Pharmacotherapy A Pathophysiologic Approach, 9th Ed.

9. Venous Thromboembolism

Daniel M. Witt and Nathan P. Clark


 Images The risk of venous thromboembolism (VTE) is related to several easily identifiable factors, including age, major surgery (particularly orthopedic procedures of the lower extremities), previous VTE, trauma, malignancy, prolonged immobility or limb paralysis, and hypercoagulable states.

 Images The diagnosis of VTE should be confirmed by objective testing.

 Images At the time of hospital admission, all patients should receive prophylaxis against VTE that corresponds to their level of risk. Prophylaxis should be continued throughout the period of risk.

 Images In the absence of contraindications, the treatment of VTE should initially include a rapid-acting anticoagulant (e.g., unfractionated heparin, low-molecular-weight heparin, fondaparinux, or rivaroxaban). If the patient is transitioned to warfarin therapy, the rapid-acting anticoagulant should be overlapped with warfarin for at least 5 days and until the patient’s international normalized ratio is greater than 2.

 Images Most patients with an uncomplicated deep vein thrombosis (DVT), with or without pulmonary embolism (PE), can be safely treated as an outpatient.

 Images Emerging anticoagulants such as rivaroxaban may mark a significant advancement in the treatment of VTE.

 Images Anticoagulant therapies require meticulous and systematic monitoring as well as ongoing patient education.

 Images Bleeding is the most common adverse effect associated with anticoagulant drugs. A patient’s risk of major hemorrhage is related to the intensity and stability of therapy, concurrent drug use, history of prior bleeding, prior history of stroke, renal or hepatic impairment, thrombocytopenia, recent surgery or trauma, and increasing age.

 Images Anticoagulation therapies for VTE should be continued for a minimum of 3 months. The duration of anticoagulation therapy should be based on the patient’s risks of VTE recurrence and major bleeding, and preference regarding continued treatment.


Venous thromboembolism (VTE) is a potentially fatal disorder and a significant health problem in our aging society.1 Although young, otherwise healthy adults can have VTE, it most frequently occurs in patients who are hospitalized, undergo major surgery, are immobile for a lengthy period of time, or have a hypercoagulable disorder. VTE is manifested as deep vein thrombosis (DVT) and/or pulmonary embolism (PE) and results from clot formation within the venous circulation, most commonly the legs, but clots can also form in the arms, and in the mesenteric and cerebral veins (Fig. 9-1).1 DVT is rarely fatal, but death from PE can occur within minutes after symptom onset, before effective treatment can be given. Beyond the symptoms produced by the acute event, the complications of VTE, such as the postthrombotic syndrome and chronic thromboembolic pulmonary hypertension (CTPH), also cause substantial disability and suffering.1


FIGURE 9-1 Venous circulation.

Anticoagulant drugs used in the prevention and treatment of VTE require precise dosing and meticulous monitoring.2,3 Systematic approaches to anticoagulant therapy management substantially reduce the risks, but bleeding remains an all too common and serious complication of administering anticoagulant drugs. Consequently, preventing VTE in at-risk patients is paramount to improving outcomes.46 When there is a suspicion of VTE, the rapid and accurate diagnosis of the disorder is critical to making appropriate treatment decisions.7 The optimal use of anticoagulant drugs requires an in-depth knowledge of their pharmacology and pharmacokinetic properties, and a comprehensive approach to patient management.2,3


The true incidence of VTE in the general population is unknown because a substantial portion of patients have clinically silent disease. The annual incidence of symptomatic objectively diagnosed VTE is estimated at 2 to 3 per 1,000, increasing with age from 0.1 per 1,000 in adolescence to 8 per 1,000 in those over 80 years.8 The incidence of VTE in specific high-risk patient populations has been extensively studied.46 Patients sustaining multiple traumas or having orthopedic procedures involving the lower extremities are at particularly high risk as are those with a prior history of VTE and/or who have metastatic cancer. Likewise, the incidence of VTE after a myocardial infarction, stroke, and spinal cord injury is high. Several disorders of hypercoagulability have also been linked to a high lifetime incidence of VTE.46


Images A number of factors increase the risk of developing VTE (Table 9-1). The majority of these risk factors can be easily identified in clinical practice. Stasis of blood favors clotting in part through reduced clearance of activated clotting factors from sites of clot formation.9 The rate of blood flow in the venous circulation, particularly in the deep veins of the lower extremities, is relatively slow. Contraction of the calf and thigh muscles works with valves in the deep veins of the leg to facilitate the flow of blood back to the heart and lungs; thus, damage to the venous valves and periods of prolonged immobility result in venous stasis.8 Vessel obstruction, from either external compression or a thrombus, also promotes clot propagation. Reduced venous blood flow explains, at least in part, why numerous medical conditions and surgical procedures are associated with an increased risk of VTE (Table 9-1). Increased blood viscosity, seen in myeloproliferative disorders such as polycythemia vera, may also contribute to thrombus formation.8

TABLE 9-1 Risk Factors for Venous Thromboembolism


A growing list of inherited and acquired diseases has been linked to hypercoagulability (see Table 9-1). Activated protein C (aPC) resistance is the most common genetic disorder of hypercoagulability (prevalence rate of 2% to 7% in whites) accounting for many cases of DVT in unselected patients.10 Most aPC resistance results from a mutation on factor V that renders it resistant to degradation by aPC.10This mutation is known as factor V Leiden, named after the city of Leiden, Holland, where the defect was first described. The prothrombin G20210A mutation is the second most frequent genetic risk factor, occurring in about 2% to 4% of whites and imparting about a threefold increased risk of VTE.10 This mutation increases circulating prothrombin, but the degree of VTE risk does not rise proportionally. Enhanced thrombin generation has been observed, but the mechanism whereby this disorder increases VTE risk is not completely understood.11 Although the rarity (present in <1% of the population) of inherited deficiencies of the natural anticoagulants protein C, protein S, and antithrombin precludes accurate quantification of their effect on the risk of initial VTE, many experts believe the lifetime risk is high.12 Conversely, excessively high concentrations of factors VIII, IX, and XI or fibrinogen also increase the risk of VTE.13 Given the prevalence of these inherited abnormalities in the general population, some patients have multiple genetic defects that have additive effects in terms of increasing the lifetime thrombotic risk.13

Acquired disorders of hypercoagulability may result from malignancy, the presence of antiphospholipid antibodies, and estrogen use.13 The strong link between cancer and thrombosis has long been recognized.14 Tumor cells secrete a number of procoagulant substances that activate the coagulation cascade. Furthermore, patients with cancer often have suppressed levels of endogenous anticoagulants (protein C, protein S, and antithrombin). It has been postulated that cancer cells use thrombotic mechanisms to recruit a blood supply (angiogenesis), metastasize, and create barriers against host defense mechanisms.15 Antiphospholipid antibodies are a heterogeneous group of antibodies targeting proteins that bind phospholipids.11 These include antibodies that prolong phospholipid-based clotting assays, known as lupus anticoagulants, as well as anticardiolipin and β2-glycoprotein (β2-gp) I antibodies. Antiphospholipid antibodies are found in up to 5% of normal healthy populations but are more common in patients with autoimmune disorders such as systemic lupus erythematosus and inflammatory bowel disease.16 Transient and low titers of these antibodies are common and do not present a risk for thrombosis. A diagnosis of antiphospholipid antibody syndrome requires the presence of moderate- or high-titer antibodies or lupus anticoagulants measured at least 12 weeks apart in a patient with a history of arterial or venous thrombosis or recurrent miscarriage.17 The precise mechanism by which antiphospholipid antibodies provoke thrombosis remains to be definitively determined. Contributing factors include complement activation, inhibition of proteins C and fibrinolysis, platelet activation, and increased tissue factor expression.16 Estrogen-containing contraception, estrogen replacement therapy, and the selective estrogen receptor modulators are all linked to venous thrombosis.18 Estrogens increase serum clotting factor concentrations and induce aPC resistance. Women with underlying hypercoagulability are at particularly high risk of developing venous thrombosis while taking estrogens.18

In many cases, VTE is the result of converging combinations of inherited and acquired thrombotic risk factors. Thus, many individuals with congenital hypercoagulable conditions experience a first VTE only after being placed in situations of high risk for thrombosis such as orthopedic surgery, immobilization, the use of estrogen-containing oral contraceptives, or pregnancy.19


The arrest of bleeding following vascular injury is an amazingly complex process that is essential to life (Fig. 9-2). In the late 1800s, Dr. Rudolf Virchow, a German pathologist, recognized the role played by blood vessels, circulating elements in the blood, and the speed of blood flow in the regulation of clot formation.20 As described previously, alterations in any one of these elements, known today as Virchow’s triad, may lead to pathologic clot formation. Through the years various descriptive models of blood clot formation have been proposed and continuously refined based on new information and fresh thinking derived from increasingly sophisticated experimental methodology. Still many questions remain and a complete understanding of in vivo thrombus formation remains incomplete.21


FIGURE 9-2 Overview of hemostasis.

Hemostasis is the complex process responsible for maintaining the integrity of the pressurized circulatory system following damage to blood vessels.22 Hemostatic clots remain localized to the vessel wall and do not greatly impair blood flow in the vessel. In contrast, thrombotic clots such as those causing VTE result in impairment of blood flow and even complete occlusion of the vessel. Under normal circumstances, endothelial cells forming the intima of vessels maintain blood flow by physically separating extravascular collagen and tissue factor from platelets and clotting factors (namely, activated factor VII [VIIa]), respectively, thus preventing activation of hemostasis. Vascular injury allows key components of the coagulation process to seal the breach through the interaction of activated platelets recruited to the site of injury and the clotting factor cascade initiated by tissue factor and culminating in the formation of thrombin and ultimately fibrin clot (see Fig. 9-2).22 In contrast to physiologic hemostasis, pathologic venous thrombosis often occurs in the absence of gross vein wall disruption and may be triggered by circulating microparticles bearing tissue factor rather than the tissue factor expressed in the vessel wall. Venous clots often occur in areas of disturbed blood flow (e.g., valve cusps in the deep veins of the legs) and are mainly composed of fibrin with platelets and trapped red blood cells.21 Platelet thrombus and fibrin clot formation overlap temporally and occur nearly simultaneously,21 but are discussed separately below for simplicity.

Platelet Pathway

Platelets become actively involved in thrombus formation via two distinct pathways acting in parallel or separately. In one pathway, exposure of circulating blood to subendothelial collagen following vascular injury initiates platelet activation; in the other pathway platelets are activated by thrombin generated by tissue factor derived from the vessel wall or present in flowing blood.22 Various adhesion proteins (e.g., von Willebrand factor, fibrinogen) appear to play critical roles in platelet–vessel wall interactions.21 A platelet thrombus develops as activated platelets recruit unstimulated platelets, some of which are also activated while others remain loosely associated but do not undergo activation and may ultimately break away from the growing thrombus.22 Activation causes platelet α and dense granules to release their cargo of components such as adenosine diphosphate (ADP) and calcium ions that are critical for sustaining further thrombus formation into the environment surrounding the developing clot.22 Activated platelets accumulating in the thrombus also express P-selectin, an adhesion molecule that facilitates capture of blood-borne microparticles bearing tissue factor. Accumulation of tissue factor in the platelet thrombus is important as tissue factor is the primary initiator of fibrin clot formation via the coagulation cascade (Fig. 9-3).22


FIGURE 9-3 Model of pathologic thrombus formation: (A) activated platelets adhere to vascular endothelium; (B) activated platelets express P-selectin; (C) pathologic microparticles express active tissue factor and are present at a high concentration in the circulation—these microparticles accumulate, perhaps by binding to activated platelets expressing P-selectin; (D) tissue factor can lead to thrombin generation, and thrombin generation leads to platelet thrombus formation and fibrin generation. (Adapted from Furie and Furie.22)

Tissue Factor Pathway

The tissue factor pathway triggers coagulation by generating a small (picomolar) amount of thrombin. Before this small amount of thrombin is generated, the tissue factor pathway proceeds inefficiently through factor IX or X because factors VIII and V, the circulating procofactors required in the tenase and prothrombinase complexes, are not yet available in their most active cofactor form (Fig. 9-4). However, once formed, this small amount of thrombin converts factors VIII and V to their active cofactor forms, factors VIIIa and Va, respectively.22 The tenase and prothrombinase complexes now efficiently proceed to generate a large burst of thrombin. Even though the tissue factor pathway is rapidly downregulated, or inhibited, by the action of tissue factor pathway inhibitor, thrombin generation proceeds without replenishing active tissue factor.22


FIGURE 9-4 Coagulation cascade: (A) During the initiation phase the tissue factor–factor VIIa complex triggers blood coagulation by generating small amounts of thrombin. The tissue factor–factor VIIa complex activates factors IX and X. Factor IXa binds to factor VIII to form a tenase complex that inefficiently activates factor X to form factor Xa. Factor Xa, generated by the tissue factor–factor VIIa complex or the inefficient tenase complex, binds factor V on membrane surfaces to form a prothrombinase complex that inefficiently converts prothrombin to thrombin. The initiation phase is quickly downregulated by tissue factor pathway inhibitor. (B) During the amplification phase, the small amount of thrombin generated during initiation activates factors VIII and V, leading to a large burst of thrombin via highly efficient tenase and prothrombinase complexes. Thrombin converts fibrinogen to fibrin that forms strands that ultimately create an intricate web that traps red blood cells, platelets, and other cells to form a fibrin clot. Thrombin also activates factor XIII that cross-links fibrin to form a stabilized clot. (Adapted from Furie and Furie.22)

Fibrin Thrombus Formation

The final step in thrombus formation is the thrombin-mediated conversion of fibrinogen to form fibrin monomers. As fibrin monomers reach a critical concentration, they begin to precipitate and polymerize to form fibrin strands. Factor XIIIa, which is also activated by the action of thrombin, covalently bonds these strands to one another (Fig. 9-4) to form an extensive meshwork that surrounds and encases the aggregated platelet thrombus and red blood cells to form a stabilized clot that seals the site of vascular injury and prevents blood loss.23 Coagulation reactions are eventually terminated when this expanding meshwork of platelets and fibrin “paves over” the initiation site and additional activated factors are unable to diffuse through the overlying layer of clot.24

Endogenous Control of Thrombus Formation

Normally, a number of tempering mechanisms control coagulation (see Fig. 9-2).24 Without effective self-regulation, the coagulation cascade would continue unabated causing vascular occlusion at the site of injury. The intact endothelium adjacent to the damaged tissue actively secretes several antithrombotic substances.9 As its name implies, thrombomodulin modulates thrombin activity by converting protein C to its active form (aPC). When joined with its cofactor protein S, aPC enzymatically inactivates factors Va and VIIIa regulating the functionality of the prothrombinase and tenase complexes, respectively.9 The physiologic role of aPC is to prevent coagulation reactions from spreading to healthy, uninjured vessel walls. Antithrombin is a circulating protein that inhibits thrombin and factor Xa. Heparan sulfate, a heparin-like compound secreted by endothelial cells, exponentially accelerates antithrombin activity.9 By a similar mechanism, heparin cofactor II also inhibits thrombin. As described previously, tissue factor pathway inhibitor plays an important role by regulating the initiation of the coagulation cascade.22 When these self-regulatory mechanisms are intact, the formation of fibrin clot is limited to the zone of tissue injury. However, disruptions in the system, previously described hypercoagulable states, often result in pathologic thrombosis.19


