Thromboresistance of Balloon-Injured Porcine Carotid Arteries After Local Gene Transfer of Human Tissue Factor Pathway Inhibitor
Background—Tissue factor pathway inhibitor (TFPI) is an endogenous inhibitor of factor Xa and the coagulant initiator complex tissue factor/factor VIIa.
Methods and Results—To study the effects of TFPI gene transfer on thrombus formation, balloon-injured porcine carotid arteries were treated locally with an adenovirus encoding human TFPI (Ad-TFPI) or control virus. Gene transfer of TFPI was confirmed by detection of human TFPI in the conditioned medium of porcine carotid arteries kept in culture after in vivo transduction. When carotid flow was measured with Doppler probe 5 days after surgery, cyclic flow variations (CFVs) developed in 7 of 8 control pigs after constriction of the injured carotid artery by 40%, and all control-treated arteries occluded after 70% constriction. In contrast, CFVs were observed in only 1 of 8 Ad-TFPI–treated pigs after 40% constriction, and only 3 of 8 occluded after constriction by 70% (P=0.0027 and P=0.007, respectively). None of the 5 TFPI-transduced arteries open after 70% constriction developed CFVs during an incremental epinephrine infusion.
Conclusions—Compared with baseline, systemic hemostatic variables and platelet aggregation were unimpaired, suggesting that TFPI gene transfer can prevent arterial thrombosis in the presence of severe shear stress and without detectable hemostatic impairment.
Tissue factor, a transmembrane protein receptor, is the cellular initiator of thrombin generation and blood coagulation in hemostasis and thrombosis. After its exposure following vessel injury or cytokine activation, tissue factor binds to circulating factor VIIa and, in the extrinsic pathway of blood coagulation, activates factor X, which in the prothrombinase complex converts prothrombin to thrombin.1 Tissue factor–driven thrombin generation plays a pivotal role in thrombosis, possibly in restenosis after percutaneous revascularization, and contributes to the thrombogenicity of the atherosclerotic plaque.2 3
Recently, recombinant tissue factor pathway inhibitor (rTFPI) has been studied in animal models as an approach to the prevention of thrombosis and restenosis after arterial injury.4 5 6 At physiological concentrations, TFPI binds to factor Xa, and this complex associates with and inhibits tissue factor/factor VIIa. Higher concentrations of TFPI can inhibit tissue factor/factor VIIa in the absence of factor Xa.7 Although attractive in principle, it is uncertain whether short-term administration of rTFPI will achieve lasting vasoprotection after percutaneous revascularization, particularly at sites of increased tissue factor burden.4 8 Moreover, the systemic doses of recombinant TFPI capable of preventing arterial thrombosis and potentially restenosis are substantial (100 μg · kg−1 · min−1 in Reference 66 ) and may entail significant hemorrhagic risks.
Here, we investigated in a porcine model whether local gene transfer of human TFPI can prevent platelet-driven thrombosis and flow reduction at sites of severe carotid injury and increased shear stress and whether antithrombotic protection would occur without systemic hemostatic impairment.
Preparation of Recombinant Adenovirus
The cDNA encoding human TFPI9 was ligated into the shuttle plasmid pACCMVpLpA10 and cotransfected into 293 cells with the rescue plasmid pJM17 (Microbix Biosystems). The recombinant adenovirus Ad-TFPI was plaque purified, as established by polymerase chain reaction of viral plaques and by an ELISA for human TFPI in the conditioned medium of the 293 cells infected with individual clones of Ad-TFPI or Ad-RR, a viral construct identical to Ad-TFPI but lacking a foreign gene. High-titer adenovirus was purified from 911 cells (IntroGene) with modification of a previously described procedure,11 including a 30-minute digestion with benzonase (100 U/mL, American International Chemical) and the addition of 2 CsCl (density 1.34 mg/mL) equilibrium centrifugation steps at 180 000g for 6 hours at 4°C. Purified virions were suspended in sucrose (2% wt/vol) and MgCl2 (2 mmol/L) in PBS, desalted by sepharose CL4B chromatography (Pharmacia), supplemented with 5% glycerol, and stored at −80°C. The concentration of infectious viral particles was determined as described elsewhere.10 Ad-RR was prepared and titrated in identical fashion. Viral preparations used were endotoxin free (<0.10 EU/mL) in a limulus amoebocyte lysate assay (BioWhittaker).
