The Role of the Fibrinolytic Joseph

G. Vermylen

System

and Dalton A. F. Chamone

A LTHOUGH can lyse

the discovery that blood clots spontaneously goes back to the 18th and 19th centuries’,’ the detailed analysis of the reactions involved in fibrinolysis started after 1930, when it became apparent that media from cultures of fl-hemolytic streptococci contained a factor that was capable of dissolving fibrin.3 This factor was called “streptococcal fibrinolysin.” Milstone4 found that a human globulin, “lytic factor,” is an essential cofactor for streptococcal fibrinolysin. Christensen5 further studied this reaction and concluded that lytic factor is an inactive proteolytic enzyme, which is activated by streptococcal fibrinolysin. He could then introduce a more rational nomenclature, replacing “lytic factor” by plasminogen and “streptococcal fibrinolysin” by streptokinase.6 Thus, streptokinase was the first recognized plasminogen activator, although the mechanism by which activation occurs is extremely complex (see Dr. Castellino’s article in this symposium). In 1947, Astrup and Permin’ showed that the fibrinolytic activity of human and animal tissues8,9 is due to the activation of plasminogen. Various tissues contain plasminogen activator, at least part of which originates in the endothelial cells of the tissue’s blood vessels.” Plasminogen activator is also present in blood,” and again, very likely originates, at least in part, from the endothelial cells (see below). In 1958, Astrup12 introduced the concept of the hemostatic balance: fibrin is formed by the process of blood clotting and is removed by the process of fibrinolysis. The balance between the effects of these two processes therefore regulates the amount of fibrin present at any time. According to this concept, thrombosis would result from enhanced fibrin deposition or from decreased fibrin removal. A discussion of the role of the fibrinolytic system in thromboembolism thus leads to an evaluation of the concept of the hemostatic balance, and more particularly, to the examination of the following points: (A) Is there evidence of increased fibrinolytic activity in situations with enhanced fibrin formation? In other words, can the human organism maintain the hemostatic balance and

Progress

in Cardiovascular

Diseases,

Vol. XXI,

in Thromboembolism

No. 4 (January/February),

prevent thrombosis even when the clotting system has been triggered? (B) Do patients with thrombosis exist in whom the only detectable abnormality is a defective fibrinolytic activity? Before starting a detailed discussion of the hemostatic balance in normal and pathologic conditions, let us consider very briefly the present concept of physiologic fibrinolysis. PHYSIOLOGIC

FIBRINOLYSIS’3,‘4

Plasmin is a relatively nonspecific protease, capable of digesting a whole number of protein substrates. Therefore, a major question to be asked is, what are the mechanisms involved in limiting plasmin’s in vivo activity rather specifically to fibrin? Like several reactions of coagulation, fibrinolysis is essentially a surface-linked phenomenon. Tissue activator,” and very likely also plasma plasminogen activator,16 have a very high affinity for fibrin. Native NH,-terminal glutamic acid plasminogen, and to a larger extent, autocatalytically altered NH2-terminal lysine plasminogen, also have affinity for a fibrin surface.” The attachment of plasminogen to fibrin occurs via one (in NH,-terminal glutamic acid plasminogen)” or two (in NH,-terminal lysine plasminogen)” binding sites, which are also involved in the binding of plasminogen to insolubilized lysine and have therefore been called the lysinebinding sites. Lysine or related aminoacids, epsilon aminocaproic acid (6-aminohexanoic acid) or tranexamic acid, can therefore occupy these sites and thus prevent the binding of plasmiFrom the Center for Department of Medical

Thrombosis Research.

and Yascular University

Campus Gasthuisberg. Leuven, Belgium. D.A.F.C. is a Research Fellow supported Cao de Amparo b Pesquisa do Estado Brazil. Reprint Vermylen, Research, Leuven,

requests should be addressed M.D.. Center for Thrombosis Department of Medical Research, Campus Gasthuisberg. Herestraat

Leuven, Belgium. c: 1979 by Grune & Stratton, 0033~620/79/2104-0003$02.00/O

1979

Research, of Leuven.

by the de Sire

FundaPauio.

to Joseph G. and Vascular University (of 49. B-3000.

Inc.

255

VERMYLEN

256

nogen to fibrin. This phenomenon explains the marked antifibrinolytic properties of these aminoacids. Recently, a new physiologic inhibitor of fibrinolysis, antiplasmin, has been described.*’ ” This a,-globulin is the main immediate antiplasmin in human plasma and is present at a concentration of 1 PM (plasminogen concentration, 1.5 @I). Therefore, it is only when more than 70% of the plasma plasminogen is rapidly converted to plasmin, that free plasmin activity is generated, leading to an overt plasma proteolytic state and important fibrinogenolysis. This free proteolytic enzyme is then progressively captured by cu,-macroglobulin. Antiplasmin binds to plasmin via the lysinebinding sites of the latter.“.“3 Antiplasmin therefore has little affinity for plasmin, if the latter protein is already bound by its lysine-binding sites to fibrin. On the other hand, plasmin generated away from a fibrin surface would be instantaneously neutralized by antiplasmin. In conclusion, the nonspecific proteolytic activity of plasmin would essentially be limited to fibrin in vivo, in view of (A) the specific adsorption of activator and of plasminogen onto the fibrin surface resulting in local generation of plasmin and (B) the fact that plasmin, adsorbed to fibrin, in contrast to plasmin in the fluid phase, largely escapes the action of antiplasmin. THE

HEMOSTATIC

BALANCE

IN THE

RESTING

CONDITION

Is There Evidence for Ongoing Deposition or Formation in the Normal Subject?

