Blood Damage in Ventricular Assist Devices

Viscosity Modeling for Blood and Blood Analog Fluids in Narrow Gap and High Reynolds Numbers Flows

For the optimization of ventricular assist devices (VADs), flow simulations are crucial. Typically, these simulations assume single-phase flow to represent blood flow. However, blood consists of plasma and blood cells, making it a multiphase flow. Cell migration in such flows leads to a heterogeneous cell distribution, significantly impacting flow dynamics, especially in narrow gaps of less than 300 μm found in VADs. In these areas, cells migrate away from the walls, forming a cell-free layer, a phenomenon not usually considered in current VAD simulations. This paper addresses this gap by introducing a viscosity model that accounts for cell migration in microchannels under VAD-relevant conditions. The model is based on local particle distributions measured in a microchannels with a blood analog fluid. We developed a local viscosity distribution for flows with particles/cells and a cell-free layer, applicable to both blood and analog fluids, with particle volume fractions of up to 5%, gap heights of 150 μm, and Reynolds numbers around 100. The model was validated by comparing simulation results with experimental data of blood and blood analog fluid flow on wall shear stresses and pressure losses, showing strong agreement. This model improves the accuracy of simulations by considering local viscosity changes rather than assuming a single-phase fluid. Future developments will extend the model to physiological volume fractions up to 40%.

Effect of Particle Migration on the Stress Field in Microfluidic Flows of Blood Analog Fluids at High Reynolds Numbers

In the present paper, we investigate how the reductions in shear stresses and pressure losses in microfluidic gaps are directly linked to the local characteristics of cell-free layers (CFLs) at channel Reynolds numbers relevant to ventricular assist device (VAD) applications. For this, detailed studies of local particle distributions of a particulate blood analog fluid are combined with wall shear stress and pressure loss measurements in two complementary set-ups with identical flow geometry, bulk Reynolds numbers and particle Reynolds numbers. For all investigated particle volume fractions of up to 5%, reductions in the stress and pressure loss were measured in comparison to a flow of an equivalent homogeneous fluid (without particles). We could explain this due to the formation of a CFL ranging from 10 to 20 μm. Variations in the channel Reynolds number between Re = 50 and 150 did not lead to measurable changes in CFL heights or stress reductions for all investigated particle volume fractions. These measurements were used to describe the complete chain of how CFL formation leads to a stress reduction, which reduces the apparent viscosity of the suspension and results in the Fåhræus-Lindqvist effect. This chain of causes was investigated for the first time for flows with high Reynolds numbers (Re∼100), representing a flow regime which can be found in the narrow gaps of a VAD.

Flow simulation-based particle swarm optimization for developing improved hemolysis models

The improvement and development of blood-contacting devices, such as mechanical circulatory support systems, is a life saving endeavor. These devices must be designed in such a way that they ensure the highest hemocompatibility. Therefore, in-silico trials (flow simulations) offer a quick and cost-effective way to analyze and optimize the hemocompatibility and performance of medical devices. In that regard, the prediction of blood trauma, such as hemolysis, is the key element to ensure the hemocompatibility of a device. But, despite decades of research related to numerical hemolysis models, their accuracy and reliability leaves much to be desired. This study proposes a novel optimization path, which is capable of improving existing models and aid in the development of future hemolysis models. First, flow simulations of three, turbulent blood flow test cases (capillary tube, FDA nozzle, FDA pump) were performed and hemolysis was numerically predicted by the widely-applied stress-based hemolysis models. Afterward, a multiple-objective particles swarm optimization (MOPSO) was performed to tie the physiological stresses of the simulated flow field to the measured hemolysis using an equivalent of over one million numerically determined hemolysis predictions. The results show that our optimization is capable of improving upon existing hemolysis models. However, it also unveils some deficiencies and limits of hemolysis prediction with stress-based models, which will need to be addressed in order to improve its reliability.

Numerical study on the performance of centrifugal blood pump with superhydrophobic surface

AIM

In order to reduce the blood damage of an artificial heart pump and optimize its hydraulic performance, a centrifugal blood pump with superhydrophobic characteristics is proposed in this study.

