To estimate the incremental effectiveness of higher vs lower oxygen saturation target ranges for extremely preterm infants and evaluate how local baseline risks may influence the risk-benefit trade-off.
Approach:
Study Design: A decision analytical modeling study using individual patient microsimulation based on data from the NeOProM Collaboration and Vermont Oxford Network.
Model Structure: Simulated infants were assigned gestational age and distributed into performance subgroups based on center-based risks of mortality and treated retinopathy of prematurity (ROP).
Outcomes: Outcomes included treated ROP, surgical necrotizing enterocolitis, chronic lung disease at 36 weeks, and mortality before discharge.
Key Findings:
No difference in primary composite outcome of death or disability at 18-24 months between higher (91%-95%) and lower (85%-89%) oxygen saturation target ranges as reported in the NeOProM trials.
Higher oxygen saturation target range was associated with decreased risks of surgical necrotizing enterocolitis and mortality according to the study findings.
Lower oxygen saturation target range was associated with decreased risk of treatment for retinopathy of prematurity and chronic lung disease at 36 weeks as indicated by the results.
Interpretation:
Trade-offs between risks and benefits of different oxygen saturation targets should be informed by individual patient and center risks as indicated by the study.
Limitations:
Baseline risks differing from the NeOProM trials could change expected treatment effects.
Risks for important neonatal outcomes vary greatly between centers and over time.
Conclusion:
The study provides insights into the effectiveness of oxygen saturation targets in extremely preterm infants, emphasizing the need for individualized approaches based on local risks.
