Poststroke Cognitive Impairment and the Risk of Recurrent Stroke and Mortality: Systematic Review and Meta-Analysis

Abstract

Background: Poststroke cognitive impairment (PSCI) occurs in about 60% of patients with stroke in the first year after stroke. However, the question regarding risks of recurrent stroke and mortality in patients with PSCI remains controversial. The goal of this study was to conduct a meta-analysis of published literature to estimate the risks of stroke recurrence and mortality associated with PSCI.

Methods and results: Electronic databases were screened for eligible studies published from 1990 to 2023. The primary end points of this study were recurrent stroke and mortality. Pooled estimates were calculated as hazard ratios (HR) with 95% CIs. Meta-regression analyses evaluated moderating effects of PSCI severity, study design, and study period on recurrent stroke and mortality. Pooled data from 27 studies comprised 39 412 patients with ischemic stroke. Nine studies evaluated the association between PSCI and risk of stroke recurrence that showed the hazard of recurrent stroke risk was significantly higher in patients with PSCI compared with those without it (HR, 1.59 [95% CI, 1.29-1.94]; I²=52.2%). Eighteen studies examined the impact of PSCI on mortality risk. The pooled hazard of mortality was significantly higher in the group with PSCI relative to the non-PSCI group (HR, 2.07 [95% CI, 1.65 -2.59]; I²=89.3%). Meta-regressions showed that the average effect of PSCI on mortality risk differed across study period and study design.

Conclusions: Based on this meta-analysis PSCI was statistically significantly associated with increased risks of recurrent stroke and all-cause mortality. Poststroke neurocognitive assessment may identify patients at a higher risk who may require more aggressive interventions for secondary prevention.

Keywords: dementia; mortality risk; poststroke cognitive impairment; stroke recurrence.

Figure 1. Preferred Reporting Items for Systematic Reviews and Meta‐Analysis.

PSCI indicates poststroke cognitive impairment.

Figure 2. Forest plots for observed hazard ratios of recurrent stroke (pooled analyses).

The plot shows a log transformed x axis. df indicates degrees of freedom; I ², percentage of variation across studies due to heterogeneity rather than chance; Q, Cochran's Q measure of heterogeneity; RE, random effects model; and τ², estimated amount of total heterogeneity. (Pooled hazard ratio estimate for the fixed effects model was 1.37 [95% CI, 1.22–1.55]; Q=18.51, df=8, P=0.02; I²=56%).

Figure 3. Forest plot of the observed hazard ratios of mortality (pooled analyses).

The plot shows a log transformed x axis. df indicates degrees of freedom; I ², percentage of variation across studies due to heterogeneity rather than chance; τ², estimated amount of total heterogeneity; Q, Cochran's Q measure of heterogeneity; and RE, random effects model. (Pooled hazard ratio estimate for the fixed effects model was 1.47 [95% CI, 1.40–1.54]; Q=82.16, df=17, P<0.001; I ²=79.3%).