The effect of fluid shear and co-adsorbed proteins on the stability of immobilized fibrinogen and subsequent platelet interactions

Fluid Dynamic Study of the Penn State Pediatric Total Artificial Heart

Penn State University is developing a pediatric total artificial heart (TAH) as a bridge-to-transplant device that supports infants and small children with single ventricle anomalies or biventricular heart failure to address high waitlist mortality rates for pediatric patients with severe congenital heart disease (CHD). Two issues with mechanical circulatory support devices are thrombus formation and thromboembolic events. This in vitro study characterizes flow within Penn State's pediatric total artificial heart under physiological operating conditions. Particle image velocimetry (PIV) is used to quantify flow within the pump and to calculate wall shear rates (WSRs) along the internal pump surface to identify potential thrombogenic regions. Results show that the diastolic inflow jets produce sufficient wall shear rates to reduce thrombus deposition potential along the inlet side of the left and right pumps. The inlet jet transitions to rotational flow, which promotes wall washing along the apex of the pumps, prevents flow stasis, and aligns flow with the outlet valve prior to systolic ejection. However, inconsistent high wall shear rates near the pump apex cause increased thrombogenic potential. Strong systolic outflow jets produce high wall shear rates near the outlet valve to reduce thrombus deposition risk. The right pump, which has a modified outlet port angle to improve anatomical fit, produces lower wall shear rates and higher thrombus susceptibility potential (TSP) compared to the left pump. In summary, this study provides a fluid dynamic understanding of a new pediatric total artificial heart and indicates thrombus susceptibility is primarily confined to the apex, consistent with similar pulsatile heart pumps.

Extracting Mural and Volumetric Growth Patterns of Platelet Aggregates on Engineered Surfaces by Use of an Entity Tracking Algorithm

Thrombosis is a major complication that can occur in both blood-contacting devices and regions and in regions of vascular damage. Microfluidic devices are popular templates to model various thrombogenic settings and to assess conditions that lead to bulk channel occlusion. However, area-averaged measurements miss the opportunity to extract real-time information on thrombus evolution and early dynamics of thrombus formation and propagation, which result in late-stage bulk channel occlusion. To clarify these dynamics, we have developed a standalone tracking algorithm that uses consecutive image connectivity and minimal centroid distance mappings to uniquely index all appearing thrombi in fluorescence time-lapse videos http://links.lww.com/ASAIO/A887 , and http://links.lww.com/ASAIO/A888 . This leads to measurements of all individual aggregates that can in turn be studied as ensembles. We applied tracking to fluorescence time-lapse videos http://links.lww.com/ASAIO/A887 , and http://links.lww.com/ASAIO/A888 of thrombosis across both collagen-functionalized substrate and across the surface of a roughened titanium alloy (Ti6Al4V) at a shear rate of 4000 s -1 . When comparing ensemble-averaged measurements to area-averaged metrics, we unveil immediate, steady thrombus growth at early phases on collagen surfaces and unstable thrombus attachment to roughened Ti6Al4V surfaces on Ti6Al4V surfaces. Additionally, we introduce tracked thrombus eccentricity and fluorescence intensity as additional volumetric measures of thrombus growth that relate back to the primary thrombosis mechanism at play. This work advocates for the complementation of surface macrostate metrics with characteristic thrombus microstate growth patterns to accurately predict critical thrombosis events.

Thrombogenicity of biomaterials depends on hemodynamic shear rate

BACKGROUND

While it is well recognized that different biomaterials induce thrombosis at low shear rates, the effect of high shear rates may be quite different. We hypothesize that the amount of thrombus formation on a given material can be greatly influenced by the local shear rate.

METHODS

We tested this hypothesis with two different whole blood perfusion loop assays to quantify biomaterial thrombogenicity as a function of shear stress. One assay uses obstructive posts (pins) of material positioned centrally in a tube perfused at high shear rate of >5000/s for 24 h. A second assay uses a parallel plate chamber to perfuse low (<150/s), medium (~500/s), and high shear rates over 96 h. We evaluated the thrombogenicity of seven different biomaterials including stainless steel, acrylic, ceramic, Dacron, polytetrafluoroethylene (PTFE), silicone, and polyvinyl chloride (PVC).

