Pathophysiology and management of coagulation disorders in critical care medicine - Thesis

(1)

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Pathophysiology and management of coagulation disorders in critical care

medicine

de Jonge, E.

Publication date 2000

Document Version Final published version

Link to publication

Citation for published version (APA):

de Jonge, E. (2000). Pathophysiology and management of coagulation disorders in critical care medicine.

General rights

It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulations

If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.

(2)

Pathophysiologyy and Management

I f ff Coagulation Disorders

inn Critical Care Medicine

(3)

PATHOPHYSIOLOGYY AND MANAGEMENT

OFF COAGULATION DISORDERS

(4)

Thiss thesis was prepared at the Departments of Intensive Care and Vascular Medicinee and at the Laboratory of Experimental Internal Medicine, Academic Medicall Center, University of Amsterdam, The Netherlands

Coverr design: Chris Bor

Dockingg of tissue factor and factor Vila.

(5)

PATHOPHYSIOLOGYY AND MANAGEMENT

OFF COAGULATION DISORDERS

INN CRITICAL CARE MEDICINE

ACADEMISCHH PROEFSCHRIFT

terr verkrijging van de graad van doctor aann de Universiteit van Amsterdam,

opp gezag van de Rector Magnificus prof.. dr. J.J.M. Franse

tenn overstaan van een door het College voor Promotiess ingestelde commissie, in het openbaar

tee verdedigen in de Aula der Universiteit opp woensdag 11 oktober 2000, te 14.00 uur

door r

Evertt de Jonge geborenn te 's Gravenhage

(6)

Promotiecommissie: :

Promotores: :

Prof.. Dr. M. Levi Prof.. Dr. H.R. Büller Co-promotores: :

Dr.. T. van der Poll Dr.. J. Kesecioglu Overigee leden:

Prof.. Dr. E. Briët

Prof.. Dr. S.J.H, van Deventer Prof.. Dr. H. Obertop

Prof.. Dr. J.A. Romijn Prof.. Dr. P. Speelman Prof.. Dr. A. Trouwborst

Faculteitt der Geneeskunde, Universiteit van Amsterdam

Thee publication of this thesis was supported by grants from Aventis, Braun, Cook,, Fesenius, Janssen-Cilag, Leo Pharma and Novo Nordisk.

(7)

VoorVoor mijn vader en moeder VoorVoor Arja

(8)

(9)

Contents s

Chapterr 1

Generall introduction and outline of the thesis 9

Chapterr 2

Currentt drug treatment strategies for disseminated intravascular coagulation n

DrugsDrugs 1998;55:767-777 15

Chapterr 3

Coagulationn abnormalities in sepsis: relation with inflammatory responses

CurrentCurrent Opinion in Critical Care, October 2000, in press 33 Chapterr 4

Tissuee factor pathway inhibitor (TFPI) dose-dependently inhibits coagulationn activation without influencing the fibrinolytic and cytokine responsee during human endotoxemia

BloodBlood 2000;95:1124-1129 49

Chapterr 5

Tissuee factor pathway inhibitor (TFPI) does not influence inflammatory pathwayss during human endotoxemia

submittedsubmitted for publication 69

Chapterr 6

Activationn of coagulation by administration of recombinant factor Vila

(10)

Chapterr 7

Thee in vivo kinetics of tissue factor mRNA expression during human endotoxemia:: relationship with activation of coagulation

BloodBlood 2000;96:554-559 95

Chapterr 8

Effectss of different plasma substitutes on blood coagulation. A comparative review w

acceptedaccepted for publication in Critical Care Medicine 115 Chapterr 9

Impairedd haemostasis by intravenous administration of a gelatin-based plasmaa expander in human subjects

ThrombosisThrombosis and Haemostasis 1998;79:286-290 131 Chapterr 10

Decreasedd circulating levels of von Willebrand factor after intravenous administrationn of a rapidly degradable hydroxyethyl starch (HES 200/0.5/6) inn healthy human subjects

submittedsubmitted for publication 143

Chapterr 11

Pharmacologicall strategies to decrease excessive blood loss in cardiacc surgery: a meta-analysis of clinically relevant end points

LancetLancet 1999;354:1940-1947 157

Summaryy 181

Samenvattingg 185

(11)

ChapterChapter 1

Generall Introduction

and d

(12)

ChapterChapter 1

Generall introduction

Coagulationn disorders, which may present as either localized or systemic thrombosiss or increased bleeding tendency, are problems physicians frequently encounterr when treating intensive care patients. Bleeding may be caused by surgicall or accidental trauma or by a variety of medical conditions. Although hemorrhagee can result from undue stress on a normal hemostatic system, it also mayy result from or be aggravated by acquired coagulation defects. Some of these coagulationn defects are a direct result of medical treatments like anticoagulants, fibrinolyticc drugs, platelet aggregation inhibitors or artificial plasma substitutes. Acquiredd coagulation disorders also may result from conditions like liver failure,, myeloproliferative disorders or hypothermia. Furthermore, a complex hemorrhagicc syndrome has emerged following open heart surgery with

extracorporeall circulation.1 It should be noted that all of these conditions

associatedd with increased bleeding tendency are likely to be seen in the intensive caree unit. Paradoxically, not only increased bleeding tendency but also hypercoagulabilityy is a frequent complication of critically ill patients, for examplee leading to the clinical syndrome of disseminated intravascular

coagulationn (DIC).2

Influencee of coagulation disorders on patient outcome and costs of treatment t

Coagulationn disorders may have a significant impact on outcome and costs of treatmentt of critically ill patients. Abnormal prothrombin time and thrombocytopeniaa have repeatedly been identified as risk factors for death in

intensivee care patients.3'5 The precise mechanism by which abnormal

coagulationn and thrombocytopenia are related to a poor prognosis is unknown. It mayy be speculated that the associated bleeding tendency may lead to shock and organn failure or that increased blood transfusion is responsible for the higher

mortality.66 Alternatively, prolonged coagulation times and thrombocytopenia

couldd be present in patients with the worst prognosis due to consumption by DIC.. Indeed, in patients with sepsis, the severity of DIC has been shown to

correlatee with poor outcome.7

Knowledgee of the pathophysiology and treatment options of coagulation disorderss is not only important to improve patient outcome, it also may affect thee economics of intensive care medicine. Treatment of coagulation disorders

frequentlyfrequently consists of replacement therapy with plasma, coagulation factors or 10 0

(13)

GeneralGeneral introduction platelets,, all of which are very expensive. Blood and coagulation products have

beenn reported to account for more than 16% of total intensive care costs8 and

theree are anecdotal reports of treatments with purified coagulation factors at expensess up to € 700,000 for one patient [Wester JP, personal communication]. Evidencee of the benefit of expensive treatments, e.g. antithrombin replacement

therapyy in patients with DIC, is frequently not or only partly present.9

Aimm and outline of this thesis

Too improve patient outcome and decrease the costs of treatment more research onn the pathophysiology and treatment of coagulation disorders is necessary. Aim off this thesis is to contribute to this research. Two major subjects are addressed. First,, we present a series of studies on the role of tissue factor (TF) and its major endogenouss inhibitor tissue factor pathway inhibitor (TFPI) in the pathophysiologyy of disseminated intravascular coagulation (DIC). DIC is a frequentfrequent complication of a variety of disease states like sepsis and trauma and

mayy have a major impact on patient outcome.2;? In the second part of this thesis

wee focus on the effects of some treatments that are frequently given to intensive caree patients on blood coagulation and post-operative blood loss.

ChapterChapter 2 and 3 give an overview of the pathophysiology of DIC, the relations

betweenn activation of coagulation and inflammatory responses and the treatment strategiess presently available.

Inn Chapter 4 and 5 we studied the influence of a 6-hr infusion of recombinant TFPII on the endotoxin-induced coagulant, fibrinolytic and inflammatory responsee in healthy human subjects. Two different doses of TFPI were investigatedd in a randomized, placebo-controlled, cross-over study.

ChapterChapter 6 addresses the question whether activation of coagulation can induce a

proinflammatoryy response in humans. Six healthy human volunteers received a boluss injection of recombinant factor Vila (rVIIa) preceded by either a subcutaneouss injection of recombinant NAPc2, a specific inhibitor of tissue factor,, or placebo. The study was designed as a randomized, placebo controlled, cross-overr study. We investigated the effects of rVIIa and rNAPc2 on the activationn of coagulation and on the cytokine response.

