Contact Us

Ilae Facebook Ilae Twitter

Roadmap for Classification of the Epilepsies

ILAE logo

Classification of the epilepsies: New concepts for discussion and debate —
Special report of the ILAE Classification Task Force of the Commission for Classification and Terminology

The ILAE Commission on Classification and Terminology has focused on updating the Classification of the Epilepsies over many years, publishing a major new Organization in 2010 in Epilepsia (1). The intention behind this work is to integrate concepts that reflect current insights into the epilepsies given many landmark discoveries in the last two decades. There has been considerable feedback about this document with further invitations for commentary, discussion documents and revision following the publication of an online document in 2013 (2).

Here, the Task Force on Classification publishes a roadmap to establish an updated Classification of the Epilepsies, an important paper for the epilepsy community to herald the launch of Epilepsia Open. The aim is to achieve a classification that is used in clinical practice for diagnosis of patients with seizure disorders and that reflects current thinking. This document focuses primarily on a framework for classification to allow different levels of classification according to the health care settings and resources available to clinicians. The classification of the epilepsies emphasizes the importance of considering etiology at all levels. We also present some suggestions regarding terminology and invite your feedback on these concepts and terms.

The concepts will be integrated with the recently suggested seizure classification (3). We will consider your valuable comments and insights in the development of a classification to be implemented in epilepsy practice. We look forward to your thoughts.

Document | Supplement
Please send your comments to who will post them here for the entire community to consider.

The deadline for submitting has been extended to 11 September 2016. Comments will be posted below 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
  3. http://www.ilae.org/Visitors/Centre/Class-Seizure.cfm


5 October 2016

Thanks you for giving opportunity to me.

Epilepsy with static encephalopathy
Epilepsy with progressive encephalopathy or with episodic encephalopathy likely mitochondrial disease,urea cycle, storage disease, other enzymes deficiencies diseases, Other Neurometabolic diseases
Genetic epilepsy (hereditary or non hereditary)
Autoimmune disease and epilepsy (autoimmune encephalitis, SREAT, Rasmussen encephalitis, Neuropsychiatry lupus)
Epilepsy due to vascular disease (arterial stroke/venous thrombosis, PRES,RCVS , CNS vasculitis and A-V malformation)
Epilepsy due to parenchymal structural lesions or malformation
Epilepsy as a spectrum of other neurodegenerative diseases like PME, NBIA, complicated HSP, Alzheimer's disease and CJD
Epilepsy with infective disease or post infective sequale
Reflex epilepsy

Surendra Khosya

15 September 2016

Thank you for the opportunity to comment on the recent publication in Epilepsia Open entitled “Classification of the Epilepsies: New Concepts for Discussion and Debate”.

We would like a clarification in the paragraph with regards to “symptomatic generalized epilepsy often maps to epileptic encephalopathy”, page 6.

Specifically, reference is made to the difference between “secondary generalized epilepsy” and “secondarily generalized epilepsy”.  Although the definition of the latter is stated in this same paragraph, the former is not.  The former should be defined, as we believe this term is not used as widely as the latter, and many readers may not know what this term refers to.  For example, how does secondary generalized epilepsy differ from primary generalized epilepsy.

Thank you.

Yours sincerely,
Eleftherios S. Papathanasio

12 September 2016

Dear Colleagues and Friends,

First of all let me congratulate for the great work and the huge effort, which the group has put into this. The Paper has been greatly improved and the categorization of etiologies is clinically useful and important. The framework of seizures as symptoms of a disease is well acknowledged and appreciated.

My personal concern, which I share with many colleagues around the world, but especially in the German speaking community, is the eradication of the term "idiopathic" which was and still is very useful since centuries.

To replace "idiopathic generalized" with "genetic generalized" is misleading, conceptually wrong, and not to the good of patients:

  1. There is a genetic predisposition and often a positive family history, but the genes involved are definitely not a single one. The same holds true for hypertension, atherosclerosis, stroke etc. No serious clinician would call these conditions or diseases "genetic hypertension" or "genetic stroke" unless there is a clearly identifiable gene responsible. A better name then is "familial" which can be easily translated into all languages. The term idiopathic reflects this polygenetic background much better than "genetic", which always implies, that one knows the defect and the defect is mainly responsible for the disease (causation), which is not the case with IGEs
  2. The concept does not change and the group of diseases which just change the names from IGE to GGE, does not bring any advantage in research, teaching and clinical treatment. The explanation that we call it genetic, instead of idiopathic, despite we do not know the genes involved, will not be understood by our colleagues in Neurology and not even in undergraduate students.
  3. Several million people died in central Europe under the Nazi regime. A large part of them under the eugenic laws. In my Hospital several children and adults with IGE were killed under the law to keep the population "genetically clean". Whenever we use the term genetic in the very broad sense as in GGE, these things and the terrible history should be kept in mind. We have also a political historical responsibility in this case.
  4. To explain a patient the diagnosis of GGE will be extremely time consuming and often misinterpreted as truly "genetic". In many societies "genetic" diseases exclude from marriage and other activities in normal social life.
  5. "Genetic" will always be associated with a feeling of guilt in the person. Maybe leading to decisions in life, where the patient may say: No I do not want children, because on my "bad genes". It would also mean that we have to provide genetic counseling for every patient with IGE, which is clearly an overshoot and practically impossible and meaningless.

I hope sincerely that this minor conceptual flaw will be corrected. For some it may be only cosmetic, for some it may be difficult to give in, but for the patients a big relief to take away the burden of the the stigma.

With best regards

Eugen Trinka

Return to top

11 September 2016

We agree with Scheffer and all in many points of the roadmap for the development of an updated , relevant classification of the epilepsies.

