DESIGN OF AXIAL AND CENTRIFUGAL PUMPS FOR LVAD AND THE HEMODYNAMICS OF A DISEASED HEART WITH SUCH PUMPS IN LOOP

Abstract

Chronic heart failure (CHF) is the leading cardiovascular disease in which the ventricle loses its ability to fill with or to eject the blood, causing millions of deaths every year. Further, due to the shortage of heart donors, it is not possible to perform heart transplantation treatment for every critical patient. Under such a scenario, a left ventricular assist device (LVAD) effectively treats end‐stage heart failures aiding the operation of the heart by battery driven mechanical pumps. LVADs not only increase the life expectancy they also improve the quality of life for an end stage heart patient. LVADs are under continuous development with inputs from multiple discipline. A thorough understanding of the hemodynamics of healthy and ailing heart as well as heart and LVAD combination is essential for such developments. In this regard, computational methods can provide very useful information economically and within a reasonable period of time. In this work, numerical investigations have been performed to propose improved design of pumps to be used as LVAD. Further, the efficacy of such pumps to support ailing heart has also been studied computationally. A computational fluid dynamic (CFD) based numerical investigation has been carried out to suggest an optimum design of centrifugal flow LVAD with a novel hemodynamic levitation technique for the impeller. Several design features of the pump have been explored to generate multiple options. For instance, design features like blade profile, blade number, blade tip width and inclusion of splitter blades have been investigated to propose a better design that minimizes the hemolytic complications associated with existing LVADs. The blood has been considered as non-Newtonian fluid and the hemodynamics inside the LVAD has been modeled using Bird-Carreu rheological model. In the end, an optimum design of centrifugal blood pump for the assistance of failed ventricle is proposed that can effectively pump the blood from the left ventricle to the ascending aorta. The range of desired pressure head (90‐140 mm Hg, as per the human physiology) for the most possible cases of blood flow rates (2‐8 L/min) can be achieved at 1600 and 1700 rpm without any complication of hemolysis and thrombus formation. The proposal can be adopted by LVAD designers to have a hemodynamically tuned efficient product. Next to this, numerical efforts have been made to improve the existing design of axial flow pumps for the LVAD. The up-gradation in the existing design has been targeted to improve the blood flow path to avoid clot or thrombus formation. Further, emphasis has also been given to ii reduce shear generation and blood damage. The performance of several design versions of axial LVAD in terms of hydraulic and hemolytic aspects has been tested at a wide range of operating speeds. To capture the change in the flow field near the rotating blade, a transient blade row model was employed. The proposal of the spiral blade impeller was found to be more compatible as per the hemolytic performance. It considerably reduces the blood damage to two times lesser value than that by the helical blade version and also improves the quality blood flow field. It also generates around 26% lesser shear with uniform pressure distribution than the existing design. The spiral blade provides a guiding path to the blood particle and avoids the mixing of different bloodstreams, thus reducing the eddy losses. Finally, a patient-specific simulations have been carried out to numerically investigate the influence of the proposed LVAD on the hemodynamics of the left heart. Two different computational domains with left heart have been simulated over the entire cardiac cycle (Case-I: Healthy heart without LVAD and Case-II: Diseased heart with LVAD). The blood flow was simulated by implementing Bird-Carreau non- Newtonian model. The obtained results are then compared with the fluidics of a healthy left heart without any support of LVAD. A significant change in hemodynamics is noticed after LVAD implantation; Heart loses its rhythmic blood wave (pulsatile flow output). In particular, major changes in the fluidics are observed inside the aortic region. For instance, post LVAD implantation, approximately eight times more wall shear stress is noticed at the aorta during the ventricular systole. However, the measured hemolysis index was well below the permissible limit specified by the United States Food and Drug Administration (US FDA) for long-term implantation. The change in hemodynamics is observed in the left heart and is also noticed in the blood pump. The flow characteristics due to the tubular assumption of pump inlet are observed to be very regular and uniform compared to the realistic case (inlet from the ventricle through inflow cannula). The observations of this study provide information about critical zones to be monitored in case of a patient-specific heart with LVAD.