The ionic contrast medium ioxaglate interferes with thrombin-mediated feedback activation of factor V, factor VIII and platelets

Transcription

1 Journal of Thrombosis and Haemostasis, 1: ORIGINAL ARTICLE The ionic contrast medium ioxaglate interferes with thrombin-mediated feedback activation of factor V, factor VIII and platelets R. AL DIERI, S. BÉGUIN and H. C. HEMKER Cardiovascular Research Institute (CARIM), Synapse BV, Maastricht University, Maastricht, the Netherlands Summary. Clinical observation shows that radiographic contrast media (CM) may influence thrombus formation. In the search for the underlying mechanism, we have shown that the ionic CM ioxaglate is a potent inhibitor of thrombin generation in platelet-poor and platelet-rich plasma, whereas the influence of the non-ionic contrast medium iodixanol is minimal. Ioxaglate boosts the inhibitory effect of the platelet GPIIb/IIIa antagonist abciximab and the effects of ioxaglate and heparin are additive. Ioxaglate inhibits the clotting of fibrinogen and the activation of factors V and VIII, and of platelets by thrombin. It does not inhibit hydrolysis of small chromogenic thrombin substrates, nor does it influence the heparin-catalyzed inactivation of thrombin by antithrombin. We assume therefore that ioxaglate interferes with the binding of macromolecular substrates to the anionic exosite I of thrombin. The biological correlation to the observed antithrombotic effect of ioxaglate is then to be found in the inhibition of thrombin generation via inhibition of thrombin-mediated feedback activations. Keywords: antithrombotics, contrast media, exosite I, ioxaglate, thrombin. Introduction Radiographic contrast media (CM) used in diagnostic and therapeutic angiography procedures interfere with blood coagulation and platelet function [1 5]. The effect on the hemostasis/thrombosis system is dependent on the type of the CM: ionic or non-ionic, or high, iso- or low osmolarity. Several randomized and non-randomized trials have reported a higher risk of arterial closure during the coronary angioplasty and stenting with non-ionic CM compared with the ionic ones [6 13], although the inverse has also been reported [14]. It has also been demonstrated that the use of the ionic dimer contrast agent ioxaglate is associated with a reduction in thrombotic episodes Correspondence: Professor H. Coenraad Hemker, Synapse BV, CARIM, University of Maastricht, PO Box 616, 6200 MD Maastricht, the Netherlands. Tel.: þ ; fax: þ ; hc.hemker@thrombin. com Received 22 April 2002, accepted 13 June 2002 after coronary angioplasty [1]. In a previous study, we observed that upon administration of ioxaglate (1% v/v) to six patients undergoing angiography, thrombin generation in platelet-rich plasma (PRP) was diminished [15]. This broaches the question of the possible interaction of ioxaglate with heparin and platelet antagonists that are both potent inhibitors of thrombin generation [16] and frequently administered during angiography. We therefore studied the mechanism of inhibition by ioxaglate and assessed the extent of interaction with heparin and abciximab. In order to stay as close as possible to the situation in vivo, we carried out our studies in platelet-poor plasma (PPP) and PRP, and used isolated factors only in so far as necessary to arrive at a plausible explanation of the observed inhibition. Part of the data shown here was presented in the Transcatheter Cardiovascular Therapeutics 2000 in Washington DC [17] and in the XVIII ISTH conference in Paris [18]. Materials and methods Materials Plasma PRP was prepared by centrifuging fresh citrated blood (nine parts of blood to one part of citrate solution 0.13 mol L 1 ) at 250 g, 158C for 10 min. The platelet count was adjusted to ml 1 using autologous PPP. Reference plasma: plasma was pooled from 10 to 12 healthy donors. PPP was obtained by two centrifugations at 15 8C, for 15 min at 3000 g, and a third centrifugation was made at 4 8C, for 15 min at g, and stored at 80 8C. The clotting factors and the antiproteases were checked to be in the normal range. Chemicals, reagents and buffers Chromogenic substrate used for thrombin determination was S2238 (H-D-Phe-Pip-ArgpNA.2HCl, Chromogenix AB, Malmö, Sweden). Recombinant relipidated tissue factor (rtf), polybrene- and calcium-free, was a gift from Dade Behring (Marburg-Germany). Phospholipids were added in the form of vesicles consisting of 20 mol% phosphatidylserine (PS) and 80 mol% phosphatidylcholine (PC), prepared as previously described [19](Avanti Polar Lipids Inc., AL, USA). Bovine serum albumin (lot A-7030) was purchased from Sigma Chemicals (St Louis, MO, USA). Ancrod, the fibrinogen-clotting enzyme extracted from the Malayan pit viper, was the commercial preparation Arvin 1