The fibrinolytic system is responsible for the dissolution of formed blood clots.25 An inactive proenzyme, plasminogen, is converted to an active enzyme, plasmin, that degrades the fibrin mesh into soluble end products collectively known as fibrin degradation products including D-dimer.25 The fibrinolytic system is also under the control of a series of stimulatory and inhibitory substances (see Fig. 9-2). Tissue plasminogen activator and urokinase plasminogen activator convert plasminogen to plasmin. Plasminogen activator inhibitor-1 inhibits the plasminogen activators, and α2-antiplasmin inhibits plasmin activity. Impaired activation of the fibrinolytic system has also been linked to hypercoagulability and thrombotic complications.25

Although a thrombus can form in any part of the venous circulation, the majority begin in the leg(s). Once formed, a venous thrombus may (a) remain asymptomatic, (b) spontaneously lyse, (c) obstruct the venous circulation, (d) propagate into more proximal veins, (e) embolize to the lungs resulting in PE, or (f) act in any combination of these ways.26 Thrombus isolated in veins of the calf is unlikely to embolize, but thrombus involving the popliteal and larger veins above it can break loose (embolize) and travel through the right side of the heart and lodge in the pulmonary artery or one of its branches, occluding blood flow to that part of the lung and impairing gas exchange. Without treatment, the affected portion of the lung becomes necrotic and oxygen delivery to other vital organs decreases, potentially resulting in fatal circulatory collapse.27


Clinical Presentation

Images The symptoms of DVT or PE (Tables 9-2 and 9-3) are nonspecific and objective tests are required to confirm or exclude the diagnosis.28 Patients with DVT frequently present with unilateral leg pain and swelling. Postthrombotic syndrome, a long-term complication of DVT caused by damage to the venous valves, may also result in chronic lower extremity swelling, pain, tenderness, skin discoloration, and, in the most severe cases, ulceration.7PE typically presents with chest pain, shortness of breath, tachypnea, and tachycardia. PE is a life-threatening condition that may result in cardiopulmonary collapse.29

TABLE 9-2 Clinical Presentation of Deep Vein Thrombosis


TABLE 9-3 Clinical Presentation of Pulmonary Embolism


Given that VTE can be debilitating or fatal, it is important to treat it quickly and aggressively.7 Conversely, because major bleeding induced by antithrombotic drugs can be equally harmful, it is important to avoid treatment when the diagnosis is not a reasonable certainty.28 Assessment of the patient’s status should focus on the search for risk factors in the patient’s medical history (see Table 9-1). Even in the presence of mild, seemingly inconsequential symptoms, VTE should be strongly suspected in those with multiple risk factors.

Diagnostic Studies

Radiographic contrast studies (venography and pulmonary angiography) are considered the gold standard in clinical trials because they are the most accurate and reliable method for diagnosing VTE.28However, contrast studies are also expensive invasive procedures technically difficult to perform and evaluate. Severely ill patients are often unable to tolerate the procedure, and many develop hypotension and cardiac arrhythmias.28 Further, the contrast medium is nephrotoxic and irritating to the vessel wall and may paradoxically precipitate VTE.28 For these reasons, less invasive tests, such as compression ultrasound (CUS), computed tomography (CT) scans, and the ventilation–perfusion (V/Q) scan, are used in clinical practice for the initial evaluation of patients with suspected VTE.


D-dimer is a degradation product of fibrin clot and levels of D-dimer are usually significantly elevated in patients with acute thrombosis.28 Although D-dimer is a very sensitive marker of clot formation, it is not sufficiently specific. A variety of conditions are associated with D-dimer elevations, including recent surgery or trauma, pregnancy, increasing age, and cancer; therefore, a positive D-dimer test is not conclusive evidence of VTE diagnosis. A wide variety of D-dimer assays of varying sensitivities for VTE are available. Enzyme-linked immunosorbent assays (ELISAs) and enzyme-linked immunofluorescence assays, along with the latex immunoturbidimetric assays, are generally termed “highly sensitive,” whereas the whole blood D-dimer assay is considered “moderately sensitive.”28Appropriate use of D-dimer should include initial risk stratification via a validated clinical assessment model.28

Risk Stratification

Clinical assessment significantly improves the diagnostic accuracy of noninvasive tests such as CUS, CT scanning, and D-dimer. Simple assessment checklists can be used to determine if a patient has a high, moderate, or low probability of DVT or PE (Table 9-4).30 Patients with a high pretest probability of VTE have a greater than 50% chance of having VTE, compared with only 5% of those with low pretest probability.31 Patients with a low pretest probability of VTE should receive testing with a moderate or highly sensitive D-dimer, if available.28 If the D-dimer result is normal, VTE is ruled out. Patients with a moderate pretest probability should receive either highly sensitive D-dimer or CUS. CUS may be performed proximally (i.e., above the knee) or throughout the entire leg. A normal full leg ultrasound or highly sensitive D-dimer rules out VTE. A normal proximal ultrasound should be repeated in 1 week or paired with D-dimer testing to rule out VTE. All patients with high pretest probability of VTE should receive either proximal or full leg CUS. A normal full leg ultrasound rules out VTE, whereas a normal proximal ultrasound requires additional testing with highly sensitive D-dimer, full leg ultrasound, or repeat proximal ultrasound surveillance in 1 week.28 Patients with CUS indicating proximal DVT should receive treatment, regardless of pretest probability. Evidence of calf vein DVT after full leg ultrasound may be managed by anticoagulation or further ultrasound surveillance to assess for propagation into the proximal deep veins of the leg.7,28

TABLE 9-4 Clinical Assessment Models for Deep Vein Thrombosis and Pulmonary Embolism



General Approach to the Prevention of Venous Thromboembolism

Given that VTE is potentially fatal and costly to treat, strategies to prevent DVT in at-risk populations positively impact patient outcomes.32 Relying on the early diagnosis and treatment of VTE is unacceptable because some patients will die before treatment can be initiated. Effective prophylaxis can reduce the risk of fatal PE in high-risk surgical and medical populations, whereas early ambulation is often sufficient for those at low risk of VTE.33 Educational programs and clinical decision support systems have been shown to improve the appropriate use of VTE prevention methods.34

Images The most authoritative and well-recognized evidence-based guidelines for the prevention and treatment of VTE are the Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: Evidence-Based Clinical Practice Guidelines published by the American College of Chest Physicians (AT9). The AT9 attempt to balance concerns regarding the competing risks of symptomatic VTE and bleeding in their recommendations for thromboprophylaxis.33Compared with previous guideline editions, the AT9 have critically evaluated the relevance of surrogate end points, such as venographically detected asymptomatic DVT, and based current recommendations mainly on the risk of symptomatic VTE (Table 9-5).33 An effective VTE prophylaxis program should identify and determine each patient’s level of risk for VTE and bleeding, and select and implement regimens that optimally balance these risks in a cost-effective manner. Several pharmacologic and mechanical methods are effective for preventing VTE, and these can be used alone or in combination.46 Mechanical methods improve venous blood flow, whereas drug therapy inhibits clotting factor activity or production. Despite ongoing efforts to minimize hospital-acquired VTE, up to one third of hospitalized patients at high risk of VTE without contraindications to anticoagulant therapy still do not receive appropriate prophylaxis.35

TABLE 9-5 Guidelines for the Prevention of Venous Thromboembolism


Clinical Controversy…

In the United States interest in VTE prevention peaked after the 2008 decision by the Centers for Medicare and Medicaid Services to withhold a portion of reimbursement to hospitals for patients in whom VTE complicated hip or knee replacement surgery. These so-called “never events” proved controversial because VTE occurs even among carefully selected populations receiving adequate prophylaxis in clinical trials, and many orthopedic surgeons believe overzealous use of anticoagulants increases the risk for postoperative bleeding complications.36

General Approach to the Treatment of Venous Thromboembolism

Images Anticoagulation therapies remain the mainstay of treatment for VTE. DVT and PE are manifestations of the same disease process and are treated similarly.7 Before prescribing a full course of anticoagulation therapy for the treatment of VTE, it is imperative to establish an accurate diagnosis, thus preventing unnecessary risk of bleeding and expense to the patient (Fig. 9-5).7 Patients with high probability of VTE may need parenteral anticoagulation therapy while awaiting the results of diagnostic testing, whereas patients with intermediate probability may need parenteral anticoagulation only if diagnostic testing will be delayed more than 4 hours.7 The acute phase of VTE treatment (–7 days) requires rapidly acting anticoagulants such as unfractionated heparin (UFH), low-molecular-weight heparin (LMWH), fondaparinux, or rivaroxaban to prevent thrombus extension and embolization. The early maintenance phase (7 days to 3 months) consists of continued therapeutic anticoagulation aimed at reducing the risk of long-term sequelae such as the postthrombotic syndrome and CTPH by allowing formed clot to be slowly dissolved by endogenous thrombolytic processes. Anticoagulation therapy extending beyond 3 months is aimed at long-term secondary prevention of recurrent VTE.1,7 Anticoagulation therapy is usually initiated with an injectable anticoagulant and then transitioned to warfarin maintenance therapy (Table 9-6 and Fig. 9-5). A simpler single-drug approach using orally administered rivaroxaban has now been shown to be noninferior to this standard approach for VTE treatment for appropriately selected patients.37,38


FIGURE 9-5 Treatment of venous thromboembolism (VTE). (DVT, deep vein thrombosis; LMWH, low-molecular-weight heparin; PE, pulmonary embolism; SBP, systolic blood pressure; UFH, unfractionated heparin.)

TABLE 9-6 Guidelines for the Treatment of Venous Thromboembolism


Determining the optimal duration of anticoagulation involves weighing the risk of recurrent VTE against the risk of bleeding associated with anticoagulation therapy and determining patient preference regarding treatment duration (see Evaluation of Therapeutic Outcomes below). When bleeding risk outweighs VTE recurrence risk or if the informed patient’s preference is to stop treatment, anticoagulation therapy should be discontinued.7 In life- or limb-threatening circumstances, elimination of the obstructing thrombus may be warranted and the use of thrombolysis or thrombectomy can be considered.7Insertion of a removable filter in the inferior vena cava (IVC) is also an option in those with contraindications to anticoagulation therapy or in whom anticoagulant therapy has failed.7

Images Once the diagnosis of VTE has been objectively confirmed (see Clinical Presentation and Diagnosis above), anticoagulant therapy with UFH, LMWH, fondaparinux, or rivaroxaban should be instituted as soon as possible. Evidence indicates that currently available injectable anticoagulants can be administered in the outpatient setting in most patients with DVT and in carefully selected hemodynamically stable patients with PE (see Table 9-6).1 The decision to initiate therapy on an outpatient basis should be based on institutional resources and patient-specific variables (Table 9-7).39

TABLE 9-7 Outpatient Treatment Protocol for Deep Venous Thrombosis



Mechanical Prevention and Treatment Strategies

Graduated compression stockings and intermittent pneumatic compression (IPC) devices improve venous blood flow and reduce the risk of VTE. These methods do not increase the risk of bleeding, which makes them attractive for postoperative VTE prophylaxis. However, they are not risk free as skin breakdown and ulceration can occur.6

Graduated compression stockings have failed to reliably demonstrate a reduction in VTE in medically ill patients.6 However, they reduce the incidence of VTE (including asymptomatic and distal DVT) by approximately 65% when used after orthopedic surgery, cardiac surgery, gynecologic surgery, or neurosurgery.5 Compression stockings work by increasing the velocity of venous blood flow through the application of graded pressure, with the greatest amount applied at the ankle. This treatment option is inexpensive and safe, and an excellent choice when pharmacologic intervention is either contraindicated or difficult to adequately monitor.6 When combined with pharmacologic interventions, graduated compression stockings have an additive effect. However, they can be uncomfortable and some patients are unable to wear them because of the size or shape of their legs and adverse skin reactions.6

IPC devices utilize sequential inflation of a series of cuffs wrapped around the patient’s legs to increase the velocity of blood flow.6 The cuffs use graded pressure and inflate in 1- to 2-minute continuous cycles from the ankles to the thighs. IPC reduces the risk of VTE by more than 60% following general surgery, neurosurgery, and orthopedic surgery.5 Although IPC is well tolerated and safe to use in patients who have contraindications to pharmacologic therapies, it does have drawbacks. There is some theoretical concern that external compression may dislodge a previously formed clot, IPC is more expensive and cumbersome to use than graduated compression stockings, and some patients have difficulty sleeping and getting in and out of bed while wearing the devices.4 For optimal effectiveness, IPCs should be worn at least 18 hours/day. Battery-operated units facilitate mobility and monitoring. Like graduated compression stockings, IPCs can be used in combination with pharmacologic strategies.4

IVC filters can provide short-term protection against PE in very-high-risk patients by blocking embolization of thrombus formed below the filter.5 Percutaneous insertion of an IVC filter is a minimally invasive procedure performed using fluoroscopic imaging to verify placement. Despite the widespread use of IVC filters, only limited data support effectiveness and long-term safety in the setting of VTE prevention. Nonrandomized data suggest IVC filters reduce the short-term risk of PE but are associated with complications including DVT, filter migration, IVC occlusion, and insertion site thrombosis.5 IVC filters should be reserved for patients at highest risk for VTE in whom other prophylactic strategies cannot be used.

IVC filters also have a limited role in the management of acute VTE when anticoagulants are ineffective or unsafe, including in (a) patients with an absolute contraindication to anticoagulation therapy because of active bleeding or anticipated bleeding from a preexisting lesion; (b) patients with massive PE who survive but in whom recurrent embolism might be fatal; (c) patients who have recurrent VTE despite adequate anticoagulation therapy; and (d) demonstration of large free-floating clot loosely attached to the wall of the IVC.40 IVC filters reduce the risk of PE in the short term, but also appear to increase the long-term risk for recurrent DVT presumably as a consequence of the accumulation of thrombus on the filter, which may partially occlude the vena cava, resulting in venous stasis.40 Retrievable filters that can be removed after the period of greatest risk for PE or after a transient bleeding risk has resolved have been developed and suggested for use in patients with transient contraindications to anticoagulation therapy. In reality most (65% in one report) “retrievable” filters are not removed.40 When an IVC filter is inserted as an alternative to anticoagulation, the AT9 suggest a conventional course of anticoagulant therapy once the risk of bleeding resolves.7

Ancillary Therapy

In addition to anticoagulant therapy for patients with proximal DVT, wearing graduated compression stockings can reduce the risk of developing the postthrombotic syndrome by 50%.7 To be effective, graduated compression stockings must fit properly and provide adequate pressure at the ankle (30 to 40 mm Hg; 4 to 5.3 kPa).7 Discomfort and unflattering appearance are barriers to adherence with compression stockings, particularly during the hot summer months. However, consistent daily use should be encouraged for at least 2 years after DVT or longer if symptoms of postthrombotic syndrome persist.7

Strict bedrest was traditionally recommended following acute DVT based on the assumption that leg movement would dislodge the clot, resulting in PE. However, ambulation in conjunction with graduated compression stockings results in faster reduction in pain and swelling with no apparent increase in the rate of embolization. Early ambulation and continued high activity level also reduce the likelihood of postthrombotic syndrome.7 Patients should be encouraged to ambulate as much as their symptoms permit. If pain and swelling increase with ambulation, the patient should be instructed to lie down and elevate the affected leg until symptoms subside.


Pharmacologic options for preventing VTE have been extensively evaluated in randomized clinical trials (see Table 9-5). Appropriately selected drug therapies can significantly reduce the incidence of VTE following hip and knee replacement, hip fracture repair, general surgery, myocardial infarction, ischemic stroke, and in appropriately selected hospitalized medical patients.46 The optimal agent and dose to use for VTE prevention must be based on the patient’s level of risk for thrombosis and bleeding complications, as well as cost and availability.

