ABSTRACT
INTRODUCTION: This meta-analysis aimed to explore the association of perivascular spaces (PVS) burden with the risks of future stroke events and mortality in patients with ischemic stroke and transient ischemic attack (TIA). METHODS: We systematically searched PubMed, Embase, and Cochrane database from inception to December 31, 2023. We included eligible studies that reported adjusted estimated effects for future intracranial hemorrhage (ICH), ischemic stroke, and mortality with baseline PVS burden in patients with ischemic stroke and TIA. Data were pooled using an inverse-variance method for the fixed effects (FE) model and a restricted maximum likelihood method for the random effects (RE) model. RESULTS: Thirteen observational studies (5 prospective, 8 retrospective) were included, comprising 20,256 patients. Compared to 0-10 PVS at basal ganglia (BG-PVS), a higher burden (>10) of BG-PVS was significantly associated with an increased risk of future ICH (adjusted hazards ratio [aHR] 2.79, 95% confidence interval [CI]: 1.16-6.73, RE model; aHR 2.14, 95% CI: 1.34-3.41, FE model; I2 = 64%, n = 17,084 from four studies) followed up for at least 1 year. There was no significant association between >10 BG-PVS and ICH within 7 days after reperfusion therapy (adjusted odds ratio [aOR] 1.69, 95% CI: 0.74-3.88, RE model; aOR 1.43, 95% CI: 0.89-2.88, FE model; I2 = 67%, n = 1,176 from four studies). We did not detect a significant association of recurrent ischemic stroke, mortality, or disability with BG-PVS burden. Neither >10 PVS at centrum semiovale (CSO-PVS) nor increasing CSO-PVS burden was significantly associated with the risk of future intracranial hemorrhage or ischemic stroke recurrence. CONCLUSIONS: Current evidence suggests that a higher BG-PVS burden may be associated with an increased risk of future ICH in patients with ischemic stroke and TIA.