2 270 R. Al Dieri et al (Knoll AG, Ludwigshafen, Germany). It has been shown that at the final concentration used (1 U ml 1 plasma, causing a clotting time of s) it did not activate platelets or any clotting factors, notably factor (F)V and FVIII. Purified clotting factors were provided by R. Wagenvoord (Synapse BV. Cardiovascular Research Institute, Maastricht, the Netherlands). Activated FVand FVIII were prepared ex tempore by incubation with thrombin (2 nmol L 1 ) and Ca 2þ for 10 min. Buffer A, ph 7.35, consists of 0.02 mol L 1 Hepes, 0.15 mol L 1 NaCl and 0.5 g ml 1 bovine serum albumin. Buffer B is buffer A with 0.02 mol L 1 EDTA adjusted to ph 7.9. Unfractionated heparin (Laboratory LeoS.A, Denmark) was used. Radiographic contrast media (CM): ioxaglate (Hexabrix 1 320) as an ionic dimer and iodixanol (Visipaque 1 320) as a nonionic dimer and Abciximab (ReoPro 1 ) were obtained from the hospital apothecary. Defibrination of plasma At 37 8C, 20 ml of a 50 U ml 1 Ancrod solution was added per 1 ml of PPP. After 10 min at 37 8C and 10 min incubation on ice the fibrin clot was wound out onto a plastic spatula. Thrombin generation: the thrombogram The thrombin formed was assessed by a subsampling method [20]. The reaction mixture consisted of 240 ml of plasma in 360 ml final volume and 60 ml of buffer A containing phospholipid vesicles (1 mmol L 1 ) and activator. These were incubated for at least 5 min at 37 8C before 60 ml of CaCl 2 (0.1 mol L 1 ) was added to start the reaction. The activators used were: (i) for the extrinsic system: 15 pmol L 1 rtf; (ii) for the intrinsic system, 50 mgml 1 of kaolin. From the reaction mixture, 10 ml aliquots were transferred into cuvettes containing 490 ml S2238 (0.2 mmol L 1 in buffer B) at regular intervals (10 60 s). Either after approximately 2 min or at the time when visible color production could be observed, the reaction was stopped by 1 mol L 1 citric acid solution. Sampling time and stopping time were registered through push buttons on the pipettes, connected to a personal computer. From these times and the measured OD (405 nm), the amidolytic activity (OD min 1 ) was automatically calculated and converted into equivalent thrombin concentrations. With PRP, thrombin generation was performed similarly: 240 ml PRP (platelet count adjusted to platelets ml 1 ) was supplemented with 60 ml of buffer A ( the material to be tested), and incubated for 10 min at 37 8C. At zero time, the coagulation was triggered by 60 ml of 0.1 mol L 1 CaCl 2 [20,21]. The thrombogram is characterized by three main parameters: (i) the lag time, which is for all practical purposes equivalent to the clotting time; (ii) the peak height that renders the maximal velocity of net thrombin production that when antithrombin is in the normal range reflects the maximal prothrombinase activity attained; and (iii) the endogenous thrombin potential (ETP), which is the area under the thrombin generation curve (AUC). Thrombin and arvin clotting times Thrombin or arvin was added to normal pooled plasma preincubated with CM for 1 min, and clotting times were measured. Human thrombin and arvin were used at concentrations adjusted to obtain a clotting time of approximately 65 s in normal plasma (6 and 2 U ml 1, respectively). Results Influence of CM on the thrombogram in PPP and PRP In PPP, ioxaglate at 5% (v/v) strongly inhibited thrombin generation that was triggered with tissue factor (TF) and completely abolished intrinsic thrombin generation (Fig. 1a,b). Iodixanol at the same concentration slightly increased the thrombin generation in both systems (Fig. 1a,b). The inhibition was independent of the presence of fibrinogen (data not shown). In the presence of platelets, iodixanol at 5% (v/v) postponed thrombin generation but increased the peak height and the ETP (Fig. 1c). Ioxaglate at the same concentration prolonged the lag time much more and significantly inhibited both peak height and ETP (45%) (Fig. 1c). Interaction of CM with antithrombotic