Updated Classification of Epileptic Seizures: Position paper of the International League Against Epilepsy

The ILAE Executive Committee has appointed a working group to evaluate the real-world application of the 2017 seizure classification and to recommend updates. 

Please review all files. Translations into 10 languages are available, and comments on these translations are also welcome. We encourage national chapters to propose translations into additional languages as well.

Draft: Updated Classification of Epileptic Seizures: Position paper of the International League Against Epilepsy

Translations: Arabic | Chinese | French | German | Italian | Japanese | Portuguese | Russian | Spanish | Ukrainian

Supplementary Material: PRISMA flow diagram for new systematic reviews | Literature search and evaluation | Definitions of generalized epileptic seizure types

Public comments are now closed and available for review below.

Comments

16 October 2024

Dear ILAE Editorial Board,

On behalf of the Department of Pediatric Neurology of the University of Antioquia, we would like to express our appreciation for the recent update of the epilepsy classification, which undoubtedly represents an important advance in the management and diagnosis of this complex disease. However, after a detailed analysis, and with the sole purpose of examining the changes it represents in relation to the previous classification and its usefulness for approaching patients with a diagnosis of epilepsy, we feel it is appropriate to make some observations that could enrich its implementation. In particular, the following points deserve attention:

• The definition of impairment of consciousness should not be considered as a dichotomy, but as a sign that is instaurated in a temporal variable and that can have a graduation in the temporal variable between mild, moderate and severe, taking into account its dynamic evolution during the temporal evolution of the seizure.
• We believe that the first motor or non-motor sign in the symptomatic cluster of focal epilepsies is practically relevant for the initial therapeutic decision. Although the identification of these signs is key to the diagnosis, their influence on the choice and implementation of treatment seems to be less determinant in clinical practice.
• It would be advisable to simplify the categories of somatotopic mediators of seizures to make them more accessible and practical in regions with less training in seizure analysis. This simplification would facilitate understanding and clinical management in areas with limited training resources in this area, allowing for more efficient care.
• The division between motor and non-motor events remains useful, since motor manifestations and their subtypes can have a high regionalising value in focal semiology and in the choice of appropriate medication in the subtypes of genetic generalised epilepsies (myoclonic motor, generalised tonic-clonic, atonic, negative myoclonic, tonic-clonic myoclonic, among others).

We appreciate the attention given to our observations and hope that our recommendations will contribute to a constructive dialogue on the new classification of epilepsies. We remain at your disposal for any further clarification or discussion on this issue.

Yours sincerely,

Department of Pediatric Neurology, University of Antioquia (Colombia)

16 October 2024

I genuinely appreciate the effort of the working group to update and simplify the classification of epileptic seizures. I agree with most of the proposed changes, in particular to re-adopt the term “consciousness” that is more widely used in medical language – not only in clinical neurology – respect to “awareness”, allowing a more homogeneous language particularly in the context of differential diagnosis (e.g. with syncope and other transitory “loss of consciousness” events).

However I suggest some corrections:

1. Since during a seizure with eyelid myoclonia (without corresponding absence) there is no alteration of consciousness/responsiveness/awareness, it is my opinion that this type of seizure should be included among motor seizures ("other than tonic-clonic") [maybe coded as EM; 2.2.1.8], while eyelid myoclonus with concomitant absence will constitute a subtype of absences (with associated motor manifestations) [linked to the code EMA; 2.1.4]
2. In the “supplementary-3-definitions-of-generalized-seizures2” file, in the description of the EMA it seems that it is mandatory to observe “the extension of the head” associated with eyelid jerks and upward eyeballs deviation, however this is not always the case, thus I suggest to be less assertive “[...] with simultaneous upward deviation of the eyeballs, sometimes associated to extension of the head [...]”
3. Regarding Myoclonic Absence Seizure (MA; 2.1.3), I suggest to add a statement as “Polygraphic recordings of EMG with EEG is recommended for ictal recordings” (similarly to what has been done for GMA; 2.2.1.7).
4. [LAST-BUT-NOT-LEAST] Crucially, I want to signal that the Italian translation of the new proposal does not take into account at all the reasons (extensively discussed in the paper and very much agreeable, in my opinion) that led the authors to use again the term "consciousness" (instead of “awareness”) and to replace "focal onset seizure" and "generalized onset seizure" with "focal seizure" and "generalized seizure". In fact, in the Italian translation of table 1, Figure 1, and Figure 2 is perpetuated the OLD translation, retaining “consapevolezza” (that is not used in any other neurological field in Italy, but has been adopted “believing it aligns with ILAE position paper, despite this interpretation being incorrect”), instead of “coscienza”. Idem for Figure 1, in which there are still the terms “Esordio Focale” and “Esordio Generalizzato” that the current proposal suggest as “misleading”; thus the new terms should be “Crisi focale” and “Crisi generalizzata”.

