, and conditions that ultimately require different management. There is evidence that early treatment can reduce the risk of seizure recurrence,1 and its efficacy depends largely on the appropriate drug choice in relation to the particular clinical syndrome.2 Therefore, identification of the particular form or syndrome is the cornerstone of meaningful, optimal management of the individual patient.2,3 THE INTERNATIONAL CLASSIFICATION OF EPILEPTIC SEIZURES AND SYNDROMES: BASIC CONCEPTS AND DEFINITIONS Apart from assisting patients clinical management and prognostication, recognition of different syndromes enables effective communication between clinicians and clinical and genetic research. The current classification of epileptic seizures4 and epilepsies5 of the International League Against Epilepsy (ILAE) and also the new ILAE proposal6,7 are essentially organized around two dichotomies in terms of their topography between generalized and localized (partial or focal) and in terms of their etiology between idiopathic epilepsies (genetically determined) and symptomatic epilepsies (due to cerebral pathology). The ILAE classification is currently under revision, but is still valid. Presumed symptomatic epilepsies, in which no certain etiology can be demonstrated, are sometimes referred to as cryptogenic. Generalized epilepsies manifest with generalized seizures, whose first ictal clinical changes reflect involvement of both hemispheres and in which initial EEG changes are bilateral, as, for example, the typical absences (TA) that are associated with 3-Hz GSW (Figure 10-1). Localization-related epilepsies (also called focal or partial) manifest with focal (or partial) seizures, in which the initial activation of a group of neurons is limited to a part of one hemisphere5; the EEG shows focal abnormalities, but GSW discharges may occur as a result of secondary bilateral synchrony (SBS).8 CONVENTIONS AND LIMITATIONS OF THE ILAE CLASSIFICATION SYSTEM AND SYNDROMIC DIAGNOSIS IN CLINICAL PRACTICE Identification of the seizure type(s) is the quintessential element of an epileptic syndrome and the first step for its definition.2 However, it is not always straightforward, particularly on historical evidence (accounts of patients and witnesses) only. TA in idiopathic generalized epilepsy (IGE), for example, may share some principal clinical features with limbic complex partial seizures (CPS) (such as unresponsiveness and automatisms), and one must turn to either their onset (presence or absence of aura) or immediately post-termination (presence or absence of confusion) to make an educated clinical hypothesis. Interictal EEG studies will assist diagnosis by showing focal epileptiform activity or GSW, but even so the rate of misdiagnosis is high with the generalized epilepsies being erroneously diagnosed as focal rather than the converse.9,10 Important reasons for such misdiagnosis include usually short-lived focal clinical symptoms and signs, such as unilateral jerks and rotatory

10  The Spectrum of Epilepsies Associated with Generalized Spike-and-Wave Patterns

131

Childhood absence epilepsy

Fp2-F4

Fp1-F3 200 V 1 sec

Juvenile absence epilepsy

Fp2-F4

Fp1-F3 150 V 1 sec Symptomatic myoclonic absence epilepsy Fp2-F4 Fp1-F3 100 V 1 sec Phantom absences Absences with single myoclonic jerks

Fp2-F4

Fp1-F3 200 V 1 sec IGE of undetermined classification Fp2-F4 Juvenile myoclonic epilepsy

Fp1-F3 150 V 1 sec 200 V 1 sec Figure 101 Generalized spike-and-wave discharges in patients with idiopathic and symptomatic generalized absence epilepsies. (Reprinted with permission from Panayiotopoulos CP. A Clinical Guide to Epileptic Syndromes and Their Treatment. 2nd ed. London: Springer; 2007.)

132

THE EPILEPSIES 3

seizures in juvenile myoclonic epilepsy (JME),1013 versive absences,14 and in addition, well-recognized asymmetrical or focal interictal EEG changes.11,15,16 Such electroclinical focalities within the generalized seizures are interpretable within the model of generalized corticoreticular epilepsy, but stand uncomfortably within the inflexible and simplified generalized versus focal dichotomy of the ILAE seizure and syndrome classification system.17 Focal epilepsies may (also) present with rapidly generalized convulsions and infrequently with interictal GSW discharges as in SBS. In this chapter, we will discuss primarily the various syndromes of IGE, whose GSW is the defining electrographic trait. From the brief foreword, it follows that GSW discharges do not always indicate IGE or indeed generalized epilepsies, and Table 10-1 includes all the main epilepsies and syndromes that are associated with 34 Hz GSW. The interictal EEG alone cannot be used for establishing or excluding the diagnosis of epilepsy (including IGEs) or provide oversimplified clues for reliable syndromic diagnosis; this can only be based on a comprehensive electroclinical synthesis for the individual patient, without which EEG interpretation may be completely meaningless and even misleading.2,17

IGE
Notwithstanding the aforementioned diagnostic limitations and pitfalls, GSW discharges are the characteristic EEG building blocks of all seizures in IGE that include typical absences (TA), myoclonic seizures (MS), and GTCS. EEG background activity is normal. IGE comprise a group of genetically determined epilepsies, unrelated to any structural brain pathology, and associated with normal neurological and neuropsychological status. Suitable AED include valproic acid VPA (arguably the first choice), clonazepam (CLZ), ethosuximide (ESX), lamotrigine (LTG), levetiracetam (LEV), topiramate (TPM), and zonisamide (ZNS) (with specific indications against main seizure types),1820 whereas carbamazepine (CBZ), oxcarbazepine (OCBZ), vigabatrin (VGB), tiagabine (TGB), gabapentin (GBP), pregabalin (PGB), and phenytoin (PHT) are generally contraindicated, as they can cause seizure worsening.2,2023 IGE SYNDROMES RECOGNIZED BY THE INTERNATIONAL LEAGUE AGAINST EPILEPSY (ILAE)6 Benign myoclonic epilepsy in infancy Epilepsy with myoclonic-astatic seizures (idiopathic form) Epilepsy with myoclonic absences (idiopathic form) Childhood absence epilepsy Juvenile absence epilepsy Juvenile myoclonic epilepsy Epilepsy with GTCS only BENIGN MYOCLONIC EPILEPSY IN INFANCY (BMEI)2,2426 This is the earliest form of IGE with a prevalence of about 1% to 2% of epilepsies that start before the age of 4 years. MS start between 4 months to 3 to 4 years and are typically favored by sleep. Two-thirds of the affected children are boys. MS are frequently described by parents as spasms or head-nodding attacks,

TABLE 101

Epilepsy Syndromes and Conditions Associated with GSW Discharges


Seizures Types Interictal EEG

IGE Absence syndromes CAE, JAE Phantom TA Myoclonic syndromes BMEI JME EMA/PMA GTCS only Mixed/unclassified

TA, MS, GTCS, rarely tonic Normal background 2.54 Hz regular GSW, TA, late GTCS, rarely MS nonlocalizing focal (in JAE) Brief TA, GTCS; frequent ASE 3-6 Hz irregular GPSW MS with fragmentations MS, GTCS, TA (35%) TA, GTCS, rarely MS and absence status epilepticus GTCS TA, MS, GTCS, tonicabsences GSW at 2.56Hz, regular or irregular fragmentations

Generalized Cryptogenic/Symptomatic Epilepsies Angelman syndrome Atypical absences, tonic, Ring 20 syndrome clonic, myoclonic, Dravet syndrome negative myoclonus, Lennox-Gastaut complex partial, GTCS syndrome Tonic, atonic, atypical Myoclonic astatic absences, MS epilepsy Myoclonicastatic Syndrome of Myoclonic absences myoclonic absences

Normal or diffusely slow background Slow <2.5 Hz GSW, multifocal spikes Characteristic EEG findings: monomorphic 2 Hz SW in Angelman syndrome

Focal Symptomatic/Cryptogenic Epilepsies Various syndromes of Simple or complex partial Focal slowing/spikes; may show lobar (mainly frontal with semiology reflecting secondary bilateral synchrony and temporal lobe) the lobe of origin or epilepsies propagation Focal Idiopathic Epilepsies Benign rolandic epilepsy Rolandic seizures, rare Normal background, multifocal, brief TA rolandic or occipital spikes, Panayiotopoulos Autonomic seizures eye GSW in both rolandic and syndrome deviation and motor Panayiotopoulos Syndrome symptoms. Rolandic and autonomic seizures may coexist in mixed phenotypes; rare TA Epileptic Generalized and partial Continuous SW activity during encephalopathy with seizures mainly during nnn-RFM sleep (electrical CSWS sleep and typical or status epilepticus) atypical absences during awake De Novo Absence-like Status None Unremarkable background without interictal spikes; CSW during status

134

THE EPILEPSIES 3

and indeed differentiation from spasms may be difficult from description, so ictal video EEG recordings with dedicated EMG electrodes have to be employed. MS are associated with brief bursts of GSW, but interictal EEG is normal. MS may be provoked by photic stimulation in 20% and by unexpected acoustic or tactile stimuli in 10% of children; the latter may be easier to control. Simple febrile seizures occur in 10%, but there are no other seizure types. BMEI responds to VPA, and remission usually occurs within 1 year from onset, but 10 to 20% may develop infrequent GTCS in their early teens. Differential diagnosis includes nonepileptic conditions, such as hypnagogic jerks and benign nonepileptic myoclonus and epileptic syndromes, including severe myoclonic epilepsy of infancy Dravet syndrome), familial infantile myoclonic epilepsy (mapped to 16p13), infantile spasms (West syndrome), epilepsy with myoclonic-astatic seizures (MAE), and the myoclonic form of LennoxGastaut syndrome (LGS). EPILEPSY WITH MYOCLONIC-ASTATIC SEIZURES (MAE)2,27,28 MAE is a genetically determined (some patients fall into the spectrum of generalized epilepsy with febrile seizures plus [GEFS+] nonlesional generalized epilepsy that affects previously normal children between 2 and 5 years of age, but has variable course ranging from good responsiveness to treatment, with normal cognitive functions and even remission, to nonprogressive epileptic encephalopathy with seizure intractability and cognitive impairment. Therefore it clinically overlaps with both IGE (see also the 2001 ILAE proposal, where it features within the IGE6) and cryptogenic/symptomatic generalized epilepsies (see relevant following section). Prevalence is around 1to 2% of all childhood epilepsies, onset ranges between 7 months and 6 years, peaking between 2 and 4 years, and two-thirds of patients are boys. Myoclonic-astatic seizures are the defining seizure symptom and consist of a myoclonic jerk immediately followed by loss of muscle tone. Either component can cause falls, head nodding, or bending of the knees. More than half of patients have brief absences, often with myoclonic jerks, facial myoclonias, and atonia. In twothirds of patients, febrile and nonfebrile GTCS precede myoclonic astatic seizures by several months. Episodes of absence status may occur, usually introduced by inappropriate treatment such as CBZ; they are frequent in the cryptogenic form. Myoclonic-atonic seizures are associated with discharges of irregular GSW or polyspike waves at >2.5 Hz, with the atonic part of the seizure corresponding to the slow wave of the GSW complex and being associated with diffuse electromyography (EMG) paucity. VPA at high doses is the first drug of choice, combined if needed with LEV, ESX, CLZ, or sulthiame. MAE shares many clinical features with cryptogenic Lennox-Gastaut syndrome, particularly its myoclonic form.29 CHILDHOOD ABSENCE EPILEPSY (CAE)2,3032 This is the prototype syndrome of idiopathic absence epilepsy in childhood with prevalence of about 10% of childhood epilepsies and incidence about 7/100,000 of children with nonfebrile seizures. In contrast to BMEI and MAE, CAE predominately affects girls (two-thirds) between 4 and 9 to 10 years, with a peak at 5 to 6 years. There is probable linkage with chromosomes 1p, 8q24, 5q31.1, and 19p13.2. TA are associated with severe impairment of consciousness (unresponsiveness and interruption of ongoing activities) and are usually multiple per day, hence the

10  The Spectrum of Epilepsies Associated with Generalized Spike-and-Wave Patterns

135

term pyknolepsy. They have abrupt onset and termination and last from 4 to 20 sec (mainly around 10 sec). Automatisms occur in two-thirds of the seizures but are not stereotyped. Mild myoclonic elements of eyes, eyebrows, and eyelids may feature at the onset of an absence. More severe and sustained myoclonic jerks of facial ro limb muscles indicate other IGEs. TA are nearly always provoked by hyperventilation, even in the clinic. Other types of seizures are incompatible with CAE except for febrile seizures and solitary or infrequent GTCS (usually in adolescence after absences have remitted). The interictal EEG may show characteristic rhythmic posterior delta activity and brief, usually regular GSW. Monotherapy with VPA ESX, or LTG is successful in most cases, and remission occurs before the age of 12 years with less than 10% of patients developing infrequent GTCS in adolescence or adult life.33 As in all IGEs, VGB, TGB, CBZ, and GBP can produce worsening of absences. CAE is not the only IGE absence syndrome in the first decade of life, and meticulous differentiation from other absence syndromes of worse prognosis is clinically important. EPILEPSY WITH MYOCLONIC ABSENCES (E-MA)34,35 This rare type of generalized epilepsy was described by Tassinari36,37 and is characterized by myoclonic absences that are associated with 3-Hz regular GSW activity like the typical absences in childhood or juvenile absence epilepsies, but also with bilateral myoclonic jerks of the arms and legs that are time locked to the spike component of the ictal discharge and have a superimposed tonic contraction resulting in a characteristic stepwise upward abduction of the arms. Other seizure types may include GTCS, clonic seizures, and rarely simple absences without myoclonus. It affects less than 1% of children with epileptic disorders, mostly boys (70%), between 5 months and 13 years (median 7 years). Etiology is heterogeneous, and only one-third of patients are of normal neurological and mental state (idiopathic form). These retain normal EEG background. The others are symptomatic, with learning difficulties manifested before or after seizure onset. Early control of absences may secure normal development. The main task of differential diagnosis here is to distinguish between idiopathic and symptomatic cases (see later discussion). JUVENILE ABSENCE EPILEPSY (JAE)2,3840 The prevalence of JAE is around 2 to 3% of all epilepsies and around 8 to 10% of IGE in adults after the age of 20 years. About 70% of the patients start having absences between ages of 9 and 13 years, but age at onset may range from 5 to 20 years. Males and females are equally affected. TA are similar to those of CAE, but milder, and duration varies from 4 to 30 sec or longer. Typically infrequent GTCS and random, rather than early morning, myoclonic jerks appear in 80% and 20% of patients, respectively, 1 to 10 years after the onset of TA. Focal epileptiform abnormalities and abortive asymmetrical bursts of spike/multiple spikes are common in the interictal EEG, whereas ictal GSW discharges are typically regular at 3 Hz. Differential diagnosis involves CAE (because of the age overlap), JME (because of possible MS), and limbic CPS because of ictal automatisms during prolonged absences (30 sec or more), which make distinction very difficult on

136

THE EPILEPSIES 3

clinical grounds only. VPA is the drug of choice, but LTG and LEV are possible alternatives. Monotherapy with ESX is not recommended because the natural history of the syndrome includes GTCS, against which ESX provides no protection. JAE are a lifelong disorder, although seizures are controlled in 70 to 80% of patients, particularly if appropriate treatment is initiated early. JUVENILE MYOCLONIC EPILEPSY (JME)2,4144 JME is the quintessence of myoclonic IGE and affects 8 to 10% of patients with epilepsies without any sex difference. Inheritance is probably complex and polygenic with susceptibility loci in chromosomes 6p11-12 (EJM1) and 15q14 (EJM2). MS are the defining seizure, as they occur in all patients, typically on awakening. They start between 8 to 26 years (with peak in early teens), may occur in volleys preceding a GTCS, and are typically activated by sleep deprivation, awakening, fatigue and mental stress and excitement, and excessive alcohol intake. Triggers include photosensitivity (present in one-third of patients) and less frequently movements or intension of movement (praxis-induced), whereas reading-induced jerks are rare but well documented. TA occur in around one-third of patients between 5 and 16 years of age (peak at 10 years) and are usually mild and inconspicuous, therefore detectable mainly by video EEG. The combination of more prolonged and frequent TA and random MS favor. JAE rather than JME. GTCS occur in nearly all patients, usually months after onset of myoclonic jerks. In untreated patients, interictal EEG is usually abnormal, with generalized discharges of irregular 3- to 6-Hz spike/polyspike waves. Focal abnormalities occur in 30 to 40% of patients. A normal EEG in a patient suspected of having JME should prompt an EEG during sleep and on awakening. MS are associated with generalized, frequently asymmetrical, multiple spikes, and TA are brief with 3- to 6-Hz GSW and multiple spikes and irregular intradischarge frequency with fragmentations. Susceptibility to all seizures is probably permanent, although patients show improvement after the age of 40. Despite appropriate treatment, seizures become intractable in about 20% of patients, with those who have all three types of seizures being more likely to show pharmacoresistance. Misdiagnosis is common resulting in avoidable morbidity. Factors responsible for misdiagnosis include lack of familiarity with JME, failure to elicit a history of jerks, misinterpretation of absences or jerks as focal seizures, and high prevalence of focal EEG abnormalities. An uncertain number of patients with MS only may escape diagnosis, as they do not seek medical advice. Optimal management includes appropriate education and advice, focusing mainly on seizure precipitants. VPA is the most effective AED but with significant concerns regarding women. Control and open studies indicate that LEV is probably one of the most promising new AEDs. Clonazepam is effective for persistent myoclonic jerks, but may deprive patients from the early morning volley that alerts for an impending GTCS.45 IGE WITH GTCS ONLY (IGE/GTCS)2,46,47 By definition, this syndrome manifests with GTCS only without TA and MS throughout the history on clinical and EEG basis, but many accept rare MS or brief TA. Therefore, the true prevalence is uncertain. EEG-wise, the interictal occurrence of GSW discharges should also be a diagnostic prerequisite. The

10  The Spectrum of Epilepsies Associated with Generalized Spike-and-Wave Patterns

137

age at onset ranges from 6 to 47 years with a peak around 16 to 17 years. GTCS can occur on awakening (usual pattern), during relaxation or leisure, during sleep, and randomly. Seizure precipitants include sleep deprivation, fatigue, and excessive alcohol consumption, whereas about 15% show EEG photosensitivity. IGE with GTCS on awakening is probably a lifelong disorder with high rate of relapse on withdrawal of treatment. GTCS become rarer and more random with time, either as a result of the evolution of the disease or drug-induced modifications, and precipitants less potent. Differential diagnosis includes JME (if one accepts rare MS), with which shares the typical circadian distribution, and the syndrome of phantom absences (if one accepts TA). Differentiation from cryptogenic generalized and focal epilepsies with secondarily GTCS or secondary bilateral synchrony employs clinical and EEG criteria, and patients with poorly understood nocturnal seizures and without clear EEG indicators should not be included here. CBZ, OXC, PB, and PHT can be helpful in some patients provided that TA and MS are not part of the clinical picture. If these seizures also occur, only VPA, LEV, TPM, and LTG are recommended. IGE SYNDROMES THAT HAVE NOT BEEN RECOGNIZED BY THE ILAE48 The following syndromes are not yet officially recognized by the ILAE but have been accepted by many authorities and are discussed later. Eyelid myoclonia with absences (Jeavons syndrome) Perioral myoclonia with absences IGE with phantom absences EYELID MYOCLONIA WITH ABSENCES (ELMA OR JEAVONS SYNDROME)2,4952 This is the most distinctive reflex IGE syndrome, characterized by (a) eyelid myoclonia with and without absences; (b) eye closure-induced seizures, EEG paroxysms, or both; and (c) photosensitivity. The age at onset ranges between 2 and 14 years (with peak at 6 to 8 years), and its prevalence is around 3% of adults with epileptic disorders and 13% of those with absence IGE syndromes. There is a twofold female preponderance. Eyelid myoclonia, the clinical hallmark of this syndrome, includes brief but marked, episodic jerking of the eyelids, the eyebrows, and often of the eyeballs and head, which may be time locked to GSW and polyspike discharges, and may or may not be associated with clinically apparent impairment of awareness (eyelid myoclonia with or without absences). These seizures are brief (3 to 6 sec) and occur mainly after eye closure, many times per day, and especially on awakening. Spontaneous or photically induced GTCS are probably inevitable in the long term and are easily provoked by sleep deprivation, alcohol, and inappropriate AED modifications. Myoclonic jerks of the limbs may occur. Absence Status Epilepticus (ASE) occurs in one-fifth of patients with prominent but discontinuous eyelid myoclonia and usually mild impairment of consciousness. Video-EEG is important for the diagnosis. Photoparoxysmal responses occur in all untreated young patients, but photosensitivity declines in adulthood. ELMA is particularly resistant to AED treatment (VPA, LEV, and CLZ

138

THE EPILEPSIES 3

are the main choices), and prognosis is guarded. Despite the sometimes extreme photosensitivity, self-induction is rare. IGE WITH PHANTOM ABSENCES (PA)2,5355 Phantom absences are by definition so mild that they are inconspicuous to the patient and imperceptible to the observer. Therefore, their onset cannot be ascertained, and diagnosis is based on video EEGs, recorded after the firstusually of late onsetGTCS. The latter are infrequent and do not show consistent circadian distribution or specific precipitants. Half of patients suffer from repeated episodes of absence status epilepticus that can occur on their own or lead to a GTCS. The identification of PA often requires breath counting during hyperventilation on video EEG. PA manifest with interruption, hesitations, or errors of counting that coincide with a briefas a ruleGSW discharge. PA seem otherwise subclinical, although occasionally there may be associated eyelid blinking. It can be appreciated that without appropriate ictal video EEG studies, most of these patients may be categorized as either IGE with (random) GTCS only or IGE with GTCS and ASE. As workup of such patients may vary, true prevalence is uncertain, but it is estimated to be around 3% of adults with seizures and 10% of all adults with IGE. This may be higher, as there may be patients in the general population with phantom absences but without GTCS. Treatment as in other IGEs, and ASE, of which most patients are aware, could be terminated with out-of-hospital self-administered benzodiazepines. PERIORAL MYOCLONIA WITH ABSENCES (PMA)2,54,5658 This is a rather rare syndrome with absences that are characteristically associated with rhythmic perioral twitching and that start in childhood, but continue through adult life (prevalence probably around 9% in adults with TA). Age at onset is between 2 and 13 years (median 10 years), and women predominate (80%). These absences usually last up to 10 sec (average 4 sec), with varying frequency and severity. All patients suffer from usually infrequent GTCS that may appear before, soon after, or exceptionally many years after the onset of absences. Absence status epilepticus is very common (57%) and frequently ends with GTCS. Interictal EEG shows GSW discharges, frequently with multiple spikes and focal, nonlocalizing abnormalities. Usually irregular and fragmented 3- to 4-Hz GSW discharges with polyspikes accompany TA. As defined earlier, PMA is often resistant to appropriate antiepileptic medication. However, orofacial myoclonus may rarely occur in absences of other IGEs and does not influence the prognosis of the individual syndrome. THE SYNDROME OF DE NOVO ABSENCE-LIKE STATUS OF LATE ONSET De novo absence-like status is a form of nonconvulsive status epilepticus that occurs for the first time in nonepileptic patients of middle or old age as a result of an exogenous trigger and bears electroclinical resemblance to idiopathic absence status epilepticus.5961 Distinction from idiopathic ASE is clinically important, as biological substrates and by implication management and prognosis are different. Patients with idiopathic ASE need appropriate prophylactic treatment with AED; for those with

10  The Spectrum of Epilepsies Associated with Generalized Spike-and-Wave Patterns

139

de novo absence-like status, elimination of the offending factor may suffice, and the overall prognosis is excellent with little risk for recurrence.60 Triggers include acute withdrawal of chronic psychotropic medication (mainly benzodiazepines60,62), but also tricyclic antidepressants,63 major neuroleptics and lithium, acute or chronic alcoholism and metabolic imbalance, and toxic or other pharmacological agents.60 However, relying only on apparent triggers may be misleading, and one has to bear in mind that in some patients with IGE, episodes of true absence status may be precipitated after discontinuation of a benzodiazepine (taken as antiepileptic medication such as clobazam) or moderate alcohol consumption. The underlying pathophysiology of de novo absence-like status is uncertain, and therefore its nosological taxonomy has been controversial. It appears that the triggering or facilitating potential of these factors becomes particularly effective when acting on brains compromised already by vascular or degenerative processes, as is more frequently the case in elderly people. A contribution of a genetically determined liability cannot be excluded.

Other Epileptic Syndromes and Conditions Associated with GSW Activity


CRYPTOGENIC/SYMPTOMATIC GENERALIZED EPILEPTIC SYNDROMES AND EPILEPTIC ENCEPHALOPATHIES Common characteristics that distinguish this group of epilepsies from IGE include varying etiology and multiple seizure types, frequent developmental delay, and a general reluctance to respond to treatment. From the EEG viewpoint, background is usually abnormal, and GSW discharges are slower, 2.5 Hz or less. Absences here are associated with slow GSW and milder, with not always sharp onset and termination, and are called atypical absence seizures. Angelman Syndrome (Partial Monosomy 15q) This is a severe neurodevelopmental disorder caused by the failure of the expression of the maternal copy of the imprinted UBE3A gene on chromosome 15q11-q13. Characteristics include developmental delay with happy disposition and inappropriate laughter, mainly motor speech impairment, ataxia, hyperactivity, dysmorphic features, and seizures in up to 90% of patients with onset from 3 months to 20 years, usually in early childhood. Seizure activity may improve with age. Seizure types include myoclonic, atypical absences mostly associated with rhythmic triphasic delta waves of high amplitude,64 convulsive generalized and unilateral clonic, and complex partial with eye deviation and vomiting suggesting occipital origin.65 The interictal EEG shows rather characteristic large monomorphic slow spike-wave activity at 2 Hz over the posterior areas that becomes diffuse symmetrical bilateral and synchronous. Ring Chromosome 20 Syndrome (r20S) This rare chromosomal syndrome is characterized by variable psychomotor delay, behavioral disturbance, and frequent episodes of nonconvulsive status epilepticus

140

THE EPILEPSIES 3

(NCSE), but there are no dysmorphisms, and this may delay the diagnosis. Some patients may have normal intelligence. Focal seizures with affective symptomatology (usually terror), frontal discharges, and atypical absences with 2- to 3-Hz bilateral rhythmic slow activity and some autonomic features66 may be part of the clinical picture, which becomes typical after the age of 8 years.67 As a rule, seizures are resistant to treatment. Interictal EEG may be normal or show bilateral frontocentral theta activity. The NCSE in r20S is mainly an absence state with an associated myoclonic component. Differentiation from IGE may be difficult in patients with regular spike-wave activity, mild retardation or normal intellect, and normal or almost normal interictal EEG. Middle-aged patients may be misdiagnosed as de novo absence-like status if previous episodes are not identified, and frontal lobe status should also be considered in view of the affective and autonomic symptomatology. Severe Myoclonic Epilepsy in InfancyDravet Syndrome First described by Dravet in 1978,68 this syndrome is a genetically determined channelopathy69 and a syndromic phenotype that can occur within the frame of GEFS+.70 The first seizure is typically triggered by fever during infancy and is prolonged unilateral or generalized clonic or tonic-clonic; other nonfebrile convulsive seizures follow soon. Generalized myoclonic seizures associated with brief 3Hz GSW with polyspikes and segmental myoclonus appear in the 2nd or 3rd year, and atypical absences occur in up to 80 to 90% of the children, either concurrently with the myoclonic seizures or later. Focal and tonic seizures also occur. Episodes of absence status occur in at least 30 to 40% of cases and may persist for hours and even days (obtundation status). Photosensitivity occurs in up to 42% of children and may persist.71 Cryptogenic/Symptomatic Form of Epilepsy with Myoclonic-Astatic Seizures As opposed to the idiopathic form of MAE (see in IGE section earlier), the cryptogenic variant is characterized by frequent (up to 90% of children) episodes of generalized myoclonic/absence status associated with chaotic EEG with asynchronous SW discharges that resemble hypsarrhythmia,29 tonic seizures during sleep, progressive slowing of the EEG background, and unfavorable outcome. These children continue with seizures, severe cognitive impairment, and behavioral abnormalities, whereas ataxia and motor and linguistic disturbances may emerge. Treatment is similar to the idiopathic form with intravenous benzodiazepines being used for nonconvulsive status epilepticus. In general, myoclonic astatic seizures occur in many childhood epilepsies, particularly epileptic encephalopathies, and differential diagnosis includes benign or progressive myoclonic epilepsies, Dravet syndrome, Lennox-Gastaut syndrome, atypical benign focal epilepsy of childhood, atypical evolutions of rolandic epilepsy, Panayiotopoulos syndrome, and nonepileptic myoclonus of various neurological disorders. Lennox-Gastaut Syndrome (LGS) LGS is a severe cryptogenic or symptomatic epileptic encephalopathy with tonic, atonic, atypical absences and other seizures, slow GSW (up to 2 to 2.5 Hz) and fast

10  The Spectrum of Epilepsies Associated with Generalized Spike-and-Wave Patterns

141

rhythms during sleep, and neuropsychological decline. Onset occurs between 3 and 10 years. The slow GSW and the lack of a prominent myoclonus distinguish cryptogenic LGS from MAE; however, some forms with predominant myoclonic-atonic seizures (known as myoclonic variant of LGS) resemble MAE and may have better outcome.29 Status epilepticus occurs in up to 90% of patients and is usually nonconvulsive. Cryptogenic/Symptomatic Form of Epilepsy with Myoclonic Absence Seizures Two-thirds of patients with these characteristic absences develop mental retardation (evident before or after seizure onset) and seizure intractability (see also IGE section earlier). Such an unfavorable outcome is usually predicted by GTCS or atonic fits and, of course, by symptomatic etiology. EEG background at onset may be abnormal in symptomatic cases or deteriorate later. Brief generalized, focal, or multifocal spikes and slow waves appear in 50% of patients. EPILEPTIC ENCEPHALOPATHY WITH CONTINUOUS SPIKE-AND-WAVE DURING SLEEP (CSWS) INCLUDING LANDAU-KLEFFNER SYNDROME (LKS) This is an age-related and partly reversible childhood disorder characterized by generalized and partial seizures mainly occurring during sleep and typical or atypical absences during awake state, global or selective regression of high cognitive functions, motor impairment. The EEG is characteristic with diffuse or regional spike-wave activity which is nearly continuous and occurs during non-REM sleep.2 There are no tonic seizures, and the kind of neuropsychological impairment relates to the topography of the continuous discharges. CSWS becomes apparent a couple of years after the onset of epilepsy and is associated with seizure worsening including also diurnal episodes of absence status. Seizures and CSWS gradually resolve, but only 25% of the children attain acceptable social and professional levels. In LKS, focal spike-wave activity heavily involves the speech cortex and cannot be considered generalized. IDIOPATHIC FOCAL EPILEPSIES OF CHILDHOOD Due perhaps to the very nature of idiopathic focal or system epileptogenesis, both benign rolandic epilepsy with centrotemporal spikes (BECTS) and Panayiotopoulos syndrome (PS) manifest with bilateral regional or diffuse epileptogenicity, distinctly different to the focal ictogenesis of the symptomatic focal epilepsies.72 GSW discharge may occur, sometimes associated with mild impairment of consciousness (absences). In addition, both BECTS and PS may occasionally evolve into atypical electroclinical forms that include continuous spike-wave during sleep, absences with inhibitory phenomena, and episodes of atypical absence status, with favorable outcome in most children.7378 SYMPTOMATIC OR CRYPTOGENIC FOCAL EPILEPSIES GSW activity associated with some impairment of consciousness may occur in frontal lobe epilepsies, but also in some patients with malformations of cortical

142

THE EPILEPSIES 3

development including temporal lobe dysembryoplastic neuroepithelial tumors.79 Diagnosis here is based on the existence of other seizure types, and the possibility of coexistence of IGE and focal lobar epilepsy should be also considered.80 REFERENCES
1. First Seizure Trial Group (FIR.S.T. Group). Randomized clinical trial on the efficacy of antiepileptic drugs in reducing the risk of relapse after a first unprovoked tonic-clonic seizure. Neurology. 1993;43(3 Pt 1):478-483. 2. Panayiotopoulos CP. A Clinical Guide to Epileptic Syndromes and Their Treatment. 2nd ed. London: Springer; 2007. 3. Panayiotopoulos CP. Evidence-based epileptology, randomized controlled trials, and SANAD: a critical clinical view. Epilepsia. 2007;48(7):1268-1274. 4. Commission of Classification and Terminology of the International League Against Epilepsy. Proposal for revised clinical and electroencephalographic classification of epileptic seizures. Epilepsia. 1981;22:489-501. 5. Commission on Classification and Terminology of the International League Against Epilepsy. Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia. 1989;30:389-399. 6. Engel J Jr. A proposed diagnostic scheme for people with epileptic seizures and with epilepsy: Report of the ILAE Task Force on Classification and Terminology. Epilepsia. 2001;42:796-803. 7. Engel J Jr. Report of the ILAE classification core group. Epilepsia. 2006;47(9):1558-1568. 8. Blume WT, Pillay N. Electrographic and clinical correlates of secondary bilateral synchrony. Epilepsia. 1985;26(6):636-641. 9. Panayiotopoulos CP, Chroni E, Daskalopoulos C, Baker A, Rowlinson S, Walsh P. Typical absence seizures in adults: clinical, EEG, video-EEG findings and diagnostic/syndromic considerations. J Neurol Neurosurg Psychiatr. 1992;55(11):1002-1008. 10. Usui N, Kotagal P, Matsumoto R, Kellinghaus C, Luders HO. Focal semiologic and electroencephalographic features in patients with juvenile myoclonic epilepsy. Epilepsia. 2005;46(10):1668-1676. 11. Panayiotopoulos CP, Tahan R, Obeid T. Juvenile myoclonic epilepsy: factors of error involved in the diagnosis and treatment. Epilepsia. 1991;32(5):672-676. 12. Topcuoglu MA, Saygi S, Ciger A. Rotatory seizures in juvenile myoclonic epilepsy. Clin Neurol Neurosurg. 1997;99(4):248-251. 13. So GM, Thiele EA, Sanger T, Schmid R, Riviello JJ Jr.. Electroencephalogram and clinical focalities in juvenile myoclonic epilepsy. J Child Neurol. 1998;13(11):541-545. 14. Ferrie CD. Idiopathic generalized epilepsies imitating focal epilepsies. Epilepsia. 2005;46(Suppl 9): 91-95. 15. Aliberti V, Grunewald RA, Panayiotopoulos CP, Chroni E. Focal electroencephalographic abnormalities in juvenile myoclonic epilepsy. Epilepsia. 1994;35(2):297-301. 16. Lombroso CT. Consistent EEG focalities detected in subjects with primary generalized epilepsies monitored for two decades. Epilepsia. 1997;38(7):797-812. 17. Koutroumanidis M, Smith S. Use and abuse of EEG in the diagnosis of idiopathic generalized epilepsies. Epilepsia. 2005;46(Suppl 9):96-107. 18. Hitiris N, Brodie MJ. Evidence-based treatment of idiopathic generalized epilepsies with older antiepileptic drugs. Epilepsia. 2005;46(Suppl 9):149-153. 19. Bergey GK. Evidence-based treatment of idiopathic generalized epilepsies with new antiepileptic drugs. Epilepsia. 2005;46(Suppl 9):161-168. 20. Shorvon S, ed. Handbook of Epilepsy Treatment. 2nd ed. Oxford: Blackwell Science; 2005. 21. Chaves J, Sander JW. Seizure aggravation in idiopathic generalized epilepsies. Epilepsia. 2005;46(Suppl 9):133-139. 22. Thomas P, Valton L, Genton P. Absence and myoclonic status epilepticus precipitated by antiepileptic drugs in idiopathic generalized epilepsy. Brain. 2006;129(Pt 5):1281-1292. 23. Kwan P. Pro-myoclonic anti-epileptic drugs to avoid in IGE with myoclonic jerks. In: Panayiotopoulos CP, ed. Idiopathic Generalised Epilepsies with Myoclonic Jerks. Vol 2.Oxford: Medicinae; 2007:189-196. 24. Darra F, Fiorini E, Zoccante L, et al. Benign myoclonic epilepsy in infancy (BMEI): a longitudinal electroclinical study of 22 cases. Epilepsia. 2006;47(Suppl 5):31-35. 25. Fejerman N. Benign myoclonic epilepsy in infancy. In: Panayiotopoulos CP, ed. Idiopathic Generalised Epilepsies with Myoclonic Jerks. Vol 2. Oxford: Medicinae; 2007:55-59. 26. Capovilla G, Beccaria F, Gambardella A, Montagnini A, Avantaggiato P, Seri S. Photosensitive benign myoclonic epilepsy in infancy. Epilepsia. 2007;48(1):96-100.

