Rivaroxaban Beats Aspirin in Reducing Risk of Recurrent VTE, Even at Lower Doses

Patients with venous thromboembolism (VTE) require anticoagulation, but concerns about bleeding have complicated decisions about using higher- or lower-intensity anticoagulation or aspirin for subsequent long-term preventive treatment. The randomized, double-blind, phase III EINSTEIN CHOICE (Reduced-dosed Rivaroxaban in the Long-term Prevention of Recurrent Symptomatic Venous Thromboembolism) trial, led by Jeffrey I. Weitz, MD, of the Thrombosis and Atherosclerosis Research Institute and McMaster University in Hamilton, Ontario, Canada, found that the use of rivaroxaban (at either 10 or 20 mg) reduced the risk of recurrent VTE more than aspirin, without significantly increasing bleeding rates.1

“Many patients with VTE benefit from extended anticoagulation but are reluctant to take it because of the fear of bleeding – opting to take aspirin instead,” Dr. Weitz explained to ASH Clinical News. “This study shows that rivaroxaban is superior to aspirin and is associated with a similar bleeding risk. Therefore, unless cost is the issue, rivaroxaban is a better choice than aspirin for extended treatment.”

The study, which was published in the New England Journal of Medicine, enrolled 3,396 patients from 244 sites in 31 countries; 3,365 patients were included in the primary analysis. Adult patients were included if they had objectively confirmed symptomatic proximal deep vein thrombosis or pulmonary embolism and were treated for 6-12 months with an anticoagulant, including a vitamin K antagonist (VKA) or direct oral anticoagulant (DOAC) such as dabigatran, rivaroxaban, apixaban, or edoxaban.

Patients were excluded from the study if they had a contraindication to continued anticoagulant therapy, required extended anticoagulant therapy at therapeutic doses or antiplatelet therapy, or had a creatinine clearance of <30 mL/min or hepatic disease associated with a coagulopathy.

Patients were enrolled at least 24 hours after receiving their last dose of a DOAC or, if they were receiving a VKA, when the international normalized ratio was ≤2.5. Patients were randomized 1:1:1 to receive:

  • rivaroxaban 20 mg (n=1,107; median age = 59 years; range = 48-69 years)
  • rivaroxaban 10 mg (n=1,127; median age = 60 years; range = 48-69 years)
  • aspirin 100 mg (n=1,131; median age = 60 years; range = 48-69 years)

Each agent was administered daily with food. Patients were treated for at least 6 months, with an intended duration of 12 months.

The primary efficacy endpoint of symptomatic, recurrent, fatal or nonfatal VTE occurred in 17 patients receiving rivaroxaban 20 mg (1.5%), 13 patients receiving rivaroxaban 10 mg (1.2%), and 50 patients receiving aspirin (4.4%), which translated to hazard ratios (HRs) of:

  • 34 for rivaroxaban 20 mg versus aspirin (95% CI 0.20-0.59; p<0.001)
  • 26 for rivaroxaban 10 mg versus aspirin (95% CI 0.14-0.47; p<0.001)
  • 34 for rivaroxaban 20 mg versus 10 mg (95% CI 0.65-2.75; p=0.42)

Fatal VTE occurred in two patients each in the rivaroxaban 20 mg cohort and aspirin cohort (0.2% each) and zero patients in the rivaroxaban 10 mg cohort.

The authors reported that patients with VTE and ongoing risk factors “have an appreciable risk of recurrence.” In the aspirin cohort, 3.6 percent of patients who experienced recurrent VTE had a provoked index event and were considered at a higher risk of recurrence, whereas 5.6 percent had an unprovoked index event. However, the authors noted, “rates of recurrence in patients whose index events were provoked or unprovoked were lower in both the rivaroxaban 20 mg group (1.4% and 1.8%) and the rivaroxaban 10 mg group (0.9% and 1.5%).”

The incidence of major bleeding (primary safety endpoint) appeared to be similar among groups: 0.5 percent (n=6) of the rivaroxaban 20 mg cohort, 0.4 percent (n=5) of the rivaroxaban 10 mg cohort, and 0.3 percent (n=3) of the aspirin cohort. Clinically relevant but non-major bleeding rates were slightly higher: 2.7 percent (n=30), 2 percent (n=22), and 1.8 percent (n=20), respectively.

“Compared with aspirin, both the 20 mg and 10 mg doses of rivaroxaban reduced the relative risk of recurrent VTE by about 70 percent (approximately 3 percentage points),” the authors concluded. “Rivaroxaban was more effective than aspirin for the prevention of recurrent VTE and was associated with a similar risk of bleeding.”

Non-major bleeding that resulted in treatment interruption for >14 days occurred in 17 patients in the rivaroxaban 20 mg cohort (1.5%) and 12 in the rivaroxaban 10 mg cohort (1.1%), compared with 12 patients in the aspirin cohort (1.1%; p values not reported).

Myocardial infarction, stroke, or systemic embolism occurred in three patients in the rivaroxaban 20 mg group (0.3%), five in the rivaroxaban 10 mg group (0.4%), and seven in the aspirin group (0.6%; p values not reported). The rates of death from any cause were 0.7 percent, 0.2 percent, and 0.6 percent, respectively (p values not reported). Adverse events (including all types of bleeding) were similar in all three study groups, according to the authors.

“Only patients with equipoise regarding the need for extended anticoagulation were included. Therefore, the results may not apply to patients perceived to require ongoing anticoagulation,” Dr. Weitz said, noting a limitation of the study’s findings. In addition, therapy was given for up to 12 months, so longer-term assessment is needed. The study was supported by Bayer Pharmaceuticals, which owns the license for rivaroxaban.

In an accompanying editorial, Mark A. Crowther, MD, also of McMaster University, and Adam Cuker, MD, of the Perelman School of Medicine at the University of Pennsylvania, noted that the results of the trial suggest that it would be “helpful to evaluate the effect of reduced doses of rivaroxaban within six months after an episode of VTE.”2

“Clinicians now have good-quality evidence to support the use of long-term, reduced-intensity anticoagulation therapy with rivaroxaban after an initial treatment course at usual therapeutic doses,” Drs. Crowther and Cuker wrote. “The favorable safety and efficacy profile of reduced-intensity rivaroxaban may lead to questions about how patients with provoked VTE should be treated.”


  1. Weitz JI, Lensing AWA, Prins MH, et al. Rivaroxaban or aspirin for extended treatment of venous thromboembolism. N Engl J Med. 2017;376:1211-22.
  2. Crowther MA and Cuker A. Reduced-intensity rivaroxaban for the prevention of recurrent venous thromboembolism. N Engl J Med. 2017;376:1279-80.