Abstracts

Rationale: Closed-loop responsive direct-brain neurostimulation was investigated in the rat tetanus toxin (TX) model of MTLE to determine whether it is an appropriate model for the preclinical evaluation of new neurostimulation therapies for mesial temporal lobe epilepsy (MTLE).Methods: Adult male Sprague-Dawley rats (n=31) were unilaterally injected with TX (50ng/0.5 l) into the ventral hippocampus (VH), implanted bilaterally with bipolar microelectrodes (125 m Pt or PtIr) into the VH, and bilaterally with screw electrodes into the skull, posterior to bregma and 3mm lateral to the midline. After a 14-21d recovery period and a 5d baseline period (BP), the rats were separated into a group that received responsive stimulation (RS, n=16) and into a control group (CTL, n=15) that did not receive stimulation. Over a 5d treatment period (TP) RS rats received electrical stimulation (200ms burst, 100 A, 200Hz, 160 s PW) delivered bilaterally to the VH in response to abnormal electrographic discharges detected by a modified NeuroPace RNS System Neurostimulator. Electrographic and behavioral activity were monitored by continuous 24/7 video/EEG. Electrographic seizures (SZ) were identified using a high-speed audio scoring method (Breeden, et al., AES abstract 2012). Video corresponding to the electrographic SZ was reviewed and scored according to the Racine seizure scale. Since seizure rates can vary considerably between rats, the effect of responsive neurostimulation on seizures was assessed by calculating the percent change in SZ rate (PCSZ) between BP and TP (((SZTP- SZBP)/ SZBP)*100) for each rat, and comparing this metric between groups.Results: The primary statistical outcome, percent change in seizure rate (PCSZ), was significantly lower (p=0.043) for RS rats (-32.9% 10.3%) compared to CTL rats (0.94% 17.1%). Median PCSZ values were also lower for RS rats (-38.7%) compared to CTL rats (-5.2%). Raw seizure rates were also assessed. During the BP, the average rate of SZ per day was highly variable (69.5 12.6) but was not significantly different between RS and CTL groups (p=0.095). During the TP, the rate of SZ per day was significantly lower (p=0.009) for RS rats (32.7 13.7) compared to CTL rats (85.4 17.7).Conclusions: These results suggest that the rat TX model of MTLE is a valid animal model by which to assess the effects of new stimulation paradigms before transitioning into humans with mesial temporal onset seizures. The seizure reductions in the treated and control MTLE rats were similar to the seizure reductions that occurred at the end of the blinded evaluation period in a multicenter, randomized double blinded sham stimulation controlled trial in adult patients with partial onset seizures. Subjects treated with responsive stimulation experienced a 41.5% reduction in seizures compared to a 9.4% reduction in subjects not treated with stimulation. (Research supported by NINDS 5U01NS064049-4)