(14)

ChapterChapter 1

Inn Chapter 7 we studied the expression of TF mRNA on human monocytes duringg endotoxemia and correlated the pattern of TF mRNA expression with in

vivovivo thrombin generation.

ChapterChapter 8 presents an overview of the effects of different artificial plasma

substitutess on blood coagulation and peri-operative blood loss.

Inn Chapter 9 and 10 we studied the influence of two plasma substitutes (i.e. a modifiedd gelatin and a rapidly degradable medium molecular weight hydroxyethyll starch) on primary hemostasis and blood coagulation. Both studies weree randomized, controlled cross-over experiments in healthy human subjects.

ChapterChapter 11 provides a meta-analysis of all randomized, controlled trials that

studiedd the effcets of aprotinin, e-aminocaproic acid, tranexamic acid and desmopressinn on blood loss in cardiac surgery. Only trials were included that reportedd at least one of the following outcomes: mortality, re-thoracotomy, proportionn of patients receiving blood transfusion or peri-operative myocardial infarction. .

Thee results of our studies are summarized in Chapter 12.

(15)

GeneralGeneral introduction

References s

1.. Mammen EF, Koets MH, Washington BC, et al. Hemostasis changes during cardiopulmonaryy bypass surgery. Semin Thromb Hemost 1985; 11: 281-292.

2.. Levi M, Ten Cate H. Disseminated intravascular coagulation. N Engl J Med 1999; 341: 586-592. .

3.. Vincent JL, de Mendonca A, Cantraine F, et al. Use of the SOFA score to assess the incidencee of organ dysfunction/failure in intensive care units: results of a multicenter, prospectivee study. Working group on "sepsis-related problems" of the European Society off Intensive Care Medicine.. Crit Care Med 1998; 26: 1793-1800.

4.. Le Gall JR, Klar J, Lemeshow S, et al. The Logistic Organ Dysfunction system. A new wayy to assess organ dysfunction in the intensive care unit. ICU Scoring Group. JAMA 1996;; 276: 802-810.

5.. Lemeshow S, Teres D, Klar J, Avrunin JS, Gehlbach SH, Rapoport J. Mortality Probabilityy Models (MPM II) based on an international cohort of intensive care unit patients.. JAMA 1993; 270: 2478-2486.

6.. Hebert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical triall of transfusion requirements in critical care. Transfusion Requirements in Critical Caree Investigators, Canadian Critical Care Trials Group. N Engl J Med 1999; 340: 409-417. .

7.. Fourrier F, Chopin C, Goudemand J, et al. Septic shock, multiple organ failure, and disseminatedd intravascular coagulation. Compared patterns of antithrombin III, protein C, andd protein S deficiencies. Chest 1992; 101: 816-823.

8.. Klepzig H, Winten G, Thierolf C, Kiesling G, Usadel KH, Zeiher AM. [Treatment costs inn a medical intensive care unit: a comparison of 1992 and 1997]. Dtsch Med Wochenschrr 1998; 123: 719-725.

9.. Levi M, de Jonge E, van der Poll T, Ten Cate H. Disseminated intravascular coagulation. State-of-the-art.. Thromb Haemost 1999; 82: 695-705.

(16)

(17)

ChapterChapter 2

Currentt Drug Treatment Strategies for

Disseminatedd Intravascular Coagulation

Evertt de Jonge1, Marcel Levi2,

Christiaann P. Stoutenbeek', Sander J.H.van Deventer3

Departmentss of (1) Intensive Care, (2) Vascular Medicine and (3) Experimental Internall Medicine. Academic Medical Center, University of Amsterdam,

thee Netherlands.

(18)

ChapterChapter 2

Abstract t

Disseminatedd intravascular coagulation can be caused by a variety of diseases.. Experimental models of DIC have provided substantial insight in the pathogenesiss of this disorder, which may ultimately result in improved treatment.. Disseminated coagulation is the result of a complex imbalance of coagulationn and fibrinolysis. Simultaneously occurring tissue-factor dependent activationn of coagulation, depression of natural anticoagulant pathways and shut-downn of endogenous fibrinolysis all contribute to the clinical picture of widespreadd thrombotic deposition in the microvasculature and subsequent multiplee organ failure. Cornerstone for the treatment of DIC is the optimal managementt of the underlying disorder. At present, specific treatment of the coagulationn disorders is not based on firm evidence from controlled clinical trials.. Plasma and platelet transfusion are used in patients with bleeding or at riskrisk for bleeding and low levels of coagulation factors or thrombocytopenia. The rolee of heparin and low molecular weight heparin is controversial, but may be justifiedd in patients with active DIC and clinical signs of extensive fibrin depositionn such as in patients with meningococcal sepsis. Scarce evidence indicatee that low molecular weight heparin is as effective as unfractionated heparinn but may be associated with a decreased bleeding risk. AT III replacementt appears to be effective in decreasing the signs of DIC if high doses aree administered, but effects on survival or other clinically significant parameterss are at least uncertain. If administered AT HI supplementation should bee dosed aimed at normal or supranormal plasma levels. Results of studies on proteinn C concentrate, thrombomodulin or inhibitors of tissue factor are promising,, but the efficacy and safety of these novel strategies remains to be establishedd in appropriate clinical trials.

(19)

DrugDrug Treatment Strategies for DIC

Introduction n

Disseminatedd intravascular coagulation (DIC) is a frequent complication of aa variety of disease states such as infection, trauma, malignancies and obstetric complications.. Infection is the commonest cause of disseminated intravascular

coagulation11 and in patients with septic shock DIC is a strong predictor of

death.22 In patients with DIC, the systemic activation of blood coagulation results

inn the generation and deposition of fibrin, leading to microvascular thrombi in variouss organs and contributing to the development of multi-organ failure. Depletionn of coagulation proteins and platelets, due to the ongoing activation of

thee coagulation system, may induce severe bleeding complications.3 The

managementt of DIC is primarily directed at treating the underlying disease, but supportivee care may be important. This supportive care may consist of supplementingg the depleted coagulation factors and endogenous coagulation inhibitors,, and of the inhibition of coagulation by various anticoagulant strategiess or by manipulating the fibrinolytic system. In this review we will first brieflyy focus on the current insight in the pathogenesis of DIC, since currently availablee and future treatment strategies may be based on this knowledge. Subsequently,, the various treatment strategies for DIC will be discussed. We willl not focus on the laboratory diagnosis of DIC. An extensive review of this

subjectt has been published recently by Bick.4

Pathogenesiss of Disseminated Intravascular Coagulation

ActivationActivation of coagulation

Thee initiation of the systemic activation of coagulation is dependent on the underlyingg cause of DIC. In most cases, however, the activation of coagulation appearss to be mediated by cytokines, which are produced by the host in response too various pathogenetic insults. For example, in sepsis, the activation of coagulationn is initiated by microorganisms and their products like endotoxins andd by cytokines, mainly produced by mononuclear cells in response to these endotoxins. .

Thee derangement of the coagulation system comprises enhanced activation off coagulation, depression of inhibitory mechanisms of coagulation, and

inhibitionn of the fibrinolytic system.5 Most of the current insight in those

pathogeneticc pathways has been derived from experimental studies of bacteriemiaa or endotoxinemia in humans or primates. The intravenous administrationn of endotoxin to human subjects or primates resulted in the

(20)

ChapterChapter 2

activationn of coagulation as reflected by elevation in markers for thrombin generationn like thrombin-antithrombin complexes and fragment F 1+2.

