Most patients with acute coronary syndromes (ACS) receive oxygen therapy as part of their emergency treatment, initiated by paramedics during transfer and before their first contact with a physician. A survey among physicians involved with acute myocardial infarction cases found that 96% of their patients with ACS received oxygen therapy. About 50% of all responders believed that oxygen decreases mortality, 25% thought it helps to relieve pain and 25% thought it has no effect. Many therapies and interventions are not based on proven benefit, but on anecdotal evidence, expert opinion and tradition. This is especially true for oxygen therapy, which is usually not questioned and has been used for over 100 years. We could argue that as long as it does no harm, it does not really matter whether we continue to provide oxygen in these situations. However, is it really harmless?
From a physiological perspective, treating ACS patients with oxygen may seem reasonable. In ACS there is a lack of myocardial perfusion and consequently a lack of oxygenation of the myocardium. Therefore, it seems logical to increase the oxygenation of the blood reaching the jeopardised myocardium by administering oxygen therapy. However, another theory argues that oxygen may increase microvascular resistance, leading to reduced coronary blood flow thus reducing cardiac output and increasing radical oxygen species, which can have multiple negative effects including increased risk for arrhythmias and cell damage leading to heart failure.[2,3]
Randomised trial data on oxygen delivery in acute myocardial infarction (AMI) are also conflicting. A systematic review is therefore very useful. In the August 2013 issue of the Cochrane Database of Systematic Reviews Cabello et al report a systematic review including a total of 4 randomised trials and 430 patients. These trials randomised patients to inhaled oxygen at 4 to 6 L/min (given by mask [3 studies] or by nasal cannula [1 study] versus a control therapy (air [3 studies] or titrated oxygen aiming for an oxygen saturation of 93%-96%). The pooled relative risk (RR) for mortality (data from 3 studies) was 2.05 (95% C.I. 0.75 to 5.53). However, only 16 deaths were included in this pooled analysis. This small number of events precludes a robust conclusion. In addition the risk of bias of the included trials was high. The treatment of AMI in older studies also differed compared with current practice.
Primary percutaneous coronary intervention (PCI) is now the gold standard treatment for AMI, with thrombolysis very rarely used in Western countries. One of the studies in the review was published over 30 years ago in the pre-reperfusion era, and another trial was conducted in the thrombolysis era. Even in the most recent study, published in 2012, about a quarter of patients received thrombolysis rather than primary percutaneous coronary intervention. While reperfusion with PCI is successful to achieve revascularisation in nearly all cases, thrombolysis achieves sufficient revascularisation in only 50% of cases. How, and if, a patients get re-perfused is potentially very important, and the effect of delivering oxygen may differ depending on the method used. While reperfusion is vital, it can also lead to myocardial damage, the so-called 'reperfusion injury'. This reperfusion damage can lead to 25% of the overall myocardial damage. Reperfusion injury may be influenced by oxygen therapy, for example by increasing the concentration of radical oxygen species.
The most recent trial reported by Ranchord et al, compared routine oxygen use versus titrated oxygen therapy, which is currently recommended in international guidelines. There was only one death in the 68 patients treated with routine oxygen and 2 in 68 of those treated with the titrated oxygen therapy in this study. However, even this very recent trial suffered several limitations, which makes it difficult to draw firm conclusions from the overall meta-analysis. Over 10% of the patients were excluded after randomisation (4 in the intervention group, 8 in the control group). The outcomes of these patients were not reported. As a consequence, the analyses in this study did not follow the intention-to-treat (ITT) principle, which could introduce serious bias. Two patients were excluded from the study because they developed cardiogenic shock after randomisation, but it was unclear which group they were in and what their outcome was. Sensitivity analyses carried out by the authors of the review illustrate the effect that different scenarios could have on the findings of the meta-analysis. Assuming these 2 patients had died and were both in the regular oxygen group the relative risk of mortality from the pooled analysis was 2.42. However, assuming the two patients had died and were in the control group the relative risk changes to 0.26. Given the considerable weight that this trial contributes to the summary relative risk, the limited robustness of the results of the meta-analysis is clearly illustrated.
Given these arguments, could the twofold increase in mortality with routine oxygen therapy in this meta-analysis be a 'real' signal which did not reach statistical significance because of insufficient statistical power, or is it an artefact due to chance? We do not know. However, considering the related treatment of hyperbaric oxygen therapy (which is administration of 100% oxygen at pressures greater than one atmosphere absolute) in patients with MI, evidence from small randomised trials suggests a reduction in myocardial damage and mortality. So if hyperbaric oxygen does not cause harm, and is even potentially beneficial, might it be rather unlikely that oxygen therapy at 4-6 L/min would increase mortality?
In conclusion, we do not know whether routine oxygen administration in patients with an acute MI has any impact on outcome. Nonetheless, this systematic review challenges the status quo predicated by international guidelines on the treatment of acute coronary syndromes and highlights the need for large-scale trials.
Pascal Meier1, Shah Ebrahim2, Catherine M Otto3, Juan P Casas4
1Pascal Meier, MD (firstname.lastname@example.org), The Heart Hospital, University College London Hospitals UCLH, London, UK and Division of Cardiology, Yale Medical School, New Haven, CT, USA; 2Shah Ebrahim, MD (Shah.Ebrahim@lshtm.ac.uk), Department of Non-communicable Disease Epidemiology, and London School of Hygiene & Tropical Medicine, Keppel Street, London, UK; 3Catherine M. Otto, MD (email@example.com), Division of Cardiology, University of Washington; 4Juan P. Casas, MD (Juan.Pablo-Casas@lshtm.ac.uk), Department of Non-communicable Disease Epidemiology, London School of Hygiene & Tropical Medicine, Keppel Street, London, UK, and Institute of Cardiovascular Science, University College London, UK.
