Description of the condition
Coronary heart disease (CHD) is an important cause of death worldwide. Over seven million people every year die from CHD, accounting for 12.8% of all deaths (WHO 2011). In the United Kingdom (UK) and the United States (US) it is the leading cause of death, accounting for about one-third of all deaths in people aged 35 years or over (BHF 2007; Thom 1998). Mortality rates for cardiovascular disease and CHD in men and women have fallen in most developed countries. For example, comparing the 1982 to 1992 cohort to the 1971 to 1982 cohort in the US the rate was 31% lower for mortality from cardiovascular disease, 21% lower for incidence of CHD and 28% lower for 28-day case fatality (after adjustment for age, sex and race) (Ergin 2004). The report commissioned by the UK Department of Health estimated a reduction in the case fatality rate for acute myocardial infarction (AMI) at 29 days, from 19.1% to 16.4% (Mason 2005). This reduction was associated with both a decline in the incidence of CHD and a reduction in the case fatality rate. Approximately 45% of the reduction in CHD mortality is attributable to improvement in medical therapies for coronary disease (Capewell 2000).
A common manifestation of CHD, often the first, is acute myocardial infarction (AMI). The Third Global MI Task Force (Thygesen 2012) defines AMI as "any evidence of myocardial necrosis in a clinical setting consistent with acute myocardial ischaemia."
Myocardial ischaemia is usually the result of spontaneous complications of atherosclerosis (plaque rupture,ulceration, fissuring, erosion, or dissection) resulting in coronary thrombosis (type 1 AMI). Other categories of AMI include: those produced by underlying CHD with an ischaemic imbalance attributable to a wide range of factors including endothelial dysfunction, coronary spasm, coronary embolism, tachy-/brady-arrhythmias, hypo- and hypertension (type 2 AMI); sudden cardiac death induced by myocardial ischaemia (type 3 AMI); and AMI occurring in the context of invasive coronary procedures such as percutaneous coronary intervention (PCI), in-stent thrombosis, or coronary artery bypass grafting (CABG), categorised as subtypes 4a, 4b and 5 of AMI. By far the most common types of AMI are types 1 and 2, to such an extent that their incidence may be used as proxy variables to estimate the prevalence of CHD in the general population. Hereafter we will use the term 'AMI' to refer the type 1 and type 2 AMI.
Myocardial injury may be detected through: 1. Highly sensitive biochemical markers such as Troponin (I or T), or the MB fraction of the creatine kinase (CKMB); 2. Electrocardiographic changes; or 3. Imaging techniques such as echocardiography, magnetic resonance imaging (MRI) or radionuclide imaging (RI). To make the diagnosis of AMI (in a clinical context) the necessary conditions include a change (rise and/or fall) in cardiac biomarker values, together with at least one of the following criteria: ischaemic symptoms; typical electrocardiographic changes; or abnormalities in the structure or wall motion of the heart identified by imaging techniques.
Moreover, the recognition that acute coronary syndromes represent a spectrum of pathophysiological processes rather than a uniform type of 'heart attack' has led to publication of separate guidelines for AMI presenting with persistent ST-segment elevation (STEMI) and non-STEMI presentations, reflecting the different therapeutic options.
The in-hospital mortality rate of unselected STEMI patients according to the Euro Heart Survey published by the European Society of Cardiology varies between 6% and 14% (Mandelzweig 2006). The most serious complications of AMI are cardiogenic shock, heart failure, ventricular fibrillation and recurrent ischaemia. Around 8% of people with AMI develop cardiogenic shock (Babaev 2005), but this remains present in 29% of those people on admission to hospital. The Global Registry of Acute Coronary Events (GRACE) reported that heart failure occurred in 15.6% of people with STEMI and 15.7% of those with non-STEMI, but heart failure was present in only 13% of these patients on admission to hospital (Steg 2004). Ventricular fibrillation occurred in 1.9% of people with AMI (Goldberg 2008), and recurrent ischaemia in 21% of those with acute coronary syndromes (Yan 2010), of which about half presented in the first 24 hours. Other possible complications of AMI include pericarditis, mitral insufficiency, arrhythmias and conduction disturbances.
The cornerstone of contemporary management of people with STEMI is reperfusion therapy, with either primary percutaneous coronary intervention (PCI) or thrombolytic treatment. If less than 12 hours has elapsed from the onset of symptoms, recommended treatments in international guidelines include morphine, oxygen (O₂), nitrates and aspirin (MONA) (O'Connor 2010; O´Gara 2013; Steg G 2012). Some of these treatments have a well-established research base, while others do not (Nikolaou 2012; O'Driscoll 2008; SIGN 2010).
Description of the intervention
Inhaled oxygen at normal pressure delivered by face mask or nasal cannula, at any concentration.
How the intervention might work
Myocardial infarction occurs when the flow of oxygenated blood in the heart is interrupted for a sustained period of time. The rationale for providing supplemental oxygen to a person with AMI is that it may improve the oxygenation of the ischaemic myocardial tissue and reduce ischaemic symptoms (pain), infarct size and consequent morbidity and mortality. This pathophysiological reasoning has face validity.
Why it is important to do this review
Although it is biologically plausible that oxygen is helpful, it is also biologically plausible that it may be harmful. Potentially harmful mechanisms include the paradoxical effect of oxygen in reducing coronary artery blood flow and increasing coronary vascular resistance, measured by intracoronary Doppler ultrasonography (McNulty 2005; McNulty 2007); reduced stroke volume and cardiac output (Milone 1999); other adverse haemodynamic consequences, such as increased vascular resistance from hyperoxia; and reperfusion injury from increased oxygen free radicals (Rousseau 2005).
