From Wiki Journal Club
Jump to: navigation, search
Anderson DR, et al. "Aspirin or Rivaroxaban for VTE Prophylaxis after Hip or Knee Arthroplasty". The New England Journal of Medicine. 2018. 378(8):699-707.
PubMedFull textPDF

Clinical Question

In patients requiring VTE prophylaxis after total hip or knee arthroplasty, how do aspirin and rivaroxaban compare in terms of rates of symptomatic VTE and bleeding?

Bottom Line

In relatively low-risk patients who received 5 days of initial rivaroxaban therapy after total hip or knee arthroplasty, extended thromboprophylaxis with aspirin 81 mg/d was noninferior to rivaroxaban 10 mg/d in preventing symptomatic VTE.

Major Points

The risk of postoperative VTE in patients undergoing hip or knee arthroplasty without postoperative anticoagulation is estimated at approximately 6% by the Caprini score.[1] As a consequence, postoperative VTE prophylaxis is a standard of care for patients undergoing hip or knee arthroplasty. Several agents are used for VTE prophylaxis following orthopedic surgery, including aspirin, enoxaparin, warfarin, rivaroxaban, apixaban, and others. Randomized trials are lacking, and thus the optimal agent, dose, and duration of VTE prophylaxis remains undefined.

The Extended Venous Thromboembolism Prophylaxis Comparing Rivaroxaban to Aspirin Following Total Hip and Knee Arthroplasty II (EPCAT II) study randomized 3,427 adults undergoing elective hip or knee arthroplasty to either aspirin 81 mg/d or rivaroxaban 10 mg/d following an initial 5 days of postoperative rivaroxaban. Patients undergoing hip arthroplasty received a total of 35 days of postoperative VTE prophylaxis, while those undergoing knee arthroplasty received a total of 14 days, in accordance with guidelines. Patients were followed for the primary efficacy outcome of symptomatic acute proximal DVT or PE within 90 days of randomization; the primary safety outcome was major or clinically relevant nonmajor bleeding. At 90 days, the symptomatic VTE occurred in 0.64% of patients randomly assigned to aspirin, and in 0.70% of patients assigned to rivaroxaban (P<0.001 for noninferiority). Bleeding was observed in 1.29% of patients in the aspirin arm, and 0.99% of patients in the rivaroxaban arm, and consisted of surgical site bleeding in all cases, and typically occurred within 10 days. The authors conclude that aspirin is an efficacious and thus more cost-effective than rivaroxaban for VTE prophylaxis following 5 days of rivaroxaban after total hip or knee arthroplasty.

The EPCAT II results are consistent with those from EPCAT I (2013)[2], which randomized a similar population of patients to extended aspirin or dalteparin following 10 days of postoperative dalteparin.


As of March 2018, no guidelines have been published that reflect the results of this trial.


  • Multicenter, double-blind, randomized, controlled, noninferiority trial
  • n=3,427 patients undergoing total hip or knee arthroplasty
    • Aspirin (n=1,707)
    • Rivaroxaban (n=1,717)
  • Setting: 15 university-affiliated health centers in Canada
  • Enrollment: 2013-2016
  • Follow-up: 90 days
  • Analysis: Intention-to-treat
  • Primary Outcomes: adjudicated symptomatic VTE; major or clinically relevant nonmajor bleeding


Inclusion Criteria

  • Age >18 years
  • Any patient undergoing elective unilateral hip or knee arthroplasty

Exclusion Criteria

  • Hip or lower limb fracture or surgery within past 3 months
  • Metastatic cancer
  • Women of childbearing age if breastfeeding or not using appropriate contraception
  • History of major bleed, aspirin allergy, or liver dysfunction
  • Life expectancy <6 months
  • Active PUD or gastritis
  • CrCl <30 ml/min
  • Platelets <100K
  • Medical condition requiring chronic anticoagulation or current aspirin >100 mg/d
  • Rivaroxaban was not given post-operatively
  • Expected surgery requirement within 90 days from arthroplasty
  • Concomitant use of strong inducers/inhibitors of Pgp or CYP3A4

Baseline Characteristics

Selected characteristics taken from the aspirin group unless otherwise specified. There were no significant differences between group in baseline characteristics.

  • Mean age: 62.9 years
  • Female sex: 52.9%
  • BMI: 31.1
  • Risk factors: prior VTE (2.2%), prior surgery (3.4%), cancer (2.5%), smoker (9.5%)
  • Hemoglobin: 13.9 g/dL
  • Platelet count, median: 241K
  • Primary arthroplasty: 91.9%
  • Tranexamic acid use: 54.3%
  • Postoperative mechanical compression: pneumatic compression (50.0%), graduated stockings (42.8%), both (7.2%)
  • Anesthetic: general (28.9%), regional (70.4%), both (0.8%)
  • Operating room time: 1.4 hours
  • Estimated blood loss: 314 ml
  • Length of hospital stay: 3.5 days


  • All patients received 5 days of rivaroxaban 10 mg/d starting between 6-24 hours postoperatively
  • Patients were then randomized to continue rivaroxaban 10 mg/d or change to aspirin 81 mg/d
    • Total hip arthroplasty patients received assigned therapy for 30 days (total 35 days of VTE prophylaxis)
    • Total knee arthroplasty patients received assigned therapy for 9 days (total 14 days of VTE prophylaxis)
  • Aspirin and rivaroxaban were administered in identical-appearing capsules
  • Patients receiving preoperative low-dose aspirin received open-label aspirin in addition to randomly assigned therapy
  • Patients were followed for 90 days


Comparisons are aspirin vs. rivaroxaban.

Primary Outcomes

Symptomatic proximal DVT or PE during 90-day follow-up
0.64% vs. 0.70% (0.06% difference; 95% CI -0.55 to 0.66; P<0.001 for noninferiority and P=0.84 for superiority)
Major or clinically relevant nonmajor bleeding
1.29% vs. 0.99% (0.3% difference; 95% CI -1.07 to 0.47; P=0.43)

Secondary Outcomes

1 vs. 0 (P not provided)
Wound infection
2.6% vs. 3.4% (P=NS)
0.1% in each group (P=NS)
0 in each group (P=NS)

Subgroup Analysis

  • There was no difference in outcome when treatments were stratified by type of intervention (hip vs. knee)
  • Patients who were on chronic aspirin therapy had similar rates of VTE/PE and major and clinically relevant nonmajor bleeding compared to those who were not on chronic aspirin therapy.


  • There was unclear clinical equipoise for allowing patients previously on aspirin to continue their aspirin if randomized to rivaroxaban.
  • The title of the article is misleading and suggests a head-to-head comparison of aspirin vs. rivaroxaban in postoperative VTE prophylaxis. A true head-to-head trial would require both drugs to be started immediately after surgery (not 5 days of initial treatment with rivaroxaban).


  • Funded by the Canadian Institutes of Health Research

Further Reading

  1. Gould MK et al. Prevention of VTE in nonorthopedic surgical patients: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012. 141:e227S-e277S.
  2. Anderson DR et al. Aspirin versus low-molecular-weight heparin for extended venous thromboembolism prophylaxis after total hip arthroplasty: a randomized trial. Ann. Intern. Med. 2013. 158:800-6.