Medical Patients

Several risk assessment models have been developed to identify hospitalized and critically ill patients at high risk of VTE who would realize the greatest benefit from thromboprophylaxis. The Padua Prediction Score is a prospectively validated VTE risk assessment tool for hospitalized medical patients.6 For this tool, 3 points each are assigned for active cancer, previous VTE, reduced mobility, and thrombophilia; 2 points are assigned for trauma and/or surgery within the last month; and 1 point each is assigned for age ≥70 years, heart and/or respiratory failure, acute myocardial infarction or ischemic stroke, acute infection and/or rheumatologic disorder, body mass index ≥30 kg/m2, or ongoing hormonal treatment. Among high-risk patients (score ≥4 points) not receiving prophylaxis, VTE occurred in 11% within 90 days compared with just 0.3% of low-risk patients.6

Recommendations for preventing VTE during medical illness are summarized in Table 9-5. Compared with placebo, low-dose unfractionated heparin (LDUH), LMWH, and fondaparinux all reduce the risk of symptomatic VTE and fatal PE among high-risk medical patients.6 Hospitalized and acutely ill medical patients at high risk of VTE and low risk of bleeding should receive pharmacologic prophylaxis with LDUH, LMWH, or fondaparinux for the duration of hospitalization or until fully ambulatory. Routine use of pharmacologic prophylaxis is not warranted in low-VTE-risk populations. In medical patients at high risk of bleeding who also require VTE prophylaxis, mechanical prophylaxis may be preferred over anticoagulation therapy.6 The AT9 guidelines suggest that mechanical prophylaxis may be preferred in patients with any of the following bleeding risk factors: active gastric or duodenal ulcer, history of bleeding within 90 days, or platelet count <50 × 109/L.6 Mechanical prophylaxis should also be considered if more than one of the following are present: Age 85 years or more, hepatic failure, renal failure (creatinine clearance <30 mL/min), admission to intensive care or cardiac care units, central venous catheter, rheumatic disease, active cancer, or male sex.6

Nonorthopedic Surgery Patients

Patient risk for VTE after general surgery can be estimated using the Caprini score.5 This extensive risk assessment tool awards 1, 2, 3, or 5 points to patient-specific risk factors (e.g., age, BMI, VTE history) and procedure-related risk factors including minor or major surgery, laparoscopic or open procedures, and elective arthroplasty. Once all risk factor points are summed, VTE risk is categorized as very low (0 to 1 point), low (2 points), moderate (3 to 4 points), or high (≥5 points).5 Estimating bleeding risk associated with surgery is challenging due to wide variety of surgery types, the effect of surgical technique, and the lack of a validated clinical predication rule. Table 9-5summarizes the AT9 recommendation for preventing VTE following nonorthopedic surgery. In general, patients at high risk of VTE but at low risk of bleeding should receive prophylaxis with LDUH or LMWH in addition to graduated compression stockings or IPC. Patients at high risk of bleeding should receive IPC if the risk of VTE is moderate or high.5

Orthopedic Surgery

Total joint arthroplasty is associated with a very high risk of postoperative VTE. Recommended pharmacologic agents for the prevention of VTE following joint replacement surgery include aspirin, adjusted-dose warfarin, UFH, LMWH, fondaparinux, dabigatran, apixaban, and rivaroxaban for a minimum duration of 10 days postsurgery.4 Head-to-head trials of these agents fail to reliably demonstrate differences in clinically relevant outcomes such as symptomatic VTE, fatal PE, major hemorrhage, and surgical site complications.4 The endorsement of aspirin for VTE prevention in the AT9 is a major departure from previous guidelines that actually recommended against aspirin use in this setting.4,41 The change in recommendation reflects recognition by AT9 panelists that previous recommendations did not reflect actual clinical practice due to overreliance on surrogate end points such as asymptomatic DVT detected via screening venography.33

The AT9 cite extensive clinical experience and similar or superior properties compared with other pharmacologic options as reasons for suggesting LMWH as preferred over other agents after total joint arthroplasty despite a lack of convincing evidence of superior efficacy or safety compared with older, less costly agents.4 The risk of bleeding associated with LMWH use following orthopedic surgery relates closely to the timing of thromboprophylaxis initiation. Administration of LMWH within 2 hours preoperatively or postoperatively increases the risk of bleeding up to fivefold compared with starting 12 hours after surgery.4

Warfarin remains among the most commonly prescribed agents for VTE prevention after total joint arthroplasty offering some advantages over LMWH, including low acquisition cost and oral administration.42 Warfarin’s delayed onset of anticoagulant effect confers both a potential advantage (reduced immediate risk of postoperative bleeding) and a disadvantage (increased risk of early VTE). Previous American College of Chest Physician’s guidelines recommended a goal international normalized ratio (INR) range of 2 to 3 following orthopedic surgery while American Academy of Orthopaedic Surgery (AAOS) guidelines recommended an INR target of 2 or “one which appropriately balances the risk of PE and bleeding.”41,43 The AT9 now simply recommend “dose-adjusted warfarin” without specific guidance on target INR, and the AAOS guidelines also no longer recommend a specific INR target.4,43 Many orthopedic surgeons prefer low-intensity warfarin (e.g., INR 1.5 to 2.5) due to perceived lower risk of postoperative bleeding.42 Regardless of INR target, warfarin use following orthopedic surgery requires a well-coordinated monitoring system and timely INR testing.44 Arranging INR testing for patients following joint replacement surgery can be challenging due to limited patient mobility often requiring arrangement of home phlebotomy or use of point-of-care INR monitoring devices that erodes the cost advantage of warfarin.

Rivaroxaban, apixaban, and dabigatran offer the convenience of oral administration and fixed dosing without the need for routine coagulation testing. Clinical trials of these agents have demonstrated safety and efficacy similar to enoxaparin after total joint replacement, but they have not been studied after hip fracture surgery.4 The AT9 express a preference for apixaban or dabigatran (alternatively warfarin or rivaroxaban) in patients unwilling to use injections of LMWH, but neither dabigatran nor apixaban is FDA approved for VTE prevention after total joint arthroplasty.4 Postmarketing studies are needed to provide real-world context in a variety of patient populations to the favorable results of initial clinical trials for these new anticoagulants.

Duration of Therapy

The optimal duration of VTE prophylaxis following surgery is not well established.41 Prophylaxis should be given throughout the period of increased VTE risk. For general surgical procedures and medical conditions, once the patient is able to ambulate regularly and other risk factors are no longer present, prophylaxis can be discontinued.5,6 Because of the relatively high incidence of VTE in the first month following hospital discharge among patients who have undergone a lower extremity orthopedic procedure, extended prophylaxis following hospital discharge appears to be beneficial.4 Most clinical trials support the use of antithrombotic therapy for 21 to 35 days following total hip replacement and hip fracture repair surgeries.4

Pharmacoeconomic Considerations in VTE Prevention

The acquisition costs of graduated compression stockings, UFH, and warfarin are considerably less than those of LMWH, fondaparinux, rivaroxaban, and dabigatran. However, the drug acquisition cost is relatively small when compared with the overall cost of care.32 Economic analyses must take into account therapeutic efficacy, duration of use, complications, and monitoring costs.45

The determination of the cost-effectiveness of VTE prophylaxis is based on the premise that a reduction in future VTE events will reduce overall healthcare costs.32 The cost of providing VTE prophylaxis for 1,000 patients declines as the incidence of VTE in a given population increases. Compared with no prophylaxis, prophylaxis with UFH or LMWH in immobilized medically ill patients appears to be cost-effective, with incremental costs ranging from $65 to $2,534 per quality-adjusted life-year.6 Studies of nonorthopedic surgical populations suggest that use of graduated compression stockings, IPCs, LMWH, and LDUH is more cost-effective than no prophylaxis.5Pharmacologic prophylaxis in high-risk surgical and orthopedic populations is likely to be cost-effective compared with no prophylaxis given the high risk for postoperative VTE. In this population a systematic review showed that fondaparinux was more cost-effective than LMWH, and that warfarin and LMWH were similarly cost-effective.46 The availability of generic enoxaparin may alter these findings. Extended prophylaxis is cost-effective after total hip arthroplasty, but the results of cost analysis after total knee arthroplasty are inconclusive.46


Unfractionated Heparin

The parenteral administration of UFH is preferred for acute VTE treatment in patients with severe renal insufficiency (creatinine clearance <30 mL/min [<0.5 mL/s]).7 UFH may be administered subcutaneously (SC) with or without coagulation monitoring or by continuous IV infusion (see Table 9-8). Because the anticoagulant response to UFH is highly variable, it is standard practice to adjust the dose based on the results of coagulation tests. When UFH is administered by IV infusion, the activated partial thromboplastin time (aPTT) is generally used to monitor the anticoagulant effect, although aPTT response varies between laboratories using different reagents and instruments and methods for improving interlaboratory agreement (e.g., standardization of aPTT ratios by reference to anti-Xa levels) have had limited success.3 For these reasons, the therapeutic aPTT range at each institution should be adapted to the responsiveness of the reagent and instrument used.3 Either weight-based dosing (see Table 9-8) or fixed dosing (e.g., 5,000 unit bolus followed by 1,000 units/h continuous infusion) of UFH produces similar clinical outcomes.44However, failure to give a sufficient IV UFH dose has been shown to increase the risk of VTE recurrence not only during initial treatment but also during long-term therapy.3 IV UFH requires hospitalization with frequent aPTT monitoring and dose adjustment. Inpatient anticoagulation management services have been shown to improve patient care by increasing the proportion of aPTT values in the therapeutic range, reducing the length of hospital stay, and lowering total hospital costs when compared with usual care.47 However, some patients still fail to achieve an adequate response to UFH therapy.3 Consequently, traditional IV UFH in the acute treatment of VTE has largely been replaced by LMWH or fondaparinux. However, as clearance of LMWH, fondaparinux, and the new oral anticoagulants is dependent in some degree on renal function, UFH will continue to have a role for acute VTE treatment in patients with creatinine clearance <30 mL/min (<0.5 mL/s).7

TABLE 9-8 Weight-Baseda Dosing for Unfractionated Heparin Administered by Continuous IV Infusion


If a sufficient dose of UFH is administered SC, aPTT-guided dose titration may be unnecessary. Weight-based LMWH compared with weight-based UFH (initial dose 333 units/kg SC followed by 250 units/kg twice daily) without coagulation monitoring for the treatment of acute VTE revealed no difference in recurrent VTE, major bleeding, or death during followup. Both groups received warfarin therapy overlapped for at least 5 days and continued after LMWH or UFH was discontinued.48 UFH administered in this manner may be a less costly option for treatment of acute VTE in appropriately selected patients. For patients weighing more than 80 kg, injection volume may be problematic.

Low-Molecular-Weight Heparin

Because of improved pharmacokinetic and pharmacodynamic profiles as well as ease of use, the LMWHs have largely replaced UFH for the treatment of VTE. LMWH given SC in fixed, weight-based doses (see Table 9-9) is at least as effective as UFH given IV for the treatment of VTE.3 There appears to be similar efficacy and safety with inpatient or outpatient LMWH administration, once- or twice-daily dosing regimens, and use of different LMWH preparations. A preference for once-daily LMWH administration is suggested by the AT9, provided the once-daily regimen uses the same daily LMWH dose as the twice-daily regimen (i.e., the once-daily injection contains double the dose of each twice-daily injection).7

TABLE 9-9 FDA-Approved Venous Thromboembolism Indications and Doses for Low-Molecular-Weight Heparins


Given the predictable response and the reduced need for laboratory monitoring with LMWH, stable patients with DVT who have normal vital signs, low bleeding risk, and no other comorbid conditions requiring hospitalization can be discharged early or treated entirely on an outpatient basis (see Table 9-7).7 In one survey, 91% of patients who received outpatient DVT treatment indicated a high degree of satisfaction.39

Patients presenting with PE and no evidence of hemodynamic instability are at low risk of subsequent morbidity and mortality. Evidence suggests that patients with submassive PE who are hemodynamically stable can be managed safely as outpatients with LMWH or fondaparinux.1,7 However, hemodynamically unstable patients with PE should generally be admitted for anticoagulation therapy initiation. Rapidly reversible UFH is preferred if thrombolytic therapy or embolectomy is anticipated.49

Not all patients are appropriate candidates for outpatient VTE treatment. At a minimum, patients must be reliable or have adequate caregiver support and be willing and active participants in the outpatient management of VTE.39Table 9-10 summarizes important aspects of patient education for outpatient VTE treatment. Patients who are unable to manage or who decline home treatment should be admitted to the hospital. These patients may subsequently opt for early discharge on LMWH. Daily patient contact either in person or via telephone is essential to identify potential complications and to address questions and concerns promptly. During daily contacts patients must be asked about symptoms that may indicate bleeding, thrombus extension, and PE.39 Once acute treatment with LMWH has been transitioned to long-term warfarin therapy (usually after about 5 to 10 days) and the warfarin dose stabilizes, patient contacts can occur less frequently.

TABLE 9-10 Patient Education for Outpatient Venous Thromboembolism Therapy



Fondaparinux has been shown to be a safe and effective alternative to LMWH for the treatment of acute VTE.7 It is dosed once daily via weight-based SC injection as follows: 5 mg if <50 kg, 7.5 mg if 50 to 100 kg, and 10 mg if >100 kg.50 Compared with weight-based LMWH dosing, this flexible dosing scheme may be particularly useful with very obese patients. Careful attention should be paid to renal function when fondaparinux is used to treat VTE as the drug is contraindicated if creatinine clearance is <30 mL/min (<0.5 mL/s).50


Warfarin monotherapy is an unacceptable choice for the acute treatment of VTE because it does not produce a rapid anticoagulation effect and is associated with high incidence of recurrent thromboembolism. However, warfarin is very effective in the long-term management of VTE and should be started concurrently with rapid-acting injectable anticoagulant therapy.7 The rapid-acting injectable anticoagulant should overlap with warfarin therapy for at least 5 days and until an INR ≥2 has been achieved for at least 24 hours.7 The initial dose of warfarin should be 5 to 10 mg for most patients (see Fig. 9-6) and periodically adjusted to achieve and maintain an INR between 2 and 3. For patients sufficiently healthy to be treated as outpatients, the AT9 suggest initiating warfarin therapy with 10 mg daily for the first 2 days followed by dosing based on INR measurements rather than starting with the estimated daily maintenance dose.44


FIGURE 9-6 Initiation of warfarin therapy. (INR, international normalized ratio; PT, prothrombin time.)

Pharmacoeconomic Considerations in VTE Treatment

Hospitalization is the main cost driver in the management of VTE. Although the drug acquisition costs for the LMWHs and fondaparinux are substantially higher than those for UFH, avoiding hospitalization dramatically decreases the overall costs of DVT treatment.51 Cost-effectiveness analyses using decision modeling suggest that the treatment of DVT with LMWHs is more cost-effective than the treatment with UFH in both inpatient and outpatient settings.52 Based on this decision model, LMWH will reduce overall healthcare cost if as few as 8% of patients are treated entirely on an outpatients basis or 13% of patients are discharged from hospital early.

Despite the substantial cost savings stemming from outpatient DVT treatment from the perspective of the insurer, the reality is that some patients are unable to afford LMWH or fondaparinux prescriptions. Fixed-dose UFH as described previously may provide a lower-cost option for the outpatient treatment of VTE for selected patients who otherwise might not have been able to afford it.48 A generic formulation of enoxaparin is now available in the United States. The cost-effectiveness of rivaroxaban or dabigatran (discussed below) compared with that of standard anticoagulation therapy with warfarin initiated with parenteral anticoagulation has yet to be defined.

