From Wiki Journal Club
Jump to: navigation, search
Stub D, et al. "Air Versus Oxygen in ST-Segment Elevation Myocardial Infarction". Circulation. 2015. 131(24):2143-2150.
PubMedFull textPDF

Clinical Question

In patients who are having a ST-elevation myocardial infarction (STEMI) who are not hypoxic, does supplemental oxygen worsen myocardial infarction size?

Bottom Line

Supplemental oxygen for patients who are having a STEMI but are not hypoxic may increase both initial myocardial infarction size and increase risk for in-hospital recurrent MI.

Major Points

Supplemental oxygen for STEMI has been a long-standing accepted practice. Despite this, there has been little in the way of data regarding supplemental oxygen in non-hypoxic patients. While supplemental oxygen may improve oxygen delivery to ischemic tissue, some studies have suggested a deleterious effect of oxygen in these non-hypoxic patients due to multiple physiologic mechanisms. Increased coronary vascular resistance can lead to reduced coronary blood flow. Formation of reactive oxygen species which can lead to worsening reperfusion injury.[1][2] However, the clinical data behind this physiologic mechanism is minimal. A Cochran Review of three randomized trials looking at overall survival of patients with MI who received supplemental oxygen versus air did not show any statistical significant difference between the two, but there was concern that due to there only being a small number of deaths, the studies and subsequent meta-analysis were underpowered.[3] The 2013 ACC/AHA STEMI guidelines recommend oxygen administration for patients with an oxygen saturation <90%, respiratory distress, or other high-risk features of hypoxia.[4] However, there have not been any recommendations regarding non-hypoxic STEMI patients.

The Air Versus Oxygen in ST-Segment Elevation Myocardial Infarction (AVOID) trial randomized 638 non-hypoxic patients with STEMI confirmed on paramedic ECG to either receive or not receive supplemental oxygen (8 L/min), started by EMS. Of the 638 randomized patients, 441 were confirmed to have STEMI and included in the analysis. Initial myocardial infarction size was measured by troponin and creatine kinase. For troponin, there was no significant difference in peak troponin between the groups, however, there was a significantly higher peak creatine kinase in the supplemental oxygen group. The supplemental oxygen group also had a significant increase in the number of in-hospital recurrent myocardial infarctions (5.5% vs. 0.9%; NNH=22) and number of in-hospital major cardiac arrhythmias (40.4% vs. 31.4% NNH=11). In a subgroup of 139 individuals who underwent cardiac MRI, the median infarct size was smaller in the no oxygen group (20.3 vs. 13.1g, P=0.04), but not when assessed by proportion of LV that was infarcted (P=0.06). Ultimately, this study suggests that supplemental oxygen can may have deleterious effects in non-hypoxic STEMI patients.

In contrast with AVOID, the 2017 DETO2X-AMI[5] trial further explored the role of O2 supplementation among adults with suspected or known acute MI without hypoxemia, randomizing 6,629 patients to 6 liters/min of O2 or ambient air. There were no differences in 1-year mortality in either groups (5.0% vs. 5.1%; P=0.80), nor were there differences in reinfarction (0.5% vs. 0.5%), new AF (2.8% vs. 3.1%), second or third degree AV block (1.4% vs. 1.7%), or cardiac arrest (2.4% vs. 1.9%).


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


  • Multicenter, prospective, open label, randomized, controlled trial
  • N=441
    • Oxygen (n=218)
    • No oxygen (n=223)
  • Setting: 9 centers in Melbourne, Australia
  • Enrollment: 2011-2014
  • Follow-up: 6 months
  • Analysis: Intention-to-treat
  • Primary outcome: Myocardial injury as measured by peak cardiac troponin I and creatine kinase


Inclusion Criteria

  • Age ≥18 years
  • Chest pain <12 hours prior to assessment
  • Prehospital ECG with evidence of STEMI as determined by paramedic:
    • ST-segment elevation ≥ 0.1 mV in two contiguous limb leads OR
    • ST-segment elevation ≥ 0.2 mV in two contiguous chest leads OR
    • New left bundle branch block

Exclusion Criteria

  • Oxygen saturation <94%
  • Bronchospasm
  • Oxygen administration prior to randomization
  • Altered mental status
  • Transport to non-participating hospital
  • Physician determination of no STEMI

Baseline Characteristics

From the Oxygen arm. Of note, there was no statistically significant difference between the Oxygen and no oxygen arms.

  • Demographics: Age 63 years, 80% males
  • PMH: DM 17%, HTN 60%, HLD 56%, current or former smoker 65%, PVD 2%, CVA 5%
    • Ischemic Heart Disease: 17%, prior PCI 11%, prior CABG 11%, mediation management only 4%
  • Creatinine >120 umol/L (1.4 mg/dL): 8%
  • Status on Arrival of Paramedics: HR 74 BPM, SBP 130 mm Hg, O2 sat 98%, pain 7/10
  • STEMI complications: Arrest 5%, inotropes 5%, intubation 0%, thrombolysis 1%, Killip class ≥2 11%
  • Artery: LAD 38%, L circ 10%, RCA 46%, other 5%
    • Disease extent: Single vessel 44%, multivessel 56%, LMCA involvement 4%
  • Procedural Details:
    • Radial Intervention: 33%
    • Stent Placed: 93%
    • IABP: 3%
    • CABG: 2%
  • Time Intervals:
    • Call to hospital arrival: 55 minutes
    • Paramedic on scene to hospital arrival: 45 minutes
    • Symptom to intervention: 150 minutes
    • Hospital arrival to intervention: 54 minutes