How to cite: Meier P, Ebrahim S, Otto CM, MD, Casas JP. Oxygen therapy in acute myocardial infarction – good or bad?[editorial]. Cochrane Database of Systematic Reviews 2013;8:ED000065. dx.doi.org/10.1002/14651858.ED000065
1. Burls A, Emparanza JI, Quinn T, Cabello JB. Oxygen use in acute myocardial infarction: an online survey of health professionals' practice and beliefs. Emergency Medicine Journal 2010;27(4):283-286. dx.doi.org/10.1136/emj.2009.077370
2. McNulty PH, Robertson BJ, Tulli MA, et al. Effect of hyperoxia and vitamin C on coronary blood flow in patients with ischemic heart disease. Journal of Applied Physiology 2007;102(5):2040-2045. dx.doi.org/10.1152/japplphysiol.00595.2006
3. Frohlich GM, Meier P, White SK, Yellon DM, Hausenloy DJ. Myocardial reperfusion injury: looking beyond primary PCI. European Heart Journal 2013;34(23):1714-1722. dx.doi.org/10.1093/eurheartj/eht090
4. Cabello JB, Burls A, Emparanza JI, Bayliss S, Quinn T. Oxygen therapy for acute myocardial infarction. Cochrane Database of Systematic Reviews 2013;8:CD007160. dx.doi.org/10.1002/14651858.CD007160.pub3
5. Ranchord AM, Argyle R, Beynon R, et al. High-concentration versus titrated oxygen therapy in ST-elevation myocardial infarction: a pilot randomized controlled trial. American Heart Journal. 2012;163(2):168-175. dx.doi.org/10.1016/j.ahj.2011.10.013
6. Bennett MH, Lehm JP, Jepson N. Hyperbaric oxygen therapy for acute coronary syndrome. Cochrane Database of Systematic Reviews 2011;8:CD004818. dx.doi.org/10.1002/14651858.CD004818.pub3
Competing interests: The authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available upon request) and declare (1) no receipt of payment or support in kind for any aspect of the article; (2) PM, CMO and JPC declare no financial relationships with any entities that have an interest related to the submitted work, SE declares that the NIHR provides core funding for the Cochrane Heart Group; (3) no other relationships or activities that could be perceived as having influenced, or giving the appearance of potentially influencing, what was written in the submitted work.
Image credit: Will & Deni McIntyre/Science Photo Library
Contact the Editor in Chief, Dr David Tovey (firstname.lastname@example.org): Feedback on this editorial and proposals for future editorials are welcome.
Comment from Graham Werstiuk, 24 September 2013
I enjoyed reading the editorial "Oxygen therapy in acute myocardial infarction – good or bad?" However, a comment towards the end overlooks or minimizes the fact that there are several orders of magnitude difference between the different therapies: "However, considering the related treatment of hyperbaric oxygen therapy (which is administration of 100% oxygen at pressures greater than one atmosphere absolute) in patients with MI, evidence from small randomised trials suggests a reduction in myocardial damage and mortality. So if hyperbaric oxygen does not cause harm, and is even potentially beneficial, might it be rather unlikely that oxygen therapy at 4-6 L/min would increase mortality?"
Based on the work by McNulty et al we know that there is roughly a 20% to 30% drop in coronary blood flow in a hyperoxic environment.[1,2] The question has always been whether the increase in oxygen content was enough to offset the decrease in delivery. The ambiguous results of previous MI oxygen therapy trials seems to suggest that at the very least there is no net benefit. Intuition based on the oxygen content formula is that once the hemoglobin is fully saturated only a small nominal increase in the oxygen content of the blood is possible using conventional oxygen delivery.
In hyperbarics with a patient submerged to 3 atm the alveolar oxygen partial pressure is increased sufficiently that we could expect the dissolved oxygen in the blood to play a significant role in oxygen delivery. Perhaps enough to overcome any deficits in delivery. For example a patient subjected to 3 atm with and FiO2 of 1.0 could have an alveolar oxygen tension of 2183 mmHg (assuming normal CO2 and water vapour pressure). Even in a diseased state with an elevated A-a gradient we could expect to see almost a 30% jump in oxygen content. It seems much more reasonable that even with profound vasoconstriction the increased content will be enough to overcome the reduction of delivery.
Graham Werstiuk, Instructor, Respiratory Therapy, School of Health Sciences, Northern Alberta Institute of Technology, Canada
1. McNulty PH, King N, Scott S, Hartman G, McCann J, Kozak M, et al. Effects of supplemental oxygen administration on coronary blood flow in patients undergoing cardiac catheterization. American Journal of Physiology Heart and Circulatory Physiology 2005;288(3):H1057-62.
2. McNulty PH, Robertson BJ, Tulli MA, Hess J, Harach LA, Scott S, et al. Effect of hyperoxia and vitamin C on coronary blood flow in patients with ischemic heart disease. Journal of Applied Physiology 2007;102(5):2040-5.
Conflicts of interest: none declared.