Epilepsy: A Comprehensive Textbook
2nd Edition

Epilepsy is one of the most common disorders of the brain.²⁸ One of every ten people will have at least one epileptic seizure during a normal lifespan, and a third of these will develop epilepsy. Worldwide, epilepsy affects 50 million people. According to a World Health Organization (WHO) survey, epilepsy accounts for 1% of the global burden of disease, a figure equivalent to breast cancer in women and lung cancer in men.¹⁵

Epilepsy has been known since antiquity. An Assyrian-Babylonian textbook written over three millennia ago provides an accurate clinical description of the condition,¹² and Indian and Chinese physicians of that time were also familiar with it. The word epilepsy is derived from the Greek verb ελαμβανειν (epilamvanein) (“to be seized,” “to be taken hold of,” or “to be attacked”). In ancient Greece, as now, people spoke of “having seized” and of having had an “attack.” This terminology derived from the very ancient notion that all diseases represented attacks by the gods or evil spirits, usually as punishment. Because seizures were the most vivid example of demonic possession, epilepsy was considered to be “the sacred disease,” and by the fifth century BC, the word had gradually acquired the specific and particular meaning associated with it today.²⁵ Indeed, the battle between prejudice and acceptance, ignorance and knowledge, myth and science, and charlatanism and rational therapy has been long and difficult, and even today it has not yet been fully won. Even in comparison with all the advances made during the last century—more than at any other time in history—consider how enormous and fundamental was that first step attributed to Hippocrates in about 400 BC, that epilepsy is a disease of the brain that must be treated by diet and drugs, not religious incantations.⁹

Epilepsy is, of course, not a specific disease, or even a single syndrome, but rather a broad category of symptom complexes arising from any number of disordered brain functions that themselves may be secondary to a variety of pathologic processes. The terms convulsive disorder, seizure disorder, and cerebral seizures are used synonymously with epilepsy: They all refer to recurrent paroxysmal episodes of brain dysfunction manifested by stereotyped alterations in behavior. Modern concepts of epilepsy originate in the work of mid–19th-century physicians and scientists, the most important among them being John Hughlings Jackson.¹⁰ At a time when epilepsy denoted disorders manifested by generalized convulsions, which were believed to arise from disturbances in the medulla oblongata, Hughlings Jackson established the important concept that there were different categories of seizures, each with its own physiology and semiology. His explanation of “dreamy states” and “uncinate group of fits” as focal seizures originating from discrete areas within the cerebral cortex comes close to present-day views of limbic seizures. Similarly, his recognition of focal motor seizures (“jacksonian seizures”) not only identified the responsible locus within the brain, but also allowed him to draw inferences that have forever changed our concepts of cortical motor representation and cerebral control of voluntary movement. Hughlings Jackson, more than anyone, established a scientific approach to the study of epileptic phenomena.

Today, a large number of clinical phenomena are recognized as epileptic seizures, some of which (e.g., myoclonic and atonic seizures) are currently poorly understood and might, in fact, reflect neuronal mechanisms that are somewhat different from the pathophysiologic processes traditionally considered to be “epileptic.” A variety of conditions or epilepsies have been categorized and defined not only by the types of seizures they manifest, but also by other, associated clinical features. Specific epileptic syndromes have been identified by their characteristic seizure types, pattern of seizure recurrence, age of onset, associated neurologic and other clinical signs, electroencephalographic (EEG) findings, presence or absence of familial occurrence, and prognosis. Epilepsies and epileptic syndromes are broadly divided into idiopathic and symptomatic disorders. Idiopathic epilepsies are generally benign in the sense that they are not associated with brain lesions, neurologic abnormalities other than seizures, or mental impairment, and that they tend to be self-limited or respond readily to antiepileptic drugs. Genetic factors are important, and manifestations are typically age related. This is epilepsy sui generis (“by itself”), which conforms to the original Greek meaning of “idiopathic,” in contrast to the commonly used but incorrect meaning of “cause unknown.” Symptomatic epilepsies are those in which seizures are the consequence of an identifiable lesion or other demonstrable physical or metabolic etiology. When epilepsies are presumably symptomatic but currently of unknown specific etiology, they have been termed cryptogenic,² a term also used in epidemiologic studies to mean unknown as to whether idiopathic or symptomatic.⁴ Because of its ambiguity, “cryptogenic” is a term that should be replaced by the more accurate “probably symptomatic.”⁵

Advances in the understanding and treatment of epilepsy have occurred because of active and continuing research efforts. Indeed, it is not an exaggeration to say that many of the exciting developments in basic neuroscience in the middle and latter parts of the last century were related to epilepsy, either directly (e.g., in cellular studies of disease mechanisms) or indirectly (e.g., in investigations of cortical excitability and its control). Clinical investigation is an essential part of practice, and clinicians have played important roles in developing hypotheses that can be subjected to experimental investigation.