10  The Spectrum of Epilepsies Associated with Generalized Spike-and-Wave Patterns 27. Neubauer BA, Hahn A, Doose H, Tuxhorn I. Myoclonic-astatic epilepsy of early childhooddefinition, course, nosography, and genetics. Adv Neurol. 2005;95:147-155. 28. Oguni H, Hayashi K, Imai K, et al. Idiopathic myoclonic-astatic epilepsy of early childhoodnosology based on electrophysiologic and long-term follow-up study of patients. Adv Neurol. 2005;95:157-174. 29. Kaminska A, Ickowicz A, Plouin P, Bru MF, Dellatolas G, Dulac O. Delineation of cryptogenic Lennox-Gastaut syndrome and myoclonic astatic epilepsy using multiple correspondence analysis. Epilepsy Res. 1999;36(1):15-29. 30. Loiseau P, Panayiotopoulos CP. Childhood absence epilepsy [ILAE Web site]. 2004 Available at: http://www.ilae-epilepsy org/Visitors/Centre/ctf/childhood_absence.html. Accessed September 18, 2007. Epileptic syndromes [Epilepsy Web site]. Available at: www.epilepsy.org/ctf. 31. Grosso S, Galimberti D, Vezzosi PS, et al. Childhood absence epilepsy: evolution and prognostic factors. Epilepsia. 2005;46(11):1796-1801. 32. Hirsch E, Panayiotopoulos CP. Childhood absence epilepsy and related syndromes. In: Roger J, Bureau M, Dravet C, Genton P, Tassinari CA, Wolf P, eds. Epileptic Syndromes in Infancy, Childhood and Adolescence. 4th ed. Montrouge, France: John Libbey Eurotext; 2005:315-335. 33. Posner E. Pharmacological treatment of childhood absence epilepsy. Expert Rev Neurother. 2006;6(6):855-862. 34. Genton P, Bureau M. Epilepsy with myoclonic absences. CNS Drugs. 2006;20(11):911-916. 35. Tassinari A, Rubolli R, Michellluchi R. Epilepsy with myoclonic absences [ILAE Web site]. 2006. Available at: http://www.ilae-epilepsy org/ctf/over_frame.html. Accessed September 18, 2008. 36. Tassinari CA, Lyagoubi S, Santos V, et al. [Study on spike and wave discharges in man. II. Clinical and electroencephalographic aspects of myoclonic absences]. Rev Neurol (Paris). 1969;121(3):379-383. 37. Bureau M, Tassinari CA. Myoclonic absences: the seizure and the syndrome. Adv Neurol. 2005;95:175-183. 38. Obeid T. Clinical and genetic aspects of juvenile absence epilepsy. J Neurol. 1994;241(8):487-491. 39. Trinka E, Baumgartner S, Unterberger I, et al. Long-term prognosis for childhood and juvenile absence epilepsy. J Neurol. 2004;251(10):1235-1241. 40. Tovia E, Goldberg-Stern H, Shahar E, Kramer U. Outcome of children with juvenile absence epilepsy. J Child Neurol. 2006;21(9):766-768. 41. Panayiotopoulos CP, Obeid T, Tahan AR. Juvenile myoclonic epilepsy: a 5-year prospective study. Epilepsia. 1994;35(2):285-296. 42. Calleja S, Salas-Puig J, Ribacoba R, Lahoz CH. Evolution of juvenile myoclonic epilepsy treated from the outset with sodium valproate. Seizure. 2001;10(6):424-427. 43. Genton P, Gelisse P. Juvenile myoclonic epilepsy. Arch Neurol. 2001;58(9):1487-1490. 44. Zifkin B, Andermann E, Andermann F. Mechanisms, genetics, and pathogenesis of juvenile myoclonic epilepsy. Curr Opin Neurol. 2005;18(2):147-153. 45. Obeid T, Panayiotopoulos CP. Clonazepam in juvenile myoclonic epilepsy. Epilepsia. 1989;30(5):603-606. 46. Wolf P. Epilepsy with grand mal on awakening. In: Roger J, Bureau M, Dravet C, Dreifuss FE, Perret A, Wolf P, ed. Epileptic Syndromes in Infancy, Childhood and Adolescence. London: John Libbey & Company; 1992:329-341. 47. Janz D. Epilepsy with grand mal on awakening and sleep-waking cycle. Clin Neurophysiol. 2000;111(Suppl 2):S103-S110. 48. Panayiotopoulos CP. Syndromes of idiopathic generalized epilepsies not recognized by the International League Against Epilepsy. Epilepsia. 2005;46(Suppl 9):57-66. 49. Duncan JS, Panayiotopoulos CP, eds. Eyelid Myoclonia with Absences. London: John Libbey & Company Ltd; 1996. 50. Panayiotopoulos CP. Eyelid myoclonia with and without absences [ILAE Web site]. 2006. Available at: http://www.ilae-epilepsy org/ctf/over_frame.html. Accessed September 18, 2007. 51. Covanis A. Eyelid myoclonia and absences: Jeavons syndrome (with video-EEG sequences). In: Panayiotopoulos CP, ed. Idiopathic Generalised Epilepsies with Myoclonic Jerks. Vol 2. Oxford: Medicinae; 2007:88-92. 52. Joshi CN, Patrick J. Eyelid myoclonia with absences: routine EEG is sufficient to make a diagnosis. Seizure. 2007;16(3):254-260. 53. Panayiotopoulos CP, Koutroumanidis M, Giannakodimos S, Agathonikou A. Idiopathic generalised epilepsy in adults manifested by phantom absences, generalised tonic-clonic seizures, and frequent absence status. J Neurol Neurosurg Psychiatr. 1997;63(5):622-627. 54. Agathonikou A, Panayiotopoulos CP, Giannakodimos S, Koutroumanidis M. Typical absence status in adults: diagnostic and syndromic considerations. Epilepsia. 1998;39(12):1265-1276.

143

144

THE EPILEPSIES 3 55. Panayiotopoulos CP, Ferrie CD, Koutroumanidis M, Rowlinson S, Sanders S. Idiopathic generalised epilepsy with phantom absences and absence status in a child. Epileptic Disord. 2001;3(2):63-66. 56. Panayiotopoulos CP, Ferrie CD, Giannakodimos S, Robinson RO. Perioral myoclonia with absences: a new syndrome. In: Wolf P, ed. Epileptic Seizures and Syndromes. London: John Libbey & Company Ltd; 1994:143-153. 57. Bilgic B, Baykan B, Gurses C, Gokyigit A. Perioral myoclonia with absence seizures: a rare epileptic syndrome. Epileptic Disord. 2001;3(1):23-27. 58. Baykan B, Noachtar S. Perioral myoclonia with absences: an overlooked and misdiagnosed generalized seizure type. Epilepsy Behav. 2005;6(3):460-462. 59. Thomas P, Beaumanoir A, Genton P, Dolisi C, Chatel M. De novo absence status of late onset: report of 11 cases [see comments]. Neurology. 1992;42(1):104-110. 60. Thomas P, Andermann F. Late-onset absence status epilepticus is most often situation-related. In: Malafosse A, Genton P, Hirsch E, Marescaux C, Broglin D, Bernasconi R, ed. Idiopathic Generalized Epilepsies. London: John Libbey & Company Ltd; 1994:95-109. 61. Koutroumanidis M. NCSE in the idiopathic and other generalized epilepsies (including absence status). In: Kaplan P, Drislane F, ed. Nonconvulsive Status Epilepticus. New York: Demos Medical Publishing. In press. 62. Dunne JW, Summers QA, Stewart-Wynne EG. Non-convulsive status epilepticus: a prospective study in an adult general hospital. Q J Med. 1987;62(238):117-126. 63. Bourrat C, Garde P, Boucher M, Fournet A. [Prolonged absence states in aged patients without epileptic antecedents]. [French]. Revue Neurologique. 1986;142(8-9):696-702. 64. Laan LA, Renier WO, Arts WF, et al. Evolution of epilepsy and EEG findings in Angelman syndrome. Epilepsia. 1997;38(2):195-199. 65. Viani F, Romeo A, Viri M, et al. Seizure and EEG patterns in Angelmans syndrome. J Child Neurol. 1995;10(6):467-471. 66. Canevini MP, Sgro V, Zuffardi O, et al. Chromosome 20 ring: a chromosomal disorder associated with a particular electroclinical pattern. Epilepsia. 1998;39(9):942-951. 67. Ville D, Kaminska A, Bahi-Buisson N, et al. Early pattern of epilepsy in the ring chromosome 20 syndrome. Epilepsia. 2006;47(3):543-549. 68. Dravet C. Les epilepsies graves de lenfant. Vie Med. 1978;8:543-548. 69. Claes L, Del Favero J, Ceulemans B, Lagae L, Van Broeckhoven C, De Jonghe P. De novo mutations in the sodium-channel gene SCN1A cause severe myoclonic epilepsy of infancy. Am J Hum Genet. 2001;68(6):1327-1332. 70. Scheffer IE, Berkovic SF. Generalized epilepsy with febrile seizures plus. A genetic disorder with heterogeneous clinical phenotypes. Brain. 1997;120(Pt 3):479-490. 71. Dravet C, Bureau M, Guerrini R, Giraud N, Roger J. Severe myoclonic epilepsy in infants. In: Roger J, Bureau M, Dravet C, Dreifuss FE, Perret A, Wolf P, eds. Epileptic Syndromes in Infancy, Childhood and Adolescence. London: John Libbey & Company; 1992:75-78. 72. Koutroumanidis M. Panayiotopoulos syndrome: an important electroclinical example of benign childhood system epilepsy. Epilepsia. 2007;48(6):1044-1053. 73. Fejerman N. Atypical evolutions of benign partial epilepsies in children. Int Pediatr. 1996;11(6):351-356. 74. Fejerman N, Caraballo R, Tenembaum SN. Atypical evolutions of benign localization-related epilepsies in children: are they predictable? Epilepsia. 2000;41(4):380-390. 75. Caraballo RH, Astorino F, Cersosimo R, Soprano AM, Fejerman N. Atypical evolution in childhood epilepsy with occipital paroxysms (Panayiotopoulos type). Epileptic Disord. 2001;3(3):157-162. 76. Prats JM, Garaizar C, Garcia-Nieto ML, Madoz P. Antiepileptic drugs and atypical evolution of idiopathic partial epilepsy. Pediatr Neurol. 1998;18(5):402-406. 77. Ferrie CD, Koutroumanidis M, Rowlinson S, Sanders S, Panayiotopoulos CP. Atypical evolution of Panayiotopoulos syndrome: a case report [published with video-sequences]. Epileptic Disord. 2002;4(1):35-42. 78. Grosso S, Balestri M, Di Bartolo RM, et al. Oxcarbazepine and atypical evolution of benign idiopathic focal epilepsy of childhood. Eur J Neurol. 2006;13(10):1142-1145. 79. Raymond AA, Fish DR, Sisodiya SM, Alsanjari N, Stevens JM, Shorvon SD. Abnormalities of gyration, heterotopias, tuberous sclerosis, focal cortical dysplasia, microdysgenesis, dysembryoplastic neuroepithelial tumour and dysgenesis of the archicortex in epilepsy. Clinical, EEG and neuroimaging features in 100 adult patients. Brain. 1995;118:629-660. 80. Koutroumanidis M, Hennessy MJ, Elwes RD, Binnie CD, Polkey CE. Coexistence of temporal lobe and idiopathic generalized epilepsies. Neurology. 1999;53(3):490-495.

THE EPILEPSIES 3

11

Epilepsies Due to Monogenic Disorders of Metabolism


CHANTAL DEPONDT

Neuronal Ceroid Lipofuscinoses Gaucher Disease Sialidosis

Mitochondrial Encephalomyopathy, Lactic Acidosis, and Strokelike Episodes (MELAS) Pyridoxine-Dependent Seizures

Myoclonic Epilepsy with Ragged-Red Fibers (MERRF)

Glucose Transporter Type 1 (Glut1) Deficiency Syndrome

About 350 different metabolic diseases are known to date, most of which affect the central nervous system. Most disorders present in infancy or childhood. In adults, mitochondrial diseases represent the most important group. Although epilepsy is a frequent symptom of inborn errors of metabolism, only a minority of epilepsies are due to metabolic disorders. Seizures are rarely particular except for myoclonus, which is a frequent feature, and the electroencephalogram (EEG) is rarely specific. Therefore, the diagnosis relies on other signs and symptoms. The diagnosis can often be confirmed by biochemical analysis of blood and urine. The major importance of these disorders lies in the fact that some are treatable. Therefore, it is essential that the diagnosis be made as early as possible. If no specific treatment is available, choice of antiepileptic drug (AED) should be based on seizure type, epilepsy syndrome, and patient characteristics, as in other patients with epilepsy. Valproic acid should be used with caution in many inborn errors of metabolism. This chapter describes a number of inherited metabolic disorders in which epilepsy is a prominent symptom (Table 11-1). These specific disorders were selected among the large group of metabolic diseases for one or more of the following reasons: (1) relatively high prevalence, (2) presentation in adulthood, (3) knowledge of the underlying genetic defect, and (4) availability of a treatment. Disorders that are very rare, that present exclusively in infancy or childhood, in which epilepsy is uncommon, and for which there is no known genetic defect and no specific treatment will not be discussed.

145

TABLE 11-1
Seizure Characteristics

Major Characteristics of Metabolic Disorders


Other Main Symptoms and Signs EEG Characteristics Diagnostic Tests Gene Inheritance Treatment

NCL Myoclonus, GTCS, atypical absences, partial seizures PME, epilepsia partialis continua (adult form)

Cognitive decline, speech abnormalities, involuntary movements, ataxia, spasticity, visual loss

BG slowing; generalized SW, polySW and slow waves Occipital spikes on PS, pseudoperiodic pattern (late-infantile NCL)

PPT1 (CLN1), TPP-1 (CLN2) assays in WBC or fibroblasts; EM on lymphocytes, skin . . .: intracellular inclusions; molecular genetics

CLN1-9

AR (all types) AD (adult type)

Symptomatic Seizures: lamotrigine, valproic acid, clonazepam; avoid phenytoin and carbamazepine Enzyme replacement therapy (IV imiglucerase); substrate replacement therapy (oral miglustat); bone marrow transplantation Symptomatic

Gaucher Disease (Types 23) PME (type 3), Hepatosplenomegaly, BG slowing; Beta-glucosylceramidase nocturnal cytopenia, pulmonary generalized spikes assay in WBC or other and sharp waves nucleated cells GTCS, visual disease, dermatological seizures changes, bone disease; most prominent bulbar and pyramidal posteriorly, signs, oculomotor apraxia, increased by PS and dementia, ataxia eye closure

GBA

AR

Sialidosis PME, GTCS

Types III: visual loss, Normal or slowed BG, Urinary sialic acid-rich macular cherry red spots, brief paroxysms oligosaccharides; cataracts, cerebellar of high amplitude neuraminidase assay in ataxia, neuropathy fast activity WBC or fibroblasts; Type II: Hurler-like EM on lymphocytes, phenotype, dysostosis bone marrow cells . . .: multiplex, short stature, vacuolated cells developmental delay, mental retardation, hepatosplenomegaly

NEU1

AR

MERRF PME, GTCS, partial seizures

Maternal Sensorineural hearing loss, BG slowing; irregular Elevated lactate and pyru- MT-TK vate in blood and CSF; exercise intolerance, generalized SW, (>80% COX-RRF and respiratory A8344G) dementia, neuropathy, polySW, sharp-slow chain enzyme defect on <10% other short stature, optic waves, sometimes atrophy, cardiomyopathy activated by PS; muscle biopsy; mt genes molecular genetics on predominantly occipital focal WBC or muscle spikes or sharp waves

Symptomatic. CoQ10, L-carnitine; myoclonus: levetiracetam; avoid valproic acid

MELAS GTCS, partial seizures, Recurrent headaches and BG slowing; posterior Elevated lac and pyr in MT-TL1 Maternal convulsive and vomiting, anorexia, or diffuse blood and CSF; (>80% nonconvulsive strokelike episodes, epileptiform COX+RRF and A3243G) status exercise intolerance, discharges, respiratory chain enzyme <10% other epilepticus proximal limb weakness, sometimes defect on muscle biopsy; mt genes short stature, dementia, enhanced by PS; molecular genetics on neuropathy, PLEDs WBC, urinary sediment encephalopathy, or muscle sensorineural hearing loss Pyridoxine-Dependent Seizures Neonatal/infantile Encephalopathy, hyper-/ Abnormal sleep Pyridoxine test; elevated ALDH7A1 AR partial seizures, hypothermia, abdominal pattern; generalized pipecolic acid and (95%) GTCS, clonic, tonic, distension, vomiting, 14 Hz sharp and a-AASA in urine, atonic, myoclonic, hepatomegaly, slow activity; burst plasma and CSF infantile spasms, respiratory distress, suppression; focal, recurrent status intellectual disability multifocal or diffuse sharp waves, spikes and polyspikes

Symptomatic CoQ10, L-carnitine; L-arginine for strokelike episodes; Avoid valproic acid

Oral pyridoxine 1518 mg/kg/ day

Table continued on following page

TABLE 11-1
Seizure Characteristics

Major Characteristics of Metabolic Disorders (Continued)


Other Main Symptoms and Signs EEG Characteristics Diagnostic Tests Gene Inheritance Treatment

Glut1-Deficiency Syndrome Generalized tonic and/ Deceleration of head growth, psychomotor or clonic, atypical retardation, ataxia, absences, partial spasticity, various seizures, myoclonic, neurological paroxysmal astatic events, dysarthria, dysphasia

BG slowing or attenuation; generalized 2- to 4Hz SW; focal slowing or attenuation and epileptiform discharges; postprandial improvement

Glycorrhachia 40 mg/dL; CSF/blood glucose 0.33; reduced erythrocyte 3-OMG uptake; molecular genetics

SLC2A1

AD

Ketogenic diet; avoid barbiturates, diazepam, chloral hydrate, ethanol and methylxanthines

GTCS: generalized tonic-clonic seizures; PME: progressive myoclonic epilepsy; BG: background; SW: spike waves; polySW: polyspike waves; WBC: white blood cells; EM: electron microscopy; AR: autosomal recessive; AD: autosomal dominant; COX-RRF: COX-negative ragged red fibers; mt: mitochondrial; CoQ10: coenzyme Q10; PLEDs: periodic lateralized epileptiform discharges; CSF: cerebrospinal fluid; COX+RRF: COX-positive ragged red fibers; 3-OMG: 3-0-methyl-D-glucose.

11  Epilepsies Due to Monogenic Disorders of Metabolism

149

Neuronal Ceroid Lipofuscinoses


The neuronal ceroid lipofuscinoses (NCLs) are a group of inherited, neurodegenerative, lysosomal-storage disorders, mainly occurring in children. Their prevalence is 1 in 25,000. The NCLs are classified based on age of onset, clinical course, pathologic findings, and molecular genetics. The major types are the infantile, late-infantile, juvenile, and adult forms. In addition, four variant forms of the disease have been described: Finnish, Gypsy/Indian, and Turkish variants of late-infantile NCL and Northern epilepsy. Inheritance is usually autosomal recessive, although autosomal-dominant inheritance has been described in the adult form. Epilepsy is a feature of all types of NCL and is usually the first symptom of disease in the late-infantile form.1 Seizures can manifest as myoclonic jerks, generalized tonic-clonic seizures (primary or secondarily), atypical absences, or partial seizures. Adult NCL may present as a progressive myoclonic epilepsy2 or as epilepsia partialis continua.3 Northern epilepsy is characterized by childhood-onset generalized tonic-clonic or complex partial seizures decreasing after puberty and not accompanied by myoclonus.4 Other symptoms of the NCLs include progressive cognitive decline, speech abnormalities, involuntary movements, ataxia, and spasticity. Visual loss is a prominent feature in the infantile, late-infantile, and juvenile forms. The EEG is characterized by slowing of background activity and the presence of generalized spike-wave and polyspike-wave complexes and bursts of slow waves.5 High-voltage, polyphasic spikes in the occipital region with photic stimulation at 1 to 2 Hz, as well as a pseudoperiodic pattern, have been described in the lateinfantile form.6,7 Brain magnetic resonance imaging (MRI) demonstrates cerebral and cerebellar atrophy, T2-hyperintensity of the lobar white matter, thinning of the cerebral cortex, and thalamic T2-hypointensity.8 The diagnostic testing strategy in the NCLs depends mainly on the age of onset. Diagnosis is based on enzymatic assays, electron microscopy, and molecular genetic testing. Two types of enzymatic deficiency have been identified in the NCLs. Palmitoyl-protein thioesterase 1 (PPT1) activity is absent in NCLs caused by mutations in the CLN1/PPT1 gene, and tripeptidyl-peptidase 1 (TPP-1) activity is usually absent in NCLs caused by mutations of the CLN2/TPP1 gene. Enzymatic assays can be performed on leukocytes or fibroblasts. Enzymatic deficiencies in the other NCLs are unknown. Electron microscopy of lymphocytes and tissue biopsies (usually skin) shows typical intracellular inclusions consisting of autofluorescent lipopigment storage material. Eight genesCLN1/PPT1, CLN2/TPP1, CLN3, CLN5, CLN6, CLN7/ MFSD8, CLN8, and CLN10/CTSDare known to be associated with NCL.9 Numerous different mutations have been identified, but most genes carry a small number of common mutations. The genes at the CLN4 ad CLN9 loci have not been identified. The NCLs are characterized by extensive phenotypic and genetic heterogeneity. Therapy for the NCLs is limited to symptomatic treatment at present. Promising future treatments include enzyme replacement therapy, gene therapy, and stem cell therapy.10 Lamotrigine has been reported to be effective and well tolerated in the infantile and juvenile forms.1112 In juvenile NCL, valproic acid is a valuable

150

THE EPILEPSIES 3

alternative, and clonazepam may be useful as add-on therapy.13 Carbamazepine and phenytoin should be avoided, as they may increase seizure activity.14,15

Gaucher Disease
Gaucher disease (GD) is an autosomal recessive lysosomal disorder caused by a deficiency of the enzyme beta-glucocerebrosidase (also called acid beta-glucosidase or acid beta-glucosylceramidase) and accumulation of glucosylceramide (GL1) and other glycolipids. Three major clinical subtypes (1, 2, and 3) and two other subtypes (perinatal lethal and cardiovascular) are recognized based on clinical presentation. Type 1 (nonneuronopathic form) usually has no primary central nervous system symptoms. Types 2 (acute neuronopathic form) and 3 (subacute neuronopathic form) are characterized by the presence of primary neurologic disease. They are classically distinguished by age of onset and rate of disease progression, but it is now increasingly recognized that neuronopathic GD represents a phenotypic continuum, ranging from severely affected infants to asymptomatic adults. GD prevalence estimates vary between 1/57,000 and 1/86,000.16,17 GD is a multisystemic disorder characterized by varying degrees of hematological, skeletal, pulmonary, and neurological involvement. Hepatosplenomegaly usually precedes neurological manifestations. Neurological symptoms may include bulbar and pyramidal signs, oculomotor apraxia, dementia, and ataxia. Epilepsy is mainly a symptom of type 3 GD and usually presents as progressive myoclonic epilepsy (also called type 3a GD).18 The myoclonus may be spontaneous, stimulus sensitive, or induced by action. Other reported seizure types include generalized tonic-clonic seizures during sleep and visual seizures.19,20 EEG shows gradual background slowing and generalized spikes and sharp waves, most prominent over the posterior regions and increased by photic stimulation and eye closure.20,21 Brain MRI may show mild cerebral atrophy. Clinical findings alone are not diagnostic. A confirmatory diagnosis can be established by assay of acid beta-glucosylceramidase enzyme activity in leukocytes or other nucleated cells. In affected individuals, glucosylceramidase enzyme activity in leukocytes is 0 to 15% of normal activity. The results of biochemical testing do not reliably enable prediction of disease severity or subtype. Glucosylceramidase enzyme activity is unreliable for carrier detection given the overlap in enzyme activity levels between carriers and noncarriers. Bone marrow examination shows lipidengorged macrophages with eccentric nuclei (Gaucher cells). However, these are nonspecific, and bone marrow examination is not a reliable diagnostic test. The GBA gene is the only gene known to be associated with GD. At least 200 GBA mutations have been identified.22,23 Four common mutations account for the majority of cases. Molecular genetic testing in a proband is not necessary to confirm the diagnosis, but may be considered for genetic counseling purposes, primarily for carrier detection among at-risk relatives. Genotype-phenotype correlation in GD is poor. Although the genotypic spectrum in patients presenting with progressive myoclonic epilepsy is different from that in other patients with type 3 GD, there appears to be no specific shared genotype.24 Affected individuals should be monitored regularly, including medical history, physical and neurological examination, blood testsespecially hemoglobin concentration and platelet countassessment of spleen and liver volumes, screening for

11  Epilepsies Due to Monogenic Disorders of Metabolism

151

pulmonary hypertension, and skeletal involvement. Enzyme replacement therapy with imiglucerase, an intravenous recombinant glucosylceramidase enzyme preparation, can reverse systemic involvement. The effectiveness of enzyme replacement therapy for the treatment of neurologic disease remains to be established, although a few reports have suggested some benefit.2527 Onset of progressive myoclonic seizures while on enzyme replacement therapy appears to indicate a poor prognosis.28 Substrate reduction therapy with the oral agent miglustat is another treatment option in individuals with mild to moderate GD for whom enzyme replacement therapy is not a therapeutic option. A case report described neurologic improvement in a patient with type 3 GD and myoclonic epilepsy on combined enzyme replacement and substrate reduction therapy.29 Individuals with chronic neurologic GD and progressive disease despite enzyme replacement therapy may be candidates for bone marrow transplantation.

Sialidosis
Sialidosis is a rare autosomal recessive lysosomal disorder caused by a deficiency of the enzyme neuraminidase and accumulation of sialidated glycopeptides and oligosaccharides. Type I sialidosis (also known as the normosomatic type or cherry red spot myoclonus syndrome) usually presents in the second or third decade and is characterized by progressive visual loss, bilateral macular cherry red spots, cataracts, progressive generalized myoclonus, generalized tonic-clonic seizures, and cerebellar ataxia. The myoclonus is often induced by action. Peripheral neuropathy has also been reported.30 Intellect is usually preserved. Type II sialidosis (also known as the dysmorphic type) is the more severe, early onset form and is additionally associated with a Hurler-like phenotype, dysostosis multiplex, short stature, developmental delay, mental retardation, and hepatosplenomegaly. Reports of EEG observations in sialidosis are rare. Brief paroxysms of high amplitude fast activity on a normal or slightly slowed background were described in one study.31 Jerk-locked back-averaging shows a consistent temporal relationship between the EEG spikes and myoclonic jerks.32 Brain MRI shows progressive cerebral and pontocerebellar atrophy.33 Increased amounts of sialic acid-rich oligosaccharides are detected in urine. Lymphocytes and fibroblasts exhibit deficient neuraminidase activity. Electron microscopy of lymphocytes, bone marrow cells, and many other tissue types shows vacuolated cells. The sialidoses are caused by mutations in the NEU1 gene. Its product, neuraminidase or lysosomal sialidase, has a dual physiologic function: it participates in intralysosomal catabolism of sialated glycoconjugates and is involved in cellular immune response. More than 40 different mutations have been characterized.3436 In general, there is a close correlation between the residual enzyme activity and the clinical disease severity.37 Treatment is symptomatic. At present, there is no disease-specific treatment available, but potential treatment strategies such as gene therapy and enzyme replacement therapy are under study.38 Treatment of seizures and myoclonus is aspecific. One report described successful treatment of myoclonus with 5-hydroxytryptophan as add-on therapy.39

152

THE EPILEPSIES 3

Myoclonic Epilepsy with Ragged-Red Fibers (MERRF)


MERRF is a mitochondrial disease presenting primarily as progressive myoclonic epilepsy. It is transmitted by maternal inheritance. Onset is usually in childhood, after an initially normal early development. The disease is characterized by myoclonus, which is often the first symptom, followed by epilepsy and ataxia. Seizures are usually generalized myoclonic or tonicclonic, but partial seizures have been reported in atypical cases.40 Other common manifestations include sensorineural hearing loss, exercise intolerance, dementia, peripheral neuropathy, short stature, optic atrophy, and cardiomyopathy. Pigmentary retinopathy, ophthalmoparesis, pyramidal signs, and multiple lipomas are occasionally observed. The disease manifestations are very heterogeneous, and overlap syndromes between MERRF and mitochondrial encephalomyopathy, lactic acidosis, and strokelike episodes (MELAS) have been reported.41 Lactate and pyruvate levels in blood and cerebrospinal fluid (CSF) are commonly elevated at rest and increase excessively after moderate activity. EEG findings include slowing of background activity and bursts of atypical, irregular generalized spikewave complexes, polyspike-wave complexes or sharp-slow wave complexes, sometimes activated by photic stimulation.40,42,43 These discharges are often, but not always, related to generalized myoclonic jerks. Focal spikes or sharp waves are also seen, most commonly over the occipital regions. Brain MRI often shows brain atrophy and basal ganglia calcification. Muscle biopsy shows typical cytochrome oxidase (COX) negative ragged-red fibers. Biochemical studies in muscle usually show defects in respiratory chain enzyme activity, especially COX deficiency, but may occasionally be normal. The diagnosis is confirmed by molecular genetic testing, demonstrating mutations in the mitochondrial DNA (mtDNA) gene MT-TK, encoding tRNALys. Over 80% of affected individuals with typical findings carry the A8344G mutation.44 Three additional mutations (T8356C, G8363A, and G8361A) account for about 10% of affected individuals. The mutations appear to exert their pathogenic effects through impairment of protein synthesis.45 The remaining 10% of affected individuals may have other mutations in MT-TK or mutations in a number of other mitochondrial genes. Mutations are usually present in all tissues and can thus be detected in mtDNA from blood leukocytes. However, the occurrence of heteroplasmy (differences in cellular mutational load) can result in variations of tissue distribution of mutated mtDNA. Hence, in some cases the mutation may be undetectable in leukocytes and may only be detected in other tissues, most reliably in skeletal muscle. For the same reason, accurate prediction of phenotype in oligosymptomatic individuals or at-risk family members based on test results is not possible. There is no specific treatment for the disease. Empirical treatment with coenzyme Q10 (50 to 100 mg 3x/day) and L-carnitine (1000 mg 3/day), aiming to improve mitochondrial function, is often used.46 No controlled studies have compared the efficacy of various AEDs in MERRF. Valproic acid should be used with caution because of the increased risk of hepatotoxicity.47 A number of reports have described substantial improvement of myoclonus with levetiracetam.48,49 Potential future treatments for MERRF include selective inhibition of mutant mtDNA replication by peptide nucleic acids and import of nuclear-encoded tRNALys into mitochondria.50,51

11  Epilepsies Due to Monogenic Disorders of Metabolism

153

Mitochondrial Encephalomyopathy, Lactic Acidosis, and Strokelike Episodes (MELAS)


MELAS is another disorder caused by mutations in mtDNA and thus is transmitted by maternal inheritance. Onset is typically in childhood, though infantile and adult onset has been reported. The most common presenting symptoms are seizures, recurrent headaches, anorexia, and recurrent vomiting. Seizures can be generalized or partial and are often associated with strokelike episodes of transient hemiparesis or cortical blindness. Generalized convulsive and complex partial status epilepticus, as well as epilepsia partialis continua, have been reported.5254 Other common symptoms and signs include exercise intolerance, proximal limb weakness, short stature, encephalopathy, dementia, sensorineural hearing loss, and peripheral neuropathy. Less-common symptoms include myoclonus, ataxia, episodic coma, optic atrophy, cardiomyopathy, pigmentary retinopathy, ophthalmoplegia, diabetes mellitus, hirsutism, gastrointestinal dysmotility, and nephropathy. MELAS should be suspected based on the following features: (1) strokelike episodes before age 40; (2) encephalopathy with seizures and/or dementia; and (3) lactic acidosis and/or ragged-red fibers on muscle biopsy, or both. The diagnosis may be confirmed if at least two of the following are also present: normal early psychomotor development, recurrent headache, or recurrent vomiting.55 There is a wide variability in clinical presentation, and overlap syndromes between MELAS and other mitochondrial disorders have been reported. Lactate and pyruvate levels in blood and CSF are commonly elevated at rest and increase excessively after moderate activity. EEG commonly shows slowing of alpharhythm and posterior or diffuse epileptiform discharges, sometimes enhanced by photic stimulation.40,5658 Periodic lateralized epileptiform discharges (PLEDs) with alternating focus have been reported during episodes of complex partial status epilepticus.52,59 Basal ganglia calcifications are commonly seen on computed tomography (CT). Brain MRI during strokelike episodes shows cortico-subcortical T2- and FLAIR-hyperintense lesions in the posterior regions that slowly spread in the weeks following initial symptoms.6062 Diffusion-weighted MRI (DWI) demonstrates increased apparent diffusion coefficient (ADC) values in these lesions, distinguishing them from classic ischemic strokes.61 Muscle biopsy typically shows COX-positive ragged red fibers and an abundance of mitochondria. Biochemical analysis of respiratory chain enzymes in muscle usually shows multiple partial defects, but it can also be normal. The diagnosis is confirmed by molecular genetic testing, demonstrating mutations in the mtDNA gene MT-TL1, encoding tRNALeu. tRNALeu is essential for mitochondrial protein synthesis, specifically for the incorporation of leucine into nascent proteins. The A3243G mutation is present in over 80% of individuals with typical clinical findings.63 The remainder of cases are caused by other mutations in MT-TL1 or mutations in other mtDNA genes, most commonly MT-ND5. Mutations are usually present in all tissues and can thus be detected in leukocytes. However, the percentage of mutated mtDNA decreases in blood with age in patients harboring the A3243G mutation,64 and in some cases the pathogenic mutation may be undetectable in leukocytes because of heteroplasmy. In such cases, urinary sediment has proven the most useful among accessible tissues for detecting the A3243G mutation.65,66 A muscle biopsy is recommended in the rare instance in which the MT-TL1 A3243G mutation cannot be detected by standard techniques in

154

THE EPILEPSIES 3

leukocytes or urinary sediment from an individual with classic MELAS. Accurate prediction of phenotype in oligosymptomatic individuals or at-risk family members based on genetic test results is not possible. No specific treatment for MELAS exists. Coenzyme Q10 (50 to 100 mg 3/day) and L-carnitine (1000 mg 3/day) may be of some benefit. L-arginine has been reported to improve strokelike symptoms when given in the acute phase and to decrease frequency and severity of strokelike episodes when given between episodes.67,68 Beneficial effects of oral succinate were reported in one individual.69 Seizures usually respond to traditional AEDs, though valproic acid may aggravate seizures.70,71

Pyridoxine-Dependent Seizures
Pyridoxine-dependent seizures are a rare seizure disorder usually presenting in the neonatal period. The disorder is inherited in an autosomal-recessive manner. The importance of the disorder lies in the fact that it responds to treatment with pyridoxine (vitamin B6). The hallmark of the disease consists of intractable seizures not controlled with AEDs but responding to pyridoxine. Seizures classically present soon after birth, but onset up to the age of 3 years has been reported.7274 Seizures may be partial, generalized tonic-clonic, clonic, tonic, atonic, myoclonic, or infantile spasms. Prolonged seizures and recurrent episodes of status epilepticus are typical. Periods of encephalopathy frequently precede seizures. Systemic manifestations may also occur, including hyper- or hypothermia, abdominal distension, vomiting, hepatomegaly, respiratory distress with hypoxemia, and metabolic acidosis, all of which resolve with pyridoxine treatment.75 Intellectual disability is common. Several atypical cases have been reported in the literature, including seizures that initially respond to AEDs and then become intractable,76 seizures that only respond to pyridoxine after several months,77 and seizure-free intervals of up to several months after discontinuation of pyridoxine.74,78 EEG background activity may be normal or abnormal, but normal sleep patterns are absent. Other EEG abnormalities include generalized bursts of 1 to 4 Hz sharp and slow activity, burst suppression patterns, and focal, multifocal, or diffuse sharp waves, spikes, and polyspikes.75,79,80 The interictal EEG may remain normal in patients with later onset. Brain imaging shows a variety of abnormalities.81,82 The most typical structural abnormality is hypoplasia of the posterior part of the corpus callosum. Other findings include cerebellar hypoplasia with megacisterna magna and hydrocephalus. Untreated seizures can be associated with intraventricular and/or subarachnoid hemorrhage and white matter changes, which may be partially reversible.83,84 The diagnosis can be established on a clinical basis by administering pyridoxine 100 mg intravenously while monitoring the EEG. In individuals with pyridoxinedependent seizures, clinical seizures generally cease over several minutes. If a clinical response is not demonstrated after 10 minutes, the dose should be repeated. In some patients, doses exceeding 500 mg have been required.85,86 A corresponding change should be observed in the EEG, though in some cases the response may be delayed by several hours. Close systemic monitoring is essential, as neurologic and cardiorespiratory depression following this trial has been reported

11  Epilepsies Due to Monogenic Disorders of Metabolism

155

in some cases.77,84,87 In older children or during the interictal phase, pyridoxine can be administered orally at 15 to 30 mg/kg/day. In individuals with pyridoxinedependent seizures, clinical seizures should cease within a week. Elevated levels of pipecolic acid and a-aminoadipic semialdehyde (a-AASA) in urine, plasma, and CSF can be used as biomarkers for the disease.88,89 Mutations in the gene ALDH7A1, encoding antiquitin (ATQ1), are responsible for the disease in 95% of typical cases with elevated urinary a-AASA.88,89 ATQ-1 is an aldehyde dehydrogenase. The mutations result in absent or strongly reduced a-AASA dehydrogenase enzyme activity, leading to increased levels of 1-piperideine6-carboxylate (P6C). P6C, in turn, condenses with and inactivates pyridoxal 50 -phosphate (PLP), a cofactor of glutamic acid decarboxylase (GAD). This probably results in abnormal metabolism of neurotransmitters. At least one family not linked to the 5q31 locus that harbors the ALDH7A1 gene has been reported, indicating genetic heterogeneity.90 The main prognostic factor is age at onset: patients presenting after 1 month of age do considerably better than those with earlier onsets.75 The question of whether early initiation of treatment improves outcome is currently a matter of debate.86,91 The recommended daily dose of oral pyridoxine is 15 to 18 mg/kg, with a maximum of 500 mg/day.75 Once seizures are controlled, all AEDs can be withdrawn, and seizure control will be maintained on pyridoxine monotherapy. Because the disease manifestations may exacerbate during intercurrent illnesses such as gastroenteritis or respiratory infection, it is recommended to double the daily dose of pyridoxine for several days until the acute illness resolves. Therapy should be maintained lifelong. For couples who have a child with the disorder, it has been recommended that mothers take 50 to 100 mg of pyridoxine per day during the last half of subsequent pregnancies to reduce the severity of intellectual impairment in a possibly affected fetus.80,91

Glucose Transporter Type 1 (Glut1) Deficiency Syndrome


Glut1-deficiency syndrome is an autosomal dominant disorder usually presenting as an infantile-onset epileptic encephalopathy. Children are normal at birth and start presenting refractory seizures between ages 1 to 18 months.92 Other symptoms include deceleration of head growth, psychomotor retardation, ataxia, and spasticity. Apneic episodes, behavioral arrests, and abnormal episodic eye movements simulating opsoclonus may precede the onset of seizures by several months.93 Seizures can be generalized tonic and/or clonic, atypical absences, partial, myoclonic, or astatic.92 More rarely seizures may be absent.94,95 Other paroxysmal events including intermittent ataxia, confusion, lethargy, or somnolence; alternating hemiparesis, abnormalities of movement or posture such as myoclonus and dystonia, total body paralysis, sleep disturbances, and recurrent headaches have also been described.95 It is unclear whether these events represent epileptic or nonepileptic phenomena. Dysarthria and predominantly expressive dysphasia are common. Most patients exhibit abnormal movements, ranging from motor restlessness to dystonia. Neurologic symptoms generally fluctuate and may be triggered by fasting or fatigue. Rare cases diagnosed in adulthood have been reported.96 EEG abnormalities include generalized 2- to 4-Hz spike-wave discharges, generalized slowing or attenuation, focal epileptiform discharges, and focal

156

THE EPILEPSIES 3

slowing or attenuation. A significant proportion of cases has normal interictal EEGs.92,97,98 A reduction of epileptiform discharges during postprandial EEG recording has been reported in some cases.98,99 Brain imaging usually shows no structural abnormalities.95,100 Cerebral fluorodeoxyglucose-positron emission tomography (FDG-PET) reveals a global decrease in glucose uptake with relative preservation of basal ganglia metabolism.100 The biochemical hallmark of the disease is hypoglycorrhachia with normal blood glucose levels, measured following a 4-hour fast. The absolute CSF glucose concentration seldom, if ever, exceeds 40 mg/dL. The ratio of CSF glucose concentration to blood glucose concentration is about 0.33. CSF lactate concentration is low to normal.95 The diagnosis is supported by the presence of reduced 3-0-methyl-Dglucose (3-OMG) uptake in erythrocytes, although normal values have been reported.101,102 Glut1-deficiency syndrome is caused by mutations in the gene SLC2A1 (solute carrier family 2, facilitated glucose transporter member 1).103 Mutations are detected in about 80% of affected individuals. Whole-gene deletions have also been reported.103105 Most mutations occur de novo. Glut1 is the major glucose transporter in the mammalian bloodbrain barrier, responsible for glucose entry into the brain. It is selectively expressed in brain endothelial cells, astroglia, and erythrocytes.106 The seizures are usually refractory to conventional AEDs. The ketogenic diet is highly effective in controlling the seizures and is well tolerated in most cases; however, neurobehavioral and cognitive deficits persist.107 Seizures sometimes recur despite the diet.108,109 Barbiturates, diazepam, and chloral hydrate are known to inhibit transport of glucose and thus should be avoided.109,110 Methylxanthines (e.g., caffeine) and ethanol also inhibit glucose transport by Glut1 and may worsen symptoms of Glut1-deficiency syndrome.111,112

REFERENCES
1. Nardocci N, Verga ML, Binelli S, Zorzi G, Angelini L, Bugiani O. Neuronal ceroid-lipofuscinosis: a clinical and morphological study of 19 patients. Am J Med Genet. 1995;57:137-141. 2. Sadzot B, Reznik M, Arrese-Estrada JE, Franck G. Familial Kufs disease presenting as a progressive myoclonic epilepsy. J Neurol. 2000;247:447-454. 3. Gambardella A, Pasquinelli G, Cittadella R, et al. Kufs disease presenting as late-onset epilepsia partialis continua. Neurology. 1998;51:1180-1182. 4. Ranta S, Lehesjoki AE. Northern epilepsy, a new member of the NCL family. Neurol Sci. 2000; 21:S43-S47. 5. Sinha S, Satishchandra P, Santosh V, Gayatri N, Shankar SK. Neuronal ceroid lipofuscinosis: a clinicopathological study. Seizure. 2004;13:235-240. 6. Taratuto AL, Saccoliti M, Sevlever G, et al. Childhood neuronal ceroid-lipofuscinoses in Argentina. Am J Med Genet. 1995;57:144-149. 7. Veneselli E, Biancheri R, Perrone MV, Buoni S, Fois A. Neuronal ceroid lipofuscinoses: clinical and EEG findings in a large study of Italian cases. Neurol Sci. 2000;21:S75-81. 8. DIncerti L. MRI in neuronal ceroid lipofuscinosis. Neurol Sci. 2000;21:S71-73. 9. http://www.ucl.ac.uk/ncl. 10. Hobert JA, Dawson G. Neuronal ceroid lipofuscinoses therapeutic strategies: past, present and future. Biochim Biophys Acta. 2006;1762:945-953. 11. Aberg L, Heiskala H, Vanhanen SL, et al. Lamotrigine therapy in infantile neuronal ceroid lipofuscinosis (INCL). Neuropediatrics. 1997;28:77-79. 12. Aberg L, Kirveskari E, Santavuori P. Lamotrigine therapy in juvenile neuronal ceroid lipofuscinosis. Epilepsia. 1999;40:796-799.