As clinicians, we are comfortable with the use of the term idiopathic. We propose the term genetic for “’research classification of the epilepsies.”

About developmental and epileptic encephalopathy, in our opinion, we have to classify epilepsy not encephalopathy. If encephalopathy is an evident consequence of epilepsy, it is epileptic encephalopathy and if epilepsy is consequence of a chronic non-progressive encephalopathy like cerebral palsy, it is "encephalopathic epilepsy," but there is difference between this and progressive encephalopathy associated with epilepsy, preceding epilepsy or occurring after epilepsy with EEG encephalopathic findings.

About the term benign, epilepsy with its consequences is not benign, let us reject this term.

About comorbidities, there is a need of criteria to define what can be considered comorbidity of epilepsy.

Best regards,
Daniel Okitundo for the epilepsy community of the Center for Neuropsychopthalogy /University of Kinshasa
Democratic Republic of Congo

11 September 2016

We wish to congratulate the authors on the new epilepsy classification, which in our opinion represents a user-friendly and relatively simple tool, especially for classifying new patients in a clinical and research setting.

The classification does, however, seem to provide little depth or guidance. We would thus like to suggest that the Commission further elaborates on the classification tree in Figure 1, in order to integrate the different seizure types, electroclinical syndromes, etiologies etc. An example of a more detailed classification tree is provided below to expand on our point.

In addition, we strongly recommend that the authors provide mapping of all terms to terminology used in previous classifications, as they did for the seizure classification.

Finally, we suggest the Commission provide computer-readable codes for each of the terms.

Especially in the setting of large, multi-centre research collaborations clinicians and scientists are often faced with the use of multiple different classifications and terminologies across patient databases, greatly hampering progress. We believe that using a common, homogeneous language across different clinicians, researchers and countries would benefit patients, clinical diagnosis and multi-centre research alike.

Chantal Depondt, Roland Krause, Pasquale Striano, Holger Lerche, Sanjay Sisodiya, members of the EpiPGX Consortium.

Example of a detailed epilepsy classification tree with accompanying codes:

1. Generalized
   1A. Generalized Genetic
      1A1. Childhood Absence Epilepsy
      1A1a. Absence seizures
   1A1b. Generalized tonic-clonic seizures
   1A2. Juvenile Absence Epilepsy
    1B. Generalized Structural
       1B1. Malformation of Cortical Development
   1C. Generalized Metabolic
2. Focal
   2A. Focal Genetic
      2A1. Temporal Lobe Epilepsy
         2A1.1. ADLTLE
          2A1.2. AD Mesial TLE
      2A2. Frontal Lobe Epilepsy
   2B. Focal Structural
      2B1. TLE/mTLE
         2B1.1 HS
         2B1.1a dyscognitive seizures
         2B1.1b sGTCS
          2B1.2 Ganglioglioma
      2B2. FLE
         2B23 Stroke
          2B24 Trauma

Return to top

11 September 2016

Thank you for your important work!

Classification of epilepsy in the clinical situation is often quite difficult due to insufficient information. This is a very common situation in the clinic and to avoid giving the wrong classification, it often feels more safe to refrain from trying. The term idiopathic epilepsy has been difficult to understand for many of us but in my opinion, genetic epilepsy runs the risk of be seen as to definitive. During a discussion recently with one of my patients, who should be classified as having genetic epilepsy according to the new suggested recommendation, I had a hard time finding good argument. The patient had a very big family and no one else was known to have had epilepsy for at least three generations and I could not present any evidence of a gene mutation. For this patient the classification did not help to increase understanding or acceptance, only added concern for heredity. Since we only are in the beginning of understanding how genes interfere with different processes of disease, to my mind it would be a positive thing to emphasize this in a classification. It would be a great help if “genetic” in these cases could be combined with a word that declares this lack of knowledge, for instance, “unknown”.

Yours sincerely
Ulla Lindbom

9 September 2016

Dear members of the ILAE committee for Seizure classification 2016,

We want to thank you for your great efforts to create a new system of classification that will beneft us all. We would like to offer some suggestions that have come to mind when analysing this proposal.

  1. We think it is important to note that not all focal seizures are restricted to a single hemisphere, as there can sometimes be two or more epileptogenic zones comprising both hemispheres, but still be a focal set of seizures.
  2. How could we know if an epileptic spasm has a focal or generalized onset, if not from electroencephalography, which is not always precise.
  3. An EEG in a mesial seizure zone, or a seizure with an onset in a deep part of the brain can be confusing as it can often present a normal EEG
  4. What is the role of electroencephalography in this classification? Does it overule the clinical diagnosis or we should always give priority to semiology of the seizures?
  5. We think the term “unknown onset” is far too vague, we would like to suggest the terms “probable”, “posible” and “certain/definitive”, to define an even more accuarate (or innacuarate) prefix to the diagnosis
  6. We still think ethiology is important in the cassification of seizures, so “genetic” and “structural/metabolic” should be reconsidered as valid terms

We appreciate your efforts and thank you for letting us submit our comments in this subject.

The team of neurology at the Epilepsy Clinic at Hospital Central Dr. Ignacio Morones Prieto, San Luis Potosí, México.

9 September 2016

I am a pediatric epileptologist and I wish to say..the Task force members have really
hit the nail on its head." The framework is exactly what is required to diagnose and manage a child with epilepsy.

1. I am glad that the previous (2010) model of classifying epilepsy on basis of genetic, then structural etc. has given way to this pragmatic model of starting with seizure->epilepsy->syndrome classification with concomitant consideration of etiology and comorbidities.

2. The nomenclature of 'idiopathic' is definitely welcome as compared to 'genetic' in the case of generalised epiepsies, ast is not stigmatizing.

