ILAE Operational Classification of Seizure Types 2017 PDF

ILAE POSITION PAPER Operational classification of seizure types by the International League Against Epilepsy: Position Paper of the ILAE Commission for Classification and Terminology *Robert S. Fisher, †J. Helen Cross, ‡Jacqueline A. French, §Norimichi Higurashi, ¶Edo...

ILAE POSITION PAPER Operational classification of seizure types by the International League Against Epilepsy: Position Paper of the ILAE Commission for Classification and Terminology *Robert S. Fisher, †J. Helen Cross, ‡Jacqueline A. French, §Norimichi Higurashi, ¶Edouard Hirsch, #Floor E. Jansen, **Lieven Lagae, ††Solomon L. Mosh
e, ‡‡Jukka Peltola, §§Eliane Roulet Perez, ¶¶Ingrid E. Scheffer, and ##***Sameer M. Zuberi Epilepsia, 58(4):522–530, 2017 doi: 10.1111/epi.13670 SUMMARY The International League Against Epilepsy (ILAE) presents a revised operational clas- sification of seizure types. The purpose of such a revision is to recognize that some sei- zure types can have either a focal or generalized onset, to allow classification when the onset is unobserved, to include some missing seizure types, and to adopt more trans- parent names. Because current knowledge is insufficient to form a scientifically based classification, the 2017 Classification is operational (practical) and based on the 1981 Classification, extended in 2010. Changes include the following: (1) “partial” becomes “focal”; (2) awareness is used as a classifier of focal seizures; (3) the terms dyscognitive, simple partial, complex partial, psychic, and secondarily generalized are eliminated; Dr. Robert S. Fisher, (4) new focal seizure types include automatisms, behavior arrest, hyperkinetic, auto- past president of nomic, cognitive, and emotional; (5) atonic, clonic, epileptic spasms, myoclonic, and American Epilepsy tonic seizures can be of either focal or generalized onset; (6) focal to bilateral tonic– Society and editor of clonic seizure replaces secondarily generalized seizure; (7) new generalized seizure Epilepsia and types are absence with eyelid myoclonia, myoclonic absence, myoclonic–atonic, epilepsy.com, led the myoclonic–tonic–clonic; and (8) seizures of unknown onset may have features that can Seizure Classification still be classified. The new classification does not represent a fundamental change, but Task Force. allows greater flexibility and transparency in naming seizure types. KEY WORDS: Classification, Seizures, Focal, Generalized, Epilepsy, Taxonomy. Accepted December 21, 2016; Early View publication March 8, 2017. *Stanford Department of Neurology & Neurological Sciences, Stanford, California, U.S.A.; †UCL-Institute of Child Health, Great Ormond Street Hospital for Children, London, United Kingdom; ‡Department of The International League Against Epilepsy (ILAE), Neurology, NYU Langone School of Medicine, New York, New York, through the Commission for Classification and Terminol- U.S.A.; §Department of Pediatrics, Jikei University School of Medicine, Tokyo, Japan; ¶Unite Francis Rohmer, Strasbourg, France; #Department ogy, has developed a working classification of seizures of Pediatric Neurology, Brain Center Rudolf Magnus, University Medical and epilepsy. Following the proposed reorganization in Center, Utrecht, The Netherlands; **Pediatric Neurology, University 2010,1,2 further clarification has been discussed and feed- Hospitals KU Leuven, Leuven, Belgium; ††Saul R. Korey Department of Neurology, Department of Pediatrics and Dominick P. Purpura Department back sought from the community. One area that required Neuroscience, Einstein College of Medicine and Montefiore Medical further elucidation was the organization of seizure types. Center, Bronx, New York, U.S.A.; ‡‡Department of Neurology, Tampere A Seizure Type Classification Task Force was estab- University Hospital, Tampere, Finland; §§Pediatric Neurology and Rehabilitation Unit, CHUV, Lausanne, Switzerland; ¶¶Florey Institute lished in 2015 to prepare recommendations for classifica- and University of Melbourne, Austin Health and Royal Children’s tion of seizure types, which are summarized in this Hospital, Melbourne, Victoria, Australia; ##The Paediatric Neurosciences document. A companion document guides the intended Research Group, Royal Hospital for Children, Glasgow, United Kingdom; and ***College of Medicine, Veterinary & Life Sciences, University of use of the classification. Glasgow, Glasgow, United Kingdom Descriptions of seizure types date back at least to the Address correspondence to Robert S. Fisher, Neurology, SNHC, Room time of Hippocrates. Gastaut3,4 proposed a modern classi- 4865, 213 Quarry Road, Palo Alto, CA 94304, U.S.A. E-mail: robert. [email protected] fication in 1964. Various basic frameworks for seizure classification can be considered. Manifestations of certain Wiley Periodicals, Inc. © 2017 International League Against Epilepsy seizures are age-specific and depend on the maturation of 522  523 Operational Classification of Seizure Types 1981 and subsequently modified1,2 as a starting point for the revised operational classification. Key Points The ILAE has constructed a revised classification of Methods seizure types; the classification is operational and not What is a seizure type? based on fundamental mechanisms A seizure is defined as “a transient occurrence of signs Reasons for revision include clarity of nomenclature, and/or symptoms due to abnormal excessive or synchronous ability to classify some seizure types as either focal or neuronal activity in the brain.”25 It is the clinician’s first task generalized, and classification when onset is unknown to determine that an event has the characteristics of a seizure Seizures are divided into those of focal, generalized, and not one of the many imitators of seizures.26 The next unknown onset, with subcategories of motor, non- step is classification into a seizure type. motor, with retained or impaired awareness for focal The Task Force operationally defines a seizure type as a seizures useful grouping of seizure characteristics for purposes of communication in clinical care, teaching, and research. Mention of a seizure type should bring to mind a specific the brain. Previous classifications have been based on entity, albeit sometimes with subcategories and variations