Fibrin

The concept that the endothelium is covered by a thin layer of fibrin, acting both as a variable barrier to diffusion and as an anticoagulant surface was first postulated by Nolf.24 Present evidence for this concept, however, is virtually nonexistent. The lack of continuous fibrin deposition on the endothelium, however, does not exclude the possibility of continuous ongoing fibrin formation. Soluble fibrin oligomers would then be removed from the circulation by the reticuloendothelial system before they had become sufficiently large to become insoluble and be depos-

AND

CHAMONE

ited on the vesselwall. Alternatively, they could be lysed as rapidly as they formed. There arc. however, several arguments against continuous fibrin formation being an important physiologic process. The Measurement of Fibrinogen or Prothrombin Survival in Normal tndividuals During Heparin Administration or in Patients With a Major Congenital Coagulation Disturbance The half-disappearance time of radiolabeled fibrinogen in normal man is 4.14 * 0.56 days.“ The factors involved in the catabolism of tibrinogen are incompletely understood. However, it is well established that in patients with enhanced thrombin-mediated fibrinogen activation (the “disseminated intravascular coagulation” syndrome, seebelow), the half-disappearance time of this protein is considerably shortened. It could not be excluded a priori that also in normal subjects at least part of the physiologic fibrinogen catabolism would occur via transformation to fibrin monomer and subsequent removal from the circulation. For this reason, Collen et al.*’ measured the half-disappearance time of fibrinogen in five healthy subjects before and during administration of a “therapeutic” dose of heparin. If thrombin-mediated fibrinogen activation played a significant part in the physiologic catabolism of fibrinogen, interruption of this pathway by heparin should result in a measurable prolongation of fibrinogen survival. In fact, no such prolongation was noted. The possibleexistence of continued activation of the clotting system was also studied a stage earlier, at the level of prothrombin. Shapiro and Martinez’(’ found the half-disappearance time of radiolabeled prothrombin to be 2.8 1 + 0.5 1 days in normal subjects. Results in four hemophilic patients did not differ significantly from normal. The authors concluded that continuous physiologic activation of the blood coagulation mechanism plays only a small part, if any, in the normal catabolism of prothrombin. The Measurement of Circulating Fibrinopeptide A in Normal Individuals Upon proteolysis of fibrinogen by thrombin. there is an initial rapid releaseof fibrinopeptide

FIBRINOLYTIC

SYSTEM

IN THROMBOEMBOLISM

257

A. Nossel and coworkers have developed a specific radioimmunoassay for this peptide.” They found that plasma fibrinopeptide A levels in 30 normal men were below 2 “g/ml, with a mean of 0.5 rig/ml. Based on a 3-5 min halfdisappearance time, one may calculate that when the fibrinopeptide A level is 0.5 rig/ml, this would amount to thrombin proteolysis of 40-67 mg of fibrinogen in 24 hr, which would represent about 2%-3% of the normal fibrinogen catabolism. This finding again indicates that in normal individuals, thrombin proteolysis is not a major determinant of fibrinogen catabolism. This section may therefore be concluded by stating that under normal conditions, systemic intravascular fibrin deposition or formation must be either nonexistent or extremely limited. This does not necessarily exclude the possibility that localized fibrin deposition may be an everyday occurrence, an expression of the “wear and tear” of life. Is There Evidence for Ongoing Fibrinogenolysis in the Normal

Fibrinolysis Subject?

or

There are several points in this regard that deserve discussion:the presence of plasminogen activators or fibrinogen derivatives in blood and the turnover of fibrinogen and plasminogen. Plasminogen Activators in Blood Within human blood, several activators of plasminogen may be present or formed. One is probably continuously secreted in small quantities by the endothelial cells. Secretion is probably enhanced, since increased levels are found, in association with mental stress and anxiety,” severe muscular exercise or adrenaline” or venous occlusion.” Blood plasminogen activator may also be formed by an indirect mechanism after the activation of factor XII (Hageman factor). There is controversy concerning the precise steps and components involved in the factor-XII-dependent pathway.” Also, the physiologic significance of this mechanism is completely unknown. Human blood, even drawn in the resting state without previous venous occlusion and despite avoidance of surfaces that activate factor XII, contains a small amount of plasminogen activator. A potential for continuous plasminogen

activation in vivo may therefore exist. But is there really ongoing plasminogen activation in the normal individual? The plasminogen activator of plasma is very labile; in addition, a specific inhibitor of plasminogen activator also exists in plasma.32Generated plasmin would very rapidly be bound by antiplasmin. Therefore, despite the fact that plasminogen activator can be demonstrated, there is somedoubt as to whether there is ongoing fibrin0 (geno) lysis in the resting state. Plasma plasminogen activator is usually measuredin the euglobulin fraction, which also contains plasminogen and fibrinogen, but is devoid of antiplasmin and the inhibitor of plasminogen activator. Fibrinogen Derivatives in Blood Studies of fibrinogen derivatives in normal blood have provided some information. Measurement of jibrinogen-related material in serum. Using different approaches (sensitive immunologic assays such as Merskey’s tanned red cell hemagglutination inhibition immunoassay33or the staphylococcal clumping testJ4), it usually is possibleto measure fibrinogen-related material in normal serum. Since these nonclottable fibrinogen derivatives most probably represent fibrin(ogen) degradation products, their presence may reflect ongoing fibrin(ogen) olysis in vivo. However, this may not be considered a conclusive argument for the following reasons: (1) The fibrinogen-related material may have been formed during the preparation of serum; by taking very strict precautions, Merskey3’ could considerably reduce the amount of fibrinogenrelated material in serum, indicating that it mainly originated from in vitro fibrin(ogen) degradation (2) Other systems besidesthe plasminogenplasmin system could be responsible for the degradation of fibrinogen (e.g., leukocytic proteases). Measurement of degradation of the COOHterminal part of the Aa chains of jibrinogen. When fibrinogen is exposed to plasmin, the first degradation to occur is located in the COOH-terminal part of the Aa chains of this molecule. Evidence for ongoing generation of plasmin in the circulation would thus be

