What it is
Studies comparing aspirin with anticoagulants for preventing venous thromboembolism (VTE), blood clots in the veins, after hip or knee replacement have had conflicting results. In this comparative-effectiveness noninferiority trial, patients having elective hip or knee arthroplasty were randomly assigned to aspirin (81 mg twice daily), warfarin (to a target international normalized ratio of 2.0) or rivaroxaban (10 mg daily) for 30 days. The primary effectiveness outcome (death from any cause or symptomatic VTE) and the primary safety outcome (bleeding, reoperation for wound problems or myocardial infarction) were evaluated through 6 months in a blinded fashion. Of 18,883 participants randomized, 15,317 were in the primary per-protocol analysis. The effectiveness outcome occurred in 1.26% with aspirin, 1.23% with warfarin and 0.94% with rivaroxaban, and the safety outcome in 2.91%, 3.36% and 3.48%. Aspirin was not noninferior for effectiveness: in pairwise comparisons its estimated 6-month risk was 1.24% against 0.94% for rivaroxaban, a difference of 0.30 percentage points (confidence interval -0.20 to 0.81), and 1.24% against 1.01% for warfarin, a difference of 0.23 percentage points (interval -0.34 to 0.81), with both intervals reaching past the noninferiority margin of 0.69 percentage points. Warfarin and rivaroxaban, in turn, were not noninferior to aspirin for safety. Intention-to-treat analyses gave similar results.
Why it matters
In a trial that randomized 18,883 patients having a hip or knee replaced, aspirin could not be shown to prevent symptomatic clots and deaths as well as two anticoagulants. Nor could the anticoagulants be shown to be as safe. Not noninferior is not the same as inferior: each effectiveness interval also included no difference, so the trial could not rule out a disadvantage for aspirin larger than the margin rather than showing one. The safety comparisons mirror it: in pairwise comparisons the estimated 6-month safety risk was 3.70% with rivaroxaban against 3.02% with aspirin, a difference of 0.68, and 3.65% with warfarin against 2.88%, a difference of 0.77 percentage points, and both intervals included no difference and crossed the margin. Every event rate was low, at a few per hundred patients or fewer, and each estimated difference between regimens was under one percentage point.
Underlined numbers link to their source. Every metric and quoted figure is listed under Sources and data below.
Filed underVenous Thromboembolism Diagnosis and Management, Total Knee Arthroplasty Outcomes, Orthopaedic implants and arthroplasty
