Journal Logo

Myths in EM

Myths in EM

Know Better, Do Better

The Last Gasp for Oxygen Therapy in STEMI?

May, Natalie MBCHB, MPHE

Author Information
Emergency Medicine News: May 2016 - Volume 38 - Issue 5 - p 13
doi: 10.1097/01.EEM.0000483179.88845.d4

    I remember revising for my final exams at medical school and dutifully learning strings of mnemonics. Perhaps the best embedded in my memory is the MONA treatment mnemonic for patients with acute ST-elevation myocardial infarction: morphine, oxygen, nitrates, and aspirin. But, oh, how things change in 10 years!

    Oxygen for STEMI is a stubborn myth to bust. The theory seems sensible enough. Cardiac syndromes, whether acute coronary syndrome or STEMI, damage cardiac muscle because blood supply is impaired, and so delivery of nutrients and oxygen is impaired. It must be good if we can increase the partial pressure of oxygen in the blood to reach those cardiac myocytes.

    The thing about oxygen is that it is really easy to administer. It's one of the first therapies we give to almost everyone in the hospital or prehospital, it takes less than 10 seconds to set up, and it is needle-free and harm-free.

    Except we are starting to see that it isn't harm-free unless the patient is actually hypoxic. By all means, administer supplemental oxygen if that's the case or the patient is in respiratory distress. But evidence suggests that supplemental oxygen may be doing more harm than good for your other STEMI patients.

    Published evidence on oxygen for cardiac chest pain has been around for a while. We've been skeptical here at St. Emlyn's virtual ED ever since 2004 when Body and Hogg's shortcut review (Emerg Med J 2004;21[1]:75) of oxygen in acute uncomplicated myocardial infarction appraised a paper by Rawles, et al. from 1976. (Br Med J 1976;1[6018]:1121.) The study randomized and double-blinded groups that received oxygen versus air, and found no difference in mortality or analgesic requirements between subjects. At this stage, Body and Hogg found “no evidence of harm,” but publications have expanded on this in the past 40 years.

    Research Expansion

    The Cochrane Collaboration also tried to answer this question. (Cochrane Database Syst Rev 2013;21[8]:CD007160.) The review covers four papers, and found a pooled relative risk of death of 2.05 (95% CI 0.75-5.58) with oxygen therapy; small sample sizes and, in particular, low event rates mean this is far from definitive. At the last review, Cochrane called for a definitive randomized-controlled trial.

    Thankfully, the AVOID investigators have undertaken some strong work since then. This prospective multicenter randomized controlled trial of air versus oxygen in STEMI diagnosed by paramedics found that mean peak troponin was similar between the groups, but the mean peak in creatine kinase (CK) was higher in the oxygen group, with higher rates of recurrent MI, arrhythmia, and increased infarct size on cardiac MR scanning at six months, suggesting harm associated with oxygen use. (Circulation 2015;131[24]:2143.) So we should at least have stopped using oxygen in STEMI by 2015, right? Especially because EMN already published on this exact topic more than 12 months ago. (EMN 2015;37[2]:10.)

    Well, in case you still aren't convinced, further analysis of the AVOID data by Nehme, et al. published in March specifically looked at the effects of supplementary oxygen on biochemical and cardiac MRI measures of myocardial injury. (Heart 2016;102[6]:444.)

    The AVOID trial randomized 638 prehospital patients with suspected STEMI (determined by ECG analysis) who had SpO2 of 94% or greater to receive either supplementary facemask oxygen at 8L/min prehospital (and after transfer to hospital according to local protocols), or room air both prehospital and in-hospital as long as SpO2 remained ≥94%.

    The authors in this secondary analysis looked at the effects of the amount of oxygen exposure on their outcome measures; the median amount of oxygen was 1746L in the overall population, but this was higher in the oxygen group (2258L vs 960L). No significant differences were seen between the groups at baseline other than dyslipidaemia (present in 64.4% of subjects in the high oxygen exposure group, versus 46.1% in the room air group), and the extent of coronary artery disease and procedural findings at angiography were similar between the groups.

    The authors then adjusted for potential confounders of myocardial injury, and proposed that every 100L of oxygen exposure was associated with a 1.4% (95% CI 0.6% to 2.2%, p<0.001) and 1.2% (95% CI 0.7% to 1.8%, p<0.001) increase in the mean troponin and CK, respectively. A 1.2% (95% CI 0.1% to 2.3%, p=0.03) increase in the cardiac MR infarct mass and a 0.9% (95% CI 0.01% to 1.9%, p=0.06) increase in the infarct size as a proportion of left ventricular mass was observed in a subgroup of 139 patients undergoing a six-month cardiac MR scan.

    What Does This Mean?

    Let's remember that this is a subgroup analysis; the original paper was well designed, but the study was powered for a broader analysis of the effect of oxygen (comparing subgroups has the effect of reducing the sample size considerably). That said, the trends described in the initial paper — suggesting that oxygen therapy is associated with increased biomarker levels and size of infarct at MR scan — seem to play out as the amount of oxygen exposure is increased.

    As much as we love to give patients oxygen, there is good quality evidence to nudge us beyond “no benefit” in giving supplemental oxygen to patients with STEMI and SpO2 ≥ 94% to “potential for harm.” I'll keep on taking off those oxygen masks when I see these patients in the ED, and will be encouraging my colleagues to do so, too. And I'm going to have to come up with a more accurate mnemonic to teach to medical students.

    Share this article on Twitter and Facebook.

    Access the links in EMN by reading this on our website or in our free iPad app, both available at

    Comments? Write to us at

    Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.