agents Abciximab The GPIIb/IIIa antagonist, abciximab, inhibits platelet-dependent thrombin generation and prolongs the lag time [16]. Admixture of iodixanol (5% v/v) further retarded the thrombin generation but had no influence on the amount of thrombin formed (Fig. 2). In contrast, ioxaglate at 5% (v/v) together with abciximab (40 mgml 1 ) almost completely abolished thrombin generation, whereas each of the drugs alone caused an inhibition of 45% and 25%, respectively. Heparin TF-induced thrombin generation was measured in PPP in the presence and absence of 0.04 U ml 1 of unfractionated heparin and 5% (v/v) of ioxaglate, alone or in combination. The parameters of the thrombogram (lag time, peak height and the ETP) were affected markedly by ioxaglate and heparin together (Fig. 3); the inhibitory effects appeared to be additive. The pseudo first-order decay constant of thrombin inactivation, in the presence and absence of various concentrations of heparin, was not significantly affected by concentrations up to 20% (v/v) of either CM (data not shown). The influence of CM on the antithrombin-mediated decay of thrombin appeared to be negligible. Localization of the inhibitory effect of ioxaglate As ioxaglate has no influence on thrombin breakdown, its effect on the thrombin-time curve must be localized in prothrombin conversion. From the data of the thrombin curves and the decay constant of thrombin in the absence and presence of ioxaglate we calculated [20] the course of prothrombin conversion and noted that prothrombinase activity was lowered from 95% to 0% as the concentration of ioxaglate was increased from 5% to

3 Ioxaglate interference with feedback activation of FV, FVIII and platelets 271 Fig. 2. Effect of abciximab on the thrombogram in PRP in the absence and presence of CM (5% v/v). *, Control; *, abciximab alone (40 mgml 1 ); &, abciximab þ iodixanol; ~, abciximab þ ioxaglate. Data represent median of three independent experiments. Fig. 3. Effect of unfractionated heparin on the extrinsic thrombogram in PPP in the presence and absence of ioxaglate. Thrombin generation was measured at 0.04 U ml 1 of unfractionated heparin and at 5% v/v of ioxaglate. *, Control; &, unfractionated heparin alone; ^, ioxaglate alone; ~, unfractionated heparin þ ioxaglate. Data represent median of four independent experiments. Fig. 1. Influence of the contrast media addition on the thrombogram in PPP and PRP. (a) In defibrinated PPP initiated with rtf. (b) In defibrinated PPP initiated with contact activator. (c) In PRP initiated only with Ca 2þ. *, Control; *, iodixanol (5% v/v); ~, ioxaglate (5% v/v). Data represent median of four independent experiments. 20% v/v. It also increased the amount of residual prothrombin in serum (data not shown). We checked whether ioxaglate had a direct effect on purified prothrombinase (i.e. FXa and FVa) acting on prothrombin in the presence of procoagulant phospholipids, measured according to [19] and on intrinsic tenase (i.e. FIXa and FVIIIa acting on FX in the presence of procoagulant phospholipids measured according to [22]. These reactions all appeared to be unaffected (results not shown), so we deduced that ioxaglate inhibits the formation of prothrombinase in plasma but not the action of prothrombinase as such, or the activation of FX. In accordance with these results, FXa and FIXa did not overcome the inhibition of thrombin generation in plasma when added during the lag phase of thrombin generation in the presence of ioxaglate, unless added at non-physiologically high concentrations (>20 nmol L 1 ; results not shown). As can be seen from Fig. 4(a), the inhibition of thrombin generation by ioxaglate could be relieved by adding FV that had been previously activated by thrombin and Ca 2þ (10 min), but not by FV as such. The effect was not due to the thrombin present in the FVa preparation because the same concentration of thrombin added without FV(a) had no effect (data not shown). This suggests that ioxaglate has an influence on the activation of FV. Indeed, when FV was preincubated with thrombin in the presence of ioxaglate (20% v/v), it was unable to restore thrombin generation (Fig. 4b). This fact also held true when FV was activated by an extensively washed fibrin clot, i.e. by thrombin adsorbed onto fibrin [23], a form that is likely to be abundant at the site of a thrombus (data not shown).

