Targeted Normoxemia and Supplemental Oxygen-Free Days in Critically Injured Adults: A Stepped-Wedge Cluster Randomized Clinical Trial.

Summary

In this multicenter stepped-wedge cluster RCT (N=12,487), targeting SpO2 90–96% increased time in normoxemia, reduced hyperoxemia, and did not increase hypoxemia. Although supplemental oxygen–free days were not improved overall, time to room air was shorter and a prespecified subgroup (not ventilated at ICU admission) had slightly more oxygen-free days.

Key Findings

• Normoxemia time increased from 56.2% to 71.6%; hyperoxemia decreased from 42.4% to 26.7%; hypoxemia remained 1.1% in both groups.
• No overall increase in supplemental oxygen–free days (adjusted mean difference 0.32 days; 95% CI -0.37 to 1.00; P=0.30).
• Faster weaning to room air (adjusted hazard ratio 1.23; 95% CI 1.13–1.33) without safety signal; mortality to day 90 similar.
• Among patients not ventilated at ICU admission, oxygen-free days modestly improved (AMD 0.75 days; 95% CI 0.00–1.50).

Clinical Implications

Adopt SpO2 targets of 90–96% for critically injured adults to reduce hyperoxemia and expedite weaning to room air, with monitoring to avoid hypoxemia. Expect neutral effect on overall oxygen-free days but process-of-care gains.

Limitations

• Primary outcome was neutral overall; potential contamination and secular trends inherent to stepped-wedge designs
• Unblinded process intervention; conducted in US level I trauma centers, potentially limiting generalizability

Future Directions

Test normoxemia-targeting bundles in broader ICU populations, assess patient-centered outcomes and cost-effectiveness, and integrate automated closed-loop oxygen control.