ADJUVANT EFFECT OF ARGATROBAN ON STAPHYLOKINASE INDUCED THROMBOLYSIS OF PLATELET RICH THROMBI IN RAT MESENTERIC VENULES IN VIVO

TlrrombosisResearch, Vol. 86, No. 2,pp. 115-126,1997 Copyright @ 1997 Elsevier Science Ltd Printed in the USA. All rights reserved 0049.3848/97 $17.00 + .00

Pergamon

PII SO049-3848(97)OO055-8

ADJUVANT EFFECT OF ARGATROBAN ON STAPHYLOKINASE INDUCED THROMBOLYSIS OF PLATELET RICH THROMBI IN RAT MESENTERIC VENULES IN VWO Makiko Kawano’, Sadahiro Watanabe2, Yasuto Sasaki’, John C Giddings~ and Junichiro Yamamoto’ 1.Laboratory of Physiology, Faculty of Nutrition, Kobe Gakuin University, Kobe, Japan 2. Kobe City College of Nursing, Division of Basic Medical Science, Kobe, Japan 3, Department of Hematology, University of Wales College of Medicine, Cardiff, UK

(Received30 August 1996by Editor E Numano; revised/accepted3 March 1997)

Abstract Effective, therapeutic thrornbolysis should not only promote dissolution of fibrin but should also regulate continued thrombin-induced fibrin formation and the accumulation of platelets on the thrombotic lesion. The aim of the present study was to assess the use of a synthetic, low molecular weight thrombin inhibitor, argatroban in association with a well defined thrombolytic agent in a reproducible animal model of thrornbolysis ill v~vo, Thrombi were formed in rat mesenteric venules with a helium neon (He-Ne) laser in the presence of Evans blue and were stabilised for 10 minutes. Thrombi formed in this manner were shown by transmission electron microscopy to be composed mainly of platelets. Thrombolysis was induced with recombinant staphylokinase in the presence and absence of argatroban and the process was monitored using computerised image analysis, Co-infusion of argatroban at a dose of 2,0mg/kg/h with staphylokinase significantly enhanced the rate of thrombolysis. The results suggested that administration of the thrornbin inhibitor together with the fibrinolytic agent moderated platelet-dependent mechanisms and led to a more rapid restoration of blood vessel patency. Q1997 ElsevierScienceLtd Key words: thrombolysis, thrombosis, staphylokinase, argatroban Corresponding author: J Yamamoto PhD, Laboratory of Physiology, Faculty of Nutrition, Kobe Gakuin University, Nishi-ku, Kobe 651-21, Japan. 115

116

ARGATROBANAND THROMBOLYSIS

Vol. 86, No. 2

Thrombosis plays central role in the pathogenesis of myocardial and cerebral infarction and thrombolytic therapy provides an effective means of treatment in these cases. A variety of animal models have been proposed to study thrombotic and thrombolytic mechanisms (l-5) and many of these have been adapted especially to assess the efficacy of thrornbolytic agents for potential use in clinical practice (5-8).

Thrombi formed in arteries, and acute thrombi formed in veins, are mainly composed of platelets (9). Moreover, as well as acting on fibrinogen, thrombin appears to be an important mediator of platelet aggregation in stenosed blood vessels (10). Inhibitors of thrornbin activity have been shown to prevent or moderate platelet-rich vascular occlusion ii? vivo ( 11- 13). It appears reasonable to propose, therefore, that effective thrombolytic therapy would not only promote dissolution of fibrin but would also moderate platelet-dependent thrombogenesis. Acute plateletrich thrombi are reproducibly formed in vivo by pulsed irradiation with a helium-neon (He-Ne) laser in the presence of Evans blue (14). We have previously utilised this method to study microvascular thrombosis (13, 15-17) and we have confirmed that the synthetic antithrornbin, argatroban, inhibited acute thrombus formation in rat blood vessels ( 13, 16). In the present investigations, we have extended the use of this method to assess the effects of argatroban on staphylokinase-induced thrombolysis in vivo.

MATERIALS AND METHODS

Male Wistar ST rats weighting 260-300g were obtained from SLC Japan. Argatroban was purchased from Daiichi Pharmaceutical Co Ltd Japan and recombinant staphylokinase was kindly donated by Yakult Co Ltd Japan.

