At 33 hospitals in the United States, a double-blind, randomized clinical trial was performed to compare one versus three prophylactic doses of synthetic surfactant in 700 to 1100 gm inborn infants. All 826 infants received an initial prophylactic dose of surfactant within 30 minutes of birth. Subsequently 410 infants received two doses of placebo (air) 12 and 24 hours later, and 416 infants received two additional doses of surfactant. By the age of 28 days, 70 infants who received one dose of surfactant and 40 infants who received three doses were dead, a 43% relative reduction in the mortality rate (30 fewer deaths; p = 0.002). By the age of 1 year after term, 87 infants who received one dose and 62 infants who received three doses were dead: a 29% relative reduction in the mortality rate (25 fewer deaths; p = 0.027). Infants who received three doses of surfactant required significantly less oxygen and less mean airway pressure for the first week of life. Necrotizing enterocolitis (9 vs 25 infants; p = 0.005), and use of high-frequency ventilation (13 vs 26 infants; p = 0.037); pancuronium (43 vs 62 infants; p = 0.045); and leukocyte transfusions (0 vs 4 infants; p = 0.042) were less frequent in the three-dose group, but air leak, bronchopulmonary dysplasia, intraventricular hemorrhage, patent ductus arteriosus, pulmonary hemorrhage, and infections were not different. These results indicate that physiologic findings, mortality rates, and probably morbidity rates are improved by two additional prophylactic doses of synthetic surfactant. (J PEDIATR 1995;126:969-78)