Thrombosis in Clinical Practice

Thrombosis in Clinical Practice

Thrombosis in Clinical Practice Edited by

Andrew D Blann, Gregory YH Lip Haemostasis Thrombosis and Vascular Biology Unit University Department of Medidne City Hospital, Birmingham, UK Alexander GG Turpie Department of Medidne McMaster University Hamilton, Canada

LONDON AND NEW YORK

© 2005 Taylor & Francis, an imprint of the Taylor & Francis Group First published in the United Kingdom in 2005 by Taylor & Francis, an imprint of the Taylor & Francis Group, 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN TeL:+44 (0)20 7017 6000 Fax.: +44 (0) 20 7017 6699 E-mail: [email protected] Website: http://www.tandf.co.uk/medicine This edition published in the Taylor & Francis e-Library, 2005. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to http://www.ebookstore.tandf.co.uk/.” All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of the publisher or in accordance with the provisions of the Copyright, Designs and Patents Act 1988 or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1P 0LP. Although every effort has been made to ensure that all owners of copyright material have been acknowledged in this publication, we would be glad to acknowledge in subsequent reprints or editions any omissions brought to our attention. Although every effort has been made to ensure that drug doses and other information are presented accurately in this publication, the ultimate responsibility rests with the prescribing physician. Neither the publishers nor the authors can be held responsible for errors or for any consequences arising from the use of information contained herein. For detailed prescribing information or instructions on the use of any product or procedure discussed herein, please consult the prescribing information or instructional material issued by the manufacturer. A CIP record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Data available on application ISBN 0-203-64035-7 Master e-book ISBN

ISBN 0-203-67765-X (Adobe e-Reader Format) ISBN 1 84214 163 5 (Print Edition) Distributed in North and South America by Taylor & Francis 2000 NW Corporate Blvd Boca Raton, FL 33431, USA Within Continental USA Tel: 800 272 7737; Fax: 800 374 3401 Outside Continental USA Tel: 561 994 0555; Fax: 561 361 6018 E-mail: [email protected] Distributed in the rest of the world by Thomson Publishing Services Cheriton House North Way Andover, Hampshire SP10 5BE, UK Tel: +44(0)1264 332424 E-mail: [email protected] Composition by Newgen Imaging Systems (P) Ltd, Chennai, India

Contents Contributors

vi

Preface

x

1 The pathophysiology of thrombosis Lina Badimon and Juan Jose Badimon 2 Epidemiology of coagulopathy Andrew J Catto 3 Commonly used anticoagulant and antiplatelet drugs: aspirin, heparin, and warfarin Andrew D Blann 4 Bleeding risk and hemorrhage: why it happens and what to do about it Gualtiero Palareti, Benilde Cosmi, and Cristina Legnani 5 Thrombophilia and venous thrombosis Pieter W Kamphuisen, Harry R Büller, and Frits R Rosendaal 6 Antithrombotic therapy for atrial fibrillation Ioannis Karalis and Gregory YH Lip 7 Valves Walter Ageno and Alexander GG Turpie 8 Percutaneous coronary intervention Freek WA Verheugt 9 Coronary artery disease: coronary revascularization Petr Widimsky, Zbynĕk Straka, and Martin Pĕnička 10 Peripheral arterial disease M Burress Welborn III, Franklin S Yau, and G Patrick Clagett 11 Antithrombotic therapy for ischemic stroke Tarvinder S Dhanjal and Matthew Walters 12 Management of venous thromboembolism during pregnancy Ian A Greer and Andrew J Thomson 13 Thrombophilia Ian Jennings 14 Systemic thrombosis in children M Patricia Massicotte, Paul Monagle, and Anthony K Chan 15 New drugs and directions Bernd Jilma Index

1 18 36

66 85 105 130 140 152 190 230 252 270 294 323

351

Contributors Walter Ageno Associate Professor of Medicine Department of Clinical Medicine University of Insubria Varese, Italy Juan Jose Badimon Cardiovascular Biology Research Laboratory Mount Sinai School of Medicine New York, USA Lina Badimon Director The Cardiovascular Research Center, CSIC-ICCC Barcelona, Spain Andrew D Blann Senior Lecturer and Consultant Clinical Scientist Haemostasis, Thrombosis and Vascular Biology Unit University Department of Medicine Birmingham, UK Harry R Büller Academic Medical Center Department of Vascular Medicine Meibergdreef Amsterdam The Netherlands Andrew J Catto Senior Lecturer, Honorary Consultant Physician and MRC Clinician Scientist Academic Unit of Molecular Vascular Medicine Leeds, UK Anthony K Chan Associate Professor Department of Pediatrics McMaster University Hamilton, Ontario, Canada

G Patrick Clagett Department of Surgery University of Texas Dallas, Texas, USA Benilde Cosmi Department of Angiology and Blood Coagulation University Hospital S. Orsola-Malpighi Bologna, Italy Tarvinder S Dhanjal British Heart Foundation Research Fellow and Specialist Registrar in Cardiology Division of Cardiovascular Sciences Institute of Biomedical Research University of Birmingham Birmingham, UK Ian A Greer Depute Dean, Faculty of Medicine Regius Professor, Obstetrics and Gynaecology Division of Developmental Medicine Reproductive and Maternal Medicine University of Glasgow Glasgow, UK Ian Jennings UK National External Quality Assessment Scheme for Blood Coagulation Sheffield, UK Bernd Jilma Division of Haematology and Immunology Department of Clinical Pharmacology Vienna University Hospital Wien, Austria Pieter W Kamphuisen Department of Internal Medicine University Medical Center Nijmegen Nijmegen, The Netherlands Ioannis Karalis Research Fellow Haemostasis Thrombosis and Vascular Biology Unit University Department of Medicine Birmingham, UK

Cristina Legnani Department of Angiology and Blood Coagulation University Hospital S. Orsola-Malpighi Bologna, Italy Gregory YH Lip Consultant Cardiologist and Professor of Cardiovascular Medicine Director, Haemostasis Thrombosis and Vascular Biology Unit University Department of Medicine Birmingham, UK M Patricia Massicotte The Peter Olley Chair in Pediatric Thrombosis Pediatric Thrombosis Program, Clinical Director Professor, Department of Pediatrics University of Alberta Edmonton, Alberta Canada Paul Monagle Division of Laboratory Services Royal Children’s Hospital Department of Pediatrics University of Melbourne Melbourne, Australia Gualtiero Palareti Department of Angiology and Blood Coagulation University Hospital S. Orsola-Malpighi Bologna, Italy Martin Pĕnička Cardiovascular Center Aalst Aalst, Belgium Frits R Rosendaal Department of Clinical Epidemiology, and Hemostasis and Thrombosis Research Center Leiden The Netherlands Zbynĕk Straka Cardiocentre Charles University Srobarova Prague, Czech Republic

Andrew J Thomson Consultant in Obstetrics & Gynaecology and Honorary Senior Lecturer Department of Obstetrics and Gynaecology Royal Alexandra Hospital Paisley, Scotland Alexander GG Turpie Department of Medicine McMaster University Hamilton, Canada Freek WA Verheugt Professor of Cardiology Heartcenter Department of Cardiology University Medical Center St Radboud Nijmegen, The Netherlands Matthew Walters Clinical Scientist Gardiner Institute Western Infirmary Glasgow University of Glasgow Glasgow, UK M Burress Welborn III Assistant Professor of Surgery UT Southwestern Medical Center Division of Vascular Surgery Dallas, Texas, USA Petr Widimsky Cardiocenter Charles University Srobarova Prague, Czech Republic Franklin S Yau Department of Surgery University of Texas Dallas, Texas, USA

Preface As thrombosis will cause (via cardiovascular disease, cancer, and connective tissue diseases) the majority of morbidity and mortality in the Western and developed world, then reduction of this major risk factor is clearly a problem that must be addressed. This will be even more pertinent as the new and developing world is rapidly catching up and acquiring the disease profile of the old. This book tries to address/highlight the questions relating to the pathophysiology and clinical practice related to thrombosis. Our initial questions, therefore, begin with “how can we detect thrombosis?”, and “how can we prevent them?” In order to answer the first question, we need first to get to grips with the mechanistic nuts and bolts of thrombosis, and Chapters 1 and 2 are designed to help out in this respect. Chapter 3 provides a historical perspective of antithrombotic therapies. However, overambitious use of antithrombotic agents can lead to hemorrhage—a topic discussed in Chapter 4. Most episodes of thrombosis, and almost all of the life-threatening events, occur in the arteries—thromboses in the venous circulation are discussed in Chapter 5. Chapters that follow provide a systematic and comprehensive view of the various conditions that lead to (arterial) thrombosis in adult life and different ways to treat them. For example, atrial fibrillation is covered in Chapter 6, artificial valves in Chapter 7, coronary interventions in Chapters 8 and 9 and peripheral artery disease in Chapter 10. Stroke is addressed in Chapter 11, and Chapter 12 discusses problems associated with pregnancy. Despite addressing all the possible risk factors for thrombosis (such as smoking and diabetes), some individuals are still prone to thrombosis. This relatively new syndrome—thrombophilia, often the result of a genetic polymorphism—is discussed in Chapter 13. Chapter 14 looks at problems associated with thrombosis in children and acknowledging the shortcomings of existing therapies. We conclude with Chapter 15 on new drugs and directions. This latter chapter comes at a time when the next few years will see the introduction of entirely new classes of antithrombotic agents. Our expectant readers are physicians, general practitioners, nurse practitioners, and other healthcare workers who care for patients presenting with thrombosis-related problems, and thus, the scope is necessarily wide. Evermindful of the fast-moving pace of this area, we have included as many up-to-date guidelines as are feasible. We thank our excellent colleagues for their help, encouragement, and contribution. Andrew D Blann Gregory YH Lip Birmingham, United Kingdom Alexander GG Turpie Hamilton, Canada January 2005

1 The pathophysiology of thrombosis Lina Badimon and Juan Jose Badimon Introduction Arterial and venous thrombosis and thromboembolism cause life-threatening episodes, disabilities, reduction in quality of life, and death. Whereas arterial thrombi are predominantly formed by platelets, venous thrombi are intravascular deposits composed predominantly of fibrin and red cells, with a variable platelet and leukocyte component [1]. Growing thrombi may locally occlude the lumen, or embolize and be washed away by the blood flow to occlude distal vessels. However, thrombi may be physiologically and spontaneously lysed by mechanisms that block thrombus propagation. Thrombus size, location, and composition are regulated by hemodynamic forces (mechanical effects), thrombogenicity of exposed substrate (local molecular effects), relative concentration of fluid phase and cellular blood components (local cellular effects), and the efficiency of the physiologic mechanisms of control of the system, mainly fibrinolysis [2, 3]. Mechanisms of thrombus formation Physiologically, blood constituents do not attach nor adhere to intact endothelial structures. However, the exposure of subendothelial structures, or foreign cardiovascular devices, to flowing blood initiates complex mechanisms that trigger platelet activation and deposition, fibrin formation and deposition, leukocyte accretion, and erythrocyte entrapping in variable flow-dependent characteristics. The dynamics of platelet deposition and thrombus formation are modulated by the type of injury and the local geometry at the site of damage [2–4] (Table 1.1). Platelets The recognition of subendothelial ligands by platelets involves (a) adhesion, activation, and adherence to recognition sites on the thromboactive substrate (extracellular matrix proteins; e.g. von Willebrand factor, collagen, fibronectin, vitronectin, laminin), (b) spreading of the platelet on the surface, and (c) aggregation of platelets with each other to form a platelet plug or white thrombus. The efficiency of the platelet recruitment will depend on the underlying substrate and local geometry. A final step of recruitment of other blood cells also occurs; erythrocytes, neutrophils, and occasionally monocytes are found on evolving mixed thrombus (Figure 1.1). Platelet function depends on adhesive interactions and most of

Thrombosis in clinical practice

2

Table 1.1 Factors modulating thrombus formation. Nature of the exposed substrate • Degree of injury (mild vs severe arterial injury) • Composition of atherosclerotic plaque • Residual mural thrombus Local fluid dynamics • Shear stress • Tensile stress Systemic thrombogenic factors Hypercholesterolemia Catecholamines (smoking, cocaine, stress, etc.) Smoking Diabetes Homocysteine Lipoprotein (a) Infections (Chlamidia pneumoniae, Helicobacter pylori, Cytomegalovirus) Hypercoagulable state (fibrinogen, von Willebrand factor, Tissue factor, factor VII) Defective fibrinolytic state, etc.

the glycoproteins on the platelet membrane surface are receptors for adhesive proteins. Many of these receptors have been identified, cloned, sequenced, and classified within large gene families that mediate a variety of cellular interactions [5] (Table 1.2). The understanding of the biochemical events involved in platelet activation has progressed significantly. Most platelet aggregation agonists seem to act through the hydrolysis of platelet membrane phosphatidylinositol by phospholipase C, which results in the mobilization of free calcium from the platelet-dense tubular system. Exposed matrix from the vessel wall and thrombin generated by the activation of the coagulation cascade as well as circulating epinephrine are powerful platelet agonists. Adenosine diphosphate (ADP) is a platelet agonist that may be released from hemolyzed red cells in the area of vessel injury. Each agonist stimulates the discharge of calcium and promotes the subsequent release of its granule contents. Platelet-related ADP and serotonin stimulate adjacent platelets, further enhancing the process of platelet activation. Arachidonate, which is released from the platelet membrane by the stimulatory effect of collagen, thrombin, ADP, and serotonin, is another platelet agonist. Arachidonate is converted to thromboxane A2 by the sequential effects of cyclooxygenase and thromboxane synthetase. Thromboxane A2 not only promotes further platelet aggregation but is also a potent vasoconstrictor [6].

The pathophysiology of thrombosis

3

The most abundant family of platelet surface receptors is the integrin family, which includes GPIIb/IIIa, GPIa/IIa, GPIc/IIa, the fibronectin receptor, and the vitronectin receptor, in decreasing order of magnitude. Another gene family present in the platelet membrane glycocalyx is the leucine-rich glycoprotein family represented by the GPIb/IX complex, receptor for von Willebrand factor (vWF) on unstimulated platelets that mediates adhesion to subendothelium and GPV. Other gene families include the selectins (such as GMP-140) and the immunoglobulin domain protein (HLA class I antigen and platelet/endothelial cell adhesion molecule 1, PECAM-1). Unrelated to any other gene family is the GPIV (IIIa) (5). The GPIb/IX complex consists of two disulfide-linked subunits (GPIbα and GPIbβ) tightly (not covalently) complexed with GPIX in a 1:1 heterodimer. GPIbβ and GPIX are transmembrane glycoproteins and form the larger globular domain. The major role of GPIb/IX is to bind immobilized vWF on the exposed vascular subendothelium and initiate adhesion of platelets. GPIb does not bind

Figure 1.1 Diagram of platelet-vessel wall and platelet-platelet interaction with the participation of the coagulation and spontaneous fibrinolysis systems. (Continuous lines: pathways of activation; Dashed lines: pathways of inhibition.)

Thrombosis in clinical practice

4

soluble vWF in plasma. Apparently, it undergoes a conformation change upon binding to the extracellular matrix and then exposes a recognition sequence for GPIb/IX. The cytoplasmic domain of GPIb/IX has a major function in linking the plasma membrane to the intracellular actin filaments of the cytoskeleton and functions to stabilize the membrane and to maintain the platelet shape [7]. Randomly distributed on the surface of resting platelets are about 50,000 molecules of GPIIb/IIIa. The complex is composed of one molecule of GPIIb (disulfide-linked large and light chains) and one of GPIIIa (single polypeptide chain). It is a Ca2+-dependent heterodimer, noncovalently associated on the platelet membrane [8]. Calcium is required for maintenance of the complex and for binding of adhesive proteins. On activated platelets, the GPIIb/IIIa is a receptor for fibrinogen, fibronectin, vWF, vitronectin, and thrombospondin [9, 10].

Table 1.2 Platelet membrance glycoprotein receptors. Glycoprotein receptor

Function

Ligand

GPIIb/IIIa

• Aggregation, adhesion at high shear rate

Fg, vWF, Fn, Ts, Vn

Receptor Vn

• Adhesion

Vn, vWF, Fn, Fg, Ts

GPIa/IIa

• Adhesion

C

GPIc/IIa

• Adhesion

Fn

GPIcN/IIa

• Adhesion

Ln

GPIb/IX

• Adhesion

vWF, T

GPV

• Unknown

Substrate T

GPIV (GPIIIb)

• Adhesion

Ts, C

GMP-140 (PADGEM)

• Interaction with leucocytes

Unknown

PECAM-1 (GPIIa)

• Unknown

Unknown

Fg: fibrinogen; vWF: von Willebrand factor; Fn: fibronectin; Ts; thrombospondin; Vn: vitronectin; C; collagen; Ln: laminin; T: thrombin; PECAM-1: platelet/endothelial cell adhesion molecule 1.

Thrombin plays an important role in the pathogenesis of arterial thrombosis. It is one of the most potent known agonists for platelet activation and recruitment. The thrombin receptor has 425 amino acids with seven transmembrane domains and a large NH2terminal extracellular extension that is cleaved by thrombin to produce a “tethered” ligand that activates the receptor to initiate signal transduction [11, 12]. Thrombin is a critical enzyme in early thrombus formation, cleaving fibrinopeptides A and B from fibrinogen to yield insoluble fibrin, which effectively anchors the evolving thrombus. Both free and fibrin-bound fibrin thrombin are able to convert fibrinogen to fibrin allowing propagation of thrombus at the site of injury. Therefore, platelet activation triggers intracellular signaling and expression of platelet membrane receptors for adhesion and initiation of cell contractile processes that induce shape change and secretion of the granular contents. The expression of the integrin IIb/IIa

The pathophysiology of thrombosis

5

(αIIbβ3) receptors for adhesive glycoprotein ligands (mainly fibrinogen and vWF) in the circulation initiates platelet-to-platelet interaction. The process becomes perpetuated by the arrival of platelets brought by the circulation. Most of the glycoproteins in the platelet membrane surface are receptors for adhesive proteins or mediate cellular interactions. It has been shown that vWF binds to platelet membrane glycoproteins in both adhesion (platelet-substrate interaction) and aggregation (platelet-platelet interaction), leading to thrombus formation at high shear rates [7, 13, 14]. In Figure 1.2, a simplified diagram shows different platelet membrane receptors involved in platelet activation. Ligand binding to the different membrane receptors triggers platelet activation with different relative potencies. A lot of interest has been recently generated on the platelet ADP-receptors (P2Y12, P2Y1R, P2X1R) because of available pharmacological inhibitors [15].

Figure 1.2 Simplified diagram of platelet receptors and ligands. Activation of the coagulation system The activation of the coagulation cascade leads to the generation of thrombin, which is a powerful platelet agonist that contributes to platelet recruitment in addition to catalyzing the formation and polymerization of fibrin. Fibrin is essential to the stabilization of the platelet thrombus and its withstanding removal forces by flow, shear, and high intravascular pressure (Figure 1.3). The proteins which compose the clotting enzymes do not collide and interact on a random basis in the plasma but interact in complexes in a highly efficient manner on platelet and endothelial surfaces. The major regulatory events in the coagulation (activation, inhibition, generation of anticoagulant proteins) occur on membrane surfaces. It is interesting to note that venous thrombosis, which is predominantly constituted by fibrin clots occurs in areas of stasis and low-shear-rate conditions typical of the venous system. Therefore, the low local-shear-rate conditions and flow recirculations developing

Thrombosis in clinical practice

6

in the poststenotic areas may explain fibrin accumulation. Tissue factor (TF) appears to be a major procoagulant factor in the vascular space immediately underlying the endothelial lining of arteries, a site that might be readily accessible upon local injury or upon rupture of an atherosclerotic plaque. The blood coagulation system involves a sequence of reactions integrating zymogens (proteins susceptible to be activated to enzymes via limited proteolysis) and cofactors (nonproteolytic enzyme activators) in three groups: (a) the contact activation (generation of factor XIa via the Hageman factor), (b) the

Figure 1.3 Analysis of early thrombosis on human atherosclerotic plaques perfused in the Badimon perfusion chamber for 3 minutes. Immunofluorescence analysis of a mural growing thrombus on an atherosclerotic plaque. Red (rhodamine) marks fibrin-fibrinogen deposition and green (fluorescein) marks platelet deposition. conversion of factor X to factor Xa in a complex reaction requiring the participation of factors IX and VIII, and (c) the conversion of prothrombin to thrombin and fibrin formation [16] (Figure 1.4). The triggering surfaces for in vivo initiation of contact activation have been suggested to be sulfatides and glycosaminoglycans of the vessel wall. The physiologic role of this system is unclear, however, because the absence of Hageman factor, prekallikrein, or highmolecular-weight kininogen does not induce a clinically apparent pathology. Factor XI deficiency is associated with abnormal bleeding. Activated factor XI induces the activation of factor IX in the presence of Ca2+. Factor IXa forms a catalytic complex with factor VIII on a membrane surface and efficiently activates factor X in the presence of Ca2+. Factor IX is a vitamin K-dependent enzyme, as are factor VII, factor X, prothrombin, and protein C.

The pathophysiology of thrombosis

7

Factor VIII forms a noncovalent complex with vWF in plasma and its function in coagulation is the acceleration of the effects of IXa on the activation of X to Xa. Absence of factor VIII or IX produces the hemophilic syndromes. The TF pathway, previously known as extrinsic coagulation pathway, through the TFfactor VII complex in the presence of Ca2+ induces the formation of Xa. A second TFdependent reaction catalyzes the transformation of IX into IXa. TF is an integral membrane protein that serves to initiate the activation of factors IX and X and to localize the reaction to cells on which TF is expressed. Other cofactors include factor VIIIa, which binds to platelets and forms the binding site for IXa, thereby forming the machinery for the activation of X, and factor Va, which binds to platelets and provides a binding site for Xa. The human genes for these cofactors have been cloned and sequenced. In physiologic conditions, no cells in contact with blood contain active TF, although cells such as monocytes and polymorphonuclear leukocytes can be induced to synthesize and express TF [16]. Activated Xa converts prothrombin into thrombin. The complex which catalyzes the

Figure 1.4 Simplified diagram of the coagulation cascade and thrombosis. formation of thrombin consists of factors Xa and Va in a 1:1 complex. The activation results in the cleavage of fragment F1.2 and formation of thrombin from fragment 2. The interaction of the four components of the “prothrombinase complex” (Xa, Va, phospholipid, and Ca2+) yield a more efficient reaction [17]. Activated platelets provide procoagulant surface for the assembly and expression of both intrinsic Xase and prothrombinase enzymatic complexes [18]. These complexes respectively catalyze the activation of factor X to factor Xa and prothrombin to thrombin. The expression of activity is associated with the binding of both the proteases, factor IXa and factor Xa, and the cofactors, VIIIa and Va, to procoagulant surfaces. The binding of IXa and Xa is promoted by VIIIa and Va, respectively, such that Va and likely VIIIa provide the equivalent of receptors for the proteolytic enzymes. The surface of the

Thrombosis in clinical practice

8

platelet expresses the procoagulant phospholipids that bind coagulation factors and contribute to the procoagulant activity of the cell. Blood clotting is blocked at the level of the prothrombinase complex by the physiologic anticoagulant-activated protein C and oral anticoagulants. Oral anticoagulants prevent posttranslational synthesis of γ-carboxyglutamic acid groups on the vitamin K-dependent clotting factors, preventing binding of prothrombin and Xa to the membrane surface. Activated protein C cleaves factor Va to render it functionally inactive. Loss of Va decreases the role of thrombin formation to negligible levels [19]. Thrombin acts on multiple substrates, including fibrinogen, factor XIII, factor V, factor VIII, and protein C, in addition to its effects on platelets. It plays a central role in hemostasis and thrombosis. The catalytic transformation of fibrinogen into fibrin is essential in the formation of the hemostatic plug and in the formation of arterial thrombi. It binds to the fibrinogen central domain and cleaves fibrinopeptides A and B, resulting in fibrin monomer and polymer formation [18–20]. The fibrin mesh holds the platelets together and contributes to the attachment of the thrombus to the vessel wall. The presence of a residual mural thrombus predisposes to recurrent thrombotic episodes. Two main contributing factors for the development of rethrombosis have been identified. First, a residual mural thrombus may encroach into the vessel lumen resulting in increased shear rate in an artery or areas of stasis in venous thrombi, which facilitates the activation and deposition of platelets on the lesion. Second, the presence of a fragmented thrombus appears to be one of the most powerful thrombogenic surfaces. Thus, following lysis, thrombin becomes exposed to the circulating blood, leading to activation of the platelets and coagulation, further enhancing thrombosis. In these conditions, the antithrombin activity of heparin is limited for three main reasons. First, a residual thrombus contains active thrombin bound to fibrin, which is thus poorly accessible to the large heparin-antithrombin III complexes; second, a plateletrich arterial thrombus releases large amounts of platelet factor 4, which may inhibit heparin; third, fibrin II monomer, formed by the action of thrombin on fibrinogen, may also inhibit heparin. Conversely, molecules of hirudin and other specific antithrombins are at least ten times smaller than the heparin-antithrombin III complex, have no natural inhibitors, and therefore have greater accessibility to thrombin bound to fibrin. Spontaneous anticoagulation and fibrinolysis The control of the coagulation reactions occurs by diverse mechanisms, such as hemodilution and flow effects, proteolytic feedback by thrombin, inhibition by plasma proteins (such as antithrombin III (ATIII)) and endothelial celllocalized activation of an inhibitory enzyme (protein C), and fibrinolysis. Although ATIII readily inactivates thrombin in solution, its catalytic site is inaccessible while bound to fibrin, and it may still cleave fibrinopeptides even in the presence of heparin. Thrombin has a specific receptor in endothelial cell surfaces, thrombomodulin, that triggers a physiologic anticoagulative system [21, 22]. The thrombin-thrombomodulin complex serves as a receptor for the vitamin K-dependent protein C that is activated and released from the endothelial cell surface. Activated protein C blocks the effects of factors V and VIII and limits thrombin effects. Endogenous fibrinolysis represents a repair mechanism such as

The pathophysiology of thrombosis

9

endothelial cell regrowth and vessel recanalization. Fibrinolysis involves catalytic activation of zymogens, positive and negative feedback control, and inhibitor blockade. Thrombus formation in different locations Thrombosis of left ventricular and left atrial chambers Intracavitary mural thrombi develop frequently in patients with acute myocardial infarction, chronic left ventricular infarction, chronic left ventricular aneurysm, dilated cardiomyopathy, and atrial fibrillation. The clinical significance of thrombosis in the cardiac chambers derives from its potential for systemic embolism, which also depends on dynamic forces of the circulation [23, 24]. In the first few days after acute myocardial infarction, leukocyte infiltration separates endothelial cells from their basal lamina. The resulting exposure of subendothelial tissue to blood serves as the nidus for thrombus development. Specific endocardial abnormalities have also been identified histologically in surgical and postmortem specimens from patients with left ventricular aneurysms and at necropsy in patients with idiopathic dilated cardiomyopathy. Both experimental and clinical [25, 26] studies have emphasized the importance of wall motion abnormalities in the development of left ventricular thrombi, and it seems clear that stasis of blood in regions of akinesis or dyskinesis is the essential factor. Similarly, stasis is important in the development of atrial thrombi [27] when effective mechanical atrial activity is impaired, as occurs in atrial fibrillation, atrial enlargement, mitral stenosis, and cardiac failure. Stasis is associated to conditions of low shear rate in which activation of coagulation factors rather than of platelet leads to fibrin formation and constitutes the predominant pathogenic mechanism in the development of intracavitary thrombi. Although a hypercoagulable state is controversial, a hypercoagulable tendency in this condition was suggested, and it is conceivable that a systemic procoagulant tendency arises during the acute stage of myocardial infarction and predisposes to thromboembolic events. More relevant is experimental evidence that the surface of a fresh thrombus is itself highly thrombogenic, producing at least a local, if not a systemic, hypercoagulable state [28]. The problems of thromboembolism originating from the cardiac chambers prompt consideration of the balance between the effects of regional injury, stasis, and procoagulant factors that favor thrombus formation, and dynamic forces of the circulation, which are responsible for the migration of thrombotic material into the systemic circulation. Even though stasis favors thrombus formation within the sac of a left ventricular aneurysm, isolation from dynamic circulatory forces protects against embolic migration. In diffuse dilated cardiomyopathy, on the other hand, mural thrombus is not isolated from the circulation and the embolic risk is higher.

Thrombosis in clinical practice

10

Prosthetic heart valves and thromboembolism Mechanical prosthesis. The pathologic events leading to thromboembolism already develop during surgery. The damaged peri-valvular tissue and prosthetic materials activate platelets as soon as blood starts flowing across the valve, and immediate platelet deposition results [29]. Prosthetic materials activate the intrinsic clotting system, with subsequent fibrin formation [30]. In addition, the abnormal hemodynamic characteristics of mechanical prosthetic devices promote mainly fibrin generation and, less importantly, platelet activation. Finally, the process of fibrin thrombus formation can be facilitated in areas with stasis and decreased blood flow such as in the left atrium during atrial fibrillation and in the left ventricle during low cardiac output state secondary to left ventricular dysfunction [31]. Biologic prosthesis. Bioprosthetic valves are considerably less thrombogenic mainly because of the biologic properties of materials used in their construction and also because of characteristic axial flow profile and leaflet pliability. Venous thrombosis and pulmonary embolism Venous thrombi usually form in regions of slow or disturbed flow and begin as small deposits that frequently arise in large venous sinuses in the calf, in valve cusp pockets either in the deep veins of the calf or thigh, or in venous segments that have been exposed to direct trauma [32–34], The major predisposing factors to venous thrombosis are activation of blood coagulation and venous stasis, whereas in arterial thrombosis it is vascular wall damage, usually of atherosclerotic origin [35–38]. Nevertheless, wall damage may predispose to venous thrombosis in special circumstances. Arterial thrombosis The formation of a thrombus within an artery with obstruction of blood flow and oxygen supply to the target organ produces the acute ischemic syndromes. These thrombotic episodes largely occur in response to atherosclerotic lesions that have slowly progressed to a high-risk inflammatory/prothrombotic stage. Although distinct from one another, the atherosclerotic and thrombotic processes appear to be closely interrelated as the causal presentation of ischemic syndromes in a complex multifactorial process named atherothrombosis [39]. Atherosclerotic disease in the coronary artery system may manifest in the form of stable or unstable angina, acute myocardial infarction, or sudden cardiac death. Atherosclerotic disease in the cerebral arterial system (including the intracranial and extracranial arteries) can manifest in the form of transient ischemic attack (TIA) and cerebral ischemic infarction. In contrast to that of acute coronary syndromes, the pathophysiologic process leading to acute cerebral ischemia is less well defined but appears to involve multiple etiologic mechanisms. Atherosclerotic disease of the abdominal or leg arteries can result in acute and chronic ischemia of the extremities and usually involves thrombosis, embolism, or both, originating from atherosclerotic plaques.

The pathophysiology of thrombosis

11

Thrombosis in the acute coronary syndromes Fissuring or rupture of an atherosclerotic plaque in the coronary arteries plays a fundamental role in the development of the acute coronary syndromes, as has been clearly shown in patients who died suddenly or shortly after an episode of unstable angina or myocardial infarction. Disrupted atherosclerotic plaques are commonly associated with the formation of mural or occlusive thrombi usually anchored to fissures in the ruptured or ulcerated plaque. Angiographic studies have documented the presence of intraluminal thrombi both in unstable angina and acute myocardial infarction. The incidence of thrombi in unstable angina varied significantly among different studies, in part related to the interval between anginal symptoms and arteriographic study. The shorter this interval, the higher the likelihood of finding occlusive thrombi. Severity of vessel wall damage is probably the most important of all the previously mentioned “thrombogenic risk factors” in the acute coronary syndromes. It is likely that when injury to the vessel wall is mild, the thrombogenic stimulus is relatively limited and the resulting thrombotic occlusion transient, as occurs in unstable angina. On the other hand, deep vessel injury secondary to plaque rupture and ulceration results in exposure of collagen, lipids, and other elements of the vessel media, leading to relatively persistent thrombotic occlusion and myocardial infarction. Although a substantial proportion of episodes of unstable angina and acute myocardial infarction are caused by plaque fissuring or rupture with superimposed thrombosis, other mechanisms that alter the balance between myocardial oxygen supply and demand need to be considered [2, 3]. In patients with stable coronary disease, angina commonly results from increases in myocardial oxygen demand beyond the ability of stenosed coronary arteries to increase its delivery. In contrast, unstable angina, non-Q wave myocardial infarction, and Q wave myocardial infarction represent a continuum of the disease process and are usually characterized by an abrupt reduction in coronary flow. In unstable angina, disruption of an atherosclerotic plaque may lead to acute changes in plaque morphology and reduction in coronary flow. Transient episodes of thrombotic vessel occlusion at the site of plaque injury may occur, leading to angina at rest. This thrombus is usually labile, resulting in only temporary vessel occlusion. In addition to plaque disruption, other mechanisms contribute to reduced coronary flow. Platelets attach to the damaged endothelium or to exposed media and release vasoactive substances including thromboxane A2 and serotonin, leading to aggregation of neighboring platelets and vasoconstriction. Alterations in perfusion and myocardial oxygen supply probably account for two-thirds of episodes of unstable angina; the rest may be caused by transient increases in myocardial oxygen demand. In non-Q wave infarction, the angiographic morphology of the responsible lesion is similar to that seen in unstable angina, suggesting that plaque disruption is common to both syndromes. About one-fourth of patients with non-Q wave infarction have a completely occluded infarct-related vessel at early angiography, with the distal territory usually supplied by collaterals. The presence of ST segment elevation in the electrocardiogram, early peak in plasma creatine kinase, and high angiographic patency rate of the involved vessel suggest that complete coronary occlusion followed by early

Thrombosis in clinical practice

12

reperfusion (within the first 2 h) or resolution of vasospasm, are pathogenetically important in non-Q wave infarction. In Q wave infarction, plaque rupture is commonly associated with deep arterial injury or ulceration, resulting in the formation of a fixed and persistent thrombus, leading to abrupt cessation of myocardial perfusion and necrosis. The coronary lesion responsible for the infarction is frequently only mild to moderately stenotic, which suggests that plaque rupture with superimposed thrombosis is the primary determinant of acute occlusion rather than of the severity of the underlying lesion [40]. While severe preexisting lesions often lead to complete vessel occlusion, myocardial infarction does not commonly supervene, perhaps owing to adequate collateral flow. In perhaps onefourth of patients, coronary thrombosis results from superficial intimal injury or blood stasis in areas of high-grade stenosis. In sudden cardiac death, two mechanisms predominate. Sudden death related to ischemia probably involves a rapidly progressive coronary lesion in which plaque rupture and resultant thrombosis lead to myocardial hypoperfusion and fatal ventricular arrhythmias. Absence of collateral flow to the myocardium distal to the occlusion or platelet microthrombi may contribute to the development of sudden ischemic death. Additionally, fatal ventricular arrhythmias are common in patients, after extensive myocardial infarction or other forms of cardiomyopathy, in whom a substrate for the generation and maintenance of ventricular tachycardia or fibrillation exists (Figure 1.5). Thrombosis in cerebrovascular disease Clinical manifestations of atherosclerotic disease in the cerebral circulation (intracranial and extracranial) results in a spectrum of acute

Figure 1.5 Thrombosis on an eroded (non-ruptured) human atherosclerotic coronary artery. Original preparation from the Eulalia Study on Sudden Death (H & E stain).

The pathophysiology of thrombosis

13

cerebral ischemic syndromes ranging from TIAs to full-blown cerebral infarction [41, 42]. Similar to the pathogenesis of acute coronary syndromes, thrombosis over a disrupted plaque plays a key pathogenic role in the majority of patients suffering from these vascular events. However, in contrast to acute coronary syndromes, other pathogenic mechanisms such as intracranial hemorrhage, subarachnoid hemorrhage, and cardiogenic embolism also play a role in a substantial proportion of these patients. TIAs are brief episodes of focal loss of brain function, thought to be due to ischemia, that can usually be localized to that portion of brain supplied by one vascular system, and to which no other cause can be found. Though arbitrary, by convention, episodes lasting less than 24 hours are classified as TIAs although the longer the episode, the greater the likelihood of finding a cerebral infarct by computer tomography. TIAs commonly last 2– 15 minutes and are rapid in onset. Each TIA leaves no persistent deficit, and there are often multiple attacks. Most patients have TIAs that include motor symptoms. It is common for amaurosis fugax (monocular blindness) to occur without other symptoms during the episode. In general, TIAs evolve from two causes: focal low blood flow and embolism. The mechanism of focal low flow in TIAs is not well defined. Probably, a critical stenotic or occluded artery reduces flow to a focal area of normal brain. Certainly, poor collateral circulation to the ischemic area plays a prominent role, but factors such as viscosity, vessel wall compliance, and other unknown factors are needed to explain why the reduction is transient. The presence of fibrin or fibrin-platelet mural thrombus is found overlying an atherosclerotic plaque in two-thirds of cases with hemispheric TIAs. In contrast, evidence of ulceration on the plaques, which are widely regarded as sources of emboli, is only found in part of the cases with TIA, and at a similar rate in the asymptomatic cases. That the incidence of ulceration was about the same in TIA and asymptomatic groups appeared to preclude an important role for ulceration as the etiology of repetitive TIAs; however, investigations are underway on this topic with the use of more sophisticated technologies. Intraplaque hemorrhage is present in over one-third of cases of TIAs. Future investigation into the pathogenesis of carotid intraplaque hemorrhage might indicate that this is the result of small plaque fissuring and dissection, as recently demonstrated in acute coronary syndromes. Overall, the evidence from different studies indicate that probably in a substantial proportion of patients, TIAs result from progressive luminal narrowing leading to precarious hemodynamic insufficiency. One can also infer that mural thrombi at the subocclusive stage can either contribute to the obstruction or creation of an emboli. Large plaque ulceration or intraplaque hemorrhage did not correlate well with clinical ischemic events, but probably play a role in plaque growth. It has been assumed that fragments of mural thrombi of extracranial atheromatous plaque break and cause TIAs. However, the proportion of patients with TIAs of embolic origin is not clearly defined. As discussed previously, pathologic studies in patients with TIA indicated that about 30% of endarterectomy specimens obtained demonstrated no overlying thrombi. Furthermore, probably in a substantial proportion of patients with TIAs, cardiogenic emboli (emboli from cardiac chambers) appears to be the pathogenic mechanism. Clinical data indicated that at least 25% of all cerebral ischemic events and more than a third of cerebral ischemic events in the elderly are associated with atrial fibrillation. In addition, about 1 in 3 patients with atrial fibrillation will experience a

Thrombosis in clinical practice

14

cerebral ischemic event during lifetime. These confounding data have complicated our understanding of the pathogenic mechanisms involved in transient ischemic events. Further prospective clinical studies are needed in order to assess the relative importance of various pathogenic processes in different risk groups and to formulate accurate therapeutic strategies. The clinical hallmark of a stroke (acute cerebral infarction) is the abrupt development of a focal neurologic deficit due to ischemia or bleeding in a particular territory [43]. Intracerebral or subarachnoid hemorrhage accounts for 15% of strokes. The vast majority of strokes (up to 85%) are ischemic, arising from atherosclerotic disease. The prevalence of presumed cardioembolic stroke varies between 15% and 30% suggesting either patient population differences or variability in application of diagnostic criteria. In addition to the lack of validated, reliable, clinical diagnostic criteria for differentiating between cardioembolic and no embolic ischemic stroke, a significant proportion of the patients have risks for both mechanisms of stroke. Thus, about 30% of patients with acute ischemic stroke will have a potential cardiac source of embolism, but about one-third of these patients will also have concomitant cerebrovascular atherosclerosis that could also be responsible for brain ischemia. Thrombosis in peripheral arterial disease The most characteristic symptom of peripheral arterial disease is intermittent claudication, which occurs when the oxygen demand of exercising skeletal muscle exceeds the flow reserve of the diseased limb arteries. Characteristically, intermittent claudication occurs reproducibly with exercise at a given workload and is relieved by rest. Thus, intermittent claudication represents the skeletal muscle counterpart to stable angina pectoris. Peripheral arterial disease tends to be slowly progressive and may remain asymptomatic for long periods, particularly when collateral circulation is well developed [44]. An abrupt onset of ischemic rest pain or sudden worsening of intermittent claudication may be due to thromboembolism, or thrombosis in situ complicating an atherosclerotic lesion. Embolism in a previously healthy artery is more likely when symptoms occur suddenly in a patient without prior claudication, and with risk factors for thromboembolism (such as atrial fibrillation, left ventricular failure, prosthetic heart valves, or aortic aneurysm). When this occurs, advanced ischemia is common, owing to inadequate time for development of collateral circulation. A syndrome of microembolization may also occur, commonly with atherothrombotic materials released from a proximal arterial lesion (such as an aortic aneurysm). The clinical presentation associated with microembolization is that of digital pain, cyanosis, or gangrene, with preserved pulses and skin temperature in the extremity. This may also occur as a complication of arterial catheterization. Sudden worsening of intermittent claudication, or progression of intermittent claudication to rest pain or gangrene, is often seen with thrombotic or embolic occlusion of a preexisting stenotic lesion. When in situ thrombosis complicates an existing lesion, the presumed mechanism in many cases is plaque rupture or fissuring as previously described, but is also probably due to stasis of blood flow with a severely stenotic lesion in some cases. The clinical outcome in such cases is most dependent on the adequacy of the collateral circulation.

The pathophysiology of thrombosis

15

Because peripheral arterial disease is frequently asymptomatic for long periods, the frequency of intermittent claudication is likely to underestimate the incidence of significant peripheral arterial disease. Acknowledgments The work reported in this chapter has been partially supported by grants from Plan Nacional de Salud SAF2000/0174; FIS C03–01. References [1] Badimon L, Badimon JJ, Fuster V. Pathogenesis of thrombosis. In: Fuster V, Verstraete M (eds), Thrombosis in Cardiovascular Disorders. Philadelphia, PA: WB Saunders Company; Chapter 2, 1992:17–39. [2] Badimon L, Chesebro JH, Badimon JJ. Thrombus formation on ruptured atherosclerotic plaques and rethrombosis on evolving thrombi. Circulation 1992; 86:III 74–III 85. [3] Fuster V, Badimon L, Badimon JJ, Chesebro JH. The pathogenesis of coronary artery disease and the acute coronary syndromes. (Part I). N Engl J Med 1992; 326:242–50. (Part II). N Engl J Med 1992; 326:310–18. [4] Badimon L, Badimon JJ, Fuster V. Pathogenesis of thrombosis. In: Verstraete M, Fuster V, Topol EJ (eds), Cardiovascular Thrombosis. Philadelphia, PA: Lippincot-Raven Publisher; 1998:23–44. [5] Kieffer N, Phillips DR. Platelet membrane glycoproteins: functions in cellular interactions. Annu Rev Biol 1990; 6:329–57. [6] Brass LF. The biochemistry of platelet activation. In: Hoffman R, Benz EJ Jr, Shattil SJ, Furie B, Cohen HJ (eds), Hematology: Basic Principles and Practice. New York: Churchill Livingstone; 1991:1176–97. [7] Meyer D, Girma JP. von Willebrand factor: structure and function. Thromb Haemost 1993; 70:99–104. [8] Fitzgerald LA, Phillips DR. Calcium regulation of the platelet membrane glycoprotein IIb-IIIa complex. J Biol Chem 1985; 260:11366–76. [9] Coller B. Platelet GPIIb/IIIa antagonists: the first anti-integrin receptor therapeutics. J Clin Invest 1997; 99:1467–71. [10] Ginsberg MH, Xiaoping D, O’Toole TE, Loftus JC, Plow EF. Platelet integrins. Thromb Haemost 1993; 70:87–93. [11] Vu TH, Hung DT, Wheaton VI, Coughlin SR. Molecular cloning of a functional thrombin receptor reveals a novel proteolytic mechanism of receptor activation. Cell 1991; 64:1057–68. [12] Coughlin SR. Thrombin receptor structure and function. Thromb Haemost 1993; 70:184–7. [13] Badimon L, Badimon JJ, Turitto VT, Vallabhajosula S, Fuster V. Platelet thrombus formation on collagen type I. Influence of blood rheology, von Willebrand factor and blood coagulation. Circulation 1988; 78:1431–42. [14] Badimon L, Badimon JJ, Turitto VT, Fuster V. Role of von Willebrand factor in mediating platelet-vessel wall interaction at low shear rate: the importance of perfusion conditions. Blood 1989; 73:961–7. [15] Gachet C. ADP receptors of platelets and their inhibition. Thromb Haemost 2001; 86:222–32. [16] Nemerson Y. Mechanism of coagulation. In: Williams WJ, Beutler E, Erslev AJ, Lichtman MA (eds), Hematology. New York: McGraw-Hill; 1990:1295–304.

Thrombosis in clinical practice

16

[17] Mann KG. Factor VII-activating protease: coagulation, fibrinolysis, and atherothrombosis? Circulation 2003; 107:654–5. [18] Mann KG. Thrombin formation. Chest 2003; 124 (3 Suppl.):4S–10S. [19] Mann KG, Nesheim ME, Church WR, Haley P, Krishnaswamy S. Surface dependent reactions of the vitamin K dependent enzyme complexes. Blood 1990; 76:1–16. [20] Nemerson Y, Williams WJ. Biochemistry of plasma coagulation factors. In: Williams WJ, Beutler E, Erslev AJ, Lichtman MA (eds), Hematology. New York: McGraw-Hill; 1990:1267– 84. [21] Francis CW, Marder VJ. Physiologic regulation and pathologic disorders of fibrinolysis. In: Colman RW, Hirsh J, Marder VJ, Salzman EW (eds), Hemostasis and Thrombosis: Basic Principles and Clinical Practice. Philadelphia, PA: Lippincott; 1987: 358–79. [22] Collen D, Lijnen HR. Molecular and cellular basis of fibrinolysis. In: Hoffman R, Benz EJ Jr, Shattil SJ, Furie B, Cohen HJ (eds), Hematology: Basic Principles and Practice. New York: ChurchillLivingstone; 1991:1232–42. [23] Tegeler CH, Downes TR. Thrombosis and the heart. Semin Neurol 1991; 11:339–52. [24] Adams PC, Cohen M, Chesebro JH, Fuster V. Thrombosis and embolism from cardiac chambers and infected valves. J Am Coll Cardiol 1986;8: 76B–87B. [25] Mikell FL, Asinger RW, Elsperger KJ. Regional stasis of blood in the dysfunctional left ventricle: echocardiographic detection and differentiation from early thrombosis. Circulation 1982; 66:755–63. [26] Merino A, Hauptman P, Badimon L, Badimon JJ, Cohen M, Fuster V, Goldman M. Echocardiographic “smoke” is produced by an interaction of erythrocytes and plasma proteins modulated by shear forces. J Am Coll Cardiol 1992;20: 1661–8. [27] Shrestha NK, Moreno FL, Narcisco FV, Torres L, Calleja H. Two-dimensional echocardiographic diagnosis of left atrial thrombus in rheumatic heart disease: a clinicopathologic study. Circulation 1983; 67:341–7. [28] Fuster V, Halperin JL, Left ventricular thrombi and cerebral embolism (editorial). N Engl J Med 1989; 320:392–4. [29] Fuster V, Badimon L, Badimon JJ, Chesebro JH. Prevention of thromboembolism induced by prosthetic heart valves. Semin Thromb Hemost 1988; 14:50–8. [30] Alpert JS. The thrombosed prosthetic valve. Current recommendations based on evidence from the literature. J Am Coll Cardiol 2003; 41:659–60. [31] Lengyel M, Fuster V, Keltai M, et al. Guidelines for management of left-side prosthetic valve thrombosis: a role for thrombolytic therapy. J Am Coll Cardiol 1997; 30:1521–6. [32] Mannucci PM. Venous thrombosis: the history of knowledge. Pathophysiol Haemost Thromb 2002; 32:209–12. [33] The potential role of direct thrombin inhibitors in the prevention and treatment of venous thromboembolism. Chest 2003; 124(3 Suppl.):40S–48S. [34] Tovey C, Wyatt S. Links diagnosis, investigation, and management of deep vein thrombosis. BMJ 2003; 326:1180–4. [35] Levi M, Dorffle-Melly J, Johnson GJ, Drouet L, Badimon L; Subcommittee on Animal, Cellular, and Molecular Models of Thrombosis and Haemostasis of the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis. Usefulness and limitations of animal models of venous thrombosis. Thromb Haemost 2001; 86:1331–3. [36] Kearon C. Natural history of venous thromboembolism. Circulation 2003; 107:I22–I30. [37] Anderson FA, Spencer FA. Risk factors for venous thromboembolism. Circulation 2003; 107:I9–I16. [38] White RH. The epidemiology of venous thromboembolism. Circulation 2003; 107:I4–I8. [39] Badimon L, Badimon JJ, Vilahur G, Segales E, Llorente V. Pathogenesis of the acute coronary syndromes and therapeutic implications. Pathophysiol Haemost Thromb 2002; 32:225–31. [40] Falk E, Shah PK, Fuster V. Coronary plaque disruption. Circulation 1995; 92:657–71.

The pathophysiology of thrombosis

17

[41] Adams RJ, Chimowitz MI, Alpert JS, Awad IA, Cerqueria MD, Fayad P, Taubert KA; Stroke Council and the Council on Clinical Cardiology of the American Heart Association; American Stroke Association. Coronary risk evaluation in patients with transient ischemic attack and ischemic stroke: a scientific statement for healthcare professionals from the Stroke Council and the council on Clinical Cardiology of the American Heart Association/American Stroke Association. Circulation 2003; 108:1278–90. [42] Graham CA. Transient cerebral ischemia demands urgent evaluation. Stroke 2003; 34:2451–2. [43] Hack W, Kaste M, Bogousslavsky J, Brainin M, Chamorro A, Lees K, Leys D, Kwiecinski H, Toni P, Langhorne P, Diener C, Hennerici M, Ferro J, Sivenius J, Gunnar N, Bath P, Olsen TS, Gugging M; European Stroke Initiative Executive Committee and the EUSI Writing Committee European Stroke Initiative Recommendations for Stroke Management—update 2003. Cerebrovasc Dis 2003; 16:311–37. [44] McDermott MM, Mandapat AL, Moates A, Albay M, Chiou E, Celic L, Greenland P. Knowledge and attitudes regarding cardiovascular disease risk and prevention in patients with coronary or peripheral arterial disease. Arch Intern Med 2003; 163:2157–62.

2 Epidemiology of coagulopathy Andrew J Catto Introduction Thrombosis is the commonest cause of mortality in the Western world. Thrombotic disease can be broadly classified as that arising in the venous system (at low blood flow and pressure) and in the arterial system (at high flow and pressure). A thrombus is a deposit arising from the constituents of the blood on the lining of the heart or blood vessels. Thrombi occur anywhere in the circulation, and an occlusive thrombus occupies the lumen of the vessel resulting in cessation of flow and tissue damage distal to the occlusion. Virchow, as early as 1856, proposed that three major factors determine the site and extent of thrombus. Specifically, the mechanical effects in which blood flow is predominant, the constituents of the blood, and, finally, the vessel wall. The interactions of these three factors determine the type of thrombus formed. There are basic differences between arterial and venous thrombosis such as the composition of the thrombi (platelet rich in arterial and fibrin rich in venous), and the presence of vascular wall damage (atheroma) in arterial thrombosis. However, these distinctions are not absolute and they share common underlying mechanisms. Disturbance of hemostasis is central to the pathogenesis of all thrombosis, even though it differs in nature depending on anatomical location. As hemostasis and thrombosis are similar (if not entirely identical) processes, the concept of thrombosis as “hemostasis in the wrong place” [1] provides some basis for considering how platelet function and coagulation systems might contribute to the pathogenesis of thrombosis. Arterial thrombosis There is little doubt that the development of arterial thrombosis is the product of multiple genetic and environmental risk factors in both atherosclerosis and thrombosis. Acute thrombosis occurring in a ruptured, usually lipid rich, atherosclerotic plaque is the event that precipitates the development of acute myocardial infarction (MI), cerebral thrombosis, and acute peripheral arterial occlusion. Plaque rupture is followed by platelet aggregation to exposed subendothelial von Willebrand factor (vWF) through glycoprotein complexes. A cascade of intracellular signaling events with subsequent platelet activation follows platelet adhesion to the subendothelial binding of vWF and fibrinogen to a conformationally active form of glycoprotein IIb/IIIa. Platelet aggregation results in further propagation of the expanding platelet thrombus.

Epidemiology of coagulopathy

19

The traditional view of arterial thrombus held that disorders such as myocardial infarction and ischemic stroke were largely attributed to the modifiable risk factors such as hypertension, cigarette smoking, hyperlipidemia, and obesity. However, it has been known for some time that at least 30–40% of the variance in risk of MI is attributed to factors other than these conventional risk factors [2, 3]. Furthermore, a strong body of evidence exists to support the efficacy of antithrombotic therapies in both the primary and secondary prevention of atherothrombotic events. There is also evidence from younger subjects with MI suggesting that their atherosclerotic burden is less when compared with elderly MI subjects. Consequently, the dominant pathogenic mechanisms for MI in younger patients may be more likely to involve perturbations of the hemostatic balance. What is the evidence for the hemostatic system in arterial thrombosis? Pathological studies indicate that acute thrombosis in the coronary and cerebral vasculature involves activation of both platelets and coagulation systems. Over the past two decades, there have been a considerable number of studies that support a role for the hemostatic system in the development of MI. These include both case-control and prospective studies [4–8]. However, while case-control and crosssectional studies assess the relationship between a given circulating marker and the development of arterial thrombosis, it is not possible to exclude the possibility that any observed elevation (or indeed reduction) of a given circulating factor is not the result of an acute phase reaction associated with the thrombotic event itself [9]. Although this problem is overcome in prospective studies, any association may be confounded by underlying subclinical atherosclerosis, which in turn influences the measured levels of hemostatic factor. In the case of vascular disease, any association is further complicated by interactions between the other hemostatic factors as well as the “traditional” cardiovascular risk factors such as smoking. Arguably, the firmest evidence derives from intervention studies, in which pharmacological therapies lower the level of the circulating factor. In the case of hemostatic factors, there are very few effective agents that reduce levels of hemostatic factors. For example, a number of drugs are known to lower fibrinogen levels [10, 11], but toxicity has tended to limit their effectiveness. Of the numerous individual constituents of the coagulation cascade, a small number have provided most use in our understanding of the epidemiology of thromboembolic disease (Table 2.1). Fibrinogen There is a strongly, generally consistent relationship of fibrinogen levels with the risk of MI and that of ischemic stroke. This has been demonstrated in both case-control and large well-characterized prospective studies [12–14], including the Framingham Study [15]. In the Framingham Study, there was a clear relationship between fibrinogen and incidence of ischemic heart disease (IHD) in both men and women, and stroke in men but not women (possibly an artifact arising from small numbers). They also reported an association between high fibrinogen and hypertension. In the Leigh study of IHD,

Thrombosis in clinical practice

20

incidence in males aged 40–69, there was also suggestive evidence of an interaction between systolic blood pressure and plasma fibrinogen in those in the

Table 2.1 Hemostatic and thrombotic Coagulation factors, platelet factors, and activation of coagulation Fibrinogen Factor VII Factor VIII Factor XIII von Willebrand factor Prothrombin fragment 1+2 Thrombin-antithrombin III Fibrin degradation products D-dimer Factor V Fibrinolytic factors

Tissue plasminogen activator Plasminogen activator inhibitor-1 Clot lysis time Inflammatory factors C-reactive protein Serum amyloid A Interleukins Cellular adhesion molecules Other factors Homocysteine

highest tertile of levels [16]. Furthermore, there is evidence that cholesterol and fibrinogen are also additive in coronary risk, as shown in PROCAM [17] and ECAT [18] where coronary rates were highest in individuals with elevation in both factors than elevations of any one factor alone. A number of mechanisms have been proposed to account for the effect of fibrinogen on vascular risk including an increase in fibrin formation, raised plasma viscosity, promoting platelet aggregation, and vascular and smooth muscle cell proliferation [19]. When studying fibrinogen, consideration should be taken as to the possible confounders as fibrinogen levels increase with age, menopause, hypertension, and obesity as well as positively correlating with total and LDL cholesterol [20], and inversely correlates with

Epidemiology of coagulopathy

21

exercise and hormone replacement therapy [21]. However, the most important environmentat determinant of fibrinogen levels is cigarette smoking [22], probably mediated through interleukin-6. Factor VII Factor VII (FVII) is an inactive vitamin K-dependent zymogen synthesized in the liver and secreted as a single-chain glycoprotein of 48 kD. On contact with tissue factor, it is converted to the two-chain active form by selective proteolysis by factor Xa, and the factor VIIa/ tissue factor complex then activates factor X and factor IX. Levels of FVII are influenced by multiple environmental factors including advancing age, gender, dietary lipids, body mass index, and triglyceride [23–25]. Much of the epidemiologic interest in FVII stems from the findings of the Northwick Park Heart Study (NPHS), a whole population study (as distinct from case-control study) in which cases, and those so far unaffected by clinically manifest IHD, were drawn from the same industrial population. The primary objective of NPHS was to assess hemostatic function in the pathogenesis of IHD [4]. The two variables showing clear prospective associations with subsequent IHD were factor VII coagulant activity (F:VIIC) and fibrinogen. NPHS has been one of the few studies to demonstrate prospective findings on FVII, although case-control studies have also found associations with FVII and IHD or ischemic stroke risk [26–29], including work from our Unit [30]. Some of the observed differences may be attributed to the assay and the NPHS F:VIIc assay is more sensitive to two-chain VIIa. The clotting assay, factor VIIC, reflects the total FVII antigen, but it can also be influenced by the amount of FVII that is activated (factor VIIa) in the plasma. The latter can be measured directly with an assay that utilizes truncated tissue factor as stimulant [31]. It has been suggested that the factor VIIC assay used in the NPHS is particularly sensitive to factor VIIa. There is, however, no good evidence that factor VIIa is an informative marker with respect to the prediction of cardiovascular disease [32]. Subsequent prospective studies have failed to confirm an association between FVII and risk of CHD [33–35]. FVII is not related to risk of venous thrombosis [36]. Factor VIII and von Willebrand factor Congenital deficiency or dysfunction of vWF, which is a multimeric glycoprotein circulating in blood as a noncovalent complex with procoagulant FVIII [37], results in moderate to severe bleeding and congenital FVIII deficiency is associated with a decreased mortality from IHD [38]. However, it has been recognized that vWF is involved in the pathogenesis of IHD. In prospective studies of middle aged and older adults [39], elevated FVIII activity, vWF activity, and vWF antigen have been implicated in arterial thrombosis [34, 35, 40] as well as in large case-control studies of ischemic stroke [41], in which raised vWF was shown to be an independent determinant of poststroke mortality. In a number of prospective studies, with atherosclerotic disease, vWF levels have been independently associated with risk of thrombotic complications [6, 18, 42]. The precise role of FVIII in this process is not clear. It seems likely that elevated FVIII/vWF

Thrombosis in clinical practice

22

circulates as a complex and most studies show these factors to be highly correlated. FVIII levels are influenced by ABO blood group, as subjects with non-O exhibit, a higher FVIII concentration, and increased risk of IHD [43]. Factor XIII Until recently, there was relatively little information relating changes in factor XIII (FXIII) level or activity to human arterial disease. Plasma FXIII is a tetrameric molecule composed of two A-subunits of 83.2 kD and two B-subunits of 79.7 kD that are held together noncovalently in a heterologous tetramer of 325.8 kD [44]. Two animal studies have provided evidence for a role for circulating FXIII in the pathogenesis of both MI and thromboembolic disease [45, 46]. In a canine electrically induced MI model treated with an FXIII inhibitor, fibrinolytic therapy led to increased clot lysis compared to animals not pretreated with the inhibitor. In a ferret model of pulmonary embolus, FXIII cross-linking is associated with increased resistance to exogenous tPA therapy. Both studies conclude that activated FXIII has an important role in fibrinolysis-resistant clot formation and may affect outcomes in these two pathologically distinct syndromes that have fibrin formation as a common feature. Two of the earlier published case-control studies report associations between FXIII Val34Leu and a history of MI. In a study of subjects with angiographically proven coronary artery disease and a history of MI, FXIII Leu34 was significantly less common in those with a history of MI [47]. In those subjects with Leu34 who had a previous MI, there were higher levels of the fibrinolytic inhibitor PAI-1, the PAI-1 4G/4G genotype was more common, and there was evidence of interactions with other clotting factors and with features of the insulin-resistance syndrome [48, 49]. A mixed postmortem and casecontrol study of survivors of MI in Finnish subjects again suggested a protective effect of Leu34, but it was unable to reproduce the PAI-1 4G/4G genotype associations [50]. Interestingly, this study found the prevalence of the protective allele to be lowest in the Eastern Kainuu region, which has the highest prevalence of MI, similar to that reported in the high-risk South Asian community in England [51]. Fibrin is the main protein constituent of the blood clot, which is stabilized by factor XIIIa through an amide or isopeptide bond that ligates adjacent fibrin monomers. The relationship between FXIII and the major constituent of a thrombus, fibrin, has been reviewed elsewhere [52]. Proteins C, S, and antithrombin III There is good evidence that congenital deficiencies of proteins C, S, and antithrombin III (ATIII) are associated with an increased risk of venous thrombosis. The evidence linking these deficiency states with arterial thrombosis is less well defined [53]. In an analysis of 311 largely younger patients with clinically evident ischemic stroke, deficiencies of proteins S, C, and ATIII were present in 11.9%, 2.2%, and 2.2% of cases, respectively [54]. The small numbers of subjects and heterogeneous nature of the clinical phenotypes, as well as control matching makes interpretation of these studies difficult. For example, in a recent study of young subjects with ischemic stroke, deficiencies of the naturally occurring anticoagulants is not uncommon, although congenital deficiencies are

Epidemiology of coagulopathy

23

uncommon [55–57]. There is some evidence to suggest that protein C and S deficiency may be present in 5–15% of cases with ischemic stroke [58–60]. However, in clinical practice, congenital deficiencies of proteins C, S, and ATIII are not clinically important, as the majority of patients with ischemic insults are elderly, and prothrombotic deficiencies are unlikely to be playing a major role in disease pathogenesis. Activated protein C resistance and factor V Leiden Activated protein C (APC) is a naturally occurring anticoagulant. Resistance to the anticoagulant effect of APC (APCR) is most common due to a single nucleotide polymorphism (G1619A) in the gene-encoding factor V—factor V Leiden (FVL). The procoagulant effect results from a reduced rate of factor Va inactivation. The available evidence indicates that FVL is the most common genetic risk factor associated with venous thrombosis. In relation to arterial thrombosis, the evidence that FVL plays any role is more limited. In relation to MI there is evidence for [61, 62] and against [63–65], including our own findings in ischemic stroke [66]. There is some data to suggest that approximately 20% of cases of pediatric ischemic stroke are associated with FVL. A number of studies have linked APCR with arterial thrombosis in the absence of FVL. A study of 826 adults with evidence of carotid and femoral atherosclerosis exhibited decreased response to APC, which seemed independent of FVL, supporting a role for other environmental and genetic factors [67]. Fibrinolysis in arterial thrombosis The fibrinolytic system comprises a circulating pro-enzyme plasminogen, which is converted by plasmin to the plasminogen activators tissue-plasminogen activator (tPA) and urokinase. Plasminogen activator inhibitor-1 (PAI-1) is the principal inhibitor of fibrinolysis and is the main counterbalance to natural fibrinolysis. PAI-1 is synthesized by platelets, endothelium, and vascular smooth muscle cells (SMC). There is data from several prospective studies to suggest that endogenous fibrinolytic activity at baseline was found to predict future cardiovascular morbidity and mortality. This has been shown for healthy men at risk from stroke [68] and MI [69] and also young patients suffering MI. One of the key problems associated with the interpretation of fibrinolysis studies is based in part upon whether the detected alteration in fibrinolytic function is considered to be a primary process or the result of pre (or sub) clinical atherosclerosis and a resultant dysfunctional endothelium. Other complicating factors include measurement difficulties for tPA and PAI-1 as there is significant circadian variation for the fibrinolytic parameters. There is also spontaneous transformation of PAI-1 to an inactive or latent form as well as release of PAI-1 from platelets necessitating careful and specific blood collection techniques [70]. Furthermore, as plasma tPA and PAI-1 are strongly correlated, it is difficult, if not impossible to separate the effects of these factors in an epidemiological study and depending on the variables controlled for in a regression model, the relative importance of tPA and PAI-1 tends to vary [71]. In addition, the tPA:Ag assay measures both free tPA and tPA complexed to PAI-1, which increases with high PAI-1 levels due to delayed clearance [72]. Furthermore, increased PAI-1 levels

Thrombosis in clinical practice

24

correlate with obesity, hyperinsulinemia, and raised triglycerides, which further complicate interpretation of circulating levels. The available evidence suggests that for risk prediction, the more robust marker is tPA:Ag as opposed to PAI-1:Ag or PAI-1:Ac. In the prospective Physicians Health Study (healthy men) [69] and ECAT study, both show statistically large increases in risk of coronary thrombosis for subjects with high baseline tPA:Ag and in ECAT, the relative risk associated with a 1 SD change was larger for tPA:Ag than either PAI-1:Ag or PAI1:Ac. It seems likely that the balance between net activation and net inhibition of fibrinolysis is the most important factor in fibrinolytic potential as shown in measures of global clot lysis time in NPHS [73]. Venous thrombosis Venous thromboembolic disorders are a range of conditions characterized by in situ thrombus formation and the variable presence of embolic manifestations. The clinical presentation depends to a large extent on the site and extent of thrombus formation and on the underlying cause. The spectrum encompasses thrombosis arising in the deep veins of legs (DVT) with embolism to the pulmonary vasculature (pulmonary embolism—PE). However, venous thrombosis is recognized to occur in other sites including the upper limbs, cerebral venous system, and the retinal veins. PE and DVT account for more than 250,000 hospitalizations in the US and constitute the third commonest cause of death after acute ischemic coronary syndromes and stroke. Approximately two-thirds of cases are first time episodes and the remainder arises in patients with recurrent venous thromboembolism (VTE) [74]. In the absence of genetic deficiencies, thrombosis occurs in the older population, largely in the context of environmental factors such as surgery, obesity, and underlying malignant disease. There is increasing evidence of the importance of hemostatic mechanisms in the pathogenesis of venous thrombosis. The function and regulation of coagulation and fibrinolytic systems and factors have been extensively studied. Precisely how they functionally integrate, in vivo, is still unknown. The role of inhibitory mechanisms in preventing venous thrombosis was established after the identification of the link between inherited deficiencies of antithrombin, protein C, protein S, and venous thrombosis. The general importance of the anticoagulant pathway involving protein C and protein S was established by the widespread prevalence among white populations and the clinical consequences of the phenotype of APCR and its main causative gene mutation, factor V G1691A (factor V Arg506Gln, factor V Leiden). There has been slower appreciation of the procoagulant effect of increased levels of coagulation factors in venous disease, but recent work has highlighted the likely importance of prothrombin and FVIII levels in this regard. (See Table 2.2 for the hemostatic and thrombotic markers for venous thrombosis.) Proteins C, S, and ATIII in venous thrombosis Despite their importance, proteins C, S, and ATIII deficiency together account for less than

Epidemiology of coagulopathy

25

Table 2.2 Hemostatic and thrombotic markers for venous thrombosis. Factor V Prothrombin Endothelial cell protein C/activated protein C receptor Protein C Protein S Antithrombin III Thrombomodulin Factor VIII and vWF Homocystinemia Factor VII Fibrinolytic factors?—uncertain

10% of all cases of recurrent venous thrombosis. In unselected patients with (VTE), ATIII deficiency occurs in 1.1% of subjects compared with 2.4% of selected patients [75]. Patients with ATIII deficiency appear to be at greater risk for VTE than those with protein C or S deficiency, and in one report 85% of patients with ATIII deficiency suffered a thromboembolic event by the age of 50 [76]. The prevalence of heterozygous protein C and S deficiency among the general population is low and around 5–10% of selected patients with VTE and around 50% of heterozygotes up to 50 years suffer a thrombosis [77]. APC resistance in venous thrombosis Between 20% and 60% of patients with recurrent VTE display APCR [78] and for the majority of subjects this is the result of FVL. Unlike arterial thrombosis, FVL and APCR are important risk factors for VTE. Approximately 3–5% of the general population are FVL heterozygotes (FVL homozygotes are rare) [79]. While APCR is not as strong a risk factor for VTE as the thrombophilias referred to earlier, it is associated with a 3- to 7-fold increase in risk of VTE [80], and FVL also enhances the prothrombotic tendency in the presence of other VTE risk factors, for example, approximately 60% of women experiencing a VTE during oral contraceptive use are APCR [81]. There is also a thrombotic interaction between FVL and the other heritable thrombophilias [82]. D-dimer and venous thrombosis D-dimer is a cross-linked fibrin degradation product that circulates in plasma after lysis of fibrin by plasmin. In DVT and PE patients, endogenous fibrinolysis occurs resulting in increased circulating cross-linked fibrin degradation products. D-dimer has been used as a diagnostic test for VTE, although the results need to be carefully interpreted. The utility

Thrombosis in clinical practice

26

of the D-dimers in the diagnosis of venous thrombosis has been extensively reviewed elsewhere [83, 84]. Other coagulation factors and venous thrombosis Elevated FVIII, FIX, and FXI are all linked with an increased thrombotic risk. In the Leiden Thrombophilia Study (LETS), 25% of subjects with first ever VTE had FVIII levels greater than 150% of normal compared with only 11% of healthy controls [36] and plasma levels of greater than 150 IU/dl were associated with a five times risk for an initial event as also shown in other studies [85]. Subjects who had previously had a VTE and showed higher FVIII levels were at a particularly elevated risk of VTE recurrence [86]. Interestingly, studies have also shown that plasma levels of FIX and FXI also increase risk of VTE. In LETS, subjects with FIX and FXI levels above the 90th percentile had a 2.5- and 2.2-fold risk, respectively, of VTE, compared with those beneath the 90th percentile [87, 88]. In 100 healthy individuals and 129 patients with VTE, FVIII levels correlated only with vWF but not with risk of thrombosis, although levels of tPA and PAI-1 were increased, which is interesting as these are usually associated with arterial thrombosis [89], and the lack of association of venous thrombosis with fibrinolysis was confirmed elsewhere [90]. By contrast, in the US Longitudinal Investigation of Thromboembolism Etiology (LITE) study of 159 patients with VTE, FVIII and vWF were linearly associated with increased risk of VTE, but there was no association of VTE with fibrinogen or C-reactive protein levels, or white cell count [91]. In another study, fibrinogen levels were more strongly related with risk of venous thrombosis and in one study the risk of venous thrombosis with high fibrinogen levels was increased 4-fold [92]. FXIII that has been implicated in arterial thrombosis has also been associated with reduced levels of FXIII occurring in patients with venous thrombosis [93]. Homocysteine and venous thrombosis The evidence directly linking raised homocysteine (Hcy) with risk of VTE is unknown, although there is some evidence that mild to moderate elevations in Hcy are independent risk factors for VTE. In LETS, 10% had Hcy levels above the 95th percentile with an odds ratio (OR) for VTE of 2.5 [94]. In a meta-analysis of ten case-control studies, similar findings were observed [95]. It is probable that raised Hcy is a risk factor for recurrent VTE as well. Other hemostatic factors There is a panoply of other hemostatic factors that are beyond the scope of this chapter. These include lipoprotein A [96], thrombomodulin [97] as well the platelet receptor glycoproteins IIb/IIIa [98, 99], platelet Ia/Iia [100] and Ib/IX [101]. Other inflammatory markers associated with thrombosis include C-reactive protein, serum amyloid A, cellular adhesion molecules, and the interleukins [102].

Epidemiology of coagulopathy

27

Conclusions Arterial disease occurs in a high-pressure, high-flow system with atheromatous disease as

Table 2.3 Summary of the outcome of investigations of hemostatic polymorphisms. Polymorphism

Phenotype

Association of genotype with disease

Venous thromboembolic disease Factor V G1 691A, Arg506Gln

APCR

Clear risk factor alone or in combination with other factors, genetic or acquired

Prothrombin G20210A

Altered levels, mechanism uncertain Moderate risk factor alone or in combination with other risk factors

Factor V HR2 haplotype

Slight APCR; glycosylation isoforms altered?

Weak risk factor alone or in combination with factor V G1691A?

Factor XIII Val34Leu

Increased activation rate

Preliminary results suggest 34Leu is protective against thrombosis

EPCR 23-bp repeat

Predicted altered expression on cell surface

Preliminary results suggest it a weak/moderate risk factor

Arterial occlusive disease Fibrinogen β chain— Associated with altered level 455G/A

Inconsistent

Fibrinogen β chain Bcl I,

Associated with altered level

Inconsistent

Fibrinogen a Thr313 Ala

Altered clot struture?

Inconsistent

Factor VII Arg353Gln

Altered level

Inconsistent

Factor VII HVR4

Probably altered level

Unknown

Factor VII—402G/A

Altered level

Unknown

Factor VII—401G/T

Altered level

Unknown

Factor VII—323 0/10 Altered level

Unknown

Factor XII Val34Leu Increased activation rate

Preliminary results suggest 34Leu is protective against MI

Factor V G1691A, Arg506Gln

APCR

Possibly in selected patients and in combination with acquired risk factors

Prothrombin

Altered level

Possibly in selected patients, and in

Thrombosis in clinical practice

G20210A EPCR 23-bp repeat

28

combination with acquired risk factors Predicted altered expression on cell surface

Preliminary results suggest it may be a weak risk factor

tPA Alu insertion/del Unlikely

Inconsistent

PAI-1–675 4G/5G

Inconsistent

Inconsistent

PAI-1 (CA)n

Suggestive of altered level

Unknown

PAI-1 Hind III

Suggestive of altered level

Unknown

Gp IIIa Pro33Leu

No effect on fibrinogen binding, but Inconsistent increased sensitivity to activation?

Gp IαVNTR

Unknown

Preliminary results show association

Gp Ib; Iα C3550T, Thr145Met

Unknown

Inconsistent

Gp Ia/IIa, α2 C807T

Associated with altered surface expression of receptor, altered collagen binding

Inconsistent

Figure 2.1 This figure shows the interaction between genes, environment, and the intermediate phenotype. In the illustration, the intermediate phenotype is circulating fibrinogen and the final phenotype is thrombotic stroke. a dominant feature. Hemostatic factors have been extensively studied as possible risk determinants for disease. The most consistent associations have been found for fibrinogen. Although it is formally possible that fibrinogen is merely a marker of another underlying process, such as the acute-phase reaction [103], there is reasonably strong evidence that fibrinogen is causally involved in thrombotic disease. For each of the other

Epidemiology of coagulopathy

29

hemostatic factors, though a causal role is plausible, the evidence for this is less strong. It is interesting to note that hemostatic risk factors for atherothrombotic disorders are largely those that have been associated to some extent with classical risk markers such as features of insulin resistance [104], some of which is likely to be heritable, as shown in recent twin studies [105], These observations indicate the importance of environmental influences and emphasize the complexity of the processes involved in vascular disease. When individual hemostatic factors do contribute to risk, it is highly likely to be a complex interaction with other metabolic factors. This understanding further increases the importance of lifelong risk interactions and may suggest an explanation for some of the inconsistencies in case-control studies of arterial disease. Some of the inconsistencies in hemostatic studies are likely to arise from deficiencies of study size and power, plurality of clinical endpoints (MI, unstable angina, coronary artery disease, progression of arterial disease, stroke), and these have been often selected in a postdata collection search for significance. Other difficulties include recruitment bias in the selection of patients, controls, or both. Future challenges are likely to arise from the knowledge base arising from the human genome project. The study of interindividual genetic variation and the potential for study of functional variants of candidate genes is likely to alter our understanding of the importance of these thrombotic risk factors. The genetic basis of hemostatic risk factors has been thoroughly reviewed elsewhere [106]. A summary of the currently known polymorphisms and the association with vascular disease is shown in Table 2.3. Although the goal must be risk prediction, it is most likely that genomics will provide insights into thrombotic mechanisms. It will be necessary for clinicians to integrate hemostatic factors into our evolving knowledge of the genetics, biochemistry, and epidemiology of vascular disease. References [1] Macfarlane RG. Blood coagulation and thrombosis: introduction. Br Med Bull 1955; 11:1–4. [2] Sacco RL, Ellenberg JH, Mohr JP, Tatemichi TK, Hier DB, Price TR. Infarcts of undetermined cause: the NINCDS stroke data bank. Ann Neurol 1989; 25:382–90. [3] Grover SA, Coupal L, Hu XP. Identifying adults at increased risk of coronary disease. How well do the current cholesterol guidelines work? JAMA 1995; 274:801–6. [4] Meade TW, Mellows S, Brozovic M, Miller GJ, Chakrabarti RR, North WR, et al. Haemostatic function and ischaemic heart disease: principal results of the Northwick Park Heart Study. Lancet 1986; 2:533–7. [5] Wilhelmsen L, Svärdsudd K, Korsan-Bengtsen K, Larsson B, Welin L, Tibblin G. Fibrinogen as a risk factor for stroke and myocardial infarction. N Engl J Med 1984; 311:501–5. [6] Jansson JH, Nilsson TK, Johnson O. von Willebrand factor in plasma: a novel risk factor for recurrent myocardial infarction and death. Br Heart J 1991; 66:351–5. [7] Hamsten A, Walldius G, De Faire U, Dahlen G, Szamosi A, Landou C, et al. Plasminogen activator inhibitor in plasma: risk factor for recurrent myocardial infarction. Lancet 1987; 2:3–9. [8] Ridker PM, Hennekens CH, Stampfer MJ, Manson JE, Vaughan DE. Prospective study of endogenous tissue plasminogen activator and risk of stroke. Lancet 1994; 343:940–3. [9] Carter AM, Catto AJ, Bamford JM, Grant PJ. Gender-specific associations of the fibrinogen B β 448 polymorphism, fibrinogen levels, and acute cerebrovascular disease. Arterioscler Thromb Vasc Biol 1997; 17:589–94.

Thrombosis in clinical practice

30

[10] Boda Z. Fibrinogen as an independent cardiovascular risk factor and the question of treatment with fibrinogen-lowering drugs. Orv Hetil 1997; 138:2455–60. [11] Goldbourt U, Behar S, Reicher-Reiss H, Agmon J, Kaplinsky E, Graff E, et al. Rationale and design of a secondary prevention trial of increasing serum high-density lipoprotein cholesterol and reducing triglycerides in patients with clinically manifest atherosclerotic heart disease (the Bezafibrate Infarction Prevention Trial). Am J Cardiol 1993; 71:909–15. [12] Ernst E. The role of fibrinogen as a cardiovascular risk factor. Atherosclerosis 1993; 100:1–12. [13] Danesh J, Collins R, Appleby P, Peto R. Association of fibrinogen, C-reactive protein, albumin, or leukocyte count with coronary heart disease: meta-analyses of prospective studies. JAMA 1998; 279:1477–82. [14] Maresca G, Di Blasio A, Marchioli R, Di Minno G. Measuring plasma fibrinogen to predict stroke and myocardial infarction: an update. Arterioscler Thromb Vasc Biol 1999; 19:1368–77. [15] Kannel WB, Wolf PA, Castelli WP, D’Agostino RP. Fibrinogen and risk of cardiovascular disease. JAMA 1987; 258:1183–6. [16] Stone MC, Thorp JM. Plasma fibrinogen—a major coronary risk factor. J R Coll Gen Pract 1985; 35:565–9. [17] Heinrich J, Balleisen L, Schulte H, Assmann G, van de Loo J. Fibrinogen and factor VII in the prediction of coronary risk. Results from the PROCAM study in healthy men. Arter Thromb 1994; 14:54–9. [18] Thompson SG, Kienast J, Pyke SD, Haverkate F, van de Loo JC. Hemostatic factors and the risk of myocardial infarction or sudden death in patients with angina pectoris. European Concerted Action on Thrombosis and Disabilities Angina Pectoris Study Group. N Engl J Med 1995; 332:635–41. [19] Ernst E, Resch KL. Fibrinogen as a cardiovascular risk factor: a meta-analysis and review of the literature. Ann Intern Med 1993; 118:956–63. [20] Heinrich J, Assmann G. Fibrinogen and cardiovascular risk. J Cardiovasc Risk 1995; 2:197– 205. [21] Effects of estrogen or estrogen/progestin regimens on heart disease risk factors in postmenopausal women. The Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial. The Writing Group for the PEPI Trial. JAMA 1995; 273:199–208. [22] Kannel WB, D’Agostino RB, Belanger AJ. Fibrinogen, cigarette smoking, and risk of cardiovascular disease: insights from the Framingham Study. Am Heart J 1987; 113:1006–10. [23] Mansfield MW, Heywood DM, Grant PJ. Circulating levels of factor VII, fibrinogen, and von Willebrand factor and features of insulin resistance in first-degree relatives of patients with NIDDM. Circulation 1996; 94:2171–6. [24] Heywood DM, Mansfield MW, Grant PJ. Factor VII gene polymorphisms, factor VII:C levels and features of insulin resistance in non-insulindependent diabetes mellitus. Thromb Haemost 1996; 75:401–6. [25] Mennen LI, Schouten EG, Grobbee DE, Kluft C. Coagulation factor VII, dietary fat and blood lipids: a review. Thromb Haemost 1996; 76:492–9. [26] Poller L. Thrombosis and factor VII activity. J Clin Pathol 1957; 10:348–50. [27] Carvalho de Sousa J, Azevedo J, Soria C, Barros F, Ribeiro C, Parreira F, et al. Factor VII hyperactivity in acute myocardial thrombosis. A relation to the coagulation activation. Thromb Res 1988; 51: 165–73. [28] Hoffman CJ, Miller RH, Lawson WE, Hultin MB. Elevation of factor VII activity and mass in young adults at risk of ischemic heart disease. J Am Coll Cardiol 1989;14:941–6. [29] Suzuki T, Yamauchi K, Matsushita T, Furumichi T, Furui H, Tsuzuki J, et al. Elevation of factor VII activity and mass in coronary artery disease of varying severity. Clin Cardiol 1991; 14:731–6. [30] Heywood DM, Carter AM, Catto AJ, Bamford JM, Meade TW, Grant PJ. Factor VII:C levels and FVII gene polymorphisms in acute stroke. Blood Coagul Fibrinolysis 1996; 7:378.

Epidemiology of coagulopathy

31

[31] Morrissey JH, Macik BG, Neuenschwander PF, Comp PC. Quantitation of activated factor VII levels in plasma using a tissue factor mutant selectively deficient in promoting factor VII activation. Blood 1993; 81:734–44. [32] Moor E, Silveira A, van’t Hooft F, Suontaka AM, Eriksson P, Blomback M, et al. Coagulation factor VII mass and activity in young men with myocardial infarction at a young age. Role of plasma lipoproteins and factor VII genotype. Arterioscler Thromb Vasc Biol 1995; 15:655–64. [33] Junker R, Heinrich J, Schulte H, van de LJ, Assmann G. Coagulation factor VII and the risk of coronary heart disease in healthy men. Arterioscler Thromb Vasc Biol 1997; 17:1539–44. [34] Smith FB, Lee AJ, Fowkes FG, Price JF, Rumley A, Lowe GD. Hemostatic factors as predictors of ischemic heart disease and stroke in the Edinburgh Artery Study. Arterioscler Thromb Vasc Biol 1997; 17:3321–5. [35] Folsom AR, Wu KK, Rosamond WD, Sharrett AR, Chambless LE. Prospective study of hemostatic factors and incidence of coronary heart disease: the Atherosclerosis Risk in Communities (ARIC) Study. Circulation 1997; 96:1102–8. [36] Koster T, Blann AD, Briët E, Vandenbroucke JP, Rosendaal FR. Role of clotting factor VIII in effect of von Willebrand factor on occurrence of deepvein thrombosis. Lancet 1995; 345:152–5. [37] Zimmerman TS, Ruggeri ZM. von Willebrand disease. Hum Pathol 1987; 18:140–52. [38] Triemstra M, Rosendaal FR, Smit C, Van der Ploeg HM, Briet E. Mortality in patients with hemophilia. Changes in a Dutch population from 1986 to 1992 and 1973 to 1986. Ann Intern Med 1995; 123:823–7. [39] Tracy RP, Arnold AM, Ettinger W, Fried L, Meilahn E, Savage P. The relationship of fibrinogen and factors VII and VIII to incident cardiovascular disease and death in the elderly: results from the cardiovascular health study. Arterioscler Thromb Vasc Biol 1999; 19:1776–83. [40] Rumley A, Lowe GD, Sweetnam PM, Yarnell JW, Ford RP. Factor VIII, von Willebrand factor and the risk of major ischaemic heart disease in the Caerphilly Heart Study. Br J Haematol 1999; 105:110–16. [41] Catto AJ, Carter AM, Barrett JH, Boothby M, Bamford J, Grant PJ. Elevated von Willebrand factor (vWf): an independent risk factor for cerebrovascular mortality. Blood Coagul Fibrinolysis 1995; 6:590. [42] Blann AD, McCollum CN. von Willebrand factor and soluble thrombomodulin as predictors of adverse events among subjects with peripheral or coronary atherosclerosis. Blood Coagul Fibrinolysis 1999; 10:375–80. [43] Medalie JH, Levene C, Papier C, Goldbourt U, Dreyfuss F, Oron D, et al. Blood groups, myocardial infarction and angina pectoris among 10,000 adult males. N Engl J Med 1971; 285:1348–53. [44] Schwartz ML, Pizzo SV, Hill RL, McKee PA. Human Factor XIII from plasma and platelets. Molecular weights, subunit structures, proteolytic activation, and cross-linking of fibrinogen and fibrin. J Biol Chem 1973; 248:1395–407. [45] Reed GL, Houng AK. The contribution of activated factor XIII to fibrinolytic resistance in experimental pulmonary embolism. Circulation 1999; 99:299–304. [46] Shebuski RJ, Sitko GR, Claremon DA, Baldwin JJ, Remy DC, Stern AM. Inhibition of factor XIIIa in a canine model of coronary thrombosis: effect on reperfusion and acute reocclusion after recombinant tissue-type plasminogen activator. Blood 1990; 75:1455–9. [47] Kohler HP, Stickland MH, Ossei-Gerning N, Carter A, Mikkola H, Grant PJ. Association of a common polymorphism in the factor XIII gene with myocardial infarction. Thromb Haemost 1998; 79:8–13. [48] Kohler HP, Grant PJ. Clustering of haemostatic risk factors with FXIIIVal34Leu in patients with myocardial infarction. Thromb Haemost 1998;80:862. [49] Kohler HP, Mansfield MW, Clark PS, Grant PJ. Interaction between insulin resistance and factor XIII Val34Leu in patients with coronary artery disease. Thromb Haemost 1999; 82:1202– 3.

Thrombosis in clinical practice

32

[50] Wartiovaara U, Perola M, Mikkola H, Totterman K, Savolainen V, Penttila A, et al. Association of FXIII Val34Leu with decreased risk of myocardial infarction in Finnish males. Atherosclerosis 1999; 142:295–300. [51] McCormack LJ, Kain K, Catto AJ, Kohler HP, Stickland MH, Grant PJ. Prevalence of FXIII V34L in populations with different cardiovascular risk. Thromb Haemost 1998; 80:523–4. [52] Ariens RA, Lai TS, Weisel JW, Greenberg CS, Grant PJ. Role of factor XIII in fibrin clot formation and effects of genetic polymorphisms. Blood 2002; 100:743–54. [53] De SV, Leone G, Mastrangelo S, Tripodi A, Rodeghiero F, Castaman G, et al. Clinical manifestations and management of inherited thrombophilia: retrospective analysis and followup after diagnosis of 238 patients with congenital deficiency of antithrombin III, protein C, protein S. Thromb Haemost 1994; 72:352–8. [54] Barinagarrementeria F, Cantu-Brito C, De La PA, Izaguirre R. Prothrombotic states in young people with idiopathic stroke. A prospective study. Stroke 1994; 25:287–90. [55] Douay X, Lucas C, Caron C, Goudemand J, Leys D. Antithrombin, protein C and protein S levels in 127 consecutive young adults with ischemic stroke. Acta Neurol Scand 1998; 98:124– 7. [56] Munts AG, van Genderen PJ, Dippel DW, van Kooten F, Koudstaal PJ. Coagulation disorders in young adults with acute cerebral ischaemia. J Neurol 1998; 245:21–5. [57] Margaglione M, D’Andrea G, Giuliani N, Brancaccio V, De Lucia D, Grandone E, et al. Inherited prothrombotic conditions and premature ischemic stroke: sex difference in the association with factor V Leiden. Arterioscler Thromb Vasc Biol 1999; 19:1751–6. [58] Bonduel M, Sciuccati G, Hepner M, Torres AF, Pieroni G, Frontroth JP. Prethrombotic disorders in children with arterial ischemic stroke and sinovenous thrombosis. Arch Neurol 1999; 56:967–71. [59] Nowak-Gottl U, Strater R, Heinecke A, Junker R, Koch HG, Schuierer G, et al. Lipoprotein (a) and genetic polymorphisms of clotting factor V, prothrombin, and methylenetetrahydrofolate reductase are risk factors of spontaneous ischemic stroke in childhood. Blood 1999; 94:3678– 82. [60] Kenet G, Sadetzki S, Murad H, Martinowitz U, Rosenberg N, Gitel S, et al. Factor V Leiden and antiphospholipid antibodies are significant risk factors for ischemic stroke in children. Stroke 2000; 31:1283–8. [61] Rosendaal FR, Siscovick DS, Schwartz SM, Beverly RK, Psaty BM, Longstreth WT Jr, et al. Factor V Leiden (resistance to activated protein C) increases the risk of myocardial infarction in young women. Blood 1997; 89:2817–21. [62] Doggen CJ, Cats VM, Bertina RM, Rosendaal FR. Interaction of coagulation defects and cardiovascular risk factors: increased risk of myocardial infarction associated with factor V Leiden or prothrombin 20210A. Circulation 1998; 97:1037–41. [63] Emmerich J, Poirier O, Evans A, et al. Myocardial infarction, Arg 506 to Gln factor v mutation, and activated protein C resistance. Lancet 1995; 345:321. [64] Kontula K, Ylikorkala A, Miettinen H, Vuorio A, Kauppinen-Mäkelin R, Hämäläinen L, et al. Arg506Gln factor V mutation (Factor V Leiden) in patients with ischaemic cerebrovascular disease and survivors of myocardial infarction. Thromb Haemost 1995; 73:558–60. [65] Cushman M, Rosendaal FR, Psaty BM, Cook EF, Valliere J, Kuller LH, et al. Factor V Leiden is not a risk factor for arterial vascular disease in the elderly: results from the Cardiovascular Health Study. Thromb Haemost 1998; 79:912–15. [66] Catto AJ, Carter AM, Ireland H, Bayston TA, Philippou H, Barrett J, et al. Factor V Leiden gene mutation and thrombin generation in relation to the development of acute stroke. Arterioscler Thromb Vasc Biol 1995; 15:783–5. [67] Kiechl S, Muigg A, Santer P, Mitterer M, Egger G, Oberhollenzer M, et al. Poor response to activated protein C as a prominent risk predictor of advanced atherosclerosis and arterial disease. Circulation 1999;99:614–19.

Epidemiology of coagulopathy

33

[68] Ridker PM, Hennekens CH, Stampfer MJ, Manson JE, Vaughan DE. Prospective study of endogenous tissue plasminogen activator and risk of stroke. Lancet 1994; 343:940–3. [69] Ridker PM, Vaughan DE, Stampfer MJ, Manson JE, Hennekens CH. Endogenous tissue-type plasminogen activator and risk of myocardial infarction. Lancet 1993; 341:1165–8. [70] Chandler WL, Stratton JR. Laboratory evaluation of fibrinolysis in patients with a history of myocardial infarction. Am J Clin Pathol 1994; 102:248–52. [71] Juhan-Vague I, Pyke SD, Alessi MC, Jespersen J, Haverkate F, Thompson SG. Fibrinolytic factors and the risk of myocardial infarction or sudden death in patients with angina pectoris. ECAT Study Group. European Concerted Action on Thrombosis and Disabilities. Circulation 1996; 94:2057–63. [72] Chandler WL, Alessi MC, Aillaud MF, Henderson P, Vague P, Juhan-Vague I. Clearance of tissue plasminogen activator (TPA) and TPA/plasminogen activator inhibitor type 1 (PAI-1) complex: relationship to elevated TPA antigen in patients with high PAI-1 activity levels. Circulation 1997; 96:761–8. [73] Meade TW, Ruddock V, Stirling Y, Chakrabarti R, Miller GJ. Fibrinolytic activity, clotting factors, and long-term incidence of ischaemic heart disease in the Northwick Park Heart Study. Lancet 1993; 342:1076–9. [74] Anderson FA Jr, Wheeler HB, Goldberg RJ, Hosmer DW, Patwardhan NA, Jovanovic B, et al. A population-based perspective of the hospital incidence and case-fatality rates of deep vein thrombosis and pulmonary embolism. The Worcester DVT Study. Arch Intern Med 1991; 151:933–8. [75] Heijboer H, Brandjes DP, Buller HR, Sturk A, ten Cate JW. Deficiencies of coagulationinhibiting and fibrinolytic proteins in outpatients with deep-vein thrombosis. N Engl J Med 1990; 323:1512–16. [76] Thaler E, Lechner K. Antithrombin III deficiency and thromboembolism. Clin Haematol 1981;10: 369–90. [77] Allaart CF, Poort SR, Rosendaal FR, Reitsma PH, Bertina RM, Briet E. Increased risk of venous thrombosis in carriers of hereditary protein C deficiency defect. Lancet 1993; 341:134– 8. [78] Dahlback B, Carlsson M, Svensson PJ. Familial thrombophilia due to a previously unrecognized mechanism characterized by poor anticoagulant response to activated protein C: prediction of a cofactor to activated protein C. Proc Natl Acad Sci USA 1993; 90:1004–8. [79] Koster T, Rosendaal FR, de Ronde H, Briët E, Vandenbroucke JP, Bertina RM. Venous thrombosis due to poor anticoagulant response to activated protein C: Leiden Thrombophilia Study. Lancet 1993; 342:1503–6. [80] Svensson PJ, Dahlback B. Resistance to activated protein C as a basis for venous thrombosis. N Engl J Med 1994; 330:517–22. [81] Vandenbroucke JP, Koster T, Briët E, Reitsma PH, Bertina RM, Rosendaal FR. Increased risk of venous thrombosis in oral-contraceptive users who are carriers of factor V Leiden mutation. Lancet 1994; 344:1453–7. [82] De SV, Martinelli I, Mannucci PM, Paciaroni K, Chiusolo P, Casorelli I, et al. The risk of recurrent deep venous thrombosis among heterozygous carriers of both factor V Leiden and the G20210A prothrombin mutation. N Engl J Med 1999; 341:801–6. [83] Hirsh J, Lee AY. How we diagnose and treat deep vein thrombosis. Blood 2002; 99:3102–10. [84] Lee AY, Hirsh J. Diagnosis and treatment of venous thromboembolism. Annu Rev Med 2002; 53:15–33. [85] Kraaijenhagen RA, in’t Anker PS, Koopman MM, Reitsma PH, Prins MH, van den EA, et al. High plasma concentration of factor VIIIc is a major risk factor for venous thromboembolism. Thromb Haemost 2000; 83:5–9. [86] Kyrle PA, Minar E, Hirschl M, Bialonczyk C, Stain M, Schneider B, et al. High plasma levels of factor VIII and the risk of recurrent venous thromboembolism. N Engl J Med 2000; 343:457– 62.

Thrombosis in clinical practice

34

[87] Van HV, Van der LI, Bertina RM, Rosendaal FR. High levels of factor IX increase the risk of venous thrombosis. Blood 2000; 95:3678–82. [88] Meijers JC, Tekelenburg WL, Bouma BN, Bertina RM, Rosendaal FR. High levels of coagulation factor XI as a risk factor for venous thrombosis. N Engl J Med 2000; 342:696–701. [89] Bombeli T, Jutzi M, De Conno E, Seifert B, Fehr J. In patients with deep-vein thrombosis elevated levels of factor VIII correlate only with von Willebrand factor but not other endothelial cellderived coagulation and fibrinolysis proteins. Blood Coagul Fibrinolysis 2002; 13:577–81. [90] Folsom AR, Cushman M, Heckbert SR, Rosamond WD, Aleksic N. Prospective study of fibrinolytic markers and venous thromboembolism. J Clin Epidemiol 2003; 56:598–603. [91] Tsai AW, Cushman M, Rosamond WD, Heckbert SR, Tracy RP, Aleksic N, et al. Coagulation factors, inflammation markers, and venous thromboembolism: the longitudinal investigation of thromboembolism etiology (LITE). Am J Med 2002; 113:636–42. [92] Kamphuisen PW, Eikenboom JC, Vos HL, Pablo R, Sturk A, Bertina RM, et al. Increased levels of factor VIII and fibrinogen in patients with venous thrombosis are not caused by acute phase reactions. Thromb Haemost 1999; 81:680–3. [93] Swiatkiewicz A, Jurkowski P, Kotschy M, Ciecierski M, Jawien A. Level of antithrombin III, protein C, protein S and other selected parameters of coagulation and fibrinolysis in the blood of the patients with recurrent deep venous thrombosis. Med Sci Monit 2002; 8:CR263–CR268. [94] den Heijer M, Koster T, Blom HJ, Bos GM, Briet E, Reitsma PH, et al. Hyperhomocysteinemia as a risk factor for deep-vein thrombosis. N Engl J Med 1996; 334:759– 62. [95] den Heijer M, Rosendaal FR, Blom HJ, Gerrits WB, Bos GM. Hyperhomocysteinemia and venous thrombosis: a meta-analysis. Thromb Haemost 1998; 80:874–7. [96] Stein JH, Rosenson RS. Lipoprotein Lp(a) excess and coronary heart disease. Arch Intern Med 1997; 157:1170–6. [97] Salomaa V, Matei C, Aleksic N, Sansores-Garcia L, Folsom AR, Juneja H, et al. Soluble thrombomodulin as a predictor of incident coronary heart disease and symptomless carotid artery atherosclerosis in the Atherosclerosis Risk in Communities (ARIC) Study: a case-cohort study. Lancet 1999; 353:1729–34. [98] Weiss EJ, Bray PF, Tayback M, Schulman SP, Kickler TS, Becker LC, et al. A polymorphism of a platelet glycoprotein receptor as an inherited risk factor for coronary thrombosis. N Engl J Med 1996; 334:1090–4. [99] Carter AM, Ossei-Gerning N, Grant PJ. Platelet glycoprotein IIIa PLA polymorphism in young men with myocardial infarction. Lancet 1996; 348:485–6. [100] Moshfegh K, Wuillemin WA, Redondo M, Lammle B, Beer JH, Liechti-Gallati S, et al. Association of two silent polymorphisms of platelet glycoprotein Ia/IIa receptor with risk of myocardial infarction: a case-control study. Lancet 1999; 353:351–4. [101] Gonzalez-Conejero R, Lozano ML, Rivera J, Corral J, Iniesta JA, Moraleda JM, et al. Polymorphisms of platelet membrane glycoprotein Ib associated with arterial thrombotic disease. Blood 1998; 92:2771–6. [102] Ridker PM. Fibrinolytic and inflammatory markers for arterial occlusion: the evolving epidemiology of thrombosis and hemostasis. Thromb Haemost 1997; 78:53–9. [103] Tracy RP. Epidemiological evidence for inflammation in cardiovascular disease. Thromb Haemost 1999; 82:826–31. [104] Mansfield MW, Grant PJ. Fibrinolysis and diabetic retinopathy in NIDDM. Diabetes Care 1995; 18:1577–81. [105] de Lange M, Snieder H, Ariens RA, Andrew T, Grant PJ, Spector TD. The relation between insulin resistance and hemostasis: pleiotropic genes and common environment. Twin Res 2003; 6:152–61. [106] Lane DA, Grant PJ. Role of hemostatic gene polymorphisms in venous and arterial thrombotic disease. Blood 2000; 95:1517–32.

Epidemiology of coagulopathy

35

3 Commonly used anticoagulant and antiplatelet drugs Aspirin, heparin, and warfarin Andrew D Blann Introduction As discussed in Chapter 1, because atherothrombosis is the precipitating event in myocardial infarction (MI), stroke, and critical limb ischemia [1], agents to reduce the risk of these events are desirable. The active components of a thrombus are platelets and fibrin, and, while often present, red and white blood cells may be described as minor participants, although erythrocytes can enhance platelet aggregation by releasing adenosine diphosphate (ADP). It follows that effective antithrombotic therapy should be directed towards both platelets and the coagulation pathway that concludes with fibrin formation. Aspirin, heparin, and warfarin have been, and remain, the most commonly used drugs for reducing both the short-term and long-term risk of thrombosis, be it arterial or venous. However, despite commonality of final clinical effect, these drugs have radically different modes of action. Aspirin (as acetylsalicylic acid, or a derivative) is the primary antiplatelet agent in clinical use. Its value in the secondary prevention of atherothrombotic disease and in high-risk patients (such as those with atrial fibrillation or undergoing orthopedic surgery) is undoubted [2]. Heparin and warfarin are anticoagulants and both are (at least) very effective in reducing the risk of venous thrombosis. The former is an essential cofactor for antithrombin, but also has activity in reducing the function of other coagulation proteins [3, 4], while the latter reduces the risk of thrombosis by inhibiting the synthesis of key coagulation proteins [5]. However, while these three agents are the mainstay of current clinical practice, newer drugs currently in (limited) practice, in trial, and in development may supersede the “old guard.” In Chapter 15, Jilma discusses agents operating with different mechanisms, such as the platelet ADP-receptor blocker Clopidogrel [6], direct antithrombins such as Ximelagratan [7] and Hirudin [8], and the synthetic pentasaccharide Fondaparinux [9]. In acute coronary syndromes, thrombosis can also be inhibited by drugs that interfere with the platelet receptor for fibrinogen, the gpIIb/IIIa receptor complex [10]. These issues will be revisited in other chapters.

Commonly used anticoagulant and antiplatelet drugs

37

Aspirin Historical perspective Elwood et al. [11] indicate that aspirin appears to have been the first “modern” drug to be used therapeutically, at around 400 BC (e.g. Egyptian physicians making poultices from dried myrtle leaves) although the beneficial action of willow bark has been recognized for centuries. Although isolated by Hoffman and marketed by Bayer in the late nineteenth century, and synthesized some 40 years later, firm ideas regarding the mechanisms of the antiplatelet activity of this analgesic and antipyretic did not emerge until relatively recently. The 1960s were a great decade for fundamental research in platelets and aspirin, with the development by Born and O’Brien of their techniques for assessing platelet aggregation in vitro. Quick reported prolongation of the bleeding time after aspirin ingestion [12], while O’Brien [13], Weiss [14], and others such as Evans et al. [15] demonstrated additional effects of salicylates on platelets and alluded to possible mechanisms for this such as impaired responses to stimulants. Shortly afterwards, Vane, Smith, and Willis showed that these steps involved the enzyme cyclooxygenase (COX) [16, 17]. These and other studies lead directly to the first clinical trial of aspirin that, interestingly, failed to prove a benefit [18], However, a subsequent metaanalysis of early trials [19] provided a basis for its current use [2, 20]. Mode of action Platelets can be activated via a number of routes, such as by the physical effects of increased shear forces present in turbulent blood, and via binding of specific receptors by agonists such as ADP, prostanoids, serotonin, epinephrine, thromboxane A2 (TBXA2), thrombin, and collagen [21]. Other membrane components not intimately involved in direct signalling include structural molecules such as gpIb-V-IX that binds von Willebrand factor (vWF) and gpIa-IIa, the collagen adhesion receptor. Occupancy of signalling receptors by respective ligands (e.g. gpVI by collagen) induces the activity of intracellular second messenger systems (such as tyrosine kinases, protein kinase C, phospholipase C, and G proteins), cyclic adenosine monophosphate (cAMP) [22], and the mobilization of intracellular calcium [23]. Increased cytoplasmic calcium results in the liberation of arachidonic acid from membrane phospholipids by phospholipase A2, and steps that lead to activation of gpIIb/IIIa. Arachidonic acid is converted by COX to labile endoperoxide prostaglandin intermediates (PGG2 and PGH2) and thence by thromboxane synthase to TBXA2. These events bring about structural changes in the gpIIb/IIIa receptor complex (the most abundant platelet adhesion receptor (perhaps 50,000–80,000 copies per cell), also known as integrin αIIbβ3) that allow it to bind to its major ligand, fibrinogen [24]. Major changes such as shape change and the degranulation of dense and alpha granules follow. Many of the contents of these granules, such as ADP, fibrinogen, calcium, serotonin, and TBXA2, themselves contribute to additional activation by positive feedback [25], and these lead

Thrombosis in clinical practice

38

ultimately to aggregation and (if present) adhesion (e.g. via vWF) to components of the subendothelium such as collagen, laminin, and vitronectin (Figure 3.1). Aspirin acts by irreversibly acetylating serine residue 530 (although some refer to serine 529) of the COX molecule, an enzyme present in platelets, endothelial cells, and in gastric mucosa. This amino acid is part of the active site for the substrate arachidonic acid, which it therefore cannot bind, resulting in reduced product, that is intracellular TBXA2. As this latter molecule is a powerful stimulant of addition cell activation, use of aspirin leads

Figure 3.1 Strategies to reduce the risk of thrombosis. This figure summarizes the relationships between the coagulation pathway (top right), platelets (center), and the blood vessel wall (bottom). Opportunities to inhibit thrombosis are in boxes, interactions as arrows. Warfarin acts on various soluble coagulation proteins, heparin(s) by promoting the activity of antithrombin and in inhibiting factor

Commonly used anticoagulant and antiplatelet drugs

39

Xa, and direct thrombin inhibitors (such as Ximelagratan) by interfering with thrombin. Clopidogrel blocks the ADP receptor on the platelet surface whereas aspirin enters the platelets to inhibit intracellular signalling. Finally, gpIIb/IIIa inhibitors (such as ReoPro) prevent fibrinogen binding. The figure also illustrates the fact that numerous pathways, as yet, have no well developed specific inhibitors close to the market. There include platelet signalling via the collagen and thrombin receptors, and molecules that minimize the interactions between vWF (arising from a damaged endothelium), the platelet, and the subendothelium (TBX: thromboxane; vWF: von Willebrand factor; ADP: adenosine diphosphate). to a reduction in platelet aggregation via this particular pathway. However, a weakness of aspirin is that it has little effect on other routes (e.g. the phospholipase-dependent pathway) to platelet activation. Thus, it fails to prevent thrombin-induced platelet aggregation and only partially inhibits that induced by ADP and high doses of collagen [26, 27]. Nevertheless, reduced TBXA2 has additional benefits as vasoconstriction is also reduced, providing an additional mechanism to minimized thrombosis. An additional theoretical disadvantage of aspirin is the reduction in levels of prostacyclin [16, 17], which effectively acts as an inhibitor of platelet aggregation and vasodilation. However, as the nucleate endothelium has the ability to synthesize new COX, fresh prostacyclin can be produced by this cell, and in any case it is clear that the beneficial inhibition of TBXA2 exceeds the possibly deleterious loss of prostacyclin. Aspirin may also have non-platelet effects that are desirable, such as inhibiting nuclear transcription initiators such as NFκb (implicated in the promotion of various genes with pro-inflammatory activity) [28], in protecting low-density lipoprotein cholesterol from oxidative modification [29], and in improving endothelial dysfunction in atherosclerosis [30], However, the precise value of these mechanisms in vivo and any possible contribution to a reduction in thrombosis is unclear. The drug itself is cleared rapidly from the circulation by the liver, but its effect on platelet function can (in theory) persist for up to 10 days (i.e. the maximum lifetime of the platelet), although in practice the effect is lost after 5–7 days, dependent on the dose, as fresh, aspirin-free platelets are shed by the megakaryocyte [14]. The daily

Thrombosis in clinical practice

40

administration of a low dose of aspirin, such as 20–50 mg o.d., may result in 90–98% inhibition of TBXA2 production within 3–7 days as inhibition of COX is cumulative and requires several doses for full effect [31]. Clinical effectiveness The minimum dose with the most consistent clinical benefit in numerous early clinical trials is often quoted as 160–325 mg o.d. with no apparent extra benefit at higher doses of 500–1,500 mg o.d. [2, 32, 33]. However, a more recent commentary concluded that 100 mg o.d. is adequate for the prevention of MI in subjects at high risk [34] while others recommend 75 mg o.d. [35]. One of the more recent meta-analyses of 65 trials of the effects of aspirin on vascular events in high-risk patients (excluding those with acute stroke) reported a significant effect, with a 23% odds reduction [2]. Further exact aspirinspecific extrapolations (e.g. to different subgroups of patients) are difficult as the metaanalysis considered all antiplatelet drugs (such as dipyridamole and ticlopidine) together. Nevertheless, as the great majority of participants took aspirin, it is widely believed that this drug protects those with a history of MI, acute MI, a history of stroke or transient ischemic attack, acute ischemic stroke, or other high-risk groups such as atrial fibrillation from a major vascular event [35]. Other meta-analyses have shown a benefit in subjects with unstable angina and non-ST segment elevation [33]. The International Stroke Trial in subjects with acute ischemic stroke showed the benefit of aspirin and recommended that it be started as soon as possible after onset [36]. Although aspirin is (weakly) effective in reducing the risk of venous thromboembolism (VTE) after orthopedic surgery [37, 38], there is a scarcity of head-to-head data with anticoagulants, and it plays no part in current guidelines [39, 40]. Side effects In general, low- to medium-dose aspirin is well tolerated in the majority of patients, with the principal adverse effects being gastrointestinal (GI) and hemorrhagic stroke. The former, leading to dyspepsia and bleeding, seems likely to be due to a reduction in prostacyclin, which normally offers protection. However, this damaging effect becomes significant at very high levels of aspirin, such as 900–1,300 mg o.d. The risk of GI effects has been shown to rise with increasing dose and with regular or recent use [41–43]. Two forms of COX are well recognized in man: constitutive COX-1 and inducible COX-2 [44]. It is popularly hypothesized that COX-1 is the “good” isoform generating prostaglandins for “house-keeping” functions such as GI mucosal integrity and homeostatic functions, whereas COX-2 is said to mediate inflammation and pathophysiology [45, 46]. Based on this, COX-2 inhibitors have been developed that aim to spare the GI tract, and that do not impair platelet function [47]. However, concerns have been raised, such as the possibility that inhibition of COX-2 activity may augment myocardial cell death, and that COX-2 inhibitors are prothrombotic and increase the risk of MI [48, 49]. Therefore, the value of these agents in minimizing thrombosis in cardiovascular disease is unclear. Clearly, more serious is the risk of intracranial hemorrhage, causing stroke. A metaanalysis of 16 placebo-controlled trials of aspirin for cardiac and other indications found

Commonly used anticoagulant and antiplatelet drugs

41

that aspirin increased the absolute risk of cerebral hemorrhage by 12 events per 30,000 person-years of follow-up. This risk can be minimized, but not eliminated, by administrating the lowest effective aspirin dose. It follows, therefore, that the possible value of aspirin in reducing thrombosis may be weighed against the risk of bleeding, with the greatest value being in those at highest risk [35, 50]. Despite the positive attributes of aspirin described here, several commentators have expressed the concern that it may not be broadly valuable [51]. Although recruiting patients with hypertension, the HOT trial reported no protective effect of aspirin on death, cardiovascular death, all cardiovascular events, all strokes, all MIs or silent MIs [52]. Aspirin, however, did protect against overt MI but at the cost of an increase in fatal and nonfatal major bleeds, minor bleeds, and total bleeds. There is also some evidence that dual use of aspirin and angiotensin-converting enzyme inhibitors in some subgroups of patients may be of dubious value [34, 51, 53], although this may not be true for lowdose aspirin. Nevertheless, aspirin seems to be effective in reducing the risk of colorectal cancer [54]. Other antiplatelet drugs Persantin (dipyridamole) is perhaps (historically speaking, being introduced in 1959) the second-best known antiplatelet agent. Like the quinolinone cilostazol, it acts by inhibiting phosphodiesterases, thereby preventing the degradation of cAMP to AMP [55]. The consequences of high cAMP include low intracellular calcium, which in turn minimizes cell activation, and a further bonus is in potentiating the action of prostacyclin. Despite this, in a significant proportion of cases, evidence-based medicine cannot support the widespread use of this agent in cardiological practice, although its full value in cerebrovascular disease is still unclear [2, 56, 57]. Non-steroidal anti-inflammatory drugs (NSAIDs) such as ibuprofen also function as COX inhibitors, but as most are competitive inhibitors, differing in potency, duration, and pharmacokinetics, comparison is difficult and the overall value of NSAIDs in preventing thrombosis has been questioned [58, 59]. However, there may be a place for non-aspirin drugs in those who are intolerant of aspirin due to contraindications such as GI sensitivity, or are resistant to aspirin (present in 5–10% of patients with stable coronary artery disease) [60, 61]. Combinations of different antiplatelet drugs As aspirin is effective in inhibiting only some of the metabolic pathways leading to platelet activation, the opportunity to combine it with a drug operating via a different route seems attractive. Meta-analysis indicates that addition of dipyridamole to aspirin produces no significant further reduction in vascular events compared with aspirin alone [2]. However, ADP-receptor blockade seems more fruitful. Rupprecht et al. demonstrated the synergistic and accelerated ex vivo platelet inhibitory effects of ticlopidine plus aspirin in patients after stent implantation compared with either monotherapy alone [62]. In the CURE trial of the combination of aspirin and clopidogrel, a primary outcome occurred in 11.4% of patients on aspirin alone, compared to 9.3% of patients on both agents (p<0.001) [63]. However, this was countered by an increased risk of major (3.7%

Thrombosis in clinical practice

42

vs 2.7%) and minor bleeding (5.1% vs 2.4%) in the combined group versus the aspirinonly group, although the frequency of fatal bleeds was not different. Conclusion Aspirin is generally a well-tolerated and effective agent that reduces the risk of thrombosis in a number of settings. However, its relatively weak inhibitory capacity and undesirable side effects prompt the search for better agents. Heparin Historical perspective Although Owen and Bowie [64] described references of the use of “blood thinners” from 400 BC (interestingly, a date also offered by Elwood et al. [11]), the modern era of heparin can be traced to 1880 with the description of incoagulable blood following injection of “peptone,” a partially digested protein [65]. Doyen subsequently reported extraction of a coagulant and an anticoagulant from dog liver [66]. Working in a physiology laboratory, McLean extracted a procoagulant (tissue thromboplastin) from dog hearts and livers but also noted an anticoagulant in the same tissues, naming the latter heparophosphatid [67], although it was later renamed heparin [68]. Jorpes produced a relatively safe form of heparin [69], and its clinical use began in 1937 [70, 71]. Brinkhous et al. [72] subsequently demonstrated the requirement of another substance for anticoagulation, a factor that we now recognize as antithrombin (formerly antithrombin III). Commercial heparin was thus ready for industrial/pharmacological scale production, and by 1979 its value in prophylaxis for venous thrombosis and pulmonary embolism, and possible problems, were established [3], although present concerns regarding zoonoses abound. The laboratory potential of a more efficacious fraction of whole, unfractionated heparin (UFH), that is, low-molecularweight heparin (LMWH), was also recognized in 1979 [73, 74], and clinical note was made shortly afterwards [75, 76]. Mode of action Heparin is a naturally occurring mixture of glycosaminoglycans, a family of mucopolysaccharides that consists of dermatan sulfate, chondroitin sulfate, as well as heparin sulfate. These polysaccharides are composed of long chains of repeated disaccharide units (hexosamine and glucuronic or iduronic acid), although the composition of different preparations vary markedly. In its whole, unfractionated form, with species of molecular weights ranging from 3 to 30 kDa (although most is in the range 12–15 kDa), perhaps only one-third of a standard heparin preparation has anticoagulant activity. However, the precise antithrombin cofactor activity can be accounted for by a key pentasaccharide [9]. Heparin, by itself, is not an anticoagulant: it is a cofactor in the activity of antithrombin, a 58 kDa single chain polypeptide synthesized in the liver [77]. Heparin binding to specific sites on antithrombin induces a conformational change in the latter,

Commonly used anticoagulant and antiplatelet drugs

43

exposing a site that will bind a number of serine proteases such as factor Xa, thrombin (IIa) (in approximately equal proportions) and, with less affinity, factors IXa, XIa, XIIa, kallikrein, plasmin, and C1-esterase, although almost all of its effects are against thrombin and factor Xa [4]. However, heparin fractions less than 18 monosaccharide unit (3–4 kDa) do not enhance the inhibition of thrombin by antithrombin and probably act via factor Xa inhibition. The major inhibitor of coagulation, the effects of antithrombin are accelerated some 1,000-fold in the presence of heparin. It follows, therefore, that special measures are required in subjects deficient in antithrombin (whether by genetics or excess consumption). The heparin/antithrombin and protease complex is cleared by the reticulo-endothelial system and the heparin is recycled to form fresh complexes with additional antithrombin. In high doses, heparin also prolongs the bleeding time by inhibiting platelet aggregation in vitro [78] and may exert some effect on the (rabbit) endothelium [79], although it is unclear (and possibly academic) whether or not these influence clinical efficacy. Clinical effectiveness Due to variability of effect in different individuals, mostly due to complex pharmacokinetics (such as degree of binding to numerous proteins including, platelet factor 4, vitronectin, fibronectin, and vWF [80, 81]), laboratory monitoring of the effects of heparin is necessary and can be followed with the relatively straightforward activated partial thromboplastin time (APTT), thrombin time, and to lesser extent, the prothrombin time. The APTT derives from the ability of citrated plasma to clot an artificial ‘platelet’ substitute of phospholipids and other substances to which calcium has to be added: quality control is crucial as the quality of reagents can vary markedly [82]. In practice, most clinicians aim for a degree of anticoagulation that prolongs the APTT by anything from 1.5 times to 3 times that of normal, uncoagulated blood, that is an APTT patient/control ratio of 1.5–3.0 [4, 40]. The primary use of heparin has been, and continues to be, in the treatment and prophylaxis of VTE [4, 40, 83]. As with aspirin, early crucial work was performed in the 1960s with studies looking at the effect of heparin on pulmonary embolus (PE) and deep vein thrombosis (DVT) [84, 85]. However, heparin is not absorbed from the GI tract and so must be given intravenously or subcutaneously, with the former providing a more rapid effect. Subsequent trials not only confirmed the efficacy of this agent, but also showed that continuous intravenous infusion was frequently preferable to subcutaneous or intermittent infusions [86, 87]. One “typical” trial reported a frequency of asymptomatic extension of DVT or PE of 8.2% of patients with proximal DVT taking heparin and oral anticoagulation, compared to a frequency of 39.6% in those taking anticoagulants alone [88]. There was also a reduction in the report of symptomatic events (6.7% vs 20%) but no difference in hemorrhagic complication (3% vs 5%). In practice, until recently, hospital treatment of, and prophylaxis against, VTE, generally began with heparin, moving to oral anticoagulants following discharge, for varying periods up to 6 months [89]. Heparin is also effective in the treatment of acute MI and unstable angina, with a reduced rate of left ventricular mural thrombosis [90] and recurrent MI or angina pectoris [91, 92]. Neri-Serneri et al. [93] reported a 61% reduction in re-infarction after daily

Thrombosis in clinical practice

44

subcutaneous heparin in post-MI whereas others [94] reported a 44% reduction in mortality in a similar group. These and other trials generated recommendations for the use of heparin in both VTE and coronary artery disease [4, 40]. Clinical side effects As with aspirin, the major problem with heparin is in dose-dependent excess bleeding. Early commentaries reported a rate of hemorrhagic complications of up to 30% [3]. However, a later review reported an incidence of 6.8% in patients given continuous infusion and 14.2% in patients given intermittent intravenous injections [4], emphasizing four important variables: dose, patient’s anticoagulant response, method of administration, and other patient-related factors such as renal failure or chronic alcohol use. Gallus et al. [95] reported an incidence of major bleeding of 11% in “high-risk” patients, compared to 1% in “low-risk” patients. Surprisingly, heparin-induced thrombocytopenia (HIT, generally platelet count < 140,000 cells/µL, and reversible) has been described for well over 40 years [96], with two forms (nonimmune and immune (often an IgG recognizing a heparin/platelet factor 4 complex)) being recognized [97]. When the platelet count falls, especially if it falls from normal to less than 100,000 cells/µL, cessation is mandatory and alternate anticoagulant cover (e.g. argatroban [40, 98]) is necessary. Curiously, 0.4% of patients with HIT suffer arterial thrombosis (that may follow platelet aggregation in vivo) or venous thrombosis (that may result from heparin resistance caused by the neutralizing effect of the heparininduced release of platelet factor 4). These events may be related to the generation of thrombin [99], and also to platelet activation by heparin [78, 100]. If present, HIT is likely to be apparent 3–15 days after initiation (such that a platelet count is advised on days 3–5), but may also occur very rarely within hours, with the platelet count returning to normal within 4 days of discontinuation [98]. Rates of HIT vary between less than 0.1% and 3% [40, 97, 101]. Other uncommon side-effects of heparin (especially in long-term and high-dose use) include fractures and osteopenia [102–104], hyperkalemia [105], elevations in liver enzymes [106], skin necrosis at the site of administration, alopecia, hypersensitivity (e.g. urticaria, conjunctivitis, asthma), priapism, and hypoaldosteronism [4, 40, 107], Thus, despite the proven safety and effectiveness of continuous infusions of heparin, its limitations and adverse sideeffect profile, unpredictable pharmacokinetic response, daily laboratory monitoring with dose adjustments, and requirement for hospitalization leave room for alternatives. One such alternative is a low-molecular-weight variant of the standard (unfractionated) heparin preparation. Low-molecular-weight heparin The laboratory demonstration of a more efficacious fraction of whole UFH heparin, that is LMWH [73, 74], has ushered in a new era of anticoagulation. LMWH can be manufactured from UFH by diverse routes, such as benzylation with alkaline hydrolysis, nitrous acid depolymerization or digestion, heparinase digestion, and isoamyl nitrate digestion. Each provides a different LMWH with mean molecular weights varying from 4,371 to 5,866 Da, so that each has different pharmacokinetics and activities [108–110].

