The occurrence of spikes in the EEG has long been interpreted as a marker of epileptogenicity. Recent studies in patients suggest that the contrary may be the case, i.e. that high interictal spike rates may actually be related to a low seizure frequency. Here, we correlated seizures and interictal spike periods in epidural EEG-recordings from chronically epileptic rats in the pilocarpine model, which resembles temporal lobe epilepsy as one of the most common forms of the disease in humans.

A prolonged status epilepticus (40 min) was induced by a single pilocarpine injection (340 mg/kg, i.p.) in 30-day-old rats. Control rats received saline instead of pilocarpine. Status epilepticus results in a chronic state of spontaneous recurrent seizures. After initial pilocarpine-induced seizures were observed, a telemetric EEG system was used to enable a continuous long term video-EEG monitoring of pilocarpine-treated and control rats in an environment with a fixed 12-hours light-dark-cycle. Epileptic seizures (frequency and duration) and interictal spike periods (frequency and duration) were detected by manual screening of the EEG recordings and concomitant video analysis.

Both daily seizure frequency (mean of 2.1 generalized seizures, range 0–6) and daily interictal spike period rate (mean of 8.8, range 0–52) varied substantially among animals. However, seizure duration (mean of 0.6 min, range 0–2 min) showed markedly less variance compared to the duration of spike periods (mean of 5 min, range 0–30 min). Therefore, we calculated the total time having a generalized seizure (up to 10 sec/hour) or showing interictal spike periods in the EEG (up to 4 min/hour), respectively. A subsequent regression analysis for these two variables revealed a negative correlation indicating that interictal spike periods and epileptic seizures were indeed inversely related.

Although more animals need to be included in this study, these data already support the clinical impression that a high interictal spike rate is often accompanied with a low seizure frequency. Further non-linear modelling of spike morphology and multiple regression analyses might help to increase the predictive value of interictal spikes and will certainly shed more light on the relationship between interictal and ictal epileptic discharges.