Automated oxygen control versus manual titration in preterm infants: A systematic review and meta-analysis.

OBJECTIVE

To evaluate whether closed-loop automated fraction of inspired oxygen (FiO2) control improves oxygen targeting and clinical outcomes compared with clinician-directed manual titration in preterm infants.

STUDY

DESIGN

Systematic review and random-effects meta-analysis of randomized trials searched through Feb 6, 2026. Crossover and parallel-group trials informed physiological outcomes; only longitudinal parallel-group trials informed clinical outcomes.

RESULTS

Twenty-three trials (1777 infants) were included. Automated control increased time within target range (mean difference 15.01 percentage points, 95% CI 10.90 to 19.13) and reduced time above target, below target, SpO2 < 80%, SpO2 > 98%, and manual FiO2 adjustments. Clinical outcomes remained uncertain: bronchopulmonary dysplasia risk ratio 0.80, death 1.03, necrotizing enterocolitis 0.81, and treatment-requiring retinopathy of prematurity 0.90.

CONCLUSION

Automated FiO2 control consistently improves oxygen targeting, but current randomized evidence does not show clear benefit for major neonatal outcomes.