Anticoagulation Therapy for Deep Venous Thrombosis and Pulmonary Embolism

Transcription

1 HEMATOLOGY BOARD REVIEW MANUAL PUBLISHING STAFF PRESIDENT, PUBLISHER Bruce M. White EXECUTIVE EDITOR Debra Dreger SENIOR EDITOR Becky Krumm, ELS ASSISTANT EDITOR Laurie Garrison EDITORIAL ASSISTANT Meghan Cunningham SPECIAL PROGRAMS DIRECTOR Barbara T. White, MBA PRODUCTION DIRECTOR Suzanne S. Banish PRODUCTION ASSOCIATES Tish Berchtold Klus Christie Grams PRODUCTION ASSISTANT Mary Beth Cunney ADVERTISING/PROJECT MANAGER Patricia Payne Castle NOTE FROM THE PUBLISHER: This publication has been developed without involvement of or review by the American Board of Internal Medicine. Endorsed by the Association for Hospital Medical Education The Association for Hospital Medical Education endorses HOSPITAL PHYSICIAN for the purpose of presenting the latest developments in medical education as they affect residency programs and clinical hospital practice. Anticoagulation Therapy for Deep Venous Thrombosis and Pulmonary Embolism Series Editor and Contributing Author: Richard S. Stein, MD, FACP Associate Professor of Medicine, Division of Hematology-Oncology, Vanderbilt University Medical Center, Nashville, TN Contributing Author: Anne T. Neff, MD Assistant Professor of Pathology and Medicine, Vanderbilt University Medical Center, Nashville, TN Table of Contents Introduction Warfarin Heparin Prevention of Deep Venous Thrombosis and Pulmonary Embolism Treatment of Deep Venous Thrombosis and Pulmonary Embolism Screening for Hypercoagulable States Board Review Questions Answers Explanations References Cover Illustration by Christine Schaar Copyright 2000, Turner White Communications, Inc., 125 Strafford Avenue, Suite 220, Wayne, PA , 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, mechanical, electronic, photocopying, recording, or otherwise, without the prior written permission of Turner White Communications, Inc. The editors are solely responsible for selecting content. Although the editors take great care to ensure accuracy, Turner White Communications, Inc., will not be liable for any errors of omission or inaccuracies in this publication. Opinions expressed are those of the authors and do not necessarily reflect those of Turner White Communications, Inc. Hematology Volume 1, Part 1 1

2 HEMATOLOGY BOARD REVIEW MANUAL Anticoagulation Therapy for Deep Venous Thrombosis and Pulmonary Embolism Series Editor and Contributing Author: Richard S. Stein, MD, FACP Associate Professor of Medicine Division of Hematology-Oncology Vanderbilt University Medical Center Nashville, TN Contributing Author: Anne T. Neff, MD Assistant Professor of Pathology and Medicine Vanderbilt University Medical Center Nashville, TN INTRODUCTION It has been estimated that 100,000 to 200,000 people die each year as a consequence of pulmonary embolism (PE). Although therapy for acute PE may be life saving, most deaths due to PE occur within half an hour of acute embolization. Therefore, the best means of preventing fatalities caused by PE is to prevent the occurrence of PE. Because almost all emboli arise from deep venous thrombosis (DVT) in leg veins, the critical clinical issue is the appropriate prevention and management of DVT. This review considers the standard of care regarding the prophylaxis and management of DVT in medical and surgical patients. These standards undergo continuous review, and the recommendations in this review are based on the Fifth American College of Chest Physicians Consensus Conference on Antithrombotic Therapy, published in Additionally, because heparin and warfarin are the cornerstones of antithrombotic therapy, this review considers in detail the biology underlying the use of these agents in clinical practice. Because the availability of low-molecularweight heparins (LMWHs) has had considerable impact on the therapy of DVT and has made outpatient therapy feasible for many patients, this review also considers the rationale for the clinical use of LMWH. Other recent advances in the field of antithrombotic therapy, including new considerations regarding the duration of therapy in DVT and PE and the rational work-up for suspected hypercoagulable states, are also considered. WARFARIN BIOCHEMISTRY AND MECHANISM OF ACTION Warfarin (Coumadin) is the most widely used oral anticoagulant. It exerts its anticoagulant effect by 2 Hospital Physician Board Review Manual

3 interfering with the interconversion of vitamin K and its 2,3-epoxide. Vitamin K is a cofactor for the post-translational modification of the vitamin K dependent proteins, including clotting factors II (prothrombin), VII, IX, and X. The dose-response relationship of warfarin is variable, and many drugs can influence the effects of warfarin. For example, cholestyramine reduces the anticoagulant effect of warfarin by limiting its absorption. Phenylbutazone, metronidazole, omeprazole, and trimethoprim-sulfamethoxazole potentiate the effect of warfarin by inhibiting its clearance. Conversely, the effect of warfarin is limited by barbiturates, rifampicin, and carbamazepine which induce the activity of hepatic enzymes and thereby increase the clearance of warfarin. USE OF INTERNATIONAL NORMALIZED RATIO FOR MONITORING ANTICOAGULATION Because warfarin interferes with the synthesis of clotting factors II, VII, IX, and X, it increases both the prothrombin time (PT) and the activated partial thromboplastin time (aptt). However, the PT is generally used to monitor the effect of warfarin. The reagents used to perform the PT test vary in their sensitivity to the effects of warfarin (ie, to the decreased levels of clotting factors that result from warfarin). At one time, the anticoagulant effect of warfarin was measured by comparing the patient s PT with a control PT; however, the resulting ratio varied depending on which reagents were used. If reagents are responsive to warfarin, less warfarin is needed to produce a specific increase in the PT than is required if less responsive reagents are used. 2 Recognition of this phenomenon has led to the elimination of the PT ratio to assess anticoagulation and has led to the widespread use of the international normalized ratio (INR). The INR adjusts for the sensitivity of the reagents and corrects the PT ratio by considering the sensitivity (ISI = international sensitivity index) of the reagents. 3 Use of the INR methodology has increased the effectiveness and decreased the toxicity associated with warfarin therapy. For routine anticoagulation (ie, management of DVT or PE), the target INR is 2.0 to 3.0. Higher target ranges ( ) are generally used for anticoagulation in patients with mechanical heart valves. PRINCIPLES OF WARFARIN THERAPY For the treatment of DVT/PE, warfarin is generally started at a dose of 5 mg daily, a dose likely to produce a therapeutic elevation of the INR in 4 to 5 days. This dose may be adjusted on the basis of patient weight and the perceived risk of bleeding. It should be noted that at the start of warfarin therapy, levels of factor VII, which has a short half-life, fall first, and elevations of the PT and INR at that time do not reflect true anticoagulation. This is especially true if loading doses of warfarin (20 40 mg) are used, a practice that cannot be endorsed because it has no intrinsic advantages and can produce misleading PT and INR values. Because it takes several days to achieve an anticoagulant effect with warfarin, patients with DVT or PE should be started on heparin, and heparin and warfarin therapy should overlap for several days. Overlapped therapy is not always essential, however. For patients with chronic atrial fibrillation, warfarin may be started as a single agent. Patients on warfarin therapy are generally monitored daily until the therapeutic range of anticoagulation is achieved and maintained for at least 2 days. The INR is then monitored 2 or 3 times per week for a few weeks and then less frequently. Once the INR is relatively stable, tests can be performed every 2 to 4 weeks. Because diet and drugs can affect the INR, variations in the INR may occur and indicate a need for more frequent monitoring. Managing an Elevated INR Elevations of the INR beyond the therapeutic range occur frequently in patients receiving warfarin. Treatment options to reduce the INR include reducing the warfarin dose, withholding doses of warfarin, administering vitamin K 1, and administering fresh frozen plasma. In responding to excessive elevations of the INR, several physiologic principles must be considered. First, because the effect of vitamin K depends on the synthesis of clotting factors, a therapeutic reduction in INR is often delayed by 24 hours. Second, although administration of vitamin K 1 can correct an elevated PT or INR, if doses greater than 1 mg are used, it may be very difficult to re-establish effective anticoagulation. Third, administration of fresh frozen plasma can lead to immediate correction of the PT and INR, but carries a risk of volume overload as well as a risk of transmitting viral diseases (eg, hepatitis C) a risk that is, fortunately, quite small. Fourth, although the effect of fresh frozen plasma is rapid, it is also short lived (approximately 6 hours) and additional therapy may be needed. The following recommendations from Hirsch et al 4 regarding elevations of the INR are based on clinical Hematology Volume 1, Part 1 3

4 observations rather than controlled clinical studies, and should be regarded as guidelines rather than rigid rules: If the INR is elevated above the therapeutic range but is lower than 5.0, and the patient does not have clinically significant bleeding, the warfarin dose can be lowered or the next warfarin dose omitted, and warfarin can be resumed when the INR approaches the therapeutic range. If the INR is above 5.0 but lower than 9.0 and the patient does not have clinically significant bleeding, 1, 2, or more doses of warfarin may be omitted and therapy reinstituted when the INR falls into the therapeutic range. Alternatively, a dose of warfarin can be omitted and a low dose (1 2.5 mg) of vitamin K 1 can be given subcutaneously (SC). If the INR is above 9.0 and the patient does not have clinically significant bleeding, a higher dose of vitamin K 1 (3 5 mg) can be given orally. This will likely correct the INR in 24 to 48 hours, but further anticoagulation may be difficult. If a more rapid effect is needed because of serious bleeding or the need to perform surgery, vitamin K 1 can be given at a dose of 10 mg by slow intravenous (IV) infusion with or without the addition of fresh frozen plasma. ADVERSE EFFECTS OF WARFARIN THERAPY The major complication of warfarin therapy is bleeding, which can be reduced by maintaining the INR between 2.0 to 3.0. Warfarin also crosses the placenta and can produce an embryopathy characterized by central nervous system abnormalities and fetal bleeding. Therefore, warfarin should not be used in the first trimester and should be avoided in women planning to become pregnant. Heparin is preferred when anticoagulation is indicated during pregnancy. Warfarin Necrosis Warfarin necrosis is a rare but serious complication of warfarin therapy. 5 This complication occurs within 3 to 8 days of initiating warfarin and involves extensive thrombosis of capillaries and veins in subcutaneous fat. Common sites include the breast and anterior thighs, but loss of limbs and infarction of male genitalia have been reported. The exact pathophysiology is unknown. However, the timing suggests that early in the course of administering warfarin, levels of vitamin K dependent factors that prevent clotting (eg, protein C, protein S) decrease before a reduction occurs in the levels of vitamin K dependent factors that promote clotting (ie, clotting factors II, VII, IX, and X). This would lead to a paradoxical hypercoagulable state. The association of warfarin necrosis with protein C deficiency and protein S deficiency support this theory, but cases have occurred in patients without these predisposing factors. If patients with warfarin necrosis require lifelong anticoagulation, a rational approach is to resume warfarin after the patient is first anticoagulated with heparin. HEPARIN BIOCHEMISTRY AND MECHANISM OF ACTION Heparin, which is given parenterally, is a glycosaminoglycan, the major anticoagulant effect of which depends on a pentasaccharide that binds to antithrombin (AT), previously referred to as antithrombin III (ATIII). 6 The resulting conformational change in AT increases its ability to inactivate clotting factors, including thrombin (IIa), Xa, and IXa. Unlike warfarin, which affects protein synthesis and has a delayed effect on coagulation, heparin acts immediately, a quality that increases its clinical utility. Formulations of Heparin Two forms of heparin are used in clinical practice. Unfractionated heparin (UFH) is a mixture of molecules with molecular weights ranging from 5000 to 30,000 daltons. The mean molecular weight of UFH is approximately 15,000 daltons, which corresponds to 50 monosaccharide chains. LMWHs are composed of molecules with molecular weights ranging from 1500 to 10,000 daltons. The mean molecular weight of LMWH is approximately 5000 daltons, corresponding to 15 monosaccharide chains. This difference in the number of monosaccharide chains leads to one of the major differences between UFH and LMWH. In order for the heparin-at complex to decrease the activity of factor IIa, heparin must have at least 18 saccharides (ie, a molecular weight of 5400 daltons). As a result, UFH has both anti-xa and anti-iia activity, whereas LMWH has primarily anti-xa activity. 6 Because LMWH contains some molecules of relatively higher molecular weight, this distinction is not absolute. The anti-xa:anti-iia activity of clinically effective LMWHs are 2:1 to 4:1, whereas the anti-xa:anti-iia activity of UFH is 1:1. Pharmacokinetics of Heparin There are other important biological differences between UFH and LMWH. Clearance of heparin is 4 Hospital Physician Board Review Manual

5 Table 1. Weight-Based Nomogram for the Use of Unfractionated Heparin Initial dose aptt < 35 seconds (< 1.2 control) aptt seconds ( control) aptt seconds ( control) aptt seconds (2.3 3 control) aptt > 90 seconds (> 3 control) 80 U/kg body weight bolus, then 18 U/kg body weight per hour 80 U/kg body weight bolus, then increase by 4 U/kg body weight per hour 40 U/kg body weight bolus, then increase by 2 U/kg body weight per hour No change Decrease infusion rate by 2 U/kg body weight per hour Hold infusion for 1 hour, then decrease infusion rate by 3 U/kg body weight per hour Adapted with permission from Raschke RA, Reilly BM, Guidry JR, et al. The weight based heparin dosing nomogram compared with a standard care nomogram. A randomized control study. Ann Intern Med 1993;119:875. related to molecular size, with higher molecular weight forms being cleared more rapidly. As a result, UFH generally has a shorter half-life (1 hour) than LMWH (2 4 hours). Additionally, UFH is associated with more non-specific binding to proteins and cells, including endothelial cells and platelets, than is LMWH. This non-specific binding complicates the clinical administration of UFH. Some of the proteins to which heparin binds are acute-phase reactants, which makes the dose-response relationship difficult to predict in individual patients. Additionally, this non-specific protein binding is saturable, which means that at higher doses of UFH, the half-life of UFH increases. The differences between UFH and LMWH with respect to non-specific binding to platelets are clinically important for another reason. The anti-platelet antibody that is involved in the development of heparininduced thrombocytopenia (HIT) is directed against the complex of heparin and platelet factor 4. The fact that LMWH is associated with less non-specific binding to platelets likely accounts for the fact that HIT is less common with LMWH therapy as there is less antigen to induce antibody formation. PRINCIPLES OF HEPARIN THERAPY Monitoring Anticoagulant Effects Because the dose-response relationship of UFH is relatively unpredictable, its use requires monitoring of its anticoagulant effect. The clinically effective level of UFH is 0.2 to 0.4 U/mL. However, heparin levels are rarely assayed by clinical laboratories. Instead, for patients receiving UFH, the therapeutic effect is measured indirectly by calculating the ratio of the patient s aptt to a control aptt. Although a ratio of slightly greater than 1.5:1 is often considered optimal, the clinically effective ratio depends on the reagents that are used. Each laboratory must therefore determine the therapeutic aptt ratio that correlates with heparin levels of 0.2 to 0.4 U/mL using their particular reagents. By contrast, the effect of LMWH is predictable enough that dose schedules based on weight can be used, without monitoring, in the management of DVT, PE, and unstable angina. It should be noted, however, that in patients with renal failure (creatinine clearance < 30 ml/min) or in patients with marked obesity (weight > 150 kg), the reliability of weight-based regimens of LMWH has not been established. In these circumstances, therapeutic levels for anti-xa activity need to be measured, or, alternatively, UFH should be used instead of LMWH. Dosing Considerations Heparin is not absorbed after oral administration and must be administered either IV or SC. UFH is generally given SC when the drug is given prophylactically to prevent DVT, as in patients undergoing surgery. UFH is generally given IV when used to manage DVT or PE. LMWH is generally given SC regardless of clinical indication, although different doses are used for prophylaxis as compared to treatment of existing disease. In treatment of patients with PE, achieving a therapeutic heparin level within 24 hours is associated with a decreased risk of recurrent thromboembolism. However, despite the use of several treatment nomograms 7 such as the one shown in Table 1, it is often difficult to achieve optimal levels within 24 hours of initiating UFH therapy. Because of changes in non-specific binding, daily dose adjustments are often necessary during UFH therapy. ADVERSE EFFECTS OF HEPARIN THERAPY Although long-term heparin therapy can produce Hematology Volume 1, Part 1 5

6 osteopenia and alopecia, as with all anticoagulants, the major side effect of heparin is bleeding. This can be limited by careful monitoring. Serious bleeding is generally managed by discontinuation of heparin. In a setting of life-threatening bleeding or heparin overdose, neutralization of heparin may require the use of protamine. One mg of protamine neutralizes 100 U of UFH. Protamine is administered slowly, by IV infusion, over a 10-minute period. Rarely, it may produce anaphylaxis in patients with prior exposure to protamine (eg, diabetics treated with protamine-containing insulin). In using protamine, it is important to determine how much heparin needs to be neutralized. For patients receiving UFH by continuous infusion, considerations based on the half-life of UFH suggest that between 1.5 and 2.0 times the hourly dose need to be neutralized. For each specific LMWH, neutralization recommendations for protamine are contained in the individual package insert. For example, 1 mg of protamine neutralizes 1 mg of enoxaparin. Because LMWH is generally given as a discrete SC dose, the amount of drug remaining to be neutralized can be estimated on the basis of the drug half-life. Heparin-Induced Thrombocytopenia Two forms of thrombocytopenia are associated with heparin therapy. 8 A benign, non-immune thrombocytopenia may occur in the first few days of heparin therapy. Its clinical significance is unclear because the thrombocytopenia resolves despite continuation of heparin therapy. The disease termed heparin-induced thrombocytopenia is an immunologic disorder associated with a risk of thrombosis and represents a potential disastrous consequence of heparin therapy. HIT is generally defined as a decrease in platelet count to a level lower than /mm 3, or to a level less than one-half the baseline level, occurring more than 5 days after the initiation of heparin therapy. In a large series of patients undergoing orthopedic surgery, a platelet count of less than /mm 3 was used to define HIT. The incidence of HIT was found to be 3% for UFH and 0% for LMWH. Although HIT can occur with LMWH, the incidence appears to be lower than that observed with UFH. Because of its paradoxical association with thrombosis, HIT is a potentially life-threatening complication of heparin therapy. By mechanisms that are poorly understood, the anti-platelet antibodies that form in patients with HIT lead to platelet activation and promote clotting. The white clot syndrome of arterial clotting is a well recognized manifestation of HIT; however, it is often not appreciated that approximately 80% of the clots associated with HIT are venous thromboses. Tests to confirm the diagnosis of HIT exist, and platelet activation assays to detect HIT appear to have a sensitivity and specificity of approximately 90%. However, HIT is a clinical syndrome which must be considered independent of laboratory confirmation when thrombocytopenia with or without thrombosis, or thrombocytopenia with skin lesions, occurs more than 5 days after the initiation of heparin therapy. Because of the devastating consequences of HIT with thrombosis, it is prudent to monitor the platelet count of patients on heparin therapy at least every other day during the period of high risk for HIT that is, 5 to 14 days after starting heparin therapy. If a patient develops thrombocytopenia during this period, unless an extremely strong alternative diagnosis exists, heparin therapy must be stopped and an alternate anticoagulant employed. There is incomplete consensus regarding the optimal management of HIT. However, there is agreement regarding the following facts. First, in patients who develop HIT as a result of UFH therapy, LMWH is contraindicated. Although the risk of HIT is lower with LMWH, in patients who already have HIT, LMWH can increase antibody production. 9 Secondly, warfarin alone should not be used to treat HIT because of the increased risk of venous limb gangrene. 10 However, warfarin therapy may be administered to a patient with HIT once the patient is adequately anticoagulated with a drug that reduces thrombin generation and once the platelet count has recovered to a level greater than /mm 3. In patients with confirmed or suspected HIT, alternate anticoagulant therapy is mandated because of the high risk of thrombosis, even if thrombosis is not present when heparin is discontinued. Drugs that limit thrombin generation, specifically recombinant hirudin (lepirudin), argatroban, and danaparoid are considered effective in HIT. Dosage recommendations for these drugs are presented in Table 2. As of November 2000, only hirudin and argatroban had been approved by the US Food and Drug Administration (FDA) for treatment of HIT; danaparoid had been approved for DVT prophylaxis and is often used off label in the treatment of HIT. 6 Hospital Physician Board Review Manual

7 PREVENTION OF DEEP VENOUS THROMBOSIS AND PULMONARY EMBOLISM RISK FACTORS FOR DEEP VENOUS THROMBOSIS Since the nineteenth century, it has been recognized that the major factors predisposing to DVT are blood vessel damage, blood hypercoagulability, and stasis Virchow s triad. Surgical procedures, especially orthopedic procedures involving the hip or knee, are a major risk factor for DVT, as it is estimated that 40% to 80% of such patients would develop DVT in the absence of anticoagulation. The relationship between risk category and the estimated incidence of venous thromboembolic disease (VTE) in surgical patients is shown in Table 3. Although proximal venous thrombosis is considered the major risk factor for PE, studies have suggested that up to 30% of patients with deep calf vein thrombosis have proximal extension of the clot. Thus, deep calf vein thrombosis is generally considered a surrogate marker for the risk of PE and the risk of fatal thromboembolic disease. Prevention of DVT is considered essential in decreasing the estimated 100,000 to 200,000 deaths per year that occur due to PE. PREVENTIVE STRATEGIES The specific strategy used to prevent DVT depends on the estimated risk. For patients considered low risk (ie, young patients with no risk factors for VTE undergoing minor surgical procedures), early ambulation is all that is required. 11 Patients older than 40 years undergoing major surgical procedures (eg, abdominal surgery) are considered moderate risk. For patients at moderate risk, 4 strategies are considered to have comparable efficacy. 11 These strategies are (1) the use of low-dose unfractionated heparin (LDUH), 5000 U administered 2 hours preoperatively and then every 12 hours; (2) the use of LMWH every 12 hours postoperatively at the prophylactic dose level; (3) the use of elastic stockings operatively and postoperatively; and (4) the use of intermittent pneumatic compression (IPC) operatively and postoperatively. It should be noted that when LDUH is given prophylactically, the effect of heparin is generally not monitored. For high-risk general surgery patients, it is recommended that pharmacologic methods (ie, LDUH or LMWH) be combined with IPC. 11 In selected patients, Table 2. Dosage Recommendations for Drugs Used in the Treatment of Heparin-Induced Thrombocytopenia Recombinant hirudin (lepirudin) Loading dose: 0.4 mg/kg body weight bolus IV Maintenance dose: 0.15 mg/kg body weight per hour IV, with adjustments to maintain the aptt at times the median of the normal laboratory range Argatroban Initial dose: 2 µg/kg body weight per minute IV Maintenance dose: adjust to maintain the aptt at times the median of the normal laboratory range Danaparoid Loading dose: 2250 U bolus (1500 U if body weight is < 60 kg; 3000 U if body weight is kg) Followed by 400 U/hour for 4 hours Followed by 300 U/hour for 4 hours Maintenance dose: U/hour to maintain anti-xa activity levels between 0.5 and 0.8 U/mL aptt = activated partial thromboplastin time; IV = intravenously. Data from Hirsh H, Dalen JE, Anderson DR, et al. Oral anticoagulants: mechanism of action, clinical effectiveness, and optimal therapeutic range. Chest 1998;60:499S. warfarin (rather than heparin) may be considered with the aim of maintaining the INR between 2.0 and 3.0. For high-risk orthopedic patients, recommendations are slightly different. For patients undergoing total hip replacement, 3 recommended options are as follows 11 : LMWH started 12 to 24 hours after surgery, warfarin (with a target INR of ) started preoperatively or immediately after surgery, or adjusted dose UFH started preoperatively. For patients undergoing knee replacement surgery, use of LMWH, warfarin, or IPC is recommended. The optimum duration of prophylactic therapy in patients undergoing hip or knee surgery is not known, but at least 1 week of therapy is recommended. TREATMENT OF DEEP VENOUS THROMBOSIS AND PULMONARY EMBOLISM ANTICOAGULANT THERAPY Until LMWH became available, patients with DVT or PE were managed as inpatients. To prevent clot Hematology Volume 1, Part 1 7

8 Table 3. Surgical Risk Category and the Incidence of Venous Thromboembolic Disease Proximal Clinical Fatal Calf Vein Vein Pulmonary Pulmonary Thrombosis Thrombosis Embolus Embolus Low-risk surgery Uncomplicated minor surgery* in patients age < 40 y and no risk factors for VTE Moderate-risk surgery Major surgery in patients age > 40 y and no other risk factors for VTE 2% 0.4% 0.2% 0.002% 10% 20% 2% 4% 1% 2% 0.1% 0.4% High-risk surgery Major surgery, in patients age > 60 y without additional risk factors for VTE, or major surgery in patients age y with additional risk factors for VTE 20% 40% 4% 8% 2% 4% 0.4% 1.0% Very-high-risk surgery Major surgery in patients age > 40 y with prior VTE or with malignancy; patients with elective major lower extremity surgery; patients with hip fracture, stroke, spinal cord injury, or multiple trauma 40% 80% 10% 20% 4% 10% 1% 5% MI = myocardial infarction; VTE = venous thromboembolic disease. *Minor surgery is a brief procedure, usually less than 30 minutes, that does not involve the abdominal cavity or the lower extremities. All other surgery is considered major. Risk factors for VTE include prolonged immobility, obesity, varicose veins, and congestive heart failure. Adapted with permission from Clagett GP, Anderson FA Jr, Geerts W, et al. Prevention of venous thromboembolism. Chest 1998;114(5 Suppl): 531S 60S. Table 4. Components of an Evaluation for Suspected Hypercoagulable States APC resistance screen (if positive, confirm with DNA test for Factor V Leiden Arg 506 Gln) DNA testing for prothrombin G20210A mutation Assay for antithrombin activity Assay for protein C activity Assay for protein S activity +/- immunological free protein S Clot-based assay for lupus anticoagulant ELISA for antiphospholipid antibodies Fasting total plasma homocysteine level APC = activated protein C; ELISA = enzyme-linked immunosorbent assay. extension (with the inherent risk of PE), UFH was started immediately as a bolus dose followed by a continuous infusion. The response to therapy was monitored by the aptt and dose adjustments were planned to quickly get the aptt into the therapeutic range. Warfarin was also initiated on day 1, without a loading dose, with the aim of getting the PT into therapeutic range by day 5. This approach allowed heparin to be discontinued by day 5 or 6, decreasing the risk of HIT. The introduction of LMWH, which obviates the need for monitoring the aptt, makes outpatient treatment of DVT possible. As in UFH therapy, warfarin is started on day 1 with the intent of discontinuing LMWH by day 5 or 6. The availability of LMWH does not mean that all patients with DVT should be treated as outpatients. An outpatient approach requires that the patient be able to ambulate comfortably and that facilities are in place to monitor outpatient warfarin therapy. Enoxaparin is the 8 Hospital Physician Board Review Manual

9 Table 5. Effects of Acute Thrombosis and Anticoagulant Therapy on Testing for Hypercoagulable States Measurement Acute Clot Heparin Warfarin Protein C Level may be low Test not affected Level will be low Protein S Level may be low Test not affected Level will be low Antithrombin Level may be low Level will be low Test not affected Factor V Leiden Test not affected Test not affected Test not affected Lupus anticoagulant Test may be affected False positive may occur Test not affected with high heparin levels Anticardiolipin antibody Test not affected Test not affected Test not affected only LMWH that is FDA-approved for the treatment of PE. However, only inpatient management of PE has been approved by the FDA. The main issue in the management of DVT and PE concerns the duration of warfarin therapy. Older recommendations, such as 3 months for DVT, 6 months for PE, had little empirical validation, and recent clinical studies 12,13 have led to the following recommendations, which stratify patients with respect to their risk factors for DVT/PE. 14 For patients experiencing their first episode of VTE in association with a reversible risk factor (eg, immobilization, trauma, surgery, or pharmacologic estrogen use), warfarin treatment for 3 to 6 months is recommended. Similarly, it is recommended that patients who are heterozygous for the factor V Leiden mutation (which confers resistance to activated protein C [APC]) experiencing a first episode of VTE be treated for 3 to 6 months. When there is no obvious etiology for a first event, 6 months of warfarin therapy is recommended. For patients with recurrent disease, it is recommended that warfarin therapy be continued at least 12 months, and possibly as lifetime therapy. The same recommendations apply for patients with cancer (unless a complete remission is achieved), homozygous APC resistance, antiphospholipid antibody syndrome (unless it resolves), and deficiencies of AT, protein C, or protein S. If chronic warfarin therapy is contraindicated (eg, by pregnancy), chronic heparin therapy may be substituted for warfarin. OTHER TREATMENT OPTIONS Thrombolytic therapy is generally reserved for patients with massive PE with hemodynamic instability, or for patients with extensive iliofemoral thrombosis. With the appropriate use of heparin and warfarin, the acute mortality directly due to PE is lower than 5%. As a result, it has been impossible to demonstrate a significant improvement in mortality of PE following the use of thrombolytic therapy. Patients must be evaluated carefully for the risk of bleeding prior to the use of these agents. Inferior vena cava (IVC) filters have a very limited use in the management of DVT/PE. In a large-scale randomized trial, survival following a first episode of DVT/PE was not improved by use of a filter. 15 The decrease in recurrent PE was offset by an increase in recurrent DVT. Because of this, IVC filters are part of the routine care of patients with DVT/PE only when there is a contraindication to anticoagulation in a patient with high risk for proximal venous thrombosis or PE. IVC filters may also be considered when recurrent VTE occurs despite adequate anticoagulation, although alternating medical therapy, such as changing from chronic warfarin to chronic heparin may also be considered in this situation. Finally, IVC filters are a reasonable consideration when it is felt that the next PE is likely to be fatal, for example, in patients with recurrent embolism and pulmonary hypertension or in patients who have previously required pulmonary embolectomy. SCREENING FOR HYPERCOAGULABLE STATES Deciding whom to screen for a possible genetic or acquired thrombophilic state is somewhat controversial. There is very little to be gained in screening patients who have experienced only arterial clots because an evaluation in this context rarely provides useful information regarding the presence of a hypercoagulable state. The patients with DVT/PE who are most likely to have a diagnosable hypercoagulable state are Hematology Volume 1, Part 1 9