VERMYLEN

258

provided by either (1) demonstration that a significant portion of the circulating fibrinogen has a partially degraded Aa chain, or (2) demonstration of the removed Aa chain fragment in the circulation. (1) Does a significant portion of the circulating fibrinogen have a partially degraded Aa chain? Mosesson found that human fibrinogen, prepared from fresh plasma by ethanol precipitation, consists of over 20% of high solubility material;36 this material contains degraded Aa chains, similar to those obtained by plasmic digestion of low solubility fibrinogen.37 In part, on the basis of these findings, he concluded that, under physiologic circumstances, fibrinogenolysis is a major catabolic pathway for fibrinogen in man.38 In separate work, he suggested that this continuous fibrinogen degradation may be a consequence of residual proteolytic activity of the plasmin-a,-macroglobulin complex.39 However, doubt has been expressed about the importance of this pathway. An alternative suggestion has been made, namely that the fibrinogen heterogeneity is due to partial proteolytic degradation that may have occurred during ethanol precipitation.40 Two arguments were put forward to support this suggestion. First the observation of Murano et aL4’ that when glycine was replaced by t-amino caproic acid in the fibrinogen purification method of Blomback and Blomback, also using ethanol precipitation, the yield of degraded ACYchains in the purified material was reduced, while that of intact Aa chains was increased. Secondly, is the observation of the presence of significant amounts of partially degraded plasminogen with NH,-terminal lysine in Cohn fraction III. Since this degradation occurs via plasmin generation,42 it becomes very likely that plasmin is formed during ethanol fractionation of plasma. To further quantitate the importance of the fibrinogenolytic pathway for fibrinogen catabolism, Semeraro et a1.43studied the chain composition of human fibrinogen, purified in a single step procedure with high yield by affinity chromatography. They observed that such fibrinogen contains significantly less degraded Aa chains which, however, still amount to about 11% of the total. These authors therefore had to conclude

AND

CHAMONE

that the residual ALY chain degradation they found in human fibrinogen prepared from freshly drawn blood, is at least in part due to in vivo fibrinogenolysis. Alkjaersig et al.44 analyzed the plasmas from 52 normal subjects by agarose gel chromatography and found these to contain 25.8% -t 6.4% of slightly digested material (fibrinogen first derivative). (2) Can the removed Aa chain fragment be demonstrated in the circulation? Harfenist et a1.45reported in abstract form on a radioimmunoassay for the major ACY chain fragment. Their preliminary studies showed that normal serum levels are in the range of 50-200 pmole/ml. As the half-disappearance time of this fragment is not known, it is not yet possible to quantitate on this basis the relative amount of fibrinogen catabolized via this pathway. Turnover

of Fibrinogen

and Plasminogen

Administration of a thrombolytic agent results in a marked shortening of the survival of fibrinogen or plasminogen.4hIf fibrinogenolysis were an important pathway in the normal catabolism of fibrinogen, it could be anticipated that administration of an antifibrinolytic agent such as 6-amino-hexanoic acid (t-amino caproic acid) would prolong the survival of fibrinogen. In fact, no prolongation of fibrinogen survival was observed in four of five normal subjects during inhibition of the fibrinolytic system.“5 In one subject, however, a prolongation occurred, which may be explained by the occasional activation of the fibrinolytic system in physiologic conditions by mechanismssuch as severe muscular exercise (see below) or mental stress and anxiety. This result is in agreement with the study of HarLJ’ who also noted a prolongation of the survival time of labeled fibrinogen in some normal subjects during administration of t-amino caproic acid. The half-disappearance time of radiolabeled plasminogenwas 2.2 1 + 0.29 days in 12 normal subjects4’ If plasminogen activation were a significant pathway in its physiologic catabolism, it could be anticipated that the plasminogen half-life would be prolonged in patients with depressedfibrinolytic activity. In a patient with Behcet’s syndrome and very low fibrinolytic activity (seebelow), a normal plasminogen half-

FIBRINOLYTIC

SYSTEM

IN THROMBOEMBOLISM

259

life was found.48 This suggests that the primary pathway of plasminogen catabolism is metabolic rather than fibrinolytic. To rule out the remote possibility of significant plasminogen consumption secondary to in vivo coagulation, the effect of heparin anticoagulation on plasminogen turnover was evaluated in one normal subject. The plasma radioactivity disappearance rate remained uchanged.48 The turnover of radiolabeled fibrinogen and plasminogen was also studied in untrained healthy subjects before and during strenuous physical exercise on a bicycle ergometer (repeated 4 times/day for 2 days).49 A significantly increased catabolism of fibrinogen and plasminogen was observed. The extent of Aa! chain degradation of fibrinogen in the plasma was quantitated before and 2 hr after exercise, and a significant increase in degraded Aa chains was found in the postexercise samples. These data support the concept that plasminogen activation and plasmin-induced fibrinogen degradation occur to some extent in man following strenuous physical exercise. In conclusion, there is quite considerable evidence that a limited systemic fibrinogenolysis is going on in healthy individuals. This process can be accelerated by simple physiologic procedures, such as strenuous physical exercise. THE

HEMOSTATIC

“DISSEMINATED

BALANCE

IN

INTRAVASCULAR

COAGULATION”

