Our study compared the effects of an angiotensin-converting enzyme inhibitor (captopril) versus a calcium antagonist (nifedipine) on proteinuria and renal function in patients with diabetic nephropathy. A randomized follow-up study was designed. Type 2 diabetic patients, with established diabetic nephropathy (proteinuria > 0.5 g/24 h), were treated with nifedipine (10 patients, group A) or captopril (10 patients, group B) for 6 months. Arterial blood pressure, metabolic parameters, proteinuria and renal function were measured and compared. Mean percentage differences for glomerular filtration rate, renal plasma flow and filtration fraction between the two groups were calculated. No significant differences were observed in serum glucose, glycosylated hemoglobin (hemoglobin A(1c)), Na+, K+ or albumin in either group or between groups. Blood pressure decreased significantly with both treatments and mean blood pressure was significantly lower in group A compared with group B at 6 months (Mann-Whitney U-test, P=0.03). Proteinuria was similar in both groups at randomization, but after 3 and 6 months of treatment significant reductions were observed only in the group treated with captopril (P<0.01). A significant decrease in filtration fraction was observed in group B with an increase in group A (Mann-Whitney U-test, P=0.03). Multiple regression analysis identified the therapeutic agent administered as an independent variable for decrease in proteinuria. It is concluded that antihypertensive treatment with captopril, but not with nifedipine, reduced proteinuria in patients with diabetic nephropathy, although a better mean blood pressure was obtained with nifedipine. The difference in filtration fraction variation between the two groups suggests an opposite pharmacological action on glomerular hemodynamics and perhaps an explanation for the results on proteinuria of the different drugs.