Tr”anwni&sionelectron microscopy

Mesenteric tissue was fixed with 2,5% (v/v) glutaraldehyde in O.IM phosphate buffer, pH7.4, postfixed with 1% osmium tetroxide, dehydrated through a graded series of acetone and embedded in epoxy resin. Ultra thin sections were stained with uranil acetate and lead citrate, and were examined under a JEM 2000 EX electron microscope,

Lctser-indmedthrombosiscordtl)io]jl[)oly,si,t Microvascular thrombi were produced in rat mesenteric venules by a slight modification of the method described by Yamamoto et al (15). Briefly, canulae (inner diameter O.58mm, outer diameter 0.97mm) were inserted into a femoral artery and vein of anaesthetized animals for the measurement of blood pressure and infusion of test substances respectively, The mesentery was

Vol. 86, No. 2

ARGATROBANAND THROMBOLYSIS

117

secured flat on the operating stage in Tyrode solution at 371’c.Venules were irradiated with a He-Ne laser beam at the midpoint between the vessel wall and the centreline of the lumen after the injection of Evans blue (14.2mg/kg). The diameter of the laser spot was 30~m at the focal plane and irradiation was repeated for 15 seconds at one minute intervals until the height of the thrombus reached 90% of the lumen diameter, The thrombus was then allowed to stabilize for 10 minutes before recombinant staphylokinase (rSAK) with or without argatroban (or saline as control) was infused for a further 60 minutes through the femoral veins. rSAK and argatroban were infused through separate canulae in different femoral veins.

Conlpulerisedimageanalysisoftivombolysis The process of thrombolysis was continuously recorded on a videotape recorder. Subsequently, images at fixed time intervals were transferred to a personal computer and were analysed by quantitating changes in grey scale inside an inscribed circle as shown in FIG, 1 using Macintosh - NIH- image software. The size of the thrombus was calculated from the following equation

Thrombus size= (Tn-Rn) / (T I-R1)

where T 1 was the intensity of the grey scale inside the circle (T) overlaying the thrombus immediately before infusion of the thrombolytic agent, and RI was the intensity inside the reference circle (R) at the same time point. Tn and Rn were the equivalent measurements inside the respective circles at intervals during thrombolysis. Rn values varied during the experiments and was measured at each time interval to obtain the precise thrombus size. The size of the

FIG. 1 Computer-assisted analysis of thrombus size

118

ARGATROBANAND THROMBOLYSIS

Vol. 86, No. 2

FIG. 2 Electron micrograph of a thrombus formed in a venule A) a venule without laser irradiation (control, magnification x 950), B) a thrombus in a Venule (X 1500), C) a slight injury of an endothelial cell (x 6800). Arrow shows injury, arrow head an endothelia[ cell and asterisk platelet aggregates.

Vol. 86, No. 2

ARGATROBANAND THROMBOLYSIS

119

thrombus during thrombolysis was expressed relative to the size of the same thrombus before administration of the thrombolytic agent.

StatisticalAnaly.~is The results were analysed by the unpaired Student t-test and were expressed as mean * standard error (SEM), Other technical details are given in relevant parts of the text.

RESULTS

Transmission electron microscopic observation of a He-Ne laser-inducedthrombw

A representative electron micrograph of a thrombus formed in the rat mesenteric venule is illustrated in FIG, 2, A mural thrombus mainly composed of platelets and not incorporating erythrocytes entrapped in fibrin was observed. A relatively large number of leukocytes were associated with the platelets (FIG. 2B), It is especially noteworthy that the endothelium was not denuded by the laser treatment, although slight endothelial injury could be identified (FIG. 2C)

Conlputer-assistedimageanalysis~f dle t}ll.olllbolytic ~]l-oce~’t~ The process of thrombolysis induced by rSAK is shown in FIG. 3. The rate of thrombus dissolution was assessed by analysing computer images taken at one minute intervals (FIG. 4). The mural thrombus gradually lysed and disappeared within 90 minutes after the onset of infusion. The thrombus persisted when saline was infused in place of rSAK.

Dose-dependentthrombolysisindlicedby staphylokina~’e Various concentrations of rSAK were infused to examine dose-related thrombolysis. The disappearance of the thrombus progressed evenly without intermittent rapid changes. The process was thus analysed by computer using images taken at 10 minute intervals. Both the rate of the thrombolysis and the extent of thrombus dissolution at 90 minutes significantly increased with increasing concentrations of rSAK (FIG. 5). An infusion dose of 5.8mg/kg/h was selected to examine the effect of argatroban on the thrombolytic process.