Althoughh the intrinsic (contact system-dependent) pathway of coagulation mayy be activated during sepsis, it seems not to be involved in the initiation of

DIC.8,99 This system, however, may play an important role in the pathogenesiss of

systemicc hypotension. It has been shown that thrombin generation is mediated byy the (extrinsic) tissue factor/factor Vila-dependent pathway. Indeed, tissue factorr expression by endothelial cells and blood monocytes can be induced by

endotoxinn and by cytokines like TNF.10'11 Furthermore, the importance of tissue

factorr in the pathogenesis of DIC was confirmed by observations that the coagulantt response upon bacteriemie or endotoxemia could be completely blockedd by the simultaneous administration of monoclonal antibodies that are

ablee to inhibit tissue factor or factor Vila activity.7'12'13 It may be assumed that

thrombinn plays a pivotal role in further activation of systemic coagulation. Not onlyy may thrombin affect several positive feedback loops (for example by direct activationn of factor IX, thereby resulting in even more factor IXa and subsequent factorr Xa generation), but also can thrombin act as a potent agonist for platelet activation.. Activation of platelets either by generated thrombin or as a direct of endotoxinss or cytokines may then further facilitate coagulation activation. The mechanismm of coagulation activation in DIC is summarized in figure 1.

Thee endotoxin-induced activation of the tissue-factor system and subsequentt activation of coagulation seems to be mediated by pro-inflammatory cytokiness like TNF-alpha, interleukin-1 and interleukin-6. TNF-alpha administrationn to healthy volunteers elicited a rapid activation of coagulation,

whichh was similar to that evoked by microorganisms or endotoxin.14 However

thee role of TNF in endotoxin-induced activation of coagulation became less clearr when subsequent studies showed that monoclonal antibodies directed againstt TNF activity were able to abolish the endotoxin-stimulated increase in

TNFF whereas thrombin generation was unchanged.15 In contrast to this

observationn monoclonal antibodies directed against interleukin-6 were able to completelyy block the endotoxin-induced activation of coagulation in

chimpanzees.166 In addition, it was shown that IL-6 infusion in baboons and in

humann cancer patients induced thrombin generation.1718 Hence, these data

suggestt that IL-6 rather than TNF is the primary mediator for the induction of coagulationn in sepsis. The role of other cytokines, such as IL-1 is less clear. Treatmentt of septic patients with IL-1 receptor antagonist resulted in lower thrombinn generation as reflected in decreased levels of thrombin-antithrombin

complexes.199 Also, administration of interleukin-1 to baboons resulted in

(21)

systemicc coagulation activation.20 It is, however, not clear whether this effect of IL-11 is a direct effect or an effect mediated by other IL-1 induced cytokines.

InhibitorsInhibitors of coagulation

Thee thrombin generated by the activated coagulation promotes further activationn of coagulation by a number of positive feedback loops. To balance thiss ongoing activation of coagulation the human body uses various inhibitory systems.. One of the major inhibitors of coagulation is antithrombin III (AT III). Itt rapidly binds and inactivates thrombin and factor Xa by forming thrombin-antithrombinn and factor Xa-antithrombin complexes. Antithrombin III is decreasedd after endotoxin infusion in dogs21 and during sepsis in humans due to increasedd consumption22 and degradation by elastase released from activated neutrophils.233 Low antithrombin III levels in DIC are associated with increased mortality.2 2

Inn addition to the decrease in antithrombin III, a significant down-regulation off the protein C-protein S system may occur. Activated protein C (APC) proteolyticallyy inactivates the co-factors factor Va and factor Villa, thereby rapidlyy and effectively impairing blood coagulation.24 Protein C is activated by thee complex of thrombin with the endothelial cell surface protein thrombomodulinn and activated protein C is dependent of its co-factor protein S. Theree are several explanations for an impairment of the protein C/protein S systemm in DIC. First, protein C and protein S levels are decreased in patients withh DIC,25 probably due to increased consumption. In addition pro-inflammatoryy cytokines can induce downregulation of thrombomodulin on endotheliall cells resulting in a decreased activation of protein C.26'27 Furthermore thee acute phase protein C4bBP, which can bind protein S, is increased during severee illness, leading to lower levels of the biologically active free protein S.

AA third natural anticoagulant pathway consists of tissue factor pathway inhibitorr (TFPI). Most of TFPI in the body is bound to the endothelium and can bee released into the blood, for example following heparin administration. Much off the circulating TFPI is bound to lipoproteins.28 TFPI is a direct factor Xa inhibitorr and, in a factor Xa dependent manner, produces feedback inhibition of thee factor Vila/tissue factor complex. In a primate model of sepsis TFPI levels increasedd 1.2-fold following sublethal and 2- fold following lethal E.coli infusion.299 Evidence for the importance of TFPI in sepsis was provided by a studyy in baboons that showed that TFPI infusion after the start of a lethal intravenouss E.coli infusion could prevent the activation of coagulation as well as deathh in all 5 animals studied.30

(22)

ChapterChapter 2 cytokines cytokines mononuclearmononuclear endothelial cellscells cells \\ / tissuee factor + + factorr Vll(a) factorr IX cytokines cytokines PAI-1 1 fibrinogen n plasmin n fibrinn degradation

Figuree 1. Schematic representation of the coagulation and fibrinolytic system in DIC.

Tissue-factorr dependent activation of coagulation occurs, ultimately leading to thrombin generationn and subsequent fibrinogen to fibrin conversion. The activation of coagulation iss further promoted by depression of all three natural anticoagulant pathways (i.e. TFPI, thee protein C/protein S system and antithrombin III). Simultaneously, fibrin degradation byy activation of the endogenous fibrinolytic system is impaired, due to high

concentrationss of PAI-1.

Fibrinolysis Fibrinolysis

Inn patients with DIC, deposition of fibrin in the (micro)vasculature is not onlyy due to formation of intravascular fibrin, but also due to inadequate removal.. This inadequate removal is caused by an impaired function of the fibrinolyticc system. Following endotoxin injection in healthy human subjects theree is a rapid, but short-lasting activation of fibrinolysis due to an increase in tissuee plasminogen activator (t-PA) and urokinase-type plaminogen activator (u-PA). .

Followingg this initial activation of fibrinolysis a complete and sustained inhibitionn of fibrinolysis can be observed due to an increase in plasminogen activatorr inhibitor type 1 (PAI-1). Several experimental and clinical

(23)

DrugDrug Treatment Strategies for DIC observationss have shown that at the time of maximal thrombin generation

fibrinolysisfibrinolysis is markedly inhibited. Thus, a remarkable imbalance between

coagulationn and fibrinolysis exists, resulting in a net procoagulant state.6,16

Antibodiess to tissue factor or the specific thrombin inhibitor recombinant hirudinn were able to completely block the endotoxin induced thrombin generationn in chimpanzees, but were without any effect on the activation of

fibrinolysis.fibrinolysis. Hence, inhibition of coagulation did not affect the stimulation and subsequentt inhibition of fibrinolysis, suggesting an independent regulation of

thesee two processes. Experimental studies have shown that the dysregulation of fibrinolysiss in DIC is completely mediated by TNF whereas other cytokines,

suchh as IL-6, have no effect.31

Treatmentt of DIC

Thee proper management of patients with DIC remains controversial. Unfortunately,, adequate clinical trials on DIC treatment are hardly available, probablyy due to the complexity of the syndrome and its variable and unpredictablee course. The clinical picture of simultaneously occurring systemic thromboticc depositions and bleeding due to consumption does not directly indicatee which specific therapy should be administered.

Itt is, however, well accepted that the cornerstone for the treatment of DIC is thee management of the underlying disorder. Besides, therapeutic interventions basedd on our present knowledge of the pathogenesis of DIC may be appropriate. Att present, these interventions may consist of plasma and platelet replacement therapy,, anticoagulant strategies or administration of physiologic coagulation inhibitors.. Future therapies may include interference in the fibrinolytic system.

PlasmaPlasma and platelets transfusion

Consumptionn of coagulation factors and platelets during DIC can increase thee risk of bleeding. Treatment with plasma or platelet concentrate is guided by thee clinical condition of the patient and should not be instituted on the basis of laboratoryy findings alone. Replacement may be indicated in patients with active bleeding,, in those requiring an invasive procedure or otherwise at risk for bleedingg complications. On the other hand, it has been suggested that transfusionn of blood components may also be harmful by further stimulating the activatedd coagulation system. This theory has rarely been proved to occur and simultaneouss (low-dose) heparin might be useful to prevent these

complications.32 2

(24)

ChapterChapter 2

Thee treatment with plasma is not based on evidence from controlled trials. Thee only randomised, controlled trial in neonates with DIC, comparing administrationn of fresh frozen plasma and platelets with whole blood exchange andd no specific therapy failed to show any change in outcome of DIC or

survival.333 Despite the lack of evidence most authors recommend treatment with

freshfresh frozen plasma, at least when patients are bleeding or at increased risk for

bleeding.1'34,355 To sufficiently correct the coagulation defect large volumes of

plasmaa may be needed. The use of coagulation factor concentrates may overcomee this need, however, beside the fact that these concentrates usually containn only a selected number of the various clotting factors, they may be contaminatedd with traces of activated coagulation factors and may therefore be particularlyy harmful for patients with DIC. Cryoprecipitate, which contains fibrinogenn as well as factor VIII, von Willebrand factor, factor XIII and fibronectin,, is also used as replacement therapy in DIC. However, its use is not supportedd by controlled trials. Because it is not possible to produce cryoprecipitatee without risk of Hepatitis C transmission, this product is not availablee in many countries.

