The vascular plasminogen activator as source of the fibrinolytic potential observed during cardiopulmonary bypass

THROMBOSIS RESEARCH 67; 579-588,1992 00493848/92 $5.00 + .OOPrinted in the USA. Copyright (c) 1992 Pergamon Press Ltd. All rights reserved.

THE VASCULAR PLASMINOGEN ACTIVATOR AS SOURCE OF THE FIBRINOLYTIC POTENTIAL OBSERVED DURING CARDIOPULMONARY BYPASS.

Guadalupe BaAos*, Aurora de la PeAa and Raul Izaguirre. *Biochemistry and Hematology Departments. Instituto National Cardiologia "Ignacio Chavez". Tlalpan, 14080 Mexico, D.F.

de

(Received 13.51992; accepted in revised form 14.7.1992 by Editor E. Angles Cano) (Received by Executive Editorial Office 18.8.1992)

ABSTRACT Increased fibrinolytic activity is a well recognized constant finding observed during cardiopulmonary bypass (CPB). The purpose of the present work was to study and estimate the factors involved in the plasminogen activation and prekallikrein-kallikrein systems in a population of adult patients undergoing open heart surgery with activator activity CPB. Plasminogen determinations with a fibrinolytic method as well as plasminogen activation and prekallikrein-kallikrein determinations with synthetic substrates were carried out. Our results indicate that no active fibrimolysis but a fibrinolytic potential, similar to that observed in blood obtained after venous occlusion, can be demonstrated in circulating plasma during CPB. This fibrinolytic potential is related to the presence of from vascular plasminogen activator released endothelial cells by the CPB stimulus.

INTRODUCTION Cardiopulmonary bypass (CPB) involves subjecting the patient to a considerable degree of anticoagulation, contact of his blood with foreign surfaces and, frequently, transfusion of large volumes of all of which can induce serious donors' blood or plasma, affecting hematologic complications, especially hemostasis. shown that alterations in the Several authors (l-12) have of CPB in cardiac system can occur as a result fibrinolytic Cardiopulmonary words: JW Prekallikrein.

by-pass.

579

Fibrinolysis.

Kallikrein.

580

t-PA IN CARDIOPULMONARY

BYPASS

Vol. 67, No. 5

possible outcome could be a hemorrhagic and whose surgery syndrome; some of their results are suggestive of a possible plasminogen activation by the contact system (13-15) although in some cases plasminogen activator release from the vascular tree has been shown (2). In spite of a general agreement about the fact that a number of haematological abnormalities develop in association with CPB, the observations reported by the various authors do not coincide and some questions still require a clearer answer. The purpose of the present work, therefore, was to study the alterations in the levels of some of the factors involved in the fibrinolytic and the kallikrein-kinin systems in a population of adult patients undergoing open heart surgery with extracorporeal circulation (CPB).

MATERIAL m

METHODS

Patients. Clinical cases included 130 adult patients whose ages ranged from 17 to 72 years, except two patients who were 9 and 14 years old. The average age was 40 f 14 years. The pathological conditions that required surgical intervention in these patients were: Rheumatic cardiopathy: 66 %, Congenital cardiopathy: 17 %, Ischemic cardiopathy: 13 %, and others: 4 %. Perfusion methods. CPB was performed with a bubble oxygenator system (Shiley Laboratories, Irvine Calif.) and an American Optical or Olson roller pump machine. Balanced anesthesia was administered with halothane, nitric oxide, pentrane and oxygen. The priming of the system was carried out with Crystalloid in 5 % dextrose (Normosol M., Abbot Labs.) alone, or in a mixture with fresh frozen plasma. Hematocrit was maintained between 28-30 % during the procedure. The perfusion flow rate was 2.3-2.4 l/m2/min for children and 2.0-2.2 l/m2/min for adults. Blood contact with glass was avoided. The blood dilution rate was calculated according to Bond and Parsons (15). Anticoagulation was started with an initial heparin (Richter) dose of 3 mg/kg body weight and maintained the necessary with heparin administration,to keep the activated coagulation time between 400 to 600 sec., as monitored with a semi-automated apparatus (Hemochron, Int. Technidyne Corporations, Edison, N.J., U. S. A.). At the end of the procedure, heparin was neutralized with protamine sulphate (Roche) at the dose of 1.3 mg per mg circulating heparin. Sample collection. Coagulation and fibrinolytic studies were carried out on blood collected prior CPB(pre-oxygenation sample), during the intervention (trans oxygenation samples) and about 10 min after termination of CPB and administration of protamine (post-oxygenation sample). On each occasion the blood was divided into five samples, as follows: Sample a) 4.5 ml of blood in a tube containing 0.5 ml of 3.8 % sodium citrate solution. This sample provided plasma for plasminogen activator activity, plasminogen and plasmin determinations. Sample b) 2.5 ml of blood in a polycarbonate tube containing 0.25 ml of 3.8 % sodium citrate and 0.1 ml E -amino caproic acid 10 mM, final concentration; this sample was used for estimation of kallikrein and prekallikrein. Sample c) 2.5 ml of blood in a glass tube