3. Overall, this classification road map is definitely more user friendly and acceptable to subject experts as well as pediatricians working with children with epilepsy

Kavita Srivastava

Return to top

8 September 2016

I commend the committee’s persistence and thoughtfulness in this challenging process. My only comment is that structural includes such a large range of abnormalities with different prognoses. I would recommend dividing structural into “congenital/developmental” and “acquired”. The former group would include in utero strokes, infections, ischemic injuries … which are more likely to have brain developmental effects and outcomes.

All the best,
Charles Akos Szabo

8 September 2016

Dear colleagues,

Congratulations on the road map! I really appreciate the efforts of the work group and their reasonable thoughts to develop a meaningful classification for epilepsy. I thank you for the opportunity to give a comments and suggestions.

With this road map you target the epilepsy community which is reasonable at this point. We should, however, also keep in mind that there are a lot of other physicians (neurologists, pediatricians, neuroradiologists, neurosurgeons, neuropathologists, etc.) out there who we should eventually also address. Why is it that most non-epileptologist just use the term “seizure disorder” when they are dealing with epilepsy? Most feel that the seizure and epilepsy classifications are too complicated to use them in daily practise. Why dont't follow the rules of general medicine for classification. This means dealing with three levels: 1) Symptoms, 2) Syndromes and 3) Etiology.

In epilepsy, the main symptoms are seizures. We should classify seizures according to their clinical appearance – as we do in general neurology – by using descriptive terms. In the epilepsy syndrome diagnosis we summarize all clinical and test findings. This is where we define focal or generalized epilepsies or we get more specific once we have enough information. The etiology is the third level which usually needs time and effort but eventually defines the prognosis and to some extent the therapy. There is no need to make things more complicated by using six levels. If we stick with the three levels of general medical classification, we will have a much better understanding of epilepsy in the medical community. This might eventually reduce the commonly used term “seizure disorder”.

Best regards,
Soheyl Noachtar

6 September 2016

Dear colleagues, thank you for your enormous work!

I think it would be productive to draw your attention to the following proposals:

1. The list of absence seizures should be expanded by including:

a) absence with debut in infancy (Cavazutti G.B. Epilepsia ,1980; 21:57-62. Chaix. Y and authors Epilepsia, 2003;44:944-949. Karlov VA 2010

* b) in adult (Fernander-Torre J.L.; Rebolthilo M. Seizure,2009; 18:82-83. Gendon P.,Ferlazzo E.,Thomas P. Epilepsia, 2008;49: 642-649) because of the fact that the last type is pharmacoresistant.

* Epilepsy in children and adult females and males: Manual for the Physicians. Moscow: Meditsina Publishers, 2010.- 720p. [book in Russia]

2. In etiology of epilepsy that is immune (4) it must be pointed out that immune mechanism nowadays widely included in pathogenesis in many forms of epilepsy and using corticosteroids as one of the main method of the treatment. West syndrome is the most demonstrative example of it. But as an etiological factor immune mechanism in case of immune encephalitis is the main factor. Immune encephalitis must be indentify by antibody types for choosing adequate autoimmune therapy.

Sincerely yours,
Professor Vladimir Karlov

6 September 2016

I thank ILAE task group working tirelessly in revising classification of epilepsies. I would like to use the word “genetic based epilepsies” rather than “idiopathic” because use of genetics and genetic testing is much more accepted terminology by general public.

Baldev K. Singh

6 September 2016

Here we send our comments regarding the Classification of the Epilepsies:

1. We found the diagram of Figure 1 very interesting and easy to follow, useful for general neurologists and neurologists in training

2. Level 2: we agree with the 4 proposed categories

3. We like the term "genetic". We find it clearer than the word idiopathic, which is often confused with cryptogenic or "of unknown origin."

4. We agree on the use of the term epileptic encephalopathy. However the differentiation between epileptic and developmental encephalopathyseems excessive and does not impress us to contribute in the classification of epilepsies

5. We like the use of self-limited epilepsies instead of benign

María del Carmen García

Return to top

5 September 2016

We greatly appreciated reading the paper and we congratulate the authors for the excellent work. Our comments are as follow:

1) Epilepsies classified by seizure type.
Generalised & focal epilepsies is confusing for us. Our proposition is to consider only: "generalized", "focal" and "unknown".

2) Etiology.
The term "structural" includes "infectious" and we can also extend this to "immune" and "metabolic". For the etiology, we may consider only three groups: "genetic", "structural" and "unknown".

3) Idiopathic generalized epilepsy
We don't agree to replace the term "idiopathic" by "genetic". We are afraid the term "genetic" will increase the stigma against people with epilepsy. In sub Saharan African, this diagnosis is made only on history taking and due to financial constraint, we cannot identify the gene responsible for epilepsy in these patients. The term inherited may be better compared to "genetic".

4) Use of the term "benign" epilepsy.
Not agree to use the term "self-limited" or "pharmacoresponsive". Patients and family will be confused and more anxious. Giving the long term cognitive effects, the term "self-limited" is not appropriate for us too.

5) We agree with the term "developmental epileptic encephalopathy".

Thanks to the authors.

Sincerely yours.
Callixte Kuate Tegueu

3 September 2016

My thanks to the Committee for providing this "road map" to classifying the epilepsies. I agree with almost all of your decisions with respect to epilepsy classification, and with the suggested terminology. However, I think that there is one glaring omission, and that is the group with convulsions seizures, with or without absence-like or myoclonic seizures, with 2-3 Hz bisynchronous spike and slow-wave discharges of varying frequency and duration, who present in adulthood.

