Indomethacin (Indo) is commonly used for treatment of patent ductus arteriosus (PDA) but has renal failure as a main side effect. Aspirin (ASA) is an alternative, but there are no controlled trials on its efficacy. We randomly assigned 75 premature infants suffering from respiratory distress syndrome (RDS) (mean gestational age: 29.6 +/- 2.5 wk, mean birth weight: 1295 +/- 464 g) (+/- SD) and on artificial ventilation at the start of the study (mean: 3.4 d of life), to either Indo (3 x 0.2 mg/kg/12 h) or ASA (4 x 15 mg/kg/6 h). PDA and degree of shunting were evaluated by echocardio-Doppler; side effects were carefully recorded. PDA closed in 35/38 patients from the Indo group (92%) and in 16/37 patients from the ASA group (43%) (p < 0.0001). Nineteen patients needed further treatment with Indo or surgery (17 in the ASA group and 2 in the Indo group). The only side effect observed was a decrease of uresis in the Indo group during 4 d post treatment (p < 0.01). Closing of PDA was positively correlated with gestational age, but not with time of starting Indo/ASA or grade of shunting. We conclude that ASA is not as effective in closing PDA as Indo, but has no adverse effect on uresis.