Objective:
This paper presents a comparative study of computational fluid dynamics (CFD) simulations and in vitro hemolysis examinations of a bearingless centrifugal blood pump. The outcomes of the in vitro study are analyzed with the help of CFD hemolysis models.
Methods:
Several pump prototypes were manufactured and tested. Each model was implemented in a CFD framework and simulated with different Eulerian hemolysis models. The outcomes are compared to experimental data. The model achieving the highest correlation is used to explain the in vitro outcomes in detail.
Results:
It is shown that a double-stage model achieves the best correlation. The sensitivity of the simulation is considerably lower than that of in vitro tests. The CFD model reveals that most of the cell destruction is caused in the radial gap of the pump. Further critical regions are the bottom volume and the shroud clearance gap. Only 0.5% of the priming volume is subject to overcritical shear stress.
Conclusion:
Cell compatibility can be improved by increasing the radial gap, lowering the shroud and hub clearance gaps, and increasing the fillet radius of the inlet nozzle. CFD models can be used to examine the cell damage effects and help to further improve the pump design.
Significance:
This paper compares different Eulerian CFD hemolysis models, parameter sets, and equivalent shear stresses to several in vitro hemolysis tests. The sensitivity of the models is compared to that of in vitro studies. It is shown that CFD simulations have their limitations but can help with interpreting the outcomes of in vitro studies.
