Michael B. Streiff
Venous Thromboembolism
Venous thromboembolism (VTE) is a common cause of morbidity and mortality in the United States, affecting one of every 1,000 Americans each year (1). The most common manifestations of VTE are deep venous thrombosis (DVT) and pulmonary embolism (PE). Two thirds of cases of VTE present as DVT alone; the remainder manifest as PE, with or without a demonstrable DVT (2). The 28-day case fatality rate for an initial DVT is 9.4% compared with 15% for PE (3). Therefore, adequate prevention and treatment of VTE are a priority for ambulatory care providers.
Risk Factors
Venous thrombosis occurs as a consequence of one or more of three different factors (commonly called the Virchow triad): stasis of blood flow, disruptions in the integrity of the vascular wall, and hypercoagulability of the blood. The incidence of VTE is strongly influenced by the presence or absence of risk factors that affect one or more of the Virchow triad. Age has a significant impact on the incidence of VTE. VTE is rare among children younger than 15 years (<5 episodes/100,000 per year). The incidence of VTE gradually increases during adult life up to age 59 years, after which there is a sharp increase from 280 episodes/100,000 per year to 1,800 episodes/100,000 per year among individuals 80 to 84 years old (1). Potential explanations for this dramatic difference in incidence may be the increased prevalence of malignancy and the decreased mobility among older individuals.
Ethnicity influences the risk of VTE. Asian Americans (19 events/100,000 per year) and Latinos (37 events/ 100,000 per year) have a lower incidence of VTE than Caucasians (86 events/100,000 per year) or African Americans (93 events/100,000 per year) (4). Potential reasons for these ethnic differences may be differences in the prevalence of genetic prothrombotic mutations such as factor V Leiden or the prothrombin gene G20210A mutation. Because these mutations are less common among African Americans than Caucasians, other thrombophilic mutations must be responsible for the higher incidence of VTE among African Americans (5). One report has suggested
P.846
that elevated levels of factor VIII may be a common risk factor for thrombosis among individuals of African descent (6).
|
TABLE 57.1 Prevalence of Inherited Thrombophilia |
|||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||
Unlike age and ethnicity, gender does not appear to significantly influence the incidence of VTE (3). However, a higher incidence of recurrent VTE has been noted among males in some studies (7).
Inherited prothrombotic conditions have an important influence on the risk of VTE. Surveys indicate that approximately one third of unselected patients presenting with a DVT or PE will have an inherited form of thrombophilia. Among selected patients (i.e., those who are younger than 50 years; patients who have a family history of VTE; individuals who have a history of recurrent VTE; or those who have idiopathic VTE), the prevalence of an inherited prothrombotic condition increases to 70%. Table 57.1 lists the frequencies of individual inherited prothrombotic conditions among different populations (8). The most common inherited thrombophilic condition is factor V Leiden. Factor V Leiden, a mutation in factor V that impairs the ability of activated protein C to inactivate activated forms of factor V and VIII, affects 5% of healthy Caucasians, 2% of Latinos, 1% of African Americans, and 0.5% of Asian Americans (5). Heterozygosity for factor V Leiden increases the risk of VTE fivefold, whereas homozygosity is associated with a 50-fold increase in risk. The presence of factor V Leiden can be identified using a screening assay, the activated protein C resistance assay, followed by confirmatory testing with polymerase chain reaction (PCR)-based genetic testing. The second most common inherited form of thrombophilia, the prothrombin gene G20210A mutation, is found in approximately 2% of European Americans. This mutation results in a 25% increase in prothrombin levels and a twofold increased risk for VTE. The prothrombin gene mutation can be identified only by PCR-based genetic testing (9).
Elevated levels of factors VIII, IX, and XI appear to increase the risk of VTE. Factor VIII levels >90th percentile are associated with at least a fivefold increased risk for initial and recurrent VTE (10,11). Factors IX and XI raise the risk of initial VTE twofold to threefold (12,13). Although the influence of elevated factor VIII levels on the risk of VTE has been identified in multiple independent studies, studies confirming the influence of elevated levels of factors IX and XI on VTE incidence have not been performed. Laboratory testing for coagulation factor levels is available in many clinical laboratories, but the absence of management trials using this information in therapeutic decision making and the sensitivity of test results to other factors (e.g., infection, inflammation, acute thrombosis) have reduced clinical enthusiasm for using factor level testing as a routine part of the thrombophilia evaluation. Table 57.2 provides information on factor level testing in the evaluation of thrombophilia.
Inherited deficiency of endogenous anticoagulants, such as antithrombin III, protein C, and protein S, are potent risk factors for VTE. A deficiency of antithrombin, which inactivates activated forms of factors II, X, IX, and XI, is present in one of every 5,000 in the general population and increases the risk of thrombosis 25-fold. Protein C inactivates activated forms of factors V and VIII in a complex with its cofactor, protein S. Protein C deficiency is present in one of every 500 in the population and is associated with a 10-fold increase in the risk of VTE. Protein S deficiency has a similar prevalence in the general population and similar potency as a risk factor for VTE (9). Dysfibrinogenemia is a rare inherited coagulation disorder (only 250 reported cases) that can predispose individuals to bleeding, thrombosis, or both, depending upon the site of the mutation in the fibrinogen gene. The thrombin time (usually prolonged, occasionally abnormally shortened) is a useful screening test (14). Although defects in the
P.847
fibrinolytic system and factor XII deficiency have been suggested to predispose to VTE, strong evidence supporting thrombophilia as a consequence of these abnormalities is lacking (15, 16, 17).
|
TABLE 57.2 Thrombophilia Testing Recommendations |
||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||
Elevated plasma levels of homocysteine are associated with an increased risk for venous and arterial thrombosis. Mild to moderate elevations of homocysteine are present in 5% to 7% of Americans. A common inherited cause of elevated homocysteine levels is a thermolabile mutation in N5, N10-methylene-tetrahydrofolate reductase, the enzyme responsible for maintaining adequate levels of N5-methyl-tetrahydrofolate, a cofactor necessary for the conversion of homocysteine into methionine. In the setting of folate deficiency, this abnormality can result in moderate elevations of homocysteine (15–30 µmol/L) and a twofold to threefold increased risk for venous or arterial thrombosis. Homocysteine is also metabolized by cystathionine β-synthase, which converts cystathionine into cysteine. Homozygous mutations in this enzyme result in homocystinuria, a rare disorder (one in 200,000 live births) associated with severe elevations of homocysteine (as high as 400 µmol/L) and premature atherosclerosis and recurrent VTE (18). Because folate, vitamin B12, and pyridoxine (vitamin B6) are cofactors required for homocysteine metabolism, deficiencies of these vitamins should be investigated in any patient with elevated homocysteine levels. Supplementation with folate, vitamin B12, and pyridoxine (vitamin B6) can result in reductions in homocysteine, although whether vitamin supplementation reduces the risk of VTE remains to be demonstrated.
