OBJECTIVE Recent studies in nondiabetic kidney diseases suggest that dietary supplementation with n-3 polyunsaturated fatty acids (fish oil) may have beneficial effects on albuminuria, kidney function, arterial blood pressure, and dyslipidemia. Therefore, we evaluated the long-term effect of fish oil in diabetic nephropathy. METHODS A 1-year double-blind randomized controlled study comparing fish oil (4.6 g n-3 fatty acids/day) with placebo (olive oil) was performed in an outpatient clinic in a tertiary referral center. Thirty-six normotensive IDDM patients with diabetic nephropathy were included; 18 were treated with fish oil. Seven patients dropped out (four received fish oil), and results for the remaining 29 are presented. Albuminuria (enzyme immunoassay), glomerular filtration rate (51Cr-labeled EDTA plasma clearance), 24-h ambulatory blood pressure, and lipid profile were determined every 6 months. RESULTS Albuminuria increased by 22% (1-46%) (mean [95% CI]) in the fish oil group vs. 15% (-11-49%) in the placebo group (NS). Glomerular filtration rate decreased from 116 +/- 7 to 105 +/- 7 ml.min-1.1.73 m-2 (mean +/- SE) vs. from 108 +/- 6 to 103 +/- 7, fish oil and placebo, respectively (NS). No significant changes occurred in 24-h ambulatory blood pressure: from 141 +/- 4/82 +/- 2 mmHg to 142 +/- 5/83 +/- 2 vs. from 140 +/- 4/78 +/- 2 to 144 +/- 4/80 +/- 3, fish oil and placebo, respectively (NS). In the fish oil group, serum triglycerides (median [range]) decreased from 0.97 (0.5-4.0) mmol/l to 0.8 (0.4-3.0) vs. from 1.01 (0.4-2.0) to 1.09 (0.4-2.0) in the placebo group (P < 0.05) and VLDL cholesterol decreased from 0.45 (0.23-1.88) to 0.37 (0.21-1.43) mmol/l vs. from 0.44 (0.21-0.94) to 0.41 (0.17-1.94) (P < 0.05), but total and LDL cholesterol rose in the fish oil compared with the placebo group. CONCLUSIONS Our study does not suggest that fish oil has beneficial effects on albuminuria, kidney function, blood pressure, and dyslipidemia in normotensive IDDM patients suffering from diabetic nephropathy.