anatomy, with temporal, frontal, parietal, occipital, dien- on a theme. Choices must be made by interested stakehold- cephalic, or brainstem seizures. Modern research changed ers to highlight groupings of seizure characteristics that are our view of the pathophysiologic mechanisms involved useful for specific purposes. Such stakeholders include and has shown epilepsy to be a network disease and not patients, families, medical professionals, researchers, epi- only a symptom of local brain abnormalities.5 From a net- demiologists, medical educators, clinical trialists, insurance work perspective, seizures could arise in neocortical, tha- payers, regulatory agencies, advocacy groups, and medical lamocortical, limbic, and brainstem networks. Although reporters. Operational (practical) groupings can be derived our understanding of seizure networks is evolving by those with specific interests. A pharmacologist, for rapidly,6 it is not yet sufficient to serve as a basis for sei- example, might choose to group seizures by efficacy of zure classification. In 1981, an ILAE Commission led by medications. A researcher doing a clinical trial might con- Dreifuss and Penry7 evaluated hundreds of video–elec- sider seizures as disabling or nondisabling. A surgeon might troencephalography (EEG) recordings of seizures to group by anatomy in order to predict the eligibility for and develop recommendations that divided seizures into those likely success of surgical therapy. A physician based in an of partial and generalized onset, simple and complex par- intensive care unit with predominantly unconscious patients tial seizures, and various specific generalized seizure might group seizures in part by EEG pattern.27 The principal types. This classification remains in widespread use today, aim of this classification is to provide a communication with revisions in terminology and classification of sei- framework for clinical use. Seizure types are relevant to zures and epilepsy by the ILAE,2,8–14 and with suggested clinical practice in humans; whereas, it is acknowledged insights, modifications, and criticisms by others.15–24 We that seizure types in other species, experimental and natural, chose not to develop a classification based solely on may not be reflected in the proposed classification. One goal observed behavior—instead, reflecting clinical practice, was to make the classification understandable by patients the 2017 classification is interpretive, allowing the use of and families and broadly applicable to all ages, including additional data to classify seizure types. neonates. The ILAE Commission on Classification & Ter- The intention of the 200112 and 200613 reports on minology recognizes that seizures in the neonate can have reclassification was to identify unique diagnostic entities motor manifestations, or alternatively little or no behavioral with etiologic, therapeutic, and prognostic implications, manifestations. A separate Neonatal Seizure Task Force is so that when a syndromic diagnosis could not be made, working to develop a classification of neonatal seizures. the therapy and prognosis would be based on seizure The 2017 seizure classification is not a classification of type. Such a classification would permit grouping of rea- electroencephalographic ictal or subclinical patterns. The sonably pure cohorts of patients for discovery of etiolo- guiding principle of the Seizure Type Task Force was gies, including genetic factors, research into fundamental advice from Albert Einstein to “make things as simple as mechanisms, involved networks, and clinical trials. The possible, but no simpler.” ILAE Seizure Type Classification Task Force (hereafter called “the Task Force”) chose to use the phrase “opera- Motivation for change tional classification,” because it is impossible at this time Adapting to a change in terminology can be effortful and to base a classification fully on the science of epilepsy. needs to be motivated by a rationale for change. Seizure In the absence of a full scientific classification, the Task type classification is important for several reasons. First, the Force chose to use the basic organization initiated in classification becomes a worldwide shorthand form of Epilepsia, 58(4):522–530, 2017 doi: 10.1111/epi.13670 524 R. S. Fisher et al. communication among clinicians caring for people with Structure of the classification epilepsy. Second, the classification allows grouping of The classification chart is columnar, but not hierarchical patients for therapies. Some regulatory agencies approve (meaning that levels can be skipped), so arrows intention- drugs or devices indicated for specific seizure types. A new ally are omitted. Seizure classification begins with the classification should gracefully map to existing indications determination of whether the initial manifestations of the for drug or device usage. Third, the seizure type grouping seizure are focal or generalized. The onset may be missed or might provide a useful link to specific syndromes or etiolo- obscured, in which case the seizure is of unknown onset. gies, for example, by noting an association between gelastic The words “focal” and “generalized” at the start of a seizure seizures and hypothalamic hamartoma or epileptic spasms name are assumed to mean of focal or generalized onset. with tuberous sclerosis. Fourth, the classification allows For focal seizures, the level of awareness optionally may researchers to better focus their studies on mechanisms of be included in the seizure type. Awareness is only one different seizure types. Fifth, a classification provides words potentially important feature of a seizure, but awareness is to patients to describe their disease. Motivations for revising of sufficient practical importance to justify using it as a the 1981 Seizure Classification are listed below. seizure classifier. Retained awareness means that the per- son is aware of self and environment during the seizure, 1 Some seizure types, for example, tonic seizures or epilep- even if immobile. A focal aware seizure (with or without tic spasms, can have either a focal or generalized onset. any subsequent classifiers) corresponds to the prior term 2 Lack of knowledge about the onset makes a seizure “simple partial seizure.” A focal impaired awareness unclassifiable and difficult to discuss with the 1981 sys- seizure (with or without any subsequent classifiers) corre- tem. sponds to the prior term “complex partial seizure.” 