From Wiki Journal Club
Jump to: navigation, search
Girardis M, et al. "Effect of Conservative vs Conventional Oxygen Therapy on Mortality Among Patients in an Intensive Care Unit: The Oxygen-ICU Randomized Clinical Trial". The Journal of the American Medical Association. 2016. 316(15):1553-1554.
PubMedFull text

Clinical Question

What is the impact of using a restrictive strategy for supplemental oxygen in the ICU?

Bottom Line

ICU mortality was significantly lower in patients randomized to treatment with conservative treatment with supplemental oxygen.

Major Points

Administering supplemental oxygen via nasal cannula or other devices is very common among hospitalized patients (administered in 15-15.5% of all hospitalized patients) [1]. There is in fact no quality evidence to suggest a benefit of using supplemental oxygen in patients with normal or even slightly low oxygen levels [2]. Low oxygen levels can have dramatic impacts, including cardiopulmonary arrest and irreversible damage to end organs, and several studies have demonstrated a survival benefit to providing supplemental oxygen to hypoxic patients. For example, Austin et al. demonstrated a 58% reduction in mortality when supplemental oxygen was administered and titrated to 88 to 92% in patients with an acute exacerbation of COPD [3].

While supplemental oxygen is commonly perceived as risk free, the harmful effects of hyperoxemia have been well described previously [4]. For example, the unrestricted use of oxygen in patients at risk of hypercapnia (chiefly COPD), can lead to vasodilation leading to further V/Q mismatches and ultimately a right-ward shift of the CO2 dissociation curve via the Haldane effect (not as a result of decreased respiratory drive). The end result is worsening of the hypercapnia [5]. Direct toxicity from excessive oxygen levels can lead to the generation of oxygen free radicals that have been shown to lead to CNS toxicity (confusion, seizures, etc.) in short term exposures (as seen in divers and hyperbaric oxygen therapy), or over periods of time pulmonary toxicity (tracheobronchitis, diffuse alveolar damage, and fibrosis) or ocular toxicity [6]. Despite the widespread practice, current guidelines do not set clear and unambiguous targets for a patient’s oxygenation.

This trial, the 2016 OXYGEN-ICU trial, randomized 434 ICU patients to either conservative or liberal treatment with supplemental oxygen. This study was originally planned for a sample size of 660, but this was adjusted to 480 based on subsequent analysis. The study was then stopped short of this goal following a natural disaster. The results of this study showed numerous beneficial outcomes to restricting the administration of supplemental oxygen (lower ICU mortality, lower rates of new onset shock, lower rates of bacteremia and increased mechanical ventilator free hours).

The findings of this study are good evidence for restricting oxygen, and are in line with a handful of studies demonstrating that it is safe to withhold oxygen in ICU patients without hypoxia [7]. For example, the AVOID trial (Air Versus Oxygen in Myocardial Infarction) demonstrated that providing supplemental oxygen to patients with an ST-elevation-myocardial infarction without hypoxia lead to an increase in early myocardial injury and was associated with large infarct sizes at 6 months [8]. and are promising and warrant further investigation with a larger scale multi-center trials.


Tissue oxygenation itself is measured using surrogate measures. PaO2 is a measurement of the free oxygen dissolved in plasma obtained from arterial blood, expressed in mmHg. SpO2 is a measurement of the saturation of hemoglobin obtained using a pulse oximeter, expressed as %. A final measurement is the CaO2, which is a calculation of the total oxygen content of the blood, expressed as ml/L (this measurement was not used in this paper).

  • PaO2 = Partial Pressure of Oxygen (mmHg)
  • SpO2 = Oxygen Saturation (%)
  • CaO2 = Oxygen Content (mL/L)
  • PaCO2 = Partial Pressure of Carbon Dioxide (mmHg)
  • FiO2 = Fractional Inhalation of Oxygen (%)
  • P:F ratio = PaO2:FiO2


British Thoracic Society Guidelines for Use of Supplemental Oxygen [9]

  • For critically ill patients high concentration oxygen should be administered immediately, using initial therapy with a reservoir mask at 15 L/min
    • Critically ill defined as: cardiac arrest, shock, sepsis, major trauma, near-drowning, anaphylaxis, major pulmonary hemorrhage, major head injury, carbon monoxide poisoning
  • Otherwise, recommended to initiate oxygen therapy using nasal cannula at 2-6 L/min (preferred) or simple face mask at 5-10 L/min
  • Pulse oximetry should be used in all patients with SOB
  • Oxygen should be prescribed to achieve a target saturation of 94-98% for most acutely ill patients, or 88-92% for those at risk of hypercapnic respiratory failure
  • Oxygen should be reduced in stable patients with satisfactory oxygen saturation

