Oxygen Saturation Targeting for Infants with Bronchopulmonary Dysplasia: A Pilot Randomized Trial

Rationale: The optimal target oxygen saturation (SpO2) range in infants with established bronchopulmonary dysplasia (BPD) is unknown. Objectives: To compare the incidence of intermittent hypoxemia (IH), proportion of time with hypoxemia, and secondary clinical outcomes measured up to 6 months corrected age (CA) in infants with established BPD randomized to higher (⩾96%) versus lower (90-94%) SpO2 target ranges. Methods: Fifty infants born at <30 weeks gestational age who received supplemental respiratory support at 36 weeks postmenstrual age (PMA) were randomized before 44 weeks PMA to higher (n = 22) or lower (n = 28) SpO2 target ranges. Continuous pulse oximetry data were analyzed weekly to guide titration of respiratory support until 6 months CA. Primary outcomes were the incidence of IH (SpO2 < 80% for ⩾30 s) and proportion of time with hypoxemia (<80%) over the entire study period. Secondary outcomes were hypoxemia defined using alternative durations (⩾10 and ⩾60 s) and SpO2 thresholds (<90%) and clinical and developmental outcomes assessed through 6 months CA. Post hoc analyses compared rates of hypoxemia between the two study groups from enrollment to 48 weeks PMA among infants with at least 4 weeks of study data. Results: Median duration of monitoring was 19.0 (interquartile range [IQR], 8.5-23.0) weeks, yielding 835 (IQR, 412-1,269) hours of data per participant. Over the entire study period, there was no difference between SpO2 target groups in the primary outcomes of median numbers of IH events <80% for ⩾30 seconds or time with SpO2 < 80%. Post hoc analyses of infants with at least 4 weeks of study data demonstrated higher incidence of IH events <80% and <90% for both ⩾60 and ⩾30 seconds between enrollment and 48 weeks PMA in the lower target group. Infants in the lower SpO2 target group were discharged at later PMA than infants in the higher SpO2 target group (median, 48.0 vs. 45.0 wk; P = 0.05). Conclusions: A higher (⩾96%) compared with lower (90-94%) SpO2 target strategy is unlikely to significantly reduce hypoxemia between 36-44 weeks PMA and 6 months CA. A possible decrease in IH before 48 weeks PMA and modest clinical improvements associated with the higher target range will require confirmation in future studies. Clinical trial registered with www.clinicaltrials.gov (NCT03385330).