RESULTS

For the pin assay, thrombus mass was significantly greater for stainless steel than either zirconia ceramic or acrylic (p < 0.001). Similarly, the parallel plate chamber at high shear showed that steel and PTFE (p < 0.02) occluded the chamber faster than acrylic. In contrast, a low shear parallel plate chamber revealed that stainless steel and PTFE were least thrombogenic, while silicone, Dacron, and other plastics such as acrylic were most thrombogenic. Histology revealed that high shear thrombi had a large proportion of platelets not seen in the low shear fibrin-rich thrombi.

CONCLUSION

This differential thrombogenicity based on shear rate conditions may be important in the selection of biomaterials for blood-contacting devices.

Comparison of Diazeniumdiolated Dialkylhexanediamines as Nitric Oxide Release Agents on Nonthrombogenicity in an Extracorporeal Circulation Model

When blood from a patient is circulated through extracorporeal circuits (ECCs), such as in cardiopulmonary bypass or extracorporeal life support, platelets in the blood are activated and form a thrombus. This is prevented clinically with a range of different systemic anticoagulation agents (e.g., heparin); however, this increases a patient's risk of hemorrhage. Previous work with nitric oxide (NO) releasing materials using the combined diazeniumdiolated diamine, N-N-di-N'-butyl-1,6-hexanediamine (DBHD), and a polymer-linked thrombin inhibitor, argatroban (AG), showed significant nonthrombogenicity in ECCs using a 4 h rabbit model. Herein, we evaluated if diazeniumdiolated N-N-di-N'-propyl-1,6-hexanediamine (DPHDN₂O₂), which has a slightly lower degree of lipophilicity compared to DBHDN₂O₂, would provide similar nonthrombogenicity as the AG/DBHDN₂O₂-polymer-coated circuits. While DPHDN₂O₂ releases NO at a higher flux rate than DBHDN₂O₂ when coated (within CarboSil polymer) on the inner wall of polyvinyl chloride tubing, neither coated circuit significantly affected animal hemodynamics. Both diazeniumdiolated diamines, in combination with immobilized AG or alone, significantly reduced thrombus formation similarly in the 4 h rabbit model (vs uncoated control): AG/DBHDN₂O₂: 0.12 ± 0.03 cm²; DBHDN₂O₂: 2.57 ± 0.82 cm²; AG/DPHDN₂O₂: 0.68 ± 0.22 cm²; DPHDN₂O₂: 1.87 + 1.26 cm²; uncoated control: 6.95 ± 0.82 cm². AG/DPHDN₂O₂ was no different than AG/DBHDN₂O in preserving platelet count and function. In addition, AG did not leach into the systemic circulation as the total clotting times were insignificantly different from the baseline values (AG/DPHDN₂O₂: 12.7 + 0.5 s (n = 3); AG/DBHDN₂O₂: 12.3 + 0.7 s (n = 3); baseline: 13.9 + 0.3 s (n = 13)). Based on these results, both DPHDN₂O₂ and DPHDN₂O₂ are good candidates as NO donor molecules for creating nonthrombogenic polymer coatings for ECCs.

Evaluating medical device and material thrombosis under flow: current and emerging technologies

This review highlights the importance of flow in medical device thrombosis and explores current and emerging technologies to evaluate dynamic biomaterial Thrombosis in vitro.

Direct interaction of fibrinogen with lipid microparticles modulates clotting kinetics and clot structure

Extensive studies revealed the role of blood lipid microparticles (liposomes, microvesicles) in activation of coagulation cascade. The direct interaction of fibrinogen/fibrin with lipid surfaces and its consequence for hemostasis received much less attention. We observed pronounced changes in both clot morphology and kinetics of fibrin clotting in the presence of artificial liposomes. The evidence was obtained that lipid microparticles per se present a diffusion barrier to the three-dimensional fibril assembling and pose spatial restrictions for fiber elongation. On the other hand, fibrinogen adsorption results in its high local concentration on liposome surface that accelerates fibrin polymerization. Adsorption induces Fg secondary structure alterations which may contribute to the abnormal clot morphology. In dependence on lipid composition and size of microparticles, the interplay of all the outlined mechanisms determines functionally important changes of clot morphology. The obtained results contribute to the knowledge of clotting mechanisms in the presence of artificial and natural lipid microparticles.

