Abstracts

Rationale: Interictal spikes (ISs) are biomarkers for excessive brain excitability in epilepsy. However, the relationship between IS frequency and pattern and seizure development is not well understood. In some acute seizure models, ISs increase in frequency and complexity just before seizures and appear to be the triggering event. In other epilepsy models, there appears to be little relationship. In human patients, there is a suggestion that spikes increase after seizures and then decline thereafter. If there is a relationship between seizures and ISs, it is likely to be more clearly seen in animals that exhibit clustered seizures. We hypothesize that repetitive seizures increase the frequency of ISs, and the seizure clusters or ISs slowly induce a homeostatic mechanism that may end the cluster and reduce the likelihood of subsequent seizures for days or weeks. Methods: Status epilepticus (SE) was induced in male Sprague Dawley rats by either i.p. injection of pilocarpine (n=15) or electrical stimulation of the perforant path bilaterally for 3 hours (n=8) or unilaterally for 8 hours (n=10). Animals that developed spontaneous recurrent seizures were continuously monitored with bilateral intracranial electrodes and video for periods up to 6 months. Automated spike and seizure detecting software were used for analysis, with video confirmation of seizures. Results: Adult rats that undergo either chemically or electrically induced SE develop chronic epilepsy at various times after the inducing stimulus. ISs are invariably present in those animals that develop epilepsy. In animals that exhibit clustered seizures (n=8, clusters occur at 7-14 day intervals with 10-50 seizures per cluster), ISs are often at their lowest frequency before the seizure clusters, increase in frequency during the cluster (over 1-2 days), continue to increase in frequency after the seizures subside (over 1-2 days), and then slowly decline over several days to low levels before the beginning of the next cluster. In animals with only isolated seizures, a similar pattern can be observed. Conclusions: Our data suggest that clusters of seizures are most likely to occur during periods of reduced IS firing. The seizures are associated with (or cause) an increased IS firing (especially during intense clusters) which may paradoxically reduce the likelihood of subsequent seizures. The time course of the phenomena in the rats with clustered seizures suggests that slow molecular events may be occurring that serve to both enhance some forms of brain excitability (ISs) while at the same time, reducing the tendency for subsequent seizures for substantial periods. It has been shown, for example, that repetitive seizures can induce normally glutamatergic granule cells to produce and secrete GABA and that this change in neuronal phenotype lasts for approximately one week. It is likely other homeostatic mechanisms are also involved in dampening excessive brain excitability. Identifying these mechanisms may provide new targets for innovative anti-seizure therapy.