Chapter 80 Should Thromboprophylaxis Be Used for Lower Limb Joint Replacement Surgery?
THROMBOEMBOLIC DISEASE IN JOINT REPLACEMENT SURGERY
Venous thromboembolic events (VTEs) are among the most feared complication of total hip replacement (THR) and total knee replacement (TKR) surgeries, and these patients are in the greatest surgical risk group for this complication. The development of deep venous thrombosis (DVT) with the potential to propagate a potentially lethal pulmonary embolus (PE) is the complication that presents the greatest risk for perioperative mortality after lower limb joint replacement. This has led to the adoption of routine thromboprophylaxis as a standard of care in joint replacement surgery since the late 1980s.1,2
The natural history of VTE disease, after joint replacement surgery, has been well characterized. Based on data before 1980 and from trials with placebo patients, the incidence of lower extremity DVT without thromboprophylaxis in THR and total knee arthroplasty (TKA) has been reported as 40% to 60% and 41% to 85%, respectively (Table 80-1).3–9 Many of these statistics are based on venography end points. At least 50% of these patients had either asymptomatic or distal DVTs that resolved spontaneously without clinical sequelae.10,11 In the absence of thromboprophylaxis, clinically more important proximal DVTs occurred in THR and TKA at 18% to 36% and 5% to 22%, respectively, but again many were asymptomatic. With prophylaxis, symptomatic VTE was seen in 2.4% and 1.7%, respectively, of patients within 3 months of THR or TKA in studies done between 1992 and 1996.12 Many of these thrombotic events occur after hospital discharge, and the VTE risk remains increased for at least 8 weeks after surgery.
The incidence of PEs is less certain. The incidence of asymptomatic PE ranges from 3% to 20% and symptomatic PE from 0.5% to 3%,13,14 and with the current routine use of thromboprophylaxis, fatal PE is uncommon.15 Without thromboprophylaxis, perioperative mortality in THR and TKA from PE was reported from 0.1% to 2%.5,16, 17 With thromboprophylaxis for 7 to 10 days, there is a 0.1% rate of postdischarge fatal PE at 90 days after surgery in THR.18
The ideal VTE prophylaxis includes clear effectiveness compared with placebo or other active interventions; good risk/benefit profile; good compliance by patients, nurses, and physicians; easy administration or intervention; and requiring no need for monitoring whereas being cost effective.19 The best means to attain this goal remains an evolving science.
VENOUS THROMBOEMOBLIC EVENT RISKS
The interaction of age, genetic factors, and secondary, acquired persistent or transient environment or local factors is the accepted model for the pathogenesis of venous thrombotic disease. VTE risk has a direct and major relation to age.19 At ages 20, 50, and 80 years, the prevalence of VTE is 1 in 10,000, 1 in 1000, and 1 in 100 individuals per year, respectively, with a lifetime risk of 10%.20 Ethnicity is a risk factor.21,22 VTE is more common among blacks compared with whites and Asians by 1.4- and 7-fold, respectively.
Up to 15% to 20% of the general population has demonstrable endogenous biochemical hypercoagulable states.23,24 These defects may be on a genetic or acquired basis and include defects or deficiencies in anticoagulant proteins (protein S or C, antithrombin [AT]), altered procoagulant proteins (factor V and prothrombin), or fibrinolytic processes.25 Genetic hypercoagulable states include factor V Leiden (2–4% of the white population) and prothrombin 20210 defect (2–4% of the white population), AT deficiency (1/1000), protein C deficiency (1/2000), and protein S deficiency (1/2000).26 Acquired defects include anti-phospholipid antibodies (2–5%) that can be found in up to 25% of population by age 75.23 Increased factor VIII, found in up to 10% of the well population, has a dose-related relation to VTE and is one of the more recently identified common risk factors.27 Factor V Leiden is the most common cause of inherited thrombophilia, found in up to 20% of VTE cases. In addition, local factors play a key role in thrombosis related to lower limb joint surgery and include vascular injury and thrombotic repair response, as well as venous stasis caused by prolonged immobility.