AnticoagulantAnticoagulant strategies

Heparin Heparin

Heparinn has been used as treatment for DIC since 1959.36 Animal studies

havee shown that heparin can inhibit the activation of coagulation in

experimentall septicemia but does not affect mortality.37'38 Studies of heparin for

treatmentt of DIC in humans claimed to be succesfull, but were not controlled. Althoughh one of these uncontrolled studies concluded already in 1970 that a controlledd study giving heparin to patients with Gram-negative sepsis appeared

indicated,399 to date no such trial has been performed. A retrospective analysis of

casess of DIC reported in the literature found similar survival for patients treated

andd patients not treated with heparin.40 One can conclude that there is no sound

evidencee in favor of the use of heparin as routine therapy in patients with DIC. Ann exception may be made for patients with clinical signs of extensive fibrin depositionn like purpura fulminans, acral ischemia or venous thrombosis. In suchh cases low-dose heparin (5-8 U/kg/hr) is advocated, potentially in combinationn with plasma and -if appropriate- platelet replacement.

(25)

DrugDrug Treatment Strategies for DIC Low-molecularLow-molecular weight heparin

Low-molecularr weight heparins (LMWH) are fractions of heparin with a molecularr weight between 4000 and 8000 daltons. They differ from unfractionatedd heparins (UFH) in their higher anti-factor Xa to anti-factor Ha activityy ratio. This ratio varies between 2:1 and 4:1 for LMWH as compared to 1:11 for unfractionated heparins. It has been postulated that hese LMWHs would havee a decreased bleeding risk while have at least the same antithrombotic potentiall as unfractionated heparin. Effective treatment of DIC with LMWH has

beenn reported in rabbits.42 In rats LMWH was as effective as unfractionated

heparinn in improving the symptoms of DIC after endotoxin or thromboplastin

infusion.433 Succesful treatment was also reported in two small uncontrolled

studiess in humans.44,45 Furthermore, effects of dalteparin (Fragmin®) as

anti-DICC treatment have been studied in a multi-center double-blind randomised

trial.466 The underlying cause of DIC in most of these patients was malignancy

andd only 13% of patients suffered from infectious disease. Dalteparin was given inn a dose of 75 U/kg/day, heparin in a dose of 240 U/kg/day. In this study dalteparinn showed superior efficacy as compared with unfractionated heparin in improvingg bleeding symptoms and in improving a subjective organic symptoms score.. The improvement in survival in the dalteparin group was not significant. Theree was no difference in laboratory parameters for DIC between the two treatmentt groups. Hence, based on this single study, it may be postulated that LMWHH offers the benefit of decreased bleeding complications as compared to UFHH in the treatment of DIC.

Hirudin Hirudin

Hirudinn is a potent and specific direct thrombin inhibitor. In contrast to

heparin,, its activity is not dependent on antithrombin III.47 and therefore

recombinantt hirudin is capable of inhibiting clot-bound thrombin. Hirudin

appearedd to be effective in treating DIC in animal studies,48'50 and in one series

off 5 patients with haematological malignancy and DIC.51 However, randomized

controlledd trials on the use of hirudin in patients with DIC are not available. The highh risk of bleeding with hirudin treatment, as for example shown in initial

clinicall trials may potentially limit the use of hirudin in these patients.52

(26)

ChapterChapter 2

CoagulationCoagulation inhibitors

AntithrombinAntithrombin III concentrate

Antithrombinn III (AT III) is an important physiological inhibitor of blood

coagulation.. Low AT III levels are associated with increased mortality.2

Mortalityy due to Gram-negative sepsis could be prevented by AT III infusion in baboons,, but only if adequate AT III levels were achieved early in the course of sepsis.53 3

Inn humans three controlled clinical trials on the use of antithrombin III concentratee in DIC have been performed. One trial compared AT III infusion to aa synthetic protease inhibitor (Gabexate, FOY-007) in obstetric patients with DIC.. A single infusion of AT III was more effective in controlling the symptoms

off DIC (improvement in 92 vs 60% of patients).54

Blauhutt et al. studied 51 patients with shock and DIC.55 They compared

treatmentt with AT III concentrate alone to heparin treatment and to the administrationn of AT III concentrate plus heparin. No difference in survival was observedd between the groups, but the duration of symptoms of DIC was significantlyy shortened in the groups receiving AT III with or without heparin. Bleedingg complications were increased in the group treated with AT III in combinationn with heparin.

Thee third study by Fourrier and others compared the administration of

supraphysiologicc doses of AT III to placebo in patients with septic shock.56 In

thee AT III treated patients recovery from DIC was earlier and blood transfusion requirementt was less. A trend to decreased mortality was found, but statistical significancee was not reached.

Inn a review of the trials on AT III concentrate for the treatment of DIC, Vinazzer,, one of the co-authors of the study by Blauhut et al, showed the results off a follow-up study in 170 patients with shock In this follow-up study significantlyy less patients that were treated with AT III concentrate died as comparedd to treatment with heparin. However, the information on a number of importantt methodological issues regarding this follow-up study remain

unclear.577 In all above mentioned studies AT III concentrate was given to reach

normall or supranormal plasma levels and in the study of Fourrier et al. AT III concentrationss twice as high as usually found in the plasma were attained. A recentt publication suggested that much higher doses of AT III are needed to

achievee maximal beneficial effect.58

Thee conclusion from the published studies is that AT III is able to improve disseminatedd intravascular coagulation, but that the benefit in terms of clinical

(27)

DrugDrug Treatment Strategies for DIC outcomee is less certain. Since it can not be inferred from the literature which patientss will benefit in terms of increased survival or reduced morbidity, it seemss reasonable to reserve this expensive treatment to cases were mortality attributablee to DIC is expected to be high and to patients with very active DIC leadingg to substantial morbidity. An example of such cases is a patient suffering fromfrom meningococcal sepsis with purpura fulminans and acral ischemia. In such cases,, the aim of the treatment should be normal or supranormal antithrombin

concentrations.599 Future studies will probably indicate whether treatment with

higher,, supraphysiological, doses AT III will result in more favourable clinical outcomee in patients with DIC.

ProteinProtein C concentrate and Thrombomodulin

Ass mentioned above protein C levels are decreased during sepsis and clinicall and experimental evidence indicate that depression of the protein

C/proteinn S system may contribute to a fatal outcome.25,60,61 Based on these

observations,, supplementation of protein C may be of advantage in patients with DIC.. Indeed, in baboons protein C prevented the coagulopathy and lethal

effectss of E.coli infusion.62 Activated protein C also appeared effective in a

thromboplastin-inducedd DIC model in rabbits.63 There have been several reports

off succesful treatment with protein C in sepsis, both in children64,65 and in

adults.666 However no data from controlled clinical trials are available and routine

treatmentt of DIC with protein C concentrate can not be advocated yet. Clinical studiess with activated protein C concentrate are ongoing and may yield promisingg results.

Ann alternative strategy to increase the activity of the protein C system is the infusionn of thrombomodulin. In several animal models of DIC, treatment with solublee thrombomodulin not only showed a beneficial effect on coagulation, but alsoo appeared to improve the pulmonary vascular injury and pulmonary

accumulationn of white blood cells.43,67"69 These effects were not dependent on the

thrombin-bindingg properties of thrombomodulin, but probably mediated by the

increasee in activated protein C.69 Thusfar, no studies on thrombomodulin

treatmentt in humans with DIC have been reported.

