To determine whether the duration of indomethacin administration could be shortened in infants with good early constrictive response of patent ductus arteriosus (PDA).Infants born at< 30 weeks' gestational age were assessed with echocardiography in the first 12 hours of life and treated with indomethacin (0.1 mg/kg) if the PDA was >2 mm in diameter. Randomization occurred before the second dose to either standard treatment (2 more doses of indomethacin at 0.1 mg/kg irrespective of echocardiographic findings) or to echocardiographically directed duration of indomethacin treatment (ECHO; further doses only if the PDA was>1.6 mm). Serial echocardiography was performed to day 28 of age. The primary outcome was failure of PDA closure.The infants were randomized to either the ECHO arm (n=34) or the standard treatment arm (n=40). No differences between the arms were seen in terms of failure of PDA closure, PDA reopening, need for further doses of indomethacin, or need for surgical ligation. More doses of indomethacin were given in the standard treatment arm (median, 3 doses [range, 1 to 12] vs 1 dose [range, 1 to 15]; P < .0001).Echocardiographically directed duration of indomethacin treatment is effective in achieving PDA closure and offers the potential for dose minimization.