Copyright © 2007 by The Journal of Bone and Joint Surgery, Inc.
Commentary & Perspective
Commentary & Perspective by
Vincent D. Pellegrini Jr., MD*,
University School of Medicine, Baltimore, Maryland
Posted December 2007
The subject of venous thromboembolism following total hip and knee arthroplasty is a highly emotionally charged and controversial matter. Over the past decade, guidelines promulgated by the American College of Chest Physicians1 have added fuel to the fire. Their recommendations include use of routine prophylaxis with warfarin at a target international normalized ratio of 2 to 3, fractionated heparins, or pentasaccharide. This has evoked alarm from the orthopaedic community because of the associated problem of perioperative bleeding complications related to these more intensive anticoagulants. Chest physicians prioritize the effectiveness of venous thromboembolism, while orthopaedists give deference to avoidance of postoperative hematoma and bleeding. As is typically the case, the truth probably lies somewhere between these two polar extremes, which are separated by a great deal of emotion and confusing data.
In response to this perceived imbalance between the apparent risks and benefits of routine prophylaxis, many orthopaedic investigators as well as the American Academy of Orthopaedic Surgeons have attempted to shift the emphasis of venous thromboembolism prophylaxis to the prevention of symptomatic events (pulmonary embolism and proximal thrombosis) rather than asymptomatic but venographically diagnosed disease. This is a noble and appropriate undertaking—by defining an adverse thromboembolic end point as a clinically meaningful event, it is hoped that a more reasonable risk-to-benefit ratio might be achieved. Hence the use of aspirin as primary prophylaxis has again surfaced. This exercise, however, is dependent upon even more rigorous methodological discipline than previously found in the orthopaedic literature because the prevalence of adverse bleeding or symptomatic venous thromboembolism is so much smaller than the prevalence of venographic disease. An unintended consequence is that the differentiation between useful information and confusing data becomes more difficult; unfortunately, the article by Dorr and colleagues provides data that serve more to confuse than clarify the matter.
The authors contend that a multimodal approach to prophylaxis (antiplatelet therapy, intermittent pneumatic compression devices, and early mobilization) in "low-risk" patients is "consistent with protecting patients while limiting adverse clinical outcomes" and conclude that "safe and effective prophylaxis can be achieved with use of multimodal therapy." They report on an uncontrolled retrospective review of 1179 consecutive total hip and knee arthroplasties that represent a roughly equal number of hip and knee procedures, approximately 19% of which were revision procedures, performed over a nineteen-month period. Interventions were varied and inconsistent and serve to illustrate the shortcomings of a retrospective analysis. Antiplatelet therapy in the "low-risk" group consisted of aspirin (325 mg orally twice each day) after 999 procedures and dipyridamole or clopidogrel after forty-seven procedures; use of such drugs was discontinued preoperatively only five days before operation, which is an insufficient time (by five days) to allow inactive acetylated platelets to be replaced by those with normal function. Conversely, discontinuing warfarin in the chronic user five days before operation leaves these patients in an unprotected hypercoagulable state owing to protein C and S depletion; their conditions are highly thrombogenic in the immediate preoperative period and they are at risk of forming thrombi even before the operation. Pneumatic compression was accomplished either with calf compression sleeves or foot pumps, but each has been shown to have different efficacy based on variable resulting flow velocities. "High-risk" patients did not start fractionated heparin or warfarin for up to forty-eight hours postoperatively and received only aspirin until that time, delaying effective anticoagulation with these agents by three to four days after operation. Epidural anesthesia was used in 84% of patients but no stratification was provided with respect to anticoagulation and adverse events for those who did not have this benefit. Perhaps most importantly, follow-up was obtained on all patients but information on "any venous thromboembolic complications that occurred after discharge from the hospital" was obtained from "review of office charts, and direct communication with patients when necessary." Retrospectively ascertaining the occurrence of pulmonary embolism is challenging under the best of circumstances, especially considering that readmission is not necessarily to the same hospital where the orthopaedic procedure was performed, but deriving this information primarily from a retrospective review of the office records of an orthopaedic specialty practice is highly suspect.
Analysis of the data shows significant differences only in asymptomatic proximal deep venous thrombosis discovered on ultrasound surveillance; six (4.5%) of 133 procedures in the high-risk group compared with fifteen (1.4%) of 1046 procedures in the low-risk group (p = 0.029). It is unclear whether this difference should be expected since, after all, these were "high-risk" patients, or is evident because aspirin prophylaxis was more effective in the "low-risk" group, or whether thrombi developed in these "high-risk" patients because anticoagulation was effectively delayed for up to four days. Moreover, critics of anticoagulation repeatedly contend that asymptomatic venous thrombosis is a poor surrogate for measuring efficacy in preventing pulmonary embolism. Most importantly, with respect to the clinically evident end point of pulmonary embolism, the inability to demonstrate a significant difference between the low and high-risk prophylaxis groups does not prove equivalence of the different regimens. The most significant finding and major thrust of the paper was the greater safety of aspirin multimodal prophylaxis as evidenced by the fact that all five (five of 1179 or 0.4%; p = 0.0001) wound hematomas occurred in patients on warfarin (two patients) or fractionated heparin (three patients). However, the authors report that three of these patients had previously been on aspirin; quite possibly the continued platelet-inhibiting effect of preoperative aspirin, discontinued only five days before operation, was additive with that of the warfarin or fractionated heparin and contributed to the hematomas in this group.
The authors should be commended for their interest and efforts in this challenging arena, but in an era of evidence-based medicine, their work does not support their conclusions and moves us no closer to the holy grail of ideal venous thromboembolism prophylaxis after total joint arthroplasty. Support, however, is easy to enlist for their closing statement; "it is time to perform a prospective, randomized, multicenter trial . . . " Who could disagree with that?
*The author did not receive any outside funding or grants in support of his research for or preparation of this work. Neither he nor a member of his immediate family received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, division, center, clinical practice, or other charitable or nonprofit organization with which the author, or a member of his immediate family, is affiliated or associated.
Reference
1. Callaghan JJ, Dorr LD, Engh GA, Hanssen AD, Healy WL, Lachiewicz PF, Lonner JH, Lotke PA, Ranawat CS, Ritter MA, Salvati EA, Sculco TP, Thornhill TS; American College of Chest Physicians. Prophylaxis for thromboembolic disease: recommendations from the American College of Chest Physicians—are they appropriate for orthopaedic surgery? J Arthroplasty. 2005;20:273-4.
|