Vol. 67, No. 5

t-PA IN CARDIOPULMONARY

BYPASS

581

containing 30 U of thrombin and trasylol (Bayer; 500 K.I.U./ml final concentration), for the trans-oxygenation samples, an additional 0.1 ml (1 mg) protamine sulphate. This serum sample was used to determine fibrin(ogen) degradation products. Sample ml of blood in a glass tube. This sample was centrifuged d) 5 and the serum was separated for total protein estimation. A small sample of blood was taken from a catheter for hematocrit estimates in a Cell-Dyn instrument. The samples were kept on ice from the moment of their collection, spun 10 min at 2 500 g in a refrigerated Heraeus minifuqe to separate plasma. The serum samples c) were incubated at 37 OC for at least 1 h after addition of 30 U thrombin (100 N.I.H. U/ml) to ensure complete clotting. Chemicals. E -amino caproic acid (Lederle); bovine thrombin (Hoffmann-La Roche, Switzerland); human fibrinogen (A.B. Kabi, Sweden); plasminogen-free fibrinogen was prepared by affinity chromatography on lysine-sepharose 4B (16); trasylol (Bayer); Chromogenic substrates, S-2251 (for plasmin and plasminogen), S(for preand 2322 (for plasminogen activator) and S-2302 kallikrein) from A.B. Kabi, Stockholm, Sweden. All the other chemicals were of analytical grade. Anti-fibrinogen antiserum was obtained from goats immunized with purified plasminogencfree fibrinogen and adsorbed with normal serum. Fibrinolvtic assavs. activator was Plasminogen activator activity. Plasminogen obtained from plasma by precipitating it in the euglobulin fraction as follows: titrated plasma was diluted 1:20 with icecold distilled water and the pH adjusted to 5.9 with 0.5 % (v/v) acetic acid, the mixture was allowed to rest for 30 min at 4 OC and then centrifuged at 2 000 g during 15 min at 4 OC. The discarded the inhibitors was and supernatant containing precipitate redissolved in a volume of PBS ( 10 mM Na2HP04, 5 mM KH2PC4, 5 mM KCl, 137 mM NaCl, pH 7.4) equal to the original plasma volume. This final euglobulin suspension was used for the determination of PA (plasminogen activator) by the euglobulin lysis time (ELT). Duplicate 200 ~1 of suspension were clotted with 100 ~1 of bovine thrombin ( 20 N.I.H. U/ml) and incubated at 37 oc. The ELT was measured as the time elapsed between clotting after addition of thrombin and complete clot lysis. estimates of plasmin with a final point Plasmin. Duplicate method, using the synthetic substrate S-2251, were carried out in suspension as follows: 100 1.11 of euglobulin the euglobulin suspension were incubated with 700 ~1 of PBS pH 7.4 at 37 'C for 3 min, 100 ~1 of 2 mM S-2251 substrate were then added and the 100 1.11 of 30 % 180 set with stopped at exactly reaction trichloroacetic acid (TCA). The proteins were spun down and the absorbance at 405 nm in the supernatant was determined. A blank was run using buffer instead of euglobulins and the absorbance reading from this blank was subtracted from all corresponding problem readings. Similar plasmin estimates and ELT were carried out in euglobulin suspensions obtained from normal volunteers occlusion venous after conditions and basal under (sphygmomanometer midway between systolic and diastolic pressure for 10 min) stimulation (17). This was done in order to compare the plasminogen activator activity present in plasma during both, CPB and venous occlusion stimuli.

