Abstracts

Dietary therapies are a valuable non-pharmacological option for managing drug resistant epilepsy (DRE) in pediatric patients. While the classical ketogenic diet (KD) is well established, alternatives such as the Modified Atkins Diet (mAD) and Low Glycemic Index Treatment (LGIT) have gained attention due to better tolerability. LGIT is a more flexible dietary approach that allows inclusion of lower glycemic carbohydrates (< 50 GI) while using a less restrictive fat to carbohydrate ratio than other ketogenic diets. This flexibility may support improved adherence and quality of life. Although previous meta-analyses have compared LGIT with mAD, none have systematically evaluated LGIT alone using both recent randomized trials and real-world observational data. To address this gap in the literature we conducted a meta-analysis to evaluate the efficacy, safety, and acceptability of LGIT in pediatric DRE.

We systematically searched PubMed, EMBASE and Cochrane Library through May 2025. Eligible studies included children or adolescents with DRE treated with LGIT in RCTs or observational designs. Outcomes assessed were ≥50% seizure reduction, ≥90% seizure reduction, seizure freedom at 12 and 24 weeks, and any reported adverse event (AE). A random effects meta analysis of proportions was conducted using R software (v4.4.1). Analyses were stratified by study design, outcome type, and follow up duration. All pooled outcomes are represented in forest plots. Baseline patient demographics such as age and gender were extracted. Side effects and AEs were also reviewed.

8 studies (n=369 children) were included: 5 RCTs and 3 observational cohorts. Follow up ranged from 8 to 24 weeks. LGIT overall, showed a positive clinical response. In RCTs, >50% of patients achieved ≥50% seizure reduction (50.88%), with observational studies showing slightly higher rates (61.47%). For ≥90% reduction, estimates were 17.34% in RCTs and 9.72% in observational studies. A key finding was consistent seizure freedom in RCTs: 5.92% at 12 weeks (95% CI: 0.85–13.88) and 16.85% at 24 weeks (95% CI: 15.66–18.07). This sustained effect, showing no heterogeneity, highlights LGIT’s potential as a long term therapeutic option. Seizure freedom in observational studies was lower at 4.97% but highly variable. Regarding safety and tolerability, LGIT was well accepted. Any AEs occurred in 19.97% of participants, typically mild GI issues or challenges with food acceptance. Serious AEs were rare (< 0.3%) and did not necessitate diet cessation. Patient withdrawal due to AEs remained low (< 5% overall), suggesting good tolerability. However, estimates for several safety outcomes, including serious AEs, had wide confidence intervals often including zero due to sparse reporting across studies.