Abstract
BACKGROUND AND PURPOSE: Albuminuria is a marker for endothelial dysfunction and knowledge on its association with stroke and stroke subtypes are limited. METHODS: Corresponding data from 7261 participants of the population-based HUNT2 study (1995-1997) was linked with hospital records, identified all patients registered and diagnosed with a first-time stroke. Each diagnosis was validated by reviewal of the medical record appertaining to the individual. We then applied Cox proportional hazard models to estimate the hazard ratios (HRs) for the association between albuminuria (measured as albumin-to-creatinine-ratio, ACR) and diagnosis of stroke and stroke subtypes. RESULTS: 703 (9.7%) participants developed a first ischemic stroke during a median follow-up of 15 years. Higher albuminuria was associated with a higher rate for ischemic stroke and the risk rose steadily with increasing ACR (15% increment per unit increase in ACR concentration in mg/mmol). In the fully adjusted model, the HR for all ischemic strokes was 1.56 (95% CI 1.24-1.95) for those with an ACR ≥3 mg/mmol compared to participants with an ACR < 1 mg/mmol. Overall, increasing ACR was associated with a higher risk of all ischemic stroke subtypes. This was seen to be strongest for lacunar stroke (HR 1.75, CI 1.12-2.72, p = 0.019), and also for stroke of undetermined etiology (HR 1.53, CI 1.11-2.11, p = 0.009) and those caused by atherosclerosis in the large arteries (HR 1.51, CI 0.78-2.94, p = 0.186) than for cardio-embolic stroke (HR 1.22, CI 0.64-2.3, p = 0.518). CONCLUSIONS: Albuminuria is an important risk factor, potentially already at low grade, for ischemic stroke especially for lacunar subtype. Measuring albuminuria is both cheap and readily available. This offers the opportunity to evaluate the risk for endothelial dysfunction and thus the subsequent risk for stroke and cerebral small vessel disease.