METHODS

To study the influence of superhydrophobic surface characteristics on the performance of centrifugal blood pumps, the Navier slip model is used to simulate the slip characteristics of superhydrophobic surfaces, which is realized by the user defined function of ANSYS fluent. The user defined functions with different values of slip length are verified by two benchmark solutions of laminar flow and turbulence in the pipeline. The blood pump model adopts the designed centrifugal blood pump, and its head, hydraulic efficiency and hemolysis index are calculated. The Navier slip boundary condition (a constant slip-length of 50 μm) is applied to the walls of the blood pump impeller and a volute at different positions, and the influence of the superhydrophobic surface on the performance of the blood pump at the design point Q = 6 L/min was compared and analyzed.

RESULTS

The results show that the centrifugal blood pump model used in this paper has good blood compatibility and meets the design requirements; the superhydrophobic surface can significantly reduce the scalar shear stress in the blood pump. At the design point, when the slip length is 50 μm, the mass-average scalar shear stress in the impeller area and the volute area reduction rate is about 5.9%, the hydraulic efficiency growth rate is about 3.8%, the hemolysis index reduction rate is about 18.4%, and the pressure head changes little with a growth rate of 0.3%.

CONCLUSIONS

Centrifugal blood pumps with superhydrophobic surfaces can improve the efficiency of blood pumps and reduce hemolysis. Based on these encouraging results, vitro investigations for actual blood damage would be practicable.

Biomechanical properties of erythrocytes circulating in artificial hearts measured by dielectrophoretic method

Blood damage is recognized as one of the major problems caused by non-physiological shear force induced by artificial hearts. At present, the generally accepted manifestation of mechanical blood damage is the amount of free hemoglobin released into the blood. However, there is little research on the changes of blood cell state after circulating in artificial hearts at the single-cell level. It is well known that the mechanical properties of cells are of enormous relevance in the regulation of cellular physiological and pathological processes. In this regard, it is highly needed to study the mechanical properties of blood cells affected by non-physiological shear force. In this paper, a dielectrophoresis-based method of measuring the mechanical properties of erythrocytes circulating in artificial hearts was proposed, which was quantified with some crucial parameters such as strain, elongation index (EI), and Young's modulus. Experimental results indicated that with the increase of the working time of artificial hearts, the deformability of erythrocytes decreased, the stiffness substantially increased, and the mechanical stability decreased, particularly at long exposure times. The proposed method provides a deep insight into the mechanism of subhemolytic damage at the single-cell level and has a great potential to serve as a new tool for in vitro evaluation of potential blood damage in artificial hearts.

Turbulence and turbulent flow structures in a ventricular assist device-A numerical study using the large-eddy simulation

Numerical flow simulations that analyze the turbulent flow characteristics within a turbopump are important for optimizing the efficiency of such machines. In the case of ventricular assist devices (VADs), turbulent flow characteristics must be also examined in order to improve hemocompatibility. Turbulence increases the shear stresses in the VAD flow, which can lead to an increased damage to the transported blood components. Therefore, an understanding of the turbulent flow patterns and their significance for the numerical blood damage prediction is particularly important for flow optimizations in VADs in order to identify and thus minimize flow regions where blood could be damaged due to high turbulent stresses. Nevertheless, the turbulence occurring in VADs and the local turbulent structures that lead to increased turbulent stresses have not yet been analyzed in detail in these machines. Therefore, this study aims to investigate the turbulence in an axial VAD in a comprehensive and double tracked way. First, the flow in an axial VAD was computed using the large-eddy simulation method, and it was verified that the majority of the turbulence was directly resolved by the simulation. Then, the turbulent flow state of the VAD was quantified globally. For this purpose, a self-designed evaluation method, the power loss analysis, was used. Subsequently, local flow regions and flow structures were identified where significant turbulent stresses prevail. It will be shown that the identified regions are universal and will also occur in other axial blood pumps as well, for example, in the HeartMate II.