Global Initiative for Chronic Obstructive Lung Disease (GOLD) Guidelines for use of Oxygen Therapy in patients with COPD

  • Define the key component of hospital treatment of acute exacerbations of COPD as oxygen therapy.
  • Recommend to use supplemental oxygen, titrated to maintain patients oxygen saturations at 88-92%.
  • Once oxygen is started, arterial blood gas should be checked 30-60 minutes later to confirm adequate oxygenation without developing hypercapnia.
  • Patients with PaO2 < 55mmhg or SpO2 < 88% with or without hypercapnia, confirmed twice over a three week period, should be started on oxygen therapy.
  • Patients with hypoxemia as defined above, with pulmonary hypertension or peripheral edema, suggestive of CHF or polycythemia (HCT > 55%) should also be started on long term oxygen therapy.
  • From the Global Strategy for the Diagnosis, Management and Prevention of COPD, Global Initiative for Chronic Obstructive Lung Disease (GOLD) 2016. Available from: http://goldcopd.org/

Thoracic Society of Australia and New Zealand Guidelines for Oxygen Use [10]

  • Pulse oximetry should be used whenever oxygen is used
  • In COPD patients or others with chronic respiratory failure, supplemental oxygen should be applied if SpO2 < 88%, and titrated to SpO2 of 88-92%.
  • In other medical conditions, oxygen should be administered if SpO2 < 92%, and titrated to 92-96%.

AHA guidelines on use of supplemental oxygen in acute stroke [11]

  • “Although the routine use of supplemental oxygen remains unproven, supplemental oxygen to maintain oxygen saturations >94% is recommended after cardiac arrest and is reasonable for patients with suspected stroke”

European Society of Cardiologists [12] [13]

  • Management of acute coronary syndromes
    • Oxygen should be administered at 4-8 L/min for oxygen saturation < 90%
  • Management of ST-elevation MIs
    • Oxygen should be administered at 4-8 L/min for oxygen saturation < 95%


  • Single-center, open-label, randomized clinical trial
  • N=434
    • Conservative/Intervention (n=216)
    • Conventional/Control (n=218)
  • Setting: Modena University Hospital, Modena, Italy
  • Enrollment: March 2010 to October 2012
  • Mean follow-up: 24 days (conventional group) years
  • Analysis: Intent-to-treat
  • Primary outcome: ICU mortality


Inclusion Criteria

  • 18 years old
  • Admitted to ICU
  • Expected LOS 72 hours or more

Exclusion Criteria

  • < 18 years old
  • ICU readmission
  • Decision to withhold life-sustaining treatment
  • Immunosuppression
  • Neutropenia
  • Enrollment in another study
  • Acute decompensation of COPD and ARDS with P:F ration < 150

Baseline Characteristics

Chars from Control/conventional group, unless specified

  • Age: 65
  • Female: 42.7%
  • Respiratory Failure at admission:
    • 17.4% conservative group
    • 30.2% conventional group
    • p = 0.017
  • Mechanical Ventilation at admission:
    • 16.1% conservative
    • 27% conventional
    • p = 0.023
  • Pre-existing conditions:
    • COPD: 5%
    • CKD: 6%
    • CLD: 14.2%
    • Cancer: 31.1%
  • Shock (33%)
    • Septic (21.6%)
    • Hypovolemic/ Hemorrhagic (4.1 %)
    • Cardiogenic (3.7%)
    • Mixed (3.7%)
  • Type of Patient admission:
    • Medical (39.5%)
    • Surgical (60.7%)
  • Severity of Illness
    • Measured by SAPS score
    • 22.2% of Conservative group with SAPS > 38
    • 33.3% of Conventional with SAPS > 38
    • P = 0.072
    • SAPS = Simplified Acute Physiology Score (SAPS) II


  • Supplemental oxygen to maintain PaO2 70 to 100 or SpO2 of 94% to 98%
    • Oxygen administered at lowest possible FiO2
    • FiO2 gradually reduced or discontinued whenever PaO2 > 100 or SpO2 > 98%
  • Control group/Conventional therapy: supplemental oxygen administered at a minimum of 40% (5L/min)
    • PaO2 allowed up to 150 mmHg or SpO2 97 to 100%
    • If sats dropped below 95% to 97%, the FiO2 was increased to reach desired SpO2
  • For Intubation
    • Control group received pre-oxygenation with 100% FiO2
    • Conservative group given supplemental oxygen only if sats < 94%
  • At least 1 ABG was collected per day for each patient
  • Use of NIIV, high-flow dictated by common use


Comparisons are conservative therapy vs. conventional therapy.