Adjvvant efect of argatrobanon thsombolysis Increasing concentrations of argatroban were infused in association with a constant concentration of rSAK through separate canulae in different femoral veins. The results are shown in FIG. 6. Argatroban at 20mg/kg/h significantly accelerated the rate of thrombus dissolution. There was a tendency for this adjuvant effect to be dose-related but the figures did not reach convincing statistical significance at the lower concentrations of argatroban.

120

ARGATROBANAND THROMBOLYSIS-- ~~~

Vol. 86, No. 2

FIG. 3 Thrombolytic process The thrombus formed by laser irradiation was stabilised and infusion of rSAK was begun at stage 1. The time course of thrombolysis after onset of infusion is shown.

JJ.--3-& o

I

30

rSAK

60

90

I

Time after oneet of rSAK infusion (rein)

FIG. 4 Change in thrombus size induced by rSAK Thrombus size was expressed as relative values to that just before onset of rSAK or saline infusion, ● : rSAK (5,8m~k#h) ; () : saline.

ARGATROBANAND THROMBOLYSIS

Vol. 86, No. 2

m —-

n

w

60

rSAK or saline

[

Time after onset of rSAK infueion (rein)

FIG. 5 Thrombolysisinduced by various concentrations of rSAK 0 : saline; v : staphylokinase, 3.85m@@h, v : 5.8m#kg/h; + : 67mg/k@; O : 7,7m~kg/h; . : 15,4mg/kg/h.Differences in means between rSAK and saline groups were analysed at each time interval. 3-6 rats were used in each group. *: p
1 al

.!i m

o

1

●

I

I

I

iiiaik s””

Time after onset of rSAK infuaion (rein)

FIG. 6 Adjuvand effect of argatroban on rSAK-inducedthrombolysis rSAK (58mgk~h) was infusedwithargatrobanor saline.0: saline; .: argatroban, 2.Omg/kg/h; v: 0.67mtikg/ll; T: 0.2mtik#h. Differences in means between argatromban and saline grorps were analysed at each time interval. 4-5 rats were used in eachgroup.*: P
121

ARGATROBANAND THROMBOLYSIS

Vol. 86, No. 2

DISCUSSION

Thrombolytic therapy plays an important role in the clinical management of myocardial infarction and stroke, and a variety of animal models have been described to investigate the efficacy of thrombolytic agents for potential use in these circumstances, In general, these models have utilised relatively large arteries in conjunction with physically-induced stenosis to raise shear forces in the occluding blood vessel (1, 2, 4). These models include those which promote endothelial denudation and are considered appropriate to study thrombosis and thrombolysis in advanced vascular disease which accompany vessel wall injury, In addition, animal models using microvessels have been proposed which are simple to prepare and are reproducible, For example, Kawasaki et al (6) produced thrombi in guinea pig arterioles using light irradiation at 420-490nm in conjunction with a fluorescent dye and examined the activity of two thrombolytic agents. Arterial thrombi of this nature are mainly composed of platelets and it is possible that these occlusive lesions are especially similar to cardiovascular and cerebrovascular thromboemboli in man, Moreover, acute thrombi in veins are also platelet-rich and several animal models using the venous circulation have helped to clarify primary thrombogenic mechanisms in vim (14-16, 18, 19)

In the present study, platelet-rich thrombi were formed in rat mesenteric venules using He-Ne laser irradiation in the presence of Evans blue, This dye is bound to albumin in blood and converts laser energy to heat resulting in thrombosis. Kovacs et al (14) originally developed the technique using rat mesenteric arterioles and venules and speculated that localised heat generated by the laser would result in endothelial denudation. A similar technique was utilised by Rosenblum et al (19) using pial arterioles and venules. In these instances however, endothelial denudation was not observed, In our studies, examination of blood vessels at the site of thrombus formation using transmission electron microscopy also failed to demonstrate loss of the endothelial monolayer. Some slight injury to individual endothelial cells was identified by changes in electron density but this was not thought to be fully responsible for the initiation of thrombosis as platelets were seen to adhere and aggregate on intact areas of the vessel wall that were not electron dense, It seems possible that laser irradiation activated endothelial cells and induced platelet adhesion and aggregation at the stimulated sites, It also seems likely that the primary platelet aggregates were subsequently consolidated and stabilised by fibrin formation. The precise mechanisms of thrombus generation in this system remain unclear, however. Erythrocytes were not heavily incorporated into the acute thrombus, although relatively large numbers of leukocytes were observed. The significance of this finding remains to be explored but it is noteworthy that evidence is increasing to suggest that there maybe a close relationship between platelet and leucocyte function (20-23). Furthermore, leucocyte enzymes such as elastase and cathepsin G are believed to contribute to fibrin breakdown in vim (24). Our findings confirmed that the He-Ne laser thrombosis technique would be a useful model to study thrombolytic mechanisms at an early stage of vascular disease where marked loss of endothelial