The antiphospholipid antibody syndrome (APS) is an important acquired cause of thrombophilia that increases the risk of both venous and arterial thrombosis. Other clinical manifestations of APS include recurrent miscarriages, immune thrombocytopenia or hemolytic anemia, livedo reticularis, and nonbacterial thrombotic endocarditis. APS can present as a primary autoimmune syndrome or in association with rheumatologic disorders (systemic lupus erythematosus [SLE], rheumatoid arthritis, etc.), infections (syphilis, human immunodeficiency virus, etc.), cancers, or exposure to certain medications. Antiphospholipid antibodies have been identified in 1% to 5% of asymptomatic control subjects and in as many as 30% of SLE patients (19,20).
APS is caused by antibodies directed against antiphospholipid binding proteins, such as β2-glycoprotein I, annexin V, protein C, protein S, and prothrombin. APS may result in thrombosis by disrupting the anticoagulant function of these proteins, inducing endothelial damage and activation or increasing tissue factor expression by monocytes and/or endothelial cells (19,20). In patients with SLE, APS is associated with an incidence of thromboembolism of two events per 100 person-years (21). APS also increases the risk of recurrent thromboembolism (22).
Diagnostic tests for APS include phospholipid-dependent coagulation assays (e.g., dilute activated partial thromboplastin time [aPTT], dilute prothrombin time [PT], dilute Russell viper venom time, and mixing studies with these assays) and anticardiolipin and β2-glycoprotein I antibody assays. To meet criteria for APS, patients must
P.848
have documented clinical events (thrombosis or recurrent miscarriage) and repeatedly positive laboratory tests separated by at least 6 to 8 weeks (19,20).
The clinical utility of diagnostic testing for thrombophilia remains a subject of controversy and of active investigation (23). Possible benefits of testing include identification of the potential etiology of a thrombotic event and increased attention to DVT prophylaxis during future risk periods. However, only selected thrombophilic conditions (e.g., APS, factor V Leiden homozygosity, compound heterozygosity for factor V Leiden and the prothrombin gene mutation, elevated factor VIII levels) have been demonstrated in clinical studies to increase the risk of recurrent VTE and thus potentially to influence the duration of anticoagulant therapy (11,22,24). Furthermore, some studies have not identified an increased risk for recurrent VTE associated with inherited thrombophilia (23). Therefore, clinicians should consider the benefits, risks, and cost of thrombophilia evaluation before testing. Some commonly proposed criteria for identifying high-risk patients, who are most likely to benefit from testing, include age younger than 50 years at the time of the initial event, a family history positive for VTE, recurrent episodes of VTE, idiopathic VTE or VTE in association with minimal provocation, thromboses in unusual locations (e.g., intra-abdominal, cerebral venous sinus thrombosis), and recurrent miscarriages (9). However, clinicians also may consider testing individuals with VTE who do not fulfill one of these criteria if the clinical scenario strongly suggests the presence of a thrombophilic disorder. Table 57.2 lists the recommendations regarding the timing of laboratory testing, the most appropriate assays, and factors that can affect the results of testing.
The Virchow triad is influenced by acquired intrinsic or extrinsic thrombotic stimuli. Major acquired risk factors for VTE (odds ratio for thrombosis >10) include hip or leg fractures, hip or knee replacement surgery, major general surgery, major trauma, and spinal cord injury. Moderate risk factors (odds ratio 2–10) include cancer, chemotherapy, hormone replacement or oral contraceptive therapy, central venous catheters, congestive heart failure, respiratory failure, paralytic stroke, and pregnancy. Weak clinical risk factors (odds ratio <2) include bed rest for >3 days, immobility, laparoscopic surgery, varicose veins, and obesity (25). Although the relative risk of developing VTE that is associated with many of the inherited prothrombotic conditions is greater than the relative risk of acquired thrombotic stimuli, the high prevalence of the latter group of conditions makes their overall impact as risk factors for VTE far greater than that of the inherited disorders.
Diagnosis
The diagnosis of VTE requires consideration of a patient's risk factors for VTE, the symptoms and physical findings, and the results of objective diagnostic testing. Common signs and symptoms of DVT are pain, swelling, and tenderness of the affected extremity. PE is commonly associated with dyspnea, chest pain (often pleuritic), tachypnea, tachycardia, and hypoxia. Because individual physical signs of DVT and PE have low specificity, clinical prediction rules incorporating risk factors and signs and symptoms of DVT and PE have been developed and can be useful guides for assessing the probability of disease (26,27) (Tables 57.3 and 57.4). However, these rules were developed based on data obtained from referred patients attending secondary care outpatient clinics. A prospective study in primary care patients found that clinical prediction rules may be less accurate in the primary care setting (28). Until
P.849
clinical prediction rules are developed specifically for primary care patients and additional studies are completed, DVT and PE diagnostic algorithms should retain objective radiologic imaging procedures (Figs. 57.1,57.2,57.3). The most common diagnostic test for DVT is venous duplex ultrasonography. The sensitivity and specificity of this test both are >95% in patients with symptomatic proximal DVT. However, the sensitivity and specificity of this test for distal DVT (e.g., calf vein DVT) both are only 70%. Contrast venography, once considered the diagnostic gold standard, is rarely used currently because of its invasive nature, the risks of phlebitis and contrast dye reactions, and the limited availability of the procedure.