Commonly used anticoagulant and antiplatelet drugs

45

However, there are class-common advantages over UFH (Table 3.1), most of which address mode of action and side effects. Broadly speaking, compared to UFH, LMWH has a longer subcutaneous half-life and, therefore, has potential for outpatient use, a more predictable anticoagulant response requiring less monitoring, and has better anti-Xa effect. Building on early trials [75, 76], several large studies have unequivocally demonstrated the value of LMWH. In cardiovascular

Table 3.1 Differences between LMWH and UFH. LMWH

UFH

5

15

Saccharide units

13–22

40–50

Anti-Xa/Anti-IIa activity

2:1–4:1

1:1

Platelet inhibition

++

++++

Inhibited by PF4

No

Yes

92–100%

30–50%

Intravenous

2

1

Subcutaneous

4

2

Endothelial binding

No

Yes

Dose-dependent clearance

No

Yes

Mode of clearance

Kidney

Liver/kidney

Frequency of HIT

Low (e.g. 0%)

High (e.g. 2.7%a)

Low

High

Anti-Xa assay

Routine APTT

Mean molecular weight (kDa)

Bioavailability Half-life (hours)

Frequency of osteoporosis Monitoring

LMWH: low-molecular-weight heparin; UFH: unfractionated heparin; PF4: platelet factor 4; HIT: heparin-induced thrombocytopenia; APTT; activated partial thromboplastin time, a Reference [101]. Data adapted from References [107–110] and others.

disease, the REDUCE trial randomized 625 patients undergoing angioplasty to an LMWH or standard heparin, reporting a 52% reduction in the composite end point of death, MI, repeated intervention or restenosis at 24 hours. However, at 30 weeks, there was no difference between the groups [111]. In the FRISC study, 1,506 patients with unstable angina or MI were randomized to a LMWH or placebo, with aspirin to both groups. Those in the LMWH group enjoyed a 63% reduction in death or MI at 6 days and a 21% reduction in death, MI or revascularization at 40 days [112]. However, although this benefit was present at 3 months, it was no longer apparent at 6 months [113]. The ESSENCE trial was similar in patients and design, but with an UFH arm versus an

Thrombosis in clinical practice

46

LMWH in 3,171 patients. Again, the benefits of the LMWH were clear with a 16% reduction in primary end point at 14 and 30 days [114]. The benefits of LMWH in acute MI have been emphasized in a metaanalysis [115]. Trials comparing LMWH with standard UFH in the initial treatment of DVT have been reviewed in meta-analyses [116–118]. Lensing et al. described [116] a recurrence rate of 3.1% in patients on LMWH compared to 6.6% (p<0.01) in those on standard heparin, with 0.8% and 2.8% (p<0.005) respectively suffering a major bleed, data confirmed by the other analyses. However, later trials (e.g. [119, 120]) failed to find a difference in efficacy between the two types of heparin with equivalence of adverse events. Interestingly, these meta-analyses also reported a lower mortality with LMWH due to causes unrelated to VTE in patients particularly with cancer. A subsequent, more formal meta-analysis of 629 cancer patients did indeed demonstrate a benefit in 3-month mortality for LMWH versus UFH [121]. Although an unsought for endpoint, and, therefore, of questionable value, there is nevertheless good mechanistic evidence for this benefit [122]. The value of LMWH compared to UFH now extends to prevention of thrombosis in general surgery, orthopedic surgery, hip fracture, multiple trauma, neurosurgery, and ischemic stroke [110, 123]. Apart from its clinical effectiveness and better side-effect profile, a further advantage of LMWH is in home-use by appropriate patients with DVT who are not severely ill [124, 125]. In both studies, LMWH administered b.i.d. subcutaneously was compared with standard continuous infusion in hospital. Both reported no significant difference in recurrence of thromboembolism with similar rates of bleeding. Those on LMWH in the study of Levine et al. [124] remained in hospital for a mean of 1.1 days, suffering 13 embolic events, compared to 6.5 days for those on UFH, with 17 events, of which 2 were fatal. The health economics of these figures is apparent, even after adjustment for the higher costs of the LMWH. The study of Koopman et al. [125] included quality of life data: overall, there was no difference between the LMWH and UFH groups, but those on LMWH reported better physical activity and social functioning at weeks 1–2. Data demonstrating the superiority of LMWH compared to UFH (e.g. shorter time to effective anticoagulation and more days of effective anticoagulation) continues to be published [126]. Whereas the weight of literature focuses on (unprovoked) DVT and PE (although, of course, many have well-established risk factors), use of LMWH is becoming established in cancer and in peripheral arterial occlusive disease [127–129], although more data on other groups, such as the obese and those in renal failure, is needed [130]. Conclusion Developed from UFH, LMWH is currently becoming the therapy of choice, either alone or in combination, in a large number of conditions where thrombosis is to be avoided or, at least, minimized. However, recent guidelines for the use of unfractionated heparin demonstrate some value [40, 110]. Despite the weight of literature, questions and doubts remain. For example, Bath et al. [131] reported low-dose and high-dose LMWH to be as efficacious as aspirin, with similar side-effect profiles, in the treatment of acute ischemic stroke. However, this study attempted to reduce incidence of DVT/PE, but the 6-month survival data showed no advantage because increased death from intralesion bleeding.

Commonly used anticoagulant and antiplatelet drugs

47

Furthermore, Knight et al. conclude that heparins may, in some circumstances, be prothrombotic [132]. Thus there remains scope for more efficacious, yet safe, anticoagulants. Warfarin Historical perspective Severe winters and economic depression of the early 1920s around the USA/Canada border area of North Dakota and Alberta forced farmers to feed their cattle on spoiled silage, one consequence of which was the appearance of (often fatal) bleeding in the cattle. Link and colleagues subsequently isolated the causative agent, bishydroxycoumarin (dicumarol), and synthesized others of the same chemical family, one of which was named warfarin (after the Wisconsin Alumni Research Foundation) [4, 133, 134]. As with the assimilation of heparin into clinical practice, data on the absorption, pharmacokinetics, and mode of operation (i.e. as an antagonist of vitamin K) of warfarin slowly accrued [135–137]. In one of the earliest trials, Sevitt and Gallagher demonstrated the beneficial effects of an alternative vitamin K antagonist (phenindione) after fracture of the femur [138]. Subsequent discussion of the use of warfarin and allied drugs in cardiovascular disease and in the treatment of VTE appeared during the 1970s (e.g. [139–141]). This decade also saw acknowledgment of possible problems of interaction with other drugs, such as antibiotics [142], decreased requirements with age [143], teratogencity [144], and use in cancer [145, 146], and closed with publications of promising clinical trials in the prophylaxis of thromboembolism [147, 148], and guidelines [149, 150]. Mode of action and pharmacology The correct functioning of a number of coagulation proteins requires post-translational γcarboxylation (addition of a COOH group) of up to 10 glutamic acid residues by (unsurprisingly) a microsomal carboxylase enzyme. A crucial cofactor in this carboxylation is the reduced (hydroquinone) form of vitamin K that, as a proton donor and oxygen acceptor, becomes an oxidized epoxide. An epoxide reductase regenerates the active, reduced form that is then available to participate in additional carboxylation reactions [151]. Warfarin exerts its effects by blocking the regeneration of reduced vitamin K. Consequently, carboxylation is impaired and the resulting partially carboxylated proteins, such as prothrombin, fail to bind calcium, and thus fail to bind phospholipid membranes [152, 153]. When given orally, sodium warfarin is completely absorbed from the stomach and proximal small intestine and is 100% bioavailable with a plasma peak at 0.3–4 hours [137]. It circulates highly (99%) bound to albumin, leading to different pharmacokinetics in hypoalbuminemia regardless of etiology [154, 155]. The mean half-life of some 40 hours (with hepatic metabolism and renal metabolite excretion) implies about a week for steady-state levels to be reached, although its biological action in influencing coagulation factors is much shorter [156]. The time to full anticoagulant effect (possibly several days)

Thrombosis in clinical practice

48

is a balance between the appearance in the plasma of ineffective molecules (due to warfarin) and the loss of effective molecules (by natural usage/ half-life, etc.), and relies on the different synthesis and elimination half-lives of particular molecules sensitive to warfarin [5]. Because of this slow onset, guidelines (e.g. [40]) often recommend concurrent use of UFH or LMWH to minimize the risk of thrombosis until warfarin has become therapeutic for two days. Mealtime wine and smoking do not appear to influence levels, although being metabolized by the cytochrome P450 enzyme system in the liver, warfarin is profoundly influenced by a large number of other drugs metabolized by the same pathway, but may also be influenced by drugs affecting absorption and clearance [157, 158]. Effect on coagulation The consequence of mis-carboxylation by warfarin of certain plasma proteins (Table 3.2) results in low levels and thus impaired hemostasis [5]. The most common laboratory test to assess this effect is the prothrombin time (PT), originally developed by Quick and colleagues, a test also responsive to factors VII and X. A test performed by mixing citrated plasma with calcium and thromboplastin, the latter is effectively a platelet substitute that provides phospholipids (often obtained from brain, lung, and placenta) and tissue factor to enable the activation of factor X by factor V and thus clot formation. However, the assay is dependent on the quality of the thromboplastin and thus requires considerable quality control and standardization [159]. To minimize these problems each thromboplastin batch needs a conversion factor (the international sensitivity index, or ISI) allowing the development of an internationally recognizable and comparable system for the effect of anticoagulation—the International Normalized Ratio (INR) [158, 160]. At the clinical level, this allows the effects of warfarin to be monitored and thus targeted, depending on the

Table 3.2 Characteristics of vitamin K-dependent proteins. Coagulation factor Half-life (hours) Molecular weight (kDa) Plasma concentration II (prothrombin)

42–72

71.6

2 µm

VII

4–6

50

1.0 nm

IX

18–30

57.1

0.1 µm

X

27–48

59

0.16 µm

Protein C

9

57

0.07 µm

Protein S

60

69

0.42 µm

Data from References [5, 155] and elsewhere.

diagnosis, mostly to ranges such as INR 2–3 or INR 3–4.5. These wide windows are necessary as any particular INR result can be influenced by a host of patient, handling, pharmaceutical, and methodological factors [161]. However, some studies (e.g. [162])

Commonly used anticoagulant and antiplatelet drugs

49

have shown that a lower dose of warfarin (to a target of INR 1.5) can also be effective in reducing thromboembolic events. Who should be treated? Guidelines generally suggest who should be treated, to which INR target, and for how long, and support this by referral to the risk of incurring a VTE [40, 158, 161, 163]. Thus, these guidelines recognize various conditions where different INR targets are likely to strike the optimum balance between safety (in terms of avoidance of hemorrhage) and efficacy (in terms of prevention of thrombosis) (Table 3.3). One of the earliest, most comprehensive documents [158] focused primarily on prevention (e.g. after hip or gynecologic surgery) and treatment of DVT, acute MI, prosthetic heart valves, and atrial fibrillation. Evolving from a previous document [164], the recent British Guidelines [163] also describe conditions where oral anticoagulation is not indicated (but where aspirin may be recommended) and provides a schedule for starting patients on warfarin. The current Third Edition is slightly more conservative, recommending

Table 3.3 Guidelines on oral anticoaugulation. USA Guidelines [161] Target INR 2–3 Prophylaxis of VT, treatment of VT, treatment of PE, tissue heart valves, valvular heart disease, atrial fibrillation, acute myocardial infarction, bileaflet mechanical valve in aortic position. Target INR 2.5–3.5 Acute myocardial infarction (but to prevent systemic embolization or recurrent myocardial infarction), mechanical prosthetic valves, certain patients with thrombosis, and the antiphospholipid syndrome. UK Guidelines [163] Target INR 2.5 PE, DVT, calf vein thrombosis, recurrence of VT when no longer on warfarin therapy, symptomatic inherited thrombophilia, non-rheumatic atrial fibrillation, atrial fibrillation due to rheumatic heart disease, congenital heart disease or thyrotoxicosis, cardioversion, mural thrombosis, cardiomyopathy. Target INR 3.5 Recurrence of VT while on warfarin therapy, antiphospholipid syndrome, mechanical prosthetic heart value. Oral anticoagulation not indicated Bioprosthetic valve, ischemic stroke without atrial fibrillation, retinal vein occlusion, peripheral arterial disease and grafts, coronary artery thrombosis, coronary artery graft thrombosis, coronary angioplasty and stents. VT: venous thromboembolism; PE: pulmonary embolus; DVT: deep vein thrombosis. Readers are encouraged to obtain and read both sets of guidelines to obtain full benefit. The above are abbreviated.

Thrombosis in clinical practice

50

a target INR for patients at high risk of 3.5, whereas the Second Edition recommended an INR range of 3.0–4.5 [164]. Recent American Guidelines [161], themselves building on previous recommendations [158], also provide a comprehensive list of drugs expected to interact with warfarin. Excellent as these guidelines are, other, more up-to-date works, make new suggestions. For example, some [165, 166] recently concluded that patients with atrial fibrillation at low risk should be treated with aspirin, warfarin being reserved for those at high risk. In cancer, different chemotherapy regimes are associated with exacerbation or inhibition of the effect of warfarin, and merely starting on this anticoagulant is a risk factor for cancer [167–169]. A further recent advancement has been the increasing recognition of numerous risk factors for VTE, and the acknowledgment that these may become targets for treatment [170, 171]. Risk factors such as raised body mass index, diabetes [172], hyperhomocysteinemia [173], increased plasma D-dimer [174, 175], factor V Leiden [176], and deficiencies or abnormalities in plasma participants in hemostasis [177] are more comprehensively discussed in Chapters 5 and 15. Hemorrhage—the major side effect Merely because an individual may be at (possibly remote) risk of VTE does not immediately require them to start prophylaxis with warfarin. This is because of the most serious side effect of the use of warfarin, that is (possibly fatal) hemorrhage [178]. Thus, clinical practice must consider both the risk of thrombosis versus the risk of bleeding. Indeed, in certain cases, the desire to reduce the risk of thrombotic stroke must be weighed against the risk of inducing almost exactly that clinical consequence with overanticoagulation, that is a hemorrhagic stroke. Unsurprisingly, probably the most important risk factor for bleeding is the INR itself, with a reasonably clear dose-response effect (Table 3.4) [179–182]. Other risk factors are use of antibiotics, increasing age, female sex, and the first year of anticoagulation [183– 185]. Pinede and colleagues comprehensively reviewed these risks, identifying seven areas of concern [186]. These are poor compliance and the occurrence of drug interactions (including diet), the intensity of the anticoagulant effect (i.e. the INR), variability in hypocoagulation, length of treatment, age, associated disease (e.g. liver), and genetic mechanisms (e.g. variation in cytochrome P450 CYP2C alleles). This review [186] and others [183] also discuss a further problem in assessing risk of bleeding—that is, the lack of consensus in whether or not such an event is major or serious (which may be fatal, life-threatening, or requiring transfusion), minor (e.g. hematuria, either visual or defined by dipstick) or “insignificant” (patient reported nosebleed or bruising). Generally, treatment of bleeding is the temporary discontinuation of warfarin and/or (usually oral) administration of low dose (e.g. 1 mg) vitamin K. An additional option is to infuse fresh frozen plasma or prothrombin concentrate [183]. However, in practice, it is often the degree of unnecessarily high INR that is treated, as the relationship between bleeding and INR is poor [187]. The American Guidelines [161] recommend cessation of warfarin when INR is less than 5.0 and a lack of clinically significant bleeding, with resumption when the INR is satisfactory. For an INR between 5.0 and 9.0, with no clinically significant bleeding, cessation is again recommended, but 1–2.5 mg vitamin K is to be given orally if the patient is at increased risk of bleeding. For an INR greater than

Commonly used anticoagulant and antiplatelet drugs

51

9.0 with no bleeding, more vitamin K (3–5 mg oral) is advised. For very rapid reversal in view of serious bleeding or INR greater than 20.0, 10 mg vitamin K by slow intravenous infusion, possibly supplemented with plasma or concentrate depending on the severity, is recommended. These guidelines describe only “clinically significant bleeding,” with no additional definition.

Table 3.4 Crude relationship between INR and bleeding. Mean INR

Total % of bleeding

9.1

42.4

3.75

22.4, 24.0a

3.25

13.9

2.75

21.3

2.45

6.0a

2.25

4.3, 5.9

a Patients also treated with aspirin and dipyridamole. Data pooled from References [158, 179–182].

The British Guidelines [163] recommend that an INR between 3.0 and 6.0 at target 2.5, or INR between 4.0 and 6.0 at target 3.5 is to be treated by cessation and restarting when INR is less than 5.0. This does not mention bleeding. If the INR is between 6.0 and 8.0 with no bleeding or minor bleeding, again stop warfarin and restart when INR is less than 5.0. In cases of INR greater than 8.0, with no bleeding or minor bleeding, then guidelines suggest to not only stop and restart as before, but also to give 0.5–2.5 mg vitamin K orally if there are other risk factors for bleeding. The final recommendation for a major bleed, where cessation, 5 mg vitamin K (orally or intravenous) and prothrombin complex concentrate (50 units/kg) or fresh frozen plasma (15 ml/kg) is recommended. Both sets of recommendations advocate frequent monitoring of the INR. It must be stressed that this account is a condensation of the actual guidelines and readers are referred to the originals for full details. Duration of treatment It follows that because hemorrhage is an ongoing problem with the use of warfarin, treatment should not be continued indefinitely and thus depends on the risk of thrombosis [161, 186, 188, 189]. Current opinion therefore recognizes two poles: idiopathic and secondary/precipitated. A clear example of a precipitated thrombotic event would be one following surgery or pregnancy. Here, it may be argued that treatment need only be transient and presumes that the risk will decay with time after the particular precipitating event. Alternatively, a continuing risk factor, such as a mechanical prosthetic heart valve or untreatable atrial fibrillation, may require indefinite treatment. Trials conducted in the first half of the last decade considered duration of treatment between 6 weeks and 6 months [89, 190, 191]. Subsequently, Schulman et al. [192] showed that continued treatment following a recurrent VTE resulted in fewer

Thrombosis in clinical practice

52

complications at 4 years than those treated initially for 6 months. The 1998 American Guidelines [161] recommend at least 3 months treatment for proximal vein thrombosis, at least 6 months for idiopathic proximal vein thrombosis and recurrent vein thrombosis, and 6 weeks-3 months for symptomatic calf vein thrombosis. Indefinite treatment is indicated for more than one episode of idiopathic proximal vein thrombosis and a recognized thrombophilic defect. The concurrent 1998 British Guidelines [163] make similar recommendations: 6 months for a first event of PE or proximal vein thrombosis, but 3 months for a calf vein thrombosis in nonsurgical patients with no persistent risk factors. Other recommendations include 6 weeks for a postoperative calf vein thrombosis without persistent risk factors, but continued treatment should be considered if risk factors are persistent. A recurrence after stopping warfarin requires a further episode of treatment. A recurrence on treatment requires increased intensity (i.e. to a target of INR 3.5) or alternative anticoagulant treatment. However, since these guidelines were published, additional reports have been produced. Kearon et al. [193] recruited patients with an idiopathic DVT, provided 3 months treatment aimed at INR 2.0–3.0, then randomized to placebo or continued treatment. After 24 months, the difference in recurrence was considerably less (1.3%/patient-year vs 27.4%/patient-year, p<0.001) in those whose treatment continued, although this was at a cost of increased bleeding (p=0.03). However, in a study of idiopathic DVT of similar design, Agnelli et al. [194] found that the clinical benefit associated with extending the duration of anticoagulation therapy to 1 year is not maintained after the therapy is discontinued. At the same time, Hyers et al. [40] recommended that oral anticoagulant therapy for VTE with reversible or timelimited risk factors should be continued for at least 3 months, but that patients with a first episode of idiopathic VTE should be treated for at least 6 months. Patients with a recurrent idiopathic VTE or a continuing risk factor such as cancer, antithrombin deficiency or anti-cardiolipin antibody syndrome should be treated for 12 months or longer, while duration of therapy should be individualized in patients with deficiency of proteins C and S, multiple thrombophilic conditions, and homozygous factor V Leiden. Their final recommendation is that symptomatic isolated calf vein thrombosis be treated for at least 6–12 weeks, although Pinede et al. [195] report that 6 weeks is sufficient for this group. More recently, Agnelli et al. [196] recruited patients with a PE and treated them with warfarin or acenocoumarol (INR 2.0–3.0) for 3 months. They then randomized to no additional treatment, or continuation of treatment for 6 or 12 months, effectively finding no benefit in the extended treatment arm. This study indicates that patients with PE have a substantial risk of a recurrent VTE despite being anticoagulated to INR 2.0–3.0. In the PREVENT trial [197], Ridker and colleagues recruited 508 patients who had suffered a documented idiopathic VTE and who had been treated for at least 3 months with warfarin to INR 2.0–3.0, then randomized them to discontinue or to commence a lower intensity of warfarin to INR 1.5–2.0. After mean 2.1 years, there were 7.2 recurrent events per 100 person-years in those assigned to placebo, compared to 2.6 events per 100 person-years in those assigned warfarin, a risk reduction of 64%. Despite the excess minor bleeding in the warfarin group, this study indicates that long-term, low-intensity warfarin therapy is a highly effective method of preventing recurrent VTE.

Commonly used anticoagulant and antiplatelet drugs

53

Conclusion Warfarin is one of the most effective oral anticoagulants. However, misuse confers a significant risk of bleeding, as there is no common dose, it is difficult to control, and is sensitive to numerous commonly prescribed drugs. Yet, despite these drawbacks, it will remain a powerful agent for the reduction of various VTEs in numerous different circumstances. Head-to-head comparisons and combined therapy Much of the text here has focused on studies examining only one of the three treatments. This approach, although valuable, has limitations. For example, although Smith et al. [198] unequivocally demonstrated the benefit of warfarin over placebo following a MI, such a result cannot be compared directly to an effect of aspirin or heparin. Furthermore, a combination may be more effective than a single agent, and one treatment may be superior to another. These hypotheses have been tested by numerous groups. Cohen et al. [199] randomized patients with unstable angina or non-Q wave MI to aspirin or to aspirin plus heparin followed by aspirin plus warfarin. At 14 days, there was a significant reduction in total ischemic events in the combination group versus aspirin alone (10.5% vs 27%, p=0.004) but the difference at 12 weeks was less impressive (13% vs 25%, p=0.06). Similarly, Gurfinkel et al. [200], recruiting patients with unstable angina, found that 200 mg aspirin plus a high dose of LMWH was significantly better than aspirin alone or aspirin plus regular heparin. The Thrombosis Prevention Trial [201] randomized men at a high risk of ischemic heart disease to warfarin, low-dose aspirin, both, or neither. The principal result was that a combined treatment of warfarin and aspirin is more effective in the reduction of ischemic heart disease than either agent on its own, both of which are superior to placebo. There was no overall effect on stroke, possibly because an increased risk of hemorrhagic stroke cancelled out the reduction in risk of thrombotic stroke. Unsurprisingly, this intention to treat analysis improved significantly when the data was reanalyzed according to compliance [202]. angioplasty to aspirin plus coumarins or to Patients with coronary artery disease before aspirin alone were randomized by ten Berg et al. [203]. The 30 day and 1 year composite end points (3.4% vs 6.4%, and 14.3% vs 20.3%) were significantly fewer in the combined therapy group. Huynh et al. [204] randomized patients with unstable angina or non-ST-segment elevation MI, with prior CABG, to warfarin, low dose aspirin, or both for 12 months. There was no difference in the primary end point (death or MI or unstable angina requiring hospitalization 1 year after randomization) between the groups (14.1%, 11.5%, 11.4% respectively, p=0.76). They concluded that moderate-intensity oral anticoagulation alone or combined with low-dose aspirin does not appear to be superior to low-dose aspirin. Similarly, Mohr et al. [205] also found no effective difference in event rate between high-dose aspirin (16%) and warfarin (17.8%, p=0.25) when studying ischemic stroke. The OASIS investigators [206] also directly compared high-dose aspirin to aspirin plus warfarin, but in unstable angina. Overall, in intention to treat, they found neither

Thrombosis in clinical practice

54

therapy to be superior (8.3% vs 7.6%, p=0.40), but in subanalyses, event rates were less for warfarin in a “good compliance” group (8.9% vs 6.1%, p=0.02). Fiore et al. [207] randomized patients within 14 days of an MI to aspirin alone or aspirin plus warfarin, and found no clinical benefit of the combined therapy (17.3% deaths vs 17.6%, p=0.76, 13.1% recurrent MI vs 13.3%, p=0.78). Hurlen et al. [208] randomized post-acute MI patients to moderate-dose aspirin, relatively high-dose warfarin, or to low-dose aspirin plus moderateintensity warfarin. As the event rates were 20%, 16.7%, and 15% respectively, the authors concluded that warfarin, in combination with aspirin or given alone, was superior to aspirin alone. Broadly speaking, these trials in cardiovascular disease ([199–208], Table 3.5), generally seeking common end points such as death, stroke, revascularization, or MI, provide instances of fascinating and possibly important clinical data, alongside some data that fail to support these outcomes, possibly because of differences in patients, doses, and designs. However, an important caveat is that almost all studies (generally) reported more minor and

Table 3.5 Studies directly comparing aspirin and/or warfarin in the treatment of cardiovascular disease. Reference (year)

Patients

Treatments (target Duration or range of INR)

Outcome

[199] (1994)

214 with UA or Non-Q MI

A 162.5 mg vs A plus heparin then W(2–3)

12 weeks

Combined therapy superior at 14 days but no difference at 12 weeks

[201] (1998)

4,072 men at high risk

A 75 mg, W (1.5), both, or placebo

6.3 years

Combination better than each alone; both drug above better than placebo

[203] (2000)

1,058 with CAD A 100 mg vs A 100 then Angioplasty mg plus W (2.7)

1 year

Combination superior to monotherapy

[204] (2001)

135 with ACS and Prior CABG

A 80 mg vs W (2–2.5) 1 year vs both

No difference

[205] (2001)

2,206 with ischemic stroke

A 325 mg vs W (1.4– 2.8)

2 years

No difference

[206] (2001)

3,712 with UA

A 325 mg vs A 325 mg plus W (2–2.5)

5 months

Overall, no difference, but W superior in those in good compliance

[207] (2002)

5,059 with AMI

A 162 mg vs A 81 mg plus W (1.5–2.5)

2.7 years

No difference

[208] (2002)

3,630 with AMI

A 160 mg vs W (2.8– 4.2) vs A 75 mg plus W (2–2.5)

4 years

W, alone or in combination, better than A alone

A: aspirin; W: warfarin; UA: unstable angina; MI: myocardial infarction; ACS: acute coronary syndromes; CABG: coronary artery bypass graft; AMI: acute myocardial infarction. See individual papers for fine details.

Commonly used anticoagulant and antiplatelet drugs

55

major bleeding in patients on warfarin, although these never outweighed benefits. Notably, there were no instances of warfarin being inferior to aspirin. Nevertheless, a recent review and meta-analysis concluded that the combination of moderate-intensity oral anticoagulation and aspirin is more effective and as equally safe as aspirin alone [209]. Whether or not this is extrapolatable to other cardiovascular disease remains to be seen but will be certainly eagerly awaited. Studies in other disease continue to provide perspective. Lee et al. [210] randomized 672 cancer patients who suffered a VTE to LMWH or to warfarin (target INR 2–3) and found a considerably reduced rate of recurrence in those on LMWH (8.0%) compared to warfarin (15.8%, p=0.002), with equal bleeding in the two groups. This seems to contradict a recent meta-analysis that LMWHs are possibly as effective as vitamin K antagonists in preventing a recurrent VTE [211]. Conclusions Aspirin, heparin, and warfarin, alone or in combination, have provided a reduction in the risk of (possibly fatal) thrombosis for millions of individuals worldwide, although this must be set against a risk of excess bleeding. These three drugs operate by independent mechanisms, although there is some overlap. Figure 3.1 summarizes these mechanisms. However, this chapter has deliberately focused on these three agents, and has minimized discussing of new and exciting therapeutic agents such as Fondaparinux [9, 212], Clopidogrel [6, 213, 214], and Ximelagratan [7, 215]. The future lies with these drugs and the opportunity to improve clinical practice with them is eagerly awaited. They are discussed in depth in Chapter 15. References [1] Chesebro JH, Webster MWI, Zoldhelyi P, et al. Antithrombotic therapy and the progression of coronary artery disease. Anti-platelets versus anti-thrombins. Circulation 1992; 86(Suppl. III): III-100–II-111. [2] Antithrombotic Trialists’ Collaboration. Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, stroke in high risk patients. BMJ 2002; 324:71–86. [3] Turpie AG, Hirsch J. Prophylaxis and therapy of venous thromboembolism. CRC Crit Rev Clin Lab Sci 1979; 10:247–74. [4] Hirsh J. Heparin. N Engl J Med 1991; 324: 1565–74. [5] Stirling Y. Warfarin-induced changes in procoagulant and anticoagulant proteins. Blood Coagul Fibrinolysis 1995; 6:361–73. [6] Coukell AJ, Markham A. Clopidogrel. Drugs 1997; 54:745–50. [7] Eriksson BI, Agnelli G, Cohen AT, et al. Direct thrombin inhibitor melagratan followed by oral ximelagratan in comparison with enoxaparin for prevention of venous thromboembolism after total hip or knee replacement. Thromb Haemost 2003; 89:288–96. [8] Markwardt F. Hirudin as alternative anticoagulant—a historical review. Semin Thromb Haemostas 2002; 28:405–14. [9] Turpie AGG. Overview of the clinical results of pentasaccharide in major orthopaedic surgery. Haematologica 2001; 86:59–62.

Thrombosis in clinical practice

56

[10] Topol EJ, Byzova TV, Plow EF. Platelet gpIIb/IIIa blockers. Lancet 1999; 353:227–31. [11] Elwood PC, Hughes C, O’Brien JR. Platelets, aspirin and cardiovascular disease. Postgrad Med J 1998; 74:587–91. [12] Quick AJ. Acetylsalicylate acid and haemorrhage. Lancet 1966; ii:1134–5. [13] O’Brien JR. The effects of salicylates on human platelets. Lancet 1968; i:779–83. [14] Weiss HJ, Aledort LM, Kochwa S. The effect of salicylates on the haemostatic properties of platelets in man. J Clin Invest 1968; 47:2169–80. [15] Evans G, Packham MA, Nishizawa EE, Mustard JF, Murphy EA. The effect of acetylsalicylic acid on platelet function. J Exp Med 1968; 128:877–94. [16] Vane JR. Inhibition of prostaglandins as a mechanism of action for aspirin-like drugs. Nature 1971; 231:232–5. [17] Smith JH, Willis AL. Aspirin selectively inhibits prostaglandin production in human platelets. Nature 1971; 231:235–7. [18] Elwood PC, Cochrane AL, Burr ML, et al. A randomised controlled trial of acetylsalicylic acid in the secondary prevent of myocardial infarction. BMJ 1974; 1:436–40. [19] Anon. Aspirin after myocardial infarction. Lancet 1980; i:1172–3. [20] Willard JE, Lange RA, Hillis LD. The use of aspirin in ischaemic heart disease. N Engl J Med 1992; 327:175–80. [21] Gresele P, Page C, Fuster V, Vermylen J (eds). Platelets in Thrombotic and Non-thrombotic Disorders. Cambridge: Cambridge University Press, 2002. [22] Zieve PD, Greenough WB. Adenyl cyclase in human platelets: activity and response. Biochem Biophys Res Commun 1969; 35:462–8. [23] Rink TJ, Smith SW, Tsien RY. Cytoplasmic free calcium in human platelets: calcium threshold and calcium independent activation for shape change and secretion. FEBS Lett 1982;148:21–6. [24] Shattil SJ, Ginsberg MH, Brugge JS. Adhesive signalling in platelets. Curr Opin Cell Biol 1994;6: 695–704. [25] Siess W. Molecular mechanisms of platelet activation. Physiol Rev 1989; 69:58–178. [26] Savi P, Pflieger AM, Herbert JM. cAMP is not an important messenger for ADP-induced platelet aggregation. Blood Coagul Fibrinolysis 1996; 7: 249–52. [27] Nurden AT. New thoughts on strategies for modulating platelet function through the inhibition of surface receptors. Haemostasis 1996; 26(Suppl. 4): 78–88. [28] Amann R, Peskar BA. Anti-inflammatory effects of aspirin and sodium salicylate. Eur J Pharmacol 2002; 447:1–9. [29] Steer KA, Wallace TM, Bolton CH, Hartog M. Aspirin protects low density lipoprotein from oxidative modification. Heart 1997; 77:333–7. [30] Husain S, Andrews NP, Mulcahy D, Panza JA, Quyyumi AA. Aspirin improves endothelial dysfunction in atherosclerosis. Circulation 1998; 97: 716–20. [31] Patrono C, Ciabattoni G, Patrignani P, et al. Clinical pharmacology of platelet cyclooxygenase inhibition. Circulation 1985; 54:528–32. [32] ISIS-2 (Second International Study of Infarct Survival) Collaborative Group. Randomised trial of intravenous streptokinase, oral aspirin, both, or neither among 17,187 cases of suspected acute myocardial infarction: ISIS-2. Lancet 1988; ii:349–60. [33] American College of Cardiology/American Heart Association. ACC/AHA guidelines for the management of patients with unstable angina and non-ST segment elevation. J Am Coll Cardiol 2000; 36:970–1062. [34] Lauer M. Aspirin for primary prevention of coronary events. N Engl J Med 2002; 346:1468– 73. [35] US Preventative Services Task Force. Aspirin for the primary prevention of cardiovascular events: recommendations and rationale. Ann Intern Med 2002; 136:157–60.

Commonly used anticoagulant and antiplatelet drugs

57

[36] International Stroke Trial Collaborative Group. The International Stroke Trial (IST): a randomised trial of aspirin, subcutaneous heparin, or both, or neither among 19,435 patients with acute ischaemic stroke. Lancet 1997; 349: 1569–81. [37] Antiplatelet Trialists’ Collaboration. Collaborative overview of randomised trials of antiplatelet therapy—III: reduction in venous thrombosis and pulmonary embolism by antiplatelet prophylaxis among surgical and medical patients. BMJ 1994; 308:235–46. [38] Pulmonary Embolism Prevention (PEP) Trial Collaborative Group. Prevention of pulmonary embolism and DVT with lowdose aspirin: Pulmonary Embolism Prevention (PEP) trial. Lancet 2000; 355:1295–302. [39] Sors H, Meyer G. Place of aspirin in prophylaxis of venous thrombosis. Lancet 2000; 355:1288–9. [40] Hyers TM, Agnelli G, Hull RD, et al. Antithrombotic therapy for venous thromboembolic disease. Chest 2001; 119:176S–193S. [41] Cryer B. Gastrointestinal safety of low-dose aspirin. Am J Manag Care 2002; 8:S701–S708. [42] Graham DY, Smith LJ. Aspirin and the stomach. Ann Intern Med 1986; 104:390–8. [43] Cairns JA, Gent M, Singer J, et al. Aspirin, sulfinpyrazone, or both, in unstable angina. N Engl J Med 1985; 313:1369–75. [44] Vane JR, Botting RM. A better understanding of anti-unflammatory drugs based on isoforms of cyclo-oxygenase (COX-1 and COX-2). Adv Prostaglandin Thromboxane Leukot Res 1995; 23:41–8. [45] Parente L, Perretti M. Advances in the patho physiology of constitutive and inducible cyclooxygenase: two enzymes in the spotlight. Biochem Pharmacol 2003; 65:153–9. [46] Fitzgerald GA. Cardiovascular pharmacology of nonselective nonsteroidal anti-inflammatory drugs and coxibs: clinical considerations. Am J Cardiol 2002; 89:26D–32D. [47] Homoncik M, Malec M, Marsik C, et al. Rofecoxib exerts no effect on platelet plug formation in healthy volunteers. Clin Exp Rheumatol 2003; 21:229–31. [48] Bolli R, Shinmura K, Tang XL, et al. Discovery of a new function of cyclo-oxygenase (COX)2: COX-2 is a cardioprotective protein that alleviates ischaemia/reperfusion injury and mediates the late phase of preconditioning. Cardiovasc Res 2002; 52:506–19. [49] Howes LG, Krum H. Selective cyclo-oxygenase-2 inhibitors and myocardial infarction. How strong is the link? Drug Safety 2002; 15:829–35. [50] Ridker PM, Manson JE, Gaziano M, Buring JE, Hennekens CH. Low dose aspirin therapy for chronic stable angina. Ann Intern Med 1991;114: 835–9. [51] Cleland JGE Is aspirin “the weakest link” in cardiovascular prophylaxis? The surprising lack of evidence supporting the use of aspirin for cardiovascular disease. Prog Cardiovasc Dis 2002;44: 275–92. [52] Kjeldsen SE, Kolloch RE, Leonetti G, et al. Influence of gender and age on preventing cardiovascular disease by antihypertensive treatment and acetylsalicyclic acid. J Hypertens 2000;18:629–42. [53] Jones CG, Cleland JGE Meeting report: LIDO, HOPE, MOXCON and WASH studies. Eur J Heart Failure 1999; 1:425–31. [54] Paterson JR, Lawrence JR. Salicylic acid: a link between aspirin, diet and the prevention of colorectal cancer. Quart J Med 2001; 94:445–8. [55] Fitzgerald GA. Dipyridamole. N Engl J Med 1987; 316:1247–57. [56] Gibbs CR, Lip GYH. Do we still need dipyridamole? Br J Clin Pharmacol 1998; 45:323–8. [57] De Schryver ELLM, Algra A, van Gijn J. Dipyridamole for preventing stroke and other vascular events in patients with vascular disease. (Cochrane review) In: The Cochrane Library, Issue 1, 2003. Oxford: Update Software, UK. [58] Simons LS, Mills JA. Non-steroidal anti-inflammatory drugs. N Engl J Med 1980; 312:1179– 85. [59] Cleland JGF. No reduction in cardiovascular risk with NSAIDs—including aspirin? Lancet 2002; 359:92–3.

Thrombosis in clinical practice

58

[60] Gum PA, Kottke-Marchant K, Poggio ED, et al. Profile and prevalence of aspirin resistance in patients with cardiovascular disease. Am J Cardiol 2001; 88:230–5. [61] Weber AA, Przytulski B, Schanz A, et al. Towards a definition of aspirin resistance: a typological approach. Platelets 2002; 13:37–40. [62] Rupprecht HJ, Darius H, Borkowski U, et al. Comparison of antiplatelet effects of aspirin, ticlopidine, or their combination after stent implantation. Circulation 1998; 87:1046–52. [63] Yusuf S, Zhao F, Mehta SR, Chrolavicius S, Tognoni G, Fox KK, and the Clopidogrel in Unstable Angina to Prevent Recurrent Events Trial Investigators. Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation. N Engl J Med 2001; 345:494–502. [64] Owen CA, Bowie EJ. The history of the development of oral anticoagulant drugs. In: Poller L, Hirsh J (eds), Oral Anticoagulants. London: Arnold, 1996. [65] Schmidt-Mulheim A. Beitrage zur Kenntnis des peptones under seiner physiologischen Bedeutung. Archiv fur Anatomie und Physiologie, Physiologische Abteilung 1880:33–56. [66] Doyon M. Rapports du foie avec la coagulation du sang. Journal de Physiologie et de Pathologie General 1912;14:229–40. [67] McLean J. The thromboplastic action of cephalin. Am J Physiol 1916; 41:250–7. [68] Howell WH. Heparin, an anticoagulation. Am J Physiol 1923; 63:434–5. [69] Jorpes JE. The chemistry of heparin. Biochem J 1935; 29:1817–30. [70] Crafoord C. Preliminary report on post-operative treatment with heparin as a preventative of thrombosis. Acta Chir Scand 1937; 79:407–26. [71] Best CH. Preparation of heparin and its use in the first clinical case. Circulation 1959; 19:79– 86. [72] Brinkhous KM, Smith HP, Warner ED, Seegers WH. The inhibition of blood clotting: an unidentified substance which acts in conjunction with heparin to prevent the conversion of prothrombin to thrombin. Am J Physiol 1939; 125:683–7. [73] Beeler D, Rosenberg R, Jordan R. Fractionation of low molecular weight heparin species and the interaction with antithrombin. J Biol Chem 1979; 254:2902–13. [74] Barrowcliffe TW, Johnson EA, Eggleton CA, Kemball-Cook G, Thorne DP. Anticoagulant activities of high and low molecular weight heparin fractions. Br J Haematol 1979; 41:573–83. [75] Kakkar W, Djazaeri B, Fok M, Scully MF, Westwick J. Low molecular weight heparin and prevention of postoperative deep vein thrombosis. BMJ 1982; 284:375–9. [76] Hirsh J, Ofosu F, Buchanan M. Rationale behind the development of low molecular weight heparin derivatives. Semin Thromb Haemostas 1985; 11: 13–16. [77] Roemisch J, Gray E, Hoffmann JN, Wiedermann CJ. Antithrombin: a new look at the actions of a serine protease inhibitor. Blood Coagul Fibrinolysis 2002; 13:657–70. [78] Salzman EW, Rosenberg RD, Smith MH, Lindon JN, Favreau L. Effects of heparin and heparin fractions on platelet aggregation. J Clin Invest 1980; 65:64–73. [79] Blajchman MA, Young E, Ofusu FA. Effects of unfractionated heparin, dermatan sulphate and low molecular weight heparin on vessel wall permeability in rabbits. Annals NY Acad Sci 1989; 556:245–54. [80] Lindahl U, Hook M. Glycosaminoglycans and their binding to biological macromolecules. Ann Rev Biochem 1978; 47:385–417. [81] Cipolle R, Seifert R, Neilan B, Zaske DE, Hasus E. Heparin kinetics: variables related to disposition and dosage. Clin Pharmacol Ther 1981; 29:387–93. [82] Shojania AM, Tetreault J, Turnbull JL. The variations between heparin sensitivity of different lots of activated partial thromboplastin time reagent produced by the same manufacturer. Am J Clin Pathol 1988;89:19–23. [83] Turpie AGG, Chin BSP, Lip GYH. Venous thromboembolism: treatment strategies. BMJ 2002; 325:948–50. [84] Barritt DW, Jordan SC. Anticoagulant drugs in the treatment of pulmonary embolus: a controlled clinical trial. Lancet 1960; i:1309–12.

Commonly used anticoagulant and antiplatelet drugs

59

[85] Kernohan RJ, Todd C. Heparin therapy in thromboembolic disease. Lancet 1966; i:621–3. [86] Salzman EW, Deykin D, Shaprio MR, Rosenberg R. Management of heparin therapy: controlled prospective trial. N Engl J Med 1975; 292: 1046–50. [87] Hull RD, Raskob GE, Hirsh J, et al. Continuous intravenous heparin compared with intermittent subcutaneous heparin in the initial treatment of proximal-vein thrombosis. N Engl J Med 1986; 315:1109–14. [88] Brandjes DPM, Heijboer H, Buller HR, de Rijk M, Jagt H, ten Cate JW. Acenocoumarol and heparin compared with acenocoumarol alone in the initial tratment of proximal vein thrombosis. N Engl J Med 1992; 327:1485–9. [89] Schulman S, Rhedin AS, Lindmarker P, et al. A comparison of six weeks of oral anticoagulant therapy after a first episode of venous thromboembolism. Duration of anticoagulation trial study group. N Engl J Med 1995; 332:1661–5. [90] Turpie AGG, Robinson JG, Doyle DJ, et al. Comparison of high-dose with low-dose subcutaneous heparin to prevent left ventricular mural thrombosis in patients with acute transmural anterior myocardial infarction. N Engl J Med 1989; 320:352–7. [91] Kaplan K, Davison R, Parker M, Mayberry B, Feiereisel P, Salinger M. Role of heparin after intravenous therapy for acute myocardial infarction. Am J Cardiol 1987; 59:241–4. [92] Theroux P, Ouimet H, McCans J, et al. Aspirin, heparin, or both to treat acute unstable angina? N Engl J Med 1988; 319:1 105–11. [93] Neri-Serneri GG, Rovelli F, Gensini GF, Pirelli S, Carnovali M, Fortini A. Effectiveness of low dose heparin in prevention of myocardial infarction. Lancet 1987; i:937–42. [94] The SCATI Group. Randomised controlled trial of subcutaneous calcium-heparin in acute myocardial infarction. Lancet 1989; ii: 182–6. [95] Gallus A, Jackaman J, Tillett J, Mills W, Wycherley A. Safety and efficacy of warfarin started early after submassive proximal venous thrombosis or pulmonary embolus. Lancet 1986; ii:1293–6. [96] Weismann RE, Tobin RW. Arterial embolism occurring during systemic heparin therapy. Arch Surg 1958; 76:219–27. [97] Chong BH. Heparin induced thrombocytopenia. J Thromb Haemost 2003; 1:1471–8. [98] Warkentin TE. Heparin induced thrombocytopenia. A ten year retrospective. Annu Rev Med 1999; 50:129–47. [99] Warkentin TE. Current agents for the treatment of patients with heparin induced thrombocytopenia. Curr Opin Pulm Med 2002; 8:405–12. [100] Xiao Z, Theroux P. Platelet activation with unfractionated heparin at therapeutic concentrations and comparisons with a low molecular weight heparin and with a direct thrombin inhibitor. Circulation 1998; 97:251–6. [101] Warkentin TE, Levine MN, Hirsh J, et al. Heparin-induced thrombocytopenia in patients treated with low molecular weight heparin or unfractionated heparin. N Engl J Med 1995; 332: 1330–5. [102] Squires JW, Pinch LW. Heparin induced spinal fractures. JAMA 1979; 241:2417–18. [103] Sackler JP, Liu L. Heparin induced osteoporosis. Br J Radiol 1973; 46:458–60. [104] Ginsberg JS, Kowalchuk G, Hirsch J, et al. Heparin effect on bone density. Thromb Haemost 1990;64:286–9. [105] Edes TE, Sunderrajan EV. Heparin-induced hyperkalaemia. Arch Intern Med 1985; 145: 1070–2. [106] Dukes GE Jr, Sanders SW, Russo J Jr, et al. Transaminase elevations in patients receiving bovine or porcine heparin. Ann Intern Med 1984; 100:646–50. [107] Hirsch J, Dalen JE, Deykin D, Poller L. Heparin: mechanism of action, pharmacokinetics, dosing considerations, monitoring, efficacy, safety. Chest 1992; 102(Suppl.):337S–351S. [108] Turpie AGG. Can we differentiate the low molecular weight heparins? Clin Cardiol 2000; 23(Suppl. I):I-4–I-7. [109] Offord R, Perry D. Anticoagulation. London: Science Press, 2002.

Thrombosis in clinical practice

60

[110] Hirsh J, Warkentin TE, Shaughnessy SG, et al. Heparin and low molecular weight heparin. Mechanism of action, pharmacokinetics, dosing, monitoring, efficacy, safety. Chest 2001; 119(Suppl.):64S–94S. [111] Karsch KR, Preisack MB, Rainer B, et al. Low molecular weight heparin (Reviparin) in percutaneous transluminal coronary angioplasty: results of a randomised, double blind unfractionated heparin and placebo controlled, multicentre trial. J Am Coll Cardiol 1996; 28:1437–43. [112] Fragmin during instability in coronary artery disease (FRISC) study. Low molecular weight heparin during instability in coronary artery disease. Lancet 1996; 347:561–8. [113] Fragmin and fast revascularisation during instability in coronary artery disease (FRISC II) Investigators. Long term low molecular mass heparin in unstable coronary artery disease: FRISC II prospective randomised multicentre study. Lancet 1999; 354:701–7. [114] Cohen M, Demers, Gurfinkel EP, et al. Efficacy and safety of subcutaneous enoxaparin in non-Q-wave coronary events study group: a comparison of low molecular weight heparin with unfractionated heparin for unstable coronary artery disease. N Engl J Med 1997; 337:447–52. [115] Collins R, MacMahon S, Flather M. Clinical effects of anticoagulant therapy in suspected acute myocardial infarction: systematic overview of randomised trials. BMJ 1996; 313:652–9. [116] Lensing AWA, Prins MH, Davidson BL, Hirsh J. Treatment of deep vein thrombosis with low molecular weight heparins. A meta-analysis. Arch Intern Med 1995; 155:601–7. [117] Leizorovicz A, Simonneau G, Decousus H, Boissel JP. Comparison of efficacy and safety of low molecular weight heparins and unfractionated heparin in initial treatment of deep vein thrombosis: a meta analysis. BMJ 1994; 309: 648–51. [118] Siragusa S, Cosmi B, Piovella F, Hirsch J, Ginsberg G. Low molecular weight heparins and unfractionated heparin in the treatment of patients with acute venous thromboembolism. Results of a meta-analysis. Am J Med 1996; 100:269–77. [119] The Columbus Investigators. Low molecular weight heparin in the treatment of patients with venous thromboembolism. N Engl J Med 1997; 337:657–62. [120] Simonneau G, Sors H, Charbonnier B, et al. for the THESEE study group. A comparison of low molecular weight heparin with unfractionated heparin for acute pulmonary embolism. N Engl J Med 1997; 337:663–9. [121] Gould MK, Dembitzer AD, Doyle RL, et al. Low molecular weight heparins compared with unfractionated heparin for the treatment of acute deep vein thrombosis: a meta-analysis of randomised controlled trials. Ann Intern Med 1999; 130:800–9. [122] Norrby K. Heparin and angiogenesis: a low molecular weight fraction inhibits and a high molecular weight fraction stimulates angiogenesis systematically. Haemostasis 1993; 23:141–9. [123] Kay R, Wong KS, Yu YL, et al. Low molecular weight heparin for the treatment of acute ischaemic stroke. N Engl J Med 1995; 333:1588–93. [124] Levine M, Gent M, Hirsch J, et al. A comparison of low molecular weight heparin administered primarily at home with unfractionated heparin administered in the hospital for proximal deep vein thrombosis. N Engl J Med 1996; 334: 677–81. [125] Koopman MMW, Prandoni P, Piovella F, et al. Treatment of venous thrombosis with intravenous unfractionated heparin administered in the hospital as compared with subcutaneous low molecular weight heparin administered at home. N Engl J Med 1996; 334:682–7. [126] Omran H, Hammerstingl C, Schmidt H, von der Recke G, Dieter W, Luderitz B. A prospective and randomised comparison of the safety and effects of therapeutic levels of enoxaparin versus unfractionated heparin in chronically anticoagulated patients undergoing elective cardiac catheterisation. Thromb Haemost 2003; 90:267–71. [127] Jackson MR, Clagett GP. Antithrombotic therapy in peripheral arterial occlusive disease. Chest 2001; 119:283S–299S. [128] Lee AY. The role of low molecular weight heparins in the prevention and treatment of venous thromboembolism in cancer patients. Curr Opin Pulm Med 2003; 9:351–5.

Commonly used anticoagulant and antiplatelet drugs

61

[129] Levine MN, Lee AY, Kakkar AK. From Trousseau to targeted therapy: new insights and innovations in thrombosis and cancer. J Thromb Haemost 2003; 1:1456–63. [130] Schulman S. Unresolved issues in anticoagulant therapy. J Thromb Haemost 2003; 1:1464– 70. [131] Bath PMW, Lindenstrom E, Boysen G, et al. Tinzaparin in acute ischaemic stroke (TIAST): a randomised aspirin controlled trial. Lancet 2001; 358:702–10. [132] Knight CJ, Panseart M, Wilson DJ, et al. Increased platelet responsiveness following coronary stenting. Eur Heart J 1998; 19:1239–48. [133] Link KP. The discovery of dicoumarol and its sequels. Circulation 1959; 19:97–107. [134] Overman RS, Stahmann MA, Sullivan WR, Huebuert CF, Campbell HA, Link KR Studies on the haemorrhagic sweet clover disease IV. The isolation and crystallisation of the haemorrhagic agent. J Biol Chem 1941; 141:941–55. [135] O’Reilly RA, Aggeler PM, Leong LS. Studies on the coumarin anticoagulant drugs: pharmacodynamics of Warfarin in man. J Clin Invest 1963; 42:1542–51. [136] Owren PA. Report on long term therapy with anticoagulants. Thromb Diath Haemorrh Suppl 1964; 12:131–2. [137] Deykin D. Warfarin therapy. N Engl J Med 1970; 283:691–4, 801–3. [138] Sevitt S, Gallagher NG. Prevention of venous thrombosis and pulmonary embolus in injured patients: a trial of anti-coagulant prophylaxis with phenidione in middle aged and elderly patients with fractured necks of femur. Lancet 1959; ii:981–9. [139] Taberner DA, Poller L, Burslem RW, Jones JB. Oral anticoagulants controlled by the British comparative thromboplastin versus low-dose heparin in prophylaxis of deep vein thrombosis. BMJ 1978; 1:272–4. [140] Clagett GP, Salzman EW. Prevention of venous thromboembolism in surgical patients. N Engl J Med 1974; 290:93–6. [141] Robinson DS. Anticoagulants updated—application in coronary artery disease. Cardiovasc Clin 1974; 6:9–22. [142] Self T, Mann RB. Interaction of rifampin and warfarin. Chest 1975; 67:490–1. [143] Eccles JT. Control of warfarin therapy in the elderly. Age Ageing 1975; 4:161–5. [144] Barr M Jr, Burdi AR. Warfarin associated embryopathy in a 17-week old abortus. Teratology 1976; 14:129–34. [145] Thornes RD. Adjuvant therapy of cancer via the cellular immune mechanism or fibrin by induced fibrinolysis and oral anticoagulants. Cancer 1975; 35:91–7. [146] Zacharski LR, Henderson WG, Rickles FR, et al. Rationale and experimental design for the VA cooperative study of anticoagulation (Warfarin) in the treatment of cancer. Cancer 1979; 44: 732–41. [147] Hull R, Delmore T, Genton E, et al. Warfarin sodium versus low dose heparin in the long term treatment of venous thrombosis. N Engl J Med 1979; 301:855–8. [148] Barber HM, Feil EJ, Galasko CS, et al. A comparative study of dextran-70, Warfarin and low-dose heparin for the prophylaxis of thromboembolism following total hip replacement. Postgrad Med J 1977; 53:130–3. [149] Cade JF, Hunt D, Stubbs KP, Gallus AS. Guidelines for the management of oral anticoagulant therapy in patients undergoing surgery. Med J Aust 1979; 2:292–4. [150] Sandok BA, Furlan AJ, Ehisnant JP, Sundt TM Jr. Guidelines for the manangement of transient ischaemic attacks. Mayo Clin Proc 1978; 53: 665–74. [151] Olson Re, Suttie JW. Vitamin K and γ-carboxyglutamate biosynthesis. Vitam Horm 1977; 35:59–108. [152] Nelsestuen GL. Role of γ-carboxyglutamic acid: an unusual transition required for calciumdependent binding of prothrombin to phospholipid. J Biol Chem 1976; 251:5648–56. [153] Friedman PA, Rosenberg RD, Hauschka PV, Fitz-James A. A spectrum of partially carboxylated prothrombins in the plasmas of coumarintreated patients. Biochim Biophys Acta 1977; 494: 271–6.

Thrombosis in clinical practice

62

[154] O’Reilly RA. The binding of sodium Warfarin to plasma albumin and its displacement by phenyl-butazone. Ann NY Acad Sci 1973; 226: 293–308. [155] Ganeval D, Fischer AM, Barre J, Pertuiset N, Dautzenberg MD. Pharmacokinetics of Warfarin in nephritic syndrome and effect of vitamin K dependent clotting factors. Clin Nephrol 1986; 25:75–80. [156] Andreasen P, Vessel E. Comparison of plasma levels of antipyrine, tolbutamide and Warfarin after oral and intravenous administration. Clin Pharmacol Ther 1974; 16:1059–65. [157] Routledge PA, Shetty HGM. Pharmacology of anticoagulants. In: Butchart EG, Bodnar E (eds), Thrombosis, Embolism and Bleeding. London: ICR Publishers, 1992. [158] Hirsh J, Dalen JE, Deykin D, Poller L. Oral anticoagulants: mechanism of action, clinical effectiveness and optimal therapeutic range. Chest 1992; 102(Suppl.):312S-326S. [159] Dunford SR. Thromboplastin standardisation: calibrated lyophilised plasma and coagulometerspecific international sensitivity indices. Br J Biomed Sci 1998; 55:276–80. [160] Poller L. Progress in standardisation in anticoagulant control. Haematol Rev 1987; 1:225–41. [161] Hirsh J, Dalen JE, Anderson DR, et al. Oral anticoagulants: mechanism of action, clinical effectiveness and optimal therapeutic range. Chest 1998; 114(Suppl.):445S–469S. [162] Levine M, Hirsh J, Gent M, et al. Double-blind randomised trial of a very-low-dose warfarin for prevention of thromboembolism in stage IV breast cancer. Lancet 1994; 343:886–9. [163] Haemostasis Task force for the British Committee for Standards in Haematology. Guidelines on oral anticoagulation, 3rd Edition. Br J Haematol 1998; 101:374–87. [164] British Society for Haematology. Guidelines of oral anticoagulation, 2nd Edition. J Clin Pathol 1990; 43:177–83. [165] Hart RG, Halperin JL, Pearce LA, et al. Lessons from the stroke prevention in atrial fibrillation trials. Ann Intern Med 2003; 138:831–8. [166] Lip GY, Hart RG, Conway DS. Antithrombotic therapy for atrial fibrillation. BMJ 2002; 325: 1022–5. [167] Masci G, Magagnoli M, Zucali PA, et al. Minidose warfarin prophylaxis for catheterassociated thrombosis in cancer patients: can it be safely associated with fluorouracil based chemotherapy. J Clin Oncol 2003; 21:736–9. [168] Martin LA, Mehta SD. Diminished anticoagulant effects of warfarin with concomitant mercaptopurine therapy. Pharmacotherapy 2003; 23:260–4. [169] Sculman S, Lindmarker P. Incidence of cancer after prophylaxis with warfarin against recurrent venous thromboembolism. N Engl J Med 2000; 342:1953–8. [170] Rosendaal FR. Risk factors for venous thrombotic disease. Thromb Haemost 1999;82:610– 19. [171] Bauer KA. Management of thrombophilia. J Thromb Haemost 2003; 1:1429–34. [172] Tsai AW, Cushman M, Rosamund WD, Heckbert SR, Polak JF, Folsum AR. Cardiovascular risk factors and venous thromboembolism incidence: the longitudinal investigation of thromboembolism etiology. Arch Intern Med 2002;162:1182–9. [173] Tsai AW, Cushman M, Tsai MY, et al. Serum homocysteine, thermolabile variant of methylene tetrahydrofolate reductase (MTHFR), and venous thromboembolism: Longitudinal Investigation of Thromboembolism Etiology (LITE). Am J Hematol 2003; 72:192–200. [174] Cushman M, Folsom AR, Wang L, et al. Fibrin fragment D-dimer and the risk of future venous thrombosis. Blood 2003; 101:1243–8. [175] Palareti G, Legnani C, Cosmi B, et al. Predictive value of D-dimer test for recurrent venous thromboembolism after anticoagulation withdrawal in subjects with a previous idiopathic event and in carriers of congenital thrombophilia. Circulation 2003; 108:313–18. [176] Folsom AR, Cushman M, Tsai MY, et al. A prospective study of venous thromboembolism in relation to factor V Leiden and related factors. Blood 2002; 99:2720–5. [177] Bettina RM. Molecular risk factors for thrombosis. Thromb Haemost 1999;82:601–9. [178] Levine MN, Hirsh J, Landefeld S, Raskob G. Hemorrhagic complications of anticoagulant treatment. Chest 1992; 102:352S–363S.

Commonly used anticoagulant and antiplatelet drugs

63

[179] Hull R, Delmore T, Carter C, et al. Adjusted subcutaneous heparin versus warfarin sodium in the long-term treatment of venous thrombosis. N Engl J Med 1982; 306:189–94. [180] Turpie AG, Gent M, Laupacis A, et al. A comparison of aspirin with placebo in patients treated with warfarin after heart-valve replacement. N Engl J Med 1993; 329:524–9. [181] Saour JN, Sieck JO, Mamo LA, Gallus AS. Trial of different intensities of anticoagulation in patients with prosthetic heart valves. N Engl J Med 1990; 322:428–32. [182] Altman R, Rouvier J, Gurfinkel E, Scazziota A, Turpie AG. Comparison of high-dose with lowdose aspirin in patients with mechanical heart valve replacement treated with oral anticoagulant. Circulation 1996; 94:2113–16. [183] Makris M, Watson HG. The management of coumarin-induced over-anticoagulation. Br J Haematol 2001; 114:271–80. [184] Van der Meer FJ, Rosendaal FR, Vandenbroucke JP, Briet E. Assessment of a bleeding risk index in two cohorts of patients treated with oral anticoagulants. Thromb Haemost 1996; 76:12– 16. [185] Panneerselvam S, Baglin C, Lefort W, Baglin T. Analysis of risk factors for overanticoagulation in patients receiving long-term warfarin. Br J Haematol 1998; 103:422–4. [186] Pinede L, Duhaut P, Ninet J. Management of oral anticoagulants in the treatment of venous thromboembolism. Eur J Intern Med 2001; 12:75–85. [187] Crowther MA, Donovan D, Harrison L, McGinnis J, Ginsberg J. Low-dose oral vitamin K reliably reverses over-anticoagulation due to warfarin. Thromb Haemost 1998; 79:1116–18. [188] Diuguid DL. Oral anticoagulant therapy for venous thromboembolism. N Engl J Med 1997; 336:433–4. [189] Pinede L, Cucherat P, Duhaut P, Ninet J, Boissel JP. Optimal duration of anticoagulation therapy after an episode of venous thromboembolism. Blood Coagul Fibrinolysis 2000; 11:701– 7. [190] Research Committee of the British Thoracic Society. Optimum duration of anticoagulation for deep vein thrombosis and pulmonary embolism. Lancet 1992; 340:873–6. [191] Levine MN, Hirsh J, Gent M, et al. Optimal duration of oral anticoagulation therapy: a randomised trial comparing four weeks with three months of warfarin in patients with proximal deep vein thrombosis. Thromb Haemost 1995; 74:606–11. [192] Schulman S, Granqvist S, Holmstrom M, et al. The duration of oral anticoagulant therapy after a second episode of venous thromboembolism. N Engl J Med 1997; 336:393–8. [193] Kearon C, Gent M, Hirsh J, et al. A comparison of three months of anticoagulation with extended anticoagulation for a first episode of idiopathic venous thromboembolism. N Engl J Med 1999; 340: 901–7. Erratum in: N Engl J Med 1999; 341:298. [194] Agnelli G, Prandoni P, Santamaria MG, et al. Three months versus one year of oral anticoagulant therapy for idiopathic deep venous thrombosis. Warfarin Optimal Duration Italian Trial Investigators. N Engl J Med 2001; 345:165–9. [195] Pinede L, Ninet J, Duhaut P, et al. Comparison of 3 and 6 months of oral anticoagulant therapy after a first episode of proximal deep vein thrombosis or pulmonary embolism and comparison of 6 and 12 weeks of therapy after isolated calf deep vein thrombosis. Circulation 2001; 103: 2453–60. [196] Agnelli G, Prandoni P, Becattini C, et al. Warfarin Optimal Duration Italian Trial Investigators. Extended oral anticoagulant therapy after a first episode of pulmonary embolism. Ann Intern Med 2003; 139:19–25. [197] Ridker PM, Goldhaber SZ, Danielson E, et al. PREVENT Investigators. Long-term, lowintensity warfarin therapy for the prevention of recurrent venous thromboembolism. N Engl J Med 2003; 348:1425–34. [198] Smith P, Arnesen H, Holme I. The effect of warfarin on mortality and reinfarction after myocardial infarction. N Engl J Med 1990; 323: 147–52. [199] Cohen M, Adams PC, Parry G, et al. Combination antithrombotic therapy in unstable rest angina and non-Q-wave infarction in nonprior aspirin users. Primary end points analysis from

Thrombosis in clinical practice

64

the ATACS trial. Antithrombotic Therapy in Acute Coronary Syndromes Research Group. Circulation 1994;89: 81–8. [200] Gurfinkel EP, Manos EJ, Mejail RI, et al. Low molecular weight heparin versus regular heparin or aspirin in the treatment of unstable angina and silent ischemia. J Am Coll Cardiol 1995; 26: 313–18. [201] The Medical Research Council’s General Practice Research Framework. Thrombosis Prevention Trial: randomised trial of low-intensity oral anticoagulation with warfarin and lowdose aspirin in the primary prevention of ischaemic heart disease in men at increased risk. Lancet 1998; 351:233–41. [202] Rudnicka AR, Ashby D, Brennan P, Meade T. Thrombosis prevention trial: compliance with warfarin treatment and investigation of a retained effect. Arch Intern Med 2003; 163:1454–60. [203] ten Berg JM, Kelder JC, Suttorp MJ, et al. Effect of coumarins started before coronary angioplasty on acute complications and long-term follow-up: a randomized trial. Circulation 2000;102: 386–91. [204] Huynh T, Theroux P, Bogaty P, Nasmith J, Solymoss S. Aspirin, warfarin, or the combination for secondary prevention of coronary events in patients with acute coronary syndromes and prior coronary artery bypass surgery. Circulation 2001; 103:3069–74. [205] Mohr JP, Thompson JL, Lazar RM, et al. Warfarin-Aspirin Recurrent Stroke Study Group. A comparison of warfarin and aspirin for the prevention of recurrent ischaemic stroke. N Engl J Med 2001; 345:1444–51. [206] The Organisation to Assess Strategies for Ischaemic Syndromes (OASIS) investigators. Effects of long term, moderate intensity oral anticoagulation in addition to aspirin in unstable angina. J Am Coll Cardiol 2001; 37:475–84. [207] Fiore LD, Ezekowitz MD, Brophy MT, et al. Department of Veterans Affairs Cooperative Studies Program Clinical Trial comparing combined warfarin and aspirin with aspirin alone in survivors of acute myocardial infarction: primary results of the CHAMP study. Circulation 2002; 105:557–63. [208] Hurlen M, Abdelnoor M, Smith P, Erikssen J, Arnesen H. Warfarin, aspirin, or both after myocardial infarction. N Engl J Med 2002; 347:969–74. [209] Anand SS, Yusuf S. Oral anticoagulants in patients with coronary artery disease. J Am Coll Cardiol 2003; 41:62S–69S. [210] Lee AY, Levine MN, Baker RI, et al. Low-molecularweight heparin versus a coumarin for the prevention of recurrent venous thromboembolism in patients with cancer. N Engl J Med 2003; 349: 146–53. [211] van der Heijden JF, Hutten BA, Buller HR, Prins MH. Vitamin K antagonists or lowmolecularweight heparin for the long term treatment of symptomatic venous thromboembolism. Cochrane Library, Issue 4. Oxford: Update software. [212] Turpie AG, Bauer KA, Eriksson BI, Lassen MR. Fondaparinux vs enoxaparin for the prevention of venous thromboembolism in major orthopedic surgery: a meta-analysis of 4 randomized double-blind studies. Arch Intern Med 2002; 162:1833–40. [213] Albers GW, Amarenco P. Combination therapy with clopidogrel and aspirin: can the CURE results be extrapolated to cerebrovascular patients? Stroke 2001; 32:2948–9. [214] Muller I, Massberg S, Zierhut W, et al. Effects of aspirin and clopidogrel versus oral anticoagulation on platelet function and on coagulation in patients with nonvalvular atrial fibrillation (CLAFIB). Pathophysiol Haemost Thromb 2002; 32:16–24. [215] Halperin JL. Executive Steering Committee, SPORTIF III and V Study Investigators. Ximelagratan compared with warfarin for prevention of thromboembolism in patients with nonvalvular atrial fibrillation: rationale, objectives, and design of a pair of clinical studies and baseline patient characteristics (SPORTIF III and V). Am Heart J 2003; 146:431–S.

Commonly used anticoagulant and antiplatelet drugs

65

4 Bleeding risk and hemorrhage Why it happens and what to do about it Gualtiero Palareti, Benilde Cosmi, and Cristina Legnani Introduction All antithrombotic treatments affect the hemostatic system and are therefore associated with a risk of bleeding which is the most important critical point for the safety of these treatments. The expected rate of bleeding complications, especially major ones, is the crucial factor determining the evaluation of risks/ benefits of antithrombotic agents. Accurate information on the risk of hemorrhage of the various antithrombotic treatments in the different clinical conditions is, however, not easy to achieve. Definition of major hemorrhage The adopted classification of bleeding events markedly influences the rate of complications reported in clinical studies. Unfortunately, a wide variety of classifications have been adopted in clinical studies (Table 4.1), probably accounting for the major differences in the bleeding rates reported in different studies. Though in most cases bleeding is categorized as major (including fatal) and minor, in some scientific works other categories were considered (fatal, life-threatening, serious, minor [1]), which are probably more difficult and/or arbitrary to apply. In this regard, the following points should be considered: (a) a classification of bleeding episodes is acceptable if it is easily reproducible and adequately reflects the clinical relevance of hemorrhage; (b) the most important difference in the various adopted classifications is with regard to the definition of major hemorrhage; and (c) the more clinically irrelevant events are included in classification, the higher may be the difference between observed rates in clinical studies [2]. Thus, the definition of “major hemorrhage” should reflect the severity of an outcome. If it includes hospitalization—without concomitant decrease in hemoglobin or need for blood transfusion, as adopted in some studies [3, 4]—it will overestimate the severity of bleeding, especially in elderly patients, since the decision to hospitalize a patient may be subjective and may reflect differences in countries’ health system procedures, different practices between physicians, presence of concomitant diseases and even social factors such as lack of family or social support. Some authors [3] even consider bleeding to be major when it prompts cessation of therapy; and this may also lead to an overestimation of the rate of major bleeding.

Bleeding risk and hemorrhage

67

Nonetheless, it seems odd that no attempt has been made to standardize how to define

Table 4.1 Different types of bleeding classification in studies on the effects of oral anticoagulant treatment (OAT). Author

Bleeding classification

Hull [83]

Major

• Fall in Hb≥2 g/dl • Transfusion of≥2 PRBC units • Retroperitoneal, intracranial, or in major prosthetic joints

Minor

• All the others

Major

• Requiring hospitalization, transfusion, or OAT discontinuation

Minor

• All the others

Major

• Requiring transfusion, emergency procedure, or admission to an ICU

Minor

• All the others

Major

• Requiring transfusion, hospitalization, or surgery Intracranial • All muscle and joint bleeds

Minor

• All the others

Major

• Intracranial • Retroperitoneal • Hb fall ≥2 g/dl • Transfusion ≥2 PRBC units

Minor

• Any clinically visible

Major

• Intracranial, or retroperitoneal • Requiring hospitalization or transfusion

Minor

• All the others

Fatal

• Cardiopulmonary arrest

Life-threatening

• Surgery or intervention • Irreversible sequelae • Transfusion≥3 PRBC units • Blood pressure<90 mmHg • Hematocrit≤20%

Serious

• Requiring treatment or medical evaluation

Minor

• No costs or visits or medical consequences

Major

• Fall in Hb≥2 g/dl • Transfusion≥2 PRBC units

Gurwitz [3]

Ezekowitz [84]

Van der Meer [4]

Isaacs [85]

Levine [71]

Fihn [1]

Koopman (TASMAN Study) [86]

Thrombosis in clinical practice

68

• Intracranial or retroperitoneal

Author

Bleeding classtfication • OAT discontinuation

Palareti (ISCOAT study) [16]

Minor

• Not major

Not considered as bleeding

• No clinically unusual bleeding

Major

• Fatal • Intracranial, ocular (with blindness), in major prosthetic joints, or retroperitoneal • Requiring surgery or invasive procedures • Hb reduction ≥2 g/dl and/or ≥2 PRBC units transfusion

Minor

• All the others

Not considered as bleeding

• Bruising, small ecchymoses • Self-stopping epistaxis • Occasional hemorrhoidal bleeding • Microscopic hematuria

ICU: intensive care unit; PRBC: packed red blood cells; Hb: hemoglobin.

the severity of bleeding; the lack of a uniform definition contributes to a poor comparability of results of clinical trials and observational studies as regards the safety of the various antithrombotic treatments. Clinical trials and observational studies Though clinical trials are the basis of our knowledge regarding the efficacy and risk/ benefit ratio of a specific treatment in a determined clinical condition, in the area of monitoring for drug toxicity or studying risk factors for disease, we must rely on welldesigned observational studies [5]. The information given by trials on the bleeding risks of antithrombotic treatments when administered in the daily practice may be inadequate. It can be expected, in fact, that the rate of bleeding complications recorded in experimental trials is generally lower than what may occur in daily practice due to the selection of patients, but is often very important, leading to exclusion of the oldest subjects, those with associated multiple diseases, and those needing polypharmacy. Results obtained in observational studies that prospectively include all patients, without any selection, since the commencement of their treatment seem more reliable as an indication of the risk to be expected in the daily practice. However, some methodological limitations may affect the reliability of results of these studies also. Some studies were performed before the introduction of the International Normalized Ratio (INR) system for expressing prothrombin time (PT) results; others either did not adopt the INR system [6, 7] or calculated INR values estimated on retrospective analysis of the observed PT ratios. The results of these studies are hardly comparable with those

Bleeding risk and hemorrhage

69

performed in the INR era. The rate of bleeding complications may also be underestimated in retrospective observational studies [4, 8–10] or in those that did not include an inception cohort of patients [11, 12]. As discussed later, the frequency of bleeding complications is higher during the first period of anticoagulant treatments. The rate of complications may therefore be underestimated in studies that included patients who were already being treated or had been treated in the past. In this case, in fact, a nonnegligible number of bleeding episodes may be overlooked and a spontaneous selection of patients may have occurred, due to the exclusion from treatment of those who have already suffered from bleeding complications.

Oral anticoagulant treatment (OAT) Oral anticoagulant treatment (OAT) is highly effective in reducing the risk of thromboembolism associated with various clinical conditions. According to Kearon and Hirsh [13] the estimated risk reduction of thromboembolic complications with OAT varies from 66% to 80% in the different clinical indications for the treatment. Though being a very demanding therapy, OAT is being used increasingly for its great efficacy in the prevention and treatment of thromboembolic complications of vascular disease. Bleeding is the most important complication and is a major concern for both physicians and patients, limiting the more widespread use of OAT. However, the incidence of bleeding varies widely in published studies. Levine et al. [14] recently reviewed the bleeding complications reported in published clinical trials on oral anticoagulation in patients with different indications for the treatment. In patients with prosthetic heart valves, the bleeding rates ranged from 0% to 2.2% for fatal bleeding and from 3.2% to 12.9% for major bleeding; in atrial fibrillation from 0% to 1.3% and from 0.2% to 5.8%; in ischemic heart disease from 0% to 2.9% and from 0% to 19.3%; in patients with venous thromboembolism (VTE) no fatal bleeding was reported and major bleeding ranged from 2.1% to 16.7%. As regards the observational studies, Table 4.2 shows the rates of major and fatal bleeding (expressed as percent patient-years of treatment, if not otherwise specified) recorded in some reports, less or more recently published. Obviously, the design of the different studies should be considered rather than the results of those which prospectively followed-up an inception cohort of patients being more reliable. The major determinants of oral anticoagulant-induced bleeding Intensity of anticoagulant effect The intensity of anticoagulation is the major determinant of OAT-induced bleeding. In clinical trials comparing different target zones in patients treated for deep venous thrombosis (DVT), mechanical or tissue heart valves, subjects randomized to a higher anticoagulation level (usually>3.0 INR) had about 2-fold more risk of bleeding than those treated at a lower anticoagulation intensity (approximately 2.0–3.0 INR) (see Table 4.2).

Thrombosis in clinical practice

70

Some observational studies showed that the actually achieved intensity of anticoagulation is likely to be the most important determinant of bleeding studies [4, 15]. In line with the results of other studies [4], the observational ISCOAT study [16] did not find a significant relationship between risk of bleeding and the intended anticoagulation range. However, a relationship between intensity of anticoagulation

Table 4.2 Rates of fatal and major bleeding complications recorded in observational studies on oral anticoagulant treatment. Sources

Design of the study/ type of patients

No. of patients

Bleeding Major Fatal (%/yr) (%/yr)

Landefeld [76]

Retrospective; various

Petitti [87]

Retrospective; VTE

Landefeld [6]

Retrospective; inception cohort; various

4.4a

617

0.2a

a,b

2,029

6.6

562

9.8a

1.8a

Launbjerg [11] Retrospective; various

551

2.7

0

Fihn [21]

928

Hurlen [12]

OAT (major=serious+life-threatening) Retrospective; various

13.1

0.20

c

2.2

2.1

d

1,140

0.45

0.30

261

8.1

0.45

2,026

Gitter [8]

Retrospective; various

White [51]

Retrospective (in part prospective); various Major bleeding=life-threatenmg

1,999

0.83

0.10

Van der Meer [60]

Retrospective; various

6,814

2.7

0.64

Fihn [1]

Retrospective; various Major bleeding=serious+lifethreatening

2,376

6.9

0.1

Palareti [16]

Prospective; various

2,745

1.1

0.25

Steffensen [88] Retrospective; inception cohort; various

682

5.0

0.9

Torn [37]

Retrospective; noncardiac cerebral ischemia; 2.5–4.5 INR

356

3.9

1.4

Pengo [24]

Prospective; atrial fibrillation

433

1.5

0.3

%/yr=% patient-years of treatment. a % patients. b Included major and fatal bleeding. c 2.5–4.2 INR. d 2.2–2.8 INR+150 mg ASA.

Bleeding risk and hemorrhage

71

achieved and temporally related risk of bleeding was clearly evident. In the ISCOAT study, the incidence of bleeding events at different achieved intensities of anticoagulation was investigated by dividing the number of events occurring in patients with temporally related INR values in increasing INR categories (<2, 2–2.9, 3– 4.4, 4.5–6.9, >7.0 INR) by the total number of patientyears accumulated in these categories. The lowest rate of bleeding (4.8% patient-years) was found in the 2.0–2.9 INR category, whereas many bleeding events occurred at a very low anticoagulation intensity (7.7% patient-years in the <2 INR category). Along with a further increase in INR levels there was an increase in the bleeding incidence which became exponential for INR values greater than 4.5. The multivariate analysis confirmed that the risk of bleeding was sixfold (95% CI 3.7–9.7) when INR values exceeded 4.5 (p<0.0001). The finding of a substantial risk of bleeding even in conjunction with very low INR (<2.0) is also in line with other reports [6] and suggests that some bleeds during OAT are not related to the intensity of anticoagulation but rather to the possible presence in some patients of a local bleeding source that may be unmasked by anticoagulant therapy [17]. A recent Swedish study on the risk of death in subjects on oral anticoagulation [18] reported that mortality from all causes of death was strongly related to the INR level, with a minimum risk at 2.2 INR. Importantly, high INR values were associated with an excess mortality: for one unit INR increase above 2.5, the risk increase was 2.2 (95% CI 2.1–2.2). The authors recommend the use of less intensive treatment with a target INR close to 2.2–2.3, irrespective of the indication for anticoagulation and more preventive actions to avoid episodes of high INR. INR level is also a major risk factor for intracranial hemorrhage, which is the most feared complication of OAT. Although many intracranial hemorrhages may occur within the therapeutic range, the risk greatly increases with INR level. It has been reported that the risk increases 4-fold for each unit increase in the PT ratio [19], and is particularly high for INR greater than 4.0 [20]. The variability of INR levels appears also to be associated with an increased frequency of bleeding independently of the mean INR [21, 22]. Patient characteristics Sex. Whereas some studies have noted an increased rate of bleeding in women treated with warfarin [4, 21, 23], others have not confirmed this finding [11, 16]. A particularly high incidence of major bleeding, statistically significant at univariate though not at multivariate analysis, has been reported by Pengo et al. [24] in elderly women anticoagulated for nonvalvular atrial fibrillation. Age. It has long been a matter of debate whether the risk of bleeding during OAT is higher in older patients (for reviews see [25–29]). In the ISCOAT study, patients aged 70 years or older showed a relative risk of 1.75 (p<0.01) compared to all the others [16]. Similar results have been found by several observational studies [4, 6, 11, 23], though not all [21]. In a large, prospective, multicenter, nested, case-control study [30], it was shown that the trend for overall rates of bleeding was higher, though the difference was not statistically significant, in patients aged 75 years or older (9.9% patient-years) than in matched (for sex, main indication for therapy and treating center) younger controls aged

Thrombosis in clinical practice

72

less than 70 (6.9% patient-years). However, there was a higher risk of major (2.1% patient-years vs 1.1% patient-year) and fatal (6 vs only 1, all due to intracranial bleeding) complications in elderly patients than in controls. These results are in keeping with the findings of recent studies in which the risk of life-threatening or fatal bleeding was significantly higher in older than younger patients treated with oral anticoagulants [1, 31]. Others have also reported that the risk of intracranial bleeding during OAT is higher in older persons [6, 19, 22]. In their review, Hart et al. [32] concluded that predictors of intracerebral hematoma during OAT are advanced age, prior ischemic stroke, hypertension, and intensity of anticoagulation. Pengo et al. [24] found that major bleeding occurred more frequently in patients aged 75 years or older (cumulative incidence 10.8%; 95% CI 1.8–19.8) than in younger patients (2.8%; 95% CI 0.3–5.3, p=0.006) treated for non-rheumatic atrial fibrillation; this was particularly true for major primary bleeding unrelated to organic lesions. Very recently, it has been shown that age is a risk factor for more unstable INR results and that the risk of very high INR values (≥5) increased with age by 15% every 10 years [33]. Though most physicians are aware of the higher risk of OAT in the elderly, an increasing number of elderly patients are treated with anticoagulants. We are thus faced with the dilemma that although older patients are likely to benefit the most from OAT, they have an increased risk of major bleeding complications. It is therefore important to assess the individual risk of anticoagulation-related bleeding in older patients in order to consider avoiding treatment in those at higher risk. Elderly patients on anticoagulants should be (a) treated at a low target zone; (b) monitored closely to keep their INR values within the therapeutic zone; and (c) carefully followed so that conditions which may potentially interfere with OAT (such as intercurrent illnesses, co-interventions, treatment compliance, and diet) can be monitored. It has been proven that a specific intervention consisting of patient education about warfarin, training to increase patient participation, self-monitoring of prothrombin time, and guideline-based management of warfarin dosing is effective in reducing the frequency of major bleeding in older patients [34]. Other individual factors. High plasma levels of thrombomodulin, which is a thrombin receptor located on endothelium surface that activates protein C when it is complexed to thrombin [35], have been reported to be associated with increased risk of bleeding in patients treated with oral anticoagulants [36]. Possible explanations of this association may be that high thrombomodulin levels in blood increase the effect of OAT by inhibiting factor Xa or by enhancing protein C activation. Type of treatment indication There are data showing that the risk of bleeding is higher when the indication for anticoagulant treatment is the presence of arterial disease. In the ISCOAT study [16] bleeding was particularly frequent in patients treated for cerebrovascular disease (14.5% patientyears). A high incidence of major bleeding (3.9% patient-years) has been reported in patients with ischemic stroke of arterial origin, but it was mainly confined to the first months of treatment and in elderly patients [37]. The higher incidence of bleeding in cerebrovascular or other “arterial” patients in comparison with the other types of patients (DVT, atrial fibrillation, valvular prostheses) raises the question whether the risk of bleeding during anticoagulation outweighs the benefits in those conditions [9].

Bleeding risk and hemorrhage

73

More recent data obtained in a populationbased study showed that the incidence of major complications in patients treated with warfarin for secondary prevention of stroke is lower than that reported in previous studies, likely due to the use of more moderate anticoagulation intensity [38]. Another recent study (the Stroke Prevention in Reversible Ischemia Trial, SPIRIT) was stopped early because of a high incidence of intracranial hemorrhage with a target INR of 3.0–4.5 [39]. In a still ongoing study (the European/Australasian Stroke Prevention in Reversible Ischaemia Trial, ESPRIT), an intermediate anticoagulation regimen (INR 2.0–3.0) is used—an interim analysis of the incidence of intracranial hemorrhage in this study reported an overall rate of 0.31% per year (1.21% if all of these were in the anticoagulation group), confirming that anticoagulation at a moderate intensity (INR of 2.0–3.0) is reasonably safe in patients with cerebral ischemia of arterial origin [40]. Concomitant diseases and co-medications Some patients have more than one indication for OAT, a condition that seems to increase the risk of bleeding. About one-third of the patients who presented bleeding complications during the ISCOAT study [16] had other clinical indications for OAT beyond the main one, most frequently the presence of peripheral and/or cerebral arterial disease, ischemic heart disease, and atrial fibrillation. At least one comorbid condition or general risk factor was present from the beginning of OAT in more than half of the patients who experienced bleeding. In a few cases, it was possible to correlate the occurrence of a bleeding episode with the onset of specific pro-hemorrhagic conditions, such as trauma, urinary infections and/or nephrolithiasis, heparin co-administration, thrombocytopenia, and lung disease. The history of gastrointestinal hemorrhaging is a risk factor for bleeding during OAT [41]; however, a peptic ulcer disease without previous hemorrhaging has not been associated with a higher risk of bleeding [42]. In some patients, the occurrence of VTE is associated with a condition of malignancy. These patients should be treated with warfarin for an undefined period to reduce the risk of recurrence even though their risk of bleeding during OAT is markedly higher than in patients without cancer. A recent study [43] compared the outcome of anticoagulation courses in 95 patients with malignancy versus 733 patients without malignancy. All patients were participants in a large, nation-wide population study and were prospectively followed from the initiation of their oral anticoagulant therapy. The rates of major (5.4% vs 0.9%), minor (16.2% vs 3.6%), and total (21.6% vs 4.5%) bleeding were statistically significantly higher in cancer patients than in noncancer patients. Bleeding was also a more frequent cause of early anticoagulation withdrawal in patients with malignancy (4.2% vs 0.7%; p<0.01; RR 6.2, 95% CI 1.95–19.4). In the group of patients with cancer, the bleeding rate was high across the different INR categories and was independent of the temporally associated INR. In contrast, in the group of patients without cancer the bleeding rate increased only in cases of INR values over 4.5. Similar results have recently been reported by other authors [44, 45]. Except in patients anticoagulated for VTE, who are often treated only by OAT, many other patients, especially the elderly, have multiple drug therapies. It is known that a

Thrombosis in clinical practice

74

higher frequency of bleeding is associated with the concomitant use of aspirin or other antiplatelet drugs [46] and of non-steroidal antiinflammatory drugs [47]. Presence of lesions or injuries related to the site of bleeding It has long been recognized that the presence of organic lesions, already known or still latent, is an important factor for bleeding [48]. Important remediable lesions or injuries related to the site of bleeding were detected in 42% of patients with major bleeding and in 34% of those with minor bleeding; the bleeding event led to the discovery of a remediable lesion (especially in the gastrointestinal or urogenital tracts) in 17% of patients [7]. It is more likely to diagnose previously unknown lesions when bleeding occurs at therapeutic or subtherapeutic INR values. Most frequently lesions accounting for anticoagulant-related bleeding are found in the gastrointestinal or genitourinary tracts and soft tissues [8, 16, 21, 49–51]. The concomitant use of non-steroidal antiinflammatory drugs markedly increases the risk of bleeding in the gastrointestinal tract [47]. Compliance Most authors, though not all [52], believe that noncompliance may be a problem in the elderly [53]. It has been demonstrated [54] that a reduction in mental ability/attention levels can be found in a fairly significant proportion of elderly anticoagulated patients monitored in an anticoagulation clinic. Such a condition, which seems to be more frequent with age, affects the quality of anticoagulant therapy by increasing the period of either under- or overanticoagulation and exposing the patients to a higher risk of failure or bleeding complications. We, therefore, consider it wise to recommend that anticoagulation clinics carefully assess the mental ability of elderly patients before starting OAT. Time course of anticoagulant treatment A higher frequency of bleeding early in the course of OAT has been reported in many studies [6, 16, 21, 23] but not all [49, 55]. In the ISCOAT study [16], more than one-third of all the bleeding episodes occurred within the first 90 days of each anticoagulant course, the incidence of bleeding stabilizing thereafter. The rate of hemorrhagic events during the first 90 days of treatment was as high as 11% patient-years, decreasing considerably thereafter (6.3% patient-years). In an analysis of patients with deep vein thrombosis who had been discharged from California hospitals, the incidence of rehospitalization for bleeding complications during 91 days of treatment with warfarin was 1.4%; however, the rate was 2.7 times greater during the first 30 days than afterwards [56]. Several factors may contribute to the increased risk of bleeding in the early period of anticoagulant courses. First, anticoagulant therapy can unmask a cryptic lesion, and second, the dosage adjustment of therapy may be less well controlled at the start of treatment. As clearly pointed out by Landefeld and Goldman [6], studies that examine

Bleeding risk and hemorrhage

75

noninception cohort patients and/or include patients who have resumed OAT for a second course after an interval period are likely to underestimate the true risk of bleeding by either missing early events or excluding from any second course patients who had bled in the first course. Quality of anticoagulation control The quality of monitoring anticoagulated patients is certainly an important factor affecting the risk of bleeding complications. It is a general experience, confirmed by some studies [57, 58], that the rate of bleeding is lower when patients are monitored by dedicated anticoagulation clinics. In the dedicated clinics, the special training and experience of medical/paramedical staff, proper patient education, the use of computer programs, and various other factors all help ensure optimization of anticoagulant therapy. Proper monitoring of OAT calls not just for commitment on the part of doctors and nursing staff but for patient involvement too and this is only possible by offering wellthought-out health education. Only by doing this will it be possible to optimize anticoagulant therapy with the aim of preventing the onset of new thrombi or the spread of preexisting ones, with the minimum number of complications, especially hemorrhagic. Is it possible to predict the risk of bleeding in outpatients treated with warfarin? Landefeld et al. [59] proposed a bleeding risk index to estimate the risk of major bleeding in hospitalized patients starting anticoagulant therapy. The index was based on four major predictors: number of comorbid conditions, age above 59 years, supratherapeutic anticoagulation, and worsening liver function. In a study by van der Meer et al. [60], age and achieved INR were the most important and consistent risk factors for major bleeding. The authors proposed two different indexes based on these two risk factors to assess the risk of major bleeding in individual subjects. An index (the Outpatient Bleeding Risk Index), based on four independent risk factors for major bleeding: age 65 years or greater; history of gastrointestinal bleeding; history of stroke; and one or more of four specific comorbid conditions, has been proposed and validated by Beyth et al. [41]. It has recently been confirmed that the Outpatient Bleeding Risk Index is valid to discriminate between low- and moderate-risk patients [61]. Other authors have proposed and validated a “bleeding risk prediction score” in a population of patients treated for a previous VTE [42]. The score was based on the following variables and their odds ratios (score=(1.6×age)+(1.3×sex)+(2.2× malignancy)). A score of 3 or more points, 1–3 points, or 0 points represented a high, intermediate, or low bleeding risk, respectively. The application in clinical practice of such indexes and scores may help the clinical decision making when determining the risk-benefit of starting oral anticoagulation in candidate patients or deciding the optimal duration of anticoagulant therapy in patients with previous thrombotic events.

Unfractionated heparin (UFH) and low-molecular-weight heparins (LMWHs)

Thrombosis in clinical practice

76

Heparin and its derivatives have a direct anticoagulant effect by inhibiting thrombin and activated factor X. Low doses UFH or LMWHs are given for primary prevention of VTE in conditions at risk. Higher doses of UFH or LMWHs are the standard treatment in the initial phase of acute VTE. UFH is usually given via continuous venous infusion at doses to prolong the activated partial thromboplastin time (APTT) to 1.5–2.5 times the normal value. Due to their more predictable and less variable effect on the different patients and in the same patient over time LMWHs are given subcutaneously at fixed doses in relation to the body weight. Very high doses of UFH are used during open heart surgery. Bleeding associated with heparin prophylaxis of venous thromboembolism Collins et al. [62] in their meta-analysis of the effects of prophylaxis with UFH in surgical patients reported rates of 5.8% (vs 3.8% in controls) in general surgery, 3.5% (vs 2.9%) in orthopedic surgery, and 15.2% (vs 5.1%) in urologic surgery. However, Nurmohamed et al. [63], examining studies with stronger methodology, reported rates of major bleeding of 1.7% with LMWH, of 1.3% with UFH in general surgery, and of 1.5% either with LMWH or UFH in orthopedic surgery. In a more recent meta-analysis of randomized trials on prophylaxis in internal medicine, Mismetti et al. [64] reported rates of 0.4% and 1.2% of major bleeding with LMWH and UFH, respectively. A rate of 1.9% of major bleeding with UFH or LMWH has been calculated in a meta-analysis of studies on prophylaxis in neurosurgery [65]. Bleeding associated with heparin therapy In recent reviews of clinical trials of acute VTE treatment [14, 66, 67], the rates of major bleeding events ranged from 0% to 7% and those of fatal bleeding from 0% to 2% during intravenous UFH administration, and from 0 to 3% for major bleeding and from 0% to 0.8% for fatal bleeding during LMWHs treatment. When UFH was administered in acute ischemic stroke, the rates of bleeding events ranged from 1.1% to 3.2% for major and from 0.4% to 0.5% for fatal [14]. Several studies have recently investigated the possible advantages of using LMWH instead of warfarin for a prolonged secondary prevention of VTE. In a meta-analysis of the available studies, Marchetti et al. [68] calculated rates of 0.73% and 1.4% for major bleeding and of 0.24% and 6.8% for minor bleeding for LMWH and warfarin, respectively. Risk factors for bleeding during heparin treatment The occurrence of bleeding during heparin therapy increases with dose [69], Even though there are evidences that serious bleeding during heparin treatment may occur when the anticoagulant response is within the therapeutic values, it is widely admitted that there is an association between the incidence of bleeding and the anticoagulant effect of heparin as expressed by an excessive APTT prolongation [70]. The risk of bleeding is higher if UFH at therapeutic doses is administered with intermittent than continuous intravenous

Bleeding risk and hemorrhage

77

infusion; no differences were reported between subcutaneous and continuous intravenous heparin administration [71]. An increased risk of bleeding during heparin treatment in females has been reported [72]. Higher heparin levels and higher APTT values in females than in males were recorded after the same initial doses of UFH infusion in patients treated for proximal deep vein thrombosis [73]. After achieving therapeutic APTT values, the women received significantly lower heparin doses than the men. In that study, however, the women did not suffer of more bleeding events than men; it is likely that this was attributable to the accurate and rapid adjustment of heparin doses according to the APTT values. In a recent study on LMWH treatment in patients with unstable coronary artery disease, minor bleeding episodes were more frequent in women than in men, either when LMWH doses were weight-adjusted or during a fixed-dose treatment [74]. An increase in total and major bleeding complications with aging has been reported [75]. In that study, aging was associated with a tendency to an increase in activated partial thromboplastin time after standard heparin doses and with lower heparin dose requirements. Bleeds occur more frequently in patients with recent surgery, trauma, invasive procedures, or concomitant hemostatic defects [76], Hull and colleagues [77] showed that the rate of bleeding in patients with acute VTE treated with a continuous heparin infusion at the starting dose of 40,000 U was 1.1%; however, the incidence was as high as 10.8% in those with risk factors such as recent surgery or trauma, even though they received only 30,000 U. The risk of bleeding increases also when heparin treatment is associated with aspirin administration [78], thrombolytic therapy [71], or treatment with antagonists of glycoprotein IIb/IIIa [79].

Hemorrhagic adverse effects of antiplatelet agents The inhibition of platelet function is the basis of its antithrombotic effect of antiplatelet agents; however, it also produces a mild hemostatic defect with an increase in bleeding risk. The mild hemostatic defect can produce a slight increase in mucocutaneous bleeding (e.g. increase in bruising). The risk of major bleeding associated with antiplatelet agents is difficult to estimate as it is low in individual trials (<1%/year). The Antithrombotic Trialists’ Collaboration [80] conducted a meta-analysis by pooling the results of 287 studies involving 135,000 patients in comparison of antiplatelet therapy versus control and 77,000 in comparisons of different antiplatelet regimens. The pooled results indicated an overall rate of major extracranial bleeds of 1.13% (595/47, 158) in the antiplatelet regimen group and a rate of 0.71% (333/47, 168) with an increase in risk of 1.6 (odds ratio 95% CI 1.4–1.8). Among the major bleeds, 159 were fatal (20%). No significant difference was observed between the five high-risk categories considered (previous myocardial infarction, acute myocardial infarction, previous stroke/transient ischemic attack, acute stroke, other high risk). The absolute excess of intracranial hemorrhage due to aspirin therapy is less than 1 per 1,000 patients per year in high-risk patients, with a somewhat higher risk in patients with cerebrovascular disease.

Thrombosis in clinical practice

78

The antithrombotic effect of acetyl salicylic acid (ASA) or aspirin is dose-independent in the range from 30 to 1,300 mg o.d. in relation to the almost complete thromboxane A2 inhibition at low doses [81]. On the contrary, to the inhibition of cyclooxygenase (COX) of gastroenteric toxicity is dose-dependent, due gastric mucosa and to the direct chemical toxicity. Clinical and endoscopic data indicate that the gastroenteric toxicity of ASA is dosedependent and increases with doses of 900–1,500 mg when compared to doses equal to or lower than 300 mg. The COX activity of gastric mucosa cells is not significantly influenced by a dose of 100 mg which produces a complete inhibition of platelet COX. As a result, the antithrombotic effect of ASA can be dissociated in part from gastroenteric toxicity, although major bleedings have been observed with 30–50 mg doses. Preexisting mucosal lesions due to the use of other NSAIDS and/or the presence of Helicobacter pylori infection can contribute to the increase in the risk of bleeding regardless of the dose of ASA. The relative risk of hospitalization for bleeding and/or gastroenteric perforations due to ASA (100–300 mg o.d.) is similar to that of other antiplatelet agents and it is 2.3-fold that observed in the general population as indicated by an observational study [82]. The benefit-risk ratio of antiplatelet agents therapy depends on the absolute risk of thrombotic and hemorrhagic events in the individual patient. In patients with an absolute low thrombotic risk (<1%/year), the small absolute benefit can be offset by the exposure of many subjects to a hemorrhagic risk. On the contrary, if the thrombotic risk is high (>5%/year), the absolute benefit of ASA increases significantly. For example, the absolute excess of major bleedings in acute myocardial infarction is approximately 100fold lower than the absolute number of major vascular events prevented by ASA.

References [1] Fihn SD, Callahan CM, Martin DC, Mcdonell MB, Henikoff JG, White RH. The risk for and severity of bleeding complications in elderly patients treated with warfarin. Ann Intern Med 1996; 124:970–9. [2] Graafsma YP, Prins MH, Lensing AWA, Dehaan RJ, Huisman MV, Buller HR. Bleeding classification in clinical trials: observer variability and clinical relevance. Thromb Haemost 1997; 78:1189–92. [3] Gurwitz JH, Goldberg RJ, Holden A, Knapic N, Ansell J. Age-related risks of long-term oral anticoagulant therapy. Arch Intern Med 1988;148: 1733–6. [4] Van der Meer FJM, Rosendaal FR, Vandenbroucke JP, Briet E. Bleeding complications in oral anticoagulant therapy—an analysis of risk factors. Arch Intern Med 1993; 153:1557–62. [5] Pocock SJ, Elbourne DR. Randomized trials or observational tribulations? N Engl J Med 2000; 342:1907–9. [6] Landefeld CS, Goldman L. Major bleeding in outpatients treated with warfarin: incidence and prediction by factors known at the start of outpatient therapy. Am J Med 1989; 87:144–52. [7] Landefeld CS, Rosenblatt MW, Goldman L. Bleeding in outpatients treated with warfarin: relation to the prothrombin time and important remediable lesions. Am J Med 1989; 87:153–9. [8] Gitter MJ, Jaeger TM, Petterson TM, Gersh BJ, Phil D, Silverstein MD. Bleeding and thromboembolism during anticoagulant therapy: a population-based study in Rochester, Minnesota. Mayo Clin Proc 1995; 70:725–33. [9] Dahl T, Abildgaard U, Sandset PM. Long-term anticoagulant therapy in cerebrovascular disease: does bleeding outweigh the benefit? J Intern Med 1995; 237:323–9.

Bleeding risk and hemorrhage

79

[10] Heller RF, Knapp JC, Oconnell RL, Lim LLY, Carruthers AE, Fluit JH, Macdonald JJ, Mcgrath KM, Reeves GEM, Ryall MAE. Effectiveness of anticoagulation among patients discharged from hospital on warfarin. Med J Aust 1998; 169:243–6. [11] Launbjerg J, Egeblad H, Heaf J, et al. Bleeding complications to oral anticoagulant therapy: multivariate analysis of 1,010 treatment years in 551 outpatients. J Intern Med 1991; 229:351–5. [12] Hurlen M, Erikssen J, Smith P, Arnesen H, Rollag A. Comparison of bleeding complications of warfarin and warfarin plus acetyl salicylic acid: a study in 3166 outpatients. J Intern Med 1994; 236: 299–304. [13] Kearon C, Hirsh J. Current concepts: management of anticoagulation before and after elective surgery. N Engl J Med 1997; 336:1506–11. [14] Levine MN, Raskob G, Landefeld S, Kearon C. Hemorrhagic complications of anticoagulant treatment. Chest 2001; 119:108S–121S. [15] Cannegieter SC, Rosendaal FR, Briet E. Thromboembolic and bleeding complications in patients with mechanical heart valve prostheses. Circulation 1994; 89:635–41. [16] Palareti G, Leali N, Coccheri S, Poggi M, Manotti C, D’Angelo A, Pengo V, Erba N, Moia M, Ciavarella N, Devoto G, Berrettini M, Musolesi S. Bleeding complications of oral anticoagulant treatment: an inception-cohort, prospective collaborative study (ISCOAT). Italian Study on Complications of Oral Anti-coagulant Therapy. Lancet 1996; 348:423–8. [17] Jaffin BW, Bliss CM, Lamont JT. Significance of occult gastrointestinal bleeding during anticoagulation therapy. Am J Med 1987; 83:269–72. [18] Oden A, Fahlen M. Oral anticoagulation and risk of death: a medical record linkage study. BMJ 2002; 325:1073–5. [19] Hylek EM, Singer DE. Risk factors for intracranial hemorrhage in outpatients taking warfarin. Ann Intern Med 1994; 120:897–902. [20] Rosand J, Hylek EM, ODonnell HC, Greenberg SM. Warfarin-associated hemorrhage and cerebral amyloid angiopathy—a genetic and pathologic study. Neurology 2000; 55:947–51. [21] Fihn SD, Mcdonell M, Martin D, Henikoff J, Vermes D, Kent D, White RH. Risk factors for complications of chronic anticoagulation—a multicenter study. Ann Intern Med 1993; 118:511– 20. [22] Stroke prevention in atrial fibrillation investigators. Bleeding during antithrombotic therapy in patients with atrial fibrillation. Arch Intern Med 1996; 156:401–16. [23] Petitti DB, Strom BL, Melmon KL. Duration of warfarin anticoagulant therapy and the probabilities of recurrent thromboembolism and hemorrhage. Am J Med 1986; 81:255–9. [24] Pengo V, Legnani C, Noventa F, Palareti G. Oral anticoagulant therapy in patients with nonrheumatic atrial fibrillation and risk of bleeding—a multicenter inception cohort study. Thromb Haemost 2001; 85:418–22. [25] Scott PJW. Anticoagulant drugs in the elderly: the risks usually outweigh the benefits. BMJ 1988; 2:1261–3. [26] Levine MN, Hirsh J, Landefeld S, Raskob G. Hemorrhagic complications of anticoagulant treatment. Chest 1992; 102:352S–363S. [27] Beyth RJ, Landefeld CS. Anticoagulants in older patients: a safety perspective. Drugs & Aging 1995; 6:45–54. [28] Hutten BA, Lensing AWA, Kraaijenhagen RA, Prins MH. Safety of treatment with oral anticoagulants in the elderly—a systematic review. Drugs & Aging 1999; 14:303–12. [29] Sebastian JL, Tresch DD. Use of oral anticoagulants in older patients. Drugs & Aging 2000; 16:409–35. [30] Palareti G, Hirsh J, Legnani C, Manotti C, DAngelo A, Pengo V, Moia M, Guazzaloca G, Musolesi S, Coccheri S. Oral anticoagulation treatment in the elderly—a nested, prospective, casecontrol study. Arch Intern Med 2000; 160:470–8. [31] Russmann S, Gohlke-Barwolf C, Jahnchen E, Trenk D, Roskmann H. Age-dependent differences in the anticoagulant effect of phenprocoumon in patients after heart valve surgery. Eur J Clin Pharmacol 1997; 52:31–5.

Thrombosis in clinical practice

80

[32] Hart RG, Boop BS, Anderson DC. Oral anticoagulants and intracranial hemorrhage—facts and hypotheses. Stroke 1995; 26:1471–7. [33] Froom P, Miron E, Barak M. Oral anticoagulants in the elderly. Br J Haematol 2003; 120:526–8. [34] Beyth RJ, Quinn L, Landefeld CS. A multicomponent intervention to prevent major bleeding complications in older patients receiving warfarin—a randomized, controlled trial. Ann Intern Med 2000; 133:687–95. [35] Dittman WA, Majerus PW. Structure and function of thrombomodulin: a natural anticoagulant. Blood 1990; 75:329–36. [36] Jansson JH, Boman K, Brannstrom M, Nilsson TK. High concentration of thrombomodulin in plasma is associated with hemorrhage: a prospective study in patients receiving long-term anticoagulant treatment. Circulation 1997; 96:2938–43. [37] Torn M, Algra A, Rosendaal FR. Oral anticoagulation for cerebral ischemia of arterial origin: high initial bleeding risk. Neurology 2001; 57:1993–9. [38] Petty GW, Brown RD, Whisnant JP, Sicks JD, OFallon WM, Wiebers DO. Frequency of major complications of aspirin, warfarin, and intravenous heparin for secondary stroke prevention—a population-based study. Ann Intern Med 1999; 130:14–22. [39] Franke CL, Koehler PJJ, Gorter JW, Kappelle LJ, Rinkel GJE, Tjeerdsma HC, Vangijn J, Dammers JWHH, Straatman HJS, Tenhouten R, Veering MM, Bakker SLM, Dippel D, Koudstaal PJ, Vangemert HMA, Vanswieten JC, Horn J, Kwa IH, Limburg M, Stam J, Boon AM, Lieuwens WHG, Visscher F, Bouwsma C, Rutgers AWF, Snoek JW, Brouwers PJAM, Nihom J, Solleveld H, Carbaat PAT, Hertzberger LI, Kleijweg RP, Nanningavandenneste VMH, Vandiepen AJH, Linssen WHJP, Vanneste JAL, Vos J, Weinstein HC, Schipper JP, Vandermeer WK, Berntsen PJIM, Devriesleenders EM, Geervliet JP, Tans RJJ, Feikema WJ, Lohmann HJHM, Vankasteel V, Jongebloed FA, Leyten QH, et al. A randomized trial of anticoagulants versus aspirin after cerebral ischemia of presumed arterial origin. Ann Neurol 1997; 42:857–65. [40] ESPRIT. Oral anticoagulation in patients after cerebral ischemia of arterial origin and risk of intracranial hemorrhage. Stroke 2003; 34:e45–e46. [41] Beyth RJ, Quinn LM, Landefeld CS. Prospective evaluation of an index for predicting the risk of major bleeding in outpatients treated with warfarin. Am J Med 1998; 105:91–9. [42] Kuijer PMM, Hutten BA, Prins MH, Buller HR. Prediction of the risk of bleeding during anticoagulant treatment for venous thromboembolism. Arch Intern Med 1999; 159:457–60. [43] Palareti G, Legnani C, Lee A, Manotti C, Hirsh J, DAngelo A, Pengo V, Mola M, Coccheri S. A comparison of the safety and efficacy of oral anticoagulation for the treatment of venous thromboembolic disease in patients with or without malignancy. Thromb Haemost 2000; 84:805–10. [44] Hutten BA, Prins MH, Gent M, Ginsberg J, Tijssen JGP, Buller HR. Incidence of recurrent thromboembolic and bleeding complications among patients with venous thromboembolism in relation to both malignancy and achieved international normalized ratio: a retrospective analysis. J Clin Oncol 2000; 18:3078–83. [45] Prandoni P, Lensing AWA, Piccioli A, Bernardi E, Simioni P, Girolami B, Marchiori A, Sabbion P, Prins MH, Noventa F, Girolami A. Recurrent venous thromboembolism and bleeding complications during anticoagulant treatment in patients with cancer and venous thrombosis. Blood 2002; 100:3484–8. [46] Blackshear JL, Baker VS, Holland A, Litin SC, Ahlquist DA, Hart RG, Ellefson R, Koehler J. Fecal hemoglobin excretion in elderly patients with atrial fibrillation: combined aspirin and lowdose warfarin vs conventional warfarin therapy. Arch Intern Med 1996; 156:658–60. [47] Shorr RI, Ray WA, Daugherty JR, Griffin MR. Concurrent use of nonsteroidal antiinflammatory drugs and oral anticoagulants places elderly persons at high risk for hemorrhagic peptic ulcer disease. Arch Intern Med 1993; 153:1665–70.

