Abstracts

Rationale: A newly developed algorithm was used to analyze seizure frequency in the rat tetanus toxin (Tx) model of mesial temporal lobe epilepsy. We hypothesize that a circadian pattern of seizure activity will be present in this model similar to what has been observed in many forms of human epilepsy.Methods: Adult, male Sprague-Dawley rats (n=9) were anesthetized and placed in a stereotaxic apparatus. Tx (40-50ng, 0.4-5ul) was injected into the right ventral hippocampus followed by the implantation of bipolar, platinum electrodes (250 um) ipsilateral at the same coordinates as the injection site and in the contralateral hippocampus. Two cranial screw electrodes were also implanted to collect surface EEG. All electrodes were connected to a custom headstage for chronic EEG recording. Following 2 weeks of recovery, animals entered a baseline monitoring period (5-8 days) with 24/7 collection of ambulatory 3-channel differential electrocorticograms (ECoGs). The baseline monitoring period was followed by a 5-8 day treatment period where rats received voltage controlled low frequency electrical stimulation (VCLFS, 1V peak to peak, 1Hz) delivered for 1 minute of each hour for the entire 5-8 day treatment period. Data collected were hosted on IEEG.org. An automated seizure, noise and stimulation artifact detection algorithm was run on a total of 2,916 hours (121.5 days) of data collected from the 9 rats. The algorithm detected 5,974 electrographic seizures. For each electrographic seizure detected by the algorithm, a seizure annotation was automatically added by the algorithm to the corresponding record in IEEG.org. To confirm the algorithm’s performance, a few annotations selected at random from each rat were manually reviewed to verify that the electrographic seizures detected by the algorithm were, in fact, electrographic seizures.Results: A clear circadian pattern of electrographic seizures was observed in the Tx-treated rats. Rats (nocturnally active animals) had 13.9% more electrographic seizures in the light ON cycle (periods of sleep and less activity) compared to electrographic seizures in the light OFF cycle. The average seizure rate during light OFF hours (1.86 seizures/hour) was significantly less (p<0.03, Wilcoxon rank sum test) than the average seizure rate during light ON hours (1.98 seizures/hour).Conclusions: A newly developed automated seizure detection algorithm detected and counted electrographic seizures in the continuously recorded EEG of Tx-treated rats during the light ON vs light OFF cycles (24hr). A circadian pattern of electrographic seizures was observed in the experimental tetanus-treated rats where rats had more electrographic seizures during light ON cycle compared to light OFF cycle. (Supported by NINDS 5U01NS064049)