Vagus nerve stimulation for treatment of drug-resistant epilepsy: a systematic review and meta-analysis.

To analyze the efficacy and safety of high-frequency VNS versus control (low-frequency VNS or no VNS) in patients with DRE using data from randomized controlled trials (RCTs). An electronic literature search was conducted on PubMed, EMBASE, and Cochrane Controlled Register of Trials (CENTRAL); 12 RCTs reporting seizure frequency or treatment response in studies containing a high-frequency VNS treatment arm (conventional VNS or transcutaneous VNS [tVNS]) compared to control (low-frequency VNS or no VNS) were included. Seizure frequency, treatment response (number of patients with ≥ 50% reduction in seizure frequency), quality of life (QOL), and adverse effects were analyzed. Seizure frequency was reported in 9 studies (718 patients). Meta-analysis with random-effects models favored high-frequency VNS over control (standardized mean difference = 0.82, 95%-CI = 0.39-1.24, p < .001). This remained significant for subgroup analyses of low-frequency VNS as the control, VNS modality, and after removing studies with moderate-to-high risk of bias. Treatment response was reported in 8 studies (758 patients). Random-effects models favored high-frequency VNS over control (risk ratio = 1.57, 95%-CI = 1.19-2.07, p < .001). QOL outcomes were reported descriptively in 4 studies (363 patients), and adverse events were reported in 11 studies (875 patients). Major side effects and death were not observed to be more common in high-frequency VNS compared to control. High-frequency VNS results in reduced seizure frequency and improved treatment response compared to control (low-frequency VNS or no VNS) in patients with drug-resistant epilepsy. Greater consideration for VNS in patients with DRE may be warranted to decrease seizure frequency in the management of these patients.