Objective: To describe the current use of treatments to prevent or treat patent ductus arteriosus (PDA) in preterm infants, examine the association between different treatment strategies and neonatal outcomes and review the variation in these practices between centers.Study design: Cohort study of infants born between 23 and 30 weeks gestation managed by the Pediatrix Medical Group from 1997 to 2004. We collected data on demographics, indomethacin and ligation, and outcomes of the following five groups: prophylactic indomethacin treatment: infants treated with indomethacin on day of life (DOL) 0 or 1; indicated indomethacin treatment: infants treated with indomethacin after DOL 1; PDA without treatment: infants with a PDA without report of treatment; ligation only: infants with a PDA ligation without use of indomethacin and no PDA: infants without a PDA and without treatment.Results: There were 6189 (18%) patients who received prophylactic indomethacin, 5690 (16%) patients received indicated treatment, 3886 (11%) patients had a PDA without treatment, 702 (2%) patients received ligation only and 18 136 (52%) patients had no PDA. In multivariate analysis, mortality among survivors to 2 days of age was lower (odds ratio (OR) 0.6, 95% confidence interval (CI) 0.5 to 0.7, P<0.01) and chronic lung disease, isolated intestinal perforation and severe retinopathy of prematurity (stages 3 and 4) were higher (OR 1.5, 95% CI 1.3 to 1.6, P<0.01; OR 1.5, 95% CI 1.1 to 2.0, P<0.01 and 1.4, 95% CI 1.2 to 1.6, P<0.01, respectively) in the indicated treatment group compared with the PDA without treatment group. The proportion of infants receiving prophylactic indomethacin among all infants and infants receiving indicated treatment among neonates with a report of a PDA varied by site from 0 to 59% (median 9.5%) and 0 to 100% (median 62%), respectively.Conclusions: Indomethacin use for intraventricular hemorrhage prevention and/or treatment of a PDA is common, but the selection of infants for treatment, and the decision of when and how to treat vary widely between centers. Our findings suggest the need for randomized, placebo-controlled trials of the effect of treatment of the PDA in preterm infants.