Optimizing PO2 during peripheral veno-arterial ECMO: a narrative review

Abstract

During refractory cardiogenic shock and cardiac arrest, veno-arterial extracorporeal membrane oxygenation (VA-ECMO) is used to restore a circulatory output. However, it also impacts significantly arterial oxygenation. Recent guidelines of the Extracorporeal Life Support Organization (ELSO) recommend targeting postoxygenator partial pressure of oxygen (PPOSTO2) around 150 mmHg. In this narrative review, we intend to summarize the rationale and evidence for this PPOSTO2 target recommendation. Because this is the most used configuration, we focus on peripheral VA-ECMO. To date, clinicians do not know how to set the sweep gas oxygen fraction (FSO2). Because of the oxygenator’s performance, arterial hyperoxemia is common during VA-ECMO support. Interpretation of oxygenation is complex in this setting because of the dual circulation phenomenon, depending on both the native cardiac output and the VA-ECMO blood flow. Such dual circulation results in dual oxygenation, with heterogeneous oxygen partial pressure (PO2) along the aorta, and heterogeneous oxygenation between organs, depending on the mixing zone location. Data regarding oxygenation during VA-ECMO are scarce, but several observational studies have reported an association between hyperoxemia and mortality, especially after refractory cardiac arrest. While hyperoxemia should be avoided, there are also more and more studies in non-ECMO patients suggesting the harm of a too restrictive oxygenation strategy. Finally, setting FSO2 to target strict normoxemia is challenging because continuous monitoring of postoxygenator oxygen saturation is not widely available. The threshold of PPOSTO2 around 150 mmHg is supported by limited evidence but aims at respecting a safe margin, avoiding both hypoxemia and severe hyperoxemia.

Early hyperoxia and 28-day mortality in patients on venoarterial ECMO support for refractory cardiogenic shock: a bicenter retrospective propensity score-weighted analysis

“…For each patient, the probability to show hyperoxia has been estimated using logistic regression. For ATE (average treatment effect on the entire population) analysis, weights have been attributed to each patient of the hyperoxia (overall mean PaO2 > 150 mmHg) and no hyperoxia (overall mean PaO 2 ≤ 150 mmHg) groups, making the two groups similar for the variables in the propensity score [21]. These weights were calculated using the stabilized inverse probability of treatment weighting (SIPTW).…”

Letter by Moussa et al Regarding Article, “Exposure to Arterial Hyperoxia During Extracorporeal Membrane Oxygenator Support and Mortality in Patients With Cardiogenic Shock”

“…Whether PaO 2 is obtained from a sample taken in a femoral artery or in each of the radial arteries may result in a wide difference. 3 As a result, patients can be misclassified as hypoxia, normoxia, or hyperoxia just because of the difference in sampling sites. This situation is eased when the native heart and the VA-ECMO flow mixing zone is located after the brachiocephalic trunk and is critical in Harlequin syndrome.…”

Determinants of Arterial Pressure of Oxygen and Carbon Dioxide in Patients Supported by Veno-Arterial ECMO

“…The ELSO guidelines suggest targeting post membrane-normocapnia, but without providing a cut-off value [ 15 ]. A review of the published studies found out that these recommendations were proposed without clear published data [ 16 ]. In the absence of published data, a further question would be if the P post O 2 cutoff value of 150 mmHg proposed by ELSO guidelines can prevent hyperoxemia or hypoxemia.…”