Background: Although inhaled nitric oxide (iNO) causes selective pulmonary vasodilation and improves oxygenation in newborn infants with persistent pulmonary hypertension, its effects are variable. We hypothesized (1) that the response to iNO therapy is dependent on the primary disease associated with persistent pulmonary hypertension of the newborn (PPHN) and (2) that the combination of high-frequency oscillatory ventilation (HFOV) with iNO would be efficacious in patients for whom either therapy alone had failed. Methods: To determine the relative roles of iNO and HFOV in the treatment of severe PPHN, we enrolled 205 neonates in a randomized, multicenter clinical trial. Patients were stratified by predominant disease category: respiratory distress syndrome (n = 70), meconium aspiration syndrome (n = 58), idiopathic PPHN or pulmonary hypoplasia (excluding congenital diaphragmatic hernia) ('other': n = 43), and congenital diaphragmatic hernia (n = 34); they were then randomly assigned to treatment with iNO and conventional ventilation or to HFOV without iNO. Treatment failure (partial pressure of arterial oxygen [PaO2] <60 mm Hg) resulted in crossover to the alternative treatment; treatment failure after crossover led to combination treatment with HFOV plus iNO. Treatment response with the assigned therapy was defined as sustained PaO2 of 60 mm Hg or greater. Results: Baseline oxygenation index and PaO2 were 48 ± 2 and 41 ± 1 mm Hg, respectively, during treatment with conventional ventilation. Ninety-eight patients were randomly assigned to initial treatment with HFOV, and 107 patients to iNO. Fifty-three patients (26%) recovered with the initially assigned therapy without crossover (30 with iNO [28%] and 23 with HFOV [23%]; π = 0.33). Within this group, survival was 100% and there were no differences in days of mechanical ventilation, air leak or supplemental oxygen requirement at 28 days. Of patients whose initial treatment failed, crossover treatment with the alternate therapy was successful in 21% and 14% for iNO and HFOV, respectively (π = not significant). Of 125 patients in whom both treatment strategies failed, 32% responded to combination treatment with HFOV plus iNO. Overall, 123 patients (60%) responded to either treatment alone or combination therapy. By disease category, response rates for HFOV plus iNO in the group with respiratory syndrome and the group with meconium aspiration syndrome were better than for HFOV alone or iNO with conventional ventilation (p <0.05). Marked differences in outcomes were noted among centers (percent death or treatment with extracorporeal membrane oxygenation = 29% to 75%). Conclusions: We conclude that treatment with HFOV plus iNO is often more successful than treatment with HFOV or iNO alone in severe PPHN. Differences in responses are partly related to the specific disease associated with PPHN.