Obesity has been associated with increased F2-isoprostane (F2-IsoP) levels cross-sectionally. However, the prospective association may be inverse, based on our earlier finding that elevated urinary F2-IsoP levels predict lower risk of diabetes. This earlier finding led us to hypothesize that urinary F2-IsoPs reflect the intensity of oxidative metabolism and as such predict lower risk of both diabetes and weight gain. We examined cross-sectional relationships with obesity and prospective relationships with weight gain using the data from 299 participants of the Insulin Resistance Atherosclerosis Study (IRAS), all of whom were free of diabetes at baseline. Four urinary F2-IsoPs were assayed in stored baseline urine samples using liquid chromatography with tandem mass spectrometry: iPF(2α)-III, 2,3-dinor-iPF(2α)-III, iPF(2α)-VI, and 8,12-iso-iPF(2α)-VI (F2-IsoP 1–4, respectively). Baseline F2-IsoPs were positively associated with baseline measures of obesity; the strongest associations were found with two F2-IsoPs: odds ratios (95% confidence intervals) for overall and abdominal obesity were 1.74 (1.26–2.40) and 1.63 (1.18–2.24) for F2-IsoP2 and 1.47 (1.12–1.94) and 1.64 (1.22–2.20) for F2-IsoP4. F2-IsoP2 showed the strongest and significant inverse association with weight gain during the 5-year follow-up period: increase in F2-IsoP2 equal to 1 s.d. was associated with 0.90 kg lower weight gain (P = 0.02) and the odds ratios for relative (≥5%) and absolute (≥5 kg) weight gain were 0.67 (0.47–0.96) and 0.57 (0.37–0.87), respectively. The other three F2-IsoPs were consistently inversely associated with weight gain, although not significantly, suggesting that different F2-IsoPs vary in their ability to detect the association with weight gain.