Abstracts

Rationale: Responsive neurostimualtion (RNS) is a recent development in the treatment of pharmacoresistant focal epilepsy. Vagus nerve stimulation (VNS) is a more established treatment option for focal as well as generalized epilepsy. No systematic comparisons of seizure outcomes with RNS vs. VNS have previously been reported. We performed a systematic literature review to assess long-term outcomes of RNS and VNS in patients with focal epilepsy.  Methods: Available data analyzed included the pivotal trials of VNS (E-01 to E-05 studies)1 and a larger retrospective analysis of VNS registry data.2 For RNS, the results of the pivotal trial3, long-term follow-up4, and subgroup analyses of temporal lobe epilepsy5 and neo-cortical epilepsy6 were analyzed. Responder rates (RR- percentage of patients with ≥ 50% decrease in seizure frequency) were used to compare outcomes at similar time periods from implantation (see Table).  Results: Pivotal trial data at 3 years revealed RR of 42.7% for VNS1, and 57.9% for RNS4 (χ2=4.0329, p-value=0.045) for focal-onset seizures. VNS registry data at 2 years post-implant revealed RR of 58% for complex partial seizures with VNS (VNS registry), compared to 55% for RNS5 (χ2=0.081375, p-value=0.775). At 4 years post-implant, the RR was 61% for VNS (all seizure types, VNS registry), compared to 60.8% for RNS4 (χ2=0, p-value=1.0). Seizure freedom rates were not available at similar post-implant periods for statistical comparison.  Conclusions: RNS outcomes were marginally better than VNS (p=0.045) in the pivotal trial data. However, the initial VNS trials compared low vs. high stimulation settings, and may have under-estimated seizure improvement. This analysis did not show a statistical difference in outcomes compared to RNS for complex partial seizures (p=0.775) or all seizure types (p=1.0).  It is unclear whether the attrition of poorly-controlled patients in the retrospective VNS registry analysis may have over-estimated seizure improvement and masked differences. Another confound is that several patients in the RNS trials already had VNS, which may have led to additive benefit. The available data thus do not provide a clear answer to whether VNS or RNS lead to better seizure outcomes in treatment of focal epilepsy. For generalized and poorly localized epilepsy, RNS is not indicated, while VNS may have even greater benefit than in focal epilepsy. A direct head-to-head comparison of the efficacy of VNS compared to RNS in focal epilepsy to guide clinical decision-making is critically needed.  Funding: None