Sensitivity of Porcine Plasma Clotting to Human rTFPI
The sensitivity of porcine plasma clotting to human rTFPI was studied in a prothrombin time (PT) assay. Because we do not have porcine tissue factor, rabbit (Sigma) and human (Dade Innovin) thromboplastin was used as clotting initiator. rTFPI was diluted in PBS, and 1 μL different dilutions was incubated for 1 minute at 37°C with 100 μL porcine or pooled human plasma, respectively. PT was determined in a fibrometer (BBL FibroSystem) by addition of 200 μL thromboplastin to 100 μL plasma. Simultaneously, the fibrometer was started, and the time to clot formation was recorded. PBS 1 μL without added rTFPI was used as negative control.
Gene Transfer of TFPI in a Porcine Carotid Balloon Injury Model
Repeated attempts to demonstrate human TFPI with immunohistochemistry in Ad-TFPI–infected porcine carotid arteries failed because of unresolved background problems. To evaluate TFPI production by the transduced artery, we used an ex vivo arterial culture technique.12 Initially, we determined whether the TFPI ELISA used (American Diagnostica) detected porcine TFPI. Because TFPI circulates in the blood,7 we analyzed citrated plasma from 5 pigs for the presence of endogenous TFPI. Heparin (200 U/kg), which promotes the endothelial release of TFPI,13 was administered to 3 pigs before blood was drawn. Plasma samples from 5 volunteers were tested as control.
A viral titer of 6×1010 pfu/mL has previously been shown to yield robust transgene expression in balloon-injured porcine carotid arteries.11 Therefore, after balloon angioplasty in the carotid arteries of 2 Yorkshire pigs, Ad-TFPI 6×1010 pfu/mL was delivered by 30-minute dwell to the injured vessel. The arteries were harvested after 24 hours, cut into 3-mm rings, and cultured ex vivo for an additional 4 days in DMEM supplemented with 10% FBS. Five days after angioplasty, the concentration of human TFPI in the conditioned medium from individual arterial rings was determined by ELISA. Because this ELISA does not recognize porcine TFPI (see below), no attempt was made to detect TFPI after carotid artery infection with Ad-RR. Nonetheless, as a negative control, the contralateral carotid arteries of both pigs were cultured, and the conditioned medium was analyzed for the presence of immunoreactive TFPI.
Under institutionally approved protocols, TFPI gene transfer was studied in an established porcine carotid balloon injury model.11 Sixteen male Yorkshire pigs were sedated, intubated, and anesthetized with isoflurane, followed by performance of a femoral cutdown and placement of an 8F introducer. The left carotid artery was isolated through a carotid cutdown, and the left jugular vein was cannulated for fluid support and for drawing blood to determine platelet aggregation, hemostatic variables, and a complete blood count. All pigs received 544 mg IV Baytril (Bayer-Agricultural Division) and 0.8 L lactated Ringer’s solution during surgery. The carotid blood flow velocity (CBFV, in kHz) was measured with a pulse Doppler flow probe (Dr Craig Hartley, Houston, Tex) before injury, immediately after virus delivery, and 5 days after surgery.
Beef heparin 200 U/kg (Upjohn) was given 5 minutes before angioplasty. Under fluoroscopy, a 5F balloon catheter (8×20-mm balloon) was advanced transfemorally over a guidewire into the left carotid artery. Five 30-second inflations were performed at 30-second intervals under weight-adjusted pressure: 25 to 30 kg, 6 atm; 31 to 40 kg, 6.5 atm; and 41 to 45 kg, 7 atm. Banamine 75 mg/kg (Schering-Plow) was given before angioplasty as analgesic. Immediately after angioplasty, a 23-gauge catheter was inserted into a proximal carotid arteriotomy site. Ad-TFPI or control virus (Ad-RR) 6×1010 pfu/mL was instilled for 30 minutes under 0.3 to 0.5 atm into the injured arterial segment temporarily isolated with umbilical ties. After removal of the virus, the arteriotomy site was repaired with 6–0 prolene, and CBFV was monitored for 1 to 3 hours before removal of the catheters and flow probe. After wound repair, the pigs were allowed to recover and return to their cages. Orbifloxacin (Orbax, Schering-Plow) was added daily to their food.
