Initial Oxygen Use for Preterm Newborn Resuscitation: A Systematic Review With Meta-analysis

CONTEXT: The International Liaison Committee on Resuscitation prioritized to review the initial fraction of inspired oxygen (F-IO2) during the resuscitation of preterm newborns. OBJECTIVES: This systematic review and meta-analysis provides the scientific summary of initial F-IO2 in preterm newborns (< 35 weeks' gestation) who receive respiratory support at birth. DATA SOURCES: Medline, Embase, Evidence-Based Medicine Reviews, and Cumulative Index to Nursing and Allied Health Literature were searched between January 1, 1980 and August 10, 2018. STUDY SELECTION: Studies were selected by pairs of independent reviewers in 2 stages with a Cohen's. of 0.8 and 1.0. DATA EXTRACTION: Pairs of independent reviewers extracted data, appraised the risk of bias (RoB), and assessed Grading of Recommendations Assessment, Development and Evaluation certainty. RESULTS: Ten randomized controlled studies and 4 cohort studies included 5697 patients. There are no statistically significant benefits of or harms from starting with lower compared with higher F-io2 in short-term mortality (n = 968; risk ratio = 0.83 [95% confidence interval 0.50 to 1.37]), long-term mortality, neurodevelopmental impairment, or other key preterm morbidities. A sensitivity analysis in which 1 study with a high RoB was excluded failed to reveal a reduction in mortality with initial low F-io2 (n = 681; risk ratio = 0.63 [95% confidence interval 0.38 to 1.03]). LIMITATIONS: The Grading of Recommendations Assessment, Development and Evaluation certainty of evidence was very low for all outcomes due to RoB, inconsistency, and imprecision. CONCLUSIONS: The ideal initial F-io2 for preterm newborns is still unknown, although the majority of newborns >= 32 weeks' gestation will require oxygen supplementation.