Effects of a lower versus a higher oxygenation target in intensive care unit patients with chronic obstructive pulmonary disease and acute hypoxaemic respiratory failure: a subgroup analysis of a randomised clinical trial

Background: Oxygen supplementation is ubiquitous in intensive care unit (ICU) patients with chronic obstructive pulmonary disease (COPD) and acute hypoxaemia, but the optimal oxygenation target has not been established. Methods: This was a pre-planned subgroup analysis of the Handling Oxygenation Targets in the ICU (HOT-ICU) trial, which allocated patients with acute hypoxaemia to a lower oxygenation target (partial pressure of arterial oxygen [PaO2] of 8 kPa) vs a higher target (PaO2 of 12 kPa) during ICU admission, for up to 90 days; the allocation was stratified for presence or absence of COPD. Here, we report key outcomes for patients with COPD. Results: The HOT-ICU trial enrolled 2928 patients of whom 563 had COPD; 277 were allocated to the lower and 286 to the higher oxygenation group. After allocation, the median PaO2 was 9.1 kPa (inter-quartile range 8.7–9.9) in the lower group vs 12.1 kPa (11.2–12.9) in the higher group. Data for arterial carbon dioxide (PaCO2) were available for 497 patients (88%) with no between-group difference in time-weighted average; median PaCO2 6.0 kPa (5.2–7.2) in the lower group vs 6.2 kPa (5.4–7.3) in the higher group. At 90 days, 122/277 patients (44%) in the lower oxygenation group had died vs 132/285 patients (46%) in the higher (relative risk 0.98; 95% confidence interval 0.82–1.17; P=0.67). No statistically significant differences were found in any secondary outcome. Conclusions: In ICU patients with COPD and acute hypoxaemia, a lower vs a higher oxygenation target did not reduce mortality. There were no between-group differences in PaCO2 or in secondary outcomes. Clinical trial registration: NCT 03174002, EudraCT number 2017-000632-34.