Independent VTE risk factors identified in studies of surgical populations for VTE included age older than 50 years, varicose veins, prior myocardial infarction, cancer, atrial fibrillation, ischemic stroke, diabetes mellitus, current use of estrogen-containing compounds, and chronic and acute inflammatory states including infections.28 Additional risk factors that increase the risk for surgery-related VTE include previous VTE, obesity, heart failure, and paralysis.29
VTE risk is increased, in a stepwise and fairly linear manner, with multiple risk factors. One surgical study noted a 10% to 15% increased risk for VTE with each additional risk factor identified.30 Multiple risk factors in orthopedic patients should be expected given the age, increasing numbers of patients undergoing joint replacement, and frequency of underlying VTE predispositions.
Medical and surgical patients have been risk stratified into low, intermediate, high, and highest VTE risk groups, to assist in creating robust clinical recommendation for best thromboprophylaxis practice.2,31 Patients undergoing elective joint replacement surgery fall into the highest risk category of this well-accepted clinical model. Patients at highest risk for VTE include surgical patients older than 40 years with multiple risk factors, those with THR or TKA, patients undergoing hip fracture surgery, or patients with major trauma and spinal cord injury. Risk for VTE in this aggregate group include calf DVT (40–80%), proximal DVT (10–20%), clinical PE (4–10%), and fatal PE (0.2–5%).
Prevention of Venous Thromboembolic Events: Introduction
The prophylactic measures most commonly used in surgery include low-dose unfractionated heparin (LDUFH), low-molecular-weight heparin (LMWH), the substituted pentasaccharide, fondaparinux, oral adjusted-dose vitamin K antagonists (VKAs) (international normalized ratio [INR], 2.0–3.0), and lower limb mechanical compression approaches. For joint replacement surgery, LDUFH is not an acceptable option in normal circumstances.2
A large and strong body of evidence-based studies, from randomized trials and meta-analysis, in thromboprophylaxis makes current prophylaxis recommendations robust for most common procedures including joint replacement surgery. A primary reference that is increasingly seen as the authoritative reference body that sets the North American standard of care in thromboprophylaxis is the periodic report on the “Prevention of Venous Thrombosis” for the American College of Chest Physicians Conference on Antithrombotic and Thrombolytic Therapy.2 It is strongly recommended that individual hospitals, as well as subspecialty and physician groups, should develop, adopt, and evolve formal strategies for the prevention of VTE disease for all risk categories for surgical and medical patients.
Drugs Used in Venous Thromboembolic Event Prevention
Low-Dose Unfractionated Heparin.
UFH is a heterogeneous glycosaminoglycan (5000–30,000 molecular weight [MW]) that binds to AT, altering AT so that AT then binds to and inactivates circulating factor Xa and factor IIa (thrombin). UFH then dissociates from these inactive, AT-serpin complexes that are then cleared. This anti–factor Xa and anti–factor IIa activity accounts for UFH anticoagulant activity. UFH inactivates free but not clot-bound factors IIa and Xa in a 1:1 ratio and, to a lesser extent, the other serine proteases.32
Clearance of UFH is dose dependent, with an approximate half-life of 30 to 60 minutes. This short half-life explains the need for two or three times daily dosing when prophylaxis is used. One major biophysical drawback with UFH, but not with LMWH or fondaparinux, is that UFH binds to and is made unavailable by a large number of endogenous, intravascular heparin-binding proteins (von Willebrand factor, fibronectin, platelet factor 4) that reduce its anticoagulant activity, because only free UFH is active.33 This may in part explain LMWH superiority to UFH in high-risk prothrombotic situations such as joint surgery. LDUFH has the advantage that it is relatively inexpensive compared with LMWH and fondaparinux, and is easily administered. Anticoagulant monitoring is not required for prophylaxis.
Limitations of UFH include the pharmacokinetic, biophysical properties and risk for heparin-induced thrombocytopenia (HIT).34 HIT is an immunologic disorder in which antibodies form against heparin-PF4 complex bound to the platelet and endothelial cell surface, and result in vascular cell (platelet and endothelial cell) activation, thrombocytopenia, and arterial and venous thrombosis. These pathologic antibodies form in 1% to 3% of patients receiving UFH, and although more common with full-dose UFH, the syndrome is well described with LDUFH. HIT is associated with significant morbidity (stroke, coronary events, peripheral arterial occlusions) in 10% to 15% of HIT cases. All patients receiving UFH must be followed with regular platelet counts while remaining on UFH, and any unexplained thrombocytopenia (decline in platelets of >50% from baseline or platelet count < 100,000/μL) during UFH use requires investigation and elucidation of its cause. If HIT is suspected, UFH must be stopped and replaced by an alternative anticoagulant not associated with HIT (hirudin, argatroban, or danaparoid) until the diagnosis is established.35 HIT antibodies cross-react with LMWH. Patients diagnosed with HIT should never be exposed to either UFH or LMWH.