“Disseminated intravascular coagulation” refers to clinical conditions with ongoing fibrinogen-to-fibrin conversion, excluding local thrombus formation. The name “disseminated intravascular coagulation” has not been unanimously approved, since the fibrinogen-to-fibrin transformation may not always be truly disseminated (e.g., mainly localized in a giant hemangioma), nor only intravascular (e.g., the retroplacental hematoma in premature separation of the placenta), nor really coagulation (e.g., the enhanced fibrinogen-to-fibrin conversion produced by snake-venom enzymes).50 However, agreement has not been reached on a more appropriate name. The disseminated intravascular coagulation syndrome may result in thrombosis (from excessive fibrin deposition) or in hemorrhage (from

excessive depletion of thrombin-sensitive clotting components). Thrombosis in critical organs occurs mainly when the disseminated intravascular coagulation syndrome is associated with shock (e.g., renal cortical necrosis and Sheehan’s syndrome following premature separation of the placenta), thus emphasizing the important role of blood flow for the prevention of fibrin deposition and for the removal of activated clotting components by the reticuloendothelial system.” If a hemostatic balance exists, it follows that increased fibrinogen-to-fibrin conversion should be counterbalanced by increased plasminogento-plasmin transformation in order to prevent intravascular fibrin deposition. Is intravascular coagulation associated with enhanced plasminogen-to-plasmin conversion? There are several indirect arguments for plasminogen activation in disseminated intravascular coagulation. Recently, direct proof that this phenomenon occurs could be obtained. Let us first briefly review the indirect arguments: (1) The appearance of large quantities of incoagulable circulation.

Jibrin(ogen)-related

antigen

in the

Chromatographic analysis of this materia152.53 reveals that it consistsof fibrinogen derivatives both larger and smaller than the parent molecule. The larger derivatives consist of complexes of fibrin monomer with fibrin degradation products; the smaller derivatives have the elution position of plasmin-induced fibrinogen fragments and therefore presumably reflect plasmin activity in vivo. However, the possibility that they may be derived from digestion by another proteolytic enzyme cannot be unequivocally excluded.

(2) The reduction of the level of circulating plasminogen and the more rapid disappearance

of radiolabeled plasminogenfrom the circulation in disseminated intravascular coagulation46.‘J certainly suggestan increased catabolism of this protein, most probably as a consequenceof its activation. However, an alternative explanation could be that unactivated plasminogenis cleared after adsorption onto fibrin evolving during intravascular coagulation. A disturbing feature with regard to enhanced plasminogen-to-plasmin conversion in disseminated intravascular coagulation is the lack of increased levels of plasminogen activator or of plasmin when these activities are measuredon a

260

standard substrate. The short euglobulin clot lysis time, which has been found on occasion, seems to reflect more the low amount of available substrate (hypofibrinogenemia) than an increased enzyme activity. The lack of increased levels of circulating activator or plasmin, however, does not exclude ongoing fibrinolysis. Indeed, the activator or plasmin formed may be adsorbed to and concentrated on the fibrin deposits, any residual free enzyme being rapidly bound by its inhibitor. Direct evidence for the generation of plasmin in response to intravascular fibrin formation was recently provided by Collen. He has been able to show that the plasmin-antiplasmin complex contains a neoantigen, which is absent in both plasminogen and antiplasmin.55 By quantitatively measuring this neoantigen, he could demonstrate intravascular plasmin generation in clinical situations with enhanced fibrin formation.56 This phenomenon was for instance clearly apparent in patients undergoing therapeutic defibrination by infusion of the snake-venom enzyme, reptilase. Reptilase infusion in man results in a depletion of fibrinogen, a decrease in the level of plasminogen to approximately 60% of the preinfusion value, and the appearance of fibrin(ogen) degradation products. A two to threefold increase in the fractional catabolic rate of plasminogen is observed.46 Using a latex agglutination technique for the direct determination of the plasmin-antiplasmin complex in plasma, reptilase infusion was found to result in a progressive rise in the plasmin-antiplasmin titer over a period of 6-9 hr.57 These results represent the first conclusive evidence for plasminogen activation secondary to fibrin formation in man. Since no increased levels of plasminogen activator or of free plasmin are measurable, plasminogen activation may be a local phenomenon, possibly occurring in close association with fibrin formation in the microcirculation How does intravascular formation of fibrin oligomers lead to activation of the fibrinolytic system? Two possibilities are presently being considered: (1) It has been proposed that the plasminogen activator in plasma is very strongly adsorbed to fibrin;16 since plasminogen also to some extent is adsorbed onto fibrin,” the presence of fibrin oligomers in the circulation would provide the

VERMYLEN

AND

CHAMONE

surface on which activator and plasminogen are concentrated and can interact, resulting in a local generation of plasmin and prompt degradation of the oligomer. This hypothesis therefore does not postulate enhanced formation or release of plasminogen activator in disseminated intravascular coagulation. It states that the plasma plasminogen activator, which is hardly active in physiologic conditions, would become very effective once fibrin is generated. (2) It has also been suggested that fibrin formation within the blood vessel promptly results in enhanced secretion of plasminogen activator by the endothelial cell. However, it remains completely unknown how such secretion would be induced: the role of active clotting enzymes, of peptides released during coagulation, of local hypoxia, of local release of catecholamines, or of other as yet unidentified processes as inducing agents remains undefined. There also is no evidence at present to suggest that intravascular fibrin formation triggers the factor-XII-dependent pathway leading to the formation of plasma plasminogen activator. To conclude this section, there is no doubt that the hemostatic balance plays an active role in clinical situations with important fibrinogento-fibrin conversion (disseminated intravascular coagulation); the latter is compensated by enhanced plasminogen-to-plasmin conversion and rapid removal of fibrin deposits (or oligomers). As a consequence of this hemostatic balance, ongoing fibrinogen-to-fibrin conversion is not associated with thrombotic complications (and, as with reptilase or ancrod, may even be used as a means of obtaining anticoagulation) unless local stasis allows an important local accumulation of products of coagulation. DEFECTIVE

FIBRINOLYTIC

TENDENCY

TO

ACTIVITY

AND

A

THROMBOSIS

Patients with reduced fibrinolytic activity should have a deficient hemostatic balance and an increased tendency to thrombosis. What is the present evidence that this is indeed the case‘? The four following situations merit consideration. A Low Plasminogen

Level

Congenital aplasminogenemia has not been described and may be incompatible with life. Plasminogenproduction is delayed in the human