Gaetano Cantalupo (Italy)

15 October 2024

The current proposal is better than compared with the previous version. However it is still not clear for me how it will be classified when the patient was fully aware and responsive during the seizure but totally amnestic to the event. There are two scenarious in this condition. First the patient may continue the automatisms and/or inappropriate behaviour but responsive and then amnestic or may act totally normal during the event but not remember anything at all (like Dr. Z the Jackson's famous patient. Taylor D. JNNP1980, 43, 758-767).

Çiğdem Özkara (Turkey)

14 October 2024

I really appreciate the detailed update on epileptic seizure and epilepsy syndromes.

I think the changes are very useful, but I want to point out a few things.

1. I agree that in the seizure classification, focal seizures now include tonic, and myoclonic, and epileptic spasms which were previously classified as only generalized. On the other hand, absence seizures are only described as generalized and are not recognized as focal seizures. Since atypical absence seizures can be seen in frontal lobe epilepsy, I think the atypical absence seizures should be included in not only generalized seizures but also focal seizures.
2. I agree with the idea of grouping drug-resistant epilepsy syndromes occurring in newborns and infants under the name of EIDEE (Early Infantile Developmental and Epileptic Encephalopathy). This idea includes Ohtahara syndrome and EME. These two syndromes have a common finding of burst-suppression (BS) pattern on interictal EEG, and it can be difficult to determine which one they belong to. The current proposal states that EEG findings show either BS pattern or multifocal discharges in EIDEE. Shouldn't we distinguish and classify EIDEE with BS pattern from EIDEE without BS pattern? There may be cases where the SB pattern was not confirmed because sufficient EEG tests were not performed in early infancy. However, distinguishing cases where the SB pattern was clearly recognized is an important point when considering the relationship with the subsequent course, treatment, and gene mutations.
3. Thank you for summarizing the diagnostic criteria for Dravet syndrome in an easy-to-understand manner. I thought that the wide range of ages of onset and the concise diagnostic guidelines are valuable guidelines that lead to early specific treatment through earlier diagnosis. On the other hand, I was concerned that epileptic spasms (ES) were listed as exclusionary criteria. Although rare, ES can be seen during the course of Dravet syndrome. I propose that the diagnosis of exclusion be changed to when ES is present early in the course of epilepsy.

I would appreciate your kind consideration.

Katsumi Imai (Japan)

12 October 2024

I appreciate the effort of the working group to update and simplify the classification of epileptic seizures. But still, some issues require attention to bring more clarity to seizure classification. Awareness and responsiveness are used to characterize consciousness. It may be difficult for an eyewitness to describe/appreciate responsiveness. The preferred word may be arousability which is better understood by laypersons. Another issue is whether we should change the seizure classification of well-controlled patients who were categorized using the 2017 classification. In other words, how do you transport patients from one classification to next?

Ashok Kumar (India)

11 October 2024

Despite a great deal of effort on the part of this working group, it is my view that the proposed changes are a huge step backward in the classification of seizures. The document reminds me of when I was an epilepsy fellow, back in the 1980s, where the field was full of complicated jargon that was not easily digestible. Instead of making seizure classification easy and simple to understand this new proposal makes it overly complicated, especially for non-epilepsy specialist care providers, where they now have to recognize up to 20 seizure types. That is ridiculous.

Relying mostly on seizure manifestations harkens back to clinical care nearly 100 years ago. It wasn't workable then so why should it work now? We have learned much since then but it has been through the integration of semiology with other diagnostic information, such as imaging, genetics, etc.

It is important that any classification system has merit in either diagnosis and/or treatment, which the proposed system seems to have limited value.

Overall, I see this proposal as a major step backward, and I for one feel compelled to question the value of both using it and teaching it to students, residents, and trainees.

If others feel the same way it is important for voices to speak up as we are a community that should speak with one voice.

Gary Mathern (United States)

11 October 2024

Updated Classification of Epileptic Seizures is an excellent initiative made by ILAE , most of the shortcomings and lack of explanations of 2017 classifications are carefully deleted with more elaborate and descriptive classification. I am a practicing neurologist in Northern part of India for 25 years and my patients are from Haryana, adjacent districts of Rajasthan and Punjab, more than 25000 patients were treated and being treated in our center in last couple of decades therefore a classification at hand which should give me to address the change of semiology over time during a seizure, the individual awareness during a seizure, the reactivity to external stimuli in a clinical meaning, the existence of potentially dangerous behavior, the individual memorization of whether a seizure has occurred at all, and a quick and clear communication with subsequent action to prevent the seizure to happen further and protect the patient, therefore aura is very important part to understand and awareness about it is of utmost concern in seizure management lesions and guidance for the caregivers rightly described in this position paper. Sub-classifier (2): Replace the motor versus nonmotor sub-classification within focal seizures and within seizures unknown whether focal or generalized, with a sub-classification distinguishing between seizures with observable manifestations and those without describe seizure semiology in chronological sequence, depicting the sequence of seizure phenomena and Negative myoclonus: Include the recognition of negative myoclonus within the seizure classification is an excellent approach. Lastly, the symptomatic epilepsies both acute and chronic due to CNS infection.