|
TABLE 57.3 DVT Clinical Assessment Tool |
||||||||||||||||||||||
|
||||||||||||||||||||||
|
TABLE 57.4 Wells PE Clinical Assessment Tool |
||||||||||||||||||||||
|
||||||||||||||||||||||
|
FIGURE 57.1. Algorithm for diagnosis of patients with suspected deep venous thrombosis (DVT). Neg, negative; Pos, positive; US, ultrasound. (Adapted from Wells PS, Anderson DR, Ginsberg J. Assessment of deep vein thrombosis or pulmonary embolism by the combined use of clinical model and noninvasive diagnostic tests. Semin Thromb Hemost 2000;26:643.) |
|
FIGURE 57.2. Algorithm for diagnosis of patients with suspected pulmonary embolism (PE). CT, computed tomography; Neg, negative; Pos, positive; US, ultrasound. (Adapted from Wells PS, Anderson DR, Ginsberg J. Assessment of deep vein thrombosis or pulmonary embolism by the combined use of clinical model and noninvasive diagnostic tests. Semin Thromb Hemost 2000;26:643. ) |
|
FIGURE 57.3. Ventilation/perfusion scan algorithm for diagnosis of patients with suspected pulmonary embolism (PE). Low/mod, low to moderate clinical probability; PAgram, pulmonary angiogram; PTP, pretest clinical probability. US, ultrasound. (Adapted from Wells PS, Anderson DR, Ginsberg J. Assessment of deep vein thrombosis or pulmonary embolism by the combined use of clinical model and noninvasive diagnostic tests. Semin Thromb Hemost 2000;26:643. ) |
The most common imaging studies in PE diagnosis are ventilation–perfusion (V/Q) scanning and spiral computer tomography (CT) scanning. V/Q scans are most useful when the results are normal or indicate a high probability of PE. A normal V/Q scan rules out PE, whereas a high probability scan indicates a high likelihood of PE, particularly in patients with a high pretest probability. Unfortunately, more than half of V/Q scans are read as either low (16%) or intermediate probability (41%) (29). Additional studies are necessary to rule out disease in these instances. Helical CT scanning has become an increasingly popular diagnostic tool in the investigation of PE because it is rapidly performed and often identifies alternative diagnoses. The reported sensitivity of helical CT for PE varies greatly (66%–93%), depending upon imaging protocols and technology (30). Clinical outcome studies, however, suggest that subsequent episodes of VTE are infrequent in patients with negative helical CT scans (31). Multidetector row helical CT was demonstrated to have sensitivity of 100% and specificity of 89% in one study that
P.850
compared this test to digital subtraction pulmonary angiography (32). The greater sensitivity of newer-generation helical CT scanners is likely to make this test the imaging study of choice in PE diagnosis. Figures 57.2 and 57.3 show examples of PE diagnosis algorithms incorporating V/Q and CT scanning. The topic of V/Q and CT scanning in PE diagnosis is discussed in Chapter 59.
In recent years, a large number of studies have documented the utility of D-dimer testing in the diagnosis of VTE. D-Dimers are the cross-linked fragments of fibrin produced by plasmin degradation. When used in conjunction with clinical prediction models, D-dimer assays have been successfully used to exclude the diagnosis of DVT and to reduce the need for additional diagnostic testing (33). Because many D-dimer assays are available for clinical use, only sensitive D-dimer assays validated in the diagnosis of VTE should be used for this purpose. Published thresholds for positive results should be validated in the clinician's laboratory prior to use in diagnosis.
Treatment
Therapeutic alternatives for treatment of VTE include anticoagulation, thrombolytic therapy, and vena caval interruption.
Anticoagulation
Anticoagulation remains the therapeutic option of choice for the vast majority of patients. Either unfractionated heparin (UFH) or low-molecular-weight heparin (LMWH) is used in the acute therapy for VTE. Acute therapy for VTE should be administered for at least 5 to 7 days until therapeutic levels of warfarin are achieved as reflected by an international normalized ratio (INR) of 2.0 to 3.0. The INR is calculated by dividing the patient's PT by the laboratory control PT and raising this result by an exponent, the international sensitivity index (ISI). The ISI reflects the sensitivity of an individual laboratory's PT reagent to reductions in vitamin K-dependent coagulation factors. Because different laboratories may use different reagents to measure the PT, this correction factor helps to normalize PT results for reagent differences and thus reduces variation in PT results between different laboratories. From 7 to 10 days of acute heparin therapy are often given for extensive DVT or PE (34).
Heparin
Most patients with DVT and some with PE can be safely and effectively treated as outpatients using LMWH administered in subcutaneous weight-based doses once or twice daily (for doses, see Table 57.5). UFH ordinarily should not be prescribed to outpatients. Reasons to admit a patient with DVT/PE include the presence of an extensive clot burden at high risk for progression or requiring thrombolytic therapy, significant risk factors for bleeding (e.g., thrombocytopenia or recent surgery or trauma), presence of concomitant medical disorders requiring inpatient management, and medical noncompliance that precludes the safe administration of anticoagulation on an outpatient basis. Although PE previously has been considered an absolute indication for hospital admission, several studies have demonstrated the feasibility and safety of managing hemodynamically stable patients with PE as an outpatient.
