Efficacy of scalp stimulation for multidomain cognitive impairment in patients with post-stroke cognitive impairment and dementia: A network meta-analysis and meta-regression of moderators

Background: Scalp stimulation has gained more traction for post-stroke cognitive impairment and dementia (PSCID); the interaction between stimulation targets and parameters influences the response to the stimulation. However, the most efficacious treatment for improving different domains of cognitive impairment remains unknown. Objective: We aimed to conduct a systematic review and network meta-analysis (NMA) to compare the efficacy of various scalp stimulation protocols used in PSCID treatment. Methods: Randomized controlled trials of scalp stimulation in patients with PSCID were searched in eight databases over the past 20 years. Standardized mean differences (SMDs) for global and subdomain cognitive scores were pooled in Bayesian NMA. Moderators were examined using meta-regression analysis. Results: A total of 90 trials, with 6199 patients, were included. Low-frequency repetitive transcranial magnetic stimulation (rTMS) over the unaffected dorsolateral prefrontal cortex (DLPFC) was highly suggested for alleviating global severity (SMD = 1.11, 95% CI (0.64, 1.57)). High-frequency rTMS over the left DLPFC was recommended for language use (1.85 (1.18, 2.52)), executive function (0.85 (0.36, 1.33)), orientation deficits (0.59 (0.07, 1.13)), and attention (0.85 (0.27, 1.43)). Anodal transcranial direct current stimulation over the affected DLPFC (2.03 (0.72, 3.34)) was recommended for treating memory impairment. Meta-regression analyses showed significant associations within attention, language and orientation. Conclusion: Overall, different cognitive domains have different optimal scalp stimulation prescriptions, and activating the affected key brain regions and inhibiting the unaffected area is still the most effective treatment.