Vol. 86, No. 2

ARGATROBANAND THROMBOLYSIS

123

integrity is rarely observed.

Staphylokinase is a protein with a molecular weight of 15,500 secreted from S!aphylococcm awem, The protein is not an enzyme but, like streptokinase forms an equimolar complex with the proenzyme, plasminogen, resulting in the activation of plasminogen to plasmin. In contrast to streptokinase (and urokinase), however, staphylokinase does not augment fibrinogen breakdown and catalyses relatively fibrin-specific clot lysis specially in a plasma environment (25-28). Furthermore, in preliminary studies we demonstrated that rSAK was markedly more active than either streptokinase or urokinase in a fibrinolytic test system utilising euglobulin fractions of rat plasma. For these reasons, therefore, we chose rSAK as the fibrinolytic agent for our ill vivo investigations.

In the current studies of acute thrombosis, rSAK was shown to exhibit thrombolytic potency in a time and dose dependent manner. In addition, the rate of Iysis was significantly enhanced by co-infusion of the synthetic antithrombin, argatroban. It seems likely that this dose dependent enhancement of thrombolysis by argatroban was dependent on its antithrombin activity in the circulating blood. Direct antithrombin activity was not measured in the present study but the data imply that measurements of circulating antithrombin could be helpful in the laboratory investigation of therapeutic fibrinolysis. The effective dose of argatroban, 2.Omg/kg/h,has been shown to inhibit platelet rich thrombosis in rat cerebral microvessels (16). The results support the concept that the efficiency of thrombolysis in vivo is governed not only by the dissolution of the existing thrombus but also by the formation of a new occlusive matrix.

It has been reported that platelets may be activated during thrombolysis (29) and that platelet dependent thrombin generation may occur after fibrinolytic treatments (30), Furthermore, several studies have demonstrated that thrombin is an important physiological agonist of platelet-related thrombotic mechanisms (10-13, 16, 17, 31, 32) It is reasonable to speculate, therefore, that control of thrombin generation and inhibition of platelet function would have added to the effectiveness of thrombolysis in our ill viw model. Our findings are also in keeping with reports that co-administration of thrombin inhibitors with thrombolytic agents prevents acute reocclusion following thrombolytic therapy (7, 8, 33-36). The results demonstrated that the current animal model provided a useful means to evaluate the lysis of thrombi at an early stage where vessel wall injury was minimal. Studies of this nature could help to clarify thrombolytic mechanisms in vivo and may assist with the development of new therapeutic strategies.

