Introduction:
Enlarge perivascular spaces (EPVS) are a common finding on MRI with little known about their aetiology or clinical significance. Associations of EPVS with white matter hyperintensities (WMH) and cognitive impairment have been reported. We assessed the prevalence of EPVS in the basal ganglia (BG) and centrum semiovale (CS) and associations with vascular risk factors, MRI abnormalities, and cognitive impairment in a large cohort of MRI documented lacunar stroke patients.
Methods:
All SPS3 participants enrolled in North America (N=1632) were included in the cohort. MRIs were obtained at the time of index stroke. We graded severity of EPVS on T2 in each of the right and left BG and CS as: <10, 11-20, or 20+. We assessed relationships between maximum (max) BG and CS scores and patient demographics, number of subcortical infarcts, WMH, and cognitive impairment using chi-square tests and ANOVA. Logistic regression was used to identify independent associations between max score in the BG or CS (<10 vs. 11+ points) and these factors.
Results:
Of the 1632 MRIs, 1172 had T2-imaging available. EPVS was symmetrical in right and left sides of BG and of CS so max score in each area were used for analyses. In 42% of BG and 69% of CS, the max number of unilateral EPVS was ≥11. Patients with more EPVS in the BG were older (mean 59 vs. 63 vs. 70 years, p < 0.001), had hypertension (75% vs. 84% vs. 86% p < 0.001), and lower eGFR (mean 83 vs. 80 vs. 74 ml/min/1.73m
2
, p = 0.006). More EPVS in the BG were associated with WMH severity as measured by Age Related White Matter Changes total score (median 3 vs. 6 vs. 8, p < 0.001). Patients with more EPVS in the BG more often had multiple infarcts (11% vs. 19% vs. 29%, p<0.001). All findings were similar in the CS. Age (OR 1.9 per each 10 years, 95% CI 1.7-2.1), history of hypertension (OR 1.7, 95% CI 1.2-2.3) and multiple infarcts (OR 2.4, 95% CI 1.7-3.4) were independently associated with a maximum unilateral EPVS score of 2+ in the BG. Mild cognitive impairment was not associated with EPVS in BG (p=0.3) or CS (p=0.5).
Conclusions:
In this well-defined large cohort of lacunar stroke patients, BG EPVS were associated with age, hypertension and multiple infarcts. These findings suggest that BG EPVS share similar risk factors with lacunar stroke and may be a marker for small vessel disease. Unexpectedly, EPVS were not associated with cognitive impairment.
The influence of nicergoline on cerebral, systemic and cardiac hemodynamics in patients who have had an ischemic stroke
