Population-based case-control study of white matter changes on brain imaging in transient ischemic attack and ischemic stroke.

BACKGROUND AND PURPOSE White matter changes (WMC) are a common finding on brain imaging and are associated with an increased risk of ischemic stroke. They are most frequent in small vessel stroke; however, in the absence of comparisons with normal controls, it is uncertain whether WMC are also more frequent than expected in other stroke subtypes. Therefore, we compared WMC in pathogenic subtypes of ischemic stroke versus controls in a population-based study. METHODS We evaluated the presence and severity of WMC on computed tomography and on magnetic resonance brain imaging using modified Blennow/Fazekas scale and age-related white matter changes scale, respectively, in a population-based study of patients with incident transient ischemic attack or ischemic stroke (Oxford Vascular Study) and in a study of local controls (Oxford Project to Investigate Memory and Ageing) without history of transient ischemic attack or ischemic stroke, with stratification by stroke pathogenesis (Trial of Org10172 in Acute Stroke Treatment classification). RESULTS Among 1601 consecutive eligible patients with first-ever ischemic events, 1453 patients had computed tomography brain imaging, 562 had magnetic resonance imaging, and 414 patients had both. Compared with 313 controls (all with computed tomography and 131 with magnetic resonance imaging) and after adjustment for age, sex, diabetes mellitus, and hypertension, moderate/severe WMC (age-related white matter changes scale) were more frequent in patients with small vessel events (odds ratio, 3.51 [95% confidence interval, 2.13-5.76]; P<0.0001) but not in large artery (odds ratio, 1.03 [95% confidence interval, 0.64-1.67]), cardioembolic (odds ratio, 0.87 [95% confidence interval, 0.56-1.34]), or undetermined (odds ratio, 0.90 [95% confidence interval, 0.62-1.30]) subtypes. Results were consistent for ischemic stroke and transient ischemic attack, for other scales, and for magnetic resonance imaging and computed tomography separately. CONCLUSIONS In contrast to small vessel ischemic events, WMC were not independently associated with other pathogenic subtypes, suggesting that WMC are unlikely to be an independent risk factor for nonsmall vessel events.