Modeling Oxygen Saturation Targets, Mortality, and Retinopathy Among Extremely Preterm Infants

Area of Science:

• Neonatalogy
• Pediatric Ophthalmology
• Health Economics

Background:

• Higher oxygen saturation (SpO2) targets in extremely preterm infants may decrease mortality but increase the risk of treated retinopathy of prematurity (ROP).
• Clinical centers face a trade-off when selecting SpO2 target ranges, balancing survival benefits against potential ROP complications.

Purpose of the Study:

• To estimate the trade-offs between mortality and treated ROP associated with different SpO2 targeting strategies.
• To analyze how variations in baseline risks influence these trade-offs in extremely preterm infants.

Main Methods:

• Decision analytical modeling study utilizing US national birth data and Vermont Oxford Network data (2019-2023).
• Simulated an annual birth cohort of infants born extremely preterm (gestational age 23-27 weeks), incorporating gestational age and 16 performance subgroups.
• Compared high (91%-95%) versus low (85%-89%) SpO2 target ranges, calculating the risk trade-off ratio (RTOR) of treated ROP per death averted.

Main Results:

• Universal adoption of higher SpO2 targets was associated with an estimated 638 fewer deaths and 551 additional cases of treated ROP.
• The mean RTOR for higher SpO2 targets was 0.91, indicating less than one additional case of treated ROP per death averted.
• RTORs varied significantly by center risk profiles, ranging from 0.16 in high-mortality/low-ROP centers to 2.52 in low-mortality/high-ROP centers.

Conclusions:

• The risk-benefit trade-off between mortality and treated ROP during SpO2 targeting varies by gestational age and local baseline risks.
• Decision analysis can inform the selection of individualized SpO2 target ranges for extremely preterm infants in clinical settings.
• Selective adoption of SpO2 targets based on center-specific risks may optimize outcomes, balancing survival and ROP prevention.