In this part of the country is a major concern and comprising about 40% cause of epileptic seizure so important cause is different fever related seizure including FIRES in young children therefore symptomatic and localization related epilepsy might be given a place in the new classification.

Thanks for this hard work and new guidelines.

GP Burman (India)

10 October 2024

In this expanded seizures classification, it is better to classify first whether the seizures are organic, functional or seizure mimic.

Then under the heading of organic seizures, the expanded seizures classification can be followed.

M.A. Aleem (India)

4 October 2024

I congratulate the authors on this very well written article and on their care for embracing other languages and cultures.

Please, interpret my following suggestion as a respectful comment. I believe the text is excellent and most of the changes are appropriate. However, I strongly suggest the authors and/or ILAE commission reconsider the publication of this update. The fact that the 2017 classification manuscript ended with a statement suggesting revisions is not a good reason to do so. I believe the current working group could and should conclude that modifications are not necessary at this time. Actually, I believe changes, that in practical consequences are a new classification, would bring more harm than good. As the authors say, the primary objective is the establishment of a common language for clinical practice and research. It is clear that frequent reclassifications work against this main goal and not in its favor. Another harm of reclassifications is the time doctors, students, residents and researchers waste with the classification itself instead of really studying epilepsy. Classifications are inevitably overvalued, leading students to concentrate on the words instead of the actual phenomena. After this article gets published, I will have to say to all the residents that come with an interesting case to discuss: “now we don’t say unaware, we say impaired consciousness”. It may seem wispy, but we have limited time for patient care and teaching. I would have to spend a reasonable amount of time teaching vocabulary instead of using the time to deal with actual questions and ideas. I believe classifications should only be changed when there is a very strong reason, for example: new discoveries regarding mechanisms or causes (as the pathological classifications of dementia, methylation-based classification of tumors). I believe the critics and variations in conceptions, not strongly based in new facts, should not be enough to justify the cons of a new classification.

For example, the neuroscience and epilepsy community have been discussing consciousness for a very long time, including its appropriateness as a seizure descriptor. I believe changes in the classification should not reflect what the current workgroup thinks, even if it is right, to avoid constant changes based on changes in ILAE’s commissions and the opinions of the contemporary active members. I believe ILAE should change the classification only if relevant real changes in the field occur, usually when new mechanisms are discovered. So, I agree with the authors on the consciousness issue, but the harms of a new classification (which it is for its practical consequences) outweigh the gains.

I see there are some comments against frequent changes in classification, but I believe this real-world opinion is still underrepresented here, besides being the most important one. Most of the neurologists and epileptologists I talk to think there are too many reclassifications. However, for some reason, they do not comment here.

If the reclassification remains, I have minor suggestions:
As “motor vs nonmotor” has its problems, also do “with or without observable manifestations”. If a witness detected a seizure, there must be an observable manifestation. I understood that a seizure without observable manifestations would only include purely sensitive and/or some autonomic seizures and unperceived seizures with post ictal manifestations. It would define if other people perceive the seizures, or the only one who perceives them is the patient. This new categorization could help to define social consequences of seizures. However, I believe the new classification as “with or without observable manifestations” will be confusing to implement. Moreover, it is not justified in the text. The authors present good reasons against the “motor vs non-motor”, but they don’t justify why they introduced a new and seemingly hard to implement classification (“observable”). In addition, whether manifestations are observed or not depends on the observer. A trained epileptologist will probably observe more signs than the regular seizure witness. Patient companions would probably vary a lot in terms of what manifestations they are able to catch. It makes even harder to implement a classification that implies that a manifestation is “observable”. In the second example case, even the authors seem to use “observable” inconsistently, since “auditory aura” is encompassed in “observable manifestations”. In sum, I believe “observable or not” adds an unnecessary label of complexity and should be removed. The authors may simply remove “motor vs non-motor” without an obligation to add something instead. It would be easier for healthcare providers to describe focal preserved consciousness seizures with sensory phenomena only, when that is the case, than to think they should always classify if the manifestations are “observable or not”.

“Aura” is one of those words that seem good to those who are used to using it. However, it is a very bad word for a seizure classification, because it is one of those words that constitute an unnecessary exclusive language for the specific community, it has spiritual connotations, and it is more appropriately used in migraine. Sensitive manifestations or phenomena is already used to refer to the old “aura”. I suggest changing “7. Indescribable aura” for “indescribable sensation”. In the examples, I would also recommend changing auditory and visual aura for another word (might be auditory/visual phenomena, hallucinations, or manifestations). For internal consistency, I believe phenomena should be used, because it was used in table 2.

