"Routine Oxygen Therapy in Suspected Acute Myocardial Infarction". The New England Journal of Medicine. 2017.

Clinical Question

Does the use of supplemental oxygen in patients with suspected acute myocardial infarction without hypoxemia reduce one-year all-cause mortality compared to ambient air?

Bottom Line

Routine use of supplemental oxygen in patients with suspected myocardial infarction who are not hypoxemic does not reduce one-year all-cause mortality.

Major Points

Guidelines

Current guidelines generally recommend supplemental oxygen only if hypoxemia is present. Previous studies, including the small AVOID trial and SOCCER trial, indicated possible harm or no benefit with routine oxygen therapy. This trial adds further evidence against the routine use of supplemental oxygen in non-hypoxemic myocardial infarction patients.

Design

- Multicenter, registry-based, randomized, open-label trial
- Patients with suspected myocardial infarction and oxygen saturation ≥90% were included
- Intervention: supplemental oxygen (6 L/min, 6-12 hours through an open face mask) vs. ambient air
- N=6629 patients
- Mean follow-up: 1 year

Population

- 6629 patients with suspected myocardial infarction and not hypoxemic
- Included age ≥30 years, symptoms suggestive of myocardial infarction (<6 hours), an oxygen saturation of 90% or higher

Interventions

- Patients randomly assigned to receive supplemental oxygen vs. ambient air

Outcomes

- Primary outcome: Death from any cause within 1 year after randomization
- Secondary outcomes: Death within 30 days, rehospitalization with myocardial infarction within 1 year
- Analysis based on the intention-to-treat principle

Major Points

- No significant difference in 1-year all-cause mortality (5.0% in oxygen group vs. 5.1% in ambient air group; HR 0.97, 95% CI 0.79-1.21)
- No significant difference in rehospitalization with myocardial infarction within 1 year or at 30 days
- No significant difference in the highest troponin levels during hospitalization between the groups

Criticisms

- Open-label design could introduce bias
- The primary end point does not require adjudication, and the secondary outcomes are from a registry, not centrally adjudicated
- Power calculation based on higher anticipated mortality rate than observed, making the actual power of the study lower than planned

Funding

- Supported by Swedish Heart–Lung Foundation, Swedish Research Council, and Swedish Foundation for Strategic Research

Further Reading

- Hofmann, Robin, et al. "Oxygen therapy in suspected acute myocardial infarction." The New England Journal of Medicine 377.13 (2017): 1240-1249.