Contact Us

Ilae Facebook Ilae Twitter

Seizure Type Classification

ILAE logo

The ILAE Commission on Classification and Terminology strives to continually review definitions of seizure types, syndromes, and epilepsy aetiologies, aligned to the latest developments in this field. In 2010 the Commission introduced new concepts into the classification, or as it was termed then organisation, of seizure types and the epilepsies taking a more aetiological approach than previous classifications [1]. This document and a subsequent report, published online, invited comment from the epilepsy community [2]. Debate within the epilepsy community has been well informed and passionate.

The Commission 2013-17 has re-introduced the term classification and charged two Task Forces to further develop the Classification of Seizures and Epilepsies. A Task Force on Classification is developing a framework for classification of the epilepsies and focusing on areas of terminology. This will be presented for comment in the new ILAE journal Epilepsia Open in the near future. The Seizure Type Classification Task Force has focused on developing a clinically-focused seizure type classification which is prese nted below for comment.

Based upon the comments from the epilepsy community, this Seizure Classification expands upon and slightly modifies the seizure classification described in Table 2 of the 2010 proposal [1]. The goal is to establish a final seizure classification scheme accepted by ILAE and the broad epilepsy community. The Commission on Classification & Terminology relies on comments from the wider epilepsy community to ensure that definitions and classifications are conceptually sound, acceptable and clinically meaningful. In addition, they have to translate across the many different languages in which they may be used, and work in a variety of different health care settings and in different media.

Operational Classification of Seizure Types by the International League Against Epilepsy
Please send your comments to who will post them here for the entire community to consider.

Comments are open for 2 months, i.e. until 4 June 2016. Comments will be posted on the web site and will be reviewed by the authors of the classification document prior to its finalization.

  1. Berg AT, Berkovic SF, Brodie MJ, Buchhalter J, Cross JH, van Emde Boas W, Engel J, French J, Glauser TA, Mathern GW, Moshé SL, Nordli D, Plouin P, Scheffer IE. Revised terminology and concepts for organization of seizures and epilepsies: report of the ILAE Commission on Classification and Terminology, 2005-2009. Epilepsia. 2010 Apr;51(4):676-85.
  2. http://www.ilae.org/Visitors/Centre/Organization.cfm


15 August 2016

Dear Dr. Fisher and members of the ILAE committee for Seizure classification 2016,

Congratulations on your efforts to propose a new Seizure classification, a much needed step in bringing about uniformity in naming seizures scientifically yet in a simple and reproducible method.

May I suggest a few points:

  • I agree with many others about the concern over the omission of ''automotor'' from the subheading 'motor'. The heading 'Automatisms' features in the glossary of 'Descriptors', but not in the classification. This could create much confusion.
  • In addition, however, many other motor characteristics, e.g. "dystonic" and "versive'' feature in the glossary but not under the 'motor' heading, and these features cannot be categorized under any of the currently listed subheadings like ''tonic''. In contrast, others in the 'descriptors' list like 'Figure 4' can be categorized under ''tonic'' (bilateral, asymmetrical).
  • Most importantly, this is a good opportunity to provide for mention of topographical distribution, mainly of 'motor' features. Otherwise, while someone who has Left face clonic seizures on first presentation, and right upper limb dystonic seizures several years thereafter, would continue to have seizure categorization as 'Focal motor', despite scope for easy subclassification.

It would be nice to have these points considered.

Thank you!

Garima Shukla

10 August 2016

Classifying seizures only gives meaning when the classification has clinical consequences, concerning work-up strategy, prognosis and treatment. To achieve this, at least two aspects have to be known: semiology and EEG. Classifying a seizure merely based on semiology can be misleading. A staring spell can be either a focal seizure with affected consciousness or an absence.

One could consider the following:

Axis-1: Semiology

  1. Motor signs
    1. Elementary motor signs (limited to one region of the body or unilateral or bilateral symmetric / asymmetric)
      1. Twitches (<0.2 s)
      2. Spasm (0.5-2 s)
      3. Tonic muscle activation (>3 s)
      4. Atonic muscle activation (+ duration specified)
      5. Astatic seizures (when the origin of the fall cannot be specified as tonic or atonic)
    2. Automatisms
    3. Hypermotor
    4. Hypomotor
    5. Motor or behavioral arrest
  2. Sensory signs
  3. Dyscognitive signs
  4. Autonomic signs

Axis-2: EEG
EEGs can be scored using standardized terms (IFCN Glossary; SCORE)

Based on the two axes, seizures can be classified as:

  1. Focal (unilateral network seizures)
  2. Bilateral network seizures (formerly known as generalized seizures).
  3. Seizures of unknown origin

Then, seizures can be characterized using the semiological and EEG descriptors.

Hermann Stefan and Sándor Beniczky

Return to top

1 August 2016

Thank you for the great work in seizure classification. I think this is a really practical classification, different from the 1981 one. So, I have a question: Pre-, post- and ictal bursts of sudden running had been reported initially by French semiologists at least since the 1880s, who called it “procursive seizures.” This is currently named “leaving behavior” (LB), and recently has been reported to be a lateralizing sign in MTLE (Jin and Inoue, 2009). In cases of preictal LB, usually a GCT seizure follows it. I think that this behavior, when it is a seizure, must fit into “HYPERMOTOR” sign category (Am I correct?).

A similar procursive behavior so called “jumping seizure” and “running seizure” is observed in some in vivo models of seizure, notely the audiogenic one, and has been attributed to a focal epileptiform activity in hindbrain (Jobe et al., 1972) arising from deep layers of superior colliculus and integrated in the inferior colliculus (Garcia-Cairasco et al., 1992; Tsutsui et al., 1992; Terra et al., 1992; Dutra Moraes et al., 2000, 2005; Faingold, 2012; Vinogradova et al., 2015, 2016). On the other hand, gelastic crisis is a known focal hindbrain seizure type in humans.

Thereby, the question: Is there evidence that of LB may arise from a focal epileptiform activity in hindbrain structures? This seems probable to me. Although the ictal initiation in many cases may arise in structures out of the brainstem (e.g. in MTLS), it is possible a partial recruitment of brainstem structures in any part of the process.

Best wishes,
Italo Rosal Lustosa, MD, M.Sc.

20 June 2016

May I suggest that the name of focal motor seizures which in the expanded scheme appears as “hypermotor” be replaced by the name “hyperkinetic,” in keeping with the ILAE Glossary of Descriptive Terminology? In addition to that please note that “hypermotor” is highly agrammatical, since “hyper” is derived from Greek whereas “motor” is derived from Latin. The etymologically and grammatically correct form should be either “hyperkinetic” or “supermotor.” “Supermotor” might conjure all manner of inappropriate associations in the public’s mind, so perhaps it would be more appropriate to stick with “hyperkinetic”, the opposite of “hypokinetic.’ Furthermore it is imho not appropriate to adopt “hypermotor,” a term which the arbitrary semiological classification system devised by Dr. Hans Lüders has popularized and which has never been accepted by the ILAE, in an official ILAE scheme.

Thank you.

Demetrios N. Velis

9 June 2016

Dear commission member of the ILAE Commission for Classification and Terminology,

I have following requests regarding epilepsy classification:

First we try to change name epilepsy as electro-clinical disorder with or without neuropsychiatry sign and symptoms to avoid social stigmas with epilepsy and to reduced treatment gap in seizure patients.

