PIV measurements of flow in a centrifugal blood pump: steady flow

Magnetically suspended left ventricular assist devices have only one moving part, the impeller. The impeller has absolutely no contact with any of the fixed parts, thus greatly reducing the regions of stagnant or high shear stress that surround a mechanical or fluid bearing. Measurements of the mean flow patterns as well as viscous and turbulent stresses were made in a shaft-driven prototype of a magnetically suspended centrifugal blood pump at several constant flow rates (3-9 L/min) using particle image velocimetry (PIV). The chosen range of flow rates is representative of the range over which the pump may operate while implanted. Measurements on a three-dimensional measurement grid within several regions of the pump, including the inlet, blade passage, exit volute, and diffuser are reported. The measurements are used to identify regions of potential blood damage due to high shear stress and/or stagnation of the blood, both of which have been associated with blood damage within artificial heart valves and diaphragm-type pumps. Levels of turbulence intensity and Reynolds stresses that are comparable to those in artificial heart valves are reported. At the design flow rate (6 L/min), the flow is generally well behaved (no recirculation or stagnant flow) and stress levels are below levels that would be expected to contribute to hemolysis or thrombosis. The flow at both high (9 L/min) and low (3 L/min) flow rates introduces anomalies into the flow, such as recirculation, stagnation, and high stress regions. Levels of viscous and Reynolds shear stresses everywhere within the pump are below reported threshold values for damage to red cells over the entire range of flow rates investigated; however, at both high and low flow rate conditions, the flow field may promote activation of the clotting cascade due to regions of elevated shear stress adjacent to separated or stagnant flow.     

Two-dimensional color-mapping of turbulent shear stress distribution downstream of two aortic bioprosthetic valves in vitro.

Since artificial heart valve related complications such as thrombus formation, hemolysis and calcification are considered related to flow disturbances caused by the inserted valve, a thorough hemodynamic characterization of heart valve prostheses is essential. In a pulsatile flow model, fluid velocities were measured one diameter downstream of a Hancock Porcine (HAPO) and a Ionescu-Shiley Pericardial Standard (ISPS) aortic valve. Hot-film anemometry (HFA) was used for velocity measurements at 41 points in the cross-sectional area of the ascending aorta. Three-dimensional visualization of the velocity profiles, at 100 different instants during one mean pump cycle, was performed. Turbulence analysis was performed as a function of time by calculating the axial turbulence energy within 50 ms overlapping time windows during the systole. The turbulent shear stresses were estimated by using the correlation equation between Reynolds normal stress and turbulent (Reynolds) shear stress. The turbulent shear stress distribution was visualized by two-dimensional color-mapping at different instants during one mean pump cycle. Based on the velocity profiles and the turbulent shear stress distribution, a relative blood damage index (RBDI) was calculated. It has the feature of combining the magnitude and exposure time of the estimated shear stresses in one index, covering the entire cross-sectional area. The HAPO valve showed a skewed jet-type velocity profile with the highest velocities towards the left posterior aortic wall. The ISPS valve revealed a more parabolic-shaped velocity profile during systole. The turbulent shear stresses were highest in areas of high or rapidly changing velocity gradients. For the HAPO valve the maximum estimated turbulent shear stress was 194 N m-2 and for the ISPS valve 154 Nm-2. The RBDI was the same for the two valves. The turbulent shear stresses had magnitudes and exposure times that might cause endothelial damage and sublethal or lethal damage to blood corpuscules. The RBDI makes comparison between different heart valves easier and may prove important when making correlation with clinical observations.     