P.851
Table 57.6 lists the criteria for outpatient VTE management.
|
TABLE 57.5 Low Molecular Weight Heparin and Fondaparinux Dosing for VTE |
||||||||||||||||||||
|
||||||||||||||||||||
|
TABLE 57.6 Criteria for Outpatient Management of VTE |
||||||||||||||||||||
|
||||||||||||||||||||
LMWH has predictable pharmacokinetics such that laboratory measurement to evaluate its effect is unnecessary in most patients. Situations in which monitoring can be useful include pregnancy, obesity (body weight >150 kg), and significant renal insufficiency (creatinine < 25 mL/ min). If monitoring is necessary, the aPTT is not a useful test because it is insensitive to LMWH. The anti–factor Xa assay, which determines the concentration of LMWH by virtue of its inhibitory activity on the activated form of factor X, is used to monitor LMWH therapy when necessary. Samples for anti–factor Xa assays should be drawn 4 hours after subcutaneous administration of LMWH. For twice daily doses of enoxaparin, dalteparin, or tinzaparin, the recommended therapeutic range is 0.6 to 1.0 IU/mL. A therapeutic range of 1.0 to 2.0 IU/mL should be used for once daily doses of enoxaparin, dalteparin, or tinzaparin (34,35). UFH is preferred for treatment of patients with significant renal insufficiency. However, if use of LMWH is necessary in a patient with kidney disease, enoxaparin is the LMWH of choice because Food and Drug Administration (FDA)-approved dosing guidelines for this agent are available (1 mg/kg by subcutaneous injection every 24 hours). When treating patients with body weights >150 kg, actual body weight (i.e., not ideal body weight) should be used for dose calculation. Dose capping in obese patients is to be avoided because this practice can result in subtherapeutic drug levels and increase the risk of recurrent thrombotic events.
Fondaparinux
Another alternative for acute therapy of VTE is fondaparinux, a synthetic indirect factor Xa inhibitor. Fondaparinux is administered in weight-based subcutaneous injection once daily (Table 57.5). In randomized clinical trials, fondaparinux was found to be as efficacious as enoxaparin and UFH in the treatment of DVT and PE, respectively (36,37). Unlike UFH and LMWH, heparin-induced thrombocytopenia (HIT) (see below) has not been observed during fondaparinux therapy, and case reports suggest that it may be a useful agent for treatment of HIT (38). However, exclusive renal elimination precludes the use of fondaparinux in patients with estimated creatinine clearances <30 mL/min.
Over the next few years, additional options for the acute and chronic treatment of DVT and PE undoubtedly will become available. Idraparinux, a long-acting indirect factor Xa inhibitor similar to fondaparinux that only requires once weekly dosing, currently is in testing for the prevention and treatment of VTE. The long half-life of this agent may limit its popularity until a specific reversal agent is available. Parenteral and oral compounds that directly inhibit factor Xa are in earlier stages of clinical development for VTE treatment (39).
Direct Thrombin Inhibitors
Among investigational antithrombotic agents, oral direct thrombin inhibitors have generated the most interest among clinicians. Ximelagatran, an oral prodrug of the parenteral direct thrombin inhibitor melagatran, has been demonstrated to be noninferior to enoxaparin/warfarin in the acute and chronic treatment of VTE (40). Although ximelagatran has not received FDA approval due to concerns about liver toxicity and rare cardiac events, this drug or another oral direct thrombin inhibitor likely will become available for clinical use in the next 5 to 10 years (41).
Warfarin
Warfarin is the principal agent used for chronic therapy of VTE. Warfarin can be initiated on the first day of acute therapy once therapeutic anticoagulation with UFH, LMWH, or fondaparinux is achieved. Initial warfarin doses should be chosen based upon the expected maintenance dose. Loading doses do not accelerate the pace of warfarin anticoagulation but do increase the likelihood of supratherapeutic INR values. For most patients, 5 mg is an appropriate initial dose. Elderly patients (age >70 years), patients with liver dysfunction, postoperative patients, or patients being treated with medications that reduce warfarin metabolism or disrupt vitamin K absorption or metabolism should start on no more than 2.5 mg warfarin. Use of a warfarin dosing nomogram can help simplify dosing decisions (Table 57.7) (42). The target INR for the majority of VTE patients is 2.5 (range 2.0–3.0). Possible exceptions to this rule include patients with recurrent VTE despite therapeutic anticoagulation and patients with APS and recurrent thromboembolism in whom a higher target range (2.5–3.5 or 3.0–4.0) may be appropriate (34,43).
A number of dietary and medicinal factors can influence warfarin therapy, either enhancing or inhibiting its effect (Table 57.8). Patients need not exclude all vitamin K-containing foods from their diets but should maintain moderate and consistent vitamin K intake to avoid excessive INR fluctuations (a list of the vitamin K content of foods is available at http://www.coumacarenews.com). To reduce the likelihood of drug–drug interactions with warfarin, patients should be instructed to always mention to health care providers that they are taking warfarin when they are prescribed new medications. In light of its effects on warfarin metabolism, alcohol consumption should be minimized (no more than one to two glasses of beer or wine or two ounces of liquor daily). To avoid missed doses, warfarin should be taken at the same time each day.
During the transition from another anticoagulant (e.g., LMWH or UFH) to solo warfarin therapy, the INR should be monitored at least every other day to ensure that therapy with the other anticoagulant is discontinued once the INR reaches the therapeutic range (e.g., 2.0–3.0). Once acute therapy is completed, the INR should be measured initially
P.852
at least twice weekly. Once patients achieve a stable INR on a stable dose of warfarin, the frequency of INR monitoring can be gradually reduced from weekly to monthly. Because there is a direct correlation between the frequency of INR monitoring and the time spent in the therapeutic range, less frequent monitoring is inadvisable. When dictated by the INR, small graduated dose adjustments (increases or decreases of 5%–20% of the weekly dose) should be made with weekly INR monitoring (Table 57.9). To reduce confusion, a single tablet size (e.g., 5 mg) should be used if possible. Dose schedules should assign specific doses to specific days of the week (avoid every other day or every third day schedules). Dose alterations can be made by splitting tablets (using a tablet splitter that is available from pharmacies and medical equipment stores) and by increasing or decreasing the number of days a patient receives a particular dose (e.g., 5 mg on Monday, Wednesday, and Friday; 2.5 mg on Tuesday, Thursday, Saturday, and Sunday). Dramatic changes in INR should always prompt investigation for the causative factor. Table 57.10 provides guidelines for the management of supratherapeutic INR values in the presence and absence of bleeding (43).
|
TABLE 57.7 Warfarin Initiation Guidelines |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
TABLE 57.8 Factors That Commonly Influence a Patient's Response to Warfarin |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Cancer patients are at higher risk for recurrent VTE and bleeding during treatment. A study of cancer patients demonstrated that dalteparin therapy for 6 months was associated with fewer recurrent VTE without increased bleeding compared with warfarin therapy (44). LMWH should be considered for patients likely to develop complications during warfarin therapy (e.g., cancer patients with liver metastases, poor oral intake, dramatically fluctuating INR values, etc.). Drug cost may be a significant barrier to broader use of LMWH for chronic VTE therapy.
