Commentary & Perspective

Commentary & Perspective on
"Oral Direct Thrombin Inhibitor Ximelagatran Compared with Warfarin for the Prevention of Venous Thromboembolism After Total Knee Arthroplasty"
by Clifford W. Colwell Jr., MD, et al.

Commentary & Perspective by
Vincent D. Pellegrini Jr., MD*,
University of Maryland School of Medicine, Baltimore, Maryland

Ximelagatran: Are We Keeping Our Eye on the (Right) Ball?

Colwell, Francis, and colleagues in the EXULT B study group have presented results of a well-designed prospective randomized clinical trial and purport that the oral direct thrombin inhibitor, ximelagatran, provides superior efficacy to warfarin in reducing overall venous thromboembolic events or death from all causes (22.5% compared with 31.9%, respectively; p < 0.001) following primary total knee arthroplasty performed in nearly 2000 patients. As a direct thrombin inhibitor, ximelagatran requires no monitoring or dose adjustment, is administered orally, and is an effective inhibitor of both free and clot-bound thrombin. On the strength of ximelagatran’s relative risk reduction of nearly 30% over warfarin, its practical convenience, and the fact that no significant difference was observed with regard to bleeding complications, it is quite possible that ximelagatran is the next in a recent parade of “new and improved” anticoagulants for venous thromboembolic disease after total joint arthroplasty.

Indeed, no drug is without unintended adverse consequences, and this one is no exception. Elevation of liver enzymes has been noted with a more longstanding administration of ximelagatran, and this concern was largely responsible for the recent failure of this agent to gain approval by the Food and Drug Administration in the United States. Nonetheless, in this trial, transaminase abnormalities were observed in only 0.5% of patients on the study drug at the conclusion of treatment and all elevated values normalized within four weeks. Of greater concern is the prevalence of bleeding events. While not found to be significantly different in the two groups, the occurrence of major bleeding events was nearly 2.5 times more common with ximelagatran than with warfarin (1.0% compared with 0.4%, p = 0.087; Table III). Moreover, the events included a fatal gastrointestinal bleed in one patient taking ximelagatran, four more gastrointestinal bleeds with ximelagatran compared with one with warfarin, three wound bleeds with ximelagatran compared with one with warfarin, and one expanding chronic subdural hematoma after a fall in a patient taking ximelagatran. While the authors dismiss these differences as not meeting statistical significance, one would not expect a statistical difference in a 1% event from a trial that was powered with 860 patients in each group to discern a 25% risk reduction in an event (venous thromboembolism) that is twenty to thirty times more common than major bleeding. Indeed, were the observed rates of bleeding to continue with the enrollment of only 200 additional patients in each group, the difference in major bleeding events with ximelagatran would have attained statistical significance at p < 0.05. The failure of this trial to discern any difference in major bleeding events between warfarin and ximelagatran can be arguably attributed to a study inadequately powered to answer this question rather than to any lack of real difference in bleeding between the drugs.

Of greater importance, however, is the actual derivation of the observed statistical significance of the primary trial end point and its clinical relevance to the practicing orthopaedist. Despite the observation of a significant reduction in total venous thromboembolic events detected by venography and in deaths from all causes, there was no discernible difference in the prevalence of pulmonary embolism (0.2% compared with 0.4%) or proximal deep venous thrombosis (3.1% compared with 3.4%) between ximelagatran and warfarin, respectively. Moreover, five of seven deaths with ximelagatran (one gastrointestinal bleed and four possible pulmonary emboli) compared with none of three deaths with warfarin could be attributed by the central adjudication committee to venous thromboembolism or bleeding. Yet, it is only in the next-to-last sentence of the last paragraph that the authors acknowledge that the observed statistical significance in the primary end point between the two study drugs is attributable to a pronounced reduction in deep calf-vein thrombosis with ximelagatran (21.4% compared with 31.1%; p < 0.001). With no observed differences in proximal deep venous thrombosis or pulmonary embolism, and more worrisome deaths with ximelagatran than with warfarin, one must wonder whether a significant reduction in calf-vein thrombosis attributable to ximelagatran is of any meaningful clinical consequence. Longitudinal studies have shown that the risk of venous thromboembolic complications after total joint arthroplasty continues for three months after operation1. Moreover, patients discharged from the hospital without continued anticoagulation on the basis of a negative screening study were eight times more likely to be readmitted with a clinically apparent thromboembolic event than those on outpatient warfarin treatment of calf-vein thrombosis diagnosed by venography2. Indeed, it seems that routine surveillance for deep venous thrombosis after total hip replacement is therefore unreliable and possibly even misleading, given that a negative screening study encourages discontinuation of anticoagulant prophylaxis at the time of hospital discharge. Recent guidelines from the American College of Chest Physicians recommend against routine surveillance studies and advocate for extended prophylaxis against deep venous thrombosis after total hip or knee arthroplasty3.

If current wisdom deems that extended chemoprophylaxis after total joint arthroplasty is prudent, then the value of a reduction in calf-vein thrombosis at the time of discharge is increasingly difficult to appreciate. Continued anticoagulation prevents clinically important extension and embolization of deep venous thrombi, whether present at the time of hospital discharge or not. Therefore, it would seem that we should be asking a different question. The efficacy of new agents should be measured by their ability to decrease readmission rates that are associated with clinically evident thromboembolic events. Reduction of venographic thrombi, and certainly reduction of only calf thrombi, at the time of hospital discharge is no longer an adequate surrogate for effective thromboprophylaxis. With a reduction in calf thrombi only, the current recommendations for extended prophylaxis, an absence of data on late readmissions for these patients, and a bleeding rate for ximelagatran that is nearly 2.5 times that of warfarin, this study does not justify abandoning warfarin in favor of ximelagatran as standard thromboprophylaxis after elective total knee arthroplasty.

*The authors did not receive grants or outside funding in support of their research or preparation of this manuscript. One or more of the authors received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity (Dr. Pellegrini received only research support in the 1980s and 1990s from Kabi Pharmacia, Sanofi-Synthélab, and The Kendall Company). No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the authors are affiliated or associated.


1. Pellegrini VD Jr, Clement D, Lush-Ehmann C, Keller GS, Evarts CM. Natural history of thromboembolic disease after total hip arthroplasty. Clin Orthop Rel Res. 1996;333;27-40.
2. Pellegrini VD Jr, Donaldson CT, Farber DC, Lehman EB, Evarts CM. Prevention of readmission for venous thromboembolic disease after total hip arthroplasty—The John Charnley Award. Presented at the Annual Meeting of the Hip Society; 2005 Feb 26; Washington, DC. (In press, Clin Orthop Rel Res.)
3. Geerts WH, Pineo GF, Heit JA, Bergqvist D, Lassen MR, Colwell CW, Ray JG. Prevention of venous thromboembolism: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest. 2004;126(3 Suppl):338S-400S.