InhibitorsInhibitors of tissue factor

Sincee tissue factor plays a key role in the initiation of coagulation during DIC,, inhibiting the actions of tissue factor could be of value in the treatment of intravascularr coagulation. In a rat model of DIC the infusion of recombinant

(28)

ChapterChapter 2

tissuee factor pathway inhibitor (TFPI) immediately after endotoxin administrationn significantly inhibited the consumption of coagulation factors andd platelets. Furthermore a reduced number of fibrin thrombi was formed in

liver,, lungs, kidney and spleen.70 Similar effects were found in a rabbit model of

DIC.711 Clinical trials on the use of TFPI in patients with DIC have recently been

initiated,, but results are not yet available. Also, other tissue factor-inhibiting agentss may prove to be potentially useful treatment strategies in patients with DIC. .

InterferenceInterference in the fibrinolytic system

Fibrinolyticc inhibitors, such as aprotinin or tranexaminic acid are usually contraindicatedd in patients with DIC. Although generally useful in bleeding patients,, in case of DIC these agents are thought to further block the already depressedd fibrinolytic system, thereby seriously promoting intravascular fibrin deposition.. An exception may be made in patients with a rarely occurring type off coagulation disorder associated with acute promyelocyte leukemia (AML-M3)) or sometimes with prostate carcinoma. In fact, in these situations rather primaryy hyperfibrinolysis than DIC is present, and in this situation fibrinolytic

inhibitorss may be very useful.72

Sincee the fibrinolytic shut-down in patients with DIC appears to be due to highh circulating levels of PAI-1, strategies directed against this fibrinolytic inhibitorr might be useful. Anti PAI-1 strategies have been shown to be of benefit

inn initial experimental studies,73 however, the efect of this treatment in clinical

studiess of DIC remains to be awaited. An alternative strategy to enhance fibrinolysiss is the administration of tissue type plasminogen activator (t-PA). Somee case reports have been published suggesting improvement of the clinical

conditionn of patients with meningococcemia and DIC after t-PA treatment,74,75

butt again, controlled clinical trials should be awaited before this treatment can bee advocated.

Conclusionn - Guidelines for therapy

Firstt of all, treatment of DIC should consist of optimal management of the underlyingg disease, e.g. antibiotic therapy and abcess drainage in septicemia. As mentionedd above firm evidence for any specific therapy directed at the coagulationn system for a patient with DIC is lacking. The following guidelines aree based as much as possible on the available data in the literature.

(29)

DrugDrug Treatment Strategies for DIC Inn the case of bleeding or high risk for bleeding we propose plasma and platelett replacement therapy. Depending on the levels of coagulation factors andd clotting times 2-3 units of plasma can be given initially followed by repeatedd transfusion depending on prothrombin time (PT) and activated partial thromboplastinn time (aPTT) values.

Whenn the decisision is made to give platelets transfusion one should aim at

levelss of 50-60 x 109/L. There is no evidence that treatment with heparin or

low-molecularr weight heparin is beneficial. We reserve heparin treatment for those casess with clinical signs of extensive fibrin deposition like patients with purpura fulminans.. In those patients 300-500 U/hour can be given intravenously. Finally,, we start replacement therapy with AT III in patients with severe DIC andd low levels of circulating AT III. In those circumstances replacement should bee aimed at reaching normal or supranormal AT III levels. The dose of AT III concentratee to reach normal levels can be calculated by the formula: Dose

(units)) = (100-measured AT III activity) x kg body weight.59

References s

1.. Baglin T. Disseminated intravascular coagulation: diagnosis and treatment. BMJ. 1996;312:683-7. .

2.. Fourrier F, Chopin C, Goudemand J et al. Septic shock, multiple organ failure, and disseminatedd intravascular coagulation. Compared patterns of antithrombin III, protein C,, and protein S deficiencies. Chest. 1992;101:816-23.

3.. Levi M, ten Cate H, van der Poll T et al. Pathogenesis of disseminated intravascular coagulationn in sepsis. JAMA 1993;270:975-9.

4.. Bick RL. Disseminated intravascular coagulation: objective clinical and laboratory diagnosis,, treatment, and assessment of therapeutic response. Semin Thromb Hemost 1996;22:69-88. .

5.. Levi M, van der Poll T, ten Cate H et al. The cytokine-mediated imbalance between coagulantt and anticoagulant mechanisms in sepsis and endotoxaemia. Eur J Clin Invest

1997;27:3-9. .

6.. van Deventer SJ, Buller HR, ten Cate JW et al. Experimental endotoxemia in humans: analysiss of cytokine release and coagulation, fibrinolytic, and complement pathways. Bloodd 1990;76:2520-6.

7.. Levi M, ten Cate H, Bauer KA et al. Inhibition of endotoxin-induced activation of coagulationn and fibrinolysis by pentoxifylline or by a monoclonal anti-tissue factor antibodyy in chimpanzees. J Clin Invest 1994;93:114-20.

8.. Nuijens JH, Huijbregts CC, Eerenberg-Belmer AJ et al. Quantification of plasma factor XIIa-Cl(-)-inhibitorr and kallikrein-Cl(-)-inhibitor complexes in sepsis. Blood

1988;72:1841-8. .

9.. Pixley RA, De La Cadena R, Page JD et al. The contact system contributes to

(30)

ChapterChapter 2

hypotensionn but not disseminated intravascular coagulation in lethal bacteremia. In vivo usee of a monoclonal anti-factor XII antibody to block contact activation in baboons. J Clinn Invest 1993;91:61-8.

10.. Colucci M, Balconi G, Lorenzet R et al. Cultured human endothelial cells generate tissue factorr in response to endotoxin. J Clin Invest 1983;71:1893-6.

11.. Bevilacqua MP, Pober JS, Majeau GR et al. Recombinant tumor necrosis factor induces procoagulantt activity in cultured human vascular endothelium: characterization and comparisonn with the actions of interleukin 1. Proc Natl Acad Sci USA 1986;83:4533-7. 12.. Taylor FB Jr, Chang A, Ruf W et al. Lethal E.coli septic shock is prevented by blocking

tissuee factor with monoclonal antibody. Circ Shock 1991;33:127-34.

13.. Biemond BJ, Levi M, ten Cate H et al. Complete inhibition of endotoxin-induced coagulationn in chimpanzees with monoclonal Fab fragment against factor VII/VIIa. Thrombb Haemost 1995;73:223-30.

14.. van der Poll T, Buller HR, ten Cate H et al. Activation of coagulation after administrationn of tumor necrosis factor to normal subjects. N Engl J Med 1990;322:1622-7. .

15.. van der Poll T, Levi M, van Deventer SJ et al. Differential effects of anti-tumor necrosis factorr monoclonal antibodies on systemic inflammatory responses in experimental endotoxemiaa in chimpanzees. Blood 1994;83:446-51.

16.. van der Poll T, Levi M, Hack CE et al. Elimination of interleukin 6 attenuates coagulationn activation in experimental endotoxemia in chimpanzees. J Exp Med 1994;179:1253-9. .

17.. Stouthard JM, Levi M, Hack CE et al. Interleukin-6 stimulates coagulation, not fibrinolysis,, in humans. Thromb Haemost 1996;76:738-42.

18.. Mestries JC, Kruithof EK, Gascon MP et al. In vivo modulation of coagulation and fibrinolysiss by recombinant glycosylated human interleukin-6 in baboons. Eur Cytokine Netww 1994;5:275-81.

19.. Boermeester MA, van Leeuwen PA, Coyle SM et al. Interleukin-I blockade attenuates mediatorr release and dysregulation of the hemostatic mechanism during human sepsis. Archh Surg 1995;130:739-48.

20.. Jansen PM, Boermeester MA, Fischer E et al. Contribution of interleukin-1 to activation off coagulation and fibrinolysis, neutrophil degranulation, and the release of secretory-typee phospholipase A2 in sepsis: studies in nonhuman primates after interleukin-1 alpha administrationn and during lethal bacteremia.Blood 1995;86:1027-34.

21.. Tanaka H, Kobayashi N, Maekawa T. Studies on production of antithrombin III with speciall reference to endotoxin-induced DIC in rats. Thromb Haemost 1986;56:137-43. 22.. Buller HR, ten Cate JW. Acquired antithrombin III deficiency: laboratory diagnosis,

incidence,, clinical implications, and treatment with antithrombin III concentrate. Am J Medl989;87:44S-48S. .