Computational fluid dynamics simulation of time-resolved blood flow in Budd-Chiari syndrome with inferior vena cava stenosis and its implication for postoperative efficacy assessment

BACKGROUND

This study aimed to adopt computational fluid dynamics to simulate the blood flow dynamics in inferior vena cava stenosis based on time-dependent patient-specific models of Budd-Chiari syndrome as well as a normal model. It could offer valuable references for a retrospective insight into the underlying mechanisms of Budd-Chiari syndrome pathogenesis as well as more accurate evaluation of postoperative efficacy.

METHODS

Three-dimensional inferior vena cava models of Budd-Chiari syndrome patient-specific (preoperative and postoperative) and normal morphology model were reconstructed as per magnetic resonance images using Simpleware. Moreover, computational fluid dynamics of time-resolved inferior vena cava blood flow were simulated using actual patient-specific measurements to reflect time-dependent flow rates.

FINDINGS

The assessment of the preoperative model revealed the dramatic variations of hemodynamic parameters of the stenotic inferior vena cava. Moreover, the comparison of the preoperative and postoperative models with the normal model as benchmark showed that postoperative hemodynamic parameters were markedly ameliorated via stenting, with the attenuation of overall velocity and wall shear stress, and the increase of pressure. However, the comparative analysis of the patient-specific simulations revealed that some postoperative hemodynamic profiles still bore some resemblance to the preoperative ones, indicating potential risks of restenosis.

INTERPRETATION

Computational fluid dynamics simulation of time-resolved blood flow could reveal the tight correlation between the hemodynamic characteristics and the pathological mechanisms of inferior vena cava stenosis. Furthermore, such time-resolved hemodynamic profiles could provide a quantitative approach to diagnosis, operative regimen and postoperative evaluation of Budd-Chiari syndrome with inferior vena cava stenosis.

Grid-Induced Numerical Errors for Shear Stresses and Essential Flow Variables in a Ventricular Assist Device: Crucial for Blood Damage Prediction?

Adverse events due to flow-induced blood damage remain a serious problem for blood pumps as cardiac support systems. The numerical prediction of blood damage via computational fluid dynamics (CFD) is a helpful tool for the design and optimization of reliable pumps. Blood damage prediction models primarily are based on the acting shear stresses, which are calculated by solving the Navier–Stokes equations on computational grids. The purpose of this paper is to analyze the influence of the spatial discretization and the associated discretization error on the shear stress calculation in a blood pump in comparison to other important flow quantities like the pressure head of the pump. Therefore, CFD analysis using seven unsteady Reynolds-averaged Navier–Stokes (URANS) simulations was performed. Two simple stress calculation indicators were applied to estimate the influence of the discretization on the results using an approach to calculate numerical uncertainties, which indicates discretization errors. For the finest grid with 19 × 106 elements, numerical uncertainties up to 20% for shear stresses were determined, while the pressure heads show smaller uncertainties with a maximum of 4.8%. No grid-independent solution for velocity gradient-dependent variables could be obtained on a grid size that is comparable to mesh sizes in state-of-the-art blood pump studies. It can be concluded that the grid size has a major influence on the shear stress calculation, and therefore, the potential blood damage prediction, and that the quantification of this error should always be taken into account.

Blood damage in Left Ventricular Assist Devices: Pump thrombosis or system thrombosis?

Introduction:

Despite significant technical advancements in the design and manufacture of Left Ventricular Assist Devices, post-implant thrombotic and thromboembolic complications continue to affect long-term outcomes. Previous efforts, aimed at optimizing pump design as a means of reducing supraphysiologic shear stresses generated within the pump and associated prothrombotic shear-mediated platelet injury, have only partially altered the device hemocompatibility.

Methods:

We examined hemodynamic mechanisms that synergize with hypershear within the pump to contribute to the thrombogenic potential of the overall Left Ventricular Assist Device system.

