A study by Colin Y. L. Woon, MD; Ritesh R. Shah, MD; Brandon M. Pardi, MD; Brian E. Schwartz, MD; Jeffrey M. Goldstein, MD; Nancy E. Cipparrone, MA; Wayne M. Goldstein, MD

Abstract

Thromboembolic events after total joint arthroplasty are potentially devastating complications. This study evaluated the efficacy of 4 different anticoagulants in preventing deep venous thrombosis and pulmonary embolism after total joint arthroplasty. The demographics and anticoagulant use (warfarin, enoxaparin, and aspirin with and without outpatient mechanical pumps) for patients who under- went primary unilateral total joint arthroplasties performed by a single surgeon from January 2013 to October 2014 were retrospectively reviewed. All patients underwent lower extremity ultrasound at the 3-week postoperative visit. A total of 613 primary unilateral total joint arthroplasties met the study inclusion crite- ria. There were 288 primary total knee arthroplasties and 325 primary total hip arthroplasties. The patients were 62.2% female, having a mean age of 67.6±10.6 years and a mean body mass index of 30.2±5.9 kg/m2. There were 119 patients in group 1 (aspirin alone), 40 patients in group 2 (aspirin plus pumps), 246 pa- tients in group 3 (warfarin), and 208 patients in group 4 (enoxaparin). The overall 3-week symptomatic and asymptomatic deep venous thrombosis and symptom- atic pulmonary embolism rates in the entire cohort were 5.7% and 0.3%, respec- tively. The venous thromboembolism rate was significantly affected by the antico- agulant of choice (P<.01). Compared with aspirin alone, warfarin decreased the risk of venous thromboembolism (P<.01). Increasing age led to increased risk of venous thromboembolism (P=.05). This study indicated that aspirin chemopro- phylaxis alone was not as efficacious as warfarin and enoxaparin in preventing asymptomatic and symptomatic venous thromboembolism found during routine postoperative surveillance with lower extremity ultrasound. Aspirin alone may be inadequate and should be augmented with an outpatient mechanical pump as part of multimodal prophylaxis. [Orthopedics. 2019; 42(1):48-55.]

Results

A total of 613 patients who underwent primary unilateral TJA met the study criteria. Of these, 62.2% were women (Table 1). Mean age at surgery was 67.6±10.6 years. Mean body mass index was 30.2±5.9 kg/m2. There were 288 primary TKAs and 325 primary THAs. Surgery involved the left lower limb in 48.5% of the patients. Spinal anesthesia was used for 11.4% of the patients. Intraoperative tranexamic acid was used for 71.3% of the patients. There were 119 patients in group 1 (aspirin alone), 40 patients in group 2 (aspirin plus pumps), 246 patients in group 3 (warfarin), and 208 patients in group 4 (enoxaparin). The overall 3-week symptomatic and asymptomatic DVT and symptomatic PE rates in the entire cohort were 5.7% (n=35) and 0.3% (n=2), respectively (Figure 1). Thirty-five DVTs were reported, with 80% being infrapopliteal and 20% being suprapopliteal (Figure 2). Suprapopliteal DVTs were found in 20% of group 1 (as- pirin only), 33% of group 2 (aspirin plus pumps), 0% of group 3 (warfarin), and 27% of group 4 (enoxaparin) patients, with 1 patient having PE in groups 3 and 4, respectively (P=.038). Both patients with PE had negative findings on 3-week duplex scans and 1 patient also had a negative finding on in-hospital duplex scan. The authors found that VTE rate was significantly affected by the anticoagulant of choice (P<.01) and that the transfusion rate was significantly affected by the anticoagulant (P<.01), sex (P=.03), the use of tranexamic acid (P<.01), and the type of surgery (THA vs TKA, P<.01) (Table 2). Using a binomial logistic regression model for predicting VTE (Table 3), the authors found that, compared with aspirin alone, warfarin decreased the risk of VTE (odds ratio, 0.20; 95% confidence inter- val, 0.08-0.50; P<.01), whereas increasing age increased the risk of VTE (odds ratio, 1.04; 95% confidence interval, 1.00-1.07; P=.05). Type of anesthesia and body mass index were not significant predictors of VTE. Using a binomial logistic regression model for predicting postoperative blood transfusion (Table 4), the authors found that, compared with aspirin alone, warfarin increased the risk of transfusion (odds ratio, 8.51; 95% confidence interval, 1.08- 67.27; P=.04), whereas intraoperative intravenous tranexamic acid decreased the risk of transfusion (odds ratio, 0.42; 95% confidence interval, 0.18-0.97; P=.04). Compared with THA, TKA was also less likely to require transfusion (odds ratio, 0.17; 95% confidence interval, 0.06-0.52; P<.01).

This study was published in Blue Ribbon Articles.

Need to Get in Touch?

We’re here for you.