CHAPTER 29 Deep Vein Thrombosis Prophylaxis following Unicompartmental Knee Arthroplasty
Introduction
Venous thromboembolic disease (VTED) represents a spectrum of pathology ranging from asymptomatic deep vein thrombosis (DVT) to fatal pulmonary embolism (PE). Historically, VTED was one of the most feared complications of lower extremity total joint arthroplasty (TJA), with rates of fatal PE as high as 3.4%.1 General surgical studies demonstrated that, before 1990, PE accounted for approximately 10% of in-hospital deaths.2,3 However, modern orthopaedic surgical techniques, anesthetic care, and rehabilitation protocols have vastly improved the incidence of VTED and death after TJA. The Global Orthopaedic Registry found that only 0.3% of total knee arthroplasty (TKA) patients died within 3 months after TKA.4 The National Registry for England and Wales demonstrated that the overall mortality rates at 1 year were actually 66% lower than age- and gender-matched controls from the general population.5,6 Nevertheless, VTED remains a significant concern for patients and surgeons alike. DVT is still the most frequent in-hospital complication after TKA4 and the most common reason for emergency readmission.7
Etiopathogenesis Of Vted
The etiopathogenesis of VTED is multifactorial. Known as Virchow’s triad,8 the combination of venous stasis, endothelial injury, and hypercoagulability are predispositions for VTED. Undergoing TJA exposes patients to all three of these states. Venous stasis can occur intraoperatively with manipulation of the limb, kinking of the vessels, and use of the tourniquet, and postoperatively during recuperation. A certain degree of endothelial injury inevitably happens from the physical nature of surgery and the hypoxic, hypothermic conditions during tourniquet inflation.9–11 The cause of intraoperative hypercoagulability is not certain, but the 10–15% venographic incidence of DVT in the upper or contralateral limbs clearly demonstrates an iatrogenic systemic diathesis for clotting.7 Further evidence is that over 80% of DVTs after TKA are already present by the first postoperative day.12 In addition to the numerous surgery-related components of VTED, there are many patient-based risk factors as well (Box 29–1).
Sequelae Of Vted
While fatal PE is the most devastating consequence of VTED, it is relatively rare. The 90-day rate after TKA was 0.22% after almost 27,000 patients in the Scottish Registry,13 and that for symptomatic, nonfatal PE was 0.41% in over 200,000 patients in a California database.14 Approximately 25–30% of untreated symptomatic distal DVTs can propagate to proximal veins15 and are associated with a greater embolic risk.16,17 Asymptomatic VTED may not pose an immediate clinical problem, but at least three well-recognized sequelae can cause significant morbidity. Characterized by chronic venous insufficiency, pain, swelling, and recurrent ulcers, postphlebitic or postthrombotic syndrome (PTS) can develop in 25% of patients within 3 years after a DVT.18,19 Although there is no gold standard for diagnosing PTS,20 there is evidence that the incidence of developing PTS may not necessarily be significantly elevated in patients who experience DVT after TKA,21,22 and the use of thromboprophylaxis may not be effective in allaying that risk.23,24 A second concern is recurrent VTED. An initial DVT has been shown to be an independent risk factor for subsequent DVTs.25,26 Third, chronic pulmonary hypertension leading to right ventricular hypertrophy and right heart failure is a serious condition that can develop in 3.8% of patients at 2 years after an acute episode of PE.27
Modernization Of Thromboprophylaxis
Modern developments have significantly mitigated the surgery-related risk factors for VTED after TJA. Forty years ago, the average patient underwent 2.4 hours of operation, 1650 ml of blood loss, 3 units of transfusion, 1 week of bed rest, and 3 weeks of hospitalization.1,7 Warfarin was typically started 5 days postoperatively. Without thromboprophylaxis, the rates of DVT after TKA are exceedingly high, with proximal clots occurring in as many as 22% of patients at 14 days.28 The last randomized, placebo-controlled study was over 20 years ago, and it is unlikely that another would ever be ethically approved. Although the need for some form of thromboprophylaxis is rarely questioned, there is considerable debate between two main governing bodies, the American College of Chest Physicians (ACCP) and the American Academy of Orthopaedic Surgeons (AAOS), regarding the type and extent of prevention necessary after TJA.