Bleeding risk and hemorrhage

81

[48] Wilcox CM, Truss CD. Gastrointestinal bleeding in patients receiving long-term anticoagulant therapy. Am J Med 1988; 84:683–90. [49] Forfar JC. A 7-year analysis of hemorrhage in patients on long-term anticoagulant treatment. Br Heart J 1979; 42:128–32. [50] Landefeld CS, Beyth RJ. Anticoagulant-related bleeding—clinical epidemiology, prediction, and prevention. Am J Med 1993; 95:315–28. [51] White RH, Mckittrick T, Takakuwa J, Callahan C, Mcdonell M, Fihn S. Management and prognosis of life-threatening bleeding during warfarin therapy. Arch Intern Med 1996; 156:1197–201. [52] Arnsten JH, Gelfand JM, Singer DE. Determinants of compliance with anticoagulation: a casecontrol study. Am J Med 1997; 103:11–17. [53] Lau HS, Beuning KS, Postmalim E, Kleinbeernink L, Deboer A, Porsius AJ. Non-compliance in elderly people: evaluation of risk factors by longitudinal data analysis. Pharmacy World & Science 1996; 18:63–8. [54] Palareti G, Poggi M, Guazzaloca G, Savino A, Coccheri S. Assessment of mental ability in elderly anticoagulated patients: its reduction is associated with a less satisfactory quality of treatment. Blood Coagul Fibrinolysis 1997; 8:411–17. [55] Lundstrom T, Ryden L. Hemorrhagic and thromboembolic complications in patients with atrial fibrillation on anticoagulant prophylaxis. J Intern Med 1989; 225:137–42. [56] White RH, Beyth RJ, Zhou H, Romano PS. Major bleeding after hospitalization for deepvenous thrombosis. Am J Med 1999; 107:414–24. [57] Cortelazzo S, Finazzi G, Viero P, Galli M, Remuzzi A, Parenzan L, Barbui T. Thrombotic and hemorrhagic complications in patients with mechanical heart valve prosthesis attending an anticoagulation clinic. Thromb Haemost 1993; 69:316–20. [58] Chiquette E, Amato MG, Bussey HI. Comparison of an anticoagulation clinic with usual medical care: anticoagulation control, patient outcomes, and health care costs. Arch Intern Med 1998; 158:1641–7. [59] Landefeld CS, Mcguire E, Rosenblatt MW. A bleeding risk index for estimating the probability of major bleeding in hospitalized patients starting anticoagulant therapy. Am J Med 1990; 89:569–78. [60] Van der Meer FJM, Rosendaal FR, Vandenbroucke JP, Briet E. Assessment of a bleeding risk index in two cohorts of patients treated with oral anticoagulants. Thromb Haemost 1996; 76:12– 16. [61] Wells PS, Forgie MA, Simms M, Greene A, Touchie D, Lewis G, Anderson J, Rodger MA. The outpatient bleeding risk index: validation of a tool for predicting bleeding rates in patients treated for deep venous thrombosis and pulmonary embolism. Arch Intern Med 2003; 163:917– 20. [62] Collins R, Scrimgeour A, Yusuf S, Phil D, Peto R. Reduction in fatal pulmonary embolism and venous thrombosis by perioperative administration of subcutaneous heparin—overview of results of randomized trials in general, orthopedic, and urologic surgery. N Engl J Med 1988; 318:1162–73. [63] Nurmohamed MT, Rosendaal FR, Buller HR, Dekker E, Hommes DW, Vandenbroucke JP, Briet E. Low-molecular-weight heparin versus standard heparin in general and orthopaedic surgery—a meta-analysis. Lancet 1992; 340:152–6. [64] Mismetti P, Laporte-Simitsidis S, Tardy B, Cucherat M, Buchmuller A, Juillard-Delsart D, Decousus H. Prevention of venous thromboembolism in internal medicine with unfractionated or low-molecularweight heparins: a meta-analysis of randomised clinical trials. Thromb Haemost 2000; 83:14–19. [65] Iorio A, Agnelli G. Low-molecular-weight and unfractionated heparin for prevention of venous thromboembolism in neurosurgery—a metaanalysis. Arch Intern Med 2000; 160:2327– 32.

Thrombosis in clinical practice

82

[66] Gould MK, Dembitzer AD, Sanders GD, Garber AM. Low-molecular-weight heparins compared with unfractionated heparin for treatment of acute deep venous thrombosis. A costeffectiveness analysis [see comments]. Ann Intern Med 1999; 130:789–99. [67] Dolovich LR, Ginsberg JS, Douketis JD, Holbrook AM, Cheah G. A meta-analysis comparing low-molecular-weight heparins with unfractionated heparin in the treatment of venous thromboembolism—examining some unanswered questions regarding location of treatment, product type, and dosing frequency. Arch Intern Med 2000;160:181–8. [68] Marchetti M, Pistorio A, Barone M, Serafini S, Barosi G. Low-molecular-weight heparin versus warfarin for secondary prophylaxis of venous thromboembolism: a cost-effectiveness analysis. Am J Med 2001; 111:130–9. [69] Morabia A. Heparin doses and major bleedings. Lancet 1986; 1:1278–9. [70] Levine MN, Raskob G, Landefeld S, Kearon C. Hemorrhagic complications of anticoagulant treatment. Chest 1998; 114:S511–S523. [71] Levine MN, Raskob G, Landefeld S, Hirsh J. Hemorrhagic complications of anticoagulant treatment. Chest 1995; 108:S276–S290. [72] Jick H, Slone D, Borda IT, Shapiro S. Efficacy and toxicity of heparin in relation to age and sex. N Engl J Med 1968; 279:284–6. [73] Campbell NRC, Hull RD, Brant R, Hogan DB, Pineo GF, Raskob GE. Different effects of heparin in males and females. Clin Invest Med 1998; 21:71–8. [74] Toss H, Wallentin L, Siegbahn A. Influences of sex and smoking habits on anticoagulant activity in low-molecular-weight heparin treatment of unstable coronary artery disease. Am Heart J 1999; 137:72–8. [75] Campbell NRC, Hull RD, Brant R, Hogan DB, Pineo GF, Raskob GE. Aging and heparinrelated bleeding. Arch Intern Med 1996; 156:857–60. [76] Landefeld CS, Cook EF, Flatley M, Weisberg M, Goldman L. Identification and preliminary validation of predictors of major bleeding in hospitalized patients starting anticoagulant therapy. Am J Med 1987; 82:703–13. [77] Hull RD, Raskob GE, Rosenbloom D, Panju AA, Brill-Edwards P, Ginsberg JS, Hirsh J, Martin GJ, Green D. Heparin for 5 days as compared with 10 days in the initial treatment of proximal venous thrombosis. N Engl J Med 1990; 322:1260–4. [78] Sethi GK, Copeland JG, Goldman S, Moritz T, Zadina K, Henderson WG. Implications of preoperative administration of aspirin in patients undergoing coronary artery bypass grafting. Department of Veterans Affairs Cooperative Study on Antiplatelet Therapy. J Am Coll Cardiol 1990; 15:15–20. [79] Lincoff AM, Bittl JA, Harrington RA, Feit F, Kleiman NS, Jackman JD, Sarembock IJ, Cohen DJ, Spriggs D, Ebrahimi R, Keren G, Carr J, Cohen EA, Betriu A, Desmet W, Kereiakes DJ, Rutsch W, Wilcox RG, deFeyter PJ, Vahanian A, Topol EJ. Bivalirudin and provisional glycoprotein IIb/IIIa blockade compared with heparin and planned glycoprotein IIb/IIIa blockade during percutaneous coronary intervention—REPLACE-2 Randomized Trial. JAMA 2003; 289:853–63. [80] Baigent C, Sudlow C, Collins R, Peto R. Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. BMJ 2002; 324:71–86. [81] Patrono C, Coller B, Dalen JE, FitzGerald GA, Fuster V, Gent M, Hirsh J, Roth G. Plateletactive drugs—the relationships among dose, effectiveness, and side effects. Chest 2001; 119:39S–63S. [82] Garcia-Rodriguez LA, Cattaruzzi C, Troncon MG, Agostinis L. Risk of hospitalization for upper gastrointestinal tract bleeding associated with ketorolac, other nonsteroidal antiinflammatory drugs, calcium antagonists, and other antihypertensive drugs. Arch Intern Med 1998; 158:33–9.

Bleeding risk and hemorrhage

83

[83] Hull R, Hirsh J, Jay R, Carter C, England C, Gent M, Turpie AGGG, McLoughlin D, Dodd P, Thomas M, Raskob G, Ockelford P. Different intensities of oral anticoagulant therapy in the treatment of proximal-vein thrombosis. N Engl J Med 1982; 307:1676–81. [84] Ezekowitz MD, Bridgers SL, James KE, Carliner NH, Colling CL, Gornick CC, Krausesteinrauf H, Kurtzke JF, Nazarian SM, Radford MJ, Rickles FR, Shabetai R, Deykin D. Warfarin in the prevention of stroke associated with nonrheumatic atrial fibrillation. N Engl J Med 1992; 327:1406–12. [85] Isaacs C, Paltiel O, Blake G, Beaudet W, Conochie L, Leclerc J. Age-associated risks of prophylactic anticoagulation in the setting of hip fracture. Am J Med 1994; 96:487–91. [86] Koopman MMW, Prandoni P, Piovella F, Ockelford PA, Brandjes DPM, Vandermeer J, Gallus AS, Simonneau G, Chesterman CH, Prins MH, Bossuyt PMM, Dehaes H, Vandenbelt AGM, Sagnard L, Dazemar P, Buller HR. Treatment of venous thrombosis with intravenous unfractionated heparin administered in the hospital as compared with subcutaneous lowmolecular-weight heparin administered at home. N Engl J Med 1996; 334: 682–7. [87] Petitti DB, Strom BL, Melmon KL. Prothrombin time ratio and other factors associated with bleeding in patients treated with warfarin. J Clin Epidemiol 1989; 42:759–64. [88] Steffensen FH, Kristensen K, Ejlersen E, Dahlerup JF, Sorensen HT. Major haemorrhagic complications during oral anticoagulant therapy in a Danish population-based cohort. J Intern Med 1997; 242: 497–503.

Thrombosis in clinical practice

84

5 Thrombophilia and venous thrombosis Pieter W Kamphuisen, Harry R Büller, and Frits R Rosendaal Introduction Before 1993, an inherited risk factor was detectable in only 10% of symptomatic patients with venous thrombosis. In the last 10 years, the knowledge of risk factors for venous thrombosis has increased significantly. With the discovery of several inherited coagulation abnormalities associated with an increased tendency for venous thrombosis, such as factor V Leiden and the prothrombin 20210A mutation, many patients with a first episode of venous thrombosis have a detectable disorder. The idea that clotting of the blood might lead to the development of thrombosis goes back to old Chinese medicine. The quote “When it coagulates within the pulse, the blood ceases to circulate beneficially; when the blood coagulates within the foot, it causes pain and chills” is attributed to Huang Ti, a physician of around 2650 BC, reviewed in [1], It took a long time before the pathogenesis of thrombosis was better understood. In 1686, Wiseman thought that thrombosis might be fostered by stasis of the blood flow and by a change of the nature of the blood itself [2], Rudolph Virchow further elucidated the concept of how coagulation abnormalities lead to the development of thrombosis. In 1856, he stated that the development of thrombosis was the result of changes in blood composition (hypercoagulability), reduced blood flow, or changes in the vessel wall [3]. Disturbance of this balance favors fibrin formation and may ultimately lead to the formation of occlusive thrombi. Examples of such pathophysiologic phenomenon are trauma, immobilization, pregnancy, surgery, malignancy, and infection. These are acquired risk factors for venous thrombosis that may cause tissue damage, stasis of the blood, or changes in blood composition. Although a familial tendency for venous thrombosis was observed about a century ago, it took a long time before causes of familial thrombosis were discovered, probably due to insufficient knowledge of the function of hemostasis and fibrinolysis. Insight into the balance of thrombosis and hemostasis was a prerequisite for the investigation of thrombophilia factors. Subsequently, family studies and case-control studies led to important discoveries of heritable causes of thrombosis.

1958: Thrombohemorrhagic balance In 1958, Åstrup stated that a balance exists between the process of blood clotting and the removal of this process by fibrinolysis [4]. The balance between the effects of these two

Thrombophilia and venous thrombosis

85

processes then regulates the amount of fibrin. Local hemostasis depends upon the balance between the blood clotting system and the fibrinolytic system in the wounded tissue.

1965–81: Anticoagulant deficiencies The first discoveries of hereditary thrombophilia involved deficiencies of the natural anticoagulants antithrombin, protein C, and protein S. It was in 1965 that Egeberg described a family in which several members were affected by thrombosis that was attributed to antithrombin deficiency [5]. This was the first report of an inherited cause of familial thrombosis, also called inherited thrombophilia. Since then, numerous families with antithrombin deficiency and a large number of mutations have been described [6–8]. The phenotype of antithrombin deficiency can be severe with thrombosis occurring at a young age [7, 8]. In 1981, Griffin et al. described a family with low protein C levels and a tendency to venous thrombosis [9]. Subsequent family studies confirmed the increased risk of heterozygous protein C deficiency, with most carriers experiencing thrombosis before middle age [10, 11]. There was no apparent difference between the phenotype of thrombosis and type of mutation [12]. Protein S deficiency as a cause of thrombosis was first published in 1984 by Comp and Esmon [13] and by Schwartz et al. [14]. In families with protein S deficiency and thrombosis, the majority of cases had developed venous thrombosis before the age of 40 years [15]. Although the discovery of these three anticoagulant proteins contributed importantly to the knowledge of the pathophysiology of thrombosis, two main problems arose: first, these three single gene defects are rare in the general population and even explained only about 10% of the thrombotic events among families with a predisposition of thrombosis, necessitating extensive research for other causes [16]. Second, the families that were investigated were highly selected with several members affected by thrombosis. Obviously, the estimate of thrombosis risk was too high for the general population and probably reflected the clustering of several additional defects, like factor V Leiden [17]. The results from these families cannot be extrapolated to unselected individuals. An illustrative example of this type of selection is the age of onset. The mean age of a first thrombosis was 30 years for individuals with familial thrombophilia compared to 45 years among unselected patients [17]. This difference was independent of the type of thrombophilia, such as protein C deficiency or factor V Leiden, illustrating the influence of patient selection on the clinical phenotype. This phenomenon of selection was also shown by Miletich who found that thrombosis was uncommon in healthy blood donors and their family members who had a deficiency of protein C [18]. This implied that a deficiency of protein C was probably not sufficient to develop thrombosis. In retrospect, in the earlier described thrombophilic families in which protein C deficiency co-segregated, thrombosis was also frequently found in family members who had normal protein C levels. This suggested other inherited abnormalities and supported the concept of interaction of risk factors in a multicausal disease. It was clear that population-based studies of thrombosis were necessary to solve this controversy.

Thrombosis in clinical practice

86

1988–92: Population-based case-control studies—the Leiden Thrombophilia Study In 1988, the Leiden Thrombophilia Study (LETS) was designed as a large case-control study to investigate various risk factors for venous thrombosis at the population level [19]. This study was the first with sufficient power to investigate risk factors and had in contrast to previous studies rigid criteria to minimize case-control design was chosen to study selection bias or misclassification. The genetic influences since these mutations are not subject to selection. In addition, only consecutive patients with a first objectively confirmed deep vein thrombosis were selected. Each patient was asked to find a healthy control subject (not related) of the same sex and approximately the same age (interval of 5 years). All subjects completed a standard questionnaire about the presence of acquired risk factors prior to the thrombotic event. From 1988 to 1992, a total of 474 consecutive patients with a first thrombosis

Table 5.1 Results from the Leiden Thrombophilla Study. Risk factor

Prevalence in patients (%)

Prevalence in controls (%)

OR 95% CI

Anticoagulant proteins Protein C <0.67 U/ml

4.6

0.8

3.8 1.7– 7.0

Protein S <0.67 U/ml

1.1

1.3

0.8 0.2– 3.0

Antithrombin <0.80 U/ml

1,1

0.2

5.0 0.7–34

Factor V Leiden mutation

19

3

7.9 4.4–14

Prothrombin 20210A mutation

6.2

2.3

2.8 1.4– 5.6

Factor VIII >150 IU/dl

25

11

6.2 3.4–11

Factor IX >129 U/dl

20

10

2.5 1.6– 3.9

Factor XI >120.8%

19

10

2.2 1.5– 3.2

TAFI >122 U/dl)

17

10

1.7 1.1– 2.5

Protein C inhibitor >125.5%

13

10

1.4 0.9–

Prothrombotic mutations

Elevated levels of procoagulant factors

Fibrinolytic factors

Thrombophilia and venous thrombosis

87

2.0 Other laboratory abnormalities Homocysteine >18.5 µmol/1

10

5

2.5 1.2– 5.2

APC resistance (without factor V Leiden) <0.92

36

16

4.4 2.9– 6.6

and 474 controls were included. Table 5.1 summarizes the main results of the LETS. First, the prevalence of low levels of protein C, protein S, and antithrombin was investigated [20]. The relative risk associated with protein C levels less than 0.67 U/ml was 3.8 and increased to 6.5 in the presence of a mutation. The prevalence of protein C deficiency among patients with a first episode of thrombosis was 3% compared to 0.8% of the controls. Surprisingly, a low protein S level was not associated with an increase of venous thrombosis, as measured by total protein S less than 0.67 U/ml (odds ratio 0.8) or as free protein S less than 0.57 U/ml (odds ratio 1.6). Among thrombosis patients, the prevalence of low total protein S levels was 1.1% and of low free protein S levels were found in 3.1% of the cases. These numbers were 1.3% and 2.1% respectively among the controls. Faioni et al. found a relative risk of 2.4 (95% CI 0.8–7.9) associated with low protein S levels in a population-based case-control study, showing that at population level, these levels lead only to a mild increase in thrombosis risk [21]. Antithrombin deficiency (antithrombin level <0.80 U/ml) was found in 1.1% of the thrombosis patients and in 0.2% of the controls (odds ratio 5.0), which was in agreement with other studies [22, 23]. The population-based studies of unselected patients with thrombosis made clear that the natural anticoagulant proteins are very rare both in patients and among controls.

1993–2003: Procoagulant excess This decade turned out to be the era of the procoagulant excess activity in thrombosis research. The discovery of common thrombophilia factors resulted in a better understanding of the pathophysiology of thrombosis and underlined the importance of the thrombohemorrhagic balance. Interest became focused on the procoagulant side of the hemostatic balance. Resistance to activated protein C was first described by Dahlbäck in 1993 [24]. Bertina et al. discovered the factor V Leiden mutation, the underlying mutation in most cases, in 1994 [25]. In the LETS, factor V Leiden was found in 19% of the thrombosis patients and in 3% of the controls, leading to an odds ratio of 7.9 [26]. Homozygous carriers of factor V Leiden have an estimated 80-fold increased risk [26]. More than half of the probands of selected families with familial thrombosis have factor V Leiden [27]. Therefore, factor V Leiden is the most prevalent hereditary thrombophilia factor, at least among Caucasians [28]. Among native populations in Africa, Southeast Asia, and Australia it is extremely uncommon.

Thrombosis in clinical practice

88

In 1995, it was shown that elevated factor VIII: C levels increased the risk of venous thrombosis [29]. Since hemophilia A (low factor VIII levels) is associated with an impaired clotting process, the hypothesis was raised that an excess of the procoagulant factor VIII could lead to an increased clot formation. Indeed, factor VIII levels above 150 IU/dl (the upper approximate quartile) were associated with a 6-fold increase compared to those people with levels below 100 IU/dl [29]. Compared with levels below 100 IU/dl, people with levels above 150 IU/dl have a 6.2-fold increased risk. These high levels as measured after the thrombotic event were in the majority persistent and not affected by acute phase reactions. In addition, factor VIII levels above 150 IU/dl were highly prevalent: 25% of the thrombosis patients and 11 % of the controls had such high levels [29]. These findings were confirmed by several other population-based studies [30–32]. In 1994, Falcon et al. found a high prevalence of high plasma levels of homocysteine among patients with thrombosis [33]. In the LETS in 1996, homocysteine levels above 18.5 µmol/l were found in 5% of the control population and were associated with a 2-fold increased risk [34]. In 1996, another important genetic thrombophilia factor was discovered: the prothrombin 20210A mutation [35]. A G to A transition in the 3′ untranslated region was found to be associated both with increased prothrombin levels and with the occurrence of venous thrombosis. In the LETS population, the prevalence of this mutation was 6.2% in the patients and 2.3% in the healthy controls, yielding an odds ratio of 2.8 [35]. Among selected thrombophilia families, this mutation was found in 18% of the family members [35]. In addition, increased prothrombin levels in patients with the wild-type prothrombin were also associated with an increased thrombosis risk [35]. The relation between the prothrombin mutation and venous thrombosis was confirmed by other groups [36–38]. Since the response to APC seemed to vary within each genotype of factor V Leiden, an altered response to APC in the absence of factor V Leiden could also determine the risk of thrombosis. In the LETS, a normalized APC sensitivity ratio (n-APC-SR) lower than 0.92, in the absence of factor V leiden, was associated with a 2.9-fold increased risk when compared to levels above this value [39]. Some 16% of the controls and 36% of the patients had a low n-APC-SR. The concept that high levels of procoagulant factors could lead to thrombosis, since a deficiency of these factors is associated with bleeding, led to the discovery of two other new risk factors. Since factor VIII is the cofactor of factor IX in the activation of factor X, it was hypothesized that elevated factor IX could increase the thrombosis risk. Factor IX levels greater than 129 U/dl (90th percentile in control subjects) doubled the risk of thrombosis compared to individuals with levels below this cut-off point [40]. Interestingly, the combination of high levels of both factor VIII and factor IX was associated with the highest risk of deep vein thrombosis (OR 8.2, 95% CI 3.6–18.4). Factor XI, a component of the intrinsic pathway of coagulation, contributes to the generation of thrombin, which is involved in both the formation of fibrin and protection against fibrinolysis. Factor XI levels above 120.8% (90th percentile in control subjects) were associated with a 2.2-fold increased risk of thrombosis as compared with lower levels [41]. In the beginning of this century, thrombinactivatable fibrinolysis inhibitor (TAFI) levels above the 90th percentile of the controls (>122 U/dl) increased the risk for thrombosis nearly 2-fold compared with TAFI levels below the 90th percentile (odds

Thrombophilia and venous thrombosis

89

ratio 1.7) [42]. Such levels were found in 10% of healthy controls and in 14% of patients with a first episode of deep vein thrombosis. Elevated TAFI levels did not enhance the thrombotic risk associated with factor V Leiden but seemed to interact with high factor VIII levels. High levels of protein C inhibitor, a serine proteinase inhibitor that can inhibit coagulation and fibrinolysis enzymes, were associated with a mild increased risk of thrombosis [43]. Several coagulation factors that were investigated in the LETS had a less clear association with venous thrombosis. High factor VII [44] and factor V levels [45] were unrelated to venous thrombosis, whereas high factor X [46] and fibrinogen levels [44] turned out to be mild risk factors at most. In conclusion, the first abnormalities in the clotting system that were found to be associated with an increased risk of venous thrombosis were deficiencies of the anticoagulant proteins antithrombin, protein C, and protein S. These loss of function mutations are rare in the general population. The prothrombotic abnormalities have a gain of function through subtle changes in the regulation of the gene activity. Factor V Leiden is relatively resistant to inactivation by APC and the prothrombin mutation leads to increased prothrombin levels. High levels of procoagulant factors, such as factor VIII, factor IX, and factor XI lead to prolonged formation of fibrin as a result of excessive generation of thrombin. Finally, high TAFI levels result in prolonged downregulation of fibrinolysis. Since no mutations have been found that elevate these coagulation factors, we do not know whether a gain or loss of function is responsible.

2003: Interaction, regulation, and clustering Interaction Venous thrombosis like many other diseases is multicausal. The discovery of common risk factors was a prerequisite for the study of interaction and made clear that risk factors for thrombosis result from genetic differences or differences brought about by the environment or even behavior. Plasma levels of proteins can for instance be determined by polymorphisms in the functional allele and by age or hormones. A good example of this complicated regulation is factor VIII. ABO blood group is an important genetic determinant of plasma factor VIII levels [47]. Von Willebrand factor is the carrier protein of factor VIII in plasma and also determines the factor VIII level [48]. If both blood group and von Willebrand factor are taken into account, a clear familial clustering remains, suggesting a third set of genes that regulate factor VIII levels [49]. Apart from the genetic causes, factor VIII is also influenced by environmental factors like acute phase reactions and age. It is clear that not only is thrombosis a multicausal disease, but also that the level of coagulation factors reflects a mixture of genetic and environmental determinants [50, 51]. As mentioned earlier, the mean age at first thrombosis for patients from thrombophilic families is much younger than for consecutive patients with thrombosis [17]. This phenomenon is probably due to interaction of several genetic defects. In thrombophilic families, the risk of thrombosis in a combination of protein C deficiency and factor V Leiden was much higher than for relatives with only protein C deficiency [17]. This

Thrombosis in clinical practice

90

gene-gene interaction results in variation within and between families. Homozygous disease is another example of this interaction. More commonly, a gene-environment interaction is present in patients with thrombosis. The synergistic effect of factor V Leiden and oral contraceptive use was described in 1994 [52]. The annual absolute risk of women who used oral contraceptives and were carriers of factor V Leiden was 28.5 per 10,000 people, whereas this risk was 5.7 per 10,000 women per year for those with factor V Leiden without contraceptives and 3.0 per 10,000 per year for women with contraceptives without factor V Leiden [52]. An example of environment-environment interaction is contraceptive use and age [52]. All this shows that the nature of thrombosis is complex. The model of multicausal disease is not always sufficient to explain why the clustering of these different risk factors is sufficient to cause thrombosis in one patient but not in the other. Refinement of this model by including the dynamic influence of age is more useful for an individual risk estimate [50]. In this way, we can better incorporate interaction of different risk factors. Figure 5.1 shows the hypothetical situation of a patient that is followed through life [50]. This person has a certain basic thrombosis potential, which is formed by genetic factors (in this case factor V Leiden). Through life several events lead to an increased thrombosis potential.

Figure 5.1 Models of thrombosis risk [50]. In each panel, the figure shows the thrombosis (yellow) potential of each risk factor during an individual’s

Thrombophilia and venous thrombosis

91

life and the resultant thrombosis potential (pink). Printed with permission. At the age of 30 years, the combination of several risk factors causes the thrombosis potential to exceed the thrombosis threshold and leads to clinical disease. Since increasing age itself is a risk factor for thrombosis, the threshold will be easier to reach at a later age and fewer risk factors will be needed to cause thrombosis. Clustering and regulation Since several procoagulant risk factors for thrombosis are closely related in the hemostatic system, a common genetic determinant of these coagulation factors levels could regulate these levels in addition to the environmental determinants. A significant genetic component of coagulation factors has been found in the Spanish population [53], the United Kingdom [54], and the United States [55]. Interestingly, six families with a thrombotic tendency were reported in which high levels of coagulation factors XI, IX, and VIII aggregated [56]. The inheritance pattern seemed to be dominant autosomal [56]. To date, the genetic basis of high levels is unknown. It is, however, possible that regulatory genes outside the genes of the coagulation factors regulate the protein levels. These levels would then cluster in an individual due to pleiotropic effects. The evaluation between a potential risk factor and the occurrence of thrombosis is becoming more difficult, since adjustment is needed for more and more already known thrombotic risk factors. In order to better estimate the role of possible confounders and clustering of these factors, a priori knowledge of the interrelations of procoagulant and anticoagulant factors is important. With the data of the LETS, factor analysis was conducted using principalcomponents analysis with varimax rotation [57]. The number of variables is reduced by constructing relatively independent summary factors (the socalled principal components), which explain most of the variation in the data. In large studies, where several risk factors seem to cluster, it is important to find the smallest number of principal components that still reflects the original data and variance. The newly formed principal loadings can be compared with the original variables by factor loadings, comparable to Pearsons correlation coefficients. When all the measured coagulation factors of the LETS were analyzed, three relatively separate cluster patterns were found (Figure 5.2). There was a clustering of the vitamin K-dependent factors II, VII, IX, and X, together with coagulation factors XI and XII. The second cluster consisted of factors V, VIII, IX, and fibrinogen. The third ‘cluster’ was made up only of one clotting factor, namely factor XIII subunit levels. These results show that interrelations exist between different coagulation factors in the hemostatic system. Therefore, common shared genetic mechanisms may be responsible for the clustering of these coagulation factors. Transcription factors, such as hepatocyte nuclear factor-4, may contribute to the first clustering pattern [58–60]. Factors V and VIII share a great part of homology and posttranslational modifications and this could explain the second clustering [61].

Thrombosis in clinical practice

92

Figure 5.2 Factor loading pattern of procoagulant factors and fibrinogen in 466 healthy individuals [57]. Printed with permission. By using factor analysis, a better overall estimation of the overall risk associated with coagulant factors may become possible. The described method facilitates the interpretation of epidemiological studies and hopefully the determination of the thrombosis risk for individual patients. Family studies might be helpful in unravelling the genetic basis for these findings.

Consequences of thrombophilia Nowadays, a dozen different thrombophilia factors for thrombosis have been elucidated. However, venous thrombosis is a multicausal disease in which several risk factors, both genetic and acquired, have to occur simultaneously to cause thrombosis [50, 51]. The interaction between these risk factors is dynamic instead of static, with age as an important contributor. In this complex situation, what is the contribution of inherited thrombophilia? And, now that we know so many thrombophilia factors, what is the consequence of thrombophilia? We will address this question by reviewing the influence of thrombophilia on the intensity and duration of anticoagulant therapy after a thromboembolic event, the risk of recurrence of venous thrombosis, and the type of thrombosis. Thrombophilia could further be of importance for asymptomatic individuals. Treatment of patients with thrombophilia The intensity of anticoagulant treatment of patients with thrombosis who have a thrombophilia factor usually is identical to patients without inherited defects [62]. Even in patients with deficiencies of antithrombin, protein C, or protein S, the therapeutic approach of thrombosis is generally the same. This means that irrespective of thrombophilia, the optimal intensity of the International Normalized Ratio (INR) is 2.0– 3.5, and this regimen is sufficient to prevent recurrences during therapy [63]. Recently, it was also shown that in subjects with the antiphospholipid syndrome, moderate-intensity anticoagulant therapy is adequate [64]. The optimal duration of anticoagulant therapy is

Thrombophilia and venous thrombosis

93

uncertain, but does not seem to be influenced by the common thrombophilia factors. The goal of therapy is mainly to prevent recurrences. Since factor V Leiden and the prothrombin mutation are common in patients with thrombosis, several studies have analyzed the risk of recurrent thrombosis in association with these prothrombotic defects. Neither of these mutations seems to increase the risk of recurrences, although the data are not in complete agreement [65–70]. High levels of factor VIII levels and homocysteine seem to be associated with recurrences [31, 71], but these results have to be confirmed in other studies. Recurrent venous thrombosis might be more common in patients with a deficiency of antithrombin, protein C, or protein S, but these results are based on retrospective data [72]. Given the low prevalence of these defects, it will be difficult to accurately determine the risk of recurrent thrombosis. From the other known prothrombotic defects, the effect on recurrent thrombosis is unknown. The combination of defects or homozygous factor V Leiden is probably associated with an increased risk of recurrence, although the information on patients studied so far is low [73–78]. So, apart from the antiphospholipid syndrome [79], combined or homozygous defects, and possibly antithrombin deficiency, the impact of thrombophilia on the optimal duration of therapy to prevent recurrent thrombosis is probably small [80]. Clinical manifestations of thrombophilia Thrombosis in patients with thrombophilia usually manifests as deep vein thrombosis or pulmonary embolism. In patients with thrombophilia, thrombosis can also occur at unusual sites, such as the cerebral, visceral, and axillary veins (Table 5.2). Superficial thrombophlebitis is more common in protein C or protein S deficiency. In rare cases, coumarin skin necrosis can occur [81]. Recurrence of thrombosis, a family history of thrombosis, and first episode of thrombosis at young age are more common in patients with thrombophilia. In unselected thrombosis patients with a prothrombotic defect, such as factor V Leiden or prothrombin mutation, the difference with thrombosis patients without a defect is less clear [17]. Thrombophilia in asymptomatic patients In women with the factor V Leiden or prothrombin mutation, oral contraceptive use, hormone replacement therapy, and pregnancy further increase the risk of thrombosis, but the absolute risk seems to be low. Middeldorp et al. prospectively followed asymptomatic carriers of the factor V Leiden mutation [82]. In 470 individuals, the annual incidence of venous thrombosis was 0.58%, which does not justify routine screening of family members. Also in risk situations, such as pregnancy or oral contraceptive use, the rate of thrombosis was low [82]. In pregnant asymptomatic women heterozygous for factor V Leiden or the prothrombin mutation, absolute risk of thrombosis is less than 3% [83, 84], whereas a deficiency of antithrombin, protein C, or protein S leads to a risk of 4.1% [85]. Taken together, the risk of thrombosis in asymptomatic

Thrombosis in clinical practice

94

Table 5.2 Clinical manifestations of thrombophilia. Venous thrombosis at unusual site • Mesenteric • Pelvic • Cerebral sinuses • Portal • Axillary Family history of venous thromboembolism Onset of thrombosis at young age Recurrent episodes of venous thromboembolism Warfarin-induced skin necrosis Recurrent fetal loss Thrombophlebitis Neonata purpura fulminans

carriers of thrombophilia defects seems low and does not justify screening. The optimal strategy of thrombosis prophylaxis of asymptomatic carriers is probably not different from patients without heritable thrombophilia, but this subject remains controversial as long as there are no trials comparing prolonged prophylaxis to standard prophylaxis in highrisk situations or prophylaxis versus placebo during pregnancy [86].

Implications of thrombophilia screening Testing for thrombophilia is subject to an intense debate [87, 88]. Clinicians who perform thrombophilia screening usually argue that a better understanding of the pathogenesis of thrombosis is important for both the treating physician and for the patient. Family members of the proband with a prothrombotic defect can also be screened, in order to tailor prophylactic treatment during high-risk situations [89]. Others argue against screening that, because thrombophilia does not seem to affect the usual management of thrombosis treatment, screening is not cost-effective and leads to anxiety among asymptomatic carriers or false reassurance in those without the defect [88]. Apart from the discussion whether screening should be performed, it is important how to interpret the results of studies for thrombophilia. What are the implications for an individual patient, for the family members, the treating physician, researcher, or even the society?

Thrombophilia and venous thrombosis

95

Influence of patient selection on the association of thrombophilia and thrombosis The strength of an association between an inherited coagulation defect and venous thrombosis can be influenced by the type of study and the selection of thrombosis patients and controls [51]. In cohort (follow-up) studies, quantitative estimates (i.e. absolute risks) can be obtained. In case-control studies, one can estimate relative risks (as an odds ratio) by comparing thrombosis patients with healthy individuals. This figure indicates how much higher the thrombosis risk is in the presence of a certain risk factor than in the absence of that factor. In unselected cases from populationbased studies, relative risks can be applied to all individuals with that particular risk factor, provided that cases are well selected. Population-based case-control studies can be used to calculate the attributable risk, that is, the proportion of all thrombotic events that would have been prevented by removing the risk factor. Family studies often consist of subjects that were selected because of a conspicuously high frequency of thrombosis. In these studies, the occurrence of thrombosis is compared between family members with and without the risk factor. These studies are ideal for studying the type of inheritance of a certain risk factor and to qualitatively estimate the thrombosis risks. These thrombophilia families usually have more than one thrombophilic defect, and results cannot be extrapolated to the general population. The influence of selection is well reflected in the age of onset of thrombosis that clearly differs between individuals from thrombophilia families and unselected thrombosis patients [16]. Finally, other aspects such as the objective diagnosis of thrombosis and prospective versus retrospective studies also influence the estimates of risk. Importance of a risk factor for thrombosis With so many new risk factors emerging, the question is what impact they have in daily clinical practice. In other words, how can the results from research be translated to clinically practical guidelines? First of all, we must make sure that the new risk factor is independent and clinically relevant. This requires full adjustment for potential confounders, such as age, sex, body mass index, other coagulation factors, etc. This does not apply to genetic risk factors, since these are by definition unconfounded. It is important to appreciate and interpret the differences between absolute and relative risks. The relative risks that have been calculated from case-control studies are mainly important to the researcher, whereas absolute risk estimates (the probability to develop thrombosis and the possibility to lower this probability) are relevant to the individual patient and his physician [90]. Population attributable risk estimates are also important for the population and can influence decision-making. Asymptomatic females with deficiencies of antithrombin, protein C, or protein S have an 8-fold increased relative risk of thrombosis during pregnancy [72]. The absolute risk is 4.1% (7 in 169 pregnancies). So, these deficiencies have a high magnitude of risk, but, because of the low prevalence, account for only a small percentage of the overall thrombosis risk. Likewise, oral contraceptives and hormone replacement therapy (HRT), both, increase the risk of the thrombosis approximately 4-fold. Since the baseline risk of thrombotic disease is nearly

Thrombosis in clinical practice

96

10-fold higher in the older HRT-users, their basic risk of thrombosis is larger than women who use oral contraceptives [90]. Likewise, hypertension is a moderate risk factor for congestive heart failure, but accounts for nearly half of the cases of heart failure in a population [91]. So, in general, common risk factors usually have moderate relative risks, but are important at population level when its prevalence is high. Strong risk factors are generally less important for healthy individuals. If we consider thrombophilia screening for risk factors, the measurement must be reliable (low coefficient of variation) and reflect a true representation of the risk factor. For prothrombotic mutations, such as factor V Leiden and prothrombin mutation, this is generally not a problem. Most coagulation factors, such as factor VIII, however, have a large intra-individual and inter-assay variation, as reflected in the decision to use the factor VIII measurements of three different plasma samples to calculate the probability of carriership of hemophilia A. Moreover, there is an important question as to how we should interpret the result of a measurement in terms of risk of a first thrombotic event and risk of recurrences. Most risk factors have wide ranges of values with large overlap between individuals with and without thrombosis. These risk factors typically increase with increasing levels, without a clear threshold. So, artificial cut-off values that were used in clinical research are being used now for decisionmaking for individual patients. It is unknown whether these cut-off values are practical and reliable. We do not know the sensitivity or specificity of most risk factors in order to predict future occurrences of thrombosis. Factor VIII levels can easily elevate above the cut-off value of 150 IU/dl due to acute phase reactions, as a thrombotic event. This transient rise may cause a mislabeling of a person with venous thrombosis who normally has a low factor VIII levels [92]. Since factor VIII levels may be associated with the risk of recurrences [31], a treating physician might decide to prolong anticoagulant therapy on the basis of this measurement. This shows that the results from research cannot simply be extrapolated to patient care and can even lead to wrong decisions (primum non nocere).

Implications of thrombophilia screening for the individual patient As already stated, relative risk has no value in the clinic, and only knowledge of the absolute risk of developing thrombosis may have relevance for the individual patient, and then only if this leads to the possibility of prevention. This would imply that for each patient at risk for a first episode of thrombosis, or for a recurrent event, an individualized risk profile should be available with age, sex, current risk factors, and the possibility of future risk factors, such as trauma, surgery, and pregnancy, and for each factor its strength as well as its interaction with the other factors should be known. This scenario is still far away. It is not even feasible to readily identify patients with thrombophilia unless all thrombosis patients are screened, since half of the first thrombotic events in patients with thrombophilia are not idiopathic and occur in high-risk situations. Practical recommendations have been suggested to guide screening strategies in patients with thrombosis, in which patients are divided into “strongly” and “weakly” thrombophilic categories [62]. The “strongly” thrombophilic patients include patients less than 50 years old, patients with recurrent thrombosis, or first-degree family members with a thrombotic event before 50 years of age. All other patients are “weakly” thrombophilic and should be

Thrombophilia and venous thrombosis

97

screened for the common defects such as factor V Leiden and prothrombin mutation, while the former group should also be tested for the more rare defects, such as deficiencies of protein C, protein S, and antithrombin. This strategy optimizes the likelihood of finding a prothrombotic abnormality, but does not necessarily benefit the patient. With the current knowledge, it is questionable whether the presence of a risk factor leads to any difference in clinical management, and therefore screening does not seem helpful. The most compelling question is whether based on laboratory tests, we can predict the risk of recurrence, and, while the various studies are not in complete agreement, it may well be that the risk of recurrence is not increased in the presence of prothrombotic defects. In that case, it makes more sense to base clinical strategy on clinical history, that is, the severity of the event or the age of the patient, than on laboratory tests. The next question concerns asymptomatic relatives: is it useful to screen asymptomatic individuals from a family with hereditary thrombophilia, for instance women who intend to become pregnant or want to start oral contraceptives? Again, the literature offers little assistance, except that in most cases the risk of thrombosis appears to be low. Women from families with a strong history of thrombosis may consider not using oral contraceptives.

Conclusions The last decade revealed several new risk factors that contribute to a better understanding of the pathogenesis of venous thrombosis. Well-designed, large population-based casecontrol studies were a prerequisite for the establishment of new risk factors, such as factor V Leiden, prothrombin 20210A mutation, procoagulant factors—for example, factors VIII, IX, and XI—and antifibrinolytic factors, such as TAFI. Since many persons with a thrombophilic factor are asymptomatic, a single defect is seldom sufficient to cause thrombosis. Thrombosis is thus a multicausal disease, in which genetic and environmental factors interact dynamically. The common risk factors with a high prevalence in the general population make a major contribution to the overall risk of thrombosis. These risk factors are likely to occur simultaneously in one individual. When these clusters of risk factors can be identified, preventive measures can be installed, mainly for those individuals with a genetic predisposition. This can only be assessed adequately through sufficient knowledge of important risk factors for thrombosis, their effect, and interaction with other genetic and environmental factors, and the beneficial effect of intervention. Until that time, screening for thrombophilia will remain a matter of debate.

References [1] Anning ST. The historical aspects of venous thromboembolism. Med Hist 1957; 1:28–37. [2] Wiseman R. Several Chirurgical Treatises. 2nd edn. London. R. Norton and J. Macock, 1686, book 1, p. 64 of 577. [3] Virchow R. Phlogose und Thrombose im gefäßsystem; Gesammelte Abhandlungen zur Wissenschaftlichen Medizin. Frankfurt: Staatsdruckerei; 1856.

Thrombosis in clinical practice

98

[4] Åstrup T. The haemostatic balance. Thromb Diath Haemorrh 1958; 2:347–57. [5] Egeberg O. Inherited antithrombin III deficiency causing thrombophilia. Thromb Diath Haemorrh 1965; 13:516–30. [6] Demers C, Ginsberg JS, Hirsh J, Henderson P, Blajchman MA. Thrombosis in antithrombinIIIdeficient persons. Report of a large kindred and literature review. Ann Intern Med 1992; 116:754–61. [7] Hirsh J, Piovella F, Pini M. Congenital antithrombin III deficiency. Incidence and clinical features. Am J Med 1989; 87(Suppl.):34–8. [8] Thaler E, Lechner K. Antithrombin III deficiency and thromboembolism. Clin Haematol 1981;10: 369–90. [9] Griffin JH, Evatt BL, Zimmerman TS, Kleiss AJ, Wideman C. Deficiency of protein C in congenital thrombotic disease. J Clin Invest 1981; 68:1370–3. [10] Allaart CF, Poort SR, Rosendaal FR, Reitsma PH, Bertina RM, Briët E. Increased risk of venous thrombosis in carriers of hereditary protein C deficiency defect. Lancet 1993; 341:134– 8. [11] Bovill EG, Bauer KA, Dickerman JD, Callas P, West B. The clinical spectrum of heterozygous protein C deficiency in a large New England kindred. Blood 1989; 73:712–17. [12] Reitsma PH, Bernardi F, Doig RG, Gandrille S, Greengard JS, Ireland H, et al. Protein C deficiency: a database of mutations, 1995 update. On behalf of the Subcommittee on Plasma Coagulation Inhibitors of the Scientific and Standardization Committee of the ISTH. Thromb Haemost 1995; 73:876–89. [13] Comp PC, Esmon CT. Recurrent venous thromboembolism in patients with a partial deficiency of protein S. N Engl J Med 1984; 311:1525–8. [14] Schwartz HP, Fischer M, Hopmeier P, Batard MA, Griffin JH. Plasma protein S deficiency in familial thrombotic disease. Blood 1984; 64:1297–300. [15] Engesser L, Broekmans AW, Briët E, Brommer EJ, Bertina RM. Hereditary protein S deficiency: clinical manifestations. Ann Intern Med 1987; 106: 677–82. [16] Koeleman BP, Reitsma PH, Bertina RM. Familial thrombophilia: a complex genetic disorder. Semin Hematol 1997; 34:256–64. [17] Lensen RP, Rosendaal FR, Koster T, Allaart CF, de Ronde H, Vandenbroucke JP, Reitsma PH, Bertina RM. Apparent different thrombotic tendency in patients with factor V Leiden and protein C deficiency due to selection of patients. Blood 1996; 88:4205–8. [18] Miletich J, Sherman L, Broze G Jr. Absence of thrombosis in subjects with heterozygous protein C deficiency. N Engl J Med 1987; 317:991–6. [19] van der Meer FJ, Koster T, Vandenbroucke JP, Briët E, Rosendaal FR. The Leiden Thrombophilia Study (LETS). Thromb Haemost 1997; 78:631–5. [20] Koster T, Rosendaal FR, Briët E, van der Meer FJ, Colly LP, Trienekens PH, Poort SR, Reitsma PH, Vandenbroucke JP. Protein C deficiency in a controlled series of unselected outpatients: an infrequent but clear risk factor for venous thrombosis (Leiden Thrombophilia Study). Blood 1995; 85:2756–61. [21] Faioni EM, Valsecchi C, Palla A, Taioli E, Razzari C, Mannucci PM. Free protein S deficiency is a risk factor for venous thrombosis. Thromb Haemost 1997; 78:1343–6. [22] Heijboer H, Brandjes DP, Büller HR, Sturk A, ten Cate JW. Deficiencies of coagulationinhibiting and fibrinolytic proteins in outpatients with deep-vein thrombosis. N Engl J Med 1990; 323:1512–16. [23] Mateo J, Oliver A, Borrell M, Sala N, Fontcuberta J. Laboratory evaluation and clinical characteristics of 2,132 consecutive unselected patients with venous thromboembolism—results of the Spanish Multicentric Study on Thrombophilia (EMET-Study). Thromb Haemost 1997; 77:444–51.

Thrombophilia and venous thrombosis

99

[24] Dahlbäck B, Carlsson M, Svensson PJ. Familial thrombophilia due to a previously unrecognized mechanism characterized by poor anticoagulant response to activated protein C: prediction of a cofactor to activated protein C. Proc Natl Acad Sci USA 1993; 90:1004–8. [25] Bertina RM, Koeleman BP, Koster T, Rosendaal FR, Dirven RJ, de Ronde H, van der Velden PA, Reitsma PH. Mutation in blood coagulation factor V associated with resistance to activated protein C. Nature 1994; 369:64–7. [26] Rosendaal FR, Koster T, Vandenbroucke JP, Reitsma PH. High risk of thrombosis in patients homozygous for factor V Leiden (activated protein C resistance). Blood 1995; 85:1504–8. [27] Griffin JH, Evatt B, Wideman C, Fernandez JA. Anticoagulant protein C pathway defective in majority of thrombophilic patients. Blood 1993; 82:1989–93. [28] Rees DC, Cox M, Clegg JB. World distribution of factor V Leiden. Lancet 1995; 346:1133–4. [29] Koster T, Blann AD, Briët E, Vandenbroucke JP, Rosendaal FR. Role of clotting factor VIII in effect of von Willebrand factor on occurrence of deepvein thrombosis. Lancet 1995; 345:152–5. [30] Kraaijenhagen RA, in ‘t Anker PS, Koopman MM, Reitsma PH, Prins MH, van den Ende A, Büller HR. High plasma concentration of factor VIIIc is a major risk factor for venous thromboembolism. Thromb Haemost 2000; 83:5–9. [31] Kyrle PA, Minar E, Hirschl M, Bialonczyk C, Stain M, Schneider B, et al. High plasma levels of f actor VIII and the risk of recurrent venous thromboembolism. N Engl J Med 2000; 343:457–62. [32] Kamphuisen PW, Eikenboom JCJ, Bertina RM. Elevated factor VIII levels and the risk of thrombosis. Arterioscl Thromb Vasc Bio 2001; 21:731–8. [33] Falcon CR, Cattaneo M, Panzeri D, Martinelli I, Mannucci PM. High prevalence of hyperhomocyst(e)inemia in patients with juvenile venous thrombosis. Arterioscler Thromb 1994; 14:1080–3. [34] den Heijer M, Koster T, Blom HJ, Bos GM, Briët E, Reitsma PH, Vandenbroucke JP, Rosendaal FR. Hyperhomocysteinemia as a risk factor for deepvein thrombosis. N Engl J Med 1996; 334:759–62. [35] Poort SR, Rosendaal FR, Reitsma PH, Bertina RM. A common genetic variation in the 3′untranslated region of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis. Blood 1996; 88:3698–703. [36] Cumming AM, Keeney S, Salden A, Bhavnani M, Shwe KH, Hay CR. The prothrombin gene G20210A variant: prevalence in a U.K. anticoagulant clinic population. Br J Haematol 1997; 98: 353–5. [37] Hillarp A, Zöller B, Svensson PJ, Dahlbäck B. The 20210 A allele of the prothrombin gene is a common risk factor among Swedish outpatients with verified deep venous thrombosis. Thromb Haemost 1997; 78:990–2. [38] Brown K, Luddington R, Williamson D, Baker P, Baglin T. Risk of venous thromboembolism associated with a G to A transition at position 20210 in the 3′-untranslated region of the prothrombin gene. Br J Haematol 1997; 98:907–9. [39] de Visser MC, Rosendaal FR, Bertina RM. A reduced sensitivity for activated protein C in the absence of factor V Leiden increases the risk of venous thrombosis. Blood 1999; 93:1271–6. [40] van Hylckama Vlieg A, van der Linden IK, Bertina RM, Rosendaal FR. High levels of factor IX increase the risk of venous thrombosis. Blood 2000; 95:3678–82. [41] Meijers JC, Tekelenburg WL, Bouma BN, Bertina RM, Rosendaal FR, Meijers JC, Tekelenburg WL, Bouma BN, Bertina RM, Rosendaal FR. N Engl J Med 2000; 342:696–701. [42] van Tilburg NH, Rosendaal FR, Bertina RM. Thrombin activatable fibrinolysis inhibitor and the risk for deep vein thrombosis. Blood 2000; 95: 2855–9. [43] Meijers JC, Marquart JA, Bertina RM, Bouma BN, Rosendaal FR. Protein C inhibitor (plasminogen activator inhibitor-3) and the risk of venous thrombosis. Br J Haematol 2002;118:604–9.

Thrombosis in clinical practice

100

[44] Koster T, Rosendaal FR, Reitsma PH, van der Velden PA, Briët E, Vandenbroucke JP. Factor VII and fibrinogen levels as risk factors for venous thrombosis. A case-control study of plasma levels and DNA polymorphisms—the Leiden Thrombophilia Study (LETS). Thromb Haemost 1994; 71:719–22. [45] Kamphuisen PW, Rosendaal FR, Eikenboom JCJ, Bos R, Bertina RM. Factor V antigen levels and venous thrombosis: risk profile, interaction with factor V Leiden and relation with factor VIII:Ag levels. Arterioscl Thromb Vasc Bio 2000; 20:1382–6. [46] de Visser MC, Poort SR, Vos HL, Rosendaal FR, Bertina RM. Factor X levels, polymorphisms in the promoter region of factor X, and the risk of venous thrombosis. Thromb Haemost 2001; 85:1011–17. [47] Jeremic M, Weisert O, Gedde-Dahl TW. Factor VIII (AHG) levels in 1016 regular blood donors. The effects of age, sex, and ABO blood groups. Scand J Clin Lab Invest 1976; 36:461– 6. [48] Wise RJ, Dorner AJ, Krane M, Pittman DD, Kaufman RJ. The role of von Willebrand factor multimerization and propeptide cleavage in the binding and stabilization of factor VIII. J Biol Chem 1991; 266:21948–55. [49] Kamphuisen PW, Houwing-Duistermaat JJ, van Houwelingen HC, Eikenboom, JC, Bertina RM, Rosendaal FR. Familial clustering of factor VIII and von Willebrand factor levels. Thromb Haemost 1998; 79:323–7. [50] Rosendaal FR. Venous thrombosis: a multicausal disease. Lancet 1999; 353:1167–73. [51] Rosendaal FR. Risk factors for venous thrombosis: prevalence, risk, and interaction. Semin Hematol 1997; 34:171–87. [52] Vandenbroucke JP, Koster T, Briët E, Reitsma PH, Bertina RM, Rosendaal FR. Increased risk of venous thrombosis in oral-contraceptive users who are carriers of factor V Leiden mutation. Lancet 1994; 344:1453–7. [53] Souto JC, Almasy L, Borrell M, Gari M, Martinez E, Mateo J, et al. Genetic determinants of hemostasis phenotypes in Spanish families. Circulation 2000; 101:1546–51. [54] de Lange M, Snieder H, Ariens RA, Spector TD, Grant PJ. The genetics of haemostasis: a twin study. Lancet 2001; 357:101–5. [55] Rosendaal FR, Bovill EG. Heritability of clotting factors and the revival of the prothrombotic state. Lancet 2002; 359:638–9. [56] Lavigne G, Mercie E, Queré I, Dauzat M, Gris JC. Thrombophilic families with inheritably associated high levels of coagulation factors VIII, IX and XI. J Thromb Haemost 2003; 1:2134– 9. [57] Van Hylckama Vlieg A, Callas PW, Cushman M, Bertina RM, Rosendaal FR. Inter-relation of coagulation factors and d-dimer levels in healthy individuals. J Thromb Haemost 2003; 1:516– 22. [58] Erdmann D, Heim J. Orphan nuclear receptor HNF-4 binds to the human coagulation factor VII promoter. J Biol Chem 1995; 270:22988–96. [59] Reijnen MJ, Sladek FM, Bertina RM, Reitsma PH. Disruption of a binding site for hepatocyte nuclear factor 4 results in hemophilia B Leyden. Proc Natl Acad Sci USA 1992; 89:6300–3. [60] Hung HL, High KA. Liver-enriched transcription factor HNF-4 and ubiquitous factor NF-Y are critical for expression of blood coagulation factor X. J Biol Chem 1996; 271:2323–31. [61] Kaufman RJ. Post-translational modifications required for coagulation factor secretion and function. Thromb Haemost 1998; 79:1068–79. [62] Bauer KA. The thrombophilias: well-defined risk factors with uncertain therapeutic implications. Ann Intern Med 2001; 135:367–73. [63] Kearon C, Ginsberg JS, Kovacs MJ, Anderson DR, Wells P, Julian JA, et al. Comparison of low-intensity warfarin therapy with conventional-intensity warfarin therapy for long-term prevention of recurrent venous thromboembolism. N Engl J Med 2003; 349:631–9.

Thrombophilia and venous thrombosis

101

[64] Crowther MA, Ginsberg JS, Julian JA, Denburg J, Hirsh J, Douketis J, et al. A comparison of two intensities of warfarin for the prevention of recurrent thrombosis in patients with the antiphospholipidantibody syndrome. N Engl J Med 2003; 349: 1133–8. [65] Ridker PM, Miletich JP, Stampfer MJ, Goldhaber SZ, Lindpaintner K, Hennekens CH. Factor V Leiden and risks of recurrent idiopathic venous thromboembolism. Circulation 1995; 92:2800–2. [66] Simioni P, Prandoni P, Lensing AW, Scudeller A, Sardella C, Prins MH, et al. The risk of recurrent venous thromboembolism in patients with an Arg506→Gln mutation in the gene for factor V (factor V Leiden). N Engl J Med 1997; 336:399–403. [67] Kearon C, Gent M, Hirsh J, Weitz J, Kovacs MJ, Anderson DR, et al. A comparison of three months of anticoagulation with extended anticoagulation for a first episode of idiopathic venous thromboembolism. N Engl J Med 1999; 340:901–7. [68] Eichinger S, Minar E, Hirschl M, Bialonczyk C, Stain M, Mannhalter C. The risk of early recurrent venous thromboembolism after oral anticoagulant therapy in patients with the G20210A transition in the prothrombin gene. Thromb Haemost 1999; 81:14–17. [69] Eichinger S, Pabinger I, Stumpflen A, Hirschl M, Bialonczyk C, Schneider B. The risk of recurrent venous thromboembolism in patients with and without factor V Leiden. Thromb Haemost 1997; 77:624–8. [70] Lindmarker P, Schulman S, Sten-Linder M, Wiman B, Egberg N, Johnsson H. The risk of recurrent venous thromboembolism in carriers and non-carriers of the G1691A allele in the coagulation factor V gene and the G20210A allele in the prothrombin gene. DURAC Trial Study Group. Duration of Anticoagulation. Thromb Haemost 1999;81:684–9. [71] den Heijer M, Blom HJ, Gerrits WB, Rosendaal FR, Haak HL, Wijermans PW, et al. Is hyperhomocysteinaemia a risk factor for recurrent venous thrombosis? Lancet 1995; 345:882–5. [72] van den Belt AG, Sanson BJ, Simioni P, Prandoni P, Büller HR, Girolami A, et al. Recurrence of venous thromboembolism in patients with familial thrombophilia. Arch Intern Med 1997; 157:2227–32. [73] De Stefano V, Martinelli I, Mannucci PM, Paciaroni K, Chiusolo P, Casorelli I, et al. The risk of recurrent deep venous thrombosis among heterozygous carriers of both factor V Leiden and the G20210A prothrombin mutation. N Engl J Med 1999; 341:801–6. [74] Emmerich J, Rosendaal FR, Cattaneo M, Margaglione M, De Stefano V, Cumming T, et al. Combined effect of factor V Leiden and prothrombin 20210A on the risk of venous thromboembolism—pooled analysis of 8 case-control studies including 2310 cases and 3204 controls. Study Group for Pooled-Analysis in Venous Thromboembolism. Thromb Haemost 2001;86:809–16. [75] Koeleman BP, Reitsma PH, Allaart CF, Bertina RM. Activated protein C resistance as an additional risk factor for thrombosis in protein C-deficient families. Blood 1994; 84:1031–5. [76] van Boven HH, Reitsma PH, Rosendaal FR, Bayston TA, Chowdhury V, Bauer KA, et al. Factor V Leiden (FV R506Q) in families with inherited antithrombin deficiency. Thromb Haemost 1996; 75:417–21. [77] Zöller B, Berntsdotter A, Garcia de Frutos P, Dahlbäck B. Resistance to activated protein C as an additional genetic risk factor in hereditary deficiency of protein S. Blood 1995; 85:3518–23. [78] Meinardi JR, Middeldorp S, de Kam PJ, Koopman MM, van Pampus EC, Hamulyak K, et al. The incidence of recurrent venous thromboembolism in carriers of factor V Leiden is related to concomitant thrombophilic disorders. Br J Haematol 2002; 116: 625–31. [79] Schulman S, Svenungsson E, Granqvist S. Anticardiolipin antibodies predict early recurrence of thromboembolism and death among patients with venous thromboembolism following anticoagulant therapy. Duration of Anticoagulation Study Group. Am J Med 1998; 104:332–8. [80] Bauer KA. Management of thrombophilia. J Thromb Haemost 2003; 1:1429–34. [81] Makris M, Rosendaal FR, Preston FE. Familial thrombophilia: genetic risk factors and manage-ment. J Intern Med Suppl 1997; 740:9–15.

Thrombosis in clinical practice

102

[82] Middeldorp S, Meinardi JR, Koopman MM, van Pampus EC, Hamulyak K, van der Meer J, et al. A prospective study of asymptomatic carriers of the factor V Leiden mutation to determine the incidence of venous thromboembolism. Ann Intern Med 2001; 135:322–7. [83] Lindqvist PG, Svensson PJ, Marsaal K, Grennert L, Luterkort M, Dahlbäck B. Activated protein C resistance (FV:Q506) and pregnancy. Thromb Haemost 1999; 81:532–7. [84] Martinelli I, Bucciarelli P, Margaglione M, De Stefano V, Castaman G, Mannucci PM. The risk of venous thromboembolism in family members with mutations in the genes of factor V or prothrombin or both. Br J Haematol 2000; 111:1223–9. [85] Ginsberg JS, Greer I, Hirsh J. Use of antithrombotic agents during pregnancy. Chest 2001; 119(Suppl.): 22–31. [86] Ginsberg JS, Bates SM. Management of venous thromboembolism during pregnancy. J Thromb Haemost 2003; 1:1435–42. [87] Martinelli I. Pros and cons of thrombophilia testing: pros. J Thromb Haemost 2003; 1:410–11. [88] Machin SJ. Pros and cons of thrombophilia testing: cons. J Thromb Haemost 2003; 1:412–13. [89] Martinelli I, Mannucci PM, De Stefano V, Taioli E, Rossi V, Crosti F, et al. Different risks of thrombosis in four coagulation defects associated with inherited thrombophilia: a study of 150 families. Blood 1998; 92:2353–8. [90] Rosendaal FR, Van Hylckama Vlieg A, Tanis BC, Helmerhorst FM. Estrogens, progestogens and thrombosis. J Thromb Haemost 2003; 1:1371–80. [91] Levy D, Larson MG, Vasan RS, Kannel WB, Ho KK. The progression from hypertension to congestive heart failure. JAMA 1996; 275:1557–62. [92] Kamphuisen PW, Ten Wolde M, Jacobs EM, Ullmann EF, Büller HR. Screening of high factor VIII levels is not recommended in patients with recently diagnosed pulmonary embolism. J Thromb Haemost 2003; 1:2239–40.

Thrombophilia and venous thrombosis

103

6 Antithrombotic therapy for atrial fibrillation Ioannis Karalis and Gregory YH Lip Introduction Atrial fibrillation (AF) is the commonest sustained disorder of cardiac rhythm. The prevalence of AF increases with age, affecting 2–4% of the population aged 60 and more than 10% of those aged 80 years and older (Figure 6.1). In the year 2000, 2.3 million Americans were estimated to have this arrhythmia and numbers are predicted to increase impressively in the coming decades. Although symptoms in the acute phase of AF are mainly associated with hemodynamic instability, it is the long-term consequences that make AF particularly dangerous. For more than 20 years, nonvalvular AF (NVAF) has been recognized as an independent risk factor for stroke; 1 of every 6 strokes occurs in a patient with AF and almost 10% of all the ischemic strokes are assumed to be embolic, originating from the left atrium of the heart [1]. Cross-sectional studies have also suggested that strokes associated with AF are usually worse in terms of their outcome (mortality, patient functional capacity, and quality of life) compared to those occurring without the arrhythmia [2]. Presence of AF confers an approximately 5- to 6-fold increase in stroke risk, an absolute risk of roughly 4.5% per year, although the precise level of annual stroke risk ranges from less than 1% to greater than 12%, according to the presence or absence of certain clinical and echocardiographically identifiable risk factors. There are certain criteria that can stratify our patients as having a higher or lower risk for the development of stroke. This risk stratification is not significantly altered by the different subtypes of AF (i.e. paroxysmal, persistent, permanent, and “lone,” Table 6.1) [3]. However, “lone” AF, in patients aged less than 65 years, exhibits a low risk for embolic events, irrespective of whether patients have paroxysmal, persistent, or permanent AF. In the past 15 years, a significant number of randomized clinical trials assessed the efficacy of different antithrombotic therapies (warfarin, aspirin, dipyridamole, clopidogrel, ximelagatran, or combinations of these) in stroke prevention in patients with NVAF. Despite the strong evidence base, anticoagulation therapy is still underprescribed, particularly in elderly AF patients who seem to benefit most. A clinical trialbased consensus approach to the problem is clearly needed.

Pathophysiology: thromboembolism in NVAF According to Virchow, three parameters interfere in the development of intraluminal

Antithrombotic therapy for atrial fibrillation

105

Figure 6.1 Incidence of AF increases with age. The data involve approximately 4,000 male air crew recruits followed continuously for 44 years in the Manitoba Follow-up Study [4]. Table 6.1 Classification of atrial fibrillation (AF) according to the American College of Cardiology/ American Heart Association/European Society of Cardiology (ACC/AHA/ESC). Type of AF

Characteristics

Paroxysmal

• Self-terminating AF • The episodes of AF generally last less than 7 days, usually less than 24 h • May be recurrent (patient experiencing two or more episodes)

Persistent

• Fails to self-terminate • Lasts more than 7 days • May also recur after cardioversion

Permanent

• AF present constantly for more than 1 year • Cardioversion not attempted or failed

“Lone”

• Paroxysmal, persistent or permanent AF • Absence of any identifiable structural heart disease

Data from Reference [3].

thrombosis: (i) abnormalities in blood flow, (ii) abnormalities in blood vessel wall, and (iii) hypercoagulable state (in respect to the blood constituents) (Figure 6.2). By definition AF is associated with blood stasis, especially in the left atrial appendage (LAA), while it has been shown to correlate with a prothrombotic state and endothelial dysfunction markers. The issue of endothelial damage has been suggested as a possible mechanism and

Thrombosis in clinical practice

106

Figure 6.2 Virchow’s triad and the prothrombotic state in atrial fibrillation.

Figure 6.3 Severely damaged left atrial appendage endocardial surface with thrombotic mass in a patient with atrial fibrillation and mitral valve disease [5]. demonstrated in certain circumstances [6] but its role in AF thrombosis remains less well defined (Figure 6.3). As mentioned earlier, the thromboembolic risk introduced by AF varies greatly; the presence of valvular heart disease multiplies the risk by 18 times whereas a previous history of stroke, transient ischemic attack (TIA), or other thromboembolism increases the relative risk by 2.5. Other conditions like hypertension and diabetes are also associated with a nearly 2-fold increase in risk for stroke. Abnormalities of blood flow The loss of atrial systolic function results in a reduction in stroke volume, leading to a correspondent reduction in cardiac output and increased atrial stasis. LAA peak flow velocity and spontaneous echo contrast (SEC) are two echocardiographic markers that “objectively” assess blood stasis and, not surprisingly, both have been shown to represent

Antithrombotic therapy for atrial fibrillation

107

independent risk factors for thrombus formation and future thromboembolism [7], It is probably worth mentioning at this point that LAA peak flow velocity is a quantifiable variable (with a calculated critical value, below which the risk of thromboembolism increases, being at 20 cm/sec), whereas SEC shares more qualitative characteristics: it refers to the impression of smoke-like echoes swirling in the left atrium as a result of the lack of coordinated left atrial systole and increased erythrocyte aggregation. Once it occurs it seems to be a stable condition not resolving even after proper anticoagulation therapy (Figure 6.4). Blood stasis is also involved in the increased risk of thromboembolism (about 7%) recorded in patients who get cardioverted without proper anticoagulation therapy. In these cases, apart from the dislodgment of preexisting thrombi, embolization may be the result of de novo clot formation in the poorly contracting (“stunned”) cardioverted atrium which may persist up to 3 weeks postcardioversion [8]. A poorly contracting dilated left ventricle has been associated in multiple clinical trials with an increased risk of stroke and thromboembolism, additive to the one conferred by AF itself. Pooled analysis from clinical trials incorporating echocardiography into their risk stratification scheme, indicates that the presence of left ventricular systolic dysfunction, even in the absence of congestive heart failure, increases the risk of stroke to 9.3%, compared with 4.4% for those with normal left ventricular function [9] (Figure 6.5). We also know that congestive heart failure represents a hypercoagulable state per se with increased prothrombotic and endothelial dysfunction plasma markers, but it is very likely the risk is further enhanced by the poor rheologic characteristics of the diseased left ventricle per se.

Figure 6.4 Left atrial appendage (LAA) flow and spontaneous echo contrast in a patient with atrial fibrillation (AF): (a) Pulsed Doppler

Thrombosis in clinical practice

108

tracing of LAA flow in AF; (b) Transesophageal echocardiography showing spontaneous echo contrast in an enlarged left atrium.

Figure 6.5 Left ventricular (LV) dysfunction may predict the incidence of stroke in patients with nonvalvular atrial fibrillation (NVAF). Data from three clinical trials (BAATAF: Boston Area Anticoagulation Trial for Atrial Fibrillation; SPINAF: Stroke Prevention in Non-rheumatic Atrial Fibrillation; and SPAF: Stroke Prevention in Atrial Fibrillationinvolving 1,066 patients) evaluating the role of anticoagulation in NVAF. Here the incidence of stroke rises from 4.4°/o in patients with normal left ventricular function to 9.3% in those with a moderate to severe left ventricular dysfunction [9]. Structural heart disease The left atrial diameter (together with left ventricular systolic dysfunction) had been considered in the past as an independent risk factor for ischemic stroke and systemic emboli [10]. However, in the AF Investigators metaanalysis, left atrial diameter failed to emerge as an independent risk factor for thromboembolism [9], The fact that substantial mitral valve regurgitation, which is associated with left atrial dilatation, seems to be

Antithrombotic therapy for atrial fibrillation

109

protective against the development of intracardiac thrombi, presumably due to increased turbulence in the left atrium and reduced blood stasis, is probably the most significant confounder. Other forms of valvular heart disease, especially mitral valve stenosis, significantly increase the thromboembolic risk by 18-fold. Other structural heart disease, including left ventricular hypertrophy and a dilated, poorly contracting left ventricle have been associated with increased stroke and thromboembolic risk, at least on univariate analysis—however, only the presence of moderate to severe left ventricular dysfunction on two-dimensional echocardiography emerged as an independent predictor of stroke and thromboembolism on multivariate analysis [9], Hypercoagulability The association of AF with a prothrombotic or hypercoagulable state, with activation of the hemostatic system, has been well documented in the literature. Both patients with paroxysmal AF and those with chronic AF have been found to exhibit increased concentrations of platelet activation markers (beta thromboglobulin, Platelet Factor 4), increased plasma markers of thrombogenesis (thrombinantithrombin complexes, fibrinogen, fibrin D-dimer), and evidence of endothelial damage/ dysfunction (von Willebrand factor) [11, 12], In most of the cases, these markers are associated with an increased prevalence of intracardiac thrombi. Introduction of anticoagulation therapy modifies this prothrombotic profile, in terms of lowering the levels of various markers, such as fibrin D-dimer or prothrombin fragment F1+2 and thrombin-antithrombin complexes. It is also an issue that the degree of anticoagulation appears to be important, with optimal doses of warfarin (International Normalized Ratio, INR 2.0–3.0) demonstrating better results than low doses of warfarin plus aspirin, or aspirin alone, can do. The precise mechanisms involved in the association of AF with thrombogenesis are as yet uncertain. Pulmonary vasculature endothelial dysfunction (due to abnormalities in cardiac blood flow), lung macrophage stimulation, and cytokine secretion (interleukin, IL-6) have been implicated [13], A possible role has also been discussed for neuroendocrine system activation, increased P-selectin expression (a marker of platelet activation), and serum lipoprotein A levels. The role of other risk factors: hypertension, diabetes, heart failure Hypertension is recognized as a major risk factor for the development of stroke (mainly of ischemic origin) and has been associated with a further 2-fold increase of the risk of stroke in AF patients. Although evidence on the precise mechanism of this phenomenon is lacking, it is very likely that the known effects of hypertension on endothelial dysfunction and hypercoagulable status act additively to the ones of AF. Diabetes imposes a similar very high thromboembolic risk for patients with AE As in the case of hypertension, diabetes has been associated with vascular endothelial damage/ dysfunction and increased coagulation activation markers, such as prothrombin activation fragments 1+2 and thrombin-antithrombin complexes. Many authors have documented the increased plasma levels of clotting factors including fibrinogen, factors VII, VIII, XI,

Thrombosis in clinical practice

110

XII, kallikrein, and von Willebrand factor in diabetic patients. On the other hand, the fibrinolytic/antithrombotic system seems defective, as manifested by the increased levels of plasminogen activator inhibitor type 1 and the decreased values of anticoagulant protein C. Furthermore, it is well recognized that patients with congestive heart failure are at an increased risk of stroke and venous thromboembolism. The pathophysiology of thrombogenesis in heart failure could well be explained in the context of Virchow’s original triad. In addition to “abnormal flow” through low cardiac output, dilated cardiac chambers, and poor contractility, patients with heart failure also demonstrate abnormalities of hemostasis and platelets (i.e. “abnormal blood constituents”), and endothelial dysfunction (“vessel wall abnormalities”). These abnormalities contribute to a prothrombotic or hypercoagulable state, which increases the risk of thrombosis in heart failure and impaired left ventricular systolic function [14].

Evidence from clinical trials Warfarin compared with placebo Adjusted-dose warfarin is highly efficacious for prevention of strokes in patients with AE A meta-analysis of 6 trials involving 2,900 patients (mean age 69 years, average followup period 1.6 years), assigned to either adjusted doses of warfarin or placebo, demonstrated a 62% reduction of relative risk for stroke in the warfarin group (95% CI 48–72%, Table 6.2 and Figure 6.6(a)) [15]. The relative risk reduction was similar in both primary (59% reduction) and secondary (68% reduction) prevention groups; in terms of absolute risk reduction and number needed to treat (NNT), however, the benefits in the primary prevention group were significantly lower, with an annual 2.7% reduction and 37 patients needed to treat to prevent one stroke, whereas the figures for the secondary prevention group were 8.4% per year and an NNT of 12 patients, respectively. The rate of stroke among participants who were not assigned to any anticoagulation treatment was 4.6% per year for primary prevention and 12.3% per year for secondary prevention. In terms of disabling and nondisabling strokes, the relative risk reduction offered by warfarin was almost identical, 59% and 61%, respectively. The major adverse events associated with oral anticoagulant use is the increased incidence of intracranial and extracranial hemorrhage. The frequency and severity of this complication is mainly dependent on the intensity of anticoagulation (the higher the INR the more the risk) and the patients’ age (those being above 75 years facing the greatest risk). In the meta-analysis by Hart et al. [15] the rate of intracranial hemorrhage averaged 0.3% per year for the anticoagulation group, while it was estimated at 0.1% per year for the placebo group. The relative risk for major extracranial hemorrhage was 2.4% per year for the warfarin group and 0.6% per year in patients who received placebo (Table 6.3). The optimal INR levels in terms of both getting the maximum benefit from anticoagulation

Antithrombotic therapy for atrial fibrillation

111

Figure 6.6 Effects of antithrombotic treatment on the incidence of ischemic and hemorrhagic strokes, in patients with atrial fibrillation. (a) Adjusteddose warfarin compared to placebo; (b) Aspirin compared to placebo; (c) Adjusted-dose warfarin compared to aspirin (AFASAK I, II: Copenhagen Atrial Fibrillation, Aspirin and Anticoagulation Study; BAATAF: Boston Area Anticoagulation Trial for Atrial Fibrillation; CAFA: Canadian Atrial Fibrillation Anticoagulation Study; EAFT: European Atrial Fibrillation Trial; ESPS II: European Stroke Prevention Study; LASAF: Low-dose Aspirin, Stroke and Atrial Fibrillation Pilot Study; PATAF: Prevention of Arterial Thromboembolism in Atrial Fibrilbatism; SPAF: Stroke Prevention in Atrial Fibrillation Study; SPINAF: Stroke Prevention in Non-rheumatic

Thrombosis in clinical practice

112

Atrial Fibrillation; UK-TIA: United Kingdom-TIA Study) [15]. treatment and minimizing the risk for a major bleeding event, has been an issue of controversy over the last decade. For patients with nonvalvular atrial fibrillation, INR levels of 2.0–3.0 have proven to be efficacious and relatively safe. Thus, for patients aged less than 75 years, we aim for an INR value of 2.5 (range 2.0–3.0), whereas for those aged more than 75 years and receiving warfarin for primary prevention of a stroke, less aggressive strategies, in terms of a target INR of 2.0 (range 1.6–2.5), may be appropriate to allow

Table 6.2 Antithrombotic therapies for stroke prevention in atrial fibrillation. Key results of metaanalysis of 16 randomized trials. Therapy

No. of trials

Participants

Relative risk reduction (95% CJ)

n

%

Adjusted-dose warfarin compared with placebo

6

2,900

62 (48–72)

Aspirin compared with placebo

6

3,119

22 (2–38)

Adjusted-dose warfarin compared with aspirin 5

2,837

36 (14–52)

Adjusted-dose warfarin compared with lowdose warfarin

3

893

38 (−20 to 68)

Aspirin compared with low-dose warfarin

2

934

15(−42 to 49)

Data from Reference [16].

Table 6.3 Warfarin versus placebo: prevalence of major bleeding events in atrial fibrillation trials. Study

Rate, percent/year Warfarin All

Intracranial

Placebo Extracranial

All

AFASAK I

0.5

0.2

0.3

0

BAATAF

1.6

0.2

1.4

1.6a

SPAF I

1.5

0.8

0.7

1.8

CAFA

2.1

0.4

1.7

0.5

SPINAF

1.4

0.2

1.2

1.1

Antithrombotic therapy for atrial fibrillation

113

Major bleeding was defined as bleeding requiring transfusion of blood or leading to permanent disability or to death. The rate was calculated as number of events per year of treatment. a Aspirin voluntary. AFASAKI: Atrial Fibrillation, Asprin and Anticoagulation; BAATAF; Boston Area Anticoagulation Trial for Atrial Fibrillation; SPAF I: Stroke Prevention in Atrial Fibrillation; CAFA: Canadian Atrial Fibrillation Anticoagulation; SPINAF: Stroke Prevention in Nonrheumatic Atrial Fibrillation. Data from Reference [17].

adequate (>90%) efficacy and minimize risk of bleeding. However, this has not been tested in a clinical trial setting. Aspirin compared with placebo Several clinical trials have evaluated the effects of aspirin in AF, with doses ranging from 25 mg b.i.d. to 1,200 mg o.d. A meta-analysis of six randomized trials [15], involving a total of 3,119 patients (mean age 70 years, both primary and secondary prevention), showed that aspirin reduced the incidence of stroke by 22% (95% CI 2–38%, Table 6.2 and Figure 6.6(b)). In the placebo group, the average risk of stroke was 5.2% per year for primary prevention and 12.9% per year for secondary prevention. The absolute risk reduction was 1.5% per year for primary prevention (NNT=67) and 2.5% per year for secondary prevention (NNT=40). All six trials showed a reduced stroke incidence associated with aspirin; however, this effect was not proven statistically significant but for the SPAF study [18]. What would initially seem a paradox was shown to be the result of the increased proportion of nondisabling strokes recorded in this study (52%). Further analysis of the data from the three largest trials, where the severity of the stroke was recorded, demonstrated that the effect of aspirin was statistically significant for nondisabling strokes (relative risk reduction 62%, p=0.008) but not significant for disabling strokes (relative risk reduction 17%, p>0.2). Aspirin, therefore, seems effective in preventing noncardioembolic strokes in patients with AF but represents an inadequate measure in the case of cardioembolic strokes that are, on average, more disabling and share a worse prognosis. Noncardioembolic strokes make up a substantial portion of the events recorded in younger patients with “lone” AF, whereas cardioembolic strokes are usually seen in older patients (especially women), hypertensives, and those with a poor left ventricular function. Adjusted-dose warfarin compared with aspirin Given the inability of antiplatelet therapy to prevent disabling strokes and the reluctance of many physicians to prescribe warfarin because of the potential bleeding risk, further randomized controlled trials compared the efficacy and safety of the two different strategies. As shown in Table 6.2, adjusted-dose warfarin was compared with aspirin in five nonblinded randomized trials involving 2,837 patients and an average follow-up period of 2.2 years for each participant. The meta-analysis of these trials suggested that

Thrombosis in clinical practice

114

warfarin was preferable to aspirin and reduced the relative risk for stroke in patients with AF by an average 36% (95% CI 14–52%, Figure 6.6(c)). The results, however, were not consistent in all the trials. The superiority of the adjusted-dose warfarin regimen was greater in randomized trials that involved AF patients with a higher risk for a stroke (annual rate>6%); in those cases, the incidence of a disabling cardioembolic stroke was significantly increased when not using full-dose anticoagulation therapy. On the other hand, when applied to lower risk patients, warfarin failed to demonstrate significant further reduction of the relative risk for stroke, compared to aspirin, whereas the relative risk for both intracranial and extracranial major bleedings increased by more than 2-fold. Adjusted-dose warfarin compared with other antithrombotic regimens Three trials compared adjusted-dose warfarin (target INR 2–3.5) with low or fixed doses of warfarin (INR<1.5). The hypothesis tested was that mini-doses of warfarin would prove equally effective but far less dangerous regarding the hemorrhagic risk. Pooled analysis failed to support this, since adjusted-dose warfarin was associated with a 38% relative risk reduction compared to the mini-dose group (Table 6.2). Similarly disappointing were the results in two more trials comparing full-dose warfarin with aspirin plus low, fixed doses of warfarin [19, 20]. In the first trial [19], 1,044 patients presumed to be at high risk for stroke (annual rate>6%) were randomly assigned to adjusted-dose warfarin (mean achieved INR 2.4) or aspirin, 325 mg o.d., plus low fixeddose warfarin (mean achieved INR 1.3). The trial was stopped at interim analysis after a mean follow-up period of a year, as a result of a large 74% relative risk reduction in primary events in the adjusted-dose warfarin group (p<0.001). It was clear that for the specific high-risk group of patients, full-dose warfarin offered substantially better protection. Other schemes of combining aspirin with higher anticoagulation levels have failed to further increase the efficacy of the treatment while accentuating the hemorrhagic risk and therefore are not advised. Other antiplatelet drugs One small secondary prevention trial [21] suggested that dipyridamole (in a dose of 200 mg b.i.d.) may achieve reductions similar to aspirin (25 mg b.i.d.), regarding the stroke risk of patients with AF, when compared to placebo. The authors also suggested that, when combined, dipyridamole and aspirin exert an additive favorable effect outpacing the benefits of each one of the agents when prescribed alone. However, the observed benefits were not statistically significant. Another secondary prevention trial compared adjusted-dose warfarin with indobufen, a reversible cyclooxygenase inhibitor [22]. The primary outcome of the study was the combined incidence of nonfatal stroke (including intracerebral bleeding), pulmonary or systemic embolism, nonfatal myocardial infarction, and vascular death. The authors reported no statistically significant difference among the two groups who were well matched for confounding risk factors and suggested that indobufen may apply as an alternative in cases where warfarin cannot be used.

Antithrombotic therapy for atrial fibrillation

115

Oral direct thrombin inhibitors: the drugs of a new era? Oral direct thrombin inhibitors (DTI) are being developed as potential drugs for the treatment and prevention of thrombosis. Ximelagatran (a prodrug that is rapidly converted to the active form, melagatran) is the first oral agent in the new class of direct thrombin inhibitors. Melagatran is available to be given subcutaneously in the orthopedic surgery indication, followed by oral ximelagatran. The new agents have advantages beyond warfarin in that they can be given as a fixed dose orally with few drug/food interactions and are generally well-tolerated, and need not be monitored for coagulation. The effectiveness of ximelagatran in preventing thromboembolic events in patients with AF was tested in the Stroke Prevention using an Oral Thrombin Inhibitor in Atrial Fibrillation (SPORTIF) II pilot study, where 254 patients were randomly assigned to either adjusted-dose warfarin (aiming for an INR of 2.0–3.0) or ximelagatran and monitored for up to 12 weeks [23]. The results of this study showed no statistically significant difference in the incidence of stroke and thromboembolic events between the ximelagatran (60 mg b.i.d.) and the warfarin-treated groups. SPORTIF III The SPORTIF III trial determined whether ximelagatran can offer similar protection against stroke compared to warfarin, in a randomized, open-label treatment allocation study in 23 countries, comparing a fixed dose of 36 mg b.i.d. of ximelagatran with adjusteddose warfarin (aiming for an INR of 2.0–3.0) in patients with NVAF and at least one additional risk factor for stroke including previous stroke, hypertension or heart failure (Figure 6.7). The primary endpoints were all strokes (both ischemic or hemorrhagic) and systemic embolic events, based on intention to treat. Ximelagatran was found to be noninferior to warfarin (using intention-to-treat analysis) and even reached statistical superiority (using on-treatment analyses) to warfarin in preventing stroke and systemic embolic events. The on-treatment analysis found a 2.2% events/ year in the warfarin group compared to 1.3% events/year in the ximelagatran group, with a relative risk reduction of 41% (p=0.018). Furthermore, ximelagatran has been shown to cause less bleeding than warfarin in these patients. For the combined major adverse events of stroke, systemic embolization, major bleeding, and death, patients treated with ximelagatran have shown an event rate of 4.6% compared with 6.1% in the warfarin group (a relative risk reduction of 25%, p=0.022) [24–26].

Figure 6.7 SPORTIF III: Primary objective: stroke and SEE intentionto-

Thrombosis in clinical practice

116

treat analysis (SEE: systemic embolic event) [25]. SPORTIF V SPORTIF V was a double-blind study presented at AHA in 2003 that enrolled 3,922 patients at 409 sites in North America. As with the SPORTIF III trial, the patients received either ximelagatran 36 mg b.i.d., as a fixed dose, or dose-adjusted warfarin with a target INR of 2.0–3.0, and all had NVAF with at least one additional risk factor for stroke, including previous stroke, previous systemic embolism, hypertension, age 75 years or older, and a history of congestive heart failure. The preliminary results [27] showed that the primary endpoint (any stroke or systemic embolic event) occurred in 51 patients (1.6% per year) in the ximelagatran group compared with 37 patients (1.2% per year) in the warfarin group. This difference was not statistically significant (p=0.13). Major bleeding rates were very similar between the two patient groups, occurring in 2.4% per year of the ximelagatran group versus 3.1% per year of the warfarin group (p=0.16). However, the rates of minor and major bleeding combined were significantly lower in patients treated with ximelagatran (37% per year) than in patients treated with warfarin (47% per year) (p<0.001). As in the SPORTIF III trial, serum levels of alanine-aminotransferase (ALAT) rose to greater than three times the upper limit of normal (3×ULN) in 6% of patients receiving ximelagatran, compared with 0.8% of patients receiving warfarin (p<0.001). SPORTIF III and V combined analysis The pre-specified combined analyses of the SPORTIF III and V trials [28] were recently presented at the 13th European Stroke Conference in Germany in 2004. Among the 7,329 patients randomized in the SPORTIF III and V trials, there was a combined total of 91 primary events in the ximelagatran arm (2.5%) and 93 events in the warfarin arm (2.5%), with the annualized event rates of 1.6% per year and 1.7% per year, for ximelagatran and warfarin, respectively. The pooled rate of major bleeding was 2.5% among the ximelagatran-treated patients and 3.4% among patients assigned to warfarin. Among the non-hemorrhagic adverse events, ximelagatran was associated with abnormal (defined as 3×ULN) serum ALAT levels as compared with warfarin (pooled data: 6.1% vs 0.8%; p<0.0001). This was mostly asymptomatic, however, and typically occurred 2–6 months after treatment initiation, and returned towards baseline whether or not ximelagatran treatment was continued. Left atrial appendage occlusion: is there a role for surgery in thromboprophylaxis in AF? Previous studies using transesophageal echocardiography (TOE) have demonstrated that the vast majority of intra-atrial thrombi detected in AF patients are located in the LAA. Therefore, it is theoretically possible that occlusion of the LAA using a percutaneous device would greatly reduce the risk of stroke in patients with AE, even without the use

Antithrombotic therapy for atrial fibrillation

117

of any conventional antithrombotic treatment. Proven efficient and safe, this technique could easily apply to thousands of patients undergoing an open heart operation every year, where the incidence of AF and the coexistence of risk factors for the development of a stroke are particularly high. A thoracoscopic approach could also be considered in patients with both a high stroke risk from AF and a major contraindication to warfarin therapy. Clearly, for this approach to be established as an alternative in thromboprophylaxis from stroke in AF patients, a large randomized clinical trial would be required.

Putting evidence into practice: whom to treat? Based on reports from the trials mentioned earlier, and in terms of relative risk reduction figures for thromboembolism, warfarin demonstrates the most favorable outcomes in the majority of AF patients. However, in terms of absolute risk reduction (which is the most important in clinical practice and decisionmaking), the beneficial effect of full-dose anticoagulation is not uniformly deployed to all patients with AF. Those with organic heart disease, hypertension, diabetes, or some other predictors, who tend to represent the “high-risk” group, with increased annual rates of stroke if left untreated, tend to get the most benefit from warfarin use. On the other hand, young and relatively healthy (apart from the presence of AF) patients tend to get no or minimal additional benefits when anticoagulated. Quite a few studies have discussed the reluctance of a large number of physicians to commence patients with AF on warfarin therapy [27]. This attitude has been attributed to fears (of both patients and physicians) of hemorrhagic complications in an elderly population (the median age of AF patients is calculated at 75 years) as well as to logistical problems of INR monitoring. Indeed, many physicians fail to identify the aforementioned group of patients who would benefit the most from receiving anticoagulation, a practice that results in a potential increase of otherwise avoidable strokes and arterial thromboembolism. A series of both clinical and echocardiographic parameters has been examined by numerous studies and certain of them have emerged as independent predictors of ischemic stroke (Table 6.4). The mechanisms associating these risk factors to the increased incidence of ischemic stroke in AF are only partly understood. Hypertension represents one of the best-defined consistent predictors of thromboembolism, associated with reduced atrial appendage flow velocity, SEC in the left atrium, and thrombus formation. The association of hypertension with hypercoagulability has already been discussed in the pathophysiology section; however, this constitutes only one of the different pathways through which high blood pressure contributes to thrombotic events. Whether a sustained control of hypertension substantially reduces the thrombotic risk is currently unknown; thus, any patient with a history of hypertension should be considered to be at high risk. A medical history of previous stroke or TIA and advanced age are also described as major predictors of stroke. Abnormal rheology is often affected as documented by reduced LAA flow velocities and the presence of SEC. Furthermore, atherosclerosis is

Thrombosis in clinical practice

118

almost a universal finding in the aged population, and thrombotic markers increase with age, implying an age-related prothrombotic diathesis.

Table 6.4 Independent predictors of ischemic stroke in nonvalvular atrial fibrillation. (A) Clinical parameters Consistent predictors • Advanced age • Hypertension • Previous stroke or TIA Inconsistent predictors • Diabetes mellitus • Systolic blood pressure >160 mmHga • Women, especially older than 75 years of age • Postmenopausal hormone replacement therapy • Coronary artery disease • Thyrotoxicosis Factors decreasing the risk for stroke • Moderate to severe mitral regurgitation • Regular alcohol use (>14 drinks in 2 weeks) (B) Echocardiographic parameters • Left ventricular systolic dysfunction or heart failure • Valvular heart disease, especially mitral stenosis • Aortic arch atheromatous plaque a In some analyses, systolic blood pressure >160 mmHg remained an independent predictor after adjustment for hypertension. Data from Reference [5].

Echocardiography, although not routinely used to assess the stroke risk in AF, may prove valuable in refining our stratification. Left ventricular systolic dysfunction (as demonstrated by two-dimensional transthoracic echocardiography), valvular heart disease (especially mitral stenosis), and aortic arch atheromatous plaque are considered as independent predictive factors for stroke. Analysis of the echocardiographic data of 1,066 patients associated moderate to severe left ventricular systolic dysfunction with a 2.5% relative risk increase for stroke, compared with patients who have normal or mildly impaired left ventricular function [7] (Figure 6.5). In the same analysis, left atrial diameter on M-mode echocardiography failed to predict stroke or systemic emboli, as suggested by earlier studies [18]. Complex aortic plaque is also an independent predictor

Antithrombotic therapy for atrial fibrillation

119

of stroke in patients with AE, being present in 35% of those at high risk [29]. Other “high-risk features” demonstrated by TOE include SEC or LAA flow velocities, and patients with these lesions would derive benefits, in terms of stroke risk reduction, when treated with full-dose warfarin (Figure 6.8). Combining these clinical and echocardio graphic parameters into an accurate prognostic stratification scheme has proved an extremely complex issue. Table 6.5 summarizes such attempts published over the last decade, based on analyses of prospectively followed cohorts of patients with AF in clinical trials. Although many of the criteria overlap, there is no clear consensus defining the different risk strata, resulting in confusion regarding the optimal treatment for each patient. To give an example, men older than 75 years, without a history of hypertension, diabetes, or any thromboembolic event are considered to be at high risk according to the American College of Chest Physicians (both 1998 and 2001 Consensuses) and at low risk according to the Stroke Prevention for Atrial Fibrillation (SPAF) Investigators (Table 6.5). The actual observed risk for stroke for the specific group of patients is 1.2% (95% CI 0.4–3.2%), which seems rather consistent with the “low-risk” definition. Similarly, another prospective study assessing the predictive value of different stroke risk stratification schemes in 259 elderly people with NVAF, demonstrated that the proportion of those considered to share a “high risk” may vary from 45% to 76% depending on the criteria used [31]. The uncertainty and the confusion therefore, on how to treat these patients, remains.

Figure 6.8 Transesophageal echocardiography demonstrating the presence of thrombus in the left atrium of a 75-year-old patient with atrial fibrillation (RA: right atrium; LA: left atrium; RV: right ventricle; LV: left ventricle; T:thrombus) [30]. Table 6.5 Different risk stratification schemes for primary prevention of stroke on nonvalvular atrial fibrillation. Criteria (year)

High risk

Atrial Fibrillation Investigators (1994)

• Age ≥65years • History of hypertension • Diabetes mellitus • Previous stroke or TIA

Stroke Prevention in Atrial Fibrillation (SPAF III, 1995)

• Women >75 years • SBP >160 mmHg • Previous stroke or TIA

Intermediate risk

Low risk • Age <65 years • No high-risk criteria

• History of hypertension • No high-risk criteria

• No high-risk criteria • No history of hypertension

Thrombosis in clinical practice

120

• LV dysfunctiona American College of • Age >75 years Chest Physicians • History of Consensus (1998) hypertension • Previous stroke or TIA • LV dysfunctionb • >1 moderate risk factor

• Age 65–75 years • Diabetes mellitus • Coronary artery disease • Thyrotoxicosis

• Age <65 years • No rlsk factors

Stroke Prevention in Atrial Fibrillation exploratory analysis (1999)

• Hypertension plus age ≤75 • Diabetes mellitus • No high-risk criteria

• No high-risk criteria • No moderate-risk criteria

• Age 65–75 years • Diabetes mellitus • Coronary artery disease with preserved LV function

• Age >65 years • No high- or moderate-risk factors and no clinical or echocardiographic evidence of cardiovascular disease

• Women >75 years • SBP >160 mmHg • Hypertension plus age >75 • Previous stroke or TIA

American College of • Age >75 years Chest Physicians • History of Consensus (2001) hypertension • Previous stroke or TIA or systemic embolism • LV dysfunctiona • Rheumatic mitral valve disease, prosthetic valve • >1 moderate risk factor

a Impaired left ventricular function included recent congestive heart failure or left ventricular fractional shortening ≤25% by M-mode echocardiography. b Moderate to severe left ventricular systolic dysfunction on echocardiography or recent congestive heart failure. TIA: transient ischemic attack; SBP: systolic blood pressure; LV: left ventricle, Data from Reference [16],

A practical stratification scheme is listed in Table 6.6. According to this, “low-risk” AF patients, with an annual stroke rate of about 1%, are considered to be those aged below 65 years, with no history of hypertension, diabetes, structural heart disease, or systemic thromboembolism, and can be adequately managed with aspirin alone. The use of warfarin instead, would only prevent 4 strokes per year per 1,000 patients treated (NNT=250, Table 6.7). On the other hand, the “high-risk” group, with an annual stroke rate of more than 8% if untreated, consists of hypertensives and/or diabetic patients aged more than 75 years, or patients of any age with a previous history of thromboembolism, structural heart disease, or evidence of left ventricular systolic dysfunction. The beneficial effect of warfarin (over aspirin) in this case is overwhelming, preventing up to 40 strokes per year for every 1,000 patients treated (NNT=25, Table 6.7). The AF patients who do not fit to the criteria of either one of the two previous cases represent the

Antithrombotic therapy for atrial fibrillation

121

“moderate risk” group with an annual stroke rate estimated at 4%. Clear-cut evidence is lacking here and they can be managed with either aspirin or warfarin. Clinical judgment as well as patient preferences may have a role in choosing the appropriate scheme in this case.

Table 6.6 Practical risk stratification scheme in nonvalvular atrial fibrillation—suggested anticoagulation. Risk to be reassessed regularly. (A) High risk (annual risk of CVA: 8–12%) • All patients aged 75 years and over with diabetes and/or hypertension • All patients with previous TIA, CVA, or systemic embolism • All patients with clinical evidence of valve disease, heart failure, thyroid disease, and/or impaired left ventricular function on echocardiographya Treatment: warfarin (target INR 2.0–3.0) if no contraindications and possible in practice (B) Moderate risk (annual risk of CVA: 4%) • All patients over 65 years not in high-risk group • All patients under 65 years with clinical risk factors: diabetes, hypertension, peripheral vascular disease, ischemic heart disease Treatment: either warfarin (INR 2.0–3.0) or aspirin (300 mg). In view of insufficient clear-cut evidence treatment may be decided on individual cases. Referral and echocardiography may help (C) Low risk (annual risk of CVA: 1%) • All patients under 65 years with no history of hypertension, diabetes, systemic embolism, or other clinical risk factors Treatment: aspirin (300 mg o.d.) a Echocardiocardiography not needed for routine risk assessment but refines clinical risk stratification in case of impaired left ventricular function and valve disease. A large atrium per se is not an independent risk factor on multivariate analysis. Data from Reference [32].

Table 6.7 Effect of risk stratification on stroke reduction by warfarin compared with aspirin or no antithrombotic treatment in atrial fibrillation. Risk stratification

Primary prevention

Annual stroke rate without any therapy (%)

Annual stroke rate with aspirin therapy (%)

Treatment with warfarin instead of aspirin Number needed to treat for 1 year to prevent 1 stroke (N)

Number of strokes saved yearly per 1,000 given warfarin (N)

Thrombosis in clinical practice

122

• Low risk

1

<1

250

4

• Moderate risk

4

3

83

12

• High risk

8

6

42a

24

Secondary prevention

12

10

25

40

a Calculations are based on a 40% relative risk reduction with adjusted-dose warfarin over aspirin from meta-analysis. It is possible that this overestimates the benefit of warfarin therapy in low-risk patients and importantly imderestimates it for high-risk patients, Data from Referenee [16],

It should be underlined at this point that the stratification schemes are based on clinical assessment and that the role of echocardiography is generally only supportive, by helping refine risk stratification—indeed, clinical criteria can identify most patients at moderatehigh risk. The issue of recurrent (paroxysmal) AF has already been mentioned as sharing the same risks for stroke with permanent AF. The risk stratification scheme, therefore, applies equally to these patients also.

Rhythm versus rate control For many years rhythm control (restoration of the normal sinus rhythm) was preferred by most physicians compared to rate control (controlling the ventricular rate without pursuing cardioversion) since it was considered to be associated with better outcomes. Both approaches are equally accepted today and the decision is mainly based upon the specific clinical circumstances of each case. The recently published results of two large randomized trials (RACE, AFFIRM [33, 34]), enrolling a total of 4,500 patients for a mean follow-up period of 3.5 and 2.3 years respectively, demonstrated an almost significant trend towards a lower all-cause mortality in the rate control group (Figure 6.9). The functional status and the quality of life were not affected by the choice of either therapeutic approach, while the risk of embolic events was equally distributed in both groups, with a trend (28%) towards increased ischemic strokes in patients randomized to the “rhythm control” arm (Table 6.8). In fact, most of the embolic events occurred when warfarin had been stopped or when the INR was sub-therapeutic, indicating that even when normal sinus rhythm is restored, those patients characterized as being in “high risk” for

Antithrombotic therapy for atrial fibrillation

123

Figure 6.9 All-cause mortality plotted against time in the AFFIRM study, where 4,060 patients with AF that was likely to be recurrent were randomly assigned to either rhythm or rate control. Although there was a lower trend in the rate control group, no statistical significance was observed [34]. thromboembolic events (according to the criteria mentioned earlier), still required chronic warfarin anticoagulation therapy. The problem of recurrence of the arrhythmia despite antiarrhythmic therapy (often asymptomatic and therefore undiagnosed) and the possible co-existence of other predisposing factors for thromboembolism (like complex aortic plaque or left ventricular systolic dysfunction) explain the outcomes of the studies and underline the need for chronic and adequate anticoagulation with a target INR of 2.0–3.0. Indeed, the AFFIRM stroke substudy [35] confirmed that warfarin significantly reduced the risk of stroke by 69% (p<0.0001), even in patients in sinus rhythm, and the presence of sinus rhythm per se did not ensure freedom from strokes. Patients, who are highly symptomatic, such as those with diastolic dysfunction or hypertrophic cardiomyopathy, probably represent the only cases where rhythm control is beneficial compared to rate control and cardioversion should be attempted more aggressively. For those aged over 65, especially those with one or more risk factors for stroke (hypertension,

Table 6.8 Rate versus rhythm control and ischemic strokes. n AFFIRM

4,917

Rate control (%)

Rhythm control (%) 5.7

RR (95% CI) 7.3

1.28 (0.95–1.72)

Thrombosis in clinical practice

124

RACE

522

5.5

7.9

1.44 (0.75–2.78)

STAF

266

1.0

3.0

3.01 (0.35–25.30)

PIAF

252

0.8

0.8

1.02 (0.73–2.16)

Total

5,957

5.0

6.5

1.28 (0.98–1.66)

AFFIRM: Atrial Fibrillation Follow-up Investigation of Rhythm Management; RACE: A Comparison of Rate Control and Rhythm Control in Patients with Recurrent Persistent Atrial Fibrillation (RACE); STAF: Strategies of Treatment of Atrial Fibrillation; PIAF: Pharmacological Intervention in Atrial Fibrillation. Data from Reference [36].

diabetes, left ventricular systolic dysfunction, or prior stroke), reversion to normal sinus rhythm offers no definite advantage, and rate control with adequate anticoagulation seems to be the preferable strategy. The 2001 guidelines from the 6th American College of Chest Physicians Consensus Conference on Anticoagulation* whenever cardioversion of AF is decided are summarized in Table 6.9. Transesophageal echocardiography offers an alternative in early cardioversion of patients presenting with AF of longer than 48 hours duration, with a safety profile similar to, but not safer than, conventional therapy. The technique allows cardioversion to be performed immediately without prior long-term anticoagulation, providing that no thrombi are detected in the LAA. It shares the advantages of earlier restoration of the left atrium mechanical function and it has proven to be cost-effective in terms of medical finances. However, oral anticoagulation is compulsory post-procedure and for a timespan of at least 4 weeks to prevent de novo emboli formation in the cardioverted, “stunned,” left atrium [37, 38] (Figure 6.10). In patients post-cardioversion, anticoagulation should be considered for a longer period in those with stroke risk factors or those at risk of AF recurrence, especially since many relapses occur asymptomatically.

Table 6.9 Anticoagulation for elective cardioversion. (A) Atrial fibrillation • For AF lasting >48 hours administer oral anticoagulant therapy (target INR 2.5; range 2.0–3.0) for 3 weeks before and at least 4 weeks after elective cardioversion (grade 1C+). • Alternatively, AF patients may be started on anticoagulation (intravenous heparin followed by warfarin) then undergo transesophageal echocardiography and have cardioversion performed without delay if no thrombi are seen (grade 1C). For these patients, adjusted-dose warfarin therapy should still be continued until normal sinus rhythm has been maintained for at least 4 weeks. • In AF lasting <48 hours, the risk of embolism following cardioversion appears to be low. However, anticoagulation is recommended during the pericardioversion period (grade 2C). • Treatment of potential precipitants of AF (e.g. thyrotoxicosis, pneumonia, congestive heart failure) should be completed prior to attempting elective DC cardioversion. (B) Atrial flutter and supraventricular tachycardia • Atrial flutter should be treated similarly to atrial fibrillation, regarding anticoagulation therapy

Antithrombotic therapy for atrial fibrillation

125

(grade 2C). • In the absence of prior thromboembolism, antithrombotic therapy for cardioversion of supraventricular tachycardia is not recommended (grade 2C). Data from Reference [39]. *

Since the submission of this chapter, the American College of Chest Physicians Guidelines have been updated, but remain broadly similar [39].

Figure 6.10 Schematic of TOE-guided early cardioversion protocol for patient in whom cardioversion is desired. Decision-analytic models demonstrate superior effectiveness with omission of transthoracic echocardiogram. Issues of antiarrhythmic therapy are beyond the scope of this chapter (PTT: partial thromboplastin time; TOE: Transesophageal echocardiography; LA: left atrium; LAA: left atrial

Thrombosis in clinical practice

126

appendage; RA: right atrium; RAA: right atrial appendage; and INR: International Normalized Ratio) [38]. References [1] Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: The Framingham Study. Stroke 1991; 22:983–8. [2] Lamassa M, Di Carlo A, Pracucci G, et al. Characteristics, outcome and care of stroke associated with atrial fibrillation in Europe: data from a multicenter multinational hospital-based registry (The European Community Stroke Project). Stroke 2001; 32:392–8. [3] Fuster V, Rydén LE, Asinger RW, et al. ACC/ AHA/ESC guidelines for the management of patients with atrial fibrillation: executive summary. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines and Policy Conferences (Committee to Develop Guidelines for the Management of Patients With Atrial Fibrillation). J Am Coll Cardiol 2001; 38:1231–65. [4] Krahn AD, Manfreda J, Tate RB, et al. The natural history of atrial fibrillation: incidence, risk factors, and prognosis in the Manitoba Follow-Up Study. Am J Med 1995; 98:476–84. [5] Lip GYH, Hart RG, Conway DSG. ABC of antithrombotic therapy: antithrombotic therapy for atrial fibrillation. BMJ 2002; 325:1022–5. [6] Goldsmith I, Kumar P, Carter P, et al. Atrial endocardial changes in mitral valve disease: a scanning electron microscopy study. Am Heart J 2000; 140:777–84. [7] Zabalgoitia M, Halperin JL, Pearce LA, et al. Transesophageal echocardiographic correlates of clinical risk of thromboembolism in nonvalvular atrial fibrillation. J Am Coll Cardiol 1998; 31: 1622–6. [8] Manning WJ, Silverman DI, Katz SE, et al. Impaired left atrial mechanical function after cardioversion: relation to the duration of atrial fibrillation. J Am Coll Cardiol 1994; 23:1535– 40. [9] Atrial Fibrillation Investigators. Echocardiographic predictors of stroke in patients with atrial fibrillation: a prospective study of 1066 patients from 3 clinical trials. Arch Intern Med 1998; 158: 1316–20. [10] The Stroke Prevention in Atrial Fibrillation Investigators. Predictors of thromboembolism in atrial fibrillation: II. Echocardiographic features of patients at risk. Ann Intern Med 1992;116:6– 12. [11] Heppell RM, Berkin KE, McLenachan JM, et al. Haemostatic and haemodynamic abnormalities associated with left atrial thrombosis in non-rheumatic atrial fibrillation. Heart 1997; 77:407–11. [12] Lip GY, Lowe GD, Rumley A, et al. Increased markers of thrombogenesis in chronic atrial fibrillation: effects of warfarin treatment. Br Heart J 1995; 73:527–33. [13] Lip GYH, Blann A. von Willebrand factor: a marker of endothelial dysfunction in vascular disorders? Cardiovasc Res 1997; 34:255–65. [14] Lip GYH, Gibbs CR. Does heart failure confer a hypercoagulable state? Virchow’s triad revisited. J Am Coll Cardiol 1999; 33:1424–6. [15] Hart RG, Benavente O, McBride R, et al. Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: a meta-analysis. Ann Intern Med 1999; 131:492–501. [16] Hart G, Halperin JL. Atrial fibrillation and thromboembolism: a decade of progress in stroke prevention. Ann Intern Med 1999; 131:688–95.

Antithrombotic therapy for atrial fibrillation

127

[17] Gulløv AL, Koefoed BG, Petersen, P. Bleeding during warfarin and aspirin therapy in patients with atrial fibrillation: the AFASAK 2 study. Arch Intern Med 1999; 159:1322–8. [18] Hart RG, Halperin JL, Pearce LA, et al. Lessons from the Stroke Prevention in Atrial Fibrillation Trials. Ann Intern Med 2003; 138:831–8. [19] Str oke Prevention in Atrial Fibrillation Investigator s. Adjusted-dose warfarin versus lowintensity, fixeddose warfarin plus aspirin for high-risk patients with atrial fibrillation: Stroke Prevention in Atrial Fibrillation III randomised clinical trial. Lancet 1996; 348:633–8. [20] Gulløv AL, Koefoed BG, Petersen P, et al. Fixed minidose warfarin and aspirin alone and in combination versus adjusted-dose warfarin for stroke prevention in atrial fibrillation: Second Copenhagen Atrial Fibrillation, Aspirin, and Anticoagulation Study. Arch Intern Med 1998; 158:1513–21. [21] Diener HC, Cunha L, Forbes C, et al. European Stroke Prevention Study. 2. dipyridamole and acetylsalicylic acid in the secondary prevention of stroke. J Neurol Sci 1996; 143:1–13. [22] Morocutti C, Amabile G, Fattapposta F, et al. Indobufen versus warfarin in the secondary prevention of major vascular events in nonrheumatic atrial fibrillation. SIFA (Studio Italiano Fibrillazione Atriale) Investigators. Stroke 1997; 28:1015–21. [23] Petersen P, Grind M, Adler J, et al. Ximelagatran versus warfarin for stroke prevention in patients with nonvalvular atrial fibrillation. SPORTIF II: a dose-guiding, tolerability, and safety study. J Am Coll Cardiol 2003; 41:1445–51. [24] Halperin JL and the Executive Steering Committee, on behalf of the SPORTIF III and V Study Investigators. Ximelagatran compared with warfarin for prevention of thromboembolism in patients with nonvalvular atrial fibrillation: rationale, objectives, and design of a pair of clinical studies and baseline patient characteristics (SPORTIF III and V). Am Heart J 2003; 146:431–8. [25] Executive Steering Committee on behalf of the SPORTIF III Investigators. Stroke prevention with the oral direct thrombin inhibitor ximelagatran compared with warfarin in patients with nonvalvular atrial fibrillation (SPORTIF III): randomised controlled trial. Lancet 2003; 362:1691–8. [26] Diener HC for the Executive Steering Committee on behalf of the SPORTIF Investigators. Stroke prevention with the oral direct thrombin inhibitor ximelagatran in patients with nonvalvular atrial fibrillation: pooled analysis of the SPORTIF III and V trials. Cerebrovasc Dis 2004; 17(Suppl. 5):16. [27] Executive Steering Committee on behalf of the SPORTIF V Investigators. Efficacy and safety study of the oral direct thrombin inhibitor ximelagatran compared with dose-adjusted warfarin in the prevention of stroke and systemic embolic events in patients with atrial fibrillation (SPORTIF V). Circulation 2003; 108:2723. [28] Olsson SB, for the Executive Steering Committee on behalf of the SPORTIF Investigators. Stroke prevention with the oral direct thrombin inhibitor ximelagatran in patients with nonvalvular atrial fibrillation: final results of the pooled analysis of the SPORTIF III and V trials. Eur Heart J 2004; 25 (Suppl.):Abstract P 3703. [29] Devereaux PJ, Anderson DR, Gardner MJ, et al. Differences between perspectives of physicians and patients on anticoagulation in patients with atrial fibrillation: observational study. BMJ 2001; 323:1218–22. [30] Cujec B, Marc Baltzan M. Large left atrial thrombus. NEJM 1993; 328:771. [31] Dávila-Román VG, Murphy SF, Nickerson NJ, et al. Atherosclerosis of the ascending aorta is an independent predictor of long-term neurologic events and mortality. J Am Coll Cardiol 1999; 33:1308–16. [32] Lip GYH. Thromboprophylaxis for atrial fibrillation. Lancet 1999; 353:4–6. [33] Feinberg WM, Kronmal RA, Newman AB, et al. Stroke risk in an elderly population with atrial fibrillation. J Gen Intern Med 1999; 14:56–9. [34] The Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) Investigators. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med 2002; 347:1825–33.

Thrombosis in clinical practice

128

[35] Sherman DG, Kim S, Boop B, et al. The occurrence and characteristics of stroke events in the AFFIRM study. Neurology 2003; 60(Suppl.1): Abstract A326. [36] Verheugt FWA, van Gelder IC, Wyse DG, et al. Paradoxical increase in strokes in the randomized trials on rate versus rhythm control in atrial fibrillation. J Am Coll Cardiol 2003; 41(Suppl. 1):130, Abstract 1210–19. [37] van Gelder IC, Hagens VE, Bosker HA, et al. A comparison of rate control and rhythm control in patients with recurrent persistent atrial fibrillation. N Engl J Med 2002; 347:1834–40. [38] Silverman DI, Manning WD. Role of echocardiography in patients undergoing elective cardioversion of atrial fibrillation. Circulation 1998;8: 479–86. [39] Singer DE, Albers GW, Dalen JE, et al. Antithrombotic therapy in atrial fibrillation: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004; 126(3 Suppl.): 429S–456S.

7 Valves Walter Ageno and Alexander GG Turpie Evaluation of the patient at risk Patients with valvular heart disease and in particular patients with prosthetic heart valves are at increased risk of systemic thromboembolism. In the absence of antithrombotic therapy, systemic embolism and stroke have been reported in between 5% and 50% of patients, depending upon the valve site, the type of valve replacement, and the presence of comorbid conditions [1, 2]. With the use of anticoagulants, the rate of systemic embolism has been reduced to 1–3% per year [3]. However, the intensity of anticoagulation to effectively reduce the risk of thromboembolic complications must be balanced against the risk of bleeding, which in turn is related to the level of anticoagulation used and to individual patient risk factors [4]. Therefore, risk factors that increase the incidence of systemic embolism as well as risk factors that increase the risk of bleeding complications must be considered when defining the need for starting antithrombotic therapy and the intensity of anticoagulation in patients with valvular heart disease and prosthetic heart valves. These factors include age, smoking, hypertension, diabetes, hyperlipidemia, type and severity of valve lesion, presence of atrial fibrillation, heart failure or low cardiac output, size of the left atrium (over 50 mm on echocardiography), previous thromboembolism, previous major bleeding, events or clinical conditions at high risk for bleeding, and abnormalities of the coagulation system including hepatic failure [5]. Furthermore, the type, number, and location of prostheses implanted must be considered. Mechanical prostheses are more thrombogenic than bioprostheses or homografts, and hence patients with mechanical valves require lifelong anticoagulant therapy. Moreover, the intensity of treatment varies according to the type of mechanical prosthesis implanted. First generation mechanical valves, namely the StarrEdwards caged ball valve and Bjork-Shiley standard valves, have a high thromboembolic risk [6]; single tilting disc valves have an intermediate thromboembolic risk; and the newer (second and third generation) bileaflet valves have low thromboembolic risks [7]. Finally, thromboembolic events are commoner with prosthetic mitral valves than aortic valves and in patients with double replacement valves compared with those with single prostheses [8]. Table 7.1 gives the different types of prosthetic valves and their thrombogenicity.

Thrombosis in clinical practice

130

Table 7.1 Types of prosthetic valves and thrombogenicity. Type of valve

Model

Thrombogenicity

Mechanical Caged ball

Starr-Edwards

Single tilting disc

Bjork-Shiley, Medtronic-Hall

Bileaflet

St Jude Medical, Sorin Bicarbon, Carbomedics

++++ +++ ++

Bioprosthetic Heterografts

Carpentier-Edwards, Tissue Med (Aspire), Hancock II

Homografts

+ to + + +

From Goldsmith et al. [5].