REFERENCES

1. FOLTS, J.D., CROWELL, E.B.and ROWE, G.G. Platelet aggregation in partially obstructed

124

ARGATROBANAND THROMBOLYSIS

Vol. 86, No. 2

vessels and its elimination with aspirin, Circulation 54, 365-370, 1976. 2, HASKEL, E,J., PRAGER, N.A,, SOBEL, B.E,and ABENDSCHEIN, D.R. Relative efficacy of anti thrombin compared with antiplatelet agents in accelerating coronary thrombolysis and preventing early reocclusion, Circulation 83, 1048-1056, 1991. 3. GOLD, G.K., YASUDA, T,, JANG, I., GUERRERO, J.L,, FALLON, FT., LEINBACH, R. C.and COLLEN, D. Animal models for arterial thrombolysis and prevention of reocclusion. Circulation 83 [suppl N ]: IV 26- IV-40, 1991, 4. STRONY, J,, SONG, A., RUSTERHOLTZ, L, and ADELMAN, B. Aurintricarboxylic acid prevents acute rethrombosis in a canine model of arterial thrombosis. Arterioscler Thromb Vase Biol 1.5,359-366, 1995, 5, HIRATA, Y., UMEMURA, K., TAKIGUCHI, Y, and NAKASHIMA, M, A thrombosis model for evaluating thrombolytic agents, staphylokinase, on thrombolytic activity. Blood Coagul Fibrinolysis 4, 569-575, 1993. 6. KAWASAKI, T,, KAKU, S., TAKENAKA, T., YANAGI, K. and OHSHIMA, N. Thrombolytic activity of YM866, a novel modified tissue-type plasminogen activator, in a photochemically induced platelet rich thrombosis model. J Cardiovasc Pharmacol 23,884889, 1994. 7. MELLOTT, M.J,, CONNOLLY, TM., YORK S.J. and BUSH, L.R. Prevention of reocc]usion by MCI-9038, a thrombosis, Thromb Haemost 64, 526-534, 1990, 8. SCHNEIDER, J. Heparin and thrombin inhibitor argatroban enhance fibrinolysis by infused or bolus-injected saruplase (r-scu-PA) in rabbit femoral artery thrombosis. Thromb Res 64, 677-689, 1991. 9, STOCKMANS, F,, DECKMYN, H,, GRUWEZ, J., VERMYLEN, J.and ACLAND, R, Continuous quantitative monitoring of mural, platelet-dependent, thrombus kinetics in the crushed rat femoral vein. Thromb Haemost 65, 425-431, 1991. 10, EDIT, J,F,, ALLISON, P., NOBLE, S., ASHTON, J., GOLINO, P., MCNATT, J., BUJA, L, M. and WILLERSON, J. T. Thrombin is an important mediator of platelet aggregation in stenosed canine coronary arteries with endothelial injury. J Clin Invest 84, 1827, 1989, 11. HANSON, S,R, and HARKER, L.A, Interruption of acute platelet-dependent thrombosis by the synthetic antithrombin D-phenylalanyl-L-proly l-L-arginyl chloromethyl ketone, Proc Natl Acad Sci 85,3184-3188, 1988. 12. JANG, 1,,GOLD, H,K., ZISKIND, A.A., LEINBATH R,C., FALLON, J.T, and COLLEN, D. Prevention of platelet rich arterial thrombosis by selective thrombin inhibition. Circulation 81, 219-225, 1990, 13,YAMASHITA, T., YAMAMOTO, J., SASAKI, Y. and MATSUOKA, A. The antithrombotic effect of low molecular weight synthetic thrombin inhibitor argatroban and PPACK, on He-Ne laser-induced thrombosis in rat mesenteric microvessels. Thromb Res 69, 93-100, 1993. 14. KOVACS, LB., TIGYI-SEBES, A., TROMBITAS, K. and GOROG, P. Evans blue: An ideal energy-absorbing material to produce intravascular microinjury by He-Ne gas laser. Microvasc Res 10, 107-124, 1975. 15. YAMAMOTO, J., IIZUMI, H., HIROTA, R., SHIMONAKA, K,, NAGAMATSU, Y., HORIE, N. and MORITA, S. Effect of physical training on thrombotic tendency in rats: decrease

