Thromboembolic and bleeding events after valvular intervention in patients with atrial fibrillation

Abstract

Aim To assess outcomes after cardiac surgery with biological valve replacement, valve repair or transcatheter aortic valve implantation (TAVI) in patients with atrial fibrillation (AF) in accordance with oral anticoagulant (OAC) treatment.Methods All patients in Sweden undergoing valvular intervention with AF were included. Associations between OAC exposure and cardiovascular (CV) events (composite of CV death, ischaemic stroke or systemic embolism) and major bleeding were investigated using Cox regression analysis. The analysis was separated in time periods of 0–3 and 3–12 months after discharge.Results 4730 patients were included in the first time period, 54.0% had received a surgical biological valve prosthesis, 23.8% valve repair and 22.2% TAVI. Exposure to warfarin (comparator) was 62.3%, to non-vitamin K antagonist oral anticoagulants (NOACs) 10.0% and to no OAC 27.7%. NOAC exposure was associated with similar risk of the composite CV outcome and major bleeding from 0 to 3 months. No OAC was associated with increased risk of the composite CV outcome (HR 1.71; 95% CI 1.26 to 2.32) and similar risk of major bleeding. Further analysis of the bioprosthetic valve replacement subgroup indicated increased risk of CV death when exposed to NOAC (HR 2.58; 95% CI 1.15 to 5.78) and no OAC (HR 2.82; 95% CI 1.65 to 4.82) compared with warfarin from 0 to 3 months. No differences were seen between 3 and 12 months.Conclusion In this registry-based cohort study of patients with AF with severe valvular heart disease undergoing various valvular interventions, NOAC appears to be comparable with warfarin regarding efficacy and safety. Patients not receiving OAC had higher risk of CV events. NOAC was associated with increased CV death compared with warfarin in the surgical bioprosthetic valve replacement subgroup, illustrating the importance of being cautious when extrapolating data from one patient group to another. Further studies comparing NOAC and warfarin in the early postoperative phase are warranted, especially following surgical bioprosthetic valve replacement.