Results:

Numerical simulations of blood flow in differing regions of the Left Ventricular Assist Device system, that is the diseased native left ventricle, the pump inflow cannula, the impeller, the outflow graft and the anastomosed downstream aorta, reveal that prothrombotic hemodynamic conditions might occur at these specific sites. Furthermore, we show that beyond hypershear, additional hemodynamic abnormalities exist within the pump, which may elicit platelet activation, such as recirculation zones and stagnant platelet trajectories. We also provide evidences that particular Left Ventricular Assist Device implantation configurations and specific post-implant patient management strategies, such as those allowing aortic valve opening, are more hemodynamically favorable and reduce the thrombotic risk.

Conclusion:

We extend the perspective of pump thrombosis secondary to the supraphysiologic shear stress environment of the pump to one of Left Ventricular Assist Device system thrombosis, raising the importance of comprehensive characterization of the different prothrombotic risk factors of the total system as the target to achieve enhanced hemocompatibility and improved clinical outcomes.

Effect of blade curvature on the hemolytic and hydraulic characteristics of a centrifugal blood pump

Aims:

Impeller design has a significant impact on the overall performance of a blood pump. In this study, the effect of the blade curvature was investigated by performing in silico and in vitro studies on a recently developed centrifugal blood pump.

Methods:

A computational fluid dynamics study was performed for the flow rates of 3–5 L/min at 2000 r/min. The computational fluid dynamics model was also applied on the US Food and Drug Administration (FDA) benchmark blood pump to validate our computational method. The relative hemolysis index was calculated with the Eulerian hemolysis estimation method for five impellers with the wrap angles ranging from 0° to 240°. Hydraulic experiments were conducted for the validation of computational fluid dynamics results. In addition, the curved-blade impeller (120°) and the straight-blade impeller (0°) were evaluated with in vitro hemolysis tests using human blood.

Results:

The wrap angle of 120° provided the best hydraulic and hemolytic performance. Pump achieved the physiologic operating pressures and flows with 85–115 mmHg at 2.5–5.9 L/min. Compared to the straight-blade impeller, the 120° model reduces the relative hemolysis index and the plasma-free hemoglobin near 72.8% and 56.7%, respectively. Comparison of in silico and in vitro results indicated the similar trend to the blade curvature.

Conclusion:

Introducing a blade curvature enhanced the hydrodynamic and hemolytic performance compared to the straight-blade configuration for the investigated centrifugal blood pump. The findings of this study provide new insights into centrifugal blood pump design by examining the influence of the blade curvature.

Large eddy simulation in a rotary blood pump: Viscous shear stress computation and comparison with unsteady Reynolds-averaged Navier-Stokes simulation

PURPOSE:

Numerical flow analysis (computational fluid dynamics) in combination with the prediction of blood damage is an important procedure to investigate the hemocompatibility of a blood pump, since blood trauma due to shear stresses remains a problem in these devices. Today, the numerical damage prediction is conducted using unsteady Reynolds-averaged Navier-Stokes simulations. Investigations with large eddy simulations are rarely being performed for blood pumps. Hence, the aim of the study is to examine the viscous shear stresses of a large eddy simulation in a blood pump and compare the results with an unsteady Reynolds-averaged Navier-Stokes simulation.

METHODS:

The simulations were carried out at two operation points of a blood pump. The flow was simulated on a 100M element mesh for the large eddy simulation and a 20M element mesh for the unsteady Reynolds-averaged Navier-Stokes simulation. As a first step, the large eddy simulation was verified by analyzing internal dissipative losses within the pump. Then, the pump characteristics and mean and turbulent viscous shear stresses were compared between the two simulation methods.

RESULTS:

The verification showed that the large eddy simulation is able to reproduce the significant portion of dissipative losses, which is a global indication that the equivalent viscous shear stresses are adequately resolved. The comparison with the unsteady Reynolds-averaged Navier-Stokes simulation revealed that the hydraulic parameters were in agreement, but differences for the shear stresses were found.

CONCLUSION:

The results show the potential of the large eddy simulation as a high-quality comparative case to check the suitability of a chosen Reynolds-averaged Navier-Stokes setup and turbulence model. Furthermore, the results lead to suggest that large eddy simulations are superior to unsteady Reynolds-averaged Navier-Stokes simulations when instantaneous stresses are applied for the blood damage prediction.