Conservative oxygen strategy provide no clear benefit in mechanically ventilated patients

Référence

Mackle D, Bellomo R, Bailey M, et al, for the ICU-ROX Investigators and the Australian and New Zealand Intensive Care Society Trials Group. Conservative oxygen therapy during mechanical ventilation in the ICU. N Engl J Med 2020;382(11):989-998.

Sommaire

A previous single center trial reported that a more conservative oxygen strategy that avoided high levels of partal pressure of oxygen (PaO2) was associated with lower mortality and more ventilator-free days. This multicenter randomized trial set in Australia and New Zealand identified 965 ICU patients who had received invasive mechanical or noninvasive ventilation for less than 2 hours and were expected to require mechanical ventilation during their stay. Their mean age was 58 years, 63% were men, 40% were admitted with acute brain disease, 31% were postoperative, and 17% had suspected hypoxic-ischemic encephalopathy. The median ratio of PaO2 to fraction of inspired oxygen (FiO2) was 245 in the usual care group and 259 in the conservative oxygen group. Groups were balanced at baseline and analysis was by intention to treat. Patients were randomized to receive usual care with oxygen or a conservative strategy. Usual care involved no upper limit for SpO2, the use of an FiO2 of 0.30 or higher was encouraged, and the target SpO2 was simply anything higher than 90%. The conservative strategy set a desired upper limit for SpO2 of 97%, with adjustment as necessary to keep it below 97%, and the usual low target SpO2 of greater than 90%. There was no difference between groups in the primary outcome of ventilator days.The intervention had the desired effect on oxygenation, with those in the conservative group spending significantly more time with a room air FiO2 of 0.21 (29 hours vs 1 hour) and significantly less time with an SpO2 97% or higher (27 hours vs 49 hours). At 1 month, there was no difference between groups with regard to ventilator-free days, and at 90 days (34.7% vs 32.5% with usual oxygen) and 180 days (35.7% vs 34.5% with usual oxygen) there were no significant differences in mortality. There were also no differences in severe cognitive impairment or the likelihood of being employed. In a post hoc analysis (ie, data dredging/P-hacking) of the subset of patients with hypoxic-ischemic encephalopathy, there were fewer ventilator days in the conservative strategy group.