!  "    #   "          $ # #  

%& '()" (*+,-)

"$" ) .-, $"$  /"* -)  //"*" 0 Dissertation for the Degree of Doctor of Philosophy (Faculty of Medicine) in Cardiology presented at Uppsala University in 2001

ABSTRACT

Oldgren, J. 2001. Inflammation and coagulation activity in unstable coronary artery disease and the influences of thrombin inhibition. Acta Universitatis Upsaliensis. Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1103. 68 pp. Uppsala. ISBN 91-554-5182-9.

In patients with unstable coronary artery disease, this study evaluated the degree of inflammation and coagulation activity, relations to myocardial cell damage, prognosis, and influences of randomisation to 72 h infusion with three different doses of inogatran, a direct thrombin inhibitor (n=904), or unfractionated heparin (n=305). Anticoagulant treatment effects were evaluated with aPT time. In inogatran treated patients with aPT times ³ 44 s (median), the 7-days event rate - death, myocardial infarction or refractory angina – was 11.6%, compared to 6.6% with aPT times < 44 s (p=0.01). Higher aPT times was related to improved outcome during heparin treatment. Markers of inflammation, i.e. fibrinogen and C-reactive protein (CRP), and coagulation, i.e. prothrombin fragment 1+2 (F1+2), thrombin-antithrombin complex (TAT), soluble fibrin (SF) and D-dimer were analysed in serial samples (n=320). High fibrinogen, F1+2 and D-dimer levels persisted at 30 days. Patients with myocardial damage, detected by elevated troponin, had higher levels of all markers except TAT. Ischemic events occurred at 30 days in 17% of patients with high (pre-treatment top tertile) and 8.5% of patients with lower fibrinogen levels (p=0.03), while high CRP levels only were related to increased mortality. At 30 days, patients with high compared to low pre-treatment levels of TAT or SF had 40% lower event rate. Patients with early decreased compared to raised F1+2 or TAT levels during treatment had 50% lower 30-days event rate (p<0.05). Conclusions: The aPT time is an inappropriate indicator of antithrombotic efficacy. The raise in fibrinogen in the acute phase is sustained, and indicates risk of thrombosis and new ischemic events. The pronounced CRP elevation is transient, but associated with increased mortality. Higher coagulation activity may identify patients with a thrombotic condition as the major cause of instability, who are best responders to anticoagulant therapy. However, reactivation of coagulation activity with raised risk of ischemic events is a concern at cessation of treatment. Key words: Unstable coronary artery disease, inflammation, coagulation, thrombin inhibition.

Jonas Oldgren, Department of Medical Sciences, Cardiology, University Hospital, SE-751 85 Uppsala, Sweden

© Jonas Oldgren 2001 ISSN 0282-7476 ISBN 91-554-5182-9 Printed in Sweden by Uppsala University, Tryck & Medier, Uppsala 2001

2 Man lives with arteriosclerosis, and dies of the complicating thrombosis

Jens Dedichen, 1956

3 Papers

This thesis is based on the following original papers which will be referred to by their Roman numerals:

I. Activated Partial Thromboplastin Time and Clinical Outcome after Thrombin Inhibition in Unstable Coronary Artery Disease Oldgren J, Linder R, Grip L, Siegbahn A, Wallentin L Eur Heart J 1999;20(22):1657-66.

II. Myocardial Damage, Inflammation and Thrombin Inhibition in Unstable Coronary Artery Disease Oldgren J, Wallentin L, Grip L, Linder R, Nørgaard B, Siegbahn A In progress

III. The Effect of a Low Molecular Mass Thrombin Inhibitor, Inogatran, and Heparin on Thrombin Generation and Fibrin Turnover in Patients with Unstable Coronary Artery Disease Linder R, Oldgren J, Egberg N, Grip L, Larson G, Siegbahn A, Wallentin L Eur Heart J 1999;20(7):506-18

IV. Coagulation Activity and Clinical Outcome in Unstable Coronary Artery Disease Oldgren J, Linder R, Grip L, Siegbahn A, Wallentin L Arterioscler Thromb Vasc Biol 2001;21(6):1059-64.

V. Myocardial Damage, Coagulation Activity and the Response to Thrombin Inhibition in Unstable Coronary Artery Disease Oldgren J, Siegbahn A, Grip L, Linder R, Thygesen K, Wallentin L Submitted

Reprints were made with permission from the publishers.

4 Contents

Abbreviations...... 6 Introduction and background...... 7 Definition and clinical manifestations of unstable coronary artery disease ...... 7 Atherosclerosis and plaque disruption...... 7 Inflammation and markers of inflammation ...... 8 Haemostasis and molecular markers of coagulation activity ...... 9 Myocardial infarction and biochemical markers of myocardial cell injury ...... 13 Prognosis and risk stratification in unstable coronary artery disease ...... 13 Treatment of unstable coronary artery disease ...... 14 Aims of the study ...... 17 Material and methods ...... 19 Patients and design ...... 19 aPT time and dose adjustments ...... 19 Inogatran plasma concentrations and thrombin time ...... 20 Substudy blood samples ...... 20 Markers of inflammation...... 20 Molecular markers of coagulation activity ...... 20 Marker of myocardial cell injury...... 21 End-points ...... 21 Statistics ...... 22 Results...... 24 The TRIM study...... 24 aPT time and the influence of inogatran or UF heparin ...... 25 Clinical outcome in relation to aPT times ...... 27 Baseline characteristics and the degree of inflammation and coagulation activity ...... 29 The influences of thrombin inhibition on inflammation and coagulation activity ...... 30 Signs of myocardial cell injury and inflammation and coagulation activity ...... 33 Clinical outcome in relation to the degree of inflammation and coagulation activity.. 37 Discussion ...... 44 Activated partial thromboplastin time ...... 44 The influences of thrombin inhibition on inflammation and coagulation activity ...... 46 Inflammation and coagulation activity in relation to myocardial cell injury ...... 48 The degree of inflammation and clinical outcome ...... 49 The degree of coagulation activity and clinical outcome ...... 51 Limitations ...... 53 Summary and implications ...... 54 Acknowledgements...... 56 References ...... 58

5 Abbreviations

aPT activated partial thromboplastin

ASA acetylsalicylic acid

CK-MB MB isoenzyme of creatine kinase

ECG electrocardiogram

ELISA enzyme-linked immunoassay

F1+2 prothrombin fragment 1+2

LMW low molecular weight

TAT thrombin-antithrombin complex

TRIM thrombin inhibition in myocardial ischemia

UF unfractionated

6 Introduction and background

Definition and clinical manifestations of Unstable angina and non-ST-elevation unstable coronary artery disease myocardial infarction (previously called Coronary artery disease is the major cause non-q-wave myocardial infarction) are of mortality and morbidity in developed closely related conditions and - because countries. The predominant underlying of the similarities in pathogenesis, clinical cause of coronary artery disease is presentation and initial treatment – they atherosclerosis, a process that starts early are often merged together into one entity, in life and progresses slowly and silently unstable coronary artery disease. The for decades (1). The clinical clinical presentation includes new onset manifestations of coronary artery disease of severe angina (within the latest four comprises a wide spectrum of conditions, weeks), abrupt worsening of previously ranging from chronic stable angina to the stable angina or angina at rest (5). Signs acute coronary syndromes of unstable of myocardial ischemia or necrosis, such angina, myocardial infarction and sudden as deviations of the ST-segment or the T- death. wave in standard 12-lead ECG and/or elevated levels of biochemical markers of In chronic stable angina, the ischemia myocardial cell damage, are common and usually results from an increase in supportive of the clinical diagnosis, myocardial oxygen demand that although not mandatory. outweighs the transport capacity of the stenosed coronary artery. In the acute Atherosclerosis and plaque disruption coronary syndrome, the ischemia is caused Atherosclerosis is the result of a complex by plaque disruption with exposure of the interaction between blood constituents, lipid-rich and thrombogenic core, leading disturbed blood flow, and vessel wall to the formation of a thrombus (2). The abnormalities involving several intracoronary thrombus causes aggravated pathological processes: inflammation, stenosis, transient or persistent total with increased endothelial permeability occlusion, and/or distal embolization of and monocyte recruitment; growth, with the thrombotic material (3, 4). The smooth muscle cell proliferation and severity, duration, and extent of the migration and matrix synthesis; resulting myocardial ischemia or necrosis degeneration, with lipid accumulation; will determine the clinical presentation. necrosis, by cytotoxic effect of oxidised lipids; calcification; and thrombosis (2).

7 Mature plaques typically consist of two of the thrombus may also embolize and main components: hard, collagen-rich occlude distal arteries and arterioles with sclerotic tissue and soft, lipid-rich subsequent multifocal myocardial cell atheromatous “gruel”, devoid of injury (3, 4). supporting collagen. The risk of plaque disruption depends more on the plaque Inflammation and markers of composition and vulnerability than the inflammation degree of stenosis (6). Major determinants Inflammatory processes are involved in of the liability to rupture are the size of the development of atherosclerosis (1). the atheromatous core (7), thickness and Furthermore, by destabilising the plaque collagen contents of the fibrous cap and enhancing thrombus formation, covering the core (8) and inflammation in inflammation may also be involved in the the cap (9). Plaque disruption occurs most initiation of unstable coronary artery frequently where the fibrous cap is disease (11). The potentially dangerous thinnest, often the junction between the atherosclerotic lesions are difficult to plaque and the adjacent normal vessel identify by angiography. However, at wall, i.e. the shoulder region of the plaque autopsy active inflammation is evident by (8). Mechanical and haemodynamic forces the accumulation of macrophages at the may trigger plaque disruption, e.g. sites of the plaque rupture (9). Moreover, increased blood pressure or pulse rate. elevated levels of interleukins (12) , Cigarette smoking exerts toxic effects on acute-phase proteins (13, 14), activated endothelial function and is another circulating monocytes (15) and potential trigger for acute coronary lymphocytes (16) have been reported from syndromes (10). clinical studies in unstable coronary artery disease. As the disrupted plaque exposes its thrombogenic contents to the flowing The trigger of the active inflammation in blood, thrombotic material immediately unstable coronary artery disease is not layers on top of the plaque (3). Thrombus known, although infections such as may also develop over an area of Chlamydia pneumoniae and Helicobacter endothelial denudation or erosion, without pylori have been suggested (17). The plaque disruption (10). The expanding higher degree of inflammation in patients thrombus may completely occlude the with unstable coronary artery disease coronary artery at the site of the plaque might also, at least in part, be an acute- disruption, resulting in acute myocardial phase reaction induced by the myocardial infarction or sudden death. The evolving cell damage (18-20). However, ischemia thrombus may also be partially or only in itself does not seem to elicit an acute- transiently occlusive resulting in unstable phase reaction (21). coronary artery disease. Finally, fragments

8 Markers of inflammation Haemostasis and molecular markers of Although an increased inflammatory coagulation activity activity has been documented in unstable The endothelial cells outlining the vessels coronary artery disease, there is limited prevent interaction between blood knowledge about the influence of constituents and subendothelial anticoagulant treatment on the degree of components under normal conditions. inflammation (14, 21, 22). However, the exposure of circulating blood to the disrupted or eroded plaque In the present study we have measured: initiates a series of complex reactions including platelets, leukocytes, Fibrinogen, an acute-phase protein mainly erythrocytes and haemostatic factors, synthesised by hepatocytes, with a half- producing an haemostatic plug. life of 5 days. Fibrinogen is regulated by interleukin-6 and is directly involved also Platelets in the thrombotic process, both in platelet The primary step in haemostasis is the aggregation by cross-linking the adhesion, activation and aggregation of glycoprotein IIb/IIIa-receptors on adjacent platelets (23). In response to vascular and platelets (23), and in the coagulation tissue trauma, such as plaque disruption, cascade (24), figure 1. platelets interact with exposed subendothelial structures, such as C-reactive protein, a typical acute-phase collagen. Upon activation, platelets protein, mainly synthesised in the change shape, secrete the contents of hepatocytes, with a half-life of 19 hours. a-granulae, bind soluble adhesion The release of C-reactive protein is molecules and become the surface for regulated by interleukin-6 and increased continuing platelet deposition. The levels are seen in response to infection, activated platelets express glycoprotein inflammation and tissue damage. The receptors, e.g. Ib-IX-V complex and biological function is unclear but it has conformationally changed IIb/IIIa, on been suggested that C-reactive protein their surface and is cross-linked by binding may directly interact with the of von Willebrand factor and fibrinogen, atherosclerotic process by activating the thereby aggregating to a platelet plug. complement system (25). C-reactive protein has also been shown to induce Coagulation tissue factor expression on monocytes The coagulation system is a series of (26). procoagulant and anticoagulant proteins circulating in plasma as proenzymes or procofactors (24). The activated coagulation factors are rapidly formed by proteolytic cleavage of one or two peptide

9 bonds, and the activated factor then macrophages, endothelial cells and activates another factor in the coagulation smooth muscle cells (27). In vivo, the cascade, figure 1. exposure of tissue factor to the circulating blood initiates activation of the extrinsic The lipid-rich core of the disrupted coagulation cascade (24). Factor VII binds atherosclerotic plaque has high contents tissue factor on the cell surface and the of tissue factor, a small glycoprotein tissue factor/factor VIIa-complex then expressed on stimulated monocytes, activates factors IX and X. Factor Xa

tissue factor - F VII/VIIa

F IX F X Prothrombin (F II)

heparin F Va F IXa F Xa

Antithrombin F 1+2

inogatran thrombin

fibrinogen TAT

Plasminogen

tPA soluble fibrin plasmin

fibrin D-dimer

Figure 1. Simplified illustration of the extrinsic coagulation cascade, initiated by the tissue/factor VIIa-complex. The molecular markers measured in the present study are underlined. Broken arrows indicate inhibition. TAT = thrombin-antithrombin comlex, tPA = tissue-type plasminogen activator.

10 assembles on the surface of activated Fibrinolysis platelets as part of the prothrombinase The fibrinolytic system has a major role complex, consisting of factors Xa, Va and in the balance of haemostasis (30). The calcium, and converts prothrombin to proenzyme plasminogen is converted to thrombin. Thrombin then activates the active enzyme plasmin by tissue-type platelets and converts fibrinogen to fibrin. (or urokinase-type) plasminogen activator. The fibrin is stabilised by factor XIIIa This conversion only occurs efficiently on (activated by thrombin) forming a fibrin the fibrin surface. Plasmin is the key network, thus stabilising the platelet clot. enzyme in the fibrinolytic system and Factors V, VIII and XI are also activated degrades fibrin, thus forming fibrin by thrombin, thereby autoamplifying degradation products, one of which is D- thrombin generation. Moreover, thrombin dimer. Free plasmin in the blood is very inhibits fibrinolysis by activating rapidly inactivated by a2-antiplasmin. The thrombin-activatable fibrinolysis inhibitor tissue-type plasminogen activator is (28). generated in and rapidly released from endothelial cells, and it is inactivated by a Several anticoagulant systems operate in specific plasminogen activator inhibitor, order to regulate procoagulant activity thus the balance of these two regulates the (24). Antithrombin forms complexes with fibrinolytic capacity. coagulation factors Xa, IIa (thrombin), IXa and XIa and these complexes are inactive Molecular markers of coagulation activity and removed by the liver. The inhibition The degree of coagulation activity can by antithrombin is markedly increased in reliably be determined by various assays the presence of heparin or heparin sulphate measuring molecular markers in plasma on endothelial cells. Thrombin bound to (31). For this purpose, proper blood thrombomodulin activates protein C sampling technique is critical and samples which, in the presence of protein S, should preferably be acquired with direct degrades factors Va and VIIIa on platelet venipuncture and a minimum of venous surfaces, thereby slowing the blood stasis. The blood sample should be rapidly clotting. Another important anticoagulant cooled and as quickly as possible is the tissue factor pathway inhibitor, an centrifuged. Also the timing of blood endothelium-bound protein that sampling is important since several inactivates both factors VIIa and Xa by markers have a short half-life. forming quaternary complexes in the presence of phospholipid surfaces and An activation of the coagulation system calcium, thereby inhibiting the extrinsic in the acute phase of unstable angina or coagulation pathway (29). acute myocardial infarction has been demonstrated by elevated levels of molecular markers of thrombin

