Measuring clusters of spontaneous spike-wave discharges in absence epileptic rats

Spike-wave discharges (SWDs) characterizing absence epilepsy appear in closely packed aggregated sequences, which gave rise to the name "pyknolepsy" for this disease. In WAG/Rij rats, genetically prone to absence epilepsy, spontaneous SWDs seem to occur in clusters as well. Here, we aimed to quantify the seizures' clusters. SWDs sequences were extracted from long-term (complete estrous cycle) EEG recordings of adult female WAG/Rij rats. Spectral characteristics and half-decay time of autocorrelation functions (AC-tau) were calculated for time series of i(SWD) (proportion of time occupied by spike-wave activity), measured for subsequent periods. The clusters were characterized by means of AC-tau calculated for time series of i(SWD). The absence seizures were indeed clustered in a minute range. The clustering had a non-periodical character, since no significant and consistent periodicity was found in the minute range. AC-tau correlated positively with propensity of SWDs: i.e. the aggravation of absence epilepsy led to longer sequences of paroxysms and thus to a less random distribution. AC-tau was not sensitive to various phases of the estrous cycle, but was larger in the dark than in the light periods. We suggest that AC-tau can be used to quantify aggregation of epileptic events in the search for physiological basis of its temporal clustering. (c) 2006 Elsevier B.V. All rights reserved.