Berkovic and colleagues called attention to this group in their 2003 article (C. Marini et al, J Neurol Neurosurg Psychiat 74: 192-196 2003). However, their conclusion that "adult onset IGE is a relatively … benign disorder … with seizures that are easy to control" is misleading. We have found that subclinical seizure discharges, if frequent and/or prolonged, can produce a potentially reversible, chronic epileptic encephalopathy (D.G. Fujikawa et al. Epilepsy Behav 25: 442-448 2012).

This group should not be subsumed under either juvenile myoclonic epilepsy (JME) or "epilepsy with generalized tonic-clonic seizures alone" (EGTCS) category. It should be included in this classification, and a category of idiopathic generalized epilepsy of adult onset should be created, to distinguish it from JME and EGTCS.

Denson G. Fujikawa

2 September 2016

The road map detailed by Ingrid Shaffer offers a pardigm shift to a new classification of the epilepsies, an issue that has plagued the ILAE for several decades.

We need to redefine the rationale for a re-classification, to ensure its wide acceptance and its applicability to both patient care and research. Prof. Schaffer's efforts are laudable in both these objectives as she so clearly pointed out, but would it also serve as a roadmap to further our understanding of the basic mechanisms of the epilepsies, especially their genetic underpinnings that have become increasingly to the forefront of epilepsy research?

We look forward to the next draft of the new classification of the epilepsies, having a clear roadmap by which to proceed from a leader in the genetics of epilepies.

Submitted, on behalf of the UTERP,
Paul Hwang

Return to top

1 September 2016

I welcome the ongoing evolution of thinking regarding seizures and epilepsies.

My comments are based on experience attempting to apply clinically previous classifications, audit practice at UK level and build understandable teaching programmes aimed at broad ranges of professional.

My comments are really around the overarching process:

  • I would like to see the classification review cycle move towards 1) provisional proposal 2) consultation and then 3) finalised classification that comprehensively includes all current agreed terms and principles etc (a bit more like the ICD-10 model). At the moment we seem to just have proposals and consultation
  • Again similar to ICD i like the term formulation to describe the attempt to map the individual to a classification. Classification for seizure type and epilepsy patterns in a population; multiaxial formulation for the individual!
  • I support that attempts to provide a practical tool to aid the clinician. Within BPNA PET course we have adopted the DESSCRIBE approach (example http://www.cewt.org.uk/CEWT/Epilepsy-to-go.html). This is an educational and clinical formulation tool that maps to evolving ILAE classifications but builds in the following elements which seem to be missing from current clinical approach
    • an explicit step requiring whether considered episodes are epileptic, non-epileptic or uncertain (this is important as many people with an epilepsy also have non-epileptic episodes; many people go through a phase whether it is not clear whether they are having epileptic episodes or not. Also given the new definitions of epilepsy it is important that we explicitly state that whether we consider that an individual has epilepsy or not!
    • a description axis (was built into previous multiaxial approach) We must systematically capture the descriptions of the episode rather than just tag people with the current seizure classification term. Otherwise we will just keep reinforcing are current thinking rather than populating our notes, minds and datasets to inform the next classification
    • an incremental dynamic diagnostic approach that accepts that information at any level may be incomplete, uncertain or unclear!
  • I think the infographic (fig 1.) in the proposal is confusing. Do the arrows infer causation, association and/or diagnostic process etc?

Colin Dunkley

1 September 2016

It is a highly applaudible effort of ILAE Classification Task Force on Classification and Terminology to prepare a roadmap for the development of a revised classification of the epilepsies. Putting up the same for comments and suggestions has provided a common platform for interaction of physicians dealing with epilepsy patients” world over. This creates a feeling of belongingness among physician groups and a positive approach to accept ever changing concepts based on updated information.

Being a pediatrician, I am trying to anticipate issues which might arise in practical application of the proposed classification while dealing with children with epilepsy. My feedback is as follows:

1. Returning to using "classification" rather than "organization": seems appropriate considering the vast array of heterogeneous group of physicians managing epilepsy patients.

2. Seizure Type classification: The proposed categorization into Focal, Generalized and of Unknown onset appears a bit dubious. Mostly parents panic on witnessing their child throw a seizure. Majority fail to provide details of seizure semiology. Reflex action of parents is mostly to grab their children in their lap and rush to seek medical care. In such a scenario, accurate information about seizure onset and progression usually cannot be obtained. Hence, it is usually not possible to comment on onset of seizure: whether focal or generalized. Only if the child has had multiple seizures or a prolonged seizure, an accurate account of seizure can be obtained. Here again, parents tend to describe the ictal events in form of manifestations of fully established seizure and not its onset. Hence, classifying seizure type based on onset is practically not very yielding in majority of pediatric cases. It is best to state seizure semiology witnessed in ictal phase (i.e when seizure is well established and not at onset mostly/definitely!).

3. Epilepsy classified by seizure type: The category "Generalized & Focal" seems unnecessary.

4. Epilepsy with known etiology/ Epilepsy with ascertained etiology seem more appropriate.

The term Epilepsy with specific etiology is not very clear. Every epilepsy does/would have a specific etiology, whether it is known/ ascertained at present is what needs to be communicated.

5. The term “Genetic” should be avoided as it is taken synonymous with inherited by majority and can aggravate the stigma faced by epilepsy patients.

6. The term Developmental Epileptic encephalopathy seems fine.

7. Replacement of term “Benign” needs consideration but both terms selected: “self limited” and “pharmacoresponsive” don”t seem appropriate. Patients/ attendants might feel that treatment is unnecessary if these are self limited. BECTS is known to have long term cognitive effects, hence can”t be broadly labeled as self limiting in all cases. Again use of term “pharmacoresponsive” might cause undue anxiety in parents” of children with epilepsy who are categorized into any other category which doesn”t include the term “pharmacoresponsive”. They might interpret this to mean that probably their child has an epilepsy category which is not pharmacoresponsive, which obviously is not the case always.

