Introduction: Responsive Neurostimulation (RNS) has been shown to be safe and effective in reducing seizure frequency in patients with drug resistant epilepsy (DRE). Only a few large single-center cohort studies reporting RNS outcomes exist in the literature, often industry-sponsored with limited subgroup analysis. Further characterization and analysis of patients implanted with RNS within a single center may provide valuable clinical insights.
Methods: We conducted a retrospective review of patients with DRE implanted with the RNS system at our institution between 2017 and 2024 with at least six months of follow-up and two programming sessions. Patient demographics, medical and surgical history were collected. Efficacy was assessed as the median percent change in post-implantation seizure frequency compared to baseline, responder rate was defined as >50% reduction in seizures.
Results: Thirty patients underwent RNS implantation between 2017 and 2024. 73% of patients were male (n=22), median age at implantation was 36.5 years (IQR: 25-46). Eight patients had vagal nerve stimulator (VNS) placement. Lead distribution included: (n=11) bilateral hippocampal, (n=9) cortical, (n=3) bilateral thalamic, (n=2) corticothalamic and (n=5) unilateral hippocampal. Median seizure frequency reduction was 75% (IQR: 50.0-92.3) and responder rate was 70%. 18 (60%) patients achieved an ENGEL class ≤2B or better. Cenobamate use and prior VNS placement had a statistically significant relationship with achieving a post-operative Engel classification of 2B or better, (p < 0.035 and p < .0003 respectively). There was no statistically significant difference in postoperative seizure reduction based on lead location. The cohort had no permanent morbidity, mortality or infections.
Conclusion : RNS system is an effective treatment for DRE and the responder rate in our cohort is consistent with the seizure outcomes reported in the literature. Additional studies are needed to characterize the differences between RNS responders and non-responders.
.jpg)