I believe the information that the seizures of the 14-month-old girls were (focal) spasms is equally important and influential on treatment as that of the 3-month-old boy. It seems unjustified that focal spasms are a descriptor and generalized spasms are a classifier. I do not know how I would deal with that in the classification, since focal seizures are so many.

I have read comments asking “simple and complex partial seizures” back. These are very bad terms, they are inaccurate, and require specific translation (change in the actual meaning) for people to understand. I am certain people only like them because they were used to them. It would be a step backward to go back to this kind of terms.

Ricardo Lutzky Saute

3 October 2024

I would like to express my gratitude for the outstanding updated proposal for the classification of epileptic seizures in 2024. This proposal reflects the relentless efforts of a dedicated team and is underpinned by a robust research design. Attached to this letter, you will find our team's comments and queries regarding your work, as well as three proposed projects related to this classification. We are eager to explore opportunities for close cooperation with the International League Against Epilepsy (ILAE).

Comments:

1. Consciousness vs. Awareness:
There are numerous debates surrounding the concepts of "consciousness" and "awareness" that extend beyond mere linguistic variations. Consciousness encompasses a vast domain that remains largely unexplored and intangible. A simplistic aspect of this is "memory," which was not addressed in this revision. Evaluating consciousness, particularly in younger children and infants, is more complex due to the significant influence of developmental processes on consciousness compared to awareness.

2. Observable vs. Non-observable Dichotomy:
While this distinction is recognized, it may not be as clinically useful or practical as the differentiation between motor and non-motor functions. Several considerations regarding this rephrasing include:

• Terminology: Alternatives such as "observable-unobservable distinction" or "observable-latent variable dichotomy" may better convey the conceptual divide.
• Conceptual vs. Empirical: It is crucial to differentiate between conceptual distinctions and their empirical applicability; while theoretically valid, the observable-nonobservable dichotomy may have limited practical utility in clinical settings compared to other frameworks.
• Continuum vs. Dichotomy: Some scholars argue that observable and non-observable phenomena exist on a continuum rather than as strict categories, suggesting that certain variables may possess both observable and non-observable components.
• Context-Dependent: The relevance of this dichotomy may vary based on specific contexts, research questions, and theoretical frameworks.

In summary, while the observable-nonobservable dichotomy serves as a conceptual distinction in literature, its practical utility and terminology may differ based on context and research focus.

3. Epileptic Negative Myoclonus:
This electroclinical entity can only be effectively studied using video-EEG monitoring with an EMG extra-electrode, which is often unavailable in resource-limited countries.

Queries:

1. Regarding the working group, how was selection bias addressed? Were participants matched in terms of experience, knowledge, training, etc.?
2. Given that one of the main objectives of the 2017 version was to ensure practicality and feasibility in resource-limited countries—where English is not always the native language—was a linguist expert included in your working groups?
3. In the review process, were papers distributed randomly between the two raters (SB and ET)?
4. Could “non-motor” be replaced with "non-convulsive"? This terminology seems more applicable.

Proposals:

1. We propose developing structured interviews and an educational module for patients and caregivers to enhance information gathering. Designing this as a smartphone application could be advantageous, and we believe collaborating with ILAE would be beneficial.
2. We suggest an AI-assisted program aimed at developing an application that will assist physicians in classifying seizures more effectively according to the updated ILAE guidelines.
3. Additionally, we propose integrating this updated epilepsy classification by ILAE with the latest version of ICD.

Thank you for considering our feedback and proposals. We look forward to your response and hope for a fruitful collaboration.

Sincerely,

Mahmoud Mohammadi, Reza Shervin Badv, Zahra Rezaei (Iran)

2 October 2024

Epileptic seizure are hard to understand until now, specially the patients and the community. Epilepsy awareness are known only for it's word, and basic information, but in reality, specification of epilepsy are far to understand. The doctors in public hospitals did not define to the patient the classification of epilepsy he/she has. Just basic check up only perform, but communication between the doctor and the patient, by asking or revealing is not practiced to define epilepsy and do's and dont's of AED the patient would take. Encouraging and helping PWE to spread awareness are only pinned during Epilepsy Awareness Week, but nothing else. If I am right or not, sorry for my comments but just to express my side, our side, as a person with epilepsy.

Thank you.

Jun Enriquez

28 September 2024

I wish to support the comments by Marian Galovic and Pasquale Striano, the ILAE should carefully select the information to be presented in Figures 1 and 2 and tailor them to the intended audience.

I strongly agree with the subclassification of generalized seizures, and also with delegating them to the expanded version. However, having MYOCLONIC seizures in Figure 1 could be helpful, as most first generalized seizures are seen by generalists, rather than epileptologists.

