Contact Us

Ilae Facebook Ilae Twitter

Organization of the Epilepsies

Request for Comments

The ILAE Commission on Classification and Terminology has prepared a new manuscript on the organization of the epilepsies, based on the 2010 proposal, and revised to incorporate many comments that the Commission has received. The proposal and a quick overview are linked below, followed by comments received from November 2013 to January 2014. A task force appointed by the Executive Committe is reviewing the comments, and preparing a final proposal.

Organization of the Epilepsies full document | Quick overview
Please send your comments to who will post them here for the entire community to consider.


The following are comments made on or before 5 January. Read more recent comments here.

5 January 2014

The proposed classification (generalized and partial) is easier to comprehend than the old classification.

The generalized epilepsies is complete; but where would you classify the ones with focal onset? I think it should be a separate entity under generalized epilepsies because it reveals the most possible etiology.

The partial epilepsy with preserved alertness and with impaired alertness is better understood than the simple and complex partial seizures.

The epilepsy syndromes according to the age of onset is very helpful.

Thank you for coming up with this important update. Congratulations to the authors!

Very truly yours,
Maria Lina Renales

5 January 2014

Thank you for sending this document on the recent organization of epilepsy. This is an interesting paper. I must commend the authors for the innovative and thorough approach to this classification of epilepsy.

I agree with the terminology of 'organization' based on the reasons advanced by the authors. In addition, the nosology and neurobiological basis for epilepsy terminology at the moment are inadequately understood to allow for thorough classifications. I agree with the present organization, it is explicit, self descriptive and easy to apply to clinical scenarios.

The major challenge to the implementation of the current organization is the lack of neuro-imaging and genetic facilities in poor resource countries in sub-Saharan Africa as this will mitigate against the application of this document to evaluation and management of our patients. I guess neurologists in these regions (SSA) would have to rely on clinical skills/acumen (which has always been the case) to place patients with epilepsy in the appropriate categories.

The infectious category under structural/metabolic is particularly relevant to SSA where significant number of our patients with previously classified symptomatic epilepsy belong. Furthermore, most of the PWE previously categorized as idiopathic actually belong to the genetic group.

The 1981 classification included 'status epilepticus' – where is this emergency in the recent 'organization'? What of the patients with Reflex seizures like reading, photosensitive, musicogenic, catamenial, etc. epilepsies? Where would they be placed? Are there genetic basis for these epilepsies?

Thank you.

5 January 2014

This is an excellent document, Thanks for this opportunity.

1- I think, term of "Structural " is not enough to define microstructural abnormalities such as non detectable dysplasias, traumatic or posttraumatic seizures (without brain lesions in neuroimaging).

2- Some seizures are developmental problems as West syndrome, absance seizures, rolandic seizures, occipital seizures or " benign neonatal – infantile seizures" etc. In classification, terms of "Developmental seizures" or another definition for these seizures groups may be used.

3- Infectious seizures are probably structural or microstructural because of neuronal death or neuronal dysfunctions without any detectable radiological findings. Also in inflammatory area, there are some cytokines which may cause neurostimulation, neuronal dysfunction or apoptosis. Hence, these types of seizures are structural - microstructural or immune. I think we don't need "infectious" as a category.

4- Some seizures such as paraneoplastic seizures, are mediated by neurostimulation. It may be included in classification.

5- Probably there are some other reasons as paraneoplastic seizures, reflex seizures etc. "Others" category may be used for rare cause of epilepsies or for unclassified etiologic categories.

6- In focal seizure classification: "Evolving to bilateral convulsive-seizures" is a good definition. However some seizures are asymmetric (Eg: temporal or extratemporal seizures). I think asymmetric tonic and migratory seizures may be included in seizure types.

Ozgur Duman

Return to top

5 January 2014

I would like to endorse the comments of dr Jacqueline French including both the congratulations on all the hard work and the suggestions for improvement/modification.

Many of these comments echo my commentary in gray matters about the new classification in its prior iteration in 2010. While the revisions are an improvement, issues remain.

The disappearance of IGE is of particular concern. From a therapeutic point of view, it really makes little sense from a therapeutic point of view to lump Childhood absence and Juvenile myoclonic epilepsy (2 classic examples of IGE) with Dravet syndrome even if all are genetically mediated.

The new revisions take into account many of the comments and critiques of the 2010 proposal. But more attention needs to be paid to the utility and usability of the revised classification in clinical practice and in clinical research.

Shlomo Shinnar

5 January 2014

It is interesting to read the proposal for the new classification of epilepsy. In my opinion, it is much more structured than before, good job!

It is also good to see that "benign" is eliminated in Rolandic epilepsy. It took me (and the rest of our study group) several years of reseach to prove that it is not that benign as it is always suggested. I've some suggestions for references, as most references mentioned in the proposal were on literature of gen-mutations and former classifications.

Kind regards,
Geke Overvliet

Early onset of cortical thinning in children with rolandic epilepsy.
Overvliet GM, Besseling RM, Jansen JF, van der Kruijs SJ, Vles JS, Hofman PA, Ebus SC, de Louw A, Aldenkamp AP, Backes WH.
Neuroimage Clin. 2013 Mar 22;2:434-9.
Aberrant functional connectivity between motor and language networks in rolandic epilepsy.
Besseling RM, Overvliet GM, Jansen JF, van der Kruijs SJ, Vles JS, Ebus SC, Hofman PA, de Louw AJ, Aldenkamp AP, Backes WH.
Epilepsy Res. 2013 Dec;107(3):253-62.
Reduced functional integration of the sensorimotor and language network in rolandic epilepsy.
Besseling RM, Jansen JF, Overvliet GM, van der Kruijs SJ, Vles JS, Ebus SC, Hofman PA, Louw Ad, Aldenkamp AP, Backes WH.
Neuroimage Clin. 2013 Jan 18;2:239-46.
Reduced Structural Connectivity between Sensorimotor and Language Areas in Rolandic Epilepsy.
Besseling RM, Jansen JF, Overvliet GM, van der Kruijs SJ, Ebus SC, de Louw A, Hofman PA, Vles JS, Aldenkamp AP, Backes WH.
PLoS One. 2013 Dec 23;8(12):e83568.
 Impaired language performance as a precursor or consequence of Rolandic epilepsy?
Overvliet GM, Aldenkamp AP, Klinkenberg S, Vles JS, Hendriksen J.
J Neurol Sci. 2011 May 15;304(1-2):71-4.
Clinical evaluation of language fundamentals in Rolandic epilepsy, an assessment with CELF-4.
Overvliet GM, Besseling RM, van der Kruijs SJ, Vles JS, Backes WH, Hendriksen JG, Ebus SC, Jansen JF, Hofman PA, Aldenkamp AP.
Eur J Paediatr Neurol. 2013 Jul;17(4):390-6

5 January 2014

Thank you very much for asking me to comment on the ILAE report. I have looked over it and have found it excellent. However, I have a few comments.

Generalized seizures and focal seizures
The classification of generalized seizures is simple and well organized. Epileptic spasms are unfortunately excluded from generalized seizures because they can be controlled by focal resection or hemispherectomy. The concept of generalized seizures should imply that the seizure discharges involve both hemispheres instantaneously, but presumably triggered by focal or multifocal "pacemaker" foci; cortical and subcortical networks are driven through the "pacemaker" focus, which may be focal in one hemisphere or multifocal in both hemispheres. In such concept, the disappearance of epileptic spasms by focal resection may not justify them excluding from generalized seizures.

Regarding the classification of focal seizures, the example of complex partial seizures vs. focal seizures with impaired consciousness or awareness and those vs. focal dyscognitive seizures are well illustrated the difference in the classification system. I hope that all representative types of focal seizures are better to be replaced by the new descriptions like Table 2 which replaces complex partial seizures into those of the new classification system. I hope that newly introduced seizure classification system is better to be more organized.

Generalized and focal epilepsies: not always a dichotomy
Most of seizures, or epileptic EEG abnormality which are at times difficult to classify into either generalized or focal are seen in childhood epilepsies in which the maturation of brain would affect the synchronization or generalization because once the patients grow up to adolescent or young adult, the seizures and epileptic EEG abnormality appear to be more fixed and static. For example, patients with Dravet syndrome show both (unstable) focal and generalized seizures during early childhood. However, once, they grow up to adolescents, they tend to have generalized tonic-clonic seizures and no polymorphous seizure types which they used to have during young.

Thus, generalized and focal dichotomy can still explain the fuzzy problem if utilizing the old neurophysiological concept of secondary bilateral synchrony together with Gloor’s generalized cortico-reticular hypothesis.

Etiologically based diagnoses
Immune and infectious etiologies are newly introduced in the new system. However, the inclusion diseases may be responsible to both acute and remote symptomatic seizures. Epilepsy in some of the disease may be secondarily immune-mediated. Thus, immune and infectious etiologies are better to be combined to immune and infectious etiology.

Table 3
Early onset childhood occipital epilepsy (Panayiotopoulos type) is incorrect. Panayiotopoulos syndrome is not an occipital epilepsy but epilepsy with age-dependent unstable seizure focus (having occipital seizures during young and more anterior onset seizures later on, along with the shifting of predominant epileptic EEG foci). As there is no good alternative terminology to characterize the epilepsy, Panayiotopoulos syndrome should be left.

Epilepsy with myoclonic atonic (previously astatic) seizures.  Does the description of myoclonic atonic seizures imply either myoclonic (and/or) atonic seizures or myoclonic-atonic seizures?  I think the former is correct because the patients can have myoclonic-astatic, atonic or myoclonic-atonic seizures.

Hirokazu Oguni

4 January 2014

I have read this report and though it helpful. This proposed organization is clear and easy to use in daily work. Although it will take some time to accept the term “dyscognitive” for me, I think “dyscognitive” is better than “complex partial seizure”. I have no further comments.

Xiaoqiu Shao

4 January 2014

I appreciate the opportunity to comment on the “organization of the epilepsies” proposal.

I would prefer the use of the term “classification” over “organization” to help emphasize the implicit continuity of the current proposal with previous attempts to “classify” seizures.

Over all, I think the proposal addresses an important need to provide an organizational framework for our advancing understanding of the etiology of subtypes of epileptic seizures.  The proposed change from the term “partial” to “focal” is welcomed as “focal” is a more intuitive term.  The discussion of the overlapping features of focal and generalized seizures is important, and again reflects a progression of ideas and understanding of epileptic seizures. 

The broad approach of epilepsy diagnosis as electroclinical syndromes and  etiologically-based diagnoses is also reasonable.  I have no specific comments on the outline of ideas presented, only to state agreement with the observation that individual cases will likely overlap in classifications of electroclinical syndromes and/or etiologically-based diagnoses.

In general, I have one major concern, in that the proposal includes major changes from previous classifications.  There will be many practical issues in applying the new “organization” of the epilepsies, but one primary issue will be changes in clinical AED trials.  Phase II and III studies of AEDs typically rely heavily on past classifications of “partial” and “generalized” epilepsy.  Also, organization of seizures into various etiologies often has no direct implication for medical treatment.  For example, with rare exceptions (such as Dravet syndrome), a genetic etiology of epilepsy is treated the same medically regardless of whether a genetic etiology is determined or not.  Therefore, I strongly recommend ongoing assessment of the proposed organization from the perspective of clinical AED trials, which I think would offer an important testing ground of the proposed ideas, both as a help in using the organization for research-based outcomes, as well as practical application of applying the organization for clinical diagnosis and treatment.

Ed Hogan

Return to top

4 January 2014


I have read this material over twice.  It still does not tell a person how to diagnose a victim who has complex partial seizures that can change or secondarily generalize from a complex partial seizure into a generalized seizure that causes the body to turn and if lying on a bed, can cause the person to come very close to smothering as the seizure presses the face into the bed.  This is considered a genetic seizure in our family, as it has been seen in three generations, and written about by members of the previous three  generations.  It also has the similarities of Jeavons Syndrome as well.  I do believe that you need to look more closely at the genetic epilepsies that are under this document considered unknown.  It is not good to have an umbrella term used with genetic epilepsy.  As Berkovic and Scheffer have pointed out if the correct diagnosis is achieved, it will be easier to medicate the victim. It continues to be important to gather good family histories, and to absorb the body language of the family whose child or young adult may have epilepsy that is genetic (Janz, D, 1993).

Otherwise the material is great. 

I believe that we need to think more positively about finding what drugs will work with refractory epilepsy.  Phenobarbital, if used at a lower degree than usual, sometimes does work, when nothing else does.  To that I can testify.  In persons who have some type of seizure on most drugs with which they are presented, the need to rule out generic drugs is very important.  Myzantoin  (1940's) and Meberal (2012)  when used with Mysoline did control this seizure form, but they were removed from the market.  Since Meberal turns into phenobarbital in the system, it would seem that it should have remained available to persons who used it, but it did not.  That has created a problem for a lot of people.

Nancy C Schumacher

4 January 2014

Thank you for asking for comments about this excellent document.

- I think that the term "classification" is still more appropriate than "organization.

- Tumor-associated epilepsies should deserve a special attention in the lesion-structural chapter, owing to their unique aspects.

Best regards and a happy New Year,
Bruno Jandolo

4 January 2014

I am so grateful to get an opportunity to comment on the new proposal.

I would like to stress on the idea of the new title – ‘organization’, to my opinion the previous term classification is more appropriate.

The other thing that I would like to comment on is the opened term unclassified type. This will make one confuse is the patient is clearly an epileptic patients or it is uncleared diagnosis of a clinical or electrophysiological presentation.

We can first apply a criteria for involving such patient in this category then according this criteria we can give groups of subtypes like what Kavita Srivastava said in this direction in his previous comment.

Fakhir S. Al-Ani

4 January 2014

Thank you for asking for comments and for your excellent effort.

I think the new classification of partial epilepsy is confusing. The old classification of simple, complex and secondary generalization is simple and clear. Otherwise I agree on the other modifications

Tayseer Zein

4 January 2014

I should first of all thank the team for their commitment to simplify epilepsy classification. I am a mental health nurse currently pursuing a degree in nursing. The only comment I can make is that the document is rich with accurate information and it is easy to understand. This document must be made available to academicians in Africa popularization. Despite reading the current classification, I have not heard it being taught to students and not even be used in clinical areas. Clinicians and Nurses are still using the old classification.

Mutinta Mweemba

Return to top

4 January 2014

I would like to submit the comments on the Classification and Terminology as ILAE's request.

The classification of the epilepsies you made is an excellent one. In your refined organization, the second domain concerning etiology is quite interesting.  Among them, genetic aetiology could be an imaginary one in the poor resource setting like our country, Myanmar. Again unknown aetiology under the umbrella of aetiologically based diagnoses and unclassified epilepsies might be overlapping. Nevertheless these six non-exclusive categories would have many positive effects on therapeutic approaches.

I would like to appreciate the League’s Commission on Classification and Terminology for their wonderful endeavour.

Sincerely yours,
Nwe Nwe Win

3 January 2014

I think that the committee did an excellent job and that the summary is quite clear.

Under the etiologies section-genetics, the committee might consider adding a couple of sentences regarding the mechanism of somatic mosaic mutation as an underlying genetic cause of a focal/structural epilepsy. The causative somatic mosaic mutation for Sturge-Weber syndrome was recently reported and it seems likely that this mechanism will also be found to underlie other focal/structural epilepsies. Considering this form of genetics is key to research, and finding the mutation we hope will impact treatment.

Best regards,
Anne Comi

3 January 2014

Sorry to be late. Thank you very much for the opportunity to comment the excellent work of the ILAE task force.

I agree with most of the statesments and concepts illustrated in this topic, but I have only some concerns that are
shared by many other collegues:

  • The term "dyskognitive" seems to be unclear. In my opinion "seizures without impairment of consciousness" is more
  • Regarding the use of the term "antiepileptic drug", I agree with the comment of J. Engel Jr ( correct form: antiseizure drug)

Best regards
Eduardo Cumbo

3 January 2014

I would like to support "infectious" etiology since it is quite common in developing countries as remote symptomatic epilepsies.
What about another etiological category - hypoxic/ischaemic!

Many children with cerebral palsy due to birth asphyxia got epilepsy and I'd suggest as hypoxic/ischaemic etiology.

Kyaw linn

Return to top

2 January 2014

I am thankful to get an opportunity to comment on the new proposal.

1. I beg to differ on the new title - ‘organization’, I think the previous term classification is more appropriate. The term organization can be reserved for the age wise categorization of electro-clinical syndromes.

2. The meaning of complex focal in terms of impaired cognition is well understood and has been in popular usage. The new terms suggested as alternatives have confusing connotations.I would suggest to retain this term.

3. The hierarchy suggested : ( A) Electroclinical Syndrome- ( B) Etiology- (C) Unclassified looks very simplistic.
This leaves a large chunk of patients unclassified, we can form some subgroups which will help us in delineating further treatment strategies. I used this same approach in my paper presented at 8th AOEC very few out of 1000 patients were left completely unclassified.These subgroups, even though not well defined, are inhomogeneous, still help in deciding drugs and other therapies.

(a) Probably Symptomatic epilepsy- with normal IQ, epilepsy,EEG shows focality and MRI is normal.

(b)Symptomatic focal epilepsy- developmental delay of unknown cause with epilepsy,

(c) Epileptic encephalopathy - not fitting in West Syndrome / LGS / CSWS / LKS etc. These could also be called ‘complex focal with essential generalisation’ as suggested by Brian Neville.

(d) Generalized epilepsies - not fitting into known syndromes (CAE, JME etc).

Also, In the unclassified group, there should be an attempt to at least classify the seizure type: Focal / generalised / unclassified seizure type with further detailing.

4. I would also suggest that few categories (like post ischemic/ post traumatic/ post infectious etc.) may be added in the structural group to facilitate further classification, rather than having them a separate etiologies.In our country, perinatal HIE and hypoglycemia are the major causes of epilepsy.

5. I agree with previous comments that ‘idiopathic’ should not be equated with ‘genetic’

6. I would like to add that the unclassified group may be due to either ‘not completely worked up’ or ‘adequately worked up’ but the commission needs to elaborate on what is the extent of ‘adequate’ workup for a given category of patients. e.g. In our country, there is dearth of facilities providing metabolic and genetic tests, that too at affordable rates. Sometimes, even MRI may not be possible because of financial reasons.

I think the Axis approach (Engel,2001, 2006) works best (we are currently using the same):
Axis1: Type of seizure
Axis 2: Type of Epilepsy
Axis 3: Electro-clinical Syndrome
Axis4: Etiology
Axis 5: Severity and co-morbidities

Thanks and Wishing you a very Happy New Year to all.
Kavita Srivastava

1 January 2014

Thank you for asking for comments in this topic.

Really for neurologists or neurosurgeons working in this topic, has been so difficult to make a good classification of this kind of diseases, this concern is greater with the patients itself.

This paper with the "new" concepts shows a really giant effort to try help us in the understanding but specially in the work for helping others.

My suggestion is to try keep the definition of the word "symptomatic" because many of epilepsies can not be ascribed to a single cause or in some cases despite exhaustive search we are not able to find an specific cause; any way looked as a better explanation the necessity to keep on mind the search for one.

Fernando Estévez

Return to top

1 January 2014

The Commission fine-tuned an excellent document and it should be acknowledged for the great work done regarding this complex matter. Clearly, a multiscale classification would be much more suitable for the complex world of Epilepsies but the Classification must be an easy-to-use and ‘acceptable agreement’ with strong practical implications, especially for non-specialists in the field.

Some comments:

  1. The terms "idiopathic" and "benign" continues to be controversial. If we look at the Classification other non-neurological diseases (e.g, myopathies, scoliosis, arthritis), the term "idiopathic" refers to genetic (not necessarily inherited) conditions with a strong genetic background. Lastly, if we have Genetic Generalized Epilepsies why should we not have also Genetic Focal Epilepsies? The term "self-limiting" may be somehow misleading as well as "benign". These terms are not strictly needed for classification purposes although they are extremely useful in terms of communication with the parents/patients.

    In summary, it seems that there is so added value in just switching "idiopathic" and "benign" to "genetic" and "self-limiting", respectively. The same can be applied to ‘dyscognitive’ which is not properly synonymous of ‘impaired consciousness’. We could just keep the term ‘complex’ for patients who do not have awareness of his/her seizures. This concept has practical implications for physicians and patients’ life.

  2. We do not feel that the introduction of the ‘Immune’ Epilepsy is justified at this stage. Epilepsy is not a constant symptom in autoimmune encephalopathies and when present it is always associated to ataxia, cognitive impairment and additional neurological manifestations. Eventually, Rasmussen Encephalitis and celiac-associated epilepsy should also be included among this chapter.
  3. Within the lesional-structural (i.e., symptomatic) chapter, tumor-associated epilepsy deserves special attention for their unique evolutive and physiopathological aspects (the so-called ‘small-world’ network suggested by Smit et al., 2008) which also explains the poor efficacy of currently available antiseizure drugs.
  4. Table 1. We feel that Eyelid Myoclonia should be included within myoclonic rather than absence seizures. This is important also for pharmacological management as they are better responsive to antimyoclonic anticonvulsants.
  5. Table 3: the statement ‘these are examples of syndromes and not a complete list...’ appears a little elusive. The Commission would make even a better job if they could indicate which syndromes are accepted or not. We would introduce sporadic (i.e., non-familial) ‘Idiopathic Lateral Temporal Lobe Epilepsy (or IPEAF)’ and ‘Nocturnal Frontal Lobe Epilepsy (NFLE).

    -We would delete the term ‘Early-onset’ and ‘Late-onset’ used for Childhood Occipital Epilepsies. Panayiotopoulos syndrome vs Childhood Occipital Epilepsy of Gastaut is sufficiently clear.
    -We also do not agree that ‘Febrile Seizures’ is an electroclinical syndrome. This statement may have implications in terms of management and pharmacological treatment.

  6. There is no mention in the document of Reflex Seizures and Reflex Epilepsies. The latter group accounts for up 5% of all Epilepsies and some of them carry many practical issues, e.g., use of special lens for patients with photosensitive seizures. In addition, they represents an unique opportunity for studying the physiopatological basis of many forms of epilepsies.

A final comment: it should be clearly stated whether this is or not THE New classification. This situation is generating confusion among general practitioners, medical students, patients, and ... even epileptologists. Some reviewers of peer-reviewed papers often requires to use the new terminology. Please clarify that in your final version of the document.  

Thanks again for your excellent work and we wish you all a Happy New Year.

Pasquale & Salvatore Striano

29 December 2013

This is a concise and helpful classification. The authors and contributors should be thanked for their effort and time. I only have one suggestion. The term "generalized" might give the false impression that further workup (imaging or surgical workup) is not needed to delineate the focus of seizure generation. I hope that an approach similar to the one described below might be considered in the future in order to account for the known or suspected mechanism of seizure onset.

Generalized and focal seizures, as defined in the manuscript, refer to an electroclinical phenotype based on EEG features and clinical behavior and not to a mechanism of seizure generation. According to the manuscript, generalized seizures involve networks in both hemispheres. The only manner by which electrical networks in both hemispheres are, or appear to be, simultaneously activated without defying the chemicophysical laws of nature is by receiving simultaneous orders from a maestro, or via rapid spread from such a maestro, whereby the electrical spread escapes our detection techniques. Such a maestro can be a cortical focus or a subcortical circuit. In addition, theoretically, generalized seizures might variably originate from different hyperexcitable foci at different times in the same patient. While the terms "focal and generalized" upon which a whole literature on response to AEDs was built, cannot be just abandoned, it might be helpful to further characterize them based on the known or suspected mechanism of seizure generation along the lines of the following classification:

  1. Cortical epileptogenic focus
  2. Multiple cortical epileptogenic foci
  3. Subcortical epileptogenic mechanism
  4. Undetermined

According to this proposed additional classification, childhood absence seizure is an epilepsy syndrome that manifests with a generalized seizure phenotype and that has a subcortical mechanism of seizure onset. A patient with tuberous sclerosis or structural Lennox Gastaut might experience both generalized and focal electroclinical seizure phenotypes, but mechanistically, his or her epilepsy will be classified under either "cortical epileptogenic focus" or "cortical epileptogenic foci". Such a classification respects our current understanding of the pathophysiology of seizure generation and alludes to surgical candidacy without abandoning the terminology of the electroclinical phenotypes of "generalized or focal seizures" delineated in table 1.

Thank you for giving me the chance to give my opinion and Happy New Year,
Makram Obeid

27 December 2013

Thank you for asking me to comment on the ILAE report. I have read it and have found it excellent. However, I have a few comments:

Page 6: Childhood epilepsy with centro-temporal spikes may be misleading. Children can have structural epilepsy with centrotemporal spikes. A name, which has been successively more used in literature over the last years is Rolandic epilepsy.
Table 3: Self-limited epilepsy with centrotemporal spikes (ECTS) is the same as Rolandic epilepsy, but the last-mentioned name is more simplel to use.

Page 6: Epileptic encephalopathy depends on an underlying brain dysfunction resulting in both seizure manifestations, electrographic abnormalities, and cognitive and behavioral disorders. Consequently, can we really say that the epileptiform activity as such influence the mental disturbances???

Page 8 and 9: Under (2) Structural, the last paragraph, ‘genetic’ is mentioned. ‘Genetic’ should also be mentioned with respect to immune, infectious and unknown etiologies.

Orvar Eeg-Olofsson

27 December 2013

Thank you for the opportunity to comment the new report of the ILAE Commission on Classification and Terminology related to the "Organization on Epilepsies".

First, I agree that whatever the final document contains, it should be called a classification. My main concerns refer to the etiological categories. Great numbers of physicians don't have the ability to specify genetic, metabollic, immunology factor as etyologic, we suggest to keep term symptomatic, with possibility to specify overknown cause (genetic, metabollic, immunology, infectious, traumatic, hypoxic-ischemic, vascular etc).

Also, we think that if we plan to keep term focal motor, autonomic, why shouldn't we keep term focal sensitive, focal sensory, focal psychic etc. instead suggested term aura.

With best regards,
Almasa Kapidzic

Return to top

23 December 2013

In my view, this refinement of the proposal is an improvement of the 2010 version. I only have a few comments and suggestions on details and terminology rather than on the overall proposed structure. These few details may, however, be relevant since the document will end up as an official ILAE Position Paper.

1. In Abstract as well as in the text, the authors write that "The etiology may fall into two or more categories." I assume they mean "one or more categories."

2. There are a couple of statements that are difficult to interpret. (i) P 4. Last paragraph. "In order to make an epilepsy diagnosis, in addition to a careful clinical evaluation, it is necessary to have access to EEG an MRI studies where appropriate." What is meant by "where appropriate"? I guess that the authors would agree that although EEG and MRI are desirable and often needed, there are situations where an epilepsy diagnosis can be made without these examinations (e.g. seizure witnessed by an experienced health care professional) as well as there are settings where such resources are unavailable. Could this statement be modified "it may be necessary.." "it is sometimes..." "it is often/generally..."?? Similarly p 6 on Structural etiology. While no one would question that "identification of a subtle structural lesion requires appropriate MRI studies using specific epilepsy protocols." this could perhaps be rephrased slightly to indicate that many structural etiologies can be identified with less sophisticated imaging methods.

3. Page 9 Metabolic causes: Seizure inducing metabolic conditions can be permanent or more transient (uremia is given as an example in the document). For the latter category a short discussion clarifying that a distinction should be made between acute symptomatic/provoked and unprovoked seizures would be useful (c.f. the section on infectious etiology where this distinction is stressed).

4. Page 10 Unclassified epilepsies. Why not add "despite adequate work-up" to the sentence "For a third of the epilepsies the specific diagnosis in terms of syndrome or etiology remains unknown."

5. Terminology: The report suggests substituting the term secondary generalized tonic-clonic with Focal seizure evolving to a bilateral convulsive seizure. Why not simplify this to Focal evolving to generalized seizure"

6. Would it be worth considering adding as an appendix a table with short vignettes/examples and how these would be classified/organized depending on the available level of information?

Torbjörn Tomson

20 December 2013

I'd like to make some succinct suggestions on classification of the epilepsies, which is a project I wholly support.

1. ‘Consciousness’ should not be used as a differentiating factor, as has been the case with simple and complex partial seizures; this is because it is a vague concept which is inadequately assessed even in ideal experimental conditions, let alone clinical situations. If reference is sought to facets of consciousness, let them be discernible features such as response to commands, memory of events, or postural muscle tone. Using anything more than the phenomenology and immediate deductions from that would step perilously into assumptions about an area which we know too little about. 

2. Phenomenological symptom clusters could be employed to guide categorisation of the semiology, and to move away from presumptive statements about lesion location. For instance instead of a ‘frontal lobe picture’ we might instead use a different term not presuming location. The features which are included within each umbrella term should be decided by analysis of clustering of symptoms to derive probabilistically optimised clinical tools. The analysis of clustering could be achieved by meta-analytic synthesis of symptoms followed by computer analysis techniques such as machine learning (specifically K-means algorithm). In this way we will have robust and update-able information from which to practice.

Felix May

18 December 2013

The terms used for classification of etiology in the 1989 classification (“idiopathic,” “cryptogenic,” and “symptomatic”) were problematic, and the Commission is to be congratulated for trying to address this problem.  However, translating the term “idiopathic” into “genetic” is a mistake.  It ignores the complexity in the role of genetics in the epilepsies and is likely to mislead both physicians and patients. 

