Purpose: Racial minorities have a higher prevalence of vascular risk factors (VRF), as well as higher incidence and greater severity of stroke. Enlarged perivascular spaces (EPVS) are considered a marker of small vessel disease reflecting the long term effect of VRF. We aim to determine the prevalence and burden of EPVS in racial minorities, and examine their association with VRF and stroke severity.
Methods: We included 894 consecutive stroke patients who underwent brain MRI admitted to an academic stroke center. EPVS were rated in the basal ganglia (BG), centrum semiovale (CSO) and midbrain (MB) following a standardized method, and dichotomized for analyses (mild <20 vs severe ≥20). Race was assessed based on US census procedures and categorized as White, Black or Other racial groups (ORG). Stroke severity was determined at time of admission using the NIH stroke scale. We used univariate analysis to assess the relation of VRF and EPVS severity, and multivariable logistic regression analysis to relate EPVS and stroke severity.
Results: EPVS were present in the entire sample (219 White, 455 Black, 220 ORG). Severe EPVS prevalence for Whites, Blacks and ORG was for BG 19.6%, 18%, and 15.4%; for CSO, 38.8%, 42.6%, and 33.6%; and for MB, 88.6%, 84.8%, and 84.1%, respectively. In univariate analyses, increasing age was associated with severe EPVS in the BG and CSO in all racial groups (p=< .0001); White and Black women were more likely to have severe EPVS in the BG and CSO (p<0.05). Hypertension was associated with EPVS severity in the BG in the entire sample (p=0.0004), with severe EPVS in MB and BG among Blacks, and with severe EPVS in BG and CSO in ORG (p<0.0001). Dyslipidemia was associated with severity of EPVS in BG only in ORG (p=0.01), while diabetes showed an inverse relation with EPVS severity in BG among Blacks only (p=0.003). EPVS severity was not associated with stroke severity when evaluating different NIHSS thresholds.
Conclusion: Race was not related to EPVS severity in our study, but traditional risk factors were, supporting their role as markers of small vessel disease. Severe EPVS were more frequent in MB, CSO and less common in BG. Further studies are required to clarify the relation of EPVS with stroke severity, assessing other measures and functional outcomes.