Data from: Association of white matter hyperintensity markers on MRI and long-term risk of mortality and ischemic stroke: The SMART-MR study

Objective To determine whether white matter hyperintensity (WMH) markers on MRI are associated with long-term risk of mortality and ischemic stroke. Methods We included consecutive patients with manifest arterial disease enrolled in the SMART-MR study. We obtained WMH markers (volume, type and shape) from brain MRI scans performed at baseline using an automated algorithm. During follow-up, occurrence of death and ischemic stroke was recorded. Using Cox regression, we investigated associations of WMH markers with risk of mortality and ischemic stroke, adjusting for demographics, cardiovascular risk factors and cerebrovascular disease. Results We included 999 patients (59±10 years; 79% male) with a median follow-up of 12.5 years (range 0.2-16.0 years). A greater periventricular or confluent WMH volume was independently associated with a greater risk of vascular death (hazard ratio (HR) 1.29, 95% CI 1.13-1.47 for a 1 unit increase in natural log-transformed WMH volume) and ischemic stroke (HR=1.53, 95% CI 1.26-1.86). A confluent WMH type was independently associated with a greater risk of vascular (HR=2.05, 95% CI 1.20-3.48) and nonvascular death (HR=1.65, 95% CI 1.01-2.73), and ischemic stroke (HR=4.01, 95% CI 1.72-9.35). A more irregular shape of periventricular or confluent WMH, as expressed by an increase in concavity index, was independently associated with a greater risk of vascular (HR=1.20, 95% CI 1.05-1.38 per standard deviation increase) and nonvascular death (HR=1.21, 95% CI 1.03-1.42), and ischemic stroke (HR=1.28, 95% CI 1.05-1.55). Conclusions WMH volume, type and shape are associated with long-term risk of mortality and ischemic stroke in patients with manifest arterial disease.Supplemental Figure e-1. Examples of white matter hyperintensities (WMH) on FLAIR images with corresponding visualizations in the automated algorithm