3 Retrospective seizure descriptions often do not specify a Impaired awareness during any part of the seizure renders level of consciousness, and altered consciousness, it a focal impaired awareness seizure. In addition, focal sei- although central to many seizures, is a complicated con- zures are subgrouped as those with motor and nonmotor cept. signs and symptoms at the onset. If both motor and nonmo- 4 Some terms in current use do not have high levels of tor signs are present at the seizure start, the motor signs community acceptance or public understanding, such as will usually dominate, unless non-motor (e.g., sensory) “psychic,” “partial,” “simple partial,” “complex partial,” symptoms and signs are prominent. and “dyscognitive.” Focal aware or impaired awareness seizures optionally 5 Some important seizure types are not included. may be further characterized by one of the listed motor onset or nonmotor onset symptoms, reflecting the first prominent sign or symptom in the seizure, for example, Results focal impaired awareness automatism seizure. Seizures Classification of seizure types should be classified by the earliest prominent motor onset or Figure 1 depicts the basic and Figure 2 depicts the nonmotor onset feature, except that a focal behavior arrest expanded 2017 seizure classification. The two represent the seizure is one for which cessation of activity is the dominant same classification, with collapse of the subcategories to feature throughout the seizure, and any significant impair- form the basic version. Use of one versus the other depends ment of awareness during the course of the seizure causes a on the desired degree of detail. Variations on the individual focal seizure to be classified as having impaired awareness. seizure theme can be added for focal seizure types according Classification according to onset has an anatomic basis, to level of awareness. whereas classification by level of awareness has a Figure 1. The basic ILAE 2017 operational classification of seizure types. 1 Definitions, other seizure types and descriptors are listed in the accompanying paper and glossary of terms. 2Due to inadequate information or inability to place in other categories. Epilepsia ILAE Epilepsia, 58(4):522–530, 2017 doi: 10.1111/epi.13670  525 Operational Classification of Seizure Types Figure 2. The expanded ILAE 2017 operational classification of seizure types. The following clarifications should guide the choice of seizure type. For focal seizures, specification of level of awareness is optional. Retained awareness means the person is aware of self and environment during the seizure, even if immobile. A focal aware seizure corresponds to the prior term simple partial seizure. A focal impaired aware- ness seizure corresponds to the prior term complex partial seizure, and impaired awareness during any part of the seizure renders it a focal impaired awareness seizure. Focal aware or impaired awareness seizures optionally may further be characterized by one of the motor-onset or nonmotor-onset symptoms below, reflecting the first prominent sign or symptom in the seizure. Seizures should be clas- sified by the earliest prominent feature, except that a focal behavior arrest seizure is one for which cessation of activity is the dominant feature throughout the seizure. A focal seizure name also can omit mention of awareness when awareness is not applicable or unknown and thereby classify the seizure directly by motor onset or nonmotor-onset characteristics. Atonic seizures and epileptic spasms would usually not have specified awareness. Cognitive seizures imply impaired language or other cognitive domains or positive features such as d
eja vu, hallucinations, illusions, or perceptual distortions. Emotional seizures involve anxiety, fear, joy, other emotions, or appearance of affect without subjective emotions. An absence is atypical because of slow onset or termination or significant changes in tone supported by atypical, slow, generalized spike and wave on the EEG. A seizure may be unclassified due to inadequate information or inability to place the type in other categories. 1Definitions, other seizure types and descriptors are listed in the accompanying paper and glossary of terms. 2 Degree of awareness usually is not specified. 3Due to inadequate information or inability to place in other categories. Epilepsia ILAE behavioral basis, justified by the practical importance of onset may be omitted when a subsequent term generates an impaired awareness. Both methods of classification are unambiguous seizure name. available and can be used in concert. Brief behavioral arrest The classification of an individual seizure can stop at any at the start of a seizure often is imperceptible, and so it is not level: a “focal onset” or “generalized onset” seizure, with no used as a classifier unless dominant throughout the seizure. other elaboration, or a “focal sensory seizure,” “focal motor The earliest (anatomic) classifier will not necessarily be the seizure,” “focal tonic seizure,” or “focal automatism sei- most significant behavioral feature of a seizure. For exam- zure,” and so on. Additional classifiers are encouraged, and ple, a seizure might start with fear and progress to vigorous their use may depend on the experience and purposes of the focal clonic activity resulting in falling. This seizure would person classifying the seizure. The terms focal onset and still be a focal emotional seizure (with or without impair- generalized onset are for purposes of grouping. No infer- ment of awareness), but free text description of the ensuing ence is made that each seizure type exists in both groups; features would be very useful. including absence seizures in the generalized-onset cate- A focal seizure name can omit mention of awareness gory does not imply existence of “focal absence” seizures. when awareness is not applicable or unknown, thereby clas- When the primacy of one versus another key symptom or sifying the seizure directly by motor onset or nonmotor sign is unclear, the seizure can be classified at a level above onset characteristics. The