Hemodynamic Consequences and hypo oxygenation in Peripheral Extracorporeal Membrane Oxygenation Support Systems: A Computational Study
Presenting Author: Farhan Khodaee1
Institute for Medical Engineering and Science, Massachusetts Institutes of Technology, USA
Additional Authors: Farhad Rikhtegar Nezami 1, Elazer R. Edelman1,2, and Steven P. Keller1,2
1 Institute for Medical Engineering and Science, Massachusetts Institutes of Technology, USA
2 Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, USA
BACKGROUND/PURPOSE: Extracorporeal membrane oxygenation (ECMO) is rapidly emerging as a means of mechanical circulatory support for cardiogenic shock patients despite limited understanding of its effect on circulation and end-organ perfusion. ECMO profoundly alters physiologic patterns of aortic blood flow as blood is shunted from the venous system and passed through an oxygenator and then returned, with a significantly higher velocity, to the circulation via either a cannula inserted into the aortic arch (Central) or femoral artery (Peripheral). This study aims to quantify the effect of ECMO support level on the pulsatility of blood flow and perfusion of vital organs.
METHOD: An idealized geometry, constructed averaging in vivo images and post-mortem measurements of aorta, was used which includes major aortic outlets. In a polyhedral computational grid, a multiphase turbulent model for incompressible Newtonian blood flow were solved using ANSYS Fluent, wherein a varying level of ECMO support were evaluated (90%, 75%, and 50% of total cardiac output). Dynamic boundary conditions at the outlets were set using a 3-element Windkessel model, and hemodynamic parameters, stress metrics, and end-organ perfusion were analyzed.
RESULTS: Decreasing the proportion of ECMO flow resulted in shifting the mixing cloud downstream and therefore increasing the cerebral tissue, or even renal, hypo oxygenation risk in case of pulmonary deficiency. For low ECMO support levels (<=75%), proximal arteries did not receive any oxygenated blood. However, increasing the ECMO flow rate resulted in ample perfusion to proximal arteries, and suppression of blood flow pulsatility.
CONCLUSION: ECMO has been rapidly embraced to provide mechanical circulatory support for patients in cardiogenic shock. Yet, the ECMO circuit profoundly disrupts physiological hemodynamic patterns. We demonstrate that retrograde perfusion provided by peripheral cannulation markedly changes cerebral blood flow, and consequently leads to cerebral hypoxemia. Our findings, highlighting the expediency of computational tools to scrutinize life support systems, warrant systematic studies to mechanistically understand the clinical impact of cannulation strategy and its potential consequences of end-organ malperfusion.