BACKGROUND:

The incidence of patent ductus arteriosus is higher in premature infants than in full-term infants and is highest in premature infants who have respiratory distress syndrome. It is generally agreed that intervention in an asymptomatic infant with a small left-to-right shunt is unnecessary, since the patent ductus almost invariably closes spontaneously and thus does not require surgery. A few infants will demonstrate signs of a large shunt during the course of respiratory distress syndrome. Many of these infants will improve with medical management of congestive heart failure, but others require surgical closure. A third group of babies with respiratory distress have severe progressive pulmonary disease requiring ventilatory support. There was disagreement as to whether elimination of the patent ductus in these infants resulted in decreased mortality. A variety of therapeutic approaches was being used, and there was no convincing evidence of the superiority of one treatment over another.

DESIGN NARRATIVE:

Trial A was a randomized, double-blind trial in which indomethacin plus usual medical therapy was compared with a placebo plus medical therapy. Where this regimen was unsuccessful, the code was broken, and infants who received indomethacin were treated surgically. Infants who had received placebo in Trial A were entered, if there were no contraindications to indomethacin, into Trial B. In Trial B, infants were randomized to surgery or indomethacin therapy. Those in whom indomethacin treatment was unsuccessful were treated surgically. The Recruitment and Intervention Phase began in April 1979. All patients were enrolled by March 31, 1981, and followed for one year after enrollment.