Free hemoglobin increases von Willebrand factor-mediated platelet adhesion in vitro: implications for circulatory devices

Intravascular hemolysis occurs in patients on extracorporeal membrane oxygenation. High levels of free acellular adult hemoglobin (free HbA) are associated with clotting in this mechanical device that can result in thrombotic complications. Adsorption of fibrinogen onto the surface of biomaterial correlates with platelet adhesion, which is mediated by von Willebrand factor (VWF). Because free Hb interacts with VWF, we studied the effect of hemoglobin (Hb) on platelet adhesion to fibrin(ogen) under conditions of different hydrodynamic forces. This effect was investigated using purified human HbA and fibrinogen, extracellular matrix, collagen, or purified plasma VWF as surface-coated substrates to examine flow-dependent platelet adhesion. Antibodies and VWF-deficient plasma were also used. Free Hb (≥50 mg/dL) effectively augmented platelet adhesion, and microthrombi formation on fibrin(ogen), extracellular matrix, and collagen at high shear stress. The effect of free Hb was effectively blocked by anti-glycoprotein Ibα (GPIbα) antibodies or depletion of VWF. Unexpectedly, free Hb also promoted firm platelet adhesion and stable microthrombi on VWF. Lastly, we determined that Hb interacts directly with the A1 domain. This study is the first to demonstrate that extracellular Hb directly affects the GPIbα-VWF interaction in thrombosis, and describes another mechanism by which hemolysis is connected to thrombotic events.

Platelet adhesion to polyurethane urea under pulsatile flow conditions

Platelet adhesion to a polyurethane urea surface is a precursor to thrombus formation within blood-contacting cardiovascular devices, and platelets have been found to adhere strongly to polyurethane surfaces below a shear rate of approximately 500 s(-1). The aim of the current work is to determine the properties of platelet adhesion to the polyurethane urea surface as a function of time-varying shear exposure. A rotating disk system was used to study the influence of steady and pulsatile flow conditions (e.g., cardiac inflow and sawtooth waveforms) for platelet adhesion to the biomaterial surface. All experiments were conducted with the same root mean square angular rotation velocity (29.63 rad/s) and waveform period. The disk was rotated in platelet-rich bovine plasma for 2 h, with adhesion quantified by confocal microscopy measurements of immunofluorescently labeled bovine platelets. Platelet adhesion under pulsating flow was found to decay exponentially with increasing shear rate. Adhesion levels were found to depend upon peak platelet flux and shear rate, regardless of rotational waveform. In combination with flow measurements, these results may be useful for predicting regions susceptible to thrombus formation within ventricular assist devices.

The Use of Fluid Mechanics to Predict Regions of Microscopic Thrombus Formation in Pulsatile VADs

We compare the velocity and shear obtained from particle image velocimetry (PIV) and computational fluid dynamics (CFD) in a pulsatile ventricular assist device (VAD) to further test our thrombus predictive methodology using microscopy data from an explanted VAD. To mimic physiological conditions in vitro, a mock circulatory loop is used with a blood analog that matched blood's viscoelastic behavior at 40% hematocrit. Under normal physiologic pressures and for a heart rate of 75 bpm, PIV data is acquired and wall shear maps are produced. The resolution of the PIV shear rate calculations are tested using the CFD and found to be in the same range. A bovine study, using a model of the 50 cc Penn State V-2 VAD, for 30 days at a constant beat rate of 75 beats per minute (bpm) provides the microscopic data whereby after the 30 days, the device is explanted and the sac surface analyzed using scanning electron microscopy (SEM) and, after immunofluorescent labeling for platelets and fibrin, confocal microscopy. Areas are examined based on PIV measurements and CFD, with special attention to low shear regions where platelet and fibrin deposition are most likely to occur. Data collected within the outlet port in a direction normal to the front wall of the VAD shows that some regions experience wall shear rates less than 500 s^-1, which increases the likelihood of platelet and fibrin deposition. Despite only one animal study, correlations between PIV, CFD, and in vivo data show promise. Deposition probability is quantified by the thrombus susceptibility potential, a calculation to correlate low shear and time of shear with deposition.

