Read the full EBM Focus and earn CME credit.
Reference - AVOID trial (Circulation 2015 May 22 early online)(level 2 [mid-level] evidence)
Supplemental oxygen therapy is commonly given as part of the initial treatment to all patients with ST-elevation myocardial infarction (STEMI), even when the oxygen saturation is normal, though studies supporting its use are limited. Current guidelines recommend administration of oxygen in patients with STEMI presenting with hypoxia, but exact definitions and recommendations vary (Circulation. 2013 Jan 29;127(4):e362-425, Eur Heart J. 2012 Oct;33(20):2569-619). A previous Cochrane review found routine oxygen therapy may not reduce mortality in patients with suspected or proven acute myocardial infarction (Cochrane Database Syst Rev 2013 Aug 21;(8):CD007160), but only 4 trials (all with wide confidence intervals) were included in this analysis. A recent randomized trial compared oxygen supplementation at 8 L/min vs. no oxygen supplementation in 441 adults with normal oxygen saturation who had chest pain for < 12 hours and confirmed STEMI. Patients with suspected STEMI on prehospital electrocardiogram and oxygen saturation > 94% (638) were randomized by paramedics who initiated therapy before hospital admission. Paramedics also administered aspirin 300 mg orally to all patients. Only patients later confirmed to have STEMI by emergent coronary angiography were included in the analysis.
Patients in the no oxygen group were started on oxygen if their oxygen saturation went below 94% before or on arrival to the cardiac catheterization lab, which occurred in 7.7% of patients randomized to no oxygen. All patients with no contraindications were invited for contrast enhanced cardiac magnetic resonance (CMR) imaging at 6 months, and patients not electing for CMR had 6-month follow-up telephone assessment. Although mean peak troponin I levels did not differ significantly between groups, oxygen supplementation was associated with increased mean peak creatine kinase levels compared to no oxygen supplementation (1,948 units/L vs. 1,543 units/L, p = 0.01). In the 32% of patients completing the 6-month CMR assessment, supplemental oxygen was also associated with an increase in median infarct size (20.3 g vs. 13.1 g, p = 0.04). Furthermore, a greater number of patients receiving supplemental oxygen had recurrent in-hospital myocardial infarctions (5.5 % vs. 0.9%, p = 0.006), in-hospital cardiac arrhythmia (40.4% vs. 31.4%, p = 0.05), and 6-month major cardiovascular adverse events (21.9% vs. 15.4%, p = 0.08). There were no significant differences in mortality during hospital stay or at 6 months.
A major limitation of previous studies evaluating oxygen supplementation is the randomization of patients upon reaching the hospital, after oxygen therapy is routinely administered. The randomization method employed by this trial accounted for pre-hospital interventions by providing sealed, opaque, externally numbered randomization envelopes to ambulances. In this study, oxygen was provided by face mask at 8 L/minute, but oxygen is often provided by nasal cannulation at 2-4 L/minute, making generalizations difficult. It is unknown if the adverse effects of oxygen observed in this trial also pertain to lower level oxygen administration. Finally, the primary outcome of this trial was myocardial injury assessed by peak cardiac troponin I and creatine kinase levels, and the trial was powered to detect differences in these surrogate outcomes. Although the rate of in-hospital recurrent myocardial infarctions was significantly higher in the group receiving oxygen therapy, the 6-month rate of recurrent myocardial infarctions was no longer significant (p = 0.07). This trial was underpowered to detect differences in clinical endpoints, however, and further evaluation is necessary to determine if the observed increase in myocardial injury translates into an increase in clinical outcomes. Overall, the results of this trial call into question the routine use of oxygen therapy, suggesting oxygen supplementation in patients with normal oxygen saturation levels may increase myocardial injury in patients having STEMI.
For more information, see the ST-elevation myocardial infarction (STEMI) topic in DynaMed.