582

t-PA IN CARDIOPULMONARY

BYPASS

Vol. 67, No. 5

The effect of temperature upon enzyme activity was investigated in 7 euglobulin suspensions, by incubating the samples at 37 OC, during an hour previous to the assay for plasmin.Plasminogen activator was measured in duplicate, with the final point method, using the synthetic substrate S-2322, following the procedure recommended by the manufacturer (Kabi).Plasminogen was measured the method substrate S-2251, following with the chromogenic (18).Plasma kallikrein and Friberger et al described by prekallikrein were directly measured in plasma, using a mixture of 15 pl plasma, 0.8 ml 50 mM Tris-HCl buffer pH 7.8 and 0.1 ml 2 After 3 min incubation at 37 OC, the mM S-2302 substrate. reaction was stopped with 0.1 ml TCA. Duplicates were run for all plasma tests. Prekallikrein was estimated by activation of 15 1.11 of plasma diluted 1:4 with Tris-HCl buffer pH 7.8, mixed with an equal volume of kaolin suspension ( 10 mg/ml). After activating the plasma for 7 min at 4 OC, aliquots of 15 ~1 were then allowed to react as for kallikrein measurements. A plasma sample from a healthy donor was tested in a run parallel to the problems as of which was well as a blank without plasma, the reading degradation products subtracted from all problems.Fibrin(ogen) (FDP). FDPs were measured by the technique of inhibition of agglutination (19). Red cells group 0 Rhesus positive from a donor healthy and a serum-adsorbed goat antiserum against fibrinogen, prepared in our laboratory, were used.Serum proteins were measured by means of an automatic Biuret method.

RESULTS Plasminogen activator (PA) activity. Table 1 shows the PA activity of pre-, trans- and post-oxygenation samples from 117 patients as measured by the euglobulin clot lysis method. In all pre-oxygenation samples ELT exceeded 6 hours, except in 8 cases with an ELT ranging from 30 minutes to 2 h 45 min; the average for these was 1 h 45 min. All trans-oxygenation ELT were shorter than 2 h, except 10 cases which averaged 2 h 47 min, the range was 15 min to 1 h 45 min, with an average of 45 min + . All postoxygenation ELT were longer than 6 h, except 3 ( 2 h, 2 h 20 min and 2 h 45 min ). Since some blood samples presented certain degree of hemolysis, a test was carried out to estimate the effect of blood lysates upon ELT. After centrifugation, an aliquot of packed red cells was measured and lysis was induced by addition of distilled water. Different amounts of this lysate were added to samples of plasma from normal subjects, taken under basal conditions and after venous occlusion. No significant alteration was observed (results not shown). Plasmin. As shown in Table 1, plasmin was undetected in the euglobulin suspension of 26 pre-oxygenation samples. Although there was a significant difference between basal and transoxygenation absorbances (p < 0.02), the actual levels are small, compared with those found in the lysates . Plasmin was measured

Vol. 67, No. 5

t-PA IN CARDIOPULMONARY

BYPASS

583

TABLE 1 FIBRINOLYTIC PARAMETERS IN PRE, TRANS AND POST EXTRACORPOREAL CIRCULATION SAMPLES (Mean i S.D.) PRE

TRANS 30

Euglobulin lysis time(minutes) n=117

(min) 45

341+56

POST 60

+50 + 23

Plasmin Abs 405 n= 26

0.047*0.02

Activator Abs 405 n=23

0.06f0.03

0.065+0.04

Kallikrein Abs 405 n=22

0.06+0.05

0.07+0.02

Prekallikrein Abs 405 n= 16

0.6620.27

O-498+0.28

Plasminogen % n= 6

99+8.7

F.D.P. pg/ml n=34

2Ok1.7

Proteins g/d1 n= 11

5.8kO.5

Hematocrit % n= 11

36.9f4.8

* p < 0.02, ** p < 0.01,

*0.063+0.03 LYSATES: 0.49 + 0.3

+ 73+8

** 7729

** 751f-7.6

40 f 36

>180

0.06?0.03

0.071to.05

0.065f0.03

0.63kO.24

** 77211

23 f 4

t4.1+0.5

*4.5+0.6

**31+4.7**30.5+2.9**30.9+2.2

35.824

f4.220.9

t4.2a0.6

fp < 0.001

in 23 trans lysates, i.e. after coagulation and clot lysis. The levels of plasmin in these samples were high, as expected, and the difference between these and basal levels, as determined in the euglobulin suspension, was highly significant ( p < 0.001 ). The results for the pre- and trans- oxygenation samples were similar to those observed in euglobulin suspensions obtained respectively; a similar occlusion venous after before and relationship was obtained between the lysates of post-occlusion

