(696b) Hemolysis Prediction from CFD Simulations of Turbulent Blood Flow in a Functioning and Malfunctioning Bi-Leaflet Artificial Heart Valve

Artificial heart valves are commonly used devices for treating valvular defects. However, these prosthetic devices may expose the blood to turbulent flow conditions leading to high stress that can damage blood cells. The purpose of this research is to simulate blood flow in both a functioning and malfunctioning bi-leaflet artificial heart valve and predict the damage caused to red blood cells (RBCs) using eddy analysis. The magnitude of the stress and exposure time blood experiences, as determined by the size and spatial distribution of turbulent flow eddies derived from computational fluid dynamics (CFD) simulations, is used to calculate the expected percent hemolysis. Using CFD software ANSYS DesignModeler, two prosthetic heart valve models were constructed. In one valve, both leaflets were kept open in a fully functioning position, while in the other, one leaflet was kept mostly closed to replicate a serious problem that can occur in an implanted valve. Blood flow simulations and calculations were done using ANSYS Fluent and validated with experimental findings available in the literature. Results from the CFD simulations provided the spatial distribution of Kolmogorov length scales that were used for eddy analysis in the heart valves. The Kolmogorov length scale (KLS), or the dissipative length scale, was used to characterize eddies in the flow field. The analysis is centered on the hypothesis that only some of the turbulent flow eddies – those with size comparable to or smaller than the size of RBCs – are the ones that contribute to cell damage. This is because smaller KLS reflect higher stress levels, which causes more damage to RBCs at the surface of an eddy. This CFD-based research utilized the total surface area of small eddies in the blood as a way to predict the amount of hemolysis experienced by the blood cells. The spatial distribution of these small eddies provides an indication of the extent of exposure of the RBCs to these stresses. The flow fields, stresses, and eddy distributions were then compared between the two valves. Early results indicate that hemolysis levels are low, suggesting the need for further study of subhemolytic damage. However, hemolysis predictions could be used as a comparative analysis to determine what situations may cause increased damages. For example, additional preliminary results have shown an increase in hemolysis levels in the malfunctioning heart valve when compared to a functioning heart valve under the same flow conditions of simulation.