Associations between cardiovascular risk factors, stroke severity, and post-stroke cognition: a moderated mediation analysis

Abstract

AbstractIntroductionCognitive impairment may affect one third of the stroke survivors. Cardiovascular risk factors have been described to be risk factors for lower cognition after stroke. However, most previous studies only used multivariate regression models to learn the association. The aim of our study was to investigate whether the effect of cardiovascular risk factors on cognition after stroke was mediated by stroke severity, the estimated effect of direct and indirect pathways, and the moderated association.MethodIn this incident cross-sectional study, 300 stroke patients received cognitive test within seven days after stroke. Cognitive tested was performed by the Chinese version of Mini-Mental State Examination (MMSE). A second stage dual moderated mediation model was used the select moderation variables. Finally, we constructed a structural equation model to test the indirect effects of cardiovascular and demographic factors on cognition stroke severity, the direct effects of predictors on cognition, and the moderated effects of hypertension.ResultsAge (estimate, -0.114; 95% bias-corrected CI, -0.205, -0.032; P<0.001), female (estimate, -2.196; 95% bias-corrected CI, -4.359, -0.204; P=0.009), lower education (estimate, -0.893; 95% bias-corrected CI, -1.662, --0.160; P<0.001), stroke severity (estimate, -1.531; 95% bias-corrected CI, -3.015, -0.095), hypertension (estimate, -2.242; 95% bias-corrected CI, -4.436, -0.242; P=0.003) and atrial fibrillation (estimate, -4.930; 95% bias-corrected CI, -12.864, -0.126; P=0.048) were directly associated with lower cognitive function after stroke. We found no evidence that cardiovascular risk factors indirectly correlated cognitive function through stroke severity. The combination of hypertension could alleviate the negative effect of atrial fibrillation on cognition (estimate, -3.928; 95% bias-corrected CI, -7.954, 0.029; P=0.009).ConclusionsWe explored the complex relationship between cardiovascular risk factors, stroke severity, and cognitive function after stroke. Using our method, researchers using other dataset could repeat the analysis and achieve a better understanding of the relationship. Future researchers are needed to find whether the moderated associations were casual or modifiable.