Mean Oxygen Levels

Median FiO2
36% vs 39%
Median PaO2
87 mmHg vs 102 mmHg

Primary Outcomes

ICU Mortality
11.6 % vs 20.2% (95% CI, 0.017-0.150; P=0.01)

Secondary Outcomes

New organ failure
19.0% vs 25.7% (p = 0.09)
Respiratory failure
6.5% vs 6.4% (p = 0.98)
Renal failure
12.0% vs 9.6% (p = 0.42)
Shock/Circulatory failure
3.7% vs 10.6% (p = 0.006)
Liver failure
1.9% vs 6.4% (p = 0.02)
New onset bacteremia
5.1% vs 10.1% (p = 0.049)
New onset respiratory infection
13.9% vs 17.0% (p = 0.37)
New onset surgical site infection, in surgical patients
7.2% vs 9.1% (p = 0.68)
Mechanical ventilator free hours
72 vs 48 (p = 0.02)
Hospital length of stay
21 days vs 21 days
ICU length of stay
6 days vs 6 days

Subgroup Analysis

  • Sample size too small and precluded sub-group analysis. Only included 31 patients with stroke and 19 with acute MI, for example.
  • Post hoc analysis, the conservative group had lower risk for ICU mortality among patients with respiratory failure who received mechanical ventilation (absolute risk reduction, 0.05 [95% CI, 0.00-0.09]; relative risk, 0.67 [95% CI, 0.46-0.96])

Adverse Events

None reported.


  • This is not a study that can be blinded.
  • One of the most striking differences between the 2 groups is the difference in illness severity among the 2 groups. While not statistically significant, the conventional group tended to be sicker (with higher SAPS scores). This alone could have accounted for higher ICU mortality in the conventional group.
    • Along these same lines, significantly more patients in the conventional group were on mechanical ventilation on admission. Additionally, significantly more patients were diagnosed with respiratory failure on admission in the conventional group
  • It may be that the increased attention needed to titrate down oxygen levels was behind lower mortality. For example, perhaps altered mental status or hypotension could have been caught sooner in the intervention group, which could have influenced mortality completely distinctly from oxygen levels.
  • Finally, this was a single center study in Italy, which was halted early for a natural disaster.


This study was supported by the National Fund for Scientific Research of the University of Modena and Reggio Emilia.

Dr Singer reported serving as the data monitoring chair for a phase 2 study sponsored by InflaRx, on the antibiotic advisory board for Bayer, and on sepsis advisory boards for Biotest and Merck. No other disclosures were reported.

Further Reading

  1. O'Driscoll BR et al. British Thoracic Society emergency oxygen audits. Thorax 2011. 66:734-5.
  2. Abernethy AP et al. Effect of palliative oxygen versus room air in relief of breathlessness in patients with refractory dyspnoea: a double-blind, randomised controlled trial. Lancet 2010. 376:784-93.
  3. Austin MA et al. Effect of high flow oxygen on mortality in chronic obstructive pulmonary disease patients in prehospital setting: randomised controlled trial. BMJ 2010. 341:c5462.
  4. Cousins JL et al. Acute oxygen therapy: a review of prescribing and delivery practices. Int J Chron Obstruct Pulmon Dis 2016. 11:1067-75.
  5. Abdo WF & Heunks LM Oxygen-induced hypercapnia in COPD: myths and facts. Crit Care 2012. 16:323.
  6. Thomson L & Paton J Oxygen toxicity. Paediatr Respir Rev 2014. 15:120-3.
  7. Damiani E et al. Arterial hyperoxia and mortality in critically ill patients: a systematic review and meta-analysis. Crit Care 2014. 18:711.
  8. Stub D et al. Air Versus Oxygen in ST-Segment-Elevation Myocardial Infarction. Circulation 2015. 131:2143-50.
  9. O'Driscoll BR et al. BTS guideline for emergency oxygen use in adult patients. Thorax 2008. 63 Suppl 6:vi1-68.
  10. Beasley R et al. Thoracic Society of Australia and New Zealand oxygen guidelines for acute oxygen use in adults: 'Swimming between the flags'. Respirology 2015. 20:1182-91.
  11. Jauch EC et al. Guidelines for the early management of patients with acute ischemic stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2013. 44:870-947.
  12. Hamm CW et al. ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: The Task Force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC). Eur. Heart J. 2011. 32:2999-3054.
  13. Steg PG et al. ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation. Eur. Heart J. 2012. 33:2569-619.