Rationale

Major orthopedic procedures including hip and knee arthroplasty and hip fracture surgery confer the highest risk for venous thromboembolic events and remain a challenge globally.¹

Mechanical compression methods – including graduated compression stockings, intermittent pneumatic compression, and foot‐pumps – are widely available, variable in expense, and have very few contraindications. These measures can be employed as monotherapy in patients with contraindications to anticoagulant therapy, or in conjunction with anticoagulants in higher‐risk patients. The 2012 American College of Chest Physicians (ACCP) guidelines recommend the use of several anticoagulants for orthopedic surgery.² These include low‐dose unfractionated heparin (LDUH), low‐molecular‐weight heparin (LMWH), more recent novel oral anticoagulants, aspirin, and vitamin K antagonists (VKAs), exemplified by warfarin.

New oral anticoagulants (NOACs) are broadly divided into direct thrombin inhibitors (dabigatran) or direct factor Xa inhibitors (fondaparinux, rivaroxaban, and apixaban).

Clinical comment

Evidence has been contradictory regarding the effectiveness of mechanical thromboprophylaxis.

LDUH has largely been surpassed by the LMWHs. NOACs may have even greater thrombo‐prophylactic efficacy and ease of administration, but this is balanced by a higher rate of bleeding events. The ideal anticoagulant should have high efficacy, safety, low levels of bleeding, rapid onset of action, fixed dosing, and no requirement for therapeutic monitoring.

Available literature and quality of the evidence

• Systematic reviews/meta‐analyses: 2 (level I).
  
• Systematic reviews/meta‐analyses with methodological limitations: 1 (level II).
  
• Case series: 2 (level IV).

Findings

Pooled data from low‐quality randomized controlled trials (RCTs) comparing mechanical compression to no thromboprophylaxis show a relative risk reduction of >50% for both DVT and PE in arthroplasty and hip fracture surgery (pulmonary embolism [PE] risk ratio [RR] = 0.4, 95% confidence interval [CI]: 0.17–0.92; DVT RR = 0.46; 95% CI: 0.35–0.61).²

A meta‐synthesis identified six good‐quality systematic reviews that compared anticoagulants with LMWH.³ The risk for symptomatic DVT was reduced with factor Xa inhibitors compared to LMWH (four fewer events per 1000 patients), albeit with an increase in major bleeding events (two per 1000 patients). Dabigatran had similar outcomes to LMWH. Conclusions about differences between NOACs could not be ascertained. In a meta‐analysis of six RCTs on arthroplasty patients, the combination of pharmacologic and mechanical prophylaxis conferred a lower risk for DVT (relative risk 0.48, 95% CI 0.32–0.72) compared to pharmacologic prophylaxis alone.⁴

Asian patients undergoing hip and knee arthroplasty have been shown to have a noticeable low prevalence of DVT and PE. A recent large case series highlighted the prevalence of DVT at 6.6%, and proximal DVT of 0.4%, with no PE in patients undergoing total knee arthroplasty with only mechanical prophylaxis.⁵ Another case series in Asian patients undergoing total hip arthroplasty with only mechanical prophylaxis revealed a DVT prevalence rate of 4.8%, 1.6% with proximal DVT, 0.7% with asymptomatic PE, and no symptomatic PE.⁶ The authors recommend mechanical compression devices only in Asian patients.