Low-Molecular-Weight Heparins.
LMWH is a thromboprophylactic drug of choice for joint replacement surgery. LMWHs are fractionated, low-molecular-weight moieties (3000–5000 MW) of UFH, produced by chemical and physical methods, that are approximately one-third as long as native UFH.32 LMWHs when compared with UFH have a number of desirable pharmacologic and clinical properties that include: (1) superior and predictable pharmacokinetics that allow once or twice daily subcutaneous (SC) dosing based on body weight, (2) no laboratory monitoring, (3) more favorable benefit/risk ratio, and (4) less anti–factor IIa and more anti–factor Xa activity. Because of these properties and its demonstrated improved clinical efficacy in preventing VTE disease, LMWH has replaced UFH for many clinical indications including joint replacement surgery thromboprophylaxis.36
Like UFH, LMWH induces its anticoagulant effect through complex formation with AT and factor Xa, and to a lesser extent, thrombin. The anti–factor IIa/anti–factor Xa ratio is 1:1 for UFH, but 1:2 to 1:5 for different LMWHs. A number of pharmaceutical formulations are available in North America, including nadroparin, enoxaparin, dalteparin, and tinzaparin. They are prepared by different methods resulting in slightly different pharmacokinetic and anticoagulant properties, although clinical efficacy is similar for all indications other than acute coronary syndromes.37 Bleeding risk is similar or slightly lower with LMWH compared with UFH, and there is no true antidote for its anticoagulant action, unlike UFH.
Biophysical properties of LMWH are its major advantage. Because LMWH does not bind to intravascular proteins or cells, they have predictable pharmacokinetics including dose–response relations and an increased plasma half-life. Peak anticoagulant effect is seen at 4 to 8 hours after SC injection. Half-life of this renally cleared agent is much longer than UFH at 3 to 6 hours, and any measurable anticoagulant effect of LMWH is gone in 24 hours in the absence of kidney dysfunction. Depending on the formulation and supporting clinical evidence, LMWHs are administered either once or twice a day for thromboprophylaxis.37
HIT is much less common with LMWH use than UFH, although the precise incidence is not clear. As an example, one randomized, double-blind study of patients after hip surgery found that thrombocytopenia occurred in 9 of 332 patients (2%) receiving UFH compared with none receiving LMWH.38 Because of LMWH cross-reactivity with HIT antibodies, patients with a history of HIT should never be exposed to either LMWH.
Pentasaccharide.
Fondaparinux (Arixtra), a synthetic heparin analog, is a highly sulfated, pentasaccharide sequence, derived from the AT binding region of UFH. It is the newest, currently available anticoagulant for joint surgery thromboprophylaxis and has demonstrated improved efficacy compared with LMWH.39,40 The anticoagulant action of fondaparinux is via a mechanism similar to LMWH in which the drug binds to AT and causes AT to bind to and inactivate factor Xa, but not thrombin. Its action is directed against factor Xa only. Fondaparinux, like LMWH, does not bind to or interact with other plasma proteins or cells resulting in predictable and stable pharmacokinetics. Its half-life at 17 hours is much longer than UFH (0.5–1 hour) or LMWH (4–6 hours), and prophylactic doses (2.5 mg SC) are given daily. Although clinical HIT has not been described with fondaparinux, HIT antibodies do develop with it use.41 Bleeding with prophylactic fondaparinux is slightly more common than with LMWH or VKA, and like LMWH, this agent is not reversible.15,42
Oral Anticoagulants.
VKA has been available in clinical practice since the 1950s and is commonly used for VTE prophylaxis in lower limb joint replacement surgery. Certain anticoagulant properties including long half-life, delayed onset of action, and reversibility have made it a favored drug for joint replacement thromboprophylaxis in North America.43 The VKA warfarin induces an anticoagulant state by interfering with the vitamin K–dependent, post-translational modification of hepatically synthesized blood clotting factors II, VII, IX, and X. This results in nearnormal circulating levels of hypofunctional clotting factors II, VII, IX, and X.
