Oxygen therapy is a common treatment in ICUs, traditionally given liberally to avoid hypoxaemia and its complications. However, excessive oxygen can also be harmful, making it important to balance oxygen levels.
While many clinical trials have shown no significant difference in outcomes between conservative and liberal oxygen strategies in critically ill patients, some studies in specific populations, such as patients with severe COVID-19 or ventilated children, suggest possible benefits of a conservative approach.
The UK-ROX trial was designed to evaluate whether conservative oxygen therapy (targeting SpO₂ of 88–92%) improves 90-day mortality in mechanically ventilated adult ICU patients receiving supplemental oxygen compared to usual care. The trial was conducted in 97 UK ICUs, enrolling 16,500 mechanically ventilated patients receiving supplemental oxygen.
Participants were randomised into two groups: 8,258 received conservative oxygen therapy, aiming to maintain SpO₂ at 90% with the lowest possible oxygen fraction, while 8,242 received usual oxygen therapy managed according to clinician discretion.
The primary outcome was all-cause mortality at 90 days. Secondary outcomes included ICU and hospital length of stay among survivors, days alive and free from organ support at 30 days, and mortality at additional time points.
Among 16,500 randomised patients, primary outcome data were available for 16,394 (8,211 conservative oxygen, 8,183 usual therapy). Groups were similar in age (median 60 years) and gender (38.2% female). The conservative group received 29% less supplemental oxygen. By 90 days, mortality was 35.4% in the conservative group versus 34.9% in the usual therapy group. After adjustment, the risk difference was 0.7 percentage points, indicating no significant difference. Secondary outcomes, including ICU/hospital stay duration, organ support–free days, and mortality at other time points, also showed no significant differences.
Overall, in this trial, conservative oxygen therapy aiming to minimise oxygen exposure did not reduce 90-day all-cause mortality compared with usual oxygen therapy. The slight adjusted increase in mortality was not statistically significant, and no differences were found in subgroup or secondary analyses. These results align with previous trials and meta-analyses, though this study compared conservative therapy to usual care rather than protocolised liberal oxygen. The usual care group received relatively conservative oxygen levels compared to historical norms, reflecting recent ICU practice trends.
Achieving strict arterial oxygen targets remains challenging, partly because SpO₂ often exceeds upper limits even without extra oxygen. While oxygen exposure was lower in the conservative group, the separation between groups was smaller than in previous studies due to the more cautious usual care. Patient characteristics may influence oxygen therapy effects, with evidence suggesting the benefits of lower targets in some (e.g., COVID-19 patients) and higher targets in others (e.g., sepsis), a heterogeneity analysis planned for future reporting.
Source: JAMA
Image Credit: iStock
References:
Martin DS, Gould DW, Shahid T et al. (2025) Conservative Oxygen Therapy in Mechanically Ventilated Critically Ill Adult Patients. The UK-ROX Randomized Clinical Trial. JAMA.
