Preincubation of Dacron grafts with recombinant tissue factor pathway inhibitor decreases their thrombogenicity in vivo

Effect of upstream priming on transient downstream platelet-substrate interactions

Upstream exposure of platelets to activating proteins 'primes' platelets for increased downstream adhesion, though the mechanics of platelet translocation before permanently arresting are not well understood. To investigate platelet translocation on platelet-binding proteins, primed platelets' transient contacts with immobilized proteins were recorded and analyzed. Using a microfluidic channel, representative of a vascular graft, platelet-activating proteins were covalently attached to the upstream priming, center, and downstream capture positions. Image sequences of platelet interactions with the center protein were captured as platelet-rich plasma (PRP) was perfused through the channel. There was an increase in both platelet pause events and net platelet adhesion on von Willebrand factor, collagen, or fibrinogen following upstream exposure to the same protein. Upstream priming also caused a decrease in average platelet velocity. The duration of transient platelet arrests on the protein-coated surface and the distance that platelets travel between pause events depended on the protein with which they were interacting. The most significant increase in platelet pause events frequency and decrease in average velocity occurred on immobilized von Willebrand factor, compared to the control with no upstream priming. These results demonstrate that platelet priming increases downstream platelet-protein interactions prior to permanent adhesion.

Two-wavelength near-infrared fluorescence for the quantitation of drug antiplatelet effects in large animal model systems

OBJECTIVE

Intraoperative imaging of intravascular thrombi is limited by the inability of visible light to penetrate thick-walled vessels. Near-infrared (NIR) light has relatively high tissue penetration and low autofluorescence and scatter, offering significant advantages. We hypothesized that the development of 700-nm NIR fluorophores for platelet labeling, in conjunction with existing 800-nm NIR fluorophores, would permit simultaneous and separable quantitation of intravascular thrombi and measurement of the antiplatelet effect of drugs.

METHODS

We synthesized a series of lipophilic, cationic, polymethine indocyanine dyes (MHI-86, 94, 106, and 114) that emit at approximately 700 nm. Platelet uptake was optimized in vitro and the bioactivity and blood half-life of labeled platelets was characterized in vitro and in vivo. FeCl(3)-induced injury of the femoral arteries and intravascular thrombus formation was performed in 35-kg Yorkshire pigs. A combination of 700-nm and 800-nm NIR fluorophore-labeled platelets was used in conjunction with the fluorescence-assisted resection and exploration imaging system to image and quantify the antiplatelet effect of cilostazol and acetylsalicylic acid.

RESULTS

MHI-114 was incorporated at nearly 4.1 × 10(6) molecules per platelet without affecting platelet function. When infused into pigs, the signal-to-background ratio of MHI-114-labeled platelets exhibited a blood half-life of 16.4 ± 2.2 (mean ± SEM; n = 3) minute and generated a signal-to-background ratio of 2.5 ± 0.5 (mean ± SEM; n = 3) at the site of thrombi. Using dual-NIR-labeled platelet populations, cilostazol and acetylsalicylic acid were found to cause a reduction in platelet incorporation into thrombi of 51 ± 2% and 10 ± 1% (mean ± SEM; n = 3), respectively, relative to vehicle-only treated control thrombi.

CONCLUSIONS

New platelet-avid 700-nm NIR fluorophores permit simultaneous two-wavelength NIR fluorescence imaging and quantitation of intravascular thrombi in intact vessels approaching the size of humans and can be used to study the antiplatelet effect of drugs.

The use of antithrombotic therapies in reducing synthetic small-diameter vascular graft thrombosis

BACKGROUND

Thrombosis of synthetic small-diameter bypass grafts remains a major problem. The aim of this article is to review the antithrombotic strategies that have been used in an attempt to reduce graft thrombogenicity.

METHODS

A PubMed/MEDLINE search was performed using the search terms "vascular graft thrombosis," "small-diameter graft thrombosis," "synthetic graft thrombosis" combined with "antithrombotic," "antiplatelet," "anticoagulant," "Dacron," "PTFE," and "polyurethane."

RESULTS

The majority of studies on antithrombotic therapies have used either in vitro models or in vivo animal experiments. Many of the therapies used in these settings do show antithrombotic efficacy against synthetic graft materials. There is however, a distinct lack of human in vivo studies to further delineate the performance and limitations of therapies displaying good antithrombotic characteristics.

CONCLUSION

Very few antithrombotic therapies have translated into clinical use. More human in vivo studies are required to assess the efficacy and safety of such therapies.