10 patients who are younger than 50 years, those who have recurrent thrombosis, and those with a positive family history of clots before age 50 years. Identifying a genetic abnormality may influence duration or intensity of anticoagulant therapy. It may also prove advantageous in that family members at risk may be screened so that preventive measures can be employed (eg, avoiding oral contraceptives). Asymptomatic family members with a genetic defect predisposing them to hypercoagulation should be educated regarding the symptoms of thromboembolism and the need for immediate medical attention should symptoms occur. Additionally, physicians, as well as affected family members, should understand that additional prophylaxis may be indicated if such persons undergo elective surgical procedures or are placed in other highrisk situations (eg, prolonged bed rest). Testing for hypercoagulable states should include the most common abnormalities, such as resistance to APC (factor V Leiden), the prothrombin G20210A mutation, and antithrombin deficiency (Table 4). The first 2 can be detected with DNA testing. APC resistance screens are sensitive, but should be confirmed with DNA testing if positive. A fasting homocysteine level and tests for antiphospholipid antibodies and lupus anticoagulant are important because a positive result from these tests dictates a deviation in therapy from the standard recommendations. Hyperhomocystinemia typically responds to folate therapy; the presence of antiphospholipid antibodies or lupus anticoagulants may demand higher levels of anticoagulation and longer duration of therapy (for as long as the antibody remains present). Another important caveat in hypercoagulable state testing is the significant possibility of false negative and false positive values of certain tests if the tests are obtained while the patient is in the acute stages of thrombosis or on anticoagulation therapy (Table 5). Even if a diagnosis of a genetic or acquired defect is made, no change in acute management of VTE is needed; therefore, it is reasonable to wait several weeks before obtaining tests to evaluate for a hypercoagulable state. Warfarin affects protein C and protein S measurements and should be discontinued at least 2 weeks prior to performing a protein C or protein S assay. Heparin does not affect protein C or protein S measurements, but may affect other hypercoagulation assays. Tests for other congenital hypercoagulable states exist, but those disorders are quite rare; therefore, these tests yield little in a general evaluation. BOARD REVIEW QUESTIONS Choose the single best answer for each question. 1. A 63-year-old man with chronic lung disease is admitted to the hospital with chest pain and found to have unstable angina. Coronary arteriography is performed and critical narrowing of the left main coronary artery is found. The patient is placed on heparin and cardiac surgery is planned. The admission chest radiograph shows pneumonia and the patients is treated with a cephalosporin antibiotic. Heparin is continued. On hospital day 8, the platelet count, which was /mm 3 at admission, is noted to have fallen to /mm 3. The most appropriate action at this point in his treatment is: A) Stop heparin administration and continue to observe B) Stop heparin administration and institute warfarin C) Stop heparin administration and start hirudin administration D) Continue heparin administration but stop the cephalosporin antibiotic E) Continue heparin and transfuse platelets 2. A 62-year-old woman has recently started on warfarin as therapy for recurrent venous thromboses. She has no concurrent medical problems. She is seen on an urgent basis because of increased bruising. Her INR is 7.0. Proper management includes which one of the following options? A) Omit 1 or 2 doses of warfarin and resume therapy at a lower dose when the INR is between 2.0 and 3.0 B) Give vitamin K 10 mg IV slowly over a 10-minute period C) Give fresh frozen plasma D) Continue warfarin but add vitamin K 1 mg SC E) Discontinue warfarin; when INR is normal, start low-molecular-weight heparin 10 Hospital Physician Board Review Manual

11 3. A 30-year-old woman in good health is admitted to the hospital for biopsy of an axillary lymph node. She will be discharged after the procedure. Which one of the following options is appropriate anticoagulation for this patient? A) Low-dose unfractionated heparin, 5000 U 2 hours preoperatively and every 12 hours thereafter for 1 week B) Low-molecular-weight heparin, 30 mg SC every 12 hours for 5 days, starting 12 hours postoperatively C) Warfarin, at a dose designed to keep the INR between 1.5 and 2.0 for 1 week. D) No anticoagulation ANSWERS 1. C 2. A 3. D EXPLANATIONS 1. (C) Although the etiology of the thrombocytopenia is unclear in this patient, heparin-induced thrombocytopenia (HIT) must be considered in a patient who experiences a fall in platelet count to a level less than /mm 3 or less than one-half the baseline value. Because of the high morbidity of this condition in terms of limb loss, there is little