FIBRINOLYTIC

SYSTEM

IN THROMBOEMBOLISM

fetus and will not reach adult levels until 7-10 mo of age.s8 Ambrus and coworkers believe that lack of plasminogen is an important feature in the development of hyaline membrane disease of newborns. Indeed, fibrin in the alveoli and alveolar ducts may impede oxygenation and gas exchange, and furthermore, would act as an inhibitor of surfactant.59 This fibrin could not be cleared if there exists a relative lack of plasminogen. Ambrus et a1.60 have therefore performed a double-blind, randomized study in which 500 premature infants were treated with plasminogen or placebo intravenously within 60 min of birth. They found a substantial decrease in severe clinical respiratory distress, death caused by hyaline membrane disease, and total mortality in the plasminogen-treated infants as compared to the controls. This study may attest to a vital function of the fibrinolytic system, at least during the first few hours after birth. Although the site of synthesis of plasminogen has not been established unequivocally, the plasminogen level is considerably depressed in patients with cirrhosis of the liver. However, the level of antiplasmin is at least as reduced.6’ The balance is further tipped in favor of an enhanced instead of reduced fibrinolytic activity as a consequence of excess availability and delayed clearance of plasminogen activator.6’.62 In fact, slow ongoing activation of the fibrinolytic system in cirrhosis was documented by studies demonstrating an accelerated half-disappearance time of radiolabeled plasminogen.54 A Low Baseline Activator

Level

of Plasminogen

The radiolabeled fibrinogen scanning technique has revealed how frequent an occurrence deep-vein thrombosis is following general surgery. This technique has also permitted an analysis of which clinical features or laboratory tests best identify preoperatively those patients who will subsequently develop deep-vein thrombosis. Clayton and coworkers63collected data in women about to undergo gynecologic surgery and correlated these with the occurrence of postoperative deep-vein thrombosis. On the basis of three clinical features (age, percentage overweight for height, presenceof varicose veins) and two laboratory tests (the euglobulin clot lysis time and the level of fibrin(ogen)-related antigen in serum) they could calculate an index that

261

permitted a separation between patients subsequently developing deep-vein thrombosis or not. For our purpose, it is of interest that the two laboratory tests that helped in subdividing the patients both are related to fibrinolysis. Of the mentioned clinical and laboratory parameters, a prolongation of the euglobulin clot lysis time was the most discriminating as regards identification of patients who would develop thrombosis. This finding suggeststhat a reduced baseline level of plasminogen activator is associated with an increased tendency to thrombosis. On the other hand, the slight increase of fibrin(ogen)-related antigen in serum may indicate that, in some of the patients, thrombosis may already have been developing preoperatively, in line with the results of other studies. The general validity of the prognostic index developed by Clayton et al. can be questioned since it was derived from and then reapplied to the same patient group. Rakoczi and coworkers64have applied Clayton’s index to a comparable group of gynecologic patients undergoing abdominal hysterectomy and found that the index clearly separated the thrombosis group from the nonthrombisis group. Only a relatively small overlap was noted, thus confirming that prediction of postoperative leg vein thrombosis from a simple set of preoperative clinical and laboratory data is feasible. Again, the euglobulin clot lysis time was the most significant discriminator between both groups. Therefore, it may be concluded that a lower level of plasma plasminogen activator preoperatively is associatedwith an increasedrisk of developing postoperative deep-vein thrombosis and that therefore the fibrinolytic system may play an important part in preventing deep-vein thrombosis. Of interest in this regard is the study of Knight and Dawson.” These authors found that the reduction in fibrinolytic activity, which normally follows surgical operation, was prevented by intermittent compressionof the arms during and after surgery. This procedure was believed to releasevascular plasminogen activator from its stores in the arms and in any event reduced the incidence of deepvenous thrombosis in the legs to half that in control patients. This study once more emphasizes that fibrinolysis may be important in preventing thrombosis. Cutaneous vasculitis is an inflammatory state of the blood vesselswith fibrinoid necrosis and eventual occlusion, often by a thrombus. A

262

persistent reduction of the blood fibrinolytic activity has been demonstrated by Cunliffe@’ in 32 of 52 patients with this syndrome. Whether the measured changes in fibrinolysis represent cause or consequence is difficult to decide. One of the modern theories on the pathogenesis of atherosclerosis, formulated by Ross,“’ states that this disorder results from repeated endothelial injury. Platelets would adhere to the exposed subendothelial surface and secrete a mitogenic factor, which induces smooth muscle cells in the media to proliferate and to form the basic structure of the atherosclerotic plaque. About 20 yr earlier, Astrup6’ also suggested that atherosclerosis resulted from repeated endothelial injury. In his view, impaired fibrinolysis may lead to the persistence of fibrin deposits at sites of luminal injury, the incorporation of the fibrin into the vessel wall, and ultimately, the development of degenerative atherosclerotic changes, The theories of Ross and Astrup may not be entirely mutually exclusive. To substantiate Astrup’s proposal, attempts have been made to determine a possible relation between impaired blood fibrinolysis and coronary artery disease. These investigations, summarized by Rosing et a1.69have produced conflicting results, partially because of the differences in the assay systems used by the various investigators and the possible effect of age on fibrinolytic activity. Also, fibrinolytic activity usually has been assayed using single blood samples obtained from subjects at rest. Such a procedure is a relatively insensitive means of assessing the integrity of the blood fibrinolytic system, since abnormalities in the response of the system may be detectable only when the system is stressed. Although intense exercise constitutes a desirable means of testing the capacity of the blood fibrinolytic system, this type of stress is impractical for subjects with coronary artery disease. However, Rosing et a1.“9 considered that the diurnal fluctuations of fibrinolytic activity normally occurring in man and presumably representing a response to internal physiologic stimuli, also provide an index of the functional capacity of the fibrinolytic system. They therefore determined the diurnal patterns of plasma euglobulin fibrinolytic activity, estimated by the fibrin plate method, in groups of young normal and older normal subjects, in subjects with type IV hyperlipoproteinemia, and

