Abstracts

Rationale: Deep brain stimulation (DBS) as a means for seizure control has recently gained much attention. We have designed and implemented an automated, just-in-time stimulation, seizure control scheme using a seizure prediction method coupled with deep brain stimulation in a rat model of chronic epilepsy. The scheme was tested in terms of its efficacy to control seizures, as well as its effect on synchronization of brain dynamics. The proposed technique of just-in-time stimulation resulted in a significant (>50%) reduction of seizure frequency as compared to a baseline period with no stimulation. Importantly, we show that a sustained successful seizure control depends significantly upon the site of stimulation. This study provides initial evidence for the use of multi-site closed-loop feedback control systems in control of epileptic seizures combining methods from seizure prediction and deep brain stimulation.Methods: Two adult male Sprague-Dawley rats were used in this study. Each rat was made chronically epileptic following a prolonged episode of status epilepticus using the lithium-pilocarpine model. Rats were implanted with 10 individual Tungsten microwires. Five phases of recordings, each of weeks in duration, were performed on each rat while EEG was recorded continuously with simultaneous online real-time nonlinear dynamical analysis. During Phase 1, a baseline EEG recording was performed per rat. During Phase 2, automated closed loop control was employed for each rat in which one-minute biphasic DBS trains were delivered to the thalamus at seizure warning time points determined by our seizure prediction algorithm. Stimulation was off during Phase 3. During Phase 4, one-minute periodic biphasic DBS trains with identical stimulation parameters as in phase 2 stimulated the left hippocampus. Stimulation was off during Phase 5. Seizure frequency per phase was retrospectively estimated per rat in the study.Results: The automated DBS control scheme reduced seizure frequency during phases of stimulation in both rats tested, with increase of seizure frequency towards baseline during periods of no stimulation. Interestingly, hippocampal stimulation appeared to have a more drastic effect than thalamic stimulation with respect to sustaining seizure control. Due to the use of identical stimulation parameters for both stimulation locations, these results show the importance of the timing and multi-site stimulation for seizure control. Conclusions: In this small-scale but very long-term, real-time, in-vivo DBS study with two epileptic rodents, we have shown that: 1) The development of an effective closed-loop seizure control system with recorded and analyzed EEG as input and electrical stimulation as output is feasible and promising 2) Location of stimulation is important for the sustained success of a seizure control system. Conduct of larger scale studies to shed more light on the above important conjectures for ictogenesis and treatment of epilepsy is currently underway. (Supported by NSF Cyber Systems grant ECS-0601740)