These factors have markedly different half-lives, from 6 to 60 hours. It is the half-lives of these factors that dictate the anticoagulant effect of VKA. The anticoagulant efficacy of VKA most closely correlates with the functional level of factor II (prothrombin) that has the longest half-life (60 hours).44
VKA is rapidly absorbed from the gut and is 90% albumin bound in circulation. Only free, unbound VKA is biologically active. The half-life of VKA is 36 to 48 hours, and it is the hepatic metabolism by microsomal enzymes, including cytochrome P450 variants, that explains, in part, the marked interindividual variation in dosage required for drug effect. Numerous factors affect the biologic activity of VKA, including patient age, activity level, diet, drug absorption, albumin level, vitamin K intake, general state of health, and interfering drugs. The list of drugs known to interact by either potentiating or attenuating the VKA effect is lengthy, and knowledge of their identity and expected interactions is important given its narrow therapeutic index.45 Diet can also impact warfarin effect because ingestion of foodstuffs high in vitamin K will counteract the effects of warfarin within 6 to 12 hours.46
The average oral dosage of VKA is 5 mg/day, and for patients older than 70 years is 4 mg/day. The use of VKA dosing nomograms should be adopted and have been convincingly demonstrated to accelerate the attainment of a target INR whereas simplifying management for healthcare providers.47 The optimal INR target for treatment and prophylaxis of VTE indications is 2.5 (2.0–3.0). For any given indication, there is a trend to lower bleeding risk with no loss in clinical efficacy the lower the INR is kept within the therapeutic range. Target INR should be 2.0 to 3.0 as per published trials as opposed to lower targets. There is no rationale to use a loading dose for the initiation of VKA therapy.43 Given the long half-lives of both VKA (36 hours) and factor II (60 hours), this means that to attain full anticoagulant effect, the INR must be within the therapeutic range for at least 4 to 6 days for all of the vitamin K–dependent proteins to decrease to 20% to 30% of normal.
VKA is generally well tolerated, with few side effects other than the bleeding risk. Gastrointestinal (GI) upset and hair thinning may be seen in 10% to 15% of cases. An outpatient bleeding risk index has been developed that includes independent risk factors: age older than 65, history of GI bleeding, history of stroke.48 Low-, intermediate-, and high-risk patients had, respectively, 3%, 12%, and 53% probability of bleeding. The bleeding risk is directly correlated with INR; it begins to increase quite markedly when the INR increases to more than 3.5, and an INR greater than 5 requires active intervention (hold drug, oral or SC vitamin K1, fresh-frozen plasma) to reverse the drug effect.
Aspirin.
Although aspirin decreases the frequency of VTE after orthopedic surgery when analyzed in meta-analysis, this efficacy is significantly less than that obtained using other anticoagulant agents.49 A large randomized trial involving 4088 patients undergoing THR and TKR received placebo or 160 mg/day aspirin for 35 days, in addition to other thromboprophylactic measures prescribed at the discretion of the treating physician.50 Fatal PE and DVT were both significantly decreased by aspirin: relative risk reduction (RRR) of 40% and absolute risk reduction of 4 fatal PEs per 1000 patients. Wound-related bleeding, GI bleeding, and the need for transfusion were significantly more common in the aspirin-treated group. Thus, although aspirin may have some activity in preventing VTE, its efficacy and safety profile are markedly inferior to other available measures, precluding its use as monotherapy.51 Aspirin cannot currently be recommended for the prophylaxis of VTE.2
New Anticoagulant Agents.
Further improvements in thromboprophylaxis are going to come from the evolution of better pharmacologic agents with not only an increased efficacy in VTE prevention but a lower bleeding risk and improved biologic properties including reversibility and longer half-lives. The quest for this holy grail in anticoagulant therapy among pharmaceutical companies is ongoing. Examples of agents that have demonstrable efficacy with similar benefit/risk profiles when compared with LMWH and have been studied in THR and TKA include the direct thrombin inhibitors parenteral recombinant hirudin, parenteral melagatran, and oral ximelagatran and the new oral direct anti Xa inhibitor rivaroxaban.2,52, 53, 145, 146 Emerging drugs include idraparinux, a long-acting pentasaccharide congener to fondaparinux, currently in trials, that can be given once a week and the biotinylation of pentasaccharides so that their anticoagulant action can be reversed with an infusion of avidin.39 None of these agents has been approved for thromboprophylaxis use in North America.