  1. Epileptic event vs non epileptic events
  2. If epileptic then associated with encephalopathy or without encephalopathy
  3. Encephalopathy static, progressive or associated with motor,cognitive,behavior or other neuropsychiatry sign and symptoms
  4. Encephalopathy reversible with or without deficits
  5. If without encephalopathy then other associated cognitive,behavior other neuropsychiatry sign and symptoms
Dr Surendra Khosya

6 June 2016

Summary of points raised in debate at Asian-Oceanian Epilepsy Congress, Hong Kong, May 2016

Focal v generalized (concept)


  • 1981: …the first clinical changes indicate initial involvement of both hemispheres.
  • 2016: …originating at some point within, and rapidly engaging, bilaterally distributed networks. … not necessarily include the entire cortex.


  1. Many focal seizures do this. A network is defined by its connections. If I send an email from the computer in my office, it originates in a “local area network” but that network is not limited to my hospital – it is part of a much bigger network called the internet. An email can originate locally and then propagate to the other hemisphere (the Northern Hemisphere) just as a seizure can begin locally and propagate to the other hemisphere. The email is not originating in a network confined to one hemisphere and neither is the seizure.
  2. Focal hypermotor seizures may begin with bilateral motor activity. The network is bilaterally distributed and engagement is rapid.
  3. Intracranial EEG also shows rapid engagement of bilaterally distributed networks (see SEEG from both insular cortices in a focal seizure).


  • 1981: the first clinical (and EEG) changes indicate initial activation of a system of neurons limited to part of one cerebral hemisphere.
    BUT the clinical (and EEG) changes may be bilateral
  • 2016: Originating within networks limited to one hemisphere.
    BUT the networks are not limited to one hemisphere (see above)


  • Focal = originating in a part of a network where that part is limited to one hemisphere.
  • Generalized = all other seizures (if there are only 2 possibilities (focal and generalized), then we need define only one – everything else is the other).

Focal and Generalized (words)

Focal: Gastaut in 1972 rejected focal “to emphasise that the participating neuronal system may be distributed to a variable extent... It cannot be represented geometrically as a locus, even less a focus”. i.e. network rather than focus. But in 2016, we have gone back to focal.

  • 1981 Partial (focal, local)
  • 2016 Focal

Generalized: is not generalized (the seizure does not start everywhere at once)
Suggestion: Use “unilateral (or “local”) and “bilateral” [“onset”]. This would make teaching so much easier.

Corresponding focal and generalized seizure types

The desire to provide focal and generalized versions of several seizure types and seizure groupings is helpful. However, myoclonic seizures in JME are often unilateral and this may lead to incorrect diagnosis of focal epilepsy. Having a focal myoclonic (distinct from focal clonic) facilitates that error.

Motor (word)

“The motor event could consist of an increase (positive) or decrease (negative) in muscle contraction to produce a movement.”
BUT automatisms satisfy this definition
Suggestion: Use “Convulsive,” defined as “sudden movement of a type not resembling normal movements”

Unknown onset

If we can’t tell if focal or generalized onset:

  • 2016: Unknown onset tonic-clonic

BUT we don’t have to state what we don’t know unless someone assumes that we do know!
Suggest: Tonic-clonic [not otherwise specified]
(It’s hard to imagine an emergency department resident calling the neurologist and saying “There is a patient here with an unknown onset tonic-clonic seizure”)

“…Lack of knowledge about the onset makes a seizure unclassifiable with the 1981 system.” This is incorrect:
1981: Unclassified = inadequate information or seizures that don’t fit into the classification
2016: Unknown onset = inadequate information
2016: Unclassified = inadequate information or unusual clinical features
Unknown onset and unclassified are major categories, at least in the figures.

Suggestion: Don’t classify what we can’t classify. Remove the unknown onset and unclassified categories and subcategories from the figure.

One hemisphere to Both hemispheres

  • 1981: Partial seizures evolving to secondarily generalized seizures
    (abbrev. = secondarily generalized seizure)

BUT Using generalized for a focal seizure might confuse?

  • 2016: Focal to bilateral tonic-clonic seizure
    (abbrev. = bilateral tonic-clonic)

BUT bilateral tonic-clonic does not say focal onset
We now have generalized T-C and Bilateral T-C that have very different meanings.
Suggest: Unilateral/focal onset tonic-clonic
AND Not all secondarily generalized sz are tonic-clonic.
(2013 was “convulsive” – it’s unclear why it was changed)

Consciousness v Awareness


  • Consciousness = awareness and/or responsiveness (excl. paralysis, aphasia , apraxia).
  • Awareness = contact with events + memory.
  • Impaired consciousness is defined as the inability to respond normally …[due to] altered awareness and/or responsiveness ”

BUT How do you test “contact with events?”

  • Awareness (Shorthand) = awareness + responsiveness + memory + consciousness.
  • Awareness (Glossary) = an aspect of consciousness pertaining to knowledge of one’s surroundings.
  • Consciousness = a sense of self as a unique entity + awareness + responsiveness + memory
  • Responsiveness = appropriate reaction by movement or speech


  • Consciousness = Responsiveness + memory
    • Responsiveness is what matters to the patient
    • Both can be tested
    • Avoids philosophy and metaphysics

Note: 2016: Responsiveness rejected because it may be due to immobilization and often not tested BUT definition can be defined to exclude unresponsiveness due to immobilization, as it was in 1981)
Note: 2016: Absence (Glossary): includes responsiveness but not memory or awareness.

The elusive term to describe a focal seizure with impaired consciousness

Attempts so far:

  • Complex/simple partial sz
  • Focal sz with/without impaired consciousness
  • Focal dyscognitive sz
  • Focal unresponsive/responsive sz
  • 2016: Focal unaware/aware sz. BUT a patient may be unaware they have had an unaware seizure or they may be aware they have had an unaware seizure and they will certainly be aware they have had an aware seizure. This is confusing!

Suggestion: Keep thinking!
Until we find a good word, use Focal sz with impaired consciousness. (The headache doctors seem to manage OK using “migraine with/without aura”.)

Suggestion: We already have a word for a generalized non-motor seizure with impaired consciousness i.e. Absence. We can redefine “Absence” as “a non-motor seizure with impaired consciousness.”
Focal non-motor unaware seizure becomes Focal absence
Absence becomes Generalized absence

A few factual errors

  • GTC = “bilateral symmetric tonic contraction and then bilateral clonic contraction…”
    BUT GTC are often asymmetric (there are many references)
  • Drop attacks are now atonic BUT they are often tonic seizures.
  • Gustatory seizures are classified as autonomic BUT should be sensory.
  • Limbic is now focal with impaired awareness BUT consciousness not always impaired.
  • Freeze becomes focal arrest BUT focal arrest is not in classification

Ernest Somerville

Return to top

4 June 2016

I greatly appreciate the efforts that have been made to resolve the many issues that were raised during the open discussions of the previous versions, and many of the comments made to date have reflected the appreciation for the efforts made. A number of the correspondents, however, have raised a number of concerns. Although some have focused on the word choice for the new epilepsy lexicon, which will always be debated, a number have also raised questions about the purpose and practical application of this vocabulary throughout the international epilepsy community. It is primarily on these latter issues that I would like to focus.

