Controversial Issues The following commentaries on the Shorvon and Berg/Scheffer articles in this issue, and the responses from Shorvon and Berg/Scheffer, were invited by the Editors-in-Chief
Responses to Commentaries
What is at stake in a classification?
Article first published online: 14 JUN 2011
© 2011 International League Against Epilepsy
Volume 52, Issue 6, pages 1205–1208, June 2011
How to Cite
Berg, A. T. and Scheffer, I. E. (2011), What is at stake in a classification?. Epilepsia, 52: 1205–1208. doi: 10.1111/j.1528-1167.2011.03110.x
- Issue published online: 14 JUN 2011
- Article first published online: 14 JUN 2011
When I use a word … It means just what I choose it to mean Humpty Dumpty
Lewis Carroll (1832–1898)
We agree with Shorvon (2011) that cause is a major dimension by which the epilepsies should be viewed. In fact, this view has been a central tenet of classifications of epilepsy going back to those originally proposed by Gastaut, (1969a,b) and even earlier. It is also a major focus of the recent report from the ILAE’s Commission on Classification and Terminology (Berg et al., 2010). It is thus surprising to hear Shorvon’s assessment that the causes of epilepsy have been largely ignored in classification efforts. Shorvon apparently fails to acknowledge that the epileptologists who developed previous classifications carefully considered etiology (cause) as evidenced by the words we (and he) choose to use today. In the seminal 1989 classification (Commission, 1989), the concept of epileptic syndromes and their classification according to cause was formally established by the use of the terms idiopathic, cryptogenic, and symptomatic. Although these terms were useful in the past when our understanding of the causes of epilepsy was limited, their continued use detracts from our ability to communicate precisely and effectively. Consequently, the Commission on Classification and Terminology of the International League Against Epilepsy (ILAE) proposed new terminology and organizing concepts to reflect current knowledge and understandings afforded us by advances in molecular genetics, neuroimaging, and neurophysiologic techniques and to allow flexibility for inclusion of new insights gained through future advances (Berg et al., 2010). As pointed out by Moshé (2011), the Commission report was made widely available for comment prior to submission to Epilepsia in 2009, and we carefully considered the many suggestions we received—including Shorvon’s, which consisted largely of his current commentary and the accompanying table.
Shorvon’s proposal raises many concerns. For one, it fails to reflect the fact that electroclinical syndromes are often not associated with a single cause (e.g., Osborne et al., 2010). It does not advance matters to classify, for example, West syndrome—a disorder due to a large number of diverse causes—as “symptomatic.” The converse is true too; a specific causal factor is not necessarily associated with a single clinical expression. In fact, modern genetic studies have taught us that the old model of one gene–one disease has been over turned. Particularly well-developed examples of this principle in our field can be found with SLC2A1, the gene that codes for the GLUT-1 transporter in the brain (Suls et al., 2008, 2009; Mullen et al., 2010) and SCN1A, the gene that codes for the alpha1 subunit of the neuronal voltage-gated sodium channel (Mulley et al., 2005; Vahedi et al., 2009; Zuberi et al., 2011). Shorvon’s table also creates confusion in identifying clinical syndromes as causes (e.g., West syndrome as a cause of epilepsy).
Another concern addressed by our colleagues in their commentaries is the creation of a new etiological category for “provoked epilepsy,” which seems to combine acute provoked seizures (defined in the Epidemiology Commission’s report—Beghi et al., 2010) and reflex epilepsies, which themselves are due to a multitude of different causes. As all of the other commentaries point out, this category is not helpful and ignores the fundamental distinction between seizures and epilepsies.
There is much to be done in developing scientifically sound as well as clinically meaningful classifications of the epilepsies and their causes. We do not find that Shorvon’s proposal helps to overcome the deficiencies of past approaches. Furthermore, it does not appear to incorporate the tremendous advances in the neurosciences that have taken place in the last decade and that continue to occur. Responses to his proposal by others touch on several important themes that need further discussion.
- 1There remains confusion concerning the distinct concepts of a classification (a system for organizing knowledge, e.g., the periodic table of the elements) and an element that is classified within the organizational framework (e.g., sodium, or in our case, a specific electroclinical syndrome such as childhood absence epilepsy). It is here we are concerned that Beghi (2011) confuses one with the other when he refers to the boundaries and criteria for specific diagnostic entities. We are entirely sympathetic to these concerns; however, this is not the primary issue at stake in developing a rational classification of the epilepsies.
