In this month's Journal, the unanswered question of duration of prophylaxis for venous thromboembolic disease following elective total joint arthroplasty is addressed. The longstanding issue of which agent to use for such prophylaxis has been superceded in contemporary practice by the question of how long such therapy should be administered. The importance of this question is underscored by the currently standard four-day hospital stay for hip and knee arthroplasty coupled with a recognition of the protracted risk of thromboembolic disease that extends conservatively for three months following the index procedure.

Comp and colleagues present a timely study of the efficacy of extended postdischarge fractionated heparin prophylaxis against thromboembolic disease following elective primary total hip and knee arthroplasty¹. In a well-designed prospective, controlled trial, three weeks of a single 40-mg daily dose of enoxaparin reduced the venographic prevalence of deep vein thrombosis following total hip arthroplasty from 23.2% (49/211) in the placebo group to 8.0% (18/224; odds ratio 3.62; p < 0.001). A similar reduction in proximal venous thrombosis was observed after hip replacement, but significant efficacy was not generally seen after total knee arthroplasty. Perhaps most importantly, readmission for symptomatic thromboembolic events was reduced from 10.4% (22/211) in placebo-treated patients to only 1.3% (3/224) in enoxaparin-treated patients. Once again, no meaningful reduction in readmission for thromboembolic disease was observed in the knee arthroplasty cohort with rates of 5.4% (12/221) and 3.3% (7/217) in the placebo group and the enoxaparin group, respectively. Three patients, all randomized to the postdischarge placebo group, experienced symptomatic pulmonary emboli, two following knee replacement and one after hip arthroplasty.

While this is a well-done trial, it is important to note a few points regarding the article as it appears in The Journal. The authors suggest that the aggregate rate of venous thromboembolic disease for both hip and knee arthroplasty is significantly reduced by extended enoxaparin treatment, even though the specific statistical analysis of the subset of knee arthroplasty patients did not demonstrate a significant reduction in the prevalence of deep venous thrombosis. Such a conclusion, based on commingling of the knee and hip arthroplasty data, is misleading. Indeed, the nature of venous thromboembolic disease after hip arthroplasty is quite different from that after knee arthroplasty; hence the need to stratify such groups in a prospective trial. Furthermore, questions about compliance with the unsupervised administration of the study drug at home leave doubts as to the actual amount of drug received by subjects in the study. Finally, the use of continuous passive motion in an uncontrolled manner in the knee replacement cohort introduces an element of inconsistency with an intervention that has been shown by some investigators to independently reduce the prevalence of deep venous thrombosis after knee replacement. Notwithstanding these concerns, the study presents valid and timely data to the orthopaedic community and is appropriately provocative in raising the question of the duration of routine prophylaxis for venous thromboembolic disease after total joint arthroplasty.