Thrombolysis and Thrombectomy

Most cases of VTE require only anticoagulation therapy. In rare cases removal of the occluding thrombus by either pharmacologic or surgical means may be warranted. Consensus panel recommendations regarding either thrombolysis or thrombectomy in the management of VTE are based on low-quality evidence, and more study is clearly needed to clarify their precise role.7

Thrombolytic agents are proteolytic enzymes that enhance the conversion of plasminogen to plasmin that subsequently degrades the fibrin clot matrix.7 Thrombolytic therapy for DVT improves early venous patency, but this does not necessarily translate into improved long-term outcomes for all patients.7 There is some evidence from pooled study analyses suggesting that thrombolytic therapy is superior to anticoagulation therapy alone in preventing postthrombotic morbidity.7 Catheter-directed instillation of a thrombolytic agent directly into the clot is increasingly being used. The risk of bleeding associated with catheter-directed drug administration appears to be less than that associated with systemic administration.7 Patients who present within 14 days of symptom onset with extensive proximal DVT, with good functional status, low bleeding risk, and a life expectancy of a year or more are candidates for thrombolysis (Table 9-11). For DVT, catheter-directed thrombolysis is preferred provided appropriate expertise and resources are available. Catheter-based fragmentation of thrombus, with or without aspiration of the thrombus fragments, can be combined with catheter-directed thrombolysis and the combination has been associated with shorter treatment times and reduced cost. The same duration and intensity of anticoagulation therapy is recommended as for DVT patients who do not receive thrombolysis.7 Patients with DVT that involves the iliac and common femoral veins are at highest risk for postthrombotic syndrome and may have the greatest potential to benefit from thrombus removal strategies. In patients with impending venous gangrene despite optimal anticoagulant therapy, thrombus removal is indicated; for all other patients with acute DVT, the AT9 suggest anticoagulation therapy alone over either catheter-directed or systemic thrombolysis.7

TABLE 9-11 Thrombolysis for the Treatment of Venous Thromboembolism


In the management of acute PE successful clot dissolution with thrombolytic therapy reduces elevated pulmonary artery pressure and normalizes right ventricular dysfunction. However, the risk of death from PE should outweigh the risk of serious bleeding associated with thrombolytic therapy. Patients being considered for thrombolytic therapy should be screened carefully for contraindications relating to bleeding risk (see Table 9-11).7 Thrombolytic therapy is considered necessary in addition to aggressive interventions such as volume expansion, vasopressor therapy, intubation, and mechanical ventilation for patients with massive PE manifested by shock and cardiovascular collapse (about 5% of patients diagnosed with PE).7 Thrombolytic therapy in these patients should be administered without delay to reduce the risk of progression to multisystem organ failure and death. While lifesaving in the acute phase of massive PE with hypotension, the hemodynamic benefit of thrombolysis is comparable to that of UFH after a few days.53

The benefit of thrombolytic therapy in patients with PE without hemodynamic compromise is less clear and rapid risk stratification is required to determine whether the patient may benefit from thrombolysis or embolectomy in addition to anticoagulation therapy.49 Risk stratification helps inform the intensity of initial treatment, low-risk patients being discharged early or managed as outpatients and high-risk patients receiving intensive surveillance in the intensive care unit and/or advanced therapies such as thrombolysis.54 Three key components of risk stratification are clinical evaluation, determination of cardiac biomarker levels such as troponin, and assessment of right ventricular size and function.7 The Pulmonary Embolism Severity Index (PESI) is a prognostic tool utilizing 11 routinely available clinical parameters: demographics (age and gender), comorbid illnesses (cancer, heart failure, and chronic lung disease), and clinical findings (pulse, systolic blood pressure, respiratory rate, temperature, mental status, and arterial oxygen saturation), that stratifies patients into 5 risk classes with classes I and II considered low risk.54 The AT9 suggest that patients with acute PE who initially present without hypotension be risk stratified predominantly by clinical signs indicating instability including a decrease in systolic blood pressure that still remains >90 mm Hg, tachycardia, elevated jugular venous pressure, clinical evidence of poor tissue perfusion, hypoxemia, and failure to improve on anticoagulant therapy.7 Patients with one or more of these clinical features are at high risk for PE-related morbidity and mortality and may benefit from thrombolytic therapy, provided bleeding risk is acceptable, even in the absence of hemodynamic compromise (see Table 9-11).7 The optimal role of thrombolysis in the management of PE requires further study.

In rare circumstances surgical thrombectomy for extensive ileofemoral DVT may be necessary, but catheter-directed thrombolysis is preferred if bleeding risk is acceptable.7 For treatment of acute PE, catheter-based embolectomy might be particularly suitable for patients who have contraindications to thrombolytic therapy, have failed thrombolytic therapy, or in whom death is likely before onset of thrombolysis. In the absence of contraindications, catheter-based embolectomy in PE is usually combined with thrombolytic therapy unless bleeding risk is high.7 Surgical embolectomy is reserved for massive PE and hemodynamic instability when thrombolysis is contraindicated and for when thrombolysis has failed clinically or will not have sufficient time to take effect.7 In cases of chronic PE—where persistent emboli produce CTPH, hypoxemia, and right-sided heart failure—surgical pulmonary thromboendarterectomy offers greater benefit than anticoagulants and may be the treatment of choice if performed by an experienced surgical team. A permanent IVC filter is usually inserted before or during the procedure and these patients need long-term warfarin therapy targeted to an INR of 2 to 3.7

Alternative Drug Treatments


Images A randomized, open-label, controlled noninferiority trial has demonstrated that oral rivaroxaban alone is noninferior to traditional therapy with warfarin (targeted INR 2 to 3) overlapped with enoxaparin at initiation for both acute DVT and PE with similar rates of recurrent VTE and clinically relevant bleeding.37,38 Bleeding meeting the prespecified definition of major was lower with rivaroxaban in the PE trial, but not in the DVT trial. Rivaroxaban was administered in a fixed dose of 15 mg twice daily for 3 weeks followed by 20 mg once daily for at least 3 months without routine coagulation monitoring. Patients with creatinine clearance <30 mL/min (<0.5 mL/s) or cancer and those requiring thrombolytic therapy were excluded from study participation and should probably not be treated with rivaroxaban until additional information is available. Replacing the effective but cumbersome combination of LMWH or fondaparinux and warfarin with a single-drug regimen holds promise for simplifying the treatment of VTE; however, the higher acquisition cost of rivaroxaban and lack of an effective reversal agent will be of concern to some patients and clinicians.

Clinical Controversy…

Results of recent randomized clinical trials in patients with acute VTE indicate that initiation of anticoagulation with a single oral anticoagulant, rivaroxaban that does not require routine therapeutic monitoring, produced efficacy and safety outcomes that were noninferior to standard therapy.37,38 These findings highlight the potential of rivaroxaban to greatly simplify the management of VTE. The lack of a reliable rivaroxaban reversal agent and limited experience managing rivaroxaban-associated bleeding will be concerning to some clinicians and patients. The cost-effectiveness of rivaroxaban for treatment of venous thrombosis also remains to be defined and assessing outcomes outside of carefully controlled clinical trials in real-world patient populations is needed.


Oral dabigatran 150 mg twice daily was compared with warfarin (targeted INR 2 to 3) in a randomized, double-blind, noninferiority trial involving patients with acute VTE.55 Both treatment groups were initially given parenteral anticoagulation therapy (UFH or LMWH). The primary outcome was the 6-month incidence of recurrent symptomatic VTE and related deaths, and the main safety end point was bleeding events. The results indicated that fixed dose of dabigatran was as effective as and had a safety profile that is similar to that of warfarin, and did not require laboratory monitoring. Patients with hemodynamically unstable PE or creatinine clearance <30 mL/min (<0.5 mL/s) were excluded as were those at high risk for bleeding. The requirement for parenteral anticoagulation at initiation of dabigatran therapy is a disadvantage compared with VTE treatment using rivaroxaban.


Oral apixaban administered as 10 mg twice daily for 7 days followed by 5 mg twice daily for 6 months is being compared with warfarin (targeted INR 2 to 3) with initial enoxaparin coverage in an ongoing double-blind, randomized clinical trial with results expected sometime in 2013.

Treatment of Venous Thromboembolism in Special Populations


The use of anticoagulation therapy for the treatment and prevention of DVT or PE during pregnancy is common.17 UFH and LMWH are preferred during pregnancy (Table 9-12) as they do not cross the placenta and evidence suggests they are safe for the fetus.17 Warfarin crosses the placenta and can result in fetal bleeding, central nervous system abnormalities, and embryopathy and should not be used for treatment of VTE during pregnancy.17 Women of childbearing age taking warfarin must be counseled regarding the fetal risks and effective contraception should be used. Dabigatran, rivaroxaban, and apixaban should be avoided in pregnancy until more information regarding safety is available.5759 Fondaparinux, rivaroxaban, dabigatran, and apixaban have not been formally evaluated in pregnant patients.2,3

TABLE 9-12 Unfractionated and Low-Molecular-Weight Heparin Use During Pregnancy


All pregnant women with a history of VTE should receive VTE prophylaxis for 6 weeks after delivery. Antenatal prophylaxis may also be indicated depending on other risk factors, such as history of multiple VTE, VTE associated with pregnancy or estrogen therapy, or known thrombophilia. Anticoagulation for acute VTE during pregnancy should continue for at least 6 weeks postpartum and for a minimum total duration of 3 months.28 Warfarin, UFH, and LMWH are safe for use during breast-feeding.60

Pediatric Patients

Although seen far more frequently in adults, VTE in pediatric patients is increasing secondary to prematurity, cancer, trauma, surgery, congenital heart disease, and systemic lupus erythematosus. Pediatric patients often develop DVTs associated with indwelling central venous catheters. In contrast to adults, pediatric patients rarely develop idiopathic VTE.61 While recommendations for anticoagulant therapy in pediatric patients are largely extrapolated from recommendations in adults, there are likely important pharmacokinetic and pharmacodynamic differences that should be taken into consideration. The majority of literature supporting pediatric recommendations is derived from uncontrolled studies, case reports, or in vitro experiments. When possible, a pediatric hematologist with experience managing VTE should manage pediatric patients.61

Anticoagulation with UFH and warfarin remains the most frequently used approach for the treatment of VTE in pediatric patients. The recommended target aPTT and INR ranges as well as the duration of therapy are extrapolated from clinical trials in adults. Recent data suggest that extrapolating aPTT range from adults to pediatric patients is unlikely to be valid. However, in the absence of supporting clinical data, extrapolation of the adult aPTT range to pediatric patients remains necessary.61 The recommended initial bolus dose of UFH is 75 to 100 units/kg given IV over 10 minutes followed by a maintenance infusion of 28 units/kg/h for infants 2 to 12 months of age and 20 units/kg/h for children aged 1 year or older.61 Subsequent infusion rate adjustments should be made every 4 to 6 hours to maintain the aPTT within the institution-specific therapeutic range. The usual warfarin starting dose is 0.2 mg/kg with a maximum of 10 mg. Infants require higher doses of warfarin per kilogram to maintain a target INR of 2 to 3 compared with teenagers and adults (mean dose 0.33 mg/kg, 0.09 mg/g, and 0.04 to 0.08 mg/kg, respectively).61 The INR target range for VTE treatment in children is 2 to 3. Frequent INR monitoring and warfarin dose adjustments are typically required. When compared with adults, only 10% to 20% of pediatric patients can be safely monitored once monthly.61 Obtaining coagulation monitoring tests in pediatric patients is problematic because many have poor or nonexistent venous access. To address this problem, many clinicians recommend using finger-stick blood samples with a portable point-of-care monitor.61 Despite need for daily injections, LMWH is an attractive alternative in pediatric patients due to low drug interaction potential and less frequent laboratory testing. Enoxaparin, dalteparin, and tinzaparin have been evaluated in pediatric patients. Most experts recommend that anti-Xa activity be monitored and the dose adjusted to maintain anti–factor Xa levels between 0.5 and 1 unit/mL (0.5 and 1 kU/L) 4 to 6 hours following SC injection. Compared with adults, children younger than 2 to 3 months of age or who weigh less than 5 kg have higher per-kilogram dose requirements to achieve a “therapeutic” response. The dose of LMWH for older children is generally similar to the weight-adjusted doses used in adults.61 Warfarin can be initiated concurrently with UFH or LMWH therapy. Therapy should be overlapped for a minimum of 5 days and until the INR is therapeutic. Warfarin should be continued for at least 3 months for provoked VTE and 6 months for idiopathic VTE.61 Thrombolysis and thrombectomy have been successfully employed in pediatric patients, but published data are very limited—routine use is not recommended.61

Patients with Cancer

VTE is a frequent complication of cancer. Furthermore, compared with patients without cancer, the natural history of VTE in patients with cancer is different having threefold higher rates of recurrent VTE and bleeding and more resistance to standard warfarin-based therapy.7,14 Warfarin therapy in cancer patients is often complicated by drug interactions (e.g., chemotherapy and antibiotics) and the need to frequently interrupt therapy for invasive procedures (e.g., thoracentesis, percutaneous biopsies, and abdominal paracentesis). Maintaining stable INR control is more difficult in this patient population because of nausea, anorexia, and vomiting.7

Randomized trials provide evidence that long-term LMWH monotherapy for VTE in patients with cancer significantly decreases the rate of recurrent VTE without increasing bleeding risks compared with traditional warfarin-based therapy.7,14 Despite recommendations from AT9, the American Society of Clinical Oncology, and the National Comprehensive Cancer Network that patients with cancer should be given LMWH monotherapy for the long-term treatment of VTE, most patients with cancer-related VTE continue to receive warfarin-based therapy.62 The reasons for this phenomenon are unclear, but possible explanations might be patient preference for oral therapy over daily injections, and/or the higher cost of LMWH. Pooled analysis of clinical trials demonstrates no survival advantage of LMWH monotherapy compared with traditional therapy with warfarin.7 Advantages of LMWH over warfarin for VTE treatment in cancer are expected to be greatest in those with one or more of the following: metastatic disease, treatment with aggressive chemotherapy, extensive VTE at presentation, liver dysfunction, poor or unstable nutritional status, or desire to avoid frequent blood draws for coagulation monitoring.7

For patients with VTE and cancer who do receive LMWH, therapy should continue for at least the first 3 to 6 months of long-term treatment, at which time further LMWH can be considered or warfarin therapy can be substituted. Anticoagulation therapy should continue for as long as the cancer is “active” and while the patient is receiving antitumor therapy.7 A risk-to-benefit assessment should be performed on a regular basis and the overall clinical status of the patient should be considered, along with the risk for bleeding, quality of life, and life expectancy.7 Because so few patients with cancer-related VTE were included in rivaroxaban and dabigatran clinical trials, an assessment of whether these anticoagulants are appropriate alternatives to warfarin or LMWH for VTE treatment in this patient population is not possible until additional information becomes available.7

Clinical Controversy…

Most VTE in patients with cancer occurs in the outpatient setting. Several recent randomized clinical trials evaluated the utility of thromboprophylaxis for patients with solid tumors receiving systemic chemotherapy.14 While these studies demonstrate that outpatient thromboprophylaxis is feasible, safe, and effective, the rates of VTE in the placebo arms of the trials argue against a broad application of prophylaxis for cancer patients. How to prospectively identify cancer patients at sufficiently high VTE risk to justify the use of thromboprophylaxis is being actively studied, but is currently poorly understood.

