Computational Fluid Dynamics Analysis of the RealHeart™ demonstrates low power consumption and low thrombosis risk

Background/Purpose: The RealHeart™ is a 4-chamber, pulsatile, total artificial heart (TAH) which mimics the pumping principle and output of the natural heart through cyclical displacement of the atrioventricular (AV) plane. The novel design aims to produce pulsatile flow with a low power requirement and minimal damage to the blood. In this work a computational model was used to assess those aims. Methods: A numerical model of the left side of the RealHeart™ was created using a commercial finite-element solver (Autodesk CFD). The motion of the AV plane, deformation of the chamber wall membranes, and valves, was incorporated by superimposing moving solid meshes on the fluid domain. The inlet boundary was constant pressure (0 mmHg). At the outlet boundary a distributed resistance simulated the peripheral vasculature, and was set to produce aortic pressure variation of 70-120mmHg. Blood viscosity was shear thinning using a Carreau model. Simulations were run for a range of heart rates (70-120 bpm) and stroke lengths (33-55% systole). The mechanical power transmitted to the blood was calculated from the hydraulic force on the AV plane and membranes. Shear stress and stagnation risks were assessed by finding regions of the pump experiencing shear stress >13 Pa and velocity <1 mm/s. Results: Operating at 100 bpm a flow of 6.96l/min was calculated, which is in good agreement with the experimental measurement 6.41l/min. Shear stress regions were small and concentrated at the valves (maximum stress=30 Pa). Low velocity regions were small and lasted only for diastole. The mechanical power transmitted to the blood in the simulated left half of the RealHeart™ was 4.6 W. Similar shear stress, stagnation and power results were observed for other heart rates. Conclusions: Based on this numerical analysis, the RealHeart™ has low risk of haemolysis and thrombosis, and low power consumption.