23.. Seitz R, Wolf M, Egbring R et al. The disturbance of hemostasis in septic shock: role of neutrophill elastase and thrombin, effects of antithrombin and plasma substitution.Eur J Haematoll 1989;43:22-8.

24.. Furie B, Furie BC. Molecular and cellular biology of blood coagulation. N Engl J Med 1992;326:800-6. .

25.. Hesselvik JF, Malm J, Dahlback B et al. Protein C, protein S and C4b-binding protein in

(31)

severee infection and septic shock. Thromb Haemost 1991;65:126-9.

26.. Nawroth PP, Handley DA, Esmon CT et al. Interleukin 1 induces endothelial cell procoagulantt while suppressing cell-surface anticoagulant activity. Proc Natl Acad Sci USAA 1986;83:3460-4.

27.. Nawroth PP, Stern DM. Modulation of endothelial cell hemostatic properties by tumor necrosiss factor. J Exp Med 1986;163:740-5.

28.. Broze GJ Jr. Tissue factor pathway inhibitor. Thromb Haemost 1995;74:90-3.

29.. Sabharwal AK, Bajaj SP, Ameri A et al. Tissue factor pathway inhibitor and von Willebrandd factor antigen levels in adult respiratory distress syndrome and in a primate modell of sepsis. Am J Respir Crit Care Med 1995;151:758-67.

30.. Creasey AA, Chang AC, Feigen L et al. Tissue factor pathway inhibitor reduces mortalityy from Escherichia coli septic shock. J Clin Invest 1993;91:2850-6.

31.. Biemond BJ, Levi M, ten Cate H et al. Plasminogen activator and plasminogen activator inhibitorr I release during experimental endotoxaemia in chimpanzees: effect of interventionss in the cytokine and coagulation cascades. Clin Sci 1995;88:587-94. 32.. Wong VK, Hitchcock W, Mason WH et al. Meningococcal infection in children: a

revieww of 100 cases. Pediatr Infect Dis J 1989;8:224-7.

33.. Gross SJ, Filston HC. Controlled study of treatment for disseminated intravascular coagulationn in the neonate. J Pediatr 1982;100:445-8.

34.. Rubin RN, Colman RW. Disseminated intravascular coagulation. Approach to treatment. Drugss 1992;44:963-71.

35.. Hiller E, Heim M. Indikationen fur die therapie mit frischgefrorenem plasma. Dtsch Med Wochenschrr 1989;114:1371-4.

36.. Little JR. Purpura fulminans treated succesfully with anticoagulation: report of a case. JAMAA 1959;169:36-40.

37.. Gaskins RA Jr, Dalldorf FG. Experimental meningococcal septicemia. Effect of heparin therapy.. Arch Pathol Lab Med. 1976;100:318-24.

38.. Corrigan JJ Jr, Kiernat JF. Effect of heparin in experimental gram-negative septicemia. J Infectt Dis 1975;131:138-43.

39.. Corrigan JJ Jr, Jordan CM. Heparin therapy in septicemia with disseminated intravascularr coagulation. Effect on mortality and on correction of hemostatic defects. N Engll J Med 1970;283:778-82.

40.. Corrigan JJ Jr. Heparin therapy in bacterial septicemia. J Pediatrics 1977;91:695-700. 41.. Feinstein DI. Diagnosis and management of disseminated intravascular coagulation: the

rolee of heparin therapy. Blood 1982;60:284-7.

42.. Tazawa S, Ichikawa K, Misawa K et al. Effects of low molecular weight heparin on a severelyy antithrombin Ill-decreased disseminated intravascular coagulation model in rabbits.. Thromb Res 1995;80:391-8.

43.. Takahashi Y, Hosaka Y, Imada K et al. Human urinary soluble thrombomodulin (MR-33)) improves disseminated intravascular coagulation without affecting bleeding time in rats:: comparison with low molecular weight heparin. Thromb Haemost 1997;77:789-95. 44.. Audibert G, Lambert H, Toulemonde F et al. Utilisation d'une heparine de bas poids

moleculaire,, la CY 222, dans Ie traitement des coagulopathies de consommation. J Mai Vasee 1987;12 suppl b:147-51.

45.. Gillis S, Dann E J, Eldor A. Low molecular weight heparin in the prophylaxis and

(32)

ChapterChapter 2

treatmentt of disseminated intravascular coagulation in acute promyelocytic leukemia. Eurr J Haematol 1995;54:59-60.

46.. Sakuragawa N, Hasegawa H, Maki M et al. Clinical evaluation of low-molecular-weight heparinn (FR-860) on disseminated intravascular coagulation (DlC)-a multicenter co-operativee double-blind trial in comparisonn with heparin. Thromb Res 1993;72:475-500. 47.. Weitz JI, Hudoba M, Massel D et al. Clot-bound thrombin is protected from inhibition

byy heparin-antithrombin III but is susceptible to inactivation by antithrombin Ill-independentt inhibitors. J Clin Invest 1990;86:385-91

48.. Freund M, Cazenave JP, Courtney M et al. Inhibition by recombinant hirudins of experimentall venous thrombosis and disseminated intravascular coagulation induced by tissuee factor in rats. Thromb Haemost 1990;63:187-92

49.. Zawilska K, Zozulinska M, Turowiecka Z et al. The effect of a long-acting recombinant hirudinn (PEG-hirudin) on experimental disseminated intravascular coagulation (DIC) in rabbits.. Thromb Res 1993;69:315-20.

50.. Dickneite G, Czech J. Combination of antibiotic treatment with the thrombin inhibitor recombinantt hirudin for the therapy of experimental Klebsiella pneumoniae sepsis. Thrombb Haemost 1994;71:768-72.

51.. Saito M, Asakura H, Jokaji H et al. Recombinant hirudin for the treatment of disseminatedd intravascular coagulation in patients with haematological malignancy. Bloodd Coagul Fibrinolysis 1995;6:60-4.

52.. The global use of strategies to open occluded coronary arteries (GUSTO) lib investigators.. A comparison of recombinant hirudin with heparin for the treatment of acutee coronary syndromes. N Engl J Med 1996;335(11):775-82.

53.. Taylor FB Jr, Emerson TE Jr, Jordan R et al. Antithrombin-111 prevents the lethal effects off Escherichia coli infusion in baboons. Circ Shock 1988;26:227-35.

54.. Maki M, Terao T, Ikenoue T et al. Clinical evaluation of antithrombin III concentrate (Bl 6.013)) for disseminated intravascular coagulation in obstetrics. Well-controlled multicenterr trial. Gynecol Obstetr Invest 1987;23:230-40.

55.. Blauhut B, Kramar H, Vinazzer H et al. Substitution of antithrombin III in shock and DIC:: a randomized study. Thromb Res 1985;39:81-9.

56.. Fourrier F, Chopin C, Huart JJ et al. Double-blind, placebo-controlled trial of antithrombinn III concentrates in septic shock with disseminated intravascular coagulation.. Chest 1993;104:882-8.

57.. Vinazzer HA. Antithrombin III in shock and disseminated intravascular coagulation. Clinn Appl Thrombosis/Haemostasis 1995;1:62-5.

58.. Kessler CM, Tang Z, Jacobs HM et al. The suprapharmacologic dosing of antithrombin concentratee for Staphylococcus aureus induced disseminated intravascular coagulation in guineaa pigs: substantial reduction in mortality and morbidity. Blood 1997;89:4393-401. 59.. Blauhut B, Necek S, Vinazzer H et al. Substitution therapy with antithrombin III in

shockk and DIC. Thromb Res 1982;27:271-78.

60.. Taylor FB Jr, Chang AC, Peer GT et al. DEGR-factor Xa blocks disseminated intravascularr coagulation initiated by Escherichia coli without preventing shock or organ damage.. Blood 1991 ;78:364-8.

61.. Fijnvandraad K, Derkx B, Peters M et al. Coagulation activation and tissue necrosis in meningococcall septic shock: severely reduced protein C levels predict a high mortality.

(33)

Thrombb Haemost 1995 ;73:15-20.

