Reported are 1-year follow-up results of a randomized clinical trial comparing three strategies of managing clinically significant patent ductus arteriosus at the time of diagnosis in premature infants: (1) immediate administration of a three-dose course of intravenously administered indomethacin in addition to usual medical therapy (fluid restriction and use of diuretics or digitalis or both), with surgery as a backup measure, (2) usual medical therapy alone initially, with indomethacin as the first and surgery as the final backup measure, and (3) usual medical therapy alone initially, with surgery alone as backup. Of primary concern were the relative merits of these three managements strategies in the terms of the long-term occurrence of a wide range of health problems. Although at the time of neonatal hospitalization there was a significant excess of bleeding episodes in infants receiving indomethacin as part of initial treatment, and a significantly higher rate of retrolental fibroplasia in the those given usual medical therapy with surgery as backup, there were no statistically significant differences at 1 year of age related to these intermediate outcomes. In other regards, too, the treatment strategies appeared interchangeable in terms of the 1-year outcome.