Duration of Anticoagulation
The duration of anticoagulation depends upon the type of VTE, the presence of transient or permanent risk factors, and whether the event prompting treatment represents an initial or recurrent episode. Thrombotic events that occur in association with transient major risk factors require a shorter duration of therapy than do other events or VTE associated with major, long-term risk factors (e.g., malignancy, antithrombin deficiency, APS, etc.). PE is generally treated for a longer duration than DVT because of its higher case fatality rate. Recurrent episodes of VTE generally warrant long-term therapy unless they are associated with transient risk factors. Any decision about the duration of anticoagulation in a given patient should consider the patient's risk for bleeding as well as for recurrent VTE. Validated tools to assess bleeding risk in association with warfarin therapy are available (Table 57.11). Table 57.12 provides guidelines for the duration of VTE therapy (34).
Complications of Anticoagulants
The primary complication of anticoagulation is bleeding. The risk of bleeding associated with UFH or LMWH
P.853
is influenced by patient age, intensity of anticoagulation, presence of renal insufficiency, and, in some studies, patient gender (female higher risk than male). The frequency of major and fatal bleeding associated with UFH therapy for VTE is 2% and 0.25%, respectively. Major (1.2%) and fatal bleeding (0.1%) are slightly less with LMWH (45). Use of fondaparinux for VTE is associated with a frequency of major bleeding and fatal bleeding of 1.2% and 0.14%, respectively (36,37).
|
TABLE 57.9 Warfarin Management Guidelines |
||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||
Determinants of bleeding risk associated with warfarin therapy include the intensity of anticoagulation, patient age, previous history of bleeding, comorbidities (hypertension, cerebrovascular disease, renal insufficiency, and cancer), and use of concomitant antithrombotic medications (e.g., aspirin, clopidogrel, nonsteroidal anti-inflammatory drugs). The risk of major bleeding is greatest in the first month of therapy and declines thereafter. Elderly patients are at greater risk of bleeding than younger patients because of their greater number of comorbidities, their use of multiple medications, and their greater vascular fragility and likelihood of falling (45). Clinical trials of chronic warfarin therapy for VTE have estimated the incidence of major bleeding to be one to two episodes per 100 patient-years (34). Given the strict selection criteria for participation in these studies, the incidence of major bleeding derived from a cohort of anticoagulation clinic patients (7.6 events per 100 patient-years) probably is a more realistic estimate for most ambulatory care populations (46). Validated assessment models that allow prediction of a patient's bleeding risk on warfarin therapy have been developed (Table 57.11) (47).
Management of bleeding during warfarin therapy depends on careful assessment by the clinician of the risks associated with anticoagulation during a bleeding episode, as well as the risk of recurrent thromboembolism if anticoagulation is discontinued. The risk of recurrent VTE in the absence of anticoagulation depends on the amount of time that has elapsed since the thrombotic event and on the presence of ongoing risk factors for thrombosis (e.g., malignancy, thrombophilia, idiopathic or recurrent VTE). The risk of recurrent VTE declines as time elapses. During the first month, absence of anticoagulation is associated with a higher risk of recurrence (40% for the 1-month time interval) than during months 2 and 3 after thrombosis (10%
P.854
for the 2-month interval) (48). Potent thrombophilic states (e.g., APS, active cancer, or idiopathic VTE) are associated with a greater risk of recurrence than VTE associated with transient risk factors.
|
TABLE 57.10 Management of Excessive Warfarin Anticoagulation |
||||||||||||||
|
||||||||||||||
The decision to withhold anticoagulation must consider the clinical severity of bleeding. Minor bleeding (easy bruisability, gum bleeding, etc.) generally does not require a change in therapy (unless associated with a supratherapeutic INR). In contrast, major life-threatening bleeding (e.g. massive hematuria, gastrointestinal bleeding, retroperitoneal bleeding, or intracranial bleeding) generally warrants prompt discontinuation of warfarin and reversal of anticoagulation so that the cause can be identified and treated. Tables 57.9 and 57.10 provide recommendations for warfarin reversal (43). Once the cause of bleeding is identified and hemostasis is achieved, anticoagulation, if still warranted based upon the risk of recurrent VTE, can be reinstituted with close monitoring. Hemorrhage during anticoagulation should not be attributed solely to the presence of anticoagulation, even when the INR is supratherapeutic, as causative lesions are often identified (49,50).
|
TABLE 57.11 Outpatient Bleeding Risk Index |
||||||||||||
|
||||||||||||
Another complication of UFH or LMWH therapy is osteoporosis (in patients receiving prolonged therapy, typically for more than a few months). Osteoporotic fractures occurred in 2.2% of patients treated with subcutaneous UFH during pregnancy (51). Osteoporosis appears to be much less common in patients treated with LMWH (52).