Prior studies have demonstrated that the rate of formation of new venographic thrombi is approximately 20% three weeks after discharge following total hip arthroplasty^2-6. In addition to demonstrating the benefit of protracted routine prophylaxis in reducing the prevalence of thromboembolic disease after total hip arthroplasty, the failure of this study to demonstrate similar efficacy following identical intervention for knee replacement patients underscores the relatively refractory nature of venous thromboembolism to conventional measures after knee arthroplasty. Indeed, the prevalence of persistent deep venous thrombosis after any contemporary prophylaxis, pharmacologic or mechanical, following knee replacement remains three to five times greater than that observed after hip arthroplasty. The nature of this differential efficacy for hip and knee replacement remains unknown. In this study and others, the prevalence of deep venous thrombosis after hip arthroplasty is reported to be below 10% while that following knee replacement with the same prophylaxis remains between 30% and 50%. Not only has asymptomatic venous thromboembolic disease after knee arthroplasty been found to be refractory to conventional prophylaxis, but readmission for symptomatic events was similarly less benefited by extended prophylaxis than after hip replacement. Venous thromboembolic disease following total knee arthroplasty is a unique disease deserving of independent study and management. The present study clearly demonstrates the benefit of prolonged postdischarge prophylaxis in reducing asymptomatic venous thromboembolism after total hip arthroplasty and suggests that it is preferable to the historical practice of solely in-hospital prophylaxis. However, this trial does not provide a rigorous case for extended prophylaxis to be adopted as the standard of care in management of clinical outcomes. Indeed, optimal management of thromboembolic risk would ideally weigh the benefits of prophylaxis against the risks of extended anticoagulation for all patients. An alternative strategy of selective anticoagulation for only documented deep venous thrombosis diagnosed by screening studies at the time of hospital discharge would limit the number of patients on outpatient anticoagulants and therefore reduce the risk of related bleeding events⁷. This concept of selective "secondary" prophylaxis enjoys even greater advantage with increasing efficacy of the primary prophylactic intervention. Specifically, with such a strategy after total hip arthroplasty, as the prevalence of residual deep venous thrombosis falls below 10% with fractionated heparins or warfarin in conjunction with epidural anesthesia, 90% of patients without documented thrombi would be free of outpatient anticoagulation at discharge. The natural history of such a cohort of patients discharged after normal contrast venography and no further anticoagulation has been reported^8,9. In a series of 546 patients who underwent screening contrast venography after total joint arthroplasty, 382 had negative studies and were discharged home without further anticoagulation; three patients (0.78%) were readmitted during the ensuing six months for symptomatic thromboembolic events (two deep venous thromboses and one nonfatal pulmonary embolism)⁹. No patients with positive venograms, all of whom were managed with therapeutic warfarin anticoagulation, experienced related thromboembolic complications. Hence the readmission rate for symptomatic thromboembolic events was 0.58% (3/513), which compares favorably with the overall readmission rate of 2.3% (10/441) reported in this issue of The Journal by Comp and colleagues. Furthermore, this strategy of selective secondary outpatient anticoagulation therapy enjoys the additional advantage of reducing outpatient anticoagulation exposure by 90% in the total hip arthroplasty population, with the attendant benefit of a decrease in the associated bleeding hazard in these elderly patients.

Routine prophylaxis is accepted in North America as the standard of care for arthroplasty patients; now the focus must turn to the duration of anticoagulant therapy to balance the related risks and benefits to our patients. Comp and colleagues should be commended for directing our attention to the urgency of this issue in contemporary orthopaedic practice.

1. Comp PC, Spiro TE, Friedman R, Whitsett TL, Johnson GJ, Gardiner GA, Landon GC, Jove M, for the Enoxaparin Clinical Trial Group. Prolonged enoxaparin therapy in the prevention of venous thromboembolic disease in patients undergoing primary hip or knee replacement. J Bone Joint Surg Am. 2001;83:336-345.

2. Bergqvist D, Benoni G, Bjorgell O, Fredin H, Hedlundh U, Nicolas S, Nilsson P, Nylander G. Low molecular-weight heparin (enoxaparin) as prophylaxis against venous thromboembolism after total hip replacement. N Engl J Med. 1996;335:696-700.

3. Dahl OE, Andreassen G, Muller C, et al. The effect of prolonged thromboprophylaxis with dalteparin on the frequency of deep vein thrombosis and pulmonary embolism 35 days after hip replacement surgery. Thromb Haemost. 1995;73:1094:A-743.

4. Lassen MR, Borris LC. Prolonged thromboprophylaxis with low molecular weight heparin after elective total hip arthroplasty: A placebo controlled study. Thromb Haemost. 1995;73:1094:A-781.

5. Planes A, Vochelle N, Darmon JY, Fagola M, Bellaud M, Huet Y. Risk of deep-venous thrombosis after hospital discharge in patients having undergone total hip replacement: double-blind randomised comparison of enoxaparin versus placebo. Lancet. 1996;348:224-8.

6. ten Cate JW, Prins MH. Major orthopaedic surgery and post-discharge DVT. Lancet. 1996;348:209-10.

7. Pellegrini VD Jr. Prevention of thromboembolic disease in patients after total joint arthroplasty by selective posthospitalization treatment. Sem Arthroplasty. 1997;8:248-57.

8. Pellegrini VD, Clement D, Lush-Ehmann C, Keller GS, Evarts CM. The John Charnley Award. Natural history of thromboembolic disease after total hip arthroplasty. Clin Orthop. 1996;333:27-40.

9. Pellegrini VD Jr, Clement D, Lush?Ehman C, Keller GS, Totterman S, Evarts CM. Venogram surveillance of deep venous thrombosis following total joint arthroplasty. Orthop Trans. 1996;20:70.