VERMYLEN

AND

CHAMONE

in subjects with coronary artery disease who had normal lipid profiles. The marked diurnal increases in fibrinolytic activity observed in the young normal subjects were significantly reduced in a large percentage of the older normal subjects and in most of the subjects with coronary artery disease or type IV hyperlipoproteinemia. Although not conclusive, the authors believed these findings to be compatible with the hypothesis that an impairment in the responsiveness of the fibrinolytic system may be related to the development of coronary artery disease. A Reduced Plasminogen

Reserve of Activator

Attempts to estimate the reserve of plasminogen activator in the vessel wall have been made by measuring the amount of activator releasedinto the blood after a standard stimulus (e.g., venous occlusion) or by direct histochemical measurement on veins obtained by biopsy. Isaacson and Nilsson” quantitated components of the fibrinolytic system in the circulating blood, the local fibrinolytic activity in blood during venous occlusion of the limbs, and the plasminogen activator content of the wall of superficial veins in 91 patients with recurrent phlebographically verified idiopathic deep venous thrombosis and in 26 with recurrent histologically verified idiopathic superficial thrombophlebitis without phlebographic abnormalities. They found a defective release of plasminogen activator from the vessel walls during venous occlusion and/or a decreased plasminogen activator content in walls of superficial veins in 85 (73%) of the patients. This association between nonacute venous thrombosis and a defect in the fibrinolytic system is closer than that between the former and any other known disturbance of the hemostatic balance. Persistent changes in the level of plasminogen and inhibitors of fibrinolysis were rare. Judging from Isaacsonand Nilsson’s findings, patients with recurrent idiopathic deep venous thrombosis often have a reduced reserve of plasminogen activator. This group of patients, however, may be heterogeneous. Steele et al.-’ have found that some patients with idiopathic recurrent venous thrombosis have a short platelet survival, while others have a normal one. The patients with short platelet survival were

FIBRINOLYTIC

SYSTEM

IN THROMBOEMBOLISM

benefited by drugs affecting platelet function, those with normal platelet survival by anticoagulants. If, and to what extent, there would be overlapping between defective fibrinolysis and increased platelet consumption in patients with idiopathic recurrent venous thrombosis, has yet to be established. The treatment of idiopathic venous thrombosis with synthetic fibrinolytic agents will be discussed elsewhere in this symposium. An Increased

Level

of Inhibitor

Nilsson et al.” observed a patient with widespread venous thrombosis associated with severely impaired fibrinolytic activity due to the presence of a potent inhibitor of fibrinolysis. Confirmatory reports in patients with occlusion of retinal veins were subsequently published by the same group.73 From an analysis of these studies, the inhibitor appeared to be directed mainly against plasminogen activator. In 1937, Behcet described a triple symptom complex consisting of recurrent ulceration in the mucous membrane of the mouth and genitalia and of ocular inflammatory changes. Since this description, many other clinical manifestations have been reported. Superficial and deep thrombophlebitis are prominent features of this disease, their incidence ranging from 12% to 40%. Thrombosis of large veins, including superior and inferior vena cava, also has been reported. Chajek et a1.74found a pronounced decreaseof the spontaneousfibrinolytic activity in a majority of patients with this disorder. This reduced fibrinolytic activity may have resulted from the presenceof an inhibitor of plasminogen activator in these patients’ sera and may help account for the enhanced thrombotic tendency in this disorder. From the findings reported in this section, it becomes apparent that a reduced fibrinolytic activity, resulting from a low plasma plasminogen, a low baseline plasma plasminogen activator, depleted stores of endothelial plasminogen activator or increased levels of inhibitor, may indeed tip the hemostatic balance towards an increased tendency to thrombosis. Somewhat unexplained is the relatively low incidence of thromboembolic phenomenaencountered during long-term treatment with antifibrinolytic drugs.”

263

Until now we have mainly discussedthe role of the fibrinolytic system in the prevention of thrombosis. Once a detectable thrombus has formed, does the fibrinolytic system still have a function to fulfill? Kakkar76 reported on 132 consecutive patients investigated during the postoperative period using the ‘X I-labeled fibrinogen test. Thrombosis, mainly in the calf veins, developed in 40 patients. In 14 of these, the thrombus lysed spontaneously within 72 hr. A significant incidence of spontaneousthrombolysis therefore was noted. Spontaneous lysis of thrombi in larger veins or of pulmonary emboli has also frequently been observed. We shall not discussthese findings here, since they are illustrated in the control groups of the clinical studies with thrombolytic agents published elsewherein this symposium. Spontaneous lysis occurs more frequently in veins than in arteries; this may be related to a different composition of arterial and venous thrombi and to the fact that the arterial intima contains lessplasminogen activator than that present within veins.” Kwaan” demonstrated that sites of thrombolysis were related to the presenceof endothelial cells containing plasminogen activator. Destruction of endothelial cells thus resulted in markedly delayed thrombolysis. Therefore, endothelial fibrinolysis determines the fate of a thrombus to a large extent. SUMMARY

The present concept of physiologic fibrinolysis was reviewed. It was concluded that the nonspecific proteolytic activity of plasmin would essentially be limited to fibrin in vivo in view of (A) the specific adsorption of activator and of plasminogen onto the fibrin surface resulting in local generation of plasmin and (B) the fact that plasmin, adsorbed to fibrin (in contrast to plasmin in the fluid phase) largely escapesfrom the action of antiplasmin. The hemostatic balance in the resting condition was discussed.It was concluded that under normal conditions, systemic intravascular fibrin deposition or formation must be either nonexistent or extremely limited. On the other hand, there is considerable evidence that a limited systemic fibrinogenolysis is going on in healthy individuals and that this processcan be accelerated by simple physiologic procedures, such as strenuous physical exercise.