Assessment of Antithrombotic Effects of Local TFPI Gene Transfer
Five days after balloon injury, the pigs were reanesthetized with isoflurane. Femoral vein blood was collected for assessment of platelet aggregation and hemostatic variables. The damaged carotid artery was exposed, and 2-0 silk suture was applied as external loop around the center of the injured carotid artery. After CBFV was recorded for 30 minutes with no tension on the silk loop, the loop was tightened until CBFV decreased by 40% and 70% relative to values before angioplasty. At each stage, CBFV was recorded for 30 minutes, and the number and severity of cyclic flow variations (CFVs) were recorded. CFVs were considered mild if CBFV was reduced to 70% to 95% of the preangioplasty baseline, moderate if between 25% and 70%, and severe if <25%.
If CFVs progressed to zero CBFV, the pigs were killed and pressure perfusion fixed with 10% buffered formaldehyde,11 and the carotid arteries were harvested for histological analysis. If the pigs did not develop CFVs after 70% constriction, epinephrine was infused to increase local shear stress and platelet activation.14 Epinephrine was administered at doubling rates from 1.5 to 38 μg/min (10 minutes per dose). In the absence of irreversible zero flow, the infusion was continued for a total of 30 minutes at 38 μg/min before the pig was killed. The constricting suture was left in situ to demonstrate the degree of arterial constriction and folding. Arteries were histologically evaluated by an experimental cardiovascular pathologist (Dr Angela Hughes, Texas Heart Institute, Houston, Tex).
Whole-Blood Aggregation Studies
Whole-blood aggregation was measured as electrical impedance (in ohms) in a Chronolog aggregometer (Chronolog) at the beginning of surgery (baseline) and the time of the thrombotic challenge (day 5). Blood was drawn into 3.8% sodium citrate (1:9 vol/vol) for aggregation to α/β thrombin (Fibrindex, OrthoDiagnostic Systems) or sodium heparin (2 U/mL) for aggregation to adenosine diphosphate (Sigma), collagen (Hormon Chemie), and the thromboxane A2 analogue U46619 (Cayman Chemical). Activated coagulation times (ACT), PT, and activated partial thromboplastin time (aPTT) were measured in a Hemochron-80 Dual Coagulation System (International Technidyne). Ear skin bleeding times (BTs) were measured before balloon angioplasty (day 0) and on day 5 by use of a Simplex II device (Organon Teknika).
Comparisons of the means of CBVF and the number of CFVs between Ad-TFPI– and control-treated pigs were performed by use of a 2-sided unpaired t test. P<0.05 was considered statistically significant. Values are given as mean±SD.
Twenty-four hours after carotid balloon angioplasty and local dwell delivery of Ad-TFPI at 6×1010 pfu/mL to 2 pigs, the injured and contralateral carotid arteries were harvested, cut in about 3-mm-long rings, and kept in individual wells in culture for an additional 4 days. Five days after delivery of Ad-TFPI, human TFPI was detected by ELISA in the conditioned medium of all but 1 arterial ring in each of the 2 arteries (Figure 1⇓). No TFPI was detected in the conditioned medium of uninjured control arteries.
The ELISA was tested for its ability to detect endogenous TFPI in porcine blood. Whereas TFPI was detected in all 5 human plasma samples (49.4±21.7 ng/mL), none was detected in the porcine plasma samples drawn with (n=3) or without (n=2) prior heparin administration. The sensitivity of porcine plasma to rTFPI in a modified prothrombin assay (see Methods) was assessed by use of rabbit or human thromboplastin as clotting initiator. Clotting of porcine plasma with human rTFPI added ex vivo was prolonged more than human plasma, suggesting that porcine factors VII/Va and Xa were at least as sensitive to human rTFPI as the respective human factors (Figure 2⇓).