11  Epilepsies Due to Monogenic Disorders of Metabolism 13. Aberg LE, Backman M, Kirveskari E, Santavuori P. Epilepsy and antiepileptic drug therapy in juvenile neuronal ceroid lipofuscinosis. Epilepsia. 2000;41:1296-1302. 14. Philippart M. Diagnosis and treatment of typical and atypical forms of lipopigment storage disorders. Am J Med Genet Suppl. 1988;5:291-298. 15. Philippart M, Messa C, Chugani HT. Spielmeyer-Vogt (Batten, Spielmeyer-Sjo gren) disease. Distinctive patterns of cerebral glucose utilization. Brain. 1994;117 (Pt 5):1085-1092. 16. Meikle PJ, Hopwood JJ, Clague AE, Carey WF. Prevalence of lysosomal storage disorders. JAMA. 1999;281:249-254. 17. Poorthuis BJ, Wevers RA, Kleijer WJ, et al. The frequency of lysosomal storage diseases in the Netherlands. Hum Genet. 1999;105:151-156. 18. Patterson MC, Horowitz M, Abel RB, et al. Isolated horizontal supranuclear gaze palsy as a marker of severe systemic involvement in Gauchers disease. Neurology. 1993;43:1993-1997. 19. Filocamo M, Mazzotti R, Stroppiano M, Grossi S, Dravet C, Guerrini R. Early visual seizures and progressive myoclonus epilepsy in neuronopathic Gaucher disease due to a rare compound heterozygosity (N188S/S107L). Epilepsia. 2004;45:1154-1157. 20. Tuzun E, Baykan B, Gurses C, Gokyigit A. Long-term follow-up of electroencephalographic and clinical findings of a case with Gauchers disease type 3a. Seizure. 2000;9:469-472. 21. Nishimura R, Omos-Lau N, Ajmone-Marsan C, Barranger JA. Electroencephalographic findings in Gaucher disease. Neurology. 1980;30:152-159. 22. Jmoudiak M, Futerman AH. Gaucher disease: pathological mechanisms and modern management. Br J Haematol. 2005;129:178-188. 23. Tsuji S, Choudary PV, Martin BM, et al. A mutation in the human glucocerebrosidase gene in neuronopathic Gauchers disease. N Engl J Med. 1987;316:570-575. 24. Park JK, Orvisky E, Tayebi N, et al. Myoclonic epilepsy in Gaucher disease: genotype-phenotype insights from a rare patient subgroup. Pediatr Res. 2003;53:387-395. 25. Altarescu G, Hill S, Wiggs E, et al. The efficacy of enzyme replacement therapy in patients with chronic neuronopathic Gauchers disease. J Pediatr. 2001;138:539-547. 26. Erikson A, Forsberg H, Nilsson M, Astrom M, Mansson JE. Ten years experience of enzyme infusion therapy of Norrbottnian (type 3) Gaucher disease. Acta Paediatr. 2006;95:312-317. 27. Vellodi A, Bembi B, de Villemeur TB, et al. Management of neuronopathic Gaucher disease: a European consensus. J Inherit Metab Dis. 2001;24:319-327. 28. Frei KP, Schiffmann R. Myoclonus in Gaucher disease. Adv Neurol. 2002;89:41-48. 29. Capablo JL, Franco R, de Cabezon AS, Alfonso P, Pocovi M, Giraldo P. Neurologic improvement in a Type 3 Gaucher disease patient treated with imiglucerase/miglustat combination. Epilepsia. 2007;48:1406-1408. 30. Steinman L, Tharp BR, Dorfman LJ, et al. Peripheral neuropathy in the cherry-red spot-myoclonus syndrome (sialidosis type I). Ann Neurol. 1980;7:450-456. 31. Panzica F, Canafoglia L, Franceschetti S, et al. Movement-activated myoclonus in genetically defined progressive myoclonic epilepsies: EEG-EMG relationship estimated using autoregressive models. Clin Neurophysiol. 2003;114:1041-1052. 32. Tobimatsu S, Fukui R, Shibasaki H, Kato M, Kuroiwa Y. Electrophysiological studies of myoclonus in sialidosis type 2. Electroencephalogr Clin Neurophysiol. 1985;60:16-22. 33. Palmeri S, Villanova M, Malandrini A, et al. Type I sialidosis: a clinical, biochemical and neuroradiological study. Eur Neurol. 2000;43:88-94. 34. Loren DJ, Campos Y, dAzzo A, et al. Sialidosis presenting as severe nonimmune fetal hydrops is associated with two novel mutations in lysosomal alpha-neuraminidase. J Perinatol. 2005;25:491-494. 35. Pattison S, Pankarican M, Rupar CA, Graham FL, Igdoura SA. Five novel mutations in the lysosomal sialidase gene (NEU1) in type II sialidosis patients and assessment of their impact on enzyme activity and intracellular targeting using adenovirus-mediated expression. Hum Mutat. 2004; 23:32-39. 36. Seyrantepe V, Poupetova H, Froissart R, Zabot MT, Maire I, Pshezhetsky AV. Molecular pathology of NEU1 gene in sialidosis. Hum Mutat. 2003;22:343-352. 37. Bonten EJ, Arts WF, Beck M, et al. Novel mutations in lysosomal neuraminidase identify functional domains and determine clinical severity in sialidosis. Hum Mol Genet. 2000;9:2715-2725. 38. Oheda Y, Kotani M, Murata M, et al. Elimination of abnormal sialylglycoproteins in fibroblasts with sialidosis and galactosialidosis by normal gene transfer and enzyme replacement. Glycobiology. 2006;16:271-280. 39. Gascon G, Wallenberg B, Daif AK, Ozand P. Successful treatment of cherry red spot-myoclonus syndrome with 5-hydroxytryptophan. Ann Neurol. 1988;24:453-455.

157

158

THE EPILEPSIES 3 40. Canafoglia L, Franceschetti S, Antozzi C, et al. Epileptic phenotypes associated with mitochondrial disorders. Neurology. 2001;56:1340-1346. 41. Zeviani M, Muntoni F, Savarese N, et al. A MERRF/MELAS overlap syndrome associated with a new point mutation in the mitochondrial DNA tRNA (Lys) gene. Eur J Hum Genet. 1993;1:80-87. 42. Serra G, Piccinnu R, Tondi M, Muntoni F, Zeviani M, Mastropaolo C. Clinical and EEG findings in eleven patients affected by mitochondrial encephalomyopathy with MERRF-MELAS overlap. Brain Dev. 1996;18:185-191. 43. So N, Berkovic S, Andermann F, Kuzniecky R, Gendron D, Quesney LF. Myoclonus epilepsy and ragged-red fibres (MERRF). 2. Electrophysiological studies and comparison with other progressive myoclonus epilepsies. Brain. 1989;112 (Pt 5):1261-1276. 44. Shoffner JM, Lott MT, Lezza AM, Seibel P, Ballinger SW, Wallace DC. Myoclonic epilepsy and raggedred fiber disease (MERRF) is associated with a mitochondrial DNA tRNA(Lys) mutation. Cell. 1990;61:931-937. 45. Masucci JP, Davidson M, Koga Y, Schon EA, King MP. In vitro analysis of mutations causing myoclonus epilepsy with ragged-red fibers in the mitochondrial tRNA(Lys) gene: two genotypes produce similar phenotypes. Mol Cell Biol. 1995;15:2872-2881. 46. DiMauro S, Hirano M, Schon EA. Mitochondrial encephalomyopathies: therapeutic approaches. Neurol Sci. 2000;21:S901-908. 47. Krahenbuhl S, Brandner S, Kleinle S, Liechti S, Straumann D. Mitochondrial diseases represent a risk factor for valproate-induced fulminant liver failure. Liver. 2000;20:346-348. 48. Crest C, Dupont S, Leguern E, Adam C, Baulac M. Levetiracetam in progressive myoclonic epilepsy: an exploratory study in 9 patients. Neurology. 2004;62:640-643. 49. Mancuso M, Galli R, Pizzanelli C, Filosto M, Siciliano G, Murri L. Antimyoclonic effect of levetiracetam in MERRF syndrome. J Neurol Sci. 2006;243:97-99. 50. Kolesnikova OA, Entelis NS, Jacquin-Becker C, et al. Nuclear DNA-encoded tRNAs targeted into mitochondria can rescue a mitochondrial DNA mutation associated with the MERRF syndrome in cultured human cells. Hum Mol Genet. 2004;13:2519-2534. 51. Taylor RW, Chinnery PF, Turnbull DM, Lightowlers RN. Selective inhibition of mutant human mitochondrial DNA replication in vitro by peptide nucleic acids. Nat Genet. 1997;15:212-215. 52. Leff AP, McNabb AW, Hanna MG, Clarke CR, Larner AJ. Complex partial status epilepticus in late-onset MELAS. Epilepsia. 1998;39:438-441. 53. Montagna P, Gallassi R, Medori R, et al. MELAS syndrome: characteristic migrainous and epileptic features and maternal transmission. Neurology. 1988;38:751-754. 54. Veggiotti P, Colamaria V, Dalla Bernardina B, Martelli A, Mangione D, Lanzi G. Epilepsia partialis continua in a case of MELAS: clinical and neurophysiological study. Neurophysiol Clin. 1995;25: 158-166. 55. Hirano M, Ricci E, Koenigsberger MR, et al. Melas: an original case and clinical criteria for diagnosis. Neuromuscul Disord. 1992;2:125-135. 56. Fujimoto S, Mizuno K, Shibata H, et al. Serial electroencephalographic findings in patients with MELAS. Pediatr Neurol. 1999;20:43-48. 57. Majamaa-Voltti KA, Winqvist S, Remes AM, et al. A 3-year clinical follow-up of adult patients with 3243A>G in mitochondrial DNA. Neurology. 2006;66:1470-1475. 58. Tulinius MH, Hagne I. EEG findings in children and adolescents with mitochondrial encephalomyopathies: a study of 25 cases. Brain Dev. 1991;13:167-173. 59. Corda D, Rosati G, Deiana GA, Sechi G. Erratic complex partial status epilepticus as a presenting feature of MELAS. Epilepsy Behav. 2006;8:655-658. 60. Iizuka T, Sakai F, Kan S, Suzuki N. Slowly progressive spread of the stroke-like lesions in MELAS. Neurology. 2003;61:1238-1244. 61. Kolb SJ, Costello F, Lee AG, et al. Distinguishing ischemic stroke from the stroke-like lesions of MELAS using apparent diffusion coefficient mapping. J Neurol Sci. 2003;216:11-15. 62. Yonemura K, Hasegawa Y, Kimura K, Minematsu K, Yamaguchi T. Diffusion-weighted MR imaging in a case of mitochondrial myopathy, encephalopathy, lactic acidosis, and strokelike episodes. AJNR Am J Neuroradiol. 2001;22:269-272. 63. Goto Y, Nonaka I, Horai S. A mutation in the tRNA(Leu)(UUR) gene associated with the MELAS subgroup of mitochondrial encephalomyopathies. Nature. 1990;348:651-653. 64. Pyle A, Taylor RW, Durham SE, et al. Depletion of mitochondrial DNA in leucocytes harbouring the 3243A-G mtDNA mutation. J Med Genet. 2007;44:69-74. 65. McDonnell MT, Schaefer AM, Blakely EL, et al. Noninvasive diagnosis of the 3243AG mitochondrial DNA mutation using urinary epithelial cells. Eur J Hum Genet. 2004;12:778-781.

11  Epilepsies Due to Monogenic Disorders of Metabolism 66. Shanske S, Pancrudo J, Kaufmann P, et al. Varying loads of the mitochondrial DNA A3243G mutation in different tissues: implications for diagnosis. Am J Med Genet A. 2004;130:134-137. 67. Koga Y, Akita Y, Nishioka J, et al. MELAS and L-arginine therapy. Mitochondrion. 2007;7:133-139. 68. Kubota M, Sakakihara Y, Mori M, Yamagata T, Momoi-Yoshida M. Beneficial effect of L-arginine for stroke-like episode in MELAS. Brain Dev. 2004;26:481-483:discussion 480. 69. Oguro H, Iijima K, Takahashi K, et al. Successful treatment with succinate in a patient with MELAS. Intern Med. 2004;43:427-431. 70. Lam CW, Lau CH, Williams JC, Chan YW, Wong LJ. Mitochondrial myopathy, encephalopathy, lactic acidosis and stroke-like episodes (MELAS) triggered by valproate therapy. Eur J Pediatr. 1997;156:562-564. 71. Lin CM, Thajeb P. Valproic acid aggravates epilepsy due to MELAS in a patient with an A3243G mutation of mitochondrial DNA. Metab Brain Dis. 2007;22:105-109. 72. Bachman DS. Late-onset pyridoxine-dependency convulsions. Ann Neurol. 1983;14:692-693. 73. Coker SB. Postneonatal vitamin B6-dependent epilepsy. Pediatrics. 1992;90:221-223. 74. Goutieres F, Aicardi J. Atypical presentations of pyridoxine-dependent seizures: a treatable cause of intractable epilepsy in infants. Ann Neurol. 1985;17:117-120. 75. Baxter P. Pyridoxine-dependent and pyridoxine-responsive seizures. Dev Med Child Neurol. 2001; 43:416-420. 76. Gospe SM, Jr., Olin KL, Keen CL. Reduced GABA synthesis in pyridoxine-dependent seizures. Lancet. 1994;343:1133-1134. 77. Bass NE, Wyllie E, Cohen B, Joseph SA. Pyridoxine-dependent epilepsy: the need for repeated pyridoxine trials and the risk of severe electrocerebral suppression with intravenous pyridoxine infusion. J Child Neurol. 1996;11:422-424. 78. Bankier A, Turner M, Hopkins IJ. Pyridoxine dependent seizuresa wider clinical spectrum. Arch Dis Child. 1983;58:415-418. 79. Mikati MA, Trevathan E, Krishnamoorthy KS, Lombroso CT. Pyridoxine-dependent epilepsy: EEG investigations and long-term follow-up. Electroencephalogr Clin Neurophysiol. 1991;78: 215-221. 80. Nabbout R, Soufflet C, Plouin P, Dulac O. Pyridoxine dependent epilepsy: a suggestive electroclinical pattern. Arch Dis Child Fetal Neonatal Ed. 1999;81:F125-F129. 81. Baxter P, Griffiths P, Kelly T, Gardner-Medwin D. Pyridoxine-dependent seizures: demographic, clinical, MRI and psychometric features, and effect of dose on intelligence quotient. Dev Med Child Neurol. 1996;38:998-1006. 82. Gospe SM, Jr., Hecht ST. Longitudinal MRI findings in pyridoxine-dependent seizures. Neurology. 1998;51:74-78. 83. Jardim LB, Pires RF, Martins CE, et al. Pyridoxine-dependent seizures associated with white matter abnormalities. Neuropediatrics. 1994;25:259-261. 84. Tanaka R, Okumura M, Arima J, Yamakura S, Momoi T. Pyridoxine-dependent seizures: report of a case with atypical clinical features and abnormal MRI scans. J Child Neurol. 1992;7:24-28. 85. Clarke TA, Saunders BS, Feldman B. Pyridoxine-dependent seizures requiring high doses of pyridoxine for control. Am J Dis Child. 1979;133:963-965. 86. Haenggeli CA, Girardin E, Paunier L. Pyridoxine-dependent seizures, clinical and therapeutic aspects. Eur J Pediatr. 1991;150:452-455. 87. Kroll JS. Pyridoxine for neonatal seizures: an unexpected danger. Dev Med Child Neurol. 1985;27:377-379. 88. Mills PB, Struys E, Jakobs C, et al. Mutations in antiquitin in individuals with pyridoxine-dependent seizures. Nat Med. 2006;12:307-309. 89. Plecko B, Paul K, Paschke E, et al. Biochemical and molecular characterization of 18 patients with pyridoxine-dependent epilepsy and mutations of the antiquitin (ALDH7A1) gene. Hum Mutat. 2007;28:19-26. 90. Bennett CL, Huynh HM, Chance PF, Glass IA, Gospe, SM Jr. Genetic heterogeneity for autosomal recessive pyridoxine-dependent seizures. Neurogenetics. 2005;6:143-149. 91. Baxter P, Aicardi J. Neonatal seizures after pyridoxine use. Lancet. 1999;354:2082-2083. 92. Leary LD, Wang D, Nordli DR, Jr., Engelstad K, De Vivo DC. Seizure characterization and electroencephalographic features in Glut-1 deficiency syndrome. Epilepsia. 2003;44:701-707. 93. De Vivo DC, Leary L, Wang D. Glucose transporter 1 deficiency syndrome and other glycolytic defects. J Child Neurol. 2002;17(Suppl 3):3S15-3S1523:discussion 13S24-15. 94. Overweg-Plandsoen WC, Groener JE, Wang D, et al. GLUT-1 deficiency without epilepsyan exceptional case. J Inherit Metab Dis. 2003;26:559-563.

159

160

THE EPILEPSIES 3 95. Wang D, Pascual JM, Yang H, et al. Glut-1 deficiency syndrome: clinical, genetic, and therapeutic aspects. Ann Neurol. 2005;57:111-118. 96. Klepper J, Willemsen M, Verrips A, et al. Autosomal dominant transmission of GLUT1 deficiency. Hum Mol Genet. 2001;10:63-68. 97. Boles RG, Seashore MR, Mitchell WG, Kollros PR, Mofidi S, Novotny EJ. Glucose transporter type 1 deficiency: a study of two cases with video-EEG. Eur J Pediatr. 1999;158:978-983. 98. von Moers A, Brockmann K, Wang D, et al. EEG features of glut-1 deficiency syndrome. Epilepsia. 2002;43:941-945. 99. Ito Y, Gertsen E, Oguni H, et al. Clinical presentation, EEG studies, and novel mutations in two cases of GLUT1 deficiency syndrome in Japan. Brain Dev. 2005;27:311-317. 100. Pascual JM, Van Heertum RL, Wang D, Engelstad K, De Vivo DC. Imaging the metabolic footprint of Glut1 deficiency on the brain. Ann Neurol. 2002;52:458-464. 101. Fujii T, Ho YY, Wang D, et al. Three Japanese patients with glucose transporter type 1 deficiency syndrome. Brain Dev. 2007;29:92-97. 102. Klepper J, Garcia-Alvarez M, ODriscoll KR, et al. Erythrocyte 3-O-methyl-D-glucose uptake assay for diagnosis of glucose-transporter-protein syndrome. J Clin Lab Anal. 1999;13:116-121. 103. Seidner G, Alvarez MG, Yeh JI, et al. GLUT-1 deficiency syndrome caused by haploinsufficiency of the blood-brain barrier hexose carrier. Nat Genet. 1998;18:188-191. 104. Vermeer S, Koolen DA, Visser G, et al. A novel microdeletion in 1(p34.2p34.3), involving the SLC2A1 (GLUT1) gene, and severe delayed development. Dev Med Child Neurol. 2007;49:380-384. 105. Wang D, Kranz-Eble P, De Vivo DC. Mutational analysis of GLUT1 (SLC2A1) in Glut-1 deficiency syndrome. Hum Mutat. 2000;16:224-231. 106. Pardridge WM, Boado RJ, Farrell CR. Brain-type glucose transporter (GLUT-1) is selectively localized to the blood-brain barrier. Studies with quantitative western blotting and in situ hybridization. J Biol Chem. 1990;265:18035-18040. 107. Klepper J, Leiendecker B, Bredahl R, et al. Introduction of a ketogenic diet in young infants. J Inherit Metab Dis. 2002;25:449-460. 108. Klepper J, Scheffer H, Leiendecker B, et al. Seizure control and acceptance of the ketogenic diet in GLUT1 deficiency syndrome: a 2- to 5-year follow-up of 15 children enrolled prospectively. Neuropediatrics. 2005;36:302-308. 109. Klepper J, Fischbarg J, Vera JC, Wang D, De Vivo DC. GLUT1-deficiency: barbiturates potentiate haploinsufficiency in vitro. Pediatr Res. 1999;46:677-683. 110. Klepper J, Florcken A, Fischbarg J, Voit T. Effects of anticonvulsants on GLUT1-mediated glucose transport in GLUT1 deficiency syndrome in vitro. Eur J Pediatr. 2003;162:84-89. 111. Ho YY, Yang H, Klepper J, Fischbarg J, Wang D, De Vivo DC. Glucose transporter type 1 deficiency syndrome (Glut1DS): methylxanthines potentiate GLUT1 haploinsufficiency in vitro. Pediatr Res. 2001;50:254-260. 112. Krauss SW, Diamond I, Gordon AS. Selective inhibition by ethanol of the type 1 facilitative glucose transporter (GLUT1). Mol Pharmacol. 1994;45:1281-1286.

THE EPILEPSIES 3

12

Rasmussens Encephalitis
TIZIANA GRANATA  CARLO ANTOZZI

Introduction Clinical Features in Typical RE Epilepsy and EEG Features Neurologic Symptoms Other Than Epilepsy Atypical RE Adolescent- and Adult-Onset RE RE Protracted Variants RE Associated with Other Diseases (Double Pathology) Bilateral RE

Pathogenesis of RE Laboratory Findings Other Than EEG Neuroimaging Blood and CSF Tests Brain Biopsy Diagnostic Clues and Differential Diagnosis Treatment Immunomodulatory Treatments Antiepileptic drugs Surgery

Introduction
Rasmussens encephalitis (RE) is an acquired progressive unihemispheric disease characterized by intractable focal seizures, often in the form of epilepsia partialis continua (EPC) with motor and cognitive deterioration. Neuroimaging shows the progressive damage of the affected hemisphere, and histopathology is consistent with a T-cell dominated encephalitis with activated microglial cells and reactive astrogliosis.1 The etiopathogenesis of RE is still not fully understood; nevertheless, it has been considered a chronic inflammatory disease since its original description in 1958.2 This definition stems from the features of its clinical course, histopathology, and the reported efficacy of immunomodulatory treatments in delaying disease progression.

Clinical Features in Typical RE


The disease is sporadic, affecting both males and females. RE typically starts in childhood or early adolescence, with a mean age of presentation at 6 years. The previous personal history is uneventful, but in about half of the patients, a trivial febrile illness in the months preceding the onset of seizures or a remote head trauma have been reported.

161

162

THE EPILEPSIES 3

In its typical form, the onset of RE is marked by epileptic seizures, either focal or secondary-generalized. In most patients, seizures are isolated, but in a proportion of cases, the presenting symptom is status epilepticus (SE) or, more rarely, EPC.34 In a few patients, slowly progressive hemiparesis may precede seizure onset.45 A rarely reported presenting symptom is hemidystonia or hemiathetosis;6 this modality of presentation is not surprising, given the early damage of the basal ganglia in RE (see later discussion), and mild movement disorders are likely to be frequently overlooked or underreported, as suggested by Andermann.7 EPILEPSY AND EEG FEATURES The main characteristic of seizures in RE is their polymorphism in a given patient. Besides simple motor seizures that are almost always present, the patient may experience postural, versive, somatosensory, autonomic, visual, auditory, and limbic seizures (Figure 12-1). This polymorphism, which may be evident from the onset and is invariably present during the disease course, can be due to the multifocal albeit hemisphericorigin of the seizures, or to the progressive enlargement of the original epileptic zone. Seizures are refractory to antiepileptic (AE) treatment, their frequency usually increases rapidly, and partial SE may recur.34 EPC, defined as spontaneous regular or irregular clonic twitching confined to one part of the body (Figure 12-2), aggravated by action or sensory stimuli, and persisting during sleep,8 occurs in most of the patients (56 to 92% according to different series) at some time during the disease course, unresponsive to AE drugs.34,9 The striking EEG feature at onset is the presence of slow focal activity mainly involving the temporal and central leads on the affected hemisphere, associated with few epileptic abnormalities on the same regions. Additional distinctive features, in the first few months, are the early evidence of ictal and interictal hemispheric multifocality, the presence of subclinical ictal discharges, and the unilateral impoverishment of the background activity and of sleep spindles (Figure 12-2).4,10 During the disease course, the background activity further flattens, and sleep organization deteriorates further; epileptic and slow activity tend to increase and spread within the affected hemisphere and the unaffected one. With time, the contralateral abnormal activity, either resulting from diffusion or seemingly asynchronous, may become more frequent than that recorded in the affected hemisphere (i.e., false lateralization).11 In any case, the seizure onset, albeit multifocal, is closely unilateral; the recording of seizure from the supposed unaffected hemisphere strongly questions the diagnosis of RE. EPC, like in other conditions, is often not clearly related to electroencephalogram (EEG) changes recorded by scalp electrodes; however, auxiliary neurophysiologic techniques may contribute to define the cortical origin of the jerks. Back-averaging may identify the spikes preceding the motor phenomena, and somatosensory-evoked potentials of the rolandic cortex are often abnormally enlarged.12 NEUROLOGIC SYMPTOMS OTHER THAN EPILEPSY Hemiparesis invariably develops during the disease course. As already mentioned, the motor deficit may be the symptom of onset, and exceptionally remains the one and only manifestation of RE.5 In the first stages of the disease, hemiparesis may be limited to the postictal phase, but rapidly becomes constant, albeit fluctuating in

12  Rasmussens Encephalitis

163

Fp2-F4 F4-C4 C4-P4 P4-O2 T4-T6 Fp1-F3 F3-C3 C3-P3 P3-O1 T3-T5 Fx-Cz Cz-Pz Orb.oris Orb.oculi

A
Fp2-F4 F4-C4 C4-P4 P4-O2 T4-T6 T6-O2 Fp1-F3 F3-C3 C3-P3 P3-O1 T3-T5 T5-O1 Fz-Cz Cz-Pz

B
1 sec

100 V

Figure 121 Left hemisphere RE. EEG recorded from the patient in two consecutive days, showing the polymorphic seizure activity: clonic jerks involving the right side of the face are associated with focal epileptic activity over the left central leads (A). A brief versive seizure is associated with an ictal discharge over the left posterior region. (EMG1: right orbicularis oris; EMG2: right orbicularis oculi)

164

THE EPILEPSIES 3

Fp2-F4 F4-C4 C4-P4 P4-O2 T4-T6 Fp1-F3 F3-C3 C3-P3 P3-O1 T3-T5 Fx-Cz Cz-Pz R tib ant R g. med

A
Fp2-F4 F4-C4 C4-P4 P4-O2 T4-T6 Fp1-F3 F3-C3 C3-P3 P3-O1 T3-T5 Fx-Cz Cz-Pz R tib ant R g. med

B
100 V 1 sec

Figure 122 Wake (A) and sleep (B) EEG recorded from a 5-year-old patient, 4 months from the
first seizure. EPC involving the right lower limb is shown. The rhythmic myoclonic jerks are continuous while awake and persist, albeit at lower frequency, during sleep. Note the asymmetry of EEG tracing: rhythmic background activity during wakefulness and sleep spindles is evident over the right hemisphere. On the left hemisphere the EEG is dominated by slow activity and continuous epileptic discharges over the central region.

12  Rasmussens Encephalitis

165

severity as it worsens with increasing seizure activity. With time, hemiparesis, sometimes associated with a dystonic component, stabilizes. In an Italian series of 12 patients, focal motor deficits appeared from 15 days to 24 months after the first seizure and invariably worsened to severe hemiparesis, leading three patients to be wheelchair bound within 3 years of onset.4 Movement disorders, which seldom mark the onset of RE, are rarely reported during the disease course, probably because they are misinterpreted or overlooked in patients whose clinical picture is dominated by epilepsy and hemiparesis. Additional neurologic symptoms include hemianopia, cortical sensory loss, and aphasia when the dominant hemisphere is affected.13 Cognitive impairment is another constant feature of RE, and as for motor deficits, its appearance may be subtle. Behavioral changes, with irritability, emotional lability, or hyperactivity, often herald the first signs of mental decline, which consist mainly of memory and attention disorders and learning difficulties. In the Italian series, these symptoms were detected within 4 to 36 months after the first seizure and progressively worsened until surgery. At the time of surgery, performed between 7 months and 14 years after disease onset, the mean IQ was 61.3 15.1 (range 44 87).4 In most patients, the progression of mental impairment seems to correlate with the severity of epilepsy and particularly with the bilateral spreading of EEG epileptic abnormalities (personal experience), or with the appearance of asynchronous contralateral foci.14 The natural history of RE is summarized in the seminal paper by Oguni from the Montreal Neurological Institute3 and in the more recent report from the Bonn group, correlating the clinical course with the development of brain damage as documented by serial MRIs.9 Both groups recognize three stages that are virtually comparable. In the prodromal stage, lasting from 0 months to 8 years, seizures manifest at low frequency, and rarely, mild hemiparesis may be present. The acute stage, which in a significant number of cases appears to be the initial clinical manifestation, lasts 4 to 8 months and is characterized by frequent seizures, EPC (not always present), and rapid neurological deterioration. Finally, the patient enters the residual stage, with stable neurological deficits and persisting, albeit less frequent, seizures.

Atypical RE
Under this heading, the less-common manifestations of RE, with regard to age at onset, association with other diseases, and bilateral involvement, are included. ADOLESCENT- AND ADULT-ONSET RE A limited, but increasing number of patients with adult-onset RE have been reported, accounting for about 10% of all RE reported cases.12,15 The overall features in childhood and adult onset are similar with regard to clinical, electrophysiological, and neuroimaging findings. The main differences consist of a more frequent posterior onset and a milder, more protracted, clinical course. Motor and mental deterioration are less severe than in typical childhood RE, hemispheric damage less pronounced, and quality of life somewhat preserved. However, in a number of cases, late-onset RE runs a malignant course, comparable to that of childhood-onset

166

THE EPILEPSIES 3

disease. In a recent paper Villani and coworkers identified two distinct patterns of disease presentation, one characterized by focal motor epilepsy (the epileptic phenotype), and the other by focal cortical myoclonus (the myoclonic phenotype). Unilateral neurologic deficits and brain atrophy were progressive in both phenotypes, but they were more prominent and detected earlier in the epileptic phenotype.12 RE PROTRACTED VARIANTS A few RE patients are reported to have an acceptable seizure frequency and not invalidating motor deficits.15 This was already described in the pioneering study by Aguilar and Rasmussen,16 who reported that some patients may have mild neurologic deficits, despite the neuropathological evidence of active encephalitis. RE ASSOCIATED WITH OTHER DISEASES (DOUBLE PATHOLOGY) Patients affected by preexisting brain lesions account for about 10% of RE cases.1 This dual pathology has been documented in low-grade tumors, cortical dysplasia, tuberous sclerosis, vascular malformations, or old ischemic lesions. The association of structural abnormalities and inflammation is still a matter of investigation. The hypothesis that structural or acquired brain lesions, and the resulting epilepsy, may alter the bloodbrain barrier permeability, allowing the entry of compounds with immunogenic or inflammatory potential, is conceivable but, at least to date, not demonstrated.17 The association of RE with definite autoimmune diseases, such as SLE,18 linear scleroderma,19 ParryRomberg syndrome,20 although rarely reported, is noteworthy, as it provides a further hint for the role of immune-mediated mechanisms in the pathogenesis of RE. BILATERAL RE The hallmark of RE is the unilateral brain damage, and this distinctive and unique characteristic should be, in the opinion of the authors, mandatory for diagnosis, as it entails the surgical indication, as discussed in the treatment section. The term bilateral RE has been used in the literature to include two different conditions: (1) the secondary involvement of the unaffected hemisphere, as indicated by spreading of focal seizures from the original focus, by the appearance of interictal epileptic abnormalities, and by the mild brain atrophy, that most probably results from Wallerian degeneration of commissure fibers;21 (2) a familiar22 and few sporadic cases of early-onset malignant epilepsy with neuropathology consistent with chronic encephalitis resembling RE;2324 and (3) rare adult-onset cases.25 The frequently observed secondary involvement of the unaffected hemisphere, of course, does not rule out the diagnosis of RE because in no cases is there evidence of truly contralateral disease. It should also be underlined that in RE patients surgically treated, even after a very protracted disease, by hemispherectomy (or hemispheric disconnection), no relapse of seizures from the contralateral hemisphere was ever recorded. Conversely, in bilateral RE, symptoms and, when available, MRI findings pointed to bilateral brain damage within the first months of the disease. In our opinion, from a pragmatic point of view, these cases should be kept separate from

12  Rasmussens Encephalitis

167

true RE (and probably named differently), given the different clinical picture, MRI findings, and, overall, the ensuing therapeutic approaches.

Pathogenesis of RE
The etiology of RE remains unknown, but our understanding of its pathogenesis improved considerably in recent years. The initiating event triggering RE remains unknown. A viral etiology was first proposed, based on the features of lymphocyte infiltration and microglial reaction within the brain and on similarities with known forms of viral encephalitis.2 However, no attempt of isolation of a pathogenic virus has been successful.26 The interest in RE has been renewed by histopathological and experimental studies showing the involvement of the immune system in the pathogenesis of the disease, providing evidence for both humoral and cellular factors. In 1994, Rogers and coworkers observed that rabbits immunized with a recombinant fragment of the glutamate receptor (GluR3) developed seizures and that the histopathological examination of their brain showed inflammatory changes reminiscent of those found in patients with RE. In parallel, the same authors detected antiGluR3 antibodies in three patients with RE and observed clinical improvement after plasma exchange in one of them.27 Further reports of transient clinical improvement after removal of antibodies have been subsequently published.13,2829 These preliminary findings fostered the search for such autoantibodies in larger series of patients with RE. AntiGluR3 antibodies, at first assayed by ELISA techniques, were found in serum and cerebrospinal fluid (CSF) not only in a proportion of patients with RE, but also in other forms of epilepsy, particularly in the catastrophic epilepsy of children. Therefore, their presence is not specific for RE but seems to correlate with epilepsy and its severity.3031 However, the importance of antiGluR3 antibodies has been recently questioned by means of different laboratory techniques (ELISA, Western blot, immunoprecipitation, immunohistochemistry, and electrophysiology).32 Two different mechanisms by which antiGluR3 antibodies might exert their damage to the brain have been proposed. Plasma or serum of GluR3-immunized rabbits promoted the death of cortical neurons in vitro by a complement-mediated mechanism. Anti-IgG and anti-MAC antibodies were localized on neuronal cell bodies of cortex sections resected from RE patients,3335 leading to the complement activation hypothesis. Other authors reported that sera and IgG fractions of GluR3-immunized rabbits elicited rapid, reversible, and voltageindependent opening of cationic channels in cultured neurons, promoting excytotoxic damage.3637 Other antibodies have been described, such as to Munc-18 (a cytosolic protein), to NMDA-GluRe2 and anti-a7nAChR in a limited number of patients,3840 but their actual role is still to be elucidated. Nevertheless, the analysis of immunoglobulin gene rearrangement in brain samples from four RE patients provided evidence for a restricted profile.41 The possibility that the antibodies described could be secondary to neuronal damage cannot be excluded. However, negative results do not rule out the possibility that humoral immune mechanisms might be involved in RE, adding to the role of T cells. The role of T cells in the pathogenesis of RE has been emphasized by several pathological studies. The main pathological features of RE include brain inflammation dominated by T cells, microglial activation, microglial nodules, neuronal loss, and astrogliosis in the affected hemisphere (Figure 12-3).42,43 The extent of T-cell

168

THE EPILEPSIES 3

Figure 123 A, Hematoxylin and eosin-stained section showing cellular aggregates composed of inflammatory cells infiltrating the parenchyma. B, Glial fibrillary acidic protein (GFAP) staining showing an intense reactive astrogliosis in the cerebral cortex; section counterstained with hematoxylin. C, Hematoxylin and eosin-stained section showing a typical perivascular inflammatory cell cuff. Scale bar: A, 25 mm; B, C, 50 mm. (Courtesy of R. Garbelli and R. Spreafico.)

infiltration and microglial activation is inversely correlated with duration of the disease, being more prominent in the earliest stages while in the latest gliosis predominates.44 The majority of infiltrating T cells have been characterized as CD8+ cells containing granules positive for granzyme B, and part of these cells were observed in close contact with MHC class I positive neurons,45 a set of features considered evidence of a cytotoxic T-cell reaction against neuronal cells. The same authors could not detect deposits of immunoglobulins or signs of complement activation in the same brain specimens. Recently, the importance of astrocytes has been investigated in detail in RE.46 In RE, astrocytic apoptosis and loss were detected both in the cortex and the white matter; interestingly, as already observed by previous studies on T cells, granzyme B+ lymphocytes were found in close contact with astrocytes with granules polarized toward the astrocytic membranes. The authors therefore suggested that astrocytes might be a target for cytotoxic T cells, ultimately leading to astrocytic degeneration. Considering their multiple functional roles, degeneration of astrocytes might enhance neuronal loss as well as contribute to seizure induction and maintenance. Several aspects of the pathogenesis of the disease remain open to investigation. Evidence for the presence of cytotoxic T cells and their reactivity against neuronal and microglial cells is robust, but the question of their specificity and the peculiarity of unihemispheric involvement remain unanswered. On the other side, evidence for pathogenic autoantibodies has been suggested at first and then questioned by more recent studies. Nevertheless, the potential contribution of humoral immunity has not been definitely ruled out in RE, at least in a subset of patients. Despite the open controversies, immunopathogenetic data have been and remain the rationale for the different immunotherapeutic approaches investigated in recent years and still used in the treatment of RE.