A significant complication of heparin therapy is heparin-induced thrombocytopenia (HIT). HIT is an
P.855
immune-mediated prothrombotic state caused by antibodies directed against heparin-bound platelet factor 4 (PF4). It occurs in approximately 1% of patients treated with intravenous UFH and is eightfold to 10-fold less common with LMWH therapy. HIT is characterized by a ≥50% decrease in the platelet count beginning most commonly 4 to 14 days after initiation of heparin therapy. Confirmatory diagnostic tests include the heparin–PF4 assay (HIT antibodies) and the serotonin release assay. Any patient suspected of having HIT should have all heparin products eliminated and should be started on treatment with a direct thrombin inhibitor (e.g., lepirudin or argatroban) because of the increased risk for venous and/or arterial thrombosis in patients with HIT. Upper- and lower-extremity duplex ultrasonography is indicated to identify asymptomatic DVT because studies have demonstrated that 50% of HIT patients have subclinical thrombosis, the presence of which alters the duration of anticoagulation therapy (see above) (53). In the absence of thrombosis, treatment with a direct thrombin inhibitor should continue until the platelet count has returned to baseline. Although fondaparinux therapy for HIT has been used in clinically stable ambulatory patients with adequate renal function, outpatients who develop HIT should be hospitalized for initial therapy. Osteoporosis and HIT have not been seen with fondaparinux.
|
TABLE 57.12 Duration of Anticoagulation for VTE |
||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||
Warfarin skin necrosis is a rare complication of warfarin therapy (estimated frequency 0.01%–0.1% of warfarin-treated patients). Warfarin skin necrosis is characterized by thrombosis of small dermal vessels that results in skin necrosis, particularly in women, in the areas of the breasts, buttocks, and thighs. Warfarin skin necrosis most commonly occurs when warfarin is used (often in large loading doses) to treat an acute thrombotic disorder in the absence of concomitant UFH or LMWH therapy and in the presence of an underlying thrombophilic state. Although the presence of protein C deficiency has been traditionally linked to warfarin skin necrosis, factor V Leiden, protein S deficiency, antithrombin deficiency, and APS also have been associated with this devastating complication. In patients who develop warfarin skin necrosis and require long-term anticoagulation, warfarin can still be used if therapy is initiated slowly and gradually over a 10-day period (starting with doses of 1–2 mg/day) once therapeutic anticoagulation with heparin is achieved and if heparin therapy is not discontinued until INR ≥2 is achieved with warfarin (54).
Thrombolytic Therapy
Anticoagulation alone rarely results in complete thrombus resolution. Thrombolytic therapy produces substantially greater clot lysis that theoretically may reduce the frequency of postthrombotic syndrome (PTS) (see below for discussion of this syndrome) (55). In a study of patients with submassive PE, systemic thrombolysis reduced the need for treatment escalation compared to anticoagulation alone, although mortality was similar (56). Consequently, despite its therapeutic potential in VTE, thrombolytic therapy is primarily reserved for patients with extensive iliofemoral thrombosis and/or massive life-threatening PE. Randomized clinical trials are needed to determine if the benefits of applying thrombolytic therapy to a broader population are worth the risks of greater bleeding. Thrombolytic drugs should be administered only to inpatients.
Vena Caval Filters
Vena caval filters are sieves generally constructed of nonferromagnetic wire that can be placed percutaneously into the inferior vena cava (IVC) using catheter-directed techniques by an interventional radiologist or vascular surgeon. Numerous case series have demonstrated that vena caval filters are an effective means of PE prevention when anticoagulation is contraindicated or when a patient experiences a PE despite therapeutic anticoagulation. However, permanent vena caval filters are clearly associated with an increased risk for DVT compared to anticoagulation alone. Other complications of permanent filters include IVC thrombosis (2%–10%); penetration of filter components through the IVC wall (0.3%), occasionally
P.856
involving adjacent anatomic structures; migration from the placement site (0.3%); and tilting (5%) or fracture (3%) of the filter, which theoretically may result in reduced performance. Acute procedure-related complications include misplacement (1.3% of insertions), pneumothorax (0.02%), hematoma (0.6%), air embolism (0.2%), inadvertent carotid artery puncture (0.04%), and arteriovenous fistula (0.02%). Based upon the published case series, fatal complications of placement are rare (0.13% of insertions). Several retrievable vena caval filters are approved for use in the United States. These devices can be removed weeks or, in some cases, months after implantation. Whether retrievable filters will perform as well as permanent filters remains unclear. If retrievable filters are demonstrated to reduce the risk of PE as effectively as permanent filters, retrievable filters should prove to be a very useful treatment option for patients with a transient absolute contraindication to anticoagulation (57).
Management of Venous Thromboembolism during Pregnancy
Pregnancy is associated with a twofold to fourfold increased risk for VTE. Therefore, primary care providers may be involved in the management of patients with VTE during pregnancy. When a pregnant patient is suspected of having DVT or PE, the diagnosis must be confirmed (duplex ultrasonography, V/Q scanning, or helical CT scanning) using fetal shielding when appropriate rather than forgoing diagnostic testing for fear of harming the fetus. A misdiagnosis resulting in inappropriately omitted or prescribed anticoagulation is far more likely to harm the mother and fetus than the effects of diagnostic testing (58). Warfarin therapy during pregnancy is associated with teratogenicity and with an increased risk for bleeding in the fetus. Therefore, UFH or LMWH is generally used for acute and chronic therapy during pregnancy. The lower risk of osteoporosis and HIT and the more favorable pharmacokinetics of LMWH have made it the preferred anticoagulant for VTE during pregnancy. Since maternal weight increases during pregnancy, LMWH doses should be adjusted accordingly as pregnancy progresses. Twice daily dosing is preferred given the shorter half-life of LMWH during pregnancy. LMWH should be discontinued 24 hours prior to the induction of labor. In the post-partum period, either LMWH or warfarin may be used for antithrombotic therapy as maternal warfarin use does not appear to result in an anticoagulant effect in breast-fed infants. Patients with an episode of VTE during pregnancy should be treated for at least 6 weeks post-partum or for the duration of therapy that is appropriate for their thrombotic episode, whichever is longer. Patients with VTE during pregnancy and patients on long-term anticoagulation for previous VTE should receive prophylactic LMWH during subsequent pregnancies (59).
Considerations in Patients with Recurrent Venous Thromboembolism
Recurrent VTE generally indicates persistent abnormalities affecting one or more factors of the Virchow triad. If a patient presents with a recurrence soon after completion of acute VTE therapy, two diagnostic entities should be strongly considered: delayed HIT and Trousseau syndrome. HIT will be apparent by the presence of a reduced platelet count and can be confirmed by laboratory testing. Avoidance of heparin exposure and therapy with a direct thrombin inhibitor are indicated (see above for a discussion of the treatment of HIT). Trousseau syndrome is a prothrombotic disorder associated with solid tumor malignancies that is characterized classically by recurrent migratory thrombophlebitis despite adequate warfarin anticoagulation, laboratory findings consistent with disseminated intravascular coagulation (including thrombocytopenia, hypofibrinogenemia, and elevated D-dimers), and nonbacterial thrombotic endocarditis (60). Anticoagulation with UFH or LWMH is the most effective treatment. Occasional resistance to LMWH therapy has been reported.