Patients with Renal Insufficiency

Patients with acute or chronic renal compromise often require anticoagulation for concomitant risk or treatment of thromboembolic disorders. With the exception of warfarin, most anticoagulants have at least some dependency on renal elimination. Accumulation of drug is possible during treatment with LMWH, fondaparinux, dabigatran, rivaroxaban, and apixaban.2,3 In addition, patients with chronic kidney disease are at increased risk of bleeding, independent of drug clearance.63

LMWHs are primarily reliant on renal elimination and should be used with caution in patients with severe renal impairment.64 Enoxaparin has specific labeling for patients with creatinine clearance <30 mL/min (<0.5 mL/s), but supporting evidence is limited to pharmacokinetic modeling analyses and bleeding and thromboembolic outcomes in this patient population are not well defined.65 UFH is preferred for acute VTE treatment when creatinine clearance is <30 mL/min (<0.5 mL/s).7

Dabigatran, rivaroxaban, and apixaban each relies on renal elimination and requires dose adjustment at varying degrees of renal impairment.5759 Use of these anticoagulants in patients with creatinine clearance <30 mL/min (<0.5 mL/s) should be avoided.


Unfractionated Heparin

UFH has been used for the prevention and treatment of thrombosis for decades. Commercially available UFH preparations are derived from bovine lung or porcine intestinal mucosa. Although some differences exist between the two sources, no differences in antithrombotic activity have been demonstrated.3


UFH is a heterogeneous mixture of sulfated mucopolysaccharides of variable lengths and pharmacologic properties (Table 9-13).3 Each heparin molecule is composed of repetitive units of D-glycosamine and uronic acid. The anticoagulant profile and clearance of each UFH molecule varies based on its length. Smaller chains are cleared less rapidly than their longer counterparts.3

TABLE 9-13 Comparison of the Chemical and Pharmacokinetic Properties of Antithrombotic Drugs Used for Venous Thrombosis


The anticoagulant effect of UFH is mediated through a specific pentasaccharide sequence on the heparin molecule that binds to antithrombin, provoking a conformational change (Fig. 9-7). Only one third of the UFH molecules possess the unique pentasaccharide sequence with affinity for antithrombin.3 The UFH–antithrombin complex is 100 to 1,000 times more potent as an anticoagulant compared with antithrombin alone. Antithrombin inhibits the activity of several clotting factors including IXa, Xa, XIIa, and thrombin. Through its action on thrombin, the UFH–antithrombin complex also inhibits the thrombin-induced activation of factors V and VIII.3 UFH prevents the growth and propagation of formed thrombus and allows endogenous thrombolytic systems to degrade the clot.


FIGURE 9-7 Pharmacologic activity of unfractionated heparin, low-molecular-weight heparins (LMWHs), and fondaparinux.

Factors IIa (thrombin) and Xa are most sensitive to inhibition by the UFH–antithrombin complex. To inactivate thrombin, the heparin molecule must form a ternary complex bridging between antithrombin and thrombin (see Fig. 9-7).3 Only molecules that contain more than 18 saccharides are able to bind to both antithrombin and thrombin simultaneously. Smaller heparin molecules cannot facilitate the interaction between antithrombin and thrombin. In contrast, the inactivation of factor Xa does not require UFH to form a bridge with antithrombin, but requires only UFH binding to antithrombin via the specific pentasaccharide sequence. UFH molecules with as few as five saccharide units are able to catalyze the inhibition of factor Xa. After it has produced its effect UFH uncouples from antithrombin and quickly recouples with another antithrombin molecule.3 Because of its relatively large size, the UFH–antithrombin complex is incapable of inactivating thrombin or factor Xa within a formed clot or bound to surfaces. At high doses, UFH also binds to heparin cofactor II, further inhibiting the activity of thrombin.3 UFH increases the release of tissue factor pathway inhibitor from vascular endothelium, augmenting its inhibitory effect on factor Xa. UFH, especially high-molecular-weight fractions, also binds to platelets and inhibits platelet aggregation.3


UFH is not reliably absorbed when taken orally as a result of its large molecular size and anionic structure. The bioavailability and biologic activity of UFH is limited by its propensity to bind to plasma proteins, platelet factor-4 (PF-4), macrophages, fibrinogen, lipoproteins, and endothelial cells. This may explain the substantial interpatient and intrapatient variability observed in the anticoagulation response to UFH.3

The SC bioavailability of UFH is dose dependent and ranges from 30% at low doses to as much as 70% at high doses. The onset of anticoagulant effect is usually evident 1 to 2 hours after SC injection and peaks at 3 hours.3When UFH is administered via the IV route, a continuous infusion is preferable.7 Intramuscular administration is discouraged because of the risk of large hematoma formation.

UFH has a dose-dependent half-life of approximately 30 to 90 minutes, but may be prolonged to as much as 150 minutes when given in high doses.3 There are two primary mechanisms for the elimination of UFH. One is a rapid, but saturable zero-order process involving enzymatic inactivation of heparin molecules bound to endothelial cells and macrophages.3 The other mechanism is renal elimination via a slower, nonsaturable first-order process. With typical therapeutic regimens, a combination of the two mechanisms eliminates UFH with the saturable mechanism predominating.3

Dose and Administration

The dose of UFH is expressed in units of activity. For the prevention of VTE, UFH is given by SC injection in the abdominal fat layer. The typical dose for prophylaxis is 5,000 units every 8 to 12 hours. When immediate and full anticoagulation is required, an IV bolus dose followed by a continuous infusion is preferred (Table 9-8).3 SC UFH (initial dose of 333 units/kg followed by 250 units/kg every 12 hours) also provides adequate therapeutic anticoagulation for the treatment of acute VTE.48

Therapeutic Monitoring

Images Administration of UFH requires close monitoring because of the unpredictable anticoagulant patient response.3 The aPTT is the most widely used test to determine the degree of therapeutic anticoagulation and, although most experts advocate using the aPTT to monitor UFH provided that institution-specific therapeutic ranges are defined, the use of aPTT has several limitations as discussed previously. The aPTT should be measured prior to the initiation of therapy to determine the patient’s baseline. With IV infusion, the aPTT response to UFH therapy should be measured 6 hours after initiation or a dose change; this is usually sufficient time to reach steady state. Promptly adjust the dose of UFH based on patient response and the institution-specific aPTT therapeutic range.3

Adverse Effects

Bleeding is the primary adverse effect associated with all anticoagulant drugs (Table 9-14).63 Low-dose SC UFH is associated with a minimal risk of major bleeding, while the rates of major bleeding for patients receiving full therapeutic doses of UFH range from 0% to 2%.63

TABLE 9-14 Risk Factors for Major Bleeding While Taking Anticoagulation Therapy


Heparin-induced thrombocytopenia (HIT) is a rare drug-induced immunologic reaction requiring immediate intervention.66 The most common complication of HIT is VTE; arterial thromboembolic events occur less frequently. Approximately 5% to 10% of patients with HIT die, usually as a result of thrombotic complications.66 Thrombocytopenia (defined as a platelet count <150 × 103/mm3 [<150 × 109/L]) is the most common clinical manifestation of HIT and occurs in up to 95% of patients with confirmed HIT if a proportional fall in platelet count of 30% to 50% even though the count stays above 150 × 103/mm3(150 × 109/L) is included in the definition.66 The characteristic onset of falling platelet count in HIT is 5 to 10 days after initiation of UFH (day 0 being the first day of UFH), particularly when administered perioperatively.66 Thrombocytopenia alone is not sufficient for diagnosing HIT; serologic confirmation of heparin antibodies using an assay available only in a few specialty laboratories is required.66 Falsely diagnosing HIT can have serious consequences including unnecessary anxiety, unnecessary withdrawal of UFH, and the use of alternative anticoagulants with higher bleeding risk. One decision analysis found that strict adherence to platelet monitoring for HIT could, at best, prevent one thrombosis per 1,000 patients screened at the cost of one major bleeding event.66 For these reasons, AT9 suggest monitoring platelet counts every 2 to 3 days from day 4 to 14 of UFH only in populations where the expected risk of HIT exceeds 1%.66 The use of a clinical prediction rule, such as the 4 Ts score (thrombocytopenia, timing of platelet count fall or thrombosis, thrombosis, other explanation for thrombocytopenia), can improve the predictive value of platelet count monitoring and heparin antibody testing (Table 9-15).66,67While a low 4 Ts score identifies patients with low probability of HIT (0% to 3%), a substantial proportion of patients with a high 4 Ts score do not have HIT either (24% to 61%).66 A prospective trial is needed to definitively determine the role of the 4 Ts score in HIT surveillance and management. A 4 Ts score should be calculated when HIT is suspected in patients receiving heparin (UFH or LMWH). If the 4 Ts score is low, no further workup is needed, whereas if the 4 Ts score is moderate or high, further workup of HIT including serologic testing should be undertaken.68 All heparin should be discontinued if new thrombosis occurs in the setting of falling platelets in conjunction with a moderate or high 4 Ts score. Alternative anticoagulation should then be initiated with a parenteral direct thrombin inhibitor.66

TABLE 9-15 The 4 Ts Probability Score for Heparin-Induced Thrombocytopenia


Long-term UFH has been reported to cause alopecia, priapism, and suppressed aldosterone synthesis with subsequent hyperkalemia.3 The use of UFH in doses ≥20,000 units/day for more than 6 months, especially during pregnancy, is associated with significant bone loss and may lead to osteoporosis.3 Few drug interactions are reported with UFH, but concurrent use with other anticoagulant, thrombolytic, and antiplatelet agents increases the risk of bleeding.3

Management of Bleeding

Hemorrhage can occur at any site in patients receiving UFH and close monitoring for signs and symptoms of bleeding is crucial.3,63 When major bleeding occurs, UFH should be immediately discontinued and the underlying source of bleeding identified and treated. IV protamine sulfate in a dose of 1 mg per 100 units of UFH up to a maximum of 50 mg can be administered via slow IV infusion to reverse the anticoagulant effects of UFH.3 Protamine sulfate neutralizes UFH in 5 minutes, and its effect persists for 2 hours. Multiple doses or prolonged infusion of protamine sulfate may be necessary if hemorrhage continues.3

Low-Molecular-Weight Heparin

LMWH fragments produced by either chemical or enzymatic depolymerization of UFH (see Table 9-13) are heterogeneous mixtures of sulfated glycosaminoglycans with approximately one third the molecular weight of UFH.3LMWH has a similar mechanism of action as UFH, but with reduced inhibitory activity against thrombin relative to factor Xa, less affinity for plasma proteins, and longer duration of activity.3 Advantages of LMWH over UFH include (a) predictable anticoagulation dose response, (b) improved SC bioavailability, (c) dose-independent clearance, (d) longer biologic half-life, (e) lower incidence of thrombocytopenia, and (f) reduced need for routine laboratory monitoring.3 Currently three LMWH products are available in the United States (see Table 9-13) with only enoxaparin being available in a generic formulation. The FDA-approved indications and doses relating to VTE for the LMWHs are product specific (Table 9-9).


LMWH prevents the growth and propagation of formed thrombi by enhancing and accelerating the activity of antithrombin through binding to a specific pentasaccharide sequence present on about one third of LMWH molecules.3The principal difference in the pharmacologic activity of LMWH and UFH is their relative inhibition of factor Xa and thrombin. Because of smaller chain lengths, LMWH has limited activity against thrombin (see Fig. 9-7). The ratio of anti–factor Xa-to-IIa activity varies between 4:1 and 2:1. By comparison, UFH has an anti–factor Xa-to-IIa activity ratio of 1:1.3


Compared with UFH, LMWH has a more predictable anticoagulation response. The improved pharmacokinetic profile of LMWH is the result of reduced binding to proteins and cells.3 The bioavailability of LMWH is about 90% when administered SC. The SC bioavailability of the available LMWH products differs only slightly. The peak anticoagulation effect is seen in 3 to 5 hours.3 The predominant mode of elimination for LMWH is renal. Consequently, their biologic half-life may be prolonged in patients with renal impairment.3 The plasma half-life of LMWH preparations is 3 to 6 hours. The clearance of LMWH is independent of dose.3

Dosing and Administration

LMWH is given in fixed or weight-based doses based on the product and indication (see Table 9-9). Doses should be based on actual body weight and studies in obese patients indicate that full weight-based doses do not lead to elevated LMWH concentrations when compared with normal subjects; consequently, dose capping is not recommended.64 The dose for enoxaparin is expressed in milligrams, whereas dalteparin and tinzaparin are expressed in units of anti–factor Xa activity. LMWH is given by SC injection as described in Table 9-10.

The dosing interval for LMWH is every 12 or 24 hours depending on the indication and product. For VTE treatment, AT9 suggest once- over twice-daily LMWH administration provided the approved once-daily regimen contains double the dose of each twice-daily injection.7 Given that the elimination half-life of LMWH is prolonged in patients with severe renal impairment, unadjusted therapeutic doses may lead to a significant accumulation in these patients.3 The enoxaparin dose should be reduced and the dosing interval extended to once daily in patients with creatinine clearance <30 mL/min (<0.5 mL/s).65 The pharmacokinetics of dalteparin and tinzaparin is less well characterized in patients with renal insufficiency, but some studies suggest that there is a lower degree of accumulation with tinzaparin.3 Data on the use of LMWH in patients with end-stage renal disease receiving hemodialysis are very limited; thus, UFH is preferred for these patients.3 Given that few published data are available regarding the use of LMWH in the setting of renal insufficiency, some experts recommend measuring anti–factor Xa activity if therapy is continued for more than a few days.3 When given in prophylactic doses, LMWH has not been shown to increase the risk of bleeding complications, irrespective of the degree of renal function impairment. For patients with creatinine clearance <30 mL/min (<0.5 mL/s) who require VTE prophylaxis, enoxaparin 30 mg once daily is recommended, but dosing recommendations are not available in the setting of renal insufficiency for dalteparin or tinzaparin.3

Therapeutic Monitoring

Because LMWH anticoagulant response is predictable when given SC, routine laboratory monitoring is unnecessary.3 Prior to initiation of LMWH, a baseline complete blood cell count with platelet count, and serum creatinine should be obtained. The complete blood cell count can be checked every 5 to 10 days during the first 2 weeks of LMWH therapy and every 2 to 4 weeks thereafter to monitor for occult bleeding. If neuraxial anesthesia has been used, patients should be closely monitored for signs and symptoms of neurologic impairment.65

Measurement of anti–factor Xa activity is the most widely used method to monitor LMWH in clinical practice. Routine anti–factor Xa activity measurement is unnecessary in patients whose condition is stable and uncomplicated.3 Although very limited data support the use of laboratory monitoring to guide LMWH therapy, measuring anti–factor Xa activity may be helpful in patients who have significant renal impairment (e.g., creatinine clearance <30 mL/min [<0.5 mL/s]), weigh less than 50 kg, are morbidly obese, or require prolonged therapy (e.g., longer than 14 days). Periodic anti–factor Xa activity monitoring may also be useful in women treated with LMWH during pregnancy because of changing pharmacokinetic variables (e.g., volume of distribution and renal function).3

When anti–factor Xa activity is used to monitor LMWH therapy, the sample should be drawn after steady state has been achieved (after the second or third dose) and approximately 4 hours after the SC injection, during the peak period of anti–factor Xa activity.3 The therapeutic range for anti–factor Xa activity is not well defined and has not been clearly correlated with efficacy or the risk of bleeding. For the treatment of VTE, an acceptable target range for the peak anti-Xa level for twice-daily enoxaparin dosing is 0.6 to 1 unit/mL (0.6 to 1 kU/L). For once-daily dosing likely peak targets are >1 unit/mL (>1 kU/L) for enoxaparin, 0.85 unit/mL (0.85 kU/L) for tinzaparin, and 1.05 units/mL (1.05 kU/L) for dalteparin.3