62.. Taylor FB Jr, Chang A, Esmon CT et al. Protein C prevents the coagulopatic and lethal effectss of Escherichia coli infusion in the baboon. J Clin Invest 1987;79:918-25. 63.. Katsura Y, Aoki K, Tanabe H et al. Characteristic effects of activated human protein C

onn tissue thromboplastin-induced disseminated intravascular coagulation in rabbits. Thrombb Res 1994;76:353-62.

64.. Gerson WT, Dickerman JD, Bovill EG et al. Severe aquired protein C deficiency in purpuraa fulminans associated with disseminated intravascular coagulation: treatment withh protein C concentrate. Pediatrics 1993;91:418-22.

65.. Dreyfus M, Magny JF, Bridey F et al. Treatment of homozygous protein C deficiency andd neonatal purpura fulminans with a purified protein C concentrate. N Engl J Med 1991;325:1565-8. .

66.. Rintala E, Seppala O, Kotilainen P et al. Protein C in the treatment of coagulopathy in meningococcall disease. Lancet 1996;347:1767.

67.. Gonda Y, Hirata S, Saitoh H et al. Antithrombotic effect of recombinant human soluble thrombomodulinn on endotoxin-induced disseminated intravascular coagulation in rats. Thrombb Res 1993;71:325-35.

68.. Aoki Y, Ohishi R, Takei R et al. Effects of recombinant human soluble thrombomodulin (rhs-TM)) on a rat model of disseminated intravascular coagulation with decreased levels off antithrombin m. Thromb Haemost 1994;71:452-5.

69.. Uchiba M, Okajima K, Murakami K et al. Effect of human urinary thrombomodulin on endotoxin-inducedd intravascular coagulation and pulmonary vascular injury in rats. Am J Hematoll 1997;54:118-23.

70.. Elsayed YA, Nakagawa K, Kamikubo YI. Et al. Effects of recombinant human tissue factorr pathway inhibitor on thrombus formation and its in vivo distribution in a rat DIC model.. Am J Clin Pathol 1996;106:574-83.

71.. Bregengard C, Nordfang O, Wildgoose P et al. The effect of two-domain tissue factor pathwayy inhibitor on endotoxin-induced disseminated intravascular coagulation in rabbits.. Blood Coagul Fibrinolysis 1993;4:699-706.

72.. Awisati G, ten Cate JW, Buller HR et al. Tranexaminic acid for control of haemorrhage inn acute promyelocytic leukaemia. Lancet;1989;ii(8655): 122-4.

73.. Levi M, Biemond BJ, van Zonneveld AJ et al. Inhibition of plasminogen activator inhibitor-11 (PAI-1) activity results in promotion of endogenous fibrinolysis and inhibitionn of thrombosis in experimental models. Circulation 1992;83:305-12.

74.. Zenz W, Muntean W, Zobel G et al Treatment of fulminant meningococcemia with recombinantt tissue plasminogen activator. Tromb Haemost 1995;74:802-3.

75.. Aiuto LT, Barone SR, Cohen PS et al. Recombinant tissue plasminogen activator restoress perfusion in meningococcal purpura fulminans. Crit Care Med 1997 ;25:1079-82. .

(34)

(35)

ChapterChapter 3

Coagulationn abnormalities in sepsis: relation with

inflammatoryy responses

Evertt de Jonge1, Marcel Levi2, Tom van der Poll3

Departmentss of (1) Intensive Care, (2) Vascular Medicine and (3) Experimental Internall Medicine. Academic Medical Center, University of Amsterdam,

thee Netherlands.

(36)

ChapterChapter 3

Abstract t

Sepsiss can be associated with profound alterations in the coagulation and fibrinolyticc system. In this article we discuss recent insights in the pathogenetic mechanismss of the activation of coagulation and we will focus on the relations off the coagulation, fibrinolytic and inflammatory pathways. Moreover, we will addresss the role of natural inhibitors of coagulation like protein C, antithrombin andd tissue factor pathway inhibitor (TFPI). Finally, we will speculate on how thesee insights may translate into new treatment strategies for sepsis.

(37)

CoagulationCoagulation and inflammatory responses in sepsis

Introduction n

Disseminatedd intravascular coagulation (DIC) is a common feature of severee infection. Massive activation of the coagulation system may result in generationn and deposition of fibrin, leading to microvascular thrombi in various organss and contributing to the development of multi-organ failure. It also results inn depletion of coagulation factors and platelets, thereby paradoxically increasingg the risk of bleeding. In patients with septic shock, DIC was shown to

bee a strong predictor of death.1 Knowledge of the mechanisms involved in the

procoagulantt changes in sepsis and in the ways by which the coagulation systemm can interact to modulate the inflammatory response has markedly increasedd and future treatment strategies may be based on these insights. In this overvieww we will focus on data mainly derived from in vivo studies in humans andd nonhuman primates that have contributed considerably to our understanding off these mechanisms and we will speculate on how this may alter treatment of sepsiss and DIC.

Activationn of coagulation

Thee activation of coagulation during severe infection appears to be mediatedd by cytokines, which are produced by the host in response to various pathogeneticc insults. For example, in sepsis, the activation of coagulation is initiatedd by microorganisms and their products like endotoxin or exotoxins and byy cytokines, produced by different cells in response to microbial mediators. Thee derangement of the coagulation system comprises enhanced activation of coagulation,, depression of inhibitory mechanisms of coagulation, and inhibition

off the fibrinolytic system.2 Most of the current insight in those pathogenetic

pathwayss has been derived from experimental studies of bacteremia or endotoxemiaa in humans or nonhuman primates. The intravenous administration off endotoxin to human subjects or primates results in the activation of coagulationn as reflected by elevation in markers for thrombin generation such as

thrombin-antithrombinn complexes and prothrombin fragment F l+2.3;4 Although

thee intrinsic (contact system-dependent) pathway of coagulation may be

activatedd during sepsis, it seems not to be involved in the initiation of DIC.5;6

Thiss system, however, may play an important role in the pathogenesis of systemicc hypotension, presumably by means of kinin generation. Thrombin generationn in DIC was shown to be mediated exclusively by the (extrinsic) tissuee factor/factor Vila-dependent pathway.

(38)

ChapterChapter 3

Thrombin n

Endotheliall cells

Prothrombin n

Cytokines s

Figuree 1. Schematic representation of the coagulation system during sepsis. The naturall inhibitors of coagulation are shown in grey boxes. Cytokine-induced tissue factor (TF)) expression on endothelial cells leads to activation of factor X and factor IX and subsequentt thrombin generation. The activation of coagulation is further promoted by relativelyy insufficient inhibition of TF by tissue factor pathway inhibitor (TFPI) and by decreasedd circulating levels of antithrombin (AT). Activated protein C (APC) levels are loww due to down-regulation of thrombomodulin (TM) and by decreased plasma levels of proteinn C and its co-factor protein S.

Underr physiological conditions, tissue factor (TF) can not be detected on

thee luminal surface of the vascular endothelium7 and only in very low quantities

onn circulating blood cells.8"10 However, during infection and after stimulation

withh endotoxin or tumor necrosis factor (TNF), TF can be rapidly induced on

bloodd mononuclear cells8;11;12 and on vascular endothelium.13"15 Our group

recentlyy found, that thrombin generation in humans injected with endotoxin was preceededd by an increase in TF mRNA levels in circulating blood cells (Franco ett al, manuscript submitted). Furthermore, the importance of TF in the pathogenesiss of DIC was confirmed by observations that the coagulant response uponn bacteremia or endotoxemia could be completely blocked by administration off monoclonal antibodies that are able to inhibit tissue factor or factor Vila 36 6

(39)

activityy .4;16;l7 The endotoxin-induced activation of the tissue factor system and

subsequentt activation of coagulation appears to be mediated by pro-inflammatoryy cytokines like TNF, interleukin (IL)-l and IL-6. TNF administrationn to healthy volunteers elicited a rapid activation of coagulation,

whichh was similar to that evoked by microorganisms or endotoxin.18 However,

endogenouss TNF appears not to be required for a procoagulant response during endotoxemia,, since neutralization of TNF activity by monoclonal antibodies directedd against TNF or a recombinant TNF receptor IgG fusion protein did not

affectt endotoxin-induced coagulation activation in humans or chimpanzees.19"21

Inn contrast to this observation, monoclonal antibodies directed against IL-6 were ablee to completely block the endotoxin-induced activation of coagulation in

chimpanzees.222 In addition, it was shown that IL-6 infusion in baboons and in

humann cancer patients induced thrombin generation.23124 Hence, these data

suggestt that IL-6 rather than TNF is the primary mediator for the induction of coagulationn in sepsis. The role of other cytokines, such as IL-1 is less clear.