No differences were observed in ACT, PT, aPTT, and BT in the pigs, which were alternatively assigned to treatment with Ad-TFPI (n=8; weight, 34.9±3.5 kg) and Ad-RR (n=8; weight, 37.9±4.4 kg) (Table 1⇓). Compared with pigs assigned to Ad-RR, platelet aggregation to thrombin at baseline was significantly greater in pigs assigned to treatment with Ad-TFPI. (Table 2⇓). Aggregation to all other agonists was not significantly different. After completion of angioplasty, CBFV decreased to a similar extent in control vector– and Ad-TFPI–treated arteries, from 16.4±3.3 to 10.5±5 and 17.1±2.2 to 10.7±3.1 kHz, respectively.
Five days after surgery, Doppler flow probes were applied to the injured carotid artery segments (see Methods), and CBFV was recorded continuously until the pigs were killed. CFVs were observed before external constriction of the artery in 4 of the 8 control vector–treated pigs and in 1 of the 8 pigs treated with Ad-TFPI. Constriction of the arteries by 40% precipitated severe CFVs in 7 of 8 Ad-RR–treated pigs but only 1 Ad-TFPI–treated pig (P=0.0027). After further constriction by 70%, all control vector–treated pigs developed severe CFVs, followed by irreversible cessation of blood flow, whereas only 3 of the Ad-TFPI–treated pigs developed zero flow (P=0.007). The presence of obstructive thrombi when zero flow developed was histologically confirmed, as shown in Table 3⇓, which also summarizes the incidence and severity of CFVs.
CBFVs in the Ad-TFPI– and control virus–treated arteries were not significantly different during the initial 30-minute observation period (Figure 3⇓) but decreased to zero in all control vector–treated animals after constriction of the injured arteries. In contrast, blood flow was preserved in all TFPI-transduced vessels after constriction by 40% and in 5 of 8 Ad-TFPI–treated pigs after carotid artery constriction by 70%.
Figure 4⇓ shows Ad-RR– and Ad-TFPI–treated porcine carotid arteries. Because the sections analyzed include those obtained at the level of the constrictor and sections distal and proximal to it, arterial sections with both conserved and distorted architecture were observed. A section through the nonconstricted segment of the Ad-TFPI–treated artery illustrates the severity of balloon injury, resulting in disruption and effacement of the internal elastic membrane.
Representative CBFV tracings of Ad-TFPI– and Ad-RR–treated pigs are shown in Figure 5⇓. Of the 5 TFPI-transduced carotid arteries without CFVs at 70% constriction, none developed CFVs during an incremental epinephrine infusion to 38 μg/min, equivalent to 1.1±0.1 μg · kg−1 · min−1 epinephrine.
Coagulation variables and BT were not significantly different between the treatment groups and between baseline (day 0) and day 5 (Table 1⇑). Except for thrombin, whole-blood aggregation to agonists was unchanged at all concentrations (Table 2⇑). Platelet aggregation to thrombin, which was higher before surgery in pigs assigned to Ad-TFPI, was not different between the 2 groups on day 5.
We report here that local transduction of porcine balloon-injured carotid arteries with a single administration of the TFPI gene exerts protection against thrombosis even when epinephrine is infused during the thrombotic challenge. These observations confirm the important role of coagulant protease complexes upstream of thrombin in the pathogenesis of injury- and shear-dependent thrombosis and illustrate the antithrombotic efficacy of gene transfer techniques targeting tissue factor and factor Xa in the damaged vessel wall.
