Abstract
Commentary
Whether “seizures beget seizures” has been a point of contention ever since Sir William Gowers coined this aphorism more than 125 years ago (1). Although there is convincing experimental evidence to support this premise, current understanding suggests that it is not clinically applicable and that, with the exception of some rare syndromes, human epilepsy is not a progressive, self-perpetuating disorder (2). Potentially clouding this knowledge is the increasing recognition that early life episodes of complex febrile seizures are associated with the later development of temporal lobe epilepsy (3) and that the number of pretreatment seizures is related to the probability of subsequent remission (4). These findings are, of course, entirely separate issues from the suggestion that one seizure increases the likelihood of another.
Unfortunately, the boundaries of these phenomena have become somewhat blurred amid the recent clamor to investigate the cellular mechanisms of epileptogenesis and to assess how these mechanisms might be exploited to prevent or delay the development of epilepsy (5). The procedure of employing acute experimental seizures as a precipitant of a subsequent epileptic state and dissecting the myriad of molecular events that occur in the latent period is a perfectly reasonable and legitimate endeavor. However, a troubling departure from this effort has involved a regression to Gowers's dictum and resulted in a largely unwritten acceptance of the theory that a single seizure or cluster of seizures can predispose to further episodes. In their enthusiasm preclinical investigators can, on occasion, lose sight of the importance of clinical relevance and, more specifically, the fact that epileptogenesis, pharmacoresistance, and seizures begetting seizures are not one and the same thing.
The recent manuscript by Evans et al. examined the effect of three successive audiogenic seizures on the GABA neurotransmitter system in the inferior colliculus of the genetically epilepsy-prone rat (GEPR). Twenty-four hours after the final seizure, the investigators observed a pattern of cellular effects that was consistent with an alteration in the subunit composition of the postsynaptic GABAA receptor, leading to a decrease in its sensitivity to GABA and an attenuation of inhibitory neurotransmission in the site of seizure origin. They deduced that compromised GABAergic inhibition in the inferior colliculus could predispose to further seizures and contribute to the phenomenon of audiogenic kindling (6–8). This has, after all, been mooted as one of the principal mechanisms of seizure susceptibility in the GEPR (9). However, the authors chose not to comment on the apparent hyperactivity of GABAergic inhibition in the inferior colliculus of seizure-naive, epilepsy-prone rats, when compared to nonepileptic control animals. Arguably, this is a more intriguing finding—one that may underlie the epileptogenic nature of the aforementioned diminution in GABAergic activity, and one that certainly has a significant bearing on how this study could or should be interpreted. Instead, the authors elected to focus on clinical relevance, suggesting that their study might explain the phenomenon of seizure clustering and have implications for epileptogenesis, pharmacological responsiveness, and the treatment of epilepsy after a single unprovoked seizure.
At this stage, the margins of disparate clinical issues begin to merge and interpretation becomes a little questionable. On the surface, these investigators have succeeded in identifying a mechanism by which seizures might beget seizures, at least in the GEPR. However, it is not appropriate to then extrapolate this observation to the clinical arena where the phenomenon does not exist or attempt to align it with any other vaguely related clinical circumstance. There is no doubt that the study provides a novel insight into the cellular consequences of audiogenic stimulation in the GEPR, but it also offers up more questions than answers. The permanence of the observed effects and how they relate to the number and/or frequency of seizures is not addressed, and the authors fail to discount the possibility that repeated exposure to intense audiogenic provocation might elicit similar changes in the inferior colliculus of normal animals, particularly as this structure represents the primary point of convergence for multiple, bilateral auditory afferents (10). Finally, they provide no direct experimental evidence that would support their proposed exacerbation of seizures. Demonstrating that seizure severity increased with successive stimulations would have added a behavioral correlate to the cellular and molecular findings and offered at least some support to the principal findings of this manuscript.
Despite the authors’ assertions to the contrary, there is little in their paper to confirm that repeated seizures are associated with enhanced epileptogenicity in the GEPR and nothing to suggest that these findings have any relevance to the exacerbation of clinical epilepsy. This investigation has elegantly demonstrated the effect of a single seizure or a brief cluster of seizures on GABA-mediated inhibition in the primary epileptogenic zone in the GEPR but any interpretation of the findings should end there. In one sense, the conclusions of this paper are a little misguided, possibly as a result of ongoing efforts to unravel the phenomena of epileptogenesis, pharmacoresistance, and self-perpetuating seizures. In another sense, however, they are in keeping with an increasing extravagance in contemporary scientific reporting. Seizures may not beget seizures but research trends can, on occasion, beget overinterpretation of results.