ACCP Guidelines
The eighth edition of the Clinical Practice Guidelines on antithrombotic agents was published in 2008 by the ACCP, an organization formed by practitioners in multiple fields such as pulmonology, critical care medicine, and cardiology.29 In their analysis, the ACCP used objectively diagnosed DVT or PE as end points and examined only randomized controlled trials (RCTs) or meta-analyses of RCTs.30 Some of their stronger recommendations for TKA were: against using aspirin (acetylsalicylic acid; ASA) or low-dose unfractionated heparin as the only means of prophylaxis (Grade 1A); continuing thromboprophylaxis for at least 10 days (Grade 1A) and extending it to 35 days (Grade 2B); using a high-risk dose of low-molecular-weight heparin (LMWH) started preoperatively or postoperatively, or fondaparinux started 6–24 hours postoperatively, or warfarin started preoperatively with a target international normalized ratio (INR) of 2.5 (range, 2–3) (Grade 1A); and using intermittent pneumatic compression devices (IPCDs) in patients with a high risk of bleeding (Grade 1A).31 Because there are no prospective RCTs comparing multimodal prophylaxis to single modalities, the ACCP did not make any recommendations for or against it. Although the ACCP accepts that DVT is not a perfect surrogate for PE and that PE is the most important outcome for patients, it believes that DVT is a valid surrogate for PE based on a consistent correlation on imaging studies and parallel reduction of both entities with antithrombotic agents.30,32 In addition, because of the relative rarity of PE and death after TJA, the sample size necessary to prove a statistical difference in either entity with thromboprophylaxis would near 30,000 patients for each arm of an RCT.6,33
Risks of Bleeding
Pharmaceutical thromboprophylaxis is not without risk of bleeding. The ACCP recommendations were based on studies with no date criterion, including those published far before the introduction of modern surgical and rehabilitation protocols. Many orthopaedists feel that the ACCP unduly focused on the prevention of all VTED at the expense of iatrogenic hemorrhage.34 The challenge in achieving the appropriate balance was aptly put by Freedman et al.33 “The decision regarding prophylaxis against thromboembolic disease depends primarily on which events one is attempting to prevent: all DVTs, proximal DVTs, all PEs, fatal PEs, death, or all of the above. When considering the issue of safety, one must determine which adverse consequences are important: minor wound-bleeding, major wound-bleeding, or major nonwound bleeding (gastrointestinal or intracerebral hemorrhage).”
AAOS Guidelines
In 2008, the AAOS published their own clinical guidelines addressing specific concerns with the ACCP recommendations.35 First, only studies with patient recruitment since 1996 were included to better reflect the true underlying risk of VTED with modern protocols. Second, in addition to RCTs, large prospective cohort studies (>100 patients) were included. Third, instead of relying on the incidence of DVT as the primary efficacy outcome, prevention of symptomatic PE was used as the main goal of thromboprophylaxis. The AAOS challenged the validity and appropriateness of DVT, especially asymptomatic cases, as a proxy for PE or representative of clinical benefit.36 Fourth, a formal consideration of the benefits and harms of potent thromboprophylaxis was made in the context of TJA. Patients who return to the operating room within 30 days after TKA have a significantly increased risk for deep infection or for requiring other major surgery.37 Each day of prolonged wound drainage is associated with a 42% increased risk of infection.38 Fifth, in order to choose the appropriate aggressiveness of prophylaxis, the AAOS emphasized the need to risk-stratify each patient in terms of PE and hemorrhage. The ACCP, in contrast, considered all TJA patients to be high-risk candidates for VTED. The most important distinctions of the AAOS guidelines are the allowance of ASA as the sole chemoprophylaxis except in patients with elevated risk for PE, a lower target INR with warfarin of ≤2.0, and the avoidance of LMWH in patients with elevated risk of bleeding. In addition, mechanical prophylaxis and early mobilization are recommended in all patients.35
