Thalamic neuromodulation for pediatric drug-resistant epilepsy: a case series evaluating deep brain stimulation and responsive neurostimulation.

Abstract

Objective: Children with generalized or multifocal drug-resistant epilepsy (DRE) without a clearly localizable and safely resectable seizure focus present a significant clinical challenge, as traditional resective interventions are often not viable. Emerging evidence supports the use of neuromodulation-particularly deep brain stimulation (DBS) and responsive neurostimulation (RNS)-for generalized and multifocal epilepsy in adults. However, safety and efficacy data for thalamic neuromodulation in pediatric populations remain limited. This single-institution case series describes, to our knowledge, the largest pediatric cohort treated with thalamic DBS or RNS for generalized or multifocal DRE, providing descriptive data on safety and seizure burden.

Methods: The authors performed a retrospective chart review of pediatric patients with DRE who underwent thalamic neuromodulation using RNS or DBS at Seattle Children's Hospital between January 2020 and July 2025 with at least 6 months of follow-up. Clinical, surgical, and seizure outcome data were collected, including seizure frequency and complications. Seizure outcomes were stratified into categorical frequency bins, and a trend analysis was performed to evaluate postoperative shifts in seizure burden.

Results: Twenty-six patients (mean age 14.5, range 6-20 years) underwent thalamic neuromodulation with DBS (n = 12) or RNS (n = 14). The centromedian nucleus was the target in 24 cases; 1 patient each underwent targeting of the anterior nucleus and pulvinar nucleus. At the last follow-up (median 30.2 months), 65.4% of patients met the responder criteria (≥ 50% seizure reduction), with a higher response in the DBS group (83.3%) compared with the RNS group (50.0%). The median seizure reduction was 75.7% for DBS and 37.5% for RNS. A significant downward shift in seizure frequency was observed postoperatively across the entire cohort (p = 0.031), including among patients with the highest baseline seizure burden. No intraoperative complications occurred. Two patients (7.7%) required device explantation, 1 due to infection and 1 due to behavioral side effects, and 1 patient discontinued therapy without explantation. No instances of sudden unexpected death in epilepsy or hardware malfunction were observed.

Conclusions: Thalamic neuromodulation using DBS and RNS was safe and well tolerated in pediatric patients with DRE, including multifocal and generalized seizure onsets. Meaningful seizure reduction was observed across a range of epilepsy phenotypes, with a trend toward greater benefit in the DBS group. These findings support thalamic neuromodulation as a promising treatment option in children with nonlesional DRE and highlight the need for prospective multicenter studies with extended follow-up.