Restrictive vs Liberal Oxygenation Target in Resuscitated Comatose Patients

Among patients who are comatose who have been resuscitated following an out-of-hospital cardiac arrest, no difference has been reported between a restrictive oxygenation target and a liberal oxygenation target regarding incidence of death, or severe disability or coma, after 90 days, according to study findings published in The New England Journal of Medicine.

The Blood Pressure and Oxygenation Targets in Post Resuscitation Care trial (BOX; ClinicalTrials.gov Identifier: NCT03141099) was an open-label, investigator-initiated, randomized trial with a 2-by-2 factorial design in which patients who are comatose who had experienced out-of-hospital cardiac arrest were assigned either to a restrictive or to a liberal oxygen target, as well as to 1 of 2 target blood pressures. The portion of BOX regarding the oxygenation strategy is reported herein.

Recognizing that the appropriate oxygenation target for mechanical ventilation among comatose survivors of out-of-hospital cardiac arrest remains unknown, the researchers sought to compare 2 oxygenation targets (ie, restrictive vs liberal) among patients who were enrolled at 2 tertiary cardiac arrest centers in Denmark.

All participants were assigned in a 1:1 ratio to either a restrictive oxygen target of a partial pressure of arterial oxygen (PaO₂) of 9 to 10 kPa (68 to 75 mm Hg) or a liberal oxygen target of a PaO₂ of 13 to 14 kPa (98 to 105 mm Hg). Patients were assigned to 1 of 2 blood-pressure targets as well.

The primary study outcome was a composite of death from any cause, or hospital discharge with severe disability or coma (Cerebral Performance Category [CPC] of 3 or 4, with categories ranging from 1 to 5 and higher values indicative of more severe disability) — whichever occurred first within 90 days following randomization.

Secondary outcomes included neuron-specific levels of the enzyme enolase at 48 hours, death from any cause, score on the Montreal Cognitive Assessment scale (range, 0 to 30, with higher scores indicative of better cognitive ability) at 3 months, score on the modified Rankin scale (range, 0 to 6, with higher scores indicative of greater disability) at 3 months, and CPC at 3 months.

A total of 789 patients were included in the intention-to-treat population — 394 in the restrictive-target group and 395 in the liberal-target group. Study randomization was conducted from March 2017 through December 2021. Participants who had been admitted to the hospital following resuscitated cardiac arrest and who experienced a sustained return of spontaneous circulation were eligible for study enrollment.

The researchers found the primary-outcome event occurred in 32.0% (126 of 394) of participants in the restrictive-target group and in 33.9% (134 of 395) of those in the liberal-target group (hazard ratio [HR], 0.95; 95% CI, 0.75-1.21; P =.69). At 90 days, 28.7% (113 of 394) of participants in the restrictive-target group and 31.1% (123 of 395) of those in the liberal-target group had died (HR, 0.93; 95% CI, 0.72-1.20).

The median category on the CPC was 1 in both of the groups. On the Montreal Cognitive Assessment scale, the median score was 27 in both of the arms. On the modified Rankin scale, the median score was 2 in the restrictive-target arm and 1 in the liberal-target arm. At 48 hours, the median neuron-specific enolase level was 17 μg/L in the restrictive-target group and 18 μg/L in the liberal-target group. Adverse events occurred at similar rates in both of the groups.

Several limitations of the current study should be noted. Although the researchers enrolled patients with cardiac arrest that had a presumed cardiac cause, the incidence of primary-outcome events was relatively low and reflected a population with a high prevalence of acute coronary syndrome. Thus, the results may not be applicable to patients who have other causes of cardiac arrest. The trial was further limited by the number of patients who could be evaluated at 90 days, which was lower than anticipated. Additionally, the open-label design of the trial might have biased choices with respect to continued life-sustaining therapies.

According to the researchers, “We found no significant difference between liberal and restrictive oxygenation targets in the composite outcome of death or survival with a poor neurologic outcome. The results were consistent in all prespecified subgroups.”

Disclosure forms provided by the authors are available with the full text of this article at NEJM.org.