Deep brain stimulation in Alzheimer’s disease: a systematic review and meta-analysis of cognitive outcomes

Abstract

Deep brain stimulation (DBS) has emerged as a potential neuromodulatory approach for Alzheimer’s disease (AD) by targeting neural circuits involved in memory and cognition. However, clinical evidence remains limited, with small studies reporting inconsistent findings across different stimulation targets and cognitive outcome measures. To systematically evaluate and quantitatively synthesize the available clinical evidence regarding the cognitive outcomes and safety of DBS in patients with Alzheimer’s disease. This systematic review and meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement. Electronic searches of PubMed, Embase, and Web of Science were performed from database inception to identify eligible clinical studies evaluating DBS in patients with AD. Random-effects meta-analyses were conducted for 12-month Alzheimer’s Disease Assessment Scale–Cognitive Subscale (ADAS-Cog) and Mini-Mental State Examination (MMSE) outcomes using Hedges’ g as the standardized effect size. Between-study heterogeneity was assessed using Cochran’s Q , I 2 , and $$\tau ^2$$ statistics. Leave-one-out sensitivity analyses, exploratory target-based subgroup analyses, and qualitative synthesis of safety outcomes were performed. Six independent studies involving 91 patients were included in the systematic review, and all contributed to quantitative analyses where outcome data were available. Meta-analysis demonstrated a significant pooled improvement in 12-month ADAS-Cog outcomes (Hedges’ g=–0.590, 95% confidence interval (CI):–0.870 to –0.311; p <0.001), whereas no significant change was observed for MMSE outcomes (Hedges’ g=–0.131, 95% CI: –0.454 to 0.192; p =0.427). Exploratory target-based subgroup analyses showed numerically different effect estimates between fornix and nucleus basalis of Meynert stimulation cohorts, although no statistically significant differences between targets were identified and these analyses were based on separate study cohorts rather than direct comparisons. Sensitivity analyses demonstrated that the pooled estimates were not substantially influenced by individual studies. Overall, DBS was technically feasible, and reported adverse events were predominantly related to surgical implantation rather than chronic stimulation. However, the available evidence was limited by small sample sizes, clinical heterogeneity, and predominance of non-randomized studies, with the only randomized sham-controlled trial not demonstrating significant benefit in its primary cognitive outcomes. Current evidence suggests a possible signal of cognitive benefit associated with DBS, particularly in ADAS-Cog outcomes; however, interpretation of pooled estimates is limited by clinical heterogeneity across stimulation targets and patient characteristics. Target-specific findings remain exploratory because no direct comparative studies between DBS targets are currently available. Larger, adequately powered randomized controlled trials are required before the clinical efficacy of DBS in Alzheimer’s disease can be established.