UK-ROX — Conservative Oxygen Therapy in Mechanically Ventilated ICU Patients

1. Publication

• Title: Conservative Oxygen Therapy in Mechanically Ventilated Critically Ill Adult Patients: The UK-ROX Randomized Clinical Trial
• Acronym: UK-ROX
• Year & Journal: JAMA, published online June 12, 2025 (print: 2025;334(5):398-408)
• Citation: UK-ROX Investigators, Martin DS, Gould DW, Shahid T, et al. Conservative Oxygen Therapy in Mechanically Ventilated Critically Ill Adult Patients: The UK-ROX Randomized Clinical Trial. JAMA. 2025;334(5):398-408. doi:10.1001/jama.2025.9663

2. Context & Rationale

Background: Hypoxemia avoidance has traditionally driven liberal oxygen use in ICU, but excess oxygen may cause harm via hyperoxia-related mechanisms. Prior trials comparing liberal and conservative strategies, and a meta-analysis of 13 trials (10,632 patients), failed to demonstrate a significant mortality difference, leaving the question unresolved due to inadequate power.

Research Question/Hypothesis: In mechanically ventilated ICU adults receiving supplemental oxygen, does conservative oxygen therapy (SpO2 target 90%, range 88-92%) reduce 90-day mortality compared with usual oxygen therapy (clinician discretion)?

Why This Matters: UK-ROX was designed as the largest trial to date on this question, specifically powered to detect a mortality difference where prior meta-analyses had been underpowered — addressing oxygen therapy, one of the most commonly administered ICU treatments, for ~184,000 UK ICU admissions annually (>30% receiving mechanical ventilation).

3. Design & Methods

• Study Type: Multicenter, pragmatic, randomized, open-label clinical trial
• Setting & Centers: 97 ICUs, United Kingdom; enrolled May 2021 – November 2024, follow-up completed February 2025
• Population:
• Inclusion: Mechanically ventilated adults receiving supplemental oxygen in the ICU
• Exclusions: Not detailed in available trial text
• Intervention: Conservative oxygen therapy — lowest FiO2 possible to maintain SpO2 at 90% (n=8258)
• Comparator: Usual oxygen therapy at treating clinician's discretion (n=8242)
• Randomization: 1:1, 16,500 total patients — among the largest ICU trials ever conducted
• Blinding: Open-label
• Statistical Power & Follow-Up: Primary outcome: 90-day all-cause mortality. Secondary: ICU/hospital length of stay, days alive and free from organ support at 30 days, mortality at other timepoints.

4. Key Results

16,500 randomized patients.

5. Internal Validity Assessment

• Randomization & Allocation: Large-scale, multicenter randomization across 97 ICUs; specific concealment mechanics not detailed in available trial text.
• Protocol Adherence & Separation: Pragmatic protocol targeting SpO2 90% in the conservative arm vs clinician discretion in usual care — achieved meaningful separation in oxygen exposure per trial design intent.
• Blinding & Detection Bias: Open-label (inherent to an oxygen-titration strategy trial); the primary endpoint (90-day mortality) is objective and not susceptible to detection bias despite lack of blinding.
• Missing Data & Sensitivity Analyses: Very large sample with completed follow-up through February 2025; specific missing-data handling not detailed in available trial text.
• Overall Internal Validity Conclusion: Strong — by far the largest trial on this specific question (16,500 patients vs the prior largest meta-analysis of 10,632), with adequate power to detect a clinically meaningful mortality difference; the clean, precise null result (tight CI around a small 0.7 percentage-point difference) provides high-confidence evidence.

6. External Validity Assessment

• Population Representativeness: Broad, pragmatic UK ICU population, mechanically ventilated and receiving supplemental oxygen — highly representative of general UK ICU practice across 97 diverse centers.
• Practice Context: Pragmatic trial design (usual care comparator reflects real clinician discretion) directly supports real-world applicability.
• Overall External Validity Conclusion: Excellent — largest, most pragmatic trial of its kind, broadly generalizable to mechanically ventilated ICU populations in similar high-resource health systems.

7. Strengths & Limitations

Strengths:

• Largest trial to date on ICU oxygen targets (16,500 patients), definitively addressing a question prior meta-analyses (13 trials, 10,632 patients) could not resolve due to inadequate power
• Pragmatic, multicenter, broad-inclusion design
• Precise, well-powered null result with a tight confidence interval
• No significant differences across any secondary outcome or predefined subgroup

Limitations:

• Open-label design (though objective primary endpoint limits impact)
• As a pragmatic trial, some heterogeneity in usual-care oxygen practice across 97 sites
• A parallel, even larger trial (MEGA-ROX, targeting 40,000 patients) is ongoing, whose pooled results with UK-ROX will further refine oxygen-target guidance

8. Interpretation & Practice Impact

• Clinical Implications: Does not support a specific conservative (SpO2 90%) oxygen target over usual clinician-directed practice for mortality reduction — both approaches produce statistically indistinguishable 90-day mortality in mechanically ventilated ICU patients.
• Mechanistic Coherence: Confirms, at a definitive scale, the pattern seen in the prior meta-analysis of smaller trials — oxygen target strategy within this range does not appear to meaningfully alter survival.
• Systems-Level Takeaway: Clinicians can reasonably continue individualized, discretion-based oxygen titration without a mandated conservative protocol, pending the pooled analysis with the ongoing MEGA-ROX trial.

9. Controversies & Subsequent Evidence

• Editorial Commentary: Independent commentary (Chacko's Critical Care Blogspot) emphasizes this as "by far the largest trial on oxygen targets to date, with adequate power to detect a mortality difference" — framing the null result as a genuinely informative, high-confidence answer rather than an underpowered non-finding.
• Guideline Integration: Complements the ongoing MEGA-ROX trial (targeting 40,000 patients); pooled results from both studies are expected to further refine international ICU oxygen-target guidance.

10. Summary & Executive Takeaway

Summary: UK-ROX, the largest ICU oxygen-target trial to date, randomized 16,500 mechanically ventilated UK ICU patients to conservative (SpO2 90%) or usual (clinician-discretion) oxygen therapy. 90-day mortality was nearly identical (35.4% vs 34.9%; adjusted RD 0.7 percentage points, 95% CI −0.7 to 2.0, P=0.28), with no differences in any secondary outcome or subgroup.

Overall Takeaway: Conservative oxygen therapy targeting SpO2 90% does not reduce mortality compared with usual clinician-directed oxygen therapy in mechanically ventilated ICU patients — a definitive, well-powered null result that should reassure clinicians that individualized, discretion-based oxygen management remains a reasonable default pending further data from the ongoing MEGA-ROX trial.

11. Bibliography

• ICU-ROX Investigators and the Australian and New Zealand Intensive Care Society Clinical Trials Group. Conservative Oxygen Therapy during Mechanical Ventilation in the ICU. N Engl J Med. 2020;382(11):989-998.
• MEGA-ROX Trial (ongoing, target n=40,000).