584

t-PA IN CARDIOPULMONARY

BYPASS

Vol. 67, No. 5

and trans-oxygenation samples ( Table 1 ). Incubated samples had slightly higher levels of plasmin, but the differences were not statistically significant (data not shown). Plasma levels of plasminogen were found to decrease significantly during the trans and post CPB periods, even after corrected for hemodilution, but remained within normal limits (150-75 %). Kallikrein and prekallikrein ( Table 1 ). Plasma kallikrein was The levels of transand postestimated in 23 patients. oxygenation kallikrein were not significantly different from basal values ( p > 0.1 ). Plasma prekallikrein was estimated in 16 patients. The results showed a significant decrease in the level of prekallikrein during CPB in relation to basal conditions ( p < 0.001 ), as well as in relation to normal levels of healthy donors. Control series of assays for kallikrein and prekallikrein were run using a pool of normal plasmas, from which curves were constructed using serial plasma dilutions, i.e. 75 %, 50 %, 25 % upon the level of The effect of heparin and undiluted. kallikrein, both at 37 OC and 4 OC, was also investigated. It was found to have no effect at the concentrations used either in vitro or in vivo, although at higher concentrations heparin did increase the levels of plasma kallikrein in vitro ( results not shown). degradation products (FDP). Preand postFibrin(ogen) oxygenation samples had normal FDP levels for the method we used ( lo-20 pg/ml ) and some increases were detected during CPB. Nevertheless, trans and post levels were scattered over a wide range and the variations became non statistically significant. Hematocrit and serum proteins decreased on account of the hemodilution concomitant Serum with the surgical procedure. proteins were significantly reduced, even after the values were corrected according to hematocrit percentage.

DISCUSSION During CPB, several elements may contribute to the activation of physiological processes and lead to a lengthening in coagulation tests, as well asableeding (13,20-23). The fibrinolytic system is affected somehow, and its activation continues to be a constant finding, often described in open heart surgery (5-8, 10-12). Since there is some controversy as to the nature and quantity of alterations during CPB, we decided to further investigate the activating mechanism of fibrinolysis. No significant levels of circulating plasmin or kallikrein in the trans- or post-oxygenation samples were found. The presence of high levels of PA activity and a decrease in plasma prekallikrein were evident in the trans-oxygenation samples. Pre- and post-oxygenation samples showed normal levels of prekallikrein and plasminogen activator activity. The possibility exists that the presence of plasmin-a2 antiplasmin complexes interferes with the detection of free plasmin in plasma, but the very low levels of FDP indicated that, in fact, no significant amount of plasmin had been generated. Moreover, the near absence of free plasmin in the euglobulin suspension, which did not contain inhibitors, indicated clearly

Vol. 67, No. 5

t-PA IN CARDIOPULMONARY

BYPASS

585

that no plasminogen activation occurred in vitro. In contrast, high levels of free plasmin in the lysates of clotted euglobulins were found, as well as a shortening of ELT, during oxygenation. These observations indicate that plasminogen activation occurs only in clotted euglobulins and that no active fibrinolysis, but a fibrinolytic potential, similar to that observed in euglobulins from post-occlusion venous blood is present in circulating plasma during CPB (17, 24). The activator support of this fibrinolytic potential is a fibrin-dependent PA (25) and not a urokinase-like PA (u-PA). The development of PA activity only after the clotting of euglobulins, i.e., in the presence of fibrin, is characteristic of tissue-type PA (t-PA). These activators are capable of plasminogen activation only when adsorbed to a fibrin surface and, therefore, their presence in plasma or euglobulins does not ensure any plasminogen activation if coagulation is inhibited. Heparinization, as that induced in our patients, did effectively inhibit coagulation and, consequently, the formation of any traces of fibrin. This is probably the reason why no detectable fibrinolysis could be found in the trans-oxygenation samples. We can assume that the tissue-type PA present in blood during CPB is released from the vascular tree during the ECC stimulation. A similar conclusion has been arrived at by Stibbe et al (26). The vascular PA provides fibrinolytic potential, but the triggcering of fibrin. of fibrinolysis depends only on the generation Effective heparinization shunts the generation of fibrin in the vascular tree and, therefore, active fibrinolysis. Otherwise, fibrin constantly generated at the wound level can be lysed by traces of vascular PA diffusing into the interstitial space; this can interfere with the initial phases of wound healing, leading to oozing and hemorrhage. Once the stimulus disappears, the release of PA by the vascular wall ceases, and due to its very short half life, the fibrinolytic potential returns to the normal level in the post-oxygenation sample. CPB involves induced hypothermia, followed by increase in body temperature to the normal level, which could favour an activation reactions, otherwise partially inhibited by of enzymatic increase in cooling. This was investigated and no significant activity was found in heated samples. plasmin The small, but not significant, increase in FDPs could be due to some activation of the contact phase. lead to plasminogen The activation of prekallikrein can activation (27,28); we did not, however, find any significant level of free plasmin or generation of plasmin activity in the trans-oxygenation sample, and this is against the possibility of plasminogen activation by kallikrein. In conclusion, the PA activity we have demonstrated in the transoxygenation sample behaves as the plasma PA found after venous occlusion stimulation. This is a tissue type PA, which, as the PA present in a variety of tissues is of endothelial cell origin.