An important change that has occurred and accelerated in the last two decades has been the growing ability to carry out basic studies in humans. For example, intracranial monitoring
techniques used to evaluate patients with intractable seizures for surgery, and the removal of brain tissue during surgical procedures now offer investigators opportunities to investigate basic physiologic, biochemical, and molecular phenomena in patients that could previously be studied only in experimental animal models of epilepsy. Similarly, modern brain imaging methods, such as functional magnetic resonance imaging (fMRI), magnetic resonance spectroscopy (MRS), positron emission tomography (PET), and single-photon emission computed tomography (SPECT), allow noninvasive study of basic biologic questions in the living, intact brain. New insights derived from these studies make it possible to determine which abnormalities found in animal models of epilepsy have counterparts in humans, which experimental observations are valid for the human condition and which are not, and which experimental data fit within reasonable conceptual frameworks for developing further hypotheses that can be tested either in humans directly or in relevant animal models.

There have been many attempts to obtain a consensus on definitions of epileptic seizure and epilepsy. Recently, the International League against Epilepsy (ILAE) has proposed new definitions for both.⁷ In this proposal, an epileptic seizure is defined as “a transient occurrence of signs and/or symptoms due to abnormal excessive or synchronous neuronal activity in the brain.” This definition acknowledges that previous concepts of decreased inhibition and increased excitation were oversimplifications, because inhibition is actually increased in some forms of epilepsy where phasic inhibition is a central element in the primary epileptogenic abnormality. An epileptic seizure can also be a natural response of the normal brain to transient disturbance in function and, therefore, not necessarily an indication of an epileptic disorder. Such seizures are often referred to as provoked, acute symptomatic, or reactive. This accounts for a greater incidence of seizures (about 7% by age 80) than epilepsy (nearly 3%).

Epilepsy is a group of neurologic conditions, the fundamental characteristics of which are recurrent, usually unprovoked, epileptic seizures. A common operational definition of epilepsy is two or more unprovoked seizures occurring more than 24 hours apart.⁴ However, the new ILAE proposal⁷ offers a more fundamental definition of epilepsy: “A chronic condition of the brain characterized by an enduring propensity to generate epileptic seizures, and by the neurobiological, cognitive, psychological, and social consequences of this condition.” This definition emphasizes the existence of a persistent intrinsic epileptogenic abnormality that is a property of the brain itself and thus present even when seizures are not occurring. This contrasts with seizures that are dependent upon acute insults or other conditions that transiently affect an otherwise normal brain. An intrinsic epileptogenic abnormality of the brain necessary for a diagnosis of epilepsy, however, can resolve spontaneously, as in some age-related idiopathic epilepsies that typically remit.

The new ILAE definition also acknowledges importantly the psychological and social consequences of epilepsy. This change recognizes that to the affected patient, epilepsy is more than seizures, and that the condition in its entirety comprises many facets, different for each individual, that contribute to disability and impaired quality of life (Fig. 1). Treatment that focuses solely on seizures often does little to lessen disability. This is most dramatically illustrated by the patient who, having undergone successful surgical resection of epileptogenic brain tissue, becomes seizure free but remains socially isolated and unemployed, with little evidence of an improved life. Therapeutic intervention can be optimal only when the multiple medical, psychological, and environmental factors that constitute epilepsy are addressed. Thus, the physician’s role is properly defined, and sometimes circumscribed, by asking a series of questions: “What are the problems that are contributing to the patient’s predicament?” “Which of these can or need to be treated?” “What will be the consequences of treatment?” and “What outcome measures will appropriately gauge the treatment’s success?”

