Venous thromboembolism, specifically pulmonary embolism, remains the most common cause of readmission and death after elective total hip and knee arthroplasty. Despite refinements in surgical technique, advances in intra-operative anesthetic management and post-operative analgesia that allow early mobilization, and expedited rehabilitation protocols with abbreviated hospital stays, the threat of fatal pulmonary embolism remains in 0.1 to 0.5% of patients. Accordingly, considerable effort has been directed at defining appropriate prophylaxis for this patient population. The use of potent anticoagulants in the perioperative period must necessarily be tempered by a consideration of the bleeding risk following major orthopaedic procedures. Consequently, the ideal chemoprophylaxis is yet to be determined and will represent a balance between mitigation of the risk of fatal pulmonary embolism and the morbidity of hematoma and secondary infection resulting from bleeding associated with anticoagulant use.

Fractionated heparins, synthetic pentasaccharides, and direct activated factor Xa inhibitors have all demonstrated substantial efficacy in reduction of venographic thrombosis when utilized as primary chemoprophylaxis after hip and knee arthroplasty. Low intensity warfarin (INR 2.0) has been associated with a prevalence of residual venographic clot that is 2 – 5 times greater than with these newer agents, yet it is highly effective as secondary prophylaxis of clinical pulmonary embolism with the benefit of major bleeding complications that are 2-3 times less frequent than with these more potent anticoagulants. Accordingly, the orthopaedic community has favored a strategy that offers a lesser bleeding risk while providing comparable protection against clinical thromboembolic events and has been slow to accept routine use of these newer agents. Warfarin remains a popular method of chemoprophylaxis for the orthopaedic surgeon in North America and, in the absence of extensive clinical event data for meaningful pulmonary embolism, many surgeons favor aspirin as prophylaxis based largely on the strength of its negligible bleeding risk.

Federal regulatory agencies (JCAHO, NQF, CMS) now mandate assessment for, and institution of, venous thromboembolism prophylaxis after total joint replacement according to prevailing guidelines. The American College of Chest Physicians (ACCP) recommends use of potent anticoagulants, with their attendant bleeding risk, while the American Academy of Orthopaedic Surgeons (AAOS) pays greater deference to avoidance of bleeding and accepts the use of aspirin and low intensity warfarin in patients undergoing total joint replacement without extraordinary risk of thromboembolism. In contrast to the substantive data for venographic thrombosis, the occurrence of clinical pulmonary embolism has not been convincingly shown to be materially different with any of these regimens while bleeding is clearly more frequent with more intensive anticoagulation as recommended by the ACCP.

In our own experience with 3,293 total hip and knee replacement patients studied prospectively over 20 years (1984-2003) assessing contrast venography as a screening tool for clinically meaningful thromboembolic events, extended low intensity (INR 2.0) warfarin prophylaxis was highly effective in prevention of readmission for pulmonary embolism. Of all patients discharged on warfarin for any reason, including a positive screening venogram for deep vein thrombosis, 0.2% (2/844) were readmitted for clinically significant pulmonary embolism or symptomatic proximal deep vein thrombosis compared with 1.6% (38/2449) readmissions in patients discharged without warfarin based on a negative screening venogram (p=0.001). No pulmonary embolism occurred in any patient on extended warfarin prophylaxis (0/844) compared with a 0.7% (17/2449) incidence of pulmonary embolism, including 4 fatal events (4/3293; 0.12%), among patients not on extended outpatient warfarin (p=0.01). One (1/1576; 0.06%) major bleed, a fatal intracranial event, occurred on outpatient warfarin therapy.

Until such time as the ideal chemoprophylaxis is identified, providing for elimination of pulmonary embolism along with a negligible bleeding risk, we favor six weeks of low intensity (INR 2.0) warfarin without routine screening as a “therapeutic compromise” in all patients after total hip and knee arthroplasty.