terms motor onset and nonmotor the questionably applicable term with additional descriptors Epilepsia, 58(4):522–530, 2017 doi: 10.1111/epi.13670 526 R. S. Fisher et al. of seizure semiology relevant to the individual seizure. Any may be referred to by the single word “unclassified” or with signs or symptoms of seizures, suggested descriptor terms additional features, including motor, nonmotor, tonic–clo- as listed in the companion paper or free text descriptions can nic, epileptic spasms, and behavior arrest. A seizure type of optionally be appended to the seizure type as descriptions, unknown onset may later become classified as either of but they do not alter the seizure type. focal or generalized onset, but any associated behaviors The seizure type “focal to bilateral tonic–clonic” is a spe- (e.g., tonic–clonic) of the previously unclassified seizure cial seizure type, corresponding to the 1981 phrase “partial will still apply. In this regard, the term “unknown onset” is a onset with secondary generalization.” Focal to bilateral placeholder—not a characteristic of the seizure, but of tonic–clonic reflects a propagation pattern of a seizure, ignorance. rather than a unitary seizure type, but it is such a common and important presentation that the separate categorization Reasons for decisions was continued. The term “to bilateral” rather than “sec- The terminology for seizure types is designed to be useful ondary generalized” was used to further distinguish this for communicating the key characteristics of seizures and to focal-onset seizure from a generalized-onset seizure. The serve as one of the key components of a larger classification term “bilateral” is used for propagation patterns and “gener- for the epilepsies, which is being developed by a separate alized” for seizures that engage bilateral networks from ILAE Classification Task Force. The basic framework of onset. seizure classification used since 1981 was maintained. Seizure activity propagates through brain networks, sometimes leading to uncertainty about whether an event is Focal versus partial a unitary seizure or a series of multiple seizures starting In 1981, the Commission declined to designate as “focal” from different networks (“multifocal”). A single unifocal a seizure that might involve an entire hemisphere, so the seizure can present with multiple clinical manifestations as term “partial” was preferred. The 1981 terminology was in a a result of propagation. The clinician will need to determine way prescient of the modern emphasis on networks, but (by observation of a continuous evolution or stereotypy “partial” conveys a sense of part of a seizure, rather than a from seizure-to-seizure) whether an event is a single seizure location or anatomic system. The term “focal” is more or a series of different seizures. When a single focal seizure understandable in terms of seizure-onset location. presents with a sequence of signs and symptoms, then the seizure is named for the initial prominent sign or symptom, Focal versus generalized reflecting the usual clinical practice of identifying the sei- In 20101 the ILAE defined focal as “originating within zure onset focus or network. For example, a seizure begin- networks limited to one hemisphere. They may be discretely ning with sudden inability to understand language followed localized or more widely distributed. Focal seizures may by impaired awareness and clonic left arm jerks would be originate in subcortical structures.” Generalized from onset classified as a “focal impaired awareness (nonmotor onset) seizures were defined as “originating at some point within, cognitive seizure” (progressing to clonic left arm jerks). and rapidly engaging, bilaterally distributed networks.” The terms in parentheses are optional. The formal seizure Classifying a seizure as having apparently generalized onset type in this example is determined by the cognitive non- does not rule out a focal onset obscured by limitations of our motor onset and presence of altered awareness during any current clinical methods, but this is more an issue of correct point of the seizure. diagnosis than of classification. Furthermore, focal seizures Generalized seizures are divided into motor and non- may rapidly engage bilateral networks, whereas classifica- motor (absence) seizures. Further subdivisions are similar tion is based on unilateral onset. For some seizure types, for to those of the 1981 classification, with the addition of example, epileptic spasms, the distinction of a focal versus myoclonic–atonic seizures, common in epilepsy with generalized onset may require careful study of a video-EEG myoclonic–atonic seizures (Doose syndrome28), recording or the type of onset may be unknown. A distinc- myoclonic–tonic–clonic seizures common in juvenile myo- tion between focal and generalized onset is a practical one, clonic epilepsy,29 myoclonic absence,30 and absence sei- and may change with advances in ability to characterize the zures with eyelid myoclonia seen in the syndrome described onset of seizures. by Jeavons and elsewhere.31 Generalized manifestations of Focality of seizure onset can be inferred by pattern seizures can be asymmetrical, rendering difficult the dis- matching to known focal-onset seizures, even when the tinction from focal-onset seizures. The word “absence” has focality is not clear strictly in terms of observable behavior. a common meaning, but an “absent stare” is not synony- A seizure is focal, for example, when it starts with d
