BACKGROUND Microalbuminuria (MA) is thought to be a marker of widespread vascular damage. It is associated with increased mortality in diabetes mellitus, hypertension and acute myocardial infarction. The aim of the present study was to evaluate the prognostic significance of MA in non-diabetic acute stroke patients. MATERIAL AND METHODS We studied 52 patients (mean age 69.3 +/- 12.5 years) diagnosed with ischemic stroke confirmed by computed tomography, who were admitted to the Stroke Unit within 24 hours after the onset of symptoms. The control group consisted of 37 age- and gender-matched subjects (mean age 65.2 +/- 5.7 years), examined 3 to 18 months after ischemic stroke. We excluded patients with diabetes mellitus, positive urinalysis, proteinuria, hepatic or renal insufficiency, neoplastic disease or clinical signs of infection. The severity of the neurological deficit was assessed by the Scandinavian Stroke Scale (SSS). The albumin excretion rate was measured in daily urine collection on the second day of hospitalization, using the immunonephelometric method. The patients were followed up for three months. RESULTS MA was found in 24 of 52 (46.1%) acute stroke patients and in 5 of 37 (13.5%) controls (p<0.05). Patients with MA scored lower on the SSS than patients without MA, both on admission and later. We found a correlation between the daily excretion of albumin and the severity of neurological deficit on admission, as expressed by the SSS score (r = -0.48, p<0.05). The 90-day mortality rate was higher in patients with MA as compared to patients without MA (45.8% vs 7.1%). Patients with MA scored lower on the Barthel Index on Day 90 (median: 65 vs 100, p<0.01). CONCLUSIONS We found that MA can be detected in about 46% of non-diabetic patients with acute ischemic stroke. Measuring the albumin excretion rate may be a reliable predictor of increased mortality 3 months after stroke.