Background :
Supplemental oxygen therapy has been used in clinical medicine for more than a century now. It finds wide application even in modern medicine and as a matter of fact there is a tendency to start delivering oxygen to any patient who is sick. This therapy has become so engrained with medical science that even lay people expect, nay even demand that their patient be administered this therapy. Cardiac patient’s particularly acute myocardial infarction (AMI) aka heart attack and heart failure represent some of the sickest populations known in the field of medicine. So it is but natural to expect / deliver supplemental oxygen in these groups of patients. While this therapy is certainly useful for patients with hypoxemia (peripheral oxygen saturation <90% or partial arterial oxygen pressure <60 mm Hg) but where is the evidence for a normoxemic patient? We now live in an era of evidence based medicine and therefore evidence of each and every bit of therapy must be investigated however logical it may seem otherwise.
Mechanism of Action :
Oxygen therapy is believed to be life-saving in a variety of clinical situations, particularly in cases with medical emergency, where its use has practically become a knee-jerk reflex reaction. In context with acute MI (AMI), wherein patients have compromised myocardial perfusion which could lead to myocardial hypoxia there could be benefit in correcting this hypoxemia. Indeed delivering oxygen in this condition appears logical, an attempt to improve oxygenation of ischemic myocardial muscle, reducing infarct size and perhaps other complications associated with it. However, in reality, there is very little evidence backing it, the practice entirely based on tradition, expert opinion or at best anecdotal evidence. On the other hand, the beneficial effects may be counter-balanced by increased production of reactive oxygen species and the related oxidative stress (contributing to re-perfusion injury) and by the paradoxical reduction in coronary blood flow and myocardial oxygen consumption resulting from hyperoxia-induced vasoconstriction in the cerebral, coronary, and systemic vasculature.
Acute Coronary Syndrome
Interestingly, a survey conducted among physicians managing AMI revealed that 96% of patients with ACS receive oxygen therapy and nearly ½ of the treating physicians felt that it reduced mortality, while another 1/4th felt that it reduced pain, only 1/4th believing it was useless (but still use it perhaps due to pressure from patients and their relatives). Thus in this situation actually inadvertent hyperoxia commonly occurs partly because of concerns to ensure sufficient oxygenation and partly because hyperoxia is not perceived to be detrimental. While, early experimental and a few small clinical studies did suggest some benefit of this strategy, most non-randomized studies have revealed that oxygen therapy in reality actually increases myocardial ischemia, increases infarct size, and even possibly increasing mortality. Even the RCTs and subsequent Cochrane analysis based on them failed to show any benefit of routine use of inhaled oxygen in acute heart attack. Moreover, Australian AVOID Study – Air Verses Oxygen In myocardial infarction study, a multi-center randomized controlled trial enrolling 441 patients revealed that routine oxygen administration (6–8 L/min) in AMI patients being transported from the ambulance to the casualty might actually be harmful. Notably, in this trial all the enrolled patients were normoximic (pulse oximetry saturation >94%). There was a 25% increase in infarct size at 6 months and significant increases in recurrent MI and arrhythmias in the oxygen therapy arm. Thus AVOID study did raise doubt about the age old practice of using oxygen in routine patients of AMI. However, since this RCT was not powered enough to assess hard clinical end points, this area continued to remain controversial.
Recently Sweden based study, DETO2X–SWEDEHEART, a multicenter, parallel-group, open-label, registry-based, RCT which enrolled 6629 patients with suspected AMI and oxygen saturation at baseline of ≥90% has been published This study has revealed that there were no differences at 1 year for all-cause mortality or re-hospitalization with MI (3.8% v 3.3%). More importantly there was no heterogeneity among any sub-group based on clinical factors, oxygen saturation levels, or final diagnosis. A subsequent meta-analysis involving 8 RCTs and a total of nearly 8000 participants also could not find any benefit in early clinical outcomes.
Heart Failure :
Heart failure remains an extremely high risk spectrum among the cardiac patients with limited long term options. Oxygen therapy is utilized in these patients despite surprisingly limited data that is available. Only 3 small studies utilizing oxygen therapy have been undertaken in patients with chronic HF. In one of them, Miak and colleagues studied the effect of oxygen therapy in patients with stable CAD (n = 12) and those with HF (n = 16). They found that extreme hyperoxia (fraction of inspired oxygen [FiO2] = 100%) was again paradoxically associated with impairment of cardiac relaxation and increased left ventricular filling pressures in both the group of patients studied. Furthermore, in the HF group this hyperoxia was associated with several negative hemodynamic changes, such as an increase in systemic vascular resistance, and reductions in cardiac output, stroke volume, and coronary sinus blood flow. In another study Park and colleagues studying 13 patients of HF who were delivered high-flow oxygen (∼5 l/min, FiO2 = 40%) also had a reduction of both cardiac output and heart rate. They also found a trend towards increased systemic vascular resistance without any change in stroke volume compared with room air delivery (FiO2 = 21%). Thus both studies revealed that that delivery of excessive supplemental oxygen can actually be detrimental to cardiac function. Lastly a study by Haque and colleagues also revealed reductions in stroke volume and cardiac output and an increase in pulmonary capillary wedge pressure with achieved hyperoxia in patients with ADHF and mild hypoxemia (baseline peripheral oxygen saturation = 92.6%). Furthermore these hemodynamic dysfunctions started early with delivery of as little as 1 l/min of supplemental oxygen (24% FiO2). These data pertain to patients of HF with reduced ejection fraction but are unlikely to be any different even in patients of HF with preserved ejection fraction.
Acute decompensated heart failure (ADHF) represents patients who are even sicker and the benefit of supplement oxygen therapy is likely to be more compared to chronic HF. Furthermore, herein the treating physicians have even fewer therapeutic options of proven therapies at their disposal and thus rely heavily on what has been classically done down the years i.e. follow the already established clinical practice. Diuretic agents and oxygen form the back-bone of the therapeutic endeavor despite the fact that there is no clear-cut data supporting their utility. Some surveys have revealed that supplemental oxygen therapy has been prescribed in at least ½ of the patients with ADHF in causality irrespective o the baseline oxygen saturation level. However, although the practice of supplementing oxygen in patients with ADHF is common, there have been no published cohort studies or RCTs establishing the role of hyperoxia in patients with ADHF. Thus, considering the findings (primarily in chronic HF patients) the recent practice guidelines are diverging from the previous consensus that oxygen should be administered routinely in patients with cardiac disease and now advocate using it in only carefully selected patients.