eja vu mous with an absence seizure, since arrest of activity also and then progresses to loss of awareness and responsive- occurs in other seizure types. ness, lip-smacking, and hand-rubbing for a minute. There is The 2017 classification allows appending of a limited nothing intrinsically “focal” in the description, but video- number of qualifiers to seizures of unknown onset, in order EEG recordings of countless similar seizures have previ- to better characterize the seizure. Seizures of unknown onset ously shown focal onsets. If the epilepsy type is known, the Epilepsia, 58(4):522–530, 2017 doi: 10.1111/epi.13670  527 Operational Classification of Seizure Types onset can be presumed even if it is unwitnessed; for exam- measurements of awareness, responsiveness, memory, and ple, an absence seizure in a person with known juvenile a sense of self as distinct from others. The 1981 classifica- absence epilepsy. tion specifically mentioned awareness and responsiveness, Clinicians have long been aware that so-called general- but not memory for the event. ized seizures, for example, absence seizures with EEG gen- Retrospective determination of state of consciousness eralized spike-waves, do not manifest equally in all parts of can be difficult. An untrained classifier might assume that a the brain. The Task Force emphasized the concept of bilat- person must be on the ground, immobile, unaware, and eral, rather than generalized, involvement of some seizures, unresponsive (e.g., “passed out”) for a seizure to show since seizures can be bilateral without involving every brain impaired consciousness. The Task Force adopted state of network. The bilateral manifestations need not be symmet- awareness as a relatively simple surrogate marker for con- ric. The term “focal to bilateral tonic–clonic” was substi- sciousness. “Retained awareness” is considered to be an tuted for “secondarily generalized.” The term “generalized” abbreviation for “seizures with no impairment of conscious- was maintained for seizures generalized from onset. ness during the event.” We employ an operational definition of awareness as knowledge of self and environment. In this Unknown onset context, awareness refers to perception or knowledge of Clinicians commonly hear about tonic–clonic seizures events occurring during a seizure, not to knowledge of for which the onset was unobserved. Perhaps, the patient whether a seizure occurred. In several languages, “unaware” was asleep, alone, or observers were too distracted by the translates as “unconscious,” in which case changing the sei- manifestations of the seizure to notice the presence of focal zure designation from “complex partial” to “impaired features. There should be an opportunity to provisionally awareness” will emphasize the importance of consciousness classify this seizure, even in the absence of knowledge about by putting its surrogate directly in the seizure title. In Eng- its origin. The Task Force therefore allowed further descrip- lish, “focal aware seizure” is shorter than is “focal seizure tion of seizures of unknown onset when key characteristics, without impairment of consciousness” and possibly better such as tonic–clonic activity or behavior arrest are observed understood by patients. As a practical issue, retained aware- during the course of the seizure. The Task Force recom- ness usually includes the presumption that the person having mends classifying a seizure as having focal or generalized the seizure later can recall and validate having retained onset only when there is a high degree of confidence (e.g., awareness; otherwise, impaired awareness may be assumed. ≥80%, arbitrarily chosen to parallel the usual allowable beta Exceptional seizures present with isolated transient epilep- error) in the accuracy of the determination; otherwise, the tic amnesia in clear awareness,38 but classification of an seizure should remain unclassified until more information is amnestic seizure as a focal aware seizure would require available. clear documentation by meticulous observers. Awareness It may be impossible to classify a seizure at all, either may be left unspecified when the extent of awareness cannot because of incomplete information or because of the unu- be ascertained. sual nature of the seizure, in which case it is called an Responsiveness may or may not be compromised during unclassified seizure. Categorization as unclassified should a focal seizure.39 Responsiveness does not equate to aware- be used only for the exceptional situation in which the clini- ness or consciousness, since some people are immobilized cian is confident that the event is a seizure but cannot further and consequently unresponsive during a seizure, but still classify the event. able to observe and recall their environment. In addition, responsiveness often is not tested during seizures. For these Consciousness and awareness reasons, responsiveness was not chosen as a primary feature The 1981 classification and the revision in 20101,10,32 for seizure classification, although responsiveness can be suggested a fundamental distinction between seizures with helpful in classifying the seizure when it can be tested, and loss or impairment of consciousness and those with no degree of responsiveness may be relevant to the impact of a impairment of consciousness. Basing a classification on seizure. The term “dyscognitive” was not carried into the consciousness (or one of its allied functions) reflects a prac- current classification as a synonym for “complex partial” tical choice that seizures with impaired consciousness because of lack of clarity and negative public and profes- should often be approached differently from those with sional feedback. unimpaired consciousness, for example, with respect to Awareness is not a classifier for generalized-onset sei- allowing driving in adults or interfering with learning. The zures, because the large majority of generalized seizures ILAE chose to retain impairment of consciousness as a key present with impaired awareness or full loss of conscious- concept in the grouping of focal seizures. However, con- ness. However, it is recognized that awareness and respon- sciousness is a complex phenomenon, with both subjective siveness can be at least partially retained during some and objective components.33 Multiple different types of generalized seizures, for example, with brief absence sei- consciousness have been described for seizures.34 Surrogate zures,40 including absence seizures with eyelid myoclonias markers35–37 for consciousness usually comprise or myoclonic seizures. Epilepsia, 58(4):522–530, 2017 doi: 10.1111/epi.13670 528 R. S. Fisher et al. Etiology awareness” can convey meaning to a lay person with no A classification of seizure types can be applied to seizures knowledge of seizure classification. Third, the words “com- of different etiologies. A posttraumatic seizure or a reflex plex” and “simple” can be misleading in some contexts. seizure may be focal with or without impairment of aware- Complex seems to imply that this seizure type is more com- ness. Knowledge of the etiology, for instance, presence of a plicated or difficult to understand than other seizure types. focal cortical dysplasia, can aid in classification of the sei- Calling a seizure “simple” may trivialize its impact to a zure type. Any seizure can become prolonged, leading to patient who does not find the manifestations and conse- status epilepticus of that seizure type. quences of the seizures to be at all simple. Supportive information Convulsion As part of the diagnostic process, a clinician will com- The term “convulsion” is a popular, ambiguous, and monly use supportive evidence to help classify a seizure, unofficial term used to mean substantial motor activity dur- even though that evidence is not part of the classification. ing a seizure. Such activity might be tonic, clonic, myoclo- Such evidence may include videos brought in by family, nic, or tonic–clonic. In some languages, convulsions and EEG patterns, lesions detected by neuroimaging, laboratory seizures are considered synonyms and the motor component results such as detection of antineuronal antibodies, gene is not clear. The word “convulsion” is not part of the 2017 mutations, or an epilepsy syndrome diagnosis known to be seizure classification, but will undoubtedly persist in popu- associated with either focal or generalized seizures or both, lar usage. such as Dravet syndrome. The seizures usually can be classi- fied on the basis of symptoms and behavior, provided that Added terms good subjective and objective descriptions are available. Use of any available supportive information to classify the sei- Aware/impaired awareness zure is encouraged. Availability of supportive information As discussed earlier, these terms designate knowledge of may not exist in the resource-poor parts of the world, which self and environment during a seizure. may lead to a less specific, but still correct classification. Hyperkinetic ICD-9, ICD-10, ICD-11, and ICD-12 Hyperkinetic seizures have been added to the focal sei- The World Health Organization International Classifica- zure category. Hyperkinetic activity comprises agitated tion of Diseases (ICD) is used for inpatient and outpatient thrashing or leg pedaling movements. Hypermotor is an ear- diagnoses, billing, research, and many other purposes.41,42 lier term introduced as part of a different proposed classifi- Concordance between ICD epilepsy diagnoses and ILAE cation by L€uders and colleagues in 1993.43 The term seizure types is desirable for clarity and consistency. This is hypermotor, which contains both Greek and Roman roots, possible only to a limited extent with existing ICD terms, was supplanted in the 2001 ILAE glossary44 and 2006 since ICD-9, ICD-10, and ICD-11 are already formulated. report2 by “hyperkinetic,” and to be both etymologically The ILAE proposals will always lead ICD standards. ICD-9 and historically consistent, “hyperkinetic” was chosen for and ICD-10 make use of old seizure terminology, including the 2017 classification. terms such as petit mal and grand mal. ICD-11 does not name seizure types at all, but focuses on epilepsy etiologies Cognitive and syndromes, as do ILAE epilepsy classifications.1 For This term replaces “psychic” and refers to specific cogni- this reason, there is no conflict between our proposed sei- tive impairments during the seizure, for example, aphasia, zure type classification and ICD-11. Efforts can be made to apraxia, or neglect. The word “impairment” is implied incorporate new classifications of seizure types and syn- because seizures never enhance cognition. A cognitive sei- dromes into the development of ICD-12. zure can also comprise positive cognitive phenomena, such as d
eja vu, jamais vu, illusions, or hallucinations. Discussion Emotional A focal nonmotor seizure can have emotional manifesta- Discontinued terms tions, such as fear or joy. The term also encompasses affec- Simple/complex partial tive manifestations with the appearance of emotions After approximately 35 years of use, the terms “simple occurring without subjective emotionality, such as may partial seizure” and “complex partial seizure” may be occur with some gelastic or dacrystic seizures. missed by some clinicians. There are several reasons for changing. First, a decision was previously made1 to globally New focal seizure types change partial to focal. Second, “complex partial” has no Some seizure types that were described previously as intrinsic meaning to the public. The phrase “focal impaired only generalized seizures now appear under seizures of Epilepsia, 58(4):522–530, 2017 doi: 10.1111/epi.13670  529 Operational Classification of Seizure Types focal, generalized and unknown onset. These include Compared to the 1981 classification, certain seizure types epileptic spasms, tonic, clonic, atonic, and myoclonic sei- now appear in multiple categories. Epileptic spasms can be zures. The list of motor behaviors constituting seizure types of focal, generalized, or unknown onset. Represented both comprises the most common focal motor seizures, but other in focal and generalized columns are atonic, clonic, myoclo- less common types, for example, focal tonic–clonic, may be nic, and tonic seizures, although the pathophysiology of encountered. Focal automatisms, autonomic, behavior these seizure types may differ for the focal onset versus gen- arrest, cognitive, emotional, and hyperkinetic are new sei- eralized-onset seizure type of that name. zure types. Focal to bilateral tonic–clonic seizure is a new A companion paper provides guidance on how to apply type as the renamed secondarily generalized seizure. the 2017 classification. Employment of the 2017 classifica- tion in the field for a few years likely will motivate minor New generalized seizure types revisions and clarifications. Relative to the 1981 classification, new generalized sei- zure types include absence with eyelid myoclonia, myoclonic–atonic, and myoclonic–tonic–clonic (although Acknowledgments clonic onset of tonic–clonic seizures was mentioned in the Funding for this study was provided by the International League Against 1981 publication). Seizures with eyelid myoclonia could Epilepsy. The lead author (RSF) was supported by the Maslah Saul MD logically have been placed under the motor category, but Chair, the James & Carrie Anderson Fund for Epilepsy, the Susan Horngren Fund, and the Steve Chen Research Fund. Dr. Mosh
e was supported by since eyelid myoclonia are most significant as features of grant 1U54NS100064. SLM is supported by Charles Frost Chair in Neuro- absence seizures, seizures with eyelid myoclonia were surgery and Neurology, grants from the National Institutes of Health (NIH) placed in the nonmotor/absence category. Seizures with NS43209, Citizens United, the U.S. Department of Defense for Research in Epilepsy (CURE), the Heffer Family and the Segal Family Foundations and eyelid myoclonia may even rarely display focal features.45 the Abbe Goldstein/Joshua Lurie and Laurie Marsh/ Dan Levitz families. Similarly, myoclonic absence seizures potentially have Special thanks are given to the Revision Task Force appointed to revise the classification after receipt of public comments. Members of this Revision features of both absence and motor seizures, and could Task Force do not necessarily concur with all details of the classification or have been placed in either group. Epileptic spasms are sei- the publication, since opinions were not always concordant. These Task zures represented in focal, generalized, and unknown onset Force members were the following: Carol D’Souza, Sheryl Haut, Ernest categories, and the distinction may require video-EEG Somerville, Michael Sperling, Andreas Schulze-Bonhage, and Elza Marcia Yacubian. Additional key comments were received from Soheyl Noachtar, recording. The term “epileptic” is implied for every seizure Kimford Meador, and Kevin Graber. type, but explicitly stated for epileptic spasms, because of the ambiguity of the single word “spasms” in neurologic use. Disclosure of Conflict of Interest What is different from the 1981 classification? Disclosures relevant to classification: Dr. Fisher has stock options from Table 1 summarizes the changes in the ILAE 2017 sei- Avails Pharmaceuticals, Cerebral Therapeutics, Zeto, and Smart Monitor, zure type classification from the 1981 classification. Note and research grants from Medtronic and the National Science Foundation that several of these changes were already incorporated (NSF). J. A. French discloses support via The Epilepsy Study Consortium, which pays Dr French’s university employer for her consultant time related into the 2010 revision of terminology and subsequent to Acorda, Alexza, Anavex, BioPharm Solutions, Concert, Eisai, Georgia revisions.1,32 Regents University, GW Pharma, Marathon, Marinus, Neurelis, Novartis, Pfizer, Pfizer-Neusentis, Pronutria, Roivant, Sage, SciFluor, SK Life Sciences, Takeda, Turing, UCB Inc., Ultragenyx, Upsher Smith, Xenon Pharmaceuticals, and Zynerba; and grants and research from Acorda, Table 1. Changes in seizure type classification from 1981 Alexza, LCGH, Eisai Medical Research, Lundbeck, Pfizer, SK Life to 2017 Sciences, UCB, Upsher-Smith, and Vertex; and grants from the National Institute of Neurological Disorders and Stroke (NINDS), Epilepsy Therapy 1. Change of “partial” to “focal” Project, Epilepsy Research Foundation, and the Epilepsy Study Consor- 2. Certain seizure types can be either of focal, generalized, tium. She is on the editorial board of Lancet Neurology, Neurology Today or unknown onset and Epileptic Disorders, and was an Associate Editor of Epilepsia, for 3. Seizures of unknown onset may have features that can still which she received a fee. Sheryl Haut is a consultant for Acorda and Neure- be classified lis. Edouard Hirsch has received honoraria for lectures and/or advice from 4. Awareness is used as a classifier of focal seizures Novartis, Eisai, and UCB. Dr. Mosh
e is the Charles Frost Chair in Neuro- 5. The terms dyscognitive, simple partial, complex partial, psychic, surgery and Neurology and funded by grants from the National Institutes of and secondarily generalized were eliminated Health (NIH) NS43209, Citizens United for Research in Epilepsy (CURE), the U.S. Department of Defense, the Heffer Family and the Segal Family 6. New focal seizure types include automatisms, autonomic, Foundations, and the Abbe Goldstein/Joshua Lurie and Laurie Marsh/Dan behavior arrest, cognitive, emotional, hyperkinetic, sensory, Levitz families, and receives from Elsevier an annual compensation for his and focal to bilateral tonic–clonic seizures. Atonic, clonic, work as Associate Editor in Neurobiology of Disease and royalties from epileptic spasms, myoclonic, and tonic seizures can be either two books he co-edited. He received a consultant’s fee from Eisai, and focal or generalized UCB. Jukka Peltola has participated in clinical trials for Eisai, UCB, and 7. New generalized seizure types include absence with eyelid Bial; received research grants from Eisai, Medtronic, UCB, and Cyberon- myoclonia, myoclonic absence, myoclonic–tonic–clonic, myoclonic– ics; received speaker honoraria from Cyberonics, Eisai, Medtronic, Orion atonic, and epileptic spasms Pharma, and UCB; received support for travel to congresses from Cyberon- ics, Eisai, Medtronic, and UCB; and participated in advisory boards for Epilepsia, 58(4):522–530, 2017 doi: 10.1111/epi.13670 530 R. S. Fisher et al. Cyberonics, Eisai, Medtronic, UCB, and Pfizer. Dr Scheffer serves on the 20. Shorvon SD. The etiologic classification of epilepsy. Epilepsia editorial boards of Neurology and Epileptic Disorders; may accrue future 2011;52:1052–1057. revenue on pending patent WO61/010176 (filed: 2008): Therapeutic Com- 21. Beghi E. The concept of the epilepsy syndrome: how useful is it in clin- pound; and has received speaker honoraria/consultant fees from ical practice? Epilepsia 2009;50(Suppl. 5):4–10. GlaxoSmithKline, Athena Diagnostics, UCB, Eisai, and Transgenomics. 22. Tuxhorn I, Kotagal P. Classification. Semin Neurol 2008;28:277–288. Dr Zuberi is Editor-in-Chief of the European Journal of Paediatric Neurol- 23. Luders H, Acharya J, Baumgartner C, et al. A new epileptic seizure ogy for which he receives an annual honorarium from Elsevier Ltd. He has classification based exclusively on ictal semiology. Acta Neurol Scand received research funding from Dravet Syndrome UK, Epilepsy Research 1999;99:137–141. UK, UCB Pharma, and Glasgow Children’s Hospital Charity. The remain- 24. Seino M. Classification criteria of epileptic seizures and syndromes. ing authors listed no disclosures relevant to the classification of seizure Epilepsy Res 2006;70(Suppl. 1):S27–S33. types. We confirm that we have read the Journal’s position on issues 25. Fisher RS, Boas WV, Blume W, et al. Epileptic seizures and epilepsy: involved in ethical publication and affirm that this report is consistent with definitions proposed by the International League against Epilepsy those guidelines. (ILAE) and the International Bureau for Epilepsy (IBE). Epilepsia 2005;46:470–472. 26. Brodtkorb E. Common imitators of epilepsy. Acta Neurol Scand Suppl References 2013;196:5–10. 27. Trinka E, Cock H, Hesdorffer D, et al. A definition and classification 1. Berg AT, Berkovic SF, Brodie MJ, et al. Revised terminology and con- of status epilepticus–Report of the ILAE Task Force on classification cepts for organization of seizures and epilepsies: report of the ILAE of status epilepticus. Epilepsia 2015;56:1515–1523. Commission on Classification and Terminology, 2005–2009. Epilepsia 28. Kelley SA, Kossoff EH. Doose syndrome (myoclonic-astatic epilepsy): 2010;51:676–685. 40 years of progress. Dev Med Child Neurol 2010;52:988–993. 2. Engel J Jr. Report of the ILAE classification core group. Epilepsia 29. Wolf P, Yacubian EM, Avanzini G, et al. Juvenile myoclonic epilepsy: 2006;47:1558–1568. a system disorder of the brain. Epilepsy Res 2015;114:2–12. 3. Gastaut H, Magnus O, Caveness W, et al. A proposed international 30. Verrotti A, Greco R, Chiarelli F, et al. Epilepsy with myoclonic classification of epileptic seizures. Epilepsia 1964;5:297–306. absences with early onset: a follow-up study. J Child Neurol 4. Gastaut H. Classification of the epilepsies. Proposal for an international 1999;14:746–749. classification. Epilepsia 1969;10(Suppl.):14–21. 31. Striano S, Capovilla G, Sofia V, et al. Eyelid myoclonia with absences 5. Blumenfeld H. What is a seizure network? Long-range network conse- (Jeavons syndrome): a well-defined idiopathic generalized epilepsy quences of focal seizures. Adv Exp Med Biol 2014;813:63–70. syndrome or a spectrum of photosensitive conditions? Epilepsia 6. Centeno M, Carmichael DW. Network connectivity in epilepsy: rest- 2009;50(Suppl. 5):15–19. ing state fMRI and EEG-fMRI contributions. Front Neurol 2014; 32. Scheffer IE. Epilepsy: a classification for all seasons? Epilepsia 5:93. 2012;53(Suppl. 2):6–9. 7. Proposal for revised clinical and electroencephalographic classifica- 33. Ali F, Rickards H, Cavanna AE. The assessment of consciousness dur- tion of epileptic seizures. From the Commission on Classification and ing partial seizures. Epilepsy Behav 2012;23:98–102. Terminology of the International League Against Epilepsy. Epilepsia 34. Luders H, Amina S, Bailey C, et al. Proposal: different types of alter- 1981;22:489–501. ation and loss of consciousness in epilepsy. Epilepsia 2014;55:1140– 8. Berg AT. Classification and epilepsy: the future awaits. Epilepsy Curr 1144. 2011;11:138–140. 35. Cavanna AE, Monaco F. Brain mechanisms of altered conscious states 9. Berg AT, Scheffer IE. New concepts in classification of the epilepsies: during epileptic seizures. Nat Rev Neurol 2009;5:267–276. entering the 21st century. Epilepsia 2011;52:1058–1062. 36. Blumenfeld H. Impaired consciousness in epilepsy. Lancet Neurol 10. Korff CM, Scheffer IE. Epilepsy classification: a cycle of evolution 2012;11:814–826. and revolution. Curr Opin Neurol 2013;26:163–167. 37. Blumenfeld H, Taylor J. Why do seizures cause loss of consciousness? 11. Berg AT, Blackstone NW. Concepts in classification and their rele- Neuroscientist 2003;9:301–310. vance to epilepsy. Epilepsy Res 2006;70(Suppl. 1):S11–S19. 38. Felician O, Tramoni E, Bartolomei F. Transient epileptic amnesia: 12. Engel J Jr. A proposed diagnostic scheme for people with epileptic sei- update on a slowly emerging epileptic syndrome. Rev Neurol (Paris) zures and with epilepsy: report of the ILAE Task Force on Classifica- 2015;171:289–297. tion and Terminology. Epilepsia 2001;42:796–803. 39. Yang L, Shklyar I, Lee HW, et al. Impaired consciousness in epilepsy 13. Engel J Jr. ILAE classification of epilepsy syndromes. Epilepsy Res investigated by a prospective responsiveness in epilepsy scale (RES). 2006;70(Suppl. 1):S5–S10. Epilepsia 2012;53:437–447. 14. Berg AT, Cross JH. Classification of epilepsies and seizures: historical 40. Porter RJ, Penry JK. Responsiveness at the onset of spike-wave bursts. perspective and future directions. Handb Clin Neurol 2012;107:99– Electroencephalogr Clin Neurophysiol 1973;34:239–245. 111. 41. Bergen DC, Beghi E, Medina MT. Revising the ICD-10 codes for epi- 15. Luders HO, Amina S, Baumgartner C, et al. Modern technology calls lepsy and seizures. Epilepsia 2012;53(Suppl. 2):3–5. for a modern approach to classification of epileptic seizures and the 42. Jette N, Beghi E, Hesdorffer D, et al. ICD coding for epilepsy: past, epilepsies. Epilepsia 2012;53:405–411. present, and future–a report by the International League Against Epi- 16. Luders H, Acharya J, Baumgartner C, et al. Semiological seizure clas- lepsy Task Force on ICD codes in epilepsy. Epilepsia 2015;56:348– sification. Epilepsia 1998;39:1006–1013. 355. 17. Panayiotopoulos CP. The new ILAE report on terminology and con- 43. Luders HO, Burgess R, Noachtar S. Expanding the international classi- cepts for the organization of epilepsies: critical review and contribu- fication of seizures to provide localization information. Neurology tion. Epilepsia 2012;53:399–404. 1993;43:1650–1655. 18. Panayiotopoulos CP. The new ILAE report on terminology and con- 44. Blume WT, Luders HO, Mizrahi E, et al. Glossary of descriptive ter- cepts for organization of epileptic seizures: a clinician’s critical view minology for ictal semiology: report of the ILAE task force on classifi- and contribution. Epilepsia 2011;52:2155–2160. cation and terminology. Epilepsia 2001;42:1212–1218. 19. Gomez-Alonso J, Bellas-Lamas P. [The new International League 45. Mourente-Diaz S, Montenegro MA, Lowe JP, et al. Unusual focal ictal Against Epilepsy (ILAE) classification of epilepsies: a step in the pattern in children with eyelid myoclonia and absences. Pediatr Neurol wrong direction?]. Rev Neurol 2011;52:541–547. 2007;37:292–295. Epilepsia, 58(4):522–530, 2017 doi: 10.1111/epi.13670