The existing ILAE Classification of Epilepsies and Epileptic Syndromes² and Classification of Epileptic Seizures³ are presently under review. An ILAE Task Force on Classification and Terminology has proposed a diagnostic scheme for use when describing individual patients⁵ (Table 1). Axis 1 consists of a detailed phenomenologic description of the ictal events, which is useful in certain situations, such as a presurgical evaluation, but otherwise can be abbreviated or omitted. Axis 2 is a new concept, recognizing seizure types as diagnostic entities based on distinct pathophysiologic and anatomic features that provide information useful for determining etiology, therapy, and prognosis. Diagnosis of a specific seizure type is especially important when diagnosis of an epilepsy syndrome is not possible. Axis 3 consists of accepted epilepsy syndromes and recognizes that a syndromic diagnosis is not always possible. An epilepsy syndrome as used here refers to “a complex of signs and symptoms that define a unique epileptic condition. This must involve more than just a seizure type: Thus, frontal lobe seizures per se, for instance, do not constitute a syndrome.” In contrast, an epileptic disease is defined as “a pathologic condition with a single, specific, well-defined etiology. Thus, progressive myoclonus epilepsy is a syndrome, but Unverricht-Lundberg myoclonic epilepsy is a disease.” Axis 4 is etiology, which includes a wide variety of genetic and nongenetic diseases associated with epilepsy, specific epilepsy genes, and acquired cerebral injuries such as trauma and infection. Axis 5, which assesses the degree of disability caused by the epilepsy, is taken from a WHO classification of impairment for neurologic disorders and is optional.

It is readily apparent from observing the diverse ictal phenomenology of patients with epilepsy that most seizures consist of evolving processes that depend on multiple pathophysiologic mechanisms and anatomic substrates. Two essential epileptogenic factors represent the net effect of many complex interrelated events. The first is an abnormality of cellular excitability that arises from mechanisms that affect membrane
depolarization and repolarization. The second is a “network defect” that derives from mechanisms underlying the development of aberrant neuronal integration, resulting in abnormal synchronization of neuronal populations and propagation of the epileptic discharge within neural pathways. Both sets of disturbances must be present before a seizure can occur.

Part of the diversity that characterizes the clinical expression of seizures is due to the fact that different brain areas are responsible for different aspects of epileptic phenomenology¹⁴ (Table 2). Generally, these different regions are spatially proximate, although they are rarely congruent. Sometimes, however, quite distant areas give rise to some of the clinical or EEG features of a particular patient’s seizures.

The irritative zone is the area of cortex that generates interictal EEG spikes. This is clearly related but frequently not identical to the ictal onset zone, which is the area of cortex that initiates seizures. Indeed, although the interictal EEG spike has provided an enormous impetus to basic laboratory research and has led to a definition of the cellular mechanisms underlying interictal discharges, it is still not known whether any given interictal spike reflects mechanisms important in triggering seizures, is a marker of inhibitory mechanisms involved in maintaining the interictal state, or is merely an interesting epiphenomenon unrelated in any direct way to seizure occurrence. In any event, in some instances the irritative zone can include multiple spike sites, some of which are located at a distance from the ictal onset zone, even in the opposite hemisphere. This became clear early in the experience with long-term, especially intracranial, EEG monitoring, in which recordings showed that interictal epileptiform discharges are commonly more widespread than had been suggested by routine EEG or than would be expected solely on the basis of the area of ictal onset. For example, in scalp recordings from patients with unilateral mesial temporal lobe epilepsy, contralateral temporal spikes are frequently encountered, although they are usually concentrated on the side of seizure onset, at least in the majority of patients who have successful surgical outcomes. Intracranial recordings frequently demonstrate widespread ipsilateral spikes, and postsurgical electrocorticograms show that many of these may remain after resection, apparently without major adverse consequences for seizure outcome. The classic scalp EEG “spike focus,” therefore, is a very limited reflection of complex underlying physiologic events. Interictal PET scans, too, consistently demonstrate that cerebral metabolic dysfunction is more extensive than clinical evidence, scalp EEG, and structural brain imaging would suggest.

In symptomatic epilepsies, the epileptogenic lesion is the pathologic substrate for the epilepsy; it can usually be identified on MRI, although EEG remains necessary to demonstrate epileptogenicity of a lesion. Seizures can arise within, adjacent to, or even sometimes distant from an epileptogenic lesion. The symptomatogenic zone is that portion of the brain responsible for producing the first clinical ictal symptoms or signs, whereas the functional deficit zone is the cortical area or areas exhibiting focal nonepileptic dysfunction. Finally, the epileptogenic zone is the total area of brain that is necessary and sufficient to generate seizures and that must be removed to abolish seizures. The fact that the epileptogenic zone cannot be defined with precision accounts for the lack of a uniformly successful outcome following resective surgery for focal seizures. The problem is greater in extratemporal than in temporal lobe epilepsy.

