(316ai) Blood Flow Analysis in Micromed Debakey Pump by Computational Fluid Dynamics | AIChE

(316ai) Blood Flow Analysis in Micromed Debakey Pump by Computational Fluid Dynamics

Authors 

Arora, D. - Presenter, Rice University
Behbahani, M. - Presenter, RWTH Aachen University
Lammel, A. - Presenter, Rice University
Benkowski, B. - Presenter, MicroMed Technologies
Hussain, F. - Presenter, University of Houston
Behr, M. - Presenter, RWTH Aachen University
Pasquali, M. - Presenter, Rice University


The MicroMed DeBakey axial blood pump is a state-of-the-art ventricular assist device (VAD) that provides long term support to end-stage heart-failure patients; it is considerably smaller than the other VADs currently on the market and weighs 95 grams. The simplicity of this pump makes it an ideal candidate for a permanent alternative to heart transplant. Achieving permanent implantation requires design modifications to improve further the hematologic and hydrodynamic performance. We are using computational fluid dynamics (CFD) to analyze how design modifications will affect the pump performance.

CFD analysis is initiated by transforming the computer aided design (CAD) geometry into a virtual prototype, through which the time-varying, incompressible and isothermal flow of blood is computed. The flow of blood, assumed to be Newtonian, is modeled by the Navier-Stokes equations, and solved by the Deformable-Spatial-Domain/Stabilized-Space-Time finite element method. A special purpose mesh-update scheme---shear slip mesh update method---is employed to handle the impeller rotation efficiently. The pump flow rate and impeller speed are ramped up from zero to the desired values, and stationary flow state is determined by analyzing the forces on the impeller. Several clinical operating conditions of the pump---combinations of impeller speeds and flow rates---are simulated on meshes of varying refinement levels. Mesh convergence is determined by analyzing the forces on the impeller. The pump characteristics, for example, regions of high shear, are studied with the help of pathlines and velocity vectors. The extent of hemolysis (in terms of normalized index of hemolysis) caused by the flow in the pump is predicted by using a tensor-based blood damage model, which tracks the strain (tensor) experienced by the red blood cells along the pathlines. Regions of high shear and slow-moving flow, which are important for predicting and thereby minimizing platelet activation and thrombosis, are also identified