Risk factors for cognitive impairment and dementia after stroke: a systematic review and meta-analysis

Background Cognitive impairment and dementia are highly prevalent among stroke survivors and represent a major burden for patients, carers, and health-care systems. We studied the risk factors for post -stroke cognitive impairment (PSCI) and dementia (PSD) beyond the well established risk factors of age and stroke severity. Methods In this systematic review and meta -analysis we conducted a systematic literature search from database inception until Sept 15, 2023. We selected prospective and retrospective cohort studies, post -hoc analyses from randomised controlled trials, and nested case -control studies of patients with acute stroke (ischaemic, haemorrhagic, and transient ischaemic attack), exploring associations between risk factors at baseline and PSCI or PSD over a follow-up period of at least 3 months. Study quality was assessed using the Newcastle -Ottawa quality assessment scale. We calculated pooled relative risks (RRs) with random -effects meta -analyses and performed subgroup, metaregression, and sensitivity analyses. This study was preregistered with PROSPERO, CRD42020164959. Findings We identified 162 eligible articles for our systematic review, of which 113 articles (89 studies, 160 783 patients) were eligible for meta -analysis. Baseline cognitive impairment was the strongest risk factor for PSCI (RR 2 center dot 00, 95% CI 1 center dot 66-2 center dot 40) and PSD (3 center dot 10, 2 center dot 77-3 center dot 47). We identified diabetes (1 center dot 29, 1 center dot 14-1 center dot 45), presence or history of atrial fibrillation (1 center dot 29, 1 center dot 04-1 center dot 60), presence of moderate or severe white matter hyperintensities (WMH; 1 center dot 51, 1 center dot 20-1 center dot 91), and WMH severity (1 center dot 30, 1 center dot 10-1 center dot 55, per SD increase) as treatable risk factors for PSCI, independent of age and stroke severity. For PSD, we identified diabetes (1 center dot 38, 1 center dot 10-1 center dot 72), presence of moderate or severe WMH (1 center dot 55, 1 center dot 01-2 center dot 38), and WMH severity (1 center dot 61, 1 center dot 20-2 center dot 14, per SD increase) as treatable risk factors. Additional risk factors included lower educational attainment, previous stroke, left hemisphere stroke, presence of three or more lacunes, brain atrophy, and low baseline functional status. Associations of risk factors with PSD were weaker in studies conducted and published more recently. We found substantial interstudy heterogeneity and evidence of reporting bias. Interpretation Our results highlight the importance of cognitive impairment in the acute phase after stroke for longterm prediction of PSCI and PSD. Treatable risk factors include diabetes, atrial fibrillation, and markers of cerebral small vessel disease (ie, white matter hyperintensities and lacunes). Future trials should explore these risk factors as potential targets for prevention of PSCI and PSD.