BackgroundGuidelines support aspirin thromboprophylaxis for primary total hip and knee arthroplasty (THA and TKA) but supporting evidence has come from high volume centers and the practice remains controversial.MethodsWe studied 4562 Medicare patients who underwent elective primary THA (1736, 38.1%) or TKA (2826, 61.9%) at 9 diverse hospitals. Thirty-day claims data were combined with data from the health system’s electronic medical records to compare rates of venous thromboembolism (VTE) between patients who received prophylaxis with: (1) aspirin alone (47.3%), (2) a single, potent anticoagulant (29%), (3) antiplatelet agents other than aspirin or multiple anticoagulants (21.5%), or (4) low-dose subcutaneous unfractionated heparin or no anticoagulation (2.2%). Sub-analyses separately evaluating THA, TKA and cases from lower volume hospitals (n = 975) were performed.ResultsThe 30-day VTE incidence was 0.6% (29/4562). VTE rates were equal in patients receiving aspirin and those receiving a single potent anticoagulant (0.5% in both groups). Patients with VTE were significantly older than patients without VTE (mean 76.5 vs. 73.1 years, P = 0.04). VTE rate did not associate with sex or hospital case volume. On bivariate analysis considering age, aspirin did not associate with greater VTE risk compared to a single potent anticoagulant (OR = 2.1, CI = 0.7–6.3) with the numbers available. Odds of VTE were increased with use of subcutaneous heparin or no anticoagulant (OR = 6.4, CI = 1.2–35.6) and with multiple anticoagulants (OR = 3.6, CI = 1.1–11.2). THA and TKA demonstrated similar rates of VTE (0.5% vs. 0.7%, respectively, P = 0.43). Of 975 cases done at lower volume hospitals, 387 received aspirin, none of whom developed VTE.ConclusionsThis study provides further support for aspirin as an effective form of pharmacological VTE prophylaxis after total joint arthroplasty in the setting of a multi-modal regimen using 30-day outcomes. VTE occurred in 0.7% of primary joint arthroplasties. Aspirin prophylaxis did not associate with greater VTE risk compared to potent anticoagulants in the total population or at lower volume hospitals.
Read full abstract