Deep brain stimulation for epilepsy: A systematic review and meta-analysis of randomized and non-randomized studies...

Deep Brain Stimulation (DBS) of the thalamus for drug-resistant epilepsy (DRE) is an emerging treatment modality. This systematic review and meta-analysis sought to evaluate the efficacy of stimulating different targets within the thalamus.

A systematic search of four databases was conducted. Rates for overall seizure reduction (SR), responder rate (RR ≥50 % SR), and seizure freedom (SF) were evaluated at a minimum time point of 12 months post-stimulation commencement in the anterior (ANT) and centro-median (CMN) thalamic nuclei. Subgroup analyses for a minimum 24 months follow up, sensitivity analyses, and funnel plots to assess for publication bias were also performed. Risk of bias was assessed using the ROBINS-I tool.

Fourty-nine articles met the inclusion criteria. The mean seizure reduction (SR) across 21 studies was 62.31 % (95 % CI: 55.99-68.62, p < 0.01). Specifically, SR was 64.28 % for ANT (95 % CI: 57.55-71.01, p < 0.01) and 69.11 % for CMN (95 % CI: 58.14-80.09, p < 0.01). Meta-analyses of 41 ANT studies and 12 CMN studies reported a response rate (RR) of 61.51 % (95 % CI: 54.11-68.9, p < 0.01) and 69.09 % (95 % CI: 54.01-84.16, p < 0.01), respectively. Overall seizure freedom (SF) was 3.57 % % for ANT (95 % CI: 1.86-5.28, p = 0.45) and 1.32 % for CMN(95 % CI: 0-4.45, p = 0.81). For ANT, RR was 67.63 % (95 % CI: 61.04-74.23) for follow-up periods longer than 24 months, and 44.05 % (95 % CI: 26.73-61.38) for periods shorter than 24 months. The SF rate for ANT was 3 % (95 % CI: 1-4 %) for follow-up under 12 months. For CMN, RR was 70 % (95 % CI: 53-87 %) for periods over 24 months, and 68 % (95 % CI: 31-100 %) for periods under 24 months. The SF rate for CMN was 1 % (95 % CI: 0-4 %) for periods under 12 months. There was no strong evidence of publication bias based on funnel plot analysis, and results were consistent across sensitivity analyses. Insufficient data precluded meta-analysis for other nuclei.

These findings demonstrate efficacy of ANT and CMN DBS for patients with DRE, defined by responder rate and seizure reduction. Further research is required to optimize patient selection, predict individual response, and assess non-seizure related outcomes.