APS is another common cause of recurrent VTE despite adequate warfarin anticoagulation. Although most patients with APS can be managed with conventional-intensity warfarin anticoagulation (INR 2.0–3.0), occasional patients require higher INR target ranges (INR 3.0–4.0). Less commonly, other causes of hypercoagulability can trigger recurrent VTE despite adequate anticoagulation and should be investigated as dictated by the clinical circumstances. Recurrent events in the same location should prompt consideration of vascular abnormalities such as May-Thurner syndrome or Paget-Schroetter syndrome (thoracic outlet syndrome, see below), which are underrecognized causes of recurrent VTE. May-Thurner syndrome is characterized by deformation and stenosis of the left iliac vein as a consequence of compression by the overlying right iliac artery. The resulting stenosis precipitates recurrent episodes of left iliofemoral DVT. Thrombolysis and stenting of the affected vein segment can result in long-lasting resolution (61). Management of the Paget-Schroetter syndrome is discussed below.
Calf Vein Thrombosis
Calf vein thrombosis consists of DVT involving the deep vessels of the legs distal to the popliteal vein (when extension into the thigh is not also present). Calf vein DVT is associated with a significant risk of recurrent VTE when treated with short durations of anticoagulation (e.g., 5-day course of UFH); therefore, warfarin anticoagulation in addition to acute therapy with heparin is recommended for 6 to 12 weeks to prevent recurrent DVT or PE (62,63).
P.857
Upper-Extremity Deep Venous Thrombosis
Upper-extremity DVT generally consists of thrombosis involving the brachial, axillary, and/or subclavian veins with occasional extension into the superior vena cava. Common risk factors for upper-extremity DVT include central vein catheterization, pacemaker and implantable cardioverter-defibrillator (ICD) placement, cancer, chemotherapy, thrombophilia, and anatomic abnormalities such as the thoracic outlet syndrome (Paget-Schroetter syndrome). Upper-extremity DVT typically is heralded by pain and swelling of the affected extremity and later by the development of superficial venous collaterals. Duplex or color flow ultrasonography is useful to confirm the diagnosis. As in lower-extremity DVT, anticoagulation is indicated, with or without thrombolytic therapy. If the DVT is catheter associated, the catheter often is removed, but it may remain in place if indications for its use remain.
Occurrence of an upper-extremity DVT without clear risk factors should prompt investigation for the Paget-Schroetter syndrome, a venous manifestation of thoracic outlet syndrome. In affected individuals, the subclavian vein is compressed by local anatomic structures (anterior scalene muscle, cervical ribs, etc.). Vascular wall damage and stasis result in thrombosis. Historically, patients may relate a recent history of heavy lifting or exertion involving the affected extremity. Venography of the upper limb vessels in stress position (abduction, external rotation) is useful to demonstrate compression of the subclavian vein (64,65). Therapy for Paget-Schroetter syndrome consists of catheter-directed thrombolysis followed by surgical correction of the anatomic abnormality and anticoagulation (66).
As with lower-extremity DVT, hypercoagulable states may play a role in the pathogenesis of upper-extremity DVT. The risk of PE associated with upper-extremity DVT (10%–15%) appears to be less than that associated with lower-extremity DVT (50%–60%), but the risk is not insignificant (64,67). PTS (see below) can also occur in the affected extremity. Therefore, at least 3 months of anticoagulation with warfarin is recommended (64).
Superficial Venous Thrombophlebitis
Superficial venous thrombophlebitis (SVT) consists of thrombosis of the superficial veins of the legs or arms. Risk factors for SVT are similar to risk factors for DVT and include peripheral venous catheters, thrombophilia, oral contraceptives or hormone replacement therapy, pregnancy, recent surgery or trauma, and a history of previous DVT or SVT. Less common etiologies include Trousseau syndrome, Buerger disease (thromboangiitis obliterans), and Behçet syndrome. SVT traditionally has been managed symptomatically with compression, elevation of the affected extremity, and anti-inflammatory agents, or less commonly with vein ligation. A study indicates that prophylactic or therapeutic doses of LMWH may be a useful option in some patients with SVT (68). Further investigation is needed to define the patient populations likely to benefit from anticoagulation and the appropriate duration of therapy. Until these studies are performed, traditional approaches to SVT probably should remain the first-line therapy with consideration of anticoagulation in patients with progressive disease. Any episode of proximal lower-extremity SVT should prompt duplex ultrasonography to rule out extension into the deep venous system given that PE has occurred in this situation. Conventional anticoagulation for at least 3 months would be appropriate in this instance (68).
Postthrombotic Syndrome (PTS)
PTS is a common cause of chronic morbidity after DVT. PTS is characterized by edema, pain, heaviness, and occasionally skin ulceration of the affected extremity that develop as a consequence of chronic venous obstruction and valvular incompetence. Symptoms worsen with standing and ambulation (69). PTS affects almost 30% of patients within 5 years of an episode of DVT (70). Risk factors for PTS include recurrent ipsilateral DVT, obesity, female gender, and increased age (69). Consistent use of knee-high compression stockings (30–40 mm Hg) has been shown to reduce the incidence of PTS by 50% (71,72). Therefore, all patients should be prescribed knee-high compression stockings after an episode of DVT. Optimally, compression stockings should be worn daily starting within several weeks of completion of acute therapy and then for at least 2 years post-DVT. Whether patients with upper-extremity DVT will benefit from similar treatment remains unclear (73). By reducing the thrombus burden and venous valvular damage, it is possible that catheter-directed thrombolysis may reduce the risk of PTS, but this has not been demonstrated in a randomized clinical trial.