One of the most pressing concerns of patients given a diagnosis of epilepsy is whether or not it is inherited.  They grossly overestimate the risk of epilepsy in their offspring [1].  Based on the results of well-designed epidemiologic studies, the true risk of epilepsy in offspring of individuals with epilepsy is less than 4% overall, and less than 10% in those with idiopathic generalized epilepsies [2,3].  What will be the impact on patients of incorporating the term “genetic” into the name of the IGEs?  What messages will physicians give to patients when they are taught to use the name “genetic generalized epilepsy”?

The Commission emphasizes that they do not wish to imply that syndromes categorized as “genetic” etiology are exclusively genetic, and describe the IGEs as disorders that “often follow complex inheritance, where they have a polygenic basis with environmental factors possibly playing a role.”  Despite this acknowledgement, incorporating the term “genetic” into the name of these disorders creates the false impression that they are exclusively genetic.

It also creates the false impression that the IGEs are homogeneous in their cause.  One of the most important lessons from genetic research on the epilepsies is that even narrowly defined electroclinical syndromes vary in their genetic causes.  This variation includes not only different genes influencing risk for the same syndrome (e.g., SCN1A, SCN2A, GABRG2 in GEFS+), but also different genetic and nongenetic mechanisms (autosomal dominant, complex, environmental) in the same disorder (e.g., lateral temporal lobe epilepsy caused by dominant mutations in LGI1 in some patients and by traumatic brain injury in others). 

Because of this lack of correspondence between clinical phenotype and genetic (or other causal) mechanisms, designation of a “genetic” etiology should be based on the identification of a specific genetic cause in an individual patient. This is the same approach used with other etiologic categories – e.g., “structural” etiology is defined based on abnormalities detected on neuroimaging in an individual patient.  The term “genetic” should not be used to refer to a group of disorders (the IGEs) that likely vary in the extent and nature of their genetic contributions.

Ruth Ottman

1. Helbig KL, Bernhardt BA, Conway LJ, Valverde KD, Helbig I, Sperling MR. Genetic risk perception and reproductive decision making among people with epilepsy. Epilepsia 2010;51:1874-1877.
2. Ottman R, Annegers JF, Hauser WA, Kurland LT. Seizure risk in offspring of parents with generalized versus partial epilepsy. Epilepsia 1989;30:157-161.
3. Peljto AL, Barker-Cummings C, Vasoli V, Leibson CL, Hauser WA, Buchhalter JR, Ottman R. Familial Risks of Epilepsy:  A Population-Based Study. Brain (in press).

18 December 2013

Sorry to be commenting to late in the process, but as you all know, this is a .hard problem.

First, let me say that I think whatever the final document contains, it should be called a classification. That was the charge, that is what is needed. The fact that every classification is a work in progress is not a good enough reason to use a somewhat misleading word such as 'organization' to salve our insecurity (Even 'separation of the epilepsies'  might be better than 'organization'!).

Early on, I thought that from the purely clinical and EEG point of view that classifying epilepsies as Focal or Non-focal would do quite nicely as a starting point for diagnosis and treatment.  Small vs Large symptoms , and Temporal vs Non-temporal localizations were, and are, also useful dichotomies. Now it seems that a multiscale classification is needed to address the needs of the multiple communities that are concerned with epilepsy diagnosis, treatment, prevention, social impact, legislation and more.

Rather that critique, amend, or supplement the excellent work you have all done so far, I will just send you this classifcation which I have been developing for the past 4 years in the hope that you may find some features that could be helpful in completing your work. It is just one page in length, which might be viewed in a positive light, and it can easily be contracted or expanded to meet the needs of a particular setting or the development of new information bearing on classification. Seizure classification document

With best wished to all for every success in the New Year,
Richard N. Harner

Return to top

18 December 2013

I am a fourth year medical student from Mexico and I have great interest in neurology. During my neurology clerkships in my third year I noticed that despite the new classification  and terminolgy is accesible to neurologists and the general public through the ILAE's website, residents and general neurologists who taught me epilepsy, still taught us the old classification. When I asked them why they wouldn't taught us the new one, they told me that the old one was quite simple and there was no signifcative difference between both of them. I mean, they are aware of the existance of the new classification, and they know it, but still they keep using and teaching the old one. My comment is that something must be done in order to make neurologists to use  the new classification, especiallly when students are taught, in this way from the beginning the clinican will be familiar with the new terminology.

Armando Sanchez Jordan

18 December 2013

Thank you for giving me an opportunity to comment on “The Organization of the Epilepsies: Report of the ILAE Commission on Classification and Terminology”.

Overall I agree the classification based on two domains “electro clinical syndrome” and “etiology”. In the 2010 article by Berg et al “Revised terminology and concepts for organization of seizures and epilepsies: Report of the ILAE Commission on Classification and Terminology, 2005–2009”, a single table was given showing the new proposed classification of epilepsy which conveyed an idea that both these domains (electro clinical syndrome & etiology) are exclusive and hierarchical rather than complementary. In other words it looked as if a particular electro clinical syndrome does not have a clear-cut etiological basis or a specific etiology cannot manifest as an electro clinical syndrome. So it is better to put these domains in two different tables or in a single table with 2 subheadings.

Expansion of etiology classes from 3 to 6 looks more appropriate.

The new definition of focal and generalized seizures require a little more explanation for it to be used for clinical practice. While interpreting the EEG a clinician may have queries like

  1. How rapidly does the seizure have to engage bilaterally distributed networks for it to be considered generalized?
  2. How localized must the onset be and how asymmetric must the seizure be for classifying a seizure as a focal seizure?

Also a note on how much the semiology helps a clinician to differentiate between generalized seizures & focal seizures evolving into convulsive seizures may be added.

I feel that the second table showing the descriptors of focal seizures may be merged with the first table. Addition of phrases like impairment of awareness and responsiveness to the already existing altered consciousness makes the entity more clear and transparent.

With best regards,
Pradeep Pankajakshan Nair

17 December 2013

Table 3. Examples of electroclinical syndromes and other epilepsies arranged according to age@ (modified from Berg et al 2010)*

In this table does every one agree that West Syndrome is a better term than infantile spasms?
As there are children with spasms who have no developmental delay.
Since the age of onset is now known to be up to two years of age should one coin a new term like childhood spasms? or jackknife convulsions or jackknife convulsion syndrome?

Mary Iype

17 December 2013

Overall the revised organization is well laid out with a sound logic. It deals well with the many gray zones that we encounter in practice.

A common issue among many of the comments is a dyslike for the term dyscognitive, which induces a psychic dyssonance not unlike fingernails on a chalkboard, and in some cases may not be entirely appropriate.

Several alternatives have already been offered. I would like to offer another, which combines the old and new

Focal simple: a unilateral, focal seizure that involves a single system (motor, somatosensory, olfactory, visual.....) Auras would typically fit this category

Focal complex: a seizure that has focal, unilateral origin but which either involves more than a single system, or involves bilateral cortices of the same system. The key concept is that there is a focal onset from which the seizure spreads to recruit other regions.

The only other comment, which may or may not be accurate, regards placing Rasmussen's in the structural category. Although there is structural change over time, data suggest an immune mediated/inflammatory response and immunotherapy can have a positive effect. Curious as to why it was not placed together with epilepsies with an immune mediated cause.

