Reducing Hyperoxia Exposure in Infants Requiring Veno-Arterial Extracorporeal Membrane Oxygenation after Cardiac Surgery

Recent studies have suggested worse outcomes in patients exposed to hyperoxia while supported on veno-arterial extracorporeal membrane oxygenation (VA-ECMO). However, there are no data regarding the effect of reducing hyperoxia exposure in this population by adjusting the fraction of inspired oxygen (FiO(2)) of the sweep gas of the ECMO circuit. A retrospective review of 143 patients less than 1 year of age requiring VA-ECMO following cardiac surgery from 2007 to 2018 was completed. 64 patients had a FiO(2) of the sweep gas < 100% with an average PaO2 of 210 mm Hg in the first 48 h of support [vs 405 mm Hg in the group with a FiO(2) = 100% (p < 0.0001)]. There was no difference in mortality at 30 days after surgery or other markers of end-organ injury with respect to whether the FiO(2) was adjusted. At least one PaO2 value < 200 mm Hg in the first 24 h on ECMO in patients with a FiO2 < 100% trended toward a significant association (OR = 0.45, 95% CI = 0.21-1.01) with decreased risk of 30-day mortality when compared to those patients with a FiO(2)= 100% and all PaO2 values > 200 mm Hg. Only 47% of patients with a FiO(2) < 100% had an average PaO2 less than 200 mm Hg which indicates that the intervention of reducing the FiO(2) of the sweep gas was not entirely effective at reducing hyperoxia exposure. Future research is needed for developing clinical protocols to avoid hyperoxia and to identify mechanisms for hyperoxia-induced injury on VA-ECMO.