Adverse Effects

As with other anticoagulants, bleeding is the most common adverse effect of LMWH therapy.63 The frequency of major bleeding is purported to be less with LMWH than with UFH, but this has not been consistently demonstrated in clinical trials.63 Although there is no proven method for reversing LMWH anticoagulation if major bleeding does occur, IV protamine sulfate can be administered. However, because of limited binding to the shorter LMWH chains, protamine sulfate neutralizes only around 60% to 75% of LMWH anticoagulant activity.3 The recommended dose of protamine sulfate is 1/1 mg of enoxaparin or 1 mg/100 anti–factor Xa units of dalteparin or tinzaparin administered in the previous 8 hours. A second protamine sulfate dose of 0.5/1 mg or 100 anti–factor Xa units can be given if bleeding continues. Smaller doses of protamine sulfate can be used if the LMWH dose was given in the previous 8 to 12 hours. The use of protamine sulfate is not recommended if LMWH was administered more than 12 hours earlier.3

Although thrombocytopenia can occur with the use of LMWH, the incidence of HIT is three times lower than that observed with UFH, perhaps due to the reduced propensity of LMWH to bind to platelets.3Because LMWH exhibits nearly 100% cross-reactivity with heparin antibodies in vitro, LMWH should be avoided in patients with an established diagnosis or history of HIT.3 The risk of osteoporosis appears to be lower with LMWH than with UFH, but both agents have the potential to produce osteopenia.3


Fondaparinux, also known as pentasaccharide, is a synthetic molecule consisting of the five critical saccharide units that bind specifically, but reversibly, to antithrombin (see Fig. 9-7). Unlike UFH or LMWH, fondaparinux selectively inhibits factor Xa activity.3


Similar to UFH and LMWH, fondaparinux prevents thrombus generation and clot formation by indirectly inhibiting factor Xa activity through its interaction with antithrombin. Fondaparinux is not destroyed during this process and is released to bind many other antithrombin molecules.3


Fondaparinux is rapidly and completely absorbed following SC administration (absolute bioavailability 100%). Peak plasma concentrations are achieved in approximately 2 hours after a single dose and 3 hours with repeated once-daily dosing. At therapeutic concentrations, fondaparinux is highly and specifically bound to antithrombin. It does not bind to red blood cells or other plasma proteins including albumin, gp, platelets, or PF-4.3 Fondaparinux is primarily eliminated unchanged in the urine and is contraindicated in patients with creatinine clearance <30 mL/min (<0.5 mL/s). The terminal elimination half-life is 17 to 21 hours.3 The anticoagulant effect of fondaparinux persists for 2 to 4 days following discontinuation of the drug in patients with normal renal function. Fondaparinux has no known pharmacokinetic drug interactions. However, concurrent use with other drugs with anticoagulant, fibrinolytic, or antiplatelet activity increases the risk of hemorrhage.50

Dosing and Administration

Fondaparinux is FDA approved for the prevention of VTE following orthopedic (hip fracture, hip and knee replacement) or abdominal surgery and for the treatment of DVT and PE (in conjunction with warfarin).50 In the setting of VTE prevention, the dose of fondaparinux is 2.5 mg injected SC once daily starting 6 to 8 hours following surgery if hemostasis has been established. It is important to avoid initiating fondaparinux too soon because there is a significant relationship between the timing of the first dose and the risk of major bleeding complications.3 Patients who weigh less than 50 kg should not be given fondaparinux for VTE prophylaxis.50 The usual duration of therapy is 5 to 9 days, but may be given as extended prophylaxis following hospital discharge for up to 35 days.4 For the treatment of DVT or PE, the dose of fondaparinux is 5 mg for patients up to 50 kg, 7.5 mg for 50 to 100 kg, and 10 mg for patients >100 kg.50

Therapeutic Monitoring

A complete blood cell count should be measured at baseline and monitored periodically to detect the possibility of occult bleeding.50 Baseline kidney function should be determined as fondaparinux is contraindicated when creatinine clearance is <30 mL/min (<0.5 mL/s). Signs and symptoms of bleeding should be monitored daily, particularly in patients with a baseline creatinine clearance between 30 and 50 mL/min (0.5 and 0.83 mL/s). If neuraxial anesthesia has been used, patients should be closely monitored for signs and symptoms of neurologic impairment.50

Fondaparinux does not alter coagulation tests such as the aPTT and prothrombin time (PT). The role of anti–factor Xa monitoring during fondaparinux is not well defined. Patients receiving fondaparinux therapy do not require routine coagulation testing.50

Adverse Effects

The primary adverse effect associated with fondaparinux therapy is bleeding.50 Similar to UFH and LMWH, fondaparinux should be used with extreme caution in patients with neuraxial anesthesia or following a spinal puncture because of the risk for spinal or epidural hematoma formation.50 Some case reports have documented successful treatment of HIT with fondaparinux, while others have implicated fondaparinux as a cause of HIT.69 A specific antidote to reverse the antithrombotic activity of fondaparinux is not currently available; if uncontrollable bleeding occurs during fondaparinux therapy, factor VIIa may be effective.3

Direct Anti-Xa Inhibitors

The introduction of LMWH and fondaparinux transformed the initial treatment of VTE from a purely inpatient endeavor to one where the majority of patients can be treated as outpatients. However, the need for daily SC injections is a significant barrier for some patients.38 Warfarin therapy is notoriously unpredictable and labor intensive, and can be stressful for patients and anticoagulation providers. These shortcomings in available anticoagulants have driven the search for replacements with rapid onset of effect that can be administered orally without the need for anticoagulant monitoring. Two such agents that target factor Xa are rivaroxaban and apixaban. Neither of these agents has been FDA approved for use in VTE treatment in the United States, but rivaroxaban has been approved for prevention of VTE following hip or knee replacement surgery.59

Pharmacology and Pharmacokinetics

Rivaroxaban and apixaban are potent and selective inhibitors of both free and clot-bound factor Xa that do not require antithrombin to exert their anticoagulant effect.2 Both drugs have good oral bioavailability (80% and 50% for rivaroxaban and apixaban, respectively) and reach peak plasma concentrations in about 3 hours. The terminal half-life is 5 to 9 hours for rivaroxaban and 9 to 14 hours for apixaban.2 Both drugs are excreted in the urine and feces and are metabolized by CYP3A4 (among others) and cytochrome P450 (CYP)–independent mechanisms. Rivaroxaban is a substrate of CYP3A4/5, CYP2J2, and the P-gp and ATP-binding cassette G2 (ABCG2) transporters. Inhibitors and inducers of these CYP450 enzymes or transporters (e.g., P-gp) may result in changes in rivaroxaban exposure.59 Both drugs should be used with caution in patients with renal dysfunction.2

Dosing and Administration

The dose of rivaroxaban for prevention of VTE following elective hip or knee replacement surgery is 10 mg orally once daily with or without food. Rivaroxaban should be initiated at least 6 to 10 hours after surgery once hemostasis has been established and continued for 12 days (knee replacement) or 35 days (hip replacement).59 The dose used in clinical trials for VTE treatment was 15 mg orally twice daily for 3 weeks, and then 20 mg once daily thereafter.37,38 When rivaroxaban doses of 20 mg are taken by patients with atrial fibrillation, administration with food is recommended.59 Apixaban is not approved for use in the United States, but European prescribing information lists the recommended dose following elective hip or knee replacement surgery at 2.5 mg orally twice daily with or without food starting 12 to 24 hours after surgery and continuing for 10 to 14 days (knee replacement) or 32 to 38 days (hip replacement).57

Therapeutic Monitoring

Because of predictable pharmacokinetics, rivaroxaban and apixaban do not require routine laboratory monitoring or dose adjustments. In rare situations (e.g., overdose, bleeding, assessment of compliance, prior to an invasive procedure, drug interactions, or assessment of drug accumulation in renal or hepatic impairment) the ability to measure coagulation using a quantitative assay might be valuable.2 However, there are currently no validated laboratory tests that can be recommended to monitor rivaroxaban or apixaban or any recommendations for dose adjustments based on observed test results.2 Periodic assessment of renal function is important during long-term therapy with rivaroxaban or apixaban, especially for patients with creatinine clearance <50 mL/min (<0.83 mL/s). These drugs should not be used in patients with creatinine clearance less than 15 mL/min (0.25 mL/s) (apixaban) or 30 mL/min (0.5 mL/s) (rivaroxaban).57,59 Patients should be observed closely and promptly evaluated for any signs or symptoms of blood loss.

Adverse Effects

Bleeding is the most common adverse effect associated with rivaroxaban and apixaban therapy.57,59 Managing bleeding associated with rivaroxaban and apixaban is complicated by the lack of a specific reversal agent.2 Patients presenting with significant bleeding during rivaroxaban or apixaban therapy should receive routine usual supportive care (fluid resuscitation, blood transfusion, maintenance of renal function, bleeding source identification, and surgical intervention if needed), and discontinuation of anticoagulation therapy.70 Because rivaroxaban and apixaban have relatively short half-lives, these measures may control bleeding in many patients, especially those with normal renal function.70 Activated charcoal may provide some benefit if drug intake occurred within a couple of hours of presentation, but hemodialysis is not expected to be effective for rivaroxaban and apixaban as they are both highly protein bound; fresh-frozen plasma (FFP) is also unlikely to provide clinical benefit.70 It is unclear whether factor VIIa administration will be useful for emergent reversal of rivaroxaban or apixaban and its use has been associated with increased risk of arterial thrombosis in nonhemophiliac patients.70 In dire circumstances, the use of nonactivated four-factor prothrombin complex concentrates (PCCs) may be reasonable based on a small study in normal volunteers receiving rivaroxaban.70 However, it is important to note that there have been no studies evaluating the effect of PCCs in bleeding humans receiving rivaroxaban or apixaban, and four-factor PCCS are currently unavailable in the United States.70The most frequent nonbleeding adverse events in clinical trials of rivaroxaban and apixaban were nausea, vomiting, and constipation.2

Use in Special Populations

There are no adequate or well-controlled studies of rivaroxaban or apixaban in pregnant women, and use is not recommended in this patient population. It is not known if rivaroxaban is excreted in human milk and breast-feeding is not recommended in women taking either rivaroxaban or apixaban.57,59 Oral administration and no need for routine coagulation monitoring make rivaroxaban and apixaban attractive alternatives in pediatric patients; however, safety and effectiveness in this population have not been established.57,59


Dabigatran is a selective, reversible, direct thrombin inhibitor given as an orally absorbable prodrug dabigatran etexilate. Dosing schedules are 150 and 220 mg once daily when used to prevent VTE (starting with a half dose given soon after surgery) and 150 mg twice daily with therapeutic LMWH for the first 5 days for treatment of VTE.2 Available data suggest that dabigatran is at least as effective and safe as LMWH in the prevention of VTE after major orthopedic surgery and it has been approved in Canada and Europe for VTE prevention after hip and knee replacement surgery, but is not yet approved for this indication in the United States, although it is on the market for stroke prevention in patients with nonvalvular atrial fibrillation.58 The need for concurrent administration with LMWH at the initiation makes dabigatran a less attractive alternative compared with rivaroxaban for treatment of VTE.


Warfarin is currently the anticoagulant of choice for long-term or extended anticoagulation. Because of its narrow therapeutic index, predisposition to drug and food interactions, and propensity to cause bleeding, warfarin requires continuous patient monitoring and education to achieve optimal outcomes.2


Warfarin exerts its anticoagulation effect by inhibiting the enzymes responsible for the cyclic interconversion of vitamin K in the liver.2 Reduced vitamin K is a cofactor required for the carboxylation of the vitamin K–dependent coagulation proteins, namely, prothrombin, VII, IX, and X, as well as the endogenous anticoagulant proteins C and S. Carboxylation of the N-terminal region of these proteins in the liver is required for biologic activity. By inhibiting the supply of vitamin K to serve as a cofactor in the production of these proteins, warfarin results in the production of partially carboxylated and decarboxylated coagulation proteins with reduced activity.2 Warfarin has no direct effect on previously circulating clotting factors or previously formed thrombus. The time required for warfarin to achieve its pharmacologic effect is dependent on the elimination half-lives of the coagulation proteins that vary between 4 and 6 hours for factor VII and 42 and 72 hours for prothrombin.2 Given that prothrombin has a 2- to 3-day half-life, antithrombotic effect is not achieved for at least 6 days after the initiation of warfarin therapy. By suppressing the production of fully functional clotting factors, warfarin prevents the initial formation and propagation of thrombus.2


Commercially available warfarin is a racemic mixture of R and S isomers. The S isomer is 2.7 to 3.8 times more potent than the R isomer.2 Warfarin is rapidly and extensively absorbed from the GI tract and reaches peak plasma concentration within 4 hours with a bioavailability of greater than 90% following oral administration. Warfarin is 99% bound to plasma proteins.71 It undergoes stereoselective metabolism via CYP1A2, 2C9, 2C19, 2C8, 2C18, and 3A4 isoenzymes in the liver, with 2C9 being the main enzyme to modulate in vivo anticoagulant activity.71 Pharmacokinetic parameters of warfarin, particularly hepatic metabolism, vary substantially between individuals leading to large interpatient differences in dose requirements. Genetic variations in the 2C9 isoenzyme and vitamin K epoxide reductase (VKOR) have been shown to correlate with warfarin dose requirements.2 Given the relatively greater potency of S-warfarin, coadministration of drugs that induce or inhibit the CYP2C isoenzymes is more likely to cause clinically significant interactions.2 These and other pharmacokinetic variations in warfarin metabolism likely explain the large interpatient dose–response seen with warfarin in clinical practice.

Dosing and Administration

The dose of warfarin is patient specific based on the desired intensity of anticoagulation and the patient’s individual response.2 There is tremendous interpatient variability with regard to the pharmacodynamic response and pharmacokinetic disposition of warfarin. In addition, there can be significant intrapatient variability in these parameters over time. Therefore, the dose of warfarin must be based on continual clinical and laboratory monitoring.2

Although the average weekly dose of warfarin is between 25 and 55 mg, some patient-related variables are associated with lower than usual dose requirement including advanced age (>65 years), elevated baseline INR, poor nutritional status, liver disease, hyperthyroidism, genetic polymorphisms in CYP2C9 and VKOR (see Personalized Pharmacotherapy below), and concurrent use of medications known to enhance the effect of warfarin.2 Prior to initiating therapy, the clinician should screen for the presence of contraindications to anticoagulation therapy and risk factors for major bleeding (see Table 9-14). It is important to collect a complete medication history, including the use of herbal and nutritional products (Tables 9-16 and 9-17).

