While hypoxia is a clearly recognized dangerous clinical entity in critically ill patients, the dangers of hyperoxia are gaining recognition. Hyperoxia has been shown to cause systemic vasoconstriction, pulmonary interstitial fibrosis, atelectasis, and tracheobronchitis. Two studies have demonstrated the harms of hyperoxia in both peri-operative (Meyhoff 2009) and STEMI patients (Stub 2015). This study sought to look at the effect of conservative arterial oxygen goals on ICU mortality.

Clinical Question

Could the application of a conservative oxygenation protocol for oxygen supplementation targeted at a physiologic PaO2 range improve mortality in critically ill ICU patients?


All patients 18 or older admitted to an ICU with an expected length of stay of 72 hours or longer.


FiO2 targeted to PaO2 70 – 100 mm Hg or SpO2 94-98%


FiO2 targeted to PaO2 < 150 mm Hg pr SpO2 97-100%


Primary: ICU mortality
Secondary: New onset organ failure, need for reoperation in surgical patients, respiratory, blood stream, or surgical site infections.


Single center, open label, parallel group, randomized trial


Pregnancy, neutropenia, immunosuppression, COPD and ARDS.

Primary Results

  • 434 patients were included for assessment
    • Conservative n = 216
    • Conventional n = 218
  • FiO2 and PaO2 were higher in the conventional group
  • The study was stopped short of full enrollment due to an earthquake

Critical Findings

  • Mortality in the ICU (Primary Outcome)
    • Conventional  44/218 (20.2)
    • Conservative 25/216 (11.6)
    • Absolute difference: 8.6%
    • NNT = 12
    • Absolute Risk Reduction: 0.086 (CI 0.017 – 0.150)
  • Secondary outcomes (shock, liver failure, bacteremia) were all increased in the conventional oxygen group


  • Large study looking at a rational approach to oxygenation in critically ill patients
  • Important, patient centered outcome (ICU mortality)
  • Intervention is simple and inexpensive to implement


  • Single center study
  • Unblinded
  • Study stopped early due to earthquake leading to lower than planned enrollment
  • Baseline differences between groups which favors the conservative FiO2 group
  • Rate of new infections may have been underestimated due to method of confirmation (microbiological samples only)

Author's Conclusions

“Among critically ill patients with an ICU length of stay of 72 hours or longer, a conservative protocol for oxygen therapy compared with conventional therapy resulted in a lower ICU mortality. However, these preliminary findings were based on unplanned early termination of the trial, and a larger multi- center trial is needed to evaluate the potential benefit of such conservative oxygen therapy in critically ill patients.”

Our Conclusions

There is a substantial mortality benefit in critically ill patients to targeting normoxia. Hyperoxia appears to be harmful resulting in increased ICU mortality.

Potential Impact To Current Practice

Providers should be mindful of the potential harms of hyperoxia in critically ill patients. Supplemental oxygen should be used judiciously in the very sickest of patients. Patients placed on supplemental O2 should be reassessed to ensure that they are getting a safe and physiologically appropriate amount of oxygen.

Bottom Line

In critically ill ICU patients, FiO2 should be titrated to maintain normoxia as supraphysiologic levels appear to increase mortality.

Read More

The Bottom Line: Normal Oxygen Versus Hyperoxia in the Intensive Care Unit (ICU) (OXYGEN-ICU)

ScanCrit: Avoid the Oxygen Reflex

REBEL EM: July 2015 REBEL Cast


Meyhoff CS et al. PROXI Trial Group. Effect of high perioperative oxygen fraction on surgical site infection and pulmonary complications after abdominal surgery: the PROXI randomized clinical trial. JAMA. 2009; 302(14):1543-1550. PMID: 19826023

Stub D et al. AVOID Investigators. Air versus oxygen in ST-segment-elevation myocardial infarction. Circulation. 2015;131(24):2143-2150. PMID: 26002889