Comparative efficacy of deep brain stimulation in refractory epilepsy with cognitive impairment: a Meta-analysis

Abstract Deep brain stimulation (DBS) has been recognized as a promising treatment for drug-resistant epilepsy (DRE); however, its effects on comorbid cognitive impairment remain insufficiently explored. This meta-analysis assessed the effectiveness of DBS in improving cognitive function (including memory, attention, executive function, and processing speed) in DRE patients with concurrent cognitive deficits, while synthesizing evidence on seizure reduction and quality of life (QoL). In accordance with PRISMA guidelines, we analyzed data from 15 prospective studies and randomized controlled trials (RCTs) encompassing 494 participants. Standardized mean differences (SMDs) were computed using a random-effects model, with heterogeneity evaluated via I² statistics. Results demonstrated significant cognitive improvements across all domains: memory (SMD = 2.05, 95% CI: 1.71–2.39), attention (SMD = 1.36, 95% CI: 1.05–1.67), executive function (SMD = 1.67, 95% CI: 1.29–2.04), and processing speed (SMD = 2.09, 95% CI: 1.80–2.38; all p < 0.001). Subgroup analyses indicated that higher stimulation parameters (voltage, pulse width) were associated with enhanced efficacy, though inter-study heterogeneity (I²=60.4–81.0%) highlighted variability in patient characteristics and DBS protocols.And another subgroup analysis showed DBS significantly improves cognition across memory, attention, executive function, and processing speed. CMN and hippocampus targets yielded the greatest benefits, while STN showed the weakest effects, highlighting target selection’s crucial role in optimizing cognitive outcomes. Although DBS exhibits substantial benefits for cognitive impairment and seizure control in DRE, heterogeneity underscores the need for optimized stimulation parameters and standardized cognitive assessments. These findings reinforce the role of DBS as a safe and effective intervention for DRE, however, additional RCTs are necessary to validate long-term outcomes and elucidate underlying mechanisms.