154 Neurostimulation in Pediatric Epilepsy: A Meta-analysis of Long-term Outcomes

INTRODUCTION: With approximately 0.5-1% of children worldwide affected by epilepsy. METHODS: A comprehensive search strategy was executed across PubMed, Scopus, Web of Science, Embase, and Cochrane CENTRAL databases until February 20, 2024. Eligible studies included observational studies, clinical trials, and case series reporting on neurostimulation outcomes in children with refractory epilepsy. Two independent reviewers conducted screening and data extraction, followed by a risk of bias assessment using appropriate tools. Statistical analyses, including narrative synthesis and meta-analysis, were performed, along with subgroup and leave-one-out analyses. RESULTS: A total of 62 studies encompassing 2,496 patients were included for VNS analysis, revealing response rates of >50% at various time points: 55% (95% CI: [44%, 66%]) at 3 months, 47% (95% CI: [40%, 55%]) at 6 months, 50% (95% CI: [43%, 56%]) at 1 year, 57% (95% CI: [44%, 70%]) at 1.5 years, 55% (95% CI: [48%, 61%]) at 2 years, 52% (95% CI: [43%, 61%]) at 3 years, 56% (95% CI: [44%, 67%]) at 4 years, 45% (95% CI: [31%, 60%]) at 5 years, 40% (95% CI: [16%, 68%]) at 6 years, and 52% (95% CI: [24%, 79%]) at 7 years, with a seizure freedom rate of 8% (95% CI: [5%, 11%]). For RNS, six studies comprising 124 patients exhibited a >50% response incidence of 67% (95% CI: [57%, 75%]), while for DBS, six studies with 84 patients demonstrated a >50% response incidence of 77% (95% CI: [66%, 86%]). CONCLUSIONS: VNS, RNS, and DBS, exhibit promising efficacy in pediatric epilepsy management, with notable response rates observed across various time intervals. Further research is warranted to elucidate optimal treatment protocols and refine patient selection criteria for maximizing clinical benefits.