A computational analysis of an in vitro vessel wall injury model

Implantation of vascular grafts or stents causes significant injury to the vessel wall. Activated coagulation factors, such as FXa are generated at the injury site. The size of the injury and the flow conditions influence the transport of these activated factors. A simulation model has been developed to evaluate surface bound coagulation inhibitors on medical devices. Tissue factor pathway inhibitor (TFPI) isa potent inhibitor of FXa in vivo. This physiologically relevant in vitro model studies the mechanism by which immobilized rTFPI effectively inhibits TF initiated thrombosis.Computational fluid dynamics was used to develop the model and validated by experiments performed in a parallel plate flow chamber. The lower plate was divided into two zones. The first zone represents fibroblasts that catalyze FXto FXa by TF-FVIIa. The second represents passively absorbed surface bound rTFPI. The efficacy of rTFPI in inhibiting FXa from the injury site was studied under both venous and moderate arterial shear rates. This model extends a previous model to include a more physiologic model of vessel wall injury in which FXa generation by TF:VIIa occurs at the wall upstream of the TFPI coated surface. The previous model estimated a uniform inlet FXa concentration of 20 nM (20% conversion of the approximate physiologic concentration of 100 nM) to the parallel plate chamber.

Evaluating surface bound rTFPI through an in vitro model of vessel wall injury

INTRODUCTION

Injury to the surrounding vessel wall is one of the major reasons for failure of implantable medical devices. The surgical procedure itself or the altered flow conditions after implantation can cause damage to the vessel wall. This damage exposes tissue factor (TF), the initiator of the extrinsic pathway of coagulation. One approach to combat thrombosis is to use an anticoagulant on the surface of the device. The primary aim of this study is to develop a simplified physiologically relevant in vitro model of vessel wall injury to study the mechanisms by which immobilized recombinant tissue factor pathway inhibitor (rTFPI) effectively inhibits TF initiated thrombosis.

MATERIALS AND METHODS

A two well chamber slide was used for the study. Fibroblasts were cultured on the upstream portion of the slide. Fibroblast cells stimulated with TNF-α acted as a source of surface TF. The downstream portion of the slide was coated with rTFPI. A mixture of FX, FVIIa and calcium was perfused over the slides to generate FXa. Effluent collected at the outlet was used to analyze the inhibition of this surface generated FXa by the rTFPI present downstream.

RESULTS AND CONCLUSIONS

Different shear rates and rTFPI densities were used to study this effect. In most cases rTFPI inhibited FXa generated upstream as a function of the wall shear rate and rTFPI dosage (surface density). This study shows the effectiveness of the surface bound inhibitor when FXa is generated from an upstream injury site and the bulk of FXa is near the wall.

Addressing thrombogenicity in vascular graft construction

Thrombosis is a major cause of poor patency in synthetic vascular grafts for small diameter vessel (< 6 mm) bypass. Arteries have a host of structural mechanisms by which they prevent triggering of platelet activation and the clotting cascade. Many of these are present in vascular endothelial cells. These mechanisms act together with perpetual feedback at different levels, providing a constantly fine-tuned non-thrombogenic environment. The arterial wall anatomy also serves to promote thrombosis as a healing mechanism when it has been severely injured. Surface modification of synthetic graft surfaces to attenuate the coagulation cascade has reduced thrombosis levels and improved patency in vitro and in animal models. Success in this endeavor is critically dependent on the methods used to modify the surface. Platelets adhere to positively charged surfaces due to their own negative charge. They also preferentially attach to hydrophobic surfaces. Therefore synthetic graft development is concerned with hydrophilic materials with negative surface charge. However, fibrinogen has both hydrophilic and hydrophobic binding sites-amphiphilic materials reduce its adhesion and subsequent platelet activation. The self-endothelializing synthetic graft is an attractive proposition as a confluent endothelial layer incorporates many of the anti-thrombogenic properties of arteries. Surface modification to promote this has shown good results in animal models. The difficulties experienced in achieving spontaneous endothelialisation in humans have lead to the investigation of pre-implantation in vitro endothelial cell seeding. These approaches ultimately aim to result in novel synthetic grafts which are anti-thrombogenic and hence suitable for coronary and distal infrainguinal bypass.

Computational modeling of factor Xa inhibition by immobilized tissue factor pathway inhibitor

Coating surfaces of implanted devices with anticoagulants can reduce thrombosis and studies using a recombinant form of endogenous tissue factor pathway inhibitor (rTFPI) are promising. The anticoagulant function of immobilized rTFPI is thought to occur primarily by its inhibition of plasma clotting factor Xa (FXa); however the kinetics of this reaction at a surface are as yet unknown. To better understand the surface inhibition reaction under flow conditions, a theoretical model was developed delineating the roles of mass transport and reaction kinetics for an in vitro parallel plate device used in prior experimental studies [Hall et al., J. Biomech. Eng. 120:484-490, 1998]. As a first approximation, the kinetics of inhibition of FXa by rTFPI reported for static, homogeneous systems was considered. The unsteady convection-diffusion equation was solved for different wall-shear rates and inlet concentrations of FXa using the computational fluid dynamics software CFD-ACE (ESI Software Group). The results show that the heterogeneous inhibition reaction is diffusion controlled prior to saturation of the rTFPI. The experimental results compare favorably with the model at the lower shear rates (100-400 s(-1)). At higher shear rates (>400 s(-1)) the theoretical results follow the same trend as the experimental results but show a greater inhibition of FXa, implying an effect of flow or shear on the inhibition reaction.