Administrationn of IL-1 to baboons resulted in systemic coagulation activation25

andd treatment of septic baboons as well as septic human patients with IL-1

receptorr antagonist significantly attenuated the activation of coagulation.25126 It

is,, however, not clear whether IL-1 has a direct effect on coagulation, or indirect,, mediated by other IL-1 induced cytokines.

Fibrinolysis s

Inn patients with DIC, deposition of fibrin in the (micro)vasculature is due nott only to increased formation (coagulation), but also to inadequate removal. Thiss inadequate removal is caused by an impaired function of the fibrinolytic system.. Septic patients with DIC invariably show some activation of fibrinolysis ass indicated by elevated levels of plasmin-a2-antiplasmin complexes and fibrin degradationn products as well as decreasing levels of plasminogen and

a2-antiplasminn due to ongoing consumption of these factors.27"29 There is, however,

increasingg evidence that the activation of the fibrinolytic system is relatively insufficientt to adequately compensate for the systemic activation of coagulation, thuss leading to widespread deposition of fibrin in the microvasculature of

variouss organs.30;31 Furthermore, in patients with septic shock, the ratio of

plasmaa levels of thrombin-antithrombin complexes to plasmin-a2-antiplasmin

complexess was higher in non-survivors as compared with survivors32"34

suggestingg that an imbalance between the coagulant and fibrinolytic response contributess to the fibrin deposition in DIC.

(40)

ChapterChapter 3

Thee pathogenetic mechanism underlying the fibrinolytic response in DIC hass been greatly elucidated in experimental studies of human endotoxemia and bacteremiaa in animals. Endotoxemia results in an early release of tissue-type plasminogenn activator (t-PA) and urokinase-type plasminogen activator (u-PA), associatedd with the generation of plasmin, as evidenced by the simultaneously occurringg transient elevation of plasmin-a2-antiplasmin complexes. However, afterr a delay of approximately 60 minutes, fibrinolysis is strongly inhibited by a

sustainedd increase in plasminogen activator inhibitor-1 (PAI-1) levels.3;4;35

Accordingly,, PAI-1 levels are high during sepsis and DIC and are positively

correlatedd with severity.36"39 The fibrinolytic response following endotoxemia is

nott influenced by anticoagulant strategies like hirudin, tissue factor pathway

inhibitorr (TFPI) and monoclonal antibodies against factor VIIa,17;40;41 indicating

thatt thrombin is not a pivotal mediator of endotoxin-induced fibrinolysis. All thesee endotoxin-induced changes in the fibrinolytic system appear to be

mediatedd by TNF and can be abolished by neutralization of TNF.19 Recently

TAFII (thrombin-activatable fibrinolytic inhibitor) has been identified as another endogenouss antifibrinolytic agent that could contribute to the inhibition of fibrinolysiss during sepsis. Activation of TAFI appears to depend on a factor XIa-generatedd thrombin driven amplification pathway, which has been shown to

existt during endotoxemia.42

Plasminogenn activators

Cytokines s

PAI-1 1

_{Fibrin n}

Plasminogen n

Plasmin n

Fibrin n

degradation n

Figuree 2. Fibrin degradation by activation of the endogenous fibrinolytic system is

impairedd during sepsis, due to high concentrations of plasminogen activator inhibitor typee 1 (PAI-1).

(41)

Naturall inhibitors of coagulation

Too protect against uncontrolled activation of coagulation, the human coagulationn system is regulated at various levels by three major anticoagulant systems:: antithrombin, the protein C system and tissue factor pathway inhibitor (TFPI).. Antithrombin (formerly known as antithrombin III) binds and inactivatess thrombin and factor Xa. Levels of antithrombin are decreased during sepsiss due to increased consumption, decreased hepatic synthesis and

degradationn by elastase released by activated neutrophils.43"45 Low antithrombin

levelss in DIC are associated with increased mortality.' A marked impairment of thee protein C system may further compromise the regulation of activated coagulation.. Protein C is activated by the complex of thrombin with the endotheliall cell surface protein thrombomodulin and activated protein C (APC) activityy is facilitated by its cofactor protein S. APC proteolytically inactivates thee cofactors factor Va and factor Villa, thereby rapidly inactivating blood

coagulation.466 The impairment of the protein C system during sepsis is the result

off increased consumption of protein S and protein C,47 and decreased activation

off protein C by downregulation of thrombomodulin on endothelial cells.I5;48

Furthermore,, protein S will complex with the acute phase response protein

C4b-bindingg protein.47 A third natural anticoagulant pathway consists of TFPI. TFPI

iss an approximately 43-KD, trivalent, Kunitz-type inhibitor that directly inhibits factorr Xa and, in a factor Xa dependent manner, produces feedback inhibition of

thee TF/factor Vila complex.49 Most of total body TFPI is located in association

withh endothelial cells and only 10-25% is found in circulating blood. Circulating

TFPII is predominantly bound to lipoproteins.50 Platelets also carry native TFPI

(aboutt 10% of the plasma pool), which is released following stimulation by

thrombin.511 In a primate model of sepsis TFPI increased 1.2-fold following

sub-lethall and 2-fold following lethal E.coli infusion.52 Evidence for the importance

off TFPI in sepsis was provided by studies in baboons that showed that TFPI couldd prevent the activation of coagulation as well as death in a lethal E.coli

model.53;544 We recently found that TFPI , in a dose-dependent manner, could

completelyy prevent the endotoxin-induced activation of coagulation in healthy

humans.40 0

(42)

ChapterChapter 3

Antiinflammatoryy effects of natural anticoagulants

Experimentall studies using lethal primate models of sepsis have shown that treatmentss with the natural inhibitors of coagulation i.e. antithrombin, TFPI and proteinn C are effective in preventing the systemic activation of coagulation as welll as the proinflammatory response and lethal effects of E.coli bacteremia. '

577

The diminished inflammatory response could be a direct result of preventing coagulationn since it has been shown that different activated coagulation factors

likee thrombin, factor Vila and Xa can activate cells to release cytokines.58

However,, alternative anticoagulant treatments with heparin59 and active

site-inhibitedd factor Xa60 also effectively prevented the activation of coagulation in

lethall primate models of sepsis, but without effect on lethality. Thus, it appears thatt other, non-anticoagulant functions of the natural coagulation inhibitors are involvedd in their modulation of the inflammatory response. Indeed, activated

proteinn C has been shown to inhibit monocyte activation by endotoxin.61 The

62 2

tissuee factor/factor Vila complex can also elicit a proinflammatory response , therebyy offering an alternative explanation why interference very proximal in thee coagulation cascade (e.g. with TFPI) does prevent sepsis-induced cytokine release,, whereas more downstream interventions (e.g. with active site degraded factorr Xa) fail to do so. The interaction of antithrombin with endothelial cells mayy be the key to the ability of antithrombin to promote survival. Whereas antithrombinn was effective in preventing death in primates challenged with

E.coliE.coli bacteremia,55556 antithrombin-heparin complex did not protect against

lethalityy in a porcine model of endotoxin-induced DIC.63 It has been suggested

thatt antithrombin modulates the inflammatory response by binding to the endotheliumm via cell surface heparan sulfate proteoglycans, possibly by

promotingg release of prostacyclin.56 Summarizing, the endogenous coagulation

inhibitorss can modulate the inflammatory response, at least partly independent fromfrom effects on coagulation. The potential importance hereof has been illustrated byy the observation that a 1-antitrypsin Pittsburgh, a mutation of the normal proteasee which is a potent inhibitor of both thrombin and activated protein C, inhibitedd the activation of coagulation in septic baboons but also decreased

survivall compared to controls.64 Apparently, there was an exaggerated

inflammatoryy response despite adequate blockade of coagulation, potentially by inhibitionn of the protein C pathway.