CFVs in stenosed arteries reflect the dynamic process of growth and embolization of platelet-rich thrombi, with the rate of flow reduction being proportional to the rate of platelet accumulation in the narrowed lumen,15 16 and have been observed in patients during the course of acute coronary syndromes.17 18 The nadir of flow during periods of cyclic flow reduction, often to zero flow, coincides with the presence of obstructive platelet-rich thrombi,15 16 as is also apparent in our study, in which obstructive thrombi were seen in all cases of zero flow. The platelet dependency of CFVs is stressed by their exquisite sensitivity to antiplatelet agents.18 19 In contrast, the spontaneous reduction in CBFV at the time of balloon injury is believed to represent vasoconstriction, to which platelet-derived mediators substantially contribute.20
Locally expressed TFPI was particularly effective against platelet-mediated thrombosis in the presence of increased shear stress and the additional challenge of an epinephrine infusion. These finding may have relevance given the strong induction of tissue factor by shear stress.21 Added to severe wall damage, constriction functions to increase shear stress and initiate CFVs. Epinephrine exerts additional shear-dependent and -independent actions on platelet aggregation.14
The antithrombotic efficacy of local TFPI gene transfer in the presence of heightened shear stress is not unexpected. Indeed, among all platelet agonists known, thrombin is the most potent, and systemic administration of the direct antithrombin hirudin has been effective in preventing platelet-rich thrombus formation in pigs, in other animal species, and in the course of acute coronary syndromes in patients.22 Because of the dual function of TFPI as a direct inhibitor of factor Xa and as an inhibitor of the catalytic tissue factor/VIIa complex, we are unable to distinguish between the relative role of factor Xa and tissue factor/factor VIIa as a target of TFPI gene therapy. Nonetheless, studies of specific factor Xa inhibitors23 and tissue factor–blocking antibodies24 have demonstrated that both factor Xa and tissue factor play essential roles in arterial thrombogenesis.
The timing for the testing the efficacy of TFPI gene therapy (ie, 5 days after injury) warrants discussion. Our previous observations in conscious pigs, in which carotid flow was continuously recorded for 10 days, indicate that in this model CFVs begin to fade 3 to 6 hours after their initiation by severe vascular injury.11 In unprotected vessels, CFVs recur after a period of quiescence of 2 to 3 days and rapidly increase in number and severity to peak 4 to 8 days after the initial insult. In agreement with these earlier observations, the incidence of total occlusion (zero flow) in our study was 100% in the control (Ad-RR) -treated pigs on day 5. Approximately 10 days after balloon injury, CFVs spontaneously fade, concurrent with endothelial healing at the site of injury.11
The antithrombotic efficacy of local TFPI gene therapy compares favorably with the dose requirement of systemically administered recombinant TFPI, which falls in the milligram-per-hour range.4 5 6 The relative inefficacy of circulating TFPI vis-à-vis severe arterial thrombosis is also reflected by the observation that unrelieved coronary thrombosis often occurs despite elevated plasma levels of endogenous TFPI.25 On the other hand, the presence of TFPI in the low picogram-per-milligram range in human carotid atherectomy specimens was shown to attenuate local tissue factor activity measured ex vivo in these specimens,26 consistent with the efficacy of TFPI expressed in situ.
Systemic hemostatic impairment after antithrombotic drug interventions, especially those targeting thrombin, has limited the dosage of antithrombotic interventions.27 Not surprisingly, Oltrona et al6 found that systemic administration of rTFPI led to markedly prolonged PT in pigs, suggesting that rTFPI at doses preventing arterial thrombosis is associated with substantial systemic bleeding risk. Lack of hemorrhagic complications and unimpaired platelet aggregation, BT, and plasma coagulation at the time of the thrombotic challenge in this study contrast those observations and extend our previous studies with COX-1 gene transfer, demonstrating that in a large animal species local antithrombotic gene therapy can be highly effective in reducing thrombosis at sites of deep arterial injury in the absence of detectable hemostatic perturbation.11
Although inflammation and thrombosis are in general associated phenomena, the influences of the adenoviral vectors on local thrombogenesis are difficult to assess. Directly related to this issue are reports suggesting that exposure to adenoviral vectors of uninjured vessels in rabbits promotes thrombus formation,28 arterial neointimal formation,29 and inflammation of experimental vein grafts.30 This issue is compounded by the uncertain purity of viral batches from different laboratories preparing non–clinical-grade vectors (eg, the concentration of mycoplasma and other contaminants) and the tissue and species dependency of viral-host interactions.