Roosy Aulakh

Return to top

31 August 2016

I meant to sit down and write down my thoughts on the entire paper but just havn't got round to do so.

I think we could drop BENIGN in view of the times that this proves to be incorrect.

May we call Jeavons Syndrome “Jeavons Syndrome” and NOT EMA. EMA suggests Epilepsy with Myoclonic Absences to me rather than Eyelid Myoclonia with Absences.

Could write more. Hope this is not too late.

Megan K Brady

29 August 2016

I was encouraged by Japanese Epilepsy Society to send you a feedback to Special report about Terminology written by Dr Scheffer.

I agree with this scheme of classification of the epilepsy depicted by Fig. 1. In our daily clinical practice, semiology and EEG findings are only easily available information. And I guess the reason why the classification of epilepsy syndrome in 1989 is still used, it is simple and it is constructed by simple clnical findings, which are easily available.

In this mail, I would like to propose the alternative term to "idiopathic".

I have felt discomfort with the term of "idiopathic epilepsy" from the first time to hear it. In other region of medicine, idiopathic is interchangeable with other terms, including essential and primary. Generally to say, their meaning is that cause is unknown, or that all other possible etiology is differentiated. Examples include essential hypertension, essential tremor, idiopathic thrombycytopenic purpura, and idiopathic pulmonary fibrosis. In this term, I found the feeling of giving up to elucidate the etiology. However, in the world of epilepsy, idiopathic does not have such negative meaning. Idiopathic epilepsies have well defined clinical features, and we can make the diagnosis of idiopathic epilepsies on the active history taking. Before I knew this positive aspect of idiopathic epilepsies, I considered idiopathic epilepsies as garbage diagnosis. I am afraid that many medical students are confused by this term of "idiopathic", which has discrepancy in the meaning.

Using the term "genetic," I hope that the etiology of idiopathic etiologies will be clarified and I wish that it will be determined most of them are caused by the mutation of specific genes. On the other hand, I understand that the term "genetic" often leads to stigma or prejudice, though I hope the truth will set us free some day.

Then, I search for another term which is superior to "idiopathic." I think that the critical feature of idiopathic epilepsies is "well defined" by clinical, historical view. Therefore, I would like to suggest term of well-defined, demarcated or featured. As English is not my first language, and I don't know much about European language, including Greek or Latin; I regrettably to say that this choice of the word is not so appropriate. So I hope other my colleague will find more suitable term which has such meaning.

I deeply thank you for read my suggestion

Sincerely yours,
Shin Nabatame

Return to top

29 August 2016

Comments on ILAE Classification from Epilepsy Society of Thailand (EST):

  1. Concept of classification: appropriate and practical for Thailand. EST
    1. Agree with the 4-diagnostic level framework.
    2. Agree with using "classification" rather than “organization”.
    3. Agree with changing "benign" to other term due to this can be misunderstood & may cause overlooking other associated co-morbidities
    4. Not agree with 2nd level diagnosis which categorized into 4 types focal/generalized/focal and generalized / unknown because this may be confusing between focal/generalized and unknown type. Should have only three types (focal, generalized and unknown)
    5. Agree with changing the term IGE to GGE & generalized epilepsy with unknown etiology
    6. Agree with the term "epileptic encephalopathy" rather than "developmental encephalopathy, epileptic encephalopathy or developmental epileptic encephalopathy." In some regions in the country, EEG cannot be done as frequent as being suggested, therefore, it is less likely to make the differentiation according to the recommendation.
  2. Identification of the etiology: there are some limitations on laboratory investigations for epilepsy in the country, especially genetic tests. Precise classification (unknown VS genetic) is unlikely.
  3. Treatment: Owing to the limit resources, history taking and thorough physical examination to make the initial (presumptive) diagnosis of the most likely causes of epilepsy has been emphasized. Commencement of treatment is normally exercised despite the lack of complete investigation. If the initial response is not achieved, further step-wise investigation will be conducted.

Apasri Lusawat

21 August 2016

Dear authors and dear all,

I greatly appreciated reading the paper and I congratulate the authors for their excellent work. My comments/suggestions are the following:

1) Epilepsies classified by seizure type (2)

  • “focal” and “generalized” addresses which seizure types classification? The “old” 1985 one ? Other?
  • Are there, indeed, focal and generalized seizures in the same patient? Why not to consider only “generalized”, “focal” and “unknown” seizures and banned the “generalized and focal” ones?
2) Concerning aetiology (4):

  • The term “structural” already embraces immune and infectious epilepsies, hence I see no raison for considering these two aetiologies (structural is also not subdivided in all the possible causes, is it?) I stand for genetic, structural and unknown, as the Berg's paper from 2010.
  • There should have an arrow directly from aetiology “genetic” to “epilepsy syndromes (3) because most, if not all, of them, are genetics.
  • I see not enough relevance for “big and ugly” ”Comorbidities” left-hand bar, unless the very few comorbities contributing for aetiology and/or syndromes are specified.
  • I see no practical importance for using any one of the terms proposed for “aetiology”

3) - I agree with the term “genetic” but I will tend to subdivide it in “hereditary”, not hereditary” and “unknown”.

  • The term “Idiopathic generalized epilepsies” (IGE) should be banned; I agree with the subdivision of “generalized epilepsies” in “genetics” and “of unknown aetiology”

4) Although agreeing with the argument, I see no major advantage or positive contribution, at least at the present time, for changing the well-known term “epileptic encephalopathies” for “development or epileptic encephalopathies.” It will add nothing to the diagnostic procedures, treatment and prognostic.

5) I could not agree more in banning the term “symptomatic generalized epilepsies”.