Characterization of erythrocyte membrane tension for hemolysis prediction in complex flows

Hemolysis is a persistent issue with blood-contacting devices. Many experimental and theoretical contributions over the last few decades have increased insight into the mechanisms of hemolysis in both laminar and turbulent flows, with the ultimate goal of developing a comprehensive, mechanistic hemolysis model. Many models assume that hemolysis scales with a resultant, scalar stress representing all components of the fluid stress tensor. This study critically evaluates this scalar stress hypothesis by calculating the response of the red blood cell membrane to different types of fluid stress (laminar shear and extension, and three turbulent shear and extension cases), each with the same scalar stress. It was found that even though the scalar stress is the same for all cases, membrane tension varied by up to three orders of magnitude. In addition, extensional flow causes constant tension, while tank-treading in shear flow causes periodic tension, with tank-treading frequency varying by three orders of magnitude among the cases. For turbulent flow, tension also depends on eddy size. It is concluded, therefore, that scalar stress alone is inadequate for scaling hemolysis. Fundamental investigations are needed to establish a new index of the fluid stress tensor that provides reliable hemolysis prediction across the wide range of complex flows that occur in cardiovascular devices.     

Mechanical degradation of polyacrylamide solutions as a model for flow induced blood damage in artificial organs

Human or animal blood is normally used as a test fluid for the in vitro evaluation of hemolysis by artificial organs. However, blood has some disadvantages (large biological variability and problems with cleaning the devices). For that reason, we searched for a reproducible technical fluid with blood-like flow characteristics that exhibits similar shear depending destruction. In this study, a direct comparison between erythrocyte damage of bovine blood and shear-induced degradation of polyacrylamide solution is given. A uniform shear field was applied to the fluids using a shear device with a plate-plate geometry. It was shown that similarities exist between erythrocytes disaggregation and breakdown of super molecular structures in polymer solutions, caused by mechanical stress. In both cases steady low shear viscositity was diminished and the elastic component of complex viscosity of blood and polymer solutions has been reduced. There is a correlation between shear-induced hemolysis of bovine blood and mechanical polymer-degradation, which depends on the applied shear stresses.     

Red Blood Cell Damage by Shear Stress

A series of careful studies has been made on blood damage in a rotational viscometer. Specific attention has been focused on the effects of solid surface interaction, centrifugal force, air interface interaction, mixing of sheared and unsheared layers, cell-cell interaction, and viscous heating. The results show that there is a threshold shear stress, 1500 dynes/cm², above which extensive cell damage is directly due to shear stress, and the various secondary effects listed above are negligible. By analysis of these results and those of prior workers it is shown that the exposure time-shear stress plane is divided into two distinct regimes. In the regime of relatively low stresses and exposure times there is relatively little damage, and the damage is dominated by solid surface interaction effects. In the other regime, at high stresses and exposure times, stress effects alone dominate and very high rates of hemolysis occur. The experimental findings of all prior workers are shown to be consistent when interpreted in this way.     

Asymptotically consistent numerical approximation of hemolysis

In a previous communication, we have proposed a numerical framework for the prediction of in vitro hemolysis indices in the preselection and optimization of medical devices. This numerical methodology is based on a novel interpretation of Giersiepen-Wurzinger blood damage correlation as a volume integration of a damage function over the computational domain. We now propose an improvement of this approach based on a hyperbolic equation of blood damage that is asymptotically consistent. Consequently, while the proposed correction has yet to be proven experimentally, it has the potential to numerically predict more realistic red blood cell destruction in the case of in vitro experiments. We also investigate the appropriate computation of the shear stress scalar of the damage fraction model. Finally, we assess the validity of this consistent approach with an analytical example and with some 3D examples.     

Enzyme release and morphological changes in leukocytes induced by mechanical trauma

Polymorphonuclear (PMN) leukocytes exposed to mechanical trauma in vitro will release enzymes both from azurophilic and specific granules at shear stress levels of between 75 and 150 dyn/cm2 for 10 min. In addition, at these shear stresses the leukocyte count in whole blood decreased only slightly and the number of ruptured leukocytes on Wright-stained blood films increased significantly. At higher shear stresses, enzyme release and leukocyte damage increased monotonically. Transmission electron microscopy evaluation of sheared PMNs revealed that remaining intact cells had minor morphological changes at stresses of 150 dyn/cm2. They were characterized by clublike cytoplasmic potrusions, spherical shape, and a circumferential distribution of cytoplasmic granules. At higher shear stresses (600 dyn/cm2) cell destruction was marked. Intact PMNs contained fewer cytoplasmic granules, a large number of vacuoles, and condensed nuclear chromatin. These studies show that PMN morphology and function are at least as sensitive to mechanical trauma as similar platelet alterations seen in other studies.     