Mechanical Approaches to Venous Thromboembolic Event Prevention
Compared with drug interventions for thromboprophylaxis, mechanical devices have been studied in fewer patients. Thus, the confidence around the recommendation for their use is weaker. A major practical issue related to using this approach is patient and health provider compliance.54 Intermittent pneumatic leg compression (IPC), graduated compression stockings (GCS), and venous foot pump (VFP) mechanically prevent venous thrombosis by augmenting blood flow and theoretically decreasing venous stasis in the deep veins of the legs. Pneumatic compression has an added biochemical anticoagulant effect by reducing plasminogen activator inhibitor-1 levels, resulting in increased endogenous fibrinolytic activity.55 Thus, mechanical leg compression approaches have both local and systemic effects.
Although IPC56–59 has been demonstrated to reduce the incidence of DVT in moderate-risk surgical patients,60 it is less effective in patients undergoing hip surgery or knee replacement, and although it reduces RRR of total DVTs and prevents calf vein thrombosis, it is not clinically effective for more important proximal DVTs.56,59 IPC use has decreased with decreasing length of hospital stay and rapid ambulation in patients undergoing orthopedic surgery.61 In a review, the duration, degree of compression, and overall compliance were significantly less than expected, even with a concerted education program, in patients undergoing THR when IPC was the only form of prophylaxis.62
GCS16,63–65 reduce venous stasis in the limb by applying decreasing degrees of compression moving proximally from the ankle to calf, but reduce the incidence of postoperative VTE only in low-risk surgical patients.66 In a meta-analysis, GCS in combination with other forms of prophylaxis does not result in any further risk reduction in VTE.67
Evidence of efficacy with VFP65,68 is less developed than the other mechanical methods, and although it does appear to reduce overall rates of DVT, it is inferior to current anticoagulant approaches.9,69 Mechanical devices are not recommended as a sole therapy for thromboprophylaxis in joint replacement surgery, and good comparative studies with these devices as part of multimodal approaches are lacking, making it difficult to make any firm recommendations despite their attractiveness because of their biomechanical action.
TOTAL HIP REPLACEMENT
Thromboprophylaxis has been recommended in all patients undergoing THR for at least 20 years. This recommendation is supported by many randomized, clinical trials between active interventions with and without placebo groups, and guidelines have been well refined. Three pharmacologic approaches have been demonstrated to be effective and safe. LMWH (SC once or twice daily), fondaparinux (SC once daily), or adjusted-dose VKA (target INR, 2.5).2,15,70–76
Mechanical prophylaxes GCS,16,63–65 IPC,56–59 and VFP65,68 have all been studied in THR. Each of these methods provides some level of protection compared with placebo with a RRR of DVT of 20% to 70%. However, this RRR is inferior compared with current pharmacologic interventions in preventing VTE, especially proximal DVTs.2,59, 63 Mechanical, lower limb methods are not recommended as a primary prophylactic modality.2
In THR, both aspirin51 and fixed-dose LDUH77 are superior to placebo in meta-analyses for the prevention of VTE. However, neither of these modalities is recommended as monotherapy because of inferior efficacy when compared with LMWH or VKA.3,53,78–80
In North America, as opposed to the European Union, thromboprophylaxis utilizing adjusted-dose VKA (target INR, 2.5) continues to the most common pharmacologic intervention used in THR.81–86
Related posts:
What Is the Optimal Treatment for Degenerative Lumbar Spinal Stenosis?
Legg-Calve-Perthes Disease: How Should It Be Treated?
Mangled Extremity: Are Scoring Systems Useful?
Femoral Neck Fractures: When Should a Displaced Subcapital Fracture Be Replaced versus Fixed?
What Is the Best Method of Rehabilitation after Flexor Tendon Repair in Zone II: Passive Mobilization or Early Active Motion? What Is the Best Suture Configuration for Repair of Flexor Tendon Lacerations?
What Are the Best Diagnostic Criteria for Lateral Epicondylitis?