Mechanical prosthetic heart valves Patients with mechanical heart valve prostheses require lifelong anticoagulation therapy. Randomized trials have shown that oral anticoagulants are effective in reducing the risk of systemic embolism in patients with mechanical prosthetic valves when given at lower intensity than has been used in the past. The last guidelines of the American College of Chest Physicians (ACCP) [3], published in 2001, recommend two intensity regimes of long-term oral anticoagulant treatment according to the site of the mechanical prosthesis and to the presence of concomitant risk factors. A lower International Normalized Ratio (INR) range between 2.0 and 3.0 is recommended for patients with a bileaflet valve (St Jude Medical or Carbomedics) or a tilting disc valve (Medtronic-Hall) in the aortic position, which are in normal sinus rhythm and have a left atrium of normal size. These recommendations, are based on the results of long-term follow-up studies [7, 9–11]. In particular, a recent study from France [9] has confirmed the efficacy of a less intense level of anticoagulation following mechanical heart valve replacement. In this study, 433 patients with mechanical prostheses were randomized to anticoagulant therapy monitored to achieve an INR of 2.0–3.0 or 3.0–4.5 and followed for 2.2 years. The majority of patients in this trial had aortic valve replacements and were in sinus rhythm. Thromboembolic outcome events, either clinical events or asymptomatic central nervous system abnormalities proven on CT scan, occurred in 10 of 185 (5.3%) patients in the low-intensity group and 9 of 192 (4.7%) patients in the high-intensity group (p=0.78). Importantly, there was a statistically significant difference in the rate of bleeding complications between the two groups. Bleeding events occurred in 34 patients (18.1%) in the lower-intensity group compared with 56 patients (29.2%) in the high-intensity group (p<0.01). An INR range between 2.5 and 3.5 is recommended for mechanical valves in the mitral position [3]. This recommendation was based on two prospective studies that demonstrated that anticoagulant therapy maintained within this target INR range was as

Valves

131

effective as a more intense level of anticoagulation, but with less bleeding. In the first study [12] there was no difference in the frequency of major embolic events in the patients treated with a high-intensity regimen (INR 9.0) compared with patients treated with a less intense (INR 2.65) anticoagulant regimen (4.0 vs 3.7 embolic episodes per 100 patient-years, respectively). However, there was significantly less bleeding in the less intense group (6.2 vs 12.1 hemorrhagic episodes per 100 patient years; p<0.02). The second study [13] compared low-intensity (INR 2.0–2.99) with high-intensity (INR 3.0– 4.5) oral anticoagulants in patients with mechanical valves, all of whom were treated with aspirin (330 mg b.i.d.) in combination with dipyridamole (75 mg o.d.). In this study, one transient ischemic attack occurred in the low-intensity group and two in the high-intensity group. There were significantly fewer bleeding events in the low-intensity group, in which three episodes occurred compared with 12 in the high-intensity group (p<0.02). The finding that a lower-intensity anticoagulation is as effective as, and safer than, higher intensities that were previously recommended was recently challenged by the results of a meta-analysis [14], in which two intensities of anticoagulation, low, defined by a mean target INR of 3.0 or lower, and high, defined by a mean target INR above 3.0, were compared. The authors identified 35 studies that included a total of 23,145 patients with a 108,792 patient-years follow-up. The combined endpoint of thromboembolic and bleeding events were significantly lower in the high-intensity group than in the lowintensity group for both aortic and mitral valve prostheses, with a relative risk of 0.94 (p=0.0067) and 0.84 (p<0.0001), respectively. The authors concluded that both aortic and mitral valves benefit from a treatment strategy with a target INR greater than 3.0. The addition of antiplatelet agents to oral anticoagulants has been advocated as another possible approach to the treatment of patients with mechanical valves, or highrisk patients with tissue valves, to reduce further the risk of major systemic embolism. The combination of aspirin and oral anticoagulants has been used in the treatment of patients with heart valve replacement with a significant reduction in embolic complications, but with an increased risk of bleeding complications [15]. In the early studies reported to date, aspirin was used in high doses (approximately 1 g o.d.), and in most cases the bleeding with the combination of high-dose aspirin and high-dose oral anticoagulants was gastrointestinal [16]. There is good evidence that gastrointestinal irritation and hemorrhage is dose-dependent over a range of 100–1,000 mg of aspirin o.d. and that the antithrombotic effects of aspirin are independent of the dose over this range. One completed study [17] compared lowdose aspirin combined with warfarin in the treatment of patients with mechanical heart valve replacements to determine whether low-dose aspirin would result in an improved antithrombotic effect, without the same high risk of bleeding that has been reported for the combination of oral anticoagulants with high-dose aspirin. This was a double-blind, randomized trial to compare the relative efficacy and safety of aspirin (100 mg o.d.) with placebo in the prevention of systemic embolism or vascular death in patients with mechanical heart valve replacement or highrisk patients with tissue valves who had atrial fibrillation or a history of thromboembolism. Three hundred and seventy patients were treated with oral anticoagulant therapy (warfarin: INR 3.0–4.5) and randomized to receive aspirin (186 patients) or placebo (184 patients) and followed for up to 4 years (average 2.5 years). The outcomes of the study were systemic embolism, valve thrombosis, vascular death, and hemorrhage. Systemic embolism or vascular death occurred in 6 (3.2%) of the aspirin-

Thrombosis in clinical practice

132

treated patients and 24 (13.0%) of the placebo-treated patients (risk reduction (RR) 77.2%; 90% confidence interval (CI) 51.7–89.2%; p=0.0002). The corresponding rates for systemic embolism or death from any cause were 13 (7.0%) and 33 (17.9%), respectively (RR 64.7%; 90% CI 39.6–79.5%; p=0.0005); for vascular death 2 (1.1%) and 13 (7.1%), respectively (RR 85.4%; 90% CI 49.8–95.9%; p=0.0015); and for death from any cause 9 (4.8%) and 22 (12.0%), respectively (RR 62.7%; 90% CI 44.5–80.5%; p=0.0048). Major bleeding events occurred in 24 (12%) of the aspirin-treated patients compared with 19 (10.3%) in the placebo-treated patients (absolute difference 2.6%; 90% CI 3.4–8.3%; p=0.2710). The results of this study demonstrated that in patients with mechanical valve replacement or high-risk patients with tissue valve replacement, the addition of aspirin (100 mg o.d.) to oral anticoagulation therapy (warfarin: INR 3.0–4.5) reduced mortality, vascular mortality, and systemic embolism, but with some increase in minor bleeding. In a recent study in patients with mechanical prostheses [18], it was shown that low-dose aspirin (100 mg o.d.) was as effective as high-dose aspirin (650 mg o.d.) in combination with oral anticoagulants at a target INR of 2.0–3.0, but with a reduced risk of bleeding. Therefore, the addition of low-dose aspirin (80–100 mg o.d.) is now recommended for patients with concomitant atrial fibrillation or other additional risk factors and patients who had thromboembolic events despite adequate oral anticoagulant therapy. Ticlopidine may also be useful as an adjunct to oral anticoagulants, but the data are less solid since the one study in which it has been evaluated was not randomized [19]. There are currently no available data on the association of aspirin and clopidogrel in this setting. The 2001 ACCP recommendations for mechanical valves are summarized in Table 7.2.

Bioprosthetic heart valves The risk of thromboembolism is less with uncomplicated bioprosthetic valves than with

Table 7.2 Antithrombotic therapy in mechanical heart valve replacement. INR Uncomplicated bileaflet aortic

2.0–3.0

Uncomplicated tilting disk aortic

2.0–3.0

Uncomplicated aortic and atrial fibrillation

2.5–3.5 or 2.0–3.0+aspirin 80–100 mg o o.d.

Uncomplicated bileaflet mitral

2.5–3.5

Uncomplicated tilting disk mitral

2.5–3.5

Additional risk factors

2.5–3.5+aspirin 80–100 mg o.d.

Systemic embolism

2.5–3.5+aspirin aspirin 80–100 mg o.d

Caged ball or caged disk valve

2.5–3.5+aspirin 80–100 mg o o.d.

Valves

133

From Stein et al. [3],

mechanical valves [20–23]. Oral anticoagulants, including warfarin, have been shown to be effective and safe when used at a targeted INR of 2.0–3.0 in such patients based on the results of one prospective clinical trial [24]. This study compared two intensities of anticoagulation to determine the safety and efficacy of a less intense anticoagulant regimen in patients following tissue valve replacement. One hundred and eight patients were randomized to standard anticoagulant control (INR 3.0–4.5), and 102 patients to a less intensive regimen (INR 2.0–2.5). Treatment was continued for 3 months. In this study there was no difference in the frequency of major systemic emboli (1.8% in both) between the two treatment groups, but there were significantly fewer major hemorrhagic complications (0.0% vs 4.5%; p=0.034) and total hemorrhagic complications (5.4% vs 14.6%; p=0.042) in the low-intensity group (INR 2.0–3.0) compared to the high-intensity group. This level of anticoagulation (INR 2.0–3.0) is recommended by the ACCP for patients with tissue valve replacement [3]. The risk of thromboembolism is limited mainly to the first 3 months postoperatively in uncomplicated patients with tissue valves, but is present indefinitely in patients with atrial fibrillation [25]. A low ejection fraction, an enlarged left atrium, previous history of venous thromboembolism, and the presence of a pacemaker also increase the risk of thromboembolic complications [26, 27]. Consequently, in uncomplicated patients with mitral bioprosthetic valves, anticoagulant therapy is recommended for 3 months whereas long-term therapy is indicated in patients with atrial fibrillation, those with an atrial thrombus detected at echocardiography, and those who develop a systemic embolus [3]. Patients with uncomplicated bioprosthetic valves in the aortic position are at very low risk of systemic embolism and some authorities therefore suggest that they do not require anticoagulant therapy, although this recommendation remains controversial [3, 28]. Long-term treatment with aspirin 80 mg o.d. following 3 months of oral anticoagulant therapy is likely to be beneficial to prevent subsequent thromboembolic events in patients with uncomplicated bioprosthetic valves [3]. The current recommendations by the ACCP for patients with tissue valves are shown in Table 7.3.

Table 7.3 Antithrombotic theraphy in bioprosthetic heart valve replacement. INR

Duration

Mitral

2.0–3.0

3 months

Aortic

2.0–3.0

3 months

Atrial fibrillation

2.0–3.0

Long-term

Left atrial thrombosis

2.0–3.0

Long-term (duration uncertain)

Permanent pacemaker

2.0–3.0

Optional

Systemic embolism

2.0–3.0

3–12 months

Normal sinus rhythm

Long-term aspirin (80 mg o.d.)

From Stein et al. [3].

Thrombosis in clinical practice

134

Starting anticoagulant treatment following heart valve replacement Oral anticoagulant treatment is usually started on the day after implantation, with or without heparinization. As the thromboembolic risk is highest in the early postoperative period, it is advisable to give heparin and to continue it until the oral anticoagulant treatment achieves the target INR. A high sensitivity to warfarin in the immediate postoperative phase during oral anticoagulation induction has been reported [29, 30]. These finding suggested the need for lower starting doses of warfarin to regularly achieve the therapeutic range (i.e. 2.5–3.0 mg instead of 5.0 mg) in most patients.

Native valve disease The incidence of systemic emboli in patients with mitral stenosis depends on the presence or absence of concomitant risk factors. Emboli can occur in approximately 8% of the patients with mitral stenosis and sinus rhythm, and in approximately 31.5% of the patients with concomitant atrial fibrillation [31]. Other risk factors include the severity of stenosis, the patient’s age, size of the left atrium, and the presence of spontaneous echocontrast or echocardiographic evidence of left atrial appendage thrombus [32], Currently, oral anticoagulation with a target INR of 2.5 (range 2–3) is recommended in all patients with a history of systemic embolism, atrial fibrillation, and a left atrial diameter greater than 55 mm [33]. In the presence of other concomitant risk factors, the therapeutic decision should be made on an individual basis. Similarly, in patients with mitral regurgitation treatment is indicated in the presence of congestive cardiac failure, marked cardiomegaly with low cardiac output, and an enlarged left atrium [5]. In the absence of cardiac failure, previous thromboemboli, or heart failure, antithrombotic therapy is not indicated in patients with isolated aortic or tricuspid valve disease [5]. Mitral valve prolapse per se does not require anticoagulant cover, although aspirin is recommended in patients with otherwise unexplained ischemic cerebrovascular events [33]. Oral anticoagulants are recommended in patients with atrial fibrillation [33].

Management of pregnant patients with prosthetic mechanical valves The management of prosthetic heart valves in pregnant women is a matter of intense debate. Warfarin can cause two potential fetal complications, teratogenicity and bleeding, and thus it is commonly not recommended during pregnancy. The absolute contraindication to the use of warfarin actually refers to the first trimester of pregnancy, and in particular between week 6 and 12 of gestation. Unfractionated heparin (UH) and low-molecular-weight heparin (LMWH) do not cross the placenta and are safe for the fetus, and are therefore an important potential alternative to warfarin during pregnancy. Unfortunately, there are no randomized trials to allow an accurate comparison of different anticoagulation strategies during pregnancy. A review of prospective and retrospective cohort studies compared maternal and fetal risks among three commonly used strategies: oral anticoagulants throughout pregnancy, UH throughout pregnancy, and

Valves

135

UH from weeks 6 to 12, then oral anticoagulants [34]. Embryopathy occurred when oral anticoagulants were administeredthroughout pregnancy (6.4%) or when UH was administered from 6 to 12 weeks replacing oral anticoagulants (3.4%). Incidence of thromboembolic complications was impressively higher with UH administered throughout pregnancy (33%) than with UH used to replace warfarin for only a limited time (9.2%) and with oral anticoagulants throughout pregnancy (3.9%). The efficacy and safety of LMWH has not been tested in nonpregnant patients with mechanical heart valve prostheses, and its use has been reported in a limited number of pregnant women. Because of inadequate data, clear-cut recommendations on anticoagulation in pregnant individuals with heart valve replacements are not available. According to the 2001 ACCP [35], three strategies are proposed: adjusted dose UH throughout pregnancy; adjusted dose LMWH therapy throughout pregnancy; and adjusted dose UH or LMWH therapy until week 13, warfarin until the middle of the third trimester, and then again UH or LMWH until delivery. All suggested strategies had the lowest grade of recommendation. The more recent guidelines of the European Society of Cardiology [36] on the management of cardiovascular diseases during pregnancy do not recommend the use of LMWH in patients with heart valve prostheses during pregnancy. Unfortunately, it is unlikely that randomized clinical trials will be done in this setting, but well-conducted registries on the use of LMWH could be of great help. A different option could be the systematic use of biological prosthetic heart valves instead of mechanical valves in all the women who want to plan a pregnancy.

Management of patients with mechanical heart valves undergoing elective surgery The management of patients with mechanical heart valves who are undergoing elective invasive procedures is of particular concern to treating physicians. The risk of thromboembolic events must be balanced against the risk of bleeding complications when warfarin dosage is reduced or discontinued and when UH or LMWH is administered as a bridging therapy before and after the procedure. There is also considerable controversy and variation in the recommendations for the prevention of thromboembolism in such patients. Kearon and Hirsh [37] suggested that the rate of thromboembolic events in patients with uncomplicated mechanical valvular prostheses following temporary discontinuation of oral anticoagulants should be expected to be around 8 cases per 100 patientyears. This risk is much lower than that of patients with a recent episode of deep venous thrombosis (40 cases per 100 patient-years) or with a recent episode of arterial thrombosis (15 cases per 100 patient-years). Improvements in prosthetic materials and valve designs have reduced the risk of thromboembolic complications [7]. In the ACCP guidelines, otherwise healthy patients with a bileaflet valve or a tilting disk valve in the aortic position are considered to be at such low risk of thromboembolism that perioperative anticoagulants are deemed unnecessary when warfarin is interrupted for short time periods [38]. On the contrary, patients with a mechanical heart valve in the mitral position remain at high risk and full dose UH or LMWH is recommended [38]. Very similar recommendations were previously published by the American College of Cardiology [39]. However, a distinction in the risk of thromboembolic events between

Thrombosis in clinical practice

136

mechanical valves in the aortic and in the mitral positions, when oral anticoagulants are temporarily interrupted, has not been universally accepted.

References [1] Larsen GL, Alexander JA, Stanford W. Thromboembolic phenomena in patients with prosthetic aortic valves who did not receive anticoagulants. Ann Thorac Surg 1977; 12:323. [2] Limet R, Lepage G, Grondin CM. Thromboembolic complications with the cloth-covered StarrEdwards aortic prosthesis in patients not receiving anticogulants. Ann Thorac Surg 1977; 23:529. [3] Stein PD, Alpert JS, Bussey HI, et al. Antithrombotic therapy in patients with mechanical or biological prosthetic heart valves. Chest 2001; 119:220S–227S. [4] Levine MN, Hirsh J, Landefeld S, et al. Hemorrhagic complications of anticoagulant treatment. Chest 1992; 102(Suppl. 4):352–63. [5] Goldsmith I, Turpie AGG, Lip GYH. Valvar heart disease and prosthetic heart valves. BMJ 2002; 325: 1228–31. [6] Sethia B, Turner MA, Lewis S, et al. Fourteen years experience with the Bjork-Shiley tilting disc prosthesis. J Thorac Cardiovasc Surg 1986; 91:350–61. [7] Horstkotte D, Schulte HD, Birks W, et al. Lower intensity anticoagulation therapy results in lower complication rates with the St. Jude Medical Prosthesis. J Thorac Cardiovasc Surg 1994; 107: 1136–45. [8] Cannegieter SC, Rosendal FR, Wintzen AR, et al. Optimal oral anticoagulant therapy in patients with mechanical heart valves. N Engl J Med 1995; 333:11–17. [9] Acar J, lung B, Boissel JP, et al. AREVA: multicenter randomized comparison of low-dose versus standard-dose anticoagulation in patients with mechanical prosthetic heart valves. Circulation 1996; 94:2107–12. [10] David TE, Gott VL, Harker LA, et al. Mechanical valves. Ann Thorac Surg 1996; 62:1567–9. [11] Akins CW. Long term results with the MedtronicHall valvular prosthesis. Ann Thorac Surg 1996; 61:806–13. [12] Saour JN, Sieck JO, Gallus AS. Trial of different intensities of anticoagulation in patients with prosthetic heart valves. N Engl J Med 1990; 332: 428–32. [13] Altman P, Rouvier J, Gurfinkel E, et al. Comparison of two levels of anticoagulant therapy in patients with substitute heart valves. J Thorac Cardiovasc Surg 1991; 101:427–31. [14] Vink R, Kraaijenhagen RA, Hutten BA, et al. The optimal intensity of vitamin K antagonists in patients with mechanical heart valves: a meta-analysis. J Am Coll Cardiol 2003; 42:2042–8. [15] Dale J, Myhre E, Storstein O, et al. Prevention of arterial thromboembolism with acetylsalicyl acid: a controlled clinical study in patients with aortic ball valves. Am Heart J 1977; 94:101–11. [16] Patrono C, Coller B, Dalen JE, et al. Platelet-active drugs. The relationship among dose, effectiveness and side effects. Chest 2001; 119:39S–63S. [17] Turpie AGG, Gent M, Laupacis A, et al. A doubleblind randomized trial of acetylsalicylic acid (100 mg) versus placebo in patients treated with oral anticoagulants following heart valve replacement. N Engl J Med 1991; 329:1365–9. [18] Altman R, Rouvier J, Gurfinkel E, Scazziota A, Turpie AGG. Comparison of high-dose with low-dose aspirin in patients with mechanical heart valve replacement treated with oral anticoagulant. Circulation 1996; 94:2113–16. [19] Hayashi JI, Nakazawa S, Oguma F, Miyamura H, Eguchi S. Combined warfarin and antiplatelet therapy after St Jude mechanical valve replacement for mitral valve disease. J Am Coll Cardiol 1994; 23: 672–7.

Valves

137

[20] Cevese PG. Long-term results of 212 xenograft valve replacements. J Cardiovasc Surg 1975; 16: 639–42. [21] Pipkin RD, Buch WS, Fogarty TS. Evaluation of aortic valve replacement with a porcine xenograft without long-term anticoagulation. J Thorac Cardiovasc Surg 1976; 71:179–86. [22] Stinson EB, Griepp RB, Oyer PE, et al. Long-term experience with porcine aortic valve xenografts. J Thorac Cardiovasc Surg 1977; 73:54–63. [23] Ionescu MI, Pakrashi BC, Mary DAS, et al. Longterm evaluation of tissue valves. J Thorac Cardiovasc Surg 1974; 68:361–79. [24] Turpie AGG, Gunstensen J, Hirsh J, et al. A randomized trial comparing two intensities of oral anticoagulant therapy following tissue heart valve replacement. Lancet 1988; i:1242–5. [25] Cohn LH, Collins JJ Jr, Rizzo RJ, et al. Twenty-year follow-up of the Hancock modified orifice porcine aortic valve. Ann Thorac Surg 1998; 66(Suppl.): S30–S31. [26] Horstkotte D, Scharf RE, Schulteiss HP. Intracardiac thrombosis: patient related and devicerelated risk factors. J Heart Valve Dis 1995; 4:114–20. [27] Louagie YA, Jamart J, Eucher P, et al. Mitral valve Carpentier-Edwards bioprosthetic replacement, thromboembolism, and anticoagulants. Ann Thorac Surg 1993; 56:931–7. [28] Moinuddeen K, Quin J, Shaw R, et al. Anticoagulation is unnecessary after biological aortic valve replacement. Circulation 1998; 98:II-95–II-99. [29] Ageno W, Turpie AGG. Exaggerated initial response to warfarin following heart valve replacement. Am J Cardiol 1999; 84:905–8. [30] Ageno W, Turpie AGG, Steidl L, et al. Comparison of a daily fixed 2.5 mg warfarin dose with a 5 mg, INR adjusted, warfarin dose initially following heart valve replacement. Am J Cardiol 2001;88: 40–4. [31] Coulshed N, Epstein EJ, McKendrick CS, et al. Systemic embolism in mitral valve disease. Br Heart J 1970; 32:26–34. [32] Chiang CW, Lo SK, Ko YS, et al. Predictors of systemic embolism in patients with mitral stenosis: a prospective study. Ann Intern Med 1998; 128:885–9. [33] Salem DN, Daudelin DH, Levine HJ, et al. Antithrombotic therapy in valvular heart disease. Chest 2001; 119:207S–219S. [34] Chang WS, Anand S, Ginsberg JS. Anticoagulation of pregnant women with mechanical heart valves: a systematic review of the literature. Arch Intern Med 2000; 160:191–6. [35] Ginsberg JS, Greer I, Hirsh J. Use of antithrombotic agents during pregnancy. Chest 2001; 119: 122S–131S. [36] The Task Force on the Management of Cardiovascular Diseases During Pregnancy of the European Society of Cardiology. Expert consensus document on management of cardiovascular diseases during pregnancy. Eur Heart J 2003; 24:761–81. [37] Kearon C, Hirsh J. Management of anticoagulation before and after elective surgery. N Engl J Med 1997; 336:1506–11. [38] Ansell J, Hirsh J, Dalen J, et al. Managing oral anticoagulant therapy. Chest 2001; 119:22S– 38S. [39] Bonow RO, Carabello B, de Leon AC Jr, et al. Guidelines for the management of patients with valvular heart disease: executive summary. A report of the American College of Cardiology/American Heart Association Task force on Practice Guidelines (Committee on Management of Patients with Valvular Heart Disease). Circulation 1998;98:1949–84.

Thrombosis in clinical practice

138

8 Percutaneous coronary intervention Freek WA Verheugt Epidemiology Ischemic heart disease is the major cause of mortality and one of the major causes of morbidity in the Western world. Current treatment consists of medical therapy with or without revascularization of occluded or narrowed coronary arteries. Coronary bypass surgery was first applied in the early sixties by René Favoloro in Cleveland and significantly contributed to quality and quantity of life in patients with symptomatic ischemic heart duced percutaneous transluminal coronary disease. In 1977, Andreas Grüntzig introangioplasty, a technique that postponed or even avoided coronary surgery in many symptomatic patients. Nowadays, percutaneous coronary intervention (PCI) largely outnumbers coronary artery bypass surgery in the Western world.

Complications of coronary intervention Complications of PCI can be either cardiac or vascular. Vascular complications mainly consist of bleeding, and cardiac consequences are cardiac death and myocardial infarction. In elective PCI, severe bleeding, depending on definition, may be seen in up to 5% of cases, and myocardial infarction can amount to the same figure, again depending on definition (CK-MB (CK, cardiac isoenzyme) or troponin release). Cardiac death is rare (30 day mortality less than 0.5%). In both ischemic and bleeding complications the hemostatic system seems to play a very important role.

Antithrombotic drugs in (semi-)elective percutaneous coronary intervention Antiplatelet agents Aspirin Aspirin given prior to PCI reduces the development of Q-wave myocardial infarction considerably, and should be standardly administered to patients undergoing PCI [1]. If there is the slightest doubt whether the patient undergoing PCI is on aspirin, aspirin

Thrombosis in clinical practice

140

should be given in a loading dose of 325 mg orally or intravenously prior to PCI. When coronary stents were introduced, stent thrombosis was common. This

Table 8.1 Randomized placebo-controlled trials with new antiplatelet agents in PCI. Trial

Year

Number of patients

Death/infarction (%)

RR

p

Followup (days)

Treatment Control Ticlopidine/clopidogrel ISAR [2]a a

FANTASTIC [6] STARS [4]

a a

MATTIS [3]

1996

517 2.0

4.2

0.28 0.0001 30

1998

473 5.8

8.3

0.70 0.38

42

1998

1,096 1.1

2.0

0.55 0.07

30

2001

350 5.1

8.1

0.63 0.36

30

2,436 2.6

4.5

0.63 0.02

20,137 5.5

8.3

0.66 0.0001 30

Total Intraveneous glycoprotein IIb/IIIa receptor antagonists Meta-analysis [8]b

2003

a All patients received a stent. Control treatment consists of aspirin and oral anticoagulation preceded by heparin. Active treatment is aspirin plus ticlopidine. b Control treatment is aspirin, active treatment is aspirin plus glycoprotein blocker. All stented patients received also ticlopidine or clopidogrel. RR: relative risk.

complication could be reduced by the use of heparin followed by warfarin. However, the bleeding complications of warfarin in combination with aspirin after PCI could be reduced by the introduction of dual antiplatelet therapy without warfarin (Table 8.1). Thienopyridines Ticlopidine and later clopidogrel have shown to effectively reduce acute and subacute stent thrombosis [2–5] (Table 8.1). Clopidogrel should be preloaded with a dose of 300 mg orally. The earlier clopidogrel is started prior to the procedure, the better the outcome [5]. The drug must be continued in a dose of 75 mg o.d. for at least 30 days [6]. Recent studies strongly suggest that a longer period of up to one year may further protect stented patients [5, 7].

Percutaneous coronary intervention

141

Glycoprotein IIb/IIIa receptor antagonists The introduction of glycoprotein IIb/IIIa receptor antagonists abciximab, tirofiban, and eptifibatide has further brought down the risk of periprocedural myocardial infarction and death (Table 8.1) [8]. Yet, when mortality alone is taken into account, it is only marginally reduced from 1.4% to 0.9%. Or, it takes 200 patients to treat with glycoprotein blockers to save one life, or $200,000 (if abciximab is used). The classic trial with glycoprotein blockers was the EPIC study [9], in which in over 2,000 patients bolus abciximab and infusion was considerably better than abciximab bolus or placebo in the prevention of periprocedural and long-term myocardial infarction and death in patients undergoing PCI on aspirin. Bleeding risk is increased by the use of abciximab but acceptable, whereas intracranial hemorrhage is not induced by glycoprotein IIb/IIIa antagonist in the setting of PCI. In the subsequent large trials on IIb/IIIa blockers in interventional cardiology, the benefit of stents became more and more clear [7]. In the EPISTENT trial [10], the role of IIb/IIIa antagonists was prospectively evaluated in patients randomized to stent or plain balloon angioplasty. It turned out that the best results were obtained in patients undergoing stent implantation on the IIb/IIIa antagonist abciximab and the worst results were observed in patients undergoing balloon angioplasty without stents while on abciximab. Interestingly, in that study it was observed that in patients who were pretreated with ticlopidine or clopidogrel, the benefit of abciximab was almost absent, whereas the benefit of abciximab was maximal in patients not pretreated with ticlopidine or clopidogrel [11]. IIb/IIIa antagonists are expensive and increase the risk of bleeding. Furthermore, they may induce thrombocytopenia. The use of clopidogrel has been extensive ever since the massive use of stenting in PCI became common. As stated earlier, recent studies have suggested that pretreatment with a 300 mg clopidogrel loading dose both in stable [6] and unstable [5, 7] patients may result in a better, earlier, and long-term success of PCI. Therefore, the role of IIb/IIIa antagonists in PCI can be challenged by a strategy of pretreatment with clopidogrel prior to PCI. The drawback of clopidogrel is the late onset of action and that is probably the key in the success of preloading. Even higher doses than 300 mg may speed up the adenosine diphosphate (ADP) receptor blockade on the blood platelet. Therefore, the ISAR-REACT trial was designed [12]. The objective of the study was to identify the role of IIb/IIIa blockers in low-risk patients undergoing PCI after massive clopidogrel loading (600 mg) 2–3 hours prior to PCI and continued on a dose of 75 mg b.i.d. For that reason, 2,159 patients were given 600 mg of clopidogrel loading within 3 hours prior to PCI, and subsequently randomized to double-blind abciximab or placebo in the cath-lab. Patients presenting within 2 hours prior to PCI were excluded for the study, as were patients with acute coronary syndromes with troponin elevation, patients with diabetes mellitus, and patients within 2 weeks of myocardial infarction. At 30 days post-PCI, abciximab increased the primary endpoint death, myocardial infarction, and target lesion revascularization by 5% (4.2% vs 4.0% for placebo, p=0.82). Death or myocardial infarction was similar, but major bleeding was also increased by 57% (abciximab 1.1% vs placebo 0.7%, p=0.38). Thrombocytopenia was seen in 0.9% of abciximab patients versus nil in the placebo patients (p=0.002) and

Thrombosis in clinical practice

142

transfusion was given to 2.4% abciximab patients versus 0.9% placebo patients (p=0.007). The data clearly show that massive clopidogrel loading makes the use of abciximab redundant in low-risk PCI patients provided clopidogrel is given up to 2 hours prior to PCI. These data heavily challenge the routine use of abciximab in low-risk PCI patients. These results corroborate well with the observations of the use of abciximab in higherrisk patients as seen in the PCI-CURE trial [10]. In that trial, acute coronary syndrome patients without STsegment elevation were pretreated randomly for a mean of 10 days of clopidogrel or placebo prior to PCI. Not only were the early and late clinical outcomes better with clopidogrel, but the need for glycoprotein receptor IIb/IIIa antagonists (mainly abciximab) was also reduced from 27% in the placebo group to 21% in the patients who had received clopidogrel (p=0.001). Also, in the recently published REPLACE-2 study 6,000 patients on abciximab use did not prevent major ischemic events in comparison to a thrombin blocker (bivalirudin), but did increase severe bleeding and thrombocytopenia [13]. So, it is quite possible that also in higher-risk patients megadose clopidogrel may reduce or even obviate the need for the expensive IIb/IIIa blockers. Furthermore, bleeding may be reduced and thrombocytopenia even prevented by not using IIb/IIIa blockers. However, this should be tested in future trials in high-risk patients, but it is unlikely that such trials will ever be carried out. Most trials with glycoprotein IIb/IIIa antagonists have been carried out with abciximab. There is only one major trial available on a head-to-head comparison between abciximab and tirofiban in all-comers for PCI showing an efficacy benefit for abciximab [14]. Anticoagulants Heparin is used routinely in arterial catheterization, although the American College of Cardiology/American Heart Association guidelines do not mention it specifically. In percutaneous intervention, unfractionated heparin is used and monitored by the Activated Clotting Time (ACT), which should be kept between 350 and 375 s. Underdosing leads to thrombotic complications [15]. In conjunction with glycoprotein IIb/IIIa blockers, ACT should be between 200 and 250 s. Many patients undergo percutaneous intervention while on low molecular weight heparin. Whether in comparison to unfractionated heparin this is more, equally, or less effective, if continued in the catheterization laboratory, is currently under investigation in the 11,000-patient SYNERGY trial. Direct thrombin blockers like hirudin and hirulog have been tested in PCI as an alternative to heparin. In the large HELVETICA trial, hirudin bolus and hirudin bolus plus infusion had been compared to heparin. Both hirudin arms did not show a significant benefit over heparin, although during treatment there was a benefit for the hirudin infusion therapy [16]. Unfortunately, hirudin, which can be given subcutaneously, was not given for a longer period of time. Bivalirudin (hirulog) was tested against heparin plus abciximab in REPLACE-2 [13] and showed similar results to abciximab and heparin, but with much less bleeding. Oral anticoagulants can be given as well to protect against ischemic complications during catheter intervention (Table 8.2). In the BAAS study [17], patients to undergo (semi-)elective coronary intervention were randomized to

Percutaneous coronary intervention

143

Table 8.2 Randomized aspirin-controlled study of coumadin plus aspirin before and after coronary angioplasty. Trial

Year

n

INR INR Aspirin Death/reinfarction target reached dose Coumadin+ Aspirin (mg aspirin o.d.)

BAAS[17] 2000 1,058 2.1– 4.8

2.7

100

20/530 (3.7%)

31/528 (5.1%)

RR

p

F/U (mos)

0.65 0.12 12 (0.50– 0.84)

INR: International Normalized Ratio; RR: relative risk; F/U: follow-up.

warfarin (median INR 2.6) plus aspirin or aspirin alone, 2 weeks ahead of the procedure until 6 months thereafter. Coumadin on top of aspirin reduced early and late ischemic complications at an acceptable increase in major bleeding. The achieved INR was strongly related to the benefits in ischemic complications as well as the risk of restenosis [18, 19].

Antithrombotic drugs in infarct percutaneous coronary intervention PCI has become more and more popular in the treatment of acute myocardial infarction. Because emergent balloon angioplasty of the occluded infarct artery is feasible, relatively safe, and effective in achieving full reperfusion (so called TIMI-3 flow) in over 90% of cases, patients with ST-segment elevation benefit from primary PCI in comparison to fibrinolytic therapy, whether they have to be transported to a PCI center, or not [20]. Whereas the previous paragraph mainly addressed the role of antithrombotic therapy in elective or semi-elective PCI, as in unstable angina, non-ST elevation myocardial infarction, or post-infarction angina, the studies detailed next only deal with PCI in acute ST-elevation myocardial infarction. Antithrombotic therapy is very important in infarct management in general, but in primary PCI in particular. Antiplatelet agents Aspirin Aspirin is standardly administered to all patients with acute myocardial infarction whether they undergo PCI or not [21].

Thrombosis in clinical practice

144

Thienopyridines When coronary stents are introduced for infarct PCI with stenting, prevention of stent thrombosis must be achieved by using clopidogrel, although specific trials in infarct PCI are lacking. Yet, clopidogrel must be given to all patients who receive a stent in primary PCI. Uploading is not advised, since intervention and subsequent stenting is not feasible in each individual patient eligible for primary PCI. Clopidogrel may enhance bleeding in these patients already on aspirin and heparin, who undergo early arterial catheterization, but no coronary intervention. Glycoprotein IIb/IIIa receptor antagonists The glycoprotein IIb/IIIa receptor antagonist abciximab has been extensively evaluated in primary PCI. The first pilot studies were promising with regard to early coronary patency prior to the actual primary PCI [22–24]. An open infarct related artery prior to primary PCI heralds a good prognosis [25]. The placebo-controlled randomized trials showed a trend towards better long-term survival [26–30] (Table 8.3). Interestingly, in some studies it was observed that the earlier the glycoprotein IIb/IIIa receptor antagonist is administered (e.g. in the emergency room or even the ambulance)—and, thus, the drug is in the patient—the better the infarct artery patency [22, 28]. These observations were confirmed in a randomized trial, where emergency room infusion of tirofiban was superior to start of treatment in the catheterization laboratory with regard to patency prior to primary PCI [31]. This approach is now generally called facilitated PCI and glycoprotein IIb/IIIa receptor antagonists may be helpful in early restoration of coronary flow prior to the actual intervention. But the figures of TIMI-3 flow achieved with these agents will not exceed 20–25%.

Table 8.3 Mortality after abciximab in primary PCI. Trial

n

Abciximab

RAPPORT [26] 1998

483

10/241(4.2%)

11/242 (4.6%) 0 .91 (0 .38– 2.18)

0.82 6

ISAR-2 [27]

2000

401

4/201(2.0%)

9/200 (4.5%) 0.44 (0 .17– 1.18)

0.16 1

ADMIRAL [28]

2001

300

5/149(3.4%)

11/151 (7.3%) 0 .46 (0 .16– 1.36)

0.13 6

CADILLAC [29]

2002 2,082 35/1,052(3.3%)

ACE [30]

2003

Total

Year

400 3,666

9/200 (4.5%) 63/1,843 (3.4%)

Placebo

38/1,030 (3.7%)

RR

p

0 .90 (0 .62– 1.53)

0.23 6

16/200 (8.0%) 0 .56 (0.26– 1.24)

0.15 6

85/1,823 (4.7%)

0 .73 (0 .53– 1.01)

0.07

F/U (mos)

Percutaneous coronary intervention

145

Anticoagulants Intravenous unfractionated heparin is currently used in primary PCI, although its efficacy has never been prospectively evaluated for this indication. Interestingly, its role has been studied in facilitation of primary PCI. The drug was given in very high doses up to 30,000 U to patients on transport to primary PCI, or to those in the emergency room awaiting primary PCI. Although it improved early patency over no or low doses of heparin (5,000 U), TIMI-3 flow over 20% was never achieved [32]. Fibrinolytic agents The only logic approach for facilitation of primary PCI would be the application of fibrinolysis. These agents are widely available and can be administered prehospital, at a referring hospital, during transport, or in the emergency room. Fibrinolytic therapy can now be administered as a bolus and achieves TIMI-3 flow in up to 60% of patients after 60–90 minutes thereafter, which is about the same time it takes to prepare the catheterization laboratory or to transport the patient from outside to the laboratory. However, there is still some reluctance among interventionalists to operate on patients on lytic drugs given the poor results from the past. Routine balloon angioplasty after lytic therapy resulted in more complications than lytic alone—especially more bleeding and unexpectedly lower patency rates [33]. These trials were carried out with streptokinase or older alteplase regimes. Angioplasty techniques have significantly improved over the years and usually consist of stenting and the use of clopidogrel, and of glycoprotein IIb/IIIa antagonists if indicated. The first trial studying lytic facilitated PCI was the PACT trial, in which low-dose bolus alteplase (50 mg) plus heparin in a bolus of 5,000 U was compared to a heparin bolus alone in 600 patients to undergo primary PCI [34]. Preangioplasty TIMI-3 flow was 32% versus 15%, respectively, and the PCI procedure was not negatively influenced by the pre-use of the low-dose lytic. Thereafter, early patency trials on the routine combination of half dose lytic plus a full dose glycoprotein IIb/IIIa antagonist did not show better reperfusion than full dose lytic alone, but did significantly increase bleeding [35]. Trials on the safety and efficacy of full dose lytic in the facilitation of primary PCI versus primary PCI alone are underway. Finally, patients with failed fibrinolytic therapy (lack of both pain relief and resolution of

Table 8.4 Recommendations for antithrombotic therapy in PCI. (A) In patients undergoing any PCI, aspirin should be given in a dose of 75–325 mg o.d. If there is doubt whether the patient is on aspirin, aspirin should be given in a loading dose of 325 mg orally or intravenously prior to the procedure, (B) In patients undergoing any PCI with stent implantation, aspirin should be combined by clopidogrel given in a loading dose of at least 300 mg orally followed by 75 mg o.d. up to at least 30 days. The earlier clopidogrel is started prior to the procedure, the better the outcome. (C) In patients undergoing high-risk PCI, intravenous glycoprotein IIb/IIIa blockers (preferably abciximab) may be added to aspirin, irrespective of the use of stent (and, thus, clopidogrel). The

Thrombosis in clinical practice

146

earlier the therapy is started, the better the angiographic and clinical outcome. However, these agents have not been largely tested in the clopidogrel era. Furthermore, they increase the rate of major extracranial bleeding. (D) In patients undergoing PCI for acute ST-elevation myocardial infarction, intravenous glycoprotein IIb/IIIa blockers (preferably abciximab) may be added to aspirin (facilitated PCI). The earlier the therapy is started, the better the angiographic and clinical outcoine. However, it increases the rate of major extracranial bleeding. In rescue PCI, it should be avoided. The role of fibrinolytic therapy in facilitated PCI is currently under investigation in large randomized trials.

ST-segment elevation) may be referred for PCI (rescue PCI). They have an occluded infarct artery for several hours and are usually highrisk patients. The benefit of rescue PCI is likely, but has not been solidly established [36]. Clopidogrel is necessary when a stent is used. Needless to say, such additional antithrombotic therapy in these patients already on aspirin, heparin, and a lytic may further increase bleeding risk. Furthermore, randomized data on the use of glycoprotein IIb/IIIa blockers in these patients are lacking.

Conclusions Antithrombotic therapy is essential in the efficacy and safety of PCI. Platelet aggregation inhibitors and anticoagulants prevent periprocedural myocardial infarction and death effectively. In most cases stents are implanted during coronary intervention, which makes long-term dual antiplatelet therapy (aspirin and clopidogrel) necessary. Finally, PCI has evolved as the optimal reperfusion treatment of acute ST-segment elevation myocardial infarction. Proper accompanying antithrombotic therapy has shown to be critical in the optimalization of percutaneous reperfusion therapy. Both percutaneous coronary intervention and antithrombotic therapy are rapidly evolving fields of clinical research. Of the latter, current recommendations are given in Table 8.4.

References [1] Schwartz L, Bourassa MG, Lespérance J, et al. Aspirin and dipyridamole in the prevention of restenosis after percutaneous transluminal coronary angioplasty. N Engl J Med 1988; 318:1714– 19. [2] Schömig A, Neumann FJ, Kastrati A, et al. A randomized comparison of antiplatelet and anticoagulant therapy after the placement of coronary-artery stents. N Engl J Med 1996; 334:1084–9. [3] Urban P, Macaya C, Rupprecht HJ, Kiemeneij F, Emanuelsson H, Fontanelli A, Pieper M, Wesseling T, Sagnard L. Randomized evaluation of anticoagulation versus antiplatelet therapy after coronary stent implantation in high-risk patients: the multicenter aspirin and ticlopidine trial after intracoronary stenting (MATTIS). Circulation 1998;98: 2126–32. [4] Leon MB, Baim DS, Popma JJ, Gordon PC, Cutlip DE, Ho KK, Giambartolomei A, Diver DJ, Lasorda DM, Williams DO, Pocock SJ, Kuntz RE. A clinical trial comparing three antithrombotic-drug regimens after coronary-artery stenting. Stent Anticoagulation Restenosis Study Investigators. N Engl J Med 1998; 339:1665–71.

Percutaneous coronary intervention

147

[5] Steinhubl SR, Berger PB, Mann JT, Fry ET, DeLago A, Wilmer C, Topol EJ. Early and sustained dual oral antiplatelet therapy following percutaneous coronary intervention: a randomized controlled study. JAMA 2002;288:2411–20. [6] Bertrand ME, Legrand V, Boland J, et al. Randomized multicenter comparison of conventional anticoagulation versus antiplatelet therapy in unplanned and elective coronary stenting: the Full Anticoagulation versus aspirin and ticlopidine (FANTASTIC) study. Circulation 1998;98: 1597–603. [7] Mehta SR, Yusuf S, Peters RJ, et al. Effects of pretreatment with clopidogrel and aspirin followed by long-term therapy in patients undergoing percutaneous coronary intervention: the PCI-CURE study. Lancey 2001; 358:527–33. [8] Karvouni E, Katritsis DG, Ioanidis JP. Intravenous glycoprotein IIb/IIIa receptor antagonists reduce mortality after percutaneous coronary intervention. J Am Coll Cardiol 2003; 41:26–32. [9] EPIC study. The EPIC Investigation. Use of a monoclonal antibody directed against the platelet glycoprotein IIb/IIIa receptor in high-risk coronary angioplasty. N Engl J Med 1994; 330:956– 61. [10] EPISTENT Investigators. Randomised placebocontrolled and balloon-angioplasty-controlled trial to assess safety of coronary stenting with the use of platelet glycoprotein IIb/IIIa blockade. Lancet 1998; 352:87–92. [11] Steinhubl SR, Ellis SG, Wolski K, Lincoff AM, Topol EJ. Ticlopidine pretreatment before coronary stenting is associated with sustained decrease in cardiac adverse events: data from the Evaluation of Platelet glycoprotein IIb/IIIa Inhibitor for Stenting (EPISTENT) trial. Circulation 2001; 103:1402–9. [12] Kastrati A, Mehilli J, Schühlen H, Dirschinger J, Dotzer F, ten Berg JM, Neumann F-J, Bollwein H, Volmer C, Gawaz M, Berger PB, Schömig A. A clinical trial of abciximab in elective percutaneous coronary intervention after pretreatment with clopidogrel. N Engl J Med 2004; 350:232–8. [13] Lincoff AM, Bittl JA, Harrington A, et al. Bivalirudin and provisional glycoprotein IIb/IIIa blockade compared with heparin and planned glycoprotein IIb/IIIa blockade during percutaneous coronary intervention: REPLACE-2 randomized trial. JAMA 2003; 289:853–63. [14] Topol EJ, Moliterno DP, Herrmann HC, et al. Comparison of two platelet glycoprotein IIb/IIIa inhibitors, tirofiban and abciximab, for the prevention of ischemic events with percutaneous coronary revascularisation. N Engl J Med 2001; 344:1888–94. [15] Chew DP, Bhatt DL, Lincoff AM, et al. Defining the optimal clotting time during percutaneous coronary intervention: aggregate results from six randomized, controlled trials. Circulation 2001; 103:961–6. [16] Serruys PW, Herrman JP, Simon R, et al. A comparison of hirudin with heparin in the prevention of restenosis after coronary angioplasty. N Engl J Med 1995; 333:757–63. [17] Ten Berg JM, Kelder JC, Suttorp MJ, Mast EG, Bal E, Ernst JMPG, Verheugt FWA, Plokker HWM. Effect of coumarins started before coronary angioplasty on acute complications and long-term follow-up: a randomized trial. Circulation 2000; 102:386–91. [18] Ten Berg JM, Kelder JC, Suttorp MJ, Mast EG, Bal E, Ernst JMPG, Verheugt FWA, Plokker HWM. Oral anticoagulant therapy during and after coronary angioplasty: the intensity and duration of anticoagulation are essential to reduce thrombotic complications. Circulation 2001; 103:2042–7. [19] Ten Berg JM, Kelder JC, Suttorp MJ, Verheugt FWA, Plokker HWM. A randomized trial assessing the effect of coumarins started before coronary angioplasty on restenosis: results of the 6-month angiographic substudy of the Balloon Angioplasty Anticoagulation Study (BAAS). Am Heart J 2003; 145:58–65. [20] Keeley EC, Boura JA, Grines CL. Primary coronary angioplasty versus intravenous fibrinolytic therapy for acute myocardial infarction: a quantitative review of 23 randomised trials. Lancet 2003; 361:13–20.

Thrombosis in clinical practice

148

[21] Antithrombotic Trialists’ Collaboration. Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. BMJ 2002; 324:71–86. [22] Van den Merkhof LFM, Liem A, Zijlstra F, Olsson H, Veen G, De Zwaan C, Bar F, Van der Zwaan C, Simoons ML, Verheugt FWA. First results of abciximab (ReoProR) in the treatment of acute myocardial infarction and primary PTCA: the GRAPE pilot study. J Am Coll Cardiol 1999; 33: 1528–32. [23] Antman EM, Giugliano RP, Gibson CM, et al. Abciximab facilitates the rate and extent of thrombolysis: results of the thrombolysis in myocardial infarction-14 (TIMI-14) trial. Circulation 1999;99: 2720–32. [24] SPEED. Trial of abciximab with and without lowdose reteplase for acute myocardial infarction. Circulation 2000; 101:2788–94. [25] Stone GW, Cox D, Gracia E, et al. Normal flow (TIMI-3) before mechanical reperfusion therapy is an independent determinant of survival in acute myocardial infarction: results from the primary angioplasty in myocardial infarction trials. Circulation 2002; 104:636–41. [26] Brener SJ, et al. Randomized, placebo controlled trial of glycoprotein IIb/IIIa blockade with primary angioplasty for acute myocardial infarction: RAPPORT. Circulation 1998; 98:734–41. [27] Neumann FJ, Kastrati A, Schmitt C, et al. Effect of glycoprotein IIb/IIIa receptor blockade with abciximab on clinical and angiographic restenosis after the placement of coronary stents following myocardial infarction. J Am Coll Cardiol 2000; 35: 915–21. [28] Montalescot G, Barragan P, Wittenberg O, et al. Platelet glycoprotein IIb/IIIa inhibition with coronary stenting for acute myocardial infarction. N Engl J Med 2001; 344:1895–903. [29] Stone GW, Grines CL, Cox DA, et al. Comparison of angioplasty with stenting, with or without abciximab, in acute myocardial infarction. N Engl J Med 2002; 346:957–66. [30] Antoniucci D, Rodriguez A, Hempel A, et al. A multicenter randomized trial comparing infarct artery stenting alone with infarct artery stenting plus abciximab in acute myocardial infarction: the abciximab and carbostent (ACE) trial [Abstract]. Eur Heart J 2003; 24:(Suppl.):712. [31] Lee DP, Herity NA, Hiatt BL, et al. Adjunctive platelet glycoprotein IIb/IIIa receptor inhibition with tirofiban before primary angioplasty improves angiographic outcomes: results of the Tirofiban Given in the Emergency Room before Angioplasty (TIGER-PA) pilot trial. Circulation 2003; 107: 1497–501. [32] Liem AL, Zijlstra F, Ottervanger JP, Hoorntje JCA, Suryapranata H, De Boer MJ, Verheugt FWA. High dose bolus heparin as pretreatment of primary angioplasty in acute myocardial infarction: the Heparin in Early Patency (HEAP) randomized trial. J Am Coll Cardiol 2000; 35:600–4. [33] Michels KB, Yusuf S. Does PTCA in acute myocardial infarction affect mortality and reinfarction rates? A quantitative overview (meta-analysis) of the randomized clinical trials. Circulation 1995; 91:476–85. [34] Ross AM, Coyne KS, Reiner JS, et al. A randomized trial comparing primary angioplasty with a strategy of short-acting thrombolysis and immediate planned rescue angioplasty in acute myocardial infaction: the PACT trial. J Am Coll Cardiol 1999; 34:1954–62. [35] Verheugt FWA. IIb/IIIa antagonists in fibrinolytic therapy for acute myocardial infarction: a bold stroke? Eur Heart J 2003; 24:1801–3.

Percutaneous coronary intervention

149

[36] Ellis SG, Da Silva ER, Spaulding CM, Nubuyoshi M, Weiner B, Talley JD. Review of immediate angioplasty after fibrinolytic therapy for acute myocardial infarction: insights from RESCUE 1, RESCUE 2 and other contemporary clinical experiences. Am Heart J 2002; 139:1046–53.

Thrombosis in clinical practice

150

9 Coronary artery disease Coronary revascularization Petr Widimsky, Zbynĕk Straka, and Martin Pĕnička Procedural and pathophysiological aspects of PCI and CABG in relation to thrombosis Percutaneous coronary intervention (PCI) Percutaneous coronary intervention is related to coronary thrombosis in two different ways: preexisting (primary) clot is fragmented and removed by PCI (primary PCI in STsegment elevation acute myocardial infarction (STEMI), PCI in unstable angina pectoris/non-ST-segment elevation acute coronary syndrome (NSTE ACS)) or a new (secondary) clot originates as an early or late complication of the procedure. The primary clot occurs naturally on unstable atherosclerotic plaque and PCI accompanied by adjunctive pharmacotherapy is the most effective treatment for such clots. The secondary clot occurs in an artery with originally stable plaque, which was dilated by PCI and thus its endothelial surface was mechanically disrupted and became thrombogenic. The PCI procedure can provoke thrombus formation not only on the arterial wall, but also on the interventional instruments: inside the guiding catheter, on the guide wire, and rarely on the balloon or stent. The factors contributing to the risk of the instrumentrelated thrombosis are hypercoagulative state of the patient blood, insufficient antithrombotic treatment during PCI, and insufficient procedural “hygiene” (e.g. even very small barely visible blood clot not removed from the guide wire or guiding catheter during their manipulation outside patient body can be the “trigger” for thrombus growth when reintroduced into the patient). Reuse of instruments might also contribute to the increased risk of thrombus formation. Coronary artery bypass graft (CABG) The widespread use of PCI in patients with primary acute thrombotic occlusion of coronary arteries has diminished enthusiasm for emergency bypass surgery. Compared to PCI, the surgical procedure is less accessible, is associated with a delay in therapy, more stress to the patient, and higher morbidity and mortality. Therefore, in the evolving phase of acute myocardial infarction (AMI) coronary bypass surgery should be reserved for selected patients only. In contrast, postinfarction mechanical defects, such as ventricular septal defect (VSD), acute mitral valve regurgitation, and left ventricular (LV) free wall rupture, remain a surgical challenge: urgent surgical repair is needed since conservative treatment is associated with extremely high mortality.

Thrombosis in clinical practice

152

Periprocedural or early postoperative myocardial infarction (MI) after CABG is an infrequent but possibly dramatic complication. Some of the causes may be the bypass occlusion resulting from thrombosis of the graft and the corresponding coronary artery as a consequence of a technical error, severe diffuse arteriosclerosis, or pathological blood clotting. In patients with stable hemodynamics, a conservative approach is possible, but urgent transfer directly to a cath-lab and early angiographic control to assess the need for reintervention (PCI or surgical bypass correction) can prevent myocardial necrosis. In case of hemodynamic deterioration, cardiopulmonary resuscitation is necessary and emergency reoperation is initiated immediately.

Revascularization procedures in the treatment of intracoronary thrombosis (“primary clot”) Primary PCI for occlusive acute coronary thrombosis (STEMI) For this generation of physicians, it is hard to imagine, that only 25 years ago it was not clear, whether coronary thrombosis is the cause or consequence of AMI [1]. The pioneering work of P. Rentrop [2] provided important angiographic evidence in favor of the current view on MI pathophysiology. This establishment of acute thrombotic occlusion of the infarct-related artery as a primary cause of acute STEMI led to several randomized clinical studies showing benefit from intracoronary [2, 3] and later intravenous thrombolysis [4, 5]. Direct coronary angioplasty as the primary reperfusion therapy for AMI had already been first described by Meyer [6] and Hartzler [7] in 1982– 83. Ten years later three randomized studies showed the superiority of primary angioplasty over thrombolysis when used with the same delay after the onset of symptoms [8–10]. This was also confirmed by a meta-analysis of additional trials [11]. Currently, there is no doubt that primary PCI is associated with the markedly lower inhospital mortality, risk of reinfarction and stroke as compared with thrombolysis even when patients need transport to other hospital. The safety and feasibility of interhospital transportation of patients with AMI from smaller hospitals without PCI facilities to the tertiary PCI centers was tested by the LIMI study in the Netherlands [12], the PRAGUE1 and PRAGUE-2 studies in the Czech Republic [13, 14], the DANAMI-2 trial in Denmark [15], and a small Air-PAMI study in the USA [16]. These studies clearly demontransport but also highly significant decrease strated not only safety and feasibility of the in death, reinfarction, and stroke rates in patients transferred to primary PCI. Thus, pri- mary PCI is currently considered the golden standard (treatment method of choice) for all patients with STEMI who can reach cath-lab within 90 minutes after the diagnostic ECG. Thanks to both PRAGUE studies (which were running in almost half of the Czech hospitals), the Czech Republic has become the leading country with the highest proportion of primary PCI among all reperfusion therapies in STEMI. In 2003, 93% of reperfusion was done by primary PCI and only 7% by thrombolysis. The guidelines of the Czech Society of Cardiology [17] and our own institutional guidelines recommend the following optimal treatment sequence:

Coronary artery disease coronary revascularization

153

1. Prehospital ECG done by the Emergency Medical System at the first contact with the patient. 2. When ST-segment elevations are present 500 mg intravenous salicylate (Aspegic®) and 5,000 U intravenous unfractionated heparin (UFH) boluses are given. 3. Nearest PCI center is informed and patient is transferred directly to the cath-lab (passing both Emergency Room and Coronary Care Unit). 4. Based on coronary angiographic finding either additional UFH (up to a total dose 150– 200 U/kg of body weight) or combination of intravenous glycoprotein IIb/IIIa receptor inhibitor (GP IIb/IIIa inhibitor) and UFH (70 U/kg) immediately prior to PCI (i.e. after coronary angiography (CAG)). 5. Immediately after the procedure, patient receives either 300 mg of clopidogrel or 500 mg of ticlopidine. 6. The arterial sheath remains in place until APTT falls under 45 seconds, then, it is removed, puncture site compressed, and low-molecular-weight heparin (LMWH) given subcutaneously in mean therapeutic dose based on body weight. 7. LMWH is given daily until patient is mobilized (usually 2 days) and then it is stopped. 8. Clopidogrel or ticlopidine is continued daily for at least one month and as for clopidogrel, longer treatment duration of up to 6–12 months is recommended. 9. A dose of 100–200 mg of aspirin is given lifelong. Procedural aspects of primary PCI. The recognition of infarct-related artery on an urgent coronary angiogram is usually not difficult: complete (100%) occlusion with sharp distal margin, without bridging homocollaterals and with TIMI-0 flow are the hallmarks (Figure 9.1). The first contrast injections to the artery during diagnostic angiography or the passage of intracoronary guide wire may already improve the flow to TIMI-1. It is not unusual that TIMI-2 or even TIMI-3 flow is found on the initial angiogram (spontaneous or aspirin+heparin-induced reperfusion). In many patients, the thrombus is directly visualized as an intraluminal filling defect distally to the obstruction site. Long duration (>6 h) of coronary artery occlusion leads frequently to progressive thrombus distal apposition with filling of the distal vessel lumen by thrombotic masses. Excessive thrombosis with significant distal embolization is the frequent cause of the “no-reflow” phenomenon after otherwise morphologically successful PCI. Several distal protection devices have been tested in small trials during the last few years [18, 19]. They can “catch” and remove thrombus from the coronary artery. However, their use prolongs the procedure and may add to vessel wall traumatization in distal segments, which otherwise would not be attacked. Their real clinical impact is not yet well defined. The fresh clot in MI is usually very fragile—thus, it is usually rather easy to pass even a soft-tip intracoronary guide wire through the thrombotic occlusion distally to the peripheral segment of the infarct-related artery. When some filling of the distal vessel is present, direct stenting (without previous balloon predilatation) is feasible. When distal segment is not visible, initial balloon dilatation is needed. In this setting, one must be absolutely sure that the guide wire is in the distal lumen of the infarct-related artery (and e.g. not in the pericardium…). The free move of the guide wire into side branches, the typical course of the

Thrombosis in clinical practice

154

Figure 9.1 Typical coronary angiogram showing acute thrombotic occlusion of the left circumflex coronary artery (left upper), passage of the intracoronary guide wire through the clot (right upper), direct stent implantation (left lower) and final result (right lower). guide wire where the vessel is expected to be in at least three different projections, and finally, the entirely free move of the deflated balloon through the thrombus to the very distal segment of the infarct-related artery are the three key indicators of the proper guide wire position. It is fair to say that the routine use of stents in most of primary PCI was not associated with decreased mortality. In some trials [20], the mortality in stented patients with AMI was even slightly higher as compared with patients treated by balloon only. In other trials [21], the mortality was similar. In general, the early and late mortality of STEMI is decreased equally by primary balloon angioplasty and by stent implantation. The main advantage of stenting over balloon angioplasty is reduced restenosis, reocclusion, and repeated target vessel revascularization rate.

Coronary artery disease coronary revascularization

155

Rescue/facilitated PCI as combined pharmacological—mechanical treatment of occlusive acute coronary thrombosis (STEMI) Rescue PCI is defined as an emergent PCI for ongoing transmural ischemia, which continues after full dose thrombolytics was given—in other words, immediate PCI of the infarct artery done for thrombolysis failure. Despite attracting cardiologists for 20 years, this concept has still not become a routine. The three inherent shortcomings of this strategy are (1) significant bleeding risk resulting from the combination of thrombolysis with intervention (fatal bleeding complications—cardiac tamponade, hemorrhagic stroke, and retroperitoneal bleeding—were present only in the “rescue angioplasty” group in the PRAGUE-1 trial); (2) the longest time-to-reperfusion from all reperfusion strategies; and (3) difficulties in precise and early recognition of “unsuccessful thrombolysis.” For these reasons, the PCI had questionable outcomes in early trials [22–25] done before the era of primary angioplasty (c. before 1995). Until recently, it was felt that rescue angioplasty plays a minor role as compared to primary PCI on one hand and thrombolysis alone on the other hand. However, this view is currently changing in light of some more recent trials [26, 27] done in centers with large experience in primary angioplasty. In particular, this growing experience might decrease or remove the limits of rescue angioplasty in several ways: 1. Bleeding complications are highly dependent on the experience of the whole team (prehospital medical staff, PCI operator, cath-lab nurses, CCU physicians, and nurses). With increasing experience (thanks to large workload of primary PCI) on all levels, the rate of such complications in “rescue angioplasty” is acceptably low and does not differ much from primary angioplasty without preceding thrombolysis. 2. Time to reperfusion can be decreased by “standby rescue angioplasty strategy” used in both the trials mentioned [26, 27]. This underlines the key importance of the team discipline for the success of any primary (and rescue) PCI program. In our center, patients with STEMI arrive directly to the catheterization laboratory without any delays on CCU, chest pain unit, or Emergency Room. The first examination (provided they have diagnostic ECG done pre-hospitally) is the coronary angiography. Everything else is done either after (in stable patients) or during (in hemodynamically unstable patients) coronary angiography. At least one cath-lab nurse is present in the cath-lab 24 hours a day and an interventional cardiologist is on-call and must begin the procedure within 30 minutes of phone call when he is outside, or within 5 minutes when he is inside the hospital. 3. Early and precise recognition of thrombolysis failure (persistent coronary occlusion despite full dose thrombolysis) is possible by immediate coronary angiography. Thus, the strategy of performing coronary angiography immediately after thrombolysis either in all patients (as in the LIMI and PRAGUE-1 trials—this is in fact the “facilitated PCI” approach—see next paragraph) or in large proportion of patients (CAPTIM trial) may be comparable to primary PCI strategy [12, 13, 27]. These improvements in rescue PCI strategy along with new drugs led to a novel concept of “facilitated PCI,” that is pharmacologically prepared primary PCI. This concept combines positive aspects of the two original reperfusion strategies—pharmacologic and

Thrombosis in clinical practice

156

mechanical. The organization is equal to primary PCI program, that is, all patients are transferred to tertiary centers with 24-h availability of PCI. The transfer is initiated immediately after diagnostic ECG was obtained and diagnosis made. The “facilitating” drug(s) are given just before or during the transfer to PCI center. Thus, the delays are much shorter when compared to “classical rescue PCI strategy.” The “facilitating drug” can be either bolus thrombolytic (tenecteplase, reteplase), glycoprotein (GP) IIb/IIIa inhibitor (abciximab, eptifibatide), or their combination. Despite being promising, this concept has not yet been demonstrated to be superior to primary PCI strategy alone. Several randomized studies on facilitated PCI are currently being carried out or are in the offing. Hence, the position of this strategy among reperfusion strategies may be clarified during 2004–05. Although there is large international experience with combinations “thrombolysis+PCI” and “GP IIb/IIIa inhibitor+PCI,” there is only very limited experience with potentially dangerous combination of “thrombolysis+GP IIb/IIIa inhibitor+PCI.” Eventhough these “lytic cocktails” showed higher pharmacologic reperfusion rates than each drug alone, the bleeding complication rates were rather high. Optimal therapy in patients with NSTE ACS Patients with NSTE ACS represent a heterogeneous group consisting both of patients with excessive short- and long-term mortality, patients at low risk, or not having ACS at all. The latter group accounts for at least 15–40% of these patients [28–31]. With recent introduction of several potent antithrombotic agents and routine stenting, the essential question occurs: what is the optimal “dose” of therapy (invasive and/or cocktail of different antithrombotics) delivered to an individual patient? For those at highest risk, the maximal therapy is warranted. In contrast, in low-risk patients, the goal is to avoid costly invasive procedures with equally expensive adjunctive pharmacotherapy that is unlikely to confer clinical benefit and may actually cause harm. Therefore, identifying subjects at highest risk and selecting optimal “dose” of therapy tailored to individual risk remains a clinical challenge. Table 9.1 shows variables associated independently with increased risk for death and cardiac ischemic events in NSTE ACS patients [32–48]. Among these parameters, the presence and the extent of ST-segment depression on admission ECG and cardiac troponin elevation are the most powerful predictors of adverse outcome. Yet, the focusing on a single, though important, variable may lead to underestimation of patient’s risk. The TIMI risk score model (Table 9.2) represents an example of the integrated approach in assessing patient’s risk [36, 49–53]. How does this evidence translate into management of patients with NSTE ACS? Antithrombotic therapy In unstratified population, early studies demonstrated the poor prognosis of untreated NSTE ACS patients with both short- and longterm mortality rates between 5% and 10% [54, 55]. Introduction of aspirin and UFH therapy has resulted in important reductions (50–70%) in the incidence of in-hospital death or MI [56–60]. Hence, aspirin and UFH should be administered to all patients with suspected NSTE ACS regardless of the intended therapeutic strategy. Yet, despite the widespread use of aspirin and UFH, the prognosis of NSTE ACS remained unsatisfactory [28, 29, 61–63]. This indicates, in the

Coronary artery disease coronary revascularization

157

first place, that the level of antithrombotic activity achieved by the coadministration of aspirin and UFH is insufficient. In this regard, recent trials demonstrated the efficacy of several new antithrombotic agents, such as LMWH [61, 64–66], clopidogrel [67], and GP IIb/IIIa inhibitors [62, 63, 68–71] in NSTE ACS patients. All three classes of antithrombotics appear to be particularly effective in patients at highest risk. For example,

Table 9.1 Clinical factors predicting risk in NSTE ACS. Medical history Prior history of angina, frequent, prolonged/ongoing (>20 min) or postinfarction angina Age >65 years Prior aspirin or beta-blocker use Prior revascularization Diabetes mellitus Three risk factors for coronary artery disease (high cholesterol, family history, or ongoing tobacco use) Physical Signs of heart failure (including S3 gallop, elevated neck veins filling, pulmonary rales) Hypotension ECG ST-segment depression Transient ST-segment elevation Deep T-wave inversions (particularly during pain) Laboratory Cardiac markers (elevated serum troponin I or T and/or creatine kinase-MB) Inflammatory markers—serum C-reactive protein NSTE ACS: non-ST-segment elevation acute coronary syndrome. Data from References [32–48].

dalteparin or nadroparin were shown to be at least as effective as UFH [65, 66] while enoxaparin is superior to UFH in high-risk patients [61, 64]. The long-term coadministration of clopidogrel and aspirin in high-risk NSTE ACS patients was associated with a 2.1% absolute risk reduction in the composite endpoint of cardiovascular death, MI, or stroke [67]. This benefit was accompanied by a 1% absolute increase in major bleeding observed especially in individuals undergoing early CABG coronary artery bypass grafting within 5 days of the loading dose of clopidogrel. Treatment with intravenous GP IIb/IIIa inhibitors produced a 1% absolute and 9%

Thrombosis in clinical practice

158

relative reduction in the composite endpoint of death and MI at 30 days at the expense of a 1% absolute increase in major bleeding [30, 62, 63, 68–76]. The benefit was seen predominantly among high-risk patients and among patients treated with an early PCI. Hence, high-risk patients not planned for immediate catheterization should receive tirofiban or eptifibatide [63, 68] while abciximab should be considered the agent of choice in those undergoing urgent PCI [77]. In contrast, in patients treated medically, abciximab is associated with adverse outcome and hence, it should be avoided [71]. Routine early invasive versus selective “ischemia-guided” invasive management While primary (or rescue/facilitated) PCI is largely used for patients with STEMI, it is less

Table 9.2 TIMI risk score for patients with NSTE ACS (P52). Points Age >65 years

1

>3 cardiovascular risk factors (family history, HTN, DM, cholesterol, nicotine)

1

Known coronary artery disease

1

Aspirin within 7 days

1

Severe angina pectoris (<24 h)

1

Cardiac markers positive

1

ST-segment deviation >0.5 mV

1

Total risk score

0–7

% risk of adverse cardiac events (death, MI, urgent revascularization) at 2 weeks TIMI risk score

Death, MI

Composite

1

3

5

2

3

8

3

5

13

4

7

20

5

12

26

6

19

41

NSTE ACS: non-ST-segment elevation acute coronary syndrome; HTN: hypertension; DM: diabetes mellitus; MI: myocardial infarction.

known that some patients with NSTE ACS do benefit even more from early revascularization. Early studies [28, 29] suggested that there was no short- or long-term

Coronary artery disease coronary revascularization

159

survival advantage associated with an early invasive strategy. Yet, the patients assigned to the invasive strategy had a shorter hospital stay, a reduced readmission rate, and less anti-anginal medication. Moreover, a prespecified subgroup analysis did show a benefit for the invasive approach in high-risk patients, including those with ST-segment depression on the qualifying ECG, prior aspirin users, and patients over 65 years old. In contrast to early studies, recent trial using modern antithrombotic agents and stents showed unequivocally a significant improvement in short- and long-term outcomes by using an early invasive approach [30, 31, 78, 79]. These include the FRISC II trial (aspirin, dalteparin, 61% stent), the TACTICSTIMI 18 trial (aspirin, heparin, tirofiban, 84% stent), the RITA-3 trial (aspirin, enoxaparin, 88% stent), and the small VINO trial (aspirin, heparin, 44% stent). Nevertheless, the reduction in composite endpoints (death, nonfatal MI) associated with an early invasive strategy was observed only in patients with high TIMI risk score (≥4), specifically those who were troponin-positive, or who had ST-segment depression on their admission ECG. Patients with elevated troponin are at a particularly high risk. Elevated troponin represents a marker of an unstable (exulcerated) plaque with intracoronary thrombosis. These lesions may progress to total occlusion at any time. For this reason, the mechanical stabilization by an immediate PCI is associated with the reduced rate of adverse outcomes as compared with delayed intervention or medical therapy alone. In contrast to high-risk patients, the early invasive and conservative strategies were comparable in their effects on the primary outcome in patients with low TIMI risk score, without elevated troponin or in aspirin users. Interestingly, the subjects without ST-segment depressions and women showed a trend to hazard for early intervention. In summary, up to 50% of patients with NSTE ACS do not appear to benefit from an early invasive strategy. Furthermore, the timing of PCI in the setting of intensive medical therapy appears to be critical. It seems that interval between 0 and 48 hours is optimal while the intervention beyond 48 hours is associated with a reduced benefit [29–31, 72, 80]. Recommendations Based on randomized trials, recent guidelines and authors’ personal experience, the following approach to patients with NSTE ACS can be recommended. NSTE ACS patients should be stratified in the emergency department into three categories: high-risk, intermediate-risk, or low-risk for adverse outcome (Table 9.3). The high-risk NSTE ACS patients clearly warrant early catheterization with maximal antithrombotic therapy (aspirin, enoxaparin, GP IIb/IIIa inhibitors, clopidogrel). Intravenous aspirin and heparin (UFH or preferably enoxaparin) should be initiated as soon as possible. Administration of clopidogrel can be initiated in the emergency room in patients in whom an early coronary bypass grafting is not anticipated, or it should be postponed till an angiogram rules out an urgent surgical anatomy. Small molecule GP IIb/IIIa inhibitor (tirofiban or eptifibatide) should be started as soon as possible

Thrombosis in clinical practice

160

Table 9.3 Risk stratification of patients with NSTE ACS at presentation. High-risk patients Rest angina within the previous 24 hours and one of the following: prior revascularization (PCI within 6 months or bypass surgery) ≥4 TIMI risk score transient or persistent ST-segment depression elevated cardiac markers ejection fraction <40% heart failure or hemodynamic instability sustained ventricular tachycardia Intermediate-risk patients accelerated angina within the previous 2 weeks presence of chest discomfort or pain diabetes mellitus deep T-wave inversions in more than five leads Low-risk patients CCS class I or II angina normal ECG or nonspecific ST-segment changes flat T-wave inversion negative cardiac markers NSTE ACS: non-ST-segment elevation acute coronary syndrome. Data from References [32–48, 51, 52, 55].

in all high-risk patients not considered for early catheterization or planned for medical therapy. Yet, the majority of these patients should be catheterized the following morning and patients presenting to smaller hospitals should be transported to centers with PCI facilities while receiving intravenous GP IIb/IIIa inhibitor infusion. GP IIb/IIIa infusion should not be interrupted before catheterization and in patients receiving PCI. It should continue for at least 12 hours after the procedure. Since the benefit of GP IIb/IIIa inhibitors is confined mainly to high-risk patients undergoing PCI and since PCI can be performed in only a cost-effective alternative is to administer abcix35–48% of catheterized patients [28–31, 80], imab at the time of early catheterization only to patients who prove to be suitable for PCI. The low-risk patients are not likely to benefit from catheterization or intensive antithrombotic therapy. These patients should be observed for 6–9 hours after the onset of symptoms with repeated ECG and cardiac markers 6–9 hours apart. In patients without

Coronary artery disease coronary revascularization

161

recurrent chest pain, normal ECG, and cardiac markers during the observation period, an exercise stress test should be performed immediately or within a couple of days after dismissal. Patients with positive stress test should be admitted or planned for catheterization whereas patients with negative ones could be managed medically. In patients at intermediate risk, the decisions about the use of the three classes of antithrombotics (LMWH, GP IIb/IIIa inhibitors, and clopidogrel) and an early invasive strategy should be made on the individual basis. Some of these patients may actually benefit from early revascularization. On the other hand, they may be managed conservatively with selective “ischemia-guided” catheterization. The aggressive therapy including clopidogrel, GP IIb/IIIa inhibitors, and early catheterization should be strongly considered in patients with diabetes mellitus [74]. In contrast, the routine early catheterization should be avoided in troponinnegative women. All these complex recommendations can be simplified as follows: 1. All patients with positive troponin and/or ST depressions should undergo CAG within 48 hours after admission. 2. For patients with negative troponin and without ST depressions, both approaches (invasive or conservative) are acceptable. Early (<48 h after admission) CABG for acute coronary thrombosis (occlusive and subocclusive) CABG is a well-established procedure for the treatment of advanced coronary artery disease. However, for most patients in the evolving phase of AMI, primary PCI has currently been a safe and feasible method of treatment. Emergency surgery in these patients is clearly associated with higher perioperative mortality and morbidity and consequently remains reserved for AMI patients only, in whom other interventional techniques failed or were not indicated. The American College of Cardiology (ACC) and American Heart Association (AHA) formulated the following recommendations for emergency CABG in 1999 [S1, S2]: • failed angioplasty with persistent pain or hemodynamic instability in patients with coronary anatomy suitable for surgery; • AMI with persistent or recurrent ischemia refractory to medical therapy in patients with coronary anatomy suitable for surgery who are not candidates for catheter intervention; • at the time of surgical repair of postinfarction mechanical defects; • some authors recommend surgical intervention also in case of cardiogenic shock with coronary anatomy suitable for surgery [S3]. Recent studies repeatedly reported CABG in the early postinfarction period in an emergent setting to be associated with a higher perioperative mortality, ranging from 9.1% to 15.4%, compared to elective bypass surgery (2–3%) [S4–S8]. Nevertheless, these outcomes appear clearly superior to those published previously, and should be ascribed to improved myocardial protection, advances in anesthetic and surgical techniques, and more aggressive use of cardiac assist devices. Some studies found that in-hospital mortality decreases with increasing time interval between AMI and surgery. A retrospective multicenter analysis of 44,365 patients who underwent CABG after MI

Thrombosis in clinical practice

162

revealed the mortality rates to be 12.1%, 13.6%, 4.3%, and 2.4% for the time intervals of within 6 hours, 6–23 hours, 1–7 days, and 8–14 days after transmural AMI, respectively [S6]. Similar findings were also reported by other authors [S4, S5]. If indications for emergency surgery such as structural complications, ongoing ischemia, or hemodynamic compromise are not present, CABG is recommended considering a delay of surgery. On the other hand, even the patients who underwent an emergency procedure early after transmural AMI, including those with hemodynamic instability and cardiogenic shock, can be operated on with acceptable outcomes [S3]. In these patients, the rationale for surgery is lifesaving: their prognosis is poor when treated conservatively. Surgery in these patients requires intensified hemodynamic support including intra-aortic balloon pump (IABP), percutaneous circulatory pulmonary support, or LV assist device [S3, S6, S7, S9]. Focus is on maximal myocardial protection. Cold or warm blood cardioplegia given antegradely and retrogradely, possibly coupled with warm substrate-enhanced blood cardioplegia used before releasing the cross-clamp can reduce reperfusion injury and infarction size [S10]. Although some authors believe that in acute situations arterial grafts may be avoided, others, in view of a superior longterm patency rate of the internal mammary artery (IMA), recommend using at least one arterial graft even in such cases. Other authors decide to proceed to total arterial revascularization, with arterial conduits prepared after emergent institution of cardiopulmonary bypass (CPB) in case of hemodynamic instability [S3, S7]. Emergency CABG is sometimes necessary for failed PCI. The reported incidence was about 5%, but with the use of stents in clinical practice it has decreased to about 1%. The basic strategy of surgical treatment includes simultaneous resuscitation and management in a catheterization laboratory, as early a start of surgery as possible, and complete myocardial revascularization at surgery [S11]. According to ACC/AHA, the indications for emergency CABG are as follows [S2]: • ongoing ischemia or threatened occlusion with significant myocardium at risk; • hemodynamic compromise; • foreign body in crucial anatomic position. AMI poses an increased risk to patients undergoing on-pump CABG, since myocardial damage from CPB and possibly insufficient myocardial protection may increase the necrosis and thus contribute to postoperative cardiac dysfunction. Off-pump coronary artery bypass (OPCAB) grafting has recently been proposed as an alternative technique to treat patients with AMI [S12]. Using OPCAB techniques, good results were also achieved in patients who underwent emergency surgery either soon after AMI or for ACS [S13–S16]. Aggressive use of IABP is recommended to reduce risk of circulatory collapse and subsequent conversion to CPB in these patients. Another focus is on ultrafast track anesthesia making the procedure less invasive [S17, S18], since allowing early extubation of the patient still in the operating room and, in indicated cases, avoiding the use of a ventilator during surgery (awake risk patients would also be achieved by CABG) [S19]. Better outcomes in these highreducing ventilation time to a minimum, starting post-operative rehabilitation early, decreasing incidence of pulmonary complications, and shortening the intensive care unit length of stay. Nevertheless, further

Coronary artery disease coronary revascularization

163

prospective randomized studies would be needed to assess the benefits from the novel techniques in AMI management. In contrast to surgery for evolving AMI, mechanical complications of AMI require prompt surgical repair to correct the defect as a lifesaving procedure. These complications mostly develop within the first week after AMI and usually lead to hemodynamic deterioration, low cardiac output, and progressive multiorgan failure. Introducing a Swan-Ganz catheter for hemodynamic monitoring of therapy and insertion of IABP can help stabilize the patient. In patients with postinfarction VSD, surgical repair with patch is indicated. Amplatz septal occluder has also been used for percutaneous occlusion of postinfarction VSD; nevertheless, only limited experience has been gained in this field. Papillary muscle rupture causing acute mitral valve regurgitation is an indication for mitral valve replacement; repair of the valve has also been reported. LV free wall rupture requires repair of the ventricle using a patch. There is also a sutureless technique with application of the glue to hold the patch in place. All of these procedures are complemented with CABG, if needed. Although the outcomes of surgical intervention for these complications are superior to those achievable with medical treatment, in-hospital postoperative mortality remains relatively high.

Intracoronary thrombosis as an early or late complication of revascularization procedures (“secondary clot”) By “secondary clot” we mean exclusively a new intracoronary thrombus, occurring as a consequence of revascularization procedures, performed in patients without preexisting intracoronary thrombus—thus this chapter deals only with chronic stable coronary artery disease. In patients with ACS we consider all thrombi as “primary clots,” changing its size and position depending on the treatment. Intracoronary thrombosis during and early (<24 h) after PCI Intracoronary thrombosis complicating elective PCI procedures in stable coronary artery disease patients is an extremely rare event. Without preexisting intraluminal clot and with proper antithrombotic medication before and during PCI, the new intracoronary thrombus formation is a rare consequence of a major coronary dissection. Difficult PCI procedures in calcified, tortuous, diffusely diseased coronary arteries carry a higher risk of complicating thrombosis. The thrombus formation can be triggered by major damage to endothelium, by large coronary dissection, by distal embolization leading to slow flow/no flow phenomenon, by long duration of the procedure, and also by iatrogenic air embolization.

Thrombosis in clinical practice

164

Figure 9.2 Subacute stent thrombosis in the right coronary artery. When intracoronary thrombosis complicates PCI during the procedure, it may be difficult sometimes to differentiate it from other causes for intraluminal “lucencies” such as dissection, cholesterol emboli, air emboli, etc. Intracoronary ultrasound imaging (or less frequently angioscopy) can be very helpful in this situation; however, it may slightly increase the risk by prolonging the procedure and introducing additional “foreign material” (IVUs or angioscopy catheter) to the damaged artery. Thrombosis occurring early after the procedure (i.e. within the first 24 h after departure from cath-lab) is an even rarer event in chronic stable coronary artery disease. It may occur in patients with suboptimal PCI results, where slow (or turbulent) flow tends to trigger thrombosis. The best prevention of the intracoronary thrombus (“secondary clot”) is a gentle, smooth, and short procedure with optimal result (0% residual stenosis, TIMI-3 flow). It is well known that the best antithrombotic therapy is the perfect flow. This holds true in all vessels of the human body including coronary arteries. Subacute stent thrombosis The first intracoronary stent in humans was implanted by U. Sigwart in 1987. However, during the next 7 years stents remained just a bailout treatment for healing major dissection complicating balloon angioplasty. The reason was the high incidence of stent thrombosis (8–10%). Subacute stent thrombosis in bare metallic stents occurs usually within several days of the procedure (Figures 9.2 and 9.3) but almost never beyond 4 weeks when the newly formed endothelium covers most of the internal stent surface. Subacute stent thrombosis is

Coronary artery disease coronary revascularization

165

Figure 9.3 After re-PCI with GP IIb/IIIa receptor blocker. a catastrophic complication with 30-day mortality rates in recent series of 20.8–26%, and in surviving patients, 6 month mortality rates of 20–25% [81–83]. Therefore, this devastating event needs to be prevented in every way possible. Initially, subacute vessel closure was thought to be the combination of stent thrombogenicity and inadequate antithrombotic/ antiplatelet treatment [84, 85]. Hence, many centers used a “cocktail” of available antithrombotic agents (heparin+aspirin+warfarin + dipyridamole+dextran), which inherently had 5–6% of severe bleeding complications. For these reasons, between 1987 and 1994 stent implantation was a much more risky procedure than coronary bypass surgery. This was completely changed by intracoronary ultrasound studies and by introduction of routine dual antiplatelet therapy (aspirin+ticlopidine or more recently clopidogrel) after stent deployment. Intracoronary ultrasound studies defined several risk factors for subacute stent thrombosis. The majority of subacute stent thrombosis occurs in arteries with stent under deployment (malapposition or underexpansion) [82, 86–89]. Postprocedural dissection, thrombus, and tissue prolapse are also associated with increased risk for thrombotic vessel closure especially when intraluminal material causes a reduction in the final lumen dimension [86]. In contrast, only 22% of patients with subacute stent thrombosis have optimum PCI results [86]. These data clearly proved that the insufficient stent expansion with inadequate postprocedural lumen dimensions is the most frequent cause for stent thrombosis. Accordingly, the concept of high-pressure stent implantation with stent: artery diameter ratio 1.1:1 or 1.2:1 (“negative” residual stenosis: e.g. −10% or even −20% diameter stenosis) markedly improved the outcomes of coronary stenting. It is of note that the preintervention patient/lesion characteristics are only poorly related to the risk of thrombotic vessel closure. Thus, final lesion site parameters—ones that can be controlled by the interventionalist—are the most critical. Several studies [90–94] have shown the superiority of ticlopidine to the previous “antithrombotic cocktail.” With the use of high-pressure stent implantation

Thrombosis in clinical practice

166

(“overdilatation”) and routine ticlopidine (and later clopidogrel) treatment for 4 weeks after stent implantation, the incidence of stent thrombosis fell down to 1% and severe bleeding complications are now also very rare [81, 90, 93–98]. Despite this success, subacute stent thrombosis, when it occurs, is frequently a catastrophic event—clinically appearing as STEMI or even sudden cardiac death. Hence, patients who report substantial angina-like chest pain at rest within the days to weeks after stent implantation should immediately undergo urgent cardiac catheterization. With recent introduction of drug-eluting stents, the subacute stent thrombosis has become again an important issue. These stents eluting rapamycin (CYPHER® stent) or paclitaxel (TAXUS® stent) decrease the risk of restenosis by slowing the rate and the extent of neointimal hyperplasia. On the other hand, inhibiting the endothelium growth slows stent covering with endothelium as compared with bare metallic stents. Thus, very late (several months after PCI) drug-eluting stent thrombosis may represent a new risk for patients treated with these stents. For the moment, however, its real incidence is not precisely known. Nevertheless, we currently recommend at least 3–6 months of clopidogrel therapy after PCI in patients receiving drug-eluting stents. This regimen should not be interrupted for minor bleeding or elective invasive or surgical procedures, particularly during the first 4 weeks after the procedure. In addition, atorvastatin and potentially simvastatin therapy should be avoided or replaced by pravastatin (see Section “Antithrombotic treatment and PCI—review of randomized clinical trials”). Early and late bypass graft occlusion The incidence of perioperative MI for CABG varies widely with the criteria used for diagnosis of perioperative ischemia; nevertheless, the rates reported by most studies are between 3% and 10%. Perioperative AMI is a significant factor of morbidity and mortality from coronary bypass operation. Among possible causes are graft failure, incomplete revascularization, insufficient myocardial protection, vasospasm, or atherosclerotic and air embolism. Establishing the actual cause may be difficult: Fabricius et al. [S20] reported that early coronarography revealed regular grafts and normal findings for the grafted artery in 42% of the patients with postoperative AMI, but the majority of perioperative ischemias were based on pathomorphological findings. Among the most frequent causes of early bypass graft failure are graft occlusion due to thrombosis as a possible consequence of incorrect anastomosis, damage to the graft during harvesting, graft displacement or overstretching, spasm, severe diffuse arteriosclerosis, low arterial pressure, and pathological blood clotting. Most myocardial infarctions are characterized by ST-changes, elevated CK (creatine kinase) and CK-MB (CK cardiac isoenzyme) levels, impaired kinetics at echocardiography, and ventricular arrhythmias. Hemodynamics are usually normal in these patients, but low cardiac output with hemodynamic deterioration and cardiogenic shock can also be present. Conservative treatment may be used in patients with stable hemodynamics, although recent papers support a more aggressive approach [S20, S21]. Early angiography in patients with perioperative ischemia after CABG can establish accurate diagnosis in most of them. This technique can be safely used in stable patients and allows early reintervention and salvage of myocardium, following both PCI and surgical

Coronary artery disease coronary revascularization

167

reintervention. Early repeat reintervention in stable patients after graft failure has a favorable outcome. In unstable patients with hemodynamic deterioration, acute thrombosis affects large and important coronary artery. Angiography is not indicated in these patients who should be transferred directly to the operating room. More aggressive use of IABP or ventricular assist device is fully justified. Nevertheless, high mortality is to be expected even with reanastomosis of the occluded coronary artery. The left IMA and saphenous vein grafting is a standard procedure in direct myocardial revascularization. Late bypass graft occlusion is reported in 10–40% of venous grafts within the first year and in 20–50% of venous grafts at 5 years after CABG. Ten years after CABG, 60% of venous grafts remain patent, about 50% being without signs of narrowing [S22, S23]. Neointimal hyperplasia and atherosclerosis are considered to be the most frequent causes of occlusion (Figures 9.4 and 9.5). In contrast, it has been reported that if the IMA is properly prepared and anastomosed to the left anterior descending artery, it has a patency of 90% at 10 years and can improve 10-year survival by 10% [S24]. The reason is that intimal hyperplasia does not occur in the IMA, and that IMA, as a less spastic graft, is the conduit of choice for myocardial revascularization. These excellent results contributed to a wider use for grafting of other arteries such as bilateral mammary artery, right gastroepiploic artery, radial artery, and inferior epigastric artery. To prevent late graft failure, crucial steps are careful perioperative tissue manipulation (“no touch technique”) during harvesting and

Figure 9.4 Degenerated saphenous graft with multiple atherothrombotic obstructions.

Thrombosis in clinical practice

168

Figure 9.5 After direct stent implantation. avoidance of graft distension and thus endothelial injury [S25]. Postoperative antiplatelet therapy should be instituted within 48 hours after surgery.

Antithrombotic treatment before, during, and after revascularization procedures Antithrombotic treatment and PCI—review of randomized clinical trials PCI is often associated with an intraluminal thrombus formation. The thrombotic vessel closure resulting in periprocedural death, MI, or urgent coronary revascularization complicates approximately 1–12% of procedures [98–101]. Hence, an adjunctive pharmacotherapy aiming at prevention of thrombus formation has become an integral part of all PCI procedures (Table 9.4). Antiplatelet therapy Acetyl salicylic acid. In patients treated with balloon angioplasty, aspirin alone can reduce the incidence of postprocedural death or MI [102–106]. The optimal dose and timing of aspirin administration has not been established. Nevertheless, a dose of 75–325 mg, given at least 2 hours before PCI is recommended [103, 107]. Yet, despite pretreatment with aspirin, abrupt coronary closure complicates 1–8% of PCI procedures [100, 108, 109]. Thienopyridines. Data from randomized clinical trials suggested that the more potent platelet inhibition achieved by adding thienopyridine derivatives to aspirin prior to PCI is associated with a low rate of ischemic complications following balloon angioplasty or

Coronary artery disease coronary revascularization

169

coronary stent deployment [67, 90–94, 104, 106, 110–112]. Clopidogrel confers similar protection against subacute stent thrombosis

Table 9.4 Dosing of major antithrombotic agents. Loading dose Abciximab

PCI 0.25 mg/kg i.v. bolus

Maintenance dose 0.125 µg/kg/min (max 10 µg/min) 12 h i.v. infusion

Eptifibatide PCI 180 µg/kg i.v. boluses 10 min apart ACS 180 µg/kg i.v, bolus

2 µg/kg/min 18–24 h i.v. infusion 2 µg/kg/min 72–96 h i.v. infusion

Tirofiban

ACS 0.4 µg/kg/min i.v. infusion/30 min

0.1 µg/kg/min 48–108 h i.v. infusion

Aspirin

500 mg i.v.

75–325 mg per os o.d. lifelong

Clopidogrel 300–600 mg per os

75 mg per os o.d. lifelong?

Ticlopidine

PCI 500 mg per os

250 mg per os b.i.d. for 14–30 days

Heparin

PCI 5,000 U to 200 U/kg i.v. bolus ACS 5,000–10,000 U i.v. bolus

i.v. infusion with target APPT between 50 and 70 s

Dalteparin

120 IU/kg s.c. b.i.d.

Enoxaparin

1 mg/kg s.c. b.i.d.

PCI: percutaneous coronary intervention; ACS: acute coronary syndrome; i.v.: intravenous; s.c.: subcutaneous.

and ischemic cardiac events as ticlopidine [67, 95, 97, 112–114]. Hence, aspirin with ticlopidine or with clopidogrel should be routinely administered to all patients undergoing coronary stenting. The recommended duration of therapy with thienopyridines following stent deployment is 30 days while aspirin continues lifelong. Clopidogrel shows faster onset of action, lower incidence of neutropenia and it is better tolerated than ticlopidine [113, 115–117]. Hence, it has become the preferred agent after stent deployment in many institution. Furthermore, in the PCI CURE substudy, long-term coadministration of clopidogrel and aspirin (mean duration 9 months) was associated with a significant reduction in the combined outcome of cardiovascular death, MI, or rehospitalization [67, 118]. Of interest, several studies demonstrated [112, 119–122] the adverse interaction between clopidogrel and atorvastatin and potentially simvastatin, which may reduce the antithrombotic effects of clopidogrel. This interaction is not present for pravastatin. Nevertheless, with future growing indications for long-term clopidogrel therapy in cardiovascular medicine, this issue warrants prospective testing. GP IIb/IIIa inhibitors. Abciximab, eptifibatide, and tirofiban in conjunction with PCI were studied in a number of trials (Table 9.5). Abdximab has the most robust data and it is the only GP IIb/IIIa inhibitor that decreased long-term mortality as compared to placebo [123]. Short- and long-term reduction of composite endpoint (death, MI, or revascularization) was observed in patients with urgent, early, or elective PCI, balloon

Thrombosis in clinical practice

170

angioplasty or stenting, NSTE ACS, or STEMI [21, 30, 77, 123–127]. The TARGET trial [77] demonstrated superiority of abciximab over tirofiban for reducing ischemic outcomes following coronary stent deployment. The RESTORE trial [128] failed to show a significant reduction in the primary endpoint with tirofiban versus placebo. Combination of

Table 9.5 Major trials using GP IIb/IIIa inhibitorsin PCI. Abciximab Urgent PCI (for STEMI)

CADILLAC, ADMIRAL, RAPPORT

Early PCI (for NSTE ACS)

EPIC, EPILOG, TARGET

Elective PCI

EPISTENT

Balloon angioplasty

EPIC, EPILOG, CADILLAC, RAPPORT

Stenting

EPISTENT, TARGET, CADILLAC, ADMIRAL

NSTE ACS

EPIC, EPILOG, TARGET

STEMI

CADILLAC, ADMIRAL, RAPPORT

Tirofiban

TARGET, RESTORE, TACTICS-TIMI 18

Eptifibatide

ESPRIT

Abbreviations in text. Data from References [67, 90–94, 104, 106, 110–112].

tirofiban and an early invasive strategy was associated with an improved outcome in high-risk patients with NSTE ACS (TACTICS-TIMI 18, Neumann et al.) [30, 80]. Eptifibatide administered in sufficient dose is superior to placebo in patients undergoing elective stenting (ESPRIT) [129]. In summary, abciximab administered periprocedurally is considered the agent of choice in high-risk patients undergoing PCI while eptifibatide can be an option in lower-risk patients. The present data do not support the general use of tirofiban as an adjunctive periprocedural therapy in PCI. Yet some high-risk NSTE ACS patients not planned for immediate intervention may benefit from tirofiban infusion. Anticoagulants Unfractionated heparin. Data supporting the administration of UFH during PCI are largely empiric and nonrandomized [130–133]. In the early pre-GP IIb/IIIa inhibitor, prethienopyridines and pre-stent era, usually a 10,000 U (up to 20,000 U) UFH i.v. bolus was administered, with an additional 5,000 U UFH i.v. bolus as needed, mainly during lengthy procedures (>2 h). In patients undergoing an elective stent deployment, several studies [134–136] have questioned the need for such high levels of anticoagulation and have suggested equivalence in clinical outcomes for lower doses of heparin (a 100 U/kg or 5,000 U bolus) when compared with a higher-dose regimen (a 15,000 U bolus) during

Coronary artery disease coronary revascularization

171

PCI. In patients receiving GP IIb/IIIa inhibitors during PCI, the total heparin dose correlates with risk for hemorrhage [126, 137, 138]. Hence, lower doses of UFH (30–60 U/kg, up to a maximum of 5,000 U) and partial reversal of heparin anticoagulation by i.v. protamine (10–20mg) are advocated [139]. Low-molecular-weight heparin. Several trials [140–144] suggested that LMWH monotherapy is safe and efficacious in the setting of elective and urgent PCI (NICE 1, REDUCE [141]), event with coadministration of GP IIa/IIIb inhibitor (NICE 4 [142, 143], CRUISE). PCI can be performed within 8 hours of the last s.c. administration of LMWH (dalteparin 120 IU/kg b.i.d., enoxaparin 1 mg/kg b.i.d.) without the need for further supplementation [31, 142, 145, 146]. An additional i.v. bolus of LMWH (dalteparin 60 IU/kg, enoxaparin 0.3 mg/kg) should be given if PCI is performed between 8 and 12 hours after the last s.c. dose [143, 147, 148]. Alternatively, UFH may be substituted for LMWH if PCI is to be performed for more than 8–12 hours following the last s.c. dose of LMWH [149]. Ongoing large trials (A to Z, SYNERGY) will provide additional information on the safety and efficacy of enoxaparin and GP IIb/IIIa inhibition in the PCI setting. Direct antithrombin therapy (hirudin, Hirulog, Argatroban, Efegatran, bivalirudin). Due to only marginal benefits when compared with UFH, and on the downside significantly higher frequency of side effects, and high costs, at present, these agents are only indicated in patients with heparin-induced thrombocytopenia who need anticoagulation. Oral anticoagulants. In summary, based on the available evidence at the time of writing, routine oral anticoagulant therapy is not indicated after PCI procedure. Antithrombotic treatment and PCI—summary of existing guidelines Neither the European nor the Czech Society of Cardiology has specific guidelines for antithrombotic therapy during PCI. Table 9.6 and the following text are adopted from the only existing guidelines: the ACC/AHA Guidelines for Percutaneous Coronary Intervention [107]. 1. Aspirin, Ticlopidine, Clopidogrel, Warfarin. Aspirin reduces the frequency of ischemic complications after coronary angioplasty. Although the minimum effective aspirin dosage in the setting of coronary angioplasty has not been established, an empiric dose of aspirin, 80–325 mg, given at least 2 hours before the PCI procedure is generally recommended. Thienopyridine derivatives ticlopidine and clopidogrel have been routinely used as alternative antiplatelet agents in aspirinsensitive patients during coronary angioplasty. In elective settings, ideally ticlopidine and clopidogrel should be given for at least 72 hours prior to the procedure in order to achieve maximum platelet inhibition. Ticlopidine has a number of important side effects, including gastrointestinal distress (20%), cutaneous rashes (4.8–15%), and abnormal liver function tests. The most severe side effect is severe neutropenia, occurring in approximately 1 % of patients. Ticlopidine-induced neutropenia is generally reversible after its discontinuation, although infrequent episodes of sepsis and death have been reported. Rare (<1:1,000), but fatal, episodes of thrombotic thrombocytopenic purpura have also been reported, and patients receiving ticlopidine should be monitored for the occurrence of this untoward sequelae. Shorter durations (10– 14 days) of ticlopidine therapy may reduce untoward side effects of therapy while maintaining therapeutic efficacy. Clopidogrel, 300 mg loading dose followed by 75 mg

Thrombosis in clinical practice

172

o.d., may be used as an alternative to ticlopidine in patients undergoing stent placement. The routine use of warfarin is no longer recommended after stent implantation, unless there are other indications for its use, such as a poor LV function, atrial fibrillation, or mechanical heart valves.

Table 9.6 ACC recommendations for pharmacologic management of patients undergoing PCI-modified by P.Widimsky for the purpose of this book. AP, UAP NSTEMI, NQMI

STEMI, QMI

Aspirin

+

+

Clopidogrel

+

+

Ticlopidin

+

+

Warfarin

–

–

GPIIb/IIIa blockers

+

+

(PW: only selected pts)

(PW: only selected pts)

UFH

+

+

LMWH

+

+

ACC: American College of Cardiology; UFH: unfractionated heparin; LMWH: low molecular weight heparin.

2. Glycoprotein IIb/IIIa Inhibitors. These agents have reduced the frequency of ischemic complications after coronary angioplasty. a. Abciximab. The data suggest that prophylactic adjunctive platelet GP IIb/IIIa blockade improves the clinical outcomes of patients who require unplanned coronary stent deployment. One putative limitation of abciximab is the potential for immune-mediated hypersensitivity reactions following subsequent readministration. With the first administration, human antichimeric antibodies (HACA) form in approximately 6% of patients. The implications of HACA, however, are unclear. Among 500 patients enrolled in the ReoPro Readministration Registry (R3), there were no cases of anaphylaxis or other allergic manifestations whether or not HACA was present, and HACA was not predictive of any other measure of complication or success. From the R3 Study, HACA has been shown to be an IgG (not IgE) immunoglobulin that does not neutralize abciximab. The more worrisome clinical phenomenon associated with readministration is the potential for increased rates of thrombocytopenia. In the 500-patient Registry, a 4.4% incidence in thrombocytopenia (to a platelet count of <100×109/L) was observed, with half of the patients developing acute profound thrombocytopenia (to a platelet count of <20×109/L). This potential complication should always be monitored when treating a patient with abciximab. b. Eptifibatide. The clinical utility of eptifibatide, a short-acting cyclic heptapeptide that also inhibits the GP IIb/IIIa receptor, appears to be comparable with abciximab.

Coronary artery disease coronary revascularization

173

c. Tirofiban. Tirofiban is a nonpeptidyl tyrosine derivative that produces a dosedependent inhibition of GP IIb/IIIa mediated platelet aggregation. The clinical effect of tirofiban during coronary angioplasty is similar to previous drugs, but supported by less evidence. Based on the numerous trials to date, intravenous GP IIb/IIIa receptor inhibitors should be considered in patients undergoing coronary angioplasty, particularly in those with unstable angina or with other clinical characteristics of high-risk. There is no consistent evidence that the GP IIb/IIIa inhibitors reduce the frequency of late restenosis. 3. Heparin. Intravenous UFH prevents clot formation at the site of arterial injury and on coronary guidewires and catheters used for coronary angioplasty. While the intensity of anticoagulation with UFH is generally determined using activated partial thromoboplastin times (APTTs), these values are less useful for monitoring anticoagulation during coronary angioplasty because higher levels of anticoagulation are needed than can be discriminated with the APTT alone. Instead, the activated clotting time (ACT) has been more useful to follow heparin therapy during coronary angioplasty. The Hemochron and Hemo Tec devices are commonly used to measure ACT values during coronary angioplasty. The Hemochron ACT generally exceeds the Hemotech ACT by 30–50 s, although considerable measurement variability exists. Empiric recommendations regarding heparin dosage during coronary angioplasty have been proposed, but ACT levels after a fixed dose of UFH may vary substantially due to differences in body size, concomitant use of other medications, including intravenous nitroglycerin, and in the presence of ACSs that increase heparin resistance. The relationship between the level of the ACT and development of ischemic complications during coronary angioplasty has been controversial. Whereas some studies have identified an inverse relationship between the initial ACT and the risk of ischemic events, others found either no relationship or a direct relationship between the degree of anticoagulation and occurrence of complications. It is generally felt that very high levels (ACTs >400–600 s) of periprocedural anticoagulation are associated with an increased risk for bleeding complications. The results of these limited studies suggests that heparin is an important component for PCI, despite dosing uncertainties and an unpredictable therapeutic response with the unfractionated preparation. Higher levels of anticoagulation with heparin are roughly correlated with therapeutic efficacy in the reduction of complications during coronary angioplasty, albeit at the expense of bleeding complications at very high levels of heparin dosing. It appears that weight-adjusted heparin dosing may provide a clinically superior anticoagulation method over fixed heparin dosing, although definitive studies are lacking. Routine use of UFH after an uncomplicated coronary angioplasty is no longer recommended, and may be associated with more frequent bleeding events, particularly when platelet GP IIb/IIIa inhibitors are used. Subcutaneous administration of UFH may provide a safer and less costly means of extending antithrombin therapy than intravenous UFH, if there are clinical reasons to continue anticoagulation, such as residual thrombus or significant residual dissections. Some patients with unstable angina are treated with LMWH prior to coronary angioplasty. Anticoagulation monitoring is not routinely possible with LMWH, and conventional dosages of UFH are currently recommended. Conventional ACT monitoring methods may underestimate the true degree of periprocedural anticoagulation with LMWH. Use of LMWH as the sole anticoagulant during PCI is not supported at this time

Thrombosis in clinical practice

174

in the absence of absolute or relative contraindications to UFH, although data from clinical trials of these agents administered alone or in conjunction with GP IIb/IIIa blockade are forthcoming. a. Heparin Dosing Guidelines. In those patients who do not receive GP IIb/IIIa inhibitors, sufficient UFH should be given during coronary angioplasty to achieve an ACT of 250– 300 s with the Hemo Tec device and 300–350 s with the Hemochron device. Weightadjusted bolus heparin (70–100 IU per kg) can be used to avoid excess anticoagulation. If the target values for ACT are not achieved after a bolus of heparin, additional heparin boluses (2,000–5,000 IU) can be given. Early sheath removal should be performed when the ACT falls to less than 150–180 s. The UFH bolus should be reduced to 50–70 IU per kg when GP IIb/IIIa inhibitors are given in order to achieve a target ACT of 200 s using either the Hemo Tec or Hemochron device. Currently recommended Target ACT for eptifibatide and tirofiban is less than 300 s during coronary angioplasty. Post-procedural heparin infusions are not recommended during GP IIb/IIIa therapy. Antithrombotic treatment and CABG—review of recent randomized clinical trials Mortality and morbidity after CABG is dependent on graft patency after surgery. Since platelets play a crucial role in pathophysiology of thrombosis, antiplatelet therapy after surgical revascularization provides safe and effective prevention of thrombotic graft occlusion. Many studies deal with benefits and risks of antiplatelet therapy after CABG and a review of key references was recently presented by Bansal [S26]. Several randomized trials demonstrated better saphenous vein bypass graft patency with aspirin therapy than the untreated group in the late 1980s [S27–S29]. In one of the recent prospective randomized studies with 5,065 patients, Mangano and colleagues present the data on the use of aspirin within 48 hours after CABG. Among the patients who either were or were not given aspirin, the mortality rates reached 1.3% and 4%, respectively (p<0.001). The patients with aspirin therapy had substantially lower rates of MI (by 48%), stroke (by 50%), renal failure (by 74%), and bowel infarction (by 62%) [S30]. In 2001, Stein and colleagues, based on the available literature data on antiplatelet therapy, elaborated guidelines that recommend aspirin therapy doses 325 mg o.d. starting 6 hours after operation. Clopidogrel is to be used in aspirin allergic patients 6 hours after surgery at a dose of 300 mg followed by 50–100 mg daily [S31]. Antiplatelet therapy has been shown to be of benefit, and early aspirin use after CABG should be standardly administered in postoperative care [S32]. Antithrombotic treatment and CABG—summary of existing guidelines According to the recommendations of the ACC and AHA guidelines for CABG surgery of 1999 [S2], antiplatelet therapy provides extra postoperative benefit. Aspirin is known to reduce the incidence of postoperative vein graft closure, but similar effect on arterial graft patency was not reported. Beneficial effect on vein graft patency was described with

Coronary artery disease coronary revascularization

175

aspirin therapy instituted 1–24 hours after operation and dosed from 100 mg o.d. to 325 mg three times daily. Ticlopidin can also be used for prevention of thrombosis of the graft, but has no advantage over aspirin. Nevertheless, it can be given as an alternative to aspirin-allergic patients. Nor does dipyridamol in combination with aspirin increase the graft patency rate; the same is true of warfarin, possibly associated with higher rates of bleeding complications. Aspirin is a drug of choice for prevention of early saphenous graft thrombosis.

Bleeding complications related to the antithrombotic treatment in conjunction with revascularization procedures Bleeding complications at the arterial puncture site Arterial access is inherently related to the risk of arterial bleeding at the puncture site. Radial artery access has more frequent but less severe complications than femoral artery access. Bleeding from the radial artery is usually easy to control and its consequences are usually more or less cosmetic. In the following text, we are focused on the femoral artery bleeding and bleeding consequences may be life because this is the most frequently used access threatening. Simple groin hematoma. A large subcutaneous or intramuscular hematoma complicates c.2% of elective PCI procedures. Its incidence is doubled (4–5%) in PCI done for ACS due to the use of intensive antithrombotic treatment. In case of an asymptomatic simple groin hematoma, no treatment is usually necessary. One must be certain, however, that bleeding has already stopped and does not continue. Groin hematoma with hypotension or hemorrhagic shock. This is an entirely different story: it may be a life-threatening event when not recognized and treated immediately. Hypotension (or shock) is a sign of major and continuing bleeding. Immediate aggressive compression of the puncture site with immediate volume expansion and blood transfusion usually stabilizes the patient quickly. Pseudoaneurysm of the femoral artery. A small bleeding from the puncture site may form a new “cavity” between the artery and skin. It is easily recognized by physical examination: sharply delineated egg-shaped, slightly pulsating resistance at the puncture site with a weak short mid-systolic murmur (1/6) over it. The artery itself is usually not damaged—it is just the puncture that has not closed spontaneously and that communicates with the cavity in front of it. The diameter of the pseudoaneurysm varies from between 2 cm and 20 cm or more. When it is less than 10 cm, it can be almost always successfully treated conservatively by prolonged repeated compression. The rare extremely large pseudoaneurysms require surgical treatment.

Thrombosis in clinical practice

176

Potentially fatal bleedings after PCI: retroperitoneal bleeding, cardiac tamponade, intracranial bleeding, gastrointestinal bleeding Retroperitoneal bleeding. It is very rare but carries high mortality. It is caused either by direct damage (dissection or perforation) of the arteries in their retroperitoneal course (abdominal aorta, iliac arteries) or it complicates an improperly “high” puncture site (when puncture needle enters, in fact, the distal iliac artery instead of femoral artery). In the latter situation, a simple bleeding (which normally would occur subcutaneously in the groin and will be easily recognized) continues unrecognized until patient develops hemorrhagic shock. The clinical diagnosis is difficult (non-specific signs of internal bleeding in a patient after PCI). Hence, each patient with suspected retroperitoneal bleeding should undergo urgent CT scan. Treatment is usually surgical. Cardiac tamponade. Fortunately, this is a very rare event. Perforation of the coronary artery by the intracoronary guide wire usually does not cause any problems to the patient. However, when not recognized properly by the operator, and balloon introduced over the extraluminally positioned guide wire, a disastrous cardiac tamponade may be a consequence. In technically well-performed procedures, this may occur in severely calcified vessels, when the dilated and crashed calcium masses perforate the arterial wall as a blade. Treatment includes implantation of intracoronary stent graft, and/or pericardial drainage, and/or cardiac surgery. However, the progress may be so fast that none of the mentioned procedures can be completed timely. Intracranial bleeding. The risk of intracranial bleeding is highest among elderly lowweight ladies. In females over 75 years and in females less than 60 kg, one must be always very careful with any antithrombotic treatment—either in conjunction with PCI or without. Intracranial bleeding is not related to PCI procedure directly; it is related “only” to the antithrombotic treatment used during and after the procedure. Gastrointestinal bleeding. Gastrointestinal bleeding is also not related to PCI but to the antithrombotic therapy accompanying the procedure. When gastrointestinal bleeding occurs in patient with a recently implanted stent, the physicians face a difficult decision: to discontinue antithrombotic medication in order to help gastrointestinal healing at the price of high risk of stent thrombosis or to continue antithrombotic medication in some controllable form (e.g. heparin) with simultaneous treatment of gastrointestinal problem (decreased risk of stent thrombosis at the price of increased risk of recurrent gastrointestinal bleeding). This must always be managed highly individually balancing the cardiac against the gastrointestinal risks of the patient (e.g. stent thrombosis in LAD supplying chronic akinetic infarct area is probably less risky than severe gastrointestinal bleeding or vice versa: stent in the left main coronary artery usually means that gastrointestinal risk is lower than stent thrombosis risk). Bleeding complications after cardiac surgery Bleeding after cardiac surgery remains a hot topic. According to the currently available data, 5–7% of cardiac surgery patients show blood loss of more than 2 L within 24 hours postoperatively and 3.6–4.2% of patients need early postoperative surgical re-exploration for bleeding [S33, S34]. Surgical sources of bleeding are identified in less than 50% of

Coronary artery disease coronary revascularization

177

cases. Nevertheless, surgical re-exploration can be associated with negative outcomes such as increased mortality, renal failure, arrhythmias, infectious complications, and prolonged mechanical ventilation [S35]. Nonsurgical bleeding is generally related to a combination of several alterations in the hemostatic system during CPB, and potentially to preoperative use of longer acting antiplatelet or antithrombotic agents [S36]. One of the possible steps to be taken in order to reduce postoperative bleeding is better management of anticoagulation during CPB to achieve higher inhibition of thrombin formation and better protection of hemostatic mechanisms involved in postoperative bleeding control [S37]. Heparinization during CPB is adjusted according to ACT. ACT can be falsely prolonged by hemodilution, hypothermia, and use of antifibrinolytic agents and may not reflect adequately the actual anticoagulation. There is evidence that if heparin is dosed according to blood heparin concentration measurements rather than according to ACT the final heparin dosage is up to 35% higher, but at the same time, consumption of coagulation factors and platelets decreases and clinically, the blood loss is reduced [S38, S39]. Antithrombin III, heparin cofactor, plays a crucial role: its deficiency may lead to inadequate heparinization during CPB. Antifibrinolytic agents, that is synthetic analogues of lysin such as tranexamid acid and ε-aminocaproic acid or more effective aprotinin with more complex activity (including anti-inflammation) and consequently called a”broad-spectrum antifibrinolytic,” have been used perioperatively in many cardiac surgery centers. Levi’s meta-analysis of 1999, focused on results from 72 studies with more than 8,000 patients, proved antifibrinolytic agents to be effective in reducing blood loss and re-exploration and aprotinin to significantly decrease mortality in on-pump surgery [S40]. A few studies deal with the use of antifibrinolytic agents in off-pump cardiac surgery. Prospective, randomized studies in patients undergoing off-pump surgery for coronary heart disease show reduction of postoperative blood loss in patients given antifibrinolytic agents [S41-S43]. Higher incidence of MI suggestive of higher frequency of bypass graft occlusion was not reported. A novel approach to life-threatening bleeding intractable by either surgery or conventional pharmacotherapy and hemotherapy is the use of recombinant activated factor VII [S44]. Supranormal-therapeutic doses of this factor have a potent procoagulation effect at sites of tissue damage (where expression of tissue factor is induced), and should act locally without causing undesirable generalized hypercoagulation. Cardiac tamponade is an uncommon complication after open heart surgery, resulting in increased morbidity and mortality. Early cardiac tamponade is more common and generally associated with excessive mediastinal bleeding and reduction of cardiac output. Unless rapidly diagnosed and treated, it leads to deterioration of the circulation. In these situations, transfer to the operating room and emergency re-exploratation with early decompression are indicated. Less frequent, late cardiac tamponade occurs more than 1 week after surgery. The mechanism involved in its development is not completely understood but seems to be related to anticoagulant therapy [S45, S46]. Progressive development of signs of low cardiac output and refractory heart failure are typically described. The standard approach is surgical drainage by a subxiphoid incision or

Thrombosis in clinical practice

resternotomy; nevertheless, pericardiocenthesis.

many

patients

can

178

be

treated

by

percutaneous

References [1] De Wood MA, et al. Prevalence of total coronary occlusion during the early hours of transmural myocardial infarction. N Engl J Med 1980; 33: 897–904. [2] Rentrop P, Blanke H, Karsch KR, Kaiser H, Kostering H, Leitz K. Selective intracoronary thrombolysis in acute myocardial infarction and unstable angina pectoris. Circulation 1981; 63: 307–16. [3] Kennedy JW, Ritchie JL, Davis K, et al. The Western Washington intracoronary streptokinase trial in acute myocardial infarction: one year survival. Circulation 1984; 70(Suppl. II): 324. [4] GISSI Study Investigators. Effectiveness of intravenous thrombolytic treatment in acute myocardial infarction. Gruppo Italiano per lo Studio della Streptochinasi nell’Infarto Miocardico (GISSI). Lancet 1986; 1:397–402. [5] Randomised trial of intravenous streptokinase, oral aspirin, both, or neither among 17,187 cases of suspected acute myocardial infarction: ISIS-2. ISIS-2 (Second International Study of Infarct Survival) Collaborative Group. Lancet 1988; 2:349–60. [6] Meyer J, Merx W, Dorr R, Lambertz H, Bethge C, Effert S. Succesful treatment of acute myocardial infarction shock by combined percutaneous transluminal coronary recanalization and percutaneous transluminal coronary angioplasty. Am Heart J 1982; 103:132–8. [7] Hartzler GO, Rutherford BD, McConnay DR, et al. Percutaneous transluminal coronary angioplasty with and without thrombolytic therapy for treatment of acute myocardial infarction. Am Heart J 1983; 106:965–73. [8] Zijlstra F, de Boer MJ, Hoorntje JCA, Reiffers S, Reiber JHC, Suryapranata H. A comparison of immediate coronary angioplasty with intravenous streptokinase in acute myocardial infarction. N Engl J Med 1993; 328:680–4. [9] Grines CL, Browne KF, Marco J, et al. A comparison of immediate angioplasty with thrombolytic therapy for acute myocardial infarction. N Engl J Med 1993; 328:673–9. [10] Gibbons RJ, Holmes DR, Reeder GS, Bailey KR, Hopfenspirger MR, Gersh BJ. Immediate angioplasty compared with the administration of a thrombolytic agent followed by conservative treatment for myocardial infarction. N Engl J Med 1993; 328:685–91. [11] Weaver WD, Simes RJ, Betriu A, et al. Comparison of primary coronary angioplasty and intravenous thrombolytic therapy for acute myocardial infarction. JAMA 1997; 278:2093–8. [12] Vermeer F, Oude Ophuis AJ, vd Berg EJ, et al. Prospective randomized comparison between thrombolysis, rescue PTCA, and primary PTCA in patients with extensive myocardial infarction admitted to a hospital without PTCA facilities: a safety and feasibility study. Heart 1999; 82:426–31. [13] Widimsky P, Groch L, Zelízko M, Aschermann M, Bednář F, Suryapranata H. Multicenter randomized trial comparing transport to primary angioplasty vs. immediate thrombolysis vs. combined strategy for patients with acute myocardial infarction presenting to a community hospital without a catheterization laboratory. The PRAGUE Study. Eur Heart J 2000; 21:823– 31. [14] Widimsky P, Budĕšínský T, Voráč D, Groch L, Zelízko M, Aschermann M, Branny M, Stásek J, Formánek P on behalf of the “PRAGUE” Study Group Investigators, et al. Long distance transport for primary angioplasty versus immediate thrombolysis in acute myocardial infarction: final results of the randomized national multicenter trial “PRAGUE-2.” Eur Heart J 2003; 24:94–104. [15] Andersen HR, Nielsen TT, Rasmussen K, et al. A comparison of coronary angioplasty with fibrinolytic therapy in acute myocardial infarction. N Engl J Med 2003; 349:733–42.