Ed Bertram

Return to top

17 December 2013

I propose a new term for classification: Dementia Epileptica. because I have felt for quite some time that the notion of psychosis is too vague, especially for intractable epilepsy. Of course, the term may not fit children with epilepsy but by and large it fits most adults with intractable temporal lobe or extratemporal lobe epilepsies.

Fred C. C. Peng

17 December 2013

Thank you for the opportunity to comment the new report of the ILAE Commission on Classification and Terminology related to the "Organization on Epilepsies".

Overall considered, I agree with the two-level (or two-domains) diagnostic approaches to epilepsy diagnoses: 1. electroclinical syndromes, and 2. Etiologies.

My main concerns refer to the etiological categories. In general, I find misleading to adopt the straightforward term "genetic" to categorize epilepsies where, as Authors state, "in most cases, the underlying genes are not yet known". Especially when applied to "idiopathic generalized epilepsies", which now – sic et simpliciter – have become "genetic generalized epilepsies" (without any real advance in their knowledge of their physiopathology to justify such a new denomination), the term "genetic" sounds too straightforward, and may even lead to misunderstandings with the patients and their relatives (who may not easily understand the difference between "inherited" (e.g., Mendelian inheritance), and "genetic").

Furthermore, it is not clear to me why in the etiologically based diagnoses, some etiologies deserved more consideration than others. If Authors decided to adopt an analytic approach by introducing for instance the category "infectious", I find no reason why they should not have introduced also other etiological categories (e.g., "traumatic", "hypoxic-ischemic", "vascular").

Francesco Brigo

17 December 2013

Thank you for asking for comments and congratulation for the interesting proposal on a new organization of epilepsies.

As already stated by many other commentators the term "dyskognitive" seems to make things still more complicated. In my opinion the Phrase "seizures without impairment of consciousness" is clear and easily understandable.

With best regards
Y. Menedo

17 December 2013

Congratulations on this thoughtful paper.

I think the proposed classification system is all in all a systematic one, which can be reasonable well. Perhaps the only comment I have is to consider / define Status epilepticus.

With best greetings,
Rudiger Köhling

Return to top

17 December 2013

With respect to the reorganisation of epilepsies, 2013, the introductory phrases '....It is important to consider epilepsy diagnosis at several different categorical levels to distinguish the manifestations from the underlying cause. Seizures are a separate category to epilepsies and epilepsies are a separate category to etiologies....' fit perfectly in the revamping of epilepsy classifications because these principles eliminate the mixing up and confusion that was present in the past between manifestations, epilepsy type and originating cause.

Sghendo Lino Joseph

16 December 2013

First I'd like to congratulate the committee on excellent continued progress. Classification is important work and always challenging.

My one comment echoes several others below against use of the term "dyscognitive."

This term is problematic for several reasons:

1. Impaired cognition does not necessarily mean that consciousness is impaired.
Patients with focal cognitive deficits (e.g. aphasia, visual-spatial deficits, etc.) may retain relatively normal alertness and responsiveness. Therefore use of "dyscognitive" as a substitute for complex partial seizures is confusing.

2. Focal seizures with impaired consciousness typically cause relatively global deficits. 
A classic error in behavioral neurology is to misdiagnose a patient with global encephalopathy as having a focal cognitive deficit such as agraphia, while in fact the main problem is in overall attention and arousal.  Recent work suggests that most focal seizures either globally impair or globally spare diverse functions (e.g. see Blumenfeld and Jackson, Epilepsia 56 (6): 1125).  Level of consciousness is the core dysfunction and should be expressed in the terms used.

3. The term "dyscognitive" is arcane sounding

We recently suggested the alternative "Focal Impaired Consciousness Seizure" (FICS) instead of "Focal dyscognitive seizure" (Blumenfeld and Jackson, Epilepsia 56 (6): 1125).  This is consistent with the new ILAE classification but takes the concerns above into account.

Hal Blumenfeld

16 December 2013

First of all, thanks for this report. Paul Valery from the French Academy said “everything simple is false and everything complex is unusable.” That is very true for Organization of the Epilepsies. This proposed organization has the advantage to be clear and usable for daily use.
Some comments, however,

  1. The term “dyscognitive” seems to be unclear even as a shorthand and “focal seizure with impaired consciousness” should be used instead
  2. The term generalized epilepsy remains very useful in particular for Genetic Generalized Epilepsy as it means specific treatment and prognostic - but in popular imagination “genetic” is “inherited” and the fundamental difference must be clearly explain to patients
  3. Concerning the etiologically based diagnoses the most important term, in my point of view, is “unknown” which reminds us the limits of our knowledge and the substantial importance of challenging ourselves to search and find the aetiology of an epilepsy via science progress and new diagnostic resources available.

Best regards,
Jerome Aupy

16 December 2013

A reasonably satisfactory classification.


  1. I would like to add further subdivision in focal epilepsy as temporal and extra-temporal . The purpose is to give importance to temporal lobe epilepsy considering its incidence and relatively refractoriness to treatment.
  2. A semiological subdivision in focal seizures can be focal motor and focal non-motor seizures as extratemporal seizures are more likely to have motor component and  temporal lobe seizures are more likely to be without any major motor manifestations.  

With sincere regards,
Atma Ram Bansal

16 December 2013

Dear authors,

Thank you for the proposal on a new organization of epilepsies.

In my view, the term “dyskognitive” is an acceptable term in principle. But it is difficult to use it in the daily work since the alternative “seizures without impairment of consciousness or awareness” is too long and complicated to use it every day. Therefore, I would propose not to change the widely accepted terms “complex-partial” versus “focal partial”.

Yvonne Weber

10 December 2013

I have two comments.

1. How can we assume that all IGEs are GGEs? I think the term IGE should be maintained.

2. The term ‘dyscognitive’ is awful. It’s a nails on the chalk board term. It is more opaque than complex partial seizure. We seem to have entered the realm of upper-level-managerial-non-transparency speak. There is nothing wrong with complex partial seizure.

Chris Plummer

Return to top

6 December, 2013

We very much appreciate this proposal for organization of epilepsies.

Our major concerns regard the following points:

Table 3.
— Childhood.
Although authors prefer to use “self-limited” instead of  “benign” (and we totally agree), it is not clear why only “Epilepsy with centro-temporal spikes” should today be considered as “selflimited”. If we introduce this new term, we strongly suggest to apply “self-limited” also to “Early onset childhood occipital epilepsy (Panayiotopoulos type)”, even if benign the term benign was not previously assigned to such condition.

— Familial epilepsy syndrome. In this organization according to age, the paragraph “Familial epilepsy syndromes”, sound out of context. We think that the paragraph “Less specific age-relationship” present in 2010 proposal, and including “familial epilepsy syndromes”, is by far more pertinent in this table.

Etiologically based diagnoses.
5) Infectious; according to authors such etiology “carries specific treatment implications”. However, this is true in particular in the first days or week following infections, when patients may present acute symptomatic seizures. Thereafter, patients may develop a true epilepsy (unprovoked seizures) whose management is the same as compared to other structural epilepsies. Hence, we strongly suggest to delete this chapter and to include infectious in the same paragraph of structural epilepsy.

Best regards,
Umberto Aguglia and Edoardo Ferlazzo

5 December, 2013

Introduction: generalised and focal seizures.
Although some seizures with focal onset in tuberous sclerosis may be categorised as “generalised” quite often a carefully taken history might decide that it is focal.  I have recently done something similar suggesting that many febrile seizures are focal.  One of the main factors is the way in which the early history is taken and this  is likely to vary between histories taken within a unit and e.g, on wider epidemiology studies.  All cysticercosis and really all invasions of the nervous system should be labelled as focal although some will have an apparently generalised onset.

