Efficacy and Safety of Non-Invasive Brain Stimulation for Patients with Parkinson’s Disease: A Systematic Review and Meta-Analysis
Main Article Content
Abstract
Background: Non-invasive brain stimulation (NIBS) is a promising adjunct for Parkinson’s disease (PD), but variation in modality, target, stimulation dose, outcome selection, and safety reporting has produced uncertain estimates of benefit and risk.
Objective: To evaluate the efficacy and safety of sham-controlled NIBS for adults with PD.
Methods: PubMed/MEDLINE, Embase, Web of Science, and CENTRAL were searched for randomized parallel-group trials published from 2011 through August 2026. The primary outcome was immediate post-intervention motor impairment. Standardized effects were expressed as Hedges’ g, with negative values favoring active NIBS. Random-effects synthesis used restricted maximum likelihood estimation and Hartung–Knapp confidence intervals. Heterogeneity, prediction intervals, target subgroups, leave-one-out analyses, extreme-effect exclusion, funnel-plot asymmetry, and arm-level safety reporting were examined.
Results: Eighteen studies involving 579 participants were included; 17 motor comparisons (n = 561) entered the primary synthesis. Active NIBS improved motor outcomes compared with sham stimulation (g = −0.94, 95% CI [−1.57, −0.30], p = .006), although heterogeneity was substantial
(I² = 83.6%; 95% prediction interval, −3.36 to 1.49). The magnetic-stimulation-only estimate was similar (g = −1.01, 95% CI [−1.66, −0.35]). Excluding two extreme effects attenuated the estimate (g = −0.61, 95% CI [−0.97, −0.26]) without changing its direction. All leave-one-out estimates favored active stimulation. One small study supported a mood effect. Safety data were incomplete: trials with complete counts reported no serious adverse events or adverse-event-related withdrawals, but missing arm-level reporting prevented comparative pooling.
Conclusions: NIBS, particularly magnetic stimulation, produced an average short-term motor benefit in PD, but marked heterogeneity and incomplete safety reporting limit certainty about effect size, durability, target superiority, and comparative safety.