Forgetting to ask or for myoclonic seizures remains an important factor in the misclassification of JME [1] and as "myoclonic or absence …" they carry independent prognostic information [2]. Further, myoclonic seizures should be accounted for, when valproic acid is to be avoided [3].

[1] Syvertsen M, Hellum MK, Hansen G, Edland A, Nakken KO, Selmer KK, Koht J (2017). Prevalence of juvenile myoclonic epilepsy in people <30 years of age-A population-based study in Norway. Epilepsia, 58(1), 105-112.
[2] Bonnett LJ, Tudur Smith C, Smith D, Williamson PR, Chadwick D, Marson AG (2014). Time to 12-month remission and treatment failure for generalised and unclassified epilepsy. J Neurol Neurosurg Psychiatry, 85(6), 603-610.
[3] Cerulli Irelli E, Cocchi E, Morano A, Gesche J, Caraballo RH, Lattanzi S, Strigaro G, Catania C, Ferlazzo E, Pascarella A, Casciato S, Quarato P, Pizzanelli C, Pulitano P, Giuliano L, Viola V, Mostacci B, Fortunato F, Marini C, Di Gennaro G, Gambardella A, Labate A, Operto FF, Giallonardo AT, Baykan B, Beier CP, Di Bonaventura C; Women With Epilepsy Treatment Options and Research (WETOR) Study Group. Levetiracetam vs Lamotrigine as first-line antiseizure medication in female patients with idiopathic generalized epilepsy. JAMA Neurol, 80(11), 1174-1181.

Bert Kleine (Germany)

24 September 2024

I would acknowledge the Commission for the huge effort. The consensus gave an acceptable frame for the nosological rules behind the Classification. I had a few comments:

• The removal of the terms non motor is not completely useful under the clinical viewpoint. It is true that some seizure types that are usually inserted among the non-motor seizures have associated motor manifestations (descriptors) but the identification of the predominant clinical manifestations as “non motor” may have relevant prognostic implications. Perhaps, the prefix “predominant” might help (e,g,“predominantly non motor generalized seizure”).
• The text does not sufficiently explain the concept of “observable” and “not observable signs” and only a table lists them. Non “observable signs” should be better characterized in the text. Perhaps the term “self-reported” should be more adapt. The class of “possibly observable” signs creates confusion. I think that they may be inserted among the observable ones. Further examples in the text, including a clinical vignette, might be useful to realize these aims.

Mario Mastrangelo (Italy)

20 September 2024

Bert Kleine (Germany)

18 September 2024

Rodrigo Riquelme (Chile)

18 September 2024

18 September 2024

16 September 2024

13 September 2024

Andreas Schulze-Bonhage (Germany)

11 September 2024

7 September 2024

Markus Leitinger (Austria)

6 September 2024

Camila Hobi (Brazil)

6 September 2024

Congratulations to the work team. It is clear and updated without ambiguity.

Sandeep Patil (India)

5 September 2024

The revised classification is comprehensive and easier to translate to the local language. I have a challenge with negative myoclonus, the identification of. I work in a resource constraint environment, where EEG assessments are not routinely available and so will tend to use clinical evaluation.

Well done to the committee.

4 September 2024

Attempting to classify seizures better is a valiant effort and is to be applauded. Kudos to the team for putting in the effort.

However, I would advocate strongly for keeping the same names as before. Changing names every few years leads to an incredibly fragmented literature and difficulty in communicating with non-epilepsy specialists who do not understand why we keep changing things (neither do I). We have community physicians who speak of petit mal, complex partial seizures, and university-based physicians who speak of dyscognitive seizures, and focal impaired awareness seizures depending on when each person trained. This terminology heterogeneity affects research in trying to put together different terminologies over the years. This also detracts new trainees who find the field very confusing and move on to other areas.

FAS -> FPC and FIAS -> FIC may seem like progress, but is a backwards step in my opinion. I am sure the team had similar discussions between this and reached this consensus, similar to all the other times in the past when the names were changed, and similar to the future times when this will change. If the price to pay for progress is added confusion, perhaps progress is not worth the price. Hope this highly qualified team does not take this as unwarranted criticism, and adds this opinion to the mix.

Zulfi Haneef

2 September 2024

Firstly, congratulations in achieving a feat and making such a detailed classification with newer data. I have two comments.

1. In the case of only autonomic semiology, where do they need to be placed?
2. Negative myoclonic seizure need to be more clarified in terms of etiology for its varied presentations.

Thank you,

Gautam Gangopadhyay (India)

29 August 2024

I congratulate the working group on this updated classification. This classification will appear on thousands of PowerPoint slides. May I humbly suggest consulting a graphic designer for the final print version of the manuscript to enhance the style and legibility of Figures 1 and 2.

