Ultra-long-term EEG recordings indicate no effect of vagus nerve stimulation on sleep in epilepsy patients

Vagus nerve stimulation (VNS) is a therapeutic option for pharmacoresistant epilepsy patients who are not eligible for surgery. While its primary aim is seizure reduction, the effects of VNS on sleep remain unclear, with previous studies reporting conflicting outcomes. This study uses ultra-long-term EEG recordings to assess the impact of VNS on sleep by comparing extensive baseline and post-VNS data.

Patients were enrolled from the PREDYct study, which involves ultra-long-term subcutaneous EEG monitoring initiated approximately three months prior to VNS activation and continuing for up to fifteen months. Patients with fewer than five complete nights of EEG during the baseline or post-VNS period were excluded. Sleep stages were classified using the EpiSight sleep classifier. For each patient, fractions of REM, N1, N2, and N3 sleep, total sleep time (TST), and wakefulness after sleep onset (WASO) were computed for each night. Changes at the individual and group-level were assessed using Wilcoxon rank-sum tests and generalized linear mixed-effects models with a Gamma distribution, respectively.

Ten patients were included, with a median of 42.5 (IQR: 33-48) complete nights during the baseline period and 61 (IQR: 28-166) nights during the post-VNS period. Significant changes between baseline and post-VNS periods were observed in three patients, with no consistent pattern across individuals. No significant group-level effects were observed.

In this small cohort, VNS therapy did not result in consistent changes in sleep parameters at the individual or group level, suggesting that VNS has no significant impact on the assessed sleep parameters in pharmacoresistant epilepsy patients.