A new formula for determining arterial oxygen saturation during venovenous extracorporeal oxygenation

Abstract

Venovenous extracorporeal membrane oxygenation (VV-ECMO) is used to treat severe forms of acute respiratory distress syndrome (ARDS). VV-ECMO management may be confusing due to the lack of information about the interplay between the determinant parameters and their impact on oxygenation. We found a relationship between arterial oxygen saturation (SaO₂) and its relevant parameters. The aim of this study was to assess the validity of this model. We report our experience in 17 patients under VV-ECMO for severe ARDS. We compared, at two different levels of pump flow, SaO₂ and the oxygen saturation measured in the pulmonary artery (SpaO₂) with the predicted saturation using the formula: SpaO₂ = (EF/CO)SmO₂ + (1 − EF/CO)SvO₂ + 10⁻²PmO₂, where PF is pump flow, R is recirculation, EF is effective flow [= (1 − R)PF], SmO₂ is saturation of the oxygenator outgoing blood, CO is cardiac output, SvO₂ is saturation of mixed venous blood, and PmO₂ is oxygen partial pressure of the oxygenator outgoing blood. There was no significant difference between predicted and measured SpaO₂: the mean predicted and measured SpaO₂ values were 90.7 ± 2.8 % and 90.4 ± 2.7 % , respectively (p = 0.696, r = 0.966). Bland-Altman analysis showed good agreement between predicted and measured SpaO₂. Predicted SpaO₂ and SaO₂ was well correlated (r = 0.80). We have presented an explicit relationship between SaO₂ and its direct determinants during VV-ECMO. Good agreement was found with the measured values of SaO₂, but the model remains to be fully validated before its use in clinical practice.