Peri-operative Adverse Outcomes in Patients with Atrial Fibrillation Taking Warfarin or Edoxaban: Analysis of the ENGAGE AF-TIMI 48 Trial

Background Peri-operative management of anticoagulated patients with atrial fibrillation (AF) is challenging. To gain information on the peri-operative management of edoxaban, we compared outcomes in patients on warfarin or edoxaban enrolled in ENGAGE AF-TIMI 48 who underwent a surgery or invasive procedure. Methods Data from patients undergoing their first surgery/procedure were analysed and results compared by anticoagulant (warfarin vs. higher- or lower-dose edoxaban regimen [HDER and LDER, respectively]). Patients were classified by procedural management: anticoagulant interrupted (last dose 4–10 days pre-procedure) or anticoagulant continued (last dose ≤ 3 days pre-procedure). Stroke/systemic embolism (SSE), major bleeding (MB), MB or clinically relevant non-MB (CRNMB) and death were assessed from 7 days pre- until 30 days post-procedure. The chi-square test was used to compare outcomes across treatment groups. Results A total of 7,193 patients (34%) underwent surgery/procedure: 3,116 had anticoagulant interrupted, 4,077 had anticoagulant continued. Among patients on warfarin, HDER and LDER who had anticoagulant interrupted, rates of SSE were 0.6, 0.5 and 0.9% (p = 0.53), rates of MB were 1.0, 1.2 and 1.1% (p = 0.94) and rates of MB or CRNMB were 3.9, 4.2 and 3.6% (p = 0.78); among patients on warfarin, HDER and LDER who had anticoagulant continued, rates of SSE were 1.1, 0.7 and 0.9% (p = 0.51), rates of MB were 3.6, 2.6 and 2.4% (p = 0.13) and rates of MB or CRNMB were 8.5, 7.9 and 6.6% (p = 0.17). Conclusion In patients requiring surgery/procedure in ENGAGE AF-TIMI 48, peri-operative rates of SSE, MB and death were not significantly different in patients who received edoxaban or warfarin.