I have read through the manuscript several times and am still not sure where this classification fits in with those that exist and how it is to be used practically. Perhaps the authors should clearly define the purpose of this classification. Specifically is it a classification/organization of seizure types to align with etiologies and pathophysiology or is it a list of approved terms (which may not translate well into the many languages of our members)? If the former, then strict adherence to a limited list of approved terms becomes less important to the conceptual framework that the authors are trying to create.

Assuming that the goal of this proposal is to categorize a seizure type based on clinical information,, the scheme of basic classification is useful within limits. But the authors may also be trying to help localize or physiology and etiology. As others have already noted, for most clinical settings, especially when the resources are limited, most of the classification is not useful as it requires more information than will be available. Even in systems with more resources, the reality is that a more extensive categorization is not possible, as much of the clinical information will not be available or will be inconclusive. For that reason it might be useful to identify levels of classification clearly (it has been done to some extent with the basic and higher level organizational diagrams, but a clearer explanation about what is considered minimally sufficient is important). As the authors clarify the role of this organization and glossary they should consider whether it is a primary aid to a front line practitioner with limited resources working with a population that may not describe the seizure well or whether it is to be used primarily in epidemiological studies or to define eligibility for prospective trials. If this organization is to be used with different levels of complexity depending on the setting, that variable application of the classification should be made clearer.

If it is to be useful at all levels of resources and complexity greater input from those who work under conditions of limited resources is essential to determine what is really possible (or is simply the identification of epilepsy the most important step?).

Further, it is not clear what advantage the more extensive classification brings to our understanding of the pathophysiology of a seizure type, as many are beyond our current understanding. Overall the classification seems to be at odds with the Newtonian observation that “nature is pleased with simplicity.” Greater organizational and lexical simplicity would no doubt enhance acceptance and utilization.

Similarly, semantic simplicity would also improve acceptance. Although one appreciates the desire for linguistic precision (which as others have noted might be lost in translation), relegating well accepted and used terms to the trash bin of the epilepsy lexicon is counterproductive. The reality is that words can have many meanings that are established by widespread common usage. One just has to consider the word “bomb.” Although one immediately thinks about an explosive device, it can also mean an artistic or financial disaster, extraordinary pulchritude, a tasty confection or little packets or herbals to place in a bath, a junked car, a long pass in American football and an explosive statement among others. Aura has a well-established meaning from years of utilization and may be less intimidating to patients than calling one a simple focal seizure. The “partial” vs. “focal” debate is similar (I have no real preference here). The discussion about preserved awareness has raised numerous issues that may not be resolvable with a single term. One wonders if, practically speaking, it is distinction really worth pursuing. As others suggested, simply describing the effect of the seizure on function may be a useful part of the clinical history, but not add much to our understanding of the pathophysiology. People with seizures of similar origins can have vastly different effects on one’s perception of the world and memory of the event.

It would be good to field test the proposal in the front lines of clinical practice in the many regions and conditions under which this classification will be used. Bring it out of the tower and into the field or clearly indicate that this classification is intended primarily for scientific and clinical research. If the latter, the further discussion can be limited to the academic community.

Edward Bertram

Return to top

4 June 2016

Our comments:

1. “Awareness” could be as in page 6: Heading “How to use the classification” or as in page 8 Heading “Reasons for decisions” subheading ”Consciousness and awareness.” Awareness as on “Added terms” “designate degrees of impairment of consciousness?”

However, on Glossary is “an aspect of consciousness pertaining to knowledge of one's surrounding.” It means that in the “Added Terms” awareness is a degree. On the glossary is a content or a quality. The same is back again in the last paragraph of the heading “Reasons for decisions” subheading ”Consciousness and awareness” on page 9. Responsiveness is not a quality of the awareness.

2. Where the “auto motor” (self-motor) seizures can be included?

3. On supportive information the classification is obliged to include EEG with Polygraphic (EMG, other) recording if the “seizures usually can be classified on the basis of behavior.” How gene mutations, imaging, etc are supporting behavior characterization?

4. Why automatisms are outside the motor? Why undressing is different from manual (if the concept is extended) “Dèja vécu” is missed, why? And why “unpleased” is not included?

Comments from PORTO (Portugal) Epilepsy Unit, reported by António Martins da Silva on behalf of a group of epileptologists: Rui Rangel, João Ramalheira, João Chaves, Joel Feitas, Pedro Guimarães, Paulo Coelho, Sara Cavaco.

3 June 2016

On 12-14 May 2016 the 20th Conference of Ukrainian League against Epilepsy took place in Lviv (Ukraine). We were visited by more than 30 foreign participants and in total 388 participants were registered, including 236 members of ULAE.

During my report named “The classification of epileptic seizures and epilepsy:past, present and future” six questions were asked to the audience and interactive voting was held. For each question, there were three possible answers: “yes,” “no” or “undecided," More than 200 ULAE members took part in the voting process. Voting results are shown below.

1. Do you support abolishment of division of focal seizures into simple and complex?

A=Yes     B=No     C=Undecided

2. Do you support renaming of focal seizures with psychic symptoms into cognitive and emotional?

A=Yes     B=No     C=Undecided

3. Do you support renaming of "focal seizure with secondary generalization" in "focal seizure evolving to bilateral tonic-clonic"?

A=Yes     B=No     C=Undecided

4. Do you support renaming of idiopathic epilepsy into genetic?

A=Yes     B=No     C=Undecided

5. Do you support renaming of symptomatic epilepsy into structural and metabolic?

6. Do you support renaming of cryptogenic epilepsy into epilepsy of unknown cause?

The Organizing Committee of the Conference hopes that these results will be taken into account by the Task Force during discussion on the final version of the classification of epileptic seizures and epilepsy.

On behalf of the Organizing Committee,
with respect,
Professor Lidiya Mar'yenko,

Return to top

3 June 2016

We welcome the new seizure classification, which in our view represents a positive change. In particular, we appreciate the efforts of the Commission to provide translation of terms from older classifications to the new classification and to support the classification by common descriptors and a comprehensive glossary of terms.

The authors stipulate that the new classification is meant for clinical use. We think it would be a missed opportunity not to encourage its use as well in research. Within the EpiPGX consortium on epilepsy pharmacogenomics, we have phenotyped over 10000 patients with epilepsy, recruited from multiple centers in seven European countries. Based on our experience in dealing with heterogeneity in the collection, storage and interpretation of clinical data within and across centers, we propose a few modifications which in our view would allow application of the new classification in a clinical research context and which would undoubtedly benefit clinical practice as well.

The mapping of terms from previous classifications is useful but seems incomplete. Guidance for translation of some other important terms such as primary generalized tonic-clonic seizures or seizures with psychic symptoms is missing. Some new terms are included in table 2 but not in figure 1 (e.g. onset unknown clonic), generating confusion.

It is not entirely clear on which elements the classification is based. In addition to the text which describes the methods for classifying a seizure, it would be useful to provide decision trees or a flowchart to assist with seizure classification. This decision tree would be based on the core data elements and take into account missing information.

Finally, we encourage the Commission to provide standardized computer-readable codes for each of the phenotype items. These would highly facilitate translation of data across different languages and IT systems. Furthermore, such codes would avoid confusion arising from differences in the order of terms used (e.g. “focal aware tonic seizure” and “focal tonic seizure with preserved awareness”).