4 272 R. Al Dieri et al Fig. 5. Effect of ioxaglate on the activation of FVIII. The intrinsic thrombogram in PPP was measured in the absence and presence of ioxaglate (10% v/v). Ioxaglate containing plasmas were supplemented with FVa or with [FVa þ FVIII(a)] (arrow). *, Control; &, ioxaglate alone; &, ioxaglate þ (FVa þ FVIIIa); ~, ioxaglate þ (FVa þ FVIII incubated with thrombin and Ca 2þ in the presence of ioxaglate) or FVa alone (indistinguishable). Data represent median of three independent experiments. Table 1 The effect of ioxaglate on the coagulation of normal plasma by human thrombin and snake venom enzyme (arvin) and on the cleavage of S2238 by thrombin Fig. 4. Effect of ioxaglate on the activation of FV. The extrinsic thrombogram in PPP was measured in the absence and presence of ioxaglate (20% v/v). (a) Three minutes after Ca 2þ addition (arrow), ioxaglate containing plasmas were supplemented with 5 nmol L 1 of FVa or FV. *, Control; &, ioxaglate alone or ioxaglate þ FV (indistinguishable); *, ioxaglate þ FVa. (b) As above but instead of FV, a preparation was added in which FV was incubated with thrombin and Ca 2þ in the presence of ioxaglate. Data represent median of three independent experiments. Ioxaglate (% v/v) Thrombin time (s) Arvin time (s) > No clot No clot No clot No clot No clot 97 2 Amidolytic activity (%) (X SD, n ¼ 3) We also investigated the effect of the addition of thrombinactivated factors on contact-induced, i.e. FVIII-dependent, thrombin generation (Fig. 5). FVa alone had a small effect but FVa and FVIIIa together significantly enhanced thrombin generation (Fig. 5). FVIIIa, like FVa was prepared ex tempore by incubation with thrombin (2 nmol L 1 ) and Ca 2þ or a thrombin-induced and washed fibrin clot. FVIII preincubated with thrombin (free or bound) in the presence of ioxaglate (10% v/v) did not restore thrombin generation (Fig. 5). It thus appears that ioxaglate inhibits the activation of FV and FVIII by thrombin and that this phenomenon is responsible for the inhibition of thrombin generation in clotting blood. Ioxaglate did not inhibit the conversion of a small MW chromogenic substrate (S2238) by thrombin however (Table 1), i.e. it does not affect the active center as such. This suggests that it interacts with the recognition of large substrates via an exosite. Indeed, ioxaglate also inhibited the clotting of fibrinogen by thrombin. Clotting induced by arvin, an enzyme on which ioxaglate has no known effect, was much less affected than was thrombin-induced clotting (Table 1). This suggests that the prolonging of the clotting time may be partly, but not entirely, due to inhibition of fibrin polymerization. Discussion Thrombotic phenomena are common during and especially after intracoronary interventions [24,25]. Invariably such procedures are accompanied by visualization of the coronary bed with radiographic contrast media. Clinical epidemiology indicates that there are differences in the frequency of thrombotic events depending on the contrast medium (CM) used (see Introduction). We tried to find a biological correlate to these observations. Thrombin generation was chosen as an indicator of possible pro- or anti-thrombotic effects because it is increased in all prothrombotic states thus far investigated and decreases under the effect of all antithrombotics thus far tested. Not only do anticoagulants decrease thrombin generation in PPP and PRP, but antiplatelet agents also decrease thrombin generation in PRP [26,27].

