Oxygen Targets in Comatose Survivors of Cardiac Arrest

This BOX trial subanalysis evaluated the effects of two oxygenation targets restrictive (PaO₂ 68–75 mm Hg) vs. liberal (PaO₂ 98–105 mm Hg), in comatose survivors of out-of-hospital cardiac arrest receiving mechanical ventilation. Involving 789 adults from two Danish cardiac arrest centers, the randomized design aimed to determine whether oxygen levels influenced 90-day outcomes. The primary composite endpoint death or discharge with severe disability or coma (Cerebral Performance Category 3 or 4) occurred in 32.0% of the restrictive group vs. 33.9% of the liberal group (HR: 0.95; 95% CI: 0.75–1.21; P=0.69), showing no statistically significant difference. Secondary outcomes (death, CPC score, modified Rankin score, Montreal Cognitive Assessment, and neuron-specific enolase levels) were similar across groups. Adverse events including infections, arrhythmias, bleeding, seizures, and renal replacement therapy occurred at comparable rates. Despite early separation of PaO₂ values and rigorous control protocols, restrictive vs. liberal oxygen targets yielded equivalent neurologic and survival outcomes, reinforcing flexibility in oxygenation strategies for post-resuscitation care.

This BOX trial subanalysis evaluated the effects of two oxygenation targets restrictive (PaO₂ 68–75 mm Hg) vs. liberal (PaO₂ 98–105 mm Hg), in comatose survivors of out-of-hospital cardiac arrest receiving mechanical ventilation. Involving 789 adults from two Danish cardiac arrest centers, the randomized design aimed to determine whether oxygen levels influenced 90-day outcomes. The primary composite endpoint death or discharge with severe disability or coma (Cerebral Performance Category 3 or 4) occurred in 32.0% of the restrictive group vs. 33.9% of the liberal group (HR: 0.95; 95% CI: 0.75–1.21; P=0.69), showing no statistically significant difference. Secondary outcomes (death, CPC score, modified Rankin score, Montreal Cognitive Assessment, and neuron-specific enolase levels) were similar across groups. Adverse events including infections, arrhythmias, bleeding, seizures, and renal replacement therapy occurred at comparable rates. Despite early separation of PaO₂ values and rigorous control protocols, restrictive vs. liberal oxygen targets yielded equivalent neurologic and survival outcomes, reinforcing flexibility in oxygenation strategies for post-resuscitation care.