Impact of aortic morphology on VA-ECMO performance: a combined statistical shape modeling and CFD approach.

Introduction: The veno-arterial extracorporeal membrane oxygenation (VA ECMO) system is a temporary support procedure to provide oxygenated blood in patients with cardiac or respiratory failure. Despite being widely used in clinical reality, the VA ECMO has some well-known flaws, like the formation of a 'watershed zone' as a consequence of the mixing of native blood from the left ventricle and retrograde oxygenated blood from the ECMO pump. A deeper knowledge of the watershed zone behavior could contribute to better defining the oxygenation balancing for the patient. Methods: From this concept, this study proposes a numerical approach combined with Statistical Shape Modeling (SSM) to evaluate the effect of aortic morphology variations on the different conditions of VA ECMO support. In particular, a new SSM approach was adopted to define 48 synthetic aortic morphologies starting from patient-specific computed tomography datasets. After assessing the morphological differences, the first 10 modes were selected to generate volumetric meshes for a Computational Fluid Dynamics (CFD) analysis. A total of 20 cases were simulated in the healthy conditions, while four cases were simulated with a 70% reduction of cardiac output and three different levels of ECMO support (0, 4, and 6 L/min). Results: From the simulated results, it was possible to assess the behavior of the watershed zone as a function of aortic morphological parameters and ECMO level. Discussion: The results confirmed the significant effect of ECMO level on the position and distribution of the zone.