We thank the Commission for taking our comments into consideration.

Chantal Depondt, Roland Krause, Holger Lerche, Pasquale Striano, Ley Sander, Graeme Sills, John Craig, Sanjay Sisodiya
on behalf of the EpiPGX consortium.

2 June 2016

I would like to thank the Task Force for its work to improve and clarify nomenclature in epileptology. I have two comments.

First, the Task Force acknowledges that impairment of mental functions is an important part of the seizure classification. The case for changing consciousness to awareness is not entirely convincing, though. In “How to use the classification,” impairment of awareness is suggested as a shorthand for the presence of an impairment in any of awareness, responsiveness, recall, or consciousness (or, equivalently, that the patient is aware if all these are intact). There are some problems with this definition; 1) it is circular, 2) it is contradicted further on in the manuscript, where it is noted that a patient may be aware but unresponsive during a seizure, and 3) it will be almost impossible to ascertain that a patient is aware during a seizure (or, according to the slightly peculiar abbreviated terminology, “has an aware seizure”) without knowing the status for all the factors awareness, responsiveness, recall, and consciousness. Of course, consciousness is an ambiguous entity, but that could also turned into an advantage, since consciousness can through its subjective and objective aspects be ascertained in a number of ways (asking a patient afterwards if she recalls what happened during the seizure, asking her to follow commands during the seizure, apply painful stimuli during the postictal phase, etc.).

Second, the proposal aims for feasibility for professionals and laypersons alike. However, the terminology for the type of seizures commonly called tonic-clonic seizures is slightly confusing. According to the proposal, we have generalized tonic-clonic seizures, focal to bilateral tonic-clonic seizures, and (generalized) clonic-to-tonic-clonic seizures. This terminology makes it difficult to name these types of seizures, especially since the exact classification often is not known. Also, it is difficult to group them together, which would make sense when discussing the risk for SUDEP, convulsive status epilepticus, etc. May I suggest the term tonic-clonic seizure (TCS), which may be subclassified into focal to TCS, generalized TCS, and (generalized) clonic to TCS.

Best regards,
Johan Bjellvi, MD

Return to top

1 June 2016

Dear Prof.Robert Fisher and all members of the ILAE Commission for Classification and Terminology,

I think highly of your attempts to develop an operational (practical) classification on seizure type.

According to the new document your mentioned, many important clincial epileptic seizure type, such as (focal) versive, which is commonly recognized as the vital localized sign, could not be included in the below of focal category.

Moreover, it is well known that an epileptic seizure with epileptic patient may always present a sequence of signs and symptoms during an ictal seizure, such as aura( epigastric aura), hand fumbling, unwareness, followed by head versive, sGTCS etc, how can we classify so many signs and symptoms?

Best regards,
Bing Qin, MD, PhD

26 May 2016

I would like to thank the ILAE Commission for Classification and Terminology for putting this document in public domain and asking comments or suggestions from the epilepsy professionals.

In my opinion, there are many cons in this “operational” classification. There are no major changes from 2010 classification, so a new classification may not be required just after 5 years and in view of lack of enough evidence. The suggestions made in this document could have been put in some other format e.g. a commentary article.

Here are few points, which I want to highlight regarding the proposed classification and terminologies:

1. The ommision of “auras” from the proposed classification is not a good idea as there are so many patients with epilepsy who present only with aura especially after surgery. This classification does not offer any place to such patients.

2. I do not agree with the term “aware” being used as a classifier. The phrases “focal aware seizures” and “focal unaware seizures” seem very trivial and they simply do not convey the intented message!! In order to fullfill the goal to make the classification understandable by patients and families and broadly applicable to all ages, the underlying scientific gist should not go away.

3. The new classification extended scheme is very confusing because of the following reasons:

  • Why is there only hypermotor subcategory in focal seizures and not others like hypomotor. There is no reason given as to why some sub-categories are omitted and others are not, in the extended scheme e.g. no mention of hyper or hypomotor in the generalized seizures.
  • Why are there only few seizure subtypes given in the unknown onset category? Can't we encounter a clonic or absence attack with unknown onset?
  • Why don't we have a non-motor subcategory in generalized category? Though such type of seizures are rare but they do happen e.g. brief absence seizures or eyelid myoclonia.

4. In the new seizure classifying terms, absence seizures have been renamed as generalized absence seizures. This gives a wrong impression that there are focal absence seizures also!!

5. There are many confusions in the glossary of terms e.g. definition of emotional seizures, confusion between definitions of hypomotor and activity arrest etc.

The consideration of ICD classification in the ILAE classification is a welcome step and should be sought. It is my humble request to the ILAE that there should be a proper representation of all the geographical areas in the committiee so that there is a greater understanding and wider acceptability of the recommendations.

Vikas Dhiman

Return to top

19 May 2016

Dear members of the ILAE Commission for Classification and Terminology,

Congratulations for trying to make the classification more simple. I have two comments to make.

  1. Inclusion of reflex epileptic seizures is welcome.
  2. I think that “impaired awareness” is not the best word to describe the impaired interaction with the environment.

Best regards,
Romina Buono

18 May 2016

Dear all:

I read with interest the new operational classification of seizure types. I am happy with the rationale for change and the proposed modifications. However, for the sake of clarity, I would like you to address the following points:

1. If research is one of the possible indications for the classification, a problem arises when the level of accuracy is dependent on supportive information; only for the purposes of research, it should be made clear if a seizure was classified based on clinical findings alone or with the support of laboratory/instrumental findings;

2. It is unclear why epidemiologists are mentioned as a category separate from researchers;

3. Figure 1: "automatisms" are not contemplated among focal, non-motor seizures;

4. ICD9, 10, 11, 12: I suggest changing " … even though ICD10 is just being introducted into US systems" into " … even though ICD10 is just being introducted into US systems and is still unavailable in other countries;" then I wonder whether we should mention ICD12 when ICD11 is still in development.

Best regards.
Ettore Beghi

12 May 2016

We highly appreciate the attempt of the Commission to develop an operational (practical) classification.

Our major concerns:

  • Automatisms are missing from the list of “focal-motor” seizures, in the “expanded scheme” although they are discussed in the text (in section “Method for classifying a seizure”) and table-1. We cannot understand why, for example “hypermotor” is included into the “expanded scheme”, but automatism is not.
  • Further classification of “focal” seizures, according to anatomic regions is widely used in clinical practice (temporal lobe seizure, frontal lobe seizure, Rolandic seizure etc.). Please consider adding this at least as an option (alternative) to the “expanded scheme”, if the goal is indeed to develop a practical classification of seizures. These terms are widely used in clinical practice, and it would be detrimental not including them into the operational classification.
  • It is not clear whether this proposal will replace the 2001-Glossary (Blume et al., 2001). Since several terms are re-defined (“adapted from” the 2001 document) in the proposal, the position of the ILAE on which document is valid, should be clear, otherwise this will generate confusion.
  • Table 2: We have concerns with the term “generalized absence”. Is it necessary to use the attribute “generalized” here? This suggests there is a “focal absence” too!
  • Practical challenge: how to classify a typical temporal lobe seizure according to the new proposal? Just consider the example of a typical case, starting with epigastric aura, then continuing with oroalimentary and gestural automatisms, dystonic posturing and loss of awareness. Would this be classified as “focal, sensory seizure with automatisms (not yet in the scheme), tonic (dystonic not in the scheme) and impaired awareness”? Would that be practical / operational?