5 Ioxaglate interference with feedback activation of FV, FVIII and platelets 273 Our results showed that the ionic CM ioxaglate strongly inhibits thrombin generation in PPP and especially in PRP, whereas the non-ionic iodixanol has much less effect and even slightly enhances thrombin generation (in the absence of platelets). Administration of heparin and/or platelet inhibitors such as abciximab often accompanies intracoronary interventions, so we investigated the combined effects of these drugs and CM. The inhibitory effect of heparin appeared to be independent of that of ioxaglate, the total inhibition being equal to the sum of the inhibitions. In PRP, this CM strongly boosted the inhibition of thrombin generation caused by abciximab. In an attempt to localize the inhibitory effect of ioxaglate on the coagulation system, we studied its effect on the prothrombin conversion by purified prothrombinase and tenase and found them not to be affected. We observed that the inhibition of thrombin generation could be short circuited by adding FVa during extrinsic thrombin generation and by adding FVIIIa and FVa in intrinsic thrombin generation. This strongly suggests a defect in the feedback activation of FV and FVIII by thrombin [28,29]. In agreement with a previous study [30], we observed that the amidolytic activity of thrombin on a small chromogenic substrate was not inhibited (Table 1). We therefore postulate that the ionic CM interferes with binding of large molecular weight thrombin substrates to exosite I (Figs 4 and 5). This assumption also explains the inhibition of prothrombinase that was observed by Fay and Parker [30] because, in their experiments, the plasma was used as a source of FV so that activation of FV by thrombin must be a rate-limiting step in their procedure. We did not investigate whether ioxaglate inhibits the thrombin-induced splitting of FV and FVIII because the parallelism between this splitting and the acquisition of biological activity has been documented extensively in the literature [31 34]. Neither did we extend our kinetic experiments with purified factors because it was our primary aim to find the biological correlate to a clinical phenomenon that necessarily occurs in whole blood, of which PPP and PRP can be considered to be reasonable representatives. In PRP, the exposure of a procoagulant phospholipid surface by the platelet is rate-limiting for thrombin generation. Platelets are activated directly by thrombin and, in a later stage, also via von Willebrand factor interacting with fibrin [21,35]. Platelets also release FV, but under our conditions these processes are not rate-limiting [29]. Exosite I is required for thrombin-induced activation of blood platelet procoagulant activity [36 38], so that the observations on PRP are readily explained by the same assumption as those in PPP. This is in apparent contrast to the conclusion of Fay and Parker that platelet aggregation is not affected by CM. Aggregation, however, is an early and reversible step in platelet activation that can be caused by a variety of agents including thrombin. The exposure of procoagulant phospholipids takes place in an advanced stage of platelet activation that is critically thrombin dependent [21,26,39]. The release reaction is an intermediate stage of platelet activation and has been reported to be inhibited by ionic CM [40]. Abciximab inhibits the exposure of procoagulant phospholipids [16] and thus acts on the same mechanism as ionic CM, which explains the synergistic effect that we observed. Heparin, boosting the natural antithrombin of plasma, acts via a completely different pathway; hence the independence of the inhibitions brought about by CM and heparin. Ionic CM also inhibits fibrin polymerization [41]. This causes prolonging of the clotting time observed with a snake venom enzyme (arvin) that itself is not inhibited by the non-ionic CM (Table 1 and data not shown). The prolonging of the thrombin clotting time by the ionic CM is much more outspoken than that of the snake venom enzyme. This is a consequence of exosite I being involved in the interaction between