11 generation, i.e. prothrombin fragment 1+2 infarction (43, 44) it is recommended to (F1+2) and thrombin-antithrombin give lower, weight-adjusted, bolus doses complex (TAT), and of thrombin activity, and infusions of UF heparin to maintain i.e. fibrinopeptide A and soluble fibrin the aPT times at 1.5 - 2.5 x control, (32-35). Elevated levels of D-dimer have corresponding to approximately 50-70 also been found in patients with unstable seconds. angina and myocardial infarction (34, 35). However, there have been conflicting Prothrombin fragment 1+2 (F1+2), a results concerning the effects on these polypeptide released from prothrombin molecular markers of coagulation activity during its conversion to thrombin by the by various thrombin inhibitors (36-41). factor Xa/Va-complex, thus reflecting thrombin generation. The normal half-life In the present study we have measured: in plasma is 60-90 minutes. F1+2 is predominantly measured by monoclonal Activated partial thromboplastin (aPT) antibody methods (31). time is easily available and currently the most widely used method for monitoring Thrombin-antithrombin complex (TAT) of treatment antithrombin agents. It is an are formed when antithrombin binds and in-vitro method that reflects surface- inhibits the active site of thrombin. This induced activation of the coagulation binding is markedly increased in the system and is therefore mainly sensitive presence of heparin or heparin-like to factors of the intrinsic pathway. Despite cofactors. TAT is considered as a marker the long experience of monitoring UF of thrombin generation and activity. The heparin treatment with aPT time, there are normal half-life in plasma is 5-15 minutes. rather limited data available on the ELISA:s are commonly used to assay TAT desirable interval for optimal (31). anticoagulant effect in the treatment of unstable coronary artery disease. Previous Soluble fibrin, also called fibrin monomer, recommendations have been similar to the is formed when the fibrinopeptides A and recommendations for treatment of venous B are cleaved from the aminoterminus of thrombosis (42). The differences between fibrinogen by thrombin. Thus, elevated arterial and venous thrombosis in levels reflect increased thrombin activity. thrombus burden and in concomitant A subsequent cross-linking of fibrin is treatment with aspirin and other platelet induced by thrombin-activated factor inhibitors such as the glycoprotein IIa/IIa- XIIIa. Normal plasma half-life for soluble inhibitors might give over-anticoagulation fibrin is 4-6 hours. There are both ELISA:s and increased bleeding risk in unstable and functional assays, the latter measuring the activity of fibrin as a co-factor for coronary artery disease (43). In the recent tissue-type plasminogen activator- AHA/ACC guidelines for unstable mediated activation of plasminogen (31). angina/non-ST-elevation myocardial

12 D-dimer, is a fibrin degradation product Elevated levels of C-reactive protein, as containing cross-linked gamma-chains. As an acute-phase reaction induced by this is the final product of the coagulation myocardial cell damage, have been and fibrinolytic cascade, D-dimer mainly reported in unstable coronary artery reflects fibrin turnover in conditions disease patients with troponin elevation without pharmacological fibrinolytic (20). However, there are limited data treatment. Normal plasma half-life is 15- concerning the inflammatory activity in 30 hours. Both semiquantitative latex patients with unstable angina without agglutination methods and more sensitive, signs of myocardial cell damage. quantitative ELISA:s are available (31). Two small studies have reported an Antithrombin, the natural inhibitor of association of troponin elevation and signs several serine proteases, including factor of coagulation activation, e.g. as indicated Xa and thrombin, is commonly measured by elevated soluble fibrin levels (48, 49). with functional assays (45). Thus, elevated troponins in patients with unstable coronary artery disease have been Myocardial infarction and biochemical suggested as surrogate markers for plaque markers of myocardial cell injury disruption and thrombus formation (50). Myocardial ischemia initiates a chain of This concept is supported by more cellular events which, if not reversed, end frequent intracoronary thrombi at with the disruption of the myocardial cell angiography in patients with elevated membranes and cell death. The loss of cell troponin levels (51-53). However, there membrane integrity leads to diffusion of are limited data on the association of intracellular markers that can be measured troponins and other molecular markers of in the systemic circulation as signs of the coagulation activity. myocardial cell injury. Until recently, the MB isoenzyme of creatine phosphokinase Prognosis and risk stratification in (CK-MB) has been the most specific unstable coronary artery disease marker of myocardial cell injury. The The patients with unstable coronary artery newly developed commercially available disease are at variable risk for recurrent methods to analyse troponins, with higher ischemic events, with death, myocardial sensitivity and specificity, along with the infarction or need for revascularisation, at prolonged time-window have further average of about 30% of the patients at improved the possibilities to detect small 3 months (54). Non-invasive risk irreversible myocardial cell injuries (46). stratification in this heterogeneous group The diagnosis of myocardial infarction has of patients might enable the clinician to recently been redefined (47) and the new tailor the therapy based on individual risk definition is based on biochemical evaluations (55) . grounds, e.g. troponins.

13 Clinical presentation clinical outcome in unstable coronary Several easy available demographic and artery disease (14, 60-64). There is limited historical features of the patient are knowledge about the time-course and the indicators of increased risk, such as higher influence of anticoagulant treatment of age, prior myocardial infarction, history changes in fibrinogen and C-reactive of congestive heart failure, presence of protein in unstable coronary artery disease diabetes mellitus, and known significant (14, 21, 22). coronary artery stenoses. Furthermore, the risk can be estimated from the clinical Coagulation activity presentation, with angina at rest, especially There are rather limited data concerning during the last 48 hours, haemodynamic the prognostic importance of elevated instability or congestive heart failure coagulation markers in patients with indicating higher risk (44). unstable coronary artery disease (31). Furthermore, possible relations between Electrocardiogram the effect of changes in coagulation Depression of the ST-segment in a markers induced by anticoagulant standard 12-lead ECG on admission or treatment and the clinical outcome in during hospital stay indicates raised risk unstable coronary artery disease remain for adverse outcome (56, 57). Similarly, to be elucidated. ischemic episodes during continuous monitoring with 12-lead electro- Treatment of unstable coronary artery cardiogram or vectorcardiography also disease indicate higher risk (58, 59). The primary goal of the treatment in unstable coronary artery disease is Myocardial cell injury (troponins) reduction of symptoms, limitation of Besides their use as diagnostic markers of myocardial cell damage and prevention of myocardial infarction, the troponins are future cardiac events. The initial powerful predictors of future cardiac management of patients with symptoms events in unstable coronary artery disease suggesting unstable coronary artery (65, 66). However, there are no data on disease includes admission to a coronary the association of troponins and molecular care unit, bed rest with continuous 12-lead markers of coagulation activity concerning ECG or vectorcardiography and repeated clinical outcome. blood samples for analyses of biochemical markers of myocardial cell damage for at Inflammation least 6-8 hours (67). Initial treatment The early levels of fibrinogen and C- include b-blockers and/or oral long-acting reactive protein have in several studies calcium antagonists, nitro-glycerine, been identified as indicators of increased angiotensin converting enzyme-inhibitors short- and long-term risk for adverse and statins (67).

14 Early coronary angiogram and UF or LMW heparin compared with ASA revascularisation have in recent studies alone have shown reductions in the short- been shown to reduce mortality, risk of term rate of death or myocardial infarction myocardial infarction and the need for of approximately 50 % to 60 % (72, 73, readmission in unstable coronary artery 78-80). UF heparin is administered as an disease patients with ST-depression on intravenous bolus followed by an infusion, admission ECG or elevation of markers and should be titrated to an aPT time of of myocardial cell injury, i.e. troponin or 1.5 - 2.5 x control (44). UF heparin has CK-MB (68, 69). several limitations, including high inter- individual variability of anticoagulant Platelet inhibition response, a non-linear dose-response, a Platelet inhibition by ASA is a well- requirement for endogenous cofactors and documented standard therapy in unstable non-specific binding to plasma proteins coronary artery disease with an average and cells (42). The LMW heparins have, reduction in occurrence of death and compared with UF heparin, increased myocardial infarction of more than 50% bioavialability by subcutaneous (70-73). ASA ³ 300 mg should be administration, less plasma protein administered as soon as possible after binding and 2-4-fold longer half-life (81). presentation and should be continued A large meta-analysis has shown modest, indefinitely with a daily dose of at least non-significant 12% reduction in the risk 75 mg (74). A thienopyridine, i.e. of death or myocardial infarction by LMW clopidogrel, should be given to patients heparin compared with UF heparin for an unable to use ASA. There are additional equal duration of therapy (82). Since the benefits by combining clopidogrel and LMW heparins are easier to administer ASA in unstable coronary artery disease and require no monitoring they have (75). In patients for whom an early become the standard treatment of unstable invasive procedure is planned, a coronary artery disease. glycoprotein IIb/IIIa-inihibitor might be added to ASA and unfractionatd (UF) or At discontinuation of UF heparin low molecular weight (LMW) heparin, treatment, the disease process may within while awaiting and during the procedure hours be reactivated with subsequent new (69, 76, 77). ischemic events (83, 84) and, similarly, clinical reactivation was evident a few Thrombin inhibition by heparins days after lowering the dose of LMW Heparin is a key component in the heparin (80). Concomitant ASA treatment antithrombotic management of unstable has been suggested to reduce the coronary disease. The results of studies reactivation (83). with the combination of ASA and either

15 Direct thrombin inhibitors Treatment with UF and LMW heparin in Inogatran (AstraZeneca, Mölndal, unstable coronary artery disease has Sweden) is a synthetic low molecular several potential shortcomings, such as weight peptidomimetic, based on the limited effect on clot-bound thrombin, tripeptide R-Phe-Pro-Arg, which in turn reactivation after cessation of treatment is based on the structure of fibrinopeptide and, despite prolonged therapy, A close to the thrombin splitting site in considerable risk for recurrent ischemic the Aa chain of fibrinogen (87). Its events. Therefore, there is an extensive biological half-life in plasma is search for new and more effective approximately one hour and it is evenly anticoagulant agents for the treatment of eliminated in urine and bile (88). It is a acute coronary syndromes. selective competitive direct thrombin inhibitor, reversibly binding the catalytic Direct thrombin inhibitors work site of thrombin, thereby inhibiting both independently of antithrombin or heparin clot-bound thrombin and thrombin in the cofactor II and are able to inhibit both fluid phase (89, 90). In vitro, inogatran fibrin-bound and free thrombin (85). inhibits platelet aggregation and has Hirudin, found in the saliva of the negligible effects on fibrinolysis (87). medicinal leech, was already in the late Inogatran was more effective than ASA 19th century characterised to have or UF heparin in a porcine model of antithrombotic properties. Recombinant coronary thrombosis (91), and seemed hirudin and several variants of well tolerated in pilot studies in healthy recombinant hirudin with exchange of volunteers and unstable coronary artery individual amino acids have been disease patients (37). Therefore, the developed. These direct thrombin evaluation of the appropriate dose, clinical inhibitors bind thrombin both at the active efficacy and safety of inogatran in a large- site and at exosite 1. Synthetic low- scale trial of unstable coronary artery molecular-weight direct thrombin disease was highly warranted. inhibitors, e.g. argatroban, efegatran and inogatran, have also been developed. These only bind the active site of thrombin (86).

16 Aims of the study

The aim of the study was to investigate the degree of coagulation activity and inflammation in unstable coronary artery disease and to evaluate the influences of thrombin inhibition with inogatran, a direct thrombin inhibitor, or UF heparin.

More specific aims were to evaluate

¨ the anticoagulant efficacy, i.e. aPT time prolongation, of different doses of inogatran or UF heparin

¨ the relation of different aPT times during thrombin inhibition and subsequent manifestations of myocardial ischemia

¨ the degree of inflammation and coagulation activity, reflected by molecular markers of coagulation activity and acute-phase proteins

¨ the influence of treatment with inogatran or UF heparin on inflammation and coagulation activity

¨ the reactivation of coagulation activity after cessation of treatment with inogatran or UF heparin

¨ the relation of myocardial cell injury and the degree of inflammation and coagulation activity

¨ the degree of inflammation and coagulation activity in relation to short-term manifestations of myocardial ischemia and long-term mortality

17 Figure 2.

TRIM - Thrombin Inhibition in Myocardial Ischemia

Unstable angina or non-Q-myocardial infarction randomisation within 24 hours from chest pain

72 hours study drug infusion with:

Inogatran Inogatran Inogatran Heparin low dose medium dose high dose (unfractionated) 302 patients 303 patients 299 patients 305 patients

Substudy Substudy Substudy Substudy inflammation inflammation inflammation inflammation & coagulation & coagulation & coagulation & coagulation 83 patients* 84 patients* 74 patients* 79 patients*

*)consecutive patients in 19 of 61 centres

Primary composite endpoint death, myocardial (re-)infarction, refractory or recurrent angina after 7 days

Secondary composite endpoints death, myocardial (re-)infarction, refractory or recurrent angina at 72 hours and 30 days death, myocardial (re-)infarction or refractory angina at 72 hours, 7 and 30 days death or myocardial (re-)infarction at 72 hours, 7 and 30 days

18 Material and methods

Patients and design The Thrombin Inhibition in Myocardial continuous infusion of 2.0, 5.0 and 10.0 Ischemia (TRIM) study (92), a dose- mg/h respectively. Heparin was finding and safety study, enrolled 1209 administered as a 5000 U intravenous unstable coronary artery disease patients bolus injection followed by infusion with (paper I) in 61 Scandinavian centres 1200 U/hour. All infusions were to be during 1994 and 1995, figure 2. The continued for 72 hours and monitored with substudy population (papers II-V) aPT time, as outlined below. ASA was consisted of 320 consecutive patients in recommended, while other platelet 19 of the 61 participating centres of the inhibitors and oral anticoagulants were not TRIM study, figure 2. Eligible for allowed. All other medication was to be inclusion were men and post-menopausal given at the discretion of the responsible women between 25 and 80 years of age physician. with unstable angina, defined as new onset of ischemic chest pain or rapid aPT time and dose adjustments deterioration in previously stable angina Blood samples for analyses of aPT time during the last four weeks, or suspicion were obtained at baseline and during of a non-ST-elevation myocardial infusion at 6, 24 and 48 hours. They were infarction. This clinical diagnosis had to immediately analysed with the routine be supported by either changes in ECG at method of the clinical chemistry rest, e.g. ST-depression or T-wave laboratory in each of the participating inversion, or previous manifestations of centres. The study drug infusion rate was coronary artery disease, e.g. previous reduced by 10% if the aPT time was 3-4 myocardial infarction, bypass surgery or times the lower reference level at each positive exercise test. centre, and by 20% if the aPT time exceeded 4 times the reference level. If the aPT time exceeded 6 times the Patients were, within 24 hours from the reference value, the infusion was qualifying episode of chest pain, rando- temporarily stopped for 1-2 hours and then mised to blinded treatment with UF re-started with a 20 % reduction. The aPT heparin or one of three different fixed time was re-measured 4 hours after all doses of inogatran. Low, medium and high adjustments, and if needed further dose dose inogatran patients received reductions were made. Upward intravenous bolus injections of 1.10, 2.75 adjustments of infusion rates were not and 5.50 mg respectively, followed by allowed.

19 Inogatran plasma concentrations and Fibrinogen thrombin time Fibrinogen was analysed by rate Blood samples were after 6 hours of nephelometry with a Beckman Array inogatran infusion obtained from 80 low, protein system (Beckman Instruments 82 medium and 71 high dose patients and Inc). The assay was performed according after 48 hours from 207 low, 199 medium to the recommendations by the and 208 high dose patients respectively. manufacturer except that goat anti-human The plasma concentration of inogatran fibrinogen (Atlantic Antibodies) was used. was determined using a liquid phase The assay was calibrated against a human chromatography-mass spectrometry plasma standard (Behring Diagnostics method after solid-phase extraction of the GmbH), reference range 2.0 to 3.6 g/L, compound from plasma (90). total CV £ 8%.