An investigation on platelet transport during thrombus formation at micro-scale stenosis

This paper reports on an investigation of mass transport of blood cells at micro-scale stenosis where local strain-rate micro-gradients trigger platelet aggregation. Using a microfluidic flow focusing platform we investigate the blood flow streams that principally contribute to platelet aggregation under shear micro-gradient conditions. We demonstrate that relatively thin surface streams located at the channel wall are the primary contributor of platelets to the developing aggregate under shear gradient conditions. Furthermore we delineate a role for red blood cell hydrodynamic lift forces in driving enhanced advection of platelets to the stenosis wall and surface of developing aggregates. We show that this novel microfluidic platform can be effectively used to study the role of mass transport phenomena driving platelet recruitment and aggregate formation and believe that this approach will lead to a greater understanding of the mechanisms underlying shear-gradient dependent discoid platelet aggregation in the context of cardiovascular diseases such as acute coronary syndromes and ischemic stroke.

The influence of device position on the flow within the Penn State 12 cc pediatric ventricular assist device

Ventricular assist devices are a commonly used heart failure therapy for adult patients as bridge-to-transplant or bridge-to-recovery tools. The application of adult ventricular assist devices in pediatric patients has led to increased thrombotic events. Therefore, we have been developing a pediatric ventricular assist device (PVAD), the Penn State 12 cc PVAD. It is designed for patients with a body weight of 5-15 kg and has a stroke volume of 12 cc. Clot formation is the major concern. It is correlated to the coagulability of blood, the blood contacting materials and the fluid dynamics within the system. The intent is for the PVAD to be a long term therapy. Therefore, the system may be oriented in different positions according to the patient's behavior. This study evaluates for the first time the impact of position on the flow patterns within the Penn State 12 cc PVAD, which may help to improve the PVAD design concerning chamber and ports geometries. The fluid dynamics are visualized by particle image velocimetry. The evaluation is based on inlet jet behavior and calculated wall shear rates. Vertical and horizontal model orientations are compared, both with a beat rate of 75, outlet pressures of 90/60 mm Hg and a flow rate of 1.3 l/min. The results show a significant change of the inlet jet behavior and the development of a rotational flow pattern. Vertically, the inlet jet is strong along the wall. It initiates a rotational flow pattern with a wandering axis of rotation. In contrast, the horizontal model orientation results show a weaker inlet jet along the wall with a nearly constant center of rotation location, which can be correlated to a higher risk of thrombotic events. In addition, high speed videography illustrates differences in the diaphragm motion during diastole. Diaphragm opening trajectories measurements determine no significant impact of the density of the blood analog fluids. Hence, the results correlate to human blood.

A thrombus susceptibility comparison of two pulsatile Penn State 50 cc left ventricular assist device designs

Left ventricular assist devices (LVADs) have proven successful as bridge to transplant devices for patients awaiting donor organs. While survival rates continue to increase, destination therapy remains hindered by thrombus formation within the device. Research has shown that thrombosis is correlated to the fluid dynamics within the device and may be a result of sustained shear rates below 500 s(-1) on the polyurethane blood sac used in the Penn State pulsatile LVAD. Particle image velocimetry is used to compare flow within two 50 cc LVAD designs to assess fluid patterns and quantify wall shear rates in regions known from in vivo studies to be susceptible to thrombus formation. The two designs differ in their front face geometry. The V-1 model has an outward-facing "dome" whereas the face of the V-2 model is flat. A thrombus susceptibility metric, which uses measured wall shear rates and exposure times, was applied to objectively compare pump designs over the entire cardiac cycle. For each design, there are regions where wall shear rates remained below 500 s(-1) for the entire cardiac cycle resulting in high thrombus susceptibility potential. Results of this study indicate that the V-2 device had an overall lower propensity for thrombus formation in the current region of interest.

