Abstracts

Deep brain stimulation (DBS) has emerged as a promising neuromodulation intervention for patients with drug-resistant epilepsy (DRE), particularly those ineligible for resective surgery. However, response to DBS varies substantially, with only a subset achieving meaningful seizure reduction. Identifying clinical and surgical predictors of DBS responsiveness is critical to optimize patient selection and guide personalized treatment strategies. 

This study investigates patient-specific and procedural factors associated with ≥50% seizure reduction following DBS, aiming to delineate characteristics that correlate with favorable outcomes.

We retrospectively analyzed clinical data from 25 patients with DRE who underwent DBS targeting thalamic nuclei with at least 1 year follow-up (1-12 years, median 3 years). Patients were stratified into responders (≥50% seizure reduction) and non-responders (< 50%). Variables analyzed included demographics, epilepsy duration, prior resection history, type of resection (e.g., temporal lobectomy or corpus callosotomy), vagus nerve stimulation (VNS) usage, DBS target (e.g., centromedian nucleus, anterior nucleus, and pulvinar), and stimulation settings. Statistical comparisons included t-tests, chi-square tests, and logistic regression.

A total of 25 patients underwent DBS for DRE with 16 patients (64%) being classified as responders. While no variable reached statistical significance, trends suggested better response among patients without prior corpus callosotomy, those targeted at the centromedian nucleus, shorter duration of epilepsy, and younger patients. Patients with lesionectomy showed improved response with DBS compared to those with temporal resections or callosal disconnection. Responders tended to have higher current of stimulation ( >4mA) in comparison with non-responders (< 4mA).