A systematic review of human studies that included non-randomised studies did not confirm that oxygen administration diminishes acute myocardial ischaemia (Nicholson 2004). Indeed, some evidence suggested that oxygen may increase myocardial ischaemia (Nicholson 2004). Another recent narrative review of oxygen therapy (Beasley 2007) also sounded a cautionary note. It referenced a randomised controlled trial (RCT) conducted in 1976 (Rawles 1976) showing that the risk ratio of death was 2.89 (95% confidence interval (CI) 0.81 to 10.27) in participants receiving oxygen compared to those breathing air. While this suggested that oxygen may be harmful, the increased risk of death could easily have been a chance finding. A recent review (Wijesinghe 2009) looked at the effect of oxygen on infarct size in people with AMI and concluded that "There is little evidence by which to determine the efficacy and safety of high flow oxygen therapy in MI. The evidence that does exist suggests that the routine use of high flow oxygen in uncomplicated MI may result in a greater infarct size and possibly increase the risk of mortality".
Despite this lack of robust evidence of effectiveness prior to the publication of our 2010 Cochrane review of the evidence, oxygen administration was widely recommended in international guidelines (AARC 2002; AHA 2005; Anderson 2007; Antman 2002; ILCOR 2005; Van de Werf 2008). Some guidelines were more cautious; for example, the European Guideline (Bassand 2007) did not recommend routine oxygen use in acute coronary syndrome (ACS) and the Scottish Intercollegiate Guidelines Network guidance (SIGN 2007) only recommended oxygen use in hypoxaemia (< 90% saturation), noting that there was no clinical evidence for its effectiveness and referring to animal models that showed a reduction in infarct size.
Guidelines published since the 2010 Cochrane review have tended to move to a more cautious position reflecting the lack of evidence. In 2010, for example, the American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular care stated that:
“EMS providers administer oxygen during the initial assessment of patients with suspected ACS. However, there is insufficient evidence to support its routine use in uncomplicated ACS. If the patient is dyspnoeic, hypoxaemic, or has obvious signs of heart failure, providers should titrate therapy, based on monitoring of oxyhaemoglobin saturation, to 94%. (Class I, LOE)" ( O'Connor 2010).
An updated SIGN guidance states:
“A Cochrane review found no conclusive evidence from randomised controlled trials to support the routine use of inhaled oxygen in patients with acute MI. There is no evidence that routine administration of oxygen to all patients with the broad spectrum of acute coronary syndromes improves clinical outcome or reduces infarction size” (SIGN 2010).
In 2011 an Addendum to the National Heart Foundation of Australia/Cardiac Society of Australia and New Zealand Guidelines for the Management of Acute Coronary Syndromes (ACS) was published and stated that:
“There is currently insufficient evidence to formulate clear recommendations about oxygen therapy . Definitive trials are needed to answer this question" (Chew 2011).
Similarly the 2012 ESC guidelines for STEMI, citing the Cochrane review, now state:
“Oxygen (by mask or nasal prongs) should be administered to those who are breathless, hypoxic, or who have heart failure.Whether oxygen should be systematically administered to patients without heart failure or dyspnoea is at best uncertain. Non-invasive. monitoring of blood oxygen saturation greatly helps when deciding on the need to administer oxygen or ventilatory support” (Steg G 2012).
The 2013 ACCF/AHA Guideline for the Management of ST-Elevation Myocardial Infarction have a similar change in emphasis:
“Few data exist to support or refute the value of the routine use of oxygen in the acute phase of STEMI, and more research is needed. A pooled Cochrane analysis of 3 trials showed a 3-fold higher risk of death for patients with confirmed acute MI treated with oxygen than for patients with acute MI managed on room air. Oxygen therapy is appropriate for patients who are hypoxaemic (oxygen saturation <90%) and may have a salutary placebo effect in others. Supplementary oxygen may, however, increase coronary vascular resistance. Oxygen should be administered with caution to patients with chronic obstructive pulmonary disease and carbon dioxide retention”. (O´Gara 2013).
The British Heart Foundation (BHF), in response to the doubts about oxygen use raised by Beasley 2007, originally stated in an article in The Guardian 2007 that "The current practice of giving high-flow oxygen is an important part of heart attack treatment. Best practice methods have been developed and refined over the years to ensure the best possible outcome for patients. There is not enough evidence to change the current use of oxygen therapy in heart attacks". Almost three years after the publication of the first Cochrane Review the use of oxygen in AMI and more in general in the coronary acute syndromes is still controversial (Shuvy 2013). We think that, given the evidence cited, it would have been more appropriate to conclude that despite decades of use there is inadequate clinical trial evidence to unequivocally support routine administration of oxygen. The BHF subsequently stated that the 2010 Cochrane review (BHF 2010) "highlights the need for more research into the effects of oxygen when it is given during a heart attack. Until recently, heart attack patients were routinely treated with oxygen but we simply do not have enough evidence to know if that treatment is beneficial or harmful.”
With the lack of collective certainty about the use of oxygen, it is time that this treatment is re-assessed. In general, practice should not be based on tradition but on proven benefit and safety. Given that the 1976 trial (Rawles 1976) was suggestive of potential harm from oxygen in suspected AMI, it is important that the evidence base for the current guidance recommending the use of oxygen be systematically reviewed and, if necessary, further research be undertaken to clarify whether this intervention does more harm than good. If the only robust evidence is suggestive of potentially serious harm, even if the result is not statistically significant, it reinforces our opinion that this intervention should not be routinely used, however sound the underpinning pathophysiological reasoning.