Coronary artery disease coronary revascularization

179

[16] Grines C, Westerhausen DR Jr, Grines LL for the Air PAMI Study Group. A randomized trial of transfer for primary angioplasty versus on-site thrombolysis in patients with high-risk myocardial infarction: the Air Primary Angioplasty in Myocardial Infarction study. J Am Coll Cardiol 2002; 39:1713–19. [17] Widimsky P, Janoušek S, Vojáček J. z @@ CKS. Doporučení pro diagnostiku a léčbu akutního infarktu myokardu (Q-typ/s elevacemi ST/s raménkovým blokem). Cor Vasa 2002;44: K123–K143. [18] Morales P, Heuser R. Embolic protection devices. J Interv Cardiol 2002; 15:485–90. [19] Walters DL, Harding SA, Palacios IF, et al. The use of mechanical devices as adjuncts to intracoronary stenting. Curr Opin Cardiol 2001; 16:300–5. [20] Grines CL, Cox DA, Stone GW for the Stent-PAMI Study Group: Coronary Angioplasty with or without Stent Implantation for Acute Myocardial Infarction. N Engl J Med 1999; 341:1949– 56. [21] Stone GW, Grines CL, Cox DA, Garcia E, Tcheng JE, Griffin JJ, Guagliumi G, Stuckey T, Turco M, Carroll JD, Rutherford BD, Lansky AJ. Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications (CADILLAC) Investigators. Comparison of angioplasty with stenting, with or without abciximab, in acute myocardial infarction. N Engl J Med 2002; 346:957–66. [22] Outcome of attempted rescue coronary angioplasty after failed thrombolysis for acute myocardial infarction. The CORAMI Study Group. Cohort of Rescue Angioplasty in Myocardial Infarction. Am J Cardiol 1994; 74:172–4. [23] Ellis SG, da Silva ER, Heyndrickx G, Talley JD, Cernigliaro C, Steg G, Spaulding C, Nobuyoshi M, Erbel R, Vassanelli C, et al. Randomized comparison of rescue angioplasty with conservative management of patients with early failure of thrombolysis for acute anterior myocardial infarction. Circulation 1994;90:2280–4. [24] McKendall GR, Forman S, Sopko G, Braunwald E, Williams DO. Value of rescue percutaneous transluminal coronary angioplasty following unsuccessful thrombolytic therapy in patients with acute myocardial infarction. Thrombolysis in Myocardial Infarction Investigators. Am J Cardiol 1995; 76:1108–11. [25] Gibson CM, Cannon CP, Greene RM, Sequeira RF, Margorien RD, Leya F, Diver DJ, Baim DS, Braunwald E. Rescue angioplasty in the thrombolysis in myocardial infarction (TIMI) 4 trial. Am J Cardiol 1997; 80:21–6. [26] Juliard JM, Himbert D, Cristofini P, Desportes JC, Magne M, Golmard JL, Aubry P, Benamer H, Boccara A, Karrillon GJ, Steg PG. A matched comparison of the combination of prehospital thrombolysis and standby rescue angioplasty with primary angioplasty. Am J Cardiol 1999; 83: 305–10. [27] Bonnefoy E, Lapostolle F, Leizorovicz A, Steg G, McFadden EP, Dubien PY, Cattan S, Boullenger E, Machecourt J, Lacroute JM, Cassagnes J, Dissait F, Touboul P. Comparison of Angioplasty and Prehospital Thrombolysis in Acute Myocardial Infarction study group. Primary angioplasty versus prehospital fibrinolysis in acute myocardial infarction: a randomised study. Lancet 2002; 360:825–9. [28] The TIMI-IIIB investigators. Effects of tissue plasminogen activator and a comparison of early invasive and conservative strategies in unstable angina and non-Q-wave myocardial infarction. Results of the TIMI-IIIB trial. Circulation 1994; 89:1545–56. [29] Boden WE, O’Rourke RA, Crawford MH, et al, for the Veterans Affairs Non-Q-Wave Infarction Strategies in Hospital (VANQWISH) trial investigators. Outcomes in patients with acute non-Q-wave myocardial infarction randomly assigned to an invasive as compared with a conservative management strategy. N EngJ Med 1998; 338: 1785–92. [30] Cannon CP, Weintraub WS, Demopoulos LA, et al., for the TACTICS-Thrombolysis in Myocardial Infarction-18 investigators. Comparison of early invasive and conservative strategies in patients with unstable coronary syndromes treated with the glycoprotein IIb/IIIa inhibitor tirofiban. N EnglJ Med 2001; 344:1879–87.

Thrombosis in clinical practice

180

[31] Fragmin and Fast Revascularization during InStability in Coronary artery disease (FRISC II) Investigators. Invasive compared with non-invasive treatment in unstable coronary artery disease: FRISC II prospective randomized multicentre study. Lancet 1999; 354:708–15. [32] Cannon CP, Braunwald E. Unstable angina. In: Braunwald E, Zipes DP, Libby P (eds), Heart Disease: A Textbook of Cardiovascular Medicine. Philadelphia, PA: W.B. Saunders Company; 2001:1232–71. [33] Braunwald E, Antman EM, Beasley JW, et al. ACC/AHA guidelines for the management of patients with unstable angina and non-ST-segment elevation myocardial infarction. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on the Management of Patients With Unstable Angina). J Am Coll Cardiol 2000; 36: 970–1062. [34] Boersma E, Pieper KS, Steyerberg EW, et al. Predictors of outcome in patients with acute coronary syndromes without persistent ST-segment elevation. Results from an international trial of 9461 patients. The PURSUIT Investigators. Circulation 2000; 101:2557–67. [35] McGuire DK, Emanuelsson H, Charnwood A, et al. Diabetes mellitus is associated with worse clinical outcomes across the spectrum of acute coronary syndromes: results from GUSTO-IIB. Circulation 1999; 100(Suppl.):I432. [36] van Miltenburg-van Zijl AJ, Simoons ML, Veeroek RJ, et al. Incidence and follow-up of Braunwald subgroups in unstable angina pectoris. J Am Coll Cardiol 1995; 25:1286–92. [37] Alexander JH, Harrington RA, Tuttle RH, et al. Prior aspirin use predicts worse outcomes in patients with non-ST-segment elevation acute coronary syndromes. Am J Cardiol 1999; 83: 1147–51. [38] Helgason CM, Bolin KM, Hof JA, et al. Development of aspirin resistance in persons with previous ischemic stroke. Stroke 1994; 25: 2331–6. [39] Weber AA, Zimmermann KC, Meyer-Kirchrath J, et al. Cyclooxygenase-2 in human platelets as a possible factor in aspirin resistance. Lancet 1999; 353:900. [40] Cannon CP, McCabe CH, Stone PH, et al. The electrocardiogram predicts one-year outcome of patients with unstable angina and non-Q-wave myocardial infarction: results of the TIMI III Registry ECG Ancillary Study. J Am Coll Cardiol 1997; 30:133–40. [41] Savonitto S, Ardissino D, Granger CB, et al. Prognostic value of the admission electrocardiogram in acute coronary syndromes. JAMA 1999; 281:707–13. [42] Hyde TA, Grench JK, Wong CK, et al. Four-year survival of patients with acute coronary syndromes without ST-segment elevation and prognostic significance of 0.5-mm ST-segment depression. Am J Cardiol 1999; 84:379–85. [43] Anderson HV, Cannon CP, Stone PH, et al. One-year results of the Thrombolysis In Myocardial Infarction (TIMI) IIIB clinical trial. A randomized comparison of tissue-type plasminogen activator versus placebo and early invasive versus early conservative strategy in unstable angina and non-Q-wave myocardial infarction. J Am Coll Cardiol 1995; 26:1643–50. [44] Antman EM, Tanasijevic MJ, Thompson B, et al. Cardiac-specific troponin I levels to predict the risk of mortality in patients with acute coronary syndromes. N Engl J Med 1996–335:1342– 9. [45] Ohman EM, Armstrong PW, Christenson RH, et al. Cardiac troponin T levels for risk stratification in acute myocardial ischemia. GUSTO IIA Investigators. N Engl J Med 1996; 335:1333–41. [46] Sabatine MS, Morrow DA, de Lemos JA, et al. Multimarker approach to risk stratification in non-ST-elevation acute coronary syndromes: simultaneous assessment of troponin I, C-reactive protein, and B-type natriuretic peptide. Circulation 2002; 105:1760–3. [47] Lindmark E, Diderholm E, Wallentin L, et al. Relationship between interleukin 6 and mortality in patients with unstable coronary artery disease:effects of an early invasive or noninvasive strategy. JAMA 2001; 286:2107–13.

Coronary artery disease coronary revascularization

181

[48] Lindahl B, Toss H, Siegbahn H, et al. Markers of myocardial damage and inflammation in relation to long-term mortality in unstable coronary artery disease. N Engl J Med 2000; 343: 1139–47. [49] Braunwald E. Unstable angina. A classification. Circulation 1989;80:410–14. [50] Calvin JE, Klein LW, VandenBerg BJ, et al. Risk stratification in unstable angina. Prospective validation of the Braunwald classification. JAMA 1995; 273:136–41. [51] Antman EM, Cohen M, Bernink PJ, et al. The TIMI risk score for unstable angina/nonSTelevation myocardial infarction: a method for prognostication and therapeutic decisionmaking. JAMA 2000; 284:835–42. [52] Lanzer P, Markert T, Frey A, et al. Coronary atherosclerosis: acute coronary syndromes. In: Lanzer P, Topol E (eds), Pan Vascular Medicine. 1st edn, Heidelberg: Springer-Verlag; 2002:746–79. [53] Sabatine MS, Antman EM. The thrombolysis in myocardial infarction risk score in unstable angina/ non-ST-segment elevation myocardial infarction. J Am Coll Cardiol 2003; 41:89S–95S. [54] Armstrong PW, Fu Y, Chang WC, et al. Acute coronary syndromes in the GUSTO-IIB trial: prognostic insights and impact of recurrent ischemia. The GUSTO-IIB Investigators. Circulation 1998;98: 1860–8. [55] Prasad A, Mathew V, Holmes D, et al. Current management of non-ST-segment-elevation acute coronary syndrome: reconciling the results on randomized controlled trial. Eur Heart J 2003; 24: 1544–53. [56] Cairns JA, Gent M, Singer J, et al. Aspirin, sulfinpyrazone, or both in unstable angina. Results of a Canadian multicenter trial. N Engl J Med 1985; 313:1369–75. [57] Theroux P, Ouimet H, McCans J, et al. Aspirin, heparin, or both to treat acute unstable angina. N Engl J Med 1988; 319:1105–11. [58] Theroux P, Waters D, Qui S, et al. Aspirin versus heparin to prevent myocardial infarction during the acute phase of unstable angina. Circulation 1993; 88:2045–8. [59] RISC Group. Risk of myocardial infarction and death during treatment with low dose aspirin and intravenous heparin in men with unstable coronary artery disease. Lancet 1990; 336:827– 30. [60] Cohen M, Adams PC, Parry G, et al. Combination antithrombotic therapy in unstable rest angina and non-Q-wave infarction in non-prior aspirin users. Primary end points analysis from the ATACS trial. Antithrombotic Therapy in Acute Coronary Syndromes Research Group. Circulation 1994; 89:81–8. [61] Cohen M, Demers C, Gurfinkel EP, et al. A comparison of low-molecular weight heparin with unfractionated heparin for unstable coronary artery disease. Efficacy and Safety of Subcutaneous Enoxaparin in Non-Q-Wave Coronary Events Study Group. N Engl J Med 1997; 337:447–52. [62] Platelet Receptor Inhibition in Ischemic Syndrome Management (PRISM) Study Investigators. A comparison of aspirin plus tirofiban with aspirin plus heparin for unstable angina. N Engl J Med 1998; 338:1498–505. [63] The Platelet Receptor inhibition in Ischemic Syndrome Management in Patients Limited by Unstable Signs and Symptoms (PRISM-PLUS) Study Investigators. Inhibition of the platelet glycoprotein IIb/IIIa receptor with tirofiban in unstable angina and non-Q-wave myocardial infarction. N Engl J Med 1998; 338:1488–97. [64] Antman EM, McCabe CH, Gurfinkel EP, et al. Enoxaparin prevents death and cardiac ischemic events in unstable angina/non-Q-wave myocardial infarction: results of the Thrombolysis In Myocardial Infarction (TIMI-11B) trial. Circulation 1999; 100:1593–601. [65] Fragmin during instability in Coronary Artery Disease (FRISC) Study Group. Low-molecular weight heparin during instability in coronary artery disease. Lancet 1996; 347:531–8. [66] Comparison of two treatment durations (6 days and 14 days) of low molecular weight heparin with a 6-day treatment of unfractionated heparin in the initial management of unstable angina or

Thrombosis in clinical practice

182

non-Q-wave myocardial infarction. FRAX.I.S. (FRAXiparine in Ischaemic Syndrome). Eur Heart J 1999; 20:1553–62. [67] Yusuf S, Zhao F, Mehta SR, et al. The clopidogrel in unstable angina to prevent recurrent events trial investigators: effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation. N Engl J Med 2001; 345:494–502. [68] The PURSUIT Trial Investigators. Inhibition of platelet glycoprotein IIb/IIIa with eptifibatide in patients with acute coronary syndromes. N Engl J Med 1998; 339:436–43. [69] The PARAGON Investigators. International, randomized, controlled trial of lamifiban (a platelet glycoprotein IIb/IIIa inhibitor), heparin, or both in unstable angina. Platelet IIb/IIIa Antagonism for the Reduction of Acute Coronary syndrome events in a Global Organization Network. Circulation 1998; 97:2386–95. [70] The Platelet IIb/IIIa Antagonism for the Reduction of Acute Coronary syndrome events in a Global Organization Network (PARAGON)-B Investigators. Randomized, placebo-controlled trial of titrated intravenous lamifiban for acute coronary syndromes. Circulation 2002; 105: 316–21. [71] Simmons ML. Effect of glycoprotein IIb/IIIa receptor blocker abciximab on outcome in patients with acute coronary syndromes without early coronary revascularization: the GUSTO IV-ACS randomized trial. Lancet 2001; 357: 1915–24. [72] Boersma E, Harrington RA, Moliterno DJ, et al. Platelet glycoprotein IIb/IIIa inhibitors in acute coronary syndromes: a meta-analysis of all major randomized clinical trials. Lancet 2002; 359: 189–98. [73] Roffi M, Chew DP, Mukherjee D, et al. Platelet glycoprotein IIb/IIIa inhibitors reduce mortality in diabetic patients with non-ST-segment elevation acute coronary syndromes. Circulation 2001 ;104: 2767–71. [74] Roffi M, Chew DP, Mukherjee D, et al. Platelet glycoprotein IIb/IIIa inhibition in acute coronary syndromes. Gradient of benefit related to the revascularization strategy. Eur Heart J 2002; 23: 1441–8. [75] Sabatine MS, Januzzi JL, Snapinn S, et al. A risk score system for predicting adverse outcomes and magnitude of benefit with glycoprotein IIb/IIIa inhibitor therapy in patients with unstable angina pectoris. Am J Cardiol 2001;88:488–92. [76] Morrow DA, Cannon CP, Rifai N, et al. Ability of minor elevations of troponins I and T to predict benefit from an early invasive strategy in patients with unstable angina and non-STsegment elevation myocardial infarction Results from a randomized trial. JAMA 2001; 286:2405–12. [77] Topol E, Moliterno DJ, Herrmann HC, et al. Comparison of two platelet glycoprotein IIb/IIIa inhibitors, tirofiban and abciximab, for the prevention of ischemic events with percutaneous coronary revascularization. N Engl J Med 2001; 344:1888–94. [78] Fox KAA, Poole-Wilson PA, Henderson RA, et al. Interventional versus conservative treatment for patients with unstable angina or non-ST-elevation myocardial infarction: the British Heart Foundation RITA-3 randomized trial. Lancet 2002; 360: 743–51. [79] Spacek R, Widimsky P, Straka Z, et al. Value offirst day angiography/angioplasty in evolving non-ST segment elevation myocardial infarction: an open multicenter randomized trial. The VINO Study. Eur Heart J 2002; 23:230–8. [80] Neumann FJ, Kastrati A, Pogatsa-Murray G, et al. Evaluation of prolonged antithrombotic pretreatment (“cooling-off” strategy) before intervention in patients with unstable coronary syndromes. JAMA 2003; 290:1593–9. [81] Cutlip DE, Baim DS, Ho KK, et al. Stent thrombosis in the modern era: a pooled analysis of multicenter coronary stent clinical trials. Circulation 2001; 103:1967–71. [82] Moussa I, Di Mario C, Reimers B, et al. Subacute stent thrombosis in the era of intravascular ultrasound-guided coronary stenting without anticoagulation: frequency, predictors and clinical outcomesJ Am Coll Cardiol 1997; 29:6–12.

Coronary artery disease coronary revascularization

183

[83] Karrillon GJ, Morice MC, Benveniste E, et al. Intracoronary stent implantation without ultrasound guidance and with replacement of conventional anticoagulation by antiplatelet therapy. 30-day clinical outcome of the French Multicenter Registry. Circulation 1996; 94:1519–27. [84] Sigwart U, Puel J, Mirkovitch V, et al. Intravascular stents to prevent occlusion and restenosis after transluminal angioplasty. N Engl J Med 1987; 316: 701–6. [85] Serruys PW, Strauss BH, Beatt KJ, et al. Angiographic follow-up after placement of a selfexpanding coronary artery stent. N Engl J Med 1991; 324:13–17. [86] Cheneau E, Leborgne L, Mintz G, et al. Predictors of subacute stent thrombosis. Results of a systematic intravascular ultrasound study. Circulation 2003; 108:43–7. [87] Nakamura S, Colombo A, Gaglione A, et al. Intracoronary ultrasound observations during stent implantation. Circulation 1994; 89:2026–34. [88] Uren NG, Schwarzacher Sp, Metz JA, et al. Predictors and outcomes of stent thrombosis: an intravascular ultrasound registry. Eur Heart J 2002; 23:124–32. [89] Colombo A, Hall P, Nakamura S, et al. Intracoronary stenting without anticoagulation accomplished with intravascular ultrasound guidance. Circulation 1995; 91:1676–88. [90] Leon MB, Baim DS, Popma JJ, et al. A clinical trial comparing three antithrombotic-drug regimens after coronary artery stenting: Stent Anticoagulation Restenosis Study Investigators. N Engl J Med 1998; 339:1665–71. [91] Urban P, Macaya C, Rupprecht HJ, et al. Randomized evaluation of anticoagulation versus antiplatelet therapy after coronary stent implantation in high-risk patients: the Multicenter Aspirin and Ticlopidine Trial after Intracoronary Stenting (MATTIS). Circulation 1998;98: 2126–32. [92] Bertrand M, Legrand V, Boland J, et al. Full anticoagulation versus ticlopidine plus aspirin after stent implantation: a randomized multi-center European study; the FANTASTIC trail [Abstract]. Circulation 1996; 94:685. [93] Albiero R, Hall P, Itoh A, et al. Results of a consecutive series of patients receiving only antiplatelet therapy after optimized stent implantation. Comparison of aspirin alone versus combined ticlopidine and aspirin therapy. Circulation 1997; 95:1145–56. [94] Schomig A, Neumann FJ, Kastrati A, et al. A randomized comparison of antiplatelet and anticoagulant therapy after the placement of coronary artery stents. N Engl J Med 1996; 334:1084–9. [95] Bertrand ME, Rupprecht HJM, Urban P, et al. Double-blind study of the safety of clopidogrel with and without a loading dose in combination with aspirin compared with ticlopidine in combination with aspirin after coronary stenting:the clopidogrel aspirin stent international cooperative study (CLASSICS). Circulation 2000;102:624–9. [96] Schuhlen H, Kastrati A, Pache J, et al. Incidence of thrombotic occlusion and major adverse cardiac events between two and four weeks after coronary stent placement: analysis of 5678 patients with a four-week ticlipodine regimen. J Am Coll Cardiol 2001; 37:2066–73. [97] Dangas G, Mehran R, Abizaid AS, et al. Combination therapy with aspririn plus clopidogrel versus aspirin plus ticlopidine for prevention of subacute thrombosis after successful native coronary stenting. Am J Cardiol 2001; 87:470–2. [98] Wilson SH, Rihal SC, Bell MR, et al. Timing of coronary stent thrombosis in patients treated with ticlopidine and aspirin. Am J Cardiol 1999; 83:1006–11. [99] DeFeyter PJ, vandenBrand M, Jaarman G, et al. Acute coronary occlusion during and after percutaneous transluminal coronary angioplasty. Circulation 1991; 83:927–36. [100] Lincoff AM, Popma JJ, Ellis SG, et al. Abrupt vessel closure complicating coronary angioplasty: clinical, angiographic and therapeutic profile.J Am Coll Cardiol 1992; 19:926–35. [101] Detre KM, Holmes DRG, Holubkov R, et al. Incidence and consequences of periprocedural occlusion: the 1985–1986 national heart, lung and blood institute percutaneous coronary angioplasty registry. Circulation 1990; 82:739–50.

Thrombosis in clinical practice

184

[102] Schwartz L, Bourassa MG, Lesperance J, et al. Aspirin and dipyridamole in the prevention of restenosis after percutaneous transluminal coronary angioplasty. N Engl J Med 1988; 318:1714– 19. [103] Lembo NJ, Black AJ, Roubin GS, et al. Effect of pretreatment with aspirin versus aspirin plus dipyridamole on frequency and type of acute complications of percutaneous transluminal coronary angioplasty. Am J Cardiol 1990; 65:422–6. [104] White CW, Chaitman B, Lassar TA, et al. Antiplatelet agents are effective in reducing the immediate complications of PTCA: results from the ticlopidine multicenter trial [Abstract]. Circulation 1987; 76(Suppl. IV):IV–400. [105] Bertrand ME, Allain H, Lablanche JM. Results of a randomized trial of ticlopidine versus placebo for prevention of acute closure and restenosis after coronary angioplasty [Abstract]. Circulation 1990; 82(Suppl. III):III–90. [106] Chesebro JH, Webster MW, Lassar TA. Coronary angioplasty:antiplatelet therapy reduces acute complications but not restenosis [Abstract]. Circulation 1989; 80(Suppl. II):II–64. [107] Smith SC Jr, Dove JT, Jacobs AK, et al. ACC/AHA guidelines or percutaneous coronary interventions (revision of the 1993 PTCA guidelines)—executive summary. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (committee to revise the 1993 guidelines for percutaneous transluminal coronary angioplasty).J Am Coll Cardiol 2001; 37:2215–39. [108] Kuntz RE, Piana R, Pomerantz RM, et al. Changing incidence and management of abrupt closure following coronary intervention in the new device era. Cathet Cardiovasc Diagn 1992; 27:183–90. [109] Scott NA, Weintraub WS, Carlin SF, et al. Recent changes in the management and outcome of acute closure after percutaneous transluminal coronary angioplasty. Am J Cardiol 1993; 71:1158–63. [110] Steinhubl SR, Lauer MS, Mukerjee DP, et al. Pretreatment with ticlopidine reduces non-Qwave myocardial infarctions following intracoronary stenting [Abstract]. J Am Coll Cardiol 1998; 31:100. [111] Steinhubl SR, Balog C, Topol EJ. Pretreatment with ticlopidine prior to stenting is associated with a substantial decrease in complications: data from the EPISTENT trial [Abstract]. Circulation 1998; 98:573. [112] Steinhubl SR, Berger PG, Mann JT, et al. Early and sustained dual oral antiplatelet therapy following percutaneous coronary intervention: a randomized controlled trial. JAMA 2002;288:2411–20. [113] Taniuchi M, Kurz HI, Lasala JM. Randomized comparison of ticlopidine and clopidogrel after intracoronary stent implantation in a broad patient population. Circulation 2001; 104:539– 43. [114] Muller C, Buttner HJ, Petersen J, et al. A randomized comparison of clopidogrel and aspirin versus ticlopidine and aspirin after the placement of coronary artery stents. Circulation 2000; 101:590–3. [115] Herbert JM, Rechel D, Vallee E, et al. Clopidogrel, a novel antithombotic agent. Cardiovas Drug Rev 1993; 11:180–98. [116] Gregorini L, Marko J, Fajadet J, et al. Ticlopidine and aspirin pretreatment reduces coagulation of platelet activation during coronary dilatation procedures. J Am Coll Cardiol 1997; 29:13–20. [117] Quinn MJ, Fitzgerald DJ. Ticlopidine and clopidogrel. Circulation 1999; 100:1667–72. [118] Mehta SR, Yusuf S. Short- and long-term oral antiplatelet therapy in acute coronary syndromes and percutaneous coronary intervention.J Am Coll Cardiol 2003; 41:79S–88S. [119] Clarke TA, Waskell LA. Clopidogrel is metabolized by human cytochrome p450 3A and inhibited by atorvastatin. Drug Metab Dispos 2003; 31:53–9. [120] Lau WC, Waskell LA, Watkins PB, et al. Atorvastatin reduces the ability of clopidogrel to inhibit platelet aggregation: a new drug-drug interaction. Circulation 2003; 107:32–7.

Coronary artery disease coronary revascularization

185

[121] Mueller C, Roskamm H, Neumann FJ, et al. A randomized comparison of clopidogrel and aspirin versus ticlopidine and aspirin after the placement of coronary artery stents. J Am Coll Cardiol 2003; 41: 969–73. [122] Neubauer H, Gunesdogan B, Hanefeld C, et al. Lipophilic statins interfere with the inhibitory effect of clopidogrel on platelet function—a flow cytometry study. Eur Heart J 2003; 24: 1744– 9. [123] Randomized placebo-controlled and balloonangioplasty-controlled trial to assess safety of coronary stenting with use of platelet glycoprotein IIb/IIIa blockade. The EPISTENT Investigators. Lancet 1998; 352:87–92. [124] Brener S, Barr L, Burchenal J, et al. Randomized, placebo-controlled trial of platelet glycoprotein IIb/IIIa blockade with primary angioplasty for acute myocardial infarction. Circulation 1998;98: 734–41. [125] Montalescot G, Barragan P, Wittenberg O, et al. Platelet glycoprotein IIb/IIIa inhibition with coronary stenting for acute myocardial infarction. N Engl J Med 2001; 344:1895–903. [126] Use of monoclonal antibody directed against the platelet glycoprotein IIb/IIIa receptor in high-risk coronary angioplasty. The EPIC Investigation. N Engl J Med 1994; 330:956–61. [127] Platelet glycoprotein IIb/IIIa blockade and lowdose heparin during percutaneous coronary revascularization. The EPILOG Investigators. N Engl J Med 1997; 336:1689–96. [128] Effects of platelet glycoprotein IIb/IIIa blockade with tirofiban on adverse cardiac events in patients with unstable angina or acute myocardial infarction undergoing coronary angioplasty. The RESTORE Investigators. Randomized Efficacy Study of Tirofiban for Outcomes and Restenosis. Circulation 1997; 96:1445–53. [129] Novel dosing regimen of eptifibatide in planned coronary stent implantation (ESPRIT): a randomized, placebo-controlled trial. The ESPRIT Investigators. Lancet 2000; 356:2037–44. [130] Rund MM, Smith DD, DeLuca SA, et al. for the IMPACT II study coordinators/investigators. Heparin during coronary angioplasty: are there any rules? [Abstract]. Cirulation 1994; 90:487. [131] Mc Garry TF, Gottlieb RS, Morganroth J, et al. The relationship of anticoagulation level and complications after successful percutaneous transluminal coronary angioplasty. Am Heart J 1992; 123:1445–51. [132] Frierson JH, Dimas AP, Simpendoerfer CC, et al. Is aggressive heparin necessary for elective PCTA? Cathet Cardiovasc Diagn 1993; 28:279–82. [133] Ferguson JJ, Dougherty KG, Gaos CM, et al. Relationship between procedural activated coagulation time and outcome after percutaneous transluminal coronary angioplasty. J Am Coll Cardiol 1994; 23:1061–5. [134] Boccara A, Beenamer H, Juliard JM, et al. A randomized trial of a fixed high dose vs. a weight adjusted low dose of intravenous heparin during coronary angioplasty. Eur Heart J 1997; 18:631–5. [135] Koch KT, Piek JJ, deWinter RJ, et al. Safety of low dose heparin in elective coronary angioplasty. Heart 1997; 77:517–22. [136] Vianer J, Fleisch M, Gunnes P, et al. Low dose heparin for routine coronary angioplasty and stenting. Am J Cardiol 1996; 78:964–6. [137] Blankenship JC, Helkamp AS, Demko SI, et al. Vascular access site complications after percutaneous coronary intervention with glycoprotein IIb/IIIa inhibitor therapy in the EPIC trial [Abstract].J Am Coll Cardiol 1997; 29:278. [138] Mandak JS, Blankenship JC, Gardner LH, et al. For the IMPACT II Investigators. Modifiable risk factors for vascular access site complications in the IMPACT II trial of angioplasty with versus without eptifibatide. J Am Coll Cardiol 1998; 31: 1518–24. [139] Kereiakes DK, Broderick TM, Whang DD, et al. Partial reversal of heparin anticoagulation by intravenous protamine in abciximab-treated patients undergoing percutaneous intervention. Am J Cardiol 1997; 80:633–4. [140] Karsch KR, Preisack MB, Baildon R, et al. Low molecular weight heparin (reviparin) in percutaneous transluminal coronary angioplasty: results of a randomized, double-blind,

Thrombosis in clinical practice

186

unfractionated heparin and placebo-controlled, multicenter trial (REDUCE trial): Reduction of Restenosis After PTCA, Early Administration of Reviparin in a Double-Blind Unfractionated Heparin and Placebo-Controlled Evaluation. J Am Coll Cardiol 1996; 28:1437–43. [141] Choussat R, Montalescot G, Collet JP, et al. A unique low dose of intravenous enoxaparin in elective percutaneous coronary intervention. J Am Coll Cardiol 2002; 40:1943–50. [142] Kereiakes DJ, Grines C, Fry E, et al. Enoxaparin and abciximab adjunctive pharmacotherapy during percutaneous coronary intervention. J Invasive Cardiol 2001; 13:272–8. [143] Kereiakes DJ, Kleiman NS, Fry E, et al. Dalteparin in combination with abciximab during percutaneous coronary intervention. Am HeartJ 2001; 141:348–52. [144] Bhatt DL, Lee BI, Casterella PJ, et al. Safety of concomitant therapy with eptifibatide and enoxaparin in patients undergoing percutaneous coronary intervention—results of the CRUISE study. J Am Coll Cardiol 2003; 41:20–5. [145] Lindahl B, Diderholm E, Lagerqvist B, et al. Mechanisms behind the prognostic value of troponin T in unstable coronary artery disease: a FRISC II substudy. J Am Coll Cardiol 2001; 38: 979–86. [146] Collet JP, Montalescot G, Lison L, et al. Percutaneous coronary intervention after subcutaneous enoxaparin pretreatment in patients with unstable angina pectoris. Circulation 2001; 103: 658–63. [147] Ferguson JJ. The use of enoxaparin and IIb/IIIa antagonists in acute coronary syndromes including PCI: final results of the NICE 3 study. J Am Coll Cardiol 2001; 37(Suppl. A):365A. [148] Martin JL, Fry ET, Serano A. Pharmacokinetic study of enoxaparin in patients undergoing coronary intervention after treatment with subcutaneous enoxaparin in acute coronary syndromes: the PEPCI Study [Abstract]. Eur Heart J 2001; 22(Suppl.):14. [149] Fox K, Antman E, Cogen M, et al. Comparison of enoxaparin versus unfractionated heparin in patients with unstable angina pectoris/non-ST segment elevation acute myocardial infarction having subsequent percutaneous coronary intervention. AmJ Cardiol 2002; 90:477–82. [S1] Ryan TJ, Antman EM, Brooks NH, et al. 1999 update: ACC/AHA guidelines for the management of patients with acute myocardial infarction: executive summary and recommendations: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Management of Acute Myocardial Infarction). Circulation 1999; 100:1016–30. [S2] Eagle KA, Guyton RA, Davidoff R, et al. ACC/AHA guidelines for coronary artey bypass graft surgery: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1991 Guidelines for Coronary Artery Bypass Graft Surgery).J Am Coll Cardiol 1999; 34:1262–347. [S3] Hirose H, Amano A, Yoshida S, et al. Surgical management of unstable patients in the evolving phase of acute myocardial infarction. Ann Thorac Surg 2000; 69:425–8. [S4] Creswell LL, Moulton MJ, Cox JL, et al. Revascularization after acute myocardial infarction. Ann Thorac Surg 1995; 60:19–26. [S5] Lee DC, Oz MC, Veinberg AD, et al. Appropriate timing of surgical intervention after transmural acute myocardial infarction. J Thorac Cardiovasc Surg 2003; 125:1 15–20. [S6] Lee DC, Oz MC, Veinberg AD, et al. Optimal timing of revascularization: transmural versus nontransmural acute myocardial infarction. Ann Thorac Surg 2001; 70:1198–204. [S7] Albes JM, Gross M, Franke U, et al. Revascularization during acute myocardial infarction:risks and benefits revisited. Ann Thorac Surg 2002; 74:102–8. [S8] Hajek T, Straka Z, Jirasek K, et al. Emergency myocardial revascularization during a developing myocardial infarction. Rozhl Chir 1999; 78:218–22. [S9] Entwistle III JWC, Bolno PB, Holmes E, et al. Improved Survival with ventricular assist device support in cardiogenic shock after myocardial infarction. Heart Surg Forum 2003; 6:316–19.

Coronary artery disease coronary revascularization

187

[S10] Lee DC, Oz MC, Veinberg AD, et al. Optimal timing of revascularization: transmural versus nontransmural acute myocardial infarction. Ann Thorac Surg 2001; 71:1198–202 (Discussion 1202–4). [S11] Barakate MS, Bannon PG, Hughes CF, et al. Emergency surgery after unsuccessful coronary angioplasty: a review of 15 years’ experience. Ann Thorac Surg 2003; 75:1400–5. [S12] Ngaage DL. Off-pump coronary artery bypass grafting: the myth, the logic and the science. Eur J Cardiothorac Surg 2003; 24:557–70. [S13] D’Ancona G, Karamanoukian H, Ricci M, et al. Myocardial revascularization on the beating heart after recent onset of acute myocardial infarction. Heart Surg Forum 2001; 4:74–9. [S14] Locker Ch, Mohr M, Paz Y, et al. Myocardial revascularization for acute myocardial infarction: benefits and drawbacks of avoiding cardiopulmonary bypass. Ann Thorac Surg 2003; 76:771–7. [S15] Straka Z, Votava J, Jirásek K, et al. Are patients with acute myocardial infarction and unstable angina suitable candidates for myocardial revascularization without extracorporeal circulation? Cor Vasa 2000;42:420–2. [S16] Straka Z, Widimsky P, Jirasek K, et al. Off-pump versus on-pump coronary surgery: final results from a prospective randomized study PRAGUE-4. Ann Thorac Surg 2004; 77:789–93. [S17] Straka Z, , Vanĕk T, et al. Routine immediate extubation for off-pump coronary bypass surgery without thoracic epidural analgesia. Ann Thorac Surg 2002; 74:1544–7. [S18] , Straka Z, Vannĕk T, et al. Less invasive cardiac anesthesia: an ultra-fast track procedure avoiding thoracic epidural analgesia. Heart Surg Forum 2003; 6:528–31. [S19] Vanĕk T, Straka Z, , et al. Thoracic epidural anesthesia for off-pump coronary artery bypass without extubation. Eur J Cardiothorac Surg 2001; 20:858–60. [S20] Fabricius AM, Gerber W, Hanke M, et al. Early angiographic control of perioperative ischemia after coronary artery bypass grafting. Eur J Cardiothorac Surg 2001; 19:853–8. [S21] Breuer M, Schutz A, Gansera B, et al. Intraoperative local fibrinolysis as emergency therapy after early coronary artery bypass thrombosis. Eur J Cardiothorac Surg 1999; 15:266–70. [S22] FitzGibbon GM, Kafka HP, Leach AJ, et al. Coronary bypass graft fate and patient outome: angiographic follow-up of 5,065 grafts related to survival and reoperation in 1,388 patients during 25 years.J Am Coll Cardiol 1996; 28:616–26. [S23] Dion R, Glineur D, Derock D, et al. Complementary saphenous grafting: long term follow up. J Thorac Cardiovasc Surg 2001; 122:296–304. [S24] Loop FD, Lytle BW, Cosgrove DM, et al. Influence of the internal mammary artery graft on 10 year survival and other cardiac events. N Engl J Med 1986; 314:1–6. [S25] Shuhaiber JH, Evans AN, Massad MG, et al. Mechanisms and future directions for prevention of vein graft failure in coronary bypass surgery. Eur J Cardiothorac Surg 2002; 22:387–96. [S26] Bansal SK, Karamanoukian HL. Antiplatelet therapy after coronary artery bypass. Heart Surg Forum 2003; 6:273–4. [S27] Goldman S, Copeland J, Moritz T, et al. Improvement in early saphenous vein graft patency after coronary artery bypass surgery with antiplatelet therapy: results of a Veterans Administration cooperative study. Circulation 1988; 77:1324–32. [S28] Goldman S, Copeland J, Moritz T, et al. Saphenous vein graft patency 1 year after coronary artery bypass surgery and effects of antiplatelet therapy: results of a Veterans Administration cooperative study. Circulation 1989; 80:1190–7. [S29] Gavaghan TP, Gebski V, Baron DW. Immediate postoperative aspirin improves vein graft patency early and late after coronary artery bypass graft surgery: a placebo-controlled, randomized study. Circulation 1991; 83:1526–33. [S30] Mangano DT. Aspirin and mortality from coronary bypass surgery. N Engl J Med 2002; 347: 1309–17.

Thrombosis in clinical practice

188

[S31] Stein PD, Dalen JE, Goldman S, Theroux P. Antithrombotic therapy in patients with saphenous vein and internal mammary artery bypass grafts. Chest 2001; 119:278S–282S. [S32] Topol EJ. Aspirin with bypass surgery. From taboo to new standard of care. N Engl J Med 2002; 347:1359–60. [S33] Despotis GJ, Filos KS, Zoys TN, et al. Factors associated with excessive postoperative blood loss and hemostatic transfusion requirements: a multivariate analysis in cardiac surgical patients. Anesth Analg 1996; 82:13–21. [S34] Dacey LJ, Munoz JJ, Baribeau YR, et al. Reexploration for hemorrhage following coronary artery bypass grafting: incidence and risk factors. Northern New England Cardiovascular Disease Study Group. Arch Surg 1998; 133:442–7. [S35] Moulton MJ, Creswell LL, Mackey ME, et al. Reexploration for bleeding is a risk factor for adverse outcomes after cardiac operations. JThorac Cardiovasc Surg 1996; 111:1037–46. [S36] Levy JH. Pharmacologic presevation of the hemostatic system during cardiac surgery. Ann Thorac Surg 2001; 72:S1814–S1820. [S37] Despotis GJ, Avidan MS, Hogue ChW. Mechanisms and attenuation of hemostatic activation during extracorporeal circulation. Ann Thorac Surg 2001; 72:S1821–S1831. [S38] Shirota K, Watanabe T, Takagi Y, et al. Maintenance of blood heparin concentration rather than activated clotting time better preserves the coagulation system in hypothermic cardiopulmonary bypass. Artif Organs 2000; 24:49–56. [S39] Despotis GJ, Joist JH, Hogue CW, et al. The impact of heparin concentration and activated clotting time monitoring on blood conservation: a prospective, randomized evaluation in patients undergoing cardiac operations. J Thorac Cardiovasc Surg 1995; 110:46–54. [S40] Levi M, Cromheecke ME, de Jonge E, et al.Pharmacological strategies to decrease excessive blood loss in cardiac surgery: a meta-analysis of clinically relevant endpoints. Lancet 1999; 354:1940–7. [S41] Casati V, Gerli C, Franco A, et al. Tranexamic acid in off-pump coronary surgery: a preliminary, randomized, double-blind, placebo controlled study. Ann Thorac Surg 2001; 72:470–5. [S42] Englberger L, Markart P, Eckstein FS, et al. Aprotinin reduces blood loss in off-pump coronary artery bypass (OPCAB) surgery. Eur J Cardiothorac Surg 2002; 22:545–51. [S43] Jares M, Vaněk T, Straka Z, et al. Tranexamic acid reduces bleeding after off-pump coronary artery bypass grafting. J Cardiovasc Surg 2003; 44: 205–8. [S44] Hendriks HG, van der Maaten JM, de Wolf J, et al. An effective treatment of severe intractable bleeding after valve repair by one single dose of activated recombinant factor VII. Anesth Analg 2001; 93:287–9. Commentary and discussion: Anesth Analg 2002; 94:1369–71. [S45] Mangi AA, Palacios IF, Torchiana DF. Catheter pericardiocentesis for delayed tamponade after cardiac valve operation. Ann Thorac Surg 2002; 73:1479–83. [S46] Kuvin JT, Harati NA, Pandian NG. Postoperative cardiac tamponade in the modern surgical era. Ann Thorac Surg 2002; 74:1148–53.

10 Peripheral arterial disease M Burress Welborn III, Franklin S Yau, and G Patrick Clagett Epidemiology Atherosclerosis of the lower extremities or peripheral vascular disease (PVD) is common and is a disease of the elderly. While pathologic changes that are precursors of atherosclerosis can be identified in children, clinically significant PVD is rare before the seventh decade. The incidence of clinically detectable atherosclerosis is surprisingly high in the elderly population. In the Rotterdam Study, the prevalence of PVD (ankle brachial index (ABI) values were <0.9) was 19.1% [1]. However, a majority of patients in this study were asymptomatic. Only 1.6% of the total study population had symptoms of intermittent claudication and only 6.3% of the patients with clinically detectable PVD had symptoms of claudication [1]. These rates are very similar to the rates of PVD and claudication found in other studies [2–4]. In the Rotterdam study, rates of clinically detectable PVD and intermittent claudication increased with increasing age (see Figure 10.1) [1]. The natural history of claudication is benign with only 14–30% of patients developing limbthreatening ischemia after 10 years of follow-up [5, 6]. In the elderly patient population, only 1% of patients are expected to have PVD which progresses to limbthreatening ischemia. Thus, peripheral arterial atherosclerosis is a common finding but, in most patients, is a benign clinical diagnosis with little risk for development of limbthreatening ischemia. Only a small portion of patients with clinically detectable PVD will require intervention. However, clinical detection of PVD is important as the diagnosis is an important marker for concurrent atherosclerosis in other vascular beds, in particular, the cardiovascular and cerebrovascular systems.

Risk factors The risk factors for PVD have been well defined. Increasing age is an independent risk factor for the development of PVD and older patients will have lower ABI values at the time of diagnosis [1, 4]. Both current and former smokers are at increased risk of development of PVD. Patients who are not current smokers have only a mild increase of risk but patients who are smoking at the time of diagnosis carry a relative risk (RR) of more than double the RR for former smokers [7, 8]. It is estimated that smoking may be directly responsible for up to 50% of cases of PVD and smoking increases the risk of PVD to a greater extent than the diagnosis of cardiovascular disease [9, 10].

Peripheral arterial disease

189

Figure 10.1 The prevalence of PVD as a function of age. Males are represented in open bars and women with hashed bars. As age increases the prevalence of PVD increases dramatically. PVD is a disease of the aged. (Reproduced with permission from Meijer WT, Hoes AW, Rutgers D, Bots ML, Hofman A, Grobbee DE. Peripheral arterial disease in the elderly: the Rotterdam Study. Arterioscler Thromb Vasc Biol 1998; 18:185–92.) The age at which patients started to smoke has also been identified as a risk factor for PVD. Patients who began smoking prior to the age of 16 have a 3-fold higher incidence of PVD versus patients who began smoking at a later age [11]. The diagnosis of hypertension is an independent risk factor for the development of PVD and the risk increases with increasing systolic hypertension [7, 8]. The risk associated with hypertension is similar to that associated with smoking. In the Framingham Study, hypertension increased the risk of development of PVD from 2.5- to 4-fold. This increase in risk was higher than that associated with smoking (2-fold) [11]. In contrast, other studies have failed to find that hypertension poses a higher risk than smoking. It is safe to assume that the risk associated with hypertension is at least as high as smoking and may be higher. Like hypertension, diabetes poses a similar risk for the development of PVD as that associated with smoking. In the Rotterdam study, diabetes increased the RR of PVD by 1.89, which was similar to smoking [7]. Diabetes is not only a risk factor for the development of PVD but also an independent predictor of progression of the disease [6].

Thrombosis in clinical practice

190

Patients with diabetes and PVD have a more malignant course of their disease and a higher percentage can be anticipated to progress to limb-threatening ischemia [6]. Interestingly, the risk associated with diet-controlled diabetes is less than the risk associated with diabetes requiring either oral hypoglycemics or insulin therapy [12]. Surprisingly, hypercholesterolemia, while a risk factor for PVD, increases the risk of PVD to lesser extent than the factors listed here. In some studies, there has been no correlation between cholesterol levels and PVD [11], In the Rotterdam study, hypercholesterolemia had a RR index of 1.19, lower than all other reversible risk factors [7]. A similar weak relationship to PVD was demonstrated in the Framingham study [8]. More important than total cholesterol are high-density lipoproteins (HDL) and lowdensity lipoproteins (LDL) levels; higher HDL levels lower risk while elevated LDL levels increase the risk of PVD [7, 13]. These risk factors do not explain the entire story. In the Rotterdam Study, the authors calculated that 70% of cases of PVD were directly related to the risk factors cited here and that all of these risk factors are potentially treatable [7]. This leaves 30% of the etiology of PVD unexplained. There is, clearly, a genetic risk that contributes to the occurrence of PVD. Young men below the age of 55 have a particularly aggressive form of the disease and virtually all of these individuals are heavy smokers. Asymptomatic first-degree relatives of these patients have a higher incidence of occult PVD compared to both the smoking and nonsmoking general population [14]. This observation indirectly supports the hypothesis that genetic predisposition plays a role in the development of PVD in this patient population and this observation is, likely, applicable to the population as a whole. The genetic forces that contribute to the development of PVD remain poorly described. However, there is overwhelming evidence that atherosclerosis is an inflammatory disease and elevation of serum markers of inflammation is associated with the development of PVD. Several serum markers of inflammation have been found to be associated with an increased risk for the development of PVD. Elevated serum levels of fibrinogen and C reactive protein are associated with the development of PVD [7, 9, 13]. Both fibrinogen and C reactive protein are acute phase reactants and are secreted during states of inflammation. The primary signal for production of these proteins is (interleukin) IL-6. IL-6 levels have been correlated with an increased risk of coronary artery disease but evidence that IL-6 levels predict the risk for PVD is unclear but suggestive [15]. Soluble receptors for proinflammatory cytokines such as tumor necrosis factor-alpha (TNFα) and IL-1β have been found to be elevated in the patients with PVD indicating overproduction of these proinflammatory cytokines, but causality is lacking [16]. Other serum factors that increase the risk factors for PVD are homocysteine, lipoprotein A, and others. However, the correlation between these factors and the development of PVD are not as strong as the inflammatory proteins [13]. There appear to be two sets of forces at work in the pathogenesis of atherosclerosis— the reversible/environmental forces and the genetic forces. The advent of advanced techniques in cellular biology has allowed a more complete understanding of the basic pathophysiology of this complex disease. Cholesterol is not the only factor that precipitates this disease but rather the disease results from a complex interplay between environmental forces, serum lipids, and inflammation. While not proven, it is suspected

Peripheral arterial disease

191

that genetics play a prominent role in the inflammatory response to cellular injury that drives these processes.

Pathophysiology The etiology of atherosclerosis was initially thought to be due to injury to the artery with loss of the endothelial barrier and accumulation of cholesterol. It is now clear that atherogenesis is a more complex process that occurs as a result of endothelial cell dysfunction. The inciting event that promotes atherogenesis is not entirely clear but is likely due to a combination of risk factors that are permissive. Hypercholesterolemia, hypertension, genetics, diabetes, and smoking act to alter endothelial cell function to promote inflammation and the formation of atheromas [17]. Hypercholesterolemia with increased LDL is pivotal to atherogenesis. Oxidized LDL (oxLDL) plays a unique role in initiation of the process. Ox-LDL has inflammatory properties, as will be discussed later, circulates in the blood, and is a pivotal promoter of atherogenesis. However, the inflammatory response to endothelial dysfunction is the critical process that drives atheroma production. Continued inflammation results in plaque instability with plaque rupture and thrombosis resulting in end organ ischemia. This process is driven at the molecular level by a system of inflammatory molecules. The initial event in atherogenesis is the upregulation of endothelial cell adhesion molecules that promote the ingress of monocytes and T lymphocytes. Alterations in flow patterns or areas of low shear stress alter the expression of the genes that promote the production of adhesion molecules [17]. The expression of E and P selectins is the first step by which inflammatory cells migrate to the intima of the artery. The selectins promote leukocyte rolling allowing the cells to maintain contact with the endothelial cell to promote binding to vascular cell adhesion molecule-1 (VCAM-1) and intercellular adhesion molecule-1 (ICAM-1). Monocytes and lymphocytes express very late antigen-1 or α4β1 (VLA-4), which is the ligand for VCAM-1; binding to VCAM-1 allows these inflammatory cells to migrate through the endothelium into the intima [18]. In animals, hypercholesterolemia or deficiency of apolipoprotein E results in expression of VCAM-1 solely at the sites of atheroma production, indicating that flow disturbances mediate the process [19, 20]. Upregulation of leukocyte adhesion molecules, particularly VCAM-1 is thought to occur by the action of nuclear factor-kappa B (NF-κB) [18]. This complex of nuclear regulatory proteins controls expression of multiple inflammatory gene complexes and is a primary pathway of inflammatory regulation. Expression of NF-κB is controlled by shear stress [21]. Expression of adhesion molecules sets the stage for the influx of inflammatory cells into the walls of at-risk arteries (see Figure 10.2). While adhesion molecules allow inflammatory cells to migrate into the wall of arteries, it is chemoattractants that direct the migration of these cells into the intima (see Figure 10.2). Monocyte chemoattractant protein-1 (MCP-1) is one such chemoattractant. MCP-1 is overexpressed in both clinical atheroma and in animal models of atherosclerosis [22]. Transgenic mice that lack functional MCP-1 and are rendered susceptible to atherosclerosis by deletion of the genes coding for apoE or the very low density lipoprotein (VLDL) receptor, show a reduction in monocyte accumulation, decreased local lipid levels, and decreased lesion formation [23, 24], Other

Thrombosis in clinical practice

192

chemoattractants that have been implicated in the atherogenesis are IL-8, modified LDL, and interferon-gamma (IFN-γ) inducible chemokines (inducible protein-10 (IP-10), monokine induced by IFN-γ (Mig), and IFN-inducible T-cell α-chemoattractant (I-TAC)) [17, 22]. These chemokines act to attract monocytes and lymphocytes to the intima where they are activated to produce cytokines such as TNFα, IL-1β, and IFN-γ. Expression of these proinflammatory proteins results in the production of more chemokines resulting in a feedforward signal to promote continued chronic inflammation.

Figure 10.2 Macrophages in atherogenesis. Endothelial dysfunction results in the expression of macrophage adhesion molecules such as selectins (not represented) and VCAM-1. Macrophages adhere to the endothelium under the influence of these adhesion molecules resulting in migration of macrophages into the intima of the arterial wall. Migration is controlled by chemoattractants such as MCP-1, IL-8 (not shown), modified LDL (not shown), and IFN-γ inducible chemokines (not shown). Once the macrophage reaches the intima, it expresses scavenger receptors which take up modified LDL resulting in formation of foam cells and the genesis

Peripheral arterial disease

193

of early atheromatous lesions. Foam cells secrete a variety of proinflammatory cytokines, growth factors, and proteases that promote inflammation and plaque progression. The secretion of matrix metalloproteinases (MMPs) degrades the surrounding extracellular matrix causing intraplaque hemorrhage or plaque ulceration with ischemic sequela. (Reproduced with permissions from Libby P. Inflammation in atherosclerosis. Nature 2002; 420:868– 74.) Nitric oxide (NO) may play an important role in the genesis of endothelial cell dysfunction and subsequent atherosclerosis. In diabetes, altered NO production plays a prominent role in the pathobiology of atheromatous disease. NO is the primary mediator of vascular relaxation and is produced by endothelial cells expressed endothelial nitric oxide synthase (eNOS). NO, in addition to its vasculomotor effects, acts to inhibit platelet aggregation, decreases expression of leukocyte adhesion molecules, inhibits leukocyte migration, and diminishes vascular smooth muscle cell migration and proliferation [18, 25]. The anti-inflammatory effects of NO occur via stimulation of the NF-κB inhibitor IκB. Diabetes inhibits production of NO by eNOS and decreases bioavailability of NO [25]. This results in failure of vessel wall relaxation and is permissive for inflammation [26]. Paradoxically, NO has been found to stimulate formation of modified ox-LDL which is a potent leukocyte chemoattractant and promotes macrophage foam cell production [27]. The complete role of NO in the pathogenesis of atherosclerosis is incompletely understood. Monocytes differentiate into macrophages after migration into the subendothelial space. These cells express cell surface receptors to scavenge modified lipoproteins, oxLDL, taking up modified lipoproteins to become the typical foam cells found in early atheromatous lesions [22, 27]. Activation of recruited macrophage cell population results from locally produced macrophage colony-stimulating factor (M-CSF) and granulocytemacrophage colony-stimulating factor (GM-CSF) (see Figure 10.2) [22]. Mice that lack M-CSF have a marked reduction in the production of atheromatous lesions even in the face of absence of apoE [28]. Once the foam cells are formed, these macrophages produce cytokines resulting in continued inflammatory cell influx and the production of growth factors that promote smooth muscle migration and proliferation. Activated macrophages produce TNFα and IL-1β that stimulate endothelial cells to produce platelet-derived growth factor (PDGF) [27]. PDGF acts on resident smooth muscle cell populations to convert this cell population to the synthetic phenotype. The synthetic

Thrombosis in clinical practice

194

phenotype produces extracellular matrix that leads to the clinically relevant fibrous plaque responsible for the manifestations of atherosclerosis. The synthetic smooth muscle cell phenotype will produce more PDGF to promote phenotype conversion in other cells. Synthetic cells and dysfunctional endothelial cells (via vascular endothelial growth factor (VEGF)) produce basic fibroblast growth factor (bFGF) and transforming growth factor-β resulting in smooth muscle cell recruitment and proliferation [27, 29]. The production of these growth factors results in a further recruitment and activation of synthetic smooth muscle cells resulting in progression of atheromatous lesions. These molecular and cellular interactions result in the formation of the lipid-laden fibrous atheromatous lesions replete with synthetic smooth muscle cells and foamy macrophages. These lesions result in arterial occlusion and/or thromboses. Atheromatous plaques experience intraplaque hemorrhage with deposit of fibrin, thrombin, and platelets. Platelets release more mitogens including PDGF and transforming growth factor-beta (TGF-β), further stimulating smooth muscle cell migration and proliferation [22]. Interplaque hemorrhage results in a burst of lesion growth. Gross plaque rupture occurs with vessel thrombosis resulting in end organ ischemia. The culmination of this process is acute end organ ischemia with myocardial infarction (MI) or stroke. This process occurs in PVD but is not a prominent clinical feature. In the peripheral vascular bed, thrombosis of plaque results in the stepwise progression of the disease with sudden worsening of symptoms between long periods of quiescence. The most clinically significant process in PVD is the slow progression to arterial occlusion. Thus, while the pathobiology of atherosclerosis is similar in the cardiac, cerebrovascular, and peripheral beds, acute thrombosis plays a more prominent role in the coronary and carotid arterial beds. The inciting process that drives endothelial dysfunction and inflammation resulting in atherosclerosis is not well understood. Various theories have been developed to explain this phenomenon. Initially, it was thought that hypercholesterolemia with resultant endothelial cell injury was primarily responsible for atheromatous disease. Clearly, hypercholesterolemia plays a prominent role in the pathogenesis of atherosclerosis; however, the weak correlation between the total cholesterol levels and development of atherosclerosis has stimulated interest in other theories of pathogenesis. The renninangiotensin system, homocysteine, and infectious organisms have all been implicated. The rennin-angiotensin system and its role in atherogenesis will be described in a following section. Hyperhomocystinemia occurs due to defect in metabolism of methionine resulting in elevations of plasma homocysteine. This defect is either genetic in origin or due to vitamin deficiencies (folate, B6, and B12), chronic renal failure, hypothyroidism, some cancers, certain drugs, and, perhaps, smoking [30]. Homozygotes lacking either cystathionine β-synthase or N5, N10methylenetetrahydrofolate reductase (MTHFR) have a syndrome of accelerated atherosclerosis and thromboembolism that begins in childhood [30]. Mild hyperhomocysteinemia is thought to occur in 5–7% of the population and link between genetic defects in homocysteine metabolism has implicated homocysteine as the mediator of PVD [30]. However, multiple meta-analysis have shown only a modest association between elevated homocysteine levels and the development of coronary artery disease and cerebrovascular disease [31–33]. While elevated homocysteine levels appear to have a modest effect on the development of coronary and cerebrovascular

Peripheral arterial disease

195

disease, there is little convincing evidence that homocysteine impacts significantly on the progression of PVD. In young patients with accelerated atherosclerosis, homocysteine levels were no different than control patients [34]. The finding of Chlamydia pneumoniae, cytomegalovirus, and Helicobacter pylori species in atheromatous plaques has sparked an interest in biologic pathogens as mediators of atherosclerosis [35]. Such infections could be the source of inflammatory signal that promotes endothelial cell dysfunction with ingress of inflammatory cells promoting atherogenesis. To date the relationship remains unclear and causality has not been proven. These observations may promise new antiatherosclerotic therapies aimed at pathologic organisms.

Chronic PVD As the epidemiologic studies have demonstrated, PVD is prevalent in the aged population. However, only a small percentage of patients have clinically significant occlusive disease. Of patients with clinically significant PVD, only a minority will require operative intervention for limb salvage or claudication. More importantly, PVD is a marker for both cardiovascular and cerebrovascular disease. Thus, the medical management of patients with PVD must be formulated with risk reduction for coronary and cerebrovascular disease in mind. Both medical and surgical management limb ischemia resulting from PVD is palliative and designed to reduce the symptoms of PVD. There is no medical or surgical therapy that reverses the process. All therapies are designed to treat the symptoms and attempt to halt the clinical progression of the disease. Most patients can be treated medically, with only the most severe cases requiring operative intervention. The advent of endovascular therapy has allowed vascular surgeons to treat PVD with less operative risk. However, most patients who will benefit from a solely endovascular approach to their disease have a low disease burden which can often be management with medical therapy. The true benefit of endovascular approaches to PVD has come from the ability to combine endovascular techniques with standard open surgical techniques to treat multilevel disease. Diagnosis and vascular laboratory studies The diagnosis of PVD is straightforward. Elements from the history, the physical exam, and noninvasive vascular laboratory studies are combined to make the diagnosis. More importantly, these elements are combined to estimate the natural history of the disease for the specific patient. Only patients with a poor natural history should be considered for operative intervention and the decision to operate is made after careful consideration of the risk-benefit ratio for the patient. Virtually all patients with PVD have concurrent coronary artery disease making this patient population high risk for surgical intervention. Mortality following PVD ranges from 1% to 5% depending on the procedure. The rates of wound infections, bleeding complications, and even amputation are significant. Thus, a thorough understanding of the risks of surgery and the natural history of the disease is

Thrombosis in clinical practice

196

paramount. It is the history, the physical, and the vascular laboratory studies that will define the natural history of the disease. The patient’s symptomatology at presentation defines the natural history of the disease. Symptoms of PVD range from mild claudication to frank ischemic tissue loss. Claudication is demand ischemia of the muscle groups of the lower extremity. It is commonly described as a cramping sensation or “charlie horse” in the legs, most commonly the calves, but can include the thigh and buttocks. Less commonly, patients will complain of a heaviness in legs or describe the legs as “going dead” after ambulating; falls are not uncommon. Other pathologic processes particularly lumbar sacral degenerative osteopathy with cord or nerve root compression can masquerade as claudication. Claudication can be differentiated for pseudoclaudication based on the relationship to symptoms. Claudication will occur at a set distance of ambulation and will be reproducible. Symptoms will resolve with rest (<10−15 min) and the patient will fully recover. Ambulation that requires more energy expenditure such as climbing stairs, walking up an incline, or walking on uneven surfaces will decrease the distance required to elicit symptoms. The distribution of the symptoms will correspond with the level of disease. Calf claudication is associated most commonly with superficial femoral artery (SFA) occlusive disease while claudication in the thigh and buttocks indicates more proximal aortoiliac occlusive disease. Further progression of disease will result in rest pain. Rest pain results in the failure to meet metabolic demands at rest in the most distal aspect of the arterial tree, the forefoot. Most commonly, patients complain of pain across the metatarsal heads and this pain often includes the toes. The pain may occur only with elevation, most commonly occurring during sleep when the leg is elevated. The patient will resort to various positioning maneuvers to ensure that the foot is in the dependent position. The most common maneuver is to hang the foot over the edge of the bed during sleep. Often limited ambulation will relieve the pain. Leg dependency and ambulation increase blood flow to the foot due to the forces of gravity and augmentation of cardiac output (ambulation). Commonly, neuropathy will masquerade as rest pain but can be differentiated based on the relationship to positioning. Neuropathy will commonly be described as a tingling or burning sensation that is continuous, not related to positioning, improved with elevation, in a sock-like distribution, and bilateral. Often patients with neuropathy will complain of a foreign body sensation when walking, described as having “rocks in their shoes.” Ultimately rest pain progresses until the pain is no longer relieved by dependency and the pain will be difficult to control even with narcotics. Progression of ischemia results in frank ischemic tissue loss. Patients can present with tissue loss without ischemic rest pain, typically occurring following injury to a foot with insufficient blood supply to heal. More commonly, patients will present with rest pain and concurrent tissue loss. Tissue loss can range from shallow ischemic ulcers to frank gangrene of the toes or forefoot. Ulceration is typically very painful, involves the foot or toes, and is defined as nonhealing if the ulceration fails to show signs of healing after 6 weeks of appropriate wound care. The most extreme form of tissue loss is gangrene, which may be dry, but patients can present with wet gangrene and a septic foot. This represents a surgical emergency and debridement of the foot with control of sepsis should proceed prior to revascularization.

Peripheral arterial disease

197

PVD involves the arterial tree above the ankle in virtually all cases, including diabetics. It is a common misconception that diabetics suffer from small vessel disease of the foot, rendering them inoperable. While diabetics have a high rate of tibial occlusive disease, they do not suffer from microvascular or arteriolar disease of the foot [36]. Virtually all diabetics with ischemic ulceration have suitable anatomy for surgical revascularization [37]. On physical exam patients will have diminished or absent pulses on palpation. The level of disease can be determined by the pulse exam. Patients with aortoiliac disease will have absent or diminished femoral pulses. Disease of the SFA is recognized by the presence of normal femoral pulses with absent popliteal pulses. Isolated tibial disease is distinguished by palpable femoral and popliteal pulses with nonpalpable pedal pulses. Patients with mild claudication may have preserved pulses on examination at rest. In such patients, loss of pulses will not occur until the patient exercises. Examination of chronically ischemic legs will reveal hair loss, hypertrophic toenails, dry scaling skin, muscular atrophy, and dependent rubor. Dependent rubor results from ischemia-induced dermal vasodilation with dermal venous pooling. The leg appears to be red or purple in color and can be confused with cellulitis. Differentiation can be easily made by elevation of the leg, which will result in the loss of discoloration in foot being replaced by pallor. Noninvasive vascular laboratory studies are critical to the diagnosis of PVD. They serve to validate the findings on history and physical exam, and can confirm or invalidate historical or physical findings that are suspect. History and physical exam is falsely positive in 44% of exams and falsely negative in 19% [38]. The primary noninvasive study is ABI performed at rest. The highest blood pressure is taken from the arms using a cuff and continuous wave Doppler. This value is used as the denominator. With a blood pressure cuff inflated at the ankle, the pressure at which Doppler signal disappears in the pedal arteries (posterior tibial and dorsalis pedis arteries) is recorded as the numerator. The higher of the pedal pressures is used to calculate the index by dividing the ankle pressure by the arm pressure. Clinical symptoms and the relationship to ABI values are shown in Figure 10.3. Patients may rarely present with rest pain and ABI values greater than 0.5, when there has been an acute thrombosis with worsening ischemia in the face of poorly developed collateral circulation. Tissue loss typically occurs when the ABI drops to less than 0.4. There can be considerable overlap between these clinical categories.

Figure 10.3 Expected values for resting ABI by clinical symptomatology (Reproduced with permission from Ouriel K. Peripheral

Thrombosis in clinical practice

198

arterial disease. Lancet 2001; 358:1257–64) [39]. The exercise stress test is critical to diagnosing claudication when ABI values are preserved or when the diagnosis is unclear. The patient is asked to walk until the pain occurs. The distance walked is recorded and ABI values are taken immediately after cessation of ambulation every minute for 5 minutes. A positive test results in a greater than 15% fall in ABI values lasting for more than 2 minutes. ABI values may be falsely elevated in patients with medial calcinosis of the tibial vessels that results in failure to compress the tibial arteries with the ankle cuff. This typically occurs in patients with diabetes or end stage renal disease. Diagnosis and quantification of the degree of ischemia is made in these cases by measuring toe pressures in conjunction with pulse volume recording. A normal toe to arm index is greater than 0.6 and tissue loss typically occurs when the toe pressure is less than 60 mmHg. Failure to heal tissue loss can be anticipated when the toe pressure is less than 60 mmHg in diabetics and less than 40 mmHg in patients without diabetes. Arteriography is not used as a diagnostic tool as it only gives anatomic information and provides little information about hemodynamic significance of the disease burden. Arteriography should only be used for preoperative planning. The advent of noninvasive methods for imaging the arterial tree has made imaging virtually risk-free. CT angiography, MR angiography, and ultrasound duplex scanning can accurately image the arterial tree without the risks associated with invasive arteriography. However, these anatomic studies provide little insight as to the natural history of the disease and remain, primarily, for operative planning. These modalities are helpful as diagnostic tools to determine if patients with mild symptoms of claudication have anatomy that is amenable to endovascular therapy, and, thus, they may be used as screening tools to look for favorable anatomy for percutaneous interventions. The natural history of the disease is determined by combining the data obtained from the history, the physical exam, and noninvasive vascular laboratory studies. Of these modalities, the ABI is the best predictor of the natural history of disease followed by the patient’s clinical history. Using these diagnostic modalities, an accurate prediction of the natural history can be formulated and the risk-benefit ratio determined. Once the riskbenefit ratio is determined, the decision to manage either medically or surgically is made. Natural history of PVD PVD can be divided into three clinical categories: intermittent claudication, rest pain, and tissue loss. The prognosis and the risk of limb loss increase with increasing severity of the disease. The most benign form of PVD is claudication. Whereas up to 20% of the elderly population will have clinically diagnosable disease, only a minority of these patients will have symptoms of intermittent claudication. The prognosis for people with claudication is quite good, and most patients will have only mild progression of the disease throughout their lifetime. Only 2–4% of patients with intermittent claudication will ultimately require major amputation [40]. The prognosis for the patients with diabetes and claudication is worse compared to those without diabetes. The critical ABI that predicts a poor prognosis is 0.5. Patients with ABI values greater than 0.5 experience

Peripheral arterial disease

199

a very slow progression towards limb-threatening ischemia with the decline being more rapid in patients presenting with ABI values less than 0.5. As expected, in this patient population survival is severely compromised and lower ABI values are associated with a higher long-term mortality rate. That benign natural history of PVD is claudication has been demonstrated in multiple studies. In a study of 2,777 patients with claudication over a 15-year period, Muluk et al. found the cumulative 10-year frequency of major and minor amputation to be only 10% [41]. In this study, the 10-year cumulative rate of revascularization for limb-threatening ischemia was 18% [41]. In a follow-up study of those patients with complete ABI records, the authors found that patients with ABI values greater than 0.5 progressed to limb-threatening ischemia at a lower rate than those with ABI values less than 0.5. For the entire study group, the 10-year cumulative risk of progression to ischemic ulceration and ischemic rest pain were 23% and 30%, respectively [6]. The average yearly decline in ABI was 0.014 and walking distance declined by 9.2 yards per year [6]. Multivariate analysis confirmed that ABI and pharmacologically treated diabetes were independent predicts of developing ischemic rest pain and/or ulceration [6]. Smoking history did not predict progression of disease by multivariate analysis but other studies have demonstrated a strong association between smoking and disease progression [40, 42]. These results are similar to older publications which have consistently demonstrated that only a quarter of the patients with intermittent claudication (IC) will have clinically significant deterioration [40, 43]. Interestingly, the historical data demonstrates that over 50% of patients with claudication will become symptom-free after 5 years [43]. Bird et al. found that 23% of patients will improve clinically during follow-up [44]. Thus, claudication is benign and only progresses to limb-threatening ischemia in a minority of patients. The prognosis for patients with diabetes is worse compared to nondiabetics. The incidence of PVD is 2- to 4-fold higher in diabetic patients compared to nondiabetic patients [45]. In the Framingham Study, diabetic patients had markedly increased rates of absent pedal pulses, depressed ABI values, and femoral bruits; diabetics were found to have a 3-fold and 8-fold increase in men and women, respectively, in the rates of intermittent claudication [45]. The progression to limb-threatening ischemia is more rapid in diabetics. Diabetics with claudication in the Aquino study were calculated to have nearly twice the rate of progression to ischemic rest pain than nondiabetics across the full range of ABI values. The calculated rate was nearly triple in diabetics when progression to ischemic ulceration was analyzed [6], After 12 years, approximately 45% of diabetics had progressed to ischemic rest pain and 60% had progressed to ischemic ulceration [6], It is unclear why diabetics have a worse natural history, but it may be related to the pattern of disease they express. By arteriography, diabetics have a higher incidence of tibial disease and disease in the deep femoral artery [46]. The small diameter of tibial vessels and the poor collateral supply around the knee may dispose diabetics to a more rapid progression of disease. While claudication is a relatively benign diagnosis when considering the fate of the limb, the diagnosis of claudication has grave implications for the survival of the patients. Patients with claudication have a markedly altered long-term survival compared to agematched controls. This is in large part due to concurrent cardiovascular and cerebrovascular disease (see Figure 10.4). Coronary arteriography has revealed the

Thrombosis in clinical practice

200

presence of significant coronary disease in over 90% of patients being evaluated for peripheral bypass surgery [47]. In the study of 2,777 claudicants by Muluk, the yearly mortality rate was 12%, with 5- and 10-year morality rates of 42% and 65%, respectively [41]. Historical data has set these rates as 30% and 50%, respectively [43]. In a subset of 236 patients, Muluk found that 66% died as a result of ischemic cardiac events [41]. To put these statistics in perspective, the expected mortality for comparable ageadjusted US male population is approximately 15% at 5 years and 25% at 10 years [41]. The diagnosis of intermittent claudication has more important implications for the patient’s long-term survival than for the fate of the extremity. Treatment of such patients should

Figure 10.4 Frequency of symptomatic disease in the three primary organ systems afflicted by atherosclerosis and their overlap, from the CAPRIE trial (Reproduced with permission from Ouriel K. Peripheral arterial disease. Lancet 2001; 358:1257–64) [39]. aim to treat systemic atherosclerosis, rather than focusing on the extremities. The symptoms of claudication may be the first manifestations of advanced coronary disease, and, thus, serve as an important portal of entry for therapy. Physicians should worry little about the fate of the extremity and focus on reducing the coronary risk to the patient. There are subpopulations of claudicants that should be considered for early revascularization. Patients without rest pain or tissue loss who have critical hemodynamic indices have a more malignant course and should be considered for early revascularization. In patients with dependent rubor (absent rest pain or tissue loss), 25% will develop limb-threatening ischemia after 4 years [48]. Patients with claudication lacking rubor have a 9% rate of progression to limb-threatening ischemia over the same time frame [48]. In another study, Bowers found that patients with claudication and critically low toe pressures (<40 mmHg) had a 34% rate of progression to rest pain or tissue loss over a 31-month period [49]. In the control group with claudication and toe pressures greater than 40 mmHg, only 9% progressed to limb-threatening ischemia [49].

Peripheral arterial disease

201

Fowl, in a study of 23 patients with claudication and ABI values less than 0.35 found that 48% of patients progressed to limb-threatening ischemia over a mean follow-up of 45 months [50]. Thus, patients with low ABI values, low toe pressures, and/or dependent rubor are a subgroup of claudicants who can be expected to progress to limb-threatening ischemia, and should be considered for early arterial reconstruction. The onset of rest pain and tissue loss heralds the onset of limb-threatening ischemia. The natural history for both rest pain and tissue loss is assumed to be grave and most expect that limb loss is inevitable if operative intervention is not undertaken. However, the natural history of rest pain and tissue loss is nebulous owing to the fact that most series include patients that undergo revascularization. Wolfe and Wyatt analyzed 20 publications reporting the results of 6,118 patients with critical leg ischemia and grouped the patients into low-risk (rest pain and/or ankle pressure >40 mmHg) and high-risk patients (tissue loss and/or ankle pressure <40 mmHg). A total of 73% of low-risk patients treated without revascularization required amputation at 1 year [51]. The results for the high-risk group was much worse with 95% requiring amputation at 1 year [51]. These results should be considered biased, as patients treated conservatively are often not candidates for revascularization. Other studies have reported lower amputation rates with conservative management. Vayssariat et al. reported that of 20 surviving patients with critical limbischemia treated conservatively, only 1 required major amputation and 8 required minor amputations [52]. These patients were not offered revascularization due to poor anatomy for operative bypass and a clinical assessment that dystrophic lesions would heal without intervention [52]. In the ischemia critica arti inferiori trial of efficacy and safety of prostanoids in the treatment of critical limbischemia, the 6-month amputation rate was 18% in patients who were not deemed candidates for revascularization [53]. The RR for amputation (with or without revascularization) was higher for patients with ulceration (RR=1.64) compared to patients with rest pain (RR=1.29) [53]. Ubbink reported a series of 111 patients with rest pain and/or tissue loss who were not candidates for vascular reconstruction. In this series, the rate of progression to amputation was 44% [54]. The true natural history of rest pain and tissue loss between the extremes of this data is culled from the literature. Most surgeons expect that between 25% and 40% of patients with ischemic rest pain will require amputation at 1 year without intervention. For patients with tissue loss, the rate of limb loss at a year may approach 80%. It is notable that not all patients with rest pain and/or tissue loss will require amputation. Some patients with ischemic rest pain will have resolution, and up to 30% of patients with mild ischemic tissue loss will heal without arterial bypass. Thus, not all patients who are not revascularization candidates should undergo mandatory amputation. If rest pain can be controlled with analgesics or tissue loss is nonprogressive, many patients can be managed conservatively for long periods without amputation. However, those patients who are revascularization candidates should undergo arterial reconstruction. Clearly, the development of critical hemodynamic indices, rest pain, and/or tissue loss are harbingers of limb loss. They also have grave implications for the survival of the patient and a prognosis that is much worse than that associated with claudication. The expected 5-year survival for such patients is between 20% and 40% [55, 56]. Patients who are not candidates for arterial reconstruction and undergo amputation have a

Thrombosis in clinical practice

202

particularly poor expected survival. It can be expected that only 70% of patients undergoing lower extremity amputation will be alive after 2 years and survival is worse in patients who undergo amputation for critical ischemia (1 year survival=57%) [56, 57]. The majority of deaths in this patient population are cardiac in origin [58]. The onset of critical limb-ischemia is a marker for severe systemic atherosclerosis and all such patients should be assumed to have significant coronary occlusive disease. Treatment of such patients should be tailored to preserve functional limbs and to treat concurrent coronary artery disease.

Medical therapy for PVD Medical therapy for PVD consists primarily of treatment of concurrent coronary artery disease. The goal is to decrease the risk of acute coronary syndromes, stabilize coronary plaque, and perhaps promote plaque remodeling and regression. The added benefit is a potential effect on the peripheral circulation stabilizing the progression of peripheral disease. Antiplatelet therapy The efficacy of antiplatelet therapy for the prevention of acute coronary syndromes and stroke is well established. The Antithrombotic Trialists’ Collaboration recently completed a meta-analysis of the efficacy of antiplatelet therapy in high-risk patient populations [59]. The meta-analysis looked at the effects of aspirin and clopidogrel in preventing nonfatal MI, nonfatal stroke, or vascular death. This study reviewed 195 trials that enrolled 135,640 patients. Antiplatelet therapy resulted in a 34% proportional reduction in the rate of nonfatal MI, a 26% proportional reduction in nonfatal MI and death, a 25% proportional reduction in nonfatal stroke, and a 15% proportional reduction in vascular deaths [59]. In patients with PVD, there was a 23% proportional reduction in serious vascular events [59]. Based on this meta-analysis there is no indication that high-dose aspirin therapy adds additional risk reduction and only increases the risk of bleeding complications [59]. The recommended dose of aspirin is between 81 and 325 mg. Clopidogrel appears to provide additional benefits with a 10% risk reduction for major vascular events when compared to aspirin [59, 60]. Combining aspirin with clopidogrel reduces the risk of coronary events in patients with unstable angina but combined therapy for high-risk patients without unstable angina awaits the results of the CHARISMA trial [61], Despite the evidence that clopidogrel adds additional benefits compared to aspirin, the cost of lifelong therapy with this medication may outweigh the benefit. All vascular patients will benefit from therapy with aspirin, and clopidogrel should be considered in the very high-risk patient with established symptomatic coronary disease. Antiplatelet therapy appears to have additional benefits in terms of the progression of extremity atherosclerosis. Several nonrandomized prospective studies have suggested that aspirin therapy may slow the angiographic progression of occlusive disease, increase walking distances, and improve resting ABI values. [62–64]. However, these studies have not been reproduced and they remain as only weak evidence that aspirin has an effect on the progression of PVD. Ticlidopine has been demonstrated to significantly reduce the

Peripheral arterial disease

203

number of revascularization procedures in patients with claudication in two prospective randomized studies [65]. Likewise, clopidogrel has been demonstrated to significantly reduce major vascular events in one prospective randomized study [60]. Ticlopidine has largely been abandoned due to its adverse risk profile leaving clopidogrel as the only therapy that has proven benefit in patients with PVD. However, as stated earlier, it is unclear if the cost of lifelong therapy with clopidogrel warrants its routine use in patients with PVD. Thus, aspirin remains the mainstay of antiplatelet therapy for patients with PVD and all patients should receive lifelong therapy with aspirin. Beta-blockers Beta-blockers are well recognized as beneficial for the reduction of mortality following MI and in the treatment of congestive heart failure. It is very clear that beta-blockade reduces mortality following vascular surgery. The first randomized, placebo-controlled study of beta-blockade during noncardiac surgery was published in 1996. A total of 200 patients were randomized to receive placebo versus atenolol two days prior to surgery with continued therapy for 7 days. In patients receiving beta-blockade, overall mortality was decreased compared to placebo at 6 months and this effect continued for 2 years [66]. The primary effect of betablockade was to reduce cardiac mortality in the first 6–8 months [66]. Further studies have supported this finding. Bisoprolol was found to reduce the incidence of cardiac events in high-risk cardiac patients with abnormal dobutamine stress echocardiography undergoing major vascular surgery. Treated patients experienced a 12% rate of cardiac events versus 32% for patients treated with standard care and the rates of cardiac death and MI was significantly reduced in the Bisoprolol-treated group compared to control at 2 years [67]. The withdrawal of beta-blockers in the early postoperative period following vascular surgery is associated with an increased cardiovascular mortality and postoperative MI [68]. Clearly, beta-blockade is cardioprotective, reduces mortality following vascular surgery, and is considered the standard of care. However, it is not known if all patients with PVD should be treated with lifelong beta-blockade. It was initially thought that betablockade would have an adverse effect on the symptoms associated with claudication. Multiple studies have demonstrated that beta-blockers have no adverse effect on claudication [69]. Almost all patients with PVD have coronary artery disease and, thus, it is reasonable to assume that all patients will benefit from beta-blocker therapy. This is conjecture, as there is little data at present that confirms this opinion. Only one observational study of patients with PVD and prior MI has demonstrated the efficacy of beta-blockade for the prevention of cardiac events in nonoperative vascular patients. In this study of 490 patients, the use of beta-blockade was associated with a 53% independent reduction in new coronary events in patients treated with beta-blockade versus those untreated (risk ratio 0.467) [70]. It is not known if vascular patients without a history of MI will benefit from long-term beta-blockade. Beta-blockade has no adverse effect on the symptoms of occlusive disease, is expected to reduce cardiac events in patients with PVD, and is well tolerated by most patients. Thus, it is reasonable to treat all patients with PVD with lifelong beta-blockade. Beta-blockade is mandatory in all patients undergoing vascular surgery, and is essential in all patients with prior MI.

Thrombosis in clinical practice

204

Statin therapy The class of statin drugs, including simvastatin, atorvastatin, and pravastatin, are 3hydroxy-3methyl-glutaryl coenzyme A (HMG-CoA) reductase inhibitors that block the enzyme responsible for endogenous cholesterol biosynthesis. There is growing evidence that statin therapy have a wide-ranging effect on the natural history of atherosclerosis. Statin-induced reduction of cholesterol levels improves endothelial function by increasing NO production, inhibiting multiple pathways that promote thrombogenicity, and blocking cholesterol-induced alteration of heparin sulfate proteoglycan [71]. It is postulated that statin therapy provides long-term benefits of plaque stabilization by reducing cholesterol uptake into the arterial wall and decreasing inflammation within the plaque, thereby lowering the risk of plaque rupture. It is well established that statin therapy results in a remarkable reduction in cardiovascular events in patients with coronary heart disease (CARE trial) [72]. In the Scandinavian Simvastatin Survival Study (4S) in 1994, there was a 30% reduction in overall mortality and a 42% reduction in cardiovascular-related mortality [73]. Stroke risk also was reduced by 20%, suggesting a salutary effect of lipidlowering therapy on cerebrovascular and perhaps systemic atherosclerosis [73]. A subsequent analysis of the 4S trial determined that the risk of new or worsening intermittent claudication was reduced by 38% in the group treated with simvastatin [74]. A meta-analysis of 698 patients with PVD treated with lipid-lowering agents and diet demonstrated a trend towards mortality reduction, and a reduction in the severity of claudication [75]. Statin use has recently been shown to improve functional outcome in patients both with and without PVD [76]. Study participants taking statins demonstrated better walking performance than patients not taking statins, regardless of their prior history of PVD [76]. Another recent study confirmed that even short-term (6-month) therapy with simvastatin at 40 mg o.d. resulted in significant improvement in pain-free walking distance, total walking distance and ABI at rest in vascular patients [77]. Despite growing evidence for the benefits of this therapy, statin therapy use continues to be poor, comprising only 38% (5–46%) of vascular patients [78]. Current recommendations for highrisk patients include annual screening of cholesterol level, initiation of statin therapy to attain a goal of lowering LDL-C to less than 100 mg/dl, total non-HDL cholesterol less than 130 mg/dl, and additional therapy to lowering triglyceride level to less than 150 mg/dl if necessary [79]. Statin therapy should be considered a mainstay of therapy for all patients with PVD. Angiotensin-converting enzyme (ACE) inhibitors New evidence has demonstrated that the reninn-angiotensin system plays a critical role in the development and progression of PVD. ACE cleaves angiotensin I to angiotensin II. Angiotensin II is thought to play a role in the pathogenesis of atherosclerosis by altering endothelial function. Angiotensin II is a potent vasoconstrictor, is prothrombotic via activation of plasminogen activator inhibitor, promotes smooth muscle migration/proliferation, and inhibits formation of the vasodilator, bradykinin [80]. Animal studies have demonstrated a direct antiatherogenic action of ACE inhibitors through an antiproliferative and antimitogenic mechanism, which benefits endothelial

Peripheral arterial disease

205

function through its plaque stabilizing and antithrombotic effects [81, 82]. Plaque stabilization results from documented reduction in the cellularity and cholesterol content of atherosclerotic plaque, perhaps leading to a reduced risk of plaque rupture [82]. The data shows that ACE inhibitors improve survival and symptoms in patients with heart failure [83]. Recent studies have focused on the cardioprotective effects of ACE inhibitors beyond what is expected solely from blood pressure control. The HOPE study evaluated 9,297 patients with evidence of heart disease or asymptomatic diabetic patients with one cardiovascular risk factor (hypertension, dyslipidemia, or smoking) and demonstrated a 17% RR reduction in all-cause mortality, with a 22% RR reduction in combined cardiovascular deaths. There was a 12% RR reduction in need for cardiac or peripheral revascularization procedures [84]. Some 44% of the patients studied (4,046) had evidence of PVD, and a subgroup analysis of these patients demonstrated that they benefited more than patients without vascular disease [84]. Current recommendations for best medical therapy regarding ACE inhibitor therapy in vascular patients are continuing to evolve. Blood pressure reduction is a cornerstone of risk factor modification and secondary prevention of cardiovascular morbidity and mortality in vascular patients. Goal of therapy should target blood pressure lower than 130/85. Initiation of ACE inhibitor therapy is expected to have salutary effects beyond simple blood pressure reduction. If a patient does not have contraindications to ACE inhibitors, such as bilateral renal artery stenosis or renal insufficiency, starting a low-dose ACE inhibitor is recommended for all patients with PVD. Medical management of claudication The mainstay of medical therapy of claudication is risk factor modification and exercise therapy. Risk factor modification entails smoking cessation, treatment of hypertension, treatment of dyslipidemia, and tight diabetic control. Medical therapy for patients with PVD was covered in depth in the previous section but a few comments are warranted. Smoking cessation is a difficult goal to attain in this patient population with recidivism being common. Patients who stop smoking are estimated to have twice the 5-year survival of patients who continue to smoke [85]. Most patients will fail to quit without a structured program that includes nicotine replacement and antidepressant therapy [86, 87], At best, sustained smoking cessation can be anticipated in less

Table 10.1 Therapeutic goals for the medical management of claudication. Pharmacologic interventions

Antihyperlipidemic pharmacotherapy Antiplatelet therapy Treatment of hyperhomocystinemia Blood sugar control Antihypertensive therapy

Life-style

Exercise program

modifications

Smoking cessation

Thrombosis in clinical practice

206

Weight loss program (Reproduced with permission from Ouriel K. Peripheral arterial disease. Lancet 2001; 358:1257– 64) [39].

than 30% of patients. Therapeutic goals for medical therapy are listed in Table 10.1. The most effective therapy for claudication is exercise therapy. Exercise therapy entails that the patient walk until claudication symptoms are severe, rest to allow recovery, and repeat the cycle for the set session time. Multiple studies have demonstrated the utility of exercise therapy. In a meta-analysis of 21 studies, exercise therapy was found to increase walking distances to the onset of pain by 179% and the distance to maximal pain was increased by 122% [88]. The greatest improvements were found in exercise programs with walking sessions greater than 30 minutes, at least 3 sessions per week, achievement of near maximal pain prior to recovery, and duration of therapy for 6 months [88]. In a meta-analysis of 10 studies, the Cochran group found that exercise therapy increased walking ability by 150% and had better results than angioplasty or antiplatelet therapy [89]. The data is clear that exercise therapy is highly successful in improving claudication symptoms and one can expect walking distances to more than double in motivated patients. Unfortunately, only motivated patients can be expected to achieve the full benefits of such therapy. However, exercise should be firstline therapy for claudication with pharmacologic therapy reserved for treatment failures. For patients with severe claudication that fail exercise therapy, pharmacologic therapy should be considered. However, pharmacologic therapy should not be considered until the patient fails exercise therapy. The additional benefits of exercise therapy in terms of sense of well-being, weight loss, and improved cardiovascular function are as important as the improvement in leg symptoms. Often, pharmacologic therapy is the path of least resistance taking no effort on behalf of either the patient or physician. In addition, pharmacologic therapy is expensive for the patient and costly for society. Two pharmacologic therapies exist for the symptoms of claudication: Pentoxifylline and Cilostazol. Pentoxifylline is a weak antithrombotic agent, where the putative mechanisms of action include an increase in red blood cell deformity, and decreases in fibrinogen concentration, platelet adhesiveness, and whole-blood viscosity [90–92]. A number of clinical trials have evaluated pentoxifylline but have shown conflicting results. Some have concluded that pentoxifylline was significantly more effective than placebo in improving treadmill-walking distance [93–98], but others could not demonstrate consistent benefit [99–104]. In many trials, patients treated with placebo also demonstrated significant improvement. Thus, the actual improvement in walking distance attributable to pentoxifylline is often unpredictable and may not be clinically important compared with the effects of placebo [105]. Nevertheless, one meta-analysis suggests that pentoxifylline improves walking distance by 29 meters compared with placebo, although the improvement was approximately 50% in the placebo group, and use of pentoxifylline improved walking distance by an additional 30% [106]. This benefit was substantially less than that achieved with a supervised exercise program (e.g. 123% increase in peak walking time) [107], In summary, the evidence for a beneficial effect of pentoxifylline is not good enough to define its role in the treatment of patients with PAOD [108].

Peripheral arterial disease

207

Cilostazol is a type III phosphodiesterase inhibitor that suppresses platelet aggregation and is a direct arterial vasodilator. Cilostazol also has beneficial effects on the triglyceride and high-density lipoprotein levels [109]. Its mechanism of action as a treatment for claudication is not fully understood but is likely multifactorial. Multiple studies have demonstrated that cilostazol is an effective therapy for the treatment of claudication. A recent meta-analysis of eight randomized trials of the efficacy of cilostazol demonstrated that cilostazol significantly increases walking distances and quality of life [109]. Cilostazol increased maximal walking distances by 50% and painfree walking distances by 67% [108]. In two of the trials, cilostazol was compared to pentoxifylline and demonstrated significant improvements in walking distances compared to pentoxifylline [109]. In these studies, pentoxifylline was no different than placebo therapy. A separate meta-analysis confirmed these findings [110]. To date there has been no direct comparison of supervised exercise therapy versus cilostazol for claudication. Based on the evidence at hand, exercise therapy appears to improve walking distances more consistently than cilostazol. Cilostazol, at present can only be recommended as second line therapy and should always be administered with concurrent exercise therapy. The effects of cilostazol therapy on the progression of PVD (beyond its effects on claudication) awaits investigations. The favorable effects on lipid profiles and platelet function may offer a role for cilostazol as a primary therapy for PVD. The CREST trial seeks to determine if cilostazol has an effect on preventing restenosis following coronary angioplasty [111]. Should cilostazol be found to have a positive effect atherogenesis, it may have role as a primary treatment for PVD in the future. Surgical therapy Surgical therapy is reserved for patients with severe occlusive disease. Based on the natural history of the disease, patients with rest pain and tissue loss should be offered interventions or will face amputation. The difficult question is which patients with claudication should be offered interventions. Patients with very short distance claudication, rapidly progressive claudication, and critical hemodynamics (ABI<0.4) should be considered for intervention if they are low risk. For patients with moderate claudication, the decision for intervention must be made cautiously. The term “life style limiting claudication” is often used as an indication for operative therapy. However, this term is too broad and should be refined to define those patients that have claudication that is severe enough to adversely affect the patient’s economic or social status. The advent of angioplasty has allowed a more liberal decision-making process but this modality does carry some risk, albeit low. Angioplasty Endovascular techniques for the treatment of PVD offer lower associated morbidity and mortality compared to open surgical revascularization. However, the lower risk is offset by diminished durability of the repair. The determination of which option, endovascular versus open surgical, to offer a patient requires knowledge of the durability of each repair, the medical status of the patient, and, to a lesser extent, patient preference. When considering the durability of endovascular repair it is helpful to divide the peripheral

Thrombosis in clinical practice

208

arterial tree into the aortoiliac segment and the infrainguinal segment. A general principle that can be applied is that single lesions and lesions of short length respond to endovascular therapy quite well with patency rates rivaling open surgical reconstruction.

Aortoiliac occlusive disease In 2000, the TransAtlantic Inter-Society Consensus (TASC) published recommendations for the endovascular management of PVD. The TASC group divided iliac lesions into four categories (A, B, C, and D) based on the morphology of the lesion [112]. The morphologic definitions of the lesions and preferred methods for management are depicted in Figure 10.5. In general, short, less than 3 cm lesions (type A), should be preferentially treated by endovascular techniques while diffuse disease or long occlusions (type D) should be preferentially treated by surgical revascularization. There is no current scientific evidence favoring endovascular or open therapy for the treatment of type B and C lesions [112]. B and C lesions can successfully be treated by either approach, but durability with endovascular approaches is expected to be less than open surgical therapy. There is no convincing data to suggest that primary stenting of iliac lesions is beneficial in terms of patency. Meta-analysis indicated a 39% reduction in long-term failure in patients treated with stent placement versus angioplasty alone [113]. However, follow-up randomized studies of percutaneous transluminal angioplasty (PTA) with selective stent placement versus primary stenting demonstrated no benefit for primary stenting [114]. Selective stenting is recommended for technical failures of PTA such as elastic recoil, dissection, or residual pressure gradient, whereas primary stenting is indicated for complex lesions, embolic lesions, and occlusions (see Table 10.2) [112]. The expected patency for iliac lesions treated by endovascular techniques is present in Table 10.3.

Infrainguinal disease The results of endovascular repair of infrainguinal disease have been less encouraging than treatment of aortoiliac disease. Like aortoiliac disease, the TASC categorized femoral-popliteal disease in terms of anatomic patterns [112]. The A through D grading system was used again and the description is found in Figure 10.6. Endovascular therapy is recommended for type A lesions and surgical therapy for type D lesions [112]. Type B lesions are treated preferentially by endovascular techniques, whereas most type C lesions are treated with surgical bypass; however, little evidence exists to guide therapy for type B and C lesions [112]. It is clear that short lesions, proximally located lesions, nonocclusive lesions, nondiabetics, claudicants, and patients with 2–3 vessels tibial runoff versus 1 vessel runoff to the foot have improved technical success and long-term patency [112, 115]. Meta-analysis revealed equivalent results for femoral-popliteal angioplasty versus femoral to popliteal bypass with synthetic grafts but inferior results to vein bypass [116]. Primary stenting of the femoral-popliteal segment does not improve patency [117]. Thus, there is no role for primary stenting of femoral-popliteal segments and stenting

Peripheral arterial disease

209

Figure 10.5 TransAtlantic InterSociety Consensus (TASC) anatomic classification of aortoiliac occlusive disease. Short, isolated iliac lesions (TASC type A) are preferentially treated by endovascular approaches. More complex lesions (TASC type B and C) are treated by either endovascular or open surgical approaches. However, insufficient data exists for recommendations. The most complex lesions (TASC type D) are treated preferentially via open surgical approach (Reproduced with permission from Dormandy JA, Rutherford RB. Management of peripheral arterial

Thrombosis in clinical practice

210

disease (PAD). TASC Working Group. TransAtlantic Inter-Society Consensus (TASC). J Vasc Surg 2000; 31 (1 Pt 2):S1–S296) [112]. Table 10.2 Indications for selective stenting following angioplasty. • Insuffieient hemodynamic result of PTA due to elastic recoil based on residual pressure gradient • Massive, lumen-obstructing dissection • Treatment of chronic occlusions • Iliac artery ulceration associated with symptoms • Restenoses after previously performed PTA • Complex lesions for which primary stenting may give more satisfactory results (Adapted with permission from Dormandy JA, Rutherford RB. Management of peripheral arterial disease (PAD). TASC Working Group. TransAtlantic Inter-Society Consensus (TASC). J Vasc Surg 2000; 31(1 Pt 2):S1–S296.)

Table 10.3 The expected technical success and patency rates following iliac angioplasty. Treatment

Claudicants

Technical success

1 yr

Patency 3 yr

5 yr

Stenosis/PTA

77

86

78

66

61

Occlusion/PTA

82

83

68

60

–

Stenosis/Stenting

78

99

90

74

72

Occlusion/Stenting

86

82

72

64

–

In general, treatment of stenotic lesion results in improved long-term patency compared to treatment of occlusive lesions. (Adapted with permission from Dormandy JA, Rutherford RB. Management of peripheral arterial disease (PAD). TASC Working Group. TransAtlantic InterSociety Consensus (TASC). J Vasc Surg 2000; 31(1 Pt 2):S1–S296.)

should be reserved to treat PTA failures such as recoil and dissection [112]. Selfexpanding stents are preferred over balloon expandable stents as they exert more radial force and are more resistant to compression by the surrounding tissue [112]. Expected patency following femoral-popliteal stenting is presented in Table 10.4.

Peripheral arterial disease

211

Recommendations Stenting of the aortoiliac segment is the preferred approach for simple lesions that are short in length and for short occlusions (TASC type A and B). Because of the low incidence of complications following aortoiliac stenting and the excellent patency that rival open repair, a more aggressive stance to the treatment of claudication can be taken with such lesions. Patients who have abnormal femoral pulses or reduced high thighpressures with moderate to severe claudication can be screened for appropriate iliac lesions anticipating endovascular therapy. The most appropriate screening study is minimally invasive such as MRA or CT angiography. However, one should anticipate that only a minority of such patients will have TASC type A or B lesions. Medical therapy for claudication should be instituted prior to

Figure 10.6 TransAtlantic InterSociety Consensus (TASC) anatomic classificatlon of femoral-popliteal occlusive disease. Short, isolated femoral-popliteal lesions (TASC type A) are preferentially treated by endovascular approaches. Moderately complex lesions (TASC type B) are treated by either endovascular or open surgical approaches. However,

Thrombosis in clinical practice

212

insufficient data exists for recommendations. The most complex lesions (TASC type C and D) are treated preferentially via open surgical approach. (Reproduced with permission from Dormandy JA, Rutherford RB. Management of peripheral arterial disease (PAD). TASC Working Group. TransAtlantic Inter-Society Consensus (TASC). J Vasc Surg 2000; 31 (1 Pt 2)S1–S296.) intervention but the decision to treat is made less rigorous for such lesions. Patients with type C or D lesions or patients with multilevel disease (aortoiliac and infrainguinal) do not respond well to endovascular therapy and are likely to need surgical intervention. For such patients the criteria to intervene should be rigorous. Treatment of femoral-popliteal disease via endovascular approaches is unusual and the criteria to treat should be rigorous. The finding of isolated TASC type A or B lesions in patients with moderate to severe claudication is rare as most patients will have extensive occurences of SFA disease prior to developing significant claudication. For this reason, the authors do not

Table 10.4 Patency following femoral-popliteal angioplasty/stenting versus femoral-popliteal bypass. Although technical success rates for femoral-popliteal angioplasty/stenting are quite good, the patency rate for such interventions is markedly diminished compared to surgical bypass. Endovascular techniques should be reserved for simple, isolated lesions of the femoral-popliteal segments. Treatment

Claudicants

Technical success

Patency 1 yr

3 yr

5 yr

PTA

72

90

61

51

48

Stent

80

98

67

58

Femoral—popliteal bypass with vein

–

–

–

–

80

Femoral-popliteal bypass with PTFE above knee

–

–

–

–

75

Femoral-popliteal bypass with PTFE below

–

–

–

–

65

Peripheral arterial disease

213

knee (Adapted with permission from Dormandy JA, Rutherford RB. Management of peripheral arterial disease (PAD), TASC Working Group, TransAtlantic Inter-Society Consensus (TASC). J Vasc Surg 2000; 31(1 Pt 2):S1–S296.)

routinely use imaging modalities to screen for such lesions in patients with clinical femoralpopliteal disease. Such patients should have exhausted and failed medical management prior to consideration for repair and most will be found to have extensive femoral-popliteal disease that is not amenable to endovascular therapy. Thus, while endovascular therapy is appropriate for TASC type A and B lesions, patients with moderate to severe claudication who are candidates for such therapy will be found only rarely. As most patients with claudication can be managed with medical therapy alone, the true benefit of endovascular therapy is as an adjunct to surgical therapy for multilevel disease in patients with severe limb-threatening ischemia. In such patients, disease in the iliac segment can be treated with endovascular techniques at the time of open surgical infrainguinal revascularization. Such an approach has been met with excellent success rates and iliac stenting has been demonstrated to improve the patency of infrainguinal bypass compared to iliac angioplasty [118]. With the advent of covered stents (Wallgraft and Viabahn) TASC type C and D can be successfully treated with endovascular techniques. Covered stents allow for aggressive dilation of severely diseased iliac segments markedly reducing the risk of rupture and may reduce neointimal hyperplasia. Treatment of TASC C and D lesions with covered stents results in reduced durability compared to aortobifemoral bypass but equivalent or better patency compared to axillobifemoral bypassing (2 year primary patency of 84%) [119]. Often concurrent open repair of the common femoral and profunda femoral arteries is required for successful treatment [120]. This form of endovascular therapy for diffuse iliac occlusive disease is appropriate for high-risk patients with limb-threatening ischemia that would only be candidates for extra-anatomic bypasses. Open surgical therapy Open surgical therapy remains the gold standard for therapy of occlusive disease. However, open surgical therapy comes at the cost of moderate rates of morbidity and low rates of mortality. Because of the associated morbidity, surgical revascularization should be reserved for patients with severe claudication or limb-threatening ischemia. Aortoiliac reconstructions Aortic reconstruction for occlusive disease offers excellent long-term patency rates but at the expense of acceptable morbidity and mortality rates. Aortobifemoral bypass with prosthetic grafts have proven to be very durable. The 5- and 10-year patency rates are 86% and 79%, respectively [112]. Aortic reconstructions require opening of either the abdominal cavity or approach to the aorta via the retroperitoneum. After the aortic

Thrombosis in clinical practice

214

anastomosis is completed, the graft is tunneled beneath the inguinal ligament to allow anastomosis to the femoral arteries in the groin. As this procedure requires exposure of the aorta, there is a high magnitude of operative stress for the patient. Mortality rates range from 1% to 3%. Recovery from this operation is often prolonged in the elderly and the young patient is unlikely to return to work for at least 2 months. Violation of the groin increases the likelihood of graft infection with infection rates as high as 1%. Other complications include impotency, major and minor wound infections, myocardial infarction, pneumonia, renal failure, etc. (see Table 10.5). Thus, the risk-benefit ratio for aortobifemoral bypass must be well established. Lowrisk young patients may benefit greatly from aortofemoral bypass by virtue of the excellent patency rates. Such patients should only be considered for operative reconstructions if they have moderate to severe claudication and have failed medical management. Moderate to high-risk patients with severe claudication should not be considered for such operations. Moderate risk patients should undergo aortofemoral bypass for limb-threatening ischemia while high-risk patients should be repaired with extra anatomic bypasses such as axillobifemoral bypass or femoral to femoral bypass for unilateral disease. Extra anatomic bypasses such as axillobifemoral bypass (5 year patency of 50%) and femoral to femoral bypass (5 year patency of 60%) offer low-risk revascularization at the expense of lower patency rates. As mentioned earlier such bypasses may be replaced in the future by covered endovascular stents. Infrainguinal reconstruction Reconstruction of the infrainguinal vessels does not result in the same durability of the repair as aortobifemoral bypass. This is likely due to the high flow rates and the large size of aortic grafts. The durability of infrainguinal bypass grafts must be considered as well as the natural history of the patient’s disease. In patients with mild claudication, the progression of the disease is so slow that the limb will outlast that graft. The durability of femoral to popliteal grafts is acceptably high to consider revascularization in patients with moderate to severe claudication. Primary patency rates of femoral-popliteal grafts vary from 70% to 80% at 5 years

Table 10.5 Complications following aortic reconstruction for aortoiliac occlusive disease Complication

Incidence (%)

Etiology/comments

Myocardial

0.8–5.2

Concurrent cardiac disease

Death

0–3.3

Usually myocardial

Intestinal ischemia

1.1

Ligation IMA—colonic Preexisting SMA disease

Renal failure

0–4.6

Preexisting renal dysfunction increases risk

Ureteral injury

1.6

Frequent association with graft complication

Spinal cord ischemia

0.25

Atheroemboli, occlusion vascular supply

Peripheral arterial disease

215

Graft infection

0.1–1.3

Higher incidence involving groin anastomosis

Aortoenteric fistula

0.1–0.5

Erosion, lack of reperitonealization, aortic false aneurysms

Lymph fistula

1.5–3.5

Division of lymphatics

False aneurysm

3–5

Infection, native artery degeneration

Altered sexual function

20

IMA: inferior mesenteric artery; SMA: superior mesenteric artery. (Adapted with permission from Dormandy JA, Rutherford RB. Management of peripheral arterial disease (PAD). TASC Working Group. TransAtlantic Inter-Society Consensus (TASC). J Vasc Surg 2000; 31(1 Pt 2):S1–S296.)

depending on the distal target and the conduit that is used (see Table 10.4). The use of prosthetic grafts should only be considered in patients who are candidates for femoral to above-the-knee popliteal bypasses. The use of any prosthetic graft is controversial as the disadvantages of the prosthetic infrainguinal grafts may outweigh the benefits. Prosthetic grafts are much more prone to infection than venous grafts, they offer no improvement in patency, when they fail they often result in increased degree of ischemia compared to preoperative levels, the dissection required for placement of the graft may injure the saphenous vein compromising its use in the future, and it is rare to find a patient with a significant claudication that has a disease-free above-the-knee popliteal vessel [121]. Therefore, in the authors’ opinion prosthetic grafts are of dubious merit for the treatment of claudication. Autogenous saphenous grafts are the preferred conduit and have equivalent patency rates whether anastomosed to the above- or below-the-knee popliteal artery. For patients with femoral-popliteal occlusive disease, surgical revascularization should be considered only for moderate to severe claudication that has failed medical therapy. The preferred conduit is autogenous ipsilateral saphenous vein. The durability of femoral to tibial bypasses is limited (primary patency 60% at best) and such bypasses should not be undertaken for claudication except in very highly selected patients, if at all. For patients with rest pain or tissue loss, the decision to operate is very straightforward—all such patients should be considered for revascularization. Revascularization is at least as safe as amputation, in terms of operative mortality. Patients who undergo amputation will often never ambulate, particularly following abovethe-knee amputation. Revascularization has been demonstrated to be more costeffective than amputation. Lastly, revascularization may offer the patient continued independence that would be lost with amputation. Even minimally ambulatory patients with unilateral amputation should be considered for contralateral revascularization in the face of limb-threatening ischemia. Such patients will often use their remaining leg to facilitate pivoting and may have limited ambulation with a walker that affords them independence. Most patients with limb-threatening ischemia will have multi-level disease and most will need tibial bypasses for repair. The conduit of choice is ipsilateral saphenous vein followed by contralateral saphenous vein and composite vein bypasses. Prosthetic grafts

Thrombosis in clinical practice

216

have very poor patency rates when used for femoral to tibial reconstructions. The use of adjunctive venous cuffs or composite prosthetic venous bypasses may improve the prosthetic graft patency but these grafts should be considered to be inherently compromised and durability should not be anticipated. Even in the best of hands, patency of femoral to tibial bypasses is rarely reported to be above 60% at 5 years. However, limb salvage rates are often much higher than this percentage and are routinely reported to be above 80% at 5 years. In more than 90% of patients, vascular reconstruction for limbthreatening ischemia will result in a reasonably durable graft with good limb salvage rates. Recommendations Healthy patients with TASC type C and D aortoiliac lesions and moderate to severe claudication can be considered for aortobifemoral bypass. In moderate-risk patients, the decision for aortic reconstruction should be delayed until the patient has evidence of severe claudication or limb-threatening ischemia. In high-risk patients with limbthreatening ischemia, extra anatomic bypass for aortoiliac disease is the safest option for limb salvage. In patients with infrainguinal disease, femoralpopliteal bypass with vein should be reserved for low-risk patients with severe claudication. Patients with mild to moderated claudication should not be considered for open surgical repair. All patients with limb-threatening ischemia should be considered for surgical arterial bypass for limb salvage.

Acute extremity arterial insufficiency The major causes of acute arterial insufficiency are arterial thrombosis, embolus, and trauma. Extreme vasospasm (e.g. ergot induced) and arterial dissection are unusual causes. Most traumatic occlusive events are associated with transection, laceration, or occlusion from external compression such as a fracture or dislocation, but in some instances, thrombosis occurs from blunt trauma. Iatrogenic vascular trauma, most often from diagnostic and therapeutic catheter placement is a common cause of acute arterial occlusion. In most patients, early surgery is required, with appropriate repair of the injured vessel. If thrombosis occurred, use of the Fogarty balloon catheter to remove thrombi is often required and is usually effective. Anticoagulation with heparin is variably used at the time of operation, but may be contraindicated because of other injuries. Outcome is related to the seriousness of associated injuries and duration of ischemia; successful vascular repair can be achieved in most cases. Nontraumatic acute occlusion is mainly embolic or thrombotic. The large majority of emboli arise from the heart in patients with valvular disease and/or atrial fibrillation, with prosthetic valves, or with mural thrombi in an infarcted or dilated left ventricle. Noncardiac sources of embolism include arterial aneurysms, ulcerated atherosclerotic plaque, recent (endo)vascular procedures, paradoxic emboli from venous thrombi, and rarely arteritis or vascular trauma. Approximately twothirds of noncerebral emboli enter vessels of the lower extremity and half of these obstruct the iliofemoral segment while

Peripheral arterial disease

217

the remainder involve the popliteal and tibial vessels. The upper extremity and renal plus visceral vessels each receive approximately 15% of emboli [122, 123]. Thrombotic occlusions of arteries are usually associated with advanced atherosclerosis, and arteries often have preexisting and developed collateral blood supply. For this reason, final occlusion may not be a dramatic event and is sometimes silent; it is not an emergent process in many patients. Thrombosis also occurs in vascular grafts and with other degenerative or inflammatory diseases or with trauma. The upper extremity better tolerates arterial occlusion because of rich collateral blood supply: gangrene or ischemic rest pain is rare in the absence of distal embolization. Hypovolemia, hyperviscosity, and hypercoagulability as observed in shock, thrombocytosis, polycythemia, and malignant disorders predispose to thrombotic arterial occlusion. Arterial thrombosis most frequently involves the lower extremities. Therapeutic management will depend on whether the occlusion is caused by embolism in a healthy artery versus thromboembolism in an atheromatous artery. Prompt embolectomy through surgical intervention is the usual technique to remove emboli from healthy arteries. The introduction of the Fogarty balloon catheter 40 years ago dramatically decreased the mortality and the amputation rate from arterial embolism. Percutaneous thromboembolectomy with the aid of an aspiration catheter or of a thrombectomy device is a recent alternative. Literature on either of these new techniques is descriptive and was recently reviewed [124, 125], To our knowledge, no randomized comparison between the different options is available. Traditionally, thromboembolism in a severely diseased artery or in a vascular graft causing acute ischemia symptoms has been the domain of the vascular surgeon as well, but optimal management needs to be determined. Heparin Patients presenting with acute limb-ischemia secondary to thromboembolic arterial occlusion usually receive prompt anticoagulation with therapeutic dosages of heparin in order to prevent clot propagation and to obviate further embolism. The logic of this common clinical practice is not questioned, even though no formal studies have established unequivocally a beneficial role of any antithrombotic agent in patients with acute embolic occlusion. The expected adverse effect of perioperative anticoagulant therapy is an increased risk of wound complications, particularly hematomas. The major role for continued anticoagulant therapy (heparin followed by oral anticoagulants) after embolization is to prevent embolic recurrence if the source of embolism cannot be eradicated or corrected. Thrombolysis Initial intervention with thrombolysis, with the aim to eliminate all thrombotic and embolic material and restore perfusion, is a potential alternative to surgical revascularization in acute limb-ischemia of thromboembolic origin. Systemic thrombolysis with intravenous administration of a thrombolytic agent was used in the 1960s and 1970s and has been completely abandoned and replaced by catheter-directed thrombolysis. With this technique, a catheter is positioned intra-arterially and advanced

Thrombosis in clinical practice

218

into the thrombus for local delivery of the thrombolytic agent. Several infusion methods can be used. Initially, streptokinase was the most widely used agent, but later it was superseded in clinical use by urokinase and recombinant tissue-type plasminogen activator (rt-PA). Dosage schemes vary considerably; an overview of reported dosages was published in a recent consensus document [126]. Although the extensive literature on catheter-directed thrombolysis is largely descriptive, five prospective randomized studies compared this treatment method to surgical intervention. In a small trial, surgical thrombectomy was compared to an intraarterial continuous infusion of 30 mg alteplase tissue plasminogen activator (TPA) over 3 hours in 20 patients with acute (>24 h but <14 days) arterial occlusion and severe leg ischemia [127]. Only patients with a need for intervention were included. Considerable lysis was obtained in 6 out of 9 patients treated with alteplase and half of them subsequently had a PTA. Two early reocclusions occurred. Thrombectomy also resulted in an immediate restitution of blood flow in 6 out of 9 cases [127]. Ouriel et al. compared initial thrombolysis complemented with PTA or/and surgery versus immediate surgery in 114 patients with limb-threatening ischemia of less than 7 days duration, due to native artery or graft occlusion [128]. Thrombolysis resulted in dissolution of the occluding thrombus in 70% of the patients. Limb salvage rate was similar in the two groups (82% at 1 year) but cumulative survival was significantly improved in patients randomized to thrombolysis due to fewer cardiopulmonary complications in hospital (84% vs 58% at 1 year, p=0.01) [128]. The STILE trial randomized 393 patients with nonembolic native artery or bypass graft occlusion in the lower limbs within the past 6 months to either optimal surgical procedure or intra-arterial catheter-directed thrombolysis with rt-PA or urokinase [129]. The primary endpoint was a composite outcome of death, major amputation, ongoing or recurrent ischemia, and major morbidity. At 1 month, the primary end point was reached for 36.1% of surgical patients and 61.7% of thrombolysis patients (p<0.0001). This difference was primarily due to ongoing/recurrent ischemia (25.7% vs 54.0%, p<0.0001); lysis was unsuccessful in 28% of the patients assigned to thrombolysis because of failure of proper catheter placement, an inexplicably high rate. However, in a secondary analysis which stratified patients by duration of ischemia, thrombolysis resulted in improved amputation-free survival at 6 months and shorter hospital stay in patients with acutely ischemic limbs (<14 days) whereas surgical revascularization was more effective for more chronic ischemia (>14days) [129]. Two additional publications analyzed the STILE trial on an intention-to-treat basis for the 30 days, 6 months, and 1-year results in patients with native artery and graft occlusion separately [130, 131]. For 237 patients with native artery occlusion, the composite clinical outcome was in favor of surgery because of a lower incidence of major amputation (0% vs 10% at 1 year, p=0.0024) and recurrent ischemia (35% vs 64% at 1 year, p<0.0001). Factors predictive of a poor outcome with lysis were femoropopliteal occlusion, diabetes, and critical ischemia. Only 20% of those patients had an onset or progression of ischemic symptoms of less than 14 days in duration: in these patients, the 1-year death/amputation rate was similar for surgery and thrombolysis. Overall, lysis failed to reestablish patency in 45% of patients, but 22% did not receive lytic agent because of problems with catheter positioning [129]. For 124 patients with bypass graft occlusion, there was also a better overall composite clinical outcome at 30 days and 1

Peripheral arterial disease

219

year in the surgical group compared to lysis, predominantly due to a reduction in ongoing/recurrent ischemia. However, 39% randomized to lysis failed catheter placement and required surgery. Following successful catheter placement, patency was reestablished by lysis in 84%. A post-study analysis indicated that limb loss at 1 year was significantly lower in patients with ischemia for less than 14 days if treated with thrombolysis compared with those treated surgically (20% vs 48%, p=0.026) [131]. The TOPAS investigators compared prospectively recombinant urokinase versus surgery in acute arterial occlusion (≤14 days) [131, 132]. In a first dose-ranging trial, they evaluated the safety and efficacy of three doses of recombinant urokinase in comparison with surgery in 213 patients [132]. The amputation-free survival rate at 1 year was 75% in 52 patients treated initially with recombinant urokinase at 4,000 IU/min and 65% in 58 surgically treated patients, a nonsignificant difference. The 4,000 IU/min dosage appeared the most appropriate thrombolytic regimen (compared with 2,000 and 6,000 IU/min) for the first 4 hours because it maximized lytic efficacy against the bleeding risk. This optimal dosage regimen (4,000 IU/min for the initial 4 h followed by 2,000 IU/min for up to 48 h) was next tested in a large multicenter trial on 544 patients [133]. Amputation-free survival rates in the urokinase group were 71.8% at 6 months and 65.0% at 1 year, as compared with respective rates of 74.8% and 69.9% in the surgery group; these differences between the two groups were not significant. Thrombolysis reduced the need for open surgical procedures (315 vs 551 at 6 months) without increased risk of amputation or death. Overall, the randomized trials provide no clear-cut answer to the dilemma: which of the two treatments (thrombolysis or surgical intervention) to prefer? They selected heterogeneous patient populations and studied complicated end points. Two metaanalyses are available and conclude that there is a similar mortality and amputation rate for thrombolysis and surgery; thrombolysis reduces the need for open major surgical procedures but causes more bleeding and distal embolization [134, 135]. The risk of intracranial bleeding remains a major burden for thrombolytic treatment in acute limbischemia: in three American prospective randomized studies which compared thrombolysis to surgery, the intracranial bleeding rate with thrombolysis was 1.2% (STILE), 2.1% (TOPAS-I), and 1.6% (TOPASII) [129, 132, 133]. A working party reached a consensus proposal on the use of thrombolysis in the management of lower limb arterial occlusion [126]. In native artery occlusion, a management strategy incorporating thrombolysis followed by correction of the causative lesion was proposed as an appropriate strategy in patients with ischemia of less than 14 days duration. Immediate surgical revascularization is to be preferred if thrombolysis would lead to an unacceptable delay in effective reperfusion. In patients with irreversible ischemia, primary amputation is indicated. For occluded bypass grafts, the therapeutic options are either surgical revision and thrombectomy, catheter-directed thrombolysis or insertion of a new graft. Factors to consider in therapeutic decision-making are the age and nature of the graft, the duration and degree of ischemia, and the availability of vein for a new distal bypass. In patients with a recent occlusion of a well-established graft, the working party proposed thrombolytic therapy as a primary treatment modality. Thrombolysis may eventually clear the thrombosed outflow vessels as well. However, the patency rate 1 year after successful lysis of thrombosed grafts is low (±20%) and the

Thrombosis in clinical practice

220

question is whether the ultimate yield justifies the laborintensive and expensive lytic procedure [136],

Conclusions Over the ensuing decades, the proportion of Americans over the age of 65 will continue to increase. With the increasing numbers will come increases in the incidence of PVD. The medical community must meet this challenge and ensure that patients with PVD are appropriately treated with prevention in mind. The risk to the extremity is low but concurrent risk to the cardiovascular and cerebrovascular systems is quite high. Except in the minority of patients with contraindications, all vascular patients should be treated with aspirin, betablockers, statins, and ACE inhibitors to reduce cardiac and stroke risk. Only a minority of patients will require operative intervention for PVD. Most can be managed with medical risk factor modification and exercise therapy. Pharmacologic therapy for claudication should only be offered after failure of medical therapy. Surgery, including angioplasty, should be reserved for patients with extremes of the disease.