586

t-PA IN CARDIOPULMONARY

BYPASS

Vol. 67, No. 5

ACKNOWLEDGEMENTS We are indebted to Prof. F. Quijano, former Head of the Dept. of Surgery, Instituto National de Cardiologla, for allowing us to carry out this study, and to the dynamic perfusion team of the for their efficient collaboration in the Dept. of Surgery, collection of blood samples. Mrs. Eva Hilda Gonzalez, Head of the the provided facilities for laboratory Chemical Pathology estimation of plasma proteins.

REFERENCES l-

VON KAULLA KN, SWAN H. : Clotting deviations in man during cardiac bypass: fibrinolysis and circulating anticoagulant. J. Thor. Surs. 36: 519-533, 1958.

2-

GANS H, LILLEHEI CW, KRIVIT W. : Problems in hemostasis during open-heart surgery; I- On the release of plasminogen Ann. Surs. 154: 915-924, 1961. activator.

3-

ANDERSEN MN, extracorporeal

4-

KEVY SV, GLICKMAN RM, BERNHARD WF, DIAMOND LK, GROSS RE. : The pathogenesis and control of the hemorrhagic defect in open heart surgery. Surs. Gvn. Obstet. 123: 313-318, 1966.

5-

after DERMAN UM, RAND PW, BARKER N. : Fibrinolysis cardiopulmonary bypass and its relationship to fibrinogen. J. Thorac. Cardiovasc. Sure. 51 : 223-228, 1966.

6-

GANS H SUBRAMANIAN V, CASTANEDA JS, LILLEHEI CW.: theoretical And practical (clinical) considerations concerning proteolytic enzymes and their inhibitors with particular reference to changes in the plasminogen-plasmin system observed during assisted circulation in man. Ann. N. Y. Acad. Sci. 146: 721-736, 1968.

7-

STIBBE J, KLUFT C, BROMMER EJP, GOMES M, DE JONG DS, NAUTA J . : Enhanced fibrinolytic activity during cardiopulmonary bypass in open-heart surgery in man is caused by extrinsic (tissue-type) plasminogen activator. Eur. J. Clin. Invest. 14: 375-382, 1984.

8-

HOLLOWAY DS, SUMMARIA L, SANDESARA J, VAGHER JP, ALEXANDER JC, CAPRINI JA. : Decreased platelet number and function and increased fibrinolysis contribute to postoperative bleeding in cardiopulmonary bypass patients. Thromb. Haemost. 59 (1): 62-67, 1988.

9-

VAN OEVEREN W, HARDER MP, ROOZENDAAL KJ, EIJSMAN L, WILDEVUUR CRH. : Aprotinin protects platelets against the initial effect of cardiopulmonary bypass. J. Thorac. Cardiovasc. Surs. 99: 788-797, 1990.

MENDELOW M. : Fibrinolysis during and after circulation. Suroerv. 84: 649-657, 1963.

