BACKGROUND:Venoarterial extracorporeal membrane oxygenation (V-A ECMO) improves perfusion and oxygenation in patients with cardiogenic shock. However, it can also result in supranormal oxygen exposure. Recent evidence suggests hyperoxia may be harmful, particularly in critically ill patients. The aim of this study was to describe oxygen exposure in patients receiving V-A ECMO after acute myocardial infarction and to investigate the association between hyperoxia and in-hospital mortality. METHODS AND DESIGN:We conducted a retrospective, cohort study of consecutive patients receiving V-A ECMO at a single tertiary level ECMO centre. We compared the mean and peak arterial oxygen tensions over the first 72 h after V-A ECMO initiation (n = 30) with those from a convenience sample of patients treated with an intra-aortic balloon pump (IABP) (n = 30) for cardiogenic shock. RESULTS:Sixty patients admitted between January 2012 and March 2018 were included in the study. Patients on V-A ECMO had significantly higher arterial oxygen tensions during the first three days than those with an IABP, at 0-24 h; V-A ECMO: 286.51 mmHg (135.76) vs IABP: 103.48 mmHg (15.22), p < 0.01.Thirteen of 30 (44.8%) patients in the V-A ECMO cohort manifested extreme hyperoxia (PaO2 ≥300 mmHg) in the first 24 hrs, compared with none in the IABP population. Within the V-A ECMO group, there was no significant association between extreme hyperoxia and in-hospital mortality (P = 0.19), duration of mechanical ventilation (P = 0.63), or troponin levels (P = 0.16) in the first 24 hrs. CONCLUSION:Severe hyperoxia is common in patients receiving V-A ECMO after acute myocardial infarction, and this continues for at least 72 h. We found no association between extreme hyperoxia and clinical outcomes.