2146Continued vs. interrupted oral anticoagulation in patients with atrial fibrillation undergoing transfemoral transcatheter aortiv valve implantation

Abstract

Introduction: The role of continued vs. interrupted oral anticoagulation (OAC) in patients with atrial fibrillation (AF) undergoing transfemoral transcatheter aortic valve implantation (TF‐AVI) is uncertain. Purpose: The aim of this single center registry based investigation was to evaluate the risk of periprocedural bleeding and thromboembolic complications in patients undergoing TF‐AVI under continued OAC. Methods: Consecutive patients with AF and on OAC at admission (n=551) treated between 2011 and 2016 were stratified according to interrupted OAC (iOAC) vs. continued (cOAC) vs. continued novel oral anticoagulants (NOAC) at the time of TF‐AVI. VARC‐2 defined bleeding and stroke were the primary end points. 30‐day and 1‐year mortality was calculated using Kaplan‐Meier statistics. Results: Patients with iOAC (n=299), cOAC (n=117) and NOAC (n=182) had comparable baseline characteristics including age (80 (76; 83) vs. 80 (77; 83) vs. 80 (77; 84), p=0.250), gender (male: 42.5% vs. 47.9% vs. 48.9%, p=0.33) and STS‐Score (6.9 (4.2; 10.8) vs. 6.7 (4.4; 10.8) vs. 5.9 (3.6; 9.5), p=0.072). The proportion of patients having a CHA2DS2‐VASc‐Score ≥3 (p=0.791) and HASBLED‐Score ≥3 (p=0.185) were not different between groups. The majority of patients received a self‐expandable valve (70.9% vs. 59.8% vs. 67%, p=0.094). VARC‐2 defined stroke did not differ between groups (4.1% vs. 3.5% vs. 2.2%, p=0.527) including major stroke (3.8% vs. 1.7% vs. 1.1%), minor stroke (0% vs. 0.9% vs. 0.5%) and TIA (0.3% vs. 0.9% vs. 0.5%). VARC‐2 defined bleeding did also not differ between iOAC, cOAC and NOAC (34.8% vs. 38.8% vs. 27.5%, p=0.097) including life threatening bleeding (3.8% vs. 4.3% vs. 2.7%) and major bleeding (14.8% vs. 12.1% vs. 11.0%). Renal failure occurred more often in iOAC compared to cOAC and NOAC (15.8% vs. 9.5% vs. 7.7%, p=0.02). 30‐day mortality tended to be lower in NOAC (iOAC: 4.3% vs. iOAC: 3.4% vs. NOAC: 0.5%, p=0.061). 1‐year‐mortality was lower in NOAC (cOAC: 20.1% vs. iOAC: 13.7% vs. NOAC: 8.8%, p=0.015). Conclusion: TF‐AVI under continued OAC seems to be safe and effective. In particular, continued OAC using NOAC was associated with a favourable outcome compared to interrupted OAC. An adequately powered randomized clinical trial is necessary to prove this hypothesis.