  • Screened by paramedics for inclusion and allocated to study arm based upon computer-randomized envelopes carried by ambulance
  • 1:1 randomization to a group:
    • Oxygen - 8 L/min by facemask, continued until transfer from cath lab to CCU
    • No oxygen - No oxygen placed unless O2 sat <94%
  • All patients received aspirin 300 mg orally by paramedics, other anti-platelet therapy, choice of anticoagulation, and PCI was at discretion of treating physician and hospital protocol
  • Troponin I & creatinine kinase sampling was performed at baseline then every 6 hours for 24 hours and every 12 hours for total of 72 hours after admission
  • Contrast-enhanced cardiac MRI at 6 months was offered to all included patients who agreed to travel and had no contraindications


Comparisons are oxygen vs. no oxygen.

Primary Outcomes

Myocardial infarct size as determined by enzymes
A larger geometric mean denoted a larger infarct size.
Geometric mean peak of troponin I: 57.4 vs. 48.0 mcg/L (mean ratio 1.20; 95% CI 0.92-1.56; P=0.18)
Geometric mean peak of creatine kinase: 1948 vs. 1543 U/L (mean ratio 1.26; 95% CI 1.05-1.52; P=0.01)

Secondary Outcomes

In-hospital events
All-cause mortality: 1.8% vs. 4.5% (P=0.11)
Cardiac mortality: 1.8% vs. 3.1%
Bleeding mortality: 0 vs. 0.8%
Sepsis mortality: 0 vs. 0.4%
Recurrent MI: 5.5% vs. 0.9% (P=0.006; NNH=22)
Stroke/TIA: 1.4% vs. 0.4% (P=0.30)
Cardiogenic shock: 9.2% vs. 9.0% (P=0.94)
CABG: 2.3% vs. 4.0% (P=0.30)
Major bleeding: 4.1% vs. 2.7% (P=0.41)
Arrhythmia (sustained VT, NSVT, atrial tachyarrhythmia): 40.4% vs. 31.4% (P=0.05; NNH=11)
6 month follow-up events
14 individuals were lost to follow-up.
All-cause mortality: 3.8% vs. 5.9% (P=0.32)
Cardiac mortality: 2.9% vs. 4.1%
Bleeding mortality: 0 vs. 0.9%
Sepsis mortality: 0 vs. 0.5%
Renal failure mortality: 0.5% vs. 0
Cancer mortality: 0 vs. 0.5%
Recurrent MI: 7.6% vs. 3.6% (P=0.07)
Stroke/TIA: 2.4% vs. 1.4% (P=0.43)
Repeat revascularization: 11.0% vs. 7.2% (P=0.17)
Major cardiac event (all-cause mortality, recurrent MI, repeat revascularization, stroke): 21.9% vs. 15.4% (P=0.08)

Additional Analyses

Infarct size by cardiac MRI
Among a subsample of 61 and 66 participants
Median size: 20.3 vs. 13.1 g (P=0.04; note that there was no significant difference when adjusting for LV size)
Infarct proportion of LV mass: 12.6% vs. 9.0% (P=0.08)


  • Not blinded.
  • 14 individuals were lost to follow up between discharge and 6 months. Differential loss to followup may have biased the results.
  • The intervention administered a greater flow of oxygen than may be typically administered at some medical centers for non-hypoxic STEMI patients.
  • Use of cardiac biomarkers to measure infarct size, while correlated with eventual cardiac MRI size at 6 months is still a surrogate endpoint.
  • Creatine kinase is a less specific biomarker than troponin I for myocardial injury so while CK was significantly different, troponin I was not which is less suggestive of a difference between the two arms.[6]
  • No difference between cardiac MRI infarct sizes when adjusting for LV size.[6]


This study was funded by grants from the Alfred Foundation, FALCK Foundation, and Paramedics Australia.

Further Reading

  1. Farquhar H et al. Systematic review of studies of the effect of hyperoxia on coronary blood flow. Am. Heart J. 2009. 158:371-7.
  2. Mak S et al. Effect of hyperoxia on left ventricular function and filling pressures in patients with and without congestive heart failure. Chest 2001. 120:467-73.
  3. Cabello JB et al. Oxygen therapy for acute myocardial infarction. Cochrane Database Syst Rev 2010. :CD007160.
  4. O'Gara PT et al. 2013 ACCF/AHA guideline for the management of ST-elevation myocardial infarction: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Circulation 2013. 127:e362-425.
  5. Hofmann R, et al. "Oxygen therapy in suspected acute myocardial infarction." The New England Journal of Medicine. ePub 2017-08-28.
  6. 6.0 6.1 Bulluck H & Hausenloy DJ Letter by Bulluck and Hausenloy Regarding Article, "Air Versus Oxygen in ST-Segment-Elevation Myocardial Infarction". Circulation 2016. 133:e28.