Targeted Oxygen for Initial Resuscitation of Preterm Infants

Importance: The most effective initial fraction of inspired oxygen (Fio2) for resuscitating preterm newborns is unknown. Objective: To compare outcomes of newborns born at 23 to 28 weeks' gestation resuscitated with initial Fio2 of 0.6 vs 0.3. Design, Setting, and Participants: Randomized clinical trial conducted in 31 maternity hospitals in 6 countries. Consent by waiver was obtained in Australia, certain institutions in India, and Malaysia; prospective informed consent was obtained in certain institutions in India and all institutions in Singapore, Spain, and the US. Infants due at 23 to 28 weeks' gestation were randomized shortly before birth. Those with congenital abnormalities affecting oxygenation, neurodevelopment, or survival were excluded. Randomization was conducted from September 2018 to September 2024, with follow-up expected to close in September 2026. Intervention: Infants were randomized (1:1) to receive an initial Fio2 of 0.6 or 0.3; Fio2 was titrated to meet standard targets for oxygen saturation by pulse oximetry in the first 10 minutes or for clinical needs. Clinicians and those assessing outcomes were not blinded to group assignment. Main Outcomes and Measures: The primary outcome was death and brain injury at 36 weeks' corrected gestational age; secondary outcomes were the individual components of the primary outcome. Results: A total of 1641 newborns were randomized. The primary analysis included 728 newborns randomized to receive an Fio2 of 0.6 and 741 to an Fio2 of 0.3 after excluding 172 newborns, mostly for birth after 28 weeks' gestation and transfer to another hospital before birth (54% female). Rates of escalation to Fio2 of 1.0 were similar between the groups (Fio2 of 0.6: 41%; Fio2 of 0.3: 38%). Primary outcome information was ascertained in 1423 newborns (96.9%). Death or brain injury at 36 weeks' corrected gestational age occurred in 330 of 703 newborns (46.9%) assigned to the Fio2 of 0.6 group vs 344 of 720 (47.8%) assigned to the Fio2 of 0.3 group (relative risk, 0.98 [95% CI, 0.89-1.09]). Conclusions and Relevance: Initiating resuscitation of preterm infants with Fio2 of 0.6 vs 0.3 did not affect the risk of death or brain injury by 36 weeks' corrected gestational age. These results lay a foundation for future trials evaluating the effectiveness and safety of using higher initial Fio2 levels for preterm infant resuscitation. Trial Registration: ANZCTR Identifier: ACTRN12618000879268.