Does pre-existing dementia lead to adverse outcomes after stroke? A systematic review and meta-analysis

Objective: The aim of this systematic review and meta-analysis was to assess the impact of pre-stroke dementia (PSD) on stroke outcomes. Methods: Embase, PubMed, Cochrane Central Register of Controlled Trials (CENTRAL), Web of Science, and Scopus were searched from inception till 15th February 2025. Outcomes pooled were short-term mortality (6 months), poor functional outcomes, recurrence, and discharge to home. Results: 20 studies were eligible. The pooled prevalence of PSD was 10.4% (95% CI, 7.5%–14.4%). Meta-analysis of crude and adjusted data showed a significant association between PSD and short-term mortality, long-term mortality, and poor functional outcomes. Pooled analysis showed decreased odds of discharge to home with PSD but without any significant effect on stroke recurrence. A separate analysis of ischemic stroke patients undergoing reperfusion therapy showed a significant association between PSD and short-term mortality on analysis of crude but not adjusted data. There was also an increased risk of symptomatic intra-cerebral hemorrhage (ICH) with PSD but the risk of all ICH did not achieve statistical significance. Most outcomes had high inter-study heterogeneity. Conclusions: Pre-stroke dementia is associated with increased mortality and poorer functional recovery after stroke, but this effect diminishes after adjustment for confounders. Dementia alone should not deter the use of reperfusion therapy, although careful monitoring for symptomatic ICH is advised. PSD should form an integral part of stroke assessment and prognostication and clinicians should incorporate dementia status into individualized treatment and rehabilitation plans, balancing the potential benefits and risks of aggressive therapies.