In 67 patients thrombin time was C-reactive protein measured after 6 hours infusion to C-reactive protein was analysed with the correlate anticoagulant effect of inogatran Immulite system, a chemiluminescent to plasma concentrations. enzyme-labelled immunometric assay based on ligand-labelled monoclonal Substudy blood samples antibody and separation by anti-ligand- Venous blood samples were obtained, coated solid phase (Immulite CRP, preferably by direct veni-puncture, in Diagnostic Products Corporation) (93). citrated tubes for the analyses of markers The detection limit is 0.1 mg/L and the of coagulation and inflammation and in reference range 0.1 to 4.6 mg/L. Total CV heparin tubes for the analysis of troponin 5.6% at 2 mg/L. T. The first 2 mL blood were disposed and samples were within 30 minutes Molecular markers of coagulation centrifuged at 2000G for twenty minutes. activity Aliquots of 500 mL plasma in Eppendorf Markers of coagulation activity were tubes were frozen and stored at -70 °C analysed pre-treatment, during study drug until analysis. infusion after 6 hours, 24 and 72 hours, and after 76 and 96 hours, i.e. 4 and 24 Markers of inflammation hours after cessation of treatment, and 30 Markers of inflammation were analysed days after randomisation. pre-treatment, during study drug infusion after 24 and 72 hours, and after 96 hours, Prothrombin fragment 1+2 (F1+2) i.e. 24 hours after cessation of treatment, A sandwich ELISA technique (Enzygnost and 30 days after randomisation. F1+2, Behringwerke AG) was used for the determination of F1+2 (94), with reference intervals for healthy individuals 0.4-1.5

20 nmol/L and intra- and inter-assay is 0.012 µg/L and the reference range 0.0 variations (CV) between 5 and 9% (39). to 0.02 µg/L, total CV 12% at 0.22 µg/L. The discriminator value for myocardial Thrombin-antithrombin-complex (TAT) cell injury recommended by the A sandwich ELISA technique (Enzygnost manufacturer was 0.1 µg/L (98). TAT, Behringwerke AG) was used for the determination of TAT (95) with reference End-points intervals for healthy individuals 1.2 - 5.0 The primary composite end-point in the mg/L and intra- and inter-assay variations main TRIM study was the occurrence of (CV) between 5 and 9% (39). death, nonfatal myocardial infarction (or reinfarction), refractory angina or Soluble fibrin recurrence of angina after 7 days. Soluble fibrin was assessed by a Secondary end-points were the same chromogenic method (SF, Chromogenix) composite at 72 hours (the end of infusion) where the stimulatory effect of SF on the and at 30 days, or the occurrence of tissue-type plasminogen activator catalysed death, myocardial (re-)infarction or conversion of plasminogen to plasmin is refractory angina (papers I-V) at 72 hours, exploited, upper reference limit 25 nmol/L, 7 days and 30 days, or the occurrence of CV 7% (96). death or myocardial (re-)infarction at 72 hours, 3 days or 30 days (paper II). D-dimer A sandwich ELISA technique (TintElize Nonfatal myocardial (re-)infarction was D-dimer, Biopool) was used for the defined as the occurrence of at least two analysis of D-dimer, reference range 10- of the following: 130 mg/L, CV 10% (97). 1) typical ischemic chest pain, 2) diagnostic ECG with a new Q-wave Antithrombin and/or ST-elevation or Plasma antithrombin was analysed by a 3) typical elevation of cardiac enzymes, functional, chromogenic method (AT 400, i.e. CK-MB (mass) ³ the upper reference Chromogenix) using factor Xa inhibition level on one occasion, or catalytic activity (45). Reference interval 0.8-1.2 IE/mL, of CK, CK-B or CK-MB ³ twice the upper CV <3%. reference level on one occasion, or catalytic activity of CK, CK-B or CK-MB Marker of myocardial cell injury ³ the upper reference level on two Troponin T was analysed pre-treatment occasions. and after 6 and 12 hours with a sandwich ELISA technique (ELISA Troponin(e) T) A new infarction (regarded as an event) on an ES 300 analyser (Boehringer was differentiated from an ongoing Mannheim GmbH). The detection limit infarction at inclusion by normal cardiac

21 enzymes during the first six hours or the The levels of aPT times and markers of reappearance of the criteria above after inflammation and coagulation activity and normalisation of initially elevated their relation to short-term clinical enzymes. Refractory angina was defined outcome were evaluated for the substudy as chest pain lasting ³ 5 minutes with population, and separately for the transient changes in the ECG, despite inogatran treated patients (all three groups maximal ongoing medication, leading to combined) and the UF heparin treated an additional coronary intervention. An patients. independent end-point committee evaluated all end-points (99). The change in the levels of the respective coagulation marker after 6 hours of Long-term follow-up data were obtained infusion compared to the pre-treatment from 286 of the 320 patients at a median level was calculated individually and of 29 months (min 12 months, max 50 patients were grouped as either early months). This information was obtained decreasing or increasing in coagulation from hospital records and local or national activity. The latter group included 3, 19, registries. If data in these sources were 54 and 6 patients with unchanged levels missing the information was collected by of F1+2, TAT, soluble fibrin or D-dimer, telephone interview. respectively. The 10-12 patients with a missing blood sample in any of the Statistics coagulation markers, either pre-treatment Fisher exact test (2-sided) or chi-square or at 6 hours, were excluded from the tests as appropriate were used to judge analyses of changes in coagulation activity significance of differences in proportions. during treatment in relation to clinical Differences in the levels of markers were outcome. judged with non-parametric tests, between-group comparisons with Mann The maximum change in levels of each Whitney or Kruskal Wallis tests as coagulation marker from cessation of appropriate, and within-group differences treatment, at 72 hours, to samples taken between different time-points with at 4 and 24 hours thereafter, was Wilcoxon signed rank tests. Correlations individually calculated to evaluate were evaluated with Spearman rank order reactivation in coagulation activity after coefficients. The influences of pre- discontinuation of study drug. treatment levels of different markers of Multiple logistic regression analyses were inflammation and coagulation activity on performed evaluating the influence of the probability of death during long-term molecular markers of inflammation, follow-up were evaluated with the log rank coagulation activity or aPT times, together test.

22 with all relevant baseline characteristics – age, gender, weight, smoking habits, congestive heart failure, diabetes mellitus, hypertension, stable angina, previous myocardial infarction and previous angioplasty or bypass surgery – on the clinical end-points of the study.

Possible interactions between inogatran or UF heparin treatment on the markers of coagulation and the risk of ischemic events were evaluated with multiple logistic regression analyses using interaction terms.

23 Results

The TRIM study In the main TRIM study (92), there were 30 days. However , the composite of no significant differences between UF death and myocardial (re-)infarction and heparin and the combination of the three its combinations with refractory and inogatran groups, or between any of the recurrent angina were less common in the three inogatran groups concerning the UF heparin group compared to the primary end-point, a composite of death, combined three inogatran groups during myocardial (re-)infarction, refractory or the 72 hours study drug infusion, table 1. recurrent angina after 7 days, table 1. The study drug was given according to Neither were there any differences protocol in 88 % of the patients, in a between the four treatment groups in the similar manner in all four treatment same (secondary) composite end-point at groups.

Table 1. Primary and secondary end-points at 72 hours, 7 and 30 days

Inogatran Inogatran Inogatran Inogatran UF low dose medium dose high dose combined heparin n=302 n=303 n=299 n=904 n=305

Primary end-point 7 days D+MI+Ref+Rec 138 (45.7%) 139 (45.9 %) 136 (45.5 %) 413 (45.7 %) 125 (41.0 %)

Secondary end-points 7 days D+MI 12 (4.0 %) 13 (4.3 %) 20 (6.7 %) 45 (5.0 %) 8 (2.6 %) D+MI+Ref 25 (8.3 %) 27 (8.9 %) 31 (10.4 %) 83 (9.2 %) 25 (8.2 %)

Secondary end-points 72 hours D+MI 11 (3.6 %) 6 (2.0 %) 12 (4.0 %) 29 (3.2 %) 2 (0.7 %)* D+MI+Ref 15 (5.0%) 14 (4.6 %) 20 (6.7 %) 49 (5.4 %) 8 (2.6 %)* D+MI+Ref+Rec 19 (39.4%) 114 (37.6 %) 108 (36.1 %) 341 (37.7 %) 90 (29.5 %)**

Secondary end-points 30 days D+MI 23 (7.6 %) 25 (8.3 %) 27 (9.0 %) 75 (8.3 %) 18 (5.9 %) D+MI+Ref 38 (12.6 %) 37 (12.2 %) 39 (13.0 %) 114 (12.6 %) 36 (11.8 %) D+MI+Ref+Rec 156 (51.7 %) 158 (52.2 %) 159 (53.2 %) 473 (52.3 %) 146 (47.9 %)

D = death, MI = myocardial (re-)infarction, Ref = refractory angina, Rec = recurrent angina *p<0.05; **p<0.01; comparison between combined inogatran groups and UF heparin

24 Troponin substudy aPT time and the influence of inogatran In a separate TRIM substudy (100), or UF heparin (paper I) Lüscher and colleagues related troponins During inogatran infusion there was a to clinical outcome in 516 patients. Patents dose-response relationship concerning with elevated levels of troponin T, i.e. median aPT times comparing the low, troponin T ³ 0.1 mg/L either pre- medium and high dose groups, figure 3. treatment or after 6 hours, had an increased The aPT time levels were fairly stable mortality at 30 days compared with throughout the infusion period. Dose- patients with normal levels, 3.2 % versus reductions were only carried out in four 0.4 %, p=0.01. Similarly, the composite per cent or less in each group. Because of of death and myocardial (re-)infarction at the overlap in aPT times between the three 30 days was observed in 27 (11 %) of the groups and in the absence of significant patients with elevated troponin T as differences in clinical outcome, all compared to 12 (4 %) of the patients inogatran patients were combined into one without troponin elevation, p=0.006 (100). group in the further analyses of the

130 aPT time (s) 120 110 100 90 UF inogatran heparin 80 high all 70 inogatran inogatran medium patients 60 inogatran low 50 40 30 20

10 0 6 24 48 0 6 24 48 0 6 24 48 0 6 24 48 0 6 24 48 time

Figure 3. Distribution of activated partial thromboplastin (aPT) times at different time-points before and during infusion in the four randomised treatment groups – low, medium, high dose inogatran and heparin – and for the combined group of all inogatran patients. Box-plots contain median, 1st and 3rd quartiles and in the whiskers 10th and 90th percentiles.

25 relations between aPT times and clinical The UF heparin treated patients, n=305, outcome. Median aPT times for the had a clearly different pattern in aPT time combined inogatran group, n=904, were response with a peak, median 69 seconds, 29 seconds on admission and during after 6 hours of infusion and a much larger infusion 44, 44 and 45 seconds, at 6, 24 dispersion in aPT times than among the and 48 hours respectively. inogatran treated patients, figure 3. The dispersion diminished and the median aPT There was a positive correlation between times decreased to 54 seconds after 24 plasma-concentration of inogatran and hours and 49 seconds after 48 hours, in aPT times at 6 hours (data not shown) and part because of frequent dose reductions, at 48 hours, figure 4. Thrombin time had, which were carried out in 41 % of the as compared to aPT time, a better and more patients. Still, the median aPT times in linear correlation to plasma concentrations patients without any dose reduction also of inogatran at 6 hours. decreased by almost 20 % between 6 and 24 hours.

100 aPT time (s)

75

50

25

0 plasma conc. 0 0.5 1.0 1.5 2.0 2.5 mmol/L

Figure 4. Correlation of plasma concentration of inogatran and activated partial thromboplastin time (aPTT) at 48 hours, n=581, Spearman rank correlation coefficient r=0.7, p < 0.001

26 Clinical outcome in relation to aPT hours, 7 and 30 days respectively, figure times (paper I) 5. Similarly, aPT times above the median after 24 hours inogatran infusion was Inogatran treatment associated with higher adverse event rate. Adverse ischemic events, i.e. death, myocardial (re-)infarction or refractory Patients with aPT time above the median angina, were more frequent in the group after 6 hours of inogatran infusion had of patients with aPT time above the lower median body weight, higher age and median at 6 hours. This difference in slightly higher pre-treatment aPT time, clinical outcome was already seen after the median 30 vs. 28 seconds, p<0.001. 72 hours of infusion therapy, table 2. A Congestive heart failure was also more cluster of ischemic events was observed common in patients with high aPT time. after cessation of treatment and to a similar A multivariate logistic regression analysis degree in patients with aPT time levels - including aPT time on admission and at above or below the median. 6 hours, serum creatinine and all relevant baseline characteristics - demonstrated Dividing the group of inogatran treated that age and aPT time above the median patients into four groups according to at 6 hours remained independent quartiles of aPT time at 6 hours indicated predictors of death , myocardial direct relationship between higher aPTT (re-)infarction or refractory angina up to and deteriorating clinical outcome, with seven days in patients treated with an approximately doubled event rate in the inogatran. top quartile as compared to the bottom 2 quartile, p<0.05 (chi test for trend) at 72

Table 2. Clinical outcome in relation to aPT times after 6 hours inogatran infusion

aPT time after 6 hours < 44 seconds ³ 44 seconds n=423 n=464

Death, MI or refractory angina, 72 hours 15 (3.5%) 33 (7.1%) 0.02 Death, MI or refractory angina, 7 days 28 (6.6%) 54 (11.6%) 0.01 Death, MI or refractory angina, 30 days 44 (10.4%) 68 (14.7%) 0.06

MI = myocardial (re-)infarction., Statistics: Chi2 test

27 UF heparin treatment patients with aPT time above, and 7 The event rate during UF heparin infusion (4.8 %) of the 145 patients with aPT time was low, with no significant difference in below the median at 24 hours had an adverse events in relation to aPT times adverse event during the UF heparin above or below median at 6 hours. In infusion. This benefit was lost after contrast to the finding in inogatran treated cessation of treatment, because of patients, aPT times above the median at clustering of events within 24 hours, and 24 hours were related to favourable no difference in clinical outcome was seen outcome. Only 1 (0.6 %) of the 155 at seven days.

% death, MI, refractory angina

> 51 s 15 44-51 s p<0.05 38-43 s

10 p<0.05 < 38 s

p<0.05

5

0 3 7 14 21 30 days

Figure 5. Composite of death, myocardial (re-)infarction or refractory angina during seven days in inogatran treated patients, in relation to aPT times at 6 hours divided in quartiles; < 38 seconds (n=208), 38-43 seconds (n=215), 44-51 seconds (n=215), > 51 seconds (n=247). Statistics: Chi2 test for trend

28 Baseline characteristics and the degree activity, F1+2, TAT, soluble fibrin and D- of inflammation and coagulation dimer, and furthermore weak correlations activity (papers II-IV) between pre-treatment levels of markers Baseline characteristics for the substudy of inflammation and coagulation. patients in the four randomised treatment groups are presented in table 3. There were Patients older than the median age in the no significant differences between patients substudy, 66 years, had significant higher in the four treatment groups in pre- pre-treatment levels of all the markers of treatment levels of fibrinogen, C-reactive inflammation and coagulation, except protein, antithrombin, F1+2, TAT, soluble antithrombin. Women had significantly fibrin, D-dimer levels according to the higher pre-treatment levels of Kruskal-Wallis test. Pre-treatment levels antithrombin, F1+2 and D-dimer, but of markers of inflammation, fibrinogen significantly lower soluble fibrin levels and C-reactive protein, were significantly than men. No significant differences were correlated. There were also significant found in pre-treatment levels of correlations between pre-treatment levels fibrinogen, C-reactive protein or TAT in of all molecular markers of coagulation relation to gender.

Table 3. Baseline characteristics and pre-treatment levels of markers of inflammation and coagulation activity

Inogatran Inogatran Inogatran UF low dose medium dose high dose Heparin n=83 n=84 n=74 n=79

Age, years 64 (58; 73) 66 (58; 74) 66 (59; 74) 68 (58; 72) Gender (male) 78 % 67 % 65 % 78 % Current smokers 19 % 19 % 15 % 23 % Stable angina >4 weeks 70 % 70 % 68 % 57 % Hypertension 42 % 38 % 42 % 33 % Congestive heart failure 12 % 13 % 18 % 13 % Previous MI 47 % 49 % 54 % 34 % Diabetes mellitus 23 % 15 % 22 % 9 %

Fibrinogen, g/L 3.5 (3.0; 4.1) 3.4 (2.7; 4.2) 3.5 (3.1; 4.3) 3.3 (2.8; 3.6) C-reactive protein, mg/L 4.0 (2.2; 7.5) 3.9 (2.5; 6.6) 4.5 (2.9; 7.6) 3.5 (2.3; 6.2) F1+2, nmol/L 1.30 (0.95; 1.78) 1.30 (0.97; 1.82) 1.32 (1.01; 1.78) 1.41 (1.10; 1.84) TAT, mg/L 3.20 (2.40; 5.20) 3.05 (2.40; 5.23) 3.25 (2.38; 5.98) 3.60 (2.60; 7.15) Soluble fibrin, nmol/L 11 (9; 14) 11 (9; 14) 12 (9; 15) 11 (9; 15) D-dimer, mg/L 108 (74; 169) 118 (75; 197) 114 (67; 200) 122 (72; 197) Antithrombin, IU/mL 1.01 (0.92; 1.12) 1.07 (0.95; 1.16) 1.05 (0.94; 1.17) 1.01 (0.93; 1.09)

MI = myocardial (re-)infarction. Values are medians (1st and 3rd quartiles) or proportions.

29 Patients with pre-treatment levels in the independent predictor of high fibrinogen upper tertile of fibrinogen, F1+2, TAT, levels. No independent predictors for high soluble fibrin or D-dimer had in higher pre-treatment levels of TAT or C-reactive proportion concomitant diseases, for protein levels were found in the instance stable angina pectoris >4 weeks, multivariate logistic regression models. diabetes mellitus and congestive heart failure. There were no significant The influences of thrombin inhibition differences concerning smoking habits, on inflammation (paper II) and hypertension, previous myocardial coagulation activity (paper III) infarction, angioplasty or bypass surgery between patients with high or low pre- Fibrinogen treatment levels of markers of There were no significant differences in inflammation or coagulation activity. fibrinogen levels between the four treatment groups, neither during the 72 Multivariate logistic regression analyses hours of anticoagulant treatment, nor at identified age as independent predictor of 24 hours thereafter. Taking all the groups high pre-treatment levels, i.e. in the top together, there was a significant increase tertile, of antithrombin, F1+2, soluble in fibrinogen during the first 24 hours, fibrin, D-dimer and fibrinogen. which seemed to reach its maximum after Furthermore, female sex was an 72-96 hours, table 4. After 30 days the independent predictor of high F1+2 levels. fibrinogen levels still remained Congestive heart failure was an significantly higher than pre-treatment.

Table 4. Time-course of fibrinogen and C-reactive protein

Fibrinogen, g/L C-reactive protein, mg/L n=314 n=306

Pre-treatment 3.4 (2.9; 4.0) 4.0 (2.4; 6.9)

24 hours 3.7 (3.1; 4.5)* 6.6 (3.3; 19.7)*

72 hours 3.9 (3.3; 5.0)* 6.9 (3.6; 16.3)*

96 hours 4.0 (3.3; 5.0)* 6.5 (3.2; 13.3)*

30 days 3.6 (3.0; 4.3)* 3.2 (2.3; 6.3)†

Values are medians (1st and 3rd quartile), *p<0.001 and †p=0.08 for the difference compared to pre-treatment levels, Wilcoxon signed rank test.