Factor Xa inhibition by immobilized recombinant tissue factor pathway inhibitor

The recombinant form of the endogenous anticoagulant, tissue factor pathway inhibitor (rTFPI), is a potent inhibitor of Factor Xa (FXa) and the tissue factor-factor VIIa (TF:VIIa) complex. Surface-immobilized rTFPI reduces the thrombogenicity and intimal hyperplasia associated with synthetic vascular grafts in animal models and specifically reduces fibrin deposition on collagen-impregnated Dacron grafts from native blood in an in vitro flow model. The FXa inhibitory capacity of rTFPI in the bulk phase has been demonstrated in static systems and immobilized rTFPI reduces fibrin deposition in whole blood in vitro and animal studies; however, the specific mode of this anticoagulation has not been studied. Therefore, a comparison was made between the FXa binding capacity of two forms of immobilized rTFPI, i.e., passively adsorbed and covalently bound. The rTFPI-coated surfaces were evaluated using a parallel-plate flow reactor and comparing the amount of FXa exiting the flow chamber after exposure to an rTFPI-coated versus an uncoated plate. The results demonstrate that adsorbed rTFPI exhibits increased binding capacity (1.5-3.6 times) the expected stoichiometry via interactions with the C-terminus, whereas covalently-bound rTFPI interacts with FXa in a 1:1 stoichiometry. Thus, the results imply that specific FXa inhibition is a key component of the anticoagulant effect of rTFPI-coated surfaces and that passive adsorption of rTFPI to glass surfaces produces a more effective coating than covalent binding of rTFPI.

In vitro flow evaluation of recombinant tissue factor pathway inhibitor immobilized on collagen impregnated Dacron

Recombinant tissue factor pathway inhibitor (rTFPI) was immobilized on collagen impregnated (CI) knitted Dacron surfaces and its resistance to fibrin deposition evaluated following exposure to nonanticoagulated whole blood. Recombinant TFPI readily adsorbed to the CI Dacron surface and maintained its inhibitory activity. Under static conditions, rTFPI treated CI Dacron showed little fibrin deposition when compared with untreated surfaces. Treated samples exposed to flowing native blood at wall shear rates of 100 or 200 sec(-1) also demonstrated reduced fibrin deposition (up to 56%) compared with untreated samples. To assess the relative roles of the contact and tissue factor pathways in fibrin formation on artificial grafts, flow studies were performed with whole blood containing corn trypsin inhibitor, a potent inhibitor of FXIIa and contact activation. Corn trypsin inhibitor reduced fibrin deposition on untreated CI Dacron by 40%. Immobilized rTFPI alone, or corn trypsin inhibitor in combination with immobilized rTFPI, reduced fibrin deposition by 58% and 61%, respectively. These data suggest that immobilized rTFPI slows fibrin deposition on the vascular graft material by inhibiting both the contact pathway and blood borne tissue factor procoagulant activity arising from either the alternatively spliced form of tissue factor or from tissue factor containing microparticles.

Coagulation activators and inhibitors in the neointima of polyester vascular grafts

The surface of synthetic vascular grafts is thrombogenic, which implies a risk for their occlusion. The aim of the study was to evaluate expression of coagulation components in the polyester vascular grafts neointima. The study was carried out on 18 dogs, which underwent replacement of the abdominal aorta with a polyester prosthesis. Grafts were removed after 1, 4 and 12 months. Immunohistochemical labeling for von Willebrand factor, tissue factor, factor XII, tissue factor pathway inhibitor, thrombomodulin, protein C, protein S and prothrombin activation fragment F1 + 2 was performed. Increasing intensity of von Willebrand factor expression was found in successive periods of the study. Factor XII was shown in the whole neointima after 1 month, whereas in the following periods its presence was limited to the luminal surface. Tissue factor expression was demonstrated after 1 month and its intensity increased in later periods. Tissue factor pathway inhibitor and thrombomodulin expression was demonstrated after 4 and 12 months. Protein C and protein S were present in all observation periods, as well as prothrombin activation fragment F1 + 2. Results indicate a high thrombotic potential of the graft neointima early after prosthesis implantation, whereas in the late postoperative follow-up increasing expression of coagulation inhibitors reduces thrombotic properties of the graft neointima.

