Abstracts

Rationale: Animal models of status epilepticus, traumatic brain injury, and neonatal hypoxia-ischemia have been developed to study epileptogenesis. To investigate epileptogenesis, it is necessary to monitor both behavior and EEG of the animals. While seizure detection systems have been developed to monitor chronic seizures their cost can be prohibitive for routine use. Therefore, we developed a relatively inexpensive EEG detection system. To validate the system we assessed seizure activity in the rat model of kainate-induced epileptogenesis. Methods: The system was composed of 4 main parts: (1) video collection, (2) EEG collection, (3) data storage, and, (4) data analysis. The first of these involved a 4-camera system (Q-See) whose signal was fed into a video capture card. Real time acquisition software was then used to compress the signal for storage on hard drive. The EEG collection was performed using a Pinnacle system (4 systems per computer). The data was stored on the computer hard drive, with back-up external drives and DVD. A data analysis code (SVERAT) was written using Visual Basic 2008 routines that displayed the video-EEG data simultaneously for 4 rats. It also allowed for analysis of the data (FFTs, seizure detection). To validate the utility of the system we used the rat kainate model of epilepsy. Adult rats were implanted with two hippocampal electrodes and a single dural electrode using stereotaxic placement. After a recovery period of one to seven days, the rats underwent video-EEG monitoring. The sampling frequency was 400 Hz. The rats (n=10) were subsequently injected with one or more doses of kainate to achieve at least 10 convulsive seizures for 3 consecutive hours. Controls included a group of rats that underwent kainate treatment without implantation (n=4) and implantation without kainate treatment (n=3). The study endpoint was the time at which the first chronic seizure (not associated directly with the kainate treatment) occurred. Results: The system, with a cost of about $3,000 per rat, provides high quality EEG while recording at up to 1000Hz. Seizures and interictal spikes were easily identifiable. Our benchmarking study confirmed our suspicion that the implantation of electrodes did decrease the time to first chronic seizure, but not significantly. Also, implantation in the absence of kainate injection did not result in chronic seizures. The time to first chronic seizure in implanted rats injected with kainate was 293 ± 134 hours. In comparison, unimplanted kainate injected rats monitored for behavior by video displayed seizures 204 ± 50 hours after kainate injection. Although many of the chronic seizures did not have a behavioral correlate, the initial electrographic seizure had a behavioral correlate in 7 of the 10 implanted rats that were treated with kainate. Conclusions: Our preliminary results suggest that a less expensive video-EEG system can be utilized to detect chronic seizures in a rat status epilepticus model of epilepsy. Funding provided by NINDS grants K08 NS05361003(AW) and RO1NS039587(MP).