thrombin and fibrinogen. This effect may be the basis of the abnormalities of the three-dimensional fibrin structure observed by Brass et al. [41]. We conclude that the assumption that ioxaglate acts as an exosite I inhibitor explains our observations; this is in agreement with the suggestion of Li and Gabriel [42], and explains the large majority of previously published results. In view of the pivotal role of thrombin in the pathogenesis of coronary thrombotic events [43,44], the capacity of a contrast agent to prevent thrombin formation is conceivably an important asset. It remains an interesting question whether the ionic CM can be used as a lead structure for the development of a class of antithrombotics acting through exosite I inhibition. References 1 Eloy R, Corot C, Belleville J. Contrast media for angiography: physicochemical properties, pharmacokinetics and biocompatibility. Clin Mater 1991; 7: Belleville J, Baguet J, Paul J, Clendinnen G, Eloy R. In vitro study of the inhibition of coagulation induced by different radiocontrast molecules. Thromb Res 1985; 38: Belleville J, Freyria AM, Pinet A et al. Contrast agents used for excretory urography impair platelet function in human patients. Thromb Res 1982; 28: Bernstein EF, Gans H. Anticoagulant activity of angiographic contrast media. Invest Radiol 1966; 1: Verebely K, Kutt H, Torack RM, McDowell F. The effects of radioopaque contrast media on the structure, and solubility of the fibrin clot. Blood 1969; 33: Piessens JH, Stammen F, Vrolix MC et al. Effects of an ionic versus a nonionic low osmolar contrast agent on the thrombotic complications of coronary angioplasty. Cathet Cardiovasc Diagn 1993; 28: Grines CL, Schreiber TL, Savas V et al. A randomized trial of low osmolar ionic versus nonionic contrast media in patients with myocardial infarction or unstable angina undergoing percutaneous transluminal coronary angioplasty. J Am Coll Cardiol 1996; 27: Lefevre T, Adjeroud N, Royer T, Glatt B, Morice MC. Influence of contrast media on the results of percutaneous transluminal coronary angioplasty with provisional stenting: a comparative study. J Invasive Cardiol 1998; 10: Qureshi NR, den Heijer P, Crijns HJ. Percutaneous coronary angioscopic comparison of thrombus formation during percutaneous coronary angioplasty with ionic and nonionic low osmolality contrast media in unstable angina. Am J Cardiol 1997; 80: Esplugas E, Cequier A, Jara F et al. Risk of thrombosis during coronary angioplasty with low osmolality contrast media. Am J Cardiol 1991; 68:

6 274 R. Al Dieri et al 11 Fleisch M, Mulhauser B, Garachemani A et al. Impact of ionic (Ioxaglate) and non-ionic (Ioversol) contrast media on PTCA-related complications. J Am Coll Cardiol 1999; 33 (Suppl. A): (abstract). 12 Scheller B, Hennen B, Pohl A, Schieffer H, Markwirth T. Acute and subacute stent occlusion; risk-reduction by ionic contrast media. Eur Heart J 2001; 22: Batchelor W, Granger CB, Kleiman NS et al. A comparison of ionic versus non-ionic contrast medium during percutaneous transluminal coronary angioplasty for acute myocardial infarction (GUSTO II). Am J Cardiol 2000; 85: Davidson CJ, Laskey WK, Jr Hermiller JB et al. Randomized trial of contrast media utilization in high-risk PTCA. The COURT Trial. Circulation 2000; 101: Al Dieri R, de Muinck E, Hemker HC, Béguin S. An ionic contrast agent inhibits platelet-dependent thrombin generation and boosts the effect of abciximab. Thromb Haemost 2001; 85: Reverter JC, Béguin S, Kessels H, Kumar R, Hemker HC, Coller BS. Inhibition of platelet-mediated, tissue factor-induced thrombin generation by the mouse/human chimeric 7E3 antibody. Potential implications for the effect of c7e3 Fab treatment on acute thrombosis and clinical restenosis. J Clin Invest 1996; 98: Al Dieri R, Béguin S, Hemker HC. Anticoagulant-and antiplatelet-effect of an ionic iodinated contrast medium, but not of a non-ionic one. Am J Cardiol 2000; 86(8A): 100i(abstract). 18 Al Dieri R, Wagenvoord R, Béguin S, Hemker HC. Localizing the inhibitory effect of Ioxaglate (an ionic dimer contrast agent) on the coagulation system. Thromb Haemost 2001; 86: 762(abstract). 