Abstracts

Rationale: Temporal lobectomy is the best treatment option for many patients with medically intractable mesial temporal lobe epilepsy (MTLE). However, some patients fail to respond to MTL resection, and for others, surgery carries an unacceptable risk for neurological morbidity. These MTLE patients may be candidates for focal responsive neurostimulation.Methods: Subjects with MTLE were identified from those participating in clinical trials of a responsive neurostimulator (RNS® System, NeuroPace). For any subject with valid seizure diary data, the % change in clinical seizure frequency was calculated by comparing the seizure rate during the first 3 month period of each year in the open label period to the pre-implant baseline period.Results: Of the 256 subjects, 105 had seizures that arose solely from the MTL. The average age at enrollment was 37.1 years, and the median baseline seizure frequency was 7.7 per month (range 3 – 216.7). 71% (n=75) of the subjects had bilateral onsets and 29% (n=30) had unilateral. The majority (70%; n=21) of unilateral subjects had left MTL onsets. Across all MTL subjects, 54% (n=57) had mesial temporal sclerosis, 46% (n=48) had undergone intracranial video-EEG monitoring, 26% (n=27) had prior VNS, and 11% (n=12) had a prior resection. Patients were implanted with 2.3 leads on average; 71 subjects had depth leads only, 28 subjects had both depth and strip leads and 6 patients had only strip leads. The majority of depth leads were placed in the MTL structures using a longitudinal approach. Cortical strip leads were either placed subtemporally or on the lateral temporal lobe. All 105 subjects had seizure data at year 1 and 78 had seizure data at year 6. Figure 1 shows the median % reduction in seizures and responder rate for years 1-6. The median % reduction for those subjects with 6 years of post-implant data was 73%; 19.2% had not had a seizure during the 3 month analysis window for year 6. To assess whether the 6 year response was due to a smaller sample size, a last observation carried forward analysis was performed for all subjects. This also showed a median % reduction of 73%. The likelihood of a favorable response was similar in those with or without sclerosis. Serious device related adverse events in ≥2% of the MTLE subjects (n=105) over the 660 implant years were device lead damage (7 subjects) and soft tissue implant site infection (11 subjects), with an overall infection rate of 3.5% per neurostimulator implant or replacement procedure. There were no serious adverse events specific to stimulation nor were there adverse neuropsychological effects (Loring et al., 2015) or mood effects (Meador, 2015).Conclusions: Responsive stimulation represents a treatment option for patients with medically intractable MTLE who are not candidates for temporal lobectomy or who have failed a prior MTL resection. Seizure reductions began with the initiation of responsive stimulation and continued to improve over several years, reaching a median % reduction of 73% at 6 years. Further, responsive stimulation of the MTL was well tolerated.