Marian Galovic (Switzerland)

28 August 2024

In the CDKL5 Deficiency Disorder (CDD), there is a very characteristic seizure type characterized by the hypermotor-tonic-spasms sequence.

It was first described in 2011, see Klein KM Yendle SC Harvey AS Antony JH Wallace G Bienvenu T Scheffer IE. A distinctive seizure type in patients with CDKL5 mutations: Hypermotor-tonic-spasms sequence. Neurology. 2011

More than half of CDD patients present either hypermotor-tonic-spasm seizures or tonic-spasm seizures (ref: Demarest ST, Olson HE, et al 2019. CDKL5 deficiency disorder: Relationship between genotype, epilepsy, cortical visual impairment, and development. Epilepsia)

These characteristic seizures in CDD patients pose challenges for seizure counting and classification in trials.

This is a request to please provide guidance in the ILAE seizure classification for how to classify these seizures: e.g. a subtype of tonic seizures, a subtype of epileptic spams, a separate type of seizure.

One good example for how to classify some sequence seizures is the classification of “2.2.2.3. Absence-to-tonic-clonic seizure” as a subtype of GTC seizures in Table 1. A possibility would be to similarly include the tonic-to-spasm seizure so often seen in CDD patients under "2.2.1. Generalized motor seizures – other than tonic-clonic”.

But wherever you place it, my request is to please consider the inclusion of the tonic-to-spasm seizure because it is so common in CDD patients and it will make our clinical trials easier if it would have a direct map into the updated ILAE seizure classification.

Thank you so much.

Ana Mingorance (UK)

28 August 2024

Thank you to the working group for the proposed changes, most of which I believe represent improvements. In addition, the authors have to be commended for providing simultaneous translations, which I think is of major importance for promoting wide implementation.

I have two comments:

1. I suggest to replace "Focal Preserved Consciousness Seizure" and "Focal Impaired Consciousness Seizure" with "Focal Conscious Seizure" and "Focal Unconscious Seizure". Although I acknowledge that the latter may not be entirely correct in all cases, I would favour simplicity.
2. The term "Generalized Negative Myoclonic Seizure" deserves some further explanation/description.

Chantal Depondt

26 August 2024

Excellent update. I appreciate the addition of myoclonic-tonic-clonic and absence-tonic-clonic along the lines of focal-tonic-clonic. There should be a category of clonic-tonic-clonic, albeit rare, but should be a part of the new expansion.

Charles Akos Szabo (US)

26 August 2024

I suggest to create a new topic with genetic epileptic syndromes like Dravet, and Alice in Wonderland. I also suggest to differentiate epilepsy crises due to their origin in the brain with their more common phenotypes as like as frontal lobe epilepsies and parietal epilepsies.

Antonio Carlos Borges (Brazil)

23 August 2024

As Chair of the 2017 Seizure Classification taskforce, I am gratified that the Updated Classification revisions are limited and based on literature, seven years of experience in the field and expert consensus. The structure and majority of terminology from the 2017 classification is retained. The revision supports the decision to classify by the first manifestation, rather than the most prominent, because the first identifies the part of brain or network pointing to the most likely site of an important lesion. The revision also continues to allow use — when available — of ancillary information and not exclusively on semiology. 

One of the changes is introduction of the term “consciousness” in place of “awareness.” Alteration of consciousness always has been a key classifier of seizures, but the term has not before been in the name of the seizure type (for example, “temporal lobe seizure,” “psychomotor seizure,” “complex partial seizure”). The 2017 Task Force thought that a typical seizure with automatisms or freezing of activity would not be perceived by the public as loss or alteration of consciousness. In the popular view, loss of consciousness conveys someone lying on the floor “out cold.” We looked for surrogates. Possibilities were awareness, responsiveness, memory, any or all of which can occur during a seizure. Responsiveness is of great operational importance because its loss is what crashes the car. However, responsiveness usually cannot be assayed retrospectively, unless someone was present to test it at the time of a seizure. Therefore, we chose awareness during the event as a surrogate. We also made note that many languages did not make a clear distinction between “awareness” and “consciousness.”

A seizure with no observable manifestations probably is sensory, emotional or autonomic. We thought it was worth specifying which of these was present at the start of the seizure, but apparently the updated version does not, and lumps them all under “without observable manifestations.”

Focal and unknown categories now mention sequential semiological classifications. Of course, these should be in the documentation of seizures, but they are not seizure CLASSIFICATIONS. It would be impossible to list all sequential symptoms and signs and all propagation patterns during seizures. The 2017 Task Force chose to list only the most important one, which is focal to bilateral tonic-clonic. That seizure type is maintained.

