1. In epidemiological studies microalbuminuria, i.e. slightly elevated urinary albumin excretion rate, predicts increased atherosclerotic vascular morbidity and mortality. This study aimed to test the hypothesis that microalbuminuria in clinically healthy subjects is associated with a systemic transvascular albumin leakiness. In animal experiments the outflux of albumin and lipids to the arterial wall are highly correlated, and both are elevated in atherosclerosis. 2. All participants were recruited at random from a population-based epidemiological study, where the upper decile of urinary albumin excretion rate was 6.6 micrograms/min. Twenty-seven patients with persistent microalbuminuria (urinary albumin excretion rate 6.6-150 micrograms/min), and 56 age- and sex-matched control subjects with persistent normoalbuminuria (UAER < or = 6.6 micrograms/min) were studied. 3. The systemic transvascular albumin leakage was measured as the fractional disappearance rate of 125I-labelled albumin from the total plasma compartment in 1 h after intravenous injection. 4. The fractional disappearance rate of albumin from the plasma compartment was higher in the microalbuminuric than in the normoalbuminuric group [5.8 (95% confidence interval 5.3-6.2; n = 27) versus 5.0 (4.6-5.5; n = 56)%/h, P < 0.05]. The positive correlation between urinary albumin excretion rate on continuous scale (logarithmically transformed) and the fractional disappearance rate of albumin from the plasma compartment [slope 0.4 (95% confidence interval 0.1-0.7; n = 83), r = 0.29, P < 0.005] was independent of age, sex, smoking status, blood pressure, body size, plasma volume, plasma albumin concentration and concentrations of blood glucose, serum insulin and serum lipids. 5. In conclusion, microalbuminuria is an independent marker of systemic transvascular albumin leakiness in clinically healthy subjects.(ABSTRACT TRUNCATED AT 250 WORDS)