6) Epilepsy is never “benign.” This term should only be used to tranquilize patients/relatives/caregivers. “Self-limited” seizures seems also not to be a good option because some of them are not so at all. I would ban both terms.

Looking forward to seeing the paper ready

Best wishes
José Pimentel

Return to top

19 August 2016

Dear all,

It is really a great job to try to get an acceptable form for "classification" of epilepsy. At the start of my comment I would like to throw light on two points of importance in that respect:

First the form of classification which we wish to have should be acceptable and understandable not only from those who are "update" with epilepsy issues like ILAE members but also by those physicians who are not very close to epilepsy and therefore not very aware of the rapid and sometimes obscure changes in terminology like internists, pediatricians, psychiatrists, gynecologists, GPs and etc..i.e. the bulk of doctors who are away from the field of Epileptology and who may constitute more than 80 percent of medical population.

Secondly the rapid and frequent changes in terminology which occurred last years had made some impact of "boredom" on all physicians; so it may be better to cut it short and reach a form of "accepted terms" which should be almost "constant" to avoid misunderstanding. The reinstatement of "focal epilepsy" and "generalized epilepsy" is a very good act since we have the "education" of those terms for long years.

My comment on epilepsy "categories" which is the second level of classification starts with a logic question: what is the value of the terms "focal epilepsy" and "generalized epilepsy" if they are only a mirror for the seizure type? i.e. if it is a focal seizure then it is a focal epilepsy or if it was a generalized seizure then it is generalized epilepsy!

I think that the story is deeper than -beg pardon-this "superficiality". We know that "generalized epilepsy" is always presenting as generalized seizures like CAE or JAE or JME etc although JME may present apparently with seemingly focal jerks! And we also know that "focal epilepsy" can present as usual with focal seizures But also can present with generalized seizures like epilepsy due to stroke or "lesion epilepsy" which is definitely focal since it is not due to channelopathies or genetic reason; it can present with simple focal seizures or complex focal seizures with or without secondary generalization and GTCS or generalized tonic seizure or any type of generalized seizures in its different combinations so focal epilepsy in this respect has focal and generalized seizures but there is no need to put them under the category of "epilepsy with generalized and focal seizures" since there is no element of "encephalopathy" or "cognitive decline" in this focal epilepsy. In that respect we really reinstated the terms focal epilepsy and generalized epilepsy as it was before and not as a mirror for seizure type.

The third type of categories is the "epilepsy with generalized and focal seizure" which seems an "odd term" or at least it is not a "fluent one" and since you mean by this category the epilepsy with different types of seizures with a decline in cognition so a name like "encephalopathic epilepsy" may be useful although the old term "cryptogenic" is not bad and is very descriptive and all know what is meant by it so a name like "cryptogenic epilepsy" is a wishful desire!

"Unclassified epilepsy" category is a good and acceptable term The term "genetic" seems to be condemned and is not easily accepted by people: it may add to the stigma of epilepsy in a lot of societies and is a further impact on families.

The term "idiopathic" is a very "beloved" one and is an open name to include a lot of true and new meanings and is informative to physicians.

The term GGE generalized genetic epilepsy may not be acceptable because of the term genetic and fear of inheritance.

The name "generalized epilepsy of unknown etiology " is a long boring term and not a fluent one whereas the name "IGE Generalized Idiopathic Epilepsy" is beautiful in field of Epileptology: it is well understood, well descriptive, well conclusive, and well accepted so Why to change it into obscure non familiar terms if we want to simplify the field of Epileptology.

Epileptic encephalopathy and developmental encephalopathy are good acceptable terms Lastly "benign and self-limited" terms: the term self-limited may discourage initiation of therapy and it may seem more "benign" than the term "benign" itself!

People are more comfortable with old term "benign" especially after being shocked by the diagnosis of epilepsy; they may regain balance and calm down on hearing a term like "benign"!"Pharmacoresponsive" is more accepted than self-limited and is not implying a "benign course"

Best regards,
Ghaieb Aljandeel

Return to top

16 August 2016

The 2016 Roadmap proposed by the ILAE taskforce has advantages of flexibility and practicality as compared to that in 2010. Of interest, it has levels of classification that bear resemblance to the proposed diagnostic scheme set forth in 2001 (Engel et al, Epilepsia 42: 796), even though there is no overlap in the membership of the respective taskforces!

It should be stressed that a higher level classification does not subsume and replace lower level ones. Information on seizure type, and whether focal, generalized, mixed, or unclear, are all important and each level should be specified. When epilepsy is focal in onset, an important issue is whether anatomic information should be included as in the 1989 classification, but something completely dropped in the 2010 organization. I cannot make out what the Taskforce proposes in this regard.

Recognizing the ever increasing understanding of etiologic mechanisms in epilepsy, the groupings of the major etiologies need not be too restrictive. For instance many pathologies are grouped under “structural” in the proposal, but it would seem important to specify malformations of cortical development and other traditional lesions such as stroke, vascular malformation, trauma, neoplasm, amongst others, for their important epidemiological and clinical purposes.

As proposed by the Task Force, etiology can be classified as unknown. This makes the separation of the 3rd tier of classification from the 4th tier “with etiology” artificial. Rather Etiology and Co-morbidities can be better looked at as separate dimensions or axes.

There was discussion of encephalopathy and developmental disability as examples of co-morbidities. An expanded schema would be helpful. Of note, we need to decide whether a co-morbidity should be restricted to one commonly accepted as epilepsy related, or any major health concern that can impact a person with epilepsy. This is complicated. A person with epilepsy can have hypertension for instance, which is not specifically epilepsy related with no direct impact on its course or management and therefore not a comorbidity. But another can have atrial fibrillation causing strokes that is the etiology of epilepsy, and continues anticoagulation which can complicate anti-seizure medication management.