30 C-reactive protein Prothrombin fragment 1+2 There were no differences in C-reactive UF heparin was more effective than protein levels between the four treatment inogatran in early suppressing the levels groups, neither during the 72 hours of of F1+2, and 95 % of the UF heparin anticoagulant treatment, nor at 24 hours patients had a decreased F1+2 level at 6 thereafter. Overall, the increase in C- hours as compared to 75 %, 79 % and 86 reactive protein levels was, compared to % of the patients in the low, medium and fibrinogen, more pronounced with a large high dose inogatran groups respectively, upward dispersion during the first 24-96 p=0.002 (UF heparin vs. inogatran groups hours, table 4. In contrast to fibrinogen, combined). After the early decrease in the the levels of C-reactive were markedly UF heparin group there was a significant decreased from day 4 to day 30, when increase in F1+2 to a level slightly above the level tended to be lower than pre- the pre-treatment level at cessation of treatment. treatment. In contrast, F1+2 levels during inogatran treatment remained reduced throughout the infusion period, all groups significantly different from the UF heparin group at 72 hours, figure 6.

Medians in relation to baseline

% F 1+2 TAT Soluble fibrin D-dimer %

+40 +40 * +30 +30

+20 +20

+10 ** +10 LDI MDI HDI 0 ** 0 *** *** ** -10 LDI -10 MDI HDI ** ** * * * -20 * -20

-30 -30 0h 6h 24h 72h 76h 96h 30d 0h 6h 24h 72h 76h 96h 30d 0h 6h 24h 72h 76h 96h 30d 0h 6h 24h 72h 76h 96h 30d infusion infusion infusion infusion

Time after start of infusion

Figure 6. The median change in relation to baseline for F 1+2, TAT, SF and fibrin D-dimers. *p<0,05 **p<0,01 ***p<0,001. Significant differences between the groups, in specific points in time are indicated by dashed lines and levels of significance.

= LDI, = MDI, = HDI, = Heparin

31 Thrombin-antithrombin complex groups without any dose-response The TAT levels decreased during infusion relationship, and therefore no further in all groups. There was a greater reduction analyses of antithrombin in relation to at 6 hours in the high dose inogatran and clinical outcome were performed (paper the UF heparin groups, than in the medium III). and low dose inogatran groups, figure 6. A dispersion upwards was seen in different Reactivation of coagulation activity after phases of the study in all groups. cessation of thrombin inhibition There was a significant increase in levels Soluble fibrin of F1+2 during the first 24 hours after Soluble fibrin levels were significantly cessation of study drug in all four decreased at 6 and 24 hours during treatment groups. F1+2 levels at 76 and infusion only in the UF heparin group. No 96 hours in the UF heparin group were dose-response relationship could be significantly increased compared to the identified in the inogatran groups, figure pre-treatment level, while F1+2 levels in 6. There was a marked upwards dispersion the inogatran groups returned to pre- from the median value during the infusion treatment levels, figure 6. The TAT levels period and up to 24 hours thereafter in all increased in a majority of the patients in inogatran groups, but not in the UF heparin each treatment group, although to group. significantly higher levels within 24 hours after cessation only in the low and medium D-dimer dose inogatran groups. There was an The D-dimer levels were unchanged after increase to significantly higher D-dimer 6 hours treatment but decreased after 24 h levels 24 hours after cessation of treatment of infusion in all four groups. No dose- in all four groups, most pronounced after response relationship in the inogatran cessation of UF heparin. groups could be identified, figure 6. The reactivation increase in levels of F1+2 Antithrombin was significantly greater in patients with In the UF heparin group, the level of pre-treatment levels in the top tertile of antithrombin decreased during infusion F1+2 in the medium and high dose and reached the lowest level four hours inogatran groups and the UF heparin after cessation of infusion. Antithrombin group. Similarly, in all four groups there increased significantly within 24 hours was a significantly greater increase in after termination of UF heparin infusion, levels of D-dimer after cessation of but was still significantly lower than on treatment in patients with high pre- admission. Antithrombin remained near treatment D-dimer levels. the admission level in all three inogatran

32 At 30 days there were no significant The F1+2 and D-dimer levels were also differences in levels of F1+2, TAT, significantly higher pre-treatment and at soluble fibrin and D-dimer between 24-96 hours in troponin-positive patients, patients in the four treatment groups. The figure 7b and c. Pre-treatment levels of 30-day level of F1+2 in the total substudy soluble fibrin tended to be higher in the cohort was higher than pre-treatment, troponin-positive patients and soluble median 1.54 vs. 1.33 nmol/L, p<0.001, fibrin levels were significantly higher at and similarly the 30-day level of D-dimer 24-96 hours. The TAT levels were not was higher, 125 mg/L vs. 116 mg/L, related to troponin T at any sampling time- p<0.001. There was also a great upward point during the study. At 30 days follow- dispersion in D-dimer at 30 days, figure up there was no significant difference in 6. In contrast, TAT was significantly lower any of the four molecular markers of at 30 days, median 3.0 mg/L as compared coagulation activity in relation to pre- to pre-treatment 3.2 mg/L, p=0.001, and treatment troponin T. no significant difference was found in levels of soluble fibrin. Similar results were found using 0.06 µg/L as cut-off limit for pre-treatment Signs of myocardial cell injury and troponin T or positive troponin T in serial inflammation (paper II) and samples within the first 12 hours. coagulation activity (paper V) Pre-treatment elevation of troponin T ³ 0.1 µg/L was found in samples form 138 (44 %) of the 317 analysed substudy patients. These troponin-positive patients had significantly higher levels of fibrinogen and C-reactive protein pre- treatment and at 24-96 hours, figure 7a. As well in troponin-positive as troponin- negative patients there was a significant further increase in levels of fibrinogen and C-reactive protein during the first 24-96 hours, figure 7a. This increase in levels of fibrinogen and C-reactive protein in troponin-negative patients was still significant when excluding patients with clinical events, i.e. death, myocardial (re-)infarction or refractory angina during the study.

33 g/L fibrinogen 6.5

p<0.001 p<0.001 p<0.001 6.0

5.5 p=0.04 p=0.1

5.0

4.5 *** *** *** 4.0 *** *** 3.5 *** **

3.0

2.5

Pre- 2.0 treatment 24 hours 72 hours 96 hours 30 days

mg/L C-reactive protein 40 73 93 82 p<0.001 p<0.001 p<0.001 35

30

25

20

15 p<0.001 p=0.2

*** *** 10 ***

*** *** *** 5 *

Pre- 0 24 hours 72 hours 96 hours 30 days treatment Figure 7a. Distribution of molecular markers of inflammation in relation to pre-treatment troponin T < 0.1 µg/L (white boxes)r or ³ 0.1 µg/L (shaded boxes). Box-plots contain median, 1st and 3rd quartiles and in the whiskers 10th and 90th percentiles. Statistics: Mann Whitney test for between- group comparisons, Wilcoxon signed rank test (within-group) compared to pre-treatment, *p<0.05, **p<0.01, ***p<0.001

34 nmol/L prothrombin fragment 1+2

3.0 p=0.009 p=0.002 p=0.3

2.5 p=0.009 p=0.005

2.0

* *** 1.5 *** *** *** 1.0 ***

0.5

0 Pre- treatment 24 hours 72 hours 96 hours 30 days

mg/L thrombin-antithrombin complex 12 15 17 16

p=0.1 p=0.2 p=0.1 p=0.3 p=0.6 10

8

6

4 *** * ** ** ** * *** * 2

0 Pre- treatment 24 hours 72 hours 96 hours 30 days

Figure 7b. Distribution of molecular markers of coagulation activity in relation to pre-treatment troponin T < 0.1 µg/L (white boxes) or ³ 0.1 µg/L (shaded boxes). Box-plots, symbols and statistics as in figure 7a

35 soluble fibrin nmol/L

30

p=0.08 p=0.05 p=0.02 p=0.001 p=0.3 25

20

15

10

5

0 Pre- treatment 24 hours 72 hours 96 hours 30 days

mg/L D-dimer 569 463 822 980 400 p=0.4

350 p=0.04 p=0.02 p=0.001 p=0.003

300

250

200

150 ** ** ** *** 100 *** ***

50

Pre- 0 treatment 24 hours 72 hours 96 hours 30 days

Figure 7c. Distribution of molecular markers of coagulation activity in relation to pre-treatment troponin T < 0.1 µg/L (white boxes) or ³ 0.1 µg/L (shaded boxes). Box-plots, symbols and statistics as in figure 7a

36 Clinical outcome in relation to the of fibrinogen in the top tertile (and without degree of inflammation and coagulation ischemic events during treatment) died or activity (papers II, IV and V) experienced a myocardial (re-)infarction or refractory angina from cessation of Pre-treatment inflammation (paper II) and treatment to 30-days follow-up, as compared coagulation activity (paper IV) to 10 (4.8 %) of the 208 patients with lower High fibrinogen levels, i.e. in the pre- pre-treatment fibrinogen levels, p=0.009. treatment top tertile, were related to increased risk of ischemic events at 30 Pre-treatment C-reactive protein levels days, table 5. However, the rate of adverse were not related to the composite of death, ischemic events tended to be lower during myocardial (re-)infarction or refractory the ongoing anticoagulant treatment in angina during or after anticoagulant patients with high pre-treatment treatment. However, high pre-treatment C- fibrinogen, but within the first four days reactive protein, i.e. in the top tertile, was after cessation of treatment there was a significantly related to increased mortality clinical reactivation with a more than at 30 days, table 5. doubled ischemic event rate. Thus, 13 of the 100 patients with pre-treatment levels

Table 5. Clinical outcome at 30 days in relation to pre-treatment levels of fibrinogen and C-reactive protein

Fibrinogen, g/L C-reactive protein, mg/L < 3.12 3.12– > 3.80 p <2.87 2.87- >5.65 p 3.80 5.65 n=105 n=108 n=101 n=102 n=102 n=102

Death, MI or ref angina 10 (9.5) 8 (7.4) 17 (17) 0.03 7 (6.9) 11 (11) 14 (14) 0.3 Death or MI 7 (6.7) 5 (4.6) 13 (13) 0.03 5 (4.9) 8 (7.8) 9 (8.8) 0.5 Death 0 1 (0.9) 3 (3.0) 0.1† 0 0 4 (3.9) 0.01†

MI = myocardial (re-)infarction, Ref = refractory. Values shown are number of patients with percentages of the group in parenthesis. P-values are calculated by Chi-square or †Fisher exact test (2-sided) for top versus bottom + middle tertiles.

37 The pre-treatment levels of F1+2 were not These results were more pronounced in related to short-term clinical outcome. the group of inogatran treated patients, High pre-treatment levels of TAT, i.e. in in which patients within the top tertile the top tertile, tended to be associated with of TAT had a better outcome both at the better clinical outcome during the end of infusion (p=0.02) and at 30 days anticoagulant treatment, table 6. Similarly, follow-up (p=0.03). Accordingly, no ischemic event was recorded during soluble fibrin levels in the top tertile were anticoagulant treatment in patients with related to better outcome at the end of high pre-treatment levels of soluble fibrin. inogatran infusion (p=0.02), and with a During treatment there was a trend to a trend to a lower event rate also at 30 days lower event rate in patients with high pre- follow-up (p=0.06). treatment levels of D-dimer, but this difference decreased during follow-up.

Table 6. Clinical outcome in relation to pre-treatment levels of coagulation markers

prothrombin fragment 1+2, nmol/L thrombin-antithrombin, mg/L <1.14 1.14-1.60 >1.60 p <2.73 2.73-4.20 >4.20 p n=103 n=105 n=103 n=104 n=107 n=101

Events 72 hours 5 (4.9%) 3 (2.9%) 4 (3.9%) 1† 4 (3.8%) 7 (6.5%) 1 (1.0%) 0.12† Events 7 days 10 (9.7%) 4 (3.8%) 11 (11%) 0.2 9 (8.7%) 10 (9.3%) 6 (5.9%) 0.4 Events 30 days 11 (11%) 10 (9.5%) 14 (14%) 0.4 12 (12%) 15 (14%) 8 (7.9%) 0.2

soluble fibrin, nmol/L D-dimer, mg/L <10 10-13 >13 p <82 82-149 >149 p n=97 n=120 n=97 n=105 n=106 n=103

Events 72 hours 5 (5.2%) 7 (5.8%) 0 0.02† 5 (4.8%) 5 (4.7%) 2 (1.9%) 0.3† Events 7 days 9 (9.3%) 11 (9.2%) 5 (5.2%) 0.2 7 (6.7%) 12 (11%) 6 (5.8%) 0.5 Events 30 days 12 (12%) 16 (13%) 7 (7.2%) 0.14 8 (7.6%) 16 (15%) 11 (11%) 0.8

Events denote a composite of death, myocardial (re-)infarction or refractory angina. Values shown are number of patients with percentages of the group in parenthesis. P-values are calculated by Chi2 or †Fisher exact test (2-sided) for top tertile versus combination of bottom + middle tertiles

38 Pre-treatment levels, i.e. in the top tertile, with unchanged or early increased levels of fibrinogen and D-dimer were predictors had an approximately doubled event rate of long-term mortality, and there was during the 30-day follow-up, table 7. furthermore a trend to a relation between Similarly, reduction in TAT levels at 6 high pre-treatment levels of C-reactive hours was seen in 189 (61 %) of the protein, F1+2 and TAT and increased patients and the event rate during follow- long-term risk of mortality, figure 8. up was more than two-fold elevated in the group of patients with unchanged or early Changes in coagulation activity during increased levels compared to the group treatment (paper IV) with decreased TAT levels, table 7. High pre-treatment levels were significantly correlated (p<0.001) to Decreased levels of soluble fibrin at 6 decreased levels of all four molecular hours were seen in 145 (47 %) of the coagulation markers respectively after 6 patients. The 7-days event rate was and 24 hours of anticoagulant treatment. doubled for patients with unchanged or In 257 (81 %) of the patients there was, as increased levels of soluble fibrin at 6 hours compared to the pre-treatment levels, a compared to the group with decreased reduction in F1+2 levels after 6 hours levels, table 7. Decreased D-dimer levels anticoagulant treatment. The 51 patients were not related to clinical outcome.

Table 7. Clinical outcome in relation to changes in levels of coagulation markers from pre-treatment to 6 hours

prothrombin fragment 1+2 thrombin-antithrombin

Change 0-6 h Decreasing Increasing Decreasing Increasing or unchanged or unchanged n=257 n=51 p n=189 n=121 p

Events 72 hours 9 (3.5%) 3 (5.9%) 0.4† 4 (2.1%) 8 (6.6%) 0.07† Events 7 days 17 (6.6%) 8 (16%) 0.03 10 (5.3%) 15 (12%) 0.02 Events 30 days 25 (9.7%) 10 (20%) 0.04 15 (7.9%) 20 (16%) 0.02

soluble fibrin D-dimer

Change 0-6 h Decreasing Increasing Decreasing Increasing or unchanged or unchanged n=145 n=164 p n=166 n=142 p

Events 72 hours 2 (1.4%) 10 (6.1%) 0.03 7 (4.2%) 5 (3.5%) 1† Events 7 days 8 (5.5%) 17 (10%) 0.12 13 (7.8%) 12 (8.5%) 0.8 Events 30 days 13 (9.0%) 33 (13%) 0.2 19 (11%) 16 (11%) 1

Events denote a composite of death, myocardial (re-)infarction or refractory angina. Values shown are number of patients with percentages of the group in parenthesis. P-values are calculated by Chi2 or †Fisher exact test (2-sided).