margin for error. Appropriate therapy consists of stopping the heparin and treating the patient with another drug that blocks the activity of thrombin. Thus, option C, stopping the heparin and starting hirudin, is correct. The patient has unstable angina, an indication for anticoagulant therapy, and HIT places the patient at risk for thrombosis. Therefore, option A, stopping the heparin, is inadequate. Using warfarin to treat HIT has been associated with an increased risk of venous limb gangrene. Therefore, although option B reflects an action that was common clinical practice in the early 1990s, it also is incorrect. Although the cephalosporin might be the cause of the problem, continuing heparin in the face of thrombocytopenia is reckless; option D is therefore incorrect. A platelet count of /mm 3 is not an indication for platelet transfusion, and continuing the heparin in this context can lead to disaster; option E is therefore incorrect, as well. 2. (A) An elevated INR is a common clinical problem in patients on anticoagulant therapy. Two questions need to be considered. First, how critical is the need to reverse the over-anticoagulation? Second, is there a need to maintain anticoagulation? This patient is not bleeding, has no medical problems placing her at increased risk of bleeding, and the INR is under 9.0. Slowly letting the patient return to the therapeutic range is reasonable. Because the patient clearly needs chronic anticoagulation, a therapy that would compromise chronic anticoagulation should not be given. Option A, omitting 1 or 2 doses of warfarin and then resuming therapy, is the simplest and best way to achieve all of the goals. Option B, giving vitamin K at a dose of 10 mg, would correct the INR to normal but would make it very difficult to get the patient back into the therapeutic range. This could be done if a patient with an elevated INR no longer needed to be on anticoagulation, but it would be wrong in this case. Option C, giving fresh frozen plasma, is indicated only in the face of critical bleeding when rapid control of the INR is needed. Not only is it the wrong answer in this case, one must remember that it affects the INR only for 6 hours, the half-life of most clotting factors. Option D, continuing warfarin and giving vitamin K at a dose of 1 mg SC, is unlikely to be effective. If vitamin K is given for an elevated INR, the dose of 1 mg will move the INR toward normal and allow the patient to resume anticoagulation. However, if this is done, the warfarin should be held until the INR is in the therapeutic range. Whereas therapeutic failure is a reason to discontinue warfarin, an INR outside the range is an indication for dose adjustment, not for changing anticoagulant agents; option E is therefore incorrect. 3. (D) Appropriate perioperative anticoagulation depends on the risk of the procedure. This patient is younger than 40 years and the procedure is neither abdominal surgery nor orthopaedic surgery. Despite the complex series of recommendations for patients undergoing more complex procedures, one should remember that simple procedures in young, healthy patients do not necessitate the risk of anticoagulation. The correct answer is therefore option D, no anticoagulation. Hematology Volume 1, Part 1 11

12 REFERENCES 1. Fifth American College of Chest Physicians Consensus Conference on Antithrombotic Therapy. Chest 1998;114 (5 suppl):439s 769S. 2. Bussey HI, Force RW, Bianco TM, et al. Reliance on prothrombin time ratio causes significant errors in anticoagulation therapy. Arch Intern Med 1992;152: Poller L. A simple nomogram for the derivation of international normalised ratios for the standardisation of prothrombin times. Thromb Hemost 1988;60: Hirsh J, Dalen JE, Anderson DR, et al. Oral anticoagulants: mechanism of action, clinical effectiveness, and optimal therapeutic range. Chest 1998;114(5 suppl): 445S 69S. 5. Faraci PA, Deterling RA, Stein AM, et al. Warfarin induced necrosis of the skin. Surg Gynecol Obstet 1978;146: Hirsh J, Levine MN. Low molecular weight heparin. Blood 1992;79: Raschke RA, Reilly BM, Guidry JR, et al. The weight based heparin dosing nomogram compared with a standard care nomogram. A randomized control study. Ann Intern Med 1993;119: Warkentin TE, Kelton JG. Heparin-induced thrombocytopenia. Ann Rev Med 1989;40: Warkentin TE. Heparin-induced thrombocytopenia: IgG-mediated platelet activation, platelet microparticle generation, and altered procoagulant/anticoagulant balance in the pathogenesis of thrombosis and venous limb gangrene complicating heparin-induced thrombocytopenia. Transfus Med Rev 1996;10: Warkentin TE, Elavathil LJ, Hayward CP, et al. The pathogenesis of venous limb gangrene associated with heparininduced thrombocytopenia. Ann Intern Med 1997;127: Clagett GP, Anderson FA Jr, Geerts W, et al. Prevention of venous thromboembolism. Chest 1998;114(5 suppl): 531S 60S. 12. Schulman S, Rhedin AS, Lindmaker P, et al. A comparison of six weeks with six months of oral anticoagulant therapy after a first episode of venous thromboembolism. Duration of Anticoagulation Trial Study Group. N Engl J Med 1995;332: 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: Hyers TM, Agnelli G, Hull RD, et al. Antithrombotic therapy for venous thromboembolic disease. Chest 1998; 114(5 suppl):561s 78S. 15. Decousus H, Leizorovicz A, Parent F, 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 Pulmonaie par Interruption Cave Study Group. N Engl J Med 1998;338: Copyright 2000 by Turner White Communications Inc., Wayne, PA. All rights reserved. 12 Hospital Physician Board Review Manual