I read Dr. Barbara Schwartz comments from Aug 1, 2016 and agree with her that our daily work is complicated by the lack of easy translation between the ICD-10 and ILAE classifications. We know that members of ILAE have given input to WHO for ICD-11, but any chance of greater concurrence might not be apparent until ICD-11 is adopted. ICD-10 specifically codes for treatment responsiveness: not intractable v intractable. Whether this should be incorporated into the ILAE schema is debatable, and would need to take into account future WHO ICD directions.

In summary, a fourth level epilepsy classification seems unnecessary and it would be more flexible for future expansions to specify etiology and co-morbidity in separate axes or dimensions. Lobar or other anatomical information for focal epilepsy would be helpful. The schema for etiologies and co-morbidities merit more discussion.

Respectfully submitted,
Norman K. So

Return to top

13 August 2016

I would like to thank ILAE Task members for giving me a chance to express my opinion for the updated ILAE classification.

The updated classification schema based on the diagnostic step from seizure type level to etiology level is a reasonable way for epileptologists to construct the diagnosis of patients with epilepsies, similar to the way they practice at the outpatient clinic. It is welcomed for the Task Force to include epilepsy classification “focal,” “generalized,” “generalized and focal epilepsies” and “unknown,” because it is practically useful on the clinical ground. At the third step, syndromic classification is considered to apply for the patients based on the other important information including the age at onset of epilepsy, circadian rhythm, aggravating factors etc. and EEG findings. The last step is to approach to the etiology of epilepsy or epilepsy syndrome. I prefer the term “Epilepsy with Specific Etiology” rather than “Epilepsy with Etiology,” because etiology of epilepsy needs to be specific. The importance of etiology for epilepsy appears to be different according to the types of etiology, for example, ring 20 chromosomal abnormality is more important than 21 trisomy for etiology of epilepsy.

It is not reasonable to refer to GLUT-1DS as a specific cause of epilepsy with myoclonic-atonic seizures according to the most recent paper (Epilepsia. 2015 Dec; 56(12):e203-8. doi: 10.1111/epi.13222. Epub 2015 Nov 5.)

In conclusion, the updated classification emphasizes diagnostic level from seizure type to etiology, similar to the previous 2001 and 2006 proposal “diagnostic scheme.” In this way to make a diagnosis of epilepsy is reasonable for epileptologists because they have used this flow of thinking at outpatient basis to see the new patients. However, this is not a classification of the epilepsies but the rational way to making the epilepsy diagnosis. The newly-introduced modification is to employ epilepsies classified by seizure type, thus we can use “focal epilepsies,” “generalized epilepsies,” “generalized and focal epilepsies,” and “unknown if generalized and focal epilepsy.”

The term “genetic”

The issue of “idiopathic vs. genetic” has remained a matter of debate. Scientifically, “genetic” etiology largely plays an important role to produce “idiopathic generalized epilepsies”. However, “genetically-related generalized epilepsy” appears to be a more formal term based on the present our limited knowledge and appears to cover the whole IGE syndromes. We would like to avoid using two classification systems, for one for patients and family, and for another for medical chart. As for JME matter, the discussion whether genetic or unknown etiology should be added depending on the positive or negative family history appears to be unfruitful according to our limited knowledge.

Developmental and epileptic encephalopathies

Although the authors refer to Dravet syndrome as an example of the fact that underlying etiology is also important to determine the developmental prognosis, it is not appropriate for Dravet syndrome to refer because patients with this syndrome suffer from recurrent postconvulsive encephalopathy due to frequent status epilepticus instead of frequent epileptic EEG abnormality. Of course, I do not intend to deny the developmental problems due to SCN1A mutation. I suggest adding the following sentence in the definition of epileptic encephalopathy; the epileptic activity itself “either solely on EEG level or seizure level” contributes to severe -------------.

The introduction of new terms “developmental encephalopathy” or “developmental epileptic encephalopathy” may have complicated the medical terminologies in clinical practice. In the former term, “static encephalopathy” has long been used in pediatric neurology field. In the latter term, it has been already well-known that the developmental prognosis in patients with symptomatic West syndrome depends largely on the pre-existing etiology rather than the control of West syndrome itself.

Symptomatic Generalized Epilepsy often maps to epileptic encephalopathy

In clinical practice, we should better to have the terminology to cover generalized epilepsies with various co-morbidity or etiology (non-epileptic encephalopathy type) other than 6 etiological categories. Is there any appropriate terminology instead of symptomatic generalized epilepsy for non-epileptic encephalopathy patients?

The term “benign”

It is scientifically correct that even children with BECT have a wide-range of co-morbidities from minimal to significant psycho-social problems and also need to be treated with antiepileptic drug for years (of course, some professionals refuse to give AEDs despite recurrent seizures). However, we have to keep in mind that learning disabilities or other psycho-social features in children with BECT are largely pre-existing co-morbidities (LD, ADD, ADHD, ASD) and may not be consequences of epileptic seizures or epileptic EEG abnormality, or epilepsy itself. Thus, epilepsy itself may be benign. “Pharmaco-responsive” and “self-limiting” hold different meanings. In idiopathic focal epilepsies in children, the latter word can be applied. The former “pharmaco-responsive” can be applied for most of patients with genetic generalized epilepsy.

Thank you very much for reading my opinion,
Hirokazu Oguni

Return to top

12 August 2016

Dear all,

Thank you for your great work on roadmap on classification of epilepsies. During the regular meeting of Georgian League Against Epilepsy (GLAE) with participation of 14 epileptologists, on August 10, 2016 the proposed scheme of new ILAE classification of the Epilepsies was carefully discussed and the following suggestions were raised:

1. Regarding the term “Organization of the epilepsies
This term is more sophisticated for epileptologists, but difficult for other medical doctors to understand. So all participants agree to return the word “Classification” for describing the epilepsies.