- 2The purposes and qualities of a useful classification were reflected throughout the various commentaries. Jackson (2011) explicitly states key qualities in his insightful comparisons to recent efforts in classification of malformations of cortical development. Classifications should teach; they should provide a framework that reflects the inherent structure in our scientific understanding. But at what level(s) and for what purpose(s)? There are many purposes for classifying in epilepsy, but ultimately, the classification of the epilepsies should serve the purpose of facilitating diagnosis and enhancing clinical care. We believe classifications in epilepsy will do that best by organizing salient clinical elements needed for teaching and directing diagnosis, along with features reflecting underlying causes and mechanisms. Further incorporation of information regarding treatment (although this is always in flux and varies from one country to another), seizure prognosis, and even information regarding, for example, comorbidities could render a classification useful not only for treatment and management but for counseling and anticipatory guidance purposes. This sort of approach has many similarities to the goals of the five-axis proposal made 10 years ago (Engel, 2001).
Wong (2011) suggests a fixed, two-tiered system, with clinical symptoms and syndromes at the first level and their cause(s) at the next. Using the classification to catalog causes associated with specific electroclinical presentations is central to our own vision of the new organization. Therefore, we certainly endorse the suggestion of Wong (2011) that, for example, one might have juvenile myoclonic epilepsy (JME) in association with a GABRA1 mutation, a malformation of cortical development, or of no known cause. Others have reported such an approach in the context of West syndrome (Osborne, 2010) and benign familial infantile epilepsy (Mulley et al., 2011). We depart somewhat from the rigidity of Wong’s approach, however, in that we think a flexible organization will allow clinicians or researchers to organize the epilepsies according to their specific needs and purposes. For example, knowing that SCN1A mutations are the most common cause of Dravet syndrome does not help one recognize that syndrome and may actually lead to an incorrect diagnosis (e.g., GEFS+ vs. Dravet syndrome). In contrast, knowing that the typical age at onset is in the first year of life, that it frequently first presents as a hemiconvulsive status in the context of a febrile illness, and that the child was developing normally up until that time immediately raises suspicion of Dravet syndrome. At that point, knowing about the role of SCN1A can then guide genetic investigations. In fact, in a clinical diagnostic setting, one generally begins with clinical presentation, not the specific diagnosis (syndrome), and then develops a clinical diagnosis. This process, in turn, helps to focus investigations on a limited number of causes. We had earlier given an example of organization by age of onset which, for electroclinical syndromes, is of substantial clinical value in guiding diagnosis (Berg et al., 2010). Others may be interested in underlying causes, specific electroencephalography (EEG) features, the occurrence of specific seizure types, and so forth.
In this context, we would also disagree with Duncan (2011) regarding the retention of the terms simple and complex partial. They do not reflect any inherent structure in our knowledge, have been the source for considerable imprecision in diagnosis of ictal phenomenology, and have been the bane of teaching medical students over many years. Their use teaches the wrong lesson. We do, however, acknowledge and agree with his point regarding the use of the term etiology meaning the study of causation rather than being a synonym for cause.
- 3Most of our colleagues seem ready to accept that the false dichotomy of idiopathic–symptomatic is inadequate for representing our growing understanding of the causes and mechanisms of epilepsy. We find the comment of Beghi (2011) about the need to test the validity and reliability difficult to understand in this context. Does he mean that we must validate the concepts of channelopathy or of focal cortical dysplasia? As for reliability, it is not a test of validity. One can be reliably wrong. Unfortunately, we can all agree to something and be wrong. We hear continuously about the need to improve epilepsy care. How are we to do that if we cannot advance our understanding and communication about the disorders that constitute epilepsy? Modern epileptology requires that epileptologists keep learning—and the learning curve has been steep in recent years. It will likely take a generational change to make a complete shift and overcome the entrenched desire to hold on to a cherished but antiquated system and terminology.
Shorvon (2011) refers to widespread disagreement expressed by the individuals who were invited by the editors-in-chief to comment on the Commission’s 2010 report. It was surprising to see the consternation expressed by some over this proposal, as the flexibility we advocate in fact allows one to recreate a version of the 1989 classification scheme, the scheme proposed by Wong, or virtually any organization that best fits one’s purpose. Articles written by authors in other journals as well as in Epilepsia who have taken the time to understand and apply these concepts suggest that the report made points that are appreciated by many clinical and basic neuroscientists engaged in moving the field of epilepsy forward.