Computational fluid dynamics design and analysis of a passively suspended Tesla pump left ventricular assist device

This article summarizes the use of computational fluid dynamics (CFD) to design a novel suspended Tesla left ventricular assist device. Several design variants were analyzed to study the parameters affecting device performance. CFD was performed at pump speeds of 6500, 6750, and 7000 rpm and at flow rates varying from 3 to 7 liters per minute (LPM). The CFD showed that shortening the plates nearest the pump inlet reduced the separations formed beneath the upper plate leading edges and provided a more uniform flow distribution through the rotor gaps, both of which positively affected the device hydrodynamic performance. The final pump design was found to produce a head rise of 77 mm Hg with a hydraulic efficiency of 16% at the design conditions of 6 LPM through flow and a 6750 rpm rotation rate. To assess the device hemodynamics the strain rate fields were evaluated. The wall shear stresses demonstrated that the pump wall shear stresses were likely adequate to inhibit thrombus deposition. Finally, an integrated field hemolysis model was applied to the CFD results to assess the effects of design variation and operating conditions on the device hemolytic performance.

The attenuation of platelet and monocyte activation in a rabbit model of extracorporeal circulation by a nitric oxide releasing polymer

Nitric oxide (NO) has been shown to reduce thrombogenicity by decreasing platelet and monocyte activation by the surface glycoprotein, P-selectin and the integrin, CD11b, respectively. In order to prevent platelet and monocyte activation with exposure to an extracorporeal circulation (ECC), a nitric oxide releasing (NORel) polymeric coating composed of plasticized polyvinyl chloride (PVC) blended with a lipophilic N-diazeniumdiolate was evaluated in a 4 h rabbit thrombogenicity model using flow cytometry. The NORel polymer significantly reduced ECC thrombus formation compared to polymer control after 4 h blood exposure (2.8 +/- 0.7 NORel vs 6.7 +/- 0.4 pixels/cm(2) control). Platelet count (3.4 +/- 0.3 NORel vs 2.3 +/- 0.3 x 10(8)/ml control) and function as measured by aggregometry (71 +/- 3 NORel vs 17 +/- 6% control) were preserved after 4 h exposure in NORel versus control ECC. Plasma fibrinogen levels significantly decreased in both NORel and control groups. Platelet P-selectin mean fluorescence intensity (MFI) as measured by flow cytometry was attenuated after 4 h on ECC to ex vivo collagen stimulation (27 +/- 1 NORel vs 40 +/- 2 MFI control). Monocyte CD11b expression was reduced after 4 h on ECC with NORel polymer (87 +/- 14 NORel vs 162 +/- 30 MFI control). These results suggest that the NORel polymer coatings attenuate the increase in both platelet P-selectin and monocytic CD11b integrin expression in blood exposure to ECCs. These NO-mediated platelet and monocytic changes were shown to improve thromboresistance of these NORel-polymer-coated ECCs for biomedical devices.

Surface-adsorbed fibrinogen and fibrin may activate the contact activation system

INTRODUCTION

This study was designed to investigate whether fibrinogen, soluble desAA-fibrin, and insoluble desAABB-fibrin are able to induce clotting by triggering the plasma contact activation system when adsorbed to polystyrene.

MATERIALS AND METHODS

The above-mentioned substances were individually prepared on polystyrene meshwork squares, and then exposed to a purified FXII solution or non-calcium containing plasma (citrated and dialyzed normal pooled plasma) in polystyrene cuvettes coated with surface-immobilized heparin, to completely block contact activation and the coagulation mechanism that might be induced by the cuvette surfaces. Sodium glass beads were used as the reference material.

RESULTS

On exposure to purified FXII solution and plasma, all the tested materials adsorbed and activated FXII to varying degrees. This activation led to the formation of FXIa in the exposed plasma, with the highest activation occurring upon exposure to glass, desAA-fibrin and desAABB-fibrin and the lowest upon exposure to fibrinogen-adsorbed or unmodified polystyrene meshwork squares. Following recalcification, in cuvettes with surface-immobilized heparin, a spectrophotometric assay showed that the surface-exposed plasma aliquots clotted within 5 min after contact with glass, within 10 to 15 min after contact with the two forms of fibrin, and somewhat longer after contact with adsorbed fibrinogen. The longest lag phase, close to 20 min, occurred in plasma exposed to unmodified polystyrene meshwork. Whole blood deposited in surface heparinized cuvettes directly from the cubital vein did not clot during the observation time (2 h).