Human mesothelial cells express tissue factor when switched to proliferating state. Pharmacological modulation in vitro

Covering the inner surface of small-diameter arterial prostheses with endothelial cells (ECs) has been proposed as a means of improving biocompatibility and thrombosis resistance. Because the availability of autologous ECs is limited, autologous human mesothelial cells (HMCs) have been suggested as a substitute for ECs. However, HMCs express tissue factor (TF) in vitro, a deleterious characteristic in vivo. We investigated the distribution of TF antigen and of its inhibitor, tissue factor pathway inhibitor, on HMCs and the effect of pharmacological agents on TF expression. TF antigen was measured by enzyme-linked immunosorbent assay and localized by confocal microscopy. Three distinct pools of TF antigen were demonstrated: within the cells, at the cell surface, and in the extracellular matrix. The effects of ilomedin (10 microg/ml) and heparin (500 U/ml), known to affect procoagulant activity, were evaluated by incubating HMCs for 24 h with or without these agents. Ilomedin, but not heparin, decreased TF antigen expression by 30% (P < 0.05). Despite the theoretical potential of HMCs as a vascular prosthesis lining, TF expression by HMCs remains a major drawback. A technique capable of blocking TF expression until the HMCs return to their resting state is needed. Genetic manipulation of HMCs may hold promise for such a technique.

Local treatment with recombinant tissue factor pathway inhibitor reduces the development of intimal hyperplasia in experimental vein grafts

BACKGROUND

Tissue factor (TF)-initiated thrombin generation has been implicated in the development of intimal hyperplasia after arterial injury. An increase in intimal TF expression has been shown to precede the development of intimal hyperplasia in vein grafts. This study examines the effects of local treatment with recombinant human tissue factor pathway inhibitor (rTFPI) in experimental vein grafts.

METHODS

Thirty-six male New Zealand white rabbits underwent bypass grafting of the carotid artery by use of the reversed ipsilateral jugular vein and were divided into four groups. Twenty animals had ex vivo incubation with rTFPI treatment (50 microg x mL(-1); n = 10) or placebo vehicle (control; n = 10). Sixteen animals received both ex vivo incubation and in vivo gel treatment with rTFPI (50 microg. mL(-1); n = 8) or without rTFPI (gel-control; n = 8). After operation, vein grafts were harvested at 3 days for immunohistochemical and Western analyses and at 28 days for histomorphologic study.

RESULTS

Western analysis demonstrated a 6.2-fold reduction in the expression of TF protein with rTFPI treatment in comparison to without rTFPI treatment. CD-18 leukocyte staining was diminished, whereas Tie-2 endothelial staining was increased in all rTFPI-treated vein grafts, compared with control and gel-control vein grafts. Intimal thickness was reduced by 21% with ex vivo rTFPI treatment compared with placebo (69 +/- 4 versus 87 +/- 5 microm; P <.05) and by 30% with the addition of rTFPI in vivo compared with gel-control (60 +/- 4 versus 86 +/- 5 microm; P <.01).

CONCLUSION

Local administration of rTFPI exerts early beneficial effects and limits the development of intimal hyperplasia in vein grafts. Therefore blocking TF-mediated pathway may offer new therapeutic options to reduce vein graft failure.

Thrombogenicity of small-diameter prosthetic grafts: relative contributions of graft-associated thrombin and factor Xa

PURPOSE

We evaluated the contributions of coagulation factors IIa (thrombin) and Xa to small-diameter prosthetic graft thrombogenicity in vivo.

METHODS

Preclotted and nonpreclotted (collagen-coated) polyester grafts were studied before and 24 hours after implantation into pig femoral arteries. After incubation of explanted grafts was performed with plasma depleted of vitamin K-dependent coagulation factors by barium chloride adsorption (Ba-plasma), graft-associated thrombin activity was determined by radioimmunoassay for fibrinopeptide A. Fibrinopeptide A levels reflect thrombin-mediated fibrin formation. Factor Xa activity was characterized by measuring activation of prothrombin added to Ba-plasma.

RESULTS

Thrombin and factor Xa were associated with the luminal surfaces of preclotted grafts before and 24 hours after implantation. Nonpreclotted grafts had negligible procoagulant activity before implantation. After 24 hours in vivo graft-associated factor Xa activity was similar in both nonpreclotted and preclotted grafts; however, more thrombin was bound to nonpreclotted coated grafts (p < 0.01).

CONCLUSIONS

The procoagulant activity of small-diameter prosthetic grafts persists for 24 hours after implantation and is attributable not only to graft-associated thrombin but also to de novo thrombin elaboration induced by factor Xa. Moreover, graft-associated procoagulant activity is not dependent on preclotting because it develops on nonpreclotted, collagen-coated grafts as well. Treatment strategies to attenuate graft thrombosis may require the inhibition of both thrombin and factor Xa.