Wolf (2011) also expresses surprise that we did not respond to the six commentaries solicited by the editors-in-chief and published when the classification came out in 2010 (Berg & Scheffer, 2011). We were not invited to comment at the time, although most of the objections had already been addressed within the report. Wolf was a member of the Classification Commission and participated in discussions until the Monreale meeting, which he references and which was attended by four other Commission members (Capovilla et al., 2009). Although the Monreale group has no official status and cannot dictate the content of League reports, the meeting’s proceedings were very helpful and influential in the Commission’s deliberations (as acknowledged in the 2010 report). There were, however, points that were not adequately developed and others with which the Commission disagreed. Following the Monreale meeting, Wolf elected not to participate in further Commission deliberations and did not respond to multiple requests for feedback on drafts of the report that were sent to him. For this reason, he was ultimately not included as an author.
We appreciate the invitation to respond to the current set of commentaries. We certainly acknowledge that there is much to do. Hiding in the recesses of archaic concepts and approaches with the excuse of honoring history does nothing to honor the achievements of the great thinkers of the past, and it does not serve the scientists, physicians, and patients of the present and future. We hope that those interested in these issues will carefully read the original report (Berg et al., 2010) and contribute to future discussions.
We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
- 2011) New classification proposals for epilepsy: A real advancement in the nosography of the disease? Epilepsia 52:1197–1198. . (
- 2010) Recommendation for a definition of acute symptomatic seizure. Epilepsia 51:671–675. , , , , , , , . (
- 2011) New concepts in classification of the epilepsies: Entering the 21st century. Epilepsia 52:1058–1062. , . (
- 2010) Revised terminology and concepts for organization of seizures and epilepsies: report of the ILAE Commission on Classification and Terminology, 2005–2009. Epilepsia 51:676–685. , , , , , , , , , , , , , . (
- 2009) Conceptual dichotomies in classifying epilepsies: partial versus generalized and idiopathic versus symptomatic (April 18–20th, 2008, Monreale, Italy). Epilepsia 50:1645–1649. , , , , , , . (
- Commission on Classification and Terminology of the International League Against Epilepsy. (1989) Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 30:389–399.
- 2011) The evolving classification of seizures and epilepsies. Epilepsia 52:1204–1205. . (
- 2001) A proposed diagnostic scheme for people with epileptic seizures and with epilepsy: report of the ILAE Task Force on Classification and Terminology. Epilepsia 42:796–803. . (
- 1969a) Classification of the epilepsies: proposal for an international classification. Epilepsia 10:S14–S21. . (
- 1969b) Clinical and electrographical classification of seizures. Epilepsia 10:S2–S13. . (
- 2011) Commentary. Epilepsia 52:1203. . (
- 2011) In support of the ILAE Commission on classification proposal. Epilepsia 52:1200–1201. . (
- 2010) Absence epilepsies with widely variable onset are a key feature of familial GLUT1 deficiency. Neurology 75:432–440. , , , , . (
- 2005) SCN1A mutations and epilepsy. Hum Mutat 25:535–542. , , , , , . (
- 2011) Proposed genetic classification of the “benign” familial neonatal and infantile epilepsies. Epilepsia 52:649–650. , , . (
- 2010) The underlying etiology of infantile spasms (West syndrome): information from the United Kingdom Infantile Spasms Study (UKISS) on contemporary causes and their classification. Epilepsia 51:2168–2174. , , , , , , , , . (
- 2011) The etiologic classification of epilepsy. Epilepsia 52:1052–1057. . (
- 2008) Paroxysmal exercise-induced dyskinesia and epilepsy is due to mutations in SLC2A1, encoding the glucose transporter GLUT1. Brain 131:1831–1844. , , , , , , , , , , , , , , , , , , , , , , , , , , . (
- 2009) Early-onset absence epilepsy caused by mutations in the glucose transporter GLUT1. Ann Neurol 66:415–419. , , , , , , , , , , , , , , . (
- 2009) Elicited repetitive daily blindness: a new phenotype associated with hemiplegic migraine and SCN1A mutations. Neurology 72:1178–1183. , , , , , , , , , , . (
- 2011) Networks and systems, and conceptualizations and research. Epilepsia 52:1198–1200. . (
- 2011) Epilepsy is both a symptom and a disease: a proposal for a two-tiered classification system. Epilepsia 52:1201–1203. . (
- 2011) Genotype-phenotype associations in SCN1A-related epilepsies. Neurology 76:594–600. , , , , , . (