A computational framework for adjusting flow during peripheral extracorporeal membrane oxygenation to reduce...

Area of Science:

• Cardiovascular Physiology
• Biomedical Engineering
• Computational Fluid Dynamics

Background:

• Peripheral veno-arterial extra corporeal membrane oxygenation (VA-ECMO) supports severe cardiac failure.
• Concomitant respiratory failure in VA-ECMO patients can lead to upper body hypoxemia due to deoxygenated blood from native circulation.
• Differential hypoxemia of the heart and brain is a critical concern in VA-ECMO management.

Purpose of the Study:

• To present a computational framework for identifying differential hypoxemia risk in VA-ECMO patients.
• To simulate various VA-ECMO scenarios to assess the impact of stroke volume and ECMO flow on blood oxygenation.
• To evaluate the potential of computational modeling in optimizing VA-ECMO therapy.

Main Methods:

• Development of a general computational framework for differential hypoxemia risk identification.
• Transient computational fluid dynamics (CFD) simulations of patient-specific VA-ECMO scenarios.
• Analysis of varying stroke volumes and ECMO flow rates in relation to patient geometry and vascular resistance.

Main Results:

• Left ventricular stroke volumes >28 mL led to all aortic arch vessels receiving poorly-oxygenated blood from the lungs, regardless of ECMO flow.
• Brachiocephalic artery perfusion was primarily from the left ventricle unless stroke volumes were <5 mL.
• A high risk of differential hypoxemia was predicted in most scenarios with residual cardiac function.

Conclusions:

• Computational modeling is valuable for optimizing VA-ECMO design and procedures.
• Personalized approaches to VA-ECMO clinical use can be enhanced through patient-specific simulations.
• Understanding flow dynamics is crucial for mitigating differential hypoxemia risk in VA-ECMO patients.