Objectives: To compare high-frequency oscillatory ventilation (HFOV) and intermittent positive pressure ventilation (IPPV) as a primary ventilation mode in preterm infants with respiratory distress syndrome. Primary end points were survival and maintenance of the randomized ventilation mode. Study design: Prospective, multicenter, randomized clinical trial. Setting: Level III neonatal intensive care units at three university children's hospitals. Patients: Ninety-six premature infants (gestational age <32 weeks) randomly assigned to HFOV or IPPV within the first 2 hours of life. All patients received a natural surfactant. No differences were found between the study groups with respect to the demographic data or the severity of respiratory distress syndrome. Infants were stratified at randomization, by birth weight, into two groups: 750 to 1000 gm (n = 32) and 1001 to 1500 gm (n = 64). The centers involved complied with a study protocol that planned a reduction in respiratory pressures when the infant's oxygen requirement had reached a fractional concentration of inspired oxygen of 0.6. Results: Five patients in the HFOV group died, and eight patients did not respond to the randomized ventilation mode; whereas four patients in the IPPV group died, and nine were switched to HFOV. No differences were found in gas exchange or ventilator support over the first 72 hours. Premature infants with a birth weight < 1000 gm had a significantly shorter course to reach fractional concentration of inspired oxygen of 0.21 while receiving IPPV than those receiving HFOV (9.3 ± 4.5 days vs 27.5 ± 10.2 days, p = 0.01). No differences were found between the groups in extraalveolar air (HFOV, seven; IPPV, seven) and intracranial bleeding (HFOV, nine; IPPV, eight). Conclusion: After surfactant treatment, HFOV, as a primary ventilation mode in premature infants with respiratory distress syndrome, is as safe and efficacious as conventional ventilation.