Percutaneous left atrial appendage closure versus nonvitamin K oral anticoagulants in patients with non-valvular atrial fibrillation and high bleeding risk

Aims: A significant number of patients with non-valvular atrial fibrillation (NVAF) are ineligible for nonvitamin K oral anticoagulants (NOACs) due to previous major bleeding or because they are at high bleeding risk (HBR). In this setting the indication for percutaneous left atrial appendage closure (LAAO) is a valuable alternative. We aimed to evaluate the efficacy and safety of NOACs versus LAAO indication in NVAF patients at HBR (HAS-BLED ≥3). Methods and results: All consecutive patients who underwent successful LAAO (n=193) and those treated with NOACs (n=189) (dabigatran, apixaban or rivaroxaban) were included. A 1:1 propensity score matching (PSM) was used to match LAAO and NOACs patients. At baseline, patients in the LAAO group had higher HAS-BLED (4.2% vs 3.3%, p<0.001) and lower CