Jingchun Wu

Computational fluid dynamics analysis of blade tip clearances on hemodynamic performance and blood damage in a centrifugal ventricular assist device.

An important challenge facing the design of turbodynamic ventricular assist devices (VADs) intended for long-term support is the optimization of the flow path geometry to maximize hydraulic performance while minimizing shear-stress-induced hemolysis and thrombosis. For unshrouded centrifugal, mixed-flow and axial-flow blood pumps, the complex flow patterns within the blade tip clearance between the lengthwise upper surface of the rotating impeller blades and the stationary pump housing have a dramatic effect on both the hydrodynamic performance and the blood damage production. Detailed computational fluid dynamics (CFD) analyses were performed in this study to investigate such flow behavior in blade tip clearance region for a centrifugal blood pump representing a scaled-up version of a prototype pediatric VAD. Nominal flow conditions were analyzed at a flow rate of 2.5 L/min and rotor speed of 3000 rpm with three blade tip clearances of 50, 100, and 200 microm. CFD simulations predicted a decrease in the averaged tip leakage flow rate and an increase in pump head and axial thrust with decreasing blade tip clearances from 200 to 50 microm. The predicted hemolysis, however, exhibited a unimodal relationship, having a minimum at 100 microm compared to 50 microm and 200 microm. Experimental data corroborate these predictions. Detailed flow patterns observed in this study revealed interesting fluid dynamic features associated with the blade tip clearances, such as the generation and dissipation of tip leakage vortex and its interaction with the primary flow in the blade-blade passages. Quantitative calculations suggested the existence of an optimal blade tip clearance by which hydraulic efficiency can be maximized and hemolysis minimized.

Elimination of adverse leakage flow in a miniature pediatric centrifugal blood pump by computational fluid dynamics-based design optimization.

Fetal bypass presents several perfusion challenges, including the need for high arterial flow rates using flexible arterial and small venous cannulae. We hypothesized that vacuum-assisted venous drainage (VAVD) would improve drainage and allow perfusion at higher flow rates which are thought to prevent placental dysfunction induced by fetal bypass. We conducted bypass for 60 minutes in 14 fetal lambs (90-105 days gestation; ∼1-1.5 kg) using a roller pump and various angled venous cannulae (8–12 Fr). VAVD at –20 mm Hg or –40 mm Hg was compared with gravity drainage. Average flow using gravity drainage was 139 ml/kg/min; after VAVD, we achieved average flows of 285 ml/kg/min (range, 109–481 ml/kg/min). VAVD at –40 mm Hg caused right atrial trauma in four fetuses; no injury was seen at –20 mm Hg. Venous air entrainment during repair of the injuries did not result in any apparent air embolism. Spontaneous pulmonary hemorrhage occurred in two fetuses at the highest flows (≥ 400 ml/kg/min). In all but one case, termination of bypass was followed by placental dysfunction within 120 minutes. VAVD can be safely applied during fetal bypass provided pressures are kept ≤ –20 mm Hg. However, the achieved higher flow rates do not prevent postbypass placental dysfunction and may indeed be detrimental to the fetus.

A Computational and Experimental Comparison of Two Outlet Stators for the Nimbus LVAD

Two designs of an outlet stalor for the Nimbus axial flow left ventricular assist device (LVAD) are analyzed at nominal operating conditions. The original stator assembly (Design 1) has significant flow separation and reversal. A second stator assembly (Design 2) replaces the original tubular outer housing with a converging-diverging throat section with the intention of locally improving the fluid dynamics. Both stator designs are analyzed using computational fluid dynamics (CFD) analysis and experimental particle imaging flow visualization (PIFV). The computational and experimental methods indicate: 1) persistent regions of flow separation in Design 1 and improved fluid dynamics in Design 2; 2) blade-to-blade velocity fields that are well organized at the blade tip yet chaotic at the blade hub for both designs; and 3) a moderate decrease in pressure recovery for Design 2 as compared with Design 1. The CFD analysis provides the necessary insight to identify a subtle, localized flow acceleration responsible for the decreased hydraulic efficiency of Design 2. In addition, the curiously low thrombogenicity of Design 1 is explained by the existence of a three-dimensional unsteady vortical flow structure that enhances boundary advection. ASAIO Journal 1999; 45:328–333.

Computational Fluid Dynamics-Based Design Optimization for an Implantable Miniature Maglev Pediatric Ventricular Assist Device

According to recently available statistics, approximately 36,000 new cases of congenital heart disease (CHD) occur each year [1]. Of these, several studies suggest that 9200, or 2.3 per 1000 live births, require invasive treatment or result in death in the first year of life [2]. The very limited options available to treat ventricular failure in these infants and young children have motivated us to develop the PediaFlow® ventricular assist system, which features a miniature rotodynamic blood pump having a magnetically levitated impeller and streamlined blood flow path. It is intended to be fully implantable, providing chronic (up to 6 months) circulatory support from birth to 2 years of age (3 kg to 15 kg body weight) with a nominal pressure rise of ∼100 mmHg for the flow range of 0.3 ∼ 2.3 L/min. By consideration of the hemodynamic requirements of this population [3], a nominal design point of 1.5 LPM with a target pressure rise of 100 mmHg was chosen for the design procedure. An important design requirement is the need for optimizing and miniaturizing the flow path to maximize hemodynamic performance while minimizing shear-stress induced blood trauma. A unique feature of magnetically levitated axial-flow blood pumps in general, and the PediaFlow® in particular, is a continuous annular fluid gap between rotor and housing. The dimensions of this gap are limited by the requirements for magnetic stiffness and motor efficiency, but must be sufficiently large to permit desired flow of blood and to avoid excessive shear stress and other flow disturbances. When shared with the impeller blades, a narrow annular flow gap generally necessitates greater rotational speeds to generate sufficient pressure rise and flow rate. However, the combination of small gap and high speed can be a formula for blood cell damage without sufficient optimization of flow path geometry including the blade profiles. Because of design tradeoffs such as these, which span across several subsystems of the PediaFlow® (electromagnetics, heat transfer, rotordynamics, etc.), our group has adopted a numerical, multidisciplinary approach to optimization. With regard to the flow path, we employed a CFD-based design optimization system developed by Wu et al. [4,5] that integrates a robust and flexible inverse blade design tool, automatic mesh generators, parameterized geometry models, and mathematical models of blood damage, integrated with commercial CFD packages. The PediaFlow® pump has evolved over four generations, denoted as PF1, PF2, PF3, and PF4. The PF3 evolved from its predecessor (PF2, [6]) by the realization that the rotor can operate above its rotordynamic critical rotational speed, which reduced the requirement for magnetic suspension stiffness. This, in turn, permitted a relatively larger annular gap; hence, smaller rotor diameter. Although PF3 demonstrated excellent in vivo biocompatibility over 72 days [7], adverse fluid–structure interaction caused an unstable operational range greater than 0.8 L/min and 18,000 revolutions per minute (rpm). This study describes the use of the aforementioned CFD-based design optimization tools for overcoming this limitation and thereby expanding the operating range and improving hydrodynamic performance in transitioning from the PF3 to a frozen PF4 design, as illustrated in Fig. ​Fig.11. Fig. 1 Two generations of the PediaFlow® ventricular assist device: PF3 and PF4. Cutaway (bottom) shows critical internal components of PF4.

Towards the development of a pediatric ventricular assist device.

The very limited options available to treat ventricular failure in children with congenital and acquired heart diseases have motivated the development of a pediatric ventricular assist device at the University of Pittsburgh (UoP) and University of Pittsburgh Medical Center (UPMC). Our effort involves a consortium consisting of UoP, Childrens Hospital of Pittsburgh (CHP), Carnegie Mellon University, World Heart Corporation, and LaunchPoint Technologies, Inc. The overall aim of our program is to develop a highly reliable, biocompatible ventricular assist device (VAD) for chronic support (6 months) of the unique and high-risk population of children between 3 and 15 kg (patients from birth to 2 years of age). The innovative pediatric ventricular assist device we are developing is based on a miniature mixed flow turbodynamic pump featuring magnetic levitation, to assure minimal blood trauma and risk of thrombosis. This review article discusses the limitations of current pediatric cardiac assist treatment options and the work to date by our consortium toward the development of a pediatric VAD.

Thermal Analysis of the PediaFlow pediatric ventricular assist device.

Accurate modeling of heat dissipation in pediatric intracorporeal devices is crucial in avoiding tissue and blood thermotrauma. Thermal models of new Maglev ventricular assist device (VAD) concepts for the PediaFlow VAD are developed by incorporating empirical heat transfer equations with thermal finite element analysis (FEA). The models assume three main sources of waste heat generation: copper motor windings, active magnetic thrust bearing windings, and eddy currents generated within the titanium housing due to the two-pole motor. Waste heat leaves the pump by convection into blood passing through the pump and conduction through surrounding tissue. Coefficients of convection are calculated and assigned locally along fluid path surfaces of the three-dimensional pump housing model. FEA thermal analysis yields a three-dimensional temperature distribution for each of the three candidate pump models. Thermal impedances from the motor and thrust bearing windings to tissue and blood contacting surfaces are estimated based on maximum temperature rise at respective surfaces. A new updated model for the chosen pump topology is created incorporating computational fluid dynamics with empirical fluid and heat transfer equations. This model represents the final geometry of the first generation prototype, incorporates eddy current heating, and has 60 discrete convection regions. Thermal analysis is performed at nominal and maximum flow rates, and temperature distributions are plotted. Results suggest that the pump will not exceed a temperature rise of 2°C during normal operation.

Monitoring development of suction in an LVAD

Excessive speed in rotary ventricular assist pumps can create negative pressure, or suction, in the left ventricle (LV). This study evaluated several parameters derived from pump current as indicators of suction and tested them using hemodynamic data from calves. A normalized 2/sup nd/ harmonic parameter and variance of the current waveform showed the most consistent changes, approaching minimum as suction developed.

High fidelity computational simulation of thrombus formation in Thoratec HeartMate II continuous flow ventricular assist device.

Continuous flow ventricular assist devices (cfVADs) provide a life-saving therapy for severe heart failure. However, in recent years, the incidence of device-related thrombosis (resulting in stroke, device-exchange surgery or premature death) has been increasing dramatically, which has alarmed both the medical community and the FDA. The objective of this study was to gain improved understanding of the initiation and progression of thrombosis in one of the most commonly used cfVADs, the Thoratec HeartMate II. A computational fluid dynamics simulation (CFD) was performed using our recently updated mathematical model of thrombosis. The patterns of deposition predicted by simulation agreed well with clinical observations. Furthermore, thrombus accumulation was found to increase with decreased flow rate, and can be completely suppressed by the application of anticoagulants and/or improvement of surface chemistry. To our knowledge, this is the first simulation to explicitly model the processes of platelet deposition and thrombus growth in a continuous flow blood pump and thereby replicate patterns of deposition observed clinically. The use of this simulation tool over a range of hemodynamic, hematological, and anticoagulation conditions could assist physicians to personalize clinical management to mitigate the risk of thrombosis. It may also contribute to the design of future VADs that are less thrombogenic.