Laboratory Findings Other Than EEG


NEUROIMAGING The refinement of imaging techniques has highly contributed to the recognition of RE, even in the early stages of the disease, and better characterization of the type and sequence of brain pathologic changes (Figure 12-4).

12  Rasmussens Encephalitis

169

Figure 124 MRI showing the progression of the disease in a patient with right-sided RE. Axial T2w images taken at 10 days (A), 9 months (B), 12 months (C), and 18 months (D) after the onset show the progression of cortical atrophy and ventricular enlargement, spreading of abnormal signal and progression of atrophy of the basal ganglia. The disease mainly progressed during the first 18 months, although further damage occurred in the following years, as shown in axial T-2w (E) and FLAIR (F) images taken 8 years after the disease onset. (Reprinted with permission from Chiapparini L, Granata T, Farina L, et al. Diagnostic imaging in 13 cases of Rasmussens encephalitis: can early MRI suggest the diagnosis? Neuroradiology. 2003;45:171183.)

Recent studies focused on the early signs of the disease and demonstrated that within the first few months, most patients disclose a combination of changes that, albeit not diagnostic, are typical enough to raise the suspicion of RE.4,47 These early signs include, (1) mild focal cortical atrophy, always involving the insular and periinsular regions; (2) ipsilateral ventricle enlargement; (3) increased cortical and/or subcortical T2 and FLAIR signal; and (4) T2 hyperintensity or atrophy of the head of the caudate nucleus. The described changes are preceded, as evaluated in the few cases with a very early examination, by focal, transient, cortical swelling. Gadolinium enhancement is not observed. During the progression of the disease, unilateral cortical and caudate atrophy progressively worsens, whereas signal abnormalities are less evident. In a study comparing MRI and histopathology findings, Bien and coworkers demonstrated that increased MRI signal correlates with an active inflammatory process, as witnessed by the high density of inflammatory cells and reactive astrocytes (i.e., with the presence of T cells, microglial nodules, and GFAP-positive astrocytes). Conversely, MRI atrophic areas without signal abnormalities correlate with tissue loss, which mostly occurs during the first year after the acute disease stage.44 The sovratentorial tissue loss is accompanied by atrophy of the contralateral, but seldom global, cerebellar hemisphere of the brainstem and thinning of the corpus callosum.47

170

THE EPILEPSIES 3

The role of functional imaging, PET, SPECT, and MR spectroscopy (MRS) is still under study. These techniques may demonstrate, even in the very early stages, areas of abnormal metabolism (PET), or perfusion (SPECT), and reduced N-acetylaspartate (NAA), but are not useful in defining the suspected inflammatory nature of the disease.4849 Moreover, PET and SPECT findings are highly influenced by seizure activity; MRS may show increased lactate, also resulting from seizure activity, potentially misleading the diagnosis. For all these reasons, data obtained by these techniques must be cautiously interpreted and integrated with clinical and MRI findings. In clinical practice, the usefulness of PET, SPECT, and MRS should be limited to confirming the unilateral brain damage. BLOOD AND CSF TESTS Blood tests are usually unrevealing. The meaning of antiglutamate receptor 3 (GluR3) antibodies and other autoantibodies has been discussed in the pathogenesis section. It is only worth stressing here that the presence or absence of these antibodies should not be considered either a clue to diagnosis, or an exclusion criteria because it is now well established that GluR3 antibodies do not discriminate between RE and other epilepsies. CSF may be normal or show nonspecific abnormalities, such as a mild increase in the white cell count and proteins. Oligoclonal bands are found in about half the patients.4,50 In summary, blood and CSF analysis are not suitable in confirming RE, but must be carried out to exclude CNS infections or other disorders. BRAIN BIOPSY Histopathology study is not required in all RE cases because other criteria allow the correct diagnosis, even in early stages, in most cases (see following discussion). It has to be considered, moreover, that histology shows nonspecific patchy chronic inflammatory changes, and that normal and abnormal tissue elements may be located in very close apposition.4243 Therefore, false-negative results may be obtained, particularly if small tissue samples are taken. In the few patients in which biopsy is required (Table 12-1), an open biopsy, in a noneloquent area where there is increased T2/FLAIR signal on MRI, should include meninges, gray, and white matter.44

Diagnostic Clues and Differential Diagnosis


The diagnosis of RE rests on clinical, neuroradiological, and electrophysiological findings that, taken together, must strongly support the hypothesis of a strictly unilateral and progressive brain disorder. A rigorous diagnostic workup is needed to exclude alternative conditions and avoid delayed, but also precipitous, therapeutic decisions. In patients with a long-lasting disease, the diagnosis is rather simple, and differential diagnosis is limited to a few conditions; conversely, the recognition of RE in its early stages, before progressive neurologic deficits and brain atrophy have developed, can be a challenge. In the last years, efforts have been made to identify the early manifestations of RE that could prompt a reasonably secure diagnosis and, consequently, an early targeted treatment.4,9 Diagnostic criteria including typical and

12  Rasmussens Encephalitis

171

TABLE 121

Diagnostic Criteria for RE by a Two-step Approach

18 step: Check for the features of Part A. 28 step: If features of Part 1 are not fulfilled, check criteria of Part B. Part A: 1. Clinical Focal seizures (with or without epilepsia partialis continua) and unilateral cortical deficit(s) 2. EEG Unihemispheric slowing with or without epileptiform activity and unilateral seizure onset 3. MRI Unihemispheric focal cortical atrophy and at least one of the following: Grey or white matter T2/FLAIR hyperintense signal Hyperintense signal or atrophy of the ipsilateral caudate head Part B: 1. Clinical Epilepsia partialis continua or progressive* unilateral cortical deficit(s) 2. MRI Progressive* unihemispheric focal cortical atrophy 3. Histopathology T-cell dominated encephalitis with activated microglial cells (typically, but not necessarily forming nodules) and reactive astrogliosis. RE can be diagnosed in presence of: all the three criteria of Part A or two out of the three criteria of Part B.
(Modified from Bien CG, Granata T, Antozzi C, et al. Pathogenesis, diagnosis and treatment of Rasmussen encephalitis. A European consensus statement. Brain. 2005;128: 454-471.) *Clinical progression: at least two sequential clinical evaluation must document a neurological deficit that must increase over the time. MRI progression: at least two sequential MRI studies must show hemiatrophy, that must increase over time.

atypical RE have been recently reviewed by a European consensus (Table 12-1); however, it should be underlined that the correct diagnosis of RE needs a wide clinical experience supported by multidisciplinary expertise. The proposed clinical criteria include a two-step diagnostic approach: (1) if all the highly characteristic clinical, EEG, and imaging criteria are fully met, the diagnosis can be established without the need for brain biopsy. In these patients the follow-up period, to ascertain the progressive course, may be very brief, or even unnecessary. (2) On the contrary, if the aforementioned criteria are not fulfilled, the progression of clinical deficits, assessed by serial evaluations, must be associated with progressive hemispheric atrophy, as documented by sequential MRIs or, alternatively, by the histopathologically proved T-cellmediated encephalitis, with activated microglial cells and astrogliosis. The diagnostic workup must exclude, on a clinical and/or laboratory basis, conditions characterized by unilateral neurologic syndromes or EPC, as well as inflammatory diseases mimicking RE. Among these conditions, the following must be considered: (1) focal or hemispheric dysplasia, neoplasia, and neurocutaneous syndromes (in particular Sturge-Weber syndrome), stroke, and hemiconvulsion-hemiplegia-epilepsy syndrome; (2) progressive neurometabolic or degenerative syndromes, namely MELAS and other mitochondriopathies, lipofuscinoses, and Alpers disease; (3) metabolic disorders, including type 1 diabetes mellitus, associated with anti-GAD

172

THE EPILEPSIES 3

65-antibodies, and renal or hepatic encephalopathy; (4) cerebral vasculitis in systemic diseases, such as lupus erythematosus or Derrys vasculitis; (5) rare infective disorders such as cat-scratch disease, but also HIV infection; and (6) finally, there are a few reports of severe epilepsy and EPC in bone marrowtransplanted patients; this possibility must be kept in mind giving the increasing indications to this treatment. A summary of differential diagnosis with related clinical and laboratory investigations are summarized in the mentioned European consensus statement.1

Treatment
RE is an inflammatory and most probably immunomediated disease, running a progressive course, characterized by severe epilepsy and worsening motor and mental deficits. These features must be kept in mind in the therapeutic decision process. The management of RE, at least theoretically, should therefore include the treatment of both seizures and the underlying inflammation. In addition, therapeutic efficacy must be assessed on both seizure reduction and neurologic improvement and weighted against side effects and potential residual deficits. Beside surgery, which is unanimously considered the only treatment able to stop seizures and disease progression (see following discussion), different drug treatments have been employed, based on the proposed etiopathogenesis of RE. Antiviral and interferon treatment was reported in a few papers in the nineties, but the paucity of data does not allow any conclusions on their effect. On the contrary, an increasing number of reports on the use of immunomodulatory treatments have accumulated in the last decade, indicating their potential role in selected patients and circumstances. The mechanisms by which these treatments may act in RE are probably complex and include, besides the antiinflammatory and immunomodulating effects, the regulation of BBB permeability (i.e., steroids) and a direct antiepileptic effect (steroids and IVIG).17 IMMUNOMODULATORY TREATMENTS The rarity of disease, the variability of disease severity, the fluctuating course, and differences in treatment schedules employed by different centers, together with the lack of controlled clinical trials due to ethical reasons, make the evaluation of treatment efficacy very arduous. Nevertheless, case reports accumulated in the last years allow the evaluation of different treatments on the basis of which prospective studies have started. As a general statement, in the vast majority of cases, as per literature reports and personal experience, immunomodulation has only a transient and partial effect on seizure activity and progression of symptoms and should therefore be limited to patients that are not, or not yet, suitable for surgery. These include late-onset (adolescent or adult) RE with slower and milder course than the typical childhood-onset form; patients with dominant hemisphere involvement and slow progression when worsening of motor and language functions resulting from surgery are not accepted from the patient and family; suspected RE in which neurologic deterioration and hemispheric atrophy have not yet occurred; and proven or suspected bilateral RE. Nevertheless, the reports of transient efficacy of immunomodulation in some of these patients are in keeping with pathogenetic data suggesting an inflammatory brain process. In some of these patients, immunomodulation with

12  Rasmussens Encephalitis

173

corticosteroids, immunosuppressants, and IVIG or plasma treatment may be of help in the management of acute deterioration of the clinical picture or in slowing down its progression. Corticosteroids are the most widely used, and in our personal experience the most effective therapy, not only as chronic treatment, but also in halting epileptic status or reducing the intensity of EPC. In the latter condition, methylprednisolone should be used in high-dose pulses (i.e., 400 mg/m2/day or, in children, 20 mg/kg/ day tapered over 10 to 12 days).5051 Some evidence exists that steroid treatment is more effective when started close to the onset of the disease.5152 As far as long-term steroid therapy is concerned, it has been recommended to start with boluses of intravenous methylprednisolone and then shift to 1 to 2 mg/kg/day of oral prednisone,5051 to be slowly tapered to the minimal effective dosage. IVIg efficacy has been reported in a limited proportion of children and adults with RE.5355 The recommended dosing scheme consists of five consecutive infusions of 0.4 g/kg/day followed by a monthly dose of 0.4 to 2.0 g/kg distributed over 1 to 5 consecutive days; the assessment of IVIG efficacy may require as long as six courses. The combination of steroids and IVIg may also be considered5051 when the two treatments alone are ineffective. Immunomodulation by Plasmapheresis (PE) or Protein A Immunoadsorption (PAI) Plasmatic treatment has been used with the rationale of removing potentially pathogenetic circulating antibodies (namely anti GluR3). A dramatic effect in blocking status epilepticus and neurologic deterioration has been reported in a few patients,28,51 whereas evidence of long-term efficacy has been only exceptionally reported.13 For this reason, as well as for technical issues (particularly in children) and costs, PEX or PAI should be reserved to phases of acute deterioration, or to assess the residual motor and mental abilities before surgery. Immunosuppression Tacrolimus The only trial so far reported on the use of immunosuppressive agents is that of Bien and coworkers.56 The authors tested the effect of tacrolimus, a T-cell inhibiting drug, in seven RE patients and compared their outcome with that of twelve historical untreated patients. Despite no significant effect on seizure frequency, tacrolimustreated patients had a better outcome in terms of motor and mental functions, as well as on the rate of brain atrophy, to the point that none of these patients was submitted to brain surgery. Conflicting or limited results, hindering any conclusion, have been reported for cyclophosphamide and azathioprine.28,51,57 ANTIEPILEPTIC DRUGS It is common experience that conventional antiepileptic drugs, including the more recent compounds, have only partial effect on seizures, and no mono- or polytherapy has been reported to be superior to others.58 AEDs should be therefore chosen on the basis of the empirical demonstration of efficacy and tolerability, avoiding heavy politherapy to limit drug interactions and potential side effects.

174

THE EPILEPSIES 3

High-dose phenobarbital remains, in our personal experience, one of the more effective treatments in reducing seizure frequency and the intensity of EPC. SURGERY RE was first described in a patient surgically treated for intractable focal seizures and hemiparesis. Fifty years later, despite the advances in understanding the pathogenesis of RE and attempts to find rational medical therapies, surgery, and namely the surgical exclusion of the affected hemisphere, still remains inevitable in almost all patients, even in those who experienced transient benefit from immunomodulation.5152 Hemispherectomy, and in the more recent years hemispherotomy (consisting in deafferentation of the hemisphere with very limited removal of tissue), is effective in halting seizures, as well as motor and mental deterioration, in over 80% of patients. Following surgery, most patients recover, at least in part, from neurologic decline, with significant improvement in both patients and family quality of life.5962 The awareness of the inevitable progression of the disease and the known positive outcome after hemispherectomy should theoretically lead to surgery in virtually all patients once the diagnosis is established. However, when facing each single patient, concerns always arise not only in the family but also in the referring physician. If the patient is a young child, the procedure can be easily proposed and accepted given the high potential of brain plasticity that will limit surgery sequelae, even in patients with involvement of the dominant hemisphere (usually in children younger than 6 years). By contrast, reluctance is almost a rule in school-aged children, adolescents, and adults, particularly in the early stages and when the dominant hemisphere is affected. In these cases, the expected surgically induced hemianopia, hemiparesis (albeit walking is preserved), loss of fine motor skills, and aphasia, make the decision on indication and timing of surgery quite difficult. The proposal of surgery in these patients requires experience, adequate follow-up, multidisciplinary discussions of pros and cons, and the psychological preparation of the patient and family. REFERENCES
1. Bien CG, Granata T, Antozzi C, et al. Pathogenesis, diagnosis and treatment of Rasmussen encephalitis. A European consensus statement. Brain. 2005;128:454-471. 2. Rasmussen T, Olszewski L, Lloyd-Smith D. Focal seizures due to chronic localized encephalitis. Neurology. 1958;8:435-445. 3. Oguni H, Andermann F, Rasmussen TB. The natural history of the syndrome of chronic encephalitis and epilepsy: a study of the MNI series of forty-eight cases. In: Andermann F, ed. Chronic Encephalitis and Epilepsy: Rasmussens Syndrome. Boston: Butterworth-Heinemann; 1991:7-35. 4. Granata T, Gobbi G, Spreafico R, et al. Rasmussens encephalitis: early characteristics allow diagnosis. Neurology. 2003;60:422-425. 5. Bien CG, Elger CE, Leitner Y, et al. Slowly progressive hemiparesis in childhood as a consequence of Rasmussen encephalitis without or with delayed-onset seizure. Eur J Neurol. 2007;14:387-390. 6. Frucht S. Dystonia, athetosis, and epilepsia partialis continua in a patient with late-onset Rasmussens encephalitis. Mov Disord. 2002;17:609-612. 7. Andermann F. Rasmussen syndrome and movement disorder. Mov Disord. 2002;17:437-438. 8. Bancaud J, Bonis A, Trottier S, Talairach J, Dulac O. Continuous partial epilepsy: syndrome and disease. Rev Neurol. 1982;138:803-814. 9. Bien CG, Widman G, Urbach H, et al. The natural history of Rasmussens encephalitis. Brain. 2002;125:1751-1759. 10. So N, Gloor P. Electroencephalographic and electrocorticographic findings in chronic encephalitis of the Rasmussen type. In: Andermann F, ed. Chronic Encephalitis and Epilepsy: Rasmussens Syndrome. Boston: Butterworth-Heinemann; 1991:37-45.

12  Rasmussens Encephalitis 11. Andrews PI, McNamara JO, Lewis DV. Clinical and electroencephalographic correlates in Rasmussens encephalitis. Epilepsia. 1997;38:189-194. 12. Villani F, Pincherle A, Antozzi C, et al. Adult-onset Rasmussens encephalitis: anatomicalelectrographic-clinical features of 7 Italian cases. Epilepsia. 2006;47(S5):41-46. 13. Antozzi C, Granata T, Aurisano N, et al. Long-term selective IgG immunoadsorption improves Rasmussens encephalitis. Neurology. 1998;51:302-305. 14. Longaretti F, Dunkley C, Varadkar S, Boyd S, Cross JH. Early EEG findings in children with Rasmussens encephalitis versus focal cortical dysplasia. Dev Med Child Neurol. 2006;48:24. 15. Hart YM, Andermann F, Fish DR, et al. Chronic encephalitis and epilepsy in adults and adolescents: a variant of Rasmussens syndrome? Neurology. 1997;48:418-424. 16. Aguilar MJ, Rasmussen T. Role of encephalitis in pathogenesis of epilepsy. Arch Neurol. 1960;2:663-676. 17. Vezzani A, Granata T. Brain inflammation in epilepsy: experimental and clinical evidence. Epilepsia. 2005;46:1724-1743. 18. Lascelles K, Dean AF, Robinson RO. Rasmussens encephalitis followed by lupus erythematosus. Dev Med Child Neurol. 2002;44:572-574. 19. Pupillo G, Andermann F, Dubeau F. Linear scleroderma and intractable epilepsy: neuropathologic evidence for a chronic inflammatory process. Ann Neurol. 1996;39:277-278. 20. Shah JR, Juhasz C, Kupsky WJ, et al. Rasmussen encephalitis associated with Parry-Romberg syndrome. Neurology. 2003;61:395-397. 21. Larionov S, Ko nig R, Urbach H, Sassen R, Elger CE, Bien CG. MRI brain volumetry in Rasmussen encephalitis: the fate of affected and unaffected hemispheres. Neurology. 2005;64: 885-887. 22. Silver K, Andermann F, Meagher-Villemure K. Familial epilepsia partialis continua with chronic encephalitis: another variant of Rasmussen syndrome? Arch Neurol. 1988;55:733-735. 23. Tobias SM, Robitaille Y, Hickey WF, Rhodes CH, Nordgren R, Andermann F. Bilateral Rasmussen encephalitis: postmortem documentation in a five-year-old. Epilepsia. 2003;44:127-130. 24. Andermann F, Farrel K. Early onset Rasmussen syndrome: a malignant, often bilateral form of the disorder. Epilepsy Res. 2006;70(suppl 1):S259-S262. 25. McLachlan RS, Girvin JP, Blume WT, Reichman H. Rasmussens chronic encephalitis in adults. Arch Neurol. 1993;50:269-274. 26. Atkins MR, Terrell W, Hulette CM. Rasmussens syndrome: a study of potential viral etiology. Clin Neuropathol. 1995;14:7-12. 27. Rogers SW, Andrews PI, Gahring LC, et al. Autoantibodies to glutamate receptor GluR3 in Rasmussens encephalitis. Science. 1994;265:648-651. 28. Andrews PI, Dichter MA, Berkovic SF, Newton MR, McNamara JO. Plasmapheresis in Rasmussens encephalitis. Neurology. 1996;46:242-246. 29. Palcoux JB, Carla H, Tardieu M, et al. Plasma exchange in Rasmussens encephalitis. Ther Apher. 1997;1:79-82. 30. Wiendl H, Bien CG, Bernasconi P, et al. GluR3 antibodies: prevalence in focal epilepsy but no specificity for Rasmussens encephalitis. Neurology. 2001;57:1511-1514. 31. Mantegazza R, Bernasconi P, Baggi F, et al. Antibodies against GluR3 peptides are not specific for Rasmussens encephalitis but are also present in epilepsy patients with severe, early onset disease and intractable seizures. J Neuroimmunol. 2002;131:179-185. 32. Watson R, Jiang Y, Bermudez I, et al. Absence of antibodies to glutamate receptor type 3 (GluR3) in Rasmussen encephalitis. Neurology. 2004;63:43-50. 33. He XP, Patel M, Whitney KD, Janumpalli S, Tenner A, McNamara JO. Glutamate receptor GluR3 antibodies and death of cortical cells. Neuron. 1998;20:153-163. 34. Whitney KD, Andrews PI, McNamara JO. Immunoglobulin G and complement immunoreactivity in the cerebral cortex of patients with Rasmussens encephalitis. Neurology. 1999;53:699-708. 35. Whitney KD, McNamara JO. GluR3 autoantibodies destroy neural cells in a complement-dependent manner modulated by complement regulatory proteins. J Neurosci. 2000;20:7307-7316. 36. Twyman RE, Gahring LC, Spiess J, Rogers SW. Glutamate receptor antibodies activate a subset of receptors and reveal an agonist binding site. Neuron. 1995;14:755-762. 37. Levite M, Fleidervish IA, Schwarz A, Pelled D, Futerman AH. Autoantibodies to the glutamate receptor kill neurons via activation of the receptor ion channel. J Autoimmun. 1999;13: 61-72. 38. Yang R, Puranam RS, Butler LS, et al. Autoimmunity to munc-18 in Rasmussens encephalitis. Neuron. 2000;28:375-383.

175

176

THE EPILEPSIES 3 39. Takahashi Y, Mori H, Mishina M, et al. Autoantibodies and cell-mediated autoimmunity to NMDAtype GluRepsilon2 in patients with Rasmussens encephalitis and chronic progressive epilepsia partialis continua. Epilepsia. 2005;46(S5):152-158. 40. Watson R, Jepson JE, Bermudez I, et al. Alpha7-acetylcholine receptor antibodies in two patients with Rasmussen encephalitis. Neurology. 2005;65:1802-1804. 41. Baranzini SE, Laxer K, Saketkhoo R, et al. Analysis of antibody gene rearrangement, usage, and specificity in chronic focal encephalitis. Neurology. 2002;58:709-716. 42. Robitaille Y. Neuropathologic aspects of chronic encephalitis. In: Andermann F, ed. Chronic Encephalitis and Epilepsy: Rasmussens Syndrome. Boston: Butterworth-Heinemann; 1991:79-110. 43. Pardo CA, Vining EP, Guo L, Skolasky RL, Carson BS, Freeman JM. The pathology of Rasmussen syndrome: stages of cortical involvement and neuropathological studies in 45 hemispherectomies. Epilepsia. 2004;45:516-526. 44. Bien CG, Urbach H, Deckert M, et al. Diagnosis and staging of Rasmussens encephalitis by serial MRI and histopathology. Neurology. 2002;58:250-257. 45. Bien CG, Bauer J, Deckwerth TL, et al. Destruction of neurons by cytotoxic T cells: a new pathogenic mechanism in Rasmussens encephalitis. Ann Neurol. 2002;51:311-318. 46. Bauer J, Elger CE, Volkmar HH, et al. Astrocytes are a specific immunological target in Rasmussens encephalitis. Ann Neurol. 2007;62:67-80. 47. Chiapparini L, Granata T, Farina L, et al. Diagnostic imaging in 13 cases of Rasmussens encephalitis: can early MRI suggest the diagnosis? Neuroradiology. 2003;45:171-183. 48. Fiorella DJ, Provenzale JM, Edward CR, Crain BJ, Al Sugair A. (18)F-fluorodeoxyglucose positron emission tomography and MR imaging findings in Rasmussen encephalitis. Am J Neuroradiol. 2001;22:1291-1299. 49. Fogarasi A, Hegyi M, Neuwirth M, et al. Comparative evaluation of concomitant structural and functional neuroimages in Rasmussens encephalitis. J Neuroimaging. 2003;13:339-345. 50. Hart YM, Cortez M, Andermann F, et al. Medical treatment of Rasmussens syndrome (chronic encephalitis and epilepsy): effect of high-dose steroids or immunoglobulins in 19 patients. Neurology. 1994;44:1030-1036. 51. Granata T, Fusco L, Gobbi G, et al. Experience with immunomodulatory treatments in Rasmussens encephalitis. Neurology. 2003;61:1807-1810. 52. Bahi-Buisson N, Villanueva V, Bulteau C, et al. Long term response to steroid therapy in Rasmussen encephalitis. Seizure. 2007;16:485-492. 53. Leach JP, Chadwick DW, Miles JB, Hart IK. Improvement in adult-onset Rasmussens encephalitis with long-term immunomodulatory therapy. Neurology. 1999;52:738-742. 54. Villani F, Spreafico R, Farina L, et al. Positive response to immunomodulatory therapy in an adult patient with Rasmussens encephalitis. Neurology. 2001;56:248-250. 55. Granata T. Rasmussens syndrome. Neurol Sci. 2003;(S4)239-243. 56. Bien CG, Gleissner U, Sassen R, Widman G, Urbach H, Elger CE. An open study of tacrolimus therapy in Rasmussens encephalitis. Neurology. 2004;62:2106-2109. 57. Krauss GL, Campbell ML, Roche KW, Huganir RL, Niedermeyer E. Chronic steroid-responsive encephalitis without autoantibodies to glutamate receptor GluR3. Neurology. 1996;46:247-249. 58. Dubeau F, Sherwin AL. Pharmacologic principles in the management of chronic focal encephalitis. In: Andermann F, ed. Chronic Encephalitis and Epilepsy: Rasmussens Syndrome. Boston: ButterworthHeinemann; 1991:179-192. 59. Villemure J-G, Andermann F, Rasmussen TB. Hemispherectomy for the treatment of epilepsy due to chronic encephalitis. In: Andermann F, ed. Chronic Encephalitis and Epilepsy: Rasmussens Syndrome. Boston: Butterworth-Heinemann; 1991:235-241. 60. Vining EP, Freeman JM, Brandt J, Carson BS, Uematsu S. Progressive unilateral encephalopathy of childhood (Rasmussens syndrome): a reappraisal. Epilepsia. 1993;34:639-650. u G, Delalande O. Persistence of ictal activity after functional 61. Thomas P, Zifkin B, Gheta hemispherectomy in Rasmussen syndrome. Neurology. 2003;60:140-142. 62. Tubbs RS, Nimje SM, Oajes WJ. Long-term follow-up in children with functional hemispherectomy for Rasmussen encephalitis. Childs Nerv Syst. 2005;21:461-465.

THE EPILEPSIES 3

13

Seizures and Epilepsy in the Elderly


ANIL MENDIRATTA  TIMOTHY A. PEDLEY

Introduction Epidemiology Etiology Acute Symptomatic Seizures Unprovoked Seizures and Epilepsy Clinical Presentation Risk of Recurrence

Treatment Considerations Pharmacokinetic Changes Pharmacodynamic Changes Choosing an AED Older Generation AEDs Newer Generation AEDs Epilepsy Surgery Epilepsy Morbidity Status Epilepticus Bone Health, Falls, and Fractures Conclusions/Recommendations

Introduction
In developed countries, demographic trends project continued increases in the number of older people in the population. In the United States, there were 36.8 million adults over the age of 65 in 2005.1 By 2030, the U.S. Department of Health and Human Services predicts that this figure will have increased to 71.5 million and account for roughly 20% of the population. In this older age group, seizures and epilepsy are the third most common neurological condition, behind only stroke and dementia. Although the primary goals of treatmentfreedom from seizures, absence of adverse drug effects, and maintenance of a high quality of lifeare the same for all patients with epilepsy, several issues specific to the elderly population need to be considered in approaching diagnosis and making treatment decisions. Diagnosis can be challenging because of the many comorbid conditions that are common in the elderly. These comorbidities, along with the medications prescribed to treat them, must be considered carefully when devising treatment strategies. Dealing with these issues will only become more challenging in the coming years, as the number of people over the age of 65 increases steadily.

177

178

THE EPILEPSIES 3

250 200 Incidence/100,000 150 100 50 0 0 20 40 Age USA Sweden Iceland


3,4,7,8

60

80

100

Figure 131 Incidence of unprovoked seizures through lifespan. (Reprinted with permission from Cloyd J, Hauser W, Towne A, et al. Epidemiological and medical aspects of epilepsy in the elderly. Epilepsy Res. 2006;68 Suppl 1:S39.)

Epidemiology
The incidence of both acute symptomatic seizures and unprovoked seizures or epilepsy is highest in people over the age of 65.24 In a British population study, nearly 25% of newly identified seizures occurred in patients 60 years of age or older.5 The incidence continues to rise with increasing age: In those older than 75, the incidence is five times that of younger adults. In the United States, about 50,000 new cases of epilepsy occur each year in this age group.6 In people over 60 years of age, the prevalence of unprovoked seizures and epilepsy is at least 1%, and it is even higher, 1.2 to 1.5%, in people over 75 (Figure 13-1). This is a prevalence rate about twice that seen in younger adults.10,11 In a study of 1, 130, 155 U.S. veterans who were at least 65 years of age, 1.8% had epilepsy.12,13 In specific populations at risk, such as nursing home residents, who have significantly higher rates of comorbid conditions associated with epilepsy, including dementia and stroke, prevalence rates as high as 3 to 9% have been reported.1416

Etiology
ACUTE SYMPTOMATIC SEIZURES Because brain injuries caused by stroke, head injury, infections, neoplasms, and metabolic disturbances (hypoglycemia, hyponatremia, uremia) are so common in the elderly, it is not surprising that the incidence of acute symptomatic seizures is likewise high in this age group.6 Cerebrovascular disease accounts for 40% to 50% of acute symptomatic seizures, metabolic disturbances for 10 to 15%, and acute head

13  Seizures and Epilepsy in the Elderly

179

Cerebrovacular Disease

Metabolic Disturbances

Toxins/Alcohol

Head Trauma Neoplasms Other CNS Infections

Figure 132 Causes of acute symptomatic seizures in the elderly.

trauma, brain infections, neoplasms and toxins/alcohol each for 5 to 10%2,6,17 (Figure 13-2). UNPROVOKED SEIZURES AND EPILEPSY The high frequency of unprovoked seizures and epilepsy in elderly persons can also be attributed in large part to the high prevalence of a history of stroke, brain tumor, and head trauma in older age groups. Even so, about 50% of cases remain cryptogenic (Figure 13-3), although this is a significantly lower percentage than that found in younger age groups, where about 70% of cases have no identifiable cause.3,17 Cerebrovascular disease is the most frequently identified antecedent of epilepsy, accounting for 30 to 40% of all new cases.6,18 Individuals with cerebrovascular disease are 20 times more likely to develop epilepsy than the general population.19 About 15% of stroke survivors will develop unprovoked seizures within the first 5 years, and the elevated risk continues for many years thereafter. Seizures may, in fact, be the first presentation of unrecognized cerebrovascular disease. This is especially true if previous strokes had occurred in clinically silent brain areas. In a review of new onset seizures in 4709 patients older than age 60, none of whom had a known history of cerebrovascular disease, dementia, tumor, or alcohol abuse, Cleary and colleagues found that the relative risk of stroke was nearly three times that of controls.20 Hypertension is also an independent risk factor for unprovoked seizures,21 although hypertension may be a surrogate for the progressive arterial changes associated with lacunes, other small strokes, and periventricular white matter lesions. Although the incidence of head trauma is highest in adolescents and young adults, a relative peak of occurrence in the elderly is largely attributable to falls. Approximately one-third of people over the age of 65 living at home, and about half of nursing home residents, have at least one fall each year.22,23 Trauma with loss of

180

THE EPILEPSIES 3

Cerebrovascular Disease

Degenerative Diseases Cryptogenic Head Trauma Neoplasms CNS Infections

Figure 133 Etiologies of epilepsy in the elderly.

consciousness is associated with a threefold increased risk of epilepsy.6 With more severe head injuries, risk is even higher. Overall, head trauma accounts for 2 to 3% of new cases of epilepsy in the elderly. Although the incidence of central nervous system infections is highest in childhood, a second peak occurs in the elderly. Survivors of central nervous system (CNS) infections have a threefold risk of developing epilepsy, and a history of infections accounts for 2 to 3% of cases.24 Alzheimers and other neurodegenerative diseases are associated with a 5- to 10fold increase in epilepsy compared to the general population.6,25 Unprovoked seizures may be seen in as many as 8 to 15% of patients with Alzheimers disease.25,26 The incidence of intracranial neoplasms increases with age, and these are strongly associated with epilepsy. However, given the progressive nature of most brain tumors, it is often difficult to distinguish acute symptomatic seizures from unprovoked seizures or epilepsy.

Clinical Presentation
As would be expected from the etiological profile, the great majority of seizures in the elderly, more than 70%, are of partial onset.18,27,28 Even generalized tonic-clonic seizures without obvious focal features are likely to have a localized or regional onset, as it is very unusual for idiopathic generalized epilepsy syndromes to present at this age. In any event, diagnosis of epilepsy in the elderly can be difficult and may be delayed. In a recent Veterans Administration cooperative study of epilepsy in elderly patients, epilepsy was an initial diagnostic consideration in 73% of patients that were eventually diagnosed with epilepsy.18 In a subset of 151 patients from this study, Spitz and colleagues found that the delay from initial symptoms to diagnosis

13  Seizures and Epilepsy in the Elderly

181

was 2.3 years, with a median time of 1 year. Only 37% of patients in this subgroup were correctly diagnosed at the time of initial evaluation. Two-thirds of patients with generalized tonic-clonic (GTC) seizures, but only one-fourth of patients with complex partial seizures, were correctly diagnosed.29 Delay in making the correct diagnosis likely derives from three characteristics of seizures in this population: 1. The manifestations of seizures in the elderly may be quite different from those of younger patients. 2. Epileptic symptoms in the elderly are often attributed inaccurately to other disorders common in this age group. 3. Adequate descriptions of the seizures are often lacking. Because the lesions associated with epilepsy in elderly people can involve any area of the brain, extratemporal neocortical epilepsies are more common than in younger adults. Consequently, seizures may manifest with a wide variety of sensory, visual, cognitive, and behavioral phenomena that are frequently atypical for the physicians experience and thus more difficult to recognize as having an epileptic basis. Classic auras are less common in the elderly, and initial manifestations of partial seizures, such as dizziness, a vague feeling related to the head, memory loss, or confusion, may be interpreted as nonspecific symptoms due to any number of possible causes. Thus, given the high frequency with which metabolic derangements, cerebrovascular events, and dementia occur in the elderly, clinical manifestation of partial seizures, as well as of postictal states, may be incorrectly attributed to these other conditions. For example, postical aphasia or hemiparesis may be diagnosed as ischemic events.30 Finally, although published data are lacking, it is said that generalized tonic-clonic seizures, the easiest type of seizure to recognize, occur with less frequency in older age groups. Falls in the elderly are common and a frequent cause of admission to a hospital.23 However, seizures are rarely considered an important etiology, especially early in the diagnostic evaluation. More often, falls are attributed to cardiovascular, cerebrovascular, or arthritic etiologies. Confusion or memory loss, which may be ictal or postical in nature, is often first considered to be a manifestation of dementia, metabolic abnormalities (e.g., dehydration) or a head injury related to a fall. Difficulty in diagnosis is compounded by the increase in social isolation in the elderly population. In 2006, the U.S. Department of Health and Human Services found that 30% of noninstitutionalized people over 65 years of age, totaling more than 10 million people, lived alone.1 As such, elderly patients are often brought to the hospital by emergency medical services without anyone available to provide information through direct observation of an episode. NONEPILEPTIC PAROXYSMAL PHENOMENA As at other ages, elderly persons can have episodic paroxysmal phenomena that mimic seizures but are nonepileptic (Table 13-1). In a study of 94 patients at least 60 years of age (mean 70 years) who were referred for video-electroencephalogram (EEG) monitoring for evaluation of paroxysmal episodes, 27 (29%) had nonepileptic events, including 13 with psychogenic seizures.31 The majority of these patients had been taking antiepileptic drugs for presumed epileptic seizures. Such findings underscore the need for video-EEG monitoring in patients whose seizures are atypical or have not responded to treatment with antiepileptic drugs (AEDs).

182

THE EPILEPSIES 3

TABLE 131

Paroxysmal Phenomena That May Mimic Seizures in Elderly Patients

Syncope Transient ischemic attacks Transient global amnesia Migraine Drop attacks Myoclonus Confusional episodes due to medication interactions or overmedication Hypoglycemia Electrolyte disturbances/dehydration REM behavior disorder Nonepileptic psychogenic seizures

Risk of Recurrence
Physicians should have a low threshold for suspecting seizures in the elderly because of the increased incidence and the higher rate of seizure recurrence in this population.3,4 In patients presenting with a first unprovoked seizure, the overall risk of recurrence with at least 2 years of follow-up is about 38% (range 25 to 52%).3235 Factors that increase the risk of recurrence include abnormal EEG findings and presence of an underlying definable etiology. Data regarding specific effects of older age on risk of recurrence are inconclusive, and there are no population-based studies. However, when seizures can be attributed to a recognized underlying neurologic condition (that is, they are symptomatic of an acquired pathogenic mechanism), risk of recurrence is roughly double that of a cryptogenic first seizure.33 In selected populations, the recurrence rate may be especially high. For example, seizures recurred in more than 80% of a small series of patients with a remote history of stroke.36 As the majority of newly diagnosed unprovoked seizures in the elderly are symptomatic (that is, they have a known antecedent cause), it is reasonable to assume that most elderly people have a high risk of seizure recurrence. Thus, in contrast to accepted treatment paradigms in younger adults, we believe that it is reasonable to begin anticonvulsant medication after the first unprovoked seizure in someone over 65, even in the absence of an underlying demonstrable lesion.

Treatment Considerations
Although it often stated that epilepsy is more easily controlled in elderly patients than in younger ones,37,38 it is not clear that this is accurate, as no well-controlled large studies specifically address this issue. Data from two recent studies looking at tolerability of anticonvulsant drugs in the elderly showed that seizure-free rates are similar to those seen in the adult population at large. In a randomized trial that

13  Seizures and Epilepsy in the Elderly

183

compared response in the elderly to treatment regimens using modest dosages of carbamazepine (dosage range 200 to 800 mg daily) and lamotrigine (dosage range 75 to 300 mg daily), Brodie and colleagues found that only 33% remained seizure free during the final 16 weeks of the study.39 Although a high rate of medication withdrawal likely contributed to the relatively low rate of seizure control, these findings suggest that seizures are, in fact, not necessarily easily controlled in this population. Similarly, in the VA cooperative study #428, which randomized elderly patients with new-onset epilepsy to treatment with gabapentin, lamotrigine, or carbamazepine, 53% of patients who remained on treatment were seizure free at 12 months.40 These data are not substantially different from seizure control rates in the general population, as illustrated by the study by Kwan and Brodie of 470 previously untreated patients with epilepsy, ranging in age from 9 to 93.41 They found that 47% of patients became seizure free on the first drug, and a total of 61% became seizure free with the second or third monotherapy agent. PHARMACOKINETIC CHANGES Significant physiological changes occur in elderly persons that affect drug pharmacokinetics, and these must be considered in determining dosing regimens. However, the degree of these changes and the extent to which they occur at a particular age are not reliably predictable due to the great degree of normal variability and the extent to which disease-related changes may be present. Absorption Absorption of drugs depends on dissolution of their particular formulations. This, in turn, is mainly related to gastric acid secretion, which often declines in the elderly.42 In addition, to varying degrees, gastric emptying slows, intestinal transit time increases, mesenteric blood flow decreases, and the intestinal absorptive surface may decrease.43 All these changes contribute to variable and often unpredictable absorption of different drugs. Overall, this combination of factors typically results in a diminished ability to absorb antiepileptic drugs, which reduces their bioavailability. These age-related changes in absorption can be compounded by frequent use of antacids, which can specifically impair absorption of phenytoin. In addition, gabapentin, which is absorbed via a saturable L-amino acid transporter system in the small intestine,44 may be particularly susceptible to physiological changes in gastrointestinal absorption. Protein Binding In healthy adults, serum albumin concentration decreases only slightly with age.45 However, in elderly patients with acute systemic and neurological illnesses, serum albumin levels can decline significantly. Suboptimal nutrition may exacerbate this. With reductions in serum albumin concentrations, the free fraction of highly protein-bound medications can increase substantially, sometimes resulting in prominent adverse effects despite little or no change in the total serum level. AEDS that are highly protein bound include tiagabine, phenytoin, valproate, diazepam, clonazepam, clobazam, and to a lesser extent, carbamazepine.46

184

THE EPILEPSIES 3

Hepatic Clearance Hepatic mass and blood flow both decline with age. As a result, liver volume is about 25% lower in people age 65 compared to young adults.47 The degree to which the various hepatic enzymes change with increasing age is not known definitively.48 The majority of drugs metabolized in the liver utilize the cytochrome P450 system. With increasing age, this system is believed to decline in function, though to a variable and unpredictable degree. The hepatic glucoronidation conjugation process is thought to be less affected by age. AEDs metabolized primarily by the cytochrome P450 system include phenytoin, phenobarbital, primidone, carbamazepine, ethosuximide, oxcarbazepine, and tiagabine. Felbamate, topiramate, valproate, and zonisamide are also partially metabolized via this pathway.46 AEDs that primarily undergo conjugation include lamotrigine, zonisamide, and valproate.46 Renal Clearance The most consistent age-related change affecting pharmacokinetics is a decline in renal function related to reduction in renal mass and loss of glomeruli. This results in lower glomerular filtration rate (GFR) and a reduced ability to handle renally excreted medications and toxins. On average, GFR declines by about 50% between the third and eighth decades of life.49 However, this is variable, and about one-third of people may not experience declines of this degree.50 Because muscle mass, the source of serum creatinine, also decreases with age, changes in serum creatinine levels frequently do not parallel the decline in GFR. The validity of common formulas used to estimate GFR has been questioned, even in otherwise healthy elderly patients.51 AEDs that are excreted primarily by the kidney include gabapentin, levetiracetam, vigabatrin, and pregabalin. Felbamate, zonisamide, and topiramate are also partially renally excreted.46 Table 13-2 summarizes the effect of older age on AED clearance. PHARMACODYNAMIC CHANGES In addition to age-related pharmokinetic changes, it is likely that actions of AEDs at the cellular level are also altered in the elderly. Although such changes have not been well characterized, the appearance of adverse central nervous system effects (e.g., drowsiness, unsteadiness) at unbound serum levels that rarely produce similar toxicity in younger adults is probably one reflection of this. As a result, it is important to anticipate a much narrower therapeutic window for AEDs in the elderly population (Figure 13-4). CHOOSING AN AED As nearly all of the currently used AEDs have similar efficacy in treating partial and secondarily generalized seizures, choice of medication depends mainly on tolerability and the presence of comorbid conditions. Nearly 90% of elderly patients in the community take at least one prescription drug,66 and nursing home residents take an average of five.67 Serum levels of AEDs that are metabolized by the cytochrome p450 enzymes are most likely to be affected, if other medications administered

13  Seizures and Epilepsy in the Elderly

185

TABLE 132

Average Changes in Apparent Oral Clearance of Older and Newer AEDs in Elderly Patients (Interindividual variation may be considerable in relation to age and other factors.)
Effect of Old Age on Drug Clearance

Drug

Carbamazepine Felbamate Gabapentin Lamotrigine Levetiracetam Oxcarbazepine Phenobarbital Phenytoin Tiagabine Topiramate Valproic acid Vigabatrin Zonisamide

Decrease Decrease Decrease Decrease Decrease Decrease Decrease Decrease Decrease Decrease Decrease Decrease No data

by by by by by by by by by by by by

2540%52 1020%53 about 3050%54 about 35%55 about 2040%56 2535%57 about 20%58 about 25%59 about 30%60 20%61 about 40%62 5085%63

(Reprinted with permission from Perucca E, Berlowitz D, Birnbaum A, et al. Pharmacological and clinical aspects of antiepileptic drug use in the elderly. Epilepsy Res. 2006;68[Suppl 1]:S49-S63.)

Therapeutic window AED Concentration

Adults

AGE

Elderly

Figure 134 Effect of age on therapeutic ranges. The elderly typically have a narrower therapeutic window, the range between the lowest effective concentration and the maximal tolerated concentration.64,65 (Reprinted with permission from Bergey GK. Initial treatment of epilepsy: special issues in treating the elderly. Neurology. 2004;63[10 Suppl 4]:S40-S48.)

186

THE EPILEPSIES 3

concurrently inhibit this pathway. Common examples include H2-blockers (cimetidine), macrolide antibiotics (erythromycin, clarithromycin), antifungal agents (ketoconazole, fluconazole), and isoniazid. Ticlopidine increases levels of carbamazepine and phenytoin. Thus, the possibility of therapeutically significant drug interactions must be a major consideration when selecting an AED. Detailed information concerning drug interactions can be found in an extensive review by Patsalos and Perucca.68

Older-Generation AEDs
Although phenytoin, phenobarbital, primidone, valproate, and carbamazepine are effective in localization-related epilepsy,69,70 tolerability concerns have generally displaced them as drugs of first choice in elderly patients. Phenobarbital and primidone have significant sedative and cognitive effects and the first VA Cooperative Trial70 showed them to be much less well tolerated than carbamazepine and phenytoin. They are also potent inducers of hepatic enzymes, and this reduces the efficacy of many drugs administered concurrently to treat comorbid conditions. Phenytoin also has troublesome features. The transition from first-order to zeroorder kinetics at therapeutic dosages often leads to widely variable serum levels, even with small changes in dosing. This greatly increases the likelihood of toxicity. Development of imbalance and ataxia at modest dosages increases the risk for falls and consequent fractures. And as already noted, medication interactions are common and often problematic. When phenytoin must be used, development of toxicity can be minimized by starting at no more than 100 mg BID and making subsequent dosage adjustments in increments of 30 mg. Drug interactions are also common with valproate because it is both a potent inhibitor of hepatic enzymes and also highly protein bound. Occasionally, encephalopathy occurs due to hyperammonemia, which can occur in the absence of hepatic enzyme abnormalities.71 Carbamazepine is another inducer of hepatic enzymes, and thus drug interactions must be anticipated. In addition, hyponatremia may be more common and less well tolerated in elderly patients taking carbamazepine.72

Newer-Generation AEDs
Two prospective trials have shown that lamotrigine and gabapentin are better tolerated than carbamazepine. In a recent VA cooperative study, patients 65 or older with newly diagnosed epilepsy were randomized to treatment with gabapentin 1500 mg per day, lamotrigine 150 mg per day, or carbamazepine 600 mg per day.40 The primary outcome measure was early termination, most often due to adverse effects, which occurred in 64.5% of patients taking carbamazepine in contrast to 51% of patients taking gabapentin and 44% of patients taking lamotrigine. Seizure-free rates and time to first seizure did not differ significantly among the three medications. An earlier study by Brodie and colleagues, in which newly diagnosed elderly patients with epilepsy were randomized to lamotrigine or carbamazepine, had

13  Seizures and Epilepsy in the Elderly

187

similar findings: 42% of patients on carbamazepine dropped out due to adverse events compared to only 18% of patients on lamotrigine.39 Somnolence occurred in 29% of patients on carbamazepine versus only 12% of patients on lamotrigine. It was also of interest that only 3% of patients on lamotrigine developed allergic rash compared to 19% of those taking carbamazepine. In general, the newer generation AEDs are less sedating and overall have more favorable adverse effect profiles than the older generation AEDs. Details of these comparisons were reviewed extensively in the 2004 AAN guidelines on the efficacy and tolerability of the newer antiepileptic drugs.73 Lamotrigine has a generally favorable adverse effect profile. It is nonsedating and does not produce significant cognitive dysfunction. It is not an inducer of hepatic enzymes, nor is it highly protein bound, factors important in minimizing the possibility of drug interactions. However, enzyme-inducing medications and hormone replacements can reduce blood levels significantly. Valproate increases lamotrigine levels twoto threefold, and caution is necessary when using the two together. One disadvantage of lamotrigine is that therapeutic dosages cannot be achieved rapidly, in part because of the increased risk of potentially severe allergic reactions. Gradual dosage titration reduces this risk to one that is no greater than that seen with older AEDs. Levetiracetam also has a favorable adverse effect profile. It is nonsedating and does not produce significant cognitive dysfunction. It is renally excreted, is not highly protein bound, and does not induce hepatic enzymes. As a result, drug interactions are rare. About 7% of patients experience irritability and behavioral side effects.74 A major advantage of levetiracetam is that therapeutic efficacy can be achieved at starting dosages. Gabapentin and pregabalin likewise have therapeutic efficacy at starting dosages. Both drugs are renally excreted, are not significantly protein bound, and do not induce hepatic enzymes. Oxcarbazepine is metabolized in the liver, but it is only a weak inducer of hepatic enzymes. Drug interactions are minimal. Like carbamazepine, hyponatremia is seen more often in elderly patients, and concomitant use of diuretics should be cautious.75 Zonisamide and topiramate are generally well tolerated, but both are associated with higher incidences of cognitive side effects. For each, metabolism is partially renal and partially hepatic; neither induces hepatic enzymes. Because of the risk of adverse cognitive effects, zonisamide and topiramate should be titrated slowly. As both are weak carbonic anhydrase inhibitors, they are associated with a small increase in the risk for renal stones. This requires ensuring adequate hydration in elderly patients who take these drugs. To summarize, lamotrigine, levetiracetam, gabapentin, pregabalin, zonisamide, and topiramate have the least potential for interaction with medications used to treat common comorbid conditions in the elderly. However, because zonisamide and topiramate are more likely to produce cognitive side effects than the other drugs, and they also increase the risk of renal stones, we prefer using lamotrigine, levetiracetam, gabapentin, or pregabalin as initial therapy in elderly patients with seizures. EPILEPSY SURGERY Neurologists have generally been reluctant to consider resective surgery in elderly patients with epilepsy because of the higher incidence of associated comorbidities,

188

THE EPILEPSIES 3

as well as a perceived increase in complication rates. In addition, a long duration of epilepsy has been reported to be negative predictor for achieving long-term seizure control.76 However, in appropriate patients, one might equally argue that the same comorbid conditions and their consequent effects on drug absorption, clearance, rate of adverse effects, and interactions justify earlier consideration of potentially curative epilepsy surgery. Reports on the efficacy of surgery in elderly patients come from a relatively small series of patients.7779 Grivas and colleagues recently published the outcome of temporal lobe resection in 52 patients older than 50 (range 50 to 71, mean 56) years.80 At follow-up of at least 12 months (mean duration 33 months), 71% remained seizure free (Engel class 1). This seizure-free rate is virtually identical to that seen in a comparable cohort of 321 patients less than 50 years (72% seizure free) operated on in the same time period. Notably, 11 of the 52 patients were older than 60, and outcome was equally good in this subgroup. Only one patient showed no improvement in postoperative seizure control (2% Engel class 4). Surgical complications were somewhat higher (7.7%) in this older cohort, and there was also a higher rate (3.8%, 2 patients) of permanent mild neurologic morbidity, compared to younger patients. No patient had severe permanent neurologic morbidity, and there were no mortalities. These data suggest that surgery should not be excluded as a treatment option simply on the basis of age.

Epilepsy Morbidity
STATUS EPILEPTICUS The incidence of status epilepticus in persons older than 60 years of age is at least twice, and possibly greater than 10 times, that seen in younger adults. The highest rates occur in patients older than 70.8183 In a series of 171 elderly patients with status epilepticus, more than half (56%) had no previous history of seizures.84 Cerebrovascular disease is the most common cause, accounting for as many as 60% of cases.81,85,86 Other causes include hypoxia, metabolic derangement, alcohol, tumor, infection, trauma, and dementia. The mortality rate associated with status epilepticus is 38% in patients over 60 years of age, but nearly 50% in those over 80. In contrast, the overall mortality rate for status epilepticus is 22%.84,85 The increased mortality rates reflect the elderly populations susceptibility to systemic illnesses (e.g., pneumonia, organ failure), as well as the often life-threatening or progressive nature of the underlying cause of the status epilepticus. Along with advancing age, etiology is a strong determinant of mortality: The highest rates occur from acute symptomatic causes, such as ischemic stroke, intracranial hemorrhage, and anoxic brain injury.87,88 Nonconvulsive status epilepticus (NCSE) represents up to one-fourth of cases in the elderly. As these patients often present with varying degrees of confusion, behavioral change, altered consciousness, and sometimes subtle motor activity, diagnosis can be challenging. It is particularly important to consider this diagnosis when evaluating elderly patients with new-onset encephalopathy. There are no established protocols for treating status epilepticus that are age specific, but the same pharmacokinetic considerations discussed earlier in this chapter must be considered. Pharmacodynamic changes and the presence of comorbid

13  Seizures and Epilepsy in the Elderly

189

conditions in elderly patients increase their susceptibility to CNS and respiratory depression from benzodiazepines and barbiturates, as well as to the cardiovascular effects of intravenous phenytoin and fosphenytoin. BONE HEALTH, FALLS, AND FRACTURES The combination of age-related diminished bone health and the increased risk of falls results in a high rate of fractures in the elderly population. Epilepsy confers an additional risk of falls and consequent fractures, as do the deleterious effects of older generation AEDs on bone health. Patients with epilepsy have an approximately twofold increase in fractures compared to controls.89 In a study of more than 8000 community-dwelling women older than 65, those taking AEDs were 75% more likely to experience a fall.90 Although part of this increase relates to the seizures themselves causing falls, loss of balance due to the increased sensitivity of the aging brain to adverse CNS effects of AEDs also plays a role. In adults, peak bone mineral density is attained between 20 and 30 years of age. After this, bone mineral density gradually declines, but it becomes most pronounced in women following onset of menopause. Older-generation AEDs, including phenytoin, carbamazepine, primidone, and phenobarbital, are commonly associated with altered bone metabolism and decreased bone density because they are potent inducers of the hepatic cytochrome P450 enzyme system.91 Valproate, an inhibitor of the P450 system, was initially believed not to affect bone health. However, it has recently been shown to be associated with reductions in bone mineral density. Polytherapy has a higher risk of abnormalities in bone metabolism than monotherapy. Multiple mechanisms have been postulated to support the association, but none is entirely satisfactory.92 To date, the newer generation AEDs, including lamotrigine, gabapentin, and levetiracetam, have not been associated with deleterious effects on bone mineral metabolism and bone mineral density.

CONCLUSIONS/RECOMMENDATIONS
The primary goals of management are the same for all patients with epilepsy, but diagnosis and treatment of elderly patients with seizures require special considerations. These arise, in large part, from age-related physiological changes and the high frequency of comorbid conditions in this population. The common coexistence of other medical disorders frequently makes diagnosis in these patients more challenging. Once a diagnosis of epilepsy has been established, AEDs must be chosen carefully to minimize potential for toxicity and interactions with other medications. Finally, close follow-up is critical, as elderly patients are at greater risk for complications, social support is often limited, and there is a higher risk of significant injuries from falls and confusion secondary to AED toxicity. We advise keeping the following principles in mind:  Maintain a high suspicion for seizures in the elderly, and when a diagnosis remains in question, have a low threshold for video-EEG monitoring.  Use newer-generation AEDs, which have fewer CNS adverse effects and decreased potential for medication interactions.

190

THE EPILEPSIES 3

 When initiating AED therapy, begin with low doses and titrate slowly to a target dosage of about one-half that of younger patients.  Follow patients closely and be vigilant for adverse effects to minimize the risk of toxicity.  In some elderly patients with pharmacoresistent epilepsy, resective surgery may be a reasonable treatment option. REFERENCES
1. A profile of older Americans. In: U.S. Department of Health and Human Services, ed. Administration on Aging. Washington, DC: Author; 2006. 2. Annegers JF, Hauser WA, Lee JR, Rocca WA. Incidence of acute symptomatic seizures in Rochester, Minnesota, 1935-1984. Epilepsia. 1995;36(4):327-333. 3. Hauser WA, Annegers JF, Kurland LT. Incidence of epilepsy and unprovoked seizures in Rochester, Minnesota: 1935-1984. Epilepsia. 1993;34(3):453-468. 4. Olafsson E, Ludvigsson P, Gudmundsson G, Hesdorffer D, Kjartansson O, Hauser WA. Incidence of unprovoked seizures and epilepsy in Iceland and assessment of the epilepsy syndrome classification: a prospective study. Lancet Neurol. 2005;4(10):627-634. 5. Sander JW, Hart YM, Johnson AL, Shorvon SD. National General Practice Study of Epilepsy: newly diagnosed epileptic seizures in a general population. Lancet. 1990;336(8726):1267-1271. 6. Hauser WA. Epidemiology of seizures and epilepsy in the elderly. In: Rowan AJ, Ramsay RE, eds. Seizures and Epilepsy in the Elderly. Boston, Butterworth-Heineman; 1997:7-20. 7. Forsgren L, Bucht G, Eriksson S, Bergmark L. Incidence and clinical characterization of unprovoked seizures in adults: a prospective population-based study. Epilepsia. 1996;37(3):224-229. 8. Sidenvall R, Forsgren L, Blomquist HK, Heijbel J. A community-based prospective incidence study of epileptic seizures in children. Acta Paediatr. 1993;82(1):60-65. 9. Cloyd J, Hauser W, Towne A, et al. Epidemiological and medical aspects of epilepsy in the elderly. Epilepsy Res. 2006;68(suppl 1):S39-S48. 10. de la Court A, Breteler MM, Meinardi H, Hauser WA, Hofman A. Prevalence of epilepsy in the elderly: the Rotterdam Study. Epilepsia. 1996;37(2):141-147. 11. Hauser WA, Annegers JF, Kurland LT. Prevalence of epilepsy in Rochester, Minnesota: 1940-1980. Epilepsia. 1991;32(4):429-445. 12. Perucca E, Berlowitz D, Birnbaum A, et al. Pharmacological and clinical aspects of antiepileptic drug use in the elderly. Epilepsy Res. 2006;68(suppl 1):S49-S63. 13. Pugh MJ, Cramer J, Knoefel J, et al. Potentially inappropriate antiepileptic drugs for elderly patients with epilepsy. J Am Geriatr Soc. 2004;52(3):417-422. 14. Galimberti CA, Magri F, Magnani B, et al. Antiepileptic drug use and epileptic seizures in elderly nursing home residents: a survey in the province of Pavia, Northern Italy. Epilepsy Res. 2006;68(1):1-8. 15. Garrard J, Cloyd J, Gross C, et al. Factors associated with antiepileptic drug use among elderly nursing home residents. J Gerontol A Biol Sci Med Sci. 2000;55(7):M384-M392. 16. Schachter SC, Cramer GW, Thompson GD, Chaponis RJ, Mendelson MA, Lawhorne L. An evaluation of antiepileptic drug therapy in nursing facilities. J Am Geriatr Soc. 1998;46(9):1137-1141. 17. Loiseau J, Loiseau P, Duche B, Guyot M, Dartigues JF, Aublet B. A survey of epileptic disorders in southwest France: seizures in elderly patients. Ann Neurol. 1990;27(3):232-237. 18. Ramsay RE, Rowan AJ, Pryor FM. Special considerations in treating the elderly patient with epilepsy. Neurology. 2004;62(5 suppl 2):S24-S29. 19. So EL, Annegers JF, Hauser WA, OBrien PC, Whisnant JP. Population-based study of seizure disorders after cerebral infarction. Neurology. 1996;46(2):350-355. 20. Cleary P, Shorvon S, Tallis R. Late-onset seizures as a predictor of subsequent stroke. Lancet. 2004;363(9416):1184-1186. 21. Ng SK, Hauser WA, Brust JC, Susser M. Hypertension and the risk of new-onset unprovoked seizures. Neurology. 1993;43(2):425-428. 22. Rubenstein LZ, Robbins AS, Schulman BL, Rosado J, Osterweil D, Josephson KR. Falls and instability in the elderly. J Am Geriatr Soc. 1998;36(3):266-278. 23. Tinetti ME, Speechley M, Ginter SF. Risk factors for falls among elderly persons living in the community. N Engl J Med. 1988;319(26):1701-1707. 24. Annegers JF, Hauser WA, Beghi E, Nicolosi A, Kurland LT. The risk of unprovoked seizures after encephalitis and meningitis. Neurology. 1988;38(9):1407-1410.

13  Seizures and Epilepsy in the Elderly 25. Hesdorffer DC, Hauser WA, Annegers JF, Kokmen E, Rocca WA. Dementia and adult-onset unprovoked seizures. Neurology. 1996;46(3):727-730. 26. Amatniek JC, Hauser WA, DelCastillo-Castaneda C, et al. Incidence and predictors of seizures in patients with Alzheimers disease. Epilepsia. 2006;47(5):867-872. 27. Dam AM, Fuglsang-Frederiksen A, Svarre-Olsen U, Dam M. Late-onset epilepsy: etiologies, types of seizure, and value of clinical investigation, EEG, and computerized tomography scan. Epilepsia. 1985;26(3):227-231. 28. Hauser WA. Seizure disorders: the changes with age. Epilepsia. 1992;33(suppl 4):S6-S14. 29. Spitz M, Bainbridge JL, Ramsay R, et al. Observations on the delay in the diagnosis of seizures in the elderly: update 2. Epilepsia. 2002;43(suppl 7):166. 30. Godfrey JW, Roberts MA, Caird FI. Epileptic seizures in the elderly: II. Diagnostic problems. Age Ageing. 1982;11(1):29-34. 31. McBride AE, Shih TT, Hirsch LJ. Video-EEG monitoring in the elderly: a review of 94 patients. Epilepsia. 2002;43(2):165-169. 32. Randomized clinical trial on the efficacy of antiepileptic drugs in reducing the risk of relapse after a first unprovoked tonic-clonic seizure. First Seizure Trial Group (FIR.S.T. Group). Neurology. 1993;43(3 Pt 1):478-483. 33. Berg AT, Shinnar S. The risk of seizure recurrence following a first unprovoked seizure: a quantitative review. Neurology. 1991;41(7):965-972. 34. Hart YM, Sander JW, Johnson AL, Shorvon SD. National General Practice Study of Epilepsy: recurrence after a first seizure. Lancet. 1990;336(8726):1271-1274. 35. Hauser WA, Rich SS, Annegers JF, Anderson VE. Seizure recurrence after a 1st unprovoked seizure: an extended follow-up. Neurology. 1990;40(8):1163-1170. 36. Lu hdorf K, Jensen LK, Plesner AM. Etiology of seizures in the elderly. Epilepsia. 1986;27(4): 458-463. 37. Mohanraj R, Brodie MJ. Diagnosing refractory epilepsy: response to sequential treatment schedules. Eur J Neurol. 2006;13(3):277-282. 38. Lu hdorf K, Jensen LK, Plesner AM. Epilepsy in the elderly: prognosis. Acta Neurol Scand. 1986;74(5):409-415. 39. Brodie MJ, Overstall PW, Giorgi L. Multicentre, double-blind, randomised comparison between lamotrigine and carbamazepine in elderly patients with newly diagnosed epilepsy. The UK Lamotrigine Elderly Study Group. Epilepsy Res. 1999;37(1):81-87. 40. Rowan AJ, Ramsay RE, Collins JF, et al. New onset geriatric epilepsy: a randomized study of gabapentin, lamotrigine, and carbamazepine. Neurology. 2005;64(11):1868-1873. 41. Kwan P, Brodie MJ. Early identification of refractory epilepsy. N Engl J Med. 2000;342(5): 314-319. 42. Schmucker DL. Aging and drug disposition: an update. Pharmacol Rev. 1985;37(2):133-148. 43. Schwartz JB. The current state of knowledge on age, sex, and their interactions on clinical pharmacology. Clin Pharmacol Ther. 2007;82(1):87-96. 44. Goa KL, Sorkin EM. Gabapentin. A review of its pharmacological properties and clinical potential in epilepsy. Drugs. 1993;46(3):409-427. 45. Campion EW, deLabry LO, Glynn RJ. The effect of age on serum albumin in healthy males: report from the Normative Aging Study. J Gerontol. 1988;43(1):M18-M20. 46. Perucca E. Pharmacokinetics. In: Engel J Jr, Pedley TA, eds. Epilepsy: A Comprehensive Textbook. Philadelphia: Lippincott-Raven Publishers; 1997:1131-1144. 47. Woodhouse KW, Wynne HA. Age-related changes in liver size and hepatic blood flow. The influence on drug metabolism in the elderly. Clin Pharmacokinet. 1988;15(5):287-294. 48. Ahronheim JC. Age-related changes in drug metabolism and action. In: Rowan AJ, Ramsay RE, eds. Seizures and Epilepsy in the Elderly. Boston: Butterworth-Heinemann; 1997:55-61. 49. Rowe JW, Andres R, Tobin JD, Norris AH, Shock NW. The effect of age on creatinine clearance in men: a cross-sectional and longitudinal study. J Gerontol. 1976;31(2):155-163. 50. Lindeman RD, Tobin J, Shock NW. Longitudinal studies on the rate of decline in renal function with age. J Am Geriatr Soc. 1985;33(4):278-285. 51. Malmrose LC, Gray SL, Pieper CF, et al. Measured versus estimated creatinine clearance in a highfunctioning elderly sample: MacArthur Foundation Study of Successful Aging. J Am Geriatr Soc. 1993;41(7):715-721. 52. Battino D, Croci D, Rossini A, Messina S, Mamoli D, Perucca E. Serum carbamazepine concentrations in elderly patients: a case-matched pharmacokinetic evaluation based on therapeutic drug monitoring data. Epilepsia. 2003;44(7):923-929.

191

192

THE EPILEPSIES 3 53. Richens A, Banfield CR, Salfi M, Nomeir A, Lin CC, Jensen P, et al. Single and multiple dose pharmacokinetics of felbamate in the elderly. Br J Clin Pharmacol. 1997;44(2):129-134. 54. Boyd RA, Turck D, Abel RB, Sedman AJ, Bockbrader HN. Effects of age and gender on single-dose pharmacokinetics of gabapentin. Epilepsia. 1999;40(4):474-479. 55. Posner J, Holdich T, Crome P. Comparison of lamotrigine pharmacokinetics in young and elderly healthy volunteers. J Pharm Med. 1991;1:121-128. 56. Patsalos PN. Clinical pharmacokinetics of levetiracetam. Clin Pharmacokinet. 2004;43(11):707-724. 57. van Heiningen PN, Eve MD, Oosterhuis B, et al. The influence of age on the pharmacokinetics of the antiepileptic agent oxcarbazepine. Clin Pharmacol Ther. 1991;50(4):410-419. 58. Messina S, Battino D, Croci D, Mamoli D, Ratti S, Perucca E. Phenobarbital pharmacokinetics in old age: a case-matched evaluation based on therapeutic drug monitoring data. Epilepsia. 2005;46(3):372-377. 59. Bachmann KA, Belloto RJ Jr. Differential kinetics of phenytoin in elderly patients. Drugs Aging. 1999;15(3):235-250. 60. Snel S, Jansen JA, Mengel HB, Richens A, Larsen S. The pharmacokinetics of tiagabine in healthy elderly volunteers and elderly patients with epilepsy. J Clin Pharmacol. 1997;37(11): 1015-1020. 61. Doose DR, Larson KL, Natarajan J, Neto W. Comparative single-dose pharmacokinetics of topiramate in elderly versus young men and women. Epilepsia. 1998;39(suppl 6):56. 62. Perucca E, Grimaldi R, Gatti G, Pirracchio S, Crema F, Frigo GM. Pharmacokinetics of valproic acid in the elderly. Br J Clin Pharmacol. 1984;17(6):665-669. 63. Haegele KD, Huebert ND, Ebel M, Tell GP, Schechter PJ. Pharmacokinetics of vigabatrin: implications of creatinine clearance. Clin Pharmacol Ther. 1988;44(5):558-565. 64. Bergey GK. Initial treatment of epilepsy: special issues in treating the elderly. Neurology. 2004;63(10 suppl 4):S40-S48. 65. Cloyd J. Commonly used antiepileptic drugs: age-related pharmacokinetics. In: Rowan AJ, Ramsay R, eds. Seizures and Epilepsy in the Elderly. Boston: Butterworth-Heinemann; 1997:219-228. 66. Helling DK, Lemke JH, Semla TP, Wallace RB, Lipson DP, Cornoni-Huntley J. Medication use characteristics in the elderly: the Iowa 65+ Rural Health Study. J Am Geriatr Soc. 1987;35(1):4-12. 67. Lackner TE, Cloyd JC, Thomas LW, Leppik IE. Antiepileptic drug use in nursing home residents: effect of age, gender, and comedication on patterns of use. Epilepsia. 1998;39(10):1083-1087. 68. Patsalos PN, Perucca E. Clinically important drug interactions in epilepsy: interactions between antiepileptic drugs and other drugs. Lancet Neurol. 2003;2(8):473-481. 69. Mattson RH, Cramer JA, Collins JF. A comparison of valproate with carbamazepine for the treatment of complex partial seizures and secondarily generalized tonic-clonic seizures in adults. The Department of Veterans Affairs Epilepsy Cooperative Study No. 264 Group. N Engl J Med. 1992;327(11):765-771. 70. Mattson RH, Cramer JA, Collins JF, et al. Comparison of carbamazepine, phenobarbital, phenytoin, and primidone in partial and secondarily generalized tonic-clonic seizures. N Engl J Med. 1985;313(3):145-151. 71. Zaret BS, Beckner RR, Marini AM, Wagle W, Passarelli C. Sodium valproate-induced hyperammonemia without clinical hepatic dysfunction. Neurology. 1982;32(2):206-208. 72. Van Amelsvoort T, Bakshi R, Devaux CB, Schwabe S. Hyponatremia associated with carbamazepine and oxcarbazepine therapy: a review. Epilepsia. 1994;35(1):181-188. 73. French JA, Kanner AM, Bautista J, et al. Efficacy and tolerability of the new antiepileptic drugs. II: treatment of refractory epilepsy: report of the Therapeutics and Technology Assessment Subcommittee and Quality Standards Subcommittee of the American Academy of Neurology and the American Epilepsy Society. Neurology. 2004;62(8):1261-1273. 74. White JR, Walczak TS, Leppik IE, et al. Discontinuation of levetiracetam because of behavioral side effects: a case-control study. Neurology. 2003;61(9):1218-1221. 75. Kutluay E, McCague K, Dsouza J, Beydoun A. Safety and tolerability of oxcarbazepine in elderly patients with epilepsy. Epilepsy Behav. 2003;4(2):175-180. 76. Janszky J, Janszky I, Schulz R, et al. Temporal lobe epilepsy with hippocampal sclerosis: predictors for long-term surgical outcome. Brain. 2005;128(Pt 2):395-404. 77. Boling W, Andermann F, Reutens D, Dubeau F, Caporicci L, Olivier A. Surgery for temporal lobe epilepsy in older patients. J Neurosurg. 2001;95(2):242-248. 78. Cascino GD, Sharbrough FW, Hirschorn KA, Marsh WR. Surgery for focal epilepsy in the older patient. Neurology. 1991;41(9):1415-1417.

13  Seizures and Epilepsy in the Elderly 79. Sirven JI, Malamut BL, OConnor MJ, Sperling MR. Temporal lobectomy outcome in older versus younger adults. Neurology. 2000;54(11):2166-2170. 80. Grivas A, Schramm J, Kral T, et al. Surgical treatment for refractory temporal lobe epilepsy in the elderly: seizure outcome and neuropsychological sequels compared with a younger cohort. Epilepsia. 2006;47(8):1364-1372. 81. DeLorenzo RJ, Hauser WA, Towne AR, et al. A prospective, population-based epidemiologic study of status epilepticus in Richmond, Virginia. Neurology. 1996;46(4):1029-1035. 82. Knake S, Rosenow F, Vescovi M, et al. Incidence of status epilepticus in adults in Germany: a prospective, population-based study. Epilepsia. 2001;42(6):714-718. 83. Vignatelli L, Tonon C, DAlessandro R. Incidence and short-term prognosis of status epilepticus in adults in Bologna, Italy. Epilepsia. 2003;44(7):964-968. 84. DeLorenzo RJ. Clinical and epidemiologic study of status epilepticus in the elderly. In: Rowan AJ, Ramsay RE, eds. Seizures and Epilepsy in the Elderly. Boston: Butterworth-Heinemann; 1997;191-205. 85. DeLorenzo RJ, Pellock JM, Towne AR, Boggs JG. Epidemiology of status epilepticus. J Clin Neurophysiol. 1995;12(4):316-325. 86. Wu YW, Shek DW, Garcia PA, Zhao S, Johnston SC. Incidence and mortality of generalized convulsive status epilepticus in California. Neurology. 2002;58(7):1070-1076. 87. Towne AR, Pellock JM, Ko D, DeLorenzo RJ. Determinants of mortality in status epilepticus. Epilepsia. 1994;35(1):27-34. 88. Waterhouse EJ, Vaughan JK, Barnes TY, et al. Synergistic effect of status epilepticus and ischemic brain injury on mortality. Epilepsy Res. 1998;29(3):175-183. 89. Vestergaard P, Tigaran S, Rejnmark L, Tigaran C, Dam M, Mosekilde L. Fracture risk is increased in epilepsy. Acta Neurol Scand. 1999;99(5):269-275. 90. Ensrud KE, Blackwell TL, Mangione CM, et al. Central nervous system-active medications and risk for falls in older women. J Am Geriatr Soc. 2002;50(10):1629-1637. 91. Pack AM. The association between antiepileptic drugs and bone disease. Epilepsy Curr. 2003;3(3):91-95. 92. Fitzpatrick LA. Pathophysiology of bone loss in patients receiving anticonvulsant therapy. Epilepsy Behav. 2004;5(Suppl 2):S3-S15.

193

THE EPILEPSIES 3

14

Psychosis of Epilepsy
ANDRES M. KANNER

Introduction Epidemiologic Aspects Clinical Features Interictal Psychosis of Epilepsy Postictal Psychotic Episodes Postictal Psychotic Symptoms Postictal Psychotic Episodes Relation Between PIPE and IPE

Ictal Psychotic Episodes Alternative Psychosis or Forced Normalization Iatrogenic Psychotic Episodes Pharmacologic Treatment of POE

Introduction
Psychotic disorders are an uncommon psychiatric comorbidity in patients with epilepsy, but their prevalence is higher than in the general population. An association between psychotic phenomena and epilepsy was described in ancient Greek and Roman literature. For example, Hippocrates and Aristotle considered Hercules to have suffered from epilepsy and to have killed his children in the midst of a fit of madness.1 Until the 19th century, epilepsy continued to be associated with madness in one form or another to the point where various authors considered psychotic episodes to be an epileptic equivalent.2 Indeed, Falret in 1854 and Morel in 1873 referred to psychotic disorders in patients with epilepsy as larval epilepsy, epileptic mania, and grands maux intellectuels.3,4 Kraepelin described the temporal relation between seizure occurrence and the development of psychotic episodes, such as postictal psychosis, and also recognized their occurrence independent of seizures (interictal psychotic episodes).5 In the 1950s, several investigators recognized that patients with epilepsy, especially those with temporal lobe epilepsy (TLE), could have a psychotic disorder that differed in many ways from the schizophrenias. Hill in 19536 and Pond in 19577 observed that these patients did not display the lack of affect and the asocial or withdrawn attitude that was typical of patients with schizophrenia. In their classic paper, Slater et al.8 also emphasized the differences described by Hill and Pond. Because the psychotic episodes in these patients included paranoid delusions with visual and auditory hallucinations Slater et al. introduced the term schizophrenia-like psychosis.

194

14  Psychosis of Epilepsy

195

Further systematic analyses of psychiatric phenomena by others continued to depict a unique psychotic disorder that, while sharing pivotal symptoms, differed in important ways from the psychoses that affected nonepileptic patients. Today, psychotic disorders identified in patients with epilepsy are commonly referred in the medical literature as psychosis of epilepsy (POE). This is a term applied to a group of psychotic disorders with distinct phenomenology and etiopathogenic mechanisms that are likely to be closely related to the seizure disorder.9 For example, in a review of the literature, Ferguson and Rayport described how episodic psychosis of epilepsy is related to seizure recurrence and remits when seizures are controlled.9,10 The purpose of this chapter is to review the clinical manifestations of the various forms of POE, their potential pathogenic mechanisms, and current management. EPIDEMIOLOGIC ASPECTS The prevalence rates of primary schizophreniform disorders in the general population range between 0.4 and 1%. In contrast, the prevalence of POE has been estimated to be between 7 and 10%.2 Unfortunately, data about POE derived from population-based studies are limited. In a retrospective study from a tertiary center, Mendez et al found psychotic disorders to be prevalent in 9% of patients with epilepsy and in 1% in migraneurs.11 In a study by Matsuura et al. carried out in Japan, psychotic disorders were found in 6% of patients with epilepsy.12 Overall, the incidence of schizophrenia-like psychosis is believed to be 6 to 12 times higher in patients with epilepsy than in the general population. The relation between psychotic disorders and epilepsy has long attracted considerable interest. At the beginning of the last century, von Meduna described an inverse relationship between the occurrence of psychotic episodes and seizures.2 Such observations ultimately led to the use of iatrogenic convulsions, initially with camphor and later with electroconvulsive therapy (ECT), for the treatment of serious psychotic disorders. In a case-control study of new-onset epilepsy in older adults, Hesdorffer et al. concluded that a prior diagnosis of schizophrenia was protective against developing epilepsy.13,14 However, this does not appear to be the case in pediatric populations. For example, Jablinsky et al. found that epilepsy in children increased the risk of schizophrenia by a factor of 2 in a case-control study of psychiatric illness in developing countries.15 The relation between febrile seizures and the risk of schizophrenia is disputed: A Danish study found a positive correlation,16 whereas no association was found in a ten-country study by the WHO.