Flash freezing of erythrocyte suspensions

Freezing whole blood in bulk usually results in severe cellular destruction through the action of ice crystals and osmotic effects in the freezing liquid. The potential of flash freezing blood aerosols onto a liquid nitrogen surface as a means of inhibiting cellular damage was studied in this work. Three commercial spraying devices were employed to spray-freeze either whole blood or concentrated erythrocyte suspensions, using hydroxyethyl starch (HES) as a cryoprotectant. The integrity and viability of the processed cells were assessed by measuring gross rheological properties and the extent of hemolysis. Cells were found to be susceptible to the very high shear stresses imposed by some of the spraying devices. Bulk freezing of blood, even in the presence of the cryoprotectant, resulted in complete cellular destruction. Whereas flash freezing was capable of substantially reducing the level of hemolysis to 12.6% and preserving the cellular deformability.     

Experimental investigation of the rheological activation of blood platelets

In order to define various aspects of platelet rheological activation, samples of whole blood and platelet-rich plasma (PRP) from the same donors were subjected for 5 min to shear rates increasing from 10 to 10000 sec-1 (shear stresses from 10(-2) to 30 Pa approximatively) in a Couette type viscometer. The following parameters were measured: erythrocyte hemolysis; lactic dehydrogenase activity; plasma B-Thromboglobulin (B-TG); adenine nucleotides, and platelet photometric aggregation. The experimental results reveal that: In whole blood, hemolysis only reached at maximum 2% of the total hemolysis. Plasma LDH activity increased regularly beyond 500 sec-1, in close correlation with B-TG plasma concentration. In contrast, ADP and ATP levels remained stable up to 1000 sec-1 then increased slowly. In PRP, the LDH, ADP and ATP levels remain practically stable up to shear rates around 5000 sec-1. In contrast, B-TG appeared to be released in plasma at shear rate values of 3000 sec-1 and its progression is only correlated with the other parameters, when the platelet lysis occurred. Finally, a rapid and complete inhibition of platelet aggregation to ADP was observed from 5000 sec-1.     

A reevaluation and discussion on the threshold limit for hemolysis in a turbulent shear flow

One major factor of red blood cell damage induced by artificial heart valves is the magnitude of turbulent shear stresses in the flow field. An often-cited threshold for hemolysis is 400 N/m(2) (Sallam and Hwang, Biorheology 21 (1984) 783). This value, however, was measured with a one-component laser Doppler anemometer and was not calculated from the major principal Reynolds shear stress, therefore underestimating the threshold. This study performed flow field measurements under similar conditions, using a two-component laser Doppler anemometer and determining major principal Reynolds shear stress, to reevaluate the hemolytic threshold as 800 N/m(2) with an exposure time of 1 ms. The Kolmogorov length scales, approximately 9 microm, are capable of inflicting direct damage to the red blood cells. These results will serve as a more accurate reference in future heart valve design and testing.     

RhoA sustains integrin alpha IIbbeta 3 adhesion contacts under high shear

The small GTPase RhoA modulates the adhesive nature of many cell types; however, despite high levels of expression in platelets, there is currently limited evidence for an important role for this small GTPase in regulating platelet adhesion processes. In this study, we have examined the role of RhoA in regulating the adhesive function of the major platelet integrin, alpha(IIb)beta(3). Our studies demonstrate that activation of RhoA occurs as a general feature of platelet activation in response to soluble agonists (thrombin, ADP, collagen), immobilized matrices (von Willebrand factor (vWf), fibrinogen) and high shear stress. Blocking the ligand binding function of integrin alpha(IIb)beta(3), by pretreating platelets with c7E3 Fab, demonstrated the existence of integrin alpha(IIb)beta(3)-dependent and -independent mechanisms regulating RhoA activation. Inhibition of RhoA (C3 exoenzyme) or its downstream effector Rho kinase had no effect on integrin alpha(IIb)beta(3) activation induced by soluble agonists or adhesive substrates, however, both inhibitors reduced shear-dependent platelet adhesion on immobilized vWf and shear-induced platelet aggregation in suspension. Detailed analysis of the sequential adhesive steps required for stable platelet adhesion on a vWf matrix under shear conditions revealed that RhoA did not regulate platelet tethering to vWf or the initial formation of integrin alpha(IIb)beta(3) adhesion contacts but played a major role in sustaining stable platelet-matrix interactions. These studies define a critical role for RhoA in regulating the stability of integrin alpha(IIb)beta(3) adhesion contacts under conditions of high shear stress.     

Wall shear stress during pulsatile flow distal to a normal porcine aortic valve

The shear stress at the wall has been of interest as one of the possible fluid dynamic factors that may be damaging in the region of prosthetic valves. The purpose of this study was to measure the axial wall shear stresses in the region of a 29 mm tissue annulus diameter porcine stent mounted prosthetic aortic valve (Hancock, Model 242). Studies were performed in an in vitro pulse duplicating system. The axial wall shear stress was calculated from velocities obtained near the wall with a laser Doppler anemometer. The largest axial wall shear stress was 29 dyn cm-2 and it occurred at the highest stroke volume used (80 ml). At a stroke volume of 50 ml, the largest axial wall shear stress was 17 dyn cm-2 and at a stroke volume of 35 ml, it was 15 dyn cm-2. Stresses of these magnitudes are far below those reported to be damaging to the endothelial surface. These stresses may be high enough, however, to affect platelet function.     

Velocity and turbulence measurements past mitral valve prostheses in a model left ventricle

Thrombogenesis and hemolysis have both been linked to the flow dynamics past heart valve prostheses. To learn more about the particular flow dynamics past mitral valve prostheses in the left ventricle under controlled experimental conditions, an in vitro study was performed. The experimental methods included velocity and turbulent shear stress measurements past caged-ball, tilting disc, bileaflet, and polyurethane trileaflet mitral valves in an acrylic rigid model of the left ventricle using laser Doppler anemometry. The results indicate that all four prosthetic heart valves studied create at least mildly disturbed flow fields. The effect of the left ventricular geometry on the flow development is to produce a stabilizing vortex which engulfs the entire left ventricular cavity, depending on the orientation of the valve. The measured turbulent shear stress magnitudes for all four valves did not exceed the reported value for hemolytic damage. However, the measured turbulent shear stresses were near or exceeded the critical shear stress reported in the literature for platelet lysis, a known precursor to thrombus formation.     

Two-component laser velocimeter measurements downstream of heart valve prostheses in pulsatile flow

Elevated turbulent shear stresses resulting from disturbed blood flow through prosthetic heart valves can cause damage to red blood cells and platelets. The purpose of this study was to measure the turbulent shear stresses occurring downstream of aortic prosthetic valves during in-vitro pulsatile flow. By matching the indices of refraction of the blood analog fluid and model aorta, correlated, simultaneous two-component laser velocimeter measurements of the axial and radial velocity components were made immediately downstream of two aortic prosthetic valves. Velocity data were ensemble averaged over 200 or more cycles for a 15-ms window opened at peak systolic flow. The systolic duration for cardiac flows of 8.4 L/min was 200 ms. Ensemble-averaged total shear stress levels of 2820 dynes/cm2 and 2070 dynes/cm2 were found downstream of a trileaflet valve and a tilting disk valve, respectively. These shear stress levels decreased with axial distance downstream much faster for the tilting disk valve than for the trileaflet valve.     