264

VERMYLEN

The hemostatic balance in disseminated intravascular coagulation was presented. It was concluded that there now exists both indirect and direct evidence that enhanced fibrinogento-fibrin conversion is counterbalanced by increased plasminogen-to-plasmin transformation. Evidence for the hypothesis that defective fibrinolytic activity leads to a tendency to thrombosis was reviewed. Situations with a low plas-

AND

CHAMONE

minogen level, a low baseline level of plasmlnogen activator, a reduced reserve of plasminogen activator, or an increased level of inhibitor were discussed. The evidence suggests that a reduced fibrinolytic activity may tip the hemobtatic balance towards an increased tendency to thrombosis. Finally, the role of the fibrinolytic system in removing already existing thrombi wah mentioned.

REFERENCES I. Morgagni JB: The Seats 111. London. 1769, p 185. Cited Arnold, London, 1965, p 3 2. Denis physiologique Behcet.

and Causes in Fearley

PS: Essai sur I’application du sang de I’homme.

Cited

in Keller

of Discascs. vol. GR: t:ibrinolysis.

de la chimic i I’itude Paris. 1X3X. Editor

F: The development

of our

of fibrinolysis. Am J Med 6:367. 1960 3. Tillett WS. Garner RL.: The fibrinolytic hemolytic streptococci. J Exp Med 5X:4X5. 1933 4. Milstone H: A factor participates in streptococcal 1941 5. Christcnscn lytic reaction due cal

fibrinolysin. 6. Christensen

in normal fibrinolysia.

LR: Streptococcal to a strum enzyme J Gen LR,

Physiol MacLeod

of serum: Characterization, inhibitors. J Gcn Physiol 7. Astrup T. Permin

2X:363. CM:

1910 L: On

S: A plasminogen blood. Proc Sot

activator in spontaneously Biol Med X2:29 I, 1953

SB (eds): Concepta.

14. ological

mechanism 1978

Diath to plas-

Fibrinolysis: London. of phyal-

P, Astrup T: Ditferencea and tissue plasminogen 2X:65. 1972 by adsorption. Nature

17. Thorsen S: Differences In the binding to fibrin of native plasminogen and plasminogen modified by proteolytlc degradation. Effect of w-aminocarboxylic acids. Biochim Biophys Acta 393:55. 1975 18. Wiman B, Wall& P: Structural relationship between “glutamic

acid”

and

“lysine”

forms

of human

plasminogen

( ollcn

‘7. plabmin 23. tion

the

N H,-terminal

altinity

ac~lvatl~~n

chromatography.

Eur

D: Identihcation

and

home

properties

human and

of a nc\r

plasma.

B. Collcn antiplasmin.

C‘ol cn dlytribution

physlo,dglcal 197’ 26. Shapiro

conditions SS.

Martincl

Ph)\iol

J: Human

mcnt 54:43.

in

A

RCi. activity

adrenaline.

I Y47

ILancet

I : 102.

I.

J C‘11n

et al: Meusurc-

blood.

2X Mact-arlane KC;, Biggs R. Obhcrvatlons : Purilication the fast-nctlng

D. Tvtgat GN. of librinogen.

I:ur

charactcrlxltion

MiillcrtT S. Clemmcnscn I The primary 111 human plasma. Biochcm J 159:545,

Wiman of human

.I

P: The \pecllic lnrcraction bct\\ccn A physiological rx)Ic of the ly\inc Thromb Rcs IO:21 3. !477

in pk~ma. Eur J Biochem 24 Nolf P: 1-a librinolysc. and

of plasminogcn

min conversion. in Gaffney PJ. Ulutin Current Fundamental and Clinical Academic, 1978. p 47

IS. Thorsen S. Glas-Greenwalt in the binding to fibrin of urokinasc activator. Thromb Diath Haemorrh 16. Fearnley CR: Fibrinolysis 172544. 1953

animal

J Biol

12. Astrup T: The haemostatic balance. Thromb Haemorrh 2:347. 195X 13. Wiman B: Biochemistry of the plasminogen

Wiman B. Collen D: Molecular fibrinolysis. Nature 272:549.

with

dcr Uterusmukosa. librlnolysln.\.

with

ct,-pla\mln inhibilor from human plasma. inhibiror which inhibl~h activator-induced Chcm ‘5 I :5Y56. I976

enzyme

the

lntcractlon ;,s studied by 50:4X9. I975

I’&-reacting plasmin inhibitor in Blochcm r>Y:209. I Y7h ?I ‘vloroi %I. Atrkl N: I\;olat1on

25.

tissue

Exp

ot

reaction in

IO. Todd AS: The histological localiaation activator. J Pathol Bacterial 78:2X1, 1959 I I. Miillert7 active human

70. activity

A proteoby streptococ-

and

organism. Nature 159:6X I. I947 8. Halban J. Frank1 0: Zur Biochemie Gynaekol Rundsch 4:471. 9. Fleisher MS, Loeb Chem 21:477, 1915

knowledge

1945 A protcolytic

activation 2X:559. lY45 PM: Fibrinolysia

their

peptidc Biochcm

I’). Wiman B, Wallcn plaaminogcn and librin. binding ‘iilc in plasminogen.