TABLE 9-16 Warfarin Dietary Supplements Interactions Involving Cytochrome P450 Metabolism


TABLE 9-17 Dietary Supplements that can Affect Platelet Function and Anticoagulation Status


For most patients, initiating therapy with 5 to 10 mg daily and adjusting the dose based on the INR response will produce therapeutic INRs in 4 to 5 days (Fig. 9-6). Lower starting doses may be acceptable based on patient-related factors such as advanced age, malnutrition, liver disease, or heart failure. Starting doses >10 mg should be avoided.2 Warfarin therapy can be safely initiated on an outpatient basis; the response to therapy should be measured every 1 to 3 days until stabilized. For patients with acute venous thrombosis, UFH, LMWH, or fondaparinux should be overlapped with warfarin therapy for at least 5 days regardless of whether the target INR has been achieved earlier.2,7

When adjusting the dose of warfarin, allow sufficient time for changes in the INR to occur. In general, maintenance dose changes should not be made more frequently than every 3 days. Doses should be adjusted by reducing or increasing the weekly dose by 5% to 25%. The full effect of dose changes may not become evident for 5 to 7 days. During maintenance therapy, patients should not have followup INR tests sooner than anticipated changes are likely to occur.2

Therapeutic Monitoring

Warfarin requires frequent laboratory monitoring to ensure optimal therapeutic outcomes and minimize bleeding complications. The PT has been used for decades to monitor the anticoagulation effects of warfarin. The PT measures the biologic activity of factors II, VII, and X and correlates well to warfarin’s anticoagulation effect. The test is performed by measuring the time required for clot formation after adding calcium and thromboplastin to citrated plasma.2 The PT is problematic to interpret because of the variable sensitivity of commercially available thromboplastin reagents. With a given blood sample, thromboplastins of differing sensitivity will produce substantially different results some of which could lead to inappropriate dosing decisions. The World Health Organization (WHO) addressed the need for standardization in the late 1970s by developing a reference thromboplastin and recommending the use of the INR to monitor warfarin therapy. The INR corrects for differences in thromboplastin reagents through the following formula:


The International Sensitivity Index (ISI) is a measure of thromboplastin responsiveness compared with the WHO reference standard.2 Each thromboplastin reagent manufactured has an ISI value that should be used to calculate the INR. Although the INR system has a number of potential problems, it is currently the best means available to interpret the PT and the preferred method for monitoring warfarin therapy.2

The recommended target INR and associated goal range is based on the therapeutic indication. For most indications including treatment of VTE, the target INR is 2.5 with an acceptable range of 2 to 3.2 A baseline INR and complete blood cell count should be obtained prior to initiating warfarin therapy. In patients with an acute thromboembolic event, an INR should be measured minimally every 3 days during the first week of therapy (daily INRs are common in hospitalized patients). Once the patient’s dose–response is established, an INR should be determined every 7 to 14 days until it stabilizes and optimally every 4 to 12 weeks thereafter.44,73

At each encounter and especially when the INR is not in range, patients on warfarin therapy should be meticulously questioned regarding their medication use and symptoms related to bleeding and thromboembolic complications. Any changes in medications, including changes in dose as well as nonprescription drug and dietary supplement use, should be carefully explored (see Tables 9-16 and 9-17). If the INR is outside the therapeutic range, dietary intake of vitamin K–rich foods should also be evaluated (Table 9-18).

TABLE 9-18 Vitamin K Content of Select Foodsa


Anticoagulation therapy management services can optimize the care of patients who take warfarin therapy by providing structured care, comprehensive patient education, and evaluation of outcomes.75 When anticoagulation management services are not available, individual clinicians should strive to implement a similar structured care process.44

Portable finger-stick INR devices are available for monitoring warfarin therapy. These devices permit clinicians to do “real-time” therapeutic drug monitoring, and enable patients to engage in self-testing and/or management at home.44 Self-monitoring, in its simplest form, requires the patient to report his or her test results to a healthcare professional who continues to make warfarin dosing decisions. Highly motivated and sophisticated patients can be trained to manage themselves, independently altering the dose of warfarin therapy based on their INR results. Patients who engage in INR self-monitoring and warfarin self-management report high levels of satisfaction with care and maintain the INR within the therapeutic range slightly more frequently than those managed by “usual care.”44 However, home INR testing and self-management is clearly not for everyone; these modalities require careful patient selection and considerable patient education.44 Finger-stick INR devices are relatively expensive, but some patients may qualify for limited coverage of the monitor and testing strips.

Adverse Effects

Warfarin’s primary adverse effect is bleeding ranging from mild to life-threatening and can occur at any site in the body.2 Although warfarin is not believed to cause bleeding per se, it can “unmask” bleeding from an existing lesion or enable massive bleeding from an ordinarily minor source. The GI tract and the nose are among the most frequent sites of bleeding, and intracranial hemorrhage is the most serious and feared complication related to warfarin therapy, often resulting in permanent disability or death.2

There are no universally accepted criteria for defining a bleeding event as major or minor. The International Society for Thrombosis and Haemostasis defines major bleeding as fatal bleeding, any bleeding into a critical anatomic space (e.g., intracranial bleeding, hemarthrosis, or intraocular bleeding), bleeding that requires transfusion of 2 or more units of whole blood or red cells, or bleeding that leads to a greater than 2 g/dL (20 g/L; 1.2 mmol/L) drop in hemoglobin concentration.76 Bleeding that does not meet the criteria for a major hemorrhage is generally considered to be a minor. Minor bleeding is common during warfarin therapy even in the most expertly managed patients.

Several risk factors for bleeding while taking anticoagulation therapy have been identified (see Table 9-14).63 Intensity of anticoagulation therapy appears to be the most powerful risk factor with the likelihood of bleeding rising steeply as the INR increases above 5.2 The risk of hemorrhage is greatest during the first few weeks of therapy; however, bleeding can occur at any time and the cumulative incidence steadily increases over time.2

Other adverse effects associated with warfarin are uncommon, but can be serious.2 The “purple toe syndrome,” manifested as a purplish discoloration of the toes, is reported in a very small percentage of patients receiving warfarin. The etiology of this unusual phenomenon is unknown, but is thought to be the result of cholesterol microembolization into the arterial circulation of the toes.2 Warfarin-induced skin necrosis is an uncommon but very serious dermatologic reaction that usually manifests in the first week of therapy as a painful maculopapular rash and ecchymosis or purpura that subsequently progresses to necrotic gangrene. It most frequently appears in areas of the body rich in SC fat, such as the breasts, thighs, buttocks, and abdomen.2 The pathogenesis of warfarin-induced skin necrosis is not clearly understood, but patients with protein C or S deficiency appear to be at greater risk.2 If the diagnosis of skin necrosis is suspected, warfarin therapy should be immediately discontinued, FFP and vitamin K administered, and full-dose UFH or LMWH therapy initiated. Restart warfarin therapy with extreme caution if at all in patients with a history of skin necrosis using small doses and gradual titration under the coverage of full-dose UFH or LMWH until therapeutic INR has been achieved.2

Management of Bleeding and Excessive Anticoagulation

The AT9 offer suggestions for the management of patients with an elevated INR.44 When the INR is >4.5 without evidence of bleeding, the INR can be lowered by withholding warfarin, adjusting the dose of warfarin, and providing some dose of vitamin K to shorten the time to return to normal INR. Although vitamin K can be administered parenterally or orally, in the absence of major bleeding, the oral route is preferred. If the INR is between 4.5 and 10 and no bleeding is present, AT9 suggest against the routine use of vitamin K as it has not been shown to affect the risk of developing subsequent bleeding or thromboembolism compared with simply withholding warfarin alone. For INRs >10 without evidence of bleeding 2.5 mg of oral vitamin K is suggested.44 Vitamin K should be used with caution in patients at high risk of recurrent thromboembolism because of the possibility of INR overcorrection. Conversely, simply withholding warfarin therapy may not lower a high INR quickly enough in patients at high risk for developing bleeding complications. Most patients with asymptomatic INR elevations can be safely managed by withholding warfarin alone. Patients with warfarin-associated major bleeding require supportive care as described previously, and in addition the AT9 suggest rapid reversal of anticoagulation with four-factor PCCs rather than FFP, but acknowledge that four-factor PCCs are currently not available in the United States. The guidelines further suggest that in addition to repletion of coagulation factors (via PCCs or FFP), 5 to 10 mg of vitamin K should be administered via slow IV injection.44

Drug–Drug and Drug–Food Interactions

The pharmacokinetic and pharmacodynamic properties of warfarin, coupled with its narrow therapeutic index, predispose this agent to numerous clinically important food and drug interactions.77 Vitamin K can reverse warfarin’s pharmacologic activity, and many foods contain sufficient vitamin K to reduce the anticoagulation effect of warfarin if a patient consumes them in large portions or repetitively within a short period of time.2 Patients should be instructed to maintain a relatively consistent intake of vitamin K–rich foods. It is important to stress consistency rather than abstinence as patients with the highest daily intake of vitamin K have been shown to have more stable INR control.78

Pharmacokinetic drug interactions with warfarin are primarily a result of alterations in hepatic metabolism. Drugs that inhibit or induce the CYP2C9, CYP1A2, and CYP3A4 isoenzymes have the greatest potential to significantly alter the response to warfarin therapy.2 Drugs that alter hemostasis, platelet function, or the clearance of clotting factors (e.g., thyroid hormone replacement) can alter the response to warfarin therapy or increase the risk of bleeding by pharmacodynamic mechanisms.77 Clinicians should advise patients on warfarin therapy to seek information about potential interactions with warfarin whenever they start to take a new drug product, dietary supplement, or herbal product, whether it is prescribed or nonprescribed. If there is a known drug interaction or doubt about its potential to alter the response to warfarin, more frequent INR testing following the initiation of the new agent is prudent.77

Use in Special Populations

Patients scheduled to undergo invasive procedures often require temporary discontinuation of warfarin therapy.79 The decision to withhold warfarin therapy should be based on the type of surgical procedure being performed and the patient’s risk of bleeding and thromboembolism. Warfarin therapy should generally not be discontinued in patients undergoing minimally invasive procedures such as dental work, cataract surgery, or minor dermatologic procedures.79 If the bleeding risk from the procedure is considerable, warfarin should be stopped 4 to 5 days prior to the procedure in order to allow the INR to return to near-normal values. Patients at high risk of thromboembolism (i.e., DVT or PE in the previous month) should be given so-called bridge therapy with UFH or a LMWH before and/or after the procedure (Table 9-19).79

TABLE 9-19 General Approach to Periprocedural Anticoagulation Therapy Management



Warfarin Pharmacogenomics

At the initiation of warfarin therapy, it is difficult to predict the specific dose that an individual will require. Warfarin dosing algorithms that incorporate pharmacogenetic information regarding CYP2C9 and VKOR polymorphisms are currently being evaluated. However, the clinical utility and cost-effectiveness of using pharmacogenetic information to guide warfarin initiation remains unproven and AT9 suggest against routine use.44


Images The appropriate initial duration of warfarin therapy to effectively treat an acute first episode of VTE for all patients is 3 months.7 Three months of therapeutic anticoagulation therapy reduces the risk of recurrent VTE to as low as can be achieved by a time-limited duration of therapy. To prevent new episodes of VTE that are not directly related to the preceding episode, continued warfarin therapy is required.1Individually tailoring warfarin maintenance duration therapy past 3 months requires careful consideration of the circumstances surrounding the initial thromboembolic event, the presence of ongoing thromboembolic risk factors and the risk of bleeding, and patient preference.7

The most important considerations in determining the risk of recurrent VTE once anticoagulation therapy is stopped is whether the initial thrombotic event was associated with a major transient or reversible risk factor (e.g., surgery, plaster cast immobilization of a leg, or hospitalization in the month prior to VTE) and the presence of active cancer.7 The estimated cumulative risk of recurrent VTE after stopping anticoagulant therapy for VTE provoked by surgery is 1% after 1 year and 3% after 5 years, and that for VTE provoked by a nonsurgical reversible risk factor is 5% after 1 year and 15% after 5 years. Three months of anticoagulation therapy is recommended in these situations.7 Patients with a first unprovoked (idiopathic) VTE have a recurrence risk of approximately 10% in the first year and approximately 30% and 50% over 5 and 10 years, respectively. These patients should be considered for extended warfarin therapy when feasible.7 Extended in this context refers to warfarin that is continued beyond 3 months without a scheduled stop date, but which may be stopped because of a subsequent increase in the risk of bleeding or change in patient preference for anticoagulation.7 Anticoagulant treatment is rarely stopped in patients with VTE and active cancer because of a high risk for recurrence.7 Factors that may lead to the decision to stop warfarin therapy after 3 months include noncompliance with therapy, initial clot even though idiopathic was isolated in calf veins, or a moderate to high risk of bleeding.7

Clinically important risk factors for bleeding include age >75 years, previous noncardioembolic stroke, history of GI bleeding, renal or hepatic impairment, anemia, thrombocytopenia, concurrent antiplatelet use (avoid if possible), noncompliance, poor anticoagulant control, serious acute or chronic illness, and the presence of a structural lesion (e.g., tumor, recent surgery) expected to be associated with bleeding. Presence of one to two bleeding risk factors suggests moderate bleeding risk while three or more risk factors suggest a high bleeding risk.80 For patients with second episode of idiopathic VTE, extended anticoagulation is recommended.7

Various secondary strategies aimed at identifying patients at very low risk of recurrence after a first idiopathic VTE are being evaluated in clinical trials. Safe withdrawal of warfarin therapy may be possible if reliable identification of these patients proves possible. Estimating an individual patient’s risk of recurrent VTE using a variety of interacting clinical, laboratory, and radiologic findings can be accomplished with increasing precision.53 Some factors that may predict lower risk of recurrence include female gender, low D-dimer levels 1 month after stopping warfarin therapy, absence of residual clot on ultrasound, absence of hereditary and acquired thrombophilia, and absence of the postthrombotic syndrome. Risk assessment derived from combining several independent risk factors for recurrence has also been investigated.7 Further validation is needed before any one factor or prediction rule using a combination of factors can justify routinely stopping warfarin after 3 months of therapy. The decision to continue extended warfarin therapy should be reassessed periodically. Patients should be involved in any decision to continue anticoagulation therapy with consideration given to the patient’s long-term prognosis, risk of bleeding, ability to adhere to anticoagulation therapy instructions, financial resources, lifestyle, and quality of life.7 Although warfarin targeted to an INR of 2 to 3 is very effective at preventing recurrence while patients are receiving therapy, when it is stopped, there is a similar risk of recurrence whether patients have been treated for 3 months or longer. The same is true for other anticoagulants used for extended therapy (e.g., LMWH, rivaroxaban).7

Increasingly, patients with VTE are being tested for hereditary and acquired hypercoagulable states (thrombophilia). The available evidence does not support an association between genetically transmitted thrombophilia and a higher chance of recurrent VTE.13 Despite a lack of convincing evidence, most experts recommend indefinite anticoagulation for individuals with the antiphospholipid antibody syndrome, homozygotes for the factor V Leiden mutation, and heterozygotes with both the factor V Leiden and prothrombin 20210 gene mutations.


Even though several clinical interventions are known to be effective in preventing and treating VTE, adherence with various consensus guidelines regarding thromboprophylaxis remains alarmingly low.81Although preventing VTE is a significant patient safety issue, there is little public awareness of the life-threatening nature of DVT and PE. A survey conducted on behalf of the American Public Health Association suggests that 75% of Americans have little or no awareness of DVT, and less than one half of respondents could identify any risk factors associated with its development.82 Recognizing the lack of public awareness, several organizations have focused on increasing consumer knowledge of the risks, signs, and symptoms of VTE through increased media visibility.

Given the number and variety of clinical conditions or circumstances that place individuals at risk for VTE, improvements in VTE prevention and care have the potential to benefit many patients. Over the past decade, the focus on quality healthcare has been emphasized by the call to accountability through the Joint Commission’s Agenda for Change, the Institute of Medicine’s report on medical errors, the National Quality Forum’s (NQF’s) endorsed safe practices, the Leapfrog Group, and the demand for value by healthcare consumers (Table 9-20).83,84 The NQF has developed national consensus standards for VTE prevention and treatment that will be applicable to a variety of healthcare settings.84 The outcomes of this effort will provide a framework for measuring the effective screening, prevention, and treatment of VTE. NQF’s recommendations include developing organizational policies that address staff education, treatment protocols, and adherence measurements to improve VTE prevention in the hospital. The ultimate goal of the NQF consensus standards is to facilitate early promulgation of VTE policies, risk assessment, prophylaxis, diagnosis, and treatment services as well as patient education and organizational accountability. To that end, the Joint Commission has developed performance measures to enforce the NQF’s recommendations.83 Four major domains have been identified: risk assessment, prevention, diagnosis, and treatment. Six measures have been selected for implementation as core VTE quality measures. It is expected that compliance and reporting on these measures eventually will be tied to payment from governmental entities such as Medicare and Medicaid (Table 9-21).