Minor concerns:

  • Is it correct to list autonomic symptoms and sensations as behavior? Table 1 is called “Common Descriptors of Behaviors…”, yet, autonomic and sensory phenomena are listed there too.
  • Lack of harmonization with the recently published ILAE classification of Status Epilepticus (Trinka et al., 2015). It would be confusing to use different terminology in ILAE documents. Consider re-naming “motor” to “with prominent motor symptoms” and “sensory” to “without prominent motor symptoms.
  • Although we agree with the definition of “myoclonic absences” in “Methods for classifying seizures”, one should emphasize that it is different from typical childhood absence with some myoclonic manifestation (usually eyes or face). A short note would be sufficient to clarify this, and prevent misinterpretation.

Sándor Beniczky and Guido Rubboli, on behalf of the Danish Epilepsy Society

5 May 2016

Dear members of the ILAE Commission for Classification and Terminology,

Congratulations to the team for trying to make the classification more and more simple and user friendly. I have two comments to make.

1. Epileptic spasms with focal onset or nonfocal onset

2. I agree with the comment of Dr S Jain that Video clippings brought by the parents/attendants be added to the description of the seizure.

Roshan Koul

Return to top

3 May 2016

Dear members of the ILAE Commission for Classification and Terminology,

The new operational classification of seizure types by ILAE has made steady progres s as compared to the previous proposal. I would be grateful if my following minor comments are considered.

1. In Table 2, drop attack is replaced by generalized/focal/onset unknown atonic.
However, drop attacks are caused not only by atonic seizures but also myoclonic seizures (more correctly myoclonic-flexor seizures) and epileptic spasms (more correctly axial epileptic spasms).

2. Generalized myoclonic seizures can occur in isolation or in conjunction with tonic or atonic activity. Myoclonus differs from clonus by being briefer and not regularly repetitive.
Myoclonic seizures seen in juvenile myoclonic epilepsy are at times regularly repetitive in brief duration. Thus, this explanation should modify the expression as, for example, not regularly repetitive in long duration.

3. Focal epileptic spasms shown in Example 9
Epileptic spasms themselves are generalized in origin presumably caused by involvement of bilateral network. “Focal” indicates that one discrete focus in one hemisphere trigger the bilateral network, while “Generalized” indicates multiple foci in bilateral hemisphere trigger the bilateral network. Thus, the concept is most likely similar to that of difference between primary and secondary bilateral synchrony in the old day. The expression of focal epileptic spasms is misleading because it gives impression that clinical manifestation of epileptic spasms themselves appears lateralized or shows focal accentuation. In case of “Epileptic spasms,” we should better to use “Focal onset epileptic spasms” or “Epileptic spasms of focal onset,” not use briefer term “Focal Epileptic spasms.” In addition, example 9, it would be better to add a sentence that resection of left parietal dysplasia terminated the epileptic spasms. It is difficult to evaluate that responsible epileptic focus is truly unilateral hemispheric or only asymmetric in both hemispheres in case of infantile spasms.

Best wishes
Hirokazu Oguni

3 May 2016

Dear Sirs and Madams,

I have the following minor comments/requests regarding the seizure classification:

  • Since the classification makes an effort to be precise in regard to the description of ictal phenomena, it should also comment on the correct description of observed lateralization. When people mention lateralization to describe focal seizures it remains often unclear if they refer to the body or the hemisphere. Hence I think that the classification should comment on that and determine the correct standard.
  • Just a minor thing: If "generalized" is mentioned in combination with absence seizures it should also be put in brackets since it is somewhat redundant.

With best wishes and many thanks,
Rosa Michaelis

2 May 2016

Dear members of the ILAE Commission for Classification and Terminology,

First of all, thanks for this work that is a major achievement for the community. Here are the very minor points that might be considered:

1. The section on "Aura" gives a clear overview of the problem of using the term "Aura." The way to write this paragraph directly describes directly the risk to a misunderstanding related to the use of the term "aura." I would suggest to discuss if the paragraph should start by a statement that an aura is a seizure type.

2. As mentioned by others, it is difficult to classify Eyelid myoclonia under the absence seizures. Using the basic scheme, this seizure type would be most likely to be in the motor generalized. Based on the most usual duration and usual level of awareness without EEG recording (not done or not available), Eyelid myoclonia are more characterized by the motor component than an "absence" component.

Stéphane Auvin

26 April 2016

Dear Commission members,

First of all thanks for your continued effort on this complicated matter and for giving the opportunity to comment.

Mutatis mutandis: this proposal does not change too much the things. A few comments:

  • A main point is that it is not clear on which basis the seizures were classified (only on clinical ground or also on the basis of EEG features)? Some seizure types, such as spasms and tonic-clonic, should be still considered only “generalized” on clinical ground, as they involve widespread bilateral network(s). We also see the risk that putting the atonic and myoclonic seizures under the category of “Focal Seizures” could increase the risk of detrimental treatment with contraindicated antiseizures drugs, such as CBZ or PHT
  • The introduction of “aware/with impaired awareness” instead of “simple/complex” does not seem a real improvement. The same is for the introduction of “bilateral TCS.”
  • The clinical relevance of the dichotomy between motor and non motor is debatable
  • Eyelid myoclonia should not be placed among absence seizures as they significantly differ for duration, level of awareness, and EEG pattern. Moreover, the initial component are eyelid myoclonia, eventually followed by the absence.
  • The term “epileptic” before spasms seems redundant as this IS the classification of seizures (otherwise it should be specified also for (epileptic) myoclonic, tonic, and atonic seizures)
  • Please consider reflex epileptic seizures, which are neither cited. The same for headache of epileptic origin

Warm regards,
Pasquale and Salvatore Striano, Italy

Return to top

25 April 2016

Dear All,

I write in regard to the new clinically-focused seizure type classification created by the Seizure Type Classification Task Force. My comments are as follows:

1. The Task Force defines an operational seizure type as a useful grouping of seizure characteristics for purposes of communication in clinical care, teaching and research by stakeholders that include patients, families, medical professionals, researchers, epidemiologists, medical educators, clinical trialists, insurance payers, regulatory agencies, advocacy groups, and medical reporters.”

The Task Force Members are very well known and some of the most outstanding epileptologists and researchers and have done a wonderful job of the very difficult task allocated to them. An important aspect, however, has been overlooked since the Task Force has no representation from parts of the world where almost 80% of PWE live today. The conditions at the grass root level, social and cultural sensibilities of the “unrepresented part of the world” are totally different and deserve to be addressed in an appropriate manner. Any recommendation by such a task force is going to affect the clinical care; teaching and research involving various stakeholders in the unrepresented part of the world to a much greater extent since majority of PWE live there today!!

2. An example highlighting the different social and cultural sensibilities being the classification of focal seizures as “Focal Aware” and “Focal Unaware.” This is confusing and can be misleading to most people in the “unrepresented” part of the world. “Focal Aware” could be understood as “being aware that the seizure is focal” and “Focal Unaware” as “being unaware that seizure is focal.” These should be classified as “Focal with intact awareness” and “Focal with impaired awareness.” Descriptive terms need not be made too short at the risk of sacrificing their main purpose.

