Computational Design and Experimental Performance Testing of an Axial-Flow Pediatric Ventricular Assist Device

Abstract

The Virginia Artificial Heart Institute continues to design and develop an axial-flow pediatric ventricular assist device (PVAD) for infants and children in the United States. Our research team has created a database to track potential PVAD candidates at the University of Virginia Children's Hospital. The findings of this database aided with need assessment and design optimization of the PVAD. A numerical analysis of the optimized PVAD1 design (PVAD2 model) was also completed using computational fluid dynamics (CFD) to predict pressure-flow performance, fluid force estimations, and blood damage levels in the flow domain. Based on the PVAD2 model and after alterations to accommodate manufacturing, a plastic prototype for experimental flow testing was constructed via rapid prototyping techniques or stereolithography. CFD predictions demonstrated a pressure rise range of 36-118 mm Hg and axial fluid forces of 0.8-1.7 N for flows of 0.5-3 l/min over 7000-9000 rpm. Blood damage indices per CFD ranged from 0.24% to 0.35% for 200 massless and inert particles analyzed. Approximately 187 (93.5%) of the particles took less than 0.14 seconds to travel completely through the PVAD. The mean residence time was 0.105 seconds with a maximum time of 0.224 seconds. Additionally, in a water/glycerin blood analog solution, the plastic prototype produced pressure rises of 20-160 mm Hg for rotational speeds of 5960 +/- 18 rpm to 9975 +/- 31 rpm over flows from 0.5 to 4.5 l/min. The numerical results for the PVAD2 and the prototype hydraulic testing indicate an acceptable design for the pump, represent a significant step in the development phase of this device, and encourage manufacturing of a magnetically levitated prototype for animal experiments.

The use of computational fluid dynamics in the development of ventricular assist devices

“…The implementations of these CFD packages are either based on the finite volume23 method or finite element method. Fluent 36, 37, 38, 39, 40, 41 and CFX 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 11, 57, 58, 59, both available from ANSYS, Inc. (Canonsburg, PA) and STAR-CD from CD-Adapco 60, 61 (Melville, New York) are the CFD software packages used most often for analysis and modelling of blood flow in VADs. In addition, AcuSolve from ACUSIM Software 62 (Mountain View, CA) and Adina from Adina R&D (Watertown, MA) are also used.…”

“…In addition to adult patients with end stage HF, paediatric patients with ventricular dysfunction (congenital or acquired) constitute another group requiring circulatory support. Development of VADs for paediatric patients has been slower, due to fewer potential patients and the complexity of paediatric pathophysiologies, and the intrinsic difficulty of producing small nonthrombogenic devices 10, 11. The widely acknowledged need for pumps specifically designed for these patient groups has lead to the development of several new devices in the last five years, in part due to funding from the NIH12.…”

Numerical Design and Experimental Hydraulic Testing of an Axial Flow Ventricular Assist Device for Infants and Children

“…9,10 Blood behaves as a Newtonian fluid for conditions of high shear stresses (Ͼ0.7 Pa) and shear rates above 100 s Ϫ1 . 9,10,14,18 The shear stresses dominating the flow paths in this computational model are large enough that the assumption of Newtonian fluid properties is acceptable. In addition, the blood analog fluid that is used in the experimental testing of the prototype is Newtonian with comparable fluid properties.…”

“…Because of the centered position of the impeller and axisymmetric flow during these CFD simulations, the radial forces exerted on the rotor were found to range from 10 Ϫ3 to 10 Ϫ2 N, similar to previous pump models. 9,10 Figure 9 depicts the scalar stress distribution through the PVAD3 model along the hub or rotor surface. As expected, the highest fluid stresses were found at the leading edge of the impeller blades along the blade tips (Figure 10).…”

Pumping Rate Study of a Left Ventricular Assist Device in a Mock Circulatory System

“…28,39 A mixture of glycerin (40%) and water (60%) is used as the fluid medium and blood analog solution. 40 In the first group, all the four test conditions are reproduced by using four HRs, which are set at 55, 60, 66, and 70 beats per minute (bpm), respectively. In the second group, the test conditions are conducted by using seven HRs, which are set at 55, 60, 65, 70, 75, 80, and 85 bpm, respectively.…”