There can be little question that at least some types of seizures, especially if prolonged or repeated frequently, can damage the brain.²⁴ Indeed, the association between chronic epilepsy and pathologic changes in the brain, especially hippocampal sclerosis, is one of the oldest and best-documented clinicopathologic correlations. The concept of an epileptic encephalopathy as a condition in which the epileptic processes themselves are believed to contribute to progressive disturbances in cerebral function is now accepted.⁵ The question of the relationship between seizures and interictal neurologic or psychiatric findings still generates controversy, however, because it is often not possible to distinguish among the effects of epilepsy per se, cerebral disease, psychological and social factors, and toxicity from chronic use of antiepileptic drugs. Whether distinct behavioral traits or profiles are associated with temporal lobe or other forms of epilepsy remains a matter of debate. Although it seems to have been impossible to devise a definitive clinical study of this issue, it would be surprising if the profound and widespread disturbance of neuronal activity caused by some epileptiform events, and the neuropathologic
findings observed in surgical and autopsy specimens of epileptogenic brain areas, never produced enduring interictal consequences, even subtle, for behavior, personality, and social functioning.

It is even likely that interictal electrical abnormalities, although traditionally considered subclinical and “silent,” also have important functional effects, because of the disruption they cause in normal information processing. Such disruption may be as much the result of excessive or abnormal inhibition as of “hyperexcitability.” Over 30 years ago, Schwartzkroin et al.²¹ asked the question, “What is the downstream effect of focal cortical epileptiform activity, and can it influence incoming sensory inputs?” To study this, they recorded field potentials and unit activity in the cuneate nucleus of cats contralateral to an experimental cortical epileptogenic focus. They discovered that when an afferent sensory volley, triggered by stimulation of a forepaw nerve, occurred in close relation to a cortical spike discharge, the amplitude of the evoked cuneate field potential was substantially and predictably decreased (Fig. 2). In other words, isolated cortical epileptiform discharges had a powerful inhibitory effect on sensory signal transmission by cuneate relay neurons. Subsequently, Shewmon and Erwin^22,23 provided suggestive evidence in humans that interictal spikes, or more likely the after-going slow waves, can similarly result in transient dysfunction as measured by longer reaction times to particular stimuli and increased nonperception of visual stimuli, especially those in the visual field contralateral to the spike discharge. Thus, epileptiform activity occurring at critical times may substantially modify or “color” the processing of sensory information.

Finally, there is now a substantial body of experimental evidence that activity-dependent behaviors of individual neurons, including those associated with seizures, rapidly initiate a cascade of changes in gene expression. These in turn may lead to further alterations in neuronal excitability within local areas or circuits, some of which are long lasting and associated with (although not yet proved to be causative of) structural remodeling. Some induced genes may relate to future seizure susceptibility, others to intrinsic protective or trophic mechanisms, and still others to plastic structural and functional changes that accompany recurrent seizures and that could, conceivably, produce changes in interictal neuropsychiatric behavior.

Depending on the patient, seizures can occur frequently or infrequently, only at night or after awakening, in a cyclic pattern suggesting hormonal influences, only with highly specific triggers, in many other permutations, and, most commonly, without any apparent predictability. However, even patients with refractory seizures have attacks relatively infrequently compared with the total time available in their lives. The factors that precipitate seizure occurrence are still poorly understood, but clinical observations indicate that specific mechanisms must govern whether a patient is in an interictal or ictal state, and how the transition from the interictal to the ictal condition is made. Many environmental and physiologic factors modulate the probability of seizure occurrence: Fever, sleep deprivation, alcohol withdrawal, highly specific triggers in reflex epilepsy, hormonal fluctuations, and even nonspecific stress in some susceptible individuals. It is unknown, however, how these perturbations translate into increased epileptic susceptibility at the cellular or molecular level. In most patients, it is not possible to identify external or internal factors that explain why a
seizure happened at a particular time, rather than a week before or a few minutes later. Changes that alter neuronal excitability or the potential for synchronous interactions among neurons are undoubtedly important but remain to be specified in detail. As but one example, consider the disturbances that can occur in the extracellular microenvironment in epileptic brain regions and, in particular, the possibility that increased potassium concentration within extracellular spaces, which have perhaps been abnormally reduced because of gliosis, cell swelling, or other pathologic factors, may serve as a mechanism for increasing the excitability level of a population of neurons to a critical threshold for seizure generation. Equally mysterious is the fundamental nature of the inherent homeostatic forces in the brain that limit ictal spread after seizures have begun and that terminate them within seconds of their onset. Undoubtedly, the basic neuronal mechanisms underlying these processes are as diverse as those leading to seizure onset. As demonstrated by the condition of epilepsia partialis continua, the mechanisms mustered by the brain to limit ictal spread are different from those utilized to terminate the seizure discharge.