A new seizure type of negative myoclonic seizures has been added, and I have no quarrel with that. Over the past years, I have received several e-mails nominating other seizure types, including tonic-atonic, atonic-tonic, absence to tonic-clonic. At some point, a classification becomes unwieldy, so inclusionary decisions had to be made.

I have a few of my own criticisms of the 2017 seizure classification. The main one is that it is not based on fundamental science and does not explain why the brain generates a discrete number of different seizure types. Our state of knowledge did not permit a scientific classification. It does not encompass the large variety of propagation patterns. The categorization as “unknown” is a feature of our ignorance, not of the seizure. Some seizures could be classified more than one way in the 2017 scheme, for example sensory versus autonomic or hallucinations as cognitive versus sensory. Many “favorite” seizure types were not included. For the most part, the revised classification does not solve these problems, but may bring terminology into simpler or more common usage. Hopefully, the energy to re-label and re-educate will be justified by greater accuracy and understanding.

Robert S. Fisher (US)

22 August 2024

In the section on "Cognitive & language phenomena," I recommend adding 2 categories: 1) Confusion, 2) Other focal cognitive deficits (eg, anosognosia, apraxia, neglect,). #1 is common and can occur in the absence of aphasia or amnesia. Further, patients can have amnesia without confusion. #2 is rare but these deficits have been reported in the literature. Since focal seizures may inhibit normal neural function, a variety of negative cognitive deficits that are topographically related to the site of cortical ictal activity. The present Somatotopic Modifiers do not account for the variety of potential ictal cognitive deficits.

It would have been nice to have a vowel in FPC to allow easy reference to focal preserved consciousness seizure and focal impaired consciousness seizures (FIC) without having to say the whole phrase when discussing with colleagues and trainees. Consider focal aware consciousness (FAC); note that this may help patients and their family understand that they can have a focal seizure where they appear conscious with eyes open and may even interact partly with the environment, but are not aware for FIC vs FAC seizures.

Thank you for your consideration.

Kimford Meador (US)

22 August 2024

Thank you for your appreciation, and it's a pleasure to assist with the revisions. I am an epileptologist from the southwest of Iran, a region with a significant prevalence of epilepsy and epilepsy syndromes among children. My experience has been deeply shaped by the numerous debates and discussions with mentors, colleagues, epilepsy fellows, adult neurology residents, and even interns and students regarding consciousness and awareness in affected children. Often, these discussions would become repetitive, prompting me to simplify the concept by suggesting they observe the child's response to stimuli. It is gratifying to see others echo my thoughts on this matter.

Moreover, I am particularly thankful for the recent definitive classification of absence seizures. Previously categorized merely as generalized non-motor seizures, this redefinition acknowledges their distinct characteristics, including symptoms like eye-blinking, which were often overlooked or underestimated. This clarification is a significant step forward and aligns with the clinical realities I face, where absence seizures present unique diagnostic and management challenges.

This updated approach not only reflects a better understanding of the condition but also promises to alleviate many of the ongoing debates among neurologists and epileptologists. By establishing clearer definitions, we can improve both daily practice and research, leading to more targeted interventions and better outcomes for our young patients. The recognition of specific seizure phenomena, like eye-blinking during absence seizures, enhances our diagnostic accuracy and enriches our understanding, ultimately benefiting the comprehensive care we strive to provide. Such advancements in our field are crucial. They not only refine our medical practices but also foster a more nuanced understanding among all healthcare providers involved in the care of children with epilepsy. This collective progress is essential for advancing our field and improving the lives of those affected by these challenging conditions. As we continue to build on these foundations, I remain hopeful for the future of epilepsy treatment and research, particularly in regions like ours where the need is profound and the impact of such advancements can be truly life-changing.

Reza Azizimalamiri (Iran)

22 August 2024

Looks very clear and sensible; I think an improvement.

Hannah Cock (UK)

21 August 2024

Dear Prof Cross, dear Prof Beniczky, dear Prof Trinka, dear colleagues of the ILAE TASK FORCE on seizure classification,

I have read your precious proposal with great interest and wish to congratulate you to this outstanding achievement.

I would like to add a few thoughts which could give way for important options.

I am aware that the ideal approach to a classification is to combine a simplicity for a broad applicability on the one hand, and a granularity which captures the complexity of real life on the other hand.

In daily practice I often wished to have a classification at hand which allowed me to address the change of semiology over time during a seizure, the individual awareness during a seizure, the reactivity to external stimuli in a clinical meaning, the existence of potentially dangerous behaviour, the individual memorization of whether a seizure has occurred at all, and a quick and clear communication and documentation tool.

This would have been helpful as it might have implications on driving a car, taking care of small children, might interfere with workplace activities, might impact on seizure counting (predisposition to underreporting), impacts on design of interventional studies, epidemiology in general and relative frequency in an individual, and neurological topography to associate specific semiological patterns to certain brain areas as a prerequisite for epilepsy surgery.