This group also pointed out that some have both general and focal features and are thus not classifiable by this system and therefore they are called a “more complex phenotype”.   These include Dravets, West and Lennox-Gastaut.  I would prefer to call these complex focal with essential generalisation.  These are all epileptic encephalopathies and it may well be that these foci have to involve both hemispheres to have an encephalopathic effect.  In a sense therefore  you have focality taking precedence over generalised seizures.  It also does not deal with the issue of lesional and non-lesional causes of  West  and Lennox-Gastaut syndromes. 
I agree with what you have said about:

A)  Electro-clinical syndromes.
I accept the expansion of (B) Aetiologically based diagnoses to 6 items and it is understood that these can be multiple.

  1. Genetic: I would leave most idiopathic generalised epilepsies until one is clearer about these being generalised genetic epilepsies.
    The likely acquired diplegia in Dravets Syndrome from 10 years seems as though it must be a metabolic effect though the mechanism is not yet clear and it is not clear whether those without SCNIA have these problems.
  2. Structural – the issue of tuberous sclerosis where lesions produce a metabolic effect related to seizures and it is of course genetic.  So long as one realises this and is prepared to triple list this disorder that is fine but it would usually be referred to as T.S.
    The mixed nature of polymicrogyria is managed well, except that the majority of children with polymicrogyria are of unknown cause.
  3. Metabolic (no comment)
  4. Immune (no comment)
  5. Infectious (no comment)
  6. Unknown (no comment)

Application of multiple aetiologies to one patient.

C) Unclasified epilepsies – future directions in epilepsy classification

My main concern therefore is over the issue of focal vs generalised seizures.

Brian Neville

1.  The organisation of the epilepsies.  Report of the ILAE Commission on Classification and Terminology.  Scheffer et al
2.  Neville B, Gindner D.  Febrile seizures are a syndrome of secondarily generalised hippocampal  epilepsy.  Dev.Med. Child Neurol 2010 Dec52(12):1151-3

Return to top

1 December, 2013

I read the proposed organization and I think it is clear. I have no further suggestions.

Best regards,
Amira Masri

1 December, 2013

Overall this is a useful document which intelligently moves forward the words we use to discuss epileptic seizures and epilepsies.

I was, however, dismayed to see that Panayiotopoulos syndrome is once again incorrectly listed as an occipital epilepsy. There is little evidence to suggest that this is the case so why propagate the error?

The term Panayiotopoulos syndrome is used widely throughout the world and has the great advantage that it discourages clinicians failing to diagnose it because the EEG fails to show occipital epileptiform abnormalities or shows extra-occipital or even generalised abnormalities.

Colin Ferrie

26 November, 2013

In Table 2, one type of focal seizure is described as "Evolving to a bilateral convulsive seizure • May include tonic, clonic or tonic and clonic components in any order"

The present ILAE classification does not account for focal seizures which generalize to atonic seizure, a phenomenon which does occur.

With kindest regards,
Kim Meador

26 November, 2013

The word "organization" while conceptually accurate, is confusing here. The purpose of the proposed changes is to develop a new classification to replace previous classifications. If this is not also called a "classification" it will appear to be something that can be used in parallel with but not necessarily in place of the previous classification system. I would label it what it is intended to be.

This document, like the 2010 publication, outlines the arguments underlying the proposed classification and terminology changes and their meaning well. However, more case examples showing the specific final diagnoses rendered through use of the newly proposed classification system would be helpful. I found myself in agreement with many concepts, but at the end was not sure exactly what I was going to list as a diagnosis in a clinical note for example of an actual patient using this system.

I agree with an earlier comment that provision of a "walk through" between final diagnoses made by the proposed classification revision and applicable ICD-9&10 codes would be helpful.  If this is going to be clinically used, coordination between it and the diagnostic code systems required for daily practice will ultimately be needed. The current ICD system uses terms from the 1989 Classification, so epilepsy care providers will continue to need to know both languages for a period of time.

- Jeff Britton

Return to top

25 November, 2013

I read the manuscript. The only things I would add, more like a wish, is that it be accompanied by ICD-10 codes.

Maria C Ramirez-Nieto

24 November, 2013

I praise the authors for their bravery. Changing classifications is like knocking down a nest of paper wasps with a baseball bat. You get stung or you damage the surroundings or both.  I would suggest the following adjustments to the classification:

The term ”Organization of the Epilepsies” is an oxymoron-Is not epilepsy the temporary disorganization of cerebral function?  You cannot logically have organization of disorganization! Therefore retain “Classification of the Epilepsies”

Retain with “cognitive changes”, avoid  “dyscongnitive”. Dys-, a prefix is Greek for “bad, disordered, or difficult” vide Dorland’s Pocket Medical Dictionary, 24 th Ed, p 192.  Cognitive is derived from Latin noun, feminine gender,  “cognitio” –knowledge, awareness recognition. “Dyscognitive” is a mongrel word, Greek mixed with Latin, and a neologism. I would suggest “with impaired thought” or just “impaired cognition”. The classification must include epilepsies with mixed  or multiple etiologies, such as MERRF which has genetic etiologies and myoclonic manifestations. In short the classification should reflect new discoveries in genetics and electrophysiology.

To the brave souls who are writing  the new classification-you will NOT make all people happy, just compose a logical classification that will satisfy 75% of your potential critics.

Edward J. Fine

22 November, 2013

I would like to draw your attention that the previous ILAE Commission on their final report had accepted my evidenced based comments and proposed “Panayiotopoulos syndrome” as the valid nomenclature (see Table 3 of their report) which is the used term in all recent publications. Therefore, unless there is an oversight, I cannot understand that this new report reverts in their Table 3 to an obsolete and erroneous terminology of the dual eponymic and descriptive term “early onset childhood occipital epilepsy (Panayiotopoulos type)”. This is against scientific evidence, a step backwards and confronts with the intention of consistency and accuracy.

Let me please remind you my previous comments.

I refer to the syndrome of “Early onset benign childhood occipital epilepsy (Panayiotopoulos type)” that the Commission maintains with a dual descriptive and eponymic nomenclature.  On two previous occasions in Epilepsia (1,2), I had made an evidence based proposal that the ILAE Commission discards the descriptive “Early onset benign childhood occipital epilepsy” keeping only “Panayiotopoulos syndrome” as they did for Dravet syndrome (discarding severe myoclonic epilepsy in infancy), Ohtahara syndrome (instead of early infantile epileptic encephalopathy with suppression-burst) and others.

“Early onset benign childhood occipital epilepsy” is not only pleonastic but also incorrect and misleading as documented from immense converging evidence from clinical, ictal and interictal EEG, magnetoenechephalography and other reports:2

  • Historically, Panayiotopoulos in his original study described two subsets of patients: 12 with EEG occipital spikes, hence the name “early onset benign childhood epilepsy with occipital paroxysms” and 9 with extraoccipital spikes or normal EEG hence the name “benign extraoccipital childhood partial seizures with ictal vomiting and excellent prognosis”. These two subsets constitute of what is now known as Panayiotopoulos syndrome.
  • Clinically in PS, onset of seizures is mainly with autonomic symptoms and particularly emesis (80%). These are not a feature of other occipital epilepsies, which otherwise the ILAE descriptive name would imply. Clinical manifestations in PS are the same irrespective of ictal or interictal EEG findings.
  • Interictal EEGs in PS mainly manifest with multifocal spikes at various locations though occipital spikes often (70%) predominate.
  • On gold standards, ictal video-EEGs have unequivocally documented anterior or posterior onsets of the seizure discharges despite similar ictal clinical manifestations.
  • Magnetoencephalography have confirmed spike localisations along the rolandic as well as the parietooccipital and calcarine fissure.
  • PS has more common features with rolandic epilepsy than the Gastaut type occipital epilepsy which is a truly pure form of rare idiopathic childhood occipital epilepsy.