CONCLUSIONS

These results indicate that domains induced by conformational changes in adsorbed fibrinogen and fibrin are capable of activating adsorbed proenzymes and that various forms of fibrin are considerably stronger activators of the contact activation system than are adsorbed fibrinogen or a polystyrene meshwork. The delayed coagulation in plasma exposed to the unmodified polystyrene meshwork can be explained by a two-step process: first, adsorption of fibrinogen, and second, activation of FXII. Under our experimental conditions, the adsorption and activation of FXII on fibrinogen and fibrin seems to be an important mechanism for triggering coagulation.

Short-term in vivo studies of surface thrombosis in a left ventricular assist system

Thrombosis continues to be a major adverse and at times fatal event in patients with left ventricular assist systems (LVAS). To assess acute thrombosis in an LVAS, multiscale analysis of surface thrombosis was performed on LVAS blood sacs retrieved after implantation in seven calves for 3 days. Two study groups were evaluated: One group was given heparin and warfarin sodium throughout the study; the second received no postoperative anticoagulation. On explantation, the blood sacs were examined for macroscopic thrombi; microscale thrombosis was assessed with the use of scanning electron microscopy. Macroscopic thrombi about 1 mm in diameter were seen in all sacs from both groups. Although macroscopic thrombi occurred in all sac regions, scanning electron microscopy revealed differences in microscale topography between the port regions and the other sac regions. The primary structure was spherical particles approximately 400 nm in diameter, found to occur at a lower density in the ports. In contrast, the highest densities of proteinaceous rough topography and fibrillar structures consistent with fibrin clot were seen in the port regions. The density distribution of these structures was different in the eight sac regions, and anticoagulation therapy appeared to have no effect on surface thrombosis in these short-term LVAS implants.

Multiscale analysis of surface thrombosis in vivo in a left ventricular assist system

Thrombosis limits the success of ventricular assist devices as the demand for alternatives to heart transplants is increasing. This study mapped the occurrence of thrombosis in a left ventricular assist system (LVAS) to better understand the biologic response to these devices. Nine calves divided into two groups were implanted with LVAS for 28 to 30 days. One group was anticoagulated, whereas the second group received no long-term anticoagulation. The blood-contacting poly(urethane urea) surfaces of blood sacs in the LVAS were examined for macroscopic thrombi upon retrieval. The sac was partitioned into eight sections and imaged for thrombi by scanning electron microscopy. No difference in thrombosis was observed macroscopically between the groups. Anticoagulation appeared to result in reduction of platelet-like structures, but the presence of fibrin-like structures remained similar between groups. Regional differences correlating with high and low shear stress regions were observed. At the macroscale, fewer thrombi were recorded in the high shear stress ports. At the microscale, features resembling fibrin were observed primarily in the ports and platelet-like features were common in lower shear stress regions. These variations in thrombosis with anticoagulation and location are likely due to varied fluid dynamics within the LVAS blood sac.

Cell attachment and biocompatibility of polytetrafluoroethylene (PTFE) treated with glow-discharge plasma of mixed ammonia and oxygen

The plasma generated from a gas mixture of NH3 plus O2 (NH3 + O2) has been used to impart unique chemical and biological characteristics to polytetrafluoroethylene (PTFE). PTFE treated with NH3 + O2 plasma was physiochemically distinct from surfaces treated with plasma of either NH3 or O2 alone, as determined by electron spectroscopy for chemical analysis (ESCA). The contact angle analysis revealed that the PTFE surfaces became less hydrophobic after plasma treatments. ESCA results indicate the presence of oxygen-containing groups and nitrogen-containing groups at the plasma-treated surfaces. PTFE treated with NH3 + O2 plasma resisted the attachment of platelets and leukocytes in a manner similar to untreated PTFE; however, the attachment of bovine aorta endothelial cells was substantially increased. Once attached, these cells grew to confluency. The increased endothelial cell attachment was higher than that observed following plasma treatment with each gas used separately, which could be attributed to the considerable amount of CF(OR)2-CF2 formed on the NH3 + O2 plasma-treated PTFE surface. At 14 days after subcutaneous implantation in rats, the PTFE wafers treated with NH3 + O2 plasma demonstrated less encapsulation and lower levels of inflammatory cells compared to controls. Collectively, the results suggest that NH3 + O2 plasma treatment imparts a unique character to PTFE and could be useful in certain in vivo applications.