Clinical Features
The traditional classification of POE has been based on the temporal relationship between the psychotic episode and occurrence of seizures. Psychotic episodes are divided into ictal (e.g., the psychotic symptoms are the clinical expression of the seizure), postictal (e.g., episodes occurring up to 120 hours after the seizure); and interictal (e.g., episodes that are independent of seizures).17 A fourth manifestation of POE is the alternative psychotic episode (APE), also known as forced normalization, in which psychotic episodes occur in patients who become seizure free after having had persistent seizures for many years.18,19 Rayport and Ferguson proposed

196

THE EPILEPSIES 3

that POE be grouped into two categories9,10: episodic psychosis of epilepsy and chronic (or nonepisodic) psychosis of epilepsy. The former has been identified primarily in patients with TLE and, according to these authors, are closely linked to the effectiveness of seizure control. Episodic psychotic episodes last from a few days to several weeks. They can have a prolonged course and carry a guarded prognosis. Rayport and Ferguson also pointed out that episodic exacerbations can occur in patients with nonepisodic disorders if the seizures worsen. Thus, Rayport and Fergusons episodic psychosis of epilepsy corresponds to postictal psychotic episodes (PIPE) and APE, whereas the nonepisodic psychoses are equivalent to the interictal psychosis of epilepsy (IPE). Finally, in addition to IPE, PIPE, and APE, any classification should also include psychotic processes resulting from iatrogenic effects of antiepileptic drugs (AEDs) or surgical treatment.

Interictal Psychosis of Epilepsy


Persons with epilepsy can have interictal psychotic disorders that are clinically indistinguishable from primary schizophreniform disorders.20 However, these patients are older at onset of psychotic symptoms than people without epilepsy. According to the Diagnostic and Statistical Manual of Mental Disorders,21 a diagnosis of schizophrenia requires symptoms from at least two of the following five categories to have been present for a minimum period of 1 month: delusions, hallucinations, disorganized speech, grossly disorganized or catatonic behavior, and negative symptoms (i.e., affective flattening, alogia, avolition). In addition, patients have to display dysfunction in social and occupational domains or in self-care for a minimum of 6 months (including at least 1 month of negative symptoms or at least two of the symptom categories listed earlier), but of lesser severity. However, as already mentioned, POE is remarkable for the absence of negative symptoms as well as a better premorbid condition. It is also only rarely accompanied by deterioration of a patients personality.8 This point is illustrated in Slaters observation that the delusions and hallucinations of patients with POE were empathizable (the patient remains in our world).8 There is general agreement about the lesser severity and also that better response to therapy can be anticipated in POE.22

Postictal Psychotic Episodes


In the past, IPE was thought to be the most frequent POE. Since the advent of videoelectroencephalogram (V-EEG), however, PIPE have been recognized with increasing frequency, and some investigators are now suggesting that their prevalence is greater than IPE. Furthermore, several studies have now demonstrated that PIPE can progress to chronic IPE 23,24 (see discussion later in chapter). Although PIPE are of limited duration and remit following treatment with low doses of antipsychotic drugs or, sometimes, even benzodiazepines, they must nonetheless be taken seriously as they are associated with increased mortality, which results from an elevated incidence of suicide. In addition, PIPE are associated with more severe forms of epilepsy that are often not amenable to surgical treatment

14  Psychosis of Epilepsy

197

(see discussion later in chapter). Postictal psychotic phenomena can present as isolated symptoms or as fully developed psychotic episodes. POSTICTAL PSYCHOTIC SYMPTOMS Kanner et al. investigated the prevalence of postictal psychiatric symptoms in 100 consecutive patients with pharmaco-resistant focal epilepsy during a 3-month period.25 The postictal period was defined as the 72 hours that followed recovery from the last seizure. Questions were asked about the frequency of occurrence of each symptom. Only symptoms that were identified after more than 50% of seizures were included in this study to reflect a habitual occurrence. To ensure that patients were reporting postictal psychiatric symptoms, each question in the survey also asked about occurrence of the same symptoms during the interictal period. When similar symptoms were identified during both interictal and postictal periods, the only symptoms considered to be postictal were those that were significantly more severe during the postictal period. These were then classified as a postictal exacerbation of interictal psychiatric symptoms. Among the 100 patients, 79 patients had seizures of temporal origin, and 21 had seizures of extratemporal origin. Half of the patients had only complex partial seizures (CPS); the other half had both CPS and generalized tonic-clonic (GTC) seizures. Fifty-two patients had a history of psychiatric conditions: depression, anxiety disorders, and attention deficit disorders. Seven patients experienced postictal psychotic symptoms after more than 50% of their seizures; the duration of the postictal psychotic symptoms, ranged between 1 and 108 hours (median: 15 hours). Such symptoms always occurred together with postictal symptoms of depression and anxiety. POSTICTAL PSYCHOTIC EPISODES As mentioned previously, postictal psychiatric symptoms may cluster into PIPE, which correspond to approximately 25% of POE.26 The prevalence of PIPE has been estimated to range between 6 and 10% among patients with pharmaco-resistant epilepsy.27,28 Since the advent of V-EEG monitoring more than four decades ago, recognition of PIPE has increased substantially. This is not surprising because the circumstances of V-EEG are such that the likelihood of psychiatric symptoms occurring is increased (due, for example, to the facilitation of frequent seizures over a short time period following discontinuation or dose reduction of AEDs). In a 1996 study, the yearly incidence of postictal psychiatric disorders among patients with focal epilepsy undergoing V-EEG was 7.9%.27 The majority (6.4%) presented as PIPE. Among the 10 patients with PIPE, four patients had a delusional psychosis, one patient had a mixed manic depressivelike psychosis, two had psychotic depression, one had a hypomanic-like psychosis, and one had a manic-like psychosis. The tenth patient presented with bizarre behavior associated with a thought disorder. In every case, the onset of symptoms lagged the last seizure by a mean period of 24 hours (range 12 to 72 hours). The mean duration of the PIPE was 69.6 hours (range 24 to 144). In five patients, psychotic episodes remitted with low doses of a neuroleptic drug (2 to 5 mg/day of haloperidol), although one patient required high doses of haloperidol (40 mg/day), and remission occurred in four without pharmacotherapy. Six of the 10 patients had an average of 2.4 PIPE prior to V-EEG, whereas in the remaining four patients, it was the first one. Other authors have reported

198

THE EPILEPSIES 3

similar findings with respect to clinical characteristics, course, and response to pharmacotherapy.2934 Common findings among the different case series include (1) a symptom-free interval between the last seizure and onset of psychiatric symptoms; (2) a relatively short duration; (even shorter durations were noted among our patients, as only two had episodes lasting more than five days, while this was true in 5 of 9 patients in Savards30 and in 7 of 14 patients in Logsdails series29; (3) affectladen symptoms; (4) clustering of symptoms into delusional and affective-like psychosis; (5) increased numbers of secondarily generalized tonic-clonic seizures before the onset of PIPE; (6) onset of PIPE after having had seizures for more than 10 years; and (7) a prompt response to low-dose neuroleptic medication or benzodiazepines. Kanemoto et al. studied the clinical differences between PIPE and acute and chronic IPE.33 They noted that patients with PIPE were more likely to experience grandiose and religious delusions in the presence of elevated moods as well as a sense of mystic fusion of the body with the universe. On the other hand, perceptual delusions or commenting voices were less frequent in PIPE, whereas feelings of impending death were common among patients with PIPE. Various investigators have tried to identify the pathogenic mechanisms of PIPE. For example, Kanner and Ostrovskaya compared 18 consecutive adults with focal seizure disorders and PIPE, and 36 patients with focal epilepsy but without PIPE. Data analysis included the number and location of ictal foci recorded by V-EEG, seizure type, etiology, age at seizure onset, duration of seizure disorder, MRI abnormalities, and psychiatric history before the index V-EEG (other than PIPE).35 Significant differences included the presence of bilateral independent ictal foci on V-EEG (these were identified as an independent predictor of the development of PIPE in univariate and multivariate analyses). The logistic regression model correctly classified 89% of patients. A second independent predictor of PIPE was the occurrence of only secondarily generalized tonic-clonic seizures. Conversely, the occurrence of PIPE and cryptogenic focal epilepsy were predictive of bilateral independent ictal foci in univariate analyses. In a recent study that included 59 consecutive patients with focal epilepsy and a history of PIPE, and 94 controls with focal epilepsy and no history of PIPE, Alper et al. found that predictors of PIPE included ambiguous or extratemporal localization, a family history of psychiatric disorders, abnormal interictal EEGs, and encephalitis.36 Other investigators have also identified bilateral independent interictal32 and ictal34 foci, as well as the presence of secondarily GTC.29,30,34 Prompt recognition of PIPE has important clinical implications. First, its recurrence can be minimized by starting an atypical neuroleptic drug at a low dose at the first sign of PIPE (e.g., risperidone 1 to 2 mg). In most cases, insomnia is the initial presenting symptom. Family members need to be informed about early symptoms so that they can assist in the timely administration of the antipsychotic drug. Patients should be maintained on a dose of the drug for 2 to 5 days, after which it can be discontinued. Patients with a family history of psychiatric symptoms, who have a cluster of secondarily GTC seizures in the course of V-EEG monitoring, and who are found to have bilateral independent ictal foci should be watched carefully for development of PIPE. Symptoms typically become evident between 12 and 72 hours after the last seizure. Occurrence of PIPE has implications with respect to the localization of ictal foci. As suggested by the data cited earlier, development of PIPE should raise the possibility of bilateral independent ictal foci. This is especially important in patients

14  Psychosis of Epilepsy

199

who are being considered for epilepsy surgery. The consequence is that such patients may require longer V-EEG monitoring and, possibly, use of intracranial electrodes. If recordings with depth or subdural electrodes are undertaken, prophylactic treatment with low-dose atypical antipsychotic drugs can avert the occurrence of PIPE during the period of invasive V-EEG monitoring.27 RELATION BETWEEN PIPE AND IPE There appears to be a bidirectional relationship between PIPE and IPE. For example, in a retrospective study of 18 consecutive adults with focal seizure disorders and PIPE and 36 patients with focal seizure disorders not accompanied by PIPE, Kanner and Ostrovskaya found that seven patients with PIPE and one control patient went on to develop IPE.24 Predictors of developing IPE in univariate logistic regression analyses included a history of PIPE, male gender and having bilateral ictal foci. Whether preventing further PIPE protects patients from developing IPE has not been established. Such risk, however, supports the recommendation to consider epilepsy surgery whenever possible, even if it is only palliative.24 Remission of IPE and PIPE has been reported after epilepsy surgery in a woman who had had temporal lobe epilepsy since the age of 15 years.37 She developed recurrent PIPE at the age of 35, and this evolved to chronic refractory IPE. Presurgical evaluation demonstrated right hippocampal atrophy and a right mesial temporal epileptogenic focus. Following a right temporal lobe resection, she has been seizure free, and the IPE has remitted. Although an association between a history of PIPE and development of IPE has been identified in nonsurgical case series,23,37 this has not been reported in studies of patients who have undergone epilepsy surgery. However, Kanemoto et al. found a significant risk of postsurgical mood disorders in patients with a history of presurgical PIPE.38 Indeed, preoperative occurrence of PIPE was five times more frequent among patients with postoperative mood disorders (38%) than among those without (7%). The relation between mood disorders and PIPE was further established in a study by Alper et al. that found a higher prevalence of mood disorders among firstand second-degree relatives as the only psychiatric variable that predicted the development of PIPE on logistic regression analyses (odds ratio = 3.49, P = 0.001).39 A bidirectional relationship between PIPE into IPE was also reported by Tarulli et al., who conducted a retrospective study of 43 patients with PIPE.23 Five (13.9%) patients developed IPE after multiple documented PIPE, whereas in one patient IPE preceded PIPE. The length of time between PIPE and IPE ranged from 7 to 96 months. Adachi et al. also reported a bidirectional relationship between IPE and PIPE (which they called bimodal psychosis) in a study of 14 patients.40 Ten patients who had PIPE went on to develop IPE, whereas four patients who had IPE that remitted later developed PIPE. Mean age at onset of epilepsy was 10.8 4.3 years, at the time of the first IPE 24.4 6.1 years, and at the time of the initial PIPE 33.8 4.5 years. These patients did not differ with respect to the epilepsyrelated characteristics found in patients with only PIPE: They had bilateral EEG abnormalities and borderline (or decreased) intellectual function. In the study by Tarulli et al., the symptomatology of PIPE and IPE was similar or identical in five of six cases, and this was also true in our study. However, differences were found in other studies. For example, Kanemoto et al. compared the clinical semiology among 30 patients with PIPE, 33 patients with acute IPE, and 25 patients

200

THE EPILEPSIES 3

with chronic IPE.41 The most striking feature that distinguished PIPE from both acute interictal and chronic psychoses was the relatively frequent occurrence of grandiose delusions as well as religious delusions in the setting of markedly elevated moods and feeling of mystic fusion of the body with the universe in pipe. In addition, patients with PIPE exhibited few schizophreniform psychotic traits such as perceptual delusions or auditory hallucinations. Further studies are needed to clarify this point.

ICTAL PSYCHOTIC EPISODES


This form of psychosis should always be considered in the differential diagnosis of PIPE and IPE. Ictal psychotic episodes are typically an expression of nonconvulsive status epilepticus (NCSE) presenting as simple partial (SPSE), complex partial (CPSE), or absence status (ASE).42 In the case of SPSE, diagnosis can be difficult, as scalp recordings may not detect ictal patterns.43 SPSE can manifest as hallucinations and illusions. Ictal hallucinations include visual and, less frequently, auditory hallucinations. Although ictal visual hallucinations are usually brief and stereotyped, there are reports of long-lasting hallucinations. In contrast to psychotic hallucinations, patients are able to realize that their hallucinations reflect unreal phenomena. This is illustrated in the case of a 54-year-old man who began to have recurrent complex visual hallucinations and illusions described as puffs of smoke, a wall filled with water and fish, and the American flag on a pole, growing larger. He said that he knew they were not real, and that he had no fear of them. These hallucinations recurred over a period of 7 weeks and remitted after administration of antiepileptic drugs (AEDs). The EEG demonstrated a focal ictal discharge over the right temporal area that was concurrent with the hallucinations.44 Elementary visual hallucinations have also been reported as illustrated in the case of a 26-year-old man with SPSE who presented with images of snowing on TV, flickering lights, and rotating colored balls in the right upper visual field that persisted for several days. Magnetoencephalography showed continuous periodic epileptiform discharges over the left posterior superior temporal region, whereas simultaneous EEG showed rhythmic theta waves and sporadic spikes over the left temporal region. Symptoms disappeared after the patient was treated with AEDs.45 Transient cortical blindness has also been reported as a manifestation of occipital lobe SPSE.46,47 Absence status epilepticus has been associated with bizarre behaviors that patients may be unaware of and are thus usually reported by relatives or friends. For example, Olnes at al. described a man who was unable to dress himself after a shower, roamed about his house until his wife could assist him, and went to bed wearing his coat and boots.48 After driving to work, he could not open his locker and when assisted he stated, I cant get my truck started. He put two cups in an empty dishwasher and ran it without detergent. He lit a cigarette as if to smoke but stared at it for several minutes. These behaviors persisted for several days. A neurological evaluation revealed that he was oriented to person, place, year, and hospital. He had difficulty concentrating and had limited short-term recall. He was unable to draw a clock face after several attempts. An EEG demonstrated generalized spikes, polyspikes, and slow waves consistent with atypical ASE. The patient was treated with intravenous lorazepam, and his mental status improved within minutes. When he was again asked to draw a clock face, he was able to do so.

14  Psychosis of Epilepsy

201

Some patients become depressed, agitated, and occasionally hostile in the midst of an ASE. Other common phenomena have been described in the following way: sensation of viewing the world through a different medium, a feeling of not being in the same world as everyone else, uncontrollable rush of thoughts, a feeling of fear of losing control of my mind, a feeling of closeness, a funny feeling that I cannot elaborate, a strange feeling of not being myself, edgy, worried, and uncomfortable, my character changes completely, I become extremely snappy . . . have a severe headache, or weird.49,50 In addition, isolated reports of ictal catatonia as an expression of NCSE have been published.51,52 For example, Kanemoto et al. described the case of a 78-year-old man who presented with a 1-week history of episodes of mutism alternating with psychomotor agitation.52 On examination, the patient was awake but had a fixed stare. Occasionally, when questioned repeatedly, the patient answered in a whisper with appropriate but fragmented words. His arms and legs were maintained in indefinite and passively placed bizarre postures. There were no focal neurological signs, and he had no history of either epilepsy or psychiatric illness. The EEG showed continuous generalized spike and wave discharges (1.5 to 2 Hz) consistent with atypical ASE. An interview with his wife revealed that he had discontinued benzodiazepines acutely. Administration of diazepam resulted in remission of the abnormal behavior, including catatonic symptoms. Without EEG recording, the ictal nature of the catatonia would likely have remained unrecognized. EEG examination should be considered in patients with catatonia, especially those that occur de novo in elderly persons. In general, the presence of unresponsiveness and automatisms should facilitate suspicion of ASE and CPSE. Only EEG confirmation, however, can definitely clarify some psychotic manifestations, such as catatonic states with unresponsiveness and mannerisms that mimic automatisms.

ALTERNATIVE PSYCHOSIS OR FORCED NORMALIZATION


In 1953, Landoldt reported an inverse relation between seizure control and the development of psychotic symptoms in patients who had suffered from persistent seizures for a long time.18 Landoldt described a normalization of EEG recordings with the appearance of psychiatric symptoms and coined the term forced normalization; Tellenbach proposed the term alternative psychosis.19 This antagonism between psychosis and epilepsy has been considered by some as the explanation for the therapeutic effect of electroconvulsive therapy (ECT) of psychotic disorders. This phenomenon is relatively rare. Landoldt reported 47 cases between 1951 and 1958. Single case studies were reported by others, and in 1988 Schmitz estimated the prevalence of alternative psychosis to be 1% among 697 patients followed at a university epilepsy center.53 Forced normalization has been reported in patients with TLE and generalized epilepsies. As with other forms of POE, Wolf identified the psychotic manifestations of an APE after a 15.2-year history of epilepsy in 23 patients.54 Both Landoldt and Wolf reported a pleomorphic clinical presentation with a paranoid psychosis without clouding of consciousness being the most frequent manifestation. As with other POE, a richness of affective symptoms has been identified. The phenomenon of forced normalization has been observed following the use of various AEDs, including phenytoin and primidone,55,56 carbamazepine and valproic

202

THE EPILEPSIES 3

acid,57 and vigabatrin.58 In these cases, the psychotic disorder was thought to result from the suppression of seizures and not from an adverse event of the AED.59 The pathogenic mechanisms mediating APE are yet to be established. Some hypotheses have been proposed. Trimble postulated that an excess of dopaminergic effect is responsible for the seizure cessation and the psychotic symptomatology.2 Rayport and Ferguson suggested that forced normalization is not the expression of seizure cessation, but, rather, of a voltage depression or suppression in neocortical derivations concurrent with ictal activity in amygdala or hippocampal structures.9,10 Depth electrode studies by Wieser appear to lend support to this hypothesis.60 The treatment of APE remains the source of debate. Ried and Mothersill, for example concluded that the treatment of alternative psychotic episodes should include the reduction and/or discontinuation of AED, until overt seizure recurrence causes remission of psychotic symptoms.59 The rapidity with which AED should be tapered is not clear. Tellenbach suggested a rapid tapering under EEG monitoring. Landoldt advocated the use of ECT or metrazol if necessary.18,19 Following the recurrence of seizures and the remission of psychotic symptoms, AEDs should be reintroduced slowly.59

IATROGENIC PSYCHOTIC EPISODES


AED-Related Psychotic Episodes Psychotic symptoms and episodes as an expression of a toxic phenomenon has been reported with most AEDs, including the first-generation AEDs ethosuximide, phenytoin, phenobarbital, and primidone.61,62 as well as the newer AEDs, like vigabatrin, topiramate, levetiracetam, and zonisamide.6365 The clinical differentiation between APE and a toxic reaction can be difficult if a seizure-free state followed the introduction of the AED. Psychotic disorders can occasionally follow the discontinuation of AEDs, particularly those with mood stabilizing properties. Ketter et al. reported the development of anxiety and depression, primarily, but also of some cases who experienced psychosis among 32 inpatients who were withdrawn from carbamazepine, phenytoin, and valproic acid.66 Acute withdrawal from benzodiazepines is well known to result in an acute psychotic episode.67 Postsurgical Psychotic Episodes Postsurgical psychotic disorders are a relatively rare complication of epilepsy surgery with prevalence rates estimated to range between 3 and 10% among patients undergoing an anterior-temporal lobe resections. Several case series have included a mixture of patients with presurgical and de novo postsurgical psychotic disorders. The first important series was that by Falconer, who reported on the prevalence of de novo postsurgical psychotic episodes among 100 patients who had undergone an anterior temporal lobe resection. Seven patients developed de novo postoperative psychosis.68 Jensen and Vaernet reported de novo psychotic disorders in 9 of 74 patients.69 Trimble calculated that the prevalence rate of postoperative de novo psychoses ranged between 3.8 and 35.7% (mean, 7.6%). He suggested that at least some of these cases may be an expression of forced normalization.70

14  Psychosis of Epilepsy

203

Many epilepsy centers currently do not consider patients with a preoperative history of psychosis as potential candidates for epilepsy surgery. Thus, the lower prevalence rates of more recent reports of postsurgical psychotic episodes are primarily de novo psychoses. De novo postsurgical psychotic episodes can mimic schizophreniform disorders, manic episodes, or present as PIPE. In a study carried out at the Rush Epilepsy Center, Kanner et al. identified de novo psychotic episodes in four patients within the first 6 months after an anterior temporal lobe resection. Manifestations consisted of manic episodes in two and paranoid episodes in two [unpublished data]. Symptoms remitted in three patients using pharmacotherapy without need for hospitalization. One patient had to be hospitalized in a psychiatric unit. Two of the patients had lesional epilepsy; in one it was caused by dysembryoplastic neuroepithelioma (DNET) and in the other by a ganglioglioma. Shaw et al. reported 11 patients with postoperative new-onset schizophreniform psychosis among 320 consecutive patients (3.2%) who underwent anterior temporal lobe resection.71 Psychotic symptoms became apparent within the first postoperative year in all patients. These 11 patients were compared to a control group of 33 postsurgical patients who did not develop psychotic symptoms. Symptomatic patients were significantly more likely to have bilateral epileptiform activity, a smaller amygdala in the nonoperated side, and pathologies other than mesial temporal sclerosis. In a study of 57 consecutive patients who underwent anterior temporal lobe resection, Leinonen et al. identified five (8.8%) who developed postoperative psychotic episodes.72 Two (3.5%) patients had experienced PIPE before surgery and continued to have similar episodes postoperatively. Among the other three patients, two (3.5%) experienced a definite and one (1.8%) a probable de novo schizophreniform psychotic disorder. Stevens identified a de novo psychotic disorder within the first 12 months after surgery in 14 patients who had undergone an anterior temporal lobe resection and who were followed for a period of 20 to 30 years.73 Some investigators have associated the risk of postsurgical psychotic episodes with a right temporal seizure focus. For example, Mace et al. reported seven consecutive patients who developed a de novo psychotic disorder following a right anterior temporal lobe resection: one developed a delusional depression and four a schizophrenic-like psychosis, and one patient was diagnosed with Capgras syndrome.74 Nonetheless, the relation between side-of-seizure focus and the risk of developing postsurgical psychosis cannot be decided on the bases of these small case series. Other authors have associated the presence of gangliogliomas or DNET with the development of de novo postsurgical psychotic disorders. Andermann et al. reported six patients from four centers who experienced a de novo postoperative psychotic disorder presenting as schizophreniform-like episodes with paranoid and depressive symptomatology.75 This association remains to be established in larger studies, however. Postsurgical psychiatric complications of anterior temporal lobe resection can present as manic episodes. For example, Carran et al. identified 16 patients who developed a de novo manic episode from a case series of 415 consecutive patients (corresponding to a prevalence of 3.8%) who had undergone anterior temporal lobe resection at the Comprehensive Epilepsy Center of Jefferson University Medical Center.76 These patients were compared to a control group of asymptomatic patients matched for age and gender and a second group of 30 patients who experienced a

204

THE EPILEPSIES 3

postsurgical depression. Manic episodes occurred within the first postoperative year and were short-lived in all but one patient. Compared to the two control groups, patients with postsurgical mania were more likely to display bilateral EEG abnormalities and to have a right temporal seizure focus, although this difference did not reach significance when compared to the depressed group. Both symptomatic postsurgical patients were more likely to have experienced GTC seizures before surgery and to have had continuing seizures postoperatively. Postictal psychotic episodes can also occur de novo after anterior temporal lobe resections. For example, Christodoulou et al. reported three cases (1%) among 282 consecutive patients who had undergone an anterior temporal lobe resection.77 All three patients had seizures predominantly from the contralateral (nonsurgical) site or had bilateral independent seizures, whereas none of the patients who failed surgery but continued to have seizures from the site of the surgery developed de novo postictal psychosis. This supports the conclusion that patients with PIPE (chronic or de novo) have bilateral independent temporal lobe dysfunction. Manchanda et al. identified four patients (1.3%) who developed a de novo PIPE among a group of 298 consecutive patients who had undergone an anterior temporal lobe resection.78 All four patients had a right-sided resection and had no preoperative psychiatric history. A history of psychosis should not be considered a contraindication to epilepsy surgery, provided the patient can cooperate during the presurgical evaluation and has a clear understanding of the nature of the surgical procedure, the potential risks and probability of seizure freedom. This view is supported by the paper of Reutens et al. on five patients with medically intractable epilepsy and chronic psychosis who had temporal lobe resection for epilepsy.79 Postoperatively, all patients were seizure free. Surgery had no apparent effect on the course or severity of the patients psychoses. However, seizure freedom allowed two of the five patients to return to work or a supervised workshop. Although it is often difficult to predict which patients will develop psychiatric complications after epilepsy surgery, several studies have suggested potential risk factors for the development of psychosis. Among these are surgery over the age of 30 years, family history of psychosis,80 and as stated earlier, the presence of a ganglioglioma or DNET.75 In addition, several studies have examined the issue of postoperative psychosis and the laterality of surgery. Most studies report psychosis to be more common in patients following right temporal lobectomy.72,75,78 The explanation for this lateralizing correlation is yet to be established.

PHARMACOLOGIC TREATMENT OF POE


The treatment of all forms of POE, with the exception of ictal psychotic episodes, requires the use of antipsychotic drugs (APD). The treatment of PIPE and alternative psychosis has already been discussed earlier and will not be reviewed in this section. We will concentrate on the pharmacologic management of IPE. In patients with this form of POE, optimal seizure control is desirable, as episodic psychotic exacerbations can accompany clusters of seizures. APDs are grouped into the first generation or conventional APD (CAPD) which include 18 drugs developed between the 1950s and the 1970s and second-generation or atypical (AAPD), which include six drugs. The mechanism of action of CAPD consists of the dopamine (DA-2) receptors blockade, both at the level of

14  Psychosis of Epilepsy

205

mesocortical, nigrostriatal, and tuberoinfundibular DA pathways.81 These mechanisms are responsible for their antipsychotic effect, but can also cause emotional blunting and cognitive symptoms that often lead to confusion with the negative symptoms of schizophrenia. Blockade at the nigrostriatal pathways can be associated with acute and chronic movement disorders, presenting as parkinsonian symptoms, as well as dystonic and dyskinetic movements, whereas blockade at the tuberoinfundibular pathways results in increased secretion of prolactin. In addition to their DA blockade properties, most of these CAPDs have muscarinic cholinergic, alpha-1 and histaminic blocking properties, responsible for anticholinergic adverse effects, weight gain, sedation, dizziness, and orthostatic hypotension. Atypical APDs are dopamine-serotonin antagonists that target DA-2 and 5HT-2A receptors.81 Their main difference with CAPDs is the absence or mild occurrence of extrapyramidal adverse events and of hyperprolactinemia. In addition, this class of drugs has less blunting of effect, and several of these AAPD have mood-stabilizing properties. Hence, AAPDs have in large part replaced the CAPDs. The six AAPDs available include clozapine, risperidone, olanzapine, ziprasidone, quietapine, and aripiprazole. The proconvulsant properties of APDs have been recognized for a long time and often led to some clinicians reluctance to use these drugs in patients with epilepsy for fear of worsening seizure frequency and severity. Among the CAPDs, the incidence rates of seizures in nonepileptic patients have ranged between 0.5 and 1.2%.82,83 The risk is higher with certain drugs and in the presence of the following factors: (1) a history of epilepsy, (2) abnormal EEG recordings, (3) history of CNS disorder, (4) rapid titration of the CAPD dose, (5) high doses of AP, and (6) the presence of other drugs that lower the seizure threshold. For example, when chlorpromazine is used at doses above 1000 mg/day, the incidence of seizures was reported to increase to 9%, in contrast to a 0.5% incidence when lower doses are taken.82 Fluphenazine, perphenazine, and trifluoperazine are the other CAPDs that can cause seizures in a dose-related manner. Haloperidol and molindone are among the CAPDs with a lower seizure risk.83 In a recent study, Alper et al. compared the incidence of seizures between (nonepileptic) patients randomized to an AAPD or placebo in the course of regulatory studies submitted to the FDA between 1985 and 2004.84 The incidence of seizures was higher among those randomized to clozapine and olanzapine, but there was no difference in seizure occurrence between patients randomized to placebo and the other four AAPDs. In nonepilepsy patients, clozapine has been reported to cause seizures in 4.4% when used at doses above 600 mg/day, whereas at a dose lower than 300 mg, the incidence of seizures has been found to be less than 1%.81 On the other hand, in patients with epilepsy, clozapine can increase the seizure frequency at any dose.85 Among the CAPDs, chlorpromazine and loxapine have the highest risk of seizure occurrence. Those with a lower seizure risk include haloperidol, molindone fluphenazine, perphenazine, and trifluoperazine at lower doses. Whether the presence of AEDs at adequate levels protects patients with epilepsy from breakthrough seizures on the introduction of APD with proconvulsant properties is yet to be established. Most APDs can cause EEG changes consisting of slowing the background activity above all when used at high doses. In addition some of these drugs, and particularly clozapine, can cause interictal sharp waves and spikes; this epileptiform activity, however, is not predictive of seizure occurrence.85 There are data suggesting

206

THE EPILEPSIES 3

that a severe disorganization of the EEG recordings is a more likely predictor of seizure occurrence. As a rule, any APD should be started at low doses and should undergo slow-dose increments to minimize the risk of seizures in patients with epilepsy. Pharmacokinetic Interaction Between AEDs and APD Induction of hepatic enzymes on the introduction of AEDs with enzyme-inducing properties (such as phenobarbital, primidone, carbamazepine, and phenytoin) may result in an increase of the clearance of most APDs. This is, in fact, the most frequent and clinically relevant pharmacokinetic interaction encountered in clinical practice. It may potentially result in recurrence of psychotic symptoms previously controlled at higher serum concentrations of APD. By the same token, discontinuation of an AED with enzyme-inducing properties may result in a decrease in the clearance of APDs, which in turn can lead to toxicity, such as extrapyramidal adverse events caused by an increase of the serum concentrations of CAPDs. Finally, certain AEDs, such as valproic acid, can inhibit the glucuronidation process of AAPDs like clozapine. Pharmacodynamic Interactions Worsening of CNS-related adverse events, including sedation, ataxic gait, and cognitive disturbances, are the most frequent consequences of the pharmacodynamic interactions between APDs and AEDs with sedative properties.85 Clozapine and carbamazepine can cause leucopenia on their own, and their combination can worsen the severity of the leucopenia.85 In addition, the combination of these two drugs has been also reported to cause an increased risk of neuroleptic malignant syndrome. Thus, this combination should be avoided. In addition all six AAPDs can cause weight gain and type II diabetes mellitus, with olanzapine having the greatest impact on weight gain and aripiprazole the least. Thus, glycemias and lipid profiles need to be monitored in all patients on a regular basis, but particularly in patients taking AEDs that are associated with an increased risk of weight gain, such as valproic acid, gabapentin, pregabalin, and carbamazepine. Electroshock therapy (ECT) is not contraindicated in patients with severe POE and should be considered in patients with epilepsy with psychotic depressive or manic episodes that fail to respond to APD.8692 Blackwood et al. found that the incidence of seizures in patients following treatment with ECT was no higher than in the general population.87 In fact, several studies have shown that ECT increases the seizure threshold by 50 to 100%,86,89,92 Krystal and Coffey have suggested the use of unilateral ECT in the nondominant frontotemporal region.91 AEDs should be withheld in the mornings before a treatment, but otherwise should be kept at baseline doses.

CONCLUSIONS
Psychotic disorders are the least frequent psychiatric comorbidities in patients with epilepsy. They are seen most often in the setting of drug-resistant epilepsy. In contrast to primary psychotic disorders, epilepsy-associated psychoses tend to be more

14  Psychosis of Epilepsy

207

responsive to pharmacotherapy using antipsychotic drugs. Although some antipsychotic drugs can lower the seizure threshold, several antipsychotic drugs, both conventional and atypical, appear to be safe. Their appropriate use should never be withheld for fear of exacerbating seizures in patients with epilepsy. Started at low dosage and using gradual titration schedules, the risk of seizures can be minimized in both nonepileptic patients and in patients with epilepsy. REFERENCES
1. Temkin O. The Falling Sickness. A History of Epilepsy from the Greeks to the Beginnings of Modern Neurology. 2nd ed. Baltimore: The John Hopkins University Press; 1971. 2. Trimble MR. Psychosis of Epilepsy. New York: Raven Press; 1991. moire sur la folie ciculaire. Bull. Acad. Med. (Paris). 1854;19:382-400. 3. Falret JP. Me pilepsie larve e. Ann. Med. Psychol. 1873;(series 5)9:155-163. 4. Morel BA. Discussion sur lE 5. Kraepelin E. Psychiatrie. 8th ed. Barth: Leipzig; 1923. 6. Hill D. Psychiatric disorders of epilepsy. Med Press. 1953;229:473-475. 7. Pond DA. Psychiatric aspects of epilepsy. J Indian Med Prof. 1957;3:1442-1451. 8. Slater E, Beard AW, Glithero E. The schizophrenia-like psychosis of epilepsy. v. Discussion and conclusions. Br J Psychiatry. 1963;109:95-150. 9. Ferguson SM, Rayport M. Psychosis in epilepsy. In: Blumer D. Psychiatric Aspects of Epilepsy. Washington, DC: American Psychiatric Press; 1984:229-270. 10. Rayport M, Ferguson SM. Psychosis of epilepsy. In: Ettinger AB, Kanner AM, eds. Psychiatric Aspects of Epilepsy: A Practical Guide to Diagnosis and Treatment. Baltimore: Lippincott, Williams and Wilkins. In press. 11. Mendez MF, Grau R, Doss RC, Taylor JL. Schizophrenia in epilepsy: seizure and psychosis variables. Neurology. 1993;43:1073-1077. 12. Matsuura M, Adachi N, Muramatsu R, et al. Intellectual disability and psychotic disorders of adult epilepsy. Epilepsia. 2005;46(Suppl 1):11-14. 13. Hesdorffer DC, Hauser WA, Annegers JF, et al. Psychiatric diagnoses preceding unprovoked seizures in adults: a population-based case-control study. Epilepsia. 1992;33(Suppl 3):16. 14. Hesdorffer DC, Hauser WA. Epidemiological considerations. In: Ettinger AB, Kanner AM, eds. Psychiatric Issues in Epilepsy: A Practical Guide to Diagnosis and Treatment. 2nd ed. Baltimore: Lippincott Williams and Wilkins; 2007:1-16. 15. Jablinsky A, Sartorius N, Ernberg G, et al. Schizophrenia: manifestations, incidence and course in different cultures. A World Health Organization ten-country study. Psychol Med Monogr Suppl. 1992;20:1-97. 16. Vestergaard M, Pedersen CB, Christensen J, et al. Febrile seizures and the risk of schizophrenia. Schizophr Res. 2005;73:343-349. 17. Kanner AM. Psychosis of epilepsy: a neurologists perspective. Epilepsy & Behav. 2000;1:219-227. 18. Landoldt H. Some clinical electroencephalographical correlations in epileptic psychosis (twilight states). Electroencephalogr Clin Neurophysiol. 1953;5:121(abstract). 19. Tellenbach H. Epilepsie als Anfallseiden und als Psychose. Uber alternative psychosen paranoider Pragung bei forcierter Normalisierung (Landoldt) des Elektroencephalogramms Epileptischer. Nervarzt. 1965;36:190. 20. Perez MM, Trimble MR, Murray NMF, Reider I. Epileptic psychosis: an evaluation of PSE profiles. Br J Psychiat. 1985;146:155-163. 21. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Association; 1994. 22. Toone BK, Garralda ME, Ron MA. The psychosis of epilepsy and the functional psychosis: a clinical and phenomenological comparison. Br. J Psychiat. 1982;141:256-261. 23. Tarulli A, Devinsky O, Alper K. Progression of postictal to interictal psychosis. Epilepsia. 2001;42(11):1468-1471. 24. Kanner AM, Ostrovskaya A. Long-term significance of postictal psychotic episodes II. Are they predictive of interictal psychotic episodes? Epilepsy & Behav. 2008;12(1):154-156. 25. Kanner AM, Soto A, Gross-Kanner H. Prevalence and clinical characteristics of postictal psychiatric symptoms in partial epilepsy. Neurology. 2004;62:708-713. 26. Dongier S. Statistical study of clinical and electroencephalographic manifestations of 536 psychotic episodes occurring in 516 epileptics between clinical seizures. Epilepsia. 1959/1960;1:117-142.