Warfarin
Warfarin (Coumadin; Bristol-Myers Squibb, Princeton, NJ) acts at multiple sites in the clotting cascade, inhibiting the synthesis of vitamin K–dependent factors II, VII, IX, and X. It is a time-honored method of anticoagulation, but has several drawbacks. Because of its effect on endogenous anticoagulants protein C and protein S, warfarin is initially prothrombotic, and even with early administration, the INR does not usually reach its target range until 36 hours. Unlike most other anticoagulants, warfarin requires frequent INR monitoring and close titration of doses in order to prevent catastrophic bleeding. This can be challenging in the presence of numerous diet and drug interactions. Contemporary rates of hemorrhage while on warfarin range from 1.2% to 3.7%,39–41 and at-risk medical patients have a 0.3% risk of bleeding for each additional month while on therapy.42,43 A meta-analysis of warfarin prophylaxis found a 45% incidence of DVT, 8.2% of asymptomatic PE, and 0.4% of symptomatic PE after TKA.44
Low-Molecular-Weight Heparin
Enoxaparin (Lovenox; Sanofi-Aventis, Bridgewater, NJ) and the less commonly used dalteparin (Fragmin; Pfizer, Brooklyn, NY) are LMWHs. Unlike commercial unfractionated heparin (molecular weight 12–15 kDa), enoxaparin is smaller and more uniform in weight (molecular weight 5 kDa).7 Its anticoagulant effect is mediated by enhanced inactivation of factor Xa and, to a lesser extent, factor IIa.45 Enoxaparin has many advantages, including a more predictable dose response, a dose-independent mechanism of clearance, a longer plasma half-life, and lower incidence of heparin-induced thrombocytopenia than unfractionated heparin.46 Unlike warfarin, no regular laboratory monitoring is necessary other than a baseline and an early platelet count to rule out heparin-induced thrombocytopenia. Several disadvantages are high cost (~$40 per dose) and painful route of administration (subcutaneous injection). Enoxaparin is associated with lower DVT and hemorrhage rates than unfractionated heparin.7,47 Compared to warfarin, enoxaparin is generally more efficacious after TKA in preventing distal DVT (24% vs. 34%), proximal DVT (2% vs. 11%), and pulmonary embolism (0% vs. 0.6%),48 but multiple studies4,7 have demonstrated a significantly higher incidence of major hemorrhage (5.2% vs. 2.3%) and clinically important operative site bleeding (6.9% vs. 3.4%).48
Aspirin
Aspirin is inexpensive and easy to administer, does not need monitoring, and has undergone a recent resurgence secondary to increased attention to bleeding complications from other agents. By inhibiting the production of thromboxane, ASA exerts an antiplatelet effect and is often used to prevent arterial tree embolism. For patients intolerant of ASA, dipyridamole (Persantine; Boehringer Ingelheim Pharmaceuticals, Ridgefield, CT) or clopidogrel bisulfate (Plavix; Bristol-Myers Squibb, Princeton, NJ) is typically substituted. To reduce the risk of gastrointestinal ulcer, a histamine2 antagonist (e.g., ranitidine) or proton pump inhibitor (e.g., omeprazole) is administered. The Antiplatelet Trialists’ Collaboration49 and Pulmonary Embolism Prevention Trial50 provided some evidence that aspirin may be effective in reducing VTED, but the data were somewhat confounded by multiple factors, including nonorthopaedic patients and concomitant anticoagulants.7 Although there are no recent well-controlled studies examining the relative effectiveness and efficacy of ASA,51 it is increasingly being used as one facet of a multimodal regimen. However, several large cohort studies have demonstrated its efficacy and safety in this context. When ASA (325 mg twice a day for 6 weeks) is combined with early mobilization, mechanical prophylaxis, and hypotensive epidural or regional anesthesia, rates of VTED and bleeding are exceedingly low, with 2.5–10.2% DVT, 0–0.1% fatal PE, and 0.2–0.5% hematoma or minor distant bleeding.51–53