As in our previous study,11 no differences were observed in the degree of inflammation between Ad-TFPI–treated and Ad-RR–treated (control) porcine carotid arteries. Given the lack of a buffer-only control group in the present study, however, we are unable to assess the contribution of adenoviral vectors to the inflammatory response triggered by severe balloon injury, thrombosis, and external manipulation of the vessel. Nonetheless, in our previous study, the incidence of histological thrombosis, CFVs, and degree of inflammation were not different between control buffer–treated (mock) and control (null) vector–treated arteries.11 Thus, published data, including our own, suggest that adenoviral vectors may either aggravate thrombosis and neointimal formation or, in the porcine carotid injury model, have no discernable prothrombotic effect.
In conclusion, our study of local TFPI gene transfer to balloon-injured carotid arteries offers new support for the important role of tissue factor/factor VIIa and its downstream proteases in the pathogenesis of platelet-mediated thrombosis. We demonstrate that a single local administration of a vector encoding human TFPI inhibits platelet-dependent thrombosis at sites of severe vascular injury and increased shear stress and that vasoprotection occurs without detectable hemorrhagic risk. Potentially, this approach may serve as a strategy to reduce the thrombotic risk after percutaneous revascularization interventions without engendering hemostatic impairment and bleeding risk. Newer vectors and the use of species-specific TFPI may extend the duration of antithrombotic protection.
Supported in part by NIH grants HL-50179 and HL-54839; the Keith Acton Meadows Foundation, Dallas, Tex; and a Texas state grant. We thank Dr Tze-Chein Wun, Searle/Monsanto Corporation, Chesterfield, Mo, for the gift of rTFPI and the cDNA encoding human full-length TFPI; Dr Robert D. Gerard (University of Michigan, Ann Arbor) for the shuttle plasmid pACCMVpLpA and control vector Ad-RR; and IntroGene, Inc (Leyden, the Netherlands) for the gift of 911 cells.
Guest Editor for this article was Désiré Collen, MD, PhD, Flanders University, Leuven, Belgium.
- Received May 18, 1999.
- Revision received August 25, 1999.
- Accepted September 7, 1999.
- Copyright © 2000 by American Heart Association
Wilcox JN, Smith KM, Schwartz SM, Gordon D. Localization of tissue factor in the normal blood vessel wall and in the atherosclerotic plaque. Proc Natl Acad Sci U S A. 1989;86:2839–2843.
Toschi V, Gallo R, Lettino M, Fallon JT, Gertz SD, Fernández-Ortiz A, Chesebro JH, Badimon L, Nemerson Y, Fuster V, Badimon JJ. Tissue factor modulates the thrombogenicity of human atherosclerotic plaques. Circulation. 1997;95:594–599.
Haskel EJ, Torr SR, Day KC, Haskel EJ, Torr SR, Day KC, Palmier MO, Wun TC, Sobel BE. Prevention of arterial reocclusion after thrombolysis with recombinant lipoprotein-associated coagulation inhibitor. Circulation. 1991;84:821–827.
Jang Y, Guzman LA, Lincoff M, Topol EJ. Influence of blockade at specific levels of the coagulation cascade on restenosis in a rabbit atherosclerotic femoral artery injury model. Circulation. 1995;92:3041–3050.
Oltrona L, Speidel CM, Recchia D, Wickline SA, Eisenberg PR, Abendschein DR. Inhibition of tissue factor-mediated coagulation markedly attenuates stenosis after balloon-induced arterial injury in minipigs. Circulation. 1997;96:646–652.
Marmur J, Thiruvikraman S, Fyfe B, Guha A, Sharma S, Ambrose J, Falloon J, Nemerson Y, Taubman MB. Identification of active tissue factor in human coronary atheroma. Circulation. 1996;94:1226–1232.
Wun TC, Kretzmer KK, Girard TJ, Miletich JP, Broze GJ. Cloning and characterization of a cDNA coding for the lipoprotein-associated coagulation inhibitor shows that its consists of three tandem Kunitz-type inhibitory domains. J Biol Chem. 1988;263:6001–6004.