39 fibrinogen C-reactive protein 12 % death 12 % death

10 10 > 3.80 g/L > 5.65 mg/L 8 8 p=0.2 6 3.12-3.80 g/L 6

2.87-5.65 mg/L 4 p=0.04 4 < 2.87 mg/L 2 < 3.12 g/L 2

months months 0 0 0 6 12 18 24 30 36 42 48 0 6 12 18 24 30 36 42 48

prothrombin fragment 1+2 thrombin-antithrombin complex 12 % death 12 % death

10 > 1.60 nmol/L 10 > 4.20 mg/L 8 8 p=0. 4 6 1.14-1.60 nmol/L 6 2.73-4.20 mg/L 4 4 p=0.07 < 2.73 mg/L

2 < 1.14 nmol/L 2

months months 0 0 0 6 12 18 24 30 36 42 48 0 6 12 18 24 30 36 42 48

soluble fibrin D-dimer 12 % death 12 % death > 149 mg/L

10 10-13 nmol/L 10 p=0.003 8 8

> 13 nmol/L 6 6 82-149 mg/L

4 p=0.2 4

2 < 10 nmol/L 2 < 82 mg/L

months months 0 0 0 6 12 18 24 30 36 42 48 0 6 12 18 24 30 36 42 48

Figure 8. Probability of death during long-term follow-up, median 29 months; min 12 months, max 50 months; in relation to pre-treatment levels of the respective marker of inflammation or coagulation divided in tertiles. Statistics: Log rank test

40 Reactivation after cessation of predominantly 160 mg, and the highest anticoagulant treatment (paper IV) recorded daily dose was 500 mg. ASA There were signs of reactivation of treatment did not protect from clinical coagulation activity within 24 hours after reactivation and there was nothing in cessation of anticoagulant treatment with favour of higher doses (101), figure 9. an increase in F1+2, TAT, SF and D-dimer levels in 85 %, 69 %, 55 % and 83 % of Myocardial cell damage and changes in the patients respectively. An adverse coagulation activity (paper V) clinical event early after cessation of Troponin-positive patients tended to have treatment, i.e. at days 4-7, occurred in 9 worse short-term clinical outcome (100), (5.0 %) of the 181 patients with signs of without relation to markers of coagulation reactivation in thrombin generation by an activity. Among the troponin-negative increase in TAT levels after patients, i.e. with pre-treatment troponin discontinuation of study drug infusion. In T < 0.1 mg/L , those with unchanged or contrast, none of the 83 patients with early increased F1+2 or TAT during unchanged or decreased TAT levels after treatment had a cluster of ischemic events cessation of treatment died or experienced within 24 hours after cessation of the study a myocardial (re-)infarction or refractory drug, figure 10. The 30-day ischemic event angina at days 4-7 (p=0.06). The median rate was 26 % in troponin-negative increase in TAT levels within 24 hours patients with unchanged or early increased after cessation of treatment was 4.6 mg/L F1+2, and 5.7 % in patients with decreased in patients with an adverse clinical event F1+2, p=0.001. Similarly, 18 % of the days 4-7 and 1.0 mg/L in patients without troponin-negative patients with unchanged an event after cessation of treatment or early increased TAT, and 4.6 % of the (p=0.03). Increased F1+2, SF and D-dimer levels after cessation of treatment were, 12 % death, MI, refractory angina ASA > 75 mg/d although not significant, related to a 38 n = 487

%, 64 % and 57 % higher adverse event 10 p = 0.02 rate respectively at days 4-7. 8

ASA £ 75 mg/d Treatment with ASA 6 n = 656 At inclusion in the study 54 % of the 4 patients were on treatment with ASA, and at cessation of the study drug 95 % of the 2 patients were on ASA, 656 of them with a 0 daily dose equal to or lower than 75 mg 0 1 2 3 4 5 6 7 days (median). In 487 patients the daily ASA Figure 9. Composite of death, myocardial infarction doses were higher than the median, or refractory angina in relation to ASA dose, Chi2-test

41 patients with decreased TAT, had an 0.06 µg/L as cut-off limit for pre- ischemic event within 30 days, p=0.005. treatment troponin T or positive troponin Similar results were found using T in serial samples within the first 12 hours.

a) Pre-treatment troponin T < 0.1 µg/L

prothrombin fragment 1+2 thrombin-antithrombin complex 25 % death, MI, refractory angina 25 % death, MI, refractory angina Increased or unchanged 0-6 h n=34 20 20

15 15 Increased or p=0.001* unchanged 0-6 h n=68 10 10 p=0.003

5 5 Decreased 0-6 h n=141 Decreased 0-6 h n=107 0 0 0 1 2 3 4 5 6 7 days 0 1 2 3 4 5 6 7 days

b) Pre-treatment troponin T ³ 0.1 µg/L

prothrombin fragment 1+2 thrombin-antithrombin complex 25 % death, MI, refractory angina 25 % death, MI, refractory angina

20 20

15 15

10 Decreased 0-6 h 10 Decreased 0-6 h n=114 n=81 Increased or unchanged 0-6 h 5 5 n=52 p=ns Increased or unchanged 0-6 h n=17 0 0 0 1 2 3 4 5 6 7 days 0 1 2 3 4 5 6 7 days

Figure 10. Composite of death, myocardial (re-)infarction or refractory angina in patients with a) pre-treatment troponin T < 0.1 µg/L or b) pre-treatment troponin T ³ 0.1 µg/L, in relation to early decreased versus unchanged or increased levels of molecular markers of thrombin generation. Vertical dotted line represents termination of anticoagulant treatment. Statistics: Chisquare or *Fisher exact test

42 Multivariate logistic regression analyses including molecular markers of coagulation activity (papers IV and V) Multivariate logistic regression analyses included all relevant baseline characteris- tics and time from start of qualifying episode of chest pain to randomisation, pre-treatment troponin T (cut-off ³ 0.1 µg/L) and unchanged/increased levels of either of F1+2 or TAT during the first 6 hours. Unchanged/increased levels of any of these two coagulation markers, together with higher age and congestive heart failure, were found to be indepen- dent predictors of death, myocardial (re-)infarction or refractory angina up to 30 days. The use of both these coagulation markers in the model indicated that only unchanged/increased levels of F1+2, age and congestive heart failure were indepen- dent predictors of short-term adverse outcome.

Logistic regression models using interaction terms did not indicate a difference between inogatran or UF he- parin treatment concerning the prediction of clinical outcome by increasing coagulation activity during treatment.

43 Discussion

Atherosclerosis, inflammation, plaque events were significantly lower in the UF disruption and subsequent intracoronary heparin group. Similar discouraging thrombus formation are the predominant results, demonstrating no or only slight causes of unstable coronary artery disease benefits with other direct thrombin (1, 2). Thus, a key component in the inhibitors, such as recombinant hirudin management of these patients is (104, 105) or the synthetic efegatran (106) antithrombotic treatment, i.e. platelet have been reported in previous clinical inhibition with ASA and antithrombin- trials of these agents in unstable coronary dependent thrombin inhibition with artery disease. heparin (67, 102). Recently, the combination of ASA and heparin with Activated partial thromboplastin time other platelet inhibitors, i.e. glycoprotein The aPT time is used to monitor the IIb/IIIa-receptor inhibitors and ADP treatment effects of thrombin inhibitors as receptor inhibitors, have been shown to well in clinical trials as in routine care of further reduce the rate of ischemic events patients with unstable coronary artery (75-77). Despite this regimen and early disease or myocardial infarction (44). revascularisation routines, these patients However, there are many differences in are at considerable risk of new ischemic dosing regimens as well as in the target events within the first months (54, 68, 69). aPT times (104, 107-110). This in-vitro Therefore, and because of the several method reflects surface-induced activation drawbacks in heparin treatment, there is of the coagulation system and is mainly an extensive search for new and more sensitive to the intrinsic pathway. effective anticoagulant agents, such as However, the in-vivo coagulation direct thrombin inhibitors. activation in unstable CAD is mainly triggered by the extrinsic tissue factor/ Inogatran (87), a synthetic low molecular factor VII-pathway (24, 27). The weight direct thrombin inhibitor, has an relationship between aPT times and anticoagulant effect independent of plasma concentrations of heparin or antithrombin and heparin cofactor II, and recombinant hirudin, currently the best ability to inhibit both fibrin-bound and evaluated direct thrombin inhibitor, is non- free thrombin (85, 103). However, none linear (111-113). There are also of the doses of inogatran in the TRIM differences in aPT times at different study (92) proved more effective than UF laboratories as there is a considerable heparin in preventing new ischemic variation in available reagents giving events. In fact, during the 72 hours of widely different results (112, 114). A study drug infusion the rates of ischemic batch-to-batch variation of a particular aPT

44 time reagent may also occur (115). In- found between higher aPT times during vivo, r-hirudin infusion induces a dose- the first 12 hours and death or myocardial dependent predictable aPT time re-infarction (117). prolongation in patients with chronic stable CAD (116). In contrast, large The results in the present study might be interindividual variations of the aPT time related to the differences in the response were seen in healthy volunteers mechanisms of action between UF heparin at identical plasma levels of r-hirudin and direct thrombin inhibitors. Heparin measured by anti-IIa-concentration (113). exerts multiple effects on the coagulation system mainly by amplifying the effects The randomised low, medium and high of the natural anticoagulant antithrombin. doses of inogatran treatment in the TRIM Antithrombin, besides inactivation of and study (92), aimed at aPT times in the low binding to thrombin, also inhibits several to medium range. The positive correlation coagulation factors such as XIIa, XIa, IXa, between aPT times and adverse events was Xa, all of which can alter the aPT time. unexpected. However, a similar Direct thrombin inhibitors, on the other experience has been made in a porcine hand, have a selective inhibitory effect on model of copper-coil-induced coronary thrombin (factor IIa). Therefore, at artery thrombosis, where inogatran comparable aPT times the thrombin dosages resulting in aPT time prolongation activity is probably much lower during of as little as 1.3 x baseline proved more treatment with a direct thrombin inhibitor effective in inhibiting thrombotic than during UF heparin treatment. occlusion than heparin in dosages However, lower aPT times during resulting in aPT time prolongation of inogatran treatment were related to better 2.0 – 5.4 x baseline (91). clinical outcome, while better outcome was seen at higher aPT times during UF In contrast, during ongoing UF heparin heparin treatment. It is thus unlikely that infusion there was an association between the aPT time, in general, is an appropriate higher aPT times and improved clinical indicator of the antithrombotic efficacy of outcome in the present study. This clinical anticoagulant treatment. Other indicators benefit was lost within the first 24 hours of the influence of different drugs on the after cessation of treatment because of thrombotic-antithrombotic balance reactivation with clustering of events. It therefore need exploration. is unknown whether the peak in aPT times at the beginning of the UF heparin infusion Thrombin time, another method to assess is advantageous. However, in a previous anticoagulant effect during thrombin trial including almost 30.000 patients inhibition, showed a more linear receiving intravenous heparin adjunctive relationship with plasma concentrations of to thrombolysis, a direct relationship was inogatran in the present study.

45 Nevertheless, the small number of samples increase in inflammatory activity, while made it impossible to relate the thrombin the lasting elevation of fibrinogen times to clinical outcome. An alternative indicates the co-existence of a chronic sensitive assay to monitor the efficacy of low-grade inflammatory condition. direct thrombin inhibitors is the ecarin clotting time (118). This method is As expected, thrombin inhibitors reduce insensitive to UF heparin and has low the degree of coagulation activity, interindividual variability and strong although different molecular markers of linear correlation to plasma concentrations coagulation activity have been measured of recombinant hirudin in healthy in the clinical studies, with somewhat volunteers (119). Further evaluation of this conflicting results for various thrombin method in clinical trials of direct thrombin inhibitors. During inogatran infusion there inhibitors thereby seems warranted. was a persistent reduction in the levels of F1+2, reflecting reduced thrombin The influences of thrombin inhibition generation, and a trend for a dose- on inflammation and coagulation response. These results are in accordance activity with significantly reduced F1+2 levels Active inflammation, reflected by elevated after 4 hours infusion in a previous small levels of interleukins (12) and acute-phase inogatran tolerability study (37). The proteins, e.g. fibrinogen and C-reactive reduced thrombin generation is probably protein (13, 14), in unstable coronary due to suppression of the feed-back artery disease has previously been mechanism by which thrombin amplifies reported. In the present study, the changes its own generation through activation of of the levels of fibrinogen and C-reactive factors V and VIII (120, 121). protein had different time-courses. The fibrinogen level started to rise at 24 hours, Similarly, in a previous study (41), seemed to reach its maximum after 72-96 reduced F1+2 levels were seen at cessation hours and remained at 30 days higher than of 3-5 days treatment with recombinant before treatment. The initial increase in hirudin in 31 unstable coronary artery levels of C-reactive protein occurred disease patients. In contrast, in another somewhat earlier and was more study (40) there was no significant pronounced. However, the C-reactive reduction in F1+2 levels after 6 or 48 hours protein level at 30 days tended to be lower in 231 patients on treatment with two than before treatment. Neither the time- different doses of recombinant hirudin. As course of levels of fibrinogen nor C- blood sampling time-points and sample reactive protein levels seemed affected by size in the latter study were similar to the the treatment with different doses of present, these results suggest differences inogatran or UF heparin. Therefore, the between inogatran and recombinant short-lasting elevation of C-reactive hirudin concerning their influence on protein seems mainly related to a transient thrombin generation.

46 The UF heparin treatment resulted, as groups. Inogatran may decrease TAT compared to inogatran, in a more levels by forming inogatran-thrombin pronounced early decrease in F1+2 levels complexes or by inhibiting thrombin in the present study, probably through generation through the above mentioned inhibition of factor Xa. The subsequent positive feed-back mechanism (120, 121). increase in F1+2 levels to levels slightly UF heparin accelerates the formation of above the pre-treatment levels at cessation TAT, but the inhibition of factor Xa and of UF heparin infusion might reflect thrombin exerted by UF heparin may also restored thrombin generation, despite the result in a decrease in TAT formation. The on-going treatment. Similarly, in 164 observed decrease in TAT levels in the UF unstable coronary artery disease patients heparin group is thus a net of two randomised to UF heparin treatment in a theoretically counteractive effects. The previous study (40), the F1+2 levels were decreased TAT levels during infusion, significantly reduced at 6 hours but then observed in both inogatran and UF heparin increased during continued UF heparin treated patients, might be interpreted as a treatment. Likewise, in the FRISC reduction in thrombin generation and coagulation substudy (39), there was an activity. In the OASIS substudy (40), early reduction in levels of F1+2, followed recombinant hirudin seemed more by significant increase from day 2 to day effective than UF heparin in the reduction 5 despite on-going treatment with of TAT levels. However, a short infusion dalteparin, a LMW heparin, in 170 (4 hours) of argatroban, another low unstable coronary artery disease patients. molecular weight direct thrombin In contrast, two studies of 40 and 36 inhibitor, did not reduce the TAT levels unstable coronary artery disease patients in 43 unstable coronary artery disease respectively, reported no effect of UF patients (122). Overall, the interactions of heparin on the levels of F1+2. These heparin with antithrombin and inogatran discrepant results might be due to the with thrombin create difficulties in the sample size or different time-points in interpretation of thrombin generation and blood sampling, with no assessment activity, and comparison between different between 90 minutes and 24 hours (38, 41). thrombin inhibitors, on the basis of TAT Thus, UF heparin seems associated with levels. an initial reduction in thrombin generation, which however is not maintained during In all four treatment groups the D-dimer prolonged treatment. levels decreased in the present study, but this reduction seemed to be delayed, with The TAT levels were decreased during unchanged levels at six hours in contrast infusion in all four treatment groups in the to the rapid decrease in the molecular present study, with the greatest reduction markers of thrombin generation. The seen in the high-dose and UF heparin decrease in D-dimer levels should

47 probably be interpreted as decrease in results suggest a persistent higher fibrin production (123), especially coagulation activity, only temporarily considering the simultaneous reduction in attenuated by anticoagulant treatment. thrombin activity as indicated by reduction Such a persistent prothrombotic state is of soluble fibrin in the UF heparin group. furthermore supported by the finding of A similar delayed decrease in levels of D- 30-day levels of F1+2 and D-dimer still dimer was noticed in a previous study (40), higher than pre-treatment. Similarly, both in recombinant hirudin and UF sustained high levels of F1+2 at 6 months heparin treated patients. after an episode of unstable angina or myocardial infarction have been reported Antithrombin decreased consistently only (127). during UF heparin infusion. This gradual decrease is a well-known phenomenon Inflammation and coagulation activity accompanying heparin treatment (38, in relation to myocardial cell injury 124), and may be a therapeutical There are limited data concerning the drawback, especially after cessation of the inflammatory activity in patients with drug (125). unstable angina in relation to signs of myocardial cell damage. Consistent with A reactivation increase in prothrombotic previous studies, the troponin-positive potential after the abrupt cessation of patients had significantly higher levels of thrombin inhibition has been observed fibrinogen and C-reactive protein pre- both clinically (83, 84) and biochemically treatment and up to at least 96 hours (37, 40, 84, 122, 126), the latter defined thereafter in the present study. This as a rapid increase in the levels of enhanced acute-phase reaction is probably molecular markers to levels as high, or induced by the myocardial cell damage even higher than, pre-treatment levels. In (18-20). the present study there were signs of reactivation of coagulation activity within In a previous study (21) of unstable angina 24 hours after cessation of study drug patients without troponin elevation, there treatment, with an increase in levels of was no increase in the levels of C-reactive F1+2, TAT and D-dimer in a majority of protein up to 96 hours after admission, the patients, in accordance with previous despite ischemic episodes during 24 hours studies. continuous ECG-monitoring in the majority of the patients. In contrast, The reactivation increase in F1+2 and D- elevated levels of C-reactive protein dimer levels after cessation of treatment despite normal troponin T on admission was significantly greater in patients with were reported in another study of unstable pre-treatment levels in the top tertile of angina patients(14). In 8 of these 20 the respective coagulation marker. These patients the C-reactive protein level was

48 doubled at 24-72 hours after admission, in small cohorts of troponin-positive and all of them had an in-hospital major unstable angina patients (48, 49) and coronary event (death or myocardial suggest that early troponin elevation in infarction) or underwent urgent patients with unstable coronary artery revascularisation. disease is associated with raised coagulation activity (50). This concept is The present study is thus the first to report also supported by more frequent a significant acute-phase response, with intracoronary thrombi at angiography in an early increase in fibrinogen and patients with elevated troponin levels (51- especially C-reactive protein, in a large 53). number of unstable angina patients without any signs of myocardial cell The degree of inflammation and clinical damage. This acute-phase reaction outcome indicates other sources of inflammatory Several epidemiological studies have activity besides myocardial cell injury in identified elevated levels of markers of the acute phase of unstable coronary artery inflammation, mainly C-reactive protein, disease (25). However, the finding of a as risk indicators for future cardiovascular marked reduction in C-reactive protein events, both in apparently healthy men early after the acute episode, without any (128-131) and women (132). Furthermore, influence of the antithrombotic treatment, both fibrinogen and C-reactive protein indicates a spontaneous resolution of the have in several studies been found to acute inflammatory response. Still, a indicate increased short- and long-term chronic low grade inflammatory activity risk for adverse clinical outcome in might be present as indicated by the unstable coronary artery disease (14, 60- remaining high fibrinogen level. 64).

There were higher levels of markers of The differences in magnitude and time- thrombin generation (F1+2), thrombin course of elevations of fibrinogen and C- activity (soluble fibrin) and fibrin turnover reactive protein levels observed in the (D-dimer) in patients with elevated present study might indicate different troponin T. The higher coagulation underlying mechanisms for their activity in troponin-positive patients associations to new ischemic events in persisted during the 72 hours treatment unstable coronary artery disease. High with UF heparin or inogatran and up to fibrinogen level was related to a trend to a 24 hours thereafter. The results were lower rate of clinical events during on- similar using different cut-off limits for going anticoagulant treatment, but also a troponin T or analyses from serial samples marked clinical reactivation after cessation of troponin T. Our data confirm earlier of treatment with significantly higher rate findings of enhanced soluble fibrin levels of ischemic events at 30 days and

49 increased long-term mortality. Fibrinogen mainly seen in the top tertile of C-reactive is directly involved in the thrombotic protein. Therefore, much of the relation process, both in platelet aggregation cross- between the C-reactive protein level and linking the glycoprotein IIb/IIIa-receptors mortality could be explained by an on adjacent platelets, and in the inflammatory reaction in the damaged coagulation cascade where it is cleaved by myocardium. thrombin to soluble fibrin which subsequently polymerises to form the Recently, polymorphism in exon 2 of the fibrin network stabilising the platelet clot C-reactive protein gene has been (24). The sustained high levels of described, although no association with C- fibrinogen might thereby be associated to reactive protein regulation or long-term risk of thrombosis and concentration is known (137). One might myocardial (re-)infarction (60). speculate that differences in the acute- phase response in unstable coronary artery On the other hand, the C-reactive protein disease might reflect differences in the level was only related to increased individual response to inflammatory mortality and not to myocardial stimuli (1, 138). Interestingly, higher C- (re-)infarction in the present as in other reactive protein levels have been reported trials (60, 63, 133, 134). Similar to these in the off-spring of patients with results, elevated levels of interleukin-6 myocardial infarction (139). Thus, a low have been related to increased mortality, grade inflammatory activity might be but not an increase in the combined end- indicated by slight elevation of C-reactive point of death and myocardial infarction, protein and long-lasting fibrinogen in patients with unstable coronary artery elevation and furthermore associated with disease (135). C-reactive protein is mainly a propensity to pronounced inflammatory regulated by interleukin-6, which is response at plaque ruptures and/or present in the atherosclerotic plaque and thromboembolic myocardial damage. secreted by both endothelial cells, smooth Furthermore, such a mechanism could muscle cells, macrophages and T-cells explain the relations between transient (136). High levels of interleukin-6 may pronounced elevation of C-reactive thereby reflect greater atherosclerotic protein at acute events and long-term burden and/or increased inflammatory mortality, as well as the relations between activity in the plaques. Thus, these plaques slight elevations of fibrinogen and C- would be more vulnerable and prone to reactive protein and the risk of new deeper fissuring, causing more severe ischemic events in the chronic stage (128- thrombotic episodes (135). However, the 132). largest increase in C-reactive protein was observed in patients with myocardial cell injury. Also the raised mortality was

50 The degree of coagulation activity and thrombin generation and activity and clinical outcome fibrin turnover were at a higher risk, as There are few studies of the influence of indicated by higher age and more anticoagulant treatment on the relation concomitant diseases such as diabetes between the molecular markers of mellitus and congestive heart failure. coagulation activity and clinical outcome Therefore, the association of high in patients with unstable coronary artery coagulation activity with improved short- disease. High levels of fibrinopeptide A term clinical outcome was intriguing. sampled after the first 24 hours in patients High baseline levels were however with unstable angina without concomitant correlated to decrease of the respective anticoagulant treatment have associated coagulation marker during treatment. An with adverse in-hospital outcome (death early decrease in molecular markers of or myocardial infarction) (140). Lower thrombin generation (F1+2 and TAT) and levels of fibrinopeptide A, sampled after thrombin activity (soluble fibrin) during 24 hours treatment, have been connected anticoagulant treatment was also related to improved TIMI flow and greater cross- to significantly improved clinical outcome sectional area on repeated angiography in during 30-days follow-up. Furthermore, 163 patients with unstable angina on among the troponin-negative patients treatment with unfractionated heparin or those with unchanged or increased recombinant hirudin (141). Another trial thrombin generation early during of 64 unstable angina or acute myocardial treatment had a marked reactivation with infarction patients, who developed in- a cluster of ischemic events within 24 hospital recurrent ischemia despite ³ 72 hours after cessation of anticoagulant hours unfractionated heparin infusion, treatment, and a higher rate of ischemic found higher levels of F1+2 and events until 30 days. Also the multivariate fibrinopeptide A before the event (38). analyses identified unchanged or early increased F1+2 or TAT, together with age, In the present study we found a relation congestive heart failure and elevated between higher baseline levels of troponin, as independent predictors of molecular markers for thrombin death, (re-)myocardial infarction or generation and activity (TAT and soluble refractory angina at 30 days. fibrin) and a lower rate of cardiac events, i.e. death, myocardial (re-)infarction or Thus, a plausible explanation for the refractory angina, during anticoagulant relation between high coagulation activity treatment. These results were more and improved short-term clinical outcome pronounced during and after treatment might be that high initial levels of markers with inogatran, then at UF heparin of coagulation activity identify patients treatment. The patients with high pre- with a thrombotic condition as the major treatment levels of molecular markers of cause of instability who are the best

51 responders to anticoagulant therapy. Such established. In the present study there were an explanation would be in accordance signs of a clinical reactivation with with the more pronounced decrease in clustering of ischemic events after coagulation activity during treatment and cessation of the 72 hours anticoagulant the decreased risk of ischemic events in treatment. There were also trends to a patients with initially high coagulation relation between elevations in thrombin activity. When considering the more generation and activity and fibrin turnover, pronounced decrease in F 1+2, TAT and as indicated by early increased F1+2, soluble fibrin during the first six hours of TAT, soluble fibrin and D-dimer levels, infusion in the UF heparin group, as and higher cardiac event rate during the compared to the inogatran groups, an first four days after cessation of explanation to the clinical superiority of anticoagulant therapy. Concomitant UF heparin during ongoing treatment treatment with ASA did not prevent might be offered (92). reactivation, as has previously been suggested (83). Moreover, low ASA doses In contrast, unchanged or early increased (£ 75 mg daily) seemed favourable, levels of molecular markers of coagulation similar to previous findings in a study of activity could indicate therapeutic failure acute coronary syndromes in the pre- with continuing thrombus formation and thrombolytic era, where 100 mg ASA increased risk for new thrombotic events. daily without loading dose reduced death Alternatively, the unchanged or early or myocardial (re-)infarction after two increased levels of molecular markers of weeks significantly better than 1000 mg coagulation activity might identify daily or placebo (142). patients with different pathogenic mechanisms without coagulation The biochemical and clinical reactivation activation and, hence, no benefit of is of course a major concern for the future anticoagulant treatment. However, the development of direct and antithrombin- cluster of ischemic events early after dependent thrombin inhibitors in the short- cessation of treatment favours the and long-term treatment of unstable interpretation of therapeutic failure with coronary artery disease. The underlying subsequent clinical consequences of a pro- plaque disruption may take several months coagulant state. to heal, and higher degree of coagulation activity may persist for six months after Although reactivation in coagulation an episode of acute coronary syndrome activity after discontinuation of UF (127), thereby suggesting an explanation heparin and direct thrombin inhibitors for the association between higher degree have previously been described (37, 40, of coagulation activity in the acute phase 84, 122, 126), a definite association to and increased long-term mortality in uns- adverse ischemic events has not been table coronary artery disease. Several

52 epidemiological studies have also, similar interindividual dispersion of these to the findings in the present study, markers. identified molecular markers of thrombin generation and activity as well as fibrin The analyses of the relations of aPT times, turnover, such as F1+2, fibrinopeptide A levels of troponin and changes in levels and D-dimer, as risk indicators for future of molecular markers of coagulation cardiovascular events in apparently activity, and clinical outcome are healthy men (143-145). Thus, improved retrospective. Accordingly, they need to long-term antithrombotic treatment in be confirmed in future clinical trials. unstable coronary artery disease seems warranted.

Limitations The analyses of F1+2, TAT and D-dimer were performed with ELISA assays, which are well established and utilised in trials in similar patient categories (35, 36, 39, 40, 122). However, the method for analysis of soluble fibrin (96), although different from older methods (146), has the disadvantage of not being entirely specific, since a variety of fibrinogen degradation products can act as co-factors for tissue-type plasminogen activator (147) and thereby influence the analysis. ELISA methods of soluble fibrin analysis may have superior performance (35). The use of Fibrinopeptide A, another sensitive molecular marker of thrombin activity, has been restricted because of its high susceptibility to sampling artefacts (31).

Although the molecular markers of coagulation activity seem promising for identifying groups of patients at high risk for subsequent ischemic events and likely to respond to anticoagulant treatment, the individual risk assessment is nevertheless difficult because of the considerable

53 Summary and implications

In unstable coronary artery disease:

¨ Inogatran - a synthetic, low molecular weight, selective, direct thrombin inhibitor - did not prove better than UF heparin in the prevention of ischemic events in unstable coronary artery disease.

¨ Inogatran treatment resulted, compared to UF heparin, in stable and more predictable prolongation of aPT times, although the correlation of inogatran plasma concentrations and aPT times was moderate.

¨ Higher aPT times during inogatran treatment were, even in the expected therapeutic range, related to increased risk of ischemic events. This might indicate a narrow therapeutic window or a methodological problem when evaluating direct thrombin inhibitors.

¨ Myocardial cell injury, as detected by elevated troponins, was associated with higher degrees of coagulation activity and inflammation, the latter only in part an acute-phase response due to the tissue damage.

¨ The elevation of fibrinogen levels in the acute phase was sustained for at least 30 days, which might reflect a low grade inflammatory condition in atherosclerotic lesions and a prothrombotic state, thus explaining the association between the fibrinogen level and the long-term risk for thrombosis and new ischemic events. This risk seems diminished during treatment with thrombin inhibitors, but is recurring early after cessation of treatment.

¨ The pronounced elevation of C-reactive protein levels in the acute phase was transient, which might indicate a propensity to pronounced inflammatory response at plaque ruptures and/or thromboembolic myocardial damage, both of which might be associated to increased mortality.

¨ UF heparin seemed more effective than inogatran in early reducing thrombin generation and activity, although this reduction was attenuated during on-going treatment.

¨ Higher thrombin generation and activity might identify patients with a thrombotic condition as the major cause of instability, who are the best responders to anticoagulant therapy with more pronounced decrease in coagulation activity during treatment and reduced short-term risk of ischemic events.

54 ¨ In contrast, unchanged or early increased levels of molecular markers of thrombin generation and activity during anticoagulant treatment may indicate therapeutic failure, with continuing thrombus formation, and increased risk for new thrombotic events.

¨ Reactivation of coagulation activity was observed within 24 hours of cessation of anticoagulant treatment, which was associated with an increase in ischemic events.

¨ Initially high coagulation activity was related to more pronounced reactivation after cessation of treatment, suggesting a persistent higher coagulation activity only temporarily attenuated by anticoagulant treatment.

¨ The coagulation activity was at 30 days still higher than pre-treatment, furthermore indicating a persistent prothrombotic state.

Implications of the study:

¨ The more pronounced early decrease in thrombin generation and activity by UF heparin treatment, as compared to inogatran, might offer an explanation to the clinical superiority during ongoing UF heparin treatment in the TRIM study.

¨ The aPT time might be an inappropriate indicator of the antithrombotic efficacy of direct and antithrombin-dependent thrombin inhibitors.

¨ The mechanisms of the inflammatory response in the acute phase of unstable coronary disease need to be further explored in order to understand its association to future cardiovascular incidents.

¨ The increased long-term mortality in relation to higher degree of coagulation activity, together with the improved short-term prognosis by treatment with thrombin inhibitors, indicate the need for a refined long-term strategy in patients with unstable coronary artery disease, i.e. continued long-term antithrombotic treatment and/or elimination of the culprit lesion by revascularisation.

55 Acknowledgements

I wish to express my sincere gratitude to all those who contributed to the TRIM study and this thesis, with special attention to

Lars Wallentin, my supervisor, for generously sharing your excellent knowledge in the field of clinical research and for your devotion, patience and extreme efficiency.

Agneta Siegbahn, my co-supervisor and outstanding guide into the mysteries of coagulation, for your extra-ordinary enthusiasm.

Rikard Linder, co-author, much appreciated friend and colleague, for pleasant collaboration in analysing and presenting the TRIM substudy.

Lars Grip, co-author, for your dedicated interest in this research field and the preparation of manuscripts.

Bjarne Nørgaard, co-author in paper II, Nils Egberg and Göran Larson, co-authors in paper III and Kristian Thygesen, co-author in paper V.

My research fellows, Bertil Lindahl, always giving wise answers, Henrik Toss, for fruitful discussions on difficulties in the field of inflammation and coagulation, Gunnar Frostfeldt, room-mate and good friend, for nice co-operation in teaching, and Christina Christersson, for sharing my interest and keeping up the research on thrombosis and thrombin inhibitors.

The research nurses involved in the TRIM study, Eva Svensson, Gunilla Lindström and Gerd Ålsjö, for taking good care of the patients, with special gratitude to Gerd for “bringing order in chaos” in the more than complicated TRIM study protocol.

All other friends, colleagues and research nurses in Lars Wallentins research group – Bertil Andrén, Eric Björklund, Jörgen Cronblad, Erik Diderholm, Catrin Henriksson, Matts Högberg, Bo Lagerqvist, Stefan James, Tomas Jernberg, Nina Johnston, Gerhard Wikström and Margareta Åkerstedt.

Eva Lindmark and Matilda Johnell, friends and PhD-students in the coagulation laboratory.

Birgitta Fahlström and Margareta Gulin for excellent technical assistance in the analyses of markers of inflammation and coagulation activity.

Johan Bring and Johan Lindbäck at Statisticon, for your great ability to give a clinical cardiologist comprehensible explanations to statistical difficulties.

56 Anders Dahlberg, Bengt Ljungberg and Sören Wilhelmsson, dept. of medicine, Nyköpings lasarett, for your terrific support and for introducing me in the field of emergency medicine and cardiology.

Per Kvidal, my clinical supervisor, for your efforts to develop my clinical skills in parallel with this research work.

The outstanding secretaries at the dept of cardiology, especially Maria Medeen, for excellent assistance, skilful help with the lay-out of this thesis and for defeating dead-lines still smiling.

The department of medical sciences and its predecessor Kjell Öberg for providing the necessary background.

All the local investigators, research nurses and staff in the coronary care units of all TRIM study centres, especially those participating in the substudy*, for your invaluable assistance in patient enrolments and careful acquisition of the substudy blood samples.

Margareta and Abiel, my parents, for your love, support and having faith in me, and special thanks to my father for linguistic assistance.

Helena and Bozo, my parents-in-law, for your outstanding support and care of my children while finishing this thesis.

My wife Elisabet, and our lovely children Sofia, Henrik and Hanna.

This study was supported by AstraZeneca AB, Mölndal, Sweden, and by grants from the Swedish Heart and Lung Foundation, Uppsala County Association against Heart and Lung Diseases, and the Faculty of Medicine, Uppsala University, Sweden.

*)The TRIM substudy centres: Bollnäs, Borås, Eksjö, Falun, Göteborg (SU-Östra), Jönköping, Kalmar, Linköping, Lund, Skellefteå, Stockholm (Karolinska & Södersjukhuset), Sundsvall, Uppsala, Sweden; Aarhus, Bispebjerg, Sønderborg, Denmark; Flekkefjord, Fredrikstad, Norway.

57 References

1. Ross R. Atherosclerosis—an 9. van der Wal AC, Becker AE, van der inflammatory disease. N Engl J Med Loos CM, Das PK. Site of intimal 1999;340(2):115-26. rupture or erosion of thrombosed coronary atherosclerotic plaques is 2. Falk E, Shah PK, Fuster V. Coronary characterized by an inflammatory plaque disruption. Circulation process irrespective of the dominant 1995;92(3):657-71. plaque morphology. Circulation 1994;89(1):36-44. 3. Falk E. Unstable angina with fatal outcome: dynamic coronary thrombosis 10. Burke AP, Farb A, Malcom GT, Liang leading to infarction and/or sudden YH, Smialek J, Virmani R. Coronary death. Autopsy evidence of recurrent risk factors and plaque morphology in mural thrombosis with peripheral men with coronary disease who died embolization culminating in total suddenly. N Engl J Med vascular occlusion. Circulation 1997;336(18):1276-82. 1985;71(4):699-708. 11. Maseri A, Liuzzo G, Biasucci LM. 4. Yamada DM, Topol EJ. Importance of Pathogenic mechanisms in unstable microembolization and inflammation in angina. Heart 1999;82 Suppl 1:I2-4. atherosclerotic heart disease. Am Heart J 2000;140(6 Suppl):S90-102. 12. Biasucci LM, Vitelli A, Liuzzo G, Altamura S, Caligiuri G, Monaco C, et 5. Braunwald E. Unstable angina. A al. Elevated levels of interleukin-6 in classification. Circulation unstable angina. Circulation 1989;80(2):410-4. 1996;94(5):874-7.