Glycoprotein (GP) Ib-IX-transfected cells roll on a von Willebrand factor matrix under flow. Importance of the GPib/actin-binding protein (ABP-280) interaction in maintaining adhesion under high shear

Adhesion of platelets to sites of vascular injury is critical for hemostasis and thrombosis and is dependent on the binding of the vascular adhesive protein von Willebrand factor (vWf) to the glycoprotein (GP) Ib-V-IX complex on the platelet surface. A unique but poorly defined characteristic of this receptor/ligand interaction is its ability to support platelet adhesion under conditions of high shear stress. To examine the structural domains of the GPIb-V-IX complex involved in mediating cell adhesion under flow, we have expressed partial (GPIb-IX), complete (GPIb-V-IX), and mutant (GPIbalpha cytoplasmic tail mutants) receptor complexes on the surface of Chinese hamster ovary (CHO) cells and examined their ability to adhere to a vWf matrix in flow-based adhesion assays. Our studies demonstrate that the partial receptor complex (GPIb-IX) supports CHO cell tethering and rolling on a bovine or human vWf matrix under flow. The adhesion was specifically inhibited by an anti-GPIbalpha blocking antibody (AK2) and was not observed with CHO cells expressing GPIbbeta and GPIX alone. The velocity of rolling was dependent on the level of shear stress, receptor density, and matrix concentration and was not altered by the presence of GPV. In contrast to selectins, which mediate cell rolling under conditions of low shear (20-200 s-1), GPIb-IX was able to support cell rolling at both venous (150 s-1) and arterial (1500-10,500 s-1) shear rates. Studies with a mutant GPIbalpha receptor subunit lacking the binding domain for actin-binding protein demonstrated that the association of the receptor complex with the membrane skeleton is not essential for cell tethering or rolling under low shear conditions, but is critical for maintaining adhesion at high shear rates (3000-6000 s-1). These studies demonstrate that the GPIb-IX complex is sufficient to mediate cell rolling on a vWf matrix at both venous and arterial levels of shear independent of other platelet adhesion receptors. Furthermore, our results suggest that the association between GPIbalpha and actin-binding protein plays an important role in enabling cells to remain tethered to a vWf matrix under conditions of high shear stress.     

A discussion on the threshold limit for hemolysis related to Reynolds shear stress.

Turbulence-related damage to blood is a major problem with the use of prosthetic devices, such as mechanical heart valves. An often-cited paper by Sallam and Hwang (1984). Biorheology 21, 783-797) quantified the threshold for hemolysis to be about 400 N m(-2), a value that has hitherto contributed to the evaluation of the potential dangerousness of a medical implantable device. We propose a discussion of the mentioned experiment, based on the application of stress analysis concepts to the original measurements: this is necessary to assess the peak turbulence shear stress value that could have been found in Sallam and Hwangs experiment, with a suitable orientation of the measurement axes. The result of our theoretical discussion is that the threshold value of 400 N m(-2) could probably be considerably underestimated: following this point of view, a 3-D stress analysis shows that the peak turbulence shear stress at the inception of hemolysis should be at least 600 N m(-2). This result, obtained on the basis of the study of RBCs' response to a turbulent environment, indicates that blood particles are probably more resistant to short-time shear stresses than it was thought.     

Continuous monitoring of human erythrocyte breakdown in the circulating blood by means of multiple light scattering

The method of hemolysis control in human circulating blood is described. The method was used for studying erythrocyte destruction resulting from osmotic swelling and shear stress in the blood flow and may be applied in the artificial blood circulation system.     

