Implantable axialflow blood pump for left ventricular support

Artificial blood pumps, either ventricular assist devices (VADs) or total artificial hearts, are currently employed for bridge to recovery, bridge to transplant, and destination therapy situations. The clinical effectiveness of VADs has been demonstrated; however, all of the currently available pumps have a limited life because of either the damage they cause to blood or their limited mechanical design life. A magnetically suspended rotary blood pump offers the potential to meet the requirements of both extending design life and causing negligible blood damage due to superior hemodynamics. Therefore, over the last few years, efforts of an interdisciplinary research team at University of Virginia have been concentrated on the design and development of a fully implantable axial flow VAD with a magnetically levitated impeller (LEV-VAD). This paper details the second generation developmental prototype (LEV-VAD2 design configuration) and includes a complete CFD analysis of device performance. Based on encouraging results of the first design stage, including a good agreement between the CFD performance estimations and the experimental measurements, a second design phase was initiated in an attempt to enhance device flow performance and suspension system capabilities. Using iterative design optimization stages, the design of the impeller and the geometry of the stationary and rotating blades have been reevaluated. A thorough CFD analysis allowed for optimization of the blood flow path such that an optimal trade-off among the hydraulic performance, specific requirements of a blood pump, and manufacturing requirements has been achieved. Per the CFD results, the LEV-VAD2 produces 6 lpm and 100 mmHg at a rotational speed of 7,000 rpm. The pressure-flow performance predictions indicate the LEV-VAD2's ability to deliver adequate flow over physiologic pressures for rotational speeds varying from 5,000 to 8,000 rpm. The blood damage numerical predictions also demonstrate acceptable levels. The axial and radial forces estimated from the computational analysis are well within the range for which the magnetic suspension and motor configuration can compensate. As a consequence of this favorable performance, the current design configuration has been selected for prototype manufacturing and further experimental testing.