Oxygenation During Venoarterial Extracorporeal Membrane Oxygenation: Physiology, Current Evidence, and a Pragmatic Approach to Oxygen Titration

Abstract

OBJECTIVES:

This review aims to: 1) identify the key circuit and patient factors affecting systemic oxygenation, 2) summarize the literature reporting the association between hyperoxia and patient outcomes, and 3) provide a pragmatic approach to oxygen titration, in patients undergoing peripheral venoarterial extracorporeal membrane oxygenation (ECMO).

DATA SOURCES:

Searches were performed using PubMed, SCOPUS, Medline, and Google Scholar.

STUDY SELECTION:

All observational and interventional studies investigating the association between hyperoxia, and clinical outcomes were included, as well as guidelines from the Extracorporeal Life Support Organization.

DATA EXTRACTION:

Data from relevant literature was extracted, summarized, and integrated into a concise narrative review. For ease of reference a summary of relevant studies was also produced.

DATA SYNTHESIS:

The extracorporeal circuit and the native cardiorespiratory circuit both contribute to systemic oxygenation during venoarterial ECMO. The ECMO circuit’s contribution to systemic oxygenation is, in practice, largely determined by the ECMO blood flow, whereas the native component of systemic oxygenation derives from native cardiac output and residual respiratory function. Interactions between ECMO outflow and native cardiac output (as in differential hypoxia), the presence of respiratory support, and physiologic parameters affecting blood oxygen carriage also modulate overall oxygen exposure during venoarterial ECMO. Physiologically those requiring venoarterial ECMO are prone to hyperoxia. Hyperoxia has a variety of definitions, most commonly Pa o ₂ greater than 150 mm Hg. Severe hypoxia (Pa o ₂ > 300 mm Hg) is common, seen in 20%. Early severe hyperoxia, as well as cumulative hyperoxia exposure was associated with in-hospital mortality, even after adjustment for disease severity in both venoarterial ECMO and extracorporeal cardiopulmonary resuscitation. A pragmatic approach to oxygenation during peripheral venoarterial ECMO involves targeting a right radial oxygen saturation target of 94–98%, and in selected patients, titration of the fraction of oxygen in the mixture via the air-oxygen blender to target postoxygenator Pa o ₂ of 150–300 mm Hg.

CONCLUSIONS:

Hyperoxia results from a range of ECMO circuit and patient-related factors. It is common during peripheral venoarterial ECMO, and its presence is associated with poor outcome. A pragmatic approach that avoids hyperoxia, while also preventing hypoxia has been described for patients receiving peripheral venoarterial ECMO.