Microscopic investigation of erythrocyte deformation dynamics

The understanding of erythrocyte deformation under conditions of high shear stress and short exposure time is central to the study of hemorheology and hemolysis within prosthetic blood contacting devices. A combined computational and experimental microscopic study was conducted to investigate the erythrocyte deformation and its relation to transient stress fields. A microfluidic channel system with small channels fabricated using polydimethylsiloxane on the order of 100 mum was designed to generate transient stress fields through which the erythrocytes were forced to flow. The shear stress fields were analyzed by three-dimensional computational fluid dynamics. Microscopic images of deforming erythrocytes were experimentally recorded to obtain the changes in cell morphology over a wide range of fluid dynamic stresses. The erythrocyte elongation index (EI) increased from 0 to 0.54 with increasing shear stress up to 123 Pa. In this shear stress range, erythrocytes behaved like fluid droplets, and deformed and flowed following the surrounding fluid. Cells exposed to shear stress beyond 123 Pa (up to 5170 Pa) did not exhibit additional elongation beyond EI=0.54. Two-stage deformation of erythrocytes in response to shear stress was observed: an initial linear elongation with increasing shear stress and a plateau beyond a critical shear stress.     

Inhibition of Hb Binding to GP1bα Abrogates Hb-Mediated Thrombus Formation on Immobilized VWF and Collagen under Physiological Shear Stress

BackgroundReports including our own describe that intravascular hemolysis increases the risk of thrombosis in hemolytic disorders. Our recent study shows that plasma Hb concentrations correlate directly with platelet activation in patients with paroxysmal nocturnal hemoglobinuria (PNH). The binding of Hb to glycoprotein1bα (GP1bα) increases platelet activation. A peptide AA1-50, designed from N-terminal amino acid sequence of GP1bα significantly inhibits the Hb binding to GP1bα as well as Hb-induced platelet activation. This study further examined if the Hb-mediated platelet activation plays any significant role in thrombus formation on subendothelium matrix under physiological flow shear stresses and the inhibition of Hb-platelet interaction can abrogate the above effects of Hb.

Methods and Results

Study performed thrombus formation assay in vitro by perfusing whole blood over immobilized VWF or collagen type I in presence of Hb under shear stresses simulating arterial or venous flow. The Hb concentrations ranging from 5 to 10 μM, commonly observed level in plasma of the hemolytic patients including PNH, dose-dependently increased thrombus formation on immobilized VWF under higher shear stress of 25 dyne/cm², but not at 5 dyne/cm². The above Hb concentrations also increased thrombus formation on immobilized collagen under both shear stresses of 5 and 25 dyne/cm². The peptide AA1-50 abrogated invariably the above effects of Hb on thrombus formation.

Conclusions and Significance

This study therefore indicates that the Hb-induced platelet activation plays a crucial role in thrombus formation on immobilized VWF or collagen under physiological flow shear stresses. Thus suggesting a probable role of this mechanism in facilitating thrombosis under hemolytic conditions.     

Investigation of platelet margination phenomena at elevated shear stress

Thrombosis is a common complication following the surgical implantation of blood contacting artificial organs. Platelet transport, which is an important process of thrombosis and strongly modulated by flow dynamics, has not been investigated under the shear stress level associated with these devices, which may range from tens to several hundred Pascal.The current research investigated platelet transport within blood under supra-physiological shear stress conditions through a micro flow visualization approach. Images of platelet-sized fluorescent particles in the blood flow were recorded within microchannels (2 cm x 100 microm x 100 microm). The results successfully demonstrated the occurrence of platelet-sized particle margination under shear stresses up to 193 Pa, revealing a platelet near-wall excess up to 8.7 near the wall (within 15 microm) at the highest shear stress. The concentration of red blood cells was found to influence the stream-wise development of platelet margination which was clearly observed in the 20% Ht sample but not the 40% Ht sample. Shear stress had a less dramatic effect on the margination phenomenon than did hematocrit. The results imply that cell-cell collision is an important factor for platelet transport under supra-physiologic shear stress conditions. It is anticipated that these results will contribute to the future design and optimization of artificial organs.