Crucial Aspects for Using Computational Fluid Dynamics as a Predictive Evaluation Tool for Blood Pumps.

Area of Science:

• Biomedical Engineering
• Computational Fluid Dynamics (CFD)
• Medical Device Safety

Background:

• Standardized validation methods for CFD in medical device safety, especially hemocompatibility, are lacking.
• This limits the reliability and comparability of simulation studies for regulatory purposes.

Purpose of the Study:

• To investigate validation and verification aspects of CFD turbulence models for medical device safety assessment.
• To compare simulation results with experimental data for a benchmark blood pump model.

Main Methods:

• Evaluated three turbulence models (laminar, k-ω SST, SBES) and three mesh refinements.
• Compared CFD predictions of pressure head, velocity, and shear stress against experimental data.
• Utilized the Food and Drug Administration critical path initiative's benchmark blood pump.

Main Results:

• All tested models achieved <6.1% average deviation in hydraulic pump characteristics at high mesh resolution.
• The SBES model accurately predicted velocity fields (error <2.9%), unlike laminar and SST models.
• Shear stress quantification required higher near-wall mesh resolutions than pressure head or velocity validation.

Conclusions:

• CFD can be validated for pressure head and velocity in blood pumps.
• Accurate shear stress prediction for hemocompatibility assessment necessitates refined mesh verification.
• Future CFD blood damage predictions require rigorous mesh analysis for credibility.