DVT Prevention: Aspirin, Rivaroxaban Similar After Arthroplasty

Troy Brown, RN

February 21, 2018

Among patients who underwent total hip or total knee arthroplasty, aspirin did not differ significantly from rivaroxaban (Xarelto, Janssen Pharmaceutica) for prevention of proximal deep-vein thrombosis or pulmonary embolism, after both groups received a 5-day course of rivaroxaban immediately after surgery, a large trial found.

"In our trial, we found that the inexpensive, widely available generic agent aspirin was not significantly different from the more expensive, direct oral anticoagulant rivaroxaban for the prevention of symptomatic, clinically important venous thromboembolism after total hip or total knee arthroplasty among patients who had received an initial 5-day postoperative course of rivaroxaban," write David R. Anderson, MD, from the Department of Medicine, Dalhousie University, Halifax, Nova Scotia, Canada, and colleagues. The researchers report their findings in an article published online today in the New England Journal of Medicine.

The multicenter, double-blind, randomized controlled trial, called Extended Venous Thromboembolism Prophylaxis Comparing Rivaroxaban to Aspirin Following Total Hip and Knee Arthroplasty II (EPCAT II), enrolled 3424 patients (1804 undergoing total hip arthroplasty and 1620 undergoing total knee arthroplasty). The mean age of participants was 62.8 years, and 47.8% were male.

All patients received rivaroxaban 10 mg orally once daily while in the hospital, beginning at least 6 hours after wound closure on the day of surgery or on postoperative day 1, and then daily through postoperative day 5. Those who underwent total knee arthroplasty were randomly assigned to receive an additional 9 days of rivaroxaban 10 mg or aspirin 81 mg orally once daily, beginning on postoperative day 6. Those who underwent total hip arthroplasty were randomly assigned to receive an additional 30 days of rivaroxaban 10 mg or aspirin 81 mg daily, beginning on postoperative day 6. The study drugs were given in identical-appearing capsules.

In addition, those who had been taking daily low-dose aspirin before randomization continued taking it as prescribed by their physician. At the time of randomization, 855 patients were receiving long-term aspirin prophylaxis: 372 patients (20.6%) in the total hip arthroplasty group and 483 patients (29.8%) in the total knee arthroplasty group.

Study participants were followed up for 90 days for the primary effectiveness outcome, symptomatic venous thromboembolism, and the primary safety outcome, bleeding complications, including major or clinically relevant nonmajor bleeding.

During the 90-day follow-up period, 11 (0.64%) of 1707 patients in the aspirin group and 12 (0.70%) of 1717 patients in the rivaroxaban group developed symptomatic proximal deep-vein thrombosis or pulmonary embolism (difference, 0.06 percentage points; 95% confidence interval [CI], −0.55 to 0.66).

Aspirin was noninferior (P < .001) but not superior (P = .84) to rivaroxaban for preventing postoperative proximal deep-vein thrombosis or pulmonary embolism, which was the primary endpoint.

One patient in the aspirin group who underwent total knee arthroplasty died from pulmonary embolism 31 days after randomization and 17 days after completing aspirin prophylaxis. No other patients died during the trial.

"There were no between-group differences in effectiveness in the subgroup of patients who were receiving long-term aspirin therapy, which suggests that there was no benefit of adding 81 mg of aspirin to either aspirin or rivaroxaban prophylaxis," the authors explain.

Bleeding Events

Rates of thromboembolism, major bleeding, and clinically relevant nonmajor bleeding were not significantly different among the 855 patients who were and the 2569 patients who were not receiving long-term aspirin therapy.

"[T]here were suggestions of more major and clinically relevant nonmajor bleeding among patients in the long-term aspirin subgroup, particularly among those who had been assigned to the aspirin group and hence were receiving a second daily dose of aspirin prophylaxis," the researchers write.

Eight patients (0.47%) in the aspirin group and five patients (0.29%) in the rivaroxaban group experienced major bleeding events (difference, 0.18 percentage points; 95% CI, −0.65 to 0.29; P = .42). Twenty-two patients (1.29%) in the aspirin group and 17 patients (0.99%) in the rivaroxaban group experienced a combination of major bleeding and clinically relevant nonmajor bleeding (difference, 0.30 percentage points; 95% CI, −1.07 to 0.47; P = .43).

All bleeding events were overt hemorrhage that occurred at the surgical site, and most occurred within 10 days after randomization.

Limitations of the trial include the frequently postoperative recruitment of patients, which meant that not the entire patient population was treated according to a standardized protocol. Therefore, it was not possible to calculate the absolute event rates of thromboembolism or bleeding complications connected to each of the two prophylaxis approaches. Also, as most bleeding events related to surgical site bleeding happened after randomization, it is hard to know whether the bleeding was mostly related to the initial postoperative rivaroxaban, the trial drug, or both.