3. In an emerging economy like India (with a population of approximately 1.2 billion), it is estimated that there are about 1027.17 million cell phone users (as of February 2016). An estimated 9 million cell phone users are added every month1, 2. Very often family members bring a video recording of a seizure when they come for the evaluation in the hospital or a clinic. Even if they don’t get it the first time, they can be told to record a seizure and bring it on the next visit. Video recording of seizure at home on a cell phone needs to be encouraged and included in the list of supportive evidence that can help in any seizure type classification scheme.

4. The new classification mentions that “as part of the diagnostic process, a clinician will commonly use supportive evidence to help to classify a seizure, even though that evidence is not part of the classification scheme. Such evidence may include EEG patterns, lesions detected by neuroimaging, laboratory results such as detection of anti-neuronal antibodies, gene mutations, or presence of an epilepsy syndrome known to be associated with either focal or generalized seizures.” Most often, clinicians working in the developing world have no access to any technology when they first examine the patient. The seizures have to be classified on the basis of the subjective and objective description that is available. It is correctly mentioned that, where supportive information is available, it should be used to secure the most accurate classification. The classification of a seizure type may change drastically from the first evaluation based on the descriptive account alone as compared to the “final impression” based on the varying level of supportive evidence(s) available in different parts of the world (on most occasions no supportive evidence is available in developing countries). This aspect of the classification procedure depending upon the variable degree of availability of supportive evidence deserves discussion in the form of a separate paragraph.

5. In the ILAE Seizure Classification 2016 basic scheme, “Generalized” category has two sub-types – Motor and Absence (presumably representing non-motor type). Further, in the ILAE Seizure Classification 2016 expanded scheme, there are Absences with myoclonic jerks and eyelid myoclonias. Both myoclonic jerks and eyelid myoclonias have a motor component. This is confusing and contradictory.

6. The previous classification of seizures that are generalized from the onset into “generalized tonic-clonic seizure” and those having “partial onset seizures getting secondarily generalized” was perhaps based not only on the seizures being focal at onset versus those being “primarily generalized”. There was a hint to the possible etiology in that classification – most focal seizures being secondary to a focal abnormality in the brain while the “primarily generalized” having a possible genetic etiology. This hint toward the underlying etiological possibility appears to get lost in the revised proposal of “focal to bilateral tonic-clonic” that seems to reflect merely a propagation pattern of a seizure.

7. As per the proposed classification, the use of one of the oldest terminology in human epilepsy – the “aura” - has been discontinued mainly because “the aura itself represents a type of seizure, but it often is considered a prelude to a seizure rather than a seizure itself. This may not be obvious to patients, who may, for example, decide to discontinue medicines because they believe themselves to be seizure-free when they only experience auras”. It has been suggested that the use of term “aura” should be replaced with “focal aware”. In comparison to “aura”, the suggested term “focal aware” is very vague, non-descriptive term that has been used to classify many other types of focal seizures. Any term like “aura” that describes in a very captivating manner what the patients experience and is as historical as “epilepsy” itself, needs to be preserved by the “connoisseurs” as part of the “epilepsy heritage”. It was in 1981 that the term “focal” was replaced by “partial”, only to make a strong comeback now as being “more understandable” within a few decades. In case “aura” as a descriptive term is allowed to be condemned, the “romance” in descriptive epileptology will be lost forever!!


1. Telecommunications statistics in India: https://en.wikipedia.org/wiki/Telecommunications_statistics_in_India

2. http://telecomtalk.info/total-gsm-subscriber-base-feb2016-india/151580/

Satish Jain

Return to top

25 April 2016

Dear Colleagues,

Kudos to the committee for their outstanding effort on a challenging subject.  

Three comments:

  1. Consider including descriptions of awareness across all seizure types.  Brief epileptic myoclonus and brief absences (≤2 sec) may not impair consciousness; absence does impair consciousness; and tonic-clonic seizures turn consciousness completely off.
  2. Graphically, consider moving “Unclassified” up and to the right, to be at the same organizational level as focal, generalized, and unknown onset.  This will emphasize that there are really four high-level choices: focal, generalized, unknown onset, and unclassified.
  3. A description of a spell is often more helpful than a label.  “The patient stares into space for 10-15 seconds and doesn’t respond” – this could be absence, focal with impaired awareness, or non-epileptic. The ambiguity is important for clinicians because it guides management, and important for researchers because it prevents misclassification. Consider emphasizing the value of an extended description in the text of this document.

Zach Grinspan

21 April 2016

Thanks for opportunity of comment. We can classify as seizure

  1. According age of onset (newborn,infant,childhood, adolescents, adult (early onset <25, late  onset >25 to 60yrs); elderly.
  2. With encephalopathy or without encephalopathy.
  3. According etiology. (Hypoxic,ischaemic, traumatic, infective, inflammatory, autoimmune, metabolic, neurodegenerative,channellopathy,vascular.Drug toxicity or withdrawal
  4. Epilepsy with family history.
  5. Epilepsy with motor; non motor sign symptom.
  6. 6.Epilepsy with behavioral and cognitive impairment.
  7. Epilepsy with autonomic impairment.
  8. Epilepsy with amnestic non amnestic feature
  9. Reflex epilepsy.
  10. Epilepsy with normal EEG and Imaging.
  11. Electroclinical seizure without motor sign symptom

Surendra Khosya

Return to top

20 April 2016

We should thank the commission for their work in this difficult area, clarifying the seizure classification and maintaining clinical relevance, which will aid applicability and adoption in the wide range of circumstances across the world.

Use of simple terminology in focal seizures aware vs unaware is helpful, particularly in training those new to these concepts and are more intuitive compared to complex partial (the meaning of which gets students confused with complex febrile seizures – complex for an entirely different reason)

Inclusion of epileptic spasms is welcome.

Regarding the subclassification of focal seizures, would it be helpful to specify the main subgroups as focal motor onset or non-motor onset i.e., clarify the temporal relationship? This is to avoid value judgements on which aspect is the main component of the seizure, since there will be subjective variation in perception e.g. is it to be based on duration, intensity, extent etc…

Use of the term “to bilateral tonic clonic” is very specific and, in situations where observers do not identify both tonic and clonic components, may lead to (a.) inability to classify as such or (b.) over-designation of tonic clonic secondary generalized semiology. Perhaps retention of “secondary generalization” would continue to serve well in this respect.