2. “Classification of seizure types
All participants came to agreement to add “Generalized & Focal” in the classification, because such cases are increasingly observed in clinical practice and until now they are usually incorrectly allocated in the category of “Unknown” seizure types.

3. “Etiology”:
All participants supposed

3.1. To accept the term “Epilepsy with known etiology,” because the following terms: “Epilepsy with etiology” and “Epilepsy with specific etiology” are somewhat vague.

3.2. The term “Generalized epilepsy with unknown etiology“ is more suitable than “Genetic“ because:

a) The term “Genetic” should have genetic confirmation in all cases that is not always possible for persons with epilepsy in developing countries that includes 80% of the world.
b) In community the term “Genetic” always perceived as an “inheritable” that makes the diagnosis poorly acceptable for patients and their families, that will result in raising epilepsy stigma.

4. We suppose that the term “Self limited” is much better than “Benign”

Best Regards,
Nana Tatishvili
Sofia Kasradze

9 August 2016

Thank you for sending this document for comment.

Simple comments

  • For educational purposes at varying levels including patients, carers, ward nurses, specialist nurses, practitioners , specialists and professors, I feel strongly and agree that a simplified algorithm of classification is the way forward. A simpler guide to diagnosis , treatment, and research. This can be adapted for teaching and expanded as needed for levels of learning.
  • I agree there should be further clarification of what is genetic and what is inherited. again a simplified algorithm/ statement even could be formulated for clarity. I ;often referred patients that idiopathic is the new “probable” genetic which in most gives the impression of being inherited, causing a concern for those who want to start families. Of course I clarify this but can quite rightly fall of deaf ears.
  • Otherwise I think this is an excellent piece of work, please keep things simple as possible, for research academics it is something they take in their stride but for explaining to patients and carers this is another story. The simpler the better.

Kind Regards and best wishes
Sally-Ann Remnant

3 August 2016


Accusant réception de votre message, en rapport à la nouvelle classification des épilepsies, j'ose espérer que la présente tiendrait compte des aspects opérationnels sur le plan électro-clinique; bien qu'il existe encore une large gamme des épilepsies non classées.

Ainsi donc, une approche approfondie de ces aspects électro-cliniques par le groupe d'experts, aiderait à une compréhension aisée des résultats attendus, en terme de finalisation de ce gigantesque travail.

Diallo Lansana Laho

Return to top

20 July 2016

Thank you for the prepublication of the new epilepsy classification framework. I think it is a significant improvement over the 2010 scheme. I do have some comments. I know that the Epileptic Encephalopathies represented an important new category reflecting our understanding of the effects of seizures on the brain and on development. However, I find it unsatisfactory since, as pointed out, some developmental syndromes begin prior to the seizure onset, in some, controlling seizures prevents progression while in others it does not, etc. Also, adults can be encephalopathic from too many seizures as well and, over time, many of these are not entirely reversible. I would put “encephalopathy” in the comorbidity class. Of course then what does it mean except cognitive and behavioral disability?

I also think that Etiology should be known, presumed, or unknown. For example, in my paper on post traumatic epilepsy without MRI changes all subjects had had a head trauma preceding (by months to years) onset of seizures but none were so called “significant” head traumas (loss of consciousness of 30 min or more or ICH on MRI). Yet the presumed etiology was the head trauma which was proven by later surgical pathology. Possibly new and improved MRI techniques would have shown changes but these are still not typically ordered as part of a seizure work up (i.e. DTI). I also had a patient with intractable frontal lobe epilepsy whose only risk factor was an earlier mumps encephalitis. Surgical pathology again showed typical viral brain and perivascular lesions. So presumed infective etiology was appropriate but not provable without surgery. When we say unknown it implies total lack of knowledge of possible cause which is almost never the case (in my experience).

The diagram is attractive but doesn’t convey many important points in the body of the text. An algorithmic diagram would be very useful for clinicians.

Finally, a classification scheme is not useful to clinicians who are trying to code their patients for visits, payment, data banks etc, if it does not directly refer to the current coding system. I personally find the ICD-10 completely useless for labelling my epilepsy patients. I generally use the ICD-9 and a converter. This is extremely important as the major EMR’s such as EPIC, Coerner, whatever Kaiser uses, etc, are being changed to be usable as data banks to track public health issues and do clinical research. Mining of these data banks will in turn determine future research funding and governmental health care initiatives and thus it is vitally important to the future of our patients that we get this right on an International level. While it will represent a lot of work I think a supplemental paper needs to be prepared to show clinicians directly if a patient is classified one way what ICD-10 code will apply and what supplemental codes should be used for the comorbidities. I would be happy to be on a task force working on this.

Congratualtions on a great job.
Barbara E Swartz

30 June 2016

I applaud the development of a tiered system of classification. This is a perfect combination of structure and flexibility that allows clinicians to transform research into practice. I am seeking clarification on a few points.

When diagnosing a type of epilepsy the manuscript refers to focal vs generalized, but what of the cases of multifocal epilepsy, hemispheric epilepsy or combinations of focal / multifocal and generalized epilepsy?

I personally like the term Epilepsy of unknown etiology. Perhaps the correlate should be Epilepsy of Known Etiology. Though typically if I know the etiology, I will simply state that etiology rather than saying Epilepsy of Known Etiology.

Finally, I will throw my support behind the term “Generalized Epilepsy of Genetic, or Unknown Etiology” as it is the most flexible for example we could easily adapt this to “Early onset generalized epilepsy with generalized tonic-clonic and absences seizures due to Glut-1 deficiency.”

I thank the committee for considering these comments.

Scott Demarest

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.