208

THE EPILEPSIES 3 27. Kanner AM, Stagno S, Kotagal P, Morris HH. Postictal psychiatric events during prolonged video-electroencephalographic monitoring studies. Arch Neurol. 1996;53:258-263. 28. Lancman ME, Craven WJ, Asconape JJ, Penry JK. Clinical management of recurrent postictal psychosis. J Epilepsy. 1994;7:47-51. 29. Logsdail SJ, Toone BK. Postictal psychosis: a clinical and phenomenological description. Br J Psychiat. 1988;152:246-252. 30. Savard G, Andermann F, Olivier A, Remilliard GM. Postictal psychosis after complex partial seizures: a multiple case study. Epilepsia. 1991;32:225-231. 31. So NK, Savard G, Andermann F, Olivier A, Remillard GM. Acute postictal psychosis: a stereo-EEG study. Epilepsia. 1990;31:188-193. 32. Devinsky O, Abrahmson H, Alper K, et al. Postictal psychosis: a case control study of 20 patients and 150 controls. Epilepsy Res. 1995;20:247-253. 33. Kanemoto K, Kawasaki J, Kawai J. Postictal psychosis: a comparison with acute interictal and chronic psychoses. Epilespia. 1996;37:551-556. 34. Umbricht D, Degreef G, Barr WB, Lieberman JA, Pollack S, Schaul N. Postictal and chronic psychosis in patients with temporal lobe epilepsy. Am J Psychiatry. 1995;152:224-231. 35. Kanner AM, Ostrovskaya A. Long-term significance of postictal psychotic episodes I. Are they predictive of bilateral ictal foci? Epilepsy & Behav. 2008 Jan;12(1):150-153. 36. Alper A, Kuznieky R, Carlson C, et al. Postictal psychosis in partial epilepsy: a case-control study. Ann Neurol. 2008;63(5):602-610. 37. Marchetti RL, Tavares AG, Gronich G, Fiore LA, Ferraz RB. Complete remission of epileptic psychosis after temporal lobectomy: case report. Arq Neuropsiquiatr. 2001;59(3-B):802-805. 38. Kanemoto K, Kim Y, Miyamoto T, Kawasaki J. Presurgical postictal and acute interictal psychoses are differentially associated with postoperative mood and psychotic disorders. J Neuropsychiatry Clin Neurosci. 2001;13(2):243-247. 39. Alper K, Devinsky O, Westbrook L, et al. Premorbid psychiatric risk factors for postictal psychosis. J Neuropsychiatry Clin Neurosci. 2001;13(4):492-499. 40. Adachi N, Kato M, Sekimoto M, et al. Recurrent postictal psychosis after remission of interictal psychosis: further evidence of bimodal psychosis. Epilepsia. 2003 Sep;44(9):1218-1222. 41. Kanemoto K, Kawasaki J, Kawai J. Postictal psychosis: a comparison with acute interictal and chronic psychoses. Epilepsia. 1996;37:551-556. 42. Kaplan PW. Behavioral manifestations of nonconvulsive status epilepticus. Epilepsy Behav. 2002; 3(2):122-139. 43. Devinsky O, Kelly K, Porter WH. Clinical and electroencephalographic features of simple partial seizures. Neurology. 1988;43:1347-1352. 44. Sowa MV, Pituck S. Prolonged spontaneous complex visual hallucinations and illusions as ictal phenomena. Epilepsia. 1989;30:524-526. 45. Oishi M, Otsubo H, Kameyama S, et al. Ictal magnetoencephalographic discharges from elementary visual hallucinations of status epilepticus. J Neurol Neurosurg Psychiatry. 2003;74(4):525-527. 46. Sawchuk KS, Churchill S, Feldman E, Drury I. Status epilepticus amauroticus. Neurology. 1997;49:1467-1469. 47. Barry E, Sussman NM, Bosley TM, Harner RN. Ictal blindness and status epilepticus amauroticus. Epilepsia. 1985;26:577-584. 48. Olnes MJ, Golding A, and Kaplan PW. Nonconvulsive status epilepticus resulting from benzodiazepine withdrawal. Ann Intern Med. 2003;139:956-958. 49. Agathonikou A, Panayiotopoulos CP, Giannakodimos S, Koutroumanidis M. Typical absence status in adults: diagnostic and syndromic considerations. Epilepsia. 1988;39:1265-1276. 50. Panayiotopoulos CP. Syndromes of idiopathic generalized epilepsies not recognized by the International League Against Epilepsy. Epilepsia. 2005;46(Suppl 9):57-66. 51. Lim J, Yagnik P, Schraeder P, Wheeler S. Ictal catatonia as a manifestation of nonconvulsive status epilepticus. J Neurol Neurosurg Psychiatry. 1986;49:833-836. 52. Kanemoto K, Miyamoto T, Abe R. Ictal catatonia as a manifestation of de novo absence status epilepticus following benzodiazepine withdrawal. Seizure. 1999;8:364-366. 53. Schmitts B, Wolf P. Psychosis in epilepsy. In: Devinsky O, Theodore, WH, eds. Epilepsy and Behavior. New York: Wiley-Liss; 1991:97-128. 54. Wolf P, Trimble MR. Biological antagonism and epileptic psychosis. Br J Psychiat. 1985;146:272-276. 55. Gibbs FA. Ictal and non-ictal psychiatric disorders in temporal lobe epilepsy. J Nerv Ment Dis. 1951;113:522-528.

14  Psychosis of Epilepsy 56. Wolf P. Acute behavioral symptomatology at disappearance of epileptiform EEG abnormality; paradoxical or forced normalization. In: Smith D, Treiman D, Trimble M, eds. Neurobehavioral Problems in Epilepsy. Advances in Neurology, Vol. 55, New York: Raven Press; 1991:127-142. 57. Pakalnis A, Drake JK, Kellum JB. Forced normalization: acute psychosis after seizure control in seven patients. Arch Neurol. 1987;44:289-292. 58. Sander JWAS, Hart YM, Trimble MR, Shorvon SD. Vigabatrin and psychosis. J Neurol Neurosurg Psychiatry. 1991;54:435-439. 59. Ried S, Mothersill IW. Forced normalization: the clinical neurologists view. In: Trimble MR, Schmitz B, eds. Forced Normalization and Alternative Psychoses of Epilepsy. Bristol, PA: Wrightson Biomedical Publishing Ltd; 1998:77-94. 60. Wieser HG. Depth recorded limbic seizures and psychopathology. Neuroscience Biobehav Rev. 1983;7:427-443. 61. Perlo VP, Schwab RS. Unrecognized dilantin intoxication. In Locke S, ed. Modern Neurology. Boston: Little Brown; 1969:589-597. 62. Rivinus TM. Psychiatric effects of the anticonvulsant regimens. J Clin Psychopharmacol. 1982;2:165-192. 63. Kanner AM, Wuu J, Faught E, Tatum WO, Fix A, French JA. A past psychiatric history may be a risk factor for topiramate-related psychiatric and cognitive adverse events. Epilepsy & Behav. 2003;4: 548-552. 64. Mula M, Trimble MR, Yuen A, Liu RS, Sander JW. Psychiatric adverse events during levetiracetam therapy. Neurology. 2003 Sep 9;61(5):704-706. 65. Miyamoto T, Kohsaka M, Koyama T. Psychotic episodes during zonisamide treatment. Seizure. 2000 Jan;9(1):65-70. 66. Ketter TA, Marlow BA, Flamini R, et al. Anticonvulsant withdrawal: emergent psychopathology. Neurology. 1994;44:55-61. 67. Sironi VA, Franzini A, Ravaghati L, Marossero F. Interictal psychoses in temporal lobe epilepsy during withdrawal of anticonvulsant therapy. J Neurol Neurosurg Psychiatry. 1979;42:724-730. 68. Taylor DC. Mental state and temporal lobe epilepsy. Epilepsia. 1972;13:727-765. 69. Jensen I, Vaernet K. Temporal lobe epilepsy: Follow-up investigation of 74 temporal lobe resected patients. Acta Neurochirurgica. 1977;37:173-200. 70. Trimble MR. Behaviour changes following temporal lobectomy, with special reference to psychosis (editorial). J Neurol Neurosurg Psychiatry. 1992;55(2):89-91. 71. Shaw P, Mellers J, Henderson M, Polkey C, David AS, Toone BK: Schizophrenia-like psychosis arising de novo following a temporal lobectomy: timing and risk factors. J Neurol Neurosurg Psychiatry. 2004;75:1003-1008. 72. Leinonen E, Tuunainen A, Lepola U: Postoperative psychoses in epileptic patients after temporal lobectomy. Acta Neurol Scand. 1994;90(6):394-399. 73. Stevens JR: Psychiatric consequences of temporal lobectomy for intractable seizures: a 20 to 30-year follow-up of 14 cases. Psychol Med. 1990;20(3):529-545. 74. Mace CJ, Trimble MR: Psychosis following temporal lobe surgery: a report of six cases. J Neurol Neurosurg Psychiatry. 1991;54(7):639-644. 75. Andermann LF, Savard G, Meencke HJ, McLachlan R, Moshe S, Andermann F: Psychosis after resection of ganglioglioma or DNET: evidence for an association. Epilepsia. 1999;40(1):83-87. 76. Carran MA, Kohler CG, OConnor MJ, Bilker WB, Sperling MR: Mania following temporal lobectomy. Neurology. 2003;61:770-774. 77. Christodoulou C, Koutroumanidid M, Hennessy MJ, et al. Postictal psychosis after temporal lobectomy. Neurology. 2002 Nov 12;59(9):1432-1435. 78. Manchanda R, Miller H, McLachlan RS: Post-ictal psychosis after right temporal lobectomy. J Neurol Neurosurg Psychiatry. 1993;56(3):277-279. 79. Reutens DC, Savard G, Andermann F, et al. Results of surgical treatment in temporal lobe epilepsy with chronic psychosis. Brain. 1977;120:1929-1936. 80. Glosser G, Zwil AS, Glosser DS, et al. Psychiatric aspects of temporal lobe epilepsy before and after anterior temporal lobectomy. J Neurol Neurosurg Psychiatry. 2000;68:53-58. 81. Stahl SM. Antipsychotic agents. In: Stahl SM, ed. Essential Pharmacology: Neuroscientific Basis and Practical Applications. 2nd ed. New York: Cambridge University Press; 2000:401-458. 82. Logothetis J. Spontaneous epileptic seizures and EEG changes in the course of phenothiazine therapy. Neurology. 1967;17:869-877. 83. Whitworth AB, Fleischhacker WW. Adverse effects of antipsychotic drugs. Int Clin Psychopharmacol. 1995;9(Suppl 5):21-27.

209

210

THE EPILEPSIES 3 84. Alper K, Schwartz KA, Kolts RL, Khan A. Seizure incidence in psychopharmacological clinical trials: an analysis of Food and Drug Administration (FDA) summary basis of approval reports. Biol Psychiatry. 2007;15;62(4):345-354. 85. Zaccara G, Messori A, Cincotta M. Clinical studies of pharmacodynamic interactions between antiepileptic drugs and other drugs. In: Majkowski J, Bourgeois B, Patsalos P, Mattson R, eds. Antiepileptic Drugs: Combination Therapy and Interactions. Cambridge, UK: Cambridge University Press; 2005:241-254. 86. Post R, Putnam F, Uhde T. Electroconvulsive therapy as an anticonvulsant: implications for its mechanisms of action in affective illness. In: Malitz S, Sackeim H, eds. Electroconvulsive Therapy: Clinical and Basic Research Issues. New York: New York Academy of Sciences; 1986. 87. Blackwood DHR, Cull RE, Freeman CP, et al. A study of the incidence of epilepsy following ECT. J Neurol Neurosurg Psychiatry. 1980;43:1098-1102. 88. Abrams R. Electroconvulsive therapy in the high-risk patient. In: Abrams R. Electroconvulsive Therapy. New York: Oxford University Press; 1997:81-113. 89. Sackeim HA. The anticonvulsant hypothesis of the mechanisms of action of ECT: current status. The Journal of ECT. 1999;15:5-26. 90. Viparelli U, Viparelli G. ECT and grand mal epilepsy. Convulsive Ther. 1992;8:39-42. 91. Coffey CE, Lucke J, Weiner RD, et al. Seizure threshold in electroconvulsive therapy (ECT). II. The anticonvulsant effect of ECT. Biol Psychiatry. 1995;37:777-788. 92. Fink M, Kellner C, Sackheim HA. Intractable seizures, status epilepticus and ECT. JECT Lett. 1999;15:282-284.

THE EPILEPSIES 3

15

Sudden Unexpected Death in Epilepsy


FERGUS J. RUGG-GUNN  LINA NASHEF

Definition Epidemiology Risk Factors Demographics Epilepsy Characteristics Antiepileptic Medication Perimortem Features Other Features

Pathophysiology of SUDEP Cerebrogenic Autonomic Control Cardiac Mechanisms Respiratory Mechanisms Suppression of Cerebral Activity SUDEP and Epilepsy Surgery Implications for Management

Patients with epilepsy have an increased mortality rate compared to the general population, to which sudden unexpected death in epilepsy (SUDEP) is a major contributor. Some groups of patients are at greater risk of SUDEP than others, although the reasons for this are only partly understood. A large number of risk factors have been proposed, and there are numerous theories as to the pathophysiological basis of SUDEP. This chapter explores the evidence behind the recognized risk factors to determine who is most at risk of SUDEP and examines theories surrounding proposed mechanisms of SUDEP and how these mechanisms are integrated with substantiated risk factor data. Areas of possible future research are also discussed.

Definition
All-cause mortality rates in patients with epilepsy are approximately two to three times higher than the general population and are age dependent.13 The major contributors of increased mortality are the causes of epilepsy, for example, traumatic brain injury, cerebrovascular disease, brain tumors, and accidents (mostly falls and drowning) or status epilepticus and SUDEP. In contrast, treatment-related mortality, from epilepsy surgery or the adverse effects of antiepileptic medication, is very rare.4 SUDEP is defined as the sudden, unexpected, witnessed or unwitnessed, nontraumatic, and nondrowning death in patients with epilepsy with or without evidence for a seizure and excluding documented status epilepticus, in which postmortem examination does not reveal a structural or toxicologic cause for

211

212

THE EPILEPSIES 3

death.5 Where autopsy is not performed, and for the purpose of epidemiological studies, sudden death occurring in benign circumstances with no known competing cause for death is classified as probable SUDEP. Cases classified as possible SUDEP are usually not included in epidemiological studies of SUDEP incidence. These are cases in which SUDEP cannot be excluded because either the information regarding circumstances of death is too limited to confirm classification as a probable or definite SUDEP or that adequate or complete documentation is available but there is a plausible competing explanation for the death.3 It must be stressed that this defines a category of sudden death in epilepsy, and not a condition, a concept that is sometimes overlooked. Despite an applicable definition, and clear guidance where there is uncertainty, significant variability in use has hampered efforts to integrate findings from multiple studies on epidemiological and risk factor data and hence establish common relevant factors.6,7

Epidemiology
Sudden unexpected death in the general population is extremely rare in young adults with an incidence of 5 to 10/100, 000 person-years, whereas the rate climbs steeply with advancing age to approximately 300/100, 000 person-years in the elderly.8 The incidence of sudden death in patients with epilepsy is significantly higher and varies markedly with the population studied.9 For example, in population-based studies, the incidence has been reported to be 0.35 and 2.7/1000 person-years, depending on the methodologies employed.10,11 This increases to between 2 and 5.9/1000 person-years in cohorts of patients attending specialist epilepsy clinics,1214 3.4/1000 person-years in pupils with epilepsy enrolled in a special residential school,15 and up to 9.3/1000 person-years in epilepsy surgery candidates.16,17 The incidence of sudden death in young adults with intractable epilepsy is, therefore, many times that of the general population, with a peak between the ages of 20 and 40 years.18 In older age groups, the relative increased incidence of SUDEP is too small to measure and is confounded by the occurrence of comorbidity such as cardiovascular, respiratory, or cerebrovascular disease. This makes it difficult to ascribe a sudden death to a pure SUDEP category but does not exclude the possibility that there is no additive risk from the epilepsy. There is limited data available on the incidence of SUDEP in children. An incidence rate of approximately 0.1 to 0.2/1000 person-years has been estimated from a number of earlier studies.19,20 Significant methodological limitations exist, however, including inadequate follow-up periods, difficulties with case-ascertainment, and assumptions regarding the prevalence of epilepsy in community-based studies. A more recent retrospective study with an 18-year follow-up period confirmed a low incidence rate of approximately 0.4/1000 person-years, despite following a cohort of children with refractory epilepsy and learning disability.21 This is approximately 10 times lower than the incidence rate for similarly affected adults; the rate for children with uncomplicated epilepsy will be lower still.22 The reason for the discrepancy between children and adults is unknown and is unlikely to be entirely due to methodological constraints, but may be as a result of a different cardiorespiratory response from a developing brain compared to a mature brain, a more intensively supervised environment, the inclusion of specific self-limiting pediatric epilepsy syndromes, lack of comorbidity, or possibly, greater autonomic stability.23

15  Sudden Unexpected Death in Epilepsy

213

Risk Factors
There is significant debate regarding risk factors for SUDEP. A large number of variables that may influence the risk of SUDEP have been proposed, and the significance of each has been discussed at length without clear consensus. Relevant and independent risk factors are difficult to establish given the nonindependence of patient, syndrome, seizures, and treatment characteristics. Multiple logistic regression analyses require large cohorts of patients to achieve statistical significance for each of the variables evaluated, and this is difficult to attain.24 Furthermore, the high variability between studies in terms of patient cohorts, definition, choice of control group, methodology, and overall study quality precludes not only a valid metaanalysis, but even a simple meaningful comparison. In a critical review of the literature regarding risk factors for SUDEP, Tellez-Zenteno noted that a clear definition of SUDEP was stated in only 65% of relevant studies. Furthermore, the low frequency of postmortem data was evident in many studies. This lack of consistency significantly undermines the suitability of such studies to comment on SUDEP causality and risk.7 The preliminary studies of risk factors in SUDEP were descriptive, providing direction for more meaningful case-control studies. The value of these observations, otherwise, is limited, however, because, in general, these studies lack suitable control groups and comprise small numbers of an often highly selected patient population. In case-control studies, the choice of control group is important and is dependent on the risk factors to be evaluated. A control group of matched living patients with epilepsy tends to favor the assessment of risk factors exploring lifestyle and clinical issues such as seizure frequency and antiepileptic drug use. A control group of nonSUDEP deaths in epilepsy is more suitable for exploring circumstances of death, such as body position, place of death, and seizures immediately prior to death. A number of studies assessing risk factors used a case-control design (Table 15-1), this often being the only option for the study of relatively uncommon occurrences, such as SUDEP. However, it also has important limitations. For example, case-control studies traditionally evaluate the probability of being exposed to a risk factor, not the risk of developing a condition. Moreover, case-control studies are subject to several sources of bias, such as selection bias of cases and controls and information and recall bias when collecting data on exposure. These biases are particularly important where an objective definition of the condition of interest is not consistently applied, as is the case in the SUDEP literature. An attempt has been made to resolve the SUDEP risk factor literature and disentangle the often contradictory results on the basis of extensive literature reviews and the implementation of a study-validity scoring system and calculation of relative risk factor ratios.6 Monte and colleagues stratified studies evaluating risk factors on the basis of fulfillment of a number of variables that were considered to be markers of good quality studies. However, the scoring system used was unvalidated and potentially misleading, with, for example, equal credence attributed to controlled and descriptive studies. Papers not achieving an arbitrary threshold were excluded, and the possibility of the same study population being used more than once was not adequately addressed.6 A more simplistic approach was used by Tellez-Zenteno, in which percentages of studies with a positive risk factor were reported. Risk factors that were reported to be significant in more than 50% of studies were a terminal

214
THE EPILEPSIES 3

TABLE 151

Case-control and Nested Case-control Studies Examining Risk Factors for SUDEP
Type of Study No. of Cases/ Controls Postmortem Performed Risk Factor Outcome (significant) Risk Factor Outcome (not significant)

Study Reference

Birnbach et al, 199139 Timmings, 199312 Leestma et al, 199732 George and Davis, 199847 Kloster and Engelskjon, 199936 Opeskin et al, 199944 Nilsson et al, 1999,30 200145

Retrospective case-control Retrospective nested case-control Retrospective nested case-control Retrospective, case-control Retrospective case-control

25/23 (non-SUDEP) 14/1820 (living) 20/19 (non-SUDEP)/ 2988 (living) 52/63 (non-SUDEP) 42/37 (non-SUDEP)

unknown

nil

nil 17/20 yes yes

Use of CBZ, male, history of GTCS Younger age, shorter duration of epilepsy

Gender, age, duration of epilepsy, IQ, GTCS within the preceding month, age of onset Number of AEDs, age, duration of epilepsy, seizure frequency Gender, LMT usage or dose

Retrospective case-control Retrospective nested case-control

44/44 (non-SUDEP) 57/171 (living)

yes 52/57 cases

Subtherapeutic levels of nil AEDs Early onset of epilepsy, Gender, age, duration of evidence of terminal epilepsy, structural brain seizure, prone position, lesions, seizure frequency, low IQ, number of AEDs, specific AEDs, nil AED levels, treatment with CBZ and PHY Increasing seizure Gender, specific AEDs (1999), symptomatic epilepsy frequency, increasing numbers of AEDs, frequent AED changes, early onset of epilepsy, use of psychotropic medication, high CBZ levels (2001)

Opeskin, 2000162 Walczak et al, 200137 Schnabel et al, 200233

Prospective, case-control Prospective case-control Retrospective case-control

10/29 (non-SUDEP) 20/80 (living) 39/102 (non-SUDEP)

yes 10/20 cases yes

nil GTCS, low IQ, multiple AEDs, long duration of epilepsy Young age, early onset of epilepsy, shorter duration of epilepsy, higher seizure frequency Younger age, female, evidence of terminal seizure, found in bed at home,

Prolactin levels Age, gender, specific AEDs, compliance, psychotropic drugs, structural brain lesions Gender, geomagnetic influence, history of GTCS

Prospective Opeskin and case-control Berkovic, 200334

50/50 (non-SUDEP)

yes

Langan et al, 200531

Retrospective, case-control

154/616 (living)

Yes

Vlooswijk et al, 200735

Retrospective, case-control

29/104 (non-SUDEP)

5/29

Hitiris et al, 200718

Retrospective, case-control

62/124 (living)

6/62

Duration of epilepsy, types of seizures, etiology of epilepsy, seizure frequency (but variance larger in SUDEP group), low IQ, psychiatric illness, number of AEDs, specific AEDs, compliance with AEDs Generalized seizures, high Gender, compliance with AEDs number of convulsive seizures in previous 3 months, >3 AEDs previously taken, current use of CBZ, supervision at night (protective), asthma (protective) Younger age, earlier Gender, seizure frequency, onset of epilepsy, seizure type, EEG data, low shorter duration of IQ, psychiatric illness, specific epilepsy, less AEDs, number of AEDs, comorbidity evidence of terminal seizure Longer duration of History of GTCS, use of CBZ, epilepsy, seizure in multiple AEDs, seizure within last 12 months the last 3 or 6 months

15  Sudden Unexpected Death in Epilepsy

215

216

THE EPILEPSIES 3

seizure, subtherapeutic antiepileptic drug (AED) levels, high seizure frequency, and a high number of AEDs.7 DEMOGRAPHICS Descriptive studies have almost universally reported that patients with SUDEP are young adults.10,11,13,14,2529 A number of biases exist, however, including, as discussed earlier, the exclusion of patients with significant comorbidity associated with increasing age, such as ischemic heart disease or cerebrovascular disease, identified on postmortem examination.13,25,29 Other examples of bias include case identification through self-referral by bereaved relatives, most commonly parents,27 and studies with only small numbers of patients.10,13 Case-control studies are less conclusive. Some studies only included defined age groups and can draw no conclusions regarding other age groups. Nevertheless, it is interesting to note that 70 to 80% of the studied population in a number of case-control studies were less than 45 years old.18,30 Data regarding age, however, is not available from a number of large studies due to age matching of control subjects.18,30,31 Of the remaining studies, the use of a cohort of non-SUDEP deaths as a control group may bias the patient group toward a younger age due to exclusion of comorbid conditions more commonly associated with advancing age, such as atherosclerosis, thromboembolism, and metastatic carcinoma,3235 although young age as an independent risk factor has not been universally reported.36,37 The likelihood of selection bias is corroborated by finding significantly less comorbidity in the SUDEP group than the non-SUDEP group.35 In studies using living control subjects, younger age was not seen more frequently in the SUDEP group, although numbers of SUDEP patients were small.12 Although a large number of descriptive studies have suggested that male gender is a significant risk factor for SUDEP,11,13,26,29,38 this has not been confirmed by the vast majority of case-control studies.3033,3537,39 In addition, a small number of both descriptive and case-control studies have reported a significantly increased standardized mortality rate in female patients, which may be attributable to a lower background rate of death in the female non-SUDEP control group.10,34 In summary, although not universally accepted, the weight of evidence at the present time suggests that sex is not a strong risk factor for SUDEP. With regard to age, incidence appears lower in children than young adults. Methodological constraints, selective definition, coexisting pathology, the absence of suitable control subjects, and lack of controlled prospective data do not allow any conclusions to be drawn regarding incidence of sudden death related to epilepsy in older age groups. EPILEPSY CHARACTERISTICS A number of case-control studies have suggested that early onset of epilepsy is a significant risk factor for SUDEP.30,33,35,36 For example, Nilsson reported an eightfold higher SUDEP risk in patients with an onset of epilepsy between the ages of 0 and 15 years, compared to patients with seizure onset after 45 years of age.30 However, whereas this may reflect a different etiological basis for the epilepsy, it may also merely be a surrogate marker for an increased cumulative lifetime risk of having seizures for a longer period of time, as suggested by other studies.18,28,29 Conversely, several reports give a shorter duration of epilepsy being associated with an increased

15  Sudden Unexpected Death in Epilepsy

217

risk of SUDEP, although this is most likely as a result of comparison with an older control population.32,33,35 Furthermore, following conditional multiple logistic regression analysis, Walczak showed that a long duration of epilepsy (>30 years) was no longer a risk factor after adjustment for seizure frequency.37 One would expect epilepsy syndrome to be a key factor in defining the risk of SUDEP. Yet, there is only limited evidence to support the association of epilepsy syndrome with an increased risk of SUDEP.30,36 Discordant results from the relatively few case-control studies that assessed this risk factor and low numbers of patients in each group preclude detailed evaluation or definitive conclusions.34 In the study reported by Nilsson, 7 out of 57 (12%) SUDEP cases had primary generalized epilepsy compared to 12 out of 171 (7%) control subjects. Statistical comparison revealed that there was a higher risk of SUDEP in patients with primary generalized epilepsy compared to patients with focal, symptomatic epilepsy, although this was only significant in men.30 Nevertheless, although idiopathic primary generalized epilepsy (IGE) is usually less refractory to treatment, individuals with IGE are well represented in SUDEP cohorts. It is possible that specific epilepsy syndrome subtypes carry an increased risk of sudden death due to phenotypic expression in other cerebral and possibly cardiac structures. For example, Rett syndrome, typically due to mutations in the MECP2 gene, is associated with brain stem immaturity and autonomic, particularly respiratory, instability in association with epilepsy. In addition to respiratory abnormalities, which include apneustic breathing and hyperventilation, patients with Rett syndrome may also present with a prolonged QT interval and reduced heart rate variability.40 The coexistence of cardiac arrhythmia and central apnea may act synergistically in the development of sudden death. Severe myoclonic epilepsy in infancy (SMEI) also appears to be associated with an increased risk of sudden death, although the pathophysiological mechanism remains unclear. A recent report of a pedigree with an SCN1A mutation and two cases of SUDEP is of interest in this respect.42 Although some of these conditions represent the most severely affected group of patients and the high mortality rate may be multifactorial, it is possible that more subtle genetic abnormalities, such as channelopathies in, for example, IGE, may also predispose patients to SUDEP. In this regard, it is important to note that a number of functional cardiac conduction abnormalities, such as long QT syndrome, are also channelopathies, and genetic susceptibility is also well documented in sinus node dysfunction and bradyarrhythmias.43 However, no epidemiologic data indicate a higher incidence of epilepsy among relatives of patients with inherited susceptibility to arrhythmia, and a family history of early sudden cardiac death is not reported in SUDEP series. Furthermore, malignant tachyarrhythmias are relatively uncommon in seizures, and ictal respiratory changes have been documented in the absence of cardiac abnormalities. The proposed cardiorespiratory mechanisms of SUDEP will be discussed in more detail later in the chapter. Clearly, scope exists for potentially useful epidemiologic studies looking at the incidence of epilepsy in families of sudden cardiac death victims; additionally, it may be helpful to study the incidence of syncope in patients with idiopathic epilepsy and their relatives. Controversy exists on whether high seizure frequency is an independent risk factor for SUDEP. Several descriptive and large case-control studies have reported an increased risk of SUDEP in patients experiencing frequent seizures.28,30,31,33,37,38 This increased risk is most marked for convulsive seizures1012,27,28,31,37 rather than nonconvulsive episodes, such as complex partial seizures.33 Moreover, on logistic

218

THE EPILEPSIES 3

regression analysis, Walczak noted that only the frequency of convulsive seizures was relevant, and not the frequency of all seizures combined.37 Conversely, high seizure frequency was not an independent risk factor in a number of other reports, although a number of methodological issues exist.12,29,34,36 For example, in a retrospective case-control study of 42 patients with SUDEP, reported by Kloster and Engelskjon, there was no reported difference in seizure frequency between the SUDEP and non-SUDEP control group. The study was undertaken at a tertiary referral center with both groups having chronic refractory epilepsy and frequent seizures.36 Other negative studies may have been similarly influenced.34 Intuitively, the severity of convulsive seizures may also be important in SUDEP, but this is more challenging to quantify and hence has not been evaluated as a risk factor. In summary, with regard to epilepsy characteristics, the most important factor, on the basis of currently available data, is the frequency of convulsive seizures. Data with respect to seizure severity does not exist at the present time, whereas data regarding the influence of the pathogenesis of the underlying epilepsy and syndromic diagnosis are also limited to a small number of specific syndromes and requires further study. ANTIEPILEPTIC MEDICATION The number of antiepileptic medications (AEDs) taken concomitantly have been reported to be an independent risk factor for SUDEP,38 even after correction for seizure frequency.30,37 This is not universally reported, however,12,18,3436 although small numbers of patients and a high frequency of polytherapy in control subjects may be contributory in these negative studies. Langan found that the risk of SUDEP increased with the number of AEDs previously taken, despite correction for seizure frequency, perhaps a surrogate for epilepsy severity. In addition, risk was also increased in those who had never been on AEDs. This is potentially an important group and needs further clarification. It may include those in whom the epilepsy is considered mild and treatment is not recommended, those who decline treatment, or those with recent onset epilepsy who have not yet been assessed or offered treatment. Risk of SUDEP is also increased in those whose treatment history was unclear, which may reflect the risk associated with the lack of treatment and uncontrolled seizures, although the reason for this was not objectively assessed.31 Despite several descriptive studies suggesting that subtherapeutic levels of AEDs are a risk factor for SUDEP,10,11,25,29 this has not been corroborated by the majority of case-control studies,34,44,45 as it is difficult to study as an independent factor. Of note is that postmortem levels of AEDs may not accurately reflect antemortem levels possibly due to, for example, redistribution and continuing metabolism.46 In a postmortem study reported by George and Davis, however, so-called subtherapeutic drug levels were detected in 69% of the 52 cases of SUDEP, in 75% of the eight cases where a seizure precipitated an accident causing death, and in 34% of the control population.47 This suggests that the increased likelihood of a seizure associated with subtherapeutic AED levels, rather than the levels per se, drives the observed elevated mortality rate. Compliance with AED treatment was first proposed as a risk factor for SUDEP in an uncontrolled study by Leestma, who found subtherapeutic AED levels in 68% of SUDEP cases.25 Therapeutic drug monitoring has traditionally been considered a surrogate for medication adherence, although due

15  Sudden Unexpected Death in Epilepsy

219

to the existence of a number of confounding factors, it is clear that the two terms are not interchangeable. For example, in patients with uncontrolled seizures, changes of dose and type of medication are commonplace, and serum levels will not be stable and may frequently be subtherapeutic, despite excellent compliance. More recent studies have attempted to address this by integrating additional clinical information, such as a previous history of noncompliance29 or an arbitrary judgment regarding the degree of compliance made by the treating physician.34 Despite this, conflicting results have been obtained (Table 15-2). Of paramount importance, however, is the understanding that the evaluation of drug levels or noncompliance as independent risk factors for SUDEP must take into account the presence, frequency, and severity of seizures. The issue of variability of AED use was recently addressed in a study by Williams and colleagues comparing hair AED concentration variability in patients with SUDEP, non-SUDEP epilepsy-related deaths, epilepsy outpatients and epilepsy inpatients. The SUDEP group showed greater hair AED concentration variability than either the outpatient or the inpatient groups, reflecting variable AED ingestion over time. However, these variations cannot distinguish prescribed changes from poor compliance or identify consistent noncompliance over time. Second, it does not provide information on drug-taking behavior immediately before death, as it takes about 5 days for a drug sequestrated into the follicle to appear at the scalp; therefore short-term noncompliance immediately before death is not assessed by this study and may have been overlooked.48 In summary, therefore, despite the existence of some evidence for serum or hair sample AED variability in SUDEP patients, the issue of AED compliance is far from being satisfactorily resolved. Perhaps one useful conclusion that can be drawn from the studies available is that none has shown a higher risk from therapeutic levels compared to those with subtherapeutic levels. Despite a number of descriptive and controlled studies, no specific AED has been clearly associated with an increased risk of SUDEP,18,29,32,36,37,44,49 although a small number of studies have implicated treatment with carbamazepine as an independent risk factor.31,50,51 For antiepileptic medication in general, proposed mechanisms include perturbed heart rate variability, lengthening of the Q-T interval on the electrocardiogram combined with a mild proarrhythmic effect of epileptic seizure discharges, or excessive postseizure brainstem inhibition producing a blunting or transient abolition of the central hypoxic and hypercarbic respiratory drive, with consequent postictal respiratory arrest.5052 Elevated serum levels of carbamazepine have been associated with an increased risk of SUDEP, even after adjustments for seizure frequency have been made. Frequent drug changes and multiple concomitant AEDs, conventional markers of severe, and unstable epilepsy increased this risk synergistically.45 On this basis, it is difficult to know whether a high carbamazepine level is an independent risk factor or is merely representative of challenging epilepsy. In summary, AED-related factors are likely to be important in relation to SUDEP risk, but the evidence available does not allow us to differentiate risks from uncontrolled seizures and challenging epilepsy from those related to specific drugs, drug combinations, and drug changes, either prescribed or otherwise. PERIMORTEM FEATURES There is evidence from both descriptive and controlled studies that a terminal convulsive seizure,11,14,25,27,29,34,36,53 being found alone in bed,14,2628,34,36 and being

220

TABLE 152

Risk Factors Evaluated by Descriptive and Case-control Studies


THE EPILEPSIES 3

Descriptive Cohort Studies Proposed Risk Factor Positive Negative Positive

Case-control Studies Negative

Young age

Male gender

Leestma et al Schwender et al 198626 Leestma et al 198911 Lip and Brodie 199213 Nashef et al 199514 Nashef et al 199827 Ficker et al 199810 Opeskin et al 200028 Lear-Kaul et al 200529 Schwender and Troncoso 198626 Lip and Brodie 199213 Tennis et al 199538 Lear-Kaul et al 200529 Leestma et al 198911

198425

Leestma et al Schnabel et al 200233 Opeskin and Berkovic 200334 Vlooswijk et al 200735

199732

Birnbach et al 199139 Timmings 199312 Kloster and Engelskjon 199936 Walczak et al 200137

Nashef et al 199514 Nashef et al 199827 Nilsson et al 200316 Ficker et al 199810

Timmings et al 199312

Long duration of epilepsy

Opeskin et al 200028 Lear-Kaul et al 200529

Nilsson et al 199930, 200145 Hitiris et al 200718

Birnbach et al 199139 Leestma et al 199732 Kloster and Engelskjon 199936 Nilsson et al 1999,30 200145 Walczak et al 200137 Schnabel et al 200233 Opeskin and Berkovic 200334 (female is sig.) Langan et al 200531 Vlooswijk et al 200735 Birnbach et al 199139 Timmings 199312 Leestma et al 199732 (shorter duration is sig.) Kloster and Engelskjon 199936 Walczak et al 200137 Schnabel et al 200233 (shorter duration is sig.) Opeskin and Berkovic 200334 Vlooswijk et al 200735 (shorter duration is sig.)

Early onset of epilepsy High seizure frequency Tennis et al 199538 Opeskin et al 200028 Lear-Kaul et al 200529 Lip and Brodie 199213

Seizure typeGTCS

Nashef et al 199827 Ficker et al 199810 Leestma et al 198911 Opeskin et al 200028 Ficker et al 199810 Leestma et al 198425 Leestma et al 198911 Nashef et al 199514 Nashef et al 199827 Lear-Kaul et al 200529 Tennis et al 199538 (in lifetime) Ficker et al 199810

Kloster and Engelskjon 199936 Nilsson et al 199930, 200145 Schnabel et al 200233 Vlooswijk et al 200735 Nilsson et al 199930, 200145 Walczak et al 200137 (GTCS only) Schnabel et al 200233 Langan et al 200531 Timmings 199312 Walczak et al 200137 Langan et al 200531 Kloster and Engelskjon 199936 Nilsson et al 199930 Kloster and Engelskjon 199936 Opeskin and Berkovic 200334

Timmings 199312 Kloster and Engelskjon 199936 Opeskin and Berkovic 200334 Vlooswijk et al 200735 Schnabel et al 200233 Opeskin and Berkovic 200334 Vlooswijk et al 200735 Hitiris et al 200718 Opeskin and Berkovic 200334
15  Sudden Unexpected Death in Epilepsy

Primary generalized epilepsy Evidence of terminal seizure

Multiple AEDs

Nilsson et al 1999,30 200145 Walczak et al 200137 Langan et al 200531 (>3 in lifetime) Lip and Brodie 199213 Schwender and Troncoso 198626 George and Davis 199847

Subtherapeutic AED levels

Leestma et al 198425 Leestma et al 198911 Ficker et al 199810 Lear-Kaul et al 200529

Timmings 199312 Kloster and Engelskjon 199936 Opeskin and Berkovic 200334 Vlooswijk et al 200735 Hitiris et al 200718 Opeskin et al 199944 Nilsson et al 200145 (high levels of CBZ sig) Opeskin and Berkovic 200334 Table continued on following page

221

222
THE EPILEPSIES 3

TABLE 152

Risk Factors Evaluated by Descriptive and Case-control Studies (Continued)


Descriptive Cohort Studies Case-control Studies Positive Negative
12

Proposed Risk Factor Positive

Negative

Specific AED use

Lear-Kaul et al 2005

Timmings 1993 (CBZ) Langan et al 200531 (CBZ)

Low IQ

Leestma et al 198911 Nashef et al 199514 Opeskin et al 200028 Leestma et al 198425 Leestma et al 198911 Schwender and Troncoso 198626 Nashef et al 199514 Nashef et al 199827 Opeskin et al 200028 Lear-Kaul et al 200529 Lear-Kaul et al 200529 Tennis et al 199538 Opeskin et al 200028

Nashef et al 199827

Walczak et al 200137

Symptomatic epilepsy/structural brain lesion Found in bed

Lear-Kaul et al 200529 Schwender and Troncoso 198626 Leestma et al 198425 Ficker et al 199810 Leestma et al 198911

Kloster and Engelskjon 199936 Leestma et al 199732 (LMT) Opeskin et al 199944 (CBZ & PHY) Nilsson et al 199930, 200145 Walczak et al 200137 Opeskin and Berkovic 200334 Vlooswijk et al 200735 Hitiris et al 200718(CBZ) Birnbach et al 199139 Opeskin and Berkovic 200334 Kloster and Engelskjon 199936 Vlooswijk et al 200735 Kloster and Engelskjon 199936 Nilsson et al 199930, 200145 Walczak et al 200137

Kloster and Engelskjon 199936* Opeskin and Berkovic 200334

Prone position Psychiatric illness/ psychotropic medication

Nashef et al 199827

Kloster and Engelskjon 199936 Nilsson et al 1999,30 200145 Walczak et al 200137 Opeskin and Berkovic 200334 Vlooswijk et al 200735

sig. = significant *Statistical comparison not made with control group, calculation based on expected number of deaths during period of sleep of 8 hours duration/24 hours.