19 Rosing J, Tans G, Govers-Riemslag JW, Zwaal RF, Hemker HC. The role of phospholipids and factor Va in the prothrombinase complex. J Biol Chem 1980; 255: Hemker HC, Willems GM, Béguin S. A computer assisted method to obtain the prothrombin activation velocity in whole plasma independent of thrombin decay processes. Thromb Haemost 1986; 56: Béguin S, Lindhout T, Hemker HC. The effect of trace amounts of tissue factor on thrombin generation in platelet rich plasma, its inhibition by heparin. Thromb Haemost 1989; 61: van Dieijen G, Tans G, Rosing J, Hemker HC. The role of phospholipid and factor VIIIa in the activation of bovine factor X. J Biol Chem 1981; 256: Kumar R, Béguin S, Hemker HC. The influence of fibrinogen and fibrin on thrombin generation evidence for feedback activation of the clotting system by clot bound thrombin. Thromb Haemost 1994; 72: Lam JY, Chesebro JH, Steele PM, Dewanjee MK, Badimon L, Fuster V. Deep arterial injury during experimental angioplasty: relation to a positive indium-111-labeled platelet scintigram, quantitative platelet deposition and mural thrombosis. J Am Coll Cardiol 1986; 8: Fischer A, Gutstein DE, Fuster V. Thrombosis and coagulation abnormalities in the acute coronary syndromes. Cardiol Clin 1999; 17 (Suppl.): viii. 26 Béguin S, Kumar R. Thrombin, fibrin and platelets: a resonance loop in which von Willebrand factor is a necessary link. Thromb Haemost 1997; 78: Hemker HC, Béguin S. Phenotyping the clotting system. Thromb Haemost 2000; 84: Béguin S, Lindhout T, Hemker HC. The mode of action of heparin in plasma. Thromb Haemost 1988; 60: Baruch D, Hemker HC, Lindhout T. Kinetics of thrombin-induced release and activation of platelet factor V. Eur J Biochem 1986; 154: Fay WP, Parker AC. Effects of radiographic contrast agents on thrombin formation and activity. Thromb Haemost 1998; 80: Rapaport SI, Hjort PF, Patch MJ. Further evidence that thrombinactivation of factor VIII is an essential step in intrinsic clotting. Scand J Clin Laboratory Invest 1965; 17 (Suppl.): Jensen AH, Béguin S, Josso F. Factor V and VIII activation in vivo during bleeding. Evidence of thrombin formation at the early stage of hemostasis. Pathol Biol (Paris) 1976; 24 (Suppl.): Suzuki K, Dahlback B, Stenflo J. Thrombin-catalyzed activation of human coagulation factor V. J Biol Chem 1982; 257: Pieters J, Lindhout T, Hemker HC. In situ-generated thrombin is the only enzyme that effectively activates factor VIII and factor V in thromboplastin-activated plasma. Blood 1989; 74: Béguin S, Kumar R, Keularts I, Seligsohn U, Coller BS, Hemker HC. Fibrin-dependent platelet procoagulant activity requires GPIb receptors and von Willebrand factor. Blood 1999; 93: De Marco L, Mazzucato M, Masotti A, Fenton JW, Ruggeri ZM. Function of glycoprotein Ib alpha in platelet activation induced by alpha-thrombin. J Biol Chem 1991; 266: Jandrot-Perrus Huisse MG, Krstenansky JL, Bezeaud A, Guillin MC. Effect of the hirudin carboxy-terminal peptide on the interaction of thrombin with platelets. Thromb Haemost 1991; 66: Jandrot-Perrus M, Clemetson KJ, Huisse MG, Guillin MC. Thrombin interaction with platelet glycoprotein Ib: effect of glycocalicin on thrombin specificity. Blood 1992; 80: van Zutphen H, Bevers EM, Hemker HC, Zwaal RF. Contribution of the platelet factor V content to platelet factor 3 activity. Br J Haematol 1980; 45: Corot C, Chronos N, Sabattier V. In vitro comparison of the effects of contrast media on coagulation and platelet activation. Blood Coagul Fibrinolysis 1996; 7: Brass O, Belleville J, Sabattier V, Corot C. Effect of ioxaglate an ionic low osmolar contrast medium on fibrin polymerization in vitro. Blood Coagul Fibrinolysis 1993; 4: Li X, Gabriel DA. Differences between contrast media in the inhibition of platelet activation by specific platelet agonists. Acad Radiol 1997; 4: Chesebro JH, Zoldhelyi P, Fuster V. Pathogenesis of thrombosis in unstable angina. Am J Cardiol 1991; 68: 2B 10B. 44 Gallo R, Padurean A, Toschi V et al. Prolonged thrombin inhibition reduces restenosis after balloon angioplasty in porcine coronary arteries. Circulation 1998; 97: