Risk Factors of Poststroke Cognitive Impairment: A Meta-Analysis

Objective To explore the potential risk factors of post stroke cognitive impairment (PSCI) by conducting a meta-analysis. Methods Literature search was performed in databases (PubMed, Embase, Web of Science, CNKI) using keywords of PSCI. Cochrane ROB tool was adopted for evaluating the quality of the included studies. Afterwards, data was independently extracted by 2 investigators. Heterogeneity was quantified across studies by Chi-squared-based Q statistic test and I2 statistic. The random-effects model or fixed-effects model was employed to compute the pooled estimates depends on whether the heterogeneity was significant (I2 > 50% or P < .05) or not. Publication bias was evaluated by the funnel plot and Egger's test. Sensitivity analysis was accomplished through eliminating studies 1 at a time to evaluate the stability of the pooled estimates. Results 23 high-quality studies with 13322 patients were included. Compared with patients with no cognitive impairment, PSCI was more likely to develop in the elderly (pooled MD = 3.58, 95% CI = [1.82, 5.34]), female (pooled RR = 1.23, 95% CI = [1.07, 1.41]), or less-educated (pooled MD = -1.63, 95% CI = [-2.96, -.31]) patients with a history of hypertension (pooled RR = 1.07, 95% CI = [1.03, 1.11]), diabetes mellitus (pooled RR = 1.10, 95% CI = [1.03, 1.17]), atrial fibrillation (pooled RR = 1.38, 95% CI = [1.10, 1.74]), or stroke (pooled RR = 1.36, 95% CI = [1.09, 1.70]). Smoking did not affect the development of PSCI in patients (pooled RR = .96, 95% CI = [.78, 1.19]). Ischemic heart disease and region represented the sources of significant heterogeneity across studies. The pooled estimates were robust, and no publication bias was seen. Conclusion Age, gender, education, hypertension, diabetes mellitus, atrial fibrillation, and stroke were the risk factors of PSCI. Controlling these risk factors can help prevent PSCI.