References [1] Meijer WT, Hoes AW, Rutgers D, Bots ML, Hofman A, Grobbee DE. Peripheral arterial disease in the elderly: The Rotterdam Study. Arterioscler Thromb Vasc Biol 1998; 18:185–92. [2] Reunanen A, Takkunen H, Aromaa A. Prevalence of intermittent claudication and its effect on mortality. Acta Med Scand 1982; 211:249–56. [3] Skau T, Jonsson B. Prevalence of symptomatic leg ischaemia in a Swedish community—an epidemiological study. Eur J Vasc Surg 1993; 7:432–7. [4] Criqui MH, Fronek A, Barrett-Connor E, Klauber MR, Gabriel S, Goodman D. The prevalence of peripheral arterial disease in a defined population. Circulation 1985; 71:510–15. [5] Cox GS, Hertzer NR, Young JR, O’Hara PJ, Krajewski LP, Piedmonte MR, et al. Nonoperative treatment of superficial femoral artery disease: long-term follow-up. J Vasc Surg 1993; 17:172– 81. [6] Aquino R, Johnnides C, Makaroun M, Whittle JC, Muluk VS, Kelley ME, et al. Natural history of claudication: long-term serial follow-up study of 1244 claudicants. J Vasc Surg 2001; 34:962–70. [7] Meijer WT, Grobbee DE, Hunink MG, Hofman A, Hoes AW. Determinants of peripheral arterial disease in the elderly: the Rotterdam study. Arch Intern Med 2000; 160:2934–8. [8] Kannel WB, McGee DL. Update on some epidemiologic features of intermittent claudication: the Framingham Study. JAm Geriatr Soc 1985; 33: 13–18. [9] Price JF, Mowbray PI, Lee AJ, Rumley A, Lowe GD, Fowkes FG. Relationship between smoking and cardiovascular risk factors in the development of peripheral arterial disease and coronary artery disease: Edinburgh Artery Study. Eur Heart J 1999; 20:344–53. [10] Fowkes FG. Epidemiology of peripheral vascular disease. Atherosclerosis 1997; 131(Suppl.):S29-S31. [11] Planas A, Clara A, Marrugat J, Pou JM, Gasol A, de Moner A, et al. Age at onset of smoking is an independent risk factor in peripheral artery disease development. JVasc Surg 2002; 35:506–9.

Peripheral arterial disease

221

[12] Fowkes FG, Housley E, Riemersma RA, Macintyre CC, Cawood EH, Prescott RJ, et al. Smoking, lipids, glucose intolerance, and blood pressure as risk factors for peripheral atherosclerosis compared with ischemic heart disease in the Edinburgh Artery Study. Am J Epidemiol 1992; 135:331–40. [13] Ridker PM, Stampfer MJ, Rifai N. Novel risk factors for systemic atherosclerosis: a comparison of C-reactive protein, fibrinogen, homocysteine, lipoprotein (a), and standard cholesterol screening as predictors of peripheral arterial disease. JAMA 2001; 285:2481–5. [14] Valentine RJ, Verstraete R, Clagett GP, Cohen JC. Premature cardiovascular disease is common in relatives of patients with premature peripheral atherosclerosis. Arch Intern Med 2000; 160:1343–8. [15] DePalma RG, Hayes VW, Cafferata HT, Mohammadpour HA, Chow BK, Zacharski LR, et al. Cytokine signatures in atherosclerotic claudicants. J Surg Res 2003; 111:215–21. [16] Fiotti N, Giansante C, Ponte E, Delbello C, Calabrese S, Zacchi T, et al. Atherosclerosis and inflammation. Patterns of cytokine regulation in patients with peripheral arterial disease. Atherosclerosis 1999; 145:51–60. [17] Ross R. Atherosclerosis—an inflammatory disease. N Engl J Med 1999; 340:115–26. [18] Libby P. Changing concepts of atherogenesis. JIntern Med 2000; 247:349–58. [19] Cybulsky MI, Gimbrone MA Jr. Endothelial expression of a mononuclear leukocyte adhesion molecule during atherogenesis. Science 1991; 251: 788–91. [20] Iiyama K, Hajra L, Iiyama M, Li H, DiChiara M, Medoff BD, et al. Patterns of vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 expression in rabbit and mouse atherosclerotic lesions and at sites predisposed to lesion formation. Circ Res 1999; 85:199–207. [21] Yamawaki H, Lehoux S, Berk BC. Chronic physiological shear stress inhibits tumor necrosis factorinduced proinflammatory responses in rabbit aorta perfused exvivo. Circulation 2003; 108:1619–25. [22] Libby P. Inflammation in atherosclerosis. Nature 2002; 420:868–74. [23] Gu L, Okada Y, Clinton SK, Gerard C, Sukhova GK, Libby P, et al. Absence of monocyte chemoattractant protein-1 reduces atherosclerosis in low density lipoprotein receptor-deficient mice. Mol Cell 1998; 2:275–81. [24] Boring L, Gosling J, Cleary M, Charo IF. Decreased lesion formation in CCR2-/- mice reveals a role for chemokines in the initiation of atherosclerosis. Nature 1998; 394:894–7. [25] Beckman JA, Creager MA, Libby P. Diabetes and atherosclerosis: epidemiology, pathophysiology, and management. JAMA 2002; 287:2570–81. [26] Creager MA, Luscher TF, Cosentino F, Beckman JA. Diabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy: Part I. Circulation 2003; 108:1527–32. [27] Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature 1993; 362:801–9. [28] Smith JD, Trogan E, Ginsberg M, Grigaux C, Tian J, Miyata M. Decreased atherosclerosis in mice deficient in both macrophage colony-stimulating factor (op) and apolipoprotein E. Proc Natl Acad Sci USA 1995; 92:8264–8. [29] Cucina A, Borrelli V, Randone B, Coluccia P, Sapienza P, Cavallaro A. Vascular endothelial growth factor increases the migration and proliferation of smooth muscle cells through the mediation of growth factors released by endothelial cells. J Surg Res 2003; 109:16–23. [30] Welch GN, Loscalzo J. Homocysteine and atherothrombosis. N Engl J Med 1998; 338:1042– 50. [31] Homocysteine and risk of ischemic heart disease and stroke: a meta-analysis. JAMA 2002;288: 2015–22. [32] Bautista LE, Arenas IA, Penuela A, Martinez LX. Total plasma homocysteine level and risk of cardiovascular disease: a meta-analysis of prospective cohort studies. J Clin Epidemiol 2002; 55:882–7.

Thrombosis in clinical practice

222

[33] Hackam DG, Anand SS. Emerging risk factors for atherosclerotic vascular disease: a critical review of the evidence. JAMA 2003; 290:932–40. [34] Valentine RJ, Kaplan HS, Green R, Jacobsen DW, Myers SI, Clagett GP. Lipoprotein (a), homocysteine, and hypercoagulable states in young men with premature peripheral atherosclerosis: a prospective, controlled analysis. J Vasc Surg 1996; 23:53–61, discussion. [35] Spence JD, Norris J. Infection, inflammation, and atherosclerosis. Stroke 2003; 34:333–4. [36] Arora S, LoGerfo FW. Lower extremity macrovascular disease in diabetes. J Am Podiatr Med Assoc 1997; 87:327–31. [37] Panneton JM, Gloviczki P, Bower TC, Rhodes JM, Canton LG, Toomey BJ. Pedal bypass for limb salvage: impact of diabetes on long-term outcome. Ann Vasc Surg 2000; 14:640–7. [38] Marinelli MR, Beach KW, Glass MJ, Primozich JF, Strandness DE Jr. Noninvasive testing vs clinical evaluation of arterial disease. A prospective study. JAMA 1979; 241:2031–4. [39] Ouriel K. Peripheral arterial disease. Lancet 2001; 358:1257–64. [40] Schmieder FA, Comerota AJ. Intermittent claudication: magnitude of the problem, patient evaluation, and therapeutic strategies. Am J Cardiol 2001; 87:3D–13D. [41] Muluk SC, Muluk VS, Kelley ME, Whittle JC, Tierney JA, Webster MW, et al. Outcome events in patients with claudication: a 15-year study in 2777 patients. J Vasc Surg 2001; 33:251–7. [42] Cronenwett JL, Warner KG, Zelenock GB, Whitehouse WM Jr, Graham LM, Lindenauer M, et al. Intermittent claudication. Current results of non-operative management. Arch Surg 1984; 119:430–6. [43] Dormandy J, Heeck L, Vig S. The natural history of claudication: risk to life and limb. Semin Vasc Surg 1999; 12:123–37. [44] Bird CE, Criqui MH, Fronek A, Denenberg JO, Klauber MR, Langer RD. Quantitative and qualitative progression of peripheral arterial disease by non-invasive testing. Vasc Med 1999; 4:15–21. [45] Abbott RD, Brand FN, Kannel WB. Epidemiology of some peripheral arterial findings in diabetic men and women: experiences from the Framingham Study. Am J Med 1990; 88:376– 81. [46] Jude EB, Oyibo SO, Chalmers N, Boulton AJ. Peripheral arterial disease in diabetic and nondiabetic patients: a comparison of severity and outcome. Diabetes Care 2001; 24:1433–7. [47] Hertzer NR, Beven EG, Young JR, O’Hara PJ, Ruschhaupt WF, III, Graor RA, et al. Coronary artery disease in peripheral vascular patients. A classification of 1000 coronary angiograms and results of surgical management. Ann Surg 1984; 199:223–33. [48] Kozol RA, Bredenberg CE, Fey JD, Oates RP. Dependent rubor as a predictor of limb risk in patients with claudication. Arch Surg 1984;119: 932–5. [49] Bowers BL, Valentine RJ, Myers SI, Chervu A, Clagett GP. The natural history of patients with claudication with toe pressures of 40 mmHg or less. J Vasc Surg 1993; 18:506–11. [50] Fowl RJ, Gewirtz RJ, Love MC, Kempczinski RF. Natural history of claudicants with critical hemodynamic indices. Ann Vasc Surg 1992; 6:31–3. [51] Wolfe JH, Wyatt MG. Critical and subcritical ischaemia. Eur J Vasc Endovasc Surg 1997; 13: 578–82. [52] Vayssariat M, Gouny P, Cheynel C, Gaitz JP, Baudot N, Nussaume O. Haemodynamics of patients with severe lower limb arterial disease: the critical aspects of critical ischaemia. Eur J Vasc Endovasc Surg 1997; 14:284–9. [53] Bertele V, Roncaglioni MC, Pangrazzi J, Terzian E, Tognoni EG. Clinical outcome and its predictors in 1560 patients with critical leg ischaemia. Chronic Critical Leg Ischaemia Group. Eur J Vasc Endovasc Surg 1999;18:401–10. [54] Ubbink DT, Spincemaille GH, Reneman RS, Jacobs MJ. Prediction of imminent amputation in patients with non-reconstructible leg ischemia by means of microcirculatory investigations. J Vasc Surg 1999; 30:114–21.

Peripheral arterial disease

223

[55] Walker SR, Yusuf SW, Hopkinson BR. A 10-year follow-up of patients presenting with ischaemic rest pain of the lower limbs. Eur J Vasc Endovasc Surg 1998; 15:478–82. [56] Luther M. Surgical treatment of chronic critical leg ischaemia. A five-year follow-up of survival, mobility, and treatment level. Eur J Surg 1998; 164:35–43. [57] Dormandy J, Heeck L, Vig S. Major amputations: clinical patterns and predictors. Semin Vasc Surg 1999; 12:154–61. [58] Cruz CP, Eidt JF, Capps C, Kirtley L, Moursi MM. Major lower extremity amputations at a Veterans Affairs hospital. Am J Surg 2003; 186:449–54. [59] Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myo-cardial infarction, and stroke in high risk patients. BMJ 2002; 324:71–86. [60] A randomised, blinded, trial of clopidogrel versus aspirin in patients at risk of ischaemic events (CAPRIE). CAPRIE Steering Committee. Lancet 1996;348:1329–39. [61] Peters RJ, Mehta SR, Fox KA, Zhao F, Lewis BS, Kopecky SL, et al. Effects of aspirin dose when used alone or in combination with clopidogrel in patients with acute coronary syndromes: observations from the Clopidogrel in Unstable angina to prevent Recurrent Events (CURE) study. Circulation 2003; 108:1682–7. [62] Hess H, Mietaschk A, Deichsel G. Drug-induced inhibition of platelet function delays progression of peripheral occlusive arterial disease. A prospective double-blind arteriographically controlled trial. Lancet 1985; 1:415–19. [63] Libretti A, Catalano M. Treatment of claudication with dipyridamole and aspirin. Monogr Atheroscler 1986; 14:207–9. [64] Giansante C, Calabrese S, Fisicaro M, Fiotti N, Mitri E. Treatment of intermittent claudication with antiplatelet agents. J Int Med Res 1990; 18:400–7. [65] Girolami B, Bernardi E, Prins MH, ten Cate JW, Prandoni P, Hettiarachchi R, et al. Antithrombotic drugs in the primary medical management of intermittent claudication: a metaanalysis. Thromb Haemost 1999; 81:715–22. [66] Mangano DT, Layug EL, Wallace A, Tateo I. Effect of atenolol on mortality and cardiovascular morbidity after noncardiac surgery. Multicenter Study of Perioperative Ischemia Research Group. N Engl J Med 1996; 335:1713–20. [67] Poldermans D, Boersma E, Bax JJ, Thomson IR, Paelinck B, van de Ven LL, et al. Bisoprolol reduces cardiac death and myocardial infarction in high-risk patients as long as 2 years after successful major vascular surgery. Eur Heart J 2001; 22:1353–8. [68] Shammash JB, Trost JC, Gold JM, Berlin JA, Golden MA, Kimmel SE. Perioperative betablocker withdrawal and mortality in vascular surgical patients. Am Heart J 2001; 141:148– 53. [69] Radack K, Deck C. Beta-adrenergic blocker therapy does not worsen intermittent claudication in subjects with peripheral arterial disease. A metaanalysis of randomized controlled trials. Arch Intern Med 1991; 151:1769–76. [70] Aronow WS, Ahn C. Effect of beta blockers on incidence of new coronary events in older persons with prior myocardial infarction and symptomatic peripheral arterial disease. Am J Cardiol 2001; 87: 1284–6. [71] Selwyn AP, Kinlay S, Libby P, Ganz P. Atherogenic lipids, vascular dysfunction, and clinical signs of ischemic heart disease. Circulation 1997; 95:5–7. [72] Sacks FM, Pfeffer MA, Moye LA, Rouleau JL, Rutherford JD, Cole TG, et al. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. Cholesterol and Recurrent Events Trial investigators. N Engl J Med 1996; 335:1001–9. [73] Randomised trial of cholesterol lowering in 4,444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994; 344:1383–9. [74] Pedersen TR, Kjekshus J, Pyorala K, Olsson AG, Cook TJ, Musliner TA, et al. Effect of simvastatin on ischemic signs and symptoms in the Scandinavian Simvastatin Survival Study (4S). Am J Cardiol 1998; 81:333–5.

Thrombosis in clinical practice

224

[75] Leng GC, Price JF, Jepson RG. Lipid lowering therapy in the treatment of lower limb atherosclerosis. Eur J Vasc Endovasc Surg 1998; 16:5–6. [76] McDermott MM, Guralnik JM, Greenland P, Pearce WH, Criqui MH, Liu K, et al. Statin use and leg functioning in patients with and without lower-extremity peripheral arterial disease. Circulation 2003; 107:757–61. [77] Mondillo S, Ballo P, Barbati R, Guerrini F, Ammaturo T, Agricola E, et al. Effects of simvastatin on walking performance and symptoms of intermittent claudication in hypercholesterolemic patients with peripheral vascular disease. Am J Med 2003; 114:359–64. [78] Burns P, Lima E, Bradbury AW. Second best medical therapy. Eur J Vasc Endovasc Surg 2002;24:400–4. [79] Grundy SM. Approach to lipoprotein management in 2001 National Cholesterol Guidelines. Am J Cardiol 2002; 90:11i–21i. [80] Bell L, Madri JA. Influence of the angiotensin system on endothelial and smooth muscle cell migration. Am J Pathol 1990; 137:7–12. [81] Chobanian AV, Haudenschild CC, Nickerson C, Drago R. Antiatherogenic effect of captopril in the Watanabe heritable hyperlipidemic rabbit. Hypertension 1990; 15:327–31. [82] Chobanian AV. The effects of ACE inhibitors and other antihypertensive drugs on cardiovascular risk factors and atherogenesis. Clin Cardiol 1990; 13:VII43–VII48. [83] Pfeffer MA, Braunwald E, Moye LA, Basta L, Brown EJ Jr, Cuddy TE, et al. Effect of captopril on mortality and morbidity in patients with left ventricular dysfunction after myocardial infarction. Results of the survival and ventricular enlargement trial. The SAVE Investigators. N Engl J Med 1992; 327:669–77. [84] Yusuf S, Sleight P, Pogue J, Bosch J, Davies R, Dagenais G. Effects of an angiotensinconvertingenzyme inhibitor, ramipril, on cardiovascular events in high-risk patients. The Heart Outcomes Prevention Evaluation Study Investigators. N Engl J Med 2000; 342:145–53. [85] Faulkner KW, House AK, Castleden WM. The effect of cessation of smoking on the accumulative survival rates of patients with symptomatic peripheral vascular disease. Med J Aust 1983; 1:217–19. [86] Silagy C, Mant D, Fowler G, Lodge M. Metaanalysis on efficacy of nicotine replacement therapies in smoking cessation. Lancet 1994; 343: 39–42. [87] Hurt RD, Sachs DP, Glover ED, Offord KP, Johnston JA, Dale LC, et al. A comparison of sustained-release bupropion and placebo for smoking cessation. N Engl J Med 1997; 337:1195– 202. [88] Gardner AW, Poehlman ET. Exercise rehabilitation programs for the treatment of claudication pain. A meta-analysis. JAMA 1995; 274:975–80. [89] Leng GC, Fowler B, Ernst E. Exercise for intermittent claudication. Cochrane Database Syst Rev 2000; CD000990. [90] Johnson WC, Sentissi JM, Baldwin D, Hamilton J, Dion J. Treatment of claudication with pentoxifylline: are benefits related to improvement in viscosity? J Vasc Surg 1987; 6:211–16. [91] Angelkort B, Kiesewetter H. Influence of risk factors and coagulation phenomena on the fluidity of blood in chronic arterial occlusive disease. Scand J Clin Lab Invest Suppl 1981; 156:185–8. [92] Angelkort B, Maurin N, Boateng K. Influence of pentoxifylline on erythrocyte deformability in peripheral occlusive arterial disease. Curr Med Res Opin 1979; 6:255–8. [93] Bollinger A, Frei C. Double blind study of pentoxifylline against placebo in patients with intermittent claudication. Pharmatherapeutica 1977; 1:557–62. [94] Di Perri T, Guerrini M. Placebo controlled double blind study with pentoxifylline of walking performance in patients with intermittent claudication. Angiology 1983; 34:40–5. [95] Roekaerts F, Deleers L. Trental 400 in the treatment of intermittent claudication: results of long-term, placebo-controlled administration. Angiology 1984; 35:396–406.

Peripheral arterial disease

225

[96] Strano A, Davi G, Avellone G, Novo S, Pinto A. Double-blind, crossover study of the clinical efficacy and the hemorheological effects of pentoxifylline in patients with occlusive arterial disease of the lower limbs. Angiology 1984; 35:459–66. [97] Lindgarde F, Jelnes R, Bjorkman H, Adielsson G, Kjellstrom T, Palmquist I, et al. Conservative drug treatment in patients with moderately severe chronic occlusive peripheral arterial disease. Scandinavian Study Group. Circulation 1989;80: 1549–56. [98] Ciocon JO, Galindo-Ciocon D, Galindo DJ. A comparison between aspirin and pentoxifylline in relieving claudication due to peripheral vascular disease in the elderly. Angiology 1997; 48:237–40. [99] Porter JM, Cutler BS, Lee BY, Reich T, Reichle FA, Scogin JT, et al. Pentoxifylline efficacy in the treatment of intermittent claudication: multicenter controlled double-blind trial with objective assessment of chronic occlusive arterial disease patients. Am Heart J 1982; 104:66– 72. [100] Dettori AG, Pini M, Moratti A, Paolicelli M, Basevi P, Quintavalla R, et al. Acenocoumarol and pentoxifylline in intermittent claudication. A controlled clinical study. The APIC Study Group. Angiology 1989; 40:237–48. [101] Gallus AS, Gleadow F, Dupont P, Walsh J, Morley AA, Wenzel A, et al. Intermittent claudication: a double-blind crossover trial of pentoxifylline. Aust N Z J Med 1985; 15:402–9. [102] Perhoniemi V, Salmenkivi K, Sundberg S, Johnsson R, Gordin A. Effects of flunarizine and pentoxifylline on walking distance and blood rheology in claudication. Angiology 1984; 35:366–72. [103] Reilly DT, Quinton DN, Barrie WW. A controlled trial of pentoxifylline (Trental 400) in intermittent claudication: clinical, haemostatic and rheological effects. N Z Med J 1987; 100:445–7. [104] Tonak J, Knecht H, Groitl H. Treatment of circulation disorders with pentoxifylline. A doubleblind study with Trental. Med Monatsschr 1977; 31:467–72. [105] Radack K, Wyderski RJ. Conservative management of intermittent claudication. Ann Intern Med 1990; 113:135–46. [106] Hood SC, Moher D, Barber GG. Management of intermittent claudication with pentoxifylline: meta-analysis of randomized controlled trials. CMA J 1996; 155:1053–9. [107] Hiatt WR, Regensteiner JG, Hargarten ME, Wolfel EE, Brass EP. Benefit of exercise conditioning for patients with peripheral arterial disease. Circulation 1990;81:602–9. [108] Anand S, Creager M. Peripheral arterial disease. Clin Evid 2002; 79–90. [109] Thompson PD, Zimet R, Forbes WP, Zhang P. Meta-analysis of results from eight randomized, placebo-controlled trials on the effect of cilostazol on patients with intermittent claudication. Am J Cardiol 2002; 90:1314–19. [110] Regensteiner JG, Ware JE Jr, McCarthy WJ, Zhang P, Forbes WP, Heckman J, et al. Effect of cilostazol on treadmill walking, community-based walking ability, and health-related quality of life in patients with intermittent claudication due to peripheral arterial disease: meta-analysis of six randomized controlled trials. J Am Geriatr Soc 2002; 50:1939–46. [111] Douglas JS, Weintraub WS, Holmes D. Rationale and design of the randomized, multicenter, cilostazol for RESTenosis (CREST) trial. Clin Cardiol 2003; 26:451–4. [112] Dormandy JA, Rutherford RB. Management of peripheral arterial disease (PAD). TASC Working Group. TransAtlantic Inter-Society Concensus (TASC). J Vasc Surg 2000; 31:S1– S296. [113] Bosch JL, Hunink MG. Meta-analysis of theresults of percutaneous transluminal angioplasty and stent placement for aortoiliac occlusive disease. Radiology 1997; 204:87–96. [114] Tetteroo E, van der GY, Bosch JL, van Engelen AD, Hunink MG, Eikelboom BC, et al. Randomised comparison of primary stent placement versus primary angioplasty followed by selective stent placement in patients with iliac-artery occlusive disease. Dutch Iliac Stent Trial Study Group. Lancet 1998; 351:1153–9.

Thrombosis in clinical practice

226

[115] Kandarpa K, Becker GJ, Hunink MG, McNamara TO, Rundback JH, Trost DW, et al. Transcatheter interventions for the treatment of peripheral atherosclerotic lesions: part I. J Vasc Interv Radiol 2001; 12:683–95. [116] Hunink MG, Wong JB, Donaldson MC, Meyerovitz MF, Harrington DP. Patency results of percutaneous and surgical revascularization for femoropopliteal arterial disease. Med Decis Making 1994; 14:71–81. [117] Vroegindeweij D, Vos LD, Tielbeek AV, Buth J, vd Bosch HC. Balloon angioplasty combined with primary stenting versus balloon angioplasty alone in femoropopliteal obstructions: a comparative randomized study. Cardiovasc Intervent Radiol 1997; 20:420–5. [118] Timaran CH, Stevens SL, Freeman MB, Goldman MH. Infrainguinal arterial reconstructions in patients with aortoiliac occlusive disease: the influence of iliac stenting. J Vasc Surg 2001; 34:971–8. [119] Ali AT, Modrall JG, Lopez J, Brawley JG, Welborn MB, Clagett GP, et al. Emerging role of endovascular grafts in complex aortoiliac occlusive disease. J Vasc Surg 2003; 38:486–91. [120] Rzucidlo EM, Powell RJ, Zwolak RM, Fillinger MF, Walsh DB, Schermerhorn ML, et al. Early results of stent-grafting to treat diffuse aortoiliac occlusive disease. J Vasc Surg 2003; 37: 1175–80. [121] Jackson MR, Johnson WC, Williford WO, Valentine RJ, Clagett GP. The effect of anticoagulation therapy and graft selection on the ischemic consequences of femoropopliteal bypass graft occlusion: results from a multicenter randomized clinical trial. J Vasc Surg 2002; 35:292–8. [122] Elliott JP Jr, Hageman JH, Szilagyi E, Ramakrishnan V, Bravo JJ, Smith RE Arterial embolization: problems of source, multiplicity, recurrence, and delayed treatment. Surgery 1980; 88:833–45. [123] Abbott WM, Maloney RD, McCabe CC, Lee CE, Wirthlin LS. Arterial embolism: a 44 year perspective. Am J Surg 1982; 143:460–4. [124] Kasirajan K, Haskal ZJ, Ouriel K. The use of mechanical thrombectomy devices in the management of acute peripheral arterial occlusive disease. J Vasc Interv Radiol 2001; 12:405– 11. [125] Morgan R, Belli AM. Percutaneous thrombectomy: a review. Eur Radiol 2002; 12:205–17. [126] Thrombolysis in the management of lower limb peripheral arterial occlusion—a consensus document. Working Party on Thrombolysis in the Management of Limb Ischemia. Am J Cardiol 1998; 81:207–18. [127] Nilsson L, Albrechtsson U, Jonung T, Ribbe E, Thorvinger B, Thorne J, et al. Surgical treatment versus thrombolysis in acute arterial occlusion: a randomised controlled study. Eur J Vasc Surg 1992; 6:189–93. [128] Ouriel K, Shortell CK, DeWeese JA, Green RM, Francis CW, Azodo MV, et al. A comparison of thrombolytic therapy with operative revascularization in the initial treatment of acute peripheral arterial ischemia. J Vasc Surg 1994;19: 1021–30. [129] Results of a prospective randomized trial evaluating surgery versus thrombolysis for ischemia of the lower extremity. The STILE trial. Ann Surg 1994; 220:251–66. [130] Weaver FA, Comerota AJ, Youngblood M, Froehlich J, Hosking JD, Papanicolaou G. Surgical revascularization versus thrombolysis for nonembolic lower extremity native artery occlusions: results of a prospective randomized trial. The STILE Investigators. Surgery versus Thrombolysis for Ischemia of the Lower Extremity. J Vasc Surg 1996; 24:513–21. [131] Comerota AJ, Weaver FA, Hosking JD, Froehlich J, Folander H, Sussman B, et al. Results of a prospective, randomized trial of surgery versus thrombolysis for occluded lower extremity bypass grafts. Am J Surg 1996; 172:105–12. [132] Ouriel K, Veith FJ, Sasahara AA. Thrombolysis or peripheral arterial surgery: phase I results. TOPAS Investigators. J Vasc Surg 1996; 23:64–73.

Peripheral arterial disease

227

[133] Ouriel K, Veith FJ, Sasahara AA. A comparison of recombinant urokinase with vascular surgery as initial treatment for acute arterial occlusion of the legs. Thrombolysis or Peripheral Arterial Surgery (TOPAS) Investigators. N Engl J Med 1998; 338:1105–11. [134] Palfreyman SJ, Booth A, Michaels JA. A systematic review of intra-arterial thrombolytic therapy for lower-limb ischaemia. Eur J Vasc Endovasc Surg 2000; 19:143–57. [135] Berridge DC, Kessel D, Robertson I. Surgery versus thrombolysis for acute limb ischaemia: initial management. Cochrane Database Syst Rev 2002; CD002784. [136] Galland RB, Magee TR, Whitman B, Earnshaw JJ, Judge C, Hamilton G, et al. Patency following successful thrombolysis of occluded vascular grafts. Eur J Vasc Endovasc Surg 2001;22: 157–60.

11 Antithrombotic therapy for ischemic stroke Tarvinder S Dhanjal and Matthew Walters Introduction This chapter will discuss the provision of thrombolytic, antiplatelet, and anticoagulant therapy for ischemic stroke in both acute and secondary prevention settings, reviewing the results of published clinical trials and postmarketing experience. Stroke remains one of the leading causes of death and disability throughout the world. It is the third commonest cause of death in developed countries, exceeded only by coronary artery disease and cancer. The incidence of stroke is 1–2 cases in 1,000 people a year in the Western world, and is probably slightly higher among African-Caribbeans than other ethnic groups. Cerebrovascular disorders are uncommon in people aged less than 40 years but there is a definite increase with age, with an incidence of 10 cases in 1,000 people aged greater than 75 in a year [1, 2]. Stroke is slightly more common in men, but women tend to have a poorer prognosis because of a higher mean age at onset. The incidence of stroke has been declining in recent decades in many Western countries because of better population control of hypertension, smoking, and other risk factors. However, the absolute number of strokes continues to increase because of an aging population which is predicted to peak in 2015. Thus the present annual incidence of 700,000 strokes in the United States is expected to rise to 1 in 100,000 in 2015 and an increase of similar magnitude is predicted elsewhere in the developed world without further advances in prevention.

The rationale for acute stroke treatment Stroke has long been shrouded in therapeutic nihilism and treatment of stroke patients regarded as the domain of physiotherapy. Medical management was essentially supportive and involved treatment of common complications such as airway obstruction, respiratory failure, swallowing problems with the risk of aspiration, dehydration, malnutrition, venous thromboembolic complications, seizures, and infections [3–6]. Despite optimal management stroke patients are at significant risk of poor outcome, the case fatality rates after a first ever stroke are 12% at 7 days, 19% at 30 days, and 31% at 1 year. Hemorrhagic stroke carries a higher risk of death than ischemic stroke [7]. The advent of effective acute treatments for acute stroke has provided an opportunity to lessen the huge burden of stroke disease and reduce the numbers of people disabled or killed by this common, devastating condition. The stroke patient is no longer beyond the help of medical science: there is an urgent need to dispel the obsolete perception of stroke

Thrombosis in clinical practice

228

as a hopeless condition and replace it with a more vigorous and aggressive approach. A wide variety of antithrombotic strategies have been evaluated in the stroke patient, ranging from acute thrombolytic treatment to secondary prevention. A wealth of evidence supports some of these interventions, whereas others remain controversial and require further study. We will consider these strategies in turn and review available evidence concerning their use. Before doing so, the main pathophysiological features of cerebral ischemia will be briefly summarized.

Pathophysiology “Time is brain” following cerebral artery occlusion, irreversible ischemic brain damage evolves over hours. In experimental models of stroke, reperfusion can rescue tissue which is functionally inactive but still viable. Ischemic stroke usually occurs as a consequence of occlusion of a cerebral artery, or less often a reduction in perfusion distal to a severe stenosis. As cerebral blood flow falls, neuronal function is affected in two stages. Initially, as blood flow falls below a critical threshold of about 20 ml blood/100 g brain/min (normal being over 50 ml/100 g/min), loss of neuronal electrical function occurs. Crucially, this is a potentially reversible stage. Irreversible damage occurs within minutes as blood flow falls below a second critical threshold of 10 ml/100 g/min; below this level, aerobic mitochondrial metabolism fails, and the inefficient anaerobic metabolism of glucose takes over, rapidly leading to lactic acidosis. Consequently, the normal energy-dependent cellular ion homeostasis fails, resulting in potassium egress, sodium, and water entering the cell, and consequent cytotoxic edema [8]. Calcium also enters the cell, exacerbating mitochondrial failure [9] and promoting neuronal death. The relationship between duration of ischemia, severity of ischemia, and tissue outcome is shown in Figure 11.1. From the identification of hypoperfused but salvageable brain tissue has emerged the concept of the “ischemic penumbra”—that is, an area of

Figure 11.1 Relationship between degree of ischemia, time, and tissue outcome [10].

Antithrombotic therapy for ischemic stroke

229

brain which has reached the reversible stage of electrical failure, but has not yet passed onto the second irreversible stage of cellular homeostatic failure. In theory, therefore, this tissue could be “rescued,” either by early reperfusion (using clot-busting drugs) or by administration of agents which retard or reverse the cascade of events which leads to ischemic cell death. Although there is evidence that the concept of the ischemic penumbra is valid, it remains unclear how long the ischemic human brain might survive—in other words the time window for intervention is unclear [11]. Recent clinical trials of thrombolysis have attempted to answer this question; these shall be discussed in the next section.

Acute interventionthrombolysis Small, controlled trials of thrombolysis in stroke patients were initiated almost 40 years ago, but the therapy was discarded because of increased risk of death [12]. The development of computed tomography in the 1980s enabled better patient selection and the issue of thrombolytic treatment for stroke was revisited in a number of studies. As a consequence of these trials, thrombolysis is now established management for selected patients with acute ischemic stroke in the United States. More recently, thrombolysis has been licensed in Europe, with continuing clinical evaluation and assessment. Thrombolysis has been used for a number of years in the management of patients with myocardial infarction and its efficacy has been demonstrated by a number of large randomized, placebo-controlled trials [13–15]. Thrombolysis leads to reperfusion of vessels occluded by coronary atherothrombosis which in turn reduces mortality and preserves left ventricular function. Most strokes are atherothrombotic or embolic (85%) while the remainder are hemorrhagic. It was hoped that the same “open artery hypothesis” would hold true for patients with ischemic or nonhemorrhagic stroke that is early reperfusion would improve both mortality and functional outcome. The potential risks of thrombolysis are predictable from what is already known of the pathophysiology of stroke. The main risk of thrombolytic treatment is bleeding, either into the infarcted brain tissue or at another site. Certain groups of patients could be thought of as at higher risk of secondary intracerebral hemorrhage as a result of thrombolysis. Elderly patients already have a higher risk of spontaneous intracerebral hemorrhage, as have hypertensives and patients with embolic rather than occlusive atherothrombotic strokes. Delayed thrombolysis may lead to reperfusion of already irreversibly infarcted tissue and increase the risk of hemorrhage. A further risk is reperfusion injury, which may occur when toxic freeradicals and inflammatory cells enter an ischemic area following spontaneous reperfusion. When trials of thrombolytics for stroke were devised these issues were considered and are reflected in the design, entry, and exclusion criteria of the studies. All studies incorporated a variable “time window” with which patients were ineligible, all patients underwent CT scanning to exclude intracerebral hemorrhage as the primary pathology, and some studies excluded large infarcts and hypertensive patients. The analysis of treatment efficacy is crucial in the design and interpretation of any trial of stroke therapy. Most studies incorporated a validated functional outcome score, for example the Barthel or Rankin score [16]. These scores describe the patients’ ability to

Thrombosis in clinical practice

230

perform activities of daily living. More detailed assessments of neurological function were also made using, for example, the Scandinavian Stroke Scale [17].

Streptokinase studies All three trials (MAST-I, ASK, MAST-E) were halted on recommendation of the Data Monitoring Committees because of unacceptable risks [18–20]. The MAST-E (Multicentre Acute Stroke Study-Europe) and ASK (Australian Stroke Study) treated patients with streptokinase (1.5 M units over 1 h) or placebo within 6 and 4 hours, respectively. In both studies there was an increase in early mortality sufficient to result in the safety committees abandoning both studies prior to completion. This correlated with a high incidence of complicating intracerebral hemorrhage. In the MAST-E study 36% (n=156) patients treated with streptokinase versus 3% (n=154) placebo patients suffered a fatal hemorrhagic transformation of infarct [18]. The MAST-I (MAST-Italy) study evaluated treatment with streptokinase and aspirin compared with streptokinase or aspirin alone or placebo. Again the results suggested an increase in early mortality which was more marked in patients receiving both streptokinase and aspirin, while aspirin alone appeared to be safe but of no clear benefit [20]. Meta-analysis of all streptokinase results failed to reveal factors which predisposed to early mortality. Streptokinase significantly increased early mortality while there may have been a very weak trend towards improved outcome in survivors. To date, no randomized controlled trial has supported the use of intravenous streptokinase as therapy for acute ischemic stroke. Further studies of streptokinase using different patient selection criteria, lower doses of thrombolytic, and prohibition of antiplatelet or anticoagulant coadministration have been proposed; however, at present, no such trial is underway. Although there remains some debate, it is highly unlikely that streptokinase will have any role in the management of acute ischemic stroke.

Recombinant tissue plasminogen activator studies Five phase III trials have evaluated the role of recombinant tissue plasminogen activator (rt-PA) in patients with acute stroke [21–25]. Approval of this treatment by the FDA in 1996 was based on the results of the National Institute of Neurological Disorders and Stroke (NINDS) rt-PA Stroke Study. There are theoretical reasons why rt-PA may be more effective: first, rt-PA is known to be more “clot specific,” that is, it causes less generalized activation of plasminogen, instead having a more selective action at the site of the clot itself; this may make hemorrhagic complications less likely. Second, rt-PA is less antigenic than streptokinase and is not associated with a fall in blood pressure during the infusion. This may be of significance as reducing blood pressure immediately after ischemic stroke has been associated with worse outcome [26]. Finally, angiographic studies in patients with acute myocardial infarction suggest rt-PA may be more effective than streptokinase in reperfusing occluded coronary vessels [27].

Antithrombotic therapy for ischemic stroke

231

The ECASS-1 (European Co-operative Acute Stroke Study), the NINDS, and ECASS2 studies evaluated rt-PA at doses of 1.1 mg/kg, 0.9 mg/kg, and 0.9 mg/kg, respectively. Patients were treated within 6 hours in the ECASS studies and 3 hours in the NINDS study. The Alteplase Thrombolysis for Acute Noninterventional Therapy in Ischemic Stroke (ATLANTIS) study used a dose of 0.9 mg/kg administered within an initial time window of 6 hours. The time window was altered twice during the study. European Cooperative Acute Stroke Study (ECASS) [21] ECASS enrolled persons within 6 h of stroke; however, most persons were treated more than 3 hours after stroke. The higher dose of rt-PA (1.1 mg/kg) used in the ECASS-1 study is equivwith myocardial infarction. A bolus of 10% of alent to that used in the treatment of patients the total dose was given over 1–2 minutes and the remainder was administered during the next hour. Dose-ranging studies have shown an increasing incidence of complicating hematoma formation in patients receiving doses greater than 0.85 mg/kg, and this could have contributed to the increased incidence of hemorrhages seen in the ECASS-1 study. The results of ECASS-1 are also notable for the large proportion of patients excluded from the target population analysis (109 out of 620 randomized). The protocol intended to exclude patients with greater than one-third middle cerebral artery (MCA) territory stroke on CT, that is, those with large infarcts, persons with evidence on computed tomography of a major infarction including diffuse swelling of the hemisphere and parenchymal hypodensity. Sixty-six such patients were randomized, constituting the largest group of protocol violators. Survival in protocol violators was significantly worse than for those meeting the entry criteria. At 3 months, the primary end points (a 15-point difference on the Barthel Index and a 1-point difference on the Modified Rankin scale) showed no significant difference between the rt-PA-treated and placebo-treated patients. However, rt-PA was associated with an increased rate of recovery and a shorter hospital stay. Mortality in the rt-PA versus placebo group was nonsignificantly higher at 30 days (17.9% vs 12.7%, p=0.08) but significantly higher at 3 months (22.4% vs 15.8%, p=0.04). Large parenchymal hemorrhages (Figure 11.2) were increased 3-fold with the rt-PA group (6.3% vs 2.4%, p=0.001). In summary, the ECASS study did not show enough benefit to justify thrombolysis with rt-PA up to 6 hours after stroke onset although an improvement in functional outcome (Barthel and Rankin score) was seen in the target population analysis of the rt-PA receiving patients. National Institute of Neurological Disorders and Stroke Study (NINDS) [22] The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study was

Thrombosis in clinical practice

232

Figure 11.2 CT scan of brain showing large parenchymal intracerebral hemorrhage. a randomized, double-blind, placebo-controlled trial consisting of two parts. In part 1, 291 persons were enrolled in a project that assessed early response to treatment, although data about long-term outcome were also collected. In part 2, 333 persons were recruited into a study to examine the effects of treatment at 3 months. Because the results of part 1 were not known before part 2 was completed, combined data were reported. Persons with ischemic stroke in either the carotid or vertebrobasilar circulation and a wide range of severity of signs were treated within 3 hours of onset of CT confirmed ischemic stroke. Patients with isolated neurological deficits such as ataxia alone, sensory loss alone, dysarthria alone, or minimal weakness that could not be assessed by the NIH Stroke Scale (NIHSS) were not enrolled. Persons with rapidly resolving neurological symptoms were also excluded. Overall, 302 persons were treated within 90 minutes of onset of stroke. Persons assigned to active treatment received rt-PA in a dose of 0.9 mg/kg up to a maximum of 90 mg; 10% of the dose was given as a bolus and the remainder was infused over 60 minutes. Blood pressure was managed closely and no anticoagulants or antiplatelet drugs were given within 24 hours of treatment. At 3 months, 50% of rt-PA treated patients had no or minimal disability compared with 38% of the controls (Figure 11.3). The odds ratio for

Antithrombotic therapy for ischemic stroke

233

Figure 11.3 Outcome (Barthel score) of patients in NINDS trial. favorable outcome with rt-PA was 1.7 (95% confidence interval (CI), 1.2–2.6). The symptomatic brain hemorrhage rate was exceptionally small in the placebo group (0.6%) but was increased 10-fold with rt-PA (6.4%, p=0.001). Despite this, mortality at 3 months was comparable between groups ((54 of 312, 17%) with rt-PA versus (64 of 312, 21%) those with placebo (p=0.3)). Second European Cooperative Acute Stroke Study (ECASS-2) [23] In the ECASS-2 trial, 800 patients were assigned randomly to treatment with either rt-PA or placebo. The aim of the ECASS-2 study was to investigate the safety and efficacy of rt-PA given in a dose of 0.9 mg/kg within 6 hours of onset of ischemic stroke. Strenuous efforts to improve the quality of CT interpretation prior to randomization were made, and significantly fewer intracranial hemorrhages were seen in comparison with the original ECASS study. The primary outcome measure was the proportion of patients reaching modified Rankin scores of 0 or 1 (i.e. little or no disability) at 3 months [28]. The study failed to demonstrate a statistically significant difference between treated and placebo groups. A post hoc analysis of the ECASS-2 data demonstrated a significant difference between the rtPA treated group and the placebo group when the Rankin score was dichotomized into 0–2 and greater than 2. In this analysis, 54.3% of the rt-PA patients achieved a 90-day outcome of 0–2, whereas only 46% of placebo patients had a Rankin score of 0–2. The authors concluded that rt-PA treatment leads to a clinically relevant improvement in outcome without increased morbidity and mortality despite increased symptomatic hemorrhage. However, the failure of the trial to demonstrate efficacy

Thrombosis in clinical practice

234

Figure 11.4 Meta-analysis of rt-PA trials in stroke. using predetermined end points has led to the widespread interpretation of the trial as neutral. Incorporation of the ECASS-2 data into a meta-analysis of trials of rt-PA in acute stroke reveals a favorable odds ratio of 0.67 (95% CI 0.56–0.80) with respect to death and disability (Figure 11.4). Alteplase Thrombolysis for Acute Noninterventional Therapy in Ischemic Stroke (ATLANTIS) [24] The ATLANTIS study was designed to assess the safety and efficacy of intravenous rtPA 0.9 mg/kg initiated within 6 hours of ischemic stroke onset. Two years after recruitment commenced, the time window was changed to 0–5 h because of adverse interim safety analysis in the 5–6 h group. After a further 3 years. the time window was further modified (3–5 h) due to the results of the NINDS study. With the exception of the time window, entry criteria were very similar to those employed by the NINDS investigators. The primary outcome measure was the percentage of patients achieving an NIHSS of 0 or 1 at 90 days. This study included 547 patients and the primary end point was almost identical in the two groups. The median baseline NIHSS was 11 in the two groups and a 90 day modified Rankin of 0–1 was achieved in 42% of the rt-PA group and 40% of the placebo patients. The ATLANTIS study was terminated prematurely in July 1998 following further interim safety analysis which concluded that “treatment was unlikely to prove beneficial.” The trial has been considered a negative one, and it is understood that the use of intravenous rt-PA beyond 3 hours after stroke onset is not supported by this study. One positive result from the ECASS-2 and ATLANTIS trials was that the rate of symptomatic intracerebral hemorrhage was 8.8% and 7.0%, respectively, not greatly increased from the 6.4% rate seen with rt-PA in the 3-h window NINDS trial.

Antithrombotic therapy for ischemic stroke

235

Patient selection It is difficult to determine which patients are most likely to benefit from intravenous rtPA therapy and which patients are at risk particularly when time limits each patient interaction. Failure to fully understand this issue constitutes Characteristics of patients with ischemic stroke who could be treated with rt-PA Diagnosis of ischemic stroke causing measurable neurological deficit. The neurological signs should not be clearing spontaneously. The neuroiogical signs should not be minor and isolated. Caution should be exercised in treating a patient with major deficits. The symptoms of stroke should not be suggestive of subarachnoid hemorrhage. Onset of symptoms <3 h before beginning treatment. No head trauma or prior stroke in previous 3 months. No myocardial infarction in the previous 3 months. No gastrointestinal or urinary tract hemorrhage in previous 21 days. No major surgery in the previous 14 days. No arterial puncture at a noncompressible site in the previous 7 days. No history of previous intracranial hemorrhage. Blood pressure not elevated (systolic <185 mmHg and diastolic <110 mmHg). No evidence of active bleeding or acute trauma (fracture) on examination. Not taking an oral anticoagulant or if anticoagulant being taken, INR <1.5. If receiving heparin in previous 48 h, APTT must be in normal range. Platelet count >100,000/mm3. Blood glucose concentration >50 mg/dL (2.7 mmol/L). No seizure with postictal residual neurological impairments. CT does not show a multilobar infarction (hypodensity >1/3 cerebral hemisphere). The patient or family understand the potential risks and benefits from treatment. Regimen for treatment of acute ischemic stroke intravenous rt-PA Infuse 0.9 mg/kg (maximum of 90 mg) over 60 min with 10% of the dose given as a bolus dose over 1 minute. Admit the patient to an intensive care unit or a stroke unit for monitoring. Perform neurological assessments every 15 min during the infusion of rt-PA and every 30 min for the next 6 h and then every hour until 24 h from treatment.

Thrombosis in clinical practice

236

If the patient develops severe headache, acute hypertension, nausea, or vomiting, discontinue the infusion (if agent is still being administered) and obtain a CT scan of brain on an emergent basis. Measure blood pressure every 15 min for the first 2 h, every 30 min for the next 6 h, and then every hour until 24 h from treatment. Increase the frequency of blood pressure measurements if a systolic blood pressure >180 mmHg or diastolic blood pressure of >105 mmHg is recorded. Consider cautious administration of antihypertensive medications to maintain blood pressure at or below these levels. Delay placement of nasogastric tubes, indwelling bladder catheters, or intra-arterial pressure catheters. a major limitation to the implementation of rt-PA therapy. Clearly, there are several factors which are critical to selecting patients who will have the best risk: benefit ratio with intravenous rt-PA. The differences in the NINDS, ECASS, and ATLANTIS trials suggest that time to treatment is critical. If the 3-h time window of treatment can be met, the indications for intravenous rt-PA therapy supported by the NINDS Study are broad. The exclusions used in the NINDS Study should be considered contraindications for noninvestigational use of rtPA in persons with acute ischemic stroke. Time to treatment, dose of thrombolytic agent, blood pressure level, severity of neurological deficit, and severity of ischemia are risk factors for developing brain hemorrhage, mainly symptomatic bleeding into an evolving infarction [29, 30]. It is possible that the most severe hemorrhages tend to occur in rapidly worsening and enlarging ischemic areas with early tissue damage, so clinical deterioration may often be wrongly attributed to hemorrhage instead of evolving infarction. Nevertheless, there is an ill-defined relationship between elevated blood pressure and intracranial bleeding after the use of thrombolytic agents in treatment of acute ischemic stroke which has been addressed by recent American Stroke Association guidelines with the recommendation of regular blood pressure checks with the possibility of labetalol and nitrate infusions [31]. The decision for treatment with intravenous rt-PA (0.9 mg/kg, maximum dose 90 mg) is to be based on several features (see box text).

Intra-arterial delivery of thrombolytic agent Alternatives to intravenous delivery of thrombolytic drug have been explored. Intraarterial thrombolysis, although more invasive and less widely available, has the advantage of a higher local drug concentration and a lower systemic concentration. Another advantage is more accurate selection of patients with demonstration of an occluded brain artery, since angiographic data is acquired during the procedure. A prospective, randomized, placebo-controlled phase II study evaluated the utility of intraarterial administration of recombinant prourokinase (r-proUK) in combination with heparin and demonstrated that the combination was successful in achieving recanalization more frequently, but increased the risk of intracranial bleeding [32]. The results prompted a second randomized, controlled, multi-center trial testing the efficacy of intraarterial thrombolysis with r-proUK among patients with stroke of less than 6 h duration

Antithrombotic therapy for ischemic stroke

237

secondary to occlusion of the MCA [33]. Heparin was given to both patients who received r-proUK and those in the control group. In the primary intent-to-treat analysis, 40% of the 121 patients treated with r-proUK and 25% of the 59 control patients had modified Rankin score of 0–2 at 90 days (p=0.043). Recanalization of the MCA was achieved in 66% of the patients treated with r-proUK and 18% of the patients in the control group (p<0.001). Intracranial hemorrhage with neurological deterioration within 24 h of treatment occurred in 10% of patients treated with r-proUK and in 2% of the control group (p=0.06). There was no difference in overall mortality between the two groups. The FDA has not approved the drug, and r-proUK is not currently available for clinical use. The feasibility of combining intravenous and intraarterial rt-PA in treatment of ischemic stroke was examined in the Emergency Management of Stroke (EMS) Bridging Trial [34]. The study suggested that this strategy, which included early intravenous administration of rt-PA in a lower dose followed by arterial administration, could achieve recanalization and might be associated with a reasonable degree of safety. A trial testing the efficacy of combined intravenous and intra-arterial thrombolysis with rt-PA is now in progress. Physicians with expertise in endovascular therapy are using intra-arterial techniques to treat patients with acute ischemic stroke secondary to occlusion of large intracranial arteries including the basilar artery or the MCAs. Most centers that are performing intraarterial thrombolysis are using rt-PA although there are limited or no data demonstrating the efficacy or safety of the intra-arterial administration of this agent. Despite limited resources and the inherent delays with this form of acute therapy the American Stroke Association has recommended that intra-arterial thrombolysis is an option for treatment of selected patients with major stroke of less than 6 hours duration due to large vessel occlusions of the MCA. It should be recognized that intra-arterial thrombolysis is not FDA approved.

Antiplatelet treatment in acute stroke Aspirin is the only antiplatelet drug evaluated for the treatment of acute ischemic stroke and is recommended early in the management at a dose of 160–325 mg o.d. Two major randomized trials (the International Stroke Trial (IST) and the Chinese Acute Stroke Trial (CAST)) have shown that starting daily aspirin promptly (<48 h after the start rather than the end of the hospital stay) in patients with suspected acute ischemic stroke reduces the immediate risk of further stroke or death in hospital, and the overall risk of death and dependency 6 months later [35, 36]. The IST tested aspirin alone (300 mg o.d.) or in combination with 1 or 2 doses of subcutaneously administered heparin in comparison to heparin alone or control. The trial demonstrated a significant reduction in recurrent events by aspirin within the first 2 weeks, but acute mortality was not reduced. A modest but significant increase in serious systemic hemorrhages was noted with aspirin during the 14-day treatment period, and a small (0.1% absolute) significant increase in the incidence of intracranial hemorrhage was noted. At 6 months, patients assigned aspirin had a significantly lower incidence of death and dependency, but there was no significant improvement in the proportion of patients free from disability [35].

Thrombosis in clinical practice

238

The CAST tested aspirin, 160 mg o.d., in a randomized, placebo-controlled trial. A significant reduction in mortality and recurrent stroke was noted with aspirin during the 28 days of treatment. A modest but not significant increase in the risk of intracranial hemorrhage and a significant increase in systemic hemorrhage were found. At the time of discharge, mortality was significantly reduced with aspirin, but the rates of long-term complete recovery or death and disability were not significantly improved [36]. Thus, about 10 deaths or recurrent strokes are avoided in every 1,000 patients treated with aspirin in the first few weeks after an ischemic stroke. The benefit of aspirin is seen in a wide range of patients irrespective of age, sex, atrial fibrillation, blood pressure, stroke subtype, and computed tomographic findings. Thus, the two studies show that giving aspirin early in acute stroke is safe, although side effects should always be considered. Other trials have shown that continuing treatment with low-dose aspirin gives protection in the longer term. Until further evidence is available, however, aspirin should be withheld from patients receiving other forms of anticoagulant (except low-dose heparin (5,000 IU b.i.d.)) or thrombolytic treatment (and for 24 h after finishing treatment). The results of the IST and CAST studies apply chiefly to patients who had a computed tomography scan to exclude intracranial hemorrhage. Although no evidence of harm was demonstrated in patients who were treated with aspirin before CT scan was performed, brain imaging is recommended prior to initiation of antiplatelet treatment in all patients with suspected stroke. The American Stroke Association recommends the following [31]: 1. Aspirin should be given within 24–48 h of stroke onset in most patients. 2. The administration of aspirin as an adjunctive therapy, within 24 h of the use of thrombolytic agents, is not recommended. 3. Aspirin should not be used as a substitute for other acute interventions, especially intravenous administration of rt-PA, for the treatment of acute ischemic stroke. 4. No recommendation can be made about the urgent administration of other antiplatelet agents.

Anticoagulant therapy Unfractionated heparin Available data demonstrate no reduction in mortality or morbidity rates with unfractionated heparin given subcutaneously to patients with acute stroke [35, 37]. There is also no reliable evidence that dose-adjusted, intravenous, unfractionated heparin reduces the risk of early recurrent stroke. In the IST study, one-quarter of the patients were randomized to subcutaneous unfractionated heparin at 5,000 IU b.i.d. and another quarter to 12,500 IU b.i.d., whereas in the remaining half heparin was avoided. Thus, in the factorial design, half of the patients in each group either received aspirin or had it withheld. Randomized, controlled trials of dose-adjusted intravenous unfractionated heparin in acute stroke that specifically address long-term, stroke-related morbidity and mortality have not been reported. One study published in 1986 reported a prospective, double-blind trial of dose-adjusted intravenous unfractionated heparin in 225 patients with partial stable carotid and vertebrobasilar distribution stroke [38], This trial showed

Antithrombotic therapy for ischemic stroke

239

that there was no difference in death at 7 days between patients who were treated with unfractionated heparin (1/112 (0.89%)) and those treated with placebo (2/113 (1.77%)). Functional activity at 7 days, 3 months, and 1 year also was not significantly different between groups. At 6 months, the proportion of patients who were dead or dependent was identical for the group that received unfractionated heparin and the group that avoided heparin (62.9% in each). Mortality at 1 year was significantly increased in the unfractionated heparintreated group compared with the placebo group (heparin 17 vs control 8; p<0.05). Despite considerable evidence that unfractionated heparin reduces the risk of pulmonary embolism and deep vein thrombosis, no randomized, clinical trial has examined whether intravenous unfractionated heparin prevents deep vein thrombosis (DVT) or pulmonary embolus (PE) in patients with acute ischemic stroke. Low-molecular-weight heparin (LMWH) LMWH or heparinoids have anti-factor Xa activity and a decreased tendency to induce thrombocytopenia compared with unfractionated heparin. A double-blind trial randomized 308 patients from 4 Hong Kong hospitals into 3 groups: placebo, nadroparin calcium at 4,100 anti-Xa IU o.d., or 4,100 anti-Xa IU b.i.d. [39]. There was no reduction in death or dependence at 3 months between the groups given nadroparin as compared with those given placebo. After 6 months, there was a significant, dose-dependent benefit in the drug-treated patients, with good outcomes in 55% of those treated with high-dose nadroparin versus only 45% in those given placebo. However, this result was not replicated in a preliminary report of a large multi-center, randomized trial in a European population (Fraxiparine in Ischemic Stroke Study) [40]. Thus, results for anti-Xa agents in acute ischemic stroke have not been consistent. The Trial of the Heparinoid ORG 10172 (danaparoid) in Acute Stroke (TOAST) was a randomized, double-blind, placebo-controlled trial in which 1,281 patients were enrolled at 36 US centers [41]. The active treatment arm received intravenous, dose-adjusted danaparoid. There was no significant difference in overall favorable outcomes at 3 months, although at 7 days, 34% of treated patients and only 28% of control subjects had very favorable outcomes (p=0.01). The Heparin in Acute Embolic Stroke Trial (HAEST) is a prospective study that compared dalteparin with aspirin given within 30 hours of atrial fibrillation-associated stroke [42]. There were no statistically significant differences between treatment groups in death and physical dependency at 3 months (66.1% in treated vs 64.8% in control participants). A prospective, randomized, nonplacebocontrolled trial of the LMWH, certoparin, given within 12 hours of acute ischemic stroke involved 400 patients (the Therapy of Patients with Acute Stroke (TOPAS) trial) [43]. There were four treatment groups: 3,000 U o.d., 3,000 U b.i.d., 5,000 U b.i.d., and 8,000 U b.i.d. No benefit was shown in the short term or at 3 months in stroke outcome between low and higher doses of certoparin. Those with a Barthel Index greater than 90 at 3 months included 61.5% in Group 1, 60.8% in Group 2, 63.3% in Group 3, and 56.3% in Group 4. Thus, LMWHs have not been shown to reduce mortality or stroke-related morbidity when used within 48 hours of onset in patients with acute ischemic stroke. Furthermore, LMWHs, when used within 48 hours of onset in patients with acute ischemic stroke, have

Thrombosis in clinical practice

240

not been shown to reduce the rate of stroke recurrence. In TOAST, reduction in recurrent stroke in patients treated with danaparoid did not reach statistical significance. Similarly, administration of nadroparin was associated with a nonsignificant reduction in early stroke recurrence. With regards to DVT and PE, in TOAST there were 2/638 with DVT (0.3%) on danaparoid versus 10/628 (1.6%) with placebo (p<0.05), but there was no significant reduction in frequency of PE [41]. It is unclear whether frequency of PE is also decreased because too few PE occurred in cohorts studied to exclude the possibility of a type II error. Summary Current joint American Academy of Neurology and American Stroke Association guidelines [44] take into account the scope of uncertainty and unanswered questions regarding the use of antithrombotic therapy in acute ischemic stroke. 1. Patients with acute ischemic stroke presenting within 48 h of symptom onset should be given aspirin (160–325 mg o.d.) to reduce stroke mortality and decrease morbidity, provided contraindications such as allergy and gastrointestinal bleeding are absent, and the patient has or will not be treated with recombinant tissue-type plasminogen activator. The data are insufficient at this time to recommend the use of any other platelet anti-aggregant in the setting of acute ischemic stroke. 2. Subcutaneous unfractionated heparin, LMWHs, and heparinoids may be considered for DVT prophylaxis in at-risk patients with acute ischemic stroke, recognizing that nonpharmacologic treatments for DVT prevention also exist. A benefit in reducing the incidence of PE has not been demonstrated. The relative benefits of these agents must be weighed against the risk of systemic and intracerebral hemorrhage. 3. Although there is some evidence that fixed-dose, subcutaneous, unfractionated heparin reduces early recurrent ischemic stroke, this benefit is negated by a concomitant increase in the occurrence of hemorrhage. Therefore, use of subcutaneous unfractionated heparin is not recommended for decreasing the risk of death or strokerelated morbidity, or for preventing early stroke recurrence. 4A. Dose-adjusted, unfractionated heparin is not recommended for reducing morbidity, mortality, or early recurrent stroke in patients with acute stroke (i.e. in the first 48 h) because the evidence indicates it is not efficacious and may be associated with increased bleeding complications. 4B. High-dose LMWH/heparinoids have not been associated with either benefit or harm in reducing morbidity, mortality, or early recurrent stroke in patients with acute stroke and are, therefore, not recommended for these goals. 5. Intravenous, unfractionated heparin or high-dose LMWH/heparinoids are not recommended for any specific subgroup of patients with acute ischemic stroke that is based on any presumed stroke mechanism or location (e.g. cardioembolic, large vessel atherosclerotic, vertebrobasilar, or “progressing” stroke) because data are insufficient. Although the LMWH, dalteparin, at high doses may be efficacious in patients with atrial fibrillation, it is not more efficacious than aspirin in this setting. Because aspirin is easier to administer, it, rather than dalteparin, is recommended for the various stroke subgroups.

Antithrombotic therapy for ischemic stroke

241

Secondary prevention of stroke Stroke or transient ischemic attack is common and likely to be fatal or cause serious disability. A second stroke will not necessarily be of the same type as the initial event, although hemorrhages tend to recur. Patients with previous stroke commonly succumb to other vascular events, in particular, myocardial infarction. Effective secondary prevention depends on giving attention to all modifiable risk factors for stroke as well as treating the causes of the initial stroke. Antiplatelet therapy A systematic review by the Antiplatelet Trialists’ Collaboration showed that among highrisk patients, antiplatelet drugs reduced the odds of any serious vascular event (nonfatal myocardial infarction, nonfatal stroke, or death from vascular causes) by about 25% [45]. The review determined that among people with a prior ischemic stroke, antiplatelet drugs avoided 38 serious vascular events for every 1,000 people treated for about 3 years. The risk of intracranial bleeding with antiplatelet treatment is small, at most 1 or 2 per 1,000 people per year in trials of long-term treatment. Likewise, the risk of nonfatal major extracranial bleeding was only about 3 per 1,000 per year. In general, the benefits of antiplatelet therapy in high-risk individuals outweigh any hazards. Medium-dose aspirin (75–325 mg o.d.) is the agent that has been most thoroughly evaluated, but direct randomized comparisons provide no clear evidence that any one dose of aspirin is more effective than another [45]. Gastrointestinal side effects (dyspepsia, constipation) are clearly dose related. One recent trial assessing different doses of aspirin in patients undergoing carotid endarterectomy confirmed previous trial evidence that adverse events are less common in patients receiving lower doses of aspirin [46]. A recent systematic review comparing thienopyridines (ticlopidine and clopidogrel) with aspirin showed a 12% absolute reduction in the odds of recurrent stroke, corresponding to seven strokes avoided per 1,000 patients treated with a thienopyridine (instead of aspirin) for 2 years [47]. The combination of aspirin and dipyridamole in the second European Secondary Prevention Study (ESPS-2) showed a small advantage over aspirin alone, but with wide CIs including the possibility of almost no extra benefit [48]. A systematic review suggested that, compared with aspirin, the combination reduces the risk of stroke but has no effect on myocardial infarction and little or no overall effect on “serious vascular events” [49]. The European and Australian Stroke Prevention in Reversible Ischemia Trial (ESPRIT) should provide further information about the benefits of adding dipyridamole to aspirin [50]. Both the thienopyridines and dipyridamole plus aspirin are more expensive than aspirin, and, given the modest benefits when compared with aspirin alone, such regimens should probably only be considered in patients with an allergy to aspirin or those with further vascular events while receiving aspirin alone. In the latter case, drugs should only be switched after reconsidering the suspected mechanism of the stroke, and further investigations should be undertaken so as to rule out other treatable causes such as severe carotid stenosis or paroxysmal atrial fibrillation. Clopidogrel is a newer thienopyridine derivative without the adverse effect profile of ticlopidine. The CAPRIE (clopidogrel versus aspirin in patients at risk of ischemic

Thrombosis in clinical practice

242

events) study showed that clopidogrel is slightly more effective than aspirin in reducing the combined outcome of stroke, myocardial infarction, and vascular death among patients with atherosclerotic vascular disease [51]. Although clopidogrel seems to be as safe as aspirin, it is considerably more expensive, and it remains to be seen whether its use in routine practice is cost-effective. Its use is justified in patients who are intolerant to aspirin or who develop a stroke while taking aspirin. The MATCH trial explored the potential benefit of adding aspirin to clopidogrel in a group of approximately 7,600 patients with cerebrovascular disease. Rates of further vascular events were similar in the clopidogrel alone and aspirin plus clopidogrel groups; however, a higher rate of bleeding complications was observed in the recipients of combination antiplatelet treatment. Aspirin plus clopidogrel is unlikely to have a significant role in the secondary preventative treatment of most stroke patients. Whether the combination will be effective in preventing stroke in more highly selected patients, such as those with significant carotid arterial disease, is not yet known. Anticoagulant therapy Anticoagulant in the form of warfarin has a role in a variety of cardiac disorders in primary and secondary prevention of stroke. Cardiac disorders that predispose to stroke and unequivocally seem to benefit from anticoagulation therapy include atrial fibrillation, mitral stenosis, and mechanical valve prosthesis. Anticoagulants are the drugs of choice for preventing stroke in high-risk patients with atrial fibrillation. A systematic review evaluated six trials comparing anticoagulants (target International Normalized Ratio (INR) about 2.0–3.0) with placebo in 2,900 patients with atrial fibrillation [52]. Anticoagulants reduced the relative risk of stroke by 62% (48–72%), corresponding to a reduction in the absolute risk of stroke of 2.7% per year for primary prevention and 8.4% per year for secondary prevention. The rate of intracranial hemorrhage averaged 0.3% per year in the group receiving anticoagulants and 0.1% in the placebo group. Warfarin (target INR 2.2–3.1) has been compared with aspirin for stroke prevention in 2,837 patients with atrial fibrillation in five trials [52]. Both agents were effective but warfarin especially. Overall, warfarin reduced the relative risk of stroke by 36% (14–52%) compared with aspirin. One trial was subsequently excluded from the meta-analysis owing to important differences in the patient population. The relative risk of reduction of stroke with warfarin was reestimated at 49% (26–65%), corresponding to an absolute reduction in risk of stroke per year of 0.6% for primary prevention and 7.0% for secondary prevention. One additional clinical trial compared warfarin (target INR 2.0–3.5) with indobufen (a reversible inhibitor of cyclo-oxygenase) but did not find any major difference in the rate of recurrent stroke between the two groups (absolute risk reduction 1.0%, 1.7–3.7%) [53]. A recent consensus statement based on the available evidence recommends warfarin both for patients of any age who have atrial fibrillation and specific risk factors for stroke (previous transient ischemic attack, stroke, other systemic embolism, hypertension, left ventricular dysfunction) and for patients older than 75 years with atrial fibrillation and no risk factors. Either warfarin or antiplatelet therapy is suggested for patients aged 65–75 with atrial fibrillation and no risk factors, depending on the status of the patient.

Antithrombotic therapy for ischemic stroke

243

Anticoagulation increases the risk of serious bleeding for patients in normal sinus rhythm. Aspirin is a reasonable option for patients with atrial fibrillation who cannot tolerate anticoagulants, although it is not as effective as anticoagulation. In a systematic review of six trials comparing antiplatelet therapy with placebo (3,337 high-risk patients with atrial fibrillation; 40% with prior stroke), aspirin reduced the overall incidence of stroke by 22% (2–38%), with a reduction in the absolute risk of stroke per year of 1.5% for primary prevention and 2.5% for secondary prevention [52]. In general, moderateintensity anticoagulation (target INR 2.0–3.0) is recommended. Therapy should be tailored to the individual, depending not only on the risk of recurrent stroke but also on bleeding risks (e.g. a tendency to fall, recent gastrointestinal bleeding, liver disease, dementia, uncontrolled hypertension) and the potential to benefit from treatment. The best time to start anticoagulation after an ischemic stroke is unclear. Aspirin does reduce the risk of recurrent ischemic stroke and may be the best initial treatment immediately after stroke [54]. When anticoagulants are being considered for long-term use, the treatment preferences of the patient should also be considered, because the benefits of warfarin in the trials may not reflect clinical practice owing to probable differences in anticoagulant monitoring and patient compliance. Indeed, evidence from several observational studies shows that warfarin is generally underused in people with atrial fibrillation at risk of stroke, and that the risk of hemorrhage may be lower than the risks associated with not prescribing warfarin when warranted. Blood pressure and angiotensin converting enzyme inhibition Immediate reduction of blood pressure may be deleterious, but long-term risk is inversely related to the blood pressure achieved. The risk of stroke doubles for every 7.5 mmHg increase in usual diastolic blood pressure; antihypertensives have been shown to reduce stroke risk by about 38% [55]. A recent consensus statement has advocated a patientcentered multidisciplinary approach to the evaluation and treatment of hypertension, particularly patients at the highest risk of stroke [56]. Hypertension is arguably the most important and treatable risk factor for stroke, and there is increasing evidence about the effectiveness of modifying blood pressure in secondary prevention of stroke. A meta-analysis of data from 9 randomized controlled trials on the effects of drugs for lowering blood pressure in survivors of stroke estimated a reduction in the relative risk of recurrent stroke of 29% (95% CI 5–47%) [57]. Whether patient characteristics such as baseline blood pressure were important variables were not shown. The authors also identified several limitations of the analyses and concluded that further evidence was needed. Evidence does, however, remain limited about when to begin antihypertensive treatment after stroke and which drugs to use, although there is limited indirect evidence from randomized trials of primary prevention to support using low-dose diuretics or low dose beta-blockers [58]. Recently, the Swedish trial in old patients with hypertension (STOP-2) published data on 6,614 hypertensive patients randomized to conventional antihypertensives (atenolol, metolprolol, pindolol, or hydrocholorthiazide plus amiloride) or newer antihypertensives (enalapril, lisinopril, felodipine, or isradipine) [59]. Both groups showed important decreases in blood pressure (about 35/17 mmHg) but no major differences in primary endpoints, including fatal and nonfatal stroke, showing

Thrombosis in clinical practice

244

that the conventional and newer antihypertensives are similar at preventing major events or death from cardiovascular disease. The results of the large scale multicenter Heart Outcomes Prevention Evaluation (HOPE) trial suggest that activation of the renin-angiotensin system is an independent risk factor in people with cardiovascular disease, and that the use of angiotensin converting enzyme inhibitors may reduce vascular risk in this population [60]. Overall, 9,297 patients with any evidence of coronary artery disease, stroke, or peripheral vascular disease were randomized to receive either ramipril 10 mg o.d. or placebo. The trial was terminated early when 13.9% of patients given ramipril had reached the primary end point (myocardial infarction, primary stroke, or death from cardiovascular causes) compared with 17.5% of patients given placebo. These results correspond to a risk reduction of 25% for death from cardiovascular disease, 20% for myocardial infarction, and 32% for stroke. The reduction in vascular events was larger than might have been expected from the size of the reductions in blood pressure, again supporting the hypothesis that angiotensin converting enzyme inhibitors act not only by reducing blood pressure. The implications of this trial for clinical practice are that if 50% of people in developed countries and 25% of people in developing countries with vascular disease were to take angiotensin converting enzyme inhibitors, 400,000 deaths and 600,000 nonfatal cardiovascular events could be prevented every year, but at a substantial cost [60]. The cost-effectiveness (and appropriate costs) of large-scale use of these drugs has not been determined. The large, randomized, placebo-controlled Perindopril Protection against Recurrent Stroke Study (PROGRESS), provided further convincing evidence for the rationale of ACE inhibition as secondary prevention [61]. This was a large, randomized, placebocontrolled study of 6,105 patients from 172 centers in Asia, Australasia, and Europe. Patients were randomly assigned active treatment (n=3,051) or placebo (n=3,054). The treatment group was further divided into those patients receiving perindopril alone or perindopril plus indapamide. Over 4 years of follow-up, active treatment reduced blood pressure by 9/4 mmHg. There were 307 (10%) patients who were assigned to active treatment who suffered a stroke, compared with 420 (14%) assigned placebo (relative risk reduction of 28% CI 17–38, p<0.0001). The combination therapy of perindopril plus indapamide reduced blood pressure by 12/5 mmHg and stroke risk by 43%, however perindopril alone reduced blood pressure by only 5/3 mmHg and produced no discernible reduction in the risk of stroke. It is surprising that with perindopril alone there was no significant reduction in stroke risk; however, the combination therapy should now be considered routinely for the secondary prevention of stroke, irrespective of blood pressure.

Antithrombotic therapy for ischemic stroke

245

Cholesterol reduction Recent guidelines recommend the use of 3-hydroxy-3methylglutaryl coenzyme A reductase inhibitors (“statins”) to reduce cholesterol concentrations after myocardial infarction, thereby reducing the risk of death from coronary artery disease and fatal or nonfatal stroke [56]. The strong association between cholesterol concentrations and future coronary heart disease shows that all people with stroke should reduce their cholesterol concentrations by dietary means [62]. A systematic review of the evidence supports cholesterol reduction with a statin in people with prior stroke, a history of coronary heart disease, and a cholesterol concentration greater than 5 mmol/1 (or low density lipoprotein cholesterol concentration greater than 3 mmol/1) [63]. The high profile MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals, clearly demonstrated marked risk reductions of stroke and myocardial infarction, in highrisk patients, irrespective of their initial cholesterol concentrations [64]. UK patients (aged 40–80 years) with coronary artery disease, other occlusive artery disease, or diabetes (with a cholesterol greater than 3.5 mmol/1) were randomly allocated to receive 40 mg simvastatin o.d. or placebo. All-cause mortality was significantly reduced due to a highly significant proportional reduction in the coronary death rate (5.7% vs 6.9%, p=0.0005). There were highly significant reductions of about one-quarter in the first event rate for nonfatal myocardial infarction or coronary death (8.7% vs 11.8%, p<0.0001), for nonfatal or fatal stroke (4.3% vs 5.7%, p< 0.0001), and for coronary or noncoronary revascularization (9.1% vs 11.7%, p< 0.0001). The authors concluded that adding simvastatin to existing treatments safely produces substantial additional benefits for a wide range of high-risk patients, irrespective of their initial cholesterol concentrations. Allocation of 40 mg of simvastatin reduces the rates of stroke and myocardial infarction by about 25% [64].

References [1] Hatano S. Experience from a multicentre stroke register: a preliminary report. Bull WHO 1976; 54: 541–53. [2] Bamford J, Dennis M, Sandercock P, et al. The frequency, causes and timing of death within 30 days of a first stroke: the Oxfordshire community stroke project. J Neurol Neurosurg Psychiatry 1990; 53: 824–9. [3] Davenport RJ, Dennis MS, Wellwood I, et al. Complications after acute stroke. Stroke 1996; 27:415–20. [4] Dromerick A, Reding M. Medical and neurological complications during inpatient stroke rehabilitation. Stroke 1994; 25:358–61. [5] Dobkin BH. Neuromedical complications in stroke patients transferred for rehabilitation before and after diagnostic related groups. J Neurol Rehabil 1987; 1:3–7. [6] Kalra L, Yu G, Wilson K, et al. Medical complications during stroke rehabilitation. Stroke 1995; 26:990–4. [7] Dennis MS, Burn JP, Sandercock PA, et al. Long-term survival after first ever stroke: the Oxfordshire community stroke project. Stroke 1993; 24:796–800.

Thrombosis in clinical practice

246

[8] Sharp FR, Swanson RA, Honkaniemi, et al. Neurochemistry and molecular biology. In: Barnett HJM, Mohr JP, Stein BM, et al. (eds), Stroke: Pathophysiology, Diagnosis, and Management. Philadelphia, PA: Churchill-Livingstone, 1998. [9] Kristian T, Siesjo BK. Calcium in ischaemic cell death. Stroke 1998; 29:705–18. [10] Jones TH, Morawetz RB, Crowell RM, Marcoux FW, FitzGibbon SJ, DeGirolami U, Ojemann RG. Thresholds of focal cerebral ischemia in awake monkeys. J Neurosurg 1981; 54:773–82. [11] Marchal G, Beaudouin V, Rioux P, et al. Prolonged persistence of substantial volumes of potentially viable brain tissue after stroke: a correlative PETCT study with voxel-based data analysis. Stroke 1996; 27:599–606. [12] Hommel M, Bogousslavsky J. Thrombolytics in acute cerebral ischaemia. Exp Opinion Investig Drugs 1994; 3:1011–20. [13] Bourke JP, Young AA, Richards DAB, et al. Second International Study of Infarct Survival. Lancet 1988; 349–60. [14] The TIMI Study Group. Thrombolysis in myocardial infarction. N Engl J Med 1989; 320:618– 27. [15] The International Study Group. GISSI-2. Lancet 1990; 336:71–5. [16] Mahoney F, Barthel D. Functional evaluation: the Barthel Index. Md Med J 1965; 14:61–5. [17] Lindstrom E, Boysen G, et al. Reliability of the Scandinavian Neurological Stroke Scale. Cerebrovasc Dis 1991; 1:103–7. [18] The Multicenter Acute Stroke Trial—Europe Study Group. Thrombolytic therapy with streptokinase in acute ischemic stroke N Engl J Med 1996; 335: 145–50. [19] Donnan GA, Hommel M, Davis SM, McNeil JJ. Streptokinase in acute ischaemic stroke. Lancet 1995; 345:578–9. [20] Candelise L, Aritzu E, Ciccone A, Ricci S, Wardlaw J, MAST-I group. Randomised controlled trial of streptokinase, aspirin, and combination of both in treatment of acute ischaemic stroke. Lancet 1995; 346:1509–14. [21] The European Co-operative Acute Stroke Study (ECASS). Intravenous thrombolysis with recombinant tissue plasminogen activator for acute hemispheric stroke. JAMA 1995; 274:1017– 25. [22] The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke (NINDS). N Engl J Med 1995; 333:1581–7. [23] The Second European Acute Stroke Study (ECASS II) Intravenous thrombolysis with recombinant tissue plasminogen activator for acute hemispheric stroke. Lancet 1998; 352:1245– 51. [24] Clark WM, Wissman S, Albers GW, et al. Recombinant tissue-type plasminogen activator (alteplase) for ischaemic stroke 3 to 5 hours after symptom onset: the ATLANTIS study: a randomised controlled trial. JAMA 1999;282:2019–26. [25] Albers GW, Bates VE, Clark WM, et al. Intravenous tissue-type plasminogen activator for treatment of acute stroke: the Standard Treatment with Alteplase to Reverse Stroke (STARS) study. JAMA 2000; 283:1145–50. [26] Wahlgren NG, McMahon DG, De Keyser J, Indredavik B, Ryman T. Intravenous Nimodipine West European Stroke Trial (INWEST) of Nimodipine in the treatment of Acute Ischaemic Stroke. Cerebrovasc Dis 1994;4:204–10. [27] Lundergan CF, Reiner JS, McCarthy WF, Coyne KS, Califf RM, Ross AM. Clinical predictors of early infarct-related artery patency following thrombolytic therapy: importance of body weight, smoking history, infarct-related artery and choice of thrombolytic regimen: the GUSTO-I experience. Global Utilization of Streptokinase and t-PA for Occluded Coronary Arteries. J Am Coll Cardiol 1998; 32:641–7. [28] Rankin J. Cerebral vascular accidents in patients over the age of 60: prognosis. Scottish Med J 1957; 2:200–15.

Antithrombotic therapy for ischemic stroke

247

[29] Alexandrov AV, Black SE, Ehrlich LE, et al. Predictors of haemorrhagic transformation occurring spontaneously and on anticoagulants in patients with acute ischaemic stroke. Stroke 1997; 28:1198–202. [30] Toni D, Fiorelli M, Bastionello S, et al. Hemorrhagic transformation of brain infarct: predictability in the first 5 hours from stroke onset and influence on clinical outcome. Neurology 1996; 46:341–5. [31] Adams PA, Adams RJ, Brott T, et al. Guidelines for the early management of patients with ischemic stroke: a scientific statement from the Stroke Council of the American Stroke Association. Stroke 2003; 34:1056–83. [32] del Zoppo GJ, Higashida RT, Furlan AJ, et al. PROACT: a phase II randomised trial of recombinant pro-urokinase by direct arterial delivery in acute middle cerebral artery stroke: PROACT Investigators: Prolyse in Acute Cerebral Thromboembolism. Stroke 1998;29:4–11. [33] Furlan AJ, Higashida R, Wechsler L, et al. PROACT II investigators. PROACT II: recombinant prourokinase in acute cerebral thromboembolism: initial trial results. Stroke 1999; 30:234. [34] Lewandowski CA, Frankel M, Tomsick TA, et al. Combined intravenous and intra-arterial rTPA versus intra-arterial therapy of acute ischemic stroke: Emergency Management of Stroke Bridging Trial. Stroke 1999; 30:2598–605. [35] International Stroke Trial Collaborative Group. The International Stroke Trial (IST): a randomised trial of aspirin, subcutaneous heparin, both, or neither among 19435 patients with acute ischaemic stroke. Lancet 1997; 349:1569–81. [36] CAST (Chinese Acute Stroke Trial) Collaborative Group. CAST: randomised placebo controlled trial of early aspirin use in 20,000 patients with acute ischaemic stroke. Lancet 1997; 349: 1641–9. [37] McCarthy ST, Turner J. Low-dose subcutaneous heparin in the prevention of deep vein thrombosis and pulmonary emboli following acute stroke. Age Ageing 1986; 15:84–8. [38] Duke RJ, Bloch FG, Turpie AGG, et al. Intravenous heparin for the prevention of stroke progression in acute partial stable stroke: a randomized controlled trial. Ann Intern Med 1986; 105:825–8. [39] Kay R, Sing Wong K, Yu YL, et al. Low-molecularweight heparin for the treatment of acute ischemic stroke. N Engl J Med 1995; 333:1588–93. [40] Hommel M, FISS bis Investigators Group Fraxiparine in Ischemic Stroke Study (FISS bis) [Abstract]. Cerebrovasc Dis 1998; 8(Suppl. 4):19A. [41] Publications Committee for the Trial of ORG 10172 in Acute Stroke Treatment (TOAST) Investigators. Low molecular weight heparinoid, ORG 10172 (Danaparoid), and outcome after acute ischemic stroke. JAMA 1998; 279:1265–72. [42] Berge E, Abdelnoor M, Nakstad PH, Sandset PM. Low molecular-weight heparin versus aspirin in patients with acute ischaemic stroke and atrial fibrillation: a double-blind randomised study. HAEST Study Group Heparin in Acute Embolic Stroke Trial. Lancet 2000; 355:1205–10. [43] Diener HC, Ringelstein EB, von Kummer R, et al. Treatment of acute ischemic stroke with the low-molecular-weight heparin certoparin. Results of the TOPAS Trial. Stroke 2001; 32:22–9. [44] Report of the Joint Stroke Guideline Development Committee of the American Academy of Neurology and the American Stroke Association. Anticoagulants and Antiplatelet Agents in Acute Ischemic Stroke. Stroke 2002; 33:1934–42. [45] Antiplatelet Trialists’ Collaboration. Collaborative overview of randomised trials of antiplatelet therapy I: prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. BMJ 1994; 308:81–106. [46] Taylor DW, Barnett HJM, Haynes RB, Ferguson GG, Sackett DL, Thorpe KE, et al. Low-dose and high-dose acetylsalicylic acid for patients under going carotid endarterectomy: a randomised controlled trial. Lancet 1999; 353:2179–84. [47] Hankey GJ, Sudlow C, Dunbabin D. Thienopyridine derivatives (ticlopidine, clopidogrel) versus aspirin for prevention of stroke and other serious vascular events in high vascular risk

Thrombosis in clinical practice

248

patients. In: Cochrane Collaboration. Cochrane Library. Issue 4. Oxford: Update Software, 1999. [48] Diener HC, Cunha L, Forbes C, Sivenius J, Smets P, Lowenthal A. European Secondary Prevention Study 2: dipyridamole and acetylsalicylic acid in the secondary prevention of stroke. J Neurol Sci 1996; 143:1–13. [49] Wilterdink JL, Easton JD. Dipyridamole plus aspirin in cerebrovascular disease. Arch Neurol 1999; 56:1087–92. [50] De Schryver E for the ESPRIT Study Group. ESPRIT: mild anticoagulation, acetylsalicylic acid plus dipyridamole or acetylsalicylic acid alone after cerebral ischaemia of arterial origin [Abstract]. Cerebrovasc Dis 1998; 8(Suppl. 4):83. [51] CAPRIE Steering Committee. A randomised, blinded trial of clopidogrel versus aspirin in patients at risk of ischaemic events (CAPRIE). Lancet 1996; 348:1329–39. [52] Hart RG, Benavente O, McBride R, Pearce LA. Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: a meta-analysis. Ann Intern Med 1999; 131:492–501. [53] Morocutti, C, Amabile G, Fattapposta F, Nicolosi A, Matteoli S, Trappolini M, et al. Indobufen versus warfarin in the secondary prevention of major vascular events in nonrheumatic atrial fibrillation. Stroke 1997; 28:1015–21. [54] Chen Z, Sandercock P, Counsell C, Dan HC, Collins R, Liu L, et al. Indications for early aspirin use in acute ischaemic stroke: a combined analysis of 40,000 randomised patients from the Chinese acute stroke trials and the international stroke trial. Stroke 2000; 31:1240–9. [55] MacMahon S, Rodgers A. The epidemiological association between blood pressure and stroke: implications for primary and secondary prevention. Hypertens Res 1994; 17(Suppl. I):23S–32S. [56] Gorelick PB, Sacco RL, Smith DB, Alberts M, Mustone-Alexander L, Rader D, et al. Prevention of a first stroke. A review of guidelines and a multidisciplinary consensus statement from the National Stroke Association. JAMA 1999;281:1112–20. [57] The INDANA (Individual Data Analysis of Antihypertensive intervention trials) Project Collaborators. Effect of antihypertensive treatment in patients having already suffered a stroke. Stroke 1997; 28:2557–62. [58] Mulrow C, Jackson R. Treating primary hypertension. In: Godlee F, Goldmann D, Donald A, Barton S (eds), Clinical Evidence, Issue 2. London: BMJ Publishing Group, 1999; 524–31. [59] Hansson L, Lindholm L, Ekbom T, Dahlof B, Lanke J, Schersten B, et al. Randomised trial of old and new antihypertensive drugs in elderly patients: cardiovascular mortality and morbidity. The Swedish trial in old patients with hypertension-2 study. Lancet 1999; 354:1751–6. [60] The Heart Outcomes Prevention Evaluation [HOPE] Study Investigators. Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in high-risk patients. N Engl J Med 2000; 342:145–53. [61] PROGRESS Collaborative Group. Randomised trial of a perindopril-based blood pressure lowering regimen among 6105 individuals with previous stroke or transient ischaemic attack. Lancet 2001; 358:1033–41. [62] Pekkanen J, Linn S, Heiss G, Suchindron CM, Leon A, Ruskind BM, et al. Ten year mortality from cardiovascular disease in relation to cholesterol level among men with and without preexisting cardiovascular disease. N Engl J Med 1990; 322: 1700–7. [63] Hebert PR, Gaziano JM, Chan KS, Hennekens CH. Cholesterol lowering with statin drugs, risk of stroke, and total mortality: an overview of randomized trials. J AMA 1997; 278:313–21. [64] Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20536 high-risk individuals: a randomised placebo-controlled trial. Lancet 2002; 360:7–22.

12 Management of venous thromboembolism during pregnancy Ian A Greer and AndrewJ Thomson Introduction Venous thromboembolic complications are leading causes of maternal mortality in the developed world. To reduce the incidence of venous thromboembolism (VTE) in pregnancy, and improve outcomes, a wider understanding of the risk factors involved and a better identification of women at risk of thrombosis coupled with effective thromboprophylaxis are required. It is also critical to obtain an objective diagnosis and provide the optimal effective and safe treatment when VTE arises. Thus, VTE is an important area in contemporary medical and obstetric practice.

Diagnosis The clinical diagnosis of VTE during pregnancy is challenging, since the symptoms and signs of deep venous thrombosis (DVT) and pulmonary thromboembolism (PTE) are commonly found in normal pregnancy. These clinical features are usually not caused by thrombosis, but instead, reflect the physiological changes of pregnancy. The symptoms and signs of DVT and PTE are listed in Table 12.1. Several studies have shown that the clinical diagnosis of VTE in pregnancy is unreliable. In one study of pregnant women presenting with suspected DVT, less than 10% actually had the diagnosis confirmed when objective testing was performed [1]; this compared to approximately 25% in studies of nonpregnant subjects [2, 3]. With regard to the diagnosis of PTE, studies in nonpregnant patients have shown that less than 35% of patients suspected of having PTE actually have the diagnosis confirmed following objective testing [4]. In pregnant patients, the clinical features are even less specific. Chan et al. [5] found that only 1.8% of women with suspected PTE in pregnancy had high-probability ventilation-perfusion (V/Q) scan results and less than 6% were treated for PTE after completion of diagnostic imaging. These results suggest that many women without VTE are being admitted to hospital, receiving anticoagulant therapy, and undergoing potentially hazardous investigations unnecessarily. However, since the mortality from untreated PTE is high—up to 30% in nonpregnant patients [6]—it is important that diagnostic imaging should be performed in pregnancy when VTE is suspected to allow appropriate anticoagulation for women who have the diagnosis confirmed and to reduce the risks, inconvenience, and costs of

Thrombosis in clinical practice

250

unnecessary and inappropriate anticoagulation for women who have the diagnosis excluded [7]. It is critical that all women with symptoms or signs that

Table 12.1 The symptoms and signs of VTE. Symptoms and signs DVT

Leg pain or discomfort (especially the left leg) Swelling Tenderness Increased temperature and edema Lower abdominal pain Elevated white cell count

PTE

Dyspnea Collapse Chest pain Hemoptysis Faintness Raised JVP Focal signs in the chest Associated symptoms and signs of DVT

could be compatible with VTE have appropriate investigation since the diagnosis has serious implications, not only for the management of the present pregnancy, but also for other aspects of her life ranging from contraception to thromboprophylaxis in future pregnancies and hormone replacement therapy in later life. In all pregnant women with clinical features of VTE, anticoagulant treatment should be considered until objective testing has been performed. Obstetric units should have protocols in place for the objective diagnosis of suspected VTE during pregnancy. As most events occur outside hospital, midwives and general practitioners must be aware of the possibility of VTE and its underlying risk factors. DVT Real time or Duplex ultrasound is the firstline diagnostic tool for DVT [8]. If ultrasound confirms the diagnosis, anticoagulant treatment should be commenced or continued. In nonpregnant subjects, the pretest clinical probability of DVT modifies both the positive predictive value and the negative predictive value of objective diagnostic tests [9, 10]. This can be extrapolated to the diagnosis of DVT in pregnancy: a negative ultrasound result with a low level of clinical suspicion, suggests that anticoagulant treatment can be discontinued or withheld. With a negative ultrasound report and a high level of clinical suspicion, anticoagulation should be continued and the ultrasound repeated in one week or X-ray venography should be considered. If repeat testing is negative, anticoagulant treatment should be discontinued [11].

Management of venous thromboembolism during pregnancy

251

PTE In the investigation and diagnosis of suspected PTE, guidelines suggest that both a V/Q lung scan and bilateral duplex ultrasound leg examinations should be performed [7]. Where the V/Q scan reports a “medium” or “high” probability of PTE, anticoagulant treatment should be continued. With a “low” probability of PTE on V/Q scan but positive ultrasound for DVT, anticoagulant treatment should be continued. When a V/Q scan reports a low risk of PTE and there are negative leg ultrasound examinations, yet there is a high level of clinical suspicion, anticoagulant treatment should continue with repeat testing in 1 week (V/Q scan and leg ultrasound examination). If the clinical probability of PTE is high, even if the V/Q scan shows “low” probability and leg ultrasound examination is negative, then alternative imaging techniques should be considered. Similarly, if the chest X-ray (CXR) has abnormalities, which lead to difficulties in the diagnosis of PTE using V/Q scanning, then alternative imaging techniques are warranted. Thus, it is usually best to perform a CXR before proceeding to V/Q scanning. These alternative techniques include pulmonary angiography, magnetic resonance imaging or helical computerized tomography (CT). While it has been suggested that helical CT testing should be restricted to the postnatal period (because of concerns surrounding the radiation doses involved), recent studies conclude that the average fetal radiation dose with helical CT is less than that with V/Q lung scanning during all trimesters [12], Further, a recent survey of North American radiology departments found that most respondents already perform CT angiography in pregnant patients suspected of having PTE, although their policies and practices were found to vary considerably [13]. It should be borne in mind, however, that maternal breast tissue is especially sensitive to radiation exposure during pregnancy. It is estimated that spiral CT is associated with exposure of the woman’s breast to radiation of 2.0–3.5 rads. The delivery of 1 rad of radiation to a woman’s breast increases her lifetime risk of developing breast cancer by 14% [14]. Thus, while helical CT scanning is associated with a lower risk of radiation for the fetus, this must be offset by the relatively high radiation dose to the mother’s thorax and, in particular, breast tissue. In pregnant patients with suspected PTE, an electrocardiogram (ECG), a CXR, and arterial blood gases (ABG) should be considered. While these investigations are useful in the clinical assessment of the patient, their value in the diagnosis of PTE is less clear [16]. The CXR is usually normal, and its value is in the exclusion of other pathology (e.g. pneumothorax and pneumonia). Radiological findings that are associated with PTE include focal infiltrate, segmental collapse, a raised diaphragm, pleural effusion, and wedge-shaped infarction (a raised diaphragm is a normal finding in pregnancy). Since the radiation dose to the fetus with a CXR is small (Table 12.2), this test should be performed in all pregnant women with suspected PTE. Regarding the ECG, one study of unselected nonpregnant patients with suspected PTE found that only tachycardia and right bundle branch block were significantly more frequent in patients with confirmed PTE than in patients who had negative objective testing [17]. Furthermore, changes, can occur in the ECG in normal pregnancy that makes the ECG more difficult to interpret. These changes include a right or left axis shift,

Thrombosis in clinical practice

252

transient ST segment and/or T wave changes, and small Q waves in lead III, which can disappear on inspiration. The ABG in PTE may show a reduced PaO2 and a normal or low PaCO2, although normal PaO2 and PaCO2 values may be found, particularly with smaller emboli. During normal pregnancy, there may be a chronic mild respiratory alkalosis; this results in lower PaCO2 and bicarbonate ion levels than in the

Table 12.2 Estimates of fetal radiation dose during diagnostic tests for VTE. Chest X-ray

<0.001 rad

Limited venography

<0.05 rad

V/Q scan (depends on isotopes used)

0.58 rad

Low-dose perfusion scanning (omitting ventilation scanning)

<0.012 rad

CT pulmonary angiographaya

1st trimester <0.002 rad 2nd trimester <0.008 rad 3rd trimester <0.013 rad

Data from Ginsberg et al. [15] except (a) from Winer-Muram et al. [12].

nonpregnant situation. In view of these changes in ABG values with normal pregnancy, and the risk of hemorrhage from the arterial puncture site in the anticoagulated patient, some authors have proposed that ABG testing is of no value in the assessment and management of patients with suspected PTE [18]. D-dimer testing D-dimer, a degradation product of crosslinked thrombus, is now used as a screening test for VTE in the nonpregnant situation. D-dimer levels are typically elevated in patients with acute VTE and, as a screening test, it has a high negative predictive value. During normal pregnancy, D-dimer levels are increased and become abnormal at term and in the postnatal period in most healthy pregnant women [19]. Furthermore, D-dimer levels are increased if there is a concomitant problem such as preeclampsia [20]. Therefore, Ddimer tests are, in general, and, particularly in pregnancy, sensitive but nonspecific markers of VTE. Thus, use of D-dimer does not usually avoid the need for objective diagnosis in pregnancy.

Initial treatment of VTE When VTE is suspected during pregnancy, treatment with low-molecular-weight heparin (LMWH) or unfractionated heparin (UFH) should be given until the diagnosis is confirmed or refuted by objective testing, unless such treatment is contraindicated. In nonpregnant patients, many well-conducted randomized trials and meta-analyses have compared intravenous UFH and subcutaneous LMWH for the treatment of acute DVT

Management of venous thromboembolism during pregnancy

253

and PTE [21, 22]. These studies show that LMWH is at least as safe and effective as UFH in the nonpregnant, but LMWH is preferable during pregnancy in terms of ease of administration and safety (see following text). Other studies in nonpregnant patients have shown that long-term LMWH (and UFH) is as effective and safe as warfarin for the prevention of recurrent VTE [23–25]. In light of these data, guideline documents [26] have proposed that, in the pregnant patient with acute VTE, two alternative approaches are acceptable: 1. intravenous UFH followed by at least 3 months of subcutaneous LMWH in therapeutic doses or adjusted-dose subcutaneous UFH; or 2. adjusted-dose subcutaneous UFH or therapeutic doses of LMWH can be used both for initial and long-term treatment. UFH is monitored using the mid-interval activated partial thromboplastin time (APTT). The dose of UFH will need to be adjusted so that the APTT ratio falls into the therapeutic target range (usually 1.5–2.5 times the average laboratory control value). The use of the APTT to monitor treatment with UFH is problematic in pregnancy. Clinical audits have shown that APTT testing is often poorly performed and is technically problematic, especially in late pregnancy, when an apparent heparin resistance occurs owing to increased levels of fibrinogen and factor VIII. These difficulties can lead to unnecessarily high doses of heparin being administered, with subsequent hemorrhagic complications. LMWHs have clear advantages over UFH. They have a better bioavailabilty, a better safety profile compared to UFH, are more convenient for the patient to use, and more convenient for the clinician to monitor. Treatment with LMWHs is monitored by measuring peak anti-Xa activity (3 h postinjection) using a chromogenic substrate assay. In this situation, a target therapeutic range of approximately 0.5–1.2 U/ml for anti-Xa should be used [7, 27]. In the nonpregnant patient with an acute VTE, LMWH is administered once daily using a weight-adjusted dose regimen. However, as the half-life of LMWH is decreased in pregnancy, because of an increased renal excretion [28], twice daily regimens are preferable to once daily dosing [29]. Since most women gain weight as pregnancy progresses, there is the potential for a change in the LMWH’s volume of distribution, and some authors have proposed that the dose of LMWH should be increased as pregnancy advances [30, 31]. Our experience of the LMWH, enoxaparin, administered in a dose of 1 mg/kg 12 hourly (Table 12.3), suggests that dose adjustments are rarely required and monitoring is not necessary (using the therapeutic range of approximately 0.5–1.2 U/ml for anti-Xa) or need only be done infrequently [26] or in unusual situations such as extremes of body weight.

Table 12.3 Initial dose of the LMWH enoxaparin for acute treatment of VTE in pregnancy. Early pregnancy weight (kg)

Initial dose of enoxaparin (mg b.i.d.)

<50

40

50–69

60

70–89

80

Thrombosis in clinical practice

254

≥90

100

Risks of anticoagulant therapy in pregnancy The use of anticoagulant therapy in pregnancy is associated with risks to both the fetus and the mother. These risks emphasize the importance of performing objective tests to confirm the initial diagnosis so that anticoagulant therapy is continued only in those women who require it. Fetal complications of anticoagulants during pregnancy Since heparins (UFH and LMWH) do not cross the placenta, they are not associated with teratogenicity or fetal bleeding [32–34], and numerous studies conclude that heparin therapy is safe for the fetus [35]. In contrast, coumarin derivatives, such as warfarin, cross the placenta and should be avoided, where possible, during pregnancy since they are associated with congenital malformations and fetal and neonatal hemorrhage [36, 37]. In the first trimester of pregnancy, warfarin can cause an embryopathy (Table 12.4) if taken between 6 and 12 weeks of gestation. This was reported in 6.4% of women with prosthetic heart valves who took warfarin throughout pregnancy [38]. It is probable that the risk can be eliminated by

Table 12.4 The features of warfarin embryopathy. Mid-facial, particularly nasal, hypoplasia Stippled chondral calcification Short proximal limbs Short phalanges Scoliosis

avoiding warfarin between 6 and 12 weeks of gestation, and the risk may be higher when the dose of warfarin is greater than 5 mg o.d. [39]. In addition to embryopathy, warfarin is associated with fetal and neonatal hemorrhage. As the fetal liver is immature and levels of vitamin-K dependent clotting factors are low (Factors II, VII, IX, and X), maternal warfarin therapy maintained in the therapeutic range (INR of 2–3) will be associated with excessive anticoagulation in the fetus. This can lead to hemorrhagic complications in the fetus and is a concern, particularly, at the time of delivery, when the combination of the anticoagulant effect and trauma of delivery can lead to bleeding in the neonate. Further, the administration of warfarin during the second and third trimesters of pregnancy may lead to neurodevelopmental problems in childhood [40].

Management of venous thromboembolism during pregnancy

255

Maternal complications of anticoagulant therapy during pregnancy The maternal complications of anticoagulant therapy taken during pregnancy include hemorrhage (with both heparin and warfarin), and osteoporosis, thrombocytopenia, and allergy (with heparin). With UFH, the rate of major bleeding in pregnant patients is 2% [43], which is similar to the reported rates of hemorrhage when heparin [42] and warfarin [40] are used for the treatment of DVT in the nonpregnant situation. We discussed earlier, that it is often very difficult to achieve an adequate APTT response with UFH during pregnancy because of the apparent heparin resistance, and this may lead to hemorrhagic complications. Bleeding complications appear to be very uncommon with LMWH [26]. Osteoporosis Heparin-induced osteoporosis is a significant problem with long-term UFH. This effect has been demonstrated in both animal and human studies and arises through decreasing rates of bone formation and increased bone resorption [44]. In patients receiving longterm UFH, symptomatic vertebral fractures occur in about 2–3% and significant reductions in bone density have been reported in up to 30% [45, 46]. Human and animal studies indicate that LMWHs appear to have substantially less risk of osteoporosis [35, 47, 48]. Monreal et al. [49] compared therapeutic doses of dalteparin and UFH, administered for up to 6 months, in non-pregnant patients with DVT: 6 of 40 (15%; 95% CI 6–30%) patients who received UFH developed spinal fractures compared with only 1 of 40 (3%; 95% CI 0–11%) receiving dalteparin. Petilla et al. [50] found that, in contrast to UFH, the LMWH dalteparin administered during pregnancy had no significant effects on bone mineral density in the lumbar spine, compared with a control group of healthy pregnant women. Despite these reassuring data, the administration of LMWH has been reported to cause osteoporosis and vertebral fracture [51]. Thrombocytopenia Heparin-induced thrombocytopenia is a rare but potentially fatal side effect [52]. An early, benign, transient thrombocytopenia can occur with initiation of UFH therapy. More seriously, approximately 3% of nonpregnant patients receiving UFH develop an idiosyncratic immune, IgG-mediated thrombocytopenia, which is frequently complicated by extension of preexisting VTE or new arterial thrombosis [53]. The risk is much lower for patients receiving LMWH. Immune-mediated thrombocytopenia should be suspected when the platelet count falls to less than 100×109/L or less than 50% of the baseline value 5–15 days after commencing heparin (or sooner with recent heparin exposure). In pregnant women who develop immune-mediated thrombocytopenia and who require ongoing anticoagulant therapy, use of the heparinoid, danaparoid sodium, is recommended [54].

Thrombosis in clinical practice

256

Allergy Allergic reactions to heparins (both UFH and LMWH) usually take the form of itchy, erythematous raised plaque-like lesions at the injection sites. Changing heparin preparations or switching from an LMWH to UFH may be helpful, although a degree of cross-reactivity can occur [55].

Maintenance treatment Following initial heparinization, maintenance of anticoagulation with warfarin is recommended in nonpregnant patients with VTE. In view of the fetal complications of warfarin administration during pregnancy outlined earlier, adjusted-dose subcutaneous, UFH or subcutaneous LMWH in therapeutic doses are employed for maintenance treatment in pregnancy [26]. Guideline documents recommend that therapeutic levels of anticoagulation should be continued throughout pregnancy because of the ongoing prothrombotic changes in the coagulation system and venous flow, and the risk of recurrent VTE during this time [7]. However, in certain clinical situations, notably patients with contraindications to warfarin [49] and patients with malignant disease [56], the dose of heparin has been successfully reduced to an intermediate dose after initial full-dose anticoagulation in an attempt to reduce the risks of anticoagulant-related bleeding and heparin-induced osteoporosis. This type of modified dosing regimen may be useful in pregnant women who require prolonged periods of anticoagulation with heparin, although there have been no comparative studies investigating these strategies in pregnancy. Further, a high recurrence rate of VTE was reported (47%) in a trial of nonpregnant patients when thromboprophylactic doses of UFH were employed after initial management with intravenous UFH [57].

Labor and delivery In an attempt to minimize bleeding complications at the time of delivery (including epidural hematoma formation during neuraxial anesthesia), heparin treatment should be discontinued 24 hours prior to elective induction of labor or delivery by Caesarean section. If spontaneous labor occurs, the woman should be advised that she should not inject any further heparin until she has been clinically assessed. If the woman is in labor and she has been administering UFH, then monitoring of the APTT should be performed and protamine sulphate may be given to reverse the heparin’s anticoagulant effect, if required. A similar strategy has been proposed for the management of LMWH therapy during labor and delivery (i.e. discontinuing the LMWH 24 h prior to elective induction of labor or Caesarean section). Bleeding complications in pregnancy associated with LMWH do not appear to be a problem. However, the anticoagulant effects of LMWH are not fully reversed with protamine sulphate [58]. For this reason, if there is considered to be a highrisk of hemorrhage in a woman requiring anticoagulation, intravenous UFH

Management of venous thromboembolism during pregnancy

257

should be employed (as with cessation of the heparin infusion, the anticoagulant effects reverse promptly). Similarly, if the woman is deemed to have a very high risk of recurrent VTE (e.g. a VTE diagnosed near term), then therapeutic intravenous UFH can be initiated and discontinued 4–6 hours prior to the expected time of delivery in order to limit the duration of time without therapeutic anticoagulation. These are essentially the only situations where UFH may be preferred over LMWH in pregnancy. Some clinicians have proposed that, in this situation, consideration should be given to the insertion of a temporary IVC filter and a planned induction of labor after reversal of anticoagulation [16, 26]. The risk of epidural or spinal hematoma formation during neuraxial instrumentation in pregnant patients receiving LMWH has not been clearly quantified [59]. In summary, guideline documents suggest that regional techniques should not be used until at least 12 hours after the previous prophylactic dose of LMWH. When a woman presents on a therapeutic regimen of LMWH (e.g. enoxaparin 1 mg/kg, 12 hourly), then regional techniques should not be employed for at least 24 hours after the last dose of LMWH. LMWH should not be given for at least 3 hours after the epidural catheter has been removed and the cannula should not be removed within 10–12 hours of the most recent injection [7]. Postpartum anticoagulants (either heparin or warfarin) should be given for at least 6 weeks or until at least 3 months of anticoagulant therapy has been completed. Both heparin (UFH and LMWH) and warfarin can be used safely during breast-feeding.

Thromboprophylaxis in pregnancy The management of women with a single previous VTE has been controversial until recently. This was because of the wide variation in risk that has been reported (1–13%) [60–64] and concerns about the hazards of long-term UFH therapy which were discussed before. The higher estimate of risk led many clinicians to employ pharmacologic prophylaxis with heparin or LMWH during pregnancy and the puerperium. However, these estimates of risk have significant limitations. For example, objective testing was not used in all cases, some of the studies were retrospective and the prospective studies had relatively small sample sizes. Brill-Edwards et al. [65] reported a prospective study of 125 pregnant women with a single previous objectively diagnosed VTE. No heparin was given antenatally, but anticoagulants, usually coumarin, following an initial short course of heparin or LMWH, was given for 4–6 weeks postpartum. The overall rate for recurrent antenatal VTE was 2.4% (95% CI 0.2–6.9%). However, none of the 44 women (95% CI 0–8.0%) who did not have an underlying thrombophilia and whose previous VTE had been associated with a temporary risk factor developed a VTE, whereas 5.9% (95% CI 1.2–16%) of the women who were found to have an underlying thrombophilia or whose previous VTE had been idiopathic had a recurrent event. As pregnancy is associated with hyperestrogenism, this should probably be considered a recurrent risk factor in women with a previous VTE on the “pill” or in pregnancy. Thus, in the woman with a previous VTE that was not pregnancy related, associated with a risk factor that is no longer present, and with no additional risk factor or underlying thrombophilia (see Table 12.5 for common risk factors for VTE in

Thrombosis in clinical practice

258

pregnancy), antenatal LMWH should not be routinely prescribed, but this strategy must be discussed with the woman and her views taken into account, especially in view of the wide CIs reported by Brill-Edwards et al. [65] (95% CI 0–8.0%). Graduated elastic compression stockings and/or

Table 12.5 Common risk factors for VTE in pregnancy. Patient factors Age over 35 years Obesity (BMI>29 kg/m2) in early pregnancy

Thrombophilia Past history of VTE (especially if idiopathic or thrombophilia associated) Gross varicose veins Significant current medical problem (e.g. nephrotic syndrome) Current infection or inflammatory process (e.g. active inflammatory bowel disease or urinary tract infection) Immobility (e.g. bed rest or lower limb fracture) Paraplegia Recent long distance travel Dehydration Intravenous drug abuse Qvarian hyperstimulation Pregnancy/obstetric factors Caesarean section particularly as an emergency in labor Operative vaginal delivery Major obstetric hemorrhage Hyperemesis gravidarum Preeclampsia

low-dose aspirin can be employed antenatally in these women. Postpartum, she should receive anticoagulant therapy for at least 6 weeks (e.g. 40 mg enoxaparin or 5,000 IU dalteparin daily or warfarin (target INR 2–3) with LMWH overlap until the INR is >2.0)± graduated elastic compression stockings (Table 12.6). In those women with a single previous VTE and an underlying thrombophilia, or where the VTE was idiopathic, pregnancy, or “pill” related, or where there are additional risk factors such as obesity or nephrotic syndrome, there is a stronger case for LMWH prophylaxis. Antenatally, these women should be considered for prophylactic doses of LMWH (e.g. 40 mg enoxaparin or 5,000 IU dalteparin daily) ± graduated elastic

Management of venous thromboembolism during pregnancy

259

compression stockings. This should be started as soon as possible following the diagnosis of pregnancy. More intense LMWH therapy in the presence of antithrombin deficiency is usually prescribed (e.g. enoxaparin 0.5–1 mg/kg 12 hourly or dalteparin 50–100 IU/kg 12 hourly), although many women with previous VTE and antithrombin deficiency will be on long-term anticoagulant therapy (see later). Postpartum anticoagulant therapy for at least 6 weeks (e.g. 40 mg enoxaparin or 5,000 IU dalteparin daily or warfarin (target INR 2–3) with LMWH overlap until the INR is ≥2.0) ± graduated elastic compression stockings is recommended (Table 12.6). In the woman with multiple previous VTE, and no identifiable thrombophilia and who is not on long-term anticoagulant therapy, there is consensus that she should receive antenatal LMWH thromboprophylaxis (e.g. 40 mg enoxaparin or 5,000 IU dalteparin daily) and graduated elastic compression stockings. This should be started as soon as possible following the diagnosis of pregnancy. Postpartum she should receive at least 6 weeks’ pharmacological

Table 12.6 Suggested thromboprophylactic strategies in pregnancy (NB specialist advice for individualized management of patients is advisable in many of these situations). Clinical situation

Suggested thromboprophylaxis strategy

Single previous VTE (not pregnancy or ‘pill’ related) associated with a transient risk factor and no additional current risk factors, such as obesity.

Antenatal: surveillance or prophylactic doses of LMWH (e.g. 40 mg enoxaparin or 5,000 IU dalteparin daily), ± graduated elastic Compression stockings. Discuss decision regarding antenatal LMWH with the woman. Postpartum: anticoagulant therapy for at least 6 weeks (e.g. 40 mg enoxaparin or 5,000 IU dalteparin daily or warfarin (target INR 2–3) with LMWH overlap until the INR is ≥ 2.0 2.0) ± graduated elastic compression stockings.

Single previous idiopathic VTE or single previous VTE with underlying thrombophilia and not on long-term anticoagulant therapy, or single previous VTE and additional current risk factor(s) (e.g. morbid obesity, nephrotic syndrome)

Antenatal: prophylactic doses of LMWH (e.g. 40 mg enoxaparin or 5,000 IU dalteparin daily) ± graduated elastic compression stockings. NB there is a strong case for more intense LMWH therapy in antithrombin deficiency (e.g. enoxaparin 0.5–1 mg/kg 12 hourly or dalteparin 50–100 IU/kg 12 hourly). Postpartum: anticoagulant therapy for at least 6 weeks (e.g. 40 mg enoxaparin or 5,000 IU dalteparin daily or warfarin (target INR 2–3) with LMWH overlap until the INR is ≥ 2.0) ± graduated elastic compression stockings.

More than one previous episode of VTE, Antenatal: prophylactic doses of LMWH (e.g. 40 mg with no thrombophilia and not on longenoxaparin or 5,000 IU dalteparin daily)+graduated term anticoagulant therapy. elastic compression stockings. Postpartum: anticoagulant therapy for at least 6 weeks (e.g. 40 mg enoxaparin or 5,000 IU dalteparin daily or warfarin (target INR 2–3) with LMWH overlap until the

Thrombosis in clinical practice

260

INR is ≥2.0)+graduated elastic compression stockings. Previous episode(s) of VTE in women receiving long-term anticoagulants (e.g. with underlying thrombophilia).