In Silico Design and In-Vivo Analysis of the Pediaflow™ Pediatric Ventricular Assist Device

Mechanical circulatory support for the smallest newborn pediatric patients has historically been limited to extracorporeal membrane oxygenation, which can only provide several days to weeks of full cardiac support; far short of the median waiting time for pediatric heart transplantation of nearly three months [1]. Recently, new technologies have been developed, including the PediaFlow pediatric ventricular assist device, to address this need. The PediaFlow device is a magnetically levitated (mag lev), mixed flow turbodynamic blood pump which has been developed in large part in silico using CFD-based inverse design optimization and closed form rotor dynamics models [2, 3]. Each prototype undergoes a series of in vitro and in vivo tests to verify the accuracy of the simulations in predicting performance and biocompatibility. The overall goal is continued refinement and progress towards an implantable pump that produces 0.3 −1.5 L/min for up to 6 months in pediatric heart failure patients from 5 to 15 kg. We describe here the design principles and test procedures for the first three prototypes as well as the predicted performance for a fourth prototype currently being prepared for testing (Figure 1).Copyright

CFD-BASED OPTIMIZATION OF MAGNETICALLY LEVITATED PEDIAFLOW VENTRICULAR ASSIST DEVICE

FLOW VENTRICULAR ASSIST DEVICE Jingchun Wu, James F Antaki, Jeff Gardiner, Dave Paden, Brad E Paden, Harvey S Borovetz. Department of CFD, LaunchPoint Technologies, Goleta, CA; Biomedical Engineering & Computer Science, Carnegie Mellon University, Pittsburgh, PA; Mechanical and Environmental Engineering, University of California, Santa Barbara, Santa Barbara, CA; Department of Bioengineering and McMcGowan Institute for Regenerative Medicine, University of Pittsburgh, Pittsburgh, PA. We have utilized a computational fluid dynamic (CFD)-based design optimization system to develop a miniature magnetically levitated pediatric ventricular assist device, the PediaFlow, in support of neonates and infants weighing 3–15 Kg. The optimization tightly integrates custom-developed 3-D inverse design methods, parameterized geometric models, automatic mesh generators and mathematical blood damage models with a commercial CFD package, CFX. The PediaFlow with a specific speed of 0.56 was designed with a mixed-flow style impeller and a single annular flow gap of 0.5mm between the rotor and housing to avoid unfavorable retrograde flow and separation. All the flow path including the inlet and outlet cannula elbows were optimized to eliminate vortices and flow separation over a wide range of conditions between 0.3 1.5 lpm. Blood damage was evaluated by the Lagrangian particle tracking technique and a new scalar shear stress equation was evaluated in this study. According to our CFD results, the PediaFlow can generate 100 mmHg at 0.5 lpm and 1.5 lpm with a rotor speed of 8,000 rpm and 11,200 rpm, respectively. The CFD predicted pressure-flow (H-Q) and efficiency-flow ( -Q) characteristics were validated by invitro testing. The smooth local flow characteristics over the entire flow range were also confirmed by flow-visualization. FEASIBILITY OF THE RABBIT MODEL IN TESTING NEONATAL/PEDIATRIC CARDIAC ASSIST SYSTEMS Egemen Tuzun, Fred Baimbridge, Jeff Conger, Gil Costas, Vivian Dimas, OH Frazier, Brano Radovancevic. Cardiovascular Research Laboratories, Texas Heart Institute, Houston, TX; Pediatric Cardiology, Texas Children’s Hospital, Houston, TX. Purpose: To evaluate the feasibility of the rabbit model in testing neonatal/pediatric cardiac assist systems. Methods: Three New Zealand white rabbits (average weight, 4.5 kg) were used. After heparinization, an inflow cannula (10 Fr) was inserted into the left ventricular apex. An outflow cannula (12 Fr) was inserted into the left ventricle and advanced transaortically into the descending aorta while the heart was beating. The Levitronix UltraMag, an extracorporeal centrifugal pump, was used for mechanical circulatory support. Arterial pressure was continuously monitored throughout the study. Results: The pump’s rotation was maintained between 1600 rpm and 1800 rpm, with a flow range of 250–300 mL/min. All animals were successfully supported for 4 hours, at which time, the studies were electively terminated. Conclusion: The rabbit is an accurate, acute neonate model for studying cardiac assist systems. Further studies are being conducted to evaluate the rabbit as a chronic neonate/pediatric model.