"This large, randomized, double-blind trial is likely to be practice-changing, because the point estimates for the relative safety and efficacy of the two treatments are precise and the included patients represent a typical population undergoing joint-replacement surgery," David Garcia, MD, from the Department of Medicine, University of Washington, Seattle, writes in an accompanying editorial.

As the randomized population included relatively few participants with previous venous thromboembolism, morbid obesity, or cancer, it is unclear how effective the two prophylaxis strategies would be in very high risk populations such as these, Garcia notes.

"However, because the EPCAT II trial was conducted in centers in which postoperative mechanical thrombosis prophylaxis was not commonly performed, only about 15% of the enrolled patients received such prophylaxis with pneumatic compression, graduated stockings, or both. Thus, it is possible that even high-risk patients would have acceptably low rates of thromboembolism if mechanical prophylaxis were combined with the medical regimens used in the EPCAT II trial," Garcia explains.

More than half of the patients received perioperative tranexamic acid, an antifibrinolytic agent that promotes hemostasis. Despite that, Garcia says the trend toward more bleeding among patients on continued aspirin therapy should be a reminder to use doses below 100 mg.

"Going forward, the very low rates of bleeding and thrombosis seen with the relatively inexpensive and user-friendly aspirin-based strategy probably mean that the EPCAT II trial has established a prophylaxis regimen against which all strategies to prevent venous thromboembolism after joint replacement will be compared," Garcia concludes.

However, Kevin Bozic, MD, American Academy of Orthopaedic Surgeons spokesperson, says the study is unlikely to be practice changing overall. Bozic is also chair of the American Joint Replacement Registry and a specialist in hip and knee replacement.

"[The findings are] very consistent with current practice. The American Academy of Orthopaedic Surgeons published a clinical practice guideline 4 years ago which looked at the body of evidence that existed at that time, and we said there was no evidence that said that one pharmacologic approach was better than another pharmacologic approach, but having a pharmacologic approach was better than not having a pharmacologic approach, which is very consistent with this study," Bozic told Medscape Medical News.

"Patients should be on a drug, but it's probably not important which drug they're on. So I don't think that it's practice changing in that regard. The other thing is the clinical approach that was used in this study — using 5 days of rivaroxaban and then switching to aspirin or rivaroxaban — is actually not a common practice. The vast majority of hip and knee surgeons in the United States, for simple primary hip or knee replacement patients that aren't medically complex, are treated with aspirin from day 1, as opposed to rivaroxaban first. Rivaroxaban is another strategy, as is low molecular weight heparin, and there are other strategies as well," Bozic explained.

"Physicians who use aspirin for VTE prophylaxis now will continue to use aspirin, and potentially the group of physicians who use rivaroxaban may switch them after 5 days, so in that regard I guess you could call it practice changing, but I think the percentage of physicians that currently use rivaroxaban as their primary VTE prophylaxis agent is small," he added.

Treatment Strategies Should Evolve as Practice Patterns Change

"Having randomized controlled trials, which are very difficult to do, and expensive and time-consuming, is critically important in helping us understand and evaluate important clinical questions like this. As practice evolves, it's important to reevaluate strategy, so what may have been appropriate for VTE prophylaxis 10 or 15 years ago may not be appropriate today because surgical techniques have changed, rehabilitation protocols have changed; for instance, patients are getting up and mobilizing on the day of surgery, and being discharged home on the day of surgery in many cases, which significantly reduces their risk for venous thromboembolism complications, whereas in an era when patients were staying in the hospital and staying in bed for many days, that increased their risk for VTE complications," Bozic continued.

"It's important to reevaluate treatment strategies as practice patterns change," Bozic concluded.

Anderson has disclosed no relevant financial relationships. Several authors have disclosed a variety of financial relationships including receipt of lecture fees, grant support, consulting fees, drugs supplied for a study, fees for preparation and presentation of educational materials, advisory board fees, fees for serving on a steering committee, travel support, fees for providing expert testimony, fees for serving on a data and safety monitoring board, stock ownership, royalties, and honoraria, from companies including Bayer Pharma, Bristol-Myers Squibb, Pfizer, Leo Pharma, Boehringer Ingelheim, Pendopharm, Victhom, Bodycad, Sanofi, Smith & Nephew, Stryker, Canadian Institutes of Health Research, Ministére de la Santè et des Services Sociaux, Octapharma, Shionogi, Alexion, Portola, Leo Pharma, McCarthy Tétrault, Daiichi Sankyo, Alnylam, DePuy Synthes, Zimmer Biomet, and MicroPort. The remaining authors have disclosed no relevant financial relationships. Garcia reports a variety of financial relationships including personal fees, grant support, and nonfinancial support from Genzyme Corporation, Pfizer, Boehringer Ingelheim, and Alexion, Bristol-Myers Squibb, Janssen, Incyte, Bayer, and Daiichi Sankyo. Bozic has disclosed no relevant financial relationships.

N Engl J Med. Published online February 22, 2018.

For more news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.