Hypoxemia during veno-venous extracorporeal membrane oxygenation. When two is not better than one

Abstract

Unwittingly, hypoxemia may persist or even supervene after a patient is placed on veno-venous extracorporeal membrane lung oxygenation (VV-ECMO) for refractory hypoxemia. According to Extracorporeal Life Support Organization (ELSO) guidelines, the threshold for adequate arterial O2 saturation is > 80 85%,(1) while a value > 88% has been considered the threshold in other guidelines.(2) Although the exact incidence is difficult to ascertain and the definition itself may vary, hypoxemia during VV-ECMO requires both systematic assessment and prompt optimization of modifiable variables, as it has been associated with increased mortality.(3) To fully understand why hypoxemia still occurs, one has to consider the principles underpinning the ability of ECMO to ensure adequate oxygen (O2) transfer across the membrane lung and into the patient’s blood. First, there is a fraction of oxygen in the fresh sweep gas that can be set, usually at 1.0. Second, a membrane lung, with an appropriate surface area available for gas exchange, needs to be working properly, allowing unimpeded blood flow around the gas-containing polymer microfibers. Third, the absolute amount of blood flowing through the oxygenator (QECMO) and its relative proportion to the patient’s own cardiac output (Qpatient) need to be considered. Finally, the fraction of oxygenated blood flowing through ECMO that does not go into the pulmonary circulation but instead recirculates into the drainage cannula impacts the oxygenating efficacy of VV-ECMO.(4) In a concept study, Schmidt et al. clearly demonstrated that blood flow through the ECMO circuit is the key determinant of blood oxygenation.(5) Furthermore, as a higher proportion of deoxygenated venous blood goes through the patient’s right heart than through the ECMO circuit, the QECMO/Qpatient quotient falls below the boundary of 0.6, and the O2 content of arterial blood will drop even if the absolute blood flow through the membrane lung is appropriate to the body surface area.(5) This is especially important if the degree of pulmonary shunt is such that any residual lung function contributing to oxygenation is negligible, which frequently occurs in patients being considered for VV-ECMO.(4) To overcome persistent hypoxemia, different strategies have been devised. Among them, the most immediate would be to increase the QECMO/Qpatient ratio. Typical ECMO rated flows, which is the maximal flow at which hemoglobin [12g/ dL] is fully saturated at the membrane outlet, are ~7L/minute. In these extreme situations, when a patient with no lung contribution and very high cardiac output has persistent severe hypoxemia or hypercarbia, adding a second oxygenator to the extracorporeal circuit, whether in parallel or in series, might be an intuitive option. In this issue of the Revista Brasileira de Terapia Intensiva, Melro et al.,(6) using a porcine model, evaluated the impact on blood oxygenation of these two circuit configurations. Additionally, decarboxylation efficacy, as well as pressure and resistance changes to the circuit imposed by the “virtual” presence of a second oxygenator, were analyzed. To achieve this goal, the authors built on their own previous work(7) by using a validated mathematical model to calculate peripheral arterial oxygen saturation, postoxygenator O2 content and arterial partial pressure of carbon dioxide (PaCO2) for different ECMO flows while keeping the remaining variables constant (pulmonary shunt fraction, ventilator fraction of inspired oxygen [FiO2], cardiac output, sweep gas flow, O2 fraction of sweep gas flow, hemoglobin concentration, O2 consumption and CO2 production). António Tralhão1 , Philip Fortuna2