TABLE 9-20 Organizations Monitoring Quality Care


TABLE 9-21 The Joint Commission’s Proposed Performance Measures for the Prevention and Treatment of Venous Thrombosis


Hopefully, through the concerted efforts of government and accrediting agencies working with hospitals and other healthcare institutions, the incidence of DVT and PE will begin to fall. Systematic approaches to this problem are needed at every level, starting with increased public and health practitioner awareness, continuing with the uniform use of effective prophylactic strategies in patients at risk, and concluding with greater accountability with precise quality measurements.





    1. Goldhaber SZ, Bounameaux H. Pulmonary embolism and deep vein thrombosis. Lancet 2012;379(9828): 1835–1846.

    2. Ageno W, Gallus AS, Wittkowsky A, Crowther M, Hylek EM, Palareti G. Oral anticoagulant therapy: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141(2 Suppl):e44S–e88S.

    3. Garcia DA, Baglin TP, Weitz JI, Samama MM. Parenteral anticoagulants: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141(2 Suppl):e24S–e43S.

    4. Falck-Ytter Y, Francis CW, Johanson NA, et al. Prevention of VTE in orthopedic surgery patients: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141(2 Suppl):e278S–e325S.

    5. Gould MK, Garcia DA, Wren SM, et al. Prevention of VTE in nonorthopedic surgical patients: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141(2 Suppl):e227S–e277S.

    6. Kahn SR, Lim W, Dunn AS, et al. Prevention of VTE in nonsurgical patients: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141(2 Suppl):e195S–e226S.

    7. Kearon C, Akl EA, Comerota AJ, et al. Antithrombotic therapy for VTE disease: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141(2 Suppl):e419S–e494S.

    8. Reitsma PH, Versteeg HH, Middeldorp S. Mechanistic view of risk factors for venous thromboembolism. Arterioscler Thromb Vasc Biol 2012;32(3):563–568.

    9. Turpie AG, Esmon C. Venous and arterial thrombosis—Pathogenesis and the rationale for anticoagulation. Thromb Haemost 2011;105(4):586–596.

   10. Margaglione M, Grandone E. Population genetics of venous thromboembolism. A narrative review. Thromb Haemost 2011;105(2):221–231.

   11. Anderson JA, Weitz JI. Hypercoagulable states. Crit Care Clin 2011;27(4):933–952, vii.

   12. Dalen JE. Should patients with venous thromboembolism be screened for thrombophilia? Am J Med 2008;121(6): 458–463.

   13. Rosendaal FR, Reitsma PH. Genetics of venous thrombosis. J Thromb Haemost 2009;7(Suppl 1):301–304.

   14. Khorana AA. Cancer and coagulation. Am J Hematol 2012;87(Suppl 1):S82–S87.

   15. Buller HR, van Doormaal FF, van Sluis GL, Kamphuisen PW. Cancer and thrombosis: From molecular mechanisms to clinical presentations. J Thromb Haemost 2007;5(Suppl 1): 246–254.

   16. Giannakopoulos B, Passam F, Rahgozar S, Krilis SA. Current concepts on the pathogenesis of the antiphospholipid syndrome. Blood 2007;109(2):422–430.

   17. Bates SM, Greer IA, Middeldorp S, Veenstra DL, Prabulos AM, Vandvik PO. VTE, thrombophilia, antithrombotic therapy, and pregnancy: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141(2 Suppl):e691S–e736S.

   18. Rosendaal FR, Helmerhorst FM, Vandenbroucke JP. Female hormones and thrombosis. Arterioscler Thromb Vasc Biol 2002;22(2):201–210.

   19. Crowther MA, Kelton JG. Congenital thrombophilic states associated with venous thrombosis: A qualitative overview and proposed classification system. Ann Intern Med 2003;138(2):128–134.

   20. Dahlback B. Advances in understanding pathogenic mechanisms of thrombophilic disorders. Blood 2008;112(1):19–27.

   21. Furie B, Furie BC. Thrombus formation in vivo. J Clin Invest 2005;115(12):3355–3362.

   22. Furie B, Furie BC. Mechanisms of thrombus formation. N Engl J Med 2008;359(9):938–949.

   23. Crawley JT, Zanardelli S, Chion CK, Lane DA. The central role of thrombin in hemostasis. J Thromb Haemost 2007;5(Suppl 1):95–101.

   24. Monroe DM, Hoffman M. What does it take to make the perfect clot? Arterioscler Thromb Vasc Biol 2006;26(1): 41–48.

   25. Rijken DC, Lijnen HR. New insights into the molecular mechanisms of the fibrinolytic system. J Thromb Haemost 2009;7(1):4–13.

   26. Kearon C. Natural history of venous thromboembolism. Circulation 2003;107(23 Suppl 1):I22–I30.

   27. Pruitt B, Lawson R. What you need to know about venous thromboembolism. Nursing 2009;39(4):22–27.

   28. Bates SM, Jaeschke R, Stevens SM, et al. Diagnosis of DVT: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141 (2 Suppl):e351S–e418S.

   29. Tapson VF. Acute pulmonary embolism. N Engl J Med 2008;358(10):1037–1052.

   30. Bauersachs RM. Clinical presentation of deep vein thrombosis and pulmonary embolism. Best Pract Res Clin Haematol 2012;25(3):243–251.

   31. Righini M, Perrier A, De Moerloose P, Bounameaux H. D-dimer for venous thromboembolism diagnosis: 20 years later. J Thromb Haemost 2008;6(7):1059–1071.

   32. Mahan CE, Holdsworth MT, Welch SM, Borrego M, Spyropoulos AC. Deep-vein thrombosis: A United States cost model for a preventable and costly adverse event. Thromb Haemost 2011;106(3):405–415.

   33. Guyatt GH, Eikelboom JW, Gould MK, et al. Approach to outcome measurement in the prevention of thrombosis in surgical and medical patients: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141(2 Suppl):e185S–e194S.

   34. Gray J, Razmus I. Improving venous thromboembolism prevention processes and outcomes at a community hospital. Jt Comm J Qual Patient Saf 2012;38(2):61–66.

   35. Schiro TA, Sakowski J, Romanelli RJ, et al. Improving adherence to best-practice guidelines for venous thromboembolism risk assessment and prevention. Am J Health Syst Pharm 2011;68(22):2184–2189.

   36. Streiff MB, Haut ER. The CMS ruling on venous thromboembolism after total knee or hip arthroplasty: Weighing risks and benefits. JAMA 2009;301(10): 1063–1065.

   37. Bauersachs R, Berkowitz SD, Brenner B, et al. Oral rivaroxaban for symptomatic venous thromboembolism. N Engl J Med 2010;363(26):2499–2510.

   38. Buller HR, Prins MH, Lensin AW, et al. Oral rivaroxaban for the treatment of symptomatic pulmonary embolism. N Engl J Med 2012;366(14):1287–1297.

   39. ASHP Therapeutic Position Statement on the Use of Low-Molecular-Weight Heparins for Adult Outpatient Treatment of Acute Deep-Vein Thrombosis. Am J Health Syst Pharm 2004;61(18):1950–1955.

   40. Anderson RC, Bussey HI. Retrievable and permanent inferior vena cava filters: Selected considerations. Pharmacotherapy 2006;26(11):1595–1600.

   41. Geerts WH, Bergqvist D, Pineo GF, et al. Prevention of venous thromboembolism: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008;133(6 Suppl):381S–453S.

   42. Anderson FA Jr, Huang W, Friedman RJ, Kwong LM, Lieberman JR, Pellegrini VD Jr. Prevention of venous thromboembolism after hip or knee arthroplasty: Findings from a 2008 survey of US orthopedic surgeons. J Arthroplasty 2012;27(5):659–666.

   43. Mont MA, Jacobs JJ. AAOS clinical practice guideline: Preventing venous thromboembolic disease in patients undergoing elective hip and knee arthroplasty. J Am Acad Orthop Surg 2011;19(12):777–778.

   44. Holbrook A, Schulman S, Witt DM, et al. Evidence-based management of anticoagulant therapy: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141(2 Suppl):e152S–e184S.

   45. Matchar DB, Mark DB. Strategies for incorporating resource allocation and economic considerations: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008;133(6 Suppl): 132S–140S.

   46. Kapoor A, Chuang W, Radhakrishnan N, et al. Cost effectiveness of venous thromboembolism pharmacological prophylaxis in total hip and knee replacement: A systematic review. Pharmacoeconomics 2010;28(7):521–538.

   47. Mamdani MM, Racine E, McCreadie S, et al. Clinical and economic effectiveness of an inpatient anticoagulation service. Pharmacotherapy 1999;19(9):1064–1074.

   48. Kearon C, Ginsberg JS, Julian JA, et al. Comparison of fixed-dose weight-adjusted unfractionated heparin and low-molecular-weight heparin for acute treatment of venous thromboembolism. JAMA 2006;296(8):935–942.

   49. Goldhaber SZ. Advanced treatment strategies for acute pulmonary embolism, including thrombolysis and embolectomy. J Thromb Haemost 2009;7(Suppl 1):322–327.

   50. Arixtra Prescribing Information. GlaxoSmithKline. September 2012,

   51. Tillman DJ, Charland SL, Witt DM. Effectiveness and economic impact associated with a program for outpatient management of acute deep vein thrombosis in a group model health maintenance organization. Arch Intern Med 2000;160(19):2926–2932.

   52. Gould MK, Dembitzer AD, Sanders GD, Garber AM. Low-molecular-weight heparins compared with unfractionated heparin for treatment of acute deep venous thrombosis. A cost-effectiveness analysis. Ann Intern Med 1999;130(10):789–799.

   53. Baglin T. What happens after venous thromboembolism? J Thromb Haemost 2009;7(Suppl 1):287–290.

   54. Aujesky D, Hughes R, Jimenez D. Short-term prognosis of pulmonary embolism. J Thromb Haemost 2009;7(Suppl 1): 318–321.

   55. Schulman S, Kearon C, Kakkar AK, et al. Dabigatran versus warfarin in the treatment of acute venous thromboembolism. N Engl J Med 2009;361(24):2342–2352.

   56. Ginsberg JS, Bates SM. Management of venous thromboembolism during pregnancy. J Thromb Haemost 2003;1(7):1435–1442.

   57. Eliquis Prescribing Information. Bristol-Myers Squibb/Pfizer. September 2012,

   58. Pradaxa Prescribing Information. Boehringer-Ingelheim. September 2012,

   59. Xarelto Prescribing Information. Janssen Pharmaceuticals Inc. September 2012,

   60. Stone SE, Morris TA. Pulmonary embolism and pregnancy. Crit Care Clin 2004;20(4):661–677, ix.

   61. Monagle P, Chan AK, Goldenberg NA, et al. Antithrombotic therapy in neonates and children: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141(2 Suppl):e737S–e801S.

   62. Delate T, Witt DM, Ritzwoller D, et al. Outpatient use of low molecular weight heparin monotherapy for first-line treatment of venous thromboembolism in advanced cancer. Oncologist 2012;17(3):419–427.

   63. Schulman S, Beyth RJ, Kearon C, Levine MN. Hemorrhagic complications of anticoagulant and thrombolytic treatment: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008;133 (6 Suppl):257S–298S.

   64. Clark NP. Low-molecular-weight heparin use in the obese, elderly, and in renal insufficiency. Thromb Res 2008;123(Suppl 1):S58–S61.

   65. Lovenox Prescribing Information. Sanofi-Aventis Pharmaceuticals. September 2012,

   66. Linkins LA, Dans AL, Moores LK, et al. Treatment and prevention of heparin-induced thrombocytopenia: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141(2 Suppl):e495S–e530S.

   67. Crowther MA, Cook DJ, Albert M, et al. The 4Ts scoring system for heparin-induced thrombocytopenia in medical-surgical intensive care unit patients. J Crit Care 2010;25(2):287–293.

   68. Cuker A, Gimotty PA, Crowther MA, Warkentin TE. Predictive value of the 4Ts scoring system for heparin-induced thrombocytopenia: A systematic review and meta-analysis. Blood 2012;120(5):4160–4167.

   69. Blackmer AB, Oertel MD, Valgus JM. Fondaparinux and the management of heparin-induced thrombocytopenia: The journey continues. Ann Pharmacother 2009;43(10): 1636–1646.

   70. Kaatz S, Kouides PA, Garcia DA, et al. Guidance on the emergent reversal of oral thrombin and factor Xa inhibitors. Am J Hematol 2012;87(Suppl 1):S141–S145.

   71. Coumadin Prescribing Information. Bristol-Myers Squibb. September 2012,

   72. Nutescu EA, Shapiro NL, Ibrahim S, West P. Warfarin and its interactions with foods, herbs and other dietary supplements. Expert Opin Drug Saf 2006;5(3):433–451.

   73. Rose AJ, Ozonoff A, Berlowitz DR, Ash AS, Reisman JI, Hylek EM. Reexamining the recommended follow-up interval after obtaining an in-range international normalized ratio value: Results from the Veterans Affairs study to improve anticoagulation. Chest 2011;140(2):359–365.

   74. Booth SL, Centurelli MA. Vitamin K: A practical guide to the dietary management of patients on warfarin. Nutr Rev 1999;57(9 Pt 1):288–296.

   75. Garcia DA, Witt DM, Hylek E, et al. Delivery of optimized anticoagulant therapy: Consensus statement from the Anticoagulation Forum. Ann Pharmacother 2008;42(7): 979–988.

   76. Seto AC, Kenyon K, Wittkowsky AK. Discrepancies in identification of major bleeding events in patients taking warfarin. Pharmacotherapy 2008;28(9):1098–1103.

   77. Holbrook AM, Pereira JA, Labiris R, et al. Systematic overview of warfarin and its drug and food interactions. Arch Intern Med 2005;165(10):1095–1106.

   78. Schurgers LJ, Shearer MJ, Hamulyak K, Stocklin E, Vermeer C. Effect of vitamin K intake on the stability of oral anticoagulant treatment: Dose–response relationships in healthy subjects. Blood 2004;104(9):2682–2689.

   79. Douketis JD, Spyropoulos AC, Spencer FA, et al. Perioperative management of antithrombotic therapy: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141 (2 Suppl): e326S–e350S.

   80. Kearon C. Balancing risks and benefits of extended anticoagulant therapy for idiopathic venous thrombosis. J Thromb Haemost 2009;7(Suppl 1):296–300.

   81. Herner SJ, Paulson DC, Delate T, Witt DM, Vondracek TG. Evaluation of venous thromboembolism risk following hospitalization. J Thromb Thrombolysis 2011;32(1):32–39.

   82. American Public Health Association. Deep-Vein Thrombosis: Advancing Awareness to Protect Patient Lives. White Paper, Public Health Leadership Conference on Deep-Vein Thrombosis. September 2009,

   83. The Joint Commission. National Consensus Standards on the Prevention and Care of Venous Thromboembolism. September 2009,

   84. The National Quality Forum. Establishing a Patient-Centered National Action Plan for Deep Vein Thrombosis Prevention, Treatment, and Research. Conference Proceedings. Washington, DC: National Quality Forum, 2007.