6. Nobuyoshi M, Tanaka M, Nosaka H, 13. Berk BC, Weintraub WS, Alexander Kimura T, Yokoi H, Hamasaki N, et al. RW. Elevation of C-reactive protein in Progression of coronary ”active” coronary artery disease. Am J atherosclerosis: is coronary spasm Cardiol 1990;65(3):168-72. related to progression? J Am Coll Cardiol 1991;18(4):904-10. 14. Liuzzo G, Biasucci LM, Gallimore JR, Grillo RL, Rebuzzi AG, Pepys MB, et 7. Gertz SD, Malekzadeh S, Dollar AL, al. The prognostic value of C-reactive Kragel AH, Roberts WC. Composition protein and serum amyloid a protein in of atherosclerotic plaques in the four severe unstable angina. N Engl J Med major epicardial coronary arteries in 1994;331(7):417-24. patients greater than or equal to 90 years of age. Am J Cardiol 15. Jude B, Agraou B, McFadden EP, 1991;67(15):1228-33. Susen S, Bauters C, Lepelley P, et al. Evidence for time-dependent activation 8. Richardson PD, Davies MJ, Born GV. of monocytes in the systemic Influence of plaque configuration and circulation in unstable angina but not stress distribution on fissuring of in acute myocardial infarction or in coronary atherosclerotic plaques. stable angina. Circulation Lancet 1989;2(8669):941-4. 1994;90(4):1662-8.

58 16. Neri Serneri GG, Prisco D, Martini F, 24. Furie B, Furie BC. Molecular and Gori AM, Brunelli T, Poggesi L, et al. cellular biology of blood coagulation. Acute T-cell activation is detectable in N Engl J Med 1992;326(12):800-6. unstable angina. Circulation 1997;95(7):1806-12. 25. Lagrand WK, Visser CA, Hermens WT, Niessen HW, Verheugt FW, Wolbink 17. Braunwald E. Unstable angina: an GJ, et al. C-reactive protein as a etiologic approach to management. cardiovascular risk factor: more than an Circulation 1998;98(21):2219-22. epiphenomenon? Circulation 1999;100(1):96-102. 18. de Beer FC, Hind CR, Fox KM, Allan RM, Maseri A, Pepys MB. 26. Ernofsson M, Tenno T, Siegbahn A. Measurement of serum C-reactive Inhibition of tissue factor surface protein concentration in myocardial expression in human peripheral blood ischaemia and infarction. Br Heart J monocytes exposed to cytokines. Br J 1982;47(3):239-43. Haematol 1996;95(2):249-57. 27. Moreno PR, Bernardi VH, Lopez- 19. Pietila K, Harmoinen A, Poyhonen L, Cuellar J, Murcia AM, Palacios IF, Ruosteenoja R. C-reactive protein in Gold HK, et al. Macrophages, smooth subendocardial and transmural muscle cells, and tissue factor in myocardial infarcts. Clin Chem unstable angina. Implications for cell- 1986;32(8):1596-7. mediated thrombogenicity in acute coronary syndromes. Circulation 20. Hoffmeister HM, Büttcher E, Ehlers R, 1996;94(12):3090-7. Szabo S, Kazmaier S, Helber U, et al. Troponin release is associated with 28. Bajzar L, Nesheim ME, Tracy PB. The activation of an acute phase profibrinolytic effect of activated inflammatory reaction in unstable protein C in clots formed from plasma angina pectoris. Eur Heart J is TAFI-dependent. Blood 2000;21(Abstr Suppl):522:P2836. 1996;88(6):2093-100.

21. Liuzzo G, Biasucci LM, Rebuzzi AG, 29. Broze GJ, Jr. Tissue factor pathway Gallimore JR, Caligiuri G, Lanza GA, inhibitor. Thromb Haemost et al. Plasma protein acute-phase 1995;74(1):90-3. response in unstable angina is not induced by ischemic injury. Circulation 30. Collen D. The plasminogen 1996;94(10):2373-80. (fibrinolytic) system. Thromb Haemost 1999;82(2):259-70. 22. Ferreiros ER, Boissonnet CP, Pizarro 31. Ardissino D, Merlini PA, Eisenberg PR, R, Merletti PF, Corrado G, Cagide A, Kottke-Marchant K, Crenshaw BS, et al. Independent prognostic value of Granger CB. Coagulation markers and elevated C-reactive protein in unstable outcomes in acute coronary syndromes. angina. Circulation Am Heart J 1998;136(4 Pt 2 Suppl):S7- 1999;100(19):1958-63. 18.

23. Ruggeri ZM, Dent JA, Saldivar E. 32. al-Nozha M, Gader AM, al-Momen Contribution of distinct adhesive AK, Noah MS, Jawaid M, Arafa M. interactions to platelet aggregation in Haemostatic variables in patients with flowing blood. Blood 1999;94(1):172- unstable angina. Int J Cardiol 8. 1994;43(3):269-77.

59 33. Merlini PA, Ardissino D, Oltrona L, 40. Flather MD, Weitz JI, Yusuf S, Pogue Broccolino M, Coppola R, Mannucci J, Sussex B, Campeau J, et al. PM. Heightened thrombin formation Reactivation of coagulation after but normal plasma levels of activated stopping infusions of recombinant factor VII in patients with acute hirudin and unfractionated heparin in coronary syndromes. Arterioscler unstable angina and myocardial Thromb Vasc Biol 1995;15(10):1675- infarction without ST elevation: results 9. of a randomized trial. OASIS Pilot Study Investigators. Organization to 34. Kruskal JB, Commerford PJ, Franks JJ, Assess Strategies for Ischemic Kirsch RE. Fibrin and fibrinogen- Syndromes. Eur Heart J related antigens in patients with stable 2000;21(17):1473-81. and unstable coronary artery disease. N Engl J Med 1987;317(22):1361-5. 41. Merlini PA, Ardissino D, Rosenberg RD, Colombi E, Agricola P, Oltrona L, 35. Lee LV, Ewald GA, McKenzie CR, et al. In vivo thrombin generation and Eisenberg PR. The relationship of activity during and after intravenous soluble fibrin and cross-linked fibrin infusion of heparin or recombinant degradation products to the clinical hirudin in patients with unstable angina course of myocardial infarction. pectoris. Arterioscler Thromb Vasc Arterioscler Thromb Vasc Biol Biol 2000;20(9):2162-6. 1997;17(4):628-33.

36. Zoldhelyi P, Bichler J, Owen WG, Grill 42. Hirsh J. Heparin. N Engl J Med DE, Fuster V, Mruk JS, et al. Persistent 1991;324(22):1565-74. thrombin generation in humans during specific thrombin inhibition with 43. Menon V, Berkowitz SD, Antman EM, hirudin. Circulation 1994;90(6):2671- Fuchs RM, Hochman JS. New heparin 8. dosing recommendations for patients with acute coronary syndromes. Am J 37. Andersen K, Dellborg M, Emanuelsson Med 2001;110(8):641-50. H, Grip L, Swedberg K. Thrombin inhibition with inogatran for unstable 44. Braunwald E, Antman EM, Beasley angina pectoris: evidence for JW, Califf RM, Cheitlin MD, Hochman reactivated ischaemia after cessation of JS, et al. ACC/AHA guidelines for the management of patients with unstable short-term treatment. Coron Artery Dis angina and non-ST-segment elevation 1996;7(9):673-81. myocardial infarction: executive summary and recommendations. A 38. Merlini PA, Ardissino D, Bauer KA, report of the American College of Oltrona L, Pezzano A, Bottasso B, et Cardiology/American Heart al. Persistent thrombin generation Association task force on practice during heparin therapy in patients with guidelines (committee on the acute coronary syndromes. Arterioscler management of patients with unstable Thromb Vasc Biol 1997;17(7):1325-30. angina).Circulation 2000;102(10):1193-209. 39. Ernofsson M, Strekerud F, Toss H, Abildgaard U, Wallentin L, Siegbahn 45. Odegard OR, Lie M, Abildgaard U. A. Low-molecular weight heparin Heparin cofactor activity measured with reduces the generation and activity of thrombin in unstable coronary artery an amidolytic method. Thromb Res disease. Thromb Haemost 1975;6(4):287-94. 1998;79(3):491-4.

60 46. Jaffe AS, Ravkilde J, Roberts R, 54. The FRISC-II Study Group. Long-term Naslund U, Apple FS, Galvani M, et low-molecular-mass heparin in unstable al. It’s time for a change to a troponin coronary-artery disease: FRISC II standard. Circulation prospective randomised multicentre 2000;102(11):1216-20. study. Lancet 1999;354(9180):701-7.

47. The Joint European Society of 55. Antman EM, Cohen M, Bernink PJ, Cardiology/American College of McCabe CH, Horacek T, Papuchis G, Cardiology Committee for the et al. The TIMI risk score for unstable redefinition of myocardial infarction angina/non-ST elevation MI: A method Myocardial infarction redefined—a for prognostication and therapeutic consensus document. Eur Heart J decision making. Jama 2000;21(18):1502-13. 2000;284(7):835-42. 48. Terres W, Kummel P, Sudrow A, 56. Holmvang L, Clemmensen P, Wagner Reuter H, Meinertz T, Hamm CW. G, Grande P. Admission standard Enhanced coagulation activation in electrocardiogram for early risk troponin T-positive unstable angina stratification in patients with unstable pectoris. Am Heart J 1998;135(2 Pt coronary artery disease not eligible for 1):281-6. acute revascularization therapy: a 49. Giannitsis E, Bardorff MM, Schweikart TRIM substudy. ThRombin Inhibition S, Wiegand U, Kampmann M, Katus in Myocardial Infarction. Am Heart J HA. Relationship of cardiac troponin T 1999;137(1):24-33. and procoagulant activity in unstable angina. Thromb Haemost 57. Savonitto S, Ardissino D, Granger CB, 2000;83(2):224-8. Morando G, Prando MD, Mafrici A, et al. Prognostic value of the admission 50. Hamm CW, Braunwald E. A electrocardiogram in acute coronary classification of unstable angina syndromes. Jama 1999;281(8):707-13. revisited. Circulation 2000;102(1):118- 22. 58. Andersen K, Eriksson P, Dellborg M. Ischaemia detected by continuous on- 51. Benamer H, Steg PG, Benessiano J, line vectorcardiographic monitoring Vicaut E, Gaultier CJ, Aubry P, et al. predicts unfavourable outcome in Elevated cardiac troponin I predicts a patients admitted with probable high-risk angiographic anatomy of the unstable coronary disease. Coron culprit lesion in unstable angina. Am Heart J 1999;137(5):815-20. Artery Dis 1996;7(10):753-60.

52. Heeschen C, van Den Brand MJ, Hamm 59. Jernberg T, Lindahl B, Wallentin L. ST- CW, Simoons ML. Angiographic segment monitoring with continuous findings in patients with refractory 12-lead ECG improves early risk unstable angina according to troponin stratification in patients with chest pain T status. Circulation and ECG nondiagnostic of acute 1999;100(14):1509-14. myocardial infarction. J Am Coll Cardiol 1999;34(5):1413-9. 53. Lindahl B, Diderholm E, Lagerqvist B, Venge P, Wallentin L. Mechanisms behind the prognostic value of troponin T in unstable coronary artery disease – a FRISC II substudy. J Am Coll Cardiol 2001;38(4):979-86.

61 60. Toss H, Lindahl B, Siegbahn A, 66. Lindahl B, Venge P, Wallentin L. Wallentin L. Prognostic influence of Relation between troponin T and the increased fibrinogen and C-reactive risk of subsequent cardiac events in protein levels in unstable coronary unstable coronary artery disease. The artery disease. FRISC Study Group. FRISC study group. Circulation Fragmin during Instability in Coronary 1996;93(9):1651-7. Artery Disease. Circulation 1997;96(12):4204-10. 67. Bertrand ME, Simoons ML, Fox KA, Wallentin LC, Hamm CW, McFadden 61. Becker RC, Cannon CP, Bovill EG, E, et al. Management of acute coronary Tracy RP, Thompson B, Knatterud GL, syndromes: acute coronary syndromes et al. Prognostic value of plasma without persistent ST segment fibrinogen concentration in patients elevation; recommendations of the Task with unstable angina and non-Q-wave Force of the European Society of myocardial infarction (TIMI IIIB Trial). Cardiology. Eur Heart J Am J Cardiol 1996;78(2):142-7. 2000;21(17):1406-32.

62. Haverkate F, Thompson SG, Pyke SD, 68. The FRISC-II Study Group. Invasive Gallimore JR, Pepys MB. Production compared with non-invasive treatment of C-reactive protein and risk of in unstable coronary-artery disease: coronary events in stable and unstable FRISC II prospective randomised angina. European Concerted Action on multicentre study. FRagmin and Fast Thrombosis and Disabilities Angina Revascularisation during InStability in Pectoris Study Group. Lancet Coronary artery disease Investigators. 1997;349(9050):462-6. Lancet 1999;354(9180):708-15.

63. Heeschen C, Hamm CW, Bruemmer J, 69. Cannon CP, Weintraub WS, Simoons ML. Predictive value of C- Demopoulos LA, Vicari R, Frey MJ, et reactive protein and troponin T in al. Comparison of early invasive and patients with unstable angina: a conservative strategies in patients with comparative analysis. CAPTURE unstable coronary syndromes treated Investigators. Chimeric c7E3 with the glycoprotein IIb/IIIa inhibitor AntiPlatelet Therapy in Unstable tirofiban. N Engl J Med angina REfractory to standard treatment 2001;344(25):1879-87. trial. J Am Coll Cardiol 2000;35(6):1535-42. 70. Lewis HD, Jr., Davis JW, Archibald DG, Steinke WE, Smitherman TC, 64. Lindahl B, Toss H, Siegbahn A, Venge Doherty JE, et al. Protective effects of P, Wallentin L. Markers of myocardial aspirin against acute myocardial damage and inflammation in relation to infarction and death in men with long-term mortality in unstable unstable angina. Results of a Veterans coronary artery disease. FRISC Study Administration Cooperative Study. N Group. Fragmin during Instability in Engl J Med 1983;309(7):396-403. Coronary Artery Disease. N Engl J Med 2000;343(16):1139-47. 71. Cairns JA, Gent M, Singer J, Finnie KJ, 65. Hamm CW, Ravkilde J, Gerhardt W, Froggatt GM, Holder DA, et al. Aspirin, Jorgensen P, Peheim E, Ljungdahl L, sulfinpyrazone, or both in unstable et al. The prognostic value of serum angina. Results of a Canadian troponin T in unstable angina. N Engl J multicenter trial. N Engl J Med Med 1992;327(3):146-50. 1985;313(22):1369-75.

62 72. Theroux P, Ouimet H, McCans J, 78. Cohen M, Adams PC, Parry G, Xiong Latour JG, Joly P, Levy G, et al. J, Chamberlain D, Wieczorek I, et al. Aspirin, heparin, or both to treat acute Combination antithrombotic therapy in unstable angina. N Engl J Med unstable rest angina and non-Q-wave 1988;319(17):1105-11. infarction in nonprior aspirin users. Primary end points analysis from the 73. The RISC Group. Risk of myocardial ATACS trial. Antithrombotic Therapy infarction and death during treatment in Acute Coronary Syndromes with low dose aspirin and intravenous Research Group. Circulation heparin in men with unstable coronary 1994;89(1):81-8. artery disease. Lancet 1990;336(8719):827-30. 79. Gurfinkel EP, Manos EJ, Mejail RI, Cerda MA, Duronto EA, Garcia CN, et 74. Wallentin LC. Aspirin (75 mg/day) al. Low molecular weight heparin after an episode of unstable coronary versus regular heparin or aspirin in the artery disease: long-term effects on the treatment of unstable angina and silent risk for myocardial infarction, ischemia. J Am Coll Cardiol occurrence of severe angina and the 1995;26(2):313-8. need for revascularization. Research Group on Instability in Coronary Artery 80. The FRISC Study Group. Low- Disease in Southeast Sweden. J Am molecular-weight heparin during Coll Cardiol 1991;18(7):1587-93. instability in coronary artery disease. Lancet 1996;347(9001):561-8. 75. Yusuf S, Zhao F, Mehta SR, Chrolavicius S, Tognoni G, Fox KK. 81. Wallentin L. Low molecular weight Effects of clopidogrel in addition to heparins: a valuable tool in the aspirin in patients with acute coronary treatment of acute coronary syndromes. syndromes without ST-segment Eur Heart J 1996;17(10):1470-6. elevation. N Engl J Med 2001;345(7):494-502. 82. Eikelboom JW, Anand SS, Malmberg K, Weitz JI, Ginsberg JS, Yusuf S. 76. The PRISM-PLUS Study Investigators. Unfractionated heparin and low- molecular-weight heparin in acute Inhibition of the platelet glycoprotein coronary syndrome without ST IIb/IIIa receptor with tirofiban in elevation: a meta-analysis. Lancet unstable angina and non-Q-wave 2000;355(9219):1936-42. myocardial infarction. Platelet Receptor Inhibition in Ischemic Syndrome Management in Patients Limited by 83. Theroux P, Waters D, Lam J, Juneau M, McCans J. Reactivation of unstable Unstable Signs and Symptoms angina after the discontinuation of (PRISM-PLUS). N Engl J Med heparin. N Engl J Med 1998;338(21):1488-97. 1992;327(3):141-5.