Vol. 86, No. 2

ARGATROBANAND THROMBOLYSIS

125

in thrombotic tendency measured by the He-Ne laser-induced thrombus formation method. Haemostasis 19,260-265.1989, 16. SASAKI, Y., MORII, S., YAMASHITA, T. and YAMAMOTO, J. Antithrombotic effect of argatroban of the pial vessels of the rat: a study with He-Ne laser-induced thrombus formation. Haemostasis 23, 104-111, 1993, 17, YAMASHITA, T., MATSUOKA, A., FUNATSU, A and YAMAMOTO, J. The antithrombotic effect of human activated protein C on He-Ne laser-induced thrombosis in rat mesenteric microvessels. Thromb Res 75, 33-40, 1994. 18. SATO, M, and OHSHIMA, N. Platelet thrombus induced in viw by filtered light and fluorescent dye in mesenteric microvessels of rat, Thromb Res 35,319-334, 1984. 19, POVLISHOCK, J.T. and ROSENBLUM, WI, Injury of brain microvessels with a helium neon laser and evans blue can elicit local platelet aggregation without endothelial denudation. Arch Pathol Lab Med 111,415-421, 1987. 20. YAMAMOTO, J., ISHII, L, OKADA, Y., YAMASHITA, T., RIDLER, CD., GOROG, P. and KOVACS, I.B. Effect of leucocyte products on platelet thrombus formation, coagulation and spontaneous thrombolysis, as measured from native human blood, in vitro, Thromb Res 71, 281-287, 1993. 21. CERLETTI, C,, EVANGELISTA, V,, MOLINO, M. and de GAETANO, G. Platelet activation by polymorphonuclear leukocytes: Role of cathepsin G and P-Selectin. Thromb Haemost 74,218-223, 1995. 22, FURIE, B. and FURIE, B.C. The molecular basis of platelet and endothelial cell interaction with neutrophils and monocytes: Role of P-Selectin and the P-Selectin Iigand, PSGL- 1. Thromb Haemost 74,224-227, 1995. 23, RUF, A, and PATSCHEKE, H. Platelet-induced neutrophil activation: platelet-expressed fibrinogen induces the oxidativeburst in neutrophils by an interaction with CD II C / CD18, Br J Haematol 90,791-796,1995. 24. NAGAMATSU, Y., FUKUDA, A., YAMAMOTO, J., TSUDA, Y. and OKADA, Y. Quantification of leucocyte elastase.and cathepsin G in plasma by a simple method. Effect of elastase in plasma levels of D-dimer and thrombomodulin. Clin Lab Haematol 1$, 29-34, 1996. 25. SAKO, T, Overproduction of staphylokinase in Esherichia coli and its characterization. Ann J Biochem 149, 557-563, 1985. 26. SAKAI, M., WATANUKI, M. and MATSUO, O. Mechanism of fibrin-specific fibrinolysis by staphylokinase. Biochem Biophys Res Commun 162,830-837, 1989. 27. MATSUO, O., OKADA, K., FUKAO, H., TOMIOKA, Y., UESHIMA, S., WATANUKI, M, and SAKAI, M. Thrombolytic properties of staphylokinase. Blood 76,925-929, 1990. 28, LIJNEN, HR., HOEF, B.V., COCK, F.D., OKADA, K., UESHIMA, S., MATSUO, O. and COLLEN, D. On the mechanism of fibrin-specific plasminogen activation by staphylokinase. J Biol Chem 266, 11826-11832, 1991. 29. FITZGERALD, D.J,, CATELLA, F., ROY, L. and FITZGERALD, GA. Marked platelet activation ill vivo after intravenous streptokinase in patients with acute myocardial infarction. Circulation 77, 142-150, 1988. 30, ARONSON, D.L., CHANG, P. and KESSELER, C.M. Platelet-dependent thrombin

126

ARGATROBANAND THROMBOLYSIS

Vol. 86, No. 2

generation after in vitro fibrinolytic treatment. Circulation 85, 1706-1712, 1992. 31, HERAS, M., CHESEBRO, J,H., WEBSTER, M.W.I:, MRUK J.S,, GRILL, DE., PENNY, W, J., BOWIE, E, J. W., BADIMON, L. and FUSTER, V. Hirudin, heparin, and placebo during deep arterial injury in the pig. Circulation 82, 1476-1484,1990. 32. HARKER, LA., HANSON, S.R. and RUNGE, M.S. Thrombin hypothesis of thrombus generation and vascular lesion formation. Am J Cardiol 75, 12B-17B, 1995. 33. FITZGERALD, D.J. and FITZGERALD, G.A. Role of thrombin and thromboxane A2 in reocclusion following coronary thrombolysis with tissue-type plasminogen activator. Proc Natl Acad Sci 86, 7585-7589, 1989. 34. IMURA, Y., STASSEN, J., VREYS, 1,, LESAFFRE, E., COLD, H,K. and COLLEN, D. Synergistic antithrombotic properties of G4120, a RGD-containing synthetic peptide, and argatroban, a synthetic thrombin inhibitor, in a hamster femoral vein platelet-rich thrombosis model. Thromb Haemost 68, 336-340, 1992. 35. CHESEBRO, J.H., ZOLDHELYI, P., BADIMON, L. and FUSTER, V. Role of thrombin in arterial thrombosis: Implications for therapy. Thromb Haemost 66, I-5, 1991, 36. COLLEN, D. and WERF, F.V, Coronary thrombolysis with recombinant staphylokinase in patients with evolving myocardial infarction, Circulation 87, 1850-1853, 1993.