Arterial Thromboembolism
Arterial thromboembolism (AT), as manifested by cerebrovascular, cardiovascular, or peripheral vascular disease, remains the most common cause of morbidity and mortality in developed countries. Unlike the venous circulation where coagulation factors play the principal role in the pathogenesis of thromboembolism, arterial thrombosis is primarily the result of platelet-rich thrombi. Consequently, the most important risk factors for arterial thrombosis affect vascular wall and platelet function. Atherosclerosis is an important risk factor for AT, which induces
P.858
endothelial dysfunction, tissue factor formation, and, ultimately, platelet plug formation that can result in local stenosis and ischemia or embolization. Other risk factors for AT include hypertension, diabetes, smoking, obesity, elevated low-density lipoprotein (LDL) cholesterol and fibrinogen levels, abnormalities of fibrinolysis, hyperhomocysteinemia, APS, and rarely malignancy-associated thrombosis including Trousseau syndrome. Whereas anticoagulation is clearly indicated in the acute treatment of unstable angina, myocardial infarction (seeChapter 62), and cardioembolic stroke (see Chapter 91) and in the prevention of thromboembolism associated with atrial fibrillation (seeChapter 64), left ventricular thrombus, APS, and malignancy-associated thrombosis, inhibiting platelet function remains preeminent in the treatment of AT (74).
Aspirin
Aspirin remains the most commonly used medication for the prevention and treatment of AT. The antithrombotic activity of aspirin is primarily due to its ability to irreversibly acetylate and thus inhibit cyclooxygenase, the enzyme that converts arachidonic acid into prostaglandin H2. This interferes with the synthesis of thromboxane A2, a potent vasoconstrictor and platelet aggregating substance (75).
Aspirin has been widely studied in doses ranging from 50 to 1,500 mg/day in the prevention and treatment of AT, including myocardial infarction, stroke, and peripheral vascular disease. The decision to use aspirin in the primary prevention of AT requires an assessment of the individual's risk of AT as well as his or her risk of gastrointestinal toxicity and bleeding associated with aspirin therapy. The risk of a major bleed associated with aspirin in doses of 81 to 325 mg/day is small (approximately one to two major bleeds per 1,000 patient-years, approximately 10-fold less than that associated with warfarin). This risk is not reduced with enteric-coated aspirin or buffered aspirin. However, the benefits of aspirin for healthy individuals at low risk for AT are equally small. Therefore, aspirin should not be used for primary prevention of AT in low-risk healthy adults (75). In contrast, three large studies have demonstrated that low doses of aspirin (75–100 mg/day) significantly reduce (by 15%–23%) vascular events in high-risk patients with one or more risk factors for vascular disease (76,77, 78).
Aspirin has also proven to have activity in the treatment of patients with pre-existing vascular disease. Among patients with stable angina, aspirin therapy results in 10 fewer vascular events per 1,000 patient-years. The benefits of aspirin are even greater among patients with unstable angina or previous myocardial infarction who experienced 50 and 36 fewer vascular events per 1,000 patient-years, respectively. Similar benefits have been seen in patients with transient ischemic attack (TIA) and stroke. It is important to emphasize that although daily doses of aspirin ranging from 50 to 1,500 mg have been demonstrated to be effective in the prevention and treatment of AT, the incidence of gastrointestinal side effects and clinical bleeding increases with high doses, and the clinical benefits of aspirin do not appear to increase at doses >100 mg/day. Therefore, daily doses of aspirin of 75 to 100 mg probably are sufficient for most patients, particularly patients deemed to be at higher risk of bleeding complications (75,79, 80, 81).
Thienopyridines
Thienopyridines represent the second class of antiplatelet agents used for AT. Ticlopidine (Ticlid) and clopidogrel (Plavix) belong to this drug class. These agents inhibit platelet aggregation induced by adenosine diphosphate. Ticlopidine has been demonstrated to be superior to aspirin in patients with stroke or TIA and of comparable efficacy in patients with a recent myocardial infarction (82,83). However, hematologic adverse effects, including neutropenia (0.8% of patients), thrombocytopenia, aplastic anemia, and thrombotic thrombocytopenic purpura (TTP; estimated frequency, one case per 1,600–5,000 patients treated) have greatly reduced the use of ticlopidine in clinical practice (75,84).
Clopidogrel has equivalent efficacy to aspirin in prevention of vascular events in patients with myocardial infarction, stroke, and peripheral arterial disease. When added to aspirin, clopidogrel is associated with a reduction in vascular events compared with aspirin alone and has become standard therapy for patients after percutaneous coronary interventions. Clopidogrel is associated with a comparable risk of bleeding and thrombocytopenia compared to aspirin. TTP rarely occurs with clopidogrel, and the incidence may not exceed that seen in the general population. When it occurs, TTP develops within the first 2 weeks of therapy (75,85). Because it does not cause neutropenia and is associated much less often with TTP (86), clopidogrel has become the thienopyridine of choice in clinical practice.
Dipyridamole
Dipyridamole affects platelet function by inhibiting phosphodiesterase that results in increased intraplatelet cyclic adenosine monophosphate levels. The European Stroke Prevention Study 2 demonstrated that the combination of dipyridamole and aspirin was superior to aspirin alone in the prevention of stroke in patients with a previous stroke or TIA (87).
Glycoprotein IIb/IIIa Inhibitors
Glycoprotein IIb/IIIa is the platelet receptor responsible for platelet aggregation. Congenital deficiency of this protein is the cause of the platelet function disorder Glanzmann thrombasthenia. As predicted by its physiologic role,
P.859
glycoprotein IIb/IIIa inhibitors have proven to be important treatments for coronary artery disease. The currently available glycoprotein IIb/IIIa inhibitors include abciximab, a chimeric mouse–human monoclonal antibody, and two synthetic receptor antagonists, tirofiban and eptifi-batide. Development of effective oral glycoprotein IIb/IIIa inhibitors is an active area of clinical investigation that may have a significant impact on the future ambulatory care of patients with AT (75). Currently, none of these drugs is suitable for use in ambulatory patients.
Specific References*
For annotated General References and resources related to this chapter, visit http://www.hopkinsbayview.org/PAMreferences.
P.860