OBJECTIVE: Microalbuminuria and impaired endothelium-dependent vasodilation are both predictors for cardiac events in patients with type 2 diabetes. The aim of the study was to evaluate whether microalbuminuria correlated with coronary endothelium-dependent vasodilation. RESEARCH DESIGN AND METHODS: We evaluated 84 patients (47 men, mean age 50.5 +/- 5.9 years) with type 2 diabetes for 9.4 +/- 3.4 years, without angiographic coronary stenosis and without major cardiovascular risk factors or other confounding factors, for endothelium investigation. Quantitative coronary angiography was used to assess coronary artery response to cold pressor testing, used to assess endothelium-dependent vasodilation, and to isosorbide dinitrate (endothelium-independent vasodilation). RESULTS: Endothelium-dependent vasodilation differed in the patients with and without microalbuminuria (changes in coronary artery diameter during cold pressor testing: -15.0 +/- 1.9% vs. -10.2 +/- 1.3%, respectively, P < 0.05) and correlated with urinary albumin excretion rate (r = -0.39, P = 0.003), diastolic blood pressure (r = 0.29, P < 0.01), and left ventricular mass index (r = -0.24, P < 0.05). Independent predictors for endothelium-dependent vasodilation were urinary albumin excretion rate (beta -0.04 [95% CI -0.07 to -0.01], P < 0.005) and left ventricular mass index (-0.26 [-0.49 to -0.05], P < 0.05). Endothelium-independent vasodilation was similar in both groups. CONCLUSIONS: Type 2 diabetic patients with microalbuminuria have a more severely impaired coronary endothelium-dependent vasodilation than those with normoalbuminuria. These data suggest a common pathophysiological process for both coronary vasomotor abnormalities and microalbuminuria.