77. The PURSUIT Trial Investigators. 84. Granger CB, Miller JM, Bovill EG, Inhibition of platelet glycoprotein IIb/ Gruber A, Tracy RP, Krucoff MW, et IIIa with eptifibatide in patients with al. Rebound increase in thrombin acute coronary syndromes. Platelet generation and activity after cessation Glycoprotein IIb/IIIa in Unstable of intravenous heparin in patients with Angina: Receptor Suppression Using acute coronary syndromes. Circulation Integrilin Therapy. N Engl J Med 1995;91(7):1929-35. 1998;339(7):436-43.

63 85. Gast A, Tschopp TB, Schmid G, Hilpert 91. Uriuda Y, Wang QD, Grip L, Ryden L, K, Ackermann J. Inhibition of clot- Sjoquist PO, Mattsson C. bound and free (fluid-phase thrombin) Antithrombotic activity of inogatran, a by a novel synthetic thrombin inhibito new low-molecular-weight inhibitor of (Ro 46-6240), recombinant hirudin and thrombin, in a closed-chest porcine heparin in human plasma. Blood model of coronary artery thrombosis. Coagul Fibrinolysis 1994;5(6):879-87. Cardiovasc Res 1996;32(2):320-7.

86. Callas DD, Hoppensteadt D, Fareed J. 92. The TRIM Study Group. A low Comparative studies on the molecular weight, selective thrombin anticoagulant and protease generation inhibitor, inogatran, vs heparin for inhibitory actions of newly developed unstable coronary artery disease. Eur site-directed thrombin inhibitory drugs. Heart J 1997;18:1416-1425. Efegatran, argatroban, hirulog, and hirudin. Semin Thromb Hemost 93. Babson AL, Olson DR, Palmieri T, 1995;21(2):177-83. Ross AF, Becker DM, Mulqueen PJ. The IMMULITE assay tube: a new 87. Teger-Nilsson AC, Bylund R, approach to heterogeneous ligand Gustafsson D, Gyzander E, Eriksson U. assay. Clin Chem 1991;37(9):1521-2. In vitro effects of inogatran, a selective low molecular weight thrombin 94. Pelzer H, Schwarz A, Stuber W. inhibitor. Thromb Res 1997;85(2):133- Determination of human prothrombin 45. activation fragment 1 + 2 in plasma with an antibody against a synthetic peptide. 88. Eriksson UG, Renberg L, Bredberg U, Thromb Haemost 1991;65(2):153-9. Teger-Nilsson AC, Regardh CG. 95. Pelzer H, Schwarz A, Heimburger N. Animal pharmacokinetics of inogatran, Determination of human thrombin- a low-molecular-weight thrombin antithrombin III complex in plasma inhibitor with potential use as an with an enzyme-linked immunosorbent antithrombotic drug. Biopharm Drug assay. Thromb Haemost Dispos 1998;19(1):55-64. 1988;59(1):101-6.

89. Chen LY, Nichols WW, Mattsson C, 96. Wiman B, Ranby M. Determination of Teger-Nilson AC, Wallin R, Saldeen soluble fibrin in plasma by a rapid and TG, et al. Aspirin does not potentiate quantitative spectrophotometric assay. effect of suboptimal dose of the Thromb Haemost 1986;55(2):189-93. thrombin inhibitor inogatran during coronary thrombolysis. Cardiovasc Res 97. Elms MJ, Bunce IH, Bundesen PG, 1995;30(6):866-74. Rylatt DB, Webber AJ, Masci PP, et al. Measurement of crosslinked fibrin 90. Gustafsson D, Elg M, Lenfors S, degradation products - an immunoassay Borjesson I, Teger-Nilsson AC. Effects using monoclonal antibodies. Thromb of inogatran, a new low-molecular- Haemost 1983;50(2):591-4. weight thrombin inhibitor, in rat models of venous and arterial thrombosis, 98. Müller-Bardorff M, Hallermayer K, thrombolysis and bleeding time. Blood Schroder A, Ebert C, Borgya A, Coagul Fibrinolysis 1996;7(1):69-79. Gerhardt W, et al. Improved troponin T ELISA specific for cardiac troponin T isoform: assay development and analytical and clinical validation. Clin Chem 1997;43(3):458-66.

64 99. Naslund U, Grip L, Fischer-Hansen J, procedures in patients with acute Gundersen T, Lehto S, Wallentin L. The myocardial ischaemia without ST impact of an end-point committee in a elevation: a randomised trial. large multicentre, randomized, placebo- Organisation to Assess Strategies for controlled clinical trial: results with and Ischemic Syndromes (OASIS-2) . without the end-point committee’s final Lancet 1999;353(9151):429-38. decision on end-points. Eur Heart J 1999;20(10):771-7. 106. Klootwijk P, Lenderink T, Meij S, Boersma H, Melkert R, Umans VA, et 100. Lüscher MS, Thygesen K, Ravkilde J, al. Anticoagulant properties, clinical Heickendorff L. Applicability of efficacy and safety of efegatran, a direct cardiac troponin T and I for early risk thrombin inhibitor, in patients with stratification in unstable coronary artery unstable angina. Eur Heart J disease. TRIM Study Group. Thrombin 1999;20(15):1101-11. Inhibition in Myocardial ischemia. Circulation 1997;96(8):2578-85. 107. OASIS Investigators. Comparison of the effects of two doses of recombinant 101. Oldgren J, Grip L, Wallentin L, and the hirudin compared with heparin in TRIM study group. Reactivation after patients with acute myocardial ischemia cessation of thrombin inhibition in without ST elevation: a pilot study. unstable coronary artery disease occurs Organization to Assess Strategies for regardless of aspirin dose. Circulation Ischemic Syndromes (OASIS) 1996;94:I-431(abstract). Investigators. Circulation 1997;96(3):769-77. 102. The Swedish National board of Health and Welfare. National guidelines for 108. Neuhaus KL, von Essen R, Tebbe U, coronary heart disease 2001. ISBN 91- Jessel A, Heinrichs H, Maurer W, et al. 7201-522-5. Safety observations from the pilot phase of the randomized r-Hirudin for 103. Weitz JI, Hudoba M, Massel D, Improvement of Thrombolysis (HIT- Maraganore J, Hirsh J. Clot-bound III) study. A study of the thrombin is protected from inhibition Arbeitsgemeinschaft Leitender by heparin-antithrombin III but is Kardiologischer Krankenhausarzte susceptible to inactivation by (ALKK). Circulation 1994;90(4):1638- antithrombin III-independent inhibitors. 42. J Clin Invest 1990;86(2):385-91. 109. Fox KA. r-hirudin in unstable angina 104. GUSTO-2B Investigators. A pectoris. Rationale and preliminary data comparison of recombinant hirudin from the Apt pilot study. Eur Heart J with heparin for the treatment of acute 1995;16 Suppl D:28-32. coronary syndromes. The Global Use of Strategies to Open Occluded 110. Antman EM. Hirudin in acute Coronary Arteries (GUSTO) IIb . N myocardial infarction. Thrombolysis Engl J Med 1996;335(11):775-82. and Thrombin Inhibition in Myocardial Infarction (TIMI) 9B trial. Circulation 105. OASIS-2 Investigators. Effects of 1996;94(5):911-21. recombinant hirudin (lepirudin) compared with heparin on death, myocardial infarction, refractory angina, and revascularisation

65 111. Bain B, Forster T, Sleigh B. Heparin 118. Nowak G, Bucha E. Quantitative and the activated partial thromboplastin determination of hirudin in blood and time—a difference between the in-vitro body fluids. Semin Thromb Hemost and in-vivo effects and implications for 1996;22(2):197-202. the therapeutic range. Am J Clin Pathol 1980;74(5):668-73. 119. Potzsch B, Hund S, Madlener K, Unkrig C, Muller-Berghaus G. Monitoring of 112. Tripodi A, Chantarangkul V, Arbini recombinant hirudin: assessment of a AA, Moia M, Mannucci PM. Effects of plasma-based ecarin clotting time hirudin on activated partial assay. Thromb Res 1997;86(5):373-83. thromboplastin time determined with ten different reagents. Thromb Haemost 120. Pieters J, Lindhout T, Hemker HC. In 1993;70(2):286-8. situ-generated thrombin is the only enzyme that effectively activates factor 113. Nurmohamed MT, Berckmans RJ, VIII and factor V in thromboplastin- Morrien-Salomons WM, Berends F, activated plasma. Blood Hommes DW, Rijnierse JJ, et al. 1989;74(3):1021-4. Monitoring anticoagulant therapy by activated partial thromboplastin time: 121. Fenton JW, 2nd, Villanueva GB, Ofosu hirudin assessment. An evaluation of FA, Maraganore JM. Thrombin native blood and plasma assays. inhibition by hirudin: how hirudin Thromb Haemost 1994;72(5):685-92. inhibits thrombin. Haemostasis 1991;21 Suppl 1:27-31. 114. Poller L, Thomson JM, Taberner DA. Use of the activated partial 122. Gold HK, Torres FW, Garabedian HD, thromboplastin time for monitoring Werner W, Jang IK, Khan A, et al. heparin therapy: problems and possible Evidence for a rebound coagulation solutions. Ric Clin Lab 1989;19(4):363- phenomenon after cessation of a 4-hour 70. infusion of a specific thrombin inhibitor in patients with unstable angina 115. Shojania AM, Tetreault J, Turnbull G. pectoris. J Am Coll Cardiol The variations between heparin 1993;21(5):1039-47. sensitivity of different lots of activated partial thromboplastin time reagent 123. Toss H, Wallentin L, Siegbahn A. produced by the same manufacturer. Effects of low-molecular-weight Am J Clin Pathol 1988;89(1):19-23. heparin treatment on fibrinolytic markers in unstable coronary artery 116. Zoldhelyi P, Webster MW, Fuster V, disease. Fibrinolysis and Proteolysis Grill DE, Gaspar D, Edwards SJ, et al. 2001;15(1):26-31. Recombinant hirudin in patients with chronic, stable coronary artery disease. 124. Marciniak E, Gockerman JP. Heparin- induced decrease in circulating Safety, half-life, and effect on antithrombin-III. Lancet coagulation parameters. Circulation 1977;2(8038):581-4. 1993;88(5 Pt 1):2015-22.

117. Granger CB, Hirsch J, Califf RM, Col 125. Grip L, Blomback M, Egberg N, Olsson J, White HD, Betriu A, et al. Activated A, Svane B, Velander M. Antithrombin partial thromboplastin time and III supplementation for patients outcome after thrombolytic therapy for undergoing PTCA for unstable angina acute myocardial infarction: results pectoris. A controlled randomized from the GUSTO-I trial. Circulation double-blind pilot study. Eur Heart J 1996;93(5):870-8. 1997;18(3):443-9.

66 126. Becker RC, Spencer FA, Li Y, Ball SP, 132. Ridker PM, Buring JE, Shih J, Matias Ma Y, Hurley T, et al. Thrombin M, Hennekens CH. Prospective study generation after the abrupt cessation of of C-reactive protein and the risk of intravenous unfractionated heparin future cardiovascular events among among patients with acute coronary apparently healthy women. Circulation syndromes: potential mechanisms for 1998;98(8):731-3. heightened prothrombotic potential. J Am Coll Cardiol 1999;34(4):1020-7. 133. Koukkunen H, Penttila K, Kemppainen A, Halinen M, Penttila I, Rantanen T, 127. Merlini PA, Bauer KA, Oltrona L, et al. C-reactive protein, fibrinogen, Ardissino D, Cattaneo M, Belli C, et interleukin-6 and tumour necrosis al. Persistent activation of coagulation factor-alpha in the prognostic mechanism in unstable angina and classification of unstable angina myocardial infarction. Circulation pectoris. Ann Med 2001;33(1):37-47. 1994;90(1):61-8. 134. James S, Armstrong P, Califf R, 128. Thompson SG, Kienast J, Pyke SD, Siegbahn A, Simoons M, Wallentin L. Haverkate F, van de Loo JC. C-reactive protein is independently Hemostatic factors and the risk of related to mortality but not to myocardial infarction or sudden death myocardial infarction in unstable in patients with angina pectoris. coronary artery syndrome. Eur Heart J European Concerted Action on 2001;22(Abstr Suppl):517:P2827. Thrombosis and Disabilities Angina Pectoris Study Group. N Engl J Med 135. Lindmark E, Diderhom E, Wallentin L, 1995;332(10):635-41. Siegbahn A. Relationship between interleukin-6 and mortality in patients 129. Ridker PM, Cushman M, Stampfer MJ, with unstable coronary artery disease. Tracy RP, Hennekens CH. Effects of an early invasive or non- Inflammation, aspirin, and the risk of invasive strategy. Jama cardiovascular disease in apparently 2001;286(17):In press. healthy men. N Engl J Med 1997;336(14):973-9. 136. Schieffer B, Schieffer E, Hilfiker- Kleiner D, Hilfiker A, Kovanen PT, 130. Koenig W, Sund M, Frohlich M, Kaartinen M, et al. Expression of Fischer HG, Lowel H, Doring A, et al. angiotensin II and interleukin 6 in C-Reactive protein, a sensitive marker human coronary atherosclerotic of inflammation, predicts future risk of plaques: potential implications for coronary heart disease in initially inflammation and plaque instability. healthy middle-aged men: results from Circulation 2000;101(12):1372-8. the MONICA (Monitoring Trends and Determinants in Cardiovascular 137. Cao H, Hegele RA. Human C-reactive Disease) Augsburg Cohort Study, 1984 protein (CRP) 1059G/C polymorphism. to 1992. Circulation 1999;99(2):237- J Hum Genet 2000;45(2):100-1. 42. 138. Yoshimoto T, Nakanishi K, Hirose S, 131. Danesh J, Whincup P, Walker M, Hiroishi K, Okamura H, Takemoto Y, Lennon L, Thomson A, Appleby P, et et al. High serum IL-6 level reflects al. Low grade inflammation and susceptible status of the host to coronary heart disease: prospective endotoxin and IL-1/tumor necrosis study and updated meta-analyses. Bmj factor. J Immunol 1992;148(11):3596- 2000;321(7255):199-204. 603.

67 139. Margaglione M, Cappucci G, Colaizzo 145. Lowe GD, Yarnell JW, Sweetnam PM, D, Vecchione G, Grandone E, Di Minno Rumley A, Thomas HF, Elwood PC. G. C-reactive protein in offspring is Fibrin D-dimer, tissue plasminogen associated with the occurrence of activator, plasminogen activator myocardial infarction in first-degree inhibitor, and the risk of major relatives. Arterioscler Thromb Vasc ischaemic heart disease in the Biol 2000;20(1):198-203. Caerphilly Study. Thromb Haemost 1998;79(1):129-33. 140. Ardissino D, Merlini PA, Gamba G, Barberis P, Demicheli G, Testa S, et al. 146. Godal HC, Abildgaard U. Gelation of Thrombin activity and early outcome soluble fibrin in plasma by ethanol. in unstable angina pectoris. Circulation Scand J Haematol 1966;3(5):342-50. 1996;93(9):1634-9. 147. Weitz JI, Cruickshank MK, Thong B, 141. Rao AK, Sun L, Chesebro JH, Fuster Leslie B, Levine MN, Ginsberg J, et al. V, Harrington RA, Schwartz D, et al. Human tissue-type plasminogen Distinct effects of recombinant activator releases fibrinopeptides A and desulfatohirudin (Revasc) and heparin B from fibrinogen. J Clin Invest on plasma levels of fibrinopeptide A 1988;82(5):1700-7. and prothrombin fragment F1.2 in unstable angina. A multicenter trial. Circulation 1996;94(10):2389-95.

142. Husted SE, Kraemmer Nielsen H, Krusell LR, Faergeman O. Acetylsalicylic acid 100 mg and 1000 mg daily in acute myocardial infarction suspects: a placebo-controlled trial. J Intern Med 1989;226(5):303-10.

143. Miller GJ, Bauer KA, Barzegar S, Cooper JA, Rosenberg RD. Increased activation of the haemostatic system in men at high risk of fatal coronary heart disease. Thromb Haemost 1996;75(5):767-71.

144. Ridker PM, Hennekens CH, Cerskus A, Stampfer MJ. Plasma concentration of cross-linked fibrin degradation product (D- dimer) and the risk of future myocardial infarction among apparently healthy men. Circulation 1994;90(5):2236-40.

68