From Wiki Journal Club
Jump to navigation Jump to search
Hofmann R, et al. "Oxygen therapy in suspected acute myocardial infarction". The New England Journal of Medicine. 2017. 377(13):1240-1249.
PubMedFull textPDF

Clinical Question

Among patients with suspected myocardial infarction without hypoxemia, did routine use of supplemental oxygen reduce all-cause mortality at 1 year when compared to ambient air?

Bottom Line

Among patients with suspected myocardial infarction without hypoxemia, the use of supplemental oxygen does not reduce all-cause mortality when compared to ambient air.

Major Points

The role of supplemental oxygen among adults with suspected acute MI without hypoxemia is uncertain and recent practice guidelines have highlighted a lack of evidence in its routine administration in MI.[1][2] Further, a 2015 meta-analysis of oxygen use showed no significant difference with end points when comparing those randomized to high-concentration oxygen versus those randomized to titrated oxygen or room air in the setting of AMI [3] The 2017 AVOID trial of oxygen and acute myocardial infarction in normoxia suggested that patients with MI assigned to oxygen may have a larger infarct on a subset of patients who underwent cardiac MRI. However, the role of oxygen among undifferentiated patients with findings concerning for MI was unclear.

Published in 2017, the Determination of the Role of Oxygen in Suspected Acute Myocardial Infarction (DETO2X-AMI) was a multicenter, randomized control trial with a total of 6,629 patients assigned to either oxygen or ambient-air group. All-cause mortality at 1 year was similar in the oxygen and ambient air groups (5.0% vs. 5.1%; HR 0.97; P=0.80), as was MI rehospitalization at 1 year (3.8% vs. 3.3%; HR P=0.33).

Taken together, AVOID and DETO2X-AMI provide strong support against the routine use of oxygen among patients with normoxia and MI or findings concerning for MI.[4]


As of October 2017, no guidelines have been published that reflect the results of this trial.


  • Multicenter, parallel-group, open-label, registry-based, randomized, controlled trial
  • N=6,629
    • Oxygen group (n=3,311)
    • Ambient-air group (n=3,318)
  • Setting: 35 hospitals in Sweden
  • Enrollment: 2013-2015
  • Mean follow-up: 1 year
  • Analysis: Intention-to-treat
  • Primary outcome: All-cause mortality at 1 year


Inclusion Criteria

  • Age ≥30 years
  • Symptoms suggestive of MI (e.g., chest pain or shortness of breath) for <6 hours
  • Oxygen saturation of ≥90% on pulse oximetry
  • ≥1 of the following:
    • EKG changes suggestive of ischemia
    • Elevated troponin T or I
  • Only Swedish citizens who had a unique personal identification number.

Exclusion Criteria

  • Ongoing oxygen therapy.
  • Cardiac arrest, including between presentation and enrollment

Baseline Characteristics

From the oxygen group.

  • Demographics: Age 68 years, male sex 68%
  • Baseline health data: BMI 27 kg/m2
  • Current smoker: 21%
  • PMH: Hypertension 48%, DM 18%
    • Prior CVD: Myocardial infarction 21%, PCI 16%, CABG 6.%
  • Cause of admission: Chest pain 94%, dyspnea 2%, cardiac arrest <0.1%
  • Medication at admission: Aspirin 27%, P2Y12 receptor inhibitor 5.3%, beta-blocker 31%, statin 27%, ACE-inhibitor or ARB 36%, CCB 19%, diuretic 16%
  • Vital signs at presentation: SBP 150 mm Hg, HR 79 BPM, oxygen saturation 97%
  • Final diagnosis:
    • MI 75%, STEMI 43%, angina pectoris 6%
    • Other cardiac diagnosis 8%, atrial fibrillation 2%, HF 1%, cardiomyopathy 1%, perimyocarditis 1%, PE <1%
    • Pulmonary disease 0.5%, PNA <1%, COPD <1%
    • Unspecified chest pain 8%
    • Other non-cardiac diagnosis 3.3%, MSK pain <1%


  • Randomization to a group:
    • Oxygen - 6 L/min for 6-12 hours delivered through an open face mask
    • Ambient Air
  • Both groups had end of follow-up as December 30, 2016, which was 365 days after the last patient underwent randomization
  • Non-trial-related management was left to the discretion of the treating clinician


Presented as oxygen vs. ambient air.

Primary Outcomes

All-cause mortality at 1 year
5.0% vs. 5.1% (HR 0.97; 95% CI 0.79 to 1.21; P=0.80)

Secondary Outcomes

Rehospitalization with MI at 1 year
3.8% vs. 3.3% (HR 1.13; 95% CI 0.88 to 1.46; P=0.33)
All-cause mortality or rehospitalization with MI at 1 year
8.3% vs. 8.0% (HR 1.03; 95% CI 0.87 to 1.22; P=0.70)

Subgroup Analysis

Presented on page 19 of the supplementary appendix.[5] There was no difference in the primary outcome in prespecified subgroups, including diagnosis of AMI, infarction type, sex, age ≥68 years, smoking status, diabetes, CKD, anemia, oxygen saturation ≥95%, prior MI, and prior PCI.


  • Open label design
  • The outcomes were not centrally adjudicated
  • The power calculation incorporated an anticipated 1 year mortality rate or 14%. As mortality was only 5% at 1 year, the study is underpowered to a degree.


Swedish Heart–Lung Foundation, the Swedish Research Council, and the Swedish Foundation for Strategic Research

Further Reading