Gomez-Foix AM, Coats WS, Baque S, Alam T, Gerard RD, Newgard CB. Adenovirus-mediated transfer of the muscle glycogen phosphorylase gene into hepatocytes confers altered regulation of glycogen metabolism. J Biol Chem. 1992;267:25129–25134.
Zoldhelyi P, McNatt J, Xu XM, Meidell R, Loose-Mitchell D, Willerson JT, Wu KK. Prevention of arterial thrombosis by adenovirus-mediated transfer of cyclooxygenase gene. Circulation. 1996;93:10–17.
Beaty CH, Howard CF, Caruso V. Potentiation with epinephrine of macaque platelet aggregation by other agonists: implications for studies on human atherosclerosis. Thromb Res. 1986;41:441–458.
Folts JD, Gallagher K, Rowe GG. Blood flow velocity reductions in stenosed canine coronary arteries: vasospasm or platelet aggregation. Circulation. 1982;65:248–255.
Golino P, Buja LM, Ashton JH, Kulkarni P, Taylor A, Willerson JT. Effect of thromboxane and serotonin receptor antagonists on intracoronary platelet deposition in dogs with experimentally stenosed coronary arteries. Circulation. 1988;78:701–711.
Anderson HV, Kirkeeide RL, Krishnaswami A, Weigelt LA, Revana M, Weisman HF, Willerson JT. Cyclic flow variations after coronary angioplasty in humans: clinical and angiographic characteristics and elimination with 7E3 monoclonal antiplatelet antibody. J Am Coll Cardiol. 1994;23:1031–1037.
Yao S-K, Rosolowsky M, Anderson HV, Golino P, McNatt JM, De Clerck F, Buja LM, Willerson JT. Combined thromboxane A2 synthase inhibition and receptor blockade are effective in preventing spontaneous and epinephrine-induced canine coronary cyclic flow variations. J Am Coll Cardiol. 1990;16:705–713.
Lam JY, Chesebro JH, Steele PM, Badimon L, Fuster V. Is vasospasm related to platelet deposition? Relationship in a porcine preparation of arterial injury in vivo. Circulation. 1987;75:243–248.
Verstraete M, Zoldhelyi P, Willerson JT. Specific thrombin inhibitors. In: Verstraete M, Fuster V, Topol EJ, eds. Cardiovascular Thrombosis: Thrombocardiology. Baltimore, Md: Lippincott-Raven Publishers; 1998.
Orvim U, Barstad RM, Vlasuk GP, Sakariassen KS. Effect of selective factor Xa inhibition on arterial thrombus formation triggered by tissue factor/factor VIIa or collagen in an ex vivo model of shear-dependent human thrombogenesis. Arterioscler Thromb Vasc Biol. 1995;15:2188–2194.
Ragni M, Cirillo P, Pascucci I, Scognamiglio A, D’Andrea D, Eramo N, Ezekowitz MD, Pawashe AB, Chiariello M, Golino P. Monoclonal antibody against tissue factor shortens tissue plasminogen activator lysis time and prevents reocclusion in a rabbit model of carotid artery thrombosis. Circulation. 1996;93:1913–1918.
Caplice NM, Mueske CS, Kleppe LS, Simari RS. Presence of tissue factor pathway inhibitor in human atherosclerotic plaques is associated with reduced tissue factor pathway activity. Circulation. 1998;98:1051–1057.
The Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO) IIa Investigators. Randomized trial of intravenous heparin versus recombinant hirudin for acute coronary syndromes. Circulation. 1994;90:1631–1637.
Newman KD, Dunn PF, Owens JW, Schulick AH, Virmani R, Sukhova G, Libby P, Dichek DA. Adenovirus-mediated gene transfer into normal rabbit arteries results in prolonged vascular cell activation, inflammation, and neointimal hyperplasia. J Clin Invest. 1995;96:2955–2965.
Channon KM, Fulton GJ, Gray JL, Annex BH, Shetty GA, Blazing MA, Peters KG, Hagen PO, George SE. Efficient adenoviral gene transfer to early venous bypass grafts: comparison with native vessels. Cardiovasc Res. 1997;35:505–513.