Background Leukoaraiosis is associated to age and hypertension, but the evidence is conflicting as to any relation with other vascular risk factors. We assessed clinical and laboratory associations and prognostic significance of leukoaraiosis in a population-based study of incident stroke and TIA.

Methods Presence of leukoaraiosis on 1775 consecutive CT of TIA or stroke patients was related to aetiological subtype of TIA or stroke (TOAST), vascular risk factors, inflammatory, thrombotic or neuronal blood biomarkers, and its prognostic significance was calculated for death, myocardial infarction and recurrent stroke.

Results Univariate and multivariate analyses showed association with age (OR 2.52, 2.37–2.84, p<0.0001), HTN (1.30, 1.05–1.62, p=0.02), and current smoking (1.57, 1.13–2.18, p=0.01). Leukoaraiosis was strongly associated with lacunar type of stroke (age and sex-adjusted OR=1.73, 1.26–2.7, p=0.001), and inversely associated with the cardioembolic type (0.67, 0.51–0.87, p=0.001). After adjusting for age and sex, we only found very weak associations with pro-thrombotic or inflammatory biomarkers, most noticeably CRP, NSE, vWF and Fibrinogen. Leukoaraiosis on CT independently predicted stroke-related or cardiac deaths (HR 1.46, 1.23–1.73, p<0.0001), but not myocardial infarction (1.03, 0.62–1.70, p=0.0.91) or recurrent stroke (1.11, 0.87–1.42, p=0.42).

Conclusions In our population-based cohort of strokes and TIA, leukoaraiosis on CT brain imaging was associated with age, pre-morbid hypertension and current smoking. Presence of leukoaraiosis on CT at entry in the study predicted fatal stroke and cardiac death, and this was not explained by any association with inflammatory or haemostatic blood biomarkers or by the association with type of incident stroke, as leukoaraiosis was associated with lacunar type of stroke, the least severe clinically.