Clinical situation

Antenatal: switch from oral anticoagulants to LMWH therapy (e.g. enoxaparin 0.5–1 mg/kg 12 hourly or dalteparin 50–100 IU/kg 12 hourly) by 6 weeks gestation+graduated elastic compression stockings. Postpartum: resume long-term anticoagulants with LMWH overlap until INR in pre-pregnancy therapeutic range+graduated elastic compression stockings.

Suggested thromboprophylaxis strategy

Thrombophilia (confirmed laboratory abnormality) but no prior VTE.

Antenatal: surveillance or prophylactic LMWH ± graduated elastic compression stockings. The indication for pharmacologic prophylaxis in the antenatal period is stronger in AT deficient women than the other thrombophilias, in symptomatic kindred compared to asymptomatic kindred and also where additional risk factors are present, Postpartum: anticoagulant therapy for at least 6 weeks (e.g. 40 mg enoxaparin or 5,000 IU dalteparin daily or warfarin (target INR 2–3) with LMWH overlap until the INR is ≤2.0) ± graduated elastic compression stockings.

Following Caesarean section or vaginal delivery.

Carry out risk assessement for VTE. If additional risk factors such as emergency section in labor, age over 35 years, high BMI, etc. present, then consider LMWH thromboprophylaxis (e.g. 40 mg enoxaparin or 5,000 IU dalteparin) ± graduated elastic compression stockings.

prophylaxis, either with LMWH, or warfarin. If she is switched to warfarin postpartum, the target INR is 2–3 and LMWH should be continued until the INR is ≥2. A longer duration of postpartum prophylaxis may be required for women with additional risk factors. When prophylactic doses of LMWH are used, the dose may require to be reduced in women with very low or very high body weight. At low body weight (<50 kg or BMI less than 20 kg/m2), lower doses of LMWH may be required (e.g. 20 mg enoxaparin daily or 2,500 IU dalteparin daily), while in obese patients (e.g. BMI>30 in early pregnancy), higher doses of LMWH may be required. The platelet count should be checked before, and 1 week after the introduction of LMWH, then on around a monthly basis to detect heparininduced thrombocytopenia [66]. The woman with previous episode(s) of VTE receiving long-term anticoagulants (e.g. with underlying thrombophilia) should switch from oral anticoagulants to LMWH by 6 weeks’ gestation, and be fitted with graduated elastic compression stockings. These women should be considered at very high risk of antenatal VTE and should receive anticoagulant prophylaxis throughout pregnancy. They should be advised, ideally, prepregnancy, of the need to switch from warfarin to LMWH as soon as pregnancy is confirmed. The dose of heparin given should be closer to that used for the treatment of VTE rather than that used for prophylaxis (e.g. enoxaparin 0.5–1 mg/kg 12 hourly or dalteparin 50–100 IU/kg 12 hourly. NB 12 hourly injections may be preferable to once daily injections in view of the increased clearance of LMWH in pregnancy), based on the early pregnancy weight [66]. The platelet count should be checked before and 1 week after the introduction of LMWH, then around once a month, although heparin-induced

Management of venous thromboembolism during pregnancy

261

thrombocytopenia is extremely unlikely to occur. Postpartum, she should resume longterm anticoagulants with LMWH overlap until INR is in the pre-pregnancy therapeutic range, plus graduated elastic compression stockings. Where a woman has thrombophilia confirmed on laboratory testing, but no prior VTE, surveillance or prophylactic LMWH with or without graduated elastic compression stockings can be used antenatally. The indication for pharmacological prophylaxis in the antenatal period is stronger in AT deficient women (where dose of LMWH of enoxaparin 0.5–1 mg/kg 12 hourly or dalteparin 50–100 IU/kg 12 hourly are usually employed) than the other thrombophilias and also in symptomatic kindred compared to asymptomatic kindred. The presence of additional risk factors, for example, obesity or immobility, may also merit consideration for antenatal thromboprophylaxis with LMWH. Postpartum, these women should receive anticoagulant therapy for at least 6 weeks (e.g. 40 mg enoxaparin or 5,000 IU dalteparin daily or warfarin (target INR 2–3) with LMWH overlap until the INR is ≥2.0) with or without graduated elastic compression stockings. These women usually require specialized and individualized advice from clinicians with expertise in the area. Women undergoing Caesarean section and vaginal delivery should also have a risk assessment for VTE [67]. In a patient undergoing Caesarean section, thromboprophylaxis (e.g. 40 mg enoxaparin or 5,000 IU dalteparin) should be prescribed if she has one or more additional risk factors, such as emergency section in labor, age over 35 years, high BMI. In patients at high risk, graduated elastic compression stockings should be used. These can also be employed if heparin is contraindicated. In women undergoing vaginal delivery, a similar strategy can be used with LMWH being prescribed if there are two or more additional minor risk factors or one major risk factor for example, morbid obesity [68]. There has been concern with regard to LMWH and epidural hematoma, through postmarketing reports to the FDA largely from the USA. These events have mostly been in elderly women (median age 75 years) undergoing orthopedic surgery. Additional factors, such as concomitant nonsteroidal anti-inflammatory agent use (which can enhance bleeding risk particularly in the elderly) or multiple puncture attempts at spinal or epidural, have also been implicated. The true incidence of epidural hematoma is impossible to determine due to lack of denominator data. In addition, practice in North America and Europe may differ, particularly, with regard to LMWH use. In Europe, enoxaparin is used in a dose of 20 mg or 40 mg daily, while in North America, 30 mg b.i.d., may be used. Such differences in patients and practice make it difficult to extrapolate the information in these reports to obstetric practice. A degree of caution must, nonetheless, be exercised in the concomitant use of LMWH and neuraxial anesthesia. In general terms, neuraxial anesthesia is not used until at least 12 hours after the previous prophylactic dose of LMWH. When a woman presents, while on a therapeutic regimen of LMWH, regional techniques should not be employed for at least 24 hours after the last dose of LMWH. LMWH should not be given for at least 3 hours after the epidural catheter has been removed and the cannula should not be removed within 10–12 hours of the most recent injection [69–71].

Thrombosis in clinical practice

262

References [1] Hull RD, Raskob GF, Carter CJ, Serial IPG in pregnancy patients with clinically suspected DVT. Clinical validity of negative findings. Ann Intern Med 1990; 112:663–7. [2] Hull RD, Hirsh J, Sackett D, et al. Diagnostic efficacy of IPG in suspected venous thrombosis: an alternative to venography. N Engl J Med 1977; 296:1497–1500. [3] Hull RD, Hirsh J, Carter C, et al. Diagnostic efficacy of IPG for clinically suspected DVT: a randomized trial. Ann Intern Med 1985; 102:21–8. [4] Wells PS, Ginsberg JS, Anderson DR, et al. Use of a clinical model for safe management of patients with suspected pulmonary embolism. Ann Intern Med 1998; 129:997–1005. [5] Chan WS, Ray JG, Murray S, et al. Suspected pulmonary embolism in pregnancy: clinical presentation, results of lung scanning, and subsequent maternal and pediatric outcomes. Arch Intern Med 2002; 162:1170–5. [6] Carson JL, Kelly MA, Duff A, et al. The clinical course of pulmonary embolism. N Engl J Med 1992; 326:1240–5. [7] Thomson AJ, Greer IA. Thromboembolic Disease in Pregnancy and the Puerperium: Acute Management. Royal College of Obstetricians and Gynaecologists, London: RCOG Press, Guideline No 28, 2001 (http:www.rcog.org.uk/guidelines.asp?PageID=106&GuidelineID=20). [8] Macklon NS. Diagnosis of deep venous thrombosis and pulmonary embolism. In: Greer IA (ed.), Bailliere’s Clinical Obstetrics and Gynaecology Thromboembolic Disease in Obstetrics and Gynaecology. London: Bailliere Tindall, 1997; 463–77. [9] Wheeler HB, Hirsh J, Wells P, et al. Diagnostic tests for deep vein thrombosis. Clinical usefulness depends on probability of disease. Arch Intern Med 1994; 154:1921–8. [10] Wells PS, Anderson DR, Bormanis J, et al. Value of assessment of pretest probability of deepvein thrombosis in clinical management. Lancet 1997; 350:1795–8. [11] Thomson AJ, Greer IA. Non-haemorrhagic obstetric shock. In: Thompson W and TambyRaja RL (eds), Emergencies in Obstetrics and Gynaecology. Bailliere’s Clin Obstet Gynaecol 2000; 14:19–41. [12] Winer-Muram HT, Boone JM, Brown HL, et al. Pulmonary embolism in pregnant patients: fetal radiation dose with helical CT. Radiology 2002; 224:487–92. [13] Schuster ME, Fishman JE, Copeland JF, et al. Pulmonary embolism in pregnant patients: a survey of practices and policies for CT pulmonary angiography. Am J Roentgenol 2003; 181:1495–8. [14] Remy-Jardin M, Remy J. Spiral CT angiography of the pulmonary circulation. Radiology 1999;212: 615–36. [15] Ginsberg JS, Hirsh J, Rainbow AJ, et al. Risks to the fetus of radiological procedures used in the diagnosis of materna/venous thromboembolic disease. Thromb Haemost 1989;61:189–96. [16] Rodger MA, Walker M, Wells PS. Diagnosis and treatment of venous thromboembolism in pregnancy. Best Pract Res Clin Haematol 2003; 16:279–96. [17] Rodger M, Makropoulos D, Turek M, et al. Diagnostic value of the electrocardiogram in suspected pulmonary embolism. Am J Cardiol 2000; 86:807–9,A10. [18] Rodger MA, Carrier M, Jones GN, et al. Diagnostic value of arterial blood gas measurement in suspected pulmonary embolism. Am J Respir Critical Care Med 2000; 162:2105–8. [19] Francalanci I, Comeglio P, Liotta AA, et al. D-dimer plasma levels during normal pregnancy measured by specific ELISA. Int J Clin Lab Res 1997; 27:65–7. [20] Francalanci I, Comeglio P, Liotta AA, et al. D-dimer in intrauterine growth restriction and gestational hypertension. Thromb Res 1995; 80:89–92. [21] Dolovich L, Ginsberg JS, Douketis JD, et al. A meta-analysis comparing low molecular weight heparins to unfractionated heparin in the treatment of venous thromboembolism: examining

Management of venous thromboembolism during pregnancy

263

some unanswered questions regarding location of treatment, product type, and dosing frequency. Arch Intern Med 2000;160:181–8. [22] Gould MK, Dembitzer AD, Doyle RL, et al. Lowmolecular-weight heparins compared with unfractionated heparin for treatment of acute deep venous thrombosis. A meta-analysis of randomized, controlled trials. Ann Intern Med 1999; 130:800–9. [23] Lopaciuk S, Bilska-Falda H, Noszczyk W, et al. Low-molecular-weight heparin versus acenocoumarol in the secondary prophylaxis of deep vein thrombosis. Thromb Haemost 1999; 81:26–31. [24] Pini M, Aiello S, Manotti C, et al. Low-molecularweight heparin versus warfarin in the prevention of recurrences after deep vein thrombosis. Thromb Haemost 1994; 72:191–7. [25] Van der Heijden JF, Hutten BA, Buller HR, et al. Vitamin K antagonists or low-molecularweight heparin for the long term treatment of symptomatic venous thromboembolism. Cochrane Database Syst Rev 2000:CD002001. [26] Bates SM, Greer IA, Hirsh J, Ginsberg JS. Use of antithrombotic agents during pregnancy: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004; 126(3 Suppl.):627S–644S. [27] Rodie VA, Thomson AJ, Stewart FM, et al. Low molecular weight heparin for the treatment of venous thromboembolism in pregnancy-case series. Br J Obst Gynaecol 2002;109:1020–4. [28] Casele HL, Laifer SA, Woelkers DA, et al. Changes in the pharmacokinetics of the low molecular weight heparin enoxaparin sodium during pregnancy. Am J Obstet Gynecol 1999; 181:1113–17. [29] Thomson AJ, Walker ID, Greer IA. Low molecular weight heparin for the immediate management of thromboembolic disease in pregnancy. The Lancet 1998; 352:1904. [30] Crowther MA, Spitzer K, Julian J, et al. Pharmacokinetic profile of a low-molecular weight (Reviparin) in pregnant patients: a prospective cohort study. Thromb Res 2000; 98:133–8. [31] Sephton V, Farquharson RG, Topping J, et al. A longitudinal study of maternal dose response to low molecular weight heparin in pregnancy. Obstet Gynecol 2003; 101:1307–11. [32] Flessa HC, Klapstrom AB, Glueck MJ, et al. Placental transport of heparin. Am J Obstet Gynecol 1965; 93:570–3. [33] Forestier F, Daffos F, Capella-Pavlovsky M. Low molecular weight heparin (PK 10169) does not cross the placenta during the second trimester of pregnancy: study by direct fetal blood sampling under ultrasound. Thromb Res 1984; 34:557–60. [34] Forestier F, Daffos F, Rainaut M, et al. Low molecular weight heparin (CY 216) does not cross the placenta during the third trimester of pregnancy. Thromb Haemost 1987; 57:234. [35] Sanson BJ, Lensing AWA, Prins MH, et al. Safety of low-molecular-weight heparin in pregnancy: a systematic review. Thromb Haemost 1999; 81:668–72. [36] Hall JAG, Paul RM, Wilson KM. Maternal and fetal sequelae of anticoagulation during pregnancy Am J Med 1980; 68:122–40. [37] Ginsberg JS, Hirsh J, Turner CD, et al. Risks to the fetus of anticoagulant therapy during pregnancy. Thromb Haemost 1989; 61:197–203. [38] Chan WS, Anand S, Ginsberg JS. Anticoagulationof pregnant women with mechanical heart valves. Arch Int Med 2000;160:191–6. [39] Vitale N, De Feo M, De Santo LS, et al. Dosedependent fetal complications of warfarin in pregnant women with mechanical heart valves. J Am Coll Cardiol 1999; 33:1642–5. [40] Wesseling J, van Driel D, Heymans HAS, et al. Coumarins during pregnancy: long term effects on growth and development in school age children. Thromb Haemost 2001; 85:609–13. [41] Ginsberg JS, Kowalchuk G, Hirsh J, et al. Heparin therapy during pregnancy: risks to the fetus and mother. Arch Intern Med 1989;149: 2233–6. [42] Hull RD, Delmore TJ, Carter CJ, et al. Adjusted subcutaneous heparin versus warfarin sodium in the long-term treatment of venous thrombosis. N Engl J Med 1982; 306:189–94. [43] Hull RD, Hirsh J, Jay R, et al. Different intensities of oral anticoagulant therapy in the treatment of proximal-vein thrombosis. N Engl J Med 1982; 307:1676–81.

Thrombosis in clinical practice

264

[44] Shaughnessy SG, Hirsh J, Bhandari M, et al. Histomorphometric evaluation of heparininduced bone loss after discontinuation of heparin treatment in rats. Blood 1999; 93:1231–6. [45] Dahlman TC. Osteoporotic fractures and the recurrence of thromboembolism during pregnancy and the puerperium in 184 women undergoing thromboprophylaxis with heparin. Am J Obstet Gynecol 1993; 168:1265–70. [46] Douketis JD, Ginsberg JS, Burrows RF, et al. The effects of long-term heparin therapy during pregnancy on bone density. Thromb Haemost 1996; 75: 254–7. [47] Muir J, Andrew M, Hirsh J, et al. Histomorphometric analysis of the effects of standard heparin on trabecular bone in vivo. Blood 1996; 88:1314–20. [48] Carlin AJ, Farquharson RG, Quenby SM, et al. Prospective observational study of bone mineral density during pregnancy: low molecular weight heparin versus control. Hum Reprod 2004;19: 1211–14. [49] Monreal M, Lafoz E, Olive A, et al. Comparison of subcutaneous unfractionated heparin with a low molecular weight heparin (Fragmin) in patients with venous thromboembolism and contraindications for coumarin. Thromb Haemost 1994; 71: 7–11. [50] Pettila V, Leinonen P, Markkola A, et al. Postpartum bone mineral density in women treated for thromboprophylaxis with unfractionated heparin or LMW heparin. Thromb Haemost 2002; 87:182–6. [51] Sivakumaran M, Ghosh K, Zaidi Y, et al. Osteoporosis and vertebral collapse following lowdose, low molecular weight heparin therapy in a young patient. Clin Lab Haematol 1996; 18:55–7. [52] Elalamy I, Potevin F, Lecrubier C, et al. A fatal lowmolecular-weight heparin-associated thrombocytopenia after hip surgery: possible usefulness of PF4-heparin ELISA test. Blood Coagul Fibrinolysis 1996; 7:665–71. [53] Warkentin TE, Levine MN, Hirsh J, et al. Heparininduced thrombocytopenia in patients treated with low-molecular-weight heparin or unfractionated heparin. N Engl J Med 1995; 332:1330–5. [54] Magnani HN. Heparin-induced thrombocytopenia (HIT): an overview of 230 patients treated with Orgaran (Org 10172) Thromb Haemost 1993; 70: 554–61. [55] Grassegger A, Fritsch P, Reider N. Delayed-type hypersensitivity and cross-reactivity to heparins and danaparoid: a prospective study. Dermatol Surg 2001; 27:47–52. [56] Lee AYY, Levine MN, Baker RI, et al. Lowmolecular-weight heparin versus a coumarin for the prevention of recurrent venous thromboembolism in patients with cancer. N Engl J Med 2003; 349: 146–53. [57] Hull RD, Delmore T, Carter C, et al. Adjusted subcutaneous heparin versus warfarin sodium in the long-term treatment of venous thrombosis. N Engl J Med 1982; 306:1676–81. [58] Crowther MA, Berry LR, Monagle PT, et al. Mechanisms responsible for the failure of protamine to inactivate low-molecular-weight heparin. Br J Haematol 2002; 116:178–86. [59] Abramovitz S, Beilin Y. Thrombocytopenia, low molecular weight heparin, and obstetric anesthesia. Anesthesiol Clin North America 2003; 21:99–109. [60] Ginsberg J, Greer IA, Hirsh J. Sixth ACCP consensus conference on antithrombotic therapy. Use of antithrombotic agents during pregnancy. Chest 2001; 119:122S–131S. [61] De Swiet M, Floyd E, Letsky E. Low risk of recurrent thromboembolism in pregnancy [letter]. Br J Hosp Med 1987; 38:264. [62] Howell R, Fidler J, Letsky E, et al. The risk of antenatal subcutaneous heparin prophylaxis: a controlled trial. Br J Obstet Gynaecol 1983; 90: 1124–8. [63] Badaracco MA, Vessey M. Recurrent venous thromboembolic disease and use of oral contraceptives. BMJ 1974; 1:215–17. [64] Tengborn L. Recurrent thromboembolism in pregnancy and puerperium: is there a need for thromboprophylaxis? Am J Obstet Gynecol 1989;160: 90–4.

Management of venous thromboembolism during pregnancy

265

[65] Brill-Edwards P, Ginsberg JS for the Recurrence Of Clot In This Pregnancy (ROCIT) Study Group. Safety of withholding antepartum heparin in women with a previous episode of venous thromboembolism. N Engl J Med 2000; 343:1439–44. [66] Scottish Intercollegiate Guidelines Network. Prophylaxis of venous thromboembolism. A National Clinical Guideline, Edinburgh, 2002. [67] Greer IA. Epidemiology, risk factors and prophylaxis of venous thrombo-embolism in obstetrics and gynaecology. In: Greer IA (ed.), Bailliere’s Clinical Obstetrics and GynaecologyThromboembolic Disease in Obstetrics and Gynaecology. London: Bailliere Tindall. 1997; 403– 30. [68] The National Institute for Clinical Excellence, Scottish Executive Health Department and Department of Health, Social Services and Public Safety: Northern Ireland. Confidential Enquiries into Maternal Deaths in the United Kingdom 1997–99. London: TSO, 2001. [69] Checketts MR, Wildsmith JAW. Central nerve block and thromboprophylaxis—is there a problem ? Br J Anaes 1999; 82:164–7. [70] Horlocker TT, Wedel DJ. Spinal and epidural blockade and perioperative low molecular weight heparin : smooth sailing on the Titanic. Anesth Analg 1998; 86:1153–6. [71] http://www.asra.com/items_of_interest/consensus_%20statements/ (Current website with consensus statement on anticoagulants and neuraxial anaesthesia from the American Society of Regional Anesthesia).

Thrombosis in clinical practice

266

13 Thrombophilia Ian Jennings Introduction Venous thrombosis has been estimated to have an annual incidence of 1 in 1,000 individuals, and associated pulmonary embolism represents a major cause of morbidity and mortality, accounting for 50,000–100,000 deaths each year in the US [1], Consequently, detection and treatment of venous thromboembolic disease is of major medical importance. The term thrombophilia, although reported to lack an internationally accepted definition [2] is generally used to describe subjects with a predisposition or tendency to thromboembolism. Thrombophilia may be an inherited or acquired condition, and Lane et al. [3] report use of the term to indicate atypical thrombosis including onset at an early age (the incidence of thrombosis increases from 1/10,000 young adults to 1/100 elderly subjects), frequent recurrence, and a strong family history. An inherited defect may be identified in approximately 25–30% of patients with thromboem-bolic disease [4], Indeed, many investigations for thrombophilia focus on inherited or genetic defects. The probability of a prothrombotic state was first hypothesized by Virchow in 1860 [5], who proposed trauma or damage to the vein, venous stasis, and changes in the blood as causes of hypercoagulability. However, it was not until 1965, with the description of a family with a deficiency of the serine protease inhibitor antithrombin (AT) [6], that laboratory screening for inherited (familial) thrombophilia began; subsequently, a wide variety of defects have been associated with inherited thrombophilia, and a similarly wide range of tests have been employed by laboratories to identify individuals and families with these defects. The pattern of tests employed by laboratories has changed as new defects have been discovered and epidemiologic studies have either confirmed or disproved links between a candidate defect and thrombophilia. Fibrinolytic tests, for example, formed a major part of the thrombophilia screen of many departments in the mid-1980s, but these tests have subsequently been shown to have little diagnostic value [7]. Defects which are currently considered to be associated with increased thrombotic risk are shown in Table 13.1. These can be classified as inherited or acquired, with further subclassification of environmental or transient acquired factors (Table 13.1); classification in terms of procoagulant and anticoagulant activity has been employed: “loss of function” defines deficiency of anticoagulant proteins C, S, and AT; “gain of function” occurs with increased levels of procoagulant factors and factor V

Thrombosis in clinical practice

294

Table 13.1 Causes of inherited and acquired thrombophilia. Congenital/inherited Loss of function

Gain of function

AT

Dysfibrinogenemia

PS

FVL

PC

PTA

Mixed/uncertain Hyperhomocysteinemia Elevated levels of FVIII Elevated levels of FIX, FXI, TAFI APC resistance in the absence of FVL Acquired Antiphospholipid syndrome Age, surgery, immobilization, malignancy, pregnancy, hormone therapy/oral contraceptive use AT: antithrombin; PS: protein S; PC: protein C; FVL: factor V Leiden; PTA: prothrombin G20210A; FVIII: factor VIII; FIX: factor IX; FXI: factor XI; TAFI: thrombin activatable fibrinolysis inhibitor; APC: activated protein C.

Leiden (FVL). Crowther and Kelton propose classification of these groups as group 1 and group 2 disorders [8]. Although each of these risk factors has been shown to be an independent risk factor for thromboembolic disease, it is important to note that precipitation of an event is often due to a combination of factors, particularly the interaction of inherited disorders and transient acquired defects.

Inherited thrombophilia Protein C Protein C (PC) is a vitamin K-dependent glycoprotein of MW approximately 62,000, synthesized in the liver, and circulating in plasma as a two-chain serine protease zymogen. PC is activated by thrombin, cleaving an Arg169-Leu170 bond on the heavy chain and releasing a small 12a.a. peptide (PC activation peptide) [3]. The activation rate is accelerated by formation of a complex between thrombin and the endothelial cell

Thrombophilia

295

membrane protein thrombomodulin [9], and is further enhanced by interaction with the endothelial cell PC receptor [10]. The procoagulant activity of plasma is reduced by activated protein C (APC), which inactivates factors Va and VIIIa by selective proteolytic cleavages. This inactivation requires a cofactor, protein S (PS), which together with calcium ions enhances the binding of APC to phospholipid membrane surfaces. Activated PC is slowly neutralized by PC inhibitor, α2-antiplasmin, and α2macroglobulin [11], and has a half-life of 20–30 min. APC also has an important role in the inflammatory process and also has a fibrinolytic role enhanced by PS [12]. Protein C deficiency The first case of PC deficiency was described by Griffin et al. in 1981 [13], in a family with a history of recurrent thromboembolic disease who displayed reduced levels of PC, at approximately 50% of normal. Heterozygous PC deficiency is associated with deep vein thrombosis with or without pulmonary embolism at a young age (<40 years) [14], although there is a higher association of PC deficiency with superficial thrombophlebitis and arterial disease [15]. Warfarin-induced skin necrosis is also a feature of this deficiency, though infrequently encountered. Homozygous PC deficiency is associated with massive and potentially fatal thromboembolism shortly after birth, with central nervous system thrombosis, ophthalmic thrombosis, and characteristic purpuric skin lesions commonly seen [16]. The prevalence of PC deficiency is a matter of some debate, determined in part by the source of the study population, and probably present in about 0.2–0.4% of the general population [17–19]. It is possible that two cohorts of PC-deficient subjects exist—those with a clinically dominant disorder and those with a clinically recessive picture; the former may be linked to co-segregation of additional risk factors. Hasstedt et al. postulated an unknown inherited defect linked to PC deficiency as a cause of thrombophilia in one pedigree [20]. Although mutations in those subjects with asymptomatic PC deficiency are equally as “bad” as those in symptomatic cases, it is likely that some mutations are associated with a milder risk of thrombosis [21]. PC deficiency can be divided into two sub-groups (type I and type II deficiency) based on results of functional and immunological assays. In type I deficiency, a concordant reduction in PC antigen and activity is seen. The majority of cases of PC deficiency fall into this category, and as much as 60% of the mutations causing type I deficiency are caused by missense mutations [3]. In type II deficiency, an abnormal PC molecule is implied by reduced activity in the presence of normal quantities of protein. Functional defects which feature impaired PC binding to substrate, PS, and calcium are included in this group [22], and may only be detected by clot-based assays for PC activity. Early studies suggested the frequency of such defects may be 40% of type II PC deficiency [23], but overestimation due to interference of APC resistance in subjects with FVL may have distorted these figures. More recent estimates suggest around 5% of type II defects fall into this category [24]. Acquired PC deficiency is seen in patients with liver disease, disseminated intravascular coagula tion, insulin-dependent diabetes, essential hypertension, and sickle-cell disease [11]. Therefore, finding of a low PC level should be followed up by assay of at least two other vitamin K-dependent clotting factors to allow confirmation of a specific deficiency.

Thrombosis in clinical practice

296

Protein S PS is a vitamin K-dependent single-chain glycoprotein of MW approximately 70,000, synthesized in the liver, endothelial cells, and megakaryocytes [25]. Approximately 60% of PS in plasma is bound to C4b-binding protein (C4b-BP), a complement system component. This binding occurs via a beta subunit of C4b-BP, and all circulating B-chain C4b-BP is bound to PS. Excess PS circulates as free PS, capable of acting as a nonenzymatic cofactor to activated PC. PS is thought to enhance the inactivation of factor Va and VIIIa on phospholipid surfaces, possibly by localizing APC at these loci [26]. Inactivation occurs as a biphasic reaction; for factor Va, full inactivation only occurs after a second cleavage at Arg 306, enhanced by PS. Factor VIIIa is inactivated in a biphasic reaction enhanced synergistically by PS and factor V (FV) [27]. PS also plays a role in the fibrinolytic effect of APC [28]. Recent studies have also indicated a possible role for bound PS in APC-independent anticoagulant activity [25]. PS is inactivated by thrombininduced proteolysis, although this occurs at a relatively slow rate. Although C4b-BP levels are increased in acute phase reactions, this increase is mostly of the alpha subunit, and ratios of free and bound PS remain relatively constant [29]. However, recent studies have indicated plasma levels of free PS may be regulated by genetic variation in the C4bBP gene [30]. Protein S deficiency The first case of PS deficiency was described by Comp and Esmon in 1984 [31]. Deficiency of PS is associated with deep vein thrombosis with or without pulmonary embolism at a young age (<45 years) and arterial disease. Purpura fulminans has been described but is a rare occurrence [32]. The incidence of PS deficiency in the thrombophilic population varies in different reports; this may be associated with the selection criteria for the study groups and also by assay variability. The probability of a thrombotic event occurring under the age of 45 in subjects heterozygous for PS deficiency in thrombophilic families is around 50% [32]. Case-control studies have failed to agree on the association between reduced PS levels and thrombotic risk. However, Faioni and colleagues [33] demonstrated a mild increase in risk of venous thromboembolism (VTE), and a prospective cohort study of subjects with PS deficiency showing a significantly higher risk of thromboembolism than in subjects with normal levels of PS confirms the link between PS deficiency and thrombophilia [34]. Mutations in the PS gene have been identified in a high proportion of subjects studied (up to 90% in one study) [35]. PS deficiency is subclassified based on the levels of total and free antigen and PS activity. A subject in whom free PS antigen was reduced but total PS antigen was normal was first described by Comp et al. [36]. Subsequent studies have identified subjects with type I deficiency (reduced total and free antigen and activity), type II deficiency (reduced PS activity, normal free PS antigen), and type III deficiency (total PS antigen normal or borderline, free PS antigen and activity reduced) [35]. Type II deficiency is rarely encountered; some early descriptions included subjects with FVL mutation, in whom PS activity is reduced [37], as a consequence of APC resistance. However, in the PS mutation database published by Gandrille et al. [38], 8 out of 126 mutations were associated with a type II phenotype. Because of variables which may

Thrombophilia

297

affect the specificity and sensitivity of functional PS assays, some authors do not recommend their use [39] despite the risk of missing some type II defects, whereas others recommend confirmation of low PS activity with immunological assays for free PS [2]. Type I and III deficiency are considered to be phenotypic variants of the same genetic disorder, and arise because of an age-related increase in total PS antigen levels, whereas free PS antigen remains unchanged [32]. Overlap of total PS levels occurs between normal and PS deficient subjects [40], free PS measurement is therefore important in diagnosis of the defect. Indeed, Makris et al. have shown free PS to be the best indicator of a PROS1 genetic defect with 100% specificity [41]. Acquired PS deficiency has been described in a number of conditions including pregnancy, oral contraceptive use, liver disease, and after orthotopic liver transplantation [42, 43]. Antithrombin Antithrombin (AT) is a serine protease inhibitor (serpin), molecular weight 58,200, synthesized in the liver, with a plasma concentration around 125 mg/l. The plasma halflife of AT has been reported to have three exponential components and is around 65 hours [44], although the half-life of infused AT concentrate may be considerably shorter than this. First described in 1939 [45], AT inhibits activated proteolytic procoagulant enzymes; it is the major plasma inhibitor of thrombin, accounting for approximately 80% of thrombin inhibitory activity, and also inhibits factors Xa, IXa, XIa, XIIa, and kallikrein. AT has two functional domains, a reactive center, and heparin binding site. Thrombin forms a 1:1 stochiometric complex with AT by cleavage of the reactive site (Arg393Ser394), and inactivation of thrombin is accelerated at least 1,000-fold by negatively charged UFH binding to both thrombin and positively charged Arg47 on the AT molecule. Low-molecular-weight heparins lack thrombin binding capacity, but still bind to AT and increase the inactivation of Xa relative to thrombin [46]. AT is therefore an important naturally occurring anticoagulant, functioning to maintain hemostatic balance in vivo. AT deficiency Deficiency of AT was first described by Egeberg in 1965, in a Norwegian family with thrombophilia [47]. Subsequent investigations have identified subgroups of AT deficiency, defined as type 1 deficiency (quantitative; in which both immunological and functional AT levels are reduced to a similar degree) and type 2, in which the activity is reduced relative to the immunological level. Type 2 deficiency can be further classified into II RS (reactive site), where molecular defects affecting the reactive site result in an overall reduction in AT activity, type II HBS (heparin binding site), in which the molecular defect disturbs the heparin binding site, causing a reduction in heparin binding activity, and type II PE (pleiotropic), where multiple functional defects may cause reduced AT activity and a mild reduction in AT concentration. This classification has been adopted by the Scientific and Standardisation Committee of the ISTH, and a database of over 120 distinct mutations has been compiled. Although the majority of these are point mutations causing type I deficiency, several gene mutations have been described for all subtypes of AT deficiency [48]. Both type I and type II deficiencies can

Thrombosis in clinical practice

298

be identified with a functional assay, although different assay substrates display differing sensitivity to specific molecular defects. Immunological assays may assist in the subclassification of AT defects. However, in the monitoring of treatment with AT concentrates, the presence of degraded AT molecules has led to discrepancies between functional and immunological levels of AT [49]. Deficiency of AT is associated with an increased risk of venous thrombosis, most commonly in the deep leg veins, iliac and femoral veins, and the superficial leg veins [48]. The age of first venous thrombotic event is often early—Thaler and Lechner described a series of 80 subjects in which 67% had a first clinical episode between 10 and 35 years [50]. A higher incidence of major VTE in subjects with AT deficiency than for patients with PC or PS deficiency was described by De Stefano et al. [15], although other studies suggest the relative risk of thrombosis is lower than for patients with PC deficiency [51]. It is estimated that approximately 1/5,000 of the general population have hereditary AT deficiency [52], and this figure rises to between 1% and 7% of subjects with thrombophilia depending on the selection of subjects [3]. The increased risk of VTE conferred by AT deficiency is indicated in population studies—the Leiden thrombophilia study reported an odds ratio of 5.0 for patients with a persistently low AT level compared to a control group [51]. Defects in type II HBS are thought to occur at a greater frequency, and given the specificity of some AT assays, may be underdiagnosed, but there is a lesser association with thrombophilia [48]. Acquired AT deficiency may arise in a large number of conditions, including DIC, preeclampsia, liver disease, nephrotic syndrome, major surgery, malignancies including acute promyelocytic leukemia and treatment with L-asparaginase. AT levels also fall during oral contraceptive use [53–57]. FVL and APC-resistance FV is a single-chain glycoprotein, molecular weight 330,000, which is composed of an amino-terminal region containing the A1-A2 domains, and the carboxy-terminal region containing the A3-C1-C2 domains [58]. FV may be activated by thrombin, factor Xa, meizothrombin, and Russell’s Viper venom. Thrombin activates factor V to Va by cleavage at three specific bonds, Arg709, Arg1018, and Arg1545; the resultant Va is composed of the amino-terminal region, devoid of most of the B domain, linked to the carboxy-terminal region by a calcium ion bridge. Activated FV not only accelerates prothrombin activation in the prothrombinase complex on phospholipid surfaces [58], but has also been shown to act synergistically with APC and PS in the inactivation of factor VIII (FVIII) [59]. In 1993, Dahlback described a family in which a poor anticoagulant response to APC was observed, and hypothesized an inherited deficiency of a cofactor to APC [60]. The phenotype of APCresistance (APCr) was subsequently associated with heterozygosity or homozygosity for a single point mutation (nucleotide position 1691 GA substitution) in the FV gene. The resulting FV molecule, with an Arg506Gln substitution, undergoes slow inactivation by APC, and was termed FVL [61]. Studies have shown inactivation of factor Va occurs via a biphasic reaction, in which rapid cleavage of the peptide bond at Arg506 by APC is followed by a slower cleavage at Arg306. FVL lacks the Arg506 cleavage site, and although the protein can still be inactivated, this occurs only through slow direct cleavage at the Arg306 site [58].

Thrombophilia

299

FVL occurs in the European population with a frequency of around 4% and is believed to have arisen from a point mutation in a single individual [62] around 21,000–34,000 years ago. FVL is most common in European subjects, generally more so in northern than southern Europe, and is not found in mongoloid populations of South East Asia and the Americas [63]. Approximately 5% of cases of APCr are not associated with the FVL defect. It is possible that some of these cases result from acquired defects, especially as a large number of conditions and variables affect the APCr ratio. Acquired APCr will be discussed later in this chapter. A mutation at the 306 APC cleavage site (FV Cambridge) will also cause APCr [64]. However, a more common mutation found in Chinese subjects, Arg306 to Gly306, is not associated with APCr [65, 66]. To date, no mutations have been found at the third cleavage site in factor Va, Arg679. Recombinant mutations of factor VIII which lack either the Arg336 or Arg562 cleavage site for APC do not exhibit APCr; it is only when both mutations are present in the same gene product that APCr becomes apparent [67]. Pseudohomozygosity for APCr has been described in which heterozygous FVL is associated with partial type 1 FV deficiency [68]. Colucci et al. [69] reported that FV levels in subjects without FVL do not influence the APCr test. However, subsequent studies have demonstrated that depletion of FV levels may be associated with a lowering of the APC sensitivity ratio [70]. Not all APCr is due to resistance of FV to degradation from APC; Castoldi et al. reported approximately 50% to result from loss of FV anticoagulant activity [71]. Inheritance of a specific FV gene haplotype, termed the HR2 haplotype, occurs more frequently in subjects with a low APCr [72]. De visser et al. [73] showed an association between compound heterozygosity for FVL and HR2 haplotype and a reduction in both APC sensitivity, and FV levels. However, the HR2 haplotype was not associated with an increased risk of venous or arterial thrombosis, and there was no clear indication of increased risk when the HR2 haplotype was co-inherited with FVL. APCr can be determined by addition of APC to plasma; specificity for FVL is provided by the addition of FV deficient plasma to the test system. Phenotypic analysis may suggest homozygous FVL, but confirmation of the homozygous genotype by genetic analysis is essential, as it rules out the presence of a FV null allele and a true heterozygous genotype. Detection of mutation may be carried out in a number of ways, including restriction enzyme digest [61], using PCR-based amplification of a limited region flanking the 1691 G-A mutation. The mutation results in loss of an Mnl 1 restriction site, and this can be visualized on agarose gels. Other methods include enzyme-linked immunosorbent assay [74], and allele-specific amplification [75]. A multiplex method has also been described for the simultaneous detection of FVL and a mutation in the prothrombin gene by a single-strand mutation detection multiplex [76], and commercial kits are available for multiplex assays. In a study by Lutz et al., six centers successfully genotyped 62 samples for the FVL mutation, suggesting complete concordance may be obtained between centers for this test [77]. However, Hofgartner and Tait described a problem rate of 0.15% for FVL testing in a survey of 42 US genetic testing laboratories [78]. Their data imply that difficulties in the diagnosis of FVL mutation may exist in testing for the genotype as well as the phenotype. This has been confirmed in the multi-center studies described by Preston [79] and Tripodi [80].

Thrombosis in clinical practice

300

Prothrombin G20210A Prothrombin is a glycoprotein of MW 72,500, synthesized in the liver, which undergoes proteolytic cleavage by factor Xa to form thrombin, the central enzyme in many procoagulant and anticoagulant processes. A transition from G to A at nucleotide position 20210 of the prothrombin gene, termed the prothrombin 20210A allele (PTA allele), was first reported by Poort in 1996 [81]. This transition, in a nontranslated region of the gene, was associated with increased levels of functionally normal factor II (87% of patients with the defect were in the highest quartile of plasma prothrombin levels) and was associated with increased venous thrombotic risk. Recently, it has been suggested that high levels of factor II, seen in subjects with the PTA transition, may increase thrombotic risk through inhibition of APC, thus increasing APCr. Prothrombin levels have been shown to correlate inversely with APCr [82]. However, Koenen et al. [83] have reported a possible role of elevated prothrombin levels in reduction of the APC-independent activity of PS. As with FVL, the G20210A transition is believed to have occurred in a single Caucasian individual, and is not the result of multiple mutations [84]. If a subject is heterozygous for the PTA allele, the risk for venous thrombosis is increased 2- to 5-fold, compared to subjects without the mutation [85, 86]. Plasma prothrombin concentrations are raised, to an average 1.3-fold higher than the average level in subjects possessing the mutation [81], but results of individuals often fall within the normal range. Consequently, measurement of plasma factor II levels cannot be used to diagnose the defect; genetic analysis, often in multiplex with FVL investigation, must be employed.

Dysfibrinogenemia Fibrinogen, a 340,000 kD protein synthesized in the hepatocytes, is cleaved by thrombin to form soluble fibrin monomers, which undergo polymerization and cross-linking in the presence of factor XIII to form insoluble fibrin. Subsequent degradation by plasmin forms fibrin degradation products (D-dimers). Dysfibrinogenemias arise from structural abnormalities in the fibrinogen molecule akin to type II plasma protein defects. Dysfibrinogenemias are rare, and the majority of subjects with identified defects are asymptomatic. However, some case studies have indicated a link between dysfibrinogenemia and thrombosis, with approximately 20% of subjects exhibiting thrombophilia. Defects associated with thrombosis include resistance to plasmin but also delayed fibrin polymerization. Although the fibrinogen clotting activityiantigen ratio is considered the confirmatory test for diagnosis of dysfibrinogenemia, the majority of inherited abnormalities can be detected by a prolonged thrombin time [87]. The thrombin time may also be abnormal in acquired conditions (liver disease, presence of heparin) and thus serve as a useful screen for interference with other thrombophilia screening tests. Consequently, the relevant rarity of this defect does not preclude screening with a rapid and relatively inexpensive laboratory test.

Thrombophilia

301

Thrombophilia due to mixed or uncertain etiology The cause of some thrombophilic conditions, including increased plasma homocysteine levels and increased concentrations of procoagulant clotting factors, may arise from a genetic defect or may be acquired. Homocysteine Homocysteine is a nonprotein sulfydryl aminoacid derived from metabolism of methionine. Homocysteine is subsequently remethylated to methionine or undergoes trans-sulfuration to cysteine depending upon methionine levels. Severe hyperhomocysteinemia may arise from homozygous deficiency of methylenetetrahydrofolate reductase (MTHFR) or cystathionine-β-synthase (CBS) deficiency. Homozygosity for a C677T mutation in the (MTHFR) gene, a relatively common polymorphism giving rise to a thermolabile variant of MTHFR, may cause a mild to moderate increase in homocysteine levels, as can heterozygosity for MTHFR and CBS deficiency, and deficiencies of folate, cobalamine, or pyridoxine. Mild to moderate hyperhomocysteinemia has been described by several groups as a risk factor for both arterial disease and venous thrombosis, although the mechanism by which this occurs is still unclear [88, 89]. Interference with endothelial metabolism has been postulated, and is a likely cause of enhanced atherosclerotic risk [90]. Homocysteine-mediated alteration in fibrin clot stability may also contribute to an acquired dysfibrinogenemia [91]. Despite a genetic link to hyperhomocysteinemia, Martinelli et al. reported a low risk of thrombosis in family members of patients with this disorder [92]. Increased levels of procoagulant factors A number of investigators have linked raised levels of procoagulant factors FVIII:C, FVII:C, FIX:C, and FXI:C with increased incidence of thrombosis [93]. FVIII levels above 150 U/dl are associated with a 5- to 6-fold increase in thrombotic risk compared to levels below 100 U/dl, and may be found in up to 11 % of the general population. Familial correlation of FVIII levels has been described, implying a possible genetic component [94], although a genetic link has not yet been confirmed, and Lavigne et al. have reported increased levels of factors VIII, IX, and XI in families with thrombophilia [95]. As FVIII is an acute phase protein, it is possible that raised levels of FVIII are a consequence of a thrombotic event, however, these levels have been shown to persist and are not related to levels of other acute phase proteins (CRP, fibrinogen) [96]. Although a number of laboratories now include VIII:C as part of their thrombophilia screen, measurement is currently likely to remain of value in epidemiological studies only. Guidelines on assay design and performance have been published for the measurement of procoagulant proteins in plasma [97, 98]. It is interesting to note that for loss of function defects (PS, PC, AT deficiency), the presence or absence of a deficient state, based on normal adult reference ranges, is generally employed in studies to explore the relative risk of the defect, despite the fact that for many assays there is an overlap in levels measured in subjects with a genetically identified defect and normal subjects [99]. In contrast, for gain of function defects,

Thrombosis in clinical practice

302

including raised levels of procoagulant clotting factors and APC resistance in the absence of FVL, relative risks are often assessed using percentiles. In these cases, the clinical relevance of an individual patient measurement is less certain, particularly as an increased risk may be inferred when levels of a clotting factor are still within the reference range.

Acquired thrombophilia Lupus anticoagulant/ antiphospholipid syndrome The antiphospholipid syndrome (APS) is defined by the presence of antiphospholipid antibodies (APA, lupus anticoagulant (LA) or anticardiolipin antibodies) in association with arterial thromboembolism or VTE, and/or recurrent, unexplained miscarriage. The syndrome was first described in subjects with SLE, in whom a circulating, aspecific coagulation inhibitor could be identified [100]. A link between the presence of this LA and thrombosis was subsequently made [101]. Primary APS has also been described in which APA and thromboembolic complications are found in the absence of SLE or related autoimmune disease. APA have been shown to be a mixture of antibodies to various phospholipid binding proteins, including β2glycoprotein 1 (β2GP1) and prothrombin [102]. The spectrum of antibodies present and their epitope specificity appears to vary widely between APS patients. Many investigators have attempted to explain the conundrum of anticoagulant activity in laboratory tests and a prothrombotic condition in vivo. A number of mechanisms are proposed for the thrombotic effect of LAs. Anti-β2GP1 antibodies form bivalent complexes with β2GP1; these immune complexes exhibit increased affinity for negatively charged phospholipid surfaces and may disrupt reactions which take place on these surfaces [102]. Interference with the anticoagulant activity of APC has been suggested, and APA certainly cause acquired APC resistance in vitro [103]. However, other proposed mechanisms also include direct vascular damage, platelet activation, and disruption of the antithrombotic effects of annexin V [104]. Antiprothrombin antibodies may act by cross- reaction with plasminogen antibodies, impairing the fibrinolytic activity of plasminogen [105]; however, Simmelink et al. [106] showed no link between anti-plasminogen antibodies and thrombosis. Field et al. have suggested that antiprothrombin antibodies enhance prothrombin binding to phospholipid surfaces, and consequently increase thrombin generation [107]. There is evidence that the association of thrombosis is more closely linked to LA than to anticardiolipin, and therefore specific and sensitive assays for LA are of particular importance [108]. Several different laboratory methods are available for detection of LA activity, and it has also been suggested that LA with stronger activity in Dilute Russell’s Viper Venom Time (DRVVT) tests than Kaolin Clotting Time (KCT) tests have a stronger association with anti-β2GP1 and also with thrombosis [109]. Although a hereditary link has been described for this syndrome, which may be due to a susceptibility gene inherited in an autosomal dominant pattern [110], APS is still primarily considered an acquired disorder.

Thrombophilia

303

Acquired APCr APCr has been described in subjects who have tested negative for FVL or other FV mutations, and is associated with an increased risk of venous thrombosis [111–113]. Heritability of resistance to APC in the absence of FVL has also been demonstrated, although the genetic factors responsible have not been identified [114]. Tans et al. [113] recently described a thrombin generation-based APCr test which predicted venous thrombotic risk in populations without FVL and acquired APCr resulting from oral contraceptive use. Other acquired risk factors may be transient (trauma, surgery, immobilization (including long-distance air travel), pregnancy, altered hormonal status through oral contraceptive use or HRT), or may be secondary to malignancy, renal, or hepatic disease.

Risk of thromboembolic disease associated with thrombophilia Neither genetic nor acquired disorders cause thromboembolic disease in all affected individuals. The annual incidence of spontaneous VTE in patients with PC, PS, or AT deficiency was 0.8% in the study of Sanson et al. [115]. Several studies have explored the relative risk of thromboembolism for a variety of thrombophilic conditions. Study design and source population are major determinants of the risk. In a retrospective cohort study, Simioni [116] reported a relative risk of VTE of 10.6 for carriers of AT, PC, or PS deficiency in families with inherited thrombophilia. The relative risk in carriers of FVL was 2.5 compared to those without the defect. Martinelli [117] reported a relative risk of VTE of 8.1 for AT deficiency, 7.3 for PC deficiency, 8.5 for PS deficiency and 2.2 for FVL in first and second degree relative of individuals with thrombotic defects. Of note, however, is the dramatic and often synergistic increase in risk when more than one risk factor is present. Thromboembolic disease is now understood to be a multifactorial disorder, either through inheritance of multiple genetic defects or a combination of hereditary and acquired risk factors, and in common with mild hemophilias, where spontaneous bleeds are rarely encountered, presence of a single genetic abnormality may not be sufficient per se to cause a thrombotic event. In a study by Dumenco et al. [118], 16% of patients with a predisposing risk factor had combined defects. Zotz et al. [119] reported the risk of pregnancy-associated thrombosis with one defect (either FVL or PTA mutation) to be as low as 1 in 400, using this frequency to justify not screening pregnant subjects with no history of thrombophilia, but identified a moderate risk of pregnancyassociated thrombosis (1 in 25) where two genetic factors are present. Martinelli [120] reports a relative risk of pregnancy-associated thrombosis of 41.3 in homozygous FVL subjects and 9.2 in double heterozygotes for FVL and the P20210A mutation compared to those with neither defect. Some defects with a low independent thrombotic risk, for example, hyperhomocysteinemia, become associated with a significant increase in risk when in combination with a further risk factor, for example, pregnancy [121]. Olds et al. [122] employed the concept of a threshold over which the phenotype of a thrombotic event will be expressed; co-inheritance of more than one genotype predisposing to thrombosis, or interaction of genetic and acquired risk factors, will shift an individual’s liability over this threshold. Age is a major determinant in this equation—according to

Thrombosis in clinical practice

304

Rosendaal, three or four risk factors may need to be present for thrombosis to occur in children, whereas the presence of only two persistent or transient risk factors will almost certainly lead to thrombosis in those over 55 [123].

Testing for thrombophilia A global laboratory test to detect “increased risk of thrombosis,” akin to the routine coagulation screen would undoubtedly be of benefit, and may help distinguish subjects with similar plasma levels of hemostatic proteins but differing thrombotic histories. Global PC pathway screens have been employed, and in one study, Gemmati [124] used a global screen to distinguish subjects with symptomatic and asymptomatic PS deficiency. Several authors have identified increased thrombin generation in subjects with thrombophilia, and a move towards standardization of assays for endogenous thrombin potential may lead to a greater estimation of individual thrombotic risk [125]. Many laboratories carry out thrombophilia screening, particularly for hereditary defects (in a UK NEQAS for Blood Coagulation questionnaire in 1999, approximately 65,000 assays for PC, AT, and PS were performed annually in UK laboratories). Recently published guidelines on thrombophilia testing [2] and recommend assays for PC, PS, AT; modified APCr test (with FV deficient plasma); and polymerase chain reaction (PCR) for PTA, together with assays to detect antiphospholipid antibodies. Routine screening tests (PT, APTT, Thrombin Time (TT)) are recommended to screen for acquired abnormalities and sample artifact (e.g. treatment or contamination with heparin) and may also detect dysfibrinogenemias. A number of preanalytical variables should be considered by laboratories carrying out thrombophilia screening. Acute thrombosis can affect levels of circulating anticoagulants [126] and treatment of the thrombotic event with heparin or coumarins will also affect laboratory results. Samples should therefore ideally be collected at least 6 weeks after cessation of anticoagulant therapy. Samples for most studies should be anticoagulated with 0.105 M trisodium citrate, and plasma separated by double centrifugation. Plasma for homocysteine assay must be separated from red cells within 1 hour of collection. Cold centrifugation of whole blood may cause activation of platelets and clotting factors; therefore, room temperature centrifugation is recommended. PC, PS, and AT levels are stable in plasma stored at −24°C for up to 3 months, for longer term storage and for screening tests, −70°C or below is recommended [127]. It is particularly important that samples from both patients and reference ranges are collected and processed in identical fashion. Table 13.2 shows methods which may be employed in a thrombophilia screen. Individual laboratories should establish their own reference ranges, for the method in local use. A minimum of 40 and ideally 120 normal healthy subjects should be used to establish a reference range [128]. The effect of age, gender, and other preanalytical variables should be taken into account, both in the determination of reference ranges and the interpretation of individual patient results. There is considerable debate over the role and value of thrombophilia screening [129– 132]. Indiscriminate testing is expensive, and cost-benefit analyses have been used to rule out population screening for FVL [129, 133]. Screening of relatives of index cases with

Thrombophilia

305

the FVL mutation is also not cost-effective; in contrast, the higher clinical penetrance of PC, PS, and AT is reported to make screening of relatives financially worthwhile [4, 134]. Torrano-Masetti et al. [135] reported a high degree of psychological stress and anxiety, emotional hypersensitivity, and inadequate ability to deal with screening among subjects investigated for thrombophilic defects, although a contrary view was expressed through another study [136]. Importantly, subjects once investigated will be labeled with a defect, and several studies have shown laboratory testing to be suboptimal, with error rates as high as 10% for phenotypic tests, and 3–6% for genotypic tests [79, 137]; consequently, some of these subjects will be erroneously labeled with a defect. Rigorous quality control measures and appreciation of pre-analytical and assay variables are important to ensure the accuracy of laboratory diagnoses.

Treatment of thrombophilia The primary argument for not screening individuals for hereditary thrombophilia is that treatment of patients with identified defects may not differ from subjects with a similar clinical history but without a laboratory abnormality. Acute VTE is generally treated with unfractionated or low-molecular-weight heparin and oral anticoagulant therapy, irrespective of the presence or absence of a known thrombotic risk (inherited or acquired). Equal efficacy is reported for body weight-dependent dosing with lowmolecular-weight heparin and laboratory controlled therapy with unfractionated heparin [130]. Subjects with AT deficiency may be more resistant to heparin, and it is suggested that some subjects may benefit from additional treatment with AT concentrate [138]. However, resistance is not considered a problem in the majority of AT deficient patients [139]. The short half-life of PC, resulting in the incidence of warfarin-induced skin necrosis with PC deficiency, and more rarely PS deficiency [140], has prompted recommendation to ensure alternative anticoagulation during the induction of warfarin therapy [138], and reports of gradual induction with low-dose oral anticoagulants or supplementation with PC concentrate [138, 141]. In most cases, recommended intensity of warfarin therapy is also irrespective of diagnosis, and British Committee for Standards in Hematology (BCSH) guidelines recommend a target INR of 2.5 for subjects with VTE [142]. In some groups, recurrence of thrombosis while on warfarin indicates a need for a more aggressive anticoagulant regimen; a higher level of intensity is recommended if thrombosis recurrence occurs despite adequate control between an INR of 2.0–3.0 [142]. Several authors recommend a higher target INR (between 3.0 and 4.0) for subjects with LA, although prospective studies suggest a lower target is effective [143].

Thrombosis in clinical practice

306

Table 13.2 Laboratory testing for thrombophilia. Defect

1st line test

Confirmation/subtyping

Comments

AT

Amidolytic assay (bovine thrombin or Xa-based assay)

Immunological assay Genetic screening

Not all molecular defects detected equally by all functional assaysa

PC

Amidolytic assay Clotting assay

Immunological assay Genetic screening

Clotting assay affected by FVL and other variablesa

PS

Clotting assay or Free PS antigen

Free PS antigen (if clotting Clotting assay assay used) Total PS affected by FVL antigen Genetic screening and other variablesa

APCr

APC test without FVL

FVL PCR

FVL

APC test with FV – deficient plasma or DNA screening

Specific for FVL

P20210A

DNA screening

–

–

LA

Prolongation of a phospholipiddependent coagulation test (e.g. DRVVT)

Evidence of an inhibitor (mixing studies) Confirmation of the phospholipid-dependence (e.g. DRVVT with high concentration phospholipid)

Recommended guidelines for LA screeningb

Dysfibrinogenemia

PT

–

–

Acquired defects (hepatic disease/ heparin therapy/heparin contamination/oral anticoagulant therapy

APTT Fgn TT

–

–

FVIII, other procoagulant factors

Clotting assay

–

–

Hyperhomocysteinemia

Homocysteine assay (by FPIA, ELISA, HPLC)

–

Some centers also screen for MTHFR C677T mutation

May detect APCr in the absence of FVL—not specific for FVL

AT: antithrombin; PC: protein C; PS: protein S; FVL: factor V Leiden; APC: activated protein C;

Thrombophilia

307

APCr: APC resistance; PCR: polymerase chain reaction; LA: lupus anticoagulant; DRVVT: dilute Russell’s Viper venom clotting time; PT: prothrombin time; APTT: activated partial thromboplastin time; TT: thrombin time; FVIII: factor VIII; MTHFR: methylene tetrahydrofolate reductase. a Jennings I, Cooper P. Screening for thrombophilia: a laboratory perspective. Br J Biomed Set 2003; 60:39–51. b Greaves M, Cohen H, Machin SJ, Mackie I. Guidelines on the investigation and management of the antiphospholipid syndrome. Br J Hematol 2000; 109:704–15.

The benefits of long-term or indefinite anticoagulant therapy in reducing the risk of recurrent thromboembolism is reported to be offset by the increased risk of hemorrhage. Prins and Marchiori summarize the reduction in risk of recurrence as approximately 90% in subjects receiving prolonged therapy and an increase in clinically important bleeding of 1–4% per year [144]. A number of studies have attempted to identify the risk of recurrence related to specific thrombophilic defects. The recurrence rate for subjects with a first VTE varies with precipitating cause. Acquired risk factors such as surgery, trauma, or pregnancy are often rapidly resolved, and in these subjects the risk of recurrence is low [145, 146]. Baglin et al. have also shown that in unselected patients who had a first episode of VTE, testing for heritable thrombophilia did not predict risk of recurrence within 2 years of withdrawing anticoagulant therapy [146]. Prandoni has shown risk of recurrent thromboembolism to be greatest in patients with persistent residual thrombosis [147], and Palareti has shown that the risk of recurrence is higher in subjects with idiopathic or thrombophilia-induced thrombosis [148]. However, this risk is most accurately predicted by the presence of elevated D-dimer at specific time points after cessation of oral anticoagulation; the negative predictive value of recurrence for a normal D-dimer level at 3 months after cessation of therapy was over 95%. In a subsequent study, the same author showed that D-dimer was equally useful in predicting recurrent risk in subjects with unprovoked VTE irrespective of the presence of an inherited thrombophilic defect [149]. Thus, the prospect for monitoring D-dimer levels after cessation of therapy may allow assessment of risk and subsequent treatment strategies for individual patients rather than patient groups. Recent trials of low-intensity anticoagulant therapy showed a 64% reduction in recurrent VTE among patients receiving warfarin and controlled to an INR between 1.5 and 2.0 compared with those receiving placebo [150]. However, Kearon et al. found no advantage of low intensity compared to conventional warfarin therapy [151]. Development of novel anticoagulants may offer alternative treatments for patients with VTE, with the advantage of improved efficacy or reduced need for laboratory monitoring [152]. At present, therefore, long-term oral anticoagulant therapy is only recommended for subjects with recurrent spontaneous thrombosis; a single thrombosis at an unusual or lifethreatening site; or a single thrombosis with AT deficiency, LA or co-inheritance of two or more genetic defects [138]. These recommendations may change as findings of ongoing studies are reported, and some of the future prospects for treatment strategies are summarized in Table 13.3.

Thrombosis in clinical practice

308

Thromboprophylaxis BCSH guidelines suggest all patients with a past history of VTE irrespective of thrombophilic status should be considered for shortterm prophylaxis at periods of risk, such as surgery. It is possible that asymptomatic individuals with the same risk factor as an affected close relative may also benefit from short-term prophylaxis [2]. Although there is still uncertainty about the mechanism and thrombotic risk associated with hyperhomocysteinemia, reduction in homocysteine levels through vitamin supplementation is effective, and an ongoing vitamin

Table 13.3 Possible future strategies for treatment of thrombophilia. Strategy

Advantage

Example

Treatment with novel therapeutic agents

Improved efficacy Reduced laboratory monitoring

Direct thrombin inhibitors e.g. ximelagatran

Identification of patients at high risk of thrombosis or recurrence

Selective long-term treatment

D-dimer screening Other global screening tests Measurement of residual thrombosis

Treatment with low intensity anticoagulant regimens

Reduced bleeding risk Low-dose warfarin

intervention trial [153] may elucidate the value of such supplementation in the reduction of VTE risk. However, if supplements of folate, cobalamin, and vitamin B6 are to be given to patients with hyperhomocysteinemia, it is important to note that levels in over 50% of subjects will rise again if treatment is discontinued [154].

Conclusion The limited clinical utility and potential psychological impact of thrombophilia screening has led to recommendations to avoid indiscriminate testing of patients with a tentative clinical or family history [130, 132]. However, selective screening of patients and firstdegree relatives with strong thrombophilic histories may be of value, particularly in support of ongoing studies of potential future therapies. There is a general consensus in the literature for laboratories to employ the following panel of tests for thrombophilia: PC, PS, AT, FVL, PTA mutation, LA screening. Other tests may be employed, although the diagnostic utility is a matter of debate. If testing for thrombophilia is to be undertaken by a diagnostic laboratory, it is important that all variables and possible pitfalls are considered, to ensure an accurate diagnosis is made. Careful selection and validation of test methods should be undertaken by individual laboratories to ensure both specificity and sensitivity. Appropriate reference ranges should be employed, and quality assurance

Thrombophilia

309

should be maintained through in-house quality control measures and review of results in external quality assessment programs. It should be remembered that approximately 50% of subjects with idiopathic thrombosis have no identifiable laboratory abnormality. This may relate to an as yet undiscovered genetic defect, a summative effect of minor abnormalities, or defects which cannot be measured by current laboratory tests such as endothelial dysfunction. A continued search for a screening test for hypercoagulability, which will identify subjects at transient or permanent risk of thromboembolic disease, may assist in future therapeutic decisions. Treatment should follow published guidelines, and further developments may arise from ongoing studies and development of new therapies.

References [1] Heit JA, Silverstein MD, Mohr DN, Petterson TM, Lohse CM, O’Fallon WM, Melton LJ. The epidemiology of venous thromboembolism in the community. Thromb Haemost 2001;86: 452– 63. [2] Haemostasis and Thrombosis Task Force, British Committee for Standards in Haematology. Investigation and management of heritable thrombophilia. Br J Haematol 2001; 114:512–28. [3] Lane DA, Mannucci PM, Bauer KA, et al. Inherited Thrombophilia: Part 1. Thromb Haemost 1996; 76:651–62. [4] Faioni EM, Mannucci PM. Diagnostic testing for familial thrombophilia. In: Rowan RM, van Assendelft OW, Preston FE (eds), Advanced Laboratory Methods in Haematology. London: Arnold, 2002. [5] Mannucci PM, Poller L. Venous thrombosis and anticoagulant therapy. Br J Haematol 2001;114: 258–70. [6] Egeberg O. On the natural blood coagulation system. Investigations of inhibitor factors based on antithrombin deficient blood. Thromb Diath Haemorrh 1965; 14:473–89. [7] Bauer KA. Conventional fibrinolytic assays for the evaluation of patients with venous thrombosis: don’t bother. Thromb Haemost 2001; 85: 377–8. [8] Crowther MA, Kelton JG. Congenital thrombophilic states associated with venous thrombosis: a qualitative overview and proposed classification system. Ann Intern Med 2003; 138:128–34. [9] Esmon NI, Owen WG, Esmon CT. Isolation of a membrane-bound cofactor for thrombincatalysed activation of protein C. J Biol Chem 1982; 257:859–64. [10] Esmon CT. The Endothelial Protein C Receptor. Thromb Haemost 2000; 83:639–43. [11] Bertina RM. Protein C antigen and activity. In: Jespersen J, Bertina RM, Haverkate F (eds), Laboratory Techniques in Thrombosis: A Manual. The Netherlands: Kluwer Academic Publishers, 1999. [12] Mosnier LO, Meijers JC, Bouma BN. The role of protein S in the activation of thrombin activatable fibrinolysis inhibitor (TAFI) and regulation of fibrinolysis. Thromb Haemost 2001;86:1040–6. [13] Griffin JH, Evatt B, Zimmerman TS, Kleiss AJ, Wideman C. Deficiency of protein C in congenital thrombotic disease. J Clin Invest 1981;68: 1370–3. [14] Broekmans AW, Veltkamp JJ, Bertina RM. Congenital protein C deficiency and venous thromboembolism: a study of three Dutch families. N Engl J Med 1983; 309:340–4. [15] De Stefano V, Leone G, Mastrangelo S, et al. Clinical manifestations and management of inherited thrombophilia: retrospective analysis and follow up after diagnosis of 238 patients with congenital deficiency of antithrombin III, protein C, protein S. Thromb Haemost 1994; 72:352–8.

Thrombosis in clinical practice

310

[16] Esmon CT, Schwarz HP. An update on clinical and basic aspects of the protein C anticoagulant pathway. Trends Cardiovasc Med 1995; 5:141–8. [17] Bertina RM. Prevalence of hereditary thrombophilia and the identification of genetic risk factors. Fibrinolysis 1988; 2:7–13. [18] Miletich J, Sherman L, Broze G. Absence of thrombosis in subjects with heterozygous protein C deficiency. N Engl J Med 1987; 317:991–6. [19] Tait RC, Walker ID, Reitsma PH, et al. Prevalence of protein C deficiency in the healthy population. Thromb Haemost 1995; 73:87–93. [20] Hasstedt SJ, Bovill EG, Callas PW, Long GL. An unknown genetic defect increases venous thrombosis risk, through interaction with protein C deficiency. Am J Hum Genet 1998; 63:569– 76. [21] Reitsma PH. Protein C deficiency: from gene defects to disease. Thromb Haemost 1997; 78:344–50. [22] Marlar RA, Mastovich S. Hereditary protein C deficiency: a review of the genetics, clinical presentation, diagnosis and treatment. Blood Coagul Fibrinolysis 1990; 1:319–30. [23] Marlar RA, Adcock DM, Madden RM. Hereditary dysfunctional protein C molecules (type II): assay characterisation and proposed classification. Thromb Haemost 1990; 63:375–9. [24] Faoini EM, Hermida J, Rovida E, Razzari C, Asti D, Zeinali S, Mannucci PM. Type II protein C deficiency: identification and molecular modelling of two natural mutants with low anticoagulant and normal amidolytic activity. Br J Haematol 2000; 108:265–71. [25] Bertina RM. Protein S antigen. In: Jespersen J, Bertina RM, Haverkate F (eds), Laboratory Techniques in Thrombosis: A Manual. The Netherlands: Kluwer Academic Publishers, 1999. [26] Cumming AM, Shiach CR. The investigation and management of inherited thrombophilia. Clin Lab Haematol 1999; 21:77–92. [27] Lu D, Kalafatis M, Mann KG, Long GL. Comparison of activated protein C/protein Smediated inactivation of human factor VIII and factor V. Blood 1996; 87:4708–17. [28] De Fouw NJ, Haverkate F, Bertina RM, Koopman J, van Wijngaarden A, van Hinsbergh VWM. The cofactor role of protein S in the acceleration of whole blood clot lysis by activated protein C in vitro. Blood 1986; 67:1189–92. [29] Garcia de Frutos P, Alim RIM, Hardig Y, Zoller B, Dahlback B. Differential regulation of a and β chains of C4b-binding protein during acute-phase response resulting in stable plasma levels of free anticoagulant protein S. Blood 1994; 84:815–22. [30] Almasy L, Soria JM, Souto JC, Coll I, Bacq D, Faure A, Mateo J, Borrell M, Munoz X, Sala N, Stone WH, Lathrop M, Fontcuberta J, Blangero J. A quantitative trait locus influencing free plasma protein S levels on human chromosome 1q: results from the Genetic Analysis of Idiopathic Thrombophilia (GAIT) project. Arterioscler Thromb Vasc Biol 2003 1;23:508–11. [31] Comp PC, Doray D, Patton D, Esmon CT. An abnormal plasma distribution of protein S occurs in functional protein S deficiency. Blood 1985; 67: 504–8. [32] Zoller B, Garcia de Frutos P, Dahlback B. Evaluation of the relationship between protein S and C4b-binding protein isoforms in hereditary protein S deficiency demonstrating type I and type III deficiencies to be phenotypic variants of the same disease. Blood 1995; 85:3524–31. [33] Faioni EM, Valsecchi C, Palla A, Taioli E, Razzari C, Mannucci PM. Free protein S deficiency is a risk factor for venous thrombosis. Thromb Haemost 1997; 78:1343–6. [34] Pabinger I, Kyrle PA, Heistinger M, Eichinger S, Wittman E, Lechner K. The risk of thromboembolism in asymptomatic patients with protein C and protein S deficiency: a prospective cohort study. Thromb Haemost 1994; 71:441–5. [35] Borgel D, Gandrille S, Aiach M. Protein S deficiency. Thromb Haemost 1997; 78:351–6. [36] Comp PC, Esmon CT. Recurrent venous thromboembolism in patients with partial deficiency of protein S. N Engl J Med 1984; 311:1525. [37] Faioni EM, Franchi F, Asti D, Sacchi E, Bernadi F, Mannucci PM. Resistance to activated protein C in nine thrombophilic families: interference in a protein S functional assay. Thromb Haemost 1993; 70:1067–71.

Thrombophilia

311

[38] Gandrille S, Borgel D, Ireland H, Lane DA, Simmonds R, Reitsma PH, Mannhalter C, Pabinger I, Saito H, Suzuki K, Formstone C, Cooper DN, Espinosa Y, Sala N, Bernardi F, Aiach M. Protein S Deficiency: A database of mutations—for Plasma Coagulation Inhibitors Subcommittee of the Scientific and Standardisation Committee of the International Society on Thrombosis and Haemostasis. Thromb Haemost 1997; 77: 1201–14. [39] De Stefano V, Finazzi G, Mannucci PM. Inherited Thrombophilia: pathogenesis, clinical syndromes and management. Blood 1996; 87:3531–44. [40] Edson JR, Vogt JM, Huesman DA. Laboratory diagnosis of inherited protein S deficiency. Am J Clin Pathol 1990; 94:176–86. [41] Makris M, Leach M, Beauchamp NJ, Daly ME, Cooper PC, Hampton KK, Bayliss P, Peake IR, Miller GJ, Preston FE. Genetic analysis, phenotypic diagnosis, and risk of venous thrombosis in families with inherited deficiencies of protein S. Blood 2000; 95:1935–41. [42] D’Angelo A, Vigano-D’Angelo S, Esmon CT, Comp PC. Acquired deficiencies of protein S. J Clin Invest 1988; 81:1445–54. [43] Schuetze SM, Linenberger M. Acquired protein S deficiency with multiple thrombotic complications after orthotopic liver transplantation. Transplantation 1999; 67:1366–9. [44] Thompson JM. Coagulation and Haemostasis: A Practical Guide. Edinburgh: ChurchillLivingstone, 1991:45–70. [45] Brikhous KM, Smith HP, Warner ED, Seegers WH. Inhibition of blood clotting. An identified substance which acts in conjunction with heparin to prevent the conversion of prothrombin to thrombin. Am J Physiol 1939; 125:683. [46] Michiels JJ, Hamulyak K. Laboratory Diagnosis of hereditary thrombophilia. Semin Thromb Haemost 1998; 24:309–20. [47] Egeberg O. Inherited antithrombin deficiency causing thrombophilia. Thromb Diath Haemorrh 1965; 13:516–30. [48] Lane DA, Bayson T, Olds RJ, Fitches AC, Cooper DN, Millar DS, Jochmans K, Perry DJ, Okajima K, Thein SL and Emmerich J. Antithrombin mutation database: 2nd (1997) update for the plasma coagulation inhibitors subcommittee of the International Society on Thrombosis and Haemostasis. Thromb Haemost 1997; 77:197–211. [49] Mannucci PM, Boyer C, Wolf M, Tripodi A, Larrieu MJ. Treatment of congenital antithrombin III deficiency with concentrates. Br J Haematol 1982; 50:531–5. [50] Thaler E, Lechner K. Antithrombin III deficiency and thromboembolism. Clin Haematol 1981;10: 369–90. [51] Van der Meer FJM, Koster T, Vandenbroucke JP, Briet E, Rosendaal FR. The Leiden Thrombophilia Study. Thromb Haemost 1997; 78:631–5. [52] Tait RC, Walker ID, Perry DJ, et al. Prevalence of antithombin III deficiency subtypes in 4000 healthy blood donors. Thromb Haemost 1991; 65:839. [53] Vinazzer H. Hereditary and acquired Antithrombin deficiency. Semin Thromb Hemost 1999; 25:257–63. [54] Conard J. Antithrombin activity and antigen. In: Jespersen J, Bertina RM, Harerkate F. (eds), Laboratory Techniques in Thrombosis: A Manual. The Netherlands: Kluwer Academic Publishers, 1999. [55] White B, Berry D. Acquired antithrombin deficiency in sepsis. Br J Haematol 2001; 112:26– 31. [56] Kauffmann RH, Veltkamp JJ, Van Tilberg NH, Van ES LA. Acquired antithrombin III deficiency and thrombosis in the nephrotic syndrome. Am J Med 1978; 65:607–13. [57] Pogliani EM, Parma M, Baragetti I, Mostarda G, Rivolta F, Maffe P, Corneo G. LAsparaginase in acute lymphoblastic leukemia treatment: the role of human antithrombin III concentrates in regulating the prothrombotic state induced by therapy. Acta Haematologica 1995; 93:5–8. [58] Rosing J, Tans G. Coagulation factor V: an old star shines again. Thromb Haemost 1997; 78:427–33.

Thrombosis in clinical practice

312

[59] Nicolaes GA, Dahlback B. Factor V and thrombotic disease: description of a janus-faced protein. Arterioscler Thromb Vasc Biol 2002; 22:530–8. [60] Dahlback B, Carlsson M, Svenson PJ. Familial thrombophilia due to a previously unrecognised mechanism characterized by poor anticoagulant response to activated protein C: prediction of a cofactor to activated protein C. Proc Natl Acad Sci USA 1993; 90:1004–8. [61] Bertina RM, Koeleman BPC, Koster T, Rosendaal FR, Dirven RJ, De Ronde H, Van de Velden PA, Reitsma PH. Mutation in blood coagulation factor V associated with resistance to activated protein C. Nature 1994; 369:64–7. [62] Cox MJ, Rees DC, Martinson IJ, Clegg JB. Evidence for the single origin of factor V Leiden. Br J Haematol 1996; 92:1022–5. [63] Zivelin A, Griffin JH, Xiao X, Papinger I, Samama M, Conard J, Brenner B, Eldor A, Sejigsohn U. A single genetic origin for a common Caucasian risk factor for venous thrombosis. Blood 1997; 89:397. [64] Williamson D, Brown K, Ludington R, Baglin C, Baglin T. Factor V Cambridge: a new mutation (Arg 306 Thr) associated with resistance to activated protein C. Blood 1998;91:1140– 4. [65] Chan WP, Lee CK, Kwong YL, Lam CK, Liang R. A novel mutation of Arg306 of factor V gene in Hong Kong Chinese. Blood 1998; 91:1135–9. [66] Shen MC, Lin JS, Woel T. Factor V Arg306 Gly mutation is not associated with activated protein C resistance and is rare in Taiwanese Chinese. Thromb Haemost 2001; 85:270–3. [67] Amano K, Mchnick DA, Mousalli M, Kaufman RJ. Mutation at either Arg336 or Arg562 in factor VIII is insufficient for complete resistance to activated protein C (APC)-mediated inactivation: implications for the APC resistance test. Thromb Haemost 1998; 79:557–63. [68] Castaman G, Tosetto A, Ruggeri M, Rodeghiero F. Pseudohomozygosity for activated protein C resistance is a risk factor for venous thrombosis. Br J Haematol 1999; 106:232–6. [69] Colucci M, Ciavarella N, Giliberti MG, Semeraro N. Resistance to activated protein C (APC): influence of factor V levels. Thromb Haemost 1994; 72: 987–8. [70] Luddington R, Jackson A, Pannerselvam S, Brown K, Baglin T. The factor V R2 allele: risk of venous thromboembolism, factor V levels and resistance to activated protein C. Thromb Haemost 2000; 83:204–8. [71] Castoldi E, Brugge JM, Nicolaes GAF, Tans G, Rosing J. Impaired anticoagulant activity of factor V accounts for a large fraction of the APC resistance associated with the FV Leiden (R506Q) and R2 (H1299R) mutations. J Thromb Haemost 2003; 1(Suppl. 1):Abstract OC256. [72] Bernardi F, Faioni EM, Castoldi E, Lunghi B, Castaman G, Sacchi E, Mannucci PM. A factor V genetic component differing from factor V R506Q contributes to the activated protein C resistance phenotype. Blood 1997; 90:1552–7. [73] de Visser MC, Guasch JF, Kamphuisen PW, Vos HL, Rosendaal FR, Bertina RM. The HR2 haplotype of factor V: effects on factor V levels, normalized activated protein C sensitivity ratios and the risk of venous thrombosis. Thromb Haemost 2000; 83: 577–82. [74] Wilde JT, O’Sullivan JJ, Roper JL, Aerts P, Horowitz M, Navot N. A novel ELISA-based primer extension assay for the detection of the factor V Leiden mutation. Br J Haematol 1999; 106:427–30. [75] Cooper PC, Cooper SM, Smith JM, Kitchen S, Makris M. Evaluation of the Roche LightCycler: a simple and rapid method for direct detection of factor V Leiden and prothrombin G20210A genotypes from blood samples without the need for DNA extraction. Blood Coagul Fibrinolysis 2003; 14:499–503. [76] Keeney S, Salden A, Hay C, Cumming A. A whole blood, multiplex PCR detection method for Factor V Leiden and the prothrombin G20210A variant. Thromb Haemost 1999; 81:464–5. [77] Lutz CT, Foster PA, Noll WW, Voelkerding KV, Press RD, McGlennen RC, Kirschbaum NE. Multicenter evaluation of PCR methods for the detection of factor V Leiden (R506Q) genotypes. Clin Chem 1998; 44:1356–8.

Thrombophilia

313

[78] Hofgartner WT, Tait JE Frequency of problems during clinical molecular-genetic testing. Am J Clin Pathol 1999;112:14–21. [79] Preston FE, Kitchen S, Jennings I, Woods TA. A UK National External Quality Assessment scheme (UK NEQAS) for molecular genetic testing for the diagnosis of familial thrombophilia. Thromb Haemost 1999; 82:1556–7. [80] Tripodi A, Peyvandi F, Chantarangkul V, Menegatti M, Mannucci PM. Relatively poor performance of clinical laboratories for DNA analyses in the detection of two thrombophilic mutations—a cause for concern. Thromb Haemost 2002;88: 690–1. [81] Poort SR, Rosendaal FR, Reitsma PH, Bertina RM. A common variation in the 3′-untranslated region of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis. Blood 1996; 88:3698–703. [82] Tripodi A, Chantarangkul V, Mannucci PM. Hyperprothrombinaemia may result in acquired activated protein C resistance. Blood 2000;96: 3295–6. [83] Koenen RR, Tans G, van Oerle R, Hamulyak K, Rosing J, Hackeng TM. The APCindependent anticoagulant activity of protein S in plasma is decreased by elevated prothrombin levels due to the prothrombin G20210A mutation. Blood 2003; 102:1686–92. [84] Zivelin A, Rosenberg N, Faier S, Kornbrot N, Peretz H, Mannhalter C, Horellou MH, Seligsohn U. A single origin for the common prothrombotic G20210A polymorphism in the prothrombin gene. Blood 1998;92:1119–24. [85] Cummings AM, Keeney S, Salden A, Bhavnani M, Shwe KH, Hay CRM. The prothrombin gene G20210A variant: prevalence in a U.K. anticoagulant clinic population. Br J Haematol 1998; 98:353–5. [86] Brown K, Luddington R, Williamson D, Baker P, Baglin T. Risk of venous thromboembolism associated with a G to A transition at position 20210 in the 3′-untranslated region of the prothrombin gene. Br J Haematol 1997; 98:907–9. [87] Cunningham MT, Brandt JT, Laposata M, Olson JD. Laboratory diagnosis of dysfibrinogenemia. Arch Pathol Lab Med 2002; 126:499–505. [88] Makris M. Hyperhomocysteinemia and thrombosis. Clin Lab Haematol 2000; 22:133–43. [89] Cattaneo M. Hyperhomocysteinemia, atherosclerosis and thrombosis. Thromb Haemost 1999;81: 165–76. [90] Eberhardt RT, Forgione MA, Cap A, Leopold JA, Rudd MA, Trolliet M, Heydrick S, Stark R, Klings ES, Moldovan NI, Yaghoubi M, Goldschmidt-Clermont PJ, Farber HW, Cohen R, Loscalzo J. Endothelial dysfunction in a murine model of mild hyperhomocyst(e)inemia. J Clin Invest 2000; 106:483–91. [91] Sauls DL, Wolberg AS, Hoffman M. Elevated plasma homocysteine leads to alterations in fibrin clot structure and stability: implications for the mechanism of thrombosis in hyperhomocysteinemia. J Thromb Haemost 2003; 1:300–6. [92] Martinelli I, Bucciarelli P, Zighetti ML, Cafro A, Mannucci PM. Low risk of thrombosis in family members of patients with hyperhomocysteinaemia. Br J Haematol 2002;117:709–11. [93] Martinelli I. Risk factors in venous thromboembolism. Thromb Haemost 2001; 86:395–403. [94] Schambeck CM, Hinney K, Haubitz I, Mansouri Taleghani B, Wahler D, Keller F. Familial clustering of high factor VIII levels in patients with venous thromboembolism. Arterioscler Thromb Vasc Biol 200121:289–92. [95] Lavigne G, Mercier E, Quere I, Dauzat M, Gris J-C Thrombophilic families with inheritably associated high levels of coagulation factors VIII, IX and XI. J Thromb Haemost 2003; 1:2134– 9. [96] O’Donnell J, Laffan M. Elevated plasma factor VIII levels—a novel risk factor for venous thromboembolism. Clin Lab 2001; 47:l-6. [97] Mariani G, Liberti G, D’Angelo T, Lo Coco L. Factor VII activity and antigen. In: Jespersen J, Bertina RM, Haverkate F (eds), Laboratory Techniques in Thrombosis: A Manual. The Netherlands: Kluwer Academic Publishers, 1999.

Thrombosis in clinical practice

314

[98] Mannucci PM, Tripodi A. Factor VIII clotting activity. In: Jespersen J, Bertina RM, Haverkate F (eds), Laboratory Techniques in Thrombosis: A Manual. The Netherlands: Kluwer Academic Publishers, 1999. [99] Pabinger I, Allaart CF, Hermans J, Briet E, Bertina RM. Hereditary protein C-deficiency: laboratory values in transmitters and guidelines for the diagnostic procedure. Report on a study of the SSC Subcommittee on Protein C and Protein S. Protein C Transmitter Study Group. Thromb Haemost 1992; 68:470–4. [100] Conley CL, Hartmann RC. A hemorrhagic disorder caused by circulating anticoagulant in patients with disseminated lupus erythematosus. J Clin Invest 1952; 31:621–2. [101] Bowie EJW, Thompson JH, Pascuzzi CA, Owen CA. Thrombosis in systemic lupus erythematosus despite circulating anticoagulants. J Lab Clin Med 1963; 62:416–30. [102] Arnout J. Antiphospholipid syndrome. Diagnostic aspects of lupus anticoagulants. Thromb Haemost 2001; 86:83–91. [103] Gennari L, Blanco A, Alberto MF, Grosso S, Lazzari MA. The concomitant presence of lupus anticoagulant, anticardiolipin and anti-beta2-glycoprotein 1 antibodies could be associated with acquired activated protein C resistance in non-systemic lupus erythematosus patients. Br J Haematol 2003; 121:527–9. [104] Rand JH. Molecular pathogenesis of the antiphospholipid syndrome. Circ Res 2002;90: 29– 37. [105] Puurunen M, Manttari M, Manninen V, Palosuo T, Vaarala O. Antibodies to prothrombin crossreact with plasminogen in patients developing myocardial infarction. Br J Haematol 1998; 100:374–9. [106] Simmelink MJ, De Groot PG, Derksen RH. A study on associations between antiprothrombin antibodies, antiplasminogen antibodies and thrombosis. J Thromb Haemost 2003; 1: 735–9. [107] Field SL, Hogg PJ, Daly EB, Dai YP, Murray B, Owens D, Chesterman CN. Lupus anticoagulants form immune complexes with prothrombin and phospholipid that can augment thrombin production in flow. Blood 1999; 94:3421–31. [108] Galli M, Luciani D, Bertolini G, Barbui T. Lupus anticoagulants are stronger risk factors for thrombosis than anticardiolipin antibodies in the antiphospholipid syndrome: a systematic review of the literature. Blood; 101:1 827–32. [109] Galli M, Dlott J, Norbis F, Ruggeri L, Cler L, Triplett DA, Barbui T. Lupus anticoagulants and thrombosis: clinical association of different coagulation and immunologic tests. Thromb Haemost 2000;84:1012–16. [110] Goel N, Ortel TL, Bali D, Anderson JP, Gourley IS, Smith H, Morris CA, DeSimone M, Branch DW, Ford P, Berdeaux D, Roubey RA, Kostyu DD, Kingsmore SF, Thiel T, Amos C, Seldin MF. Familial antiphospholipid antibody syndrome: criteria for disease and evidence for autosomal dominant inheritance. Arthritis Rheum 1999; 42:318–27. [111] Rodeghiero F, Tosetto A. Activated protein C resistance and factor V Leiden mutation are independent risk factors for venous thromboembolism. Ann Intern Med 1999; 130:643–50. [112] de Visser MC, Rosendaal FR, Bertina RM. A reduced sensitivity for activated protein C in the absence of factor V Leiden increases the risk of venous thrombosis. Blood 1999; 93:1271–6. [113] Tans G, van Hylckama Vlieg A, Thomassen MC, Curvers J, Bertina RM, Rosing J, Rosendaal FR. Activated protein C resistance determined with a thrombin generation-based test predicts for venous thrombosis in men and women. Br J Haematol 2003; 122:465–70. [114] Tosetto A, Castaman G, Cappellari A, Rodeghiero F. The VITA Project: heritability of resistance to activated protein C. Vincenza Thrombophilia and Arteriosclerosis. Thromb Haemost 2000; 84:811–14. [115] Sanson BJ, Simioni P, Tormene D, Moia M, Friederich PW, Huisman MV, Prandoni P, Bura A, Rejto L, Wells P, Mannucci PM, Girolami A, Buller HR, Prins MH. The incidence of venous thromboembolism in asymptomatic carriers of a deficiency of antithrombin, protein C, or protein S: a prospective cohort study. Blood 1999;94: 3702–6.

Thrombophilia

315

[116] Simioni P, Sanson BJ, Prandoni P, Tormene D, Friederich PW, Girolami B, Gavasso S, Huisman MV, Buller HR, Wouter ten Cate J, Girolami A, Prins MH. Incidence of venous thromboembolism in families with inherited thrombophilia. Thromb Haemost 1999;81: 198– 202. [117] Martinelli I, Mannucci PM, De Stefano V, Taioli E, Rossi V, Crosti F, Paciaroni K, Leone G, Faioni EM. Different risks of thrombosis in four coagulation defects associated with inherited thrombophilia: a study of 150 families. Blood 1998; 92:2353–8. [118] Dumenco LL, Blair AJ, Sweeney JD. The results of diagnostic studies for thrombophilia in a large group of patients with a personal or family history of thrombosis. Am J Clin Pathol 1998;110: 673–82. [119] Zotz RB, Gerhardt A, Scharf RE. Prediction, prevention, and treatment of venous thromboembolic disease in pregnancy. Semin Thromb Hemost 2003; 29:143–54. [120] Martinelli I, Legnani C, Bucciarelli P, Grandone E, De Stefano V, Mannucci PM. Risk of pregnancy-related venous thrombosis in carriers of severe inherited thrombophilia. Thromb Haemost 2001; 86:800–3. [121] Raziel A, Kornberg Y, Friedler S, Schachter M, Sela BA, Ron-El R. Hypercoagulable thrombophilic defects and hyperhomocysteinemia in patients with recurrent pregnancy loss. Am J Reprod Immunol 2001; 45:65–71. [122] Olds RJ, Fitches AC, Geary CP. The multigenic basis for venous thrombosis. Br J Haematol 2000; 109:508–11. [123] Rosendaal FR. Thrombosis in the young: epidemiology and risk factors. A focus on venous thrombosis. Thromb Haemost 1997; 78:1–6. [124] Gemmati D, Serino ML, Tognazzo S, Ongaro A, Moratelli S, Gilli G, Forini E, De Mattei M, Scapoli GL. The reduced sensitivity of the ProC Global test in protein S deficient subjects reflects a reduction in the associated thrombotic risk. Blood Coagul Fibrinolysis 2001; 12:691– 7. [125] Lawrie AS, Gray E, Leeming D, Davidson SJ, Purdy G, Iampietro R, Craig S, Rigsby P, Mackie IJ. A multicentre assessment of the endogenous thrombin potential using a continuous monitoring amidolytic technique. Br J Haematol 2003; 123:335–41. [126] Reiter W, Ehrensberger H, Steinbruckner B, Keller F. Parameters of haemostasis during acute venous thrombosis. Thromb Haemost 1995; 74: 596–601. [127] Woodhams B, Girardot O, Blanco MJ, Colesse G, Gourmelin Y. Stability of coagulation proteins in frozen plasma. Blood Coagul Fibrinolysis 2001; 12:229–36. [128] Walker ID. Reference ranges in haemostasis. In: Rowan RM, van Assendelft OW, Preston FE (eds), Advanced Laboratory Methods in Haematology. London: Arnold, 2002. [129] Mannucci PM. Genetic hypercoagulability: prevention suggests testing family members. Blood 2001;98:21–2. [130] Greaves M, Baglin T. Laboratory testing for heritable thrombophilia: impact on clinical management of thrombotic disease. Br J Haematol 2000; 109:699–703. [131] Martinelli I. Pros and cons of thrombophilia testing: pros. J Thromb Haewost 2003; 1:410– 11. [132] Machin SJ. Pros and cons of thrombophilia testing: cons. J Thromb Haemost 2003; 1:412–13. [133] Kalev M, Day T, Van de Water N, Ockelford P. Screening for a prothrombotic diathesis in patients attending family planning clinics. N Z Med J 1999; 112:358–61. [134] De Stefano V, Rossi E, Paciaroni K, Leone G. Screening for inherited thrombophilia: indications and therapeutic implications. Haematologica 2002; 87:1095–108. [135] Torrano-Masetti L, Rizzatti EG, Oliveira LCO, Inácio RA, Santos MA, Franco RF. Psychologic impact of thrombophilia screening in asymptomatic relatives of carriers of prothrombotic mutations. J Thromb Haemost 2003; 1(Suppl. 1): Abstract OC322. [136] Legnani C, Razzaboni E, Gremigni P, Ricci Bitti PE, Palareti G. Emotional impact of testing for inherited thrombophilia. J Thromb Haemost 2003; 1(Suppl. 1):Abstract OC324.

Thrombosis in clinical practice

316

[137] Jennings I, Kitchen S, Woods TAL, Preston FE. Diagnostic errors in thrombophilia screening: Evidence from the UK National Quality Assessment Scheme (UK NEQAS). Thromb Haemost 1999; (Suppl. 1):276. [138] Bauer KA. Management of thrombophilia. J Thromb Haemost 2003; 1:1429–34. [139] Schulman S, Tengborn L. Treatment of venous thromboembolism in patients with congenital deficiency of antithrombin III. Thromb Haemost 1992; 68:634–6. [140] Sallah S, Abdallah JM, Gagnon GA. Recurrent warfarin-induced skin necrosis in kindreds with protein S deficiency. Haemostasis 1998;28: 25–30. [141] De Stefano V, Mastrangelo S, Schwarz HP, Pola P, Flore R, Bizzi B, Leone G. Replacement therapy with a purified protein C concentrate during initiation of oral anticoagulation in severe protein C congenital deficiency. Thromb Haemost 1993; 70:247–9. [142] British Committee for Standards in Haematology. Guidelines on oral anticoagulation: third edition. Br J Haematol 1998; 101:374–87. [143] Meroni PL, Moia M, Derksen RH, Tincani A, McIntyre JA, Arnout JM, Koike T, Piette JC, Khamashta MA, Shoenfeld Y. Venous thromboembolism in the antiphospholipid syndrome: management guidelines for secondary prophylaxis. Lupus 2003; 12:504–7. [144] Prins MH, Marchiori A. Risk of recurrent venous thromboembolism—expanding the frontier. Thromb Haemost 2002;87:1–3. [145] Baglin T, Luddington R, Brown K, Baglin C. Incidence of recurrent venous thromboembolism in relation to clinical and thrombophilic risk factors: prospective cohort study. Lancet 2003; 362:523–6. [146] Prandoni P. Long-term clinical course of proximal deep venous thrombosis and detection of recurrent thrombosis. Semin Thromb Hemost 2001; 27:9–13. [147] Prandoni P, Lensing AW, Prins MH, Bernardi E, Marchiori A, Bagatella P, Frulla M, Mosena L, Tormene D, Piccioli A, Simioni P, Girolami A. Residual venous thrombosis as a predictive factor of recurrent venous thromboembolism. Ann Intern Med 2002; 137:955–60. [148] Palareti G, Legnani C, Cosmi B, Guazzaloca G, Pancani C, Coccheri S. Risk of venous thromboembolism recurrence: high negative predictive value of D-dimer performed after oral anticoagulation is stopped. Thromb Haemost 2002; 87:7–12. [149] Palareti G, Legnani C, Cosmi B, Valdre L, Lunghi B, Bernardi F, Coccheri S. Predictive value of D-dimer test for recurrent venous thromboembolism after anticoagulation withdrawal in subjects with a previous idiopathic event and in carriers of congenital thrombophilia. Circulation 2003; 108:313–18. [150] Ridker PM, Goldhaber SZ, Danielson E, Rosenberg Y, Eby CS, Deitcher SR, Cushman M, Moll S, Kessler CM, Elliott CG, Paulson R, Wong T, Bauer KA, Schwartz BA, Miletich JP, Bounameaux H, Glynn RJ; PREVENT Investigators. Long-term, low-intensity warfarin therapy for the prevention of recurrent venous thromboembolism. N Engl J Med 2003; 348:1425–34. [151] Kearon C, Ginsberg JS, Kovacs MJ, Anderson DR, Wells P, Julian JA, MacKinnon B, Weitz JI, Crowther MA, Dolan S, Turpie AG, Geerts W, Solymoss S, van Nguyen P, Demers C, Kahn SR, Kassis J, Rodger M, Hambleton J, Gent M; Extended Low-Intensity Anticoagulation for ThromboEmbolism Investigators. Comparison of low-intensity warfarin therapy with conventional-intensity warfarin therapy for long-term prevention of recurrent venous thromboembolism. N Engl J Med 2003; 349:631–9. [152] Eriksson H, Wahlander K, Gustafsson D, Welin LT, Frison L, Schulman S; Thrive Investigators. A randomized, controlled, dose-guiding study of the oral direct thrombin inhibitor ximelagatran compared with standard therapy for the treatment of acute deep vein thrombosis: THRIVE I. J Thromb Haemost 2003, 1:41–7. [153] Willems HP, den Heijer M, Bos GM. Homocysteine and venous thrombosis: outline of a vitamin intervention trial. Semin Thromb Hemost 2000; 26:297–304. [154] Santi R, Demicheli M, Contino L, Fiorini T, Levis A. Homocysteine plasma levels after suspension of vitamin treatment. J Thromb Haemost 2003; 1:1330–2.

14 Systemic thrombosis in children M Patricia Massicotte, Paul Monagle, and Anthony K Chan Introduction Systemic arterial and venous thromboses in children are being increasingly diagnosed as a consequence of clinical advances in the therapy of primary illnesses in children (e.g. cancer, trauma, congenital heart disease). The use of arterial and venous catheters for therapeutic and diagnostic purposes are in part responsible for the decreased mortality and morbidity associated with these primary illnesses. However, intravascular catheter placement has resulted in systemic arterial and venous thromboembolic events, which can independently cause mortality and significant morbidity. In this review, the epidemiology, evidencebased diagnostic and therapeutic recommendations with respect to catheter-related systemic arterial thromboembolic events (SATE) and systemic venous thromboembolic events (SVTE) will be discussed. A Medline search of the medical literature from 1973 to present was undertaken to identify relevant publications. Diagnostic recommendations are based on the sensitivity and specificity of the chosen diagnostic method. Therapeutic recommendations for thrombosis are based on the strength of the published evidence supporting the recommendation. Therapeutic recommendations for thrombosis are graded according to the strength of the evidence supporting the recommendation. The grading system used is that described by the American College of Chest Physicians for their antithrombotic guidelines [1]. Grade 1 recommendations have a clear benefit for the intervention in the patient population, while Grade 2 recommendations are far less certain. The recommendations are supported by the level of evidence: A—multiple randomized trials with clear treatment effect or a meta-analysis; B—single randomized trials; and C— observational studies.

Catheter-related systemic arterial thromboembolism Epidemiology The most common type of arterial thrombosis in children occurs as a result of placement of arterial catheters. Non-catheter-related arterial thrombosis may be congenital (familial hyperlipidemia and hyperhomocysteinemia) or acquired (Takayasu’s arteritis, Kawasaki’s disease, congenital heart disease, and arterial thrombosis in transplanted organs) and has been reviewed recently [1, 2]. This chapter will focus on arterial thrombosis secondary to intravascular access.

Thrombosis in clinical practice

292

There are three types of arterial catheterizations that are used in children, which may result in arterial thrombosis: umbilical arterial catheterization in neonates (recently reviewed and will not be discussed in this chapter) [1–3], cardiac catheterization, and peripheral catheterization. Arterial occlusion causes tissue ischemia resulting in tissue necrosis if the vascular occlusion persists. Organ and tissue damage at remote sites may result from embolic events occurring due to fragmentation of the original thrombus. Cardiac catheterization Children with congenital and acquired heart disease often require cardiac catheterization (CC) for diagnostic or therapeutic purposes. Most commonly, the superficial femoral artery is cannulated for the procedure although the brachial artery is occasionally accessed [4]. Adverse events occurring as a result of CC include arterial spasm and arterial thrombosis. Clinically, vascular spasm and thrombosis are indistinguishable in the initial phases of presentation with the following symptoms: decreased or absent pulses, pale or mottled limb, and decreased capillary refill. Arterial spasm usually resolves within a few hours in the absence of therapy while arterial thrombosis usually resolves following therapy (heparin, thrombolytic therapy, embolectomy). The incidence of femoral artery thrombosis following CC without thromboprophylaxis is approximately 40% [5]. The incidence is inversely proportional to patient age and weight, with infants at highest risk [5]. Other factors which affect the incidence of thrombosis are duration of procedure, catheter to vessel diameter, the use of balloon dilatation, larger sheath size than recommended for a given weight and body surface area, repeated catheter manipulations, and increased hematocrit [5–10]. Prophylactic recommendations for cardiac catheterizations During cardiac catheterization, children should receive intravenous unfractionated heparin (UFH) therapy as a bolus of 100–150 U/kg during CC (Grade 1A). The use of ASA (aspirin) alone for thromboprophylaxis for cardiac catheterization should not be used (Grade 1B). Evidence. The use of heparin as a bolus dose of 100–150 U/kg has been shown to significantly decrease the incidence of femoral artery thrombosis following CC in children from 40% to 8% [5]. Aspirin has not been shown to be efficacious for the prevention of arterial thrombosis secondary to CC [11]. Peripheral artery catheterization Children in an intensive care setting require peripheral artery catheterization for monitoring as well as blood sampling [12]. Thrombotic occlusion of the vessel and loss of patency of the arterial catheter can result from endothelial injury as a result of catheter placement [13]. The clinical presentation of peripheral artery thrombosis includes decreased temperature and color of the hand or foot, and increased capillary refill time.

Systemic thrombosis in children

293

Prophylactic recommendations for peripheral artery catheterization Continuous heparin of 1 unit/ml infused through a peripheral arterial catheter improves catheter patency (Grade 1B). Evidence. A number of clinical studies have demonstrated increased duration of catheter patency with UFH infusion [10, 14–17]. Patency is not necessarily related to the presence or absence of large vessel thrombosis, and at this time there are no specific studies which address the issue of prophylaxis for peripheral arterial thrombosis secondary to arterial vascular access. Diagnosis Angiography is accepted as the gold standard test to diagnose arterial thrombosis [18, 19]. The newer contrast agents are less hypertonic than those used previously, and thus allowing safer use in younger children. However, the use of angiography is rarely possible as most children with peripheral arterial thrombosis are extremely ill in an intensive care setting. Other objective tests most commonly used are the absence of pulses using Doppler ultrasound (DUS) and a blood pressure differential between two limbs of at least 10 mmHg [18]. Hand-held DUS can be used at the bedside to confirm the presence of arterial thrombosis. Diagnostic recommendations for arterial thrombosis Angiography is the gold standard technique with good sensitivity and specificity, and should be used, if possible, to diagnose arterial thrombosis. There are no studies confirming the sensitivity and specificity of other diagnostic techniques (DUS, blood pressure differential) in children. Outcome Arterial thrombosis has serious short- and longterm sequelae, including mortality and morbidity. The short-term outcome may include impaired viability in the involved limb resulting in loss of function and occasionally, surgical limb amputation. The long-term outcome may include leg length discrepancy, muscle wasting, claudication, and loss of arterial access, which is important in children requiring multiple cardiac catheterizations [20, 21]. The occlusion of the radial artery rarely results in loss of the hand if the ulnar artery is patent. Treatment The catheter should be removed immediately if arterial occlusion is suspected (pale or cyanosed skin, decreased or absent pulses, increased capillary time).

Thrombosis in clinical practice

294

Therapeutic recommendations Children with confirmed arterial thrombosis should be treated with therapeutic doses of heparin (Grade 1C). For life- or limbthreatening arterial thrombosis, in the absence of contraindications, the use of thrombolytic therapy should be considered (Grade 1C). The treatment options for arterial thrombosis includes anticoagulation, thrombolytic therapy, embolectomy, and reconstructive surgery. Anticoagulation Anticoagulation with UFH has been reported to be effective therapy in 70% of arterial thrombi without the use and associated risks of thrombolytic therapy, embolectomy, or reconstructive surgery [22], The use of UFH is initiated with a bolus or as a continuous infusion depending on UFH exposure in the previous 4 hours. Heparin nomograms have been validated and published for use in children (see Table 14.1). Thrombolytic therapy If the viability of the limb is in question, thrombolytic therapy is often initiated. There

Table 14.1 Dosing nomogram for heparin therapy. APPT

Anti-factor Xa

Bolus

Hold

Rate

Repeat

W

(units/mL)

(units/kg)

(min)

change (%)

APTT(h)

<45

<0.1

50

0

Increase 20

4

45–54

0.1–034

0

0

Increase 10

4

55–80

035–0.70

0

0

0

24

81–90

0.71–0.89

0

0

Decrease 10

4

91–115

0.90–1.20

0

30

Decrease 10

4

>1.20

0

60

Decrease 15

4

>115 Adapted from [1],

are few studies using thrombolytic therapy post-cardiac catheterization, all with small sample sizes (n=6–17). The use of streptokinase (SK) [22–26], urokinase (UK) [22] and tissue plasminogen activator (tPA) [27–29] have resulted in successful outcomes in 58– 100% of children. Gupta et al. reported a high rate of major bleeding secondary to thrombolytic therapy in children [30]. However, recent reports suggest that this risk can be reduced by pretreating children about to receive tPA with 10 ml/kg fresh frozen plasma (FFP), and limiting the tPA infusion (0.5 mg/kg/h) to 6 hours maximum.

Systemic thrombosis in children

295

Embolectomy The risk of arterial reocclusion in small children following embolectomy is high. If arterial occlusion persists despite UFH, small children with arterial occlusion should receive thrombolytic therapy if no contraindications are present (active bleeding, disseminated intravascular coagulation, central nervous system bleed). UFH therapy should be initiated post-embolectomy to prevent vascular reocclusion. Reconstructive surgery The main indication for reconstructive surgery is clinically significant claudication accompanied by muscle wasting and/or shortening of the limb. There are a few options for reconstructive surgery, including thrombectomy with autogenous saphenous vein patch angioplasty, direct angioplasty, segmental resection with end to end anastomosis and interposition bypass grafting [31]. The vascular anatomy of the affected leg should be established prior to surgery using DUS and possibly angiography via the contralateral femoral artery.

Systemic venous thrombosis Epidemiology Incidence There are several unique mechanisms, which protect children from SVTE. These include a reduced capacity to generate thrombin [32, 33],d increased capacity of α2macroglobulin to inhibit thrombin [34], and enhanced antithrombotic potential by the vessel wall [35, 36]. As a result, compared to adults, children have a significantly decreased incidence of SVTE, 5.3/ 10,000 hospital admissions [37–42] versus 2.5–5%, respectively [43–45]. The epidemiology of SVTE has been studied in international registries [46–48], which highlighted that SVTE occurs as a secondary complication to an underlying disorder, such as cancer, trauma/surgery, congenital heart disease, and systemic lupus erythematosus (SLE). Infants less than 1 year of age and teenagers are at greatest risk for development of SVTE [46, 49]. The most frequent non-CVL (central venous line) associated venous thromboembolism (VTE) is in the lower limb [46]. Recurrent VTE has been estimated to occur in 6% of children with VTE [50]. Clinical symptoms The clinical symptoms and complications of SVTE can be classified as acute or long term. The acute clinical symptoms include loss of CVL patency, swelling, pain, and discoloration of the related limb, swelling of the face and head with superior vena cava

Thrombosis in clinical practice

296

syndrome, and respiratory compromise with pulmonary embolism (PE) (Figures 14.1– 14.4). The long-term complications include prominent collateral circulation in the skin (face, back, chest, and neck as sequelae of upper venous VTE, and abdomen, pelvis, groin, and legs as sequelae of lower venous VTE), repeated loss of CVL patency, repeated requirement for CVL replacement, eventual loss of venous access, CVL-related sepsis, chylothorax, chylopericardium, recurrent VTE necessitating long-term anticoagulation and its risk of bleeding, and post-thrombotic syndrome (PTS). Reports in the literature associate the development of right atrial thrombosis (RAT) with the presence of a CVL [51, 52]. The incidence of

Figure 14.1 Collateral venous vessels on face and neck of a child following upper venous system thrombosis.

Figure 14.2 Collateral vessels on right anterior shoulder and arm secondary to superior vena cava thrombosis. RAT in children is unknown. Clinically overt symptoms of RAT include PE, loss of CVL patency, and persistent sepsis, including endocarditis. Etiology The most common risk factor for SVTE appears to be the presence of a CVL (acquired) although most children have several concomitant risk factors [46–48].

Systemic thrombosis in children

297

Figure 14.3 Venous collateral vessels on abdomen following femoral thrornbosis.

Figure 14.4 Superior vena cava syndrome secondary to upper venous system thrombosis.

Thrombosis in clinical practice

298

Acquired risk factors Over 50% of VTEs in children occur in the upper venous system secondary to the use of CVLs [46, 49]. Three types of CVL-related VTE are described in the literature; clots at the tips of CVLs, which impair infusion or withdrawal of blood, fibrin sleeves that are not adherent to vessel walls but may occlude CVLs (Figure 14.5) [53], and CVL-related thrombi that adhere to vessel walls, with partial or complete obstruction of vessels in which the CVL is located (Figure 14.6) [53]. There are a number of mechanisms which may be contributory to the development of CVL-related SVTE, including damage to the vessel wall by the CVL or by substances infused through the CVL (total parenteral nutrition (TPN), chemotherapy) [54, 55], disrupted blood flow due to the presence of the CVL, and thrombogenic catheter materials [56]. The use of CVLs occurs most commonly in children who require short-term intensive care, hemodialysis or long-term supportive care (TPN or chemotherapy). In published studies, the reported incidence of CVL-related SVTE is dependent upon the method of diagnosis, underlying illnesses present in the cohort and the clinical suspicion. As an example, in cohorts of children receiving longterm TPN, the incidence of CVL-related VTE is reported as 1 % based on clinical diagnosis [51] and increases to 35%, if based on ventilation perfusion scans (VQ) or echocardiography, and 75% if based on venography [57].

Figure 14.5 CVL-related VTE: clot at the tip of CVL and sheath surrounding catheter.

Systemic thrombosis in children

299

Figure 14.6 CVL-related VTE types: proximal vessel VTE. Symptomatic versus asymptomatic CVL-related SVTE In adults, asymptomatic SVTE are not necessarily treated and thrombosis experts argue that these thrombi are probably inconseprevent embolism and extension. However, in quential. Symptomatic thrombi are treated to children, asymptomatic SVTE, which are radiographically detected, are of clinical importance for a number of reasons. First, there is increasing evidence supporting the relationship between CVL-related sepsis and CVLrelated SVTE [58]. Second, PE in children is frequently not diagnosed during life due to the subtle symptoms and the presence of other illnesses, which can cause sudden respiratory compromise. However, most commonly, PE results from CVL-related SVTE [58] and can cause death [41]. Third, in a child with a patent foramen ovale and SVTE, stroke may occur as a result of the right to left intracardiac connection. Fourth, the longterm consequences of CVL-related SVTE may be significant and include death and loss of venous access. Sudden death has been described in a few case reports from rupture of an intrathoracic collateral vessel thought to have resulted from previous CVL placement years earlier [59, 60]. Loss of venous access is a devastating complication of CVL-related SVTE in children requiring vascular access for organ transplant, lifelong TPN or dialysis. Therapeutic recommendations for asymptomatic thrombosis In asymptomatic children when a proximal thrombosis is diagnosed during a radiographic procedure completed for other reasons (diagnosis of malignancy, echocardiography to determine cardiac anatomy or function, cardiac catheterization), in the absence of contraindications, anticoagulation should be strongly considered. (Expert opinion) If anticoagulation is used, the agent, dosing regimen, and duration should be as per first time SVTE (see later) (Grade 1C).

Thrombosis in clinical practice

300

Diagnosis Upper venous system thrombosis—asymptomatic SVTE The most sensitive diagnostic method for diagnosis of asymptomatic intrathoracic thrombosis was demonstrated to be venography when compared to ultrasound in a cohort of children with upper venous CVLs and acute lymphoblastic leukemia (PAARKA study) who were screened for SVTE. However, ultrasound is more sensitive in diagnosing jugular vein thrombosis compared to venogram [61]. Diagnostic recommendations for upper venous system thrombosis In a child symptomatic for upper venous system VTE, perform an ultrasound of the neck to image the jugular vessels. If positive for SVTE, treat as per therapeutic recommendations (see later). If negative for SVTE and the clinical likelihood is high for SVTE, perform a venogram of the intrathoracic vessels. If positive, treat as per therapeutic recommendations. A compromise approach would be to do an ultrasound alone; if that is positive, one can proceed to therapy. If the ultrasound is negative and the clinical suspicion is high, one should proceed to venogram. Magnetic resonance venography (MRV) has been recently studied but its diagnostic utility currently remains unclear. Lower venous system thrombosis There are no studies in children determining the sensitivity of any radiographic testing for the diagnosis of SVTE in the lower limb. Venography, currently considered the gold standard diagnostic test, is difficult especially in extremely ill children. Diagnostic recommendations for lower venous system thrombosis In a child symptomatic for lower venous system VTE, perform an ultrasound of the deep vasculature of the leg using compression. If negative and the clinical suspicion is high for SVTE, perform venography, if possible. Pulmonary embolism There are no studies in children establishing the sensitivity of radiographic testing for the diagnosis of pulmonary embolism. Pulmonary angiography, currently considered the gold standard diagnostic test has associated risks and cannot usually be performed in extremely ill children. The following radiographic tests have been used in children for the diagnosis of PE: VQ, spiral computed tomography (CT), magnetic resonance imaging (MRI), magnetic resonance angiography (MRA), and magnetic resonance venography (MRV).

Systemic thrombosis in children

301

Diagnostic recommendations for pulmonary embolism In a child symptomatic for PE, perform pulmonary angiography if possible. If unable to perform a pulmonary angiogram, VQ, MRV, MRA, or spiral CT may be performed. D-dimer testing Although in adults the D-dimer has proved to be a useful diagnostic test in certain situations for SVTE/PE, the sensitivity and specificity of D-dimer testing has not been tested in prospective studies in children with symptoms for SVTE/PE. In adults, D-dimer testing has a number of limitations associated with its use. Diagnostic recommendations for systemic venous thrombosis/pulmonary embolism using D-dimer There is no data present in children to provide recommendations re D-dimer diagnosis of SVTE/PE. Outcome Short term The short-term sequelae of SVTE are not trivial and include potential mortality, secondary to PE [41, 62–67] and morbidity, secondary to superior vena cava syndrome [66–68], and chylothorax [66, 67, 69]. Long term The incidence of long-term sequelae of SVTE (PTS, loss of venous access) requires further prospective studies. PTS occurs in children post-SVTE in the upper or lower limb [70]. PTS consists of pain, swelling, limb discoloration, and ulceration resulting from damage to venous valves in deep vessels. There is no properly validated outcome measure for PTS in children. The signs of PTS have been estimated to be present in up to 65% [70] of children post-VTE, but clinically significant PTS occurs in approximately 10–20% of children [40, 71]. Symptomatic relief of PTS in adults has been described with the use of custom compression stockings with pressure of at least 20 mmHg. Children may also benefit from custom compression stockings. Therapeutic recommendations for systemic venous thrombosis/pulmonary embolism First time SVTE/PE should be treated with UFH or low-molecular-weight heparin (LMWH) for a 5–10 day period followed by 3 months of treatment with oral anticoagulants or LMWH (e.g. CVL related) (Grade 1C). If the risk factor is a CVL and

Thrombosis in clinical practice

302

continues to remain in place after 3 months of therapy, consider prophylaxis (Grade 2C). If the SVTE/PE is idiopathic, consider more than 3 months of therapy (Grade 2C). Recurrent SVTE/PE should be treated with UFH or LMWH for a 5–10 day period followed by LMWH or oral anticoagulants for at least 3 months and until the risk factor (e.g. CVL) is gone (Grade 2C). If the recurrent SVTE/PE is idiopathic, treatment should continue indefinitely with either oral anticoagulants or LMWH (Grade 2C). Thromboprophylaxis in children with CVLs without a prior DVT is not currently recommended as there is insufficient safety and efficacy data available [72] (Grade 2C).

Antithrombotic therapy in children Unfractionated heparin The mechanism of action of heparin is to potentiate the inhibitory effects of antithrombin (AT). In normal children, AT does not approach adult levels (1 U/ml) until 6 months of age [73–75]. Levels of AT may be low due to acquired and congenital abnormalities. Children may develop an acquired deficiency post-cardiopulmonary bypass or nephrotic syndrome. Congenital deficiency of AT is rare and usually does not result in thrombotic events until puberty. In children requiring heparin who do not achieve an anticoagulant effect, the level of AT should be measured and if AT concentration is low, replaced using either AT concentrate (suggested dose: 50 U/kg q 24–48 h) or FFP (suggested dose: 2 ml/kg q 8–12 h). Optimal dosing of heparin will likely differ in children compared to adults. In vitro studies have shown that at UFH concentrations in the therapeutic range, the capacity of plasma to generate thrombin is delayed and decreased by 25% in children, compared with adults [33, 76]. Clinical studies confirming the potential decreased UFH requirement in children are urgently required to optimize safety and efficacy. Therapeutic range The recommended UFH therapeutic range for the treatment of SVTE/PE in children is extrapolated from adults. The activated partial thromboplastin time (APTT) in the range reflecting a heparin level by protamine titration of 0.2–0.4 U/ml or an anti-factor Xa level of 0.35–0.7 U/ml [77] is used as a surrogate of the UFH concentration. In children, APTT values correctly predict therapeutic heparin concentrations approximately 70% of the time [78]. Doses The doses of heparin required in pediatric patients to achieve adult therapeutic APTT values have been assessed using a weight-based nomogram (one prospective cohort study) [78]. In 90% of children, a bolus dose of 75–100 U/kg results in therapeutic APTT values (unpublished data). Age-dependent dosing is required for maintenance heparin with infants having the highest requirements (28 U/kg/h) and children over 1 year of age

Systemic thrombosis in children

303

having lower requirements (20 U/kg/h). The dosing nomogram for heparin therapy is shown in Table 14.1. Treatment of heparin-induced bleeding The effect of heparin can be 100% reversed by protamine sulphate (based on its positive charge) with the amount required based on the amount of heparin received in the previous 2 hours [1]. In most cases of bleeding, discontinuation of the heparin infusion is sufficient due to the short half-life of heparin. The dosing of protamine sulphate for heparin reversal is shown in Table 14.2. Efficacy Estimates of efficacy with UFH therapy in children are 7% extension and recurrence from one cohort study [78] and 12% from a randomized clinical trial (RCT) with limitations [50]. Further well-designed clinical studies are required to determine efficacy. Adverse effects There are at least three clinically important adverse effects of heparin: bleeding, osteoporosis, and heparin-induced thrombocytopenia (HIT). There have been no large well-designed

Table 14.2 Protamine sulphate dosing for heparin reversal. Time since last heparin dose or end of infusion (min)

Protamine per 100 units UFH received (mg, maximum 50 mg/dose)

<30

1

0.5–0.75

0.375–0.5

0.25–0.375

30–60

61–120

>120

Adapted from [1].

prospective studies carried out to determine the true incidence of adverse effects with the use of UFH therapy in children. Bleeding occurred in 1.9% (95% CI 0.1–10.2%) of children being treated for SVTE in one cohort study [78]. However, in this study, many children were treated with suboptimal amounts of heparin (compared with target APTT). There are only three case reports of pediatric heparin-induced osteoporosis [79–81]. In two of the children, steroids were used concomitantly [79, 81]. The third received highdose intravenous heparin therapy for a prolonged period [80]. The use of long-term UFH

Thrombosis in clinical practice

304

in adults is associated with osteoporosis. By extrapolation, long-term use of heparin in children should be avoided when other alternative anticoagulants are available. HIT has been reported in a number of case reports in children aged from 3 months to 15 years [82–87]. HIT occurs as a result of the production of an antibody to the combination of platelet factor 4 (PF4) and the glycosaminoglycans (GAGs) of the heparin molecule. When this antibody combines with the antigen (PF4 and GAG), extreme platelet activation occurs resulting in platelet microparticle production and thrombocytopenia. Adults with HIT are at high risk of developing both venous and/or arterial thrombosis [88]. Recent studies suggest the frequency of HIT may be increased in children in an intensive care (2.3%) compared with children in a nonintensive care setting [89, 90]. A high index of suspicion is required to diagnose HIT in children, as many patients in neonatal or pediatric intensive care units who are exposed to heparin have multiple reasons for thrombocytopenia and/or thrombosis. Danaparoid (not available in North America), hirudin, and argatroban are alternatives to heparin in children with HIT [48, 82, 84, 87, 91–93]. Practical tips for heparin therapy Heparin is contraindicated in children with active ongoing major bleeding. To minimize bleeding, the following coagulation parameters should be present: Platelet count greater than 50×109/L, Fibrinogen greater than 1 g/L, INR/PTT normal for age. If platelets are required, administer 1 unit of platelets per 5 kg body weight (maximum 5 units). If fibrinogen is less than 1 g/L, as a practical guide, administer 1 unit of cryoprecipitate per 5 kg body weight (suggested maximum dose: 5 units). Heparin should be administered through a designated intravenous, which should run continuously. Blood samples to monitor heparin therapy should be obtained from a fresh venipuncture, not a central venous or arterial line, which usually have heparin in their infusate. Low-molecular-weight heparin There are many potential advantages of LMWH therapy in children including the need for minimal monitoring (important in pediatric patients with poor or nonexistent venous access); the lack of interference by other drugs or diet such as exists for warfarin; a reduced risk of HIT; and a probable reduced risk of osteoporosis with long-term use compared with heparin. These advantages have prompted international large scale offlabel usage of LMWH exposing children to unknown safety and efficacy, and potentially rare adverse events that may be life-threatening and are currently unknown. Properly designed clinical studies are urgently required in children using LMWH. Therapeutic range Therapeutic dosing of LMWH is extrapolated from adult studies and is based on antifactor Xa levels. The recommended anti-factor Xa range to achieve therapeutic LMWH levels is 0.50–1.0 U/ml measured in a plasma sample taken 4–6 h following a

Systemic thrombosis in children

305

subcutaneous injection. The pharmacodynamic effect of LMWH is measured with an anti-factor Xa level, however, the antithrombotic activity is not accurately reflected [94]. Dosing The LMWHs enoxaparin, reviparin, dalteparin, and tinzaparin [95–98] have been studied in children to determine the doses required to achieve adult therapeutic antifactor Xa levels. In general, peak anti-factor Xa levels occur 2–6 h following a subcutaneous LMWH injection. Children greater than approximately 2–3 months of age or greater than 5 kg body weight have decreased requirements per kg, probably due to a smaller volume of distribution than infants. In young children, other explanations for the increased requirement of LMWH per body weight include altered heparin pharmacokinetics [95, 99] and/or a decreased expression of anticoagulant activity of heparin in children due to decreased plasma concentrations of antithrombin [73]. The dosing nomograms for the LMWH, enoxaparin are shown in Tables 14.3 and 14.4. Treatment of LMWH-induced bleeding One of the potential disadvantages of LMWH use is the lack of complete reversal. Reversal is carried out using equimolar concentrations of protamine sulphate, which neutralizes the antifactor IIa activity but results in only partial neutralization of the antifactor Xa activity [100]. However, in animal models, microvascular bleeding is completely reversed by protamine sulphate suggesting that partial reversal of

Table 14.3 Dosing nomogram for initiation of enoxaparin therapy. Age≤2 months

Age≥2 months to 18 years

Initial treatment dose

1.5 mg/kg/dose s.c. q 12 h

1 mg/kg/dose s.c. q 12 h

Initial prophylactic dose

0.75 mg/kg/dose s.c. q 12 h or 1.5 mg/kg/dose s.c. o.d.

0.5 mg/kg/dose s.c. q 12 h or 1 mg/kg/dose s.c. o.d.

Maximum dose

3 mg/kg/dose s.c. q 12 h

2 mg/kg/dose s.c. q12 h

Adapted from [1].

Table 14.4 Dosing nomogram for maintenance dosing of enoxaparin. Anti-factor Xa level (U/ml)

Hold next dose?

Dose change? (%)

Repeat anti-Xa level?

<0.35

No

Increase by 25

4 h post-next-morning dose

0.35–0.49

No

Increase by

4 h post-next-morning dose

Thrombosis in clinical practice

306

10 0.5–1

No

0

1×per week at 4 h post-morning dose

<1.20

No

Decrease by 20

4 h post-next-morning dose. Hold dose. Do a trough level. If trough <0.5 at 10 h post-dose, administer scheduled dose at 20% of previous dose.

Note: [The above nomogram assumes that there is no bleeding or renal compromise.] Dose adjustment may be limited by commercially available concentration of enoxaparin. Dose adjustments must be ordered in 1.0 mg increments for doses. Adapted from [1].

LMWH may be sufficient to arrest clinical bleeding [101–104]. Protocols using protamine sulphate for LMWH reversal have been published [1]. Adverse events The safety of the LMWH enoxaparin has been estimated in a single institution cohort study of 146 courses of therapeutic enoxaparin. Major bleeding occurred in 4.8% (95% CI 2–9.6%) of patients [105]. In a randomized trial with limitations (sample size not achieved), reviparin therapy in children with SVTE/PE resulted in major bleeding in 8.1% (95% CI 1.7–21.9%) [50]. There are no data on the frequency of HIT or osteoporosis secondary to LMWH use in children. Efficacy The studies estimating recurrence and/or clot extension demonstrated 1% and 5.6%, in a large cohort study [105] and a randomized trial with limitations, respectively [50]. Practical tips for LMWH therapy The use of a subcutaneous catheter (Insuflon, Unomedical, Denmark) in children to administer LMWH has been successfully described in a large cohort study [105]. The catheter may remain in situ for up to 7 days but the site should be closely observed for cellulitis or abscess formation, especially in immunocompromised children. Minimizing the discomfort of the catheter placement can be accomplished by application of a topical anesthetic cream to the skin about 45 minutes before catheter placement. Ensure that renal function is adequate (Cr level) as LMWH is cleared through the kidney and will accumulate in renal compromise/failure potentially resulting in an increased risk of bleeding. The drug depot site (with subcutaneous catheter in situ) should have firm pressure applied for approximately 5 minutes to minimize bleeding, bruising, and subcutaneous hematoma formation.

Systemic thrombosis in children

307

Oral anticoagulant therapy Age-dependent features Similar to children receiving UFH, the plasma of children receiving oral anticoagulants generates 25% less thrombin compared with plasmas from adults with similar International Normalized Ratios (INRs) [106]. This finding supports the hypothesis that children may require lower intensity anticoagulation (decreased therapeutic INR range) than adults to achieve the same effect. This hypothesis is further supported by the observation that plasma concentrations of a marker of endogenous thrombin generation, prothrombin fragment 1.2, is significantly lower in children compared with adults at similar INR values [106]. Therapeutic range There are no clinical trials assessing the optimal INR range using clinical outcomes. Therapeutic INR ranges are subsequently extrapolated from adult therapeutic ranges where large clinical outcome studies have been completed. Dose-response An initial dose of 0.2 mg/kg, with subsequent dose adjustments made according to a nomogram using INR values, was evaluated in a prospective cohort study [107]. The published age-specific, weight-adjusted doses for children vary due to the different study designs, patient populations, and possibly the small number of children studied. The largest cohort study treating children requiring oral anticoagulation (n= 263) determined that young children required an average of 0.2 mg/kg (0.33 mg/kg for <1 year, 0.15 mg/kg for >1 to <6 years) and teenagers 0.09 mg/kg of warfarin to maintain a target INR of 2–3 [108]. In adults, weight-adjusted doses for oral anticoagulants are not precisely known but are in the range of 0.04–0.08 mg/kg to maintain an INR of 2–3 [109]. The mechanisms responsible for the age dependency of oral anticoagulant doses are not completely clear. The nomogram for maintenance dosing of warfarin is shown in Table 14.5. Monitoring The monitoring of oral anticoagulant therapy in children is difficult and requires close supervision with frequent dose adjustments

Table 14.5 Long-term warfarin maintenance dose guidelines. INR

1.1–1.4

Check for compliance, if compliant increase by 20% of dose

Thrombosis in clinical practice

308

INR

1.5–1.9

Increase by 10% of dose

INR

2.0–3.0

No change

INR

3.1–3.5

Decrease dose by 10%

INR

>3.5–4.0

Administer one dose at 50% less than maintenance dose. Then restart at 20% less than the maintenance dose

INR

4.1–5.0

Hold x 1 dose then restart at 20% less than maintenance dose

INR

>5.0

If no bleeding, oral vitamin K 0.5–1.0 mg p.o. and repeat INR in 24 h

Note: These guidelines apply primarily to medically stable patients already established on longterm maintenance therapy. Medically unstable patients or those completing the loading protocol may respond differently. Close daily monitoring with individualized dose adjustment of such patients is essential until they are clearly established on maintenance therapy. Adapted from [1].

[107, 110]. Children require at least three INRs monthly for monitoring [110], in contrast to adults where, in most cases, monthly monitoring is recommended. Despite the extensive monitoring, 60% of children requiring warfarin are outside of the therapeutic range at any time [110]. Reasons contributing to the need for frequent monitoring include the effect of diet, medications, and primary medical illnesses. Breast-fed infants are very sensitive to oral anticoagulants due to the low concentrations of vitamin K in breast milk [111–114]. In contrast, some children are resistant to oral anticoagulants due to impaired absorption; requirements for TPN, which is routinely supplemented with vitamin K, and nutrient formulae, which are all supplemented with vitamin K (55–110 mg/L) to protect against hemorrhagic disease of the newborn [114–116]. Whole-blood monitors in children Although no formal quality of life studies have been carried out in children using a whole-blood monitor, families prefer INR self-testing with these devices for a number of reasons. The blood sample required is capillary, the device is easy to use following teaching, children may use the device in their home and then call the INR result to the Anticoagulation Clinic for warfarin dosing recommendations. Whole-blood monitors use various techniques to measure the time from application of fresh samples of capillary whole blood to coagulation of the sample, and report an INR value. Point-of-care monitors evaluated in children were shown to be acceptable and reliable for use in the outpatient laboratory and at home settings [117, 118]. Adverse effects of oral anticoagulants Bleeding is the main complication of oral anticoagulants. The risk of serious bleeding in children receiving oral anticoagulants for mechanical prosthetic valves is less than 3.2% per patient-year (13 case series) [1]. In one large cohort (391 warfarin years, variable target range) bleeding rate was 0.5% per patient-year [110]. In a randomized trial (n=41) target range 2–3 for 3 months, bleeding occurred in 12.2% (95% CI 4.1–26.2) [50].

Systemic thrombosis in children

309

Nonhemorrhagic complications of oral anticoagulants, such as tracheal calcification or hair loss have been described on rare occasions in young children [119]. A cohort of children with different primary illnesses (congenital heart disease, systemic lupus erythematosus, antiphospholipid antibody syndrome, nephrotic syndrome) who received therapeutic warfarin were described to have significantly reduced bone mineral density. However, this was an uncontrolled study, and the role of the underlying disorders in reducing bone density remains unclear [120]. Efficacy The incidence of clot recurrence and/or extension was documented to be 1.3% per patient-year in a large cohort study [110]. Treatment of oral anticoagulant induced bleeding If minor bleeding occurs, vitamin K (0.5–2.0 mg) may be administered to decrease/ reverse the INR. Adult studies have shown that p.o. vitamin K results in fast reversal (hours) of INR [121]. If major bleeding occurs, consider FFP (suggested dose: 20 ml/kg), or FVIIa. Practical tips for oral anticoagulant therapy Ensure a child is taking full feeds before starting warfarin or the INR may overshoot the target range. In outpatients, send a vial of intravenous vitamin K preparation home to store in the refrigerator. In the event of an INR greater than 6 and with no bleeding, after discussion with the family, the patient should take vitamin K p.o. (in children, a dose of 0.5–1 mg depending on the desired target INR and the measured INR). Thrombolytic therapy Thrombolytic agents are used commonly for arterial thrombosis in children, and to unblock CVLs. The role of thrombolysis in the treatment of VTE is controversial. Potential indications are obstructive intracardiac thrombosis, massive PE, bilateral renal vein thrombosis, acute organ dysfunction due to massive thrombosis [122]. There are three thrombolytic agents available: SK, UK, and tissue plasminogen activator (tPA). Children with acquired plasminogen deficiency have a reduced thrombolytic effect with the use of SK, UK, and tPA [123]. In in vitro studies using a fibrin clot model, supplementation of plasma with plasminogen increases the thrombolytic effect of all three agents [123]. The mechanism of action of tPA is to enzymatically alter plasminogen into the active enzyme, plasmin, which degrades cross-linked fibrin. In children, the level of plasminogen does not achieve adult levels until puberty [73–75]. In children who require thrombolytic therapy, consideration should be given to replacing plasminogen using FFP with a suggested dose of 10–20 ml/kg every 8–12 h.

Thrombosis in clinical practice

310

In children, tPA has become the thrombolytic agent of choice. The use of SK is associated with anaphylaxis with repeated use and UK was removed from the American market due to suspect manufacturing processes as per the Federal Drug Administration (FDA) [124]. Recombinant UK is now available and under study. In vitro studies have also shown tPA to have improved clot lysis versus UK and SK [123]. Dosing There have been multiple regimens of thrombolytic therapy reported in the literature with doses ranging from 0.01 mg/kg/h to 0.5 mg/kg/h administered for anywhere from a few hours to over 24 hours [125]. Although the use of local thrombolytic therapy may require decreased dosing to achieve clot lysis, there have been no studies confirming this hypothesis. Adverse effects of thrombolytic therapy The main adverse event occurring due to thrombolytic therapy is bleeding occurring in 68% of children with transfusion required in 39% [30]. Prolonged duration of thrombolytic therapy was associated with increased bleeding. A recent prospective study with a defined protocol (concomitant heparin at 10 U/kg/h, fixed tPA infusion of 0.5 mg/kg/h for a maximum of 6 hours, and 10 ml/kg FFP 30 minutes pre-tPA as a fibrinogen and plasminogen source) reported major bleeding (requiring transfusion) in 2 out of 20 patients (10%) and minor bleeding in 6 patients (30%) [126]. Prospective studies using defined protocols in larger numbers of patients are urgently required. Efficacy Successful clot lysis in children by thrombolytic agents as reported in the literature varies. The thrombolytic regimen used, age of clot, and age of patient will influence the amount of clot lysis. In a prospective study in 20 children who had received 10 ml/kg of FFP followed by 0.5 mg/kg/h of tPA for a 6 hours period, complete clot lysis occurred in 13 out of 20 children (65%, 12 arterial thrombi and 1 venous thrombus). Partial lysis occurred in 4 patients (20%, 1 arterial thrombus and 3 venous thrombi) [126]. Treatment of bleeding If minor bleeding occurs, usually from a puncture site, pressure can be applied to minimize/stop bleeding. If major bleeding occurs, discontinue thrombolytic therapy and heparin infusion, and administer cryoprecipitate 1 unit per 5 kg (suggested maximum: 5 units). Practical tips for thrombolytic therapy Thrombolytic therapy should be administered in a setting where close patient monitoring can occur (e.g. intensive care unit). In a young child who is extremely active, consider

Systemic thrombosis in children

311

sedation during therapy. Sites of line insertion (venous or arterial) should be regularly checked for signs of bleeding. Ensure that the INR/PTT are within normal for age at the initiation of therapy, the platelet count should be ideally greater than 100×109/L but at least greater than 50×109/L, and the fibrinogen should be greater than 1 g/L. Check blood work at the beginning of therapy and then every 4–6 hours to minimize bleeding. Alternative therapies for systemic venous thrombosis In addition to pharmacologic therapy, venous interruption devices (inferior vena cava filters) are used for specific clinical indications in adults. The most common indication for the use of inferior vena cava filters is to prevent PE in the presence of a contraindication to anticoagulant therapy, most often bleeding [127–129]. There is limited experience in children, however, temporary filters are more often used and removed when the source of PE is no longer present [130, 131]. The risk-benefit ratio needs to be considered individually.

Thrombosis in children: the future The diagnosis and treatment of thrombosis in children is based on adult recommendations. This approach is not ideal due to the many differences in hemostasis in children compared to adults. Diagnostic procedures with unproven sensitivity and specificity are currently used in children with symptomatic thrombosis. A large number of anticoagulants are currently used off-label exposing children to agents which has an undetermined risk-benefit ratio in children. Designed prospective clinical studies determining the best diagnostic and therapeutic regimens are urgently required to provide the best evidence-based care to children at risk for or with arterial and venous thrombosis.

References [1] Monagle P, Michelson AD, Bovill E, Andrew M. Antithrombotic therapy in children. Chest 2001; 119:344S-370S. [2] Price V, Massicotte P. Arterial thromboembolism in the pediatric population [Abstract]. Semin Thromb Hemost 2003; 29:557–65. [3] Andrew ME, Monagle P, deVeber G, Chan AK. Thromboembolic disease and antithrombotic therapy in newborns. Hematology (Am Soc Hematol Educ Program) 2001; 358–74. [4] Stanger P, Heymann MA, Tarnoff H, Hoffman JI, Rudolph AM. Complications of cardiac catheterization of neonates, infants, and children. A three-year study. Circulation 1974; 50:595– 608. [5] Freed MD, Keane JF, Rosenthal A. The use of heparinization to prevent arterial thrombosis after percutaneous cardiac catheterization in children. Circulation 1974; 50:565–9. [6] Vitiello R, McCrindle BW, Nykanen D, Freedom RM, Benson LN. Complications associated with pediatric cardiac catheterization. J Am Coll Cardiol 1998; 32:1433–40.

Thrombosis in clinical practice

312

[7] Saxena A, Gupta R, Kumar RK, Kothari SS, Wasir HS. Predictors of arterial thrombosis after diagnostic cardiac catheterization in infants and children randomized to two heparin dosages. Cathet Cardiovasc Diagn 1997; 41:400–3. [8] Bulbul ZR, Galal MO, Mahmoud E, et al. Arterial complications following cardiac catheterization in children less than 10 kg. Asian Cardiovasc Thorac Ann 2002; 129–32. [9] Rao PS, Thapar MK, Rogers JH Jr, Strong WB, Lutcher CL, Nesbit RR Jr, et al. Effect of intraarterial injection of heparin on the complications of percutaneous arterial catheterization in infants and children. Cathet Cardiovasc Diagn 1981; 7:235–46. [10] Butt W, Shann F, McDonnell G, Hudson I. Effect of heparin concentration and infusion rate on the patency of arterial catheters. Crit Care Med 1987; 15:230–2. [11] Freed MD, Rosenthal A, Fyler D. Attempts to reduce arterial thrombosis after cardiac catheterization in children: use of percutaneous technique and aspirin. Am Heart J 1974; 87:283–6. [12] Randolph AG, Cook DJ, Gonzales CA, Andrew M. Benefit of heparin in peripheral venous and arterial catheters: systemic review and meta-analysis of randomized controlled trials . BMJ 1998; 316:969–75. [13] Scheer B, Perel A, Pfeiffer UJ. Clinical review: complications and risk factors of peripheral arterial catheters used for haemodynamic monitoring in anaesthesia and intensive care medicine. Crit Care 2002; 199–204. [14] Sellden H, Nilsson K, Larsson LE, Ekstrom-Jodal B. Radial arterial catheters in children and neonates: a prospective study. Crit Care Med 1987; 15:1106–9. [15] Rais-Bahrami K, Karna P, Dolanski EA. Effect of fluids on life span of peripheral arterial lines. Am J Perinatol 1990; 7:122–4. [16] Heulitt MJ, Farrington EA, O’Shea TM, Stoltzman SM, Srubar NB, Levin DL. Double-blind, randomized, controlled trial of papaverinecontaining infusions to prevent failure of arterial catheters in pediatric patients [see comments]. Crit Care Med 1993; 21:825–9. [17] Tarry WC, Moser AJ, Makhoul RG. Peripheral arterial thrombosis in the nephrotic syndrome. Surgery 1993; 114:618–23. [18] Andrew M, David M, deVeber G, Brooker LA. Arterial thromboembolic complications in paediatric patients. Thromb Haemost 1997; 78:715–25. [19] Schmidt B, Andrew M. Neonatal thrombotic disease: prevention, diagnosis, and treatment. J Pediatr 1988; 113:407–10. [20] Wigger HJ, Bransilver BR, Blanc WA. Thromboses due to catheterization in infants and children. J Pediatr 1970–976:1–11. [21] Kern IB. Management of children with chronic femoral artery obstruction. J Pediatr Surg 1977; 12:83–90. [22] Ino T, Benson LN, Freedom RM, Barker GA, Aipursky A, Rowe RD. Thrombolytic therapy for femoral artery thrombosis after pediatric cardiac catheterization. Am Heart J 1988; 115:633– 9. [23] Wessel DL, Keane JF, Fellows KE, Robichaud H, Lock JE. Fibrinolytic therapy for femoral arterial thrombosis after cardiac catheterization in infants and children. Am J Cardiol 1986; 58:347–51. [24] Kirk CR, Qureshi SA. Streptokinase in the management of arterial thrombosis in infancy. Int J Cardiol 1989; 25:15–20. [25] Brus F, Witsenburg M, Hofhuis WJ, Hazelzet JA, Hess J. Streptokinase treatment for femoral artery thrombosis after arterial cardiac catheterisation in infants and children. Br Heart J 1990; 63:291–4. [26] Kothari SS, Kumar RK, Varma S, Saxena A. Thrombolytic therapy in infants for femoral artery thrombosis following cardiac catheterization. Indian Heart J 1996;48:246–8.

Systemic thrombosis in children

313

[27] Levy M, Benson LN, Burrows PE, Bentur Y, Strong DK, Smith J, et al. Tissue plasminogen activator for the treatment of thromboembolism in infants and children. J Pediatr 1991; 118:467–72. [28] Zenz W, Muntean W, Beitzke A, Zobel G, Riccabona M, Gamillscheg A. Tissue plasminogen activator (alteplase) treatment for femoral artery thrombosis after cardiac catheterization in infants and children. Br Heart J 1993; 70:382–5. [29] Ries M, Singer H, Hofbeck M. Thrombolysis of a modified Blalock-Taussig shunt with recombinant tissue plasminogen activator in a newborn infant with pulmonary atresia and ventricular septal defect. Br Heart J 1994; 72:201–2. [30] Gupta AA, Leaker M, Andrew M, Massicotte P, Liu L, Benson LN, et al. Safety and outcomes of thrombolysis with tissue plasminogen activator for treatment of intravascular thrombosis in children. J Pediatr 2001; 139:682–8. [31] Chaikof EL, Dodson TF, Salam AA, Lumsden AB, Smith RBI. Acute arterial thrombosis in the very young. J Vasc Surg 1992; 16:428–35. [32] Andrew M, Schmidt B, Mitchell L, Paes B, Ofosu F. Thrombin generation in newborn plasma is critically dependent on the concentration of prothrombin. Thromb Haemost 1990; 63:27–30. [33] Andrew M, Mitchell L, Vegh P, Ofosu F. Thrombin regulation in children differs from adults in the absence and presence of heparin. Thromb Haemost 1994; 72:836–42. [34] Ling X, Delorme M, Berry L, Ofosu F, Mitchell L, Paes B, et al. Alpha 2-Macroglobulin remains as important as antithrombin III for thrombin regulation in cord plasma in the presence of endothelial cell surfaces. Pediatr Res 1995; 37:373–8. [35] Xu L, Delorme M, Berry L, Brooker L, Mitchell L, Andrew M. Thrombin generation in newborn and adult plasma in the presence of an endothelial surface [Abstract]. Thromb Haemost 1991; 65:1230. [36] Nitschmann E, Berry L, Bridge S, Hatton MW, Richardson M, Monagle P, et al. Morphological and biochemical features affecting the antithrombotic properties of the aorta in adult rabbits and rabbit pups. Thromb Haemost 1998; 79: 1034–40. [37] Castaman G, Rodeghiero F, Dini E. Thrombotic complications during L-asparaginase treatment for acute lymphocytic leukemia. Haematologica 1990; 75:567–9. [38] Wise RC, Todd JK. Spontaneous, lower-extremity venous thrombosis in children. Am J Dis Child 1973; 126:766–9. [39] Bernstein D, Coupey S, Schonberg SK. Pulmonary embolism in adolescents. Am J Dis Child 1986; 140:667–71. [40] Monagle P, Adams M, Mahoney M, Ali K, Barnard D, Bernstein M, et al. Outcome of pediatric thromboembolic disease: a report from the Canadian Childhood Thrombophilia Registry. Pediatr Res 2000; 47:763–6. [41] Massicotte MP, Dix D, Monagle P, Adams M, Andrew M. Central venous catheter related thrombosis in children: analysis of the Canadian Registry of Venous Thromboembolic Complications. J Pediatr 1998; 133:770–6. [42] Andrew M, Massicotte MP, deVeber G, Leaker M, David M, Brown A, et al. 1–800-NOCLOTS: A quaternary care solution to a new tertiary care disease: Childhood thrombophilia [Abstract]. Thromb Haemost 1997; 77:727. [43] Coon WW, Willis PWI, Keller JB. Venous thromboembolism and other venous disease in the Tecumseh community health study. Circulation 1973; 48:839–46. [44] Gjores J. The incidence of venous thrombosis and its sequelae in certain districts in Sweden. Acta Chir Scand 1956; 206:1–10. [45] Carter C, Gent M. The epidemiology of venous thrombosis. In: Colman R, Hirsh J, Marder V, Salzman E (eds), Hemostasis and Thrombosis. Basic Principles and Clinical Practice. Philadelphia, PA: J.B.Lippincott Company, 1982; 805–19.

Thrombosis in clinical practice

314

[46] Andrew M, David M, Adams M, Ali K, Anderson R, Barnard D, et al. Venous thromboembolic complications (VTE) in children: first analyses of the Canadian Registry of VTE. Blood 1994; 83: 1251–7. [47] van Ommen CH, Heijboer H, Buller HR, Hirasing RA, Heijmans HS, Peters M. Venous thromboembolism in childhood: a prospective two-year registry in The Netherlands. J Pediatr 2001; 139:676–81. [48] Gibson BES, Chalmers EA, Bolton-Maggs P, Henderson DJ, Boshkov RL. Thromboembolism in childhood: a prospective 2 years study in the United Kingdom (February 2001-February 2003) [Abstract]. J Thromb Haemost 2003; 1(Suppl.). [49] Schmidt B, Andrew M. Neonatal thrombosis: report of a prospective Canadian and international registry. Pediatrics 1995; 96:939–43. [50] Massicotte MP, Julian JA, Gent M, Shields K, Marzinotto V, Szechtman B, et al. An openlabel randomized controlled trial of low molecular weight heparin compared to heparin and coumadin for the treatment of venous thromboembolic events in children: the REVIVE trial. Thromb Res 2003; 109:85–92. [51] Marsh D, Wilkerson SA, Cook LN, Pietsch JB. Right atrial thrombus formation screening using two-dimensional echocardiograms in neonates with central venous catheters. Pediatrics 1988;81:284–6. [52] Berman W Jr, Fripp RR, Yabek SM, Wernly J, Corlew S. Great vein and right atrial thrombosis in critically ill infants and children with central venous lines. Chest 1991; 99:963–7. [53] Williams EC. Catheter-related thrombosis. Clin Cardiol 1990; 13:VI34–VI36. [54] Chidi CC, King DR, Bales ET Jr. An ultrastructural study of the intimal injury by an indwelling umbilical catheter. J Pediatr Surg 1983; 18:109. [55] Wakefield A, Cohen Z, Rosenthal A, Craig M, Jeejeebhoy KN, Gotlieb A, et al. Thrombogenicity of total parenteral nutrition solutions: II. Effect on induction of endothelial cell procoagulant activity. Gastroenterology 1989; 97:1220–8. [56] Pottecher T, Forrler M, Picardat P, Krause D, Bellocq JP, Otteni JC. Thrombogenicity of central venous catheters: prospective study of polyethylene, silicone and polyurethane catheters with phlebography or post-mortem examination. Eur J Anaesthesiol 1984; 1:361–5. [57] Andrew M, Marzinotto V, Pencharz P, Zlotkin S, Burrows P, Ingram J, et al. A cross-sectional study of catheter-related thrombosis in children receiving total parenteral nutrition at home. J Pediatr 1995; 126:358–63. [58] Derish MT, Smith DW, Frankel LR. Venous catheter thrombus formation and pulmonary embolism in children. Pediatr Pulmonol 1995; 20: 349–54. [59] Nowak-Gottl U, Von Kries R, Gobel U. Neonatal symptomatic thromboembolism in Germany: two year survey. Arch Dis Child Fetal Neonatal Ed 1997; 76:F163–F167. [60] Nowak-Gottl U, Dubbers A, Kececioglu D, Koch HG, Kotthoff S, Runde J, et al. Factor V Leiden, protein C, and lipoprotein (a) in catheter-related thrombosis in childhood: a prospective study. J Pediatr 1997; 131:608–12. [61] Male C, Chait P, Ginsberg J. Comparison of venography and ultrasound for the diagnosis of deep vein thrombosis in the upper body in children: a substudy of the PARKAA trial. Thromb Haemost 2002; 87:593–8. [62] Ament J, Newth CJ. Deep venous lines and thromboembolism. Pediatr Pulmonol 1995; 20:347–8. [63] Hoyer PF, Gonda S, Barthels M, Krohn HP, Brodehl J. Thromboembolic complications in children with nephrotic syndrome. Risk and incidence. Acta Paediatr Scand 1986; 75:804–10. [64] Dollery CM, Sullivan ID, Bauraind O, Bull C, Milla PJ. Thrombosis and embolism in longterm central venous access for parenteral nutrition. Lancet 1994; 344:1043–5. [65] Uderzo C, Faccini P, Rovelli A, Arosio M, Marchi PF, Riva A, et al. Pulmonary thromboembolism in childhood leukemia: 8-years’ experience in a pediatric hematology center. J Clin Oncol 1995; 13: 2805–12.

Systemic thrombosis in children

315

[66] Le Coultre C, Oberhansli I, Mossaz A, Bugmann P, Faidutti B, Belli DC. Postoperative chylothorax in children: differences between vascular and traumatic origin. J Pediatr Surg 1991; 26:519–23. [67] Mollitt DL, Golladay ES. Complications of TPN catheter-induced vena caval thrombosis in children less than one year of age. J Pediatr Surg 1983; 18:462–7. [68] Graham L Jr, Gumbiner CH. Right atrial thrombus and superior vena cava syndrome in a child. Pediatrics 1984; 73:225–9. [69] Dhande V, Kattwinkel J, Alford B. Recurrent bilateral pleural effusions secondary to superior vena cava obstruction as a complication of central venous catheterization. Pediatrics 1983; 72:109–13. [70] Marzinotto V, Choi M, Massicotte MP, Chan AKC. Post-thrombotic syndrome in children with previous deep vein thrombosis [Abstract]. Thromb Haemost 2001; (Suppl.):OC962. [71] Kuhle S, Koloshuk B, Marzinotto V, Bauman M, Massicotte MP, Andrew MA, et al. A crosssectional study evaluating post thrombotic syndrome in children. Throm Res 2003; 11:227– 33. [72] Massicotte MP, Julian JA, Gent M, Shields K, Marzinotto V, Szechtman B, et al. An openlabel randomized controlled trial of low molecular weight heparin for the prevention of central venous line related thrombotic complications in children: The PROTEKT trial. Thromb Res 2003; 109:101–8. [73] Andrew M, Paes B, Johnston M. Development of the hemostatic system in the neonate and young infant. Am J Pediatr Hematol Oncol 1990;12: 95–104. [74] Andrew M, Paes B, Milner R, Johnston M, Mitchell L, Tollefsen DM, et al. Development of the human coagulation system in the healthy premature infant. Blood 1988; 72:1651–7. [75] Andrew M, Paes B, Milner R, Johnston M, Mitchell L, Tollefsen DM, et al. Development of the human coagulation system in the full-term infant. Blood 1987; 70:165–72. [76] Schmidt B, Ofosu FA, Mitchell L, Brooker LA, Andrew M. Anticoagulant effects of heparin in neonatal plasma. Pediatr Res 1989; 25:405–8. [77] Hirsh J. Heparin. N Engl J Med 1991; 324: 1565–74. [78] Andrew M, Marzinotto V, Massicotte P, Blanchette V, Ginsberg J, Brill-Edwards P, et al. Heparin therapy in pediatric patients: a prospective cohort study. Pediatr Res 1994; 35:78–83. [79] Avioli LV. Heparin-induced osteopenia: an appraisal. Adv Exp Med Biol 1975; 52:375–87. [80] Sackler JP, Liu L. Heparin-induced osteoporosis. Br J Radiol 1973; 46:548–50. [81] Murphy MS, John PR, Mayer AD, Buckels JA, Kelly DA. Heparin therapy and bone fractures. Lancet 1992; 340:1098. [82] Potter C, Gill JC, Scott JP, McFarland JG. Heparin-induced thrombocytopenia in a child. J Pediatr 1992; 121:135–8. [83] Klement D, Rammos S, Kries R, Kirschke W, Kniemeyer HW, Greinacher A. Heparin as a cause of thrombus progression. Heparin-associated thrombocytopenia is an important differential diagnosis in paediatric patients even with normal platelet counts. Eur J Pediatr 1996; 155:11–14. [84] Murdoch IA, Beattie RM, Silver DM. Heparininduced thrombocytopenia in children. Acta Paediatr 1993; 82:495–7. [85] Magnani HN. Heparin-induced thrombocytopenia (HIT): an overview of 230 patients treated with orgaran (Org 10172). Thromb Haemost 1993; 70:554–61. [86] Andrew M, deVeber G. Pediatric Thromboembolism and Stroke Protocols. Hamilton, BC: Decker Inc., 1999. [87] Severin T, Sutor AH. Heparin-induced thrombocytopenia in pediatrics. Semin Thromb Hemost 2001; 27:293–9. [88] Warkentin T. Heparin-induced thrombocytopenia: pathogenesis and management. Br J Haematol 2003; 535–55. [89] Smugge M. Heparin induced thrombocytopenia in children [Abstract]. Pediatrics 2002; 10:e10.

Thrombosis in clinical practice

316

[90] Newall F, Barnes C, Ignjatovic V, Monagle P. Heparin-induced thrombocytopenia in children. J Ped Child Health 2003; 39:289–92. [91] Deitcher SR, Topoulos AP, Bartholomew JR, Kichuk-Chrisant MR. Lepirudin anticoagulation for heparin-induced thrombocytopenia. J Pediatr 2002;140:264–6. [92] Ranze O, Ranze P, Magnani HN, Greinacher A. Heparin-induced thrombocytopenia in paediatric patients: a review of the literature and a new case treated with danaparoid sodium. Eur J Pediatr 1999; 158:S130–S133. [93] Cetta F, Graham L, Wrons L, Arruda J, Walenga J. Argatroban use during pediatric interventional cardiac catheterization. Catheter Cardiovasc Interven 2004; 147–9. [94] Greaves M. Limitations of the laboratory monitoring of heparin therapy. Scientific and Standardization Committee Communications: on behalf of the Control of Anticoagulation Subcommittee of the Scientific and Standardization Committee of the International Society of Thrombosis and Haemostasis. Thromb Haemost 2002; 87:163–4. [95] Massicotte P, Adams M, Marzinotto V, Brooker LA, Andrew M. Low-molecular-weight heparin in pediatric patients with thrombotic disease: a dose finding study. J Pediatr 1996; 128:313–18. [96] Massicotte MP, Adams M, Leaker M, Andrew M. A nomogram to establish therapeutic levels of the low molecular weight heparin (LMWH), clivarine in children requiring treatment for venous thromboembolism (VTE) [Abstract]. Thromb Haemost 1997; (Suppl.):282. [97] Nohe N, Flemmer A, Rumler R, Praum M, Auberger K. The low molecular weight heparin dalteparin for prophylaxis and therapy of thrombosis in childhood: a report on 4 cases. Eur J Pediatr 1999; 158:S134–S139. [98] Kuhle S, Massicotte MP, Andrew M, Dinyari M, Marzinotto V, Mitchell D, et al. A dosefinding study of Tinzaparin in pediatric patients [Abstract]. Blood 2002; 100:279a. [99] Hirsh J, Levine MN. Low molecular weight heparin. Blood 1992; 79:1–17. [100] Crowther MA, Berry LR, Monagle PT, Chan AK. Mechanisms responsible for the failure of protamine to inactivate low-molecular-weight heparin. Br J Haematol 2002; 116:178–86. [101] Massonet-Castel S, Pelissier E, Bara L, Terrier E, Abry B, Guibourt P, et al. Partial reversal of low molecular weight heparin (PK 10169) anti Xa activity by protamine sulfate: in vitro and in vivo study during cardiac surgery with extracorporeal circulation. Haemostasis 1986; 16:139. [102] Harenberg J, Wurzner B, Zimmermann R, Schettler G. Bioavailability and antagonization of the low molecular weight heparin CY 216 in man. Thromb Res 1986; 44:549–55. [103] Ryn-McKenna J, Cai L, Ofosu FA, Hirsh J, Buchanan MR. Neutralization of enoxaparineinduced bleeding by protamine sulfate. Thromb Haemost 1990; 63:271–4. [104] Tait DP. Does low molecular weight heparin cause bleeding? Thromb Haemost 1997; 78:1422–5. [105] Dix D, Andrew M, Marzinotto V, Charpentier K, Bridge S, Monagle P, et al. The use of low molecular weight heparin in pediatric patients: a prospective cohort study. J Pediatr 2000; 136:439–45. [106] Massicotte P, Leaker M, Marzinotto V, Adams M, Freedom R, Williams W, et al. Enhanced thrombin regulation during warfarin therapy in children compared to adults. Thromb Haemost 1998;80: 570–4. [107] Andrew M, Marzinotto V, Brooker LA, Adams M, Ginsberg J, Freedom R, et al. Oral anticoagulation therapy in pediatric patients: a prospective study. Thromb Haemost 1994; 71:265–9. [108] Streif W, Mitchell LG, Andrew M. Antithrombotic therapy in children. Curr Opin Pediatr 1999;11: 56–64. [109] Hirsh J. Oral anticoagulant drugs. N Engl J Med 1991; 324:1865–75. [110] Streif W, Andrew M, Marzinotto V, Massicotte P, Chan AK, Julian JA, et al. Analysis of warfarin therapy in pediatric patients: a prospective cohort study of 319 patients. Blood 1999; 94:3007–14.

Systemic thrombosis in children

317

[111] Shearer MJ, Rahim S, Barkhan P, Stimmler L. Plasma vitamin K1 in mothers and their newborn babies. Lancet 1982; 2:460–3. [112] Greer FR, Mummah-Schendel LL, Marshall S, Suttie JW. Vitamin K1 (phylloquinone) and vitamin K2 (menaquinone) status in newborns during the first week of life. Pediatrics 1988; 81:137–40. [113] Haroon Y, Shearer MJ, Rahim S, Gunn WG, McEnery G, Barkhan P. The content of phylloquinone (vitamin K1) in human milk, cows’ milk and infant formula foods determined by high-performance liquid chromatography. J Nutr 1982; 112:1105–17. [114] Von Kries R, Shearer M, McCarthy PT, Haug M, Harzer G, Gobel U. Vitamin K1 content of maternal milk: influence of the stage of lactation, lipid composition, and vitamin K1 supplements given to the mother. Pediatr Res 1987; 22:513–17. [115] Stewart S, Cianciotta D, Alexson C, Manning J. The long-term risk of warfarin sodium therapy and the incidence of thromboembolism in children after prosthetic cardiac valve replacement. J Thorac Cardiovasc Surg 1987; 93:551–4. [116] Boshkov LK, Thomas G, Kirby A, Shen I, Swanson V, Burch G, et al. Pharmacokinetics of Argatroban infusion in a 6 month old congenital cardiac patient with previously diagnosed heparin-induced thrombocytopenia (HIT) [Abstract]. Blood 2002; 194. [117] Marzinotto V, Monagle P, Chan A, Adams M, Massicotte P, Leaker M, et al. Capillary whole blood monitoring of oral anticoagulants in children in outpatient clinics and the home setting. Pediatr Cardiol 2000; 21:347–52. [118] Massicotte P, Marzinotto V, Vegh P, Adams M, Andrew M. Home monitoring of warfarin therapy in children with a whole blood prothrombin time monitor. J Pediatr 1995; 127:389–94. [119] Rosen HN, Maitland LA, Suttie JW, Manning WJ, Glynn RJ, Greenspan SL. Vitamin K and maintenance of skeletal integrity in adults. Am J Med 1993; 94:62–8. [120] Massicotte P, Julian J, Webber C, Charpentier K. Osteoporosis: a potential complication of long term warfarin therapy [Abstract]. Thromb Haemost 1999; (Suppl.):1333a. [121] Crowther M, Douketis J, Schnurr T, Steidl L, Mera V, Ultori C, et al. Oral vitamin K lowers the International Normalized Ratio more rapidly than subcutaneous Vitamin K in the treatment of Warfarin-associated coagulopathy: a randomized, controlled trial. Ann Intern Med 2002; 251–4. [122] Andrew M, Monagle P, Brooker L. Thrombolytic therapy. In: Anonymous Thromboembolic Complications During Infancy and Childhood. Hamilton, BC: Decker Inc, 2000; 357–84. [123] Andrew M, Brooker L, Leaker M, Paes B, Weitz J. Fibrin clot lysis by thrombolytic agents is impaired in newborns due to a low plasminogen concentration. Thromb Haemost 1992; 68:325– 30. [124] Food and Drug Administration (FDA). Important drug warning regarding the use of Abbokinase (urokinase). Public Health Services 1999;1401 Rockville Pike, Rockville MD 20852–1448. [125] Manco-Johnson M, Grabowski EF, Hellgreen M, Kemahli AS, Massicotte MP, Muntean W, et al. Recommendations for tPA thrombolysis in children: on behalf of the Scientific Subcommittee on Perinatal and Pediatric Thrombosis of the Scientific and Standardization Committee of the International Society of Thrombosis and Haemostasis. Thromb Haemost 2002;!57–8. [126] Browne M, Newall F, Campbell J, Savola HF, Monagle P. Thrombolytic therapy with tissue plasminogen activator (tPA), analysis of safety and outcome in children [Abstract]. J Thromb Haemost 2003. [127] Timsit JF, Reynaud P, Meyer G, Sors H. Pulmonary embolectomy by catheter device in massive pulmonary embolism. Chest 1991;100: 655–8. [128] Decousus H, Leizorovicz A, Parent F, Page Y, Tardy B, Girard P, et al. A clinical trial of vena caval filters in the prevention of pulmonary embolism in patients with proximal deep-vein thrombosis. Prevention du Risque d’Embolie Pulmonaire par Interruption Cave Study Group. N Engl J Med 1998; 338:409–15.

Thrombosis in clinical practice

318

[129] McBride WJ, Gadowski GR, Keller MS, Vane DW. Pulmonary embolism in pediatric trauma patients. J Trauma 1994; 37:913–15. [130] Khong PL, John PR. Technical aspects of insertion and removal of an inferior vena cava IVC filter for prophylactic treatment of pulmonary embolus. Pediatr Radiol 1997; 27:239–41. [131] Anton N, Chait P, Chan A, Marzinotto V, Massicotte P. Vena caval filters in children: preliminary safety and efficacy data [Abstract]. Thromb Haemost 2001; (Suppl.):P2227.

Systemic thrombosis in children

319

15 New drugs and directions Bernd Jilma Introduction Andrew Blann has reviewed in Chapter 3 the heparins, coumarin derivatives, and the antiplatelet drug aspirin. These drugs have served patients with arterial or venous thrombosis excellently during the last decades. However, they also have major drawbacks, which may at least in part be overcome by novel classes of anticoagulants or antiplatelet drugs. Attributes of ideal anticoagulants are listed in Table 15.1. This chapter shall mainly focus on the more recently developed anticoagulants which have already been introduced into clinical practice, including fondaparinux, hirudins, ximelagatran, and clopidogrel. It will also provide some perspective on anticoagulants that have entered the clinical drug development process but will deliberately exclude those which are currently in an experimental phase of development. An excellent review of these drugs has been published recently [1], Additionally, this chapter will also briefly touch on the use of glycoprotein (Gp) IIb/IIIa inhibitors and the use of novel fibrinolytic agents. After reviewing the tissue factor inhibitors, we will step down the coagulation ladder to Factor Xa inhibitors, thrombin inhibitors, discuss the use of endogenous

Table 15.1 Attributes of an ideal anticoagulant Orally Fixed Rapid Predi No routine Low food/ No Inhibits admin. dose on/offset ctable coagu drug thrombo free and PK lation interaction cytopenia clotb monitoring ound thrombin LMWH

X

UFH

X

X

X

X

X X

Xa

X

X

X

X

X

X

X

X

X

X

X

Ximelagatran X

X

X

X

X

X

X

Fondaparinux Warfarin

a Due to the relatively long half-life, fondaparinux does not have a rapid offset of action.

X

New drugs and directions

321

Figure 15.1 Simplified scheme of coagulation cascade focusing on important coagulation factors and levels of therapeutic intervention. Selected anticoagulants are depicted on the right side. TF pathway inhibitor (TFPI) blocks the quaternary TF-factor VII/FIX/FX complex, and active siteinhibited factor VIIa (FVIIa) (ASIS) and recombinant nematode peptide c2 (rNAPc2) have similar effects. Antithrombin (AT) inhibits mainly FXa and Flla, and its activity is greatly enhanced by indirect FX or Fll inhibitors such as pentasaccharides (fondaparinux), danaparoid, or low molecular weight heparins (LMWH), whereas DX-9065a directly inhibits FX. Similarly, direct thrombin inhibitors, unlike unfractionated heparin (UFH), work independently of AT, and include hirudins (bivalrudin, desirudin, lepirudin) xi/melagatran, argatroban, BIBR1048, the hirudins acting as irreversible inhibitors. Thrombomodulin (TM) allows activation of protein C by thrombin

Thrombosis in clinical practice

322

and protein C together with protein S inhibits FVa and FVIIIa, and thereby the positive feedback loop of thrombin. rhAPC: Recombinant human activated protein C. anticoagulants (see Figure 15.1 for overview), before briefly reviewing the treatment of arterial thrombosis with clopidogrel and finally the novel fibrinolytics. Common features Many of these novel drugs share common properties. As data is generally lacking, these novel anticoagulants should be administered in pregnancy only if specifically indicated, for example, heparin-associated thrombocytopenia. Currently no specific antidotes are available, which may make management of bleeding difficult. Cost-effectiveness should be considered as an important factor when new agents are being developed.

Factor VII and tissue factor inhibitors Tissue factor (TF) is considered the main trigger of coagulation activation in vivo. Tissue factor avidly binds FVII, which, when activated, induces FX activation. In the vascular system, TF is exposed when the endothelium is damaged and subendothelial TF comes into contact with blood. This can occur at sites of ruptured plaques in the arterial system, minimal trauma of veins, or the release of TF during surgery from damaged tissues [2], Alternatively, TF expression by monocytes [3] and potentially endothelial cells can be stimulated by inflammation. As TF is considered the central trigger of coagulation activation, pharmacological inhibition of TF/FVIIa, the most upstream event, is a highly attractive target. Tissue factor pathway inhibitor Tissue factor pathway inhibitor (TFPI) is an endogenous inhibitor of the TF/FVIIa complex. TFPI also inhibits FXa by binding to its active site, which contributes considerably to its strong anticoagulant potency. Its anticoagulant efficacy has been clearly established in animals [4], and recombinant human (rh)TFPI dose-dependently decreased coagulation activation in low-grade human endotoxemia [5]. In contrast, rhTFPI failed to provide benefit in septic patients [6]. Nematode anticoagulant proteins Nematode anticoagulant proteins (NAPs) from the hookworm Ancylostoma canium inhibit coagulation at picomolar concentrations (Ki-value of rNAPc2=8.4 pM). Similar to

New drugs and directions

323

TFPI, some NAPs inhibit FXa through binding to its active site, and the recombinant rNAPc2 additionally binds to a FXa protein exosite, which potently inhibits the TF/VIIa complex [7]. The recombinant rNAPc2 has a half-life of greater than 50 hours and a bioavailability of 90–100% after s.c. injection. Phase II trials have been completed in patients undergoing orthopedic surgery and percutaneous angioplasty, and a phase II trial has been launched in unstable angina (ANTHEM/TIMI32). A dosage of 3.0 µg/kg administered within 1 hour after knee-replacement surgery appeared to provide the best results, with an overall deep vein thrombosis (DVT) rate of 12.2%, a proximal DVT rate of 1.3%, and a major bleeding rate of 2.3% [8]. In patients undergoing percutaneous coronary intervention (PCI), rNAPc2 decreased thrombin generation when administered on top of heparin [9]. Active site inhibited factor VIIa Active site inhibited factor VIIa (ASIS, FFRrFVIIa) is another highly potent inhibitor of TF-induced coagulation. It is synthesized from rFVIIa which is used as a pro-hemostatic agent [10]: FFR-rFVIIa is a modified recombinant (r)FVIIa (NovoSeven®) in which the catalytic site is irreversibly inactivated by a synthetic tripeptide covalently bound to the rFVIIa molecule. The resulting modified rFVIIa retains its ability to bind to TF but loses its catalytic activity [11]. The excellent anticoagulant potency of FFR-rFVIIa (400 µg/kg) has been proven in a low-grade endotoxemia model in humans [12]. In conjunction with heparin, 200 and 400 µg/kg of FFR-rFVIIa effectively reduced thrombus formation in an ex vivo model in patients undergoing PCI [13]. Further, its anticoagulant potency has been investigated in combination with varying heparin concentrations in patients undergoing PCI (reviewed recently [11]). Phase III clinical trials with this interesting compound are awaited.

FXa inhibitors Danaparoid Danaparoid is an indirect FXa inhibitor, requiring antithrombin (AT) to exert its effect. Danaparoid has been on the market for a relatively long period of time and is indicated for the prophylaxis of DVT in patients undergoing orthopedic or major abdominal surgery. Danaparoid has been used most frequently as an effective treatment in heparininduced thrombocytopenia (HIT) (although this indication may not be licenced in all countries) [14, 15]. Its relative anti-FXa/anti-IIa activity is approximately 20:1, due to differences in half-life of the antiFXa and anti-FIIa activities. The prophylactic dose is 750 U s.c. o.d. or b.i.d., and patients with HIT may require up to 200 U/h, but bolus doses as high as 6400 U i.v. have been used in conjunction with cardiopulmonary bypass operations [16]. For DVT prophylaxis during hip-replacement surgery, its cost per day is approximately eight times more than LMWH.

Thrombosis in clinical practice

324

Fondaparinux Fondaparinux is a novel selective reversible FXa-inhibitor that, although based on the structure of heparin, is different from both heparin and LMWH [17]. Fondaparinux is a homogeneous pentasaccharide (molecular weight (MW): 1728 Dalton). The pharmacokinetics of fondaparinux are characterized by 100% bioavailability by s.c. route, lack of biometabolism, a Va similar to the blood volume (7–11 L), urinary excretion, and a plasma half-life of 14–21 h. Its binding affinity and dissociation constant (Kd) for AT, which may bind up to 97% of fondaparinux in plasma is 48±11 nmol/L [18]. Thus, low plasma concentrations of AT can be rate limiting for its activity [19]. There is a rapid onset of action with 50% maximal activity reached within 25 minutes after s.c. injection and peak concentrations are reached in 2 hours. No interactions with aspirin, warfarin, piroxicam, or digoxin have been noted [18, 20–22]. Fondaparinux does not affect prothrombin (PT) or thrombin clotting time, and has very weak effects on activated partial thromboplastin time (APTT). In contrast, fondaparinux activity can be determined by specific anti-FXa assays, but laboratory monitoring is currently not recommended, although it is unknown whether this could be beneficial. Thrombocytopenia (<100×109/L) occurred less commonly than with enoxaparin, and fondaparinux does not induce in vitro platelet aggregation in the presence of anti-heparin antibodies. Prophylaxis for venous thromboembolism The PENTATHLON study was a double-blind, randomized dose-ranging study (0.75–8 mg of fondaparinux) in elective hip-replacement patients. Fondaparinux given once daily was started 6 hours postoperatively, whereas enoxaparin (30 mg b.i.d.) was started 12–24 h postoperatively. Overall, there was a dosedependent decrease of venous thromboembolic events, but major bleeding also dosedependently increased so that the 6 and 8 mg doses were discontinued prematurely. The 3 mg dose of fondaparinux significantly decreased venous thromboembolism (VTE) as compared to enoxaparin [23]. Four randomized, double-blind phase III clinical trials designated PENTHIFRA, PENTATHLON 2000, EPHESUS, and PENTAMAKS were designed in parallel with sample sizes ranging from 1,034 to 2,309 patients. All trials were performed in patients undergoing orthopedic surgery with enoxaparin s.c. as the comparator drug. In the PENTHIFRA study, the first dose of enoxaparin (40 mg b.i.d.) was scheduled 12 hours preoperatively and the second dose 12–24 h after surgery for fracture of the femur. While fondaparinux was more effective (56% relative risk reduction (RRR) for VTE by day 11), minor bleeding events were also more frequent. Additionally, interpretation of this study is limited because of a huge number of protocol violations regarding the scheduled timing of drug administration of both substances [24]. The PENTATHLON 2000 study was carried out in patients undergoing hip replacement. Again, the first enoxaparin treatment (30 mg b.i.d.) was given 12–24 h postoperatively. The RRR (26%) for the prevention of VTE by fondaparinux was not significant [25].

New drugs and directions

325

In the EPHESUS trial, enoxaparin started with 40 mg before hip replacement, then continued with 40 mg o.d. Fondaparinux was more effective in preventing VTE with a 56% RRR [26]. The PENTAMAKS study of elective knee surgery used the same dose regimen as the PENTATHLON trial, and demonstrated a 55% RRR for VTE, but major bleeding occurred more frequently with fondaparinux (2% vs 0.2%) [27]. The PENTHIFRA-PLUS trial has been conducted to determine the optimal duration of prophylaxis with fondaparinux. Compared to patients treated for 1 week after hip-fracture surgery, the incidence of VTE was significantly lower among patients treated for 4 weeks postoperatively (1.4% vs 35.0%), resulting in a 96% RRR. Moreover, a significant reduction in the rate of symptomatic VTE with prolonged treatment (0.3% vs 2.7%, RRR 89%, p=0.02), without an increase in clinically relevant bleeding was seen. This study could set a new standard for length of VTE prophylaxis after orthopedic surgery of 4 weeks [28, 29]. Further potential indications for fondaparinux Prophylaxis of VTE after major abdominal surgery In the PEGASUS trial, fondaparinux 2.5 mg started postoperatively was compared to dalteparin 2,500 U preoperatively, followed by 5,000 U postoperatively for 7 days [30]. Twothirds of the patients in this study had cancer. VTE occurred in 4.6% of patients on fondaparinux and 6.1% on dalteparin corresponding to a non-significant RRR of 26% demonstrating noninferiority, with a minimal increase in bleeding. A post-hoc subgroup analysis suggested a significant decrease in VTE in cancer patients (RRR=41%; VTE in 4.7% vs 7.7% of patients, respectively). Treatment of VTE and pulmonary embolism The REMBRANDT trial was a phase IIb trial comparing 5–10 mg fondaparinux o.d. to dalteparin 100 U/kg b.i.d. in patients with acute VTE or PE. There were half as many recurrent thromboembolic events in the fondaparinux groups. Based on these results, the MATISSE DVT [31] and the MATISSE PE [32] trials have compared fondaparinux (7.5 mg s.c. o.d.) to enoxaparin (1 mg/kg b.i.d.) and full-dose heparin (average>30,000 U/day), respectively. Fondaparinux was noninferior in either condition [33]. Major and minor bleeding events were not different between groups. ARTEMIS was designed to investigate the prevention of VTE in acutely ill medical patients [34]. A total of 849 patients with heart failure (New York Heart Association class, NYHA III-IV), acute respiratory distress, and infection or inflammatory disease were enrolled in the study. All patients were randomized to once-daily s.c. injections with either 2.5 mg fondaparinux or placebo. There was a 50% RRR for VTE compared to placebo (10.5% control patients had VTE). None of the patients receiving fondaparinux died from fatal PE, as compared to five patients (1.5%) in the placebo group (p=0.029). In each treatment group, one major bleed occurred (p=n.s.).

Thrombosis in clinical practice

326

Cardiac patients The PENTALYSE trial was a phase IIb doseranging study comparing fondaparinux to heparin as an adjunct to thrombolytic therapy with alteplase in patients with ST-elevation myocardial infarction (STEMI). A trend toward less reocclusion of the infarct-related vessel has been observed for fondaparinux (1% vs 7% with UFH) [35]. Similarly, the PENTUA trial was a phase IIb dose-ranging study comparing fondaparinux (2.5–12 mg) to enoxaparin in unstable angina and non-STEMI, or heparin as an adjunct to thrombolytic therapy with alteplase in patients with MI. Whereas the lowest dose had a lower event rate than enoxaparin, the higher doses did not display heightened efficacy but appeared to induce additional bleeding episodes [36]. In addition, the MICHELANGELO Program has been started in patients with ACS. Conclusion A meta-analysis of studies in orthopedic surgery suggests not only a 55% RRR in VTE with fondaparinux in this indication, but also a higher frequency of overt bleeding with a bleeding index greater than 2 (2.3% vs 1.5%) as compared to enoxaparin [37]. However, most VTE events were distal DVT rather than proximal DVT or PE. On the basis of the current clinical trials, the FDA did not acknowledge the superior claim of fondaparinux in reference to enoxaparin [19]. In the approval for orthopedic surgery, the FDA stated that fondaparinux cannot be given to patients with renal failure and those with weight less than 110 lbs due to a risk of serious bleeding. As with LMWH, the epidural/spinal anesthesia or spinal puncture contraindication is also extended to fondaparinux. Idraparinux Idraparinux, another pentasaccharide, has an extra binding site for AT, thereby increasing its affinity for AT [38]. Its half-life (80 h) is considerably longer than fondaparinux, so that once weekly administration is possible. The PERSIST trial [39] randomized patients with acute symptomatic DVT (after an initial course of enoxaparin) to warfarin (INR 2– 3) or one of four doses of idraparinux (2.5–10 mg once weekly). A dose-dependent increase in major or clinically relevant bleeding was observed. Whereas the 2.5 mg group had less, the 10 mg group had significantly more bleeding events than warfarin. There was no obvious relationship between thrombotic burden and dose, so 2.5 mg dose will be pursued in future clinical trials. The VAN GOGH study is comparing idraparinux (2.5 mg once every week) to vitamin K antagonists for 3 and 6 months in patients with DVT or PE. The AMADEUS study will investigate the efficacy of idraparinux in atrial fibrillation. DX-9065a DX-9065a is a reversible, highly selective competitor for FXa. It is an effective anticoagulant when administered s.c., iv., or p.o., has a half-life of 7 hours, and is eliminated renally [40]. So far two clinical trials (phase I/II) have been published, the latter enrolling patients with coronary artery disease [41, 42].

New drugs and directions

327

Direct thrombin inhibitors Direct thrombin inhibitors, in contrast to heparins, act independently of AT. A meta-analysis of studies in patients with acute coronary syndromes (ACS) was undertaken [43] based on individual patient data from randomized trials comparing parenterally administered direct thrombin inhibitors (hirudin, bivalirudin, argatroban, efegatran, or inogatran) with heparin. In 11 randomized trials, 35,970 patients were assigned up to 7 days’ treatment with a direct thrombin inhibitor or heparin and followed up for at least 30 days. Compared with heparin, direct thrombin inhibitors were associated with a lower risk of death or myocardial infarction (MI) at the end of treatment (4.3% vs 5.1%; RRR=15%) and at 30 days (7.4% vs 8.2%; RRR=9%). This was due primarily to a reduction in MI (2.8% vs 3.5%; RRR=20%) with no apparent effect on deaths. Subgroup analyses suggested a beneficial effect of direct thrombin inhibitors on death or MI in trials of both ACS and PCI [43]. A reduction in death or MI was seen with hirudin and bivalirudin. Compared with heparin, there was an increased risk of major bleeding with hirudin but a reduction with bivalirudin. There was no excess in intracranial hemorrhage with direct thrombin inhibitors. Overall, direct thrombin inhibitors appear superior to heparin for the prevention of death or MI in patients with ACS. Hirudins Lepirudin Lepirudin is indicated for the treatment of heparin-induced thrombocytopenia. It can be given i.v. (targeted to an APTT ratio of 1.5–2.5, loading dose 0.1 mg/kg followed by 0.1 mg/kg/h according to the TIMI 9b trial [44]) or s.c. 0.5 mg/kg b.i.d. A variety of mostly higher dose regimens have recently been reviewed by Lubenow and Greinacher [45]. Still higher doses will be necessary for patients undergoing PCI or cardiopulmonary bypass. As a pivotal clinical trial (OASIS II) showed no significant difference in outcome in patients with ACS treated with lepirudin relative to heparin, lepirudin was not licensed for this indication [46]. Desirudin Desirudin 15 mg b.i.d. was compared to enoxaparin 40 mg o.d. s.c. after total hipreplacement therapy. There was a 28% RRR for all DVT and a 40% RRR for proximal DVT with desirudin [47]. It has been approved for the prevention of DVT after knee- and/or hip-replacement surgery in Europe and the United States, respectively.

Thrombosis in clinical practice

328

Bivalirudin Bivalirudin is a synthetic 20 amino acid peptide analogue of hirudin with a molecular weight of 2.18 kDa. Bivalirudin is a specific and reversible inhibitor infusable of thrombin that binds directly with both fluid-phase and clot-bound thrombin. Bivalirudin was evaluated in patients with unstable angina undergoing PCI in two randomized double-blind multicenter trials with identical protocols. All patients received 325 mg of aspirin. Additionally, patients received heparin or 2.5 mg/kg/h bivalirudin. Within 5 minutes after starting the infusion and prior to PCI a 1 mg/kg loading dose was given, after 4 hours the infusion rate was changed to 0.2 µg/kg/h for an average of 14 hours. Median activated clotting time (ACT) values were approximately 345 s after 5 and 45 minutes. Patients randomized to heparin received 175 U/kg, 5 minutes before PCI followed by 15 U/kg/h. The combined endpoints of death, MI, or revascularisation were reached in 6.2% and 7.9% of patients in the bivalirudin and heparin group, respectively, and major hemorrhage was encountered in 3.5% and 9.3% of patients, respectively [48]. The REPLACE-2 trial randomly assigned patients to receive intravenous bivalirudin (0.75 mg/kg bolus plus 1.75 mg/kg per hour for the duration of PCI), with provisional GpIIb/IIIa inhibition (n=2,999), or heparin (65 U/kg bolus) with planned GpIIb/IIIa inhibition (abciximab or eptifibatide) (n=3,011). Both groups received daily aspirin and a thienopyridine for at least 30 days after PCI. Provisional GpIIb/IIIa blockade was administered to 7.2% of patients in the bivalirudin group. Efficacy was similar between groups. In-hospital major bleeding rates were significantly reduced by bivalirudin (2.4% vs 4.1%; p<0.001). Thus, bivalirudin with provisional GpIIb/IIIa blockade is statistically noninferior to heparin plus planned GpIIb/IIIa blockade during contemporary PCI with regard to suppression of acute ischemic endpoints and is associated with less bleeding [49]. The combination of fibrinolytic therapy and heparin for acute MI fails to achieve reperfusion in 40–70% of patients, and early reocclusion occurs in a substantial number. A randomized, open-label trial compared bivalirudin with heparin in patients undergoing fibrinolysis with streptokinase for acute MI (HERO-2) [50]. Patients (n=17,073) with acute STEMI were randomly assigned an intravenous bolus and 48 hours infusion of either bivalirudin or heparin, together with a standard 1.5 million unit dose of streptokinase given directly after the antithrombotic bolus. Death rates were almost identical (10.8% and 10.9%). However, there were significantly fewer reinfarctions within 96 hours in the bivalirudin group than in the heparin group (RRR 30%). Severe bleeding occurred with similar frequency in both treatment groups. The rates of moderate and mild bleeding were significantly higher in the bivalirudin group than in the heparin group. Hence, bivalirudin did not reduce mortality compared with UFH, but did reduce the rate of reinfarction within 96 hours by 30%. Thus, bivalirudin is a new anticoagulant treatment option in patients with acute MI treated with streptokinase for which it has been licensed. Bivalirudin was as effective as heparin in reducing 30 day mortality in cardiac indications: data from a meta-analysis of four randomized trials among patients undergoing PTCA or treatment for ACS indicate that after 30–50 days of follow-up, bivalirudin was significantly more effective than heparin in reducing the incidence of

New drugs and directions

329

nonfatal MI and the combined endpoint of death or nonfatal MI. The most significant adverse events associated with bivalirudin are bleeding complications. In individual trials, bivalirudin was as well tolerated as heparin with, in general, a reduced incidence of bleeding complications. Additionally, bivalirudin provides a more consistent, predictable anticoagulant response. In 4,312 patients with unstable angina undergoing PTCA, the incidence of retroperitoneal bleeding, blood transfusion, and major hemorrhage was significantly lower in bivalirudin than heparin recipients. Data from the HERO-2 trial in patients with acute MI indicate that although bivalirudin recipients had a significantly higher incidence of mild or moderate bleeding than heparin recipients there was no difference in intracranial hemorrhage, severe bleeding, or transfusions. Data from a metaanalysis among 5,674 patients with ischemic heart disease show that bivalirudin recipients were at a significantly lower risk of hemorrhagic events than heparin recipients. Conclusions. Bivalirudin is an effective alternative to heparin in the prevention of ischemic complications in patients with unstable angina undergoing PTCA. In addition, the drug has shown potential in the treatment of patients with unstable angina not undergoing PCI. Recently, it has been approved for PCI also by the Committee for Medicinal Products for Human Use (CHMP) in Europe. For patients with MI, it is clear that bivalirudin can replace heparin in the management of MI where streptokinase is used as the thrombolytic agent. Further data are required on the efficacy of bivalirudin in patients undergoing thrombolysis with newer thrombolytics [51]. Argatroban Argatroban is a direct thrombin inhibitor synthesized to bind to the catalytic site of the thrombin molecule. It binds rapidly and reversibly to both clot-bound and soluble thrombin with a relatively short elimination half-life (~45 min) [52]. Argatroban is approved in the United States and Canada for both prophylaxis and treatment of thrombosis in patients with HIT and in the United States as an antithrombotic agent during PCI in patients with HIT or a history of HIT. In Japan, argatroban is approved for use in acute ischemic stroke and chronic peripheral occlusive disease [53]. Argatroban, given to patients with HIT and HIT with thrombosis (HITTS) in a large scale, nonrandomized, prospective trial, reduced a combined endpoint of morbidity and mortality when compared with historical controls. There is no need for dosage adjustment in patients with renal impairment because it is eliminated by hepatic metabolism. Preliminary reports document the feasibility of using argatroban for anticoagulation during hemodialysis and as adjunct to thrombolysis for treatment of MI. Current recommendations for argatroban monitoring are to use APTT for low doses and ACT (activated clotting time) for high doses. The specific inhibition of thrombin can be measured with ECT (ecarin clotting time). Dosing is initiated at 2 µg/kg/min and adjusted to maintain the APTT at 1.5–3 times the patient’s baseline. Ximelagatran The oral direct thrombin inhibitor ximelagatran is the first novel oral anticoagulant in more than 50 years. It is a pro-drug, which is given orally and bioconverted to

Thrombosis in clinical practice

330

melagatran, its active form (which can also be administered subcutaneously). Melagatran competitively and reversibly binds to the active site on thrombin, and because of its small molecular weight is able to penetrate clots and inhibit both free as well as clot-bound thrombin. Ximelagatran is rapidly absorbed (Cmax ~ 0.5 h), its bioavailability, as melagatran, averages 20% [54], and is independent of the dose administered. Ximelagatran fulfills many of the requirements of an ideal anticoagulant. Its predictable PK and PD enable use without routine coagulation monitoring or dose adjustment. Melagatran is not bound to plasma proteins and not metabolized. As it is predominantly renally excreted (80%), melagatran exposure correlates with renal function. Normal halflife is 3 hours in young healthy subjects and 4–5 h in patients. Due to its predictable pharmacokinetics and pharmacodynamics routine coagulation monitoring is not necessary. Importantly, ximelagatran has promising clinical efficacy, based on recently published papers and abstracts. Prevention of VTE in elective hip or knee replacement surgery Clinical trials have been designed to reflect different clinical practice between North America and Europe. (LMWH, started preoperatively is usually the regimen in Europe, whereas warfarin, started postoperatively is often, but not exclusively used in the United States.) The METHRO trials conducted in the European Union (EU) were the first to provide good evidence for the sequential use of subcutaneous melagatran and oral ximegalatran for the prophylaxis of VTE in orthopedic surgery. Timing and dose of prophylaxis have been shown to be important factors in determining efficacy and safety in orthopedic surgery, and the melagatran/ximelagatran clinical trial program has taken this into account (i.e. some regimens started preoperatively whereas METHRO III started postoperatively, differences in timing of administration also reflect differences in clinical practice). METHRO II, a dose-finding study in 1,876 patients, demonstrated a dosedependent decrease in total and proximal VTE for melagatran/ximelagatran initiated preoperatively, and provided significant differences between the highest dose of melagatran (3 mg)/ximelagatran (24 mg b.i.d.) versus dalteparin 5,000 U o.d. [55]. Melagatran was given s.c. before surgery, 7–11 h thereafter, followed by s.c. injections b.i.d. until oral administration could be started (1–3 days after surgery). Prophylaxis was maintained until venography 7–10 days after surgery. The highest dose of melagatran/ximelagatran (3 mg/24 mg, respectively) was associated with a significantly lower rate of VTE compared to dalteparin (15.1% vs 28.2%; p<0.0001) with a strong inverse relationship between VTE and the dose of melagatran/ximelagatran. In addition, the risk of bleeding was not significantly greater in the group taking the highest dose of melagatran/ ximelagatran than in the dalteparin group, suggesting the safety of this agent compared to standard UFH. However, a dose-response relationship was demonstrated in terms of risk of bleeding between the highest and lowest doses of melagatran/ximelagatran, which was statistically significant. In the METHRO III trial, melagatran (3 mg s.c.) initiated 4–12 h postoperatively, followed by oral ximelagatran (24 mg b.i.d.) was noninferior to enoxaparin 40 mg started preoperatively in patients with total hip or knee replacement [56]. A subgroup analysis showed that the first dose of melagatran should ideally be given 4–8 h postoperatively,

New drugs and directions

331

which is also stated in the label approved by the European Medicines Agency [57]. The incidence of total VTE with early melagatran dosing (4–8 h after operation: 27%) was significantly lower than with late dosing (>8 h postoperatively: 35%) and similar to preoperative enoxaparin (27%). The EXPRESS study was a randomized, double-blind study in 2,835 patients undergoing total hip or knee replacement [58]: one group of patients received melagatran 2 mg s.c. immediately before surgery followed by 3 mg s.c. in the evening after surgery, and then by oral ximelagatran 24 mg b.i.d. as a fixed dose. The comparator arm received enoxaparin s.c. 40 mg o.d., starting the evening before surgery. The total duration of active treatment was 8–11 days. The rate of proximal DVT plus PE was 2.3% in the ximelagatran group and 6.3% in the enoxaparin group (RRR=63%). The total rate of VTE was also significantly reduced from 26.6% in the enoxaparin group to 20.3%. However, bleeding events were also more frequently encountered with ximelagatran treatment (3.3% vs 1.2% serious bleeding events). As described earlier, due to differences in clinical practice between Europe and North America, warfarin was used as a comparator in the EXULT trials. The EXULT A study has demonstrated that ximelagatran (36 mg b.i.d.) started on the morning after surgery was more effective in reducing the overall incidence of VTE, without an increase in major bleeding. This randomized, double-blind trial compared a regimen of 7–12 days of oral ximelagatran, at a dose of 24 or 36 mg b.i.d., starting the morning after surgery, with warfarin therapy started the evening of the day of surgery. Among the 1,851 patients, oral ximelagatran at a dose of 36 mg b.i.d. was superior to warfarin with respect to the primary composite endpoint of VTE and death from all causes (20.3% vs 27.6%; p=0.003). There were no significant differences between these two groups with respect to major bleeding (incidence 0.8% and 0.7%, respectively), or the composite secondary endpoint of proximal DVT, PE, and death (2.7% vs 4.1%; p—0.17) [59]. In the EXULT B study, oral ximelagatran (36 mg b.i.d.) was started on the morning after surgery. It provided superior efficacy for the combined endpoint of total VTE and death as compared to warfarin (target INR 2.5) [60]. In all of the trials in orthopedic surgery described above, ximelagatran was used for periods of up to 12 days without routine coagulation monitoring or dose adjustment. Elevations of serum alanine aminotransferases (ALAT) were infrequent, transient, and occurred more often with LMWH than with ximelagatran [55, 59]. Long-term secondary prevention of VTE The optimal duration of oral anticoagulation after VTE is a matter of debate. The risk of recurrence of DVT in these patients is approximately 5–10% per year [61]; however, continuing warfarin treatment is not only inconvenient but also carries a yearly risk of bleeding of 3% [62]. The THRIVE III study evaluated the role of ximelagatran in the long-term secondary prevention of VTE. It was a double-blind, placebocontrolled study involving 1,233 patients who had been already treated with 6 months of standard therapy (initially heparin and warfarin, and then warfarin alone) for an episode of DVT/PE. Patients were randomized to fixed-dose ximelagatran (24 mg b.i.d.) or placebo for a further 18 months without monitoring of coagulation. The composite endpoint of allcause mortality and recurrent VTE was significantly lower in the patients assigned to

Thrombosis in clinical practice

332

ximelagatran compared to placebo (hazard ratio—0.23; 95% CI 0.14–0.39, p<0.0001), with no significant difference in bleeding events. Three deaths in the placebo group were attributable to PE [63]. The incidence of ALAT elevation >3×the upper limit of normal (ULN) was 6.0% in the ximelagatran group and 1.0% in the placebo arm of the THRIVE trial. The majority of ALAT elevations were transient, occuring most commonly during the first 6 months of therapy, and decreased whether treatment was continued or discontinued. Treatment of VTE The THRIVE treatment study included 2,491 patients with DVT and/or PE. Fixed dose ximelagatran (36 mg b.i.d.) was compared to dose-adjusted warfarin (target INR 2.0–3.0) after initial treatment with enoxaparin (1 mg/kg b.i.d.). The estimated cumulative risk of recurrent VTE was 2.1% vs 2.0% and risk of major bleeding 1.3% vs 2.2% in the ximelagatran and warfarin groups, respectively. Transient elevations of ALAT>3×ULN were seen in 9.6% of patients who received ximelagatran and 2.0% with enoxaparin/warfarin [64]. In the majority of patients, the ALAT elevations were not associated with symptoms and generally returned towards normal with or without discontinuation of the drug, consistent with findings in other studies. Prevention of stroke and systemic embolic events in nonvalvular atrial fibrillation The SPORTIF III trial included patients with nonvalvular atrial fibrillation at moderate to high risk of stroke [65]. Patients (n=3,410) were randomized to receive fixed dose ximelagatran (36 mg b.i.d.) or dose-adjusted warfarin (INR 2.0–3.0) using an open-label design. Ximelagatran was noninferior to warfarin in preventing stroke or systemic embolic events. The primary event rate was 2.3% per year with warfarin and 1.6% per year with ximelagatran (p=0.10). The “intention to treat” population showed a trend and the “on treatment” population even a significant RRR of 44%. Bleeding events were also reduced in the ximelagatran treatment arm. SPORTIF V was similar to SPORTIF III, except that treatment allocation was doubleblind, with ximelagatran-treated patients undergoing sham INR testing. In SPORTIF V, 1.6%/year primary events occurred in the ximelagatran group versus 1.2%/year in the warfarin group, during 6,405 patient-years exposure (mean 20 months). This trial confirmed the noninferiority of ximelagatran relative to warfarin found in SPORTIF III. Major bleeding events occured in 2.4% and 3.1% of patients on ximelagatran or warfarin, respectively (www.clinicaltrialresults.org). Serum ALAT levels rose transiently to greater than 3×ULN in 6.0% of patients taking ximelagatran and in 0.8% of patients taking warfarin (p<0.001), typically within the first 6 months of treatment, but generally returned towards baseline levels regardless of treatment continuation.

New drugs and directions

333

Cardiac indications The ESTEEM study was a phase II dose-finding study that assessed the effectiveness of ximelagatran and aspirin compared to aspirin alone for the prevention of death, nonfatal MI, and severe recurrent ischemia after a recent MI [66]. Within 14 days of the index event, 1,833 participants were randomized to oral ximelagatran (24 mg, 36 mg, 48 mg, or 60 mg b.i.d.) or to placebo for 6 months. All patients received aspirin (160 mg o.d.). Oral ximelagatran significantly reduced the risk for the primary endpoint compared with placebo from 16.3% (102 of 638) to 12.7% (154 of 1,245) (hazard ratio 0.76, 95% CI 0.59–0.98, p=0.036). There was no doseresponse relationship between the ximelagatran dose and outcome. Major bleeding events were rare, 1.8% (23 of 1,245) and 0.9% (6 of 638) in the combined ximelagatran and placebo groups, respectively. The efficacy and safety of ximelagatran plus aspirin are in line with those previously observed for the combination of aspirin and warfarin which appeared beneficial in this patient population [67]. Conclusions Ximelagatran is a highly effective anticoagulant and is likely to replace warfarin for use in longterm therapy. The results of the wide-ranging clinical trial program suggest that ximelagatran has the potential of becoming the anticoagulant of choice for the prevention and treatment of VTE and prevention of stroke in AF. The advantages of ximelagatran are its predictable pharmacokinetic and pharmacodynamic properties, wide therapeutic margin, fixed dose oral administration, low potential for drug interactions, lack of interaction with food/alcohol, and the lack of the need for coagulation monitoring. Ximelagatran is likely to be of benefit for many at-risk elderly patients including a large patient population with AF who are currently undertreated. In addition, the THRIVE III study could change current clinical practice towards longerterm anticoagulation for secondary prevention of VTE. On the other hand, a disadvantage of ximelagatran may be its twice-daily administration and the requirement for testing of liver function (ALAT) at least in the initial months. Because of the concerns and uncertainties over liver function and possible increase in vascular events, the USA Federal Cardiovascular and Renal Drugs Advisory Committee (FDA) in 2004 has recommended against the approval of ximelagatran for all of its application indications: the prevention of VTE in patients undergoing knee-replacement surgery; the prevention of stroke and systemic embolism association with AF and in the long-term secondary prevention after standard treatment for an episode of VTE. In Europe, the regulatory situation for ximelagatran is more complex. In May 2004, Exanta was approved by the European regulatory authorities for the short-term indication, the prevention of VTE in patients undergoing hip- or knee-replacement surgery, and it has since been made available in nine European countries and Argentina. The approved regimen based on the METHRO III study utilizes a first s.c. dose of melagatran 3 mg given 4–8 h post-surgery, followed by ximelagatran 24 mg b.i.d. However, following a review in January 2005 by the French Regulatory Authority (AFSSAPS) of the Exanta regulatory submission to allow EU licensing, the AFSSAPS

Thrombosis in clinical practice

334

requested further clinical information regarding the efficacy and safety of Exanta in AF to allow a definitive risk-benefit assessment to be made before approval was given. For VTE treatment—as with the USA FDA—the AFSSAPS did not believe the data presented in the single THRIVE treatment study provides adequate support for this use of Exanta in the treatment of VTE and are proposing a rejection of this indication. Dabigatran (BIBR 1048/BIBR 953) BIBR 1048 is a lipophilic pro-drug which is metabolized into BIBR 953, a potent small molecule thrombin inhibitor with an IC50 of 9 nM. Phase II trials have been completed in orthopedic surgery [68]. The primary objective of the BISTRO-2 study was to establish a dose-response relationship with regard to efficacy and safety of dabigatran etexilate in the prevention of VTE in patients undergoing total hip or total knee replacement. In a multicenter, parallel-group, double-blind study, 1,973 patients undergoing total hip or knee replacement were randomized to oral doses of dabigatran etexilate (50 and 150 mg b.i.d., 300 mg o.d., and 225 mg b.i.d.), with the first dose administered 1–4 h after surgery, or enoxaparin (40 mg o.d.) with the first dose administered subcutaneously prior to surgery, for a total of 6–10 days. VTE occurred in 28.5%, 17.4%, 16.6%, 13.1%, and 24% of patients assigned to dabigatran etexilate 50, 150 mg b.i.d., 300 mg o.d., 225 mg b.i.d. and enoxaparin, respectively. A significant dose-dependent decrease in the frequency of VTE was seen with increasing doses of dabigatran etexilate. When compared with enoxaparin, significantly lower VTE rates were seen in patients receiving dabigatran etexilate 150 mg b.i.d. (odds ratio 0.65, p=0.04), 225 mg b.i.d. (0.47, p=0.0007), and 300 mg o.d. (0.61, p=0.02). Major bleeding rates were significantly lower with 50 mg b.i.d. than with enoxaparin (p=0.047) but increased compared to enoxaparin with all higher doses of dabigatran etexilate, reaching statistical significance with the dabigatran etexilate 300 mg o.d. dose (absolute increase +2.6%, 95% CI 0.1–5.2). Hence, in these patients undergoing total hip or knee replacement, oral administration of dabigatran etexilate, commenced early in the postoperative period demonstrates good clinical benefit in comparison to preoperatively administered enoxaparin.

Use of endogenous coagulation inhibitors Plasma-derived AT [69] (more recently also recombinant AT [70, 71]) and Protein C concentrate [72, 73] have been used for a long time to replenish plasma pools in congenital or acquired deficiency states. Recombinant human activated protein C (rhAPC; drotrecogin alfa activated) inactivates FVa and FVIIIa. Apart from its anticoagulant effect, a number of other potential mechanisms of action have been proposed, although their relevance in vivo is unclear. It has a relatively short half-life so that steady-state concentrations are reached within 2 hours after start of infusion [3]. rhaPC (24 µg/kg/h) has recently been shown to reduce mortality by 6% (absolute) in patients with severe sepsis [74]. Interestingly, its anticoagulant effects were limited and resulted only in a 30% diminution of D-dimer levels in these patients after 4 days of treatment. Similarly, rhAPC had no measurable effects on coagulation activation in lowgrade endotoxemia [3]. Its use should therefore be restricted to patients with a high risk

New drugs and directions

335

for mortality and should follow the current summary of product characteristics. In contrast, trials with AT [75] and tissue factor pathway inhibitor (TFPI; tifacogin) [6] failed to show similar benefit (vide supra). In all three trials, heparin appeared to mitigate the effects of the natural anticoagulants, and AT without heparin appeared to be beneficial in a prespecified subgroup [75]. However, the nonstratified use of heparin in these studies does not permit any firm conclusions, and further studies examining the interactions of rhAPC and heparin have been mandated by the FDA. Protein C has been used successfully in a number of case series of purpura fulminans caused by meningococcal septicemia: under treatment with protein C, coagulation parameters normalized and mortality rates were markedly lower than those predicted [72, 73], but controlled trials in this indication are unfortunately lacking. ART123 ART123 is a soluble thrombomodulin with a long half-life (28 h and 63 h after i.v. and s.c. injection). Binding of thrombin to thrombomodulin inhibits its activity and additionally converts thrombin into a potent activator of protein C. However, ART123 neither affects routine global coagulation tests nor FV or FVIII activity. ART123 has recently been tested (phase II) in the prevention of VTE after hip-replacement surgery. ART123 was started 2–4 h postoperatively (days 1 and 6) and the 0.3 mg s.c. dose induced less major bleeding than the 0.45 mg s.c. dose [76].

Antiplatelet drugs Clopidogrel Clopidogrel, a new thienopyridine derivative similar to ticlopidine, is an inhibitor of platelet aggregation induced by ADP, and thereby prolongs the bleeding time [77, 78]. After activation by cytochrome P450 (mainly CYP1A)-mediated hepatic metabolism, clopidogrel is a selective and irreversible inhibitor of the purinergic P2Y12 receptor. Plasma protein binding is about 95% for clopidogrel and its inactive metabolite SR26334. The time of maximal concentration Tmax of the metabolite is 0.8 h, and very low levels of the inactive parent compound are present in the plasma. No dose adjustments

Table 15.2 Pivotal clinical trials that have been performed with clopidogrel. Trial acronym

Population studied

CAPRIE [79]

Prior myocardial infarction, stroke, peripheral arterial disease

CREDO [80]

Percutaneous coronary intervention

CURE [81]

Unstable angina, non STEMI

PCI-CURE [82]

Subset of CURE undergoing percutaneous coronary intervention

STEMI: ST-elevation myocardial infarction.

Thrombosis in clinical practice

336

are needed in patients with renal impairment but there is only limited data in patients with advanced cirrhosis (Child-Pugh class C or D). As concomitant administration of clopidogrel with naproxen increased occult gastrointestinal blood loss, caution is recommended when using NSAIDs. A total of four pivotal trials have been published so far (Table 15.2), which are described here. CAPRIE was a randomized trial designed to assess the relative efficacy of clopidogrel (75 mg o.d.) and aspirin (325 mg o.d.) in reducing the risk of a composite outcome cluster of ischemic stroke, MI, or vascular death [79]. Patients (n=19,185) suffering from atherosclerotic vascular disease manifested as either recent ischemic stroke, recent MI, or symptomatic peripheral arterial disease. Patients treated with clopidogrel had an annual 5.3% risk of ischemic stroke, MI, or vascular death compared with 5.8% with aspirin. These rates reflect a significant (p=0.043) RRR of 9% in favor of clopidogrel. A subgroup analysis showed that patients with peripheral arterial disease may profit most from clopidogrel relative to aspirin. Further retrospective subgroup analysis showed that clopidogrel provided potentially greater benefit in patients with a history of CABG or diabetes mellitus, and in those receiving concomitant lipid-lowering therapy [78]. Thus, long-term administration of clopidogrel to patients with atherosclerotic vascular disease was shown to be more effective than aspirin in reducing the combined risk of ischemic stroke, MI, or vascular death. Despite the use of aspirin, there is still a risk of ischemic events after PCI. The CREDO study [80] demonstrated that at 1 year, longterm clopidogrel therapy (in combination with aspirin) was associated with a 27% RRR of death, MI, or stroke (absolute reduction, 3%). Patients were randomly assigned to receive a 300 mg of clopidogrel loading dose or placebo 3–24 h before PCI. Thereafter, all patients received clopidogrel, 75 mg o.d., through day 28. From day 29 through 12 months, patients in the loading-dose group received clopidogrel, 75 mg o.d., and those in the control group received placebo. Clopidogrel pretreatment did not significantly reduce the combined risk of death, MI, or urgent target vessel revascularization at 28 days. However, in a prespecified subgroup analysis, patients who received clopidogrel at least 6 hours before PCI experienced an RRR of 39% (p=0.051) for this endpoint compared with no reduction with treatment less than 6 hours before PCI. Risk of major bleeding at 1 year increased, but not significantly (8.8% with clopidogrel vs 6.7% with placebo; p=0.07). Thus the CREDO trial showed that following PCI, long-term (1-year) clopidogrel therapy significantly reduced the risk of adverse ischemic events. The CREDO trial indicated that a loading dose of clopidogrel given at least 3 hours before the procedure would not reduce events at 28 days, but that longer intervals between the loading dose and PCI may reduce events. The CURE study [81] was performed in patients with ACS without STEMI, who were randomized to clopidogrel plus aspirin or aspirin alone. After an interim analysis, the steering committee recommended that only patients with ECG changes or elevated cardiac markers at entry should be included. The median treatment duration was 9 months (range 3–12 months). There was RRR of 20% for the combined endpoint of death from cardiovascular causes, nonfatal MI, or stroke. While major and minor bleeding were more common in the combined treatment group, there was no significant increase in lifethreatening, fatal, or intracranial bleeding. In patients with scheduled cardiopulmonary bypass, study medication was stopped (median 5 days) before the operation and no

New drugs and directions

337

excessive postoperative bleeding was observed. However, when the study drugs were stopped less than 5 days before surgery, bleeding was increased. The PCI-CURE study focused on the results in the 2,658 patients (21%) from the CURE study who underwent PCI [82]. Patients were pretreated with aspirin and clopidogrel for approximately 1 week. After PCI, most patients (>80%) in both groups received open-label clopidogrel plus aspirin for 2–4 weeks, after which the study drug was restarted for a mean of 8 months. The primary endpoint was a composite of cardiovascular death, MI, or urgent target-vessel revascularization within 30 days of PCI. In the clopidogrel group 4.5% had the primary endpoint, compared with 6.4% in the placebo group (corresponding to a 30% risk reduction, p=0.03). The benefit was observed within 14 days, with continuing benefit for 12 months. Overall (including events before and after PCI) there was a 31% reduction in cardiovascular death or MI (p=0.002). Hence, in patients with ACS receiving aspirin, a strategy of clopidogrel pretreatment followed by long-term therapy is beneficial in reducing major cardiovascular events, compared with placebo. Clopidogrel for the secondary prevention of stroke The MATCH study included high-risk patients (n=7,600) with prior transient ischemic attacks or stroke [83]. All patients received 75 mg o.d. clopidogrel and were randomized to receive 75 mg o.d. aspirin or placebo in addition. Combined treatment insignificantly reduced the RR of the primary endpoint by 6% (ischemic stroke, MI, vascular death, rehospitalization due to ischemic events), and the risk of any type of stroke by only 2%. Combination therapy was associated with a doubling in life-threatening bleeding, either gastrointestinal or cerebral (2.6% vs 1.3% with monotherapy). The WATCH study [84] randomized patients suffering from heart failure to aspirin 162 mg o.d., clopidogrel 75 mg o.d., or warfarin (INR: 2.5–3.0). The study aimed to recruit 4,500 patients and was stopped by the sponsor (Department of Veterans’ Affairs) due to poor recruitment after randomization of 1,587 patients. There was no difference in the primary endpoint rate (death, MI, stroke, 19.8–21.8%) in the three groups. Patients on aspirin were more frequently hospitalized for heart failure (plus 31%) than patients on warfarin, but patients on warfarin had a greater incidence of bleeding complications. Prospect into the near future A number of pivotal clinical trials are ongoing or are being planned with clopidogrel each including a large number of patients (1,500–40,000), the indications of which are outlined in Table 15.3.

Thrombosis in clinical practice

338

Table 15.3 Ongoing or planned pivotal trials with clopidogrel. Trial acronym

Population studied

ACTIVE

Atrial fibrillation

ARCH

Aortic arch atheroma

CAMPER

Peripheral arterial intervention

CASPAR

Peripheral arterial bypass surgery

CHARISMA

Secondary and high-risk primary prevention

CLARITY

Acute STEMI

COMMIT

Acute STEMI

MATCH

High-risk recent transient ischemic attacks or stroke

STEMI: ST-elevation myocardial infarction.

Safety issues The estimated incidence of ticlopidine-associated thrombotic thrombocytopenic purpura is 1 per 1,600–5,000 patients treated. A series of cases of thrombotic thrombocytopenic purpura were also observed for clopidogrel: in one study, 11 patients were identified by active surveillance by the medical directors of blood banks (3 patients), hematologists (6 patients), and the manufacturer of clopidogrel (2 patients) [85]: 10 of the 11 patients received clopidogrel for 14 days or less before the onset of thrombotic thrombocytopenic purpura. Although 10 of the 11 patients had a response to plasma exchange, 2 required 20 or more exchanges before clinical improvement occurred, and 2 had relapses while not receiving clopidogrel. One patient died despite undergoing plasma exchange soon after diagnosis. This demonstrates that thrombotic thrombocytopenic purpura can occur after the initiation of clopidogrel therapy, often within the first two weeks of treatment. While ticlopidine use often precipitated profound neutropenia, this is not a problem seen with clopidogrel treatment, although a case of fatal aplastic anemia has recently been reported [86]. Cost-effectiveness Increased prescription of aspirin for secondary prevention of coronary heart disease is attractive from a cost-effectiveness perspective. Various views exist about the costeffectiveness (one estimate: 27,000 USD per quality-adjusted life-year gained) [87]. One of these studies concluded that, because clopidogrel is more costly, its incremental costeffectiveness is currently unattractive, unless its use is restricted to patients who are ineligible for aspirin. It is probably wise to administer the drug in combination with aspirin to selected patients at high risk of coronary heart disease.

New drugs and directions

339

Conclusions CREDO and PCI-CURE suggest that treatment of patients who undergo PCI, particularly those suffering from ACS, should be started as early as possible. Further, CREDO and CURE suggest that a prolonged duration of clopidogrel treatment is likely to be beneficial. Evidence-based medicine supports the following prescription recommendation: clopidogrel should be administered as a loading dose (300 mg, in the case of elective PCI at least >6 h before intervention) followed by 75 mg (1 tablet o.d.) for the secondary prevention of cardiovascular diseases, particularly following ACS without STEMI or PCI for 9–12months [88]. GpIIb/IIIa inhibitors Only three infusible GpIIb/IIIa inhibitors have reached the market: the chimeric monoclonal antibody fragment abciximab [89] and the cyclic small molecule heptapeptide eptifibatide [90]. Currently, another GpIIb/IIIa inhibitor, tirofiban [91, 92], which (similar to eptifibatide) is approved for the medical therapy of patients with nonST segment elevation ACS, has not received indication for use in the PCI setting. Huge investments have been made into the development of oral GpIIb/IIIa inhibitors, but since a meta-analysis has demonstrated deleterious effects of GpIIb/IIIa inhibitors on survival, most companies have stopped their development of orally active GpIIb/IIIa inhibitors. In fact, there is a large discrepancy between the state-of-the-art technology deployed in the design and clinical development of these novel antiplatelet drugs and the 1960s methodology to assess their functional effects: mostly and probably inadequately, aggregometry has been used for the assessment of these drugs. However, this low shear stress system, working at unphysiologically low calcium concentrations, cannot adequately assess what happens in vivo under high shear rates and physiological levels of calcium [93]. Thus, one major indication for GpIIb/IIIa inhibitors which remains is the periprocedural use in patients undergoing PCI to restore laminar flow. Abciximab has the most robust data in patients undergoing PCI, particularly high-risk individuals. In PCI patients with lower risk (e.g. elective stenting), eptifibatide is a reasonable first-line option, while tirofiban is currently not recommended as an alternative to abciximab [88]. For individuals with signs and symptoms of ACS, specifically unstable angina or nonSTEMI, eptifibatide or tirofiban is recommended in high-risk patients when a conservative approach is used (PCI is not planned). Abciximab is recommended [88] only when PCI is planned and troponin is elevated. In patients with STEMI, abciximab is the only GPIIb/IIIa inhibitor evaluated in large, welldesigned investigations. For medical management in combination with a fibrinolytic agent, the role of abciximab remains unclear. For patients undergoing primary PCI for the management of STEMI, the available evidence supports the use of abciximab for angioplasty but not primary stenting [94], albeit further investigation is warranted [95]. Similarly, a double bolus eptifibatide and half-dose tPA may improve quality and speed of reperfusion [96]. Few head-to-head comparisons are available, but the TARGET study showed superiority of the mouse-human chimeric antibody abciximab (0.25 µg/kg bolus plus infusion of 0.125 mg/kg/min for 12 h) over the synthetic drug tirofiban [97]. The primary endpoint

Thrombosis in clinical practice

340

was a composite of death, nonfatal MI, or urgent target-vessel revascularization at 30 days, which was reached by 7.6% and 6% in the tirofiban and abciximab groups, respectively. Unfortunately, no direct comparisons of double bolus eptifibatide [98, 99] and abciximab have yet been made. However, abciximab has two major disadvantages: it has a relatively high rate of rapid-onset severe thrombocytopenias, and its costs are substantially higher as compared to both of the small molecule inhibitors. AJW200 AJW200 is an anti-von Willebrand factor (vWF) antibody with a relatively short halflife (-12 h) which has recently been investigated in a phase I trial [100]. AJW200 dosedependently reduces vWF:ristocetin cofactor activity (RiCoF), whereas vWF:Antigen (Ag) levels are not affected. However, an approximately 75% decrease in vWF:RiCoF was associated with a 2-fold prolongation in the closure time (measured with the platelet function analyzer PFA-100) but no change in the bleeding time in healthy volunteers. vWF is an important risk factor for arterial thrombosis [101, 102] and arterial stenosis induces high shear rates, where anti-GpIIb/IIIa inhibitors may only be effective at very high concentrations. Thus, an anti-vWF strategy may fill an important gap in our future therapeutic armamentarium.

Thrombolytic therapy Irrespective of the thrombolytic agent used, all patients require additional coadministration of aspirin (either i.v. or chewed, but not entericcoated) [103]. Accelerated alteplase (tissue plasminogen activator, tPA) with concomitant APTT adjusted i.v. heparin (bolus: 60 U/kg maximal 4,000 U; 12 U/kg/h max. 1,000 U/h for 24–48 h; target APTT 50–70 s) provides better outcome than streptokinase, but is also associated with a higher stroke risk [104–106]. A number of novel thrombolytic agents have been engineered, including reteplase and tenecteplase, which—due to improved pharmacokinetics—result in easier administration schedules. In the case of tenecteplase (TNK-tPA), specific mutations at three sites in the alteplase molecule result in 15-fold higher fibrin specificity, 80-fold reduced binding affinity to the physiological plasminogen activator inhibitor (PAI)-1 and 6-fold prolonged plasma half-life (22 vs 3.5 min). Consequently, TNK-tPA can be administered as a single intravenous bolus of 30–50 mg (0.53 mg/kg bodyweight) over 5–10 seconds, in contrast to the 90-minute accelerated infusion regimen of alteplase. TNK-tPA has demonstrated equivalent efficacy and improved safety, compared with the current gold standard alteplase, in a large mortality trial (ASSENT-2). Thus, a single bolus dose of TNK-tPA is equivalent to accelerated tPA, and is associated with less need for blood transfusions. In the ASSENT-3 trial TNK-tPA together with enoxaparin (30 mg i.v. bolus and 1 mg/kg s.c. every 12 h) reduced the risk of in-hospital reinfarction or refractory ischemia when compared to heparin. There was only a modest increase in noncerebral bleeding events in the enoxaparin group [107]. However, in the ASSENT-3 PLUS trial enoxaparin in combination with pre-hospital thrombolytic therapy significantly increased intracerebral bleeding when compared to heparin. In contrast to tenecteplase, reteplase is given as two

New drugs and directions

341

bolus injections of 10 U, 30 minutes apart [108]. However, the efficacy of both these agents needs to be compared to early PCI [109, 110]: PCI appears to be superior to inhospital thrombolysis when initiated within 2 hours (DANAMI-2 trial), but not to prehospital lysis (CAPTIM study) [111]. The primary endpoint was a composite of death, nonfatal reinfarction, and nonfatal disabling stroke at 30 days. Rescue angioplasty was done in 26% of the patients in the fibrinolysis group. The rate of the primary endpoint was 8.2% (34 patients) in the prehospital-fibrinolysis group and 6.2% (26 patients) in the primary angioplasty group. Death rates were 3.8% versus 4.8%, respectively. Thus, a combined prehospital pharmacological and mechanical reperfusion strategy might prove useful and is currently under investigation.

Conclusions Considerable limitations of old anticoagulants including heparins, warfarin, and antiplatelet drugs such as aspirin created the need for new anticoagulants. Numerous new compounds with different mechanisms of action have been developed, and some have been already approved for clinical use. Although their pharmacokinetic profile differs markedly, their final pharmacodynamics, speaking in terms of clinical efficacy, may not. Numerous trials to study the noninferiority of the novel agents compared to those currently in use are ongoing or planned. It is anticipated that a wider choice of anticoagulant and antiplatelet drugs will become available in the near future, this will broaden and enhance the treatment options available, but will necessitate revisions of treatment guidelines and new educational initiatives.

References [1] Ruef J, Katus HA. New antithrombotic drugs on the horizon. Expert Opin Investig Drugs 2003; 12: 781–97. [2] Girard TJ, Nicholson NS. The role of tissue factor/factor VIIa in the pathophysiology of acute thrombotic formation. Curr Opin Pharmacol 2001; 1:159–63. [3] Derhaschnig U, Reiter R, Knobl P, et al. Recombinant human activated protein C (rhAPC, drotrecogin alfa activated) has minimal effect on markers of coagulation, fibrinolysis and inflammation in acute human endotoxemia. Blood 2003; 102:2093–8. [4] Bajaj MS, Bajaj SP. Tissue factor pathway inhibitor: potential therapeutic applications. Thromb Haemost 1997; 78:471–7. [5] de Jonge E, Dekkers PE, Creasey AA, et al. Tissue factor pathway inhibitor dose-dependently inhibits coagulation activation without influencing the fibrinolytic and cytokine response during human endotoxemia. Blood 2000; 95:1124–9. [6] Abraham E, Reinhart K, Opal S, et al. Efficacy and safety of tifacogin (recombinant tissue factor pathway inhibitor) in severe sepsis: a randomized controlled trial. JAMA 2003; 290:238– 47. [7] Buddai SK, Toulokhonova L, Bergum PW, et al. Nematode anticoagulant protein c2 reveals a site on factor Xa that is important for macromolecular substrate binding to human prothrombinase. J Biol Chem 2002; 277:26689–98.

Thrombosis in clinical practice

342

[8] Lee A, Agnelli G, Buller H, et al. Dose-response study of recombinant factor VIIa/tissue factor inhibitor recombinant nematode anticoagulant protein c2 in prevention of postoperative venous thromboembolism in patients undergoing total knee replacement. Circulation 2001; 104:74–8. [9] Moons AH, Peters RJ, Bijsterveld NR, et al. Recombinant nematode anticoagulant protein c2, an inhibitor of the tissue factor/factor VIIa complex, in patients undergoing elective coronary angioplasty. J Am Coll Cardiol 2003; 41:2147–53. [10] Hedner U, Erhardtsen E. Potential role for rFVIIa in transfusion medicine. Transfusion 2002;42: 114–24. [11] Hedner U, Erhardtsen E. Future possibilities in the regulation of the extrinsic pathway: rFVIIa and TFPI. Ann Med 2000; 32(Suppl. 1):68–72. [12] Jilma B, Marsik C, Mayr F, et al. Pharmacodynamics of active site-inhibited factor VIIa in endotoxininduced coagulation in humans. Clin Pharmacol Ther 2002; 72:403–10. [13] Lev EI, Marmur JD, Zdravkovic M, et al. Antithrombotic effect of tissue factor inhibition by inactivated factor VIIa: an ex vivo human study. Arterioscler Thromb Vasc Biol 2002;22: 1036– 41. [14] Ibbotson T, Perry CM. Danaparoid: a review of its use in thromboembolic and coagulation disorders. Drugs 2002; 62:2283–314. [15] Acostamadiedo JM, Iyer UG, Owen J. Danaparoid sodium. Expert Opin Pharmacother 2000;!: 803–14. [16] Gillis S, Merin G, Zahger D, et al. Danaparoid for cardiopulmonary bypass in patients with previous heparin-induced thrombocytopenia. Br J Haematol 1997; 98:657–9. [17] Samama MM, Gerotziafas GT. Evaluation of the pharmacological properties and clinical results of the synthetic pentasaccharide (fondaparinux). Thromb Res 2003; 109:1–11. [18] Donat F, Duret JP, Santoni A, et al. The pharmacokinetics of fondaparinux sodium in healthy volunteers. Clin Pharmacokinet 2002; 41(Suppl. 2):1–9. [19] Walenga JM, Jeske WP, Samama MM, et al. Fondaparinux: a synthetic heparin pentasaccharide as a new antithrombotic agent. Expert Opin Investig Drugs 2002;11:397–407. [20] Mant T, Fournie P, Ollier C, et al. Absence of interaction of fondaparinux sodium with digoxin in healthy volunteers. Clin Pharmacokinet 2002; 41(Suppl. 2):39–45. [21] Ollier C, Faaij RA, Santoni A, et al. Absence of interaction of fondaparinux sodium with aspirin and piroxicam in healthy male volunteers. Clin Pharmacokinet 2002; 41(Suppl. 2):31–7. [22] Faaij RA, Burggraaf J, Schoemaker RC, et al. The synthetic pentasaccharide fondaparinux sodium does not interact with oral warfarin. Clin Pharmacokinet 2002; 41(Suppl. 2):27–9. [23] Turpie AG, Gallus AS, Hoek JA. A synthetic pentasaccharide for the prevention of deep-vein thrombosis after total hip replacement. N Engl J Med 2001; 344:619–25. [24] Eriksson BI, Bauer KA, Lassen MR, et al. Fondaparinux compared with enoxaparin for the prevention of venous thromboembolism after hip-fracture surgery. N Engl J Med 2001; 345: 1298–304. [25] Turpie AG, Bauer KA, Eriksson BI, et al. Postoperative fondaparinux versus postoperative enoxaparin for prevention of venous thromboembolism after elective hip-replacement surgery: a randomised double-blind trial. Lancet 2002; 359:1721–6. [26] Lassen MR, Bauer KA, Eriksson BI, et al. Postoperative fondaparinux versus preoperative enoxaparin for prevention of venous thromboembolism in elective hip-replacement surgery: a randomised double-blind comparison. Lancet 2002; 359:1715–20. [27] Bauer KA, Eriksson BI, Lassen MR, et al. Fondaparinux compared with enoxaparin for the prevention of venous thromboembolism after elective major knee surgery. N Engl J Med 2001; 345:1305–10. [28] Turpie AG, Eriksson BI, Lassen MR, et al. Fondaparinux, the first selective factor Xa inhibitor. Curr Opin Hematol 2003; 10:327–32.

New drugs and directions

343

[29] Eriksson BI, Lassen MR. Duration of prophylaxis against venous thromboembolism with fondaparinux after hip fracture surgery: a multicenter, randomized, placebo-controlled, doubleblind study. Arch Intern Med 2003; 163:1337–42. [30] Agnelli G, Bergqvist D, Cohen A, et al. A randomized double-blind study to compare the efficacy and safety of fondaparinux with dalteparin in the prevention of venous thromboembolism after high-risk abdominal surgery: the Pegasus study. J Thromb Haemost 2003; 1(Suppl. 1):OC006. [31] The MATISSE-DVT trial, a randomized, doubleblind study comparing once-daily fondaparinux (Arixtra®) with the low-molecular-weight heparin (LMWH) enoxaparin, twice daily, in the initial treatment of symptomatic deep-vein thrombosis (DVT). J Thromb Haemost 2003; 1(Suppl. 1): OC332. [32] The MATISSE-PE trial, a multicenter, randomized, open study comparing once-daily fondaparinux (Arixtra®) with adjusted-dose intravenous unfractionated heparin (UFH) in the initial treatment of acute symptomatic pulmonary embolism (PE). J Thromb Haemost 2003; 1(Suppl. 1):OC331. [33] The Matisse Investigators. Fondaparinux (Arixtra) in Comparison to (Low Molecular Weight) Heparin for the Initial Treatment of Symptomatic Deep Venous Thrombosis or Pulmonary Embolism—The Matisse Clinical Outcome Studies. Blood 2002; 100:302a. [34] Cohen AT, Gallus AS, Lassen MR, et al. Fondaparinux vs. placebo for the prevention of venous thromboembolism in acutely ill medical patients (artemis). J Thromb Haemost 2003; 1(Suppl. 1):OC006. [35] Coussement PK, Bassand JP, Convens C, et al. A synthetic factor-Xa inhibitor (ORG31540/ SR9017A) as an adjunct to fibrinolysis in acute myocardial infarction. The PENTALYSE study. Eur Heart J 2001; 22:1716–24. [36] Van de Werf F. New data in treatment of acute coronary syndromes. Am Heart J 2001 ;142: S16–S21. [37] Turpie AG. Overview of the clinical results of pentasaccharide in major orthopedic surgery. Haematologica 2001; 86:59–62. [38] Koopman MM, Buller HR. Short- and long-acting synthetic pentasaccharides. J Intern Med 2003; 254:335–42. [39] Persists Investigators. A novel long acting synthetic factor Xa inhibitor (idraparinux sodium) to replace warfarin for secondary prevention in deep vein thrombosis. A phase II evaluation. Blood 2002; 100:301a. [40] Herbert JM, Bernat A, Dol F, et al. DX 9065A a novel, synthetic, selective and orally active inhibitor of factor Xa: in vitro and in vivo studies. J Pharmacol Exp Ther 1996; 276:1030–8. [41] Dyke CK, Becker RC, Kleiman NS, et al. First experience with direct factor Xa inhibition in patients with stable coronary disease: a pharmacokinetic and pharmacodynamic evaluation. Circulation 2002; 105:2385–91. [42] Murayama N, Tanaka M, Kunitada S, et al. Tolerability, pharmacokinetics, and pharmacodynamics of DX-9065a, a new synthetic potent anticoagulant and specific factor Xa inhibitor, in healthy male volunteers. Clin Pharmacol Ther 1999; 66:258–64. [43] Direct thrombin inhibitors in acute coronary syndromes: principal results of a meta-analysis based on individual patients’ data. Lancet 2002; 359:294–302. [44] Antman EM. Hirudin in acute myocardial infarction. Thrombolysis and Thrombin Inhibition in Myocardial Infarction (TIMI) 9B trial. Circulation 1996;94:911–21. [45] Lubenow N, Greinacher A. Heparin-induced thrombocytopenia: recommendations for optimal use of recombinant hirudin. BioDrugs 2002;14: 109–25. [46] Effects of recombinant hirudin (lepirudin) compared with heparin on death, myocardial infarction, refractory angina, and revascularisation procedures in patients with acute myocardial ischaemia without ST elevation: a randomised trial. Organisation to Assess Strategies for Ischemic Syndromes (OASIS-2) Investigators. Lancet 1999; 353:429–38.

Thrombosis in clinical practice

344

[47] Eriksson BI, Wille-Jorgensen P, Kalebo P, et al. A comparison of recombinant hirudin with a low-molecular-weight heparin to prevent thromboembolic complications after total hip replacement. N Engl J Med 1997; 337:1329–35. [48] Bittl JA, Chaitman BR, Feit F, et al. Bivalirudin versus heparin during coronary angioplasty for unstable or postinfarction angina: final report reanalysis of the Bivalirudin Angioplasty Study. Am Heart J 2001; 142:952–9. [49] Lincoff AM, Bittl JA, Harrington RA, et al. Bivalirudin and provisional glycoprotein IIb/IIIa blockade compared with heparin and planned glycoprotein IIb/IIIa blockade during percutaneous coronary intervention: REPLACE-2 randomized trial. JAMA 2003; 289:853–63. [50] White H. Thrombin-specific anticoagulation with bivalirudin versus heparin in patients receiving fibrinolytic therapy for acute myocardial infarction: the HERO-2 randomised trial. Lancet 2001; 358:1855–63. [51] Carswell CI, Plosker GL. Bivalirudin: a review of its potential place in the management of acute coronary syndromes. Drugs 2002; 62:841–70. [52] McKeage K, Plosker GL. Argatroban. Drugs 2001; 61:515–22. [53] Kondo LM, Wittkowsky AK, Wiggins BS. Argatroban for prevention and treatment of thromboembolism in heparin-induced thrombocytopenia. Ann Pharmacother 2001; 35:440–51. [54] Gustafsson D, Elg M. The pharmacodynamics and pharmacokinetics of the oral direct thrombin inhibitor ximelagatran and its active metabolite melagatran: a mini-review. Thromb Res 2003; 109(Suppl. 1):S9–S15. [55] Eriksson BI, Bergqvist D, Kalebo P, et al. Ximelagatran and melagatran compared with dalteparin for prevention of venous thromboembolism after total hip or knee replacement: the METHRO II randomised trial. Lancet 2002; 360:1441–7. [56] Eriksson BI. Clinical experience of melagatran/ ximelagatran in major orthopaedic surgery. Thromb Res 2003; 109(Suppl. 1):S23–S29. [57] Dahl OE, Eriksson BI, Agnelli G, Cohen AT, Mouret P, Rosencher N, et al. Influence of time of first dose on the efficacy and safety of postoperative ximelagatran for prevention of thromboembolism after orthopaedic surgery [Abstract]. Pathophysiol Haemost Thromb 2004; 33(Suppl. 2):OC063. [58] Eriksson BI, Agnelli G, Cohen AT, et al. The direct thrombin inhibitor melagatran followed by oral ximelagatran compared with enoxaparin for the prevention of venous thromboembolism after total hip or knee replacement: the EXPRESS study. J Thromb Haemost 2003; 1:2490–6. [59] Francis CW, Berkowitz SD, Comp PC, et al. Comparison of ximelagatran with warfarin for the prevention of venous thromboembolism after total knee replacement. N Engl J Med 2003; 349:1703–12. [60] Colwell CW, Berkowitz SD, Comp PC, et al. Randomized, Double-Blind Comparison of Ximelagatran, an Oral Direct Thrombin Inhibitor, and Warfarin to Prevent Venous Thromboembolism (VTE) after Total Knee Replacement (TKR): EXULT B. Blood 2003; 102:39a. [61] Prandoni P, Lensing AW, Cogo A, et al. The longterm clinical course of acute deep venous thrombosis. Ann Intern Med 1996; 125:1–7. [62] Schulman S, Granqvist S, Holmstrom M, et al. The duration of oral anticoagulant therapy after a second episode of venous thromboembolism. The Duration of Anticoagulation Trial Study Group. N Engl J Med 1997; 336:393–8. [63] Schulman S, Wahlander K, Lundstrom T, et al. Secondary Prevention of Venous Thromboembolism with the Oral Direct Thrombin Inhibitor Ximelagatran. N Engl J Med 2003; 349:1713–21. [64] Francis CW, Berkowitz SD, Bounameaux H, et al. Efficacy and safety of the oral direct thrombin inhibitor ximelagatran compared with current standard therapy for acute, symptomatic deep vein thrombosis, with or without pulmonary embolism. The THRIVE treatment study. Blood 2003; 102:7a.

New drugs and directions

345

[65] Olsson SB. Stroke prevention with the oral direct thrombin inhibitor ximelagatran compared with warfarin in patients with non-valvular atrial fibrillation (SPORTIF III): randomised controlled trial. Lancet 2003; 362:16