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

Importance: Supplemental oxygen is fundamental to caring for critically injured adults but can expose them to excess inspired oxygen. Objective: To determine the safety and effectiveness of targeting normoxemia in critically ill trauma patients. Design, Setting, and Participants: This multicenter, stepped-wedge, cluster randomized clinical trial compared targeted normoxemia (defined as a peripheral oxygen saturation [Spo2] of 90% to 96%) with usual care among adult trauma patients admitted to an intensive care unit (ICU) at 8 level I trauma centers across the US. These trauma centers were randomized at 3-month intervals when they crossed over from usual care to targeting normoxemia. Eligible patients were enrolled between July 15, 2020, and November 14, 2022. All statistical analyses were performed from April 2023 to November 2024 according to intention-to-treat approach. Intervention: In the usual care group, supplemental oxygen was determined by treating clinicians. In the targeted normoxemia group, a multimodal educational and informatics intervention encouraged decreasing the supplemental oxygen administered whenever Spo2 exceeded 96%. Main Outcomes and Measures: The primary outcome was supplemental oxygen-free days (SOFDs), defined as the number of days alive and not receiving supplemental oxygen through day 28. Safety outcomes included hypoxemia (defined as Spo2 <88%) during the ICU admission, in-hospital mortality, and adverse events. Results: A total of 12 487 patients were enrolled (mean [SD] age, 51.7 [21.1] years; 8799 males [70.5%]; mean [SD] Injury Severity Score, 19.6 [12.0]). The proportion of ICU time spent in normoxemia increased from 56.2% in the usual care group to 71.6% in the targeted normoxemia group. Hyperoxemia (defined as Spo2 >96%) decreased from 42.4% in the usual care group to 26.7% in the targeted normoxemia group, and hypoxemia was similar between groups (1.1% vs 1.1%). The raw mean (SD) number of SOFDs was 19.6 (10.3) days for the targeted normoxemia group and 17.5 (10.4) days for the usual care group (adjusted mean difference [AMD], 0.32 [95% CI, -0.37 to 1.00] days; P = .30). Among patients not receiving mechanical ventilation at ICU admission, mean SOFDs were greater in the targeted normoxemia group than in the usual care group (22.6 [8.30] days vs 20.6 [8.86] days; AMD, 0.75; 95% CI, 0.00-1.50 days). The mean (SD) time for weaning to room air was 1.6 (3.2) days for the targeted normoxemia group and 2.7 (4.0) days for the usual care group (adjusted hazard ratio [AHR], 1.23; 95% CI, 1.13-1.33 days). In-hospital mortality to day 90 occurred in 563 patients (9.9%) in the targeted normoxemia and 732 patients (10.7%) in the usual care group (AHR, 1.05; 95% CI, 0.83-1.33). No adverse events were reported in either group. Conclusions and Relevance: This randomized clinical trial showed that targeting normoxemia did not increase the number of SOFDs but safely reduced supplemental oxygen use among critically ill trauma patients. Trial Registration: ClinicalTrials.gov Identifier: NCT04534959.