human blood which J lmmunol 42: IOY. tibrinolysia: activated

and

J -terminal glutamic acid to NH,-terminal lysine human plasminogen by plasmin. Thromb Res 3:515, 1973 43. Semeraro N. Collen D, Verstraete M: On the origin of the Aa chain heterogeneity of human fibrinogen. Biochim Biophys Acta 492:204, 1977 44. Alkjaersig N, Fletcher AP. Burstein R: Association between oral contraceptive use and thromboembolism: A new approach to its investigation based on plasma fibrinogen chromatography. Am J Obstet Gynecol 122:199, 1975 45. Harfenist EJ, Lauer RC, Canlield RE. et al: Isolation, characterization and radioimmunoassay of a peptide resulting from limited proteolysis of human fibrinogen by plasmin. Vth Congress of the International Society on Thrombosis and Haemostasis, Paris, 1975, abstr 390 46. Collen D, Vermylen J: Metabolism of iodine-labeled plasminogen during streptokinase and reptilase therapy in man. Thromb Res 2:239, 1973 47. Hart H Ch: De biologische halveringstijd van I3 lI-fibrinogeen. Thesis, Utrecht, 1966 48. Collen D, Tytgat G, Claeys H, et al: Metabolism of plasminogen in healthy subjects: Etrect of tranexamic acid. J Clin Invest 51:1310, 1972 49. Collen D, Semeraro N, Tricot JP, et al: Turnover of fibrinogen, plasminogen, and prothrombin during exercise in man. J Appl Physiol 42:865, 1977 50. Merskey C: Defibrination syndrome or. ? Blood 41:599, 1973 5 I McKay DC: Progress in disseminated intravascular coagulation. Calif Med: 186, 279, 1969 52. Vermylen J. Donati MB, Verstraete M: The identitication of fibrinogen derivatives in plasma and serum by agarose gel filtration. Stand J Haematol 525 (Suppl):157, 1971 53. Donati MB. Verhaeghe R, Culasso DE, et al: Molec-

265 ular size distribution of fibrinogen derivatives formed in vitro and in vivo: A chromatographic study. Thromb Haemostas 36:14. 1976 54. Collen D: Plasminogen and prothrombin metabolism in man. Thesis, Leuven. 1974 55. Collen D, De Cock F: Emergence in plasma during activation of the coagulation or fibrinolytic system of neoantigens. associated with the complexes of thrombin or plasmin with their inhibitors. Thromb Res 5:777, 1974 56. Collen D. De Cock F. Cambiaso CL, et al: A latex agglutination test for rapid quantitative estimation of the plasmin-antiplasmin complex in human plasma. Eur J Clin Invest 7:21. 1977 57. Collen D, De Cock F, Verstracte M: Plasminogen turnover and plasmin-antiplasmin complex formation in clinical conditions with primary or secondary activation of the fibrinolytic system, in Paoletti R, Sherry S (eds): Thrombosis and Urokinasc. New York, Academic, 197’7, P 37 58. Ambrus CM, Weintraub DH. Niswarder KR, et al: Studies on hyaline membrane disease: II. The ontogeny of the fibrinolysis system. Pediatrics 35:91, 1965 59. Taylor FB, Abrams MD: Effect of surface active lipoprotein on clotting and fibrinolysis and of fibrinogen on surface tension of surface active lipoprotein. Am J Med 40:346, I966 60. Ambrus CM, Choi TS, Cunnanan E, et al: Prevention of hyaline membrane disease with plasminogen. A cooperative study. JAMA 237:1837, 1977 61. Fletcher AP, Biederman 0. Moore D: Abnormal plasminogen-plasmin system activity (fibrinolysis) in patients with hepatic cirrhosis. Its cause and consequences. J Clin Invest 43:681, 1964 62. Weiner M: The librinolytic response to nicotinic acid in abnormal liver states. Am J Med Sci 246:294. 1963 63. Clayton JK. Anderson JA, McNicol GP: Preoperative prediction of postoperative deep vein thrombosis. Br Med J 2:910, 1976 64. Rakoczi I, Chamone DF, Collen D. et al: Prediction of postoperative leg vein thrombosis in gynaecological patients. Lancet 1:509, 1978 65. Knight MTN. Dawson R: ElTect of intermittent compression of the arms on deep venous thrombosis in the legs. Lancet 2:1265, 1976 66. Cunliffe WJ, Menon IS: The association between cutaneous vasculitis and decreased blood fibrinolytic activity. Br J Dermatol 84:99, 1971 67. Ross R, Glomset JA: The pathogenesis of atherosclcrosis. N Engl J Med 295:369, 1976 68. Astrup T: Biological significance of fibrinolysis. Lancet 2:565. 1956 69. Rosing DR. Redwood DR, Brakman P, et al: Impairment of the diurnal fibrinolytic response in man. Effects of ageing, type IV hyperlipoproteinemia, and coronary artery disease. Circ Res 32:752, 1973 70. lsaacson S, Nilsson IM: Defective fibrinolysis in blood and vein walls in recurrent “idiopathic” venous thrombosis. Acta Chir Stand 138:313, 1972 7 I Steele PP. Weily HS, Genton E: Platelet survival and adhesiveness in recurrent venous thrombosis. N Engl J Med 288:1148, 1973

VERMYLEN

266

72. Nilsson IM, Krook H, Sternby NH, et al: Severe thrombotic disease in a young man with bone marrow and skeletal changes and with a high content of an inhibitor in the fibrinolytic system. Acta Med Stand 169:323. 1961 73. Pandolfi M, Hedner U, Nilsson IM: Bilateral occlusion of the retinal veins in a patient with inhibition of fibrinolysis. Ann Ophthalmol 2:48 I, 1970 74. Chajek

T. Aronowski

E. lzak G: Decreased

fibrinoly-

AND

CHAMONE

sis in Behcet’s disease. Thromb Diath Haemorrh 29:610. 1973 75. Verstraete M: Haemostatic Drugs. A Critical Appraisal. The Hague, Martinus Nyhoff. 1977. pp 120 134 76. Kakkar VV, Howe CT, Flanc C, et al: Natural history of post-operative deep vein thrombosis. Lancet:230. 1969 77. Kwaan HC, Astrup T: Fibrinolytic activity in thrombosed veins. Circ Res 17:477. 1965