Abstract
BACKGROUND: Post-stroke cognitive impairment (PSCI) represents a serious post-stroke complication with poor cognitive consequences. A vascular consequence after a stroke is that the occurrence and progression of PSCI may be closely related to blood pressure (BP). Thus, we systematically reviewed and performed a meta-analysis of the literature to examine the correlations between BP and PSCI. METHODS: We systematically queried databases, including PubMed, the Cochrane Library, Embase, and Scopus, and conducted meta-analyses on studies reporting odds ratios (ORs) related to the association between BP and PSCI. Two authors autonomously assessed all titles, abstracts, and full texts and extracted data following the Meta-Analysis of Observational Studies in Epidemiology guidelines. The quality of the studies was evaluated using the modified Newcastle-Ottawa scale. RESULTS: Meta-analyses incorporated 12 articles comprising a cumulative participant cohort of 21,732 individuals. The quality assessment indicated good in five studies, fair in one study, and poor in six. Through meta-analyses, we found that hypertension, systolic or diastolic BP (SBP or DBP) was significantly associated with PSCI (OR 1.53, 95% confidence interval (CI), 1.18-1.99; p = 0.001, I2 = 66%; OR 1.13, 95% CI, 1.05-1.23; p = 0.002, I2 = 52%; OR 1.38, 95% CI, 1.11-1.72; p = 0.004, I2 = 90%, respectively). In the subgroup analysis, SBP < 120 mmHg, 120-139 mmHg, 140-159 mmHg, 160-179 mmHg, and DBP ≥ 100 mmHg highly predicted the occurrence of PSCI (OR 1.15, p = 0.0003; OR 1.26, p = 0.010; OR 1.15, p = 0.05; OR 1.02, p = 0.009; OR 1.96, p < 0.00001, respectively). However, the predictive effect of BP for PSCI declines when SBP ≥ 180 mmHg and DBP ≤ 99 mmHg (p > 0.05). Statistical heterogeneity was moderate to high, and publication bias was detected in SBP for PSCI. CONCLUSIONS: Considering the multifactorial etiology of PSCI, it is difficult to conclude that BP is an independent risk factor for PSCI. Given the restricted inclusion of studies, caution is advised when interpreting the findings from this meta-analysis. Subsequent investigations with substantial sample sizes are essential to exploring BP as a prospective target for addressing PSCI. TRIAL REGISTRATION NUMBER: CRD42023437783 from PROSPERO.