Background: Studies in animals and in humans have suggested that docosahexaenoic acid (DHA), an n-3 long-chain polyunsaturated fatty acid, might reduce the risk of bronchopulmonary dysplasia, but appropriately designed trials are lacking. Methods: We randomly assigned 1273 infants born before 29 weeks of gestation (stratified according to sex, gestational age [<27 weeks or 27 to <29 weeks], and center) within 3 days after their first enteral feeding to receive either an enteral emulsion providing DHA at a dose of 60 mg per kilogram of body weight per day or a control (soy) emulsion without DHA until 36 weeks of postmenstrual age. The primary outcome was bronchopulmonary dysplasia, defined on a physiological basis (with the use of oxygen-saturation monitoring in selected infants), at 36 weeks of postmenstrual age or discharge home, whichever occurred first. Results: A total of 1205 infants survived to the primary outcome assessment. Of the 592 infants assigned to the DHA group, 291 (49.1% by multiple imputation) were classified as having physiological bronchopulmonary dysplasia, as compared with 269 (43.9%) of the 613 infants assigned to the control group (relative risk adjusted for randomization strata, 1.13; 95% confidence interval [CI], 1.02 to 1.25; P = 0.02). The composite outcome of physiological bronchopulmonary dysplasia or death before 36 weeks of postmenstrual age occurred in 52.3% of the infants in the DHA group and in 46.4% of the infants in the control group (adjusted relative risk, 1.11; 95% CI, 1.00 to 1.23; P = 0.045). There were no significant differences between the two groups in the rates of death or any other neonatal illnesses. Bronchopulmonary dysplasia based on a clinical definition occurred in 53.2% of the infants in the DHA group and in 49.7% of the infants in the control group (P = 0.06). Conclusions: Enteral DHA supplementation at a dose of 60 mg per kilogram per day did not result in a lower risk of physiological bronchopulmonary dysplasia than a control emulsion among preterm infants born before 29 weeks of gestation and may have resulted in a greater risk.