Zarko Martinovic


  1. Martinovic Z. (2002) Panayiotopoulos Syndrome or Early-onset Benign Childhood Occipital Epilepsy. Epilepsia 43:1268-72.
  2. Martinovic Z. (2007) The new ILAE Report on Classification and Evidence-Based Commentary on Panayiotopoulos Syndrome and Autonomic Status Epilepticus. Epilepsia 48:1215-6.

Return to top

19 November, 2013

Berg et al. ´s (2010) contribution represents a major progress in terminology as well as in organization of epileptic seizures and syndromes. The presented manuscript (Scheffer et al.) adds to conceptional and practical clarity. 

For practical reasons, the reintroduction of “focal seizures with impaired consciousness or awareness or responsiveness” is to be greeted. However, the “shorthand” alternative “dyscognitive” is problematic. Impaired consciousness relates here to impaired quantitative consciousness. If disturbances of qualitative consciousness (contents of consciousness) are included, every seizure type has to be classified under this heading. “Dyscognitive” – at least for German speaking colleagues – includes every cognitive dysfunction like slowing, forced thinking, hallucinations etc.  Impaired quantitative consciousness might be described as twilight, dreamy or really unconscious state. See also Jackson´s term or the old German term “Dämmerattacke” (Meyer-Mickeleit). What about “Focal twilight seizure”?

Under focal motor seizures a category of hypermotor seizures could be added.

The reintroduction of age-dependant electro-clinical syndromes may solve many misunderstandings and confusions, especially with non-specialized neurologists. It would be helpful to delineate the term with respect to genetic and symptomatic generalized epilepsies. Why are several syndromes clearly characterized by clinical and electrical features excluded? Are electro-clinical syndromes used as a synonym for genetic as well as age-dependant epilepsies?

The discussion about etiologically based diagnoses reflects the progress in facts. To replace the term “idiopathic generalized” by “genetic generalized” represents a logic relief. What about genetic focal epilepsies?

No attempt is made to provide a list of established symptomatic focal syndromes according to the results of epilepsy surgery - Mesial TLE, Neocortical TLE , multifocal epilepsies ….

After all, congratulations.

G. Bauer

17 November, 2013

Many thanks for this. I support the direction and the changes of the proposed organisation.

I have a few comments:

  1. With regard to the statement "For all patients we should aim to use a diagnostic framework that includes identification of seizure types, syndrome and etiology"; I would add co-morbidities.
  2. With regard to the comment "In order to make an epilepsy diagnosis, in addition to a careful clinical evaluation, it is necessary to have access to EEG and MRI studies where appropriate." I would change this to "In order to make a full epilepsy diagnosis..." in order to reflect that not all health settings have access to EEG/MRI and it would be shame to suggest that clinical evaluation alone, albeit limited, cannot make a pragmatic diagnosis without these MRI/EEG modalities
  3. I also think that we still need to emphasise a multi-axial formulation, the process of mapping the individual to the emerging 'organisation'. The etiology based diagnosis and syndrome based diagnoses are both relevant for each individual and i consider both part of a wider diagnostic framework/formulation. It is worth for most individuals, stating the presence or absence of an electroclinical syndrome and the presence and absence of an etiology.

I would therefore favour continuing emphasis on full multiaxial diagnosis/formulation for each individual.

  • description of episodes?
  • is it epilepsy?
  • what are the seizure types?
  • is there an identifiable electroclinical syndrome?
  • are there identifiable causes (structural, genetic, metabolic, infectious, immune, unknown)
  • what are the relevant co-morbidities

We have being using this approach within the BPNA PET courses and it seems a pragmatic consistent way to connect the evolving individual to an evolving taxonomy!

Many thanks,
Colin Dunkley

Return to top

15 November 2013

In order to provide a complete classification of seizures, I suggest that the commission consider the definition and types of status epilepticus.

Michael Rogawski

15 November 2013

I congratulate the ILAE Commission on Classification and Terminology for their important expansion of the organization of the epilepsies to include a more detailed discussion of the etiologies; however, the report itself is confusing. It moves from the concept of "classification" to "organization" and then notes that this is intended primarily for clinical purposes, but never discusses the critical difference between a classification and a diagnostic manual. What is being discussed here, for the benefit of clinical application, is how physicians should describe individual patients with epilepsy, not how epilepsy should be classified or organized. The report would greatly benefit from a clear delineation of the important difference between classification and description, and considerable rewriting is necessary to point out when the following discussion is referring to classification or organization, and when it is referring to how physicians should describe their individual patients.

An example of the confusion is the paragraph on page 4 on seizure types, which implies that the term "complex partial seizure," which was a seizure type in the 1981 classification, has been replaced by the term "focal dyscognitive seizure," implying that the latter is also a seizure type for classification purposes. The intention of the 2010 report is to use "dyscognitive" as a way of describing a seizure experienced by an individual patient for clinical purposes, while emphasizing that this does not represent a class of seizures that in any way would be a unique diagnostic entity with implications for etiology, prognosis, or treatment. It merely means that in this patient there is an associated risk that would not exist if consciousness were not impaired. Also, the term "dyscognitive" is not new with the 2010 classification – it was the official ILAE term since the publication of the glossary in 2001.

Similarly, the paragraph at the bottom of the page seems to imply that there are generalized epilepsies and focal epilepsies, while the 2010 report referred only to epilepsies with generalized seizures and epilepsies with focal seizures when this was appropriate, but declined to perpetuate the previous false concept that suggested that epilepsies themselves could be called generalized or focal.

With respect to the important discussion of the etiologically based diagnosis (again for a diagnostic manual, not a classification), the last sentence of the last paragraph on page 8, "Where structural etiology has a well-defined genetic basis such as tuberous sclerosis complex, which is caused by mutations in the genes tuberin or hamartin, both terms, structural and genetic can be used," seems to contradict the definition of "genetic" on page 7: "The concept of genetic relates to where the epilepsy is, to the best of our knowledge, the direct result of a known or presumed genetic defect and where seizures are the core symptom of the disorder." Perhaps it would be better to clarify that there are "genetic structural" causes, as opposed to "primary genetic" causes.

The fourth paragraph on page 9 implies that when the etiology is unknown it is appropriate to make a diagnosis of "frontal lobe epilepsy" when the ILAE has said repeatedly in the past that there is no such diagnostic entity, and that epilepsies cannot be diagnosed merely by the anatomical location of the presumed site of ictal onset. I hope it is not the intention of this report to return to these old, unhelpful concepts.

A minor criticism is the use of the term "antiepileptic drug" several times in this report. I thought the ILAE had agreed that the correct term is "antiseizure drug."

Jerome (Pete) Engel, Jr.

15 November 2013

I think that this is a good advance. It would be helpful if the therapeutic implications (especially of the new definitions of focal and generalized seizures), as they are currently understood, would be outlined in an appendix since the distinction between the two seems less clearcut than it was in the past.

William Nowak

Read more recent comments here.

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.