Networks and systems, conceptualizations, and research


  • 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

When the International League Against Epilepsy (ILAE) Commission on Classification and Terminology of 1981–1989 (chaired by this commentator and by Dr. Roger) proposed an international syndrome classification (Commission, 1989), the part about symptomatic localization-related epilepsies remained unfinished as it comprised “syndromes of great individual variability which are based mainly on seizure types and other clinical features as well as anatomic localization and etiology—as far as these are known.”“Seizure types” referred to the 1981 international seizure classification, and a set of tentative descriptions according to anatomic localization was included in this classification, but a supplementary classification according to etiologies was left for future commissions. Only now, 22 years later, Dr. Shorvon has made a proposal based on etiologies (Shorvon, 2011), also giving some substance to the frequently expressed opinion that we need a “flexible” and “multidimensional” classification. This etiologic classification scheme is a laudable endeavor, although some aspects appear debatable. For example, Shorvon’s category of “provoked epilepsy,” although certainly useful, seems to belong to a different level of classification than the other categories, since it considers causative factors of seizures rather than etiologies of the epilepsies. On the whole, this proposal would have profited from a more consequent distinction between seizures and epilepsies, for example, in its critique of “acute symptomatic epilepsy,” a term that the international classifications have carefully avoided in order to underscore that acute symptomatic seizures (Hauser et al., 2010) indeed are not epilepsies.

For Shorvon, the genetic epilepsies belong to a classification according to etiologies. In contrast, “genetic epilepsies” seems not to be a useful category within a taxonomic system as it does not distinguish between idiopathic and symptomatic; as a consequence, in the report of the 2005–2009 Commission (Berg et al., 2010), the epilepsies with well-established genetic etiologies but gross structural abnormalities (like tuberous sclerosis or cortical dysplasia) are not classified as genetic epilepsies, which is awkward.

One wonders how Drs. Berg and Scheffer, chairpersons of the past and present Classification Commissions, would respond to the six invited comments published together with the 2010 Commission Report (Avanzini, 2010; Berg et al., 2010; Ferrie, 2010; Fisher, 2010; Guerrini, 2010; Shinnar, 2010; Wolf, 2010), or to the many publications on ictogenesis that have come out since the Commission Report was published (e.g., Moeller et al., 2009a,b; Salek-Haddadi et al., 2009; Vaudano et al., 2009; Betting et al., 2010; Landvogt et al., 2010; Moeller et al., 2010; Wilke et al., 2010)? Surprisingly, responses are missing in their commentary in this issue of Epilepsia (Berg & Scheffer, 2011). Rather, they repeat, without new arguments, core statements from the Commission Report. They do make it clearer, however, that they consider the 2010 Commission Report to reflect only a transitional state preceding the development of a new classification, and hesitatingly admit to the criticism that their proposal to change terms does not reflect changes of concepts. But why should terms be changed for unchanged concepts? Why not, instead, propose changes for the terms generalized and focal where the concepts are profoundly changed? The absence of a public discussion on these points is disappointing.

Of more concern is the fact that the 2010 Commission Report does not reference new research findings on which to base its recommendations. The 1981 seizure classification was based upon evaluation of all video-recorded seizures internationally available at the time. The 1989 syndrome classification was based upon critical literature reviews of all syndromes that had been proposed (Roger et al., 1985). In contrast, the 2010 Commission Report is based upon “conceptualizations” resulting from the Commission’s internal discussions, an arbitrary process that is acceptable for a discussion paper, but that needs to be followed up by something more solid. That the Commission is fascinated by new methods and possibilities of research is not sufficient. A new classification will not automatically result from new research. To understand its consequences for the nosology of the epilepsies requires meta-analysis and interpretation.

The only novel feature in the 2010 Commission Report is the recognition that all ictogenesis happens in neuronal networks, and that these are different for “focal” and “generalized” seizures. According to the Report, there are mainly two differences: (1) Generalized networks are bilaterally distributed and focal networks are limited to one hemisphere. But if the latter is correct at all, it applies to focal ictogenesis only in symptomatic focal epilepsies, not in idiopathic localization-related epilepsies (Salek-Haddadi et al., 2009). (2) The onset of focal seizures is consistent from one seizure to the next; if generalized seizure onset appears localized, onset location and lateralization are not consistent from one seizure to the other. This statement may eventually be found to be correct, but the only functional magnetic resonance imaging (fMRI) evidence we have at present (Moeller et al., 2010) does not support that position; that is, in patients whose absences started locally, all seizures had the same locus of onset.

The decisive distinction between ictogenic networks appears to be quite different from this description in the Commission Report. Ictogenesis in idiopathic epilepsies, both generalized (Moeller et al., 2010) and localization-related (Salek-Haddadi et al., 2009), is reported to use preestablished networks serving physiologic functions, and which interictally work normally (Moeller et al., 2011). These networks can be defined as subsystems of the central nervous system (CNS). Capovilla et al. (2009), as the result of the first Monreale workshop, therefore, proposed the term “system epilepsies.” This concept has been further developed in a second Monreale workshop (report in preparation). The missing link in our present understanding of “system epilepsies” is the regional distribution of neurons whose function is altered by mutations.

In contrast, in focal symptomatic epilepsies, pathologic networks are probably generated around an epileptogenic lesion and interact with physiologic circuits. However, much more research into focal ictogenesis is required. The clearest examples would be epilepsies caused by well-defined exogenous lesions of normal brains, such as seen with brain trauma or local inflammations. The establishment of pathologic focal ictogenic networks around these lesions could be studied prospectively. In addition, there is a need for investigation of ictogenic networks in symptomatic generalized epilepsies.

To be justified, a new classification needs to reflect the new, emerging nosologic understanding of epilepsy. It will result from hypothesis-driven research and open discussions of the results.


The author has no conflicts of interest to disclose. I confirm that I have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.