15  Sudden Unexpected Death in Epilepsy

223

in the prone position29,36 are independent risk factors for SUDEP. Whereas a small number of descriptive studies have not found an association (Table 15-2), all casecontrol studies that have evaluated these factors have found a positive relationship with the risk of SUDEP. For example, the descriptive study reported by Ficker failed to find an association between being in bed and SUDEP or evidence for a terminal seizure, although only nine patients were evaluated.10 In a report by Nashef, following interviews with bereaved relatives, evidence for a terminal seizure was found in 24 of 26 cases, but it is of interest that only two were witnessed. The observation that, in most studies, unwitnessed cases far outnumber those witnessed suggests that enhanced surveillance of patients with epilepsy may be protective.27 This is corroborated by a study of young patients with epilepsy at a special residential school. All sudden deaths that occurred during the period of the study were when the pupils were not under the close supervision of the school, and most were unwitnessed.15 Similar findings of a protective effect of enhanced supervision at night were also found in a large controlled study, where supervision was defined as the presence in the bedroom of an individual of normal intelligence and at least 10 years old or the use of special precautions, such as checks throughout the night or the use of a listening device.31 In some cases when a prone position was not observed, other factors that might compromise breathing were identified. For example, Nashef noted that only five of 26 people were found facedown in the pillow and a sixth with the head in carpet pile. In total, however, in 11 of 26 cases, an extrinsic or intrinsic positional obstruction to breathing amenable to intervention may have contributed.27 Moreover, it is possible that this may be an underestimate, as obstructive apnea can occur in an apparently benign position.54 OTHER FEATURES There is limited evidence for an independent relationship between learning disability and an increased risk of SUDEP. Early descriptive and population-based studies, in which learning disability was determined by observer impressions rather than by formal IQ examination, provided only weak support for this association.11,55 Most recent studies have found no clear correlation.27,3436,39 However, Walczak identified an IQ of less than 70 to be a risk factor for SUDEP, even after accounting for seizure frequency.37 Observational studies are likely to be biased. Nashef found that five of 11 patients with SUDEP had a low IQ, but the study was undertaken at a tertiary referral center with a higher-than-average background incidence of learning disability.14 Similar bias was introduced in an observational study by Opeskin, where all patients in the study region who died in institutions for the disabled were mandatory coroners cases, thus artificially elevating SUDEP cases with learning disability.28 It has been postulated that patients with learning disability are more susceptible to central apnea and positional asphyxia that may cause SUDEP as a result of prolonged postictal encephalopathy,56 decreased postictal respiratory drive, and impaired movement and righting reflexes.37 This is likely to reflect associated pathology, rather than the learning difficulty per se. Despite early reports of an increased incidence of structural lesions in patients with SUDEP,11,25,57 this has not been confirmed by more recent controlled studies.30,36,37 Although there is evidence that psychotropic medication can influence the risk of sudden death in general, there is no convincing evidence of this being particularly relevant in SUDEP.

224

THE EPILEPSIES 3

Logistic regression analysis on 18 patients with SUDEP suggested an independent association of the risk of SUDEP with the number of concomitant psychotropic medications; however, the study was observational and uncontrolled, and the principal inclusion criteria for the study included any patient prescribed regular antiepileptic medication. Thus, the study population excluded those with epilepsy on no medication and included patients without epilepsy who were prescribed AEDs for an unrelated condition, for example, migraine, neuropathic pain, or psychiatric disorders.38 A single case-control study has also reported an increased risk of SUDEP in women on neuroleptic medication and in men prescribed anxiolytic medication. The explanation for this is unclear and most likely represents an effect of confounding factors.30 A number of more recent case-control studies have failed to identify a similar association.34,35,37 In summary, although not universally accepted, the weight of evidence at present suggests that children are at less risk but that neither sex nor a particular adult age group are strong risk factors for SUDEP, with the absence of suitable control subjects and lack of controlled prospective data significant methodological constraints. Seizure frequency appears to be an independent risk factor, particularly with respect to convulsive seizures, and the number of previous or concomitant antiepileptic medications is, most likely, interrelated. SUDEP mostly occurs after a convulsive seizure and is more common in unsupervised patients more often found in bed in the prone position. Supervision is protective. There is no convincing evidence of an independent association between SUDEP and subtherapeutic AED levels or specific AEDs. Structural cerebral lesions and co-prescribed psychotropic medication are also, for the most part, likely to be unrelated to the risk of SUDEP. A number of epilepsy syndromes are associated with an increased risk of sudden death, although evidence for an association with the more common syndromes, such as idiopathic generalized epilepsy, is limited at the present time, and more research is required. Methodologically robust multicenter controlled prospective studies of risk factors for SUDEP are required to definitively evaluate the large number of proposed variables in an independent fashion using, for example, logistic regression analysis. This will be further facilitated by standardization of definitions, case ascertainment, and analysis methods.

Pathophysiology of SUDEP
Pathophysiological mechanisms of SUDEP are likely to be heterogeneous and may be multifactorial. Theories propounded have focused on autonomic disturbance particularly cardiac arrhythmias and central and obstructive apnea and neurogenic pulmonary edema. In addition, the possibility of structural or functional cardiac pathology predisposing patients with epilepsy to cardiac events has been proposed. CEREBROGENIC AUTONOMIC CONTROL The components of the central autonomic network involved in the functional relationships between cortical, subcortical, and somatic regions have been elucidated from experimental and human stimulation and lesional studies. The network comprises the insular cortex, central nucleus of the amygdala, and hypothalamus with interconnections with the mesial temporal and frontal areas.58 It has been

15  Sudden Unexpected Death in Epilepsy

225

demonstrated that limbic structures, especially the amygdala and pyriform cortex, modulate hypothalamic function, and stimulation of these foci can elicit both sympathetic and parasympathetic visceromotor autonomic responses.59 Both experimental and ictal electrical stimulation of the cingulate gyrus and orbitofrontal cortex also produce changes in respiration and heart rate.6064 Intraoperative stimulation of the left insular cortex produces bradycardia and hypotension, whereas the converse is true of right-sided stimulation.65 Unilateral hemispheric inactivation with an intracarotid amobarbital infusion produces bradycardia and reduced sympathetic tone when performed on the right side and tachycardia and reduced parasympathetic tone when applied on the left side.66,67 Similar lateralization findings have been reported in patients undergoing unilateral electroconvulsive therapy.68 Other than visual inspection of a standard 12-lead electrocardiogram (ECG), more sophisticated methods to interrogate the cardiac autonomic system have been developed, for example, measures of heart rate variability (HRV). In its simplest form, this is measured in a time domain analysis as the standard deviation of R-R wave intervals.69,70 Frequency domain analysis permits the calculation of highfrequency (HF) and low-frequency (LF) components that assess the relative contribution of parasympathetic and sympathetic autonomic activity.71 The HF component is attributed to vagal mechanisms, and the LF component reflects sympathetic activity with parasympathetic modulation.72 However, measures of HRV may be more complex than purely sympathovagal balance. For example, baroreflex sensitivity modulation may be an important influence.73 More advanced, nonlinear measures of regularity, such as entropy, have recently been used that facilitate further analyses of autonomic function.74 The balance between parasympathetic and sympathetic regulation is dynamic, and in healthy subjects, cyclic variation is commonly observed during the ventilatory cycle and in sleep.75 Reduced HRV has been reported in infants with aborted sudden infant death syndrome,76 in heart transplant patients,77 and as an independent risk factor for sudden arrhythmic death after myocardial infarction.78 CARDIAC MECHANISMS Structural Cardiac Pathology The exclusion of cardiac pathology as a contributing factor in SUDEP is challenging due to the presence of, for example, subtle abnormalities that only a detailed microscopic examination of cardiac tissue can elucidate, such as conducting system fibrosis or cardiomyopathy,79 tissue decomposition precluding the acquisition of suitable material for evaluation, lack of an appropriate control group for comparison, and the possibility of a functional rather than a structural disorder, such as ion channelopathies or preexcitation syndromes, with normal macroscopic and microscopic examinations being implicated.43 Furthermore, inconsistencies in the methodology and reporting of cardiac abnormalities in SUDEP postmortem reports prevent meaningful comparison of results.11,30,36,80 Increased cardiac weight has been observed in male SUDEP cases compared to control subjects,11 although it is likely that this is due to demographic and methodological confounding factors, such as a disproportionate prevalence of excess alcohol consumption in the study population and the derivation of expected cardiac weight from a regression equation based on body height, which has been demonstrated to be

226

THE EPILEPSIES 3

less accurate than using body weight.81 Recent studies, using more convincing methodology, have failed to replicate this earlier finding, and cardiac weight is not considered to differ between SUDEP and non-SUDEP cases.80,82,83 Minor, nonspecific pathological change presumed to be nonfatal, such as atherosclerosis, conducting system fibrosis, and diffuse myocardial fibrosis, was identified in 33% (14 out of 42) of SUDEP cases, compared to 16% (six out of 37) of control cases, in an unblinded study reported by Kloster and Engelskjon in 1999.36 This study corroborated earlier findings by Falconer, who demonstrated arterial, interstitial, and subendocardial fibrosis with myofibrillar degeneration in SUDEP cases,84 and Natelson, who identified perivascular and interstitial fibrosis and subendocardial myocyte vacuolization in five of seven SUDEP cases. In the latter study, these changes were not seen in the control group, which comprised 13 cases of hanging or drug overdose in otherwise healthy, similarly aged individuals,82 although in this unblinded study the pathologist was aware of the relevant clinical information. Interestingly, myocyte vacuolization has also been observed in rats who developed asystole following stimulation of the insular cortex.82 Myocyte vacuolization is considered a reversible pathological entity, occurring in the context of subendocardial ischemia.85 However, the patients universally had normal appearing coronary arteries. It has been postulated that neurogenic coronary vasospasm may be implicated, and that if recurrent, it may eventually progress to perivascular and interstitial fibrosis.86 This may, in turn, predispose the heart to arrhythmogenesis, particularly in the setting of considerable autonomic imbalance during seizures.87,88 The occurrence and significance of these pathological changes in SUDEP is not universally agreed, however. Opeskin failed to identify similar fibrotic changes in SUDEP cases but did demonstrate cardiac conduction system pathology and myocyte vacuolization, although these abnormalities were no more common in the SUDEP group than in the control subjects.80 Nevertheless, the pathological changes may still be important in sudden death, but not specific for SUDEP. It is acknowledged that the identification of subtle interstitial and perivascular fibrosis is subjective, and quantitative histopathological evaluations were not performed. Furthermore, the results were not corrected for seizure frequency. More recently, qualitative and quantitative histopathological assessments of myocardial fibrosis were undertaken in SUDEP and age- and gender-matched non-SUDEP cases, and although visual assessment showed significantly more fibrosis in the SUDEP cases, this was not verified on quantitation, further reinforcing the view that a purely qualitative evaluation is subjective and possibly unreliable. Furthermore, no abnormalities of the cardiac conducting system were demonstrated.89 Overall, the full characterization of the relationship between myocardial pathology and acute and recurrent seizures remains unclear at the present time. Future studies should utilize standardized histopathological qualitative and quantitative protocols with extensive tissue sampling and correlate findings with clinical data to ensure that accurate and reproducible data is obtained. Interictal At the simplest level, interictal cardiac function can be evaluated by visually assessing a standard 12-lead ECG, primarily for evidence of conduction abnormalities, although these are frequently normal9092 or show only minor, nonsignificant changes.93 However, a recent preliminary study of 128 patients with severe

15  Sudden Unexpected Death in Epilepsy

227

refractory epilepsy and learning disability revealed interictal ECG abnormalities in approximately 60% of patients, including first degree atrioventricular block and poor R-wave progression.94 Potentially, a number of important confounding factors exist, though, including whether this study population is adequately representative, whether account was made of the presence of systemic comorbidity or use of psychotropic medication, and whether the conduction abnormalities are clinically significant. Early experimental studies demonstrated that interictal epileptiform activity was associated with sympathetic and parasympathetic autonomic dysfunction in a timelocked synchronized pattern.95,96 In the first clinical reports, analysis of interictal heart rate variability in 19 patients with refractory temporal lobe epilepsy revealed frequent, high-amplitude fluctuations in heart rate that were most pronounced in poor surgical candidates.97 Reduced sympathetic tone, demonstrated by decreased LF power, has been seen in both focal and, albeit less markedly, primary generalized epilepsy.69,92,97,98 These findings have been corroborated by more recent studies,74,99,100 although the pattern of interictal autonomic disturbance in patients with epilepsy is contentious. For example, Evrengul reported an increase in LF power and a reduction of HF values consistent with an increase in the sympathetic control of the heart rate in patients with untreated generalized tonic-clonic seizures.71 It is possible that the disparity between the studies may be, at least partly, due to antiepileptic medication. Tomson showed that patients on carbamazepine had a significantly lower standard deviation of RR-intervals, LF power, and a LF/HF power ratio than matched healthy control subjects. In patients on sodium valproate, only the ratio of LF/HF power was lower.69 Interictal autonomic dysfunction associated with carbamazepine use has also been implicated in a number of similar studies. Isoja rvi evaluated 84 patients with a variety of epilepsies and observed autonomic dysfunction in only those patients taking carbamazepine.101 Confusingly, rapid withdrawal of carbamazepine has been associated with both increased sympathetic tone during sleep as measured by the LH/HF ratio51 and a significant reduction in heart-rate variability in both the time and frequency domains with, in particular, a significant reduction in LF power and sympathetic tone.102 The reason for the divergence between these studies is not clear, but it is important to note that both were small series studies and may have lacked sufficient power to convincingly demonstrate a clear effect. Furthermore, there may have been methodological differences. To address the possible modulatory effect of antiepileptic medication, patients with untreated epilepsy have been studied and shown to have both increased LF power, and thus augmented sympathetic tone71 and, in a similar study, normal HRV parameters.103 Again, it is difficult to satisfactorily explain the variance in the results, although important differences in the study population may be at least partly responsible.103 In an interesting study of patients with chronic temporal lobe epilepsy (TLE), postganglionic cardiac sympathetic innervation was quantified using (123)I-metaiodobenzylguanidine-single photon computed tomography (MIBGSPECT). Cardiac MIBG uptake was significantly less in the TLE patients than in the controls, but did not differ between subgroups with and without carbamazepine treatment. The findings are consistent with either postganglionic transsynaptic degeneration resulting from a prolonged increase in central sympathetic interictal discharges or competitive inhibition of MIBG uptake due to continuously enhanced sympathetic activity. The authors concluded that this may translate into an increased risk of cardiac instability and arrhythmias.100

228

THE EPILEPSIES 3

Of recent interest in the pursuit for pathophysiological mechanisms of SUDEP is the potential association between cardiac and epileptic ion channelopathies.43 For example, it is well known that long-QT syndrome, a defect of cardiac repolarization typically due to potassium or sodium channel mutations, produces prolongation of the action potential, propensity to malignant tachyarrhythmias, and sudden death.104 Ion channelopathies are also implicated in familial, monogenic idiopathic epilepsies, such as generalized epilepsy with febrile seizures plus.105 Direct evidence linking these cardiac inherited gene determinants and SUDEP is lacking at present; however, there is emerging evidence of genetic susceptibility in sinus node dysfunction and bradyarrhythmias, and this more complex pattern of inheritance may have more relevance to SUDEP than monogenic disorders (see review by Nashef et al, 200743). Future work in this area should include much larger studies of ECG changes and QT interval in patients with epilepsy and epidemiological studies exploring the association between syncope, inherited arrhythmias, and epilepsy. Overall, there is some evidence for interictal cardiac autonomic dysfunction in patients with both focal and generalized epilepsy, possibly modulated by antiepileptic medication, in particular carbamazepine. Conflicting reports in the literature, however, suggest that the relationship between interictal epileptiform activity, antiepileptic medication, and autonomic function has not yet been fully characterized. This may, at least partly, be due to a lack of standardized analysis methods and heterogeneous study populations. Further work in this field is of paramount importance, as the ability to stratify the risk of SUDEP in individual patients on the basis of interictal autonomic parameters would have valuable management and prognostic implications. Ictal There is extensive literature on the presence of ECG changes, for example T-wave inversion, ST segment elevation, and a prolonged QT interval, in patients with intracranial pathology, such as subarachnoid hemorrhage and cerebrovascular accident, despite normal cardiac examination at autopsy (see Samuels, 2007 for review106). It is understood that these changes are a manifestation of massive catecholamine release and autonomic dysregulation resulting in ventricular wall motion abnormalities, vasospasm, and subsequent cardiac contraction-band necrosis, rather than being due to established structural cardiac pathology, such as atheroma.107,108 Predominantly neurogenic, rather than humorally driven, autonomic dysfunction has been postulated as the cause of ECG abnormalities during convulsive seizures. Arrhythmias, conduction block, and repolarization ECG abnormalities, such as atrial fibrillation, marked sinus arrhythmia, supraventricular tachycardia, atrial and ventricular premature depolarization, bundle-branch block, high-grade atrioventricular conduction block, ST segment depression, and T-wave inversion have been reported in up to 56% of seizures. Abnormalities appear to be more common in nocturnal, prolonged, and generalized seizures than in focal seizures or those occurring during wakefulness.59,109112 Tavernor reported minor prolongation of the ictal, compared to interictal, corrected QT interval (QTc) in a group analysis of 11 patients who subsequently died of SUDEP. The same QTc prolongation was not seen in living age- and sex-matched control subjects who also had refractory epilepsy. The authors concluded that prolongation of the QT interval may provoke malignant ventricular tachyarrhythmias, which may be a potential cause of SUDEP.113

15  Sudden Unexpected Death in Epilepsy

229

However, although ictal sinus tachycardia is a frequent occurrence, atrial or ventricular tachyarrhythmias are only rarely seen.91,109,114 Sinus rate change is the most common cardiac accompaniment to ictal discharge. Sinus tachycardia has been reported in 50 to 100% of seizures and is dependent on the definition used and population studied.9092,112,114119 Although the heart rate in ictal tachycardia is typically 100 to 120 beats per minute,90 there are reports of rates exceeding 170 beats per minute, even during simple partial seizures.91,115 Ictal tachycardia is most commonly seen in the early ictal phase, soon after seizure onset,114,115,118,119 or rarely before clear evidence of electroclinical onset.112 This contrasts with ictal bradycardia, which is seen during the late ictal phase or in the immediate postictal period.120,121 Some evidence exists for right-sided lateralization and temporal lobe localization in patients with ictal tachycardia,114,116,119 corroborating the reports of early experimental and clinical stimulation studies,65,68,122 although it is important to note that most temporal lobe seizures are associated with ictal tachycardia, irrespective of lateralization. In contrast, in patients with unilateral temporal lobe epilepsy being evaluated with extensive intracranial electroencephalogram (EEG) electrodes, irrespective of lateralization of ictal onset, heart rate was seen to increase incrementally as new cortical regions anywhere in the brain were recruited.123 Although ictal tachycardia is almost universally observed, ictal bradycardia has received more attention due to the potential progression to cardiac asystole and intuitive but unproven association with SUDEP. The first report of ictal asystole was by Russell in 1906, who noted the disappearance of a young male patients pulse during a seizure.124 The published literature since that time is, unsurprisingly, mostly case reports or small-series studies, which significantly limit the number and confidence of any conclusions extracted from the data. Ictal bradycardia is observed in <5% of recorded seizures,91,114,118,125 but may occur in a higher percentage of patients because a consistent cardiac response to each apparently electroclinically identical seizure is not seen.91 A recent literature review by Britton revealed that of 65 cases of ictal bradycardia with sufficient EEG and ECG data, seizure onset was localized to the temporal lobe in 55%, the frontal lobe in 20%, the frontotemporal region in 23%, and the occipital lobe in 2% of cases. Information regarding seizure-onset lateralization was available in 56 cases. Seizure onset was lateralized to the left hemisphere in 63%, the right in 34%, and bilaterally in 4%. Interestingly, of 22 cases with EEG data available at the onset of the bradycardia, 12 showed bilateral hemispheric ictal activity, whereas six showed left, and four showed right-sided activity.121 No control group data is available, however. Nevertheless, it appears that there is a trend toward the left temporal lobe being implicated in ictal bradycardia, but this is not sufficiently specific to be valuable localizing semiological information.121,125 Of greater interest is the frequent observation of bilateral ictal activity during bradycardia.121,126,127 This may cause a more significant imbalance of parasympathetic and sympathetic dysfunction than unilateral stimulation via either corticocardiac pathways or through connections with subcortical and brainstem regions, which are frequently activated in seizures and that may potentially contribute to a bradycardic response.128,129 Ictal asystole, lasting between 4 and 60 seconds, is reported, albeit rarely, in patients with refractory epilepsy.54,91,112,126,130,131 In addition, experimental data suggests that ictal bradyarrhythmias can lead to complete heart block.96

230

THE EPILEPSIES 3

It is possible, however, that bradycardia and asystole are conditions with distinct pathophysiological bases rather than two points on a continuum of cerebrogenic arrhythmias. However, the characteristics of the epilepsy with respect to localization, lateralization, seizure types, and population demographics are identical to those of patients with ictal bradycardia, suggesting that the two entities are possibly linked. Short periods of EEG/ECG monitoring may underestimate the prevalence of ictal asystole. For example, Schuele searched a database of 6825 patients undergoing inpatient video-EEG monitoring and found ictal asystole in only 0.27% of all patients with epilepsy. In contrast, Rugg-Gunn reported on 19 patients with refractory focal epilepsy who were implanted with an ECG loop recorder for up to 18 months. Over 220,000 patient hours of ECG recording were monitored, during which time 3377 seizures (1897 complex partial or secondarily generalized tonic-clonic seizures and 1480 simple partial seizures) were reported by patients. Cardiac rhythm was captured on the implantable loop recorders in 377 seizures. Ictal bradycardia, defined as a rate of less than 40 beats per minute, was seen in 0.24% of all seizures over the study period and 2.1% of the recorded seizures. One patient developed supraventricular tachycardia (rate 120 bpm) lasting approximately 30 seconds, during a complex partial seizure. Seven of the 19 patients experienced ictal bradycardia. Four of these had severe bradycardia or periods of asystole, which led to the insertion of a permanent pacemaker. The small number of patients involved precluded statistical analysis of localization and lateralization data. There was no clear correlation between cardiac events and specific AEDs. Notably, only a small proportion of seizures for every patient were associated with significant cardiac events, despite identical seizure characteristics.91 The wider significance of these findings remains to be established but may be addressed by a larger UK-based multicenter study that is currently underway. For example, although this study showed asystole in patients with refractory focal seizures, a similar intractable group of patients with generalized epilepsy has not been studied. Nevertheless, the identification and targeting of patients at risk of ictal asystole, preferably with an interictal surrogate marker, is an important goal. Extrapolation of ictal bradyarrhythmias to a mechanistic explanation for SUDEP remains elusive. This is, at least partly, due to a lack of clinical evidence of common factors shared by patients with ictal bradyarrhythmias and SUDEP and the difficulty in ascertaining the importance of ictal bradyarrhythmias in SUDEP in relation to other proposed mechanisms, including other intrinsic cardiac abnormalities or apnea and hypoxia, which may aggravate arrhythmias. RESPIRATORY MECHANISMS It is likely that primary respiratory dysfunction is involved in an important proportion of SUDEP.54,132138 Central and obstructive apnea, excess bronchial and oral secretions, pulmonary edema, and hypoxia during seizures are all well documented.54,136,138140 Central chemoreceptive areas responsible for sensitivity to increases in carbon dioxide have been shown to be located in several areas of the brainstem, including the nucleus tractus solitarii, locus coeruleus, medullary raphe, and ventrolateral medulla, which comprises an extensive network of respiratory neurons, known collectively as the ventral respiratory group, involving the nucleus ambiguous (NA) and Bo tzinger and pre-Bo tzinger complexes.141 Changes in the arterial partial pressure of CO2 evoke rapid alterations in respiration.142

15  Sudden Unexpected Death in Epilepsy

231

This brainstem network receives supranuclear afferents from cortical areas, such as the rolandic region, or via cranial nerves, such as the trigeminal and glossopharyngeal nerves.60 Central apnea can therefore occur secondary to the ictal discharge, acting at either the cortical or medullary level or possibly as a result of secondary endogenous opioid release influencing the brainstem respiratory nuclei directly. During postictal impairment of consciousness, hypercapnia and hypoxia may be less potent respiratory stimuli. In a study of 17 patients with epilepsy who underwent polysomnography with cardiorespiratory monitoring in a supervised environment, ictal apnea of greater than 10 seconds in duration was demonstrated in 20 of 47 seizures. Oxyhemoglobin saturation decreased to less than 85% in 10 seizures. Central apnea, which may evolve during a focal or generalized seizure, was seen more frequently than obstructive apnea; however, the study was in a controlled environment, and assistance may have minimized the likelihood of obstructive apnea being observed.54 Interictal obstructive sleep apnea has been reported to be more frequent in patients with epilepsy than in the general population, although the reasons for this are unclear.143 Interestingly, transient bradycardia or sinus arrest has been seen in association with ictal apnea, suggesting that the reported seizure-related arrhythmias may be consecutive to ictal apnea.54 Similar findings have been reported in children.138 Additional reports of ictal apnea are typically case studies recorded incidentally during videoEEG telemetry.134136 In a study of 135 SUDEP cases, 15 of which were witnessed, observers described respiratory difficulties, such as apnea and obvious respiratory obstruction, in 12 patients, although the conclusions that may be drawn are significantly limited by the quality of the retrieved information and lack of additional relevant cardiorespiratory parameters.137 Witnesses have reported a delay between the seizure and time of death, which is more consistent with primary respiratory inhibition followed by respiratory arrest and the development of hypoxia and pulmonary edema, than primary ictal cardiac asystole.29 Analogous with reports of patients with ictal bradycardia and asystole, the majority of published cases of ictal central apnea had temporal lobe epilepsy, although this is likely to represent selection bias. No clear lateralizing information is available from the published literature. Neurogenic pulmonary edema, which may in itself be insufficient to be fatal, has been implicated in theories regarding respiratory dysfunction and SUDEP following a number of postmortem reports and case studies.11,29,133,140 In a sheep model of ictal sudden death, animals that died had a greater increase in pulmonary vascular pressure and hypoventilation. When airway obstruction was excluded by tracheostomy, central apnea and hypoventilation were observed in all, causing or contributing to death in two, whereas a third animal developed heart failure with significant pathologic cardiac ischemic changes.132,144 The apparent protective effect of supervision favors an important primary role for respiratory factors,31 as these can be influenced by relatively unskilled intervention, such as airway protection, repositioning, or stimulation. It is unknown what proportion of SUDEP cases may be prevented by such intervention. SUPPRESSION OF CEREBRAL ACTIVITY The possibility of progressive suppression and eventually cessation of cerebral activity as a cause of SUDEP, despite normal cardiac function, was introduced with the

232

THE EPILEPSIES 3

publication of a case report of an intracranially monitored patient who died of SUDEP. Bird described a seizure starting in sleep in one hemisphere and spreading to the other for several minutes. The EEG pattern on the original side then changed to burst-suppression with spindling spike discharges, followed by complete cessation of activity. The other hemisphere continued to show spike discharges until ceasing suddenly a few seconds later. A pulse artifact on the EEG continued for an additional 2 minutes; there was no recording of respiratory activity. Postmortem examination showed mild congestion of the lungs. It was postulated that the loss of EEG activity was not preceded by anoxia, as both hemispheres were not simultaneously affected.145 Excessive postictal brainstem inhibition due to seizure-induced release of GABA and other neuroinhibitory peptides may contribute to death in some patients. This may be compounded by antiepileptic medication.50 This endogenous seizureterminating mechanism could result in blunting of the central hypoxic and hypercarbic respiratory drive, resulting in postictal respiratory arrest, subsequent exacerbation of hypoxia, further cardiac destabilization, and death due to hypoxia and secondary cardiac arrhythmia. This is consistent with the observation that SUDEP occurs after a seizure and could be a consequence of failed reestablishment of respiration in the postictal phase.

SUDEP and Epilepsy Surgery


Compelling evidence shows that patients with poorly controlled, predominantly generalized tonic-clonic seizures are at greatest risk of SUDEP, and a seizure is frequently seen as the terminal event. Intuitively, therefore, good seizure control should translate into reduced risk of SUDEP. Sperling evaluated the mortality rates of 393 patients who underwent epilepsy surgery. The standardized mortality ratio (SMR) for patients with recurrent seizures postoperatively was 4.69, with a SUDEP incidence of 7.5/1000 patient-years, whereas in patients who became seizure free, there was no difference in mortality rate compared with an age- and sex-matched population.146 This compares with the results from Hennessy, who identified an overall postoperative SMR of 4.5 and SUDEP incidence of 2.2/1000 patient-years, although the results were not stratified for postoperative seizure outcome,147 and Salanova, who found a SMR of 1.8 in those with a good postoperative outcome versus 7.4 in those who failed surgery.148 In a large, population-based epilepsy surgery cohort, Nilsson failed to demonstrate an association between mortality rates and seizure outcomes, although there was a clear difference between patients who underwent surgery (SUDEP incidence 2.4 per 1000 patient-years) and those who failed presurgical assessment (SUDEP incidence 6.3 per 1000 patient-years).16 There has been recent interest in the tenet that a common factor predisposes to surgical failure and an increased risk of SUDEP so that patients who respond poorly to surgery also carry an increased risk of SUDEP and that, overall, surgery does not alter the risk of SUDEP.149 Proposed common factors include TLE, which extends beyond the temporal lobe into the insula, frontal orbital, or frontal operculum region, which may favor ictal arrhythmias, central apnea, and secondary generalization. This, in turn, would increase the risk of SUDEP, and the wide epileptogenic field would translate into a poor postoperative seizure outcome.149 This is supported by a study by Persson, who reported on 21 patients undergoing temporal lobe

15  Sudden Unexpected Death in Epilepsy

233

surgery and found that postoperative measures of HRV did not differ from preoperative HRV. Patients with a poor outcome after surgery had significantly lower preand postoperative HRV than patients with a good outcome, who had similar HRV measures to healthy control subjects.70,150 In contrast, Hilz found a postoperative reduction in sympathetic cardiovascular modulation and baroreflex sensitivity in patients undergoing temporal lobe resection, consistent with stabilization of cardiac autonomic control.151 Mortality studies performed in patients with vagal nerve stimulators have shown that excess mortality associated with refractory epilepsy reduced as a function of duration of use. The rate of SUDEP was 5.5 per 1000 patient-years in the first 24 months and 1.7 per 1000 patient-years thereafter, possibly reflecting gradual increase in efficacy over time. Stabilization of measures of heart rate variability post-VNS (vagal nerve stimulation) implantation152154 have paralleled the improved mortality rates, although these findings are not universal.155,156

Implications for Management


Despite a wealth of studies reporting on proposed risk factors or mechanisms of SUDEP, this information has not yet been translated into targeted therapeutic interventions and a reduced incidence of SUDEP. Despite this being a fundamental goal in the management of patients with epilepsy, there has been a paucity of studies specifically addressing preventative or therapeutic strategies. Given the disturbance in cardiac autonomic control in patients with epilepsy, there has been speculation as to whether cardiotropic medication, such as betaantagonists, may have a protective effect, although no studies have been performed in this regard.111 Experimental studies in rats with audiogenic seizures and ictal apnea have shown that selective serotonin reuptake inhibitors have a protective effect,157 although relevant confirmatory clinical studies are lacking. Of interest, however, is the recent finding of neuropathological evidence of involvement of the medullary serotonergic network in sudden infant death syndrome cases with a significantly lower density of serotonin receptor binding sites, particularly in male SIDS cases compared to controls.158 Whether pharmacological modulation of the brainstem serotonergic network or cardiac autonomic function results in a protective effect remains to be seen. The implications of the observed ictal asystole in a small cohort of patients to a larger, more representative group of epilepsy patients is unknown. If this finding is confirmed, the potential role of pacemaker insertion in preventing a proportion of SUDEP cases needs to be assessed. Supervision of patients with epilepsy has emerged as the only clinically important protective factor, independent of seizure control. The basis for this remains unclear but may relate to body positioning and alleviation of obstructive apnea or possibly brainstem arousal mechanisms.15,27,31,137 Strategies to adequately monitor patients with epilepsy at night, evaluating either cardiac, respiratory, or body movement parameters, have been developed, but issues with, for example, high false-positive rates render the devices less user friendly. As a result, there is an urgent need to develop more sophisticated, unobtrusive, reliable, and affordable monitoring equipment. In the U.K. the NICE guidelines state that tailored information and discussion between the individual, family, and/or caregivers and health care professional should

234

THE EPILEPSIES 3

take account of the small but definite risk of SUDEP.159 Several studies have suggested that the relatives of people who died from SUDEP wished they had known of the risk of sudden death.27,160 Whereas it is not certain that knowledge of the risks of epilepsy would necessarily prevent death, some evidence suggests that observation, positioning, and where necessary, stimulation after a seizure may protect against death. It is also likely that patients who know of the risks of epilepsy might be more adherent to AED regimens and the avoidance of trigger factors, thus reducing the frequency of seizures. In contrast, a cohort-controlled study of SUDEP from Australia concluded that because no clear risk factors were modifiable by practical intervention, disclosure to the patient of the possibility of SUDEP was inappropriate.161

Summary
Patients with epilepsy have an increased mortality rate compared to the general population, of which SUDEP is a major contributor. The incidence of SUDEP varies with the population studied. From the available evidence, we know with reasonable certainty that seizure frequency is an independent risk factor, particularly with respect to convulsive seizures, and the number of concomitant antiepileptic medications is, most likely, interrelated. The mechanism underlying SUDEP is likely to be multifactorial with seizures, in particular, GTC seizures, exerting a major influence on the homeostasis of central autonomic functions, which consecutively precipitate potentially life-threatening cardiac and respiratory events. We suspect that intrinsically altered physiological parameters, caused by interictal epileptic activity, a genetically determined predisposition, AEDs, or a cardiac substrate may facilitate this process, with specific perimortem events, such as a prone position during sleep or supervision variably moderating this further. Future work should involve prospectively collected multicenter cohort or casecontrol studies with standardized case ascertainment, established clinical and pathological definitions, and a large number of systemic variables to further investigate causality rather than incidence. Interictal and ictal electrophysiological, cardiorespiratory, and metabolic variables should be evaluated in a large population of patients, including in specific syndromes, to further establish the pathophysiological mechanisms of SUDEP. A key aim is to stratify the risk of SUDEP for an individual patient and, ideally, identify potential therapeutic targets. This would be augmented by general strategies for the prevention of SUDEP, including, for example, improved control of GTC seizures, seizure detection, and supervision. REFERENCES
1. Hauser WA, Annegers JF, Elveback LR. Mortality in patients with epilepsy. Epilepsia. 1980;21: 399-412. 2. Cockerell OC, Johnson AL, Sander JW, Hart YM, Goodridge DM, Shorvon SD. Mortality from epilepsy: results from a prospective population-based study. Lancet. 1994;344:918-921. 3. Annegers JF, Coan SP. SUDEP: overview of definitions and review of incidence data. Seizure. 1999;8:347-352. 4. Shorvon S. The treatment of chronic epilepsy: a review of recent studies of clinical efficacy and side effects. Curr Opin Neurol. 2007;20:159-163. 5. Nashef L. Sudden unexpected death in epilepsy: terminology and definitions. Epilepsia. 1997;38: S6-S8.

15  Sudden Unexpected Death in Epilepsy 6. Monte CP, Arends JB, Tan IY, Aldenkamp AP, Limburg M, De Krom MC. Sudden unexpected death in epilepsy patients: risk factors. A systematic review. Seizure. 2007;16:1-7. 7. Tellez-Zenteno JF, Ronquillo LH, Wiebe S. Sudden unexpected death in epilepsy: evidence-based analysis of incidence and risk factors. Epilepsy Res. 2005;65:101-115. 8. Elveback LR, Connolly DC, Kurland LT. Coronary heart disease in residents of Rochester, Minnesota. II. Mortality, incidence, and survivorship, 1950-1975. Mayo Clin Proc. 1981;56:665-672. 9. Tomson T, Walczak T, Sillanpaa M, Sander JW. Sudden unexpected death in epilepsy: a review of incidence and risk factors. Epilepsia. 2005;46(Suppl 11):54-61. 10. Ficker DM, So EL, Shen WK, et al. Population-based study of the incidence of sudden unexplained death in epilepsy. Neurology. 1998;51:1270-1274. 11. Leestma JE, Walczak T, Hughes JR, Kalelkar MB, Teas SS. A prospective study on sudden unexpected death in epilepsy. Ann Neurol. 1989;26:195-203. 12. Timmings PL. Sudden unexpected death in epilepsy: a local audit. Seizure. 1993;2:287-290. 13. Lip GY, Brodie MJ. Sudden death in epilepsy: an avoidable outcome? J R Soc Med. 1992;85:609-611. 14. Nashef L, Fish DR, Sander JW, Shorvon SD. Incidence of sudden unexpected death in an adult outpatient cohort with epilepsy at a tertiary referral centre. J Neurol Neurosurg Psychiatry. 1995;58:462-464. 15. Nashef L, Fish DR, Garner S, Sander JW, Shorvon SD. Sudden death in epilepsy: a study of incidence in a young cohort with epilepsy and learning difficulty. Epilepsia. 1995;36:1187-1194. 16. Nilsson L, Ahlbom A, Farahmand BY, Tomson T. Mortality in a population-based cohort of epilepsy surgery patients. Epilepsia. 2003;44:575-581. 17. Dasheiff RM. Sudden unexpected death in epilepsy: a series from an epilepsy surgery program and speculation on the relationship to sudden cardiac death. J Clin Neurophysiol. 1991;8:216-222. 18. Hitiris N, Suratman S, Kelly K, Stephen LJ, Sills GJ, Brodie MJ. Sudden unexpected death in epilepsy: a search for risk factors. Epilepsy Behav. 2007;10:138-141. 19. Donner EJ, Smith CR, Snead OC III. Sudden unexplained death in children with epilepsy. Neurology. 2001;57:430-434. 20. Camfield CS, Camfield CP. Good newsa population based study indicates that SUDEP is very unusual in childhood onset epilepsy. Epilepsia. 1999;40(Suppl 7):159. 21. Weber P, Bubl R, Blauenstein U, Tillmann BU, Lutschg J. Sudden unexplained death in children with epilepsy: a cohort study with an eighteen-year follow-up. Acta Paediatr. 2005;94:564-567. 22. Camfield P, Camfield C. Sudden unexpected death in people with epilepsy: a pediatric perspective. Semin Pediatr Neurol. 2005;12:10-14. 23. Sillanpa a M, Jalava M, Kaleva O, Shinnar S. Long-term prognosis of seizures with onset in childhood. N Engl J Med. 1998;338:1715-1722. 24. Schraeder PL, Delin K, McClelland RL, So EL. Coroner and medical examiner documentation of sudden unexplained deaths in epilepsy. Epilepsy Res. 2006;68:137-143. 25. Leestma JE, Kalelkar MB, Teas SS, Jay GW, Hughes JR. Sudden unexpected death associated with seizures: analysis of 66 cases. Epilepsia. 1984;25:84-88. 26. Schwender LA, Troncoso JC. Evaluation of sudden death in epilepsy. Am J Forensic Med Pathol. 1986;7:283-287. 27. Nashef L, Garner S, Sander JW, Fish DR, Shorvon SD. Circumstances of death in sudden death in epilepsy: interviews of bereaved relatives. J Neurol Neurosurg Psychiatry. 1998;64:349-352. 28. Opeskin K, Harvey AS, Cordner SM, Berkovic SF. Sudden unexpected death in epilepsy in Victoria. J Clin Neurosci. 2000;7:34-37. 29. Lear-Kaul KC, Coughlin L, Dobersen MJ. Sudden unexpected death in epilepsy: a retrospective study. Am J Forensic Med Pathol. 2005;26:11-17. 30. Nilsson L, Farahmand BY, Persson PG, Thiblin I, Tomson T. Risk factors for sudden unexpected death in epilepsy: a case-control study. Lancet. 1999;353:888-893. 31. Langan Y, Nashef L, Sander JW. Case-control study of SUDEP. Neurology. 2005;64:1131-1133. 32. Leestma JE, Annegers JF, Brodie MJ, et al. Sudden unexplained d