Despite the potential benefits of dietary treatment with marine omega 3 fatty acids in cardiovascular disease, there remains concern with respect to their potential for increased lipid peroxidation. Thus far, data from in vivo studies are inconclusive. Increased lipid peroxidation has also been associated with acute exercise in some studies, but the methods have been nonspecific. The quantitation of F-2-isoprostanes provides a more reliable and useful assessment of in vivo lipid peroxidation. We therefore aimed to assess the independent and combined effects of dietary omega 3 fatty acids and aerobic exercise training on urinary F-2-isoprostane levels in dyslipidemic non-insulin-dependent diabetic (NIDDM) patients. In a randomized controlled trial, 55 untrained, sedentary, dyslipidemic NIDDM patients were randomly assigned to a low-fat diet (30% of daily energy) with or without one daily fish meal (3.6 g omega 3 fatty acids per day) and further randomized to either a moderate (55% to 65% maximal oxygen consumption [(V) over doto(2)max]) or light (heart rate <100 bpm) exercise training program for 8 weeks. Twenty-four-hour urine samples from 49 subjects were collected for measurement of urinary F-2-isoprostanes by gas chromatography-mass spectrometry before and after intervention. The fish diets reduced urinary F-2-isoprostanes by 830 +/- 321 pmol/24 h (20%, P = .013) relative to the low-fat diet alone. This effect was independent of age, gender, and body weight change. Moderate exercise training did not alter F2-isoprostanes. These findings show that, at least in the short-term, exercise had no effect, whereas the inclusion of regular fish meals as part of a low-fat diet reduced in vivo lipid peroxidation in dyslipidemic NIDDM patients. This response could further complement the known benefits of omega 3 fatty acids and exercise favoring a reduced cardiovascular risk in diabetic patients. Copyright (C) 1999 by W.B. Saunders Company.