The complexity I would like to address is

AWARENESS A:

• A1->1 (A11) means that the patient is aware at the beginning (A1) and at the end (-1) of a seizure.
• A1->0 (A10) means that the patient is aware at the beginning (A1) but not at the end (-0) of a seizure.
• A0->0 (A00) means that the patient is not aware at the beginning (A0) nor at the end (-0) of a seizure.

RESPONSIVENESS R:

• R1->1 (R11) means that the patient is responsive at the beginning (R1) and at the end (-1) of a seizure.
• R1->0 (R10) means that the patient is responsive at the beginning (R1) but not at the end (-0) of a seizure.
• R0->0 (R00) means that the patient is not responsive at the beginning (R0) nor at the end (-0) of a seizure.
• Rx->x (Rxx) the patient does not know whether she/he is responsive, there are no witnesses so far.

BEHAVIOR as part of seizure semiology, BUT NOT DANGEROUS B:

• B1->1 (B11) means that the patient shows no dangerous behavior (e.g. putting boots into the oven) at the beginning (R1) and at the end (-1) of a seizure.
• B1->0 (B10) means that the patient shows no dangerous behavior (e.g. putting boots into the oven) at the beginning (R1) but does so at the end (-0) of a seizure.
• B0->0 (B00) means that the patient shows dangerous behavior at the beginning (B0) and at the end (-0) of a seizure (i.e., throughout the seizure).

MEMORIZATION whether a seizure has occurred at all M:

• M1 the patient remembers/”memorizes” that a seizure has occurred.
• M1d the patient directly remembers/”memorizes” that a seizure has occurred, i.e., the patient recalls parts of semiology.
• M1i the patient INdirectly remembers/”memorizes” that a seizure has occurred, i.e., the patient concludes from aspects other than semiology that a seizure must have occurred, e.g. blood and saliva on the pillow.
• M1x the patient remembers the seizure as such but it is unknown by which mechanism.

Examples:

A11 R11 B11 M1d:

This is a seizure in which the patient is fully aware from the beginning throughout the whole seizure, (A1->1, A11), always responsive, (R1->1, R11), always with adequate behaviour, (B1->1, B11), memorizes directly that the she/he had a seizure and can report this (M1d).

A10 R11 B11 M1:

This is a seizure in which the patient is fully aware only at the beginning of the seizure, (A1->0, A10), always responsive, (R1->1, R11), always with adequate behaviour (B1->1, B11), memorizes that the she/he had a seizure and can report this (M1x). The physician does not know how the patient remebers that there was a seizure.

A10 R10 Bxx M0

This is a seizure in which the patient is fully aware only at the beginning of the seizure (A1->0, A10), responsive only at the beginning of the seizure (R1->0, R10), the behaviour is not reported/ no witnesses available (Bxx), does not memorize that the she/he had a seizure.

(M0) A10 R10 Bx0 M0

This is a seizure in which the patient is fully aware only at the beginning of the seizure (A1->0, A10), responsive only at the beginning of the seizure (R1->0, R10), the behaviour is not reported at the beginning, but later shows inadequate behaviour (e.g. puts boots into the oven) (Bx0), and does not memorize that the she/he had a seizure (M0).

Clinical examples:

1. For a young mother it would be quite a success to improve from R11 Bx0 to R11 B11 as her child is no longer endangered by inadequate dangerous behaviour.
2. A pharmaceutical company may wish to include only patients with Aany Rany Bany M1 to guarantee adequate seizure reporting and counting.
3. There should be no problems at a suitable workplace for a patient who has seizures classified as A11 R11 B11 M1d.
4. In presurgical evaluation, A10 R10 B11 M1 has a different meaning than A00 R00 B10 M0.

Possibilities for extension:

In presurgical evaluation it might be highly relevant if the patient speaks during the seizure, therefore:

• S11 (S11) means that the patient speaks adequately throughout the seizure, an involvement of the speech relevant eloquent cortex does not occur.
• S1->0 (S10) means that the patient speaks adequately at the beginning, but there is impairment of speech at the end of the seizure.
• SX->X (SXX) Speech was not witnessed or tested during the seizure.
• SX->X/1 (SXX/1) Speech was not witnessed or tested during the seizure. However, speech was normal in the postictal testing.

I do not yet know what “unknown” should exactly mean regarding consciousness:

• We do not know whether the patient is aware during the seizure.
• We do not know whether the patient is responsive during the seizure.
• We do not know whether the patient memorizes her/his seizure.
• Any combination of the above.

Thank you so much for reading until this point. – Please allow uploads of PDFs to provide figures.

Kind regards. 

Markus Leitinger (Austria)

17 August 2024

If we can add reflex tonic seizure to tonic seizure classification and also reflex myoclonic seizure.

Sawsan Albazi (Iraq)