Vol. 67, No. 5

t-PA IN CARDIOPULMONARY

587

BYPASS

lo-

GRAM J, JANETZKO T, JESPERSEN J, BRUHN HD. : Enhanced effective fibrinolysis following the neutralization of heparin in open heart surgery increases the risk of postsurgical bleeding. Thromb. Haemost. 63 (2): 241-245, 1990.

ll-

GIULIANI R, SWARCER E, MARTINEZ-A E, PALUMBO G. : Fibrinduring extracorporeal dependent fibrinolytic activity circulation. Thromb. Res. 61: 369-373, 1991.

12-

PARAMO JA, RIFON J, LLORENS R, CASARES J, PALOMA MJ, ROCHA E. : Intra- and postoperative fibrinolysis in patients undergoing cardiopulmonary bypass surgery. Haemostasis. 21: 58-64,

1991.

13-

EISEN V. : Fibrinolysis and formation of biologically polypeptides. Brit. Med. Bull. 20: 205-209, 1964.

14-

WIEGERSHAUSEN B, HENNIGHAUSEN G, LANGE G. : Plasma kininogen influence of a in extracorporeal circulation and the protease inhibitor from bovine lung. Exnerientia. 26 : 11181119,

active

1970.

15-

and plasma proteins BOND AG, PARSONS RS.: Haemodilution during anaesthesia. Brit. J. Anaesth 42: 1113, 1970.

16-

DEUTSCH DG, MERTZ ET. : Plasminogen: purification from human plasma by affinity chromatography. Science. 170: 1095-1096, 1970.

17-

CLARKE RZ, ORANDI A, CLIFTON EE.: Induction of fibrinolysis by venous obstruction. Ansiolosv. 11: 369-370, 1960.

18-

FRIBERGER P,KNOS M, AURELL L, GUSTAVSSON S, CLAESON G. Methods for the determination of plasmin, antiplasmin and plasminogen by means of the substrate S-2251. Haemostasis. 7: 138-145, 1978.

19-

MERSKEY C, KLEINER CJ, JOHNSON AJ. : Quantitative estimation of split products of fibrinogen in human serum. Relation to diagnosis and treatment. Blood. 28: 1-16, 1966.

20-

BACHMAN R, MC KENNA R, COLE ER, MAIAFI HJ. mechanism after open-heart extracorporeal Thorac.Cardiovasc. Surs. 70: 76-85, 1975.

: The hemostatic

circulation.

21-

BICK RL, ARBEGAUST N, CRAWFORD L, HOLTERMANN M, ADAMS induced defects Hemostatic SCHMALHORST WR. : cardiopulmonary bypass. Vast. Surs. 9: 228-243, 1975.

22-

GRALNICK HR, FISCHER RD. : The hemostatic response heart operations. J. Thorac. Cardiovasc. Sure. 61:

J. T, by

to open909-915,

1971. 23-

TRIMBLE AS, HERST R, GRADY M, CROOKSTON JH. : Blood loss in open- heart surgery. Arch. Surserv. 93: 323-326, 1966.

588

t-PA IN CARDIOPULMONARY

BYPASS

Vol. 67, No. 5

24-

ANGLES-CAN0 E, SULTAN Y. : Human plasminogen activator (PA) activity from post-occlusion venous blood: requirement of fibrin to generate plasmin. Haemostasis. 11 (suppl. 1): 84, 1982.

25-

ANGLES-CAN0 E, SULTAN Y, BERNARD J. l'activateur du plasminogene d'origine Acad. Sci. PARIS. 289: 485-487, 1979.

26-

KLUFT C, BROMMER EJP, GOMES M, DE JONG DS, NAUTA STIBBE J plasminogen Secr&ion of extrinsic (tissue type) J. : enhanced cells causes endothelial the activator by bypass. activity cardiopulmonary fibrinolytic during Hemostasis. 11 (suppl. 1) : 32, 1982.

27-

MANDLE R JR, KAPLAN AP. : Hageman factor substrates: human plasma prekallikrein: mechanism of activation by Hageman dependent factor participation in Hageman factor and fibrinolysis. J. Biol. Chem. 252 : 6097-6104, 1977.

28-

EC. KLUFT C, TRUMPI-KALSHOVEN MM, JIE AFH, VELDHUYZEN-STOLK : Factor XII-dependent fibrinolysis: a double function of plasma kallikrein and the occurrence of a previously undescribed factor XII- and kallikrein-dependent plasminogen proactivator. Thromb. Haemostas. 41: 756-773, 1979.

: Purification vasculaire. C.

de R.