More than 125 years ago, Gowers⁸ first recognized that there is almost always a seizure-free interval between a causative cerebral injury and onset of symptomatic epilepsy. The latency of occurrence of posttraumatic seizures is the clearest example of this point. Figure 3, taken from the Vietnam Head Injury Study,¹⁹ shows the cumulative percentage of patients with posttraumatic seizures as a function of time after injury. Whereas seizures developed most often within the first 2 years of injury, more than 15% of veterans did not have their first seizure until 5 or more years later.

This latent period, which characterizes many (and perhaps all) cases of symptomatic epilepsy, raises important issues about the process of epileptogenesis. The clinical data suggest that this process is a dynamic and evolving one that progressively alters neuronal excitability, establishes critical interconnections, and perhaps requires critical structural changes before the first clinical seizure appears. A sequence that seems to be essential for at least one common form of epilepsy, mesial temporal lobe epilepsy with hippocampal sclerosis, involves an insult early in life that causes selective neuronal death, synaptic reorganization, and altered firing patterns within a defined neuronal population. Enhanced excitation and inhibition result in a propensity for hypersynchronization.⁶ Recurrent hypersynchronous discharges recruit efferent structures into this epileptogenic process until a sufficient area of brain tissue is involved to manifest as clinical epileptic seizures. A genetic predisposition may influence the pattern of cell loss and synaptic reorganization, or the effect of these anatomic changes on local excitability, or the influence of these local excitability changes on projection areas and eventual manifestation of ictal events. Although this sequence of events has been developed from research on experimental animal models of mesial temporal lobe epilepsy, what is lacking in humans, of course, is evidence of a subclinical epileptogenic abnormality preceding the appearance of clinical seizures. Thus, if it were possible to have depth electrodes placed in the hippocampus from the time of the epileptogenic injury, it is likely that epileptiform changes would be detected long before any clinical hint of epilepsy. That a similar epileptogenic evolution in fact occurs in humans is suggested by the occasional observation of focal EEG spikes in patients with a brain tumor or vascular malformation, even in the absence of clinical seizures.

Because of the great variability in latent periods seen among patients, there must be both intrinsic and acquired modifying factors unique to each individual. One of these is undoubtedly genetic background, which is an inherent determinant of a person’s susceptibility to seizures (“seizure threshold”). Another factor is the location and spatial dimension of the injury, and a third is the severity of the injury, which may be expressed in terms of the effect a specific lesion has on adjacent or displaced brain tissue.

This is the era of molecular genetics applied to human disease. Among the many individual epileptic syndromes described in humans, there are a number of idiopathic epilepsies that are considered to be familial and in which genetic determinants appear to be prominently involved. Twin studies^1,26 implicate strong genetic determinants in many types of seizures and seizure disorders, especially such ones as childhood absence epilepsy, juvenile myoclonic epilepsy, and idiopathic grand mal seizures. However, the role of genetic factors is not straightforward; to the contrary, it is quite complex. For example, seizures develop at increased rates in children of parents with
either idiopathic or symptomatic epilepsy, although the difference is greatest for children of parents with idiopathic generalized forms of epilepsy.¹⁷ Thus, there may be some degree of shared genetic susceptibility present in both the idiopathic and symptomatic epilepsies. However, there are undoubtedly specific genetic determinants of the susceptibility of the brain to seizures and epilepsy following a particular injury, as well as other genetic factors that determine the occurrence of individual familial idiopathic epilepsy syndromes. Furthermore, some inherited disorders, such as tuberous sclerosis and neurofibromatosis, are associated with brain lesions that in turn give rise to symptomatic epilepsies. Whereas mutations in single genes account for some rare epilepsy syndromes and familial diseases that cause epileptic seizures, multiple genes must determine the various neuronal functions that alter seizure threshold and predispose to development of symptomatic epilepsy. For most epileptic disorders, it remains to be determined to what degree abnormalities of single genes, or concordance of key overlapping genes, determine the phenotypic expression of any given epileptic condition.

It is also important to underscore the importance of interactions between genetic predisposition and environmental factors in the manifestation of seizures and the particular expression of a given disorder. Lennox¹³ was among the first to postulate a continuum between idiopathic and symptomatic epilepsies. He suggested that the occurrence of epileptic seizures derives from the complex interrelation of genetic factors and brain disease (Fig. 4). In any given patient, the relative contribution of genetic or acquired pathologic factors determines whether the epilepsy takes the form of an idiopathic disorder or a symptomatic one. Thus, although there is unequivocal evidence of hereditary factors in epilepsy, family studies do not clearly define a pattern of inheritance for the common forms of epilepsy. Furthermore, the data suggest a complex hereditary component in all forms of epilepsy, localization related and generalized, idiopathic and symptomatic.¹⁷

Animal models have allowed definition of a growing number of “epilepsy genes” and their encoded proteins. Information obtained from animals with single gene mutations producing some aspect of an epileptic phenotype is permitting identification of protein products responsible for the molecular aberrations underlying the abnormal excitability and synchrony responsible for recurrent epileptic seizures. The genes and functional consequences identified in fortuitous mouse models are providing important tools for defining key sites of vulnerability in the brain’s carefully regulated control of excitability.¹⁶ Creation of transgenic animal models that overexpress particular genes on the one hand, or “knockout” selected genes on the other, has become a vitally important step in studying candidate genes and understanding the molecular basis of defects that result in epileptic excitability. At the same time, a growing number of specific genes and gene mutations are being isolated and cloned from cases of human epilepsy. A major problem facing investigators is that the population of genes that encode molecules contributing to regulation of cortical excitability through membrane and synaptic functions (and are therefore possible candidate genes in epilepsy) is very large. The complexity issue is further magnified if genes for various second messenger cascades, which indirectly regulate membrane proteins involved in signal transduction, are considered. Another daunting consideration is that genetic studies have identified both animal and human seizure phenotypes in which the genetic mutation and protein product have no previously known association with epileptogenesis (e.g., the LGI1 mutation in autosomal dominant temporal lobe epilepsy with auditory features¹¹ or cystatin B in Unverricht-Lundborg disease¹⁸). Finally, epilepsy may result from structural and functional changes occurring as a reaction to a transient genetic defect or a time-limited aberration that is present early in development but may have vanished by the time clinical seizures appear.

So what is the present status of the genetic basis of epilepsy? There can be no doubt about the importance of hereditary factors in epilepsy, in some syndromes more than others. But although specific gene mutations have been identified in a number of rare monogenic forms of idiopathic epilepsy and there is no lack of possible candidate genes for others, a single causative genetic mutation will not be found for most cases of epilepsy. Rather, there is considerable evidence that in the majority of cases, the genetic influence on risk of developing epilepsy is conferred by complex “susceptibility” genes that
interact with other acquired or genetic disturbances, rather than cause epilepsy by themselves.¹⁷ Generalized epilepsy with febrile seizures plus (GEFS+) may reflect this phenomenon.^20,27 For most patients, epilepsy is the consequence of diverse alterations in the complex interactions that occur within neuronal populations and circuits and in the multiple factors that modulate excitability within these circuits, all of which are ultimately under genetic control. And one must not forget that at the clinical level, epilepsy also comprises the neurologic, behavioral, psychological, and social consequences of the molecular, cellular, and network alterations that underlie seizures.

The answer to the question “What is epilepsy?” is not simple; it is many things at different times, and the patient’s reply may be different from that of the physician. Answers are beginning to emerge, however, to both biologic and psychosocial aspects of the question. The diversity with which human epilepsy expresses itself indicates that it is not a unitary problem and that a single solution to any of its many facets is therefore unlikely. Nonetheless, the potential of new investigative tools, especially those of neuroimaging and molecular neurobiology, are providing unparalleled and heretofore unimagined insights into the mechanisms of epilepsy and epilepsy-related brain dysfunction, and are offering greater hope than ever before for prevention, effective treatment, and even cure.

This book was created in the belief that physicians today can be most effective when their practice has a sound scientific basis, and that the results of ongoing basic and clinical research increasingly affect patient management directly. That so much has been achieved is a tribute to past and present investigators; that there is still so much to do is a challenge to future ones. The conquest of epilepsy, as of any human disease, requires a sustained creative effort, with past, present, and future research representing an unbroken continuum of endeavor and achievement.