Predictors of post stroke cognitive impairment: VITATOPS cognition substudy

Introduction Post stroke cognitive impairment (PSCI) is a common complication of ischemic stroke. PSCI can involve different depending on clinical and stroke related characteristics. The aim of this study is to determine the factors associated with impairments in specific cognitive domains. Methods The Vitamins to Prevent Stroke (VITATOPS) trial is a large, multinational randomised controlled trial. In this substudy, consecutive patients admitted for ischaemic stroke or transient ischaemic attack (TIA) at a tertiary hospital in Singapore were included. PSCI was defined as impairment of any of the six cognitive subgroups – visuoconstruction, attention, verbal memory, language, visual memory and visuomotor function – that were assessed annually for up to five years. Univariate and multivariate Cox proportional hazard models were used to determine factors associated with impairments in each of these cognitive domains. Results A total of 736 patients were included in this study, of which 173 (23.5%) developed cognitive impairment. Out of the six cognitive domains, the greatest proportion of patients had an impairment in visuoconstruction (26.4%) followed by attention (19.8%), verbal memory (18.3%), language (17.5%), visual memory (17.3%) and visuomotor function (14.8%). Patients with posterior circulation cerebral infarction (POCI) as the index stroke subtype had higher rates of cognitive impairment. Further subgroup analyses show that Indian race and advanced age were predictive of language impairment, whilst fewer years of education and POCI were predictive of verbal memory impairment. POCI was predictive of visual memory impairment, and advanced age and POCI were predictive of visuomotor function impairment. Conclusion We identified visuoconstruction and attention domains to be the most affected in our Asian cohort of PSCI. Advanced age, lower levels of education, posterior circulation strokes and concomitant comorbidities such as peripheral artery disease are independent predictors of PSCI.