Thank you for your kind consideration.
Derrick Chan

20 April 2016

This is an excellent proposal and a clearly written document that is going to be very useful when accepted. I only have a couple of comments:

  1. The replacement of the term “secondarily generalized” seizure with “focal to bilateral tonic-clonic” seizure. I think the arguments for (and possibly against) this change in terminology deserve a separate short paragraph. It is important to have terms that clearly distinguish these focal seizures from the generalized tonic-clonic seizures, but the new proposed term can give the impression that when the focal seizure has evolved to a bilateral tonic-clonic seizure it is different from a “generalized tonic-clonic” seizure. I am not sure this is the intention. The emphasis on the concept of bilateral rather than generalized involvement is discussed in the section “Focal vs. generalized” but I miss the argument for using bilateral for the focal but not for what used to be called primary generalized seizures. Would it be clearer if in the Schemes in Figure 1 the term bilateral was added to Generalized motor tonic-clonic seizures? It could be done in brackets in the subheading Motor (bilateral), or below: “bilateral tonic-clonic” etc. A detail in Summary: Item 7 in the list of changes should probable read “ 7. Focal to bilateral tonic-clonic seizures replaces secondarily generalized seizures” rather than “ 7 Bilateral tonic-clonic seizures replaces secondarily generalized seizures.”
  2. Seizures presenting with hallmarks of tonic-clonic seizures but where origin is obscured are suggested to be labeled “unknown onset tonic-clonic” (under methods for classifying a seizure) or “bilateral tonic-clonic, onset unknown” (under Examples). Should one decide on one of the two for consistency? I think the fact that two different terms are used for what seems to be the same situation might be an illustration of the issue I raised above.
  3. The Glosssary: The authors may consider to limit this to terms that are still proposed to be used. If you want to keep terms that are outdated according to the new classification, at least indicate that they are no longer recommended.
  4. My last comment is probably beyond the scope of this seizure type classification, but might be worth a short comment. ILAE has a conceptual definition of seizures, and conceptual and practical definitions of epilepsy. The current proposal provides a rather detailed operational classification of seizure types including how seizures should be classified when information is incomplete. What is missing, however, are operational criteria for when an event can be classified as a seizure. Can the minimum data/information requirements for considering a transient event to be a seizure be described? If, as I suspect, this cannot be solved within the framework of the current report, I suggest including a comment acknowledging this limitation.

Torbjörn Tomson

16 April 2016

Dear all,

The basal scheme this new approach is suggesting to follow is nicely simple and certainly easy to use.

In one particular field I feel oversimplification and loss, i.e introducing "focal nonmotor aware" to include all the seizure-states well-known as aurae. The new term in my view does not mirror the richness and complexity of auras. Many of them are multimodal (comparable to Jackson's dreamy states) combining sensory-autonomic-cognitive facets.

Then, how to fit in the classic "epigastric aura" into your scheme? Where to put the ecstatic sensation? In addition, many auras neither do correspond to aware nor unaware, they are somehow inbetween; thus can you really classify/designate any of them as "aware?"

Aurae are classic, they are known ever since, the term transfers much of the fascinating nature of epilepsy (an important part of it). Of course there's a need to use it correctly (e.g contrasting prodromi). Of course, many patients do not know about auras as themselves representing seizures (until you explain and teach them…) However: the term "aura" grasps that it happens in the subjective sphere of the patient's experience. "Focal non-motor aware" is a very technical, cold term - but we're not talking about mechanical clockworks.

Why do you sacrifice the "aura?"

BS Kasper

Return to top

16 April 2016

Thank you for the opportunity to comment.

Do we really need another overhaul of the operational classification scheme? Particularly without any major new contributions from underlying neurobiology. The current scheme whilst imperfect serves well clinical, research and teaching needs.

My initial thoughts on this paper are:

  1. Classification of focal seizures is still best served by some attempt at localisation based on symtomatology, i.e. frontal, temporal etc recognising this may not be where the seizure originates, but the predominant brain region involved in manifesting symptoms. It prompts further important thought and analysis on neuroanatomical localisation. 
  2. I prefer to keep the term secondarily generalised – we all know what this means and it prompts an important search for a focal lesion in clinical practice.
  3. I find the term 'loss of contact’ more meaningful than awareness to capture the impaired interaction with the environment in a “complex partial seizure’ (I still find simple and complex partial seizures useful concepts and easily explained to patients). The term awareness like consciousness is potentially fraught with semantic difficulties.

Best regards

Khalid Hamandi

11 April 2016

Dear ILAE,

I like the new seizure classification. I read it too quickly so apologies if my comments are redundant.

There is still an unresolved issue, with respect to “unresponsive stares” a.k.a. “dialeptic seizures”. I agree that absence seizures still require generalised spike and wave or polyspike wave, but we need a term for the unresponsive stare semiology, when we do not know yet if it is an absence seizure. I think “unresponsive stare” is the best term, and it could be added to the list of focal sxs. Sometimes a frontal lobe onset seizure becomes an absence seizure with generalised spike and wave during the unresponsive stare i.e. an absence. This is then a secondarily generalised absence seizure. However if the “secondarily generalised” terminology is to be abandoned… I suppose we can call them frontal lobe onset absence seizures.

If revising the seizure classification, I think presenting the electroclinical seizure classification / terminology with ictal EEG correlates, and a parallel semiology classification / terminology without ictal EEG criteria would be a great help, as many still get muddled, e.g. calling unresponsive stares “absences” even when there is no generalised spike wave / polyspike wave accompanying them.

It is good to include epileptic spasm in generalised and focal and unknown onset groups. Perhaps for the next revision of the epilepsy terminology, it would help to abandon “infantile spasms” as an epilepsy name, (use West syndrome), and use “infantile epileptic spasms” for the epileptic spasms of West syndrome. Epileptic spasms are also important seizure types in older children.

Best wishes
William Whitehouse

11 April 2016

This is a noble effort to make the classification of seizure types more accessible to more people, who are not necessarily familiar with former terms. There is one glaring omission, and that is subclinical seizure discharges, often prolonged and/or frequent, that might or might not impair awareness in adults. These adults have what is called "idiopathic generalized epilepsy," and they are a distinct group who usually have their first generalized tonic-clonic seizure around age 20 or later (Marini et al., J Neurol Neurosurg Psychiat 2003;74:192-196; Fujikawa et al., Epilepsy Behav 2012;25:442-448). They have EEG discharges of varying duration (2-3 Hz bilaterally synchronous spike and/or polyspike and slow-wave discharges) often without clinical signs. They are not simply a form of juvenile myoclonic epilepsy (JME), first, because they may or may not have myoclonic seizures (or absence-like seizures), and secondly, because their discharges are generally 2-3 Hz and not faster discharges, as occur in JME.

Denson G. Fujikawa

Return to top

8 April 2016

Bravo – this is major step forward in clarity and respects the previous classifications. It looks easy to use and the terms are intuitively obvious.

Peter Camfield

7 April 2016

Dear ILAE,

I welcome the further clarification and proposed amendments in the new seizure classification. The revised approach, whilst retaining the scientific and philosophical rationale of the 2010 document (albeit in a rapidly changing area), will I think be more readily accessible to non-specialists and thus more likely to be implemented over time but doctors and patients alike. As well as being fundamental to support research and improvements in care, this is important given that the majority of people with epilepsy worldwide will never have access to an epilepsy subspecialist, and even in developed countries access can be patchy. Furthermore – patients and carers are the individuals we communicate with most regularly about their seizures, so having a terminology that they can readily understand is clearly desirable. To improve the care of people with epilepsy we all need to speak the same language (in our different languages!) – to benefit the individual and the epilepsy society.

As is highlighted it is already compatible with ICD11, again of importance for many reasons worldwide. Whilst status epilepticus is a different condition, that there are some parallels in approach between this document and the ILAE Taskforce on Classification of Status Epilepticus (Trinka et al, Epilepsia 2015) in relation to motor, non-motor, and awareness/consciousness will also likely support more harmonized communication with non-specialists.

As someone who does a lot of undergraduate teaching, and having sometimes struggled to adequately convey changing classifications in this context, I can also see how much more readily this new proposal would be understandable to a relatively uninformed, but scientifically inclined audience and look forward to hopefully using it.

Congratulations to the taskforce.

Dr Hannah Cock

6 April 2016


Can I overall compliment the committee on this important work. I recognise the difficulty in getting agreement across a disparate community of epileptologists, and the importance of having a seizure classification system that works across different languages.
I whole heartedly agree that it is important for the basic classification system to be as accessible as possible, and to have additional layers of complexity that can be used as necessary.

However, I have a couple of concerns, mainly around the use of "focal" to describe seizure types perviously regarded as 'generalised.'

  1. Are asymmetric tonic seizures (both sides stiff, one more than the other) classified as focal or generalised tonic seizures. Conceptually a bilateral network is being driven (hence generalised), but with lateral emphasis (hence focal). This is a common scenario – and in fact much more common than "focal" tonic seizures. The problem recurs with spasms classification.
  2. This feeds into a broader problem of what is meant by "focal tonic seizures." If the classification committee mean that one limb is held in a fixed posture (how is this different to dystonia?) this potentially dilutes the significance of "generalised tonic seizures," which likely reflect a specific network mechanism, quite distinct to the one generating sustained focal posturing.
  3. If the commission are describing a tonic seizures that are clinically "generalised' (or asymmetric) but are associated with a focal brain lesion seen on imaging, this is incorporating information from the syndrome (i.e.: Imaging results) into seizure classification. The clinical features of the patients attacks will not change when they have an MRI, but the seizure classification would. Furthermore it potentially falsely implies that each seizure is being driven from the focal lesion, when in at least some cases it appears that the seizures are being driven from bilaterally distributed networks, which have become intrinsically unstable due to the chronic interaction with the lesion. Hence the seizures are actually "generalised," clinically and on EEG, even though lesion removal may be beneficial.
  4. I am similarly concerned that spasms can be classified as "focal" on the basis of extra, non-seizure information. Of course I do not disagree that spasms can be driven from an epileptic focus, but we are discussing a classification of seizures, not syndromes. The clinical features of spasms are very similar whether there is a focal or unknown (?network based) driver. In this case, part of the problem is that epileptic spasms are both a seizure type, and an epileptic syndrome. In this updated seizure classification, syndromic information appears to have crept in. What to do in kids with TSC bilateral flexor spams, and tubers – are their spasms (the clinical events) to be classified as focal (they have tubers) or generalised (they look generalised, and in any case we don"t know which if any of the tubers is responsible).


Dr John Archer, FRACP, PhD

Return to top

6 April 2016

Dear colleagues,

Thanks for all your efforts. I have a few comments.

The terms focal myoclonic and focal atonic might be confusing and difficult to ascertain. Many patients with myoclonic seizures including JME (a generalized epilepsy syndrome), have asymmetrical, at times one sides jerks and existence of this terminology may lead some people to wrongly classify patients that may lead to wrong management.

Focal semiologic and electroencephalographic features in patients with juvenile myoclonic epilepsy.
Usui N, Kotagal P, Matsumoto R, Kellinghaus C, Lüders HO.
Epilepsia. 2005 Oct;46(10):1668-76.

Also, epileptic spasms are often a result of a focal pathology, but may be a consequence of a generalized process (e.g., in genetic or metabolic causes). If it is included in focal, it should also be included in generalized.

Also, what happens to the classification of patients with focal epilepsy and manual automatism or dystonic posture? These are motor phenomena, but none of the focal motor classifications, which are included in the proposal, cover them, specially for nonhypermotor manual automatisms.

Within generalized classification, what is the advantage of having both tonic-clonic and clonic-tonic-clonic? We know that CTC seizures are more in favor of generalized syndromes compared with TC seizures, but within the generalized group having CTC vs TC does not offer any diagnostic or therapeutic advantages. Yes, it exists, but is it really necessary? More subgroups would make future studies more difficult.

Thanks again for all your efforts.

Ali A. Asadi-Pooya, M.D.,

6 April 2016

Dear friends,

The new seizure proposal unfortunately does not take into account my mother language at all. No wonder since nobody from the German language region was considered to be educated enough to participate,. The result is that nobody will be able to translate the term awareness properly and usably into German. Dyscognitive was understandable and appropriate, awareness as a distinction and as a term is not. Therefore this classification will simply not be used in the second biggest chapter of the ILAE.

Second comment: It is not acceptable to use the term “cognitive” seizure. This is semantic suicide. Nobody in the world will use this term for an epileptic seizure.

Best regards,
BJ Steinhoff
Mit freundlichen Grüßen

6 April 2016

Dear colleagues:

Thank you very much for your continuous hard work on further organizing this. Your classification seems or appears to make more emphasis in what is in the adults with epilepsy, despite the presence of several pediatric epileptologist on your team.

Two comments:

I would replace “UNKNOWN” Epileptic Spasms. All seizures are unknown on presentation despite they could suggest either focal or generalized ONSET. You suggest to use either – after imaging – either focal or generalized, as you state in your example. I urge you to change to FOCAL OR GENERALIZED OR EITHER or BOTH.

Second, NOT all seizures that evolve bilaterally are TONIC-CLONIC. Clinical experience as documented on video-EEG clearly has revealed focal seizures can evolve BILATERALLY – even to status – without any tonic activity, I would clarify or extend this.

Please see below

Luis E. Bello-Espinosa, MD

Classification of Seizures suggestion 6 April 2016

Return to top

6 April 2016

The very first item I wondered about when reading through the proposed classification was "awareness." There have been many attempts so far to optimize the term describing what happens during this kind of seizures. Whether told by the patient, a lay witness, or professional observer, several terms have been/are in use to characterize what happens, but each is emphasizing a different facet.

To me, impaired awareness, again is not optimal to catch the core issue. There are patients who are aware of themselves having had/having a seizure albeit conciousness, responsiveness, memory … has been significantly to severely impaired. Others again don't notice their own seizures at all, i.e are not aware of them. Thus, while appreciating that awareness can be reduced to abolished during certain seizures, there's this other meaning of awareness (which i think is very helpful)

BS Kasper, MD
Erlangen Epilepsy Center, Germany

6 April 2016

Congratulations, nice message about epilepsy. Further problem is when translation in countries not english speaker, create a variation in terms that otherwise, had been coined by using, is important each nation and so federation and so, work in the important isue of translation.

Thank you anyway,
Celio Martinez, Spain

6 April 2016

To whom it may concern:

Neurologists and pediatricians have been calling laughing seizures "gelastic" in order to emphasize that these involuntary bursts of laughter are typically not induced or accompanied by an appropriate affect like mirth, amusement or happiness. This detail of seizure semiology is crucial for parents and carers to understand and interpret the patient's seizures correctly. Thus, it seems misleading to re-classify them as "focal emotional" seizures. (In very rare cases of temporal epileptogenesis, gelastic seizures can indeed be accompanied by mirth, but these are exceptions.)

Accordingly, as a behavioural descriptor (Table 1) "gelastic" should be classified under "Automatism" rather than "Emotional or affective."

Yours faithfully,
a reader who would prefer to remain anonymous

Return to top


Home | Contact Us | Privacy & Security | Login | Sitemap
Creative Commons License
Text on this website is available under a
Creative Commons Attribution-ShareAlike 4.0 International License
except all videos and images, which remain copyrighted by the International League Against Epilepsy.