Neuromodulation and sudden unexpected death in epilepsy: A systematic review of SUDEP rates across VNS, DBS, and RNS therapies

Background Sudden unexpected death in epilepsy (SUDEP) is the leading cause of epilepsy-related mortality, with rates between 6.3–9.3 per 1000 patient-years in surgical-candidate and drug-resistant epilepsy (DRE) cohorts, and lower rates in less severely affected populations. This systematic review examines SUDEP rates across vagus nerve stimulation (VNS), deep brain stimulation of the anterior nucleus of the thalamus (DBS-ANT), and responsive neurostimulation (RNS) to determine whether these neuromodulation therapies confer protection against SUDEP. Objective To synthesise available evidence on SUDEP incidence in patients treated with VNS, DBS, and RNS, and to compare rates against DRE population benchmarks. Methods A PRISMA-compliant systematic review searched PubMed, Embase, Cochrane Library, and ClinicalTrials.gov through January 2025. Studies reporting SUDEP events and person-years of exposure in neuromodulation-treated DRE patients were included. SUDEP rates per 1000 patient-years were extracted or calculated, with indirect comparisons against DRE reference populations. Results Eighteen studies comprising over 44,000 patients and 290,000 patient-years were included. VNS SUDEP rates ranged from 1.7 to 4.1 per 1000 patient-years, with the largest study (N = 40,443) demonstrating a significant temporal decline (rate ratio 0.68; 95 % CI: 0.53–0.87; p = 0.002). DBS-ANT rates were 1.62–2.0 per 1000 patient-years; RNS rates were 2.0–3.5 per 1000 patient-years. All modalities demonstrated rates substantially below untreated DRE populations (6.3–9.3 per 1000 patient-years). Greater seizure reduction correlated inversely with SUDEP rates. Conclusions All three neuromodulation modalities are associated with SUDEP rates of 1.7–2.8 per 1000 patient-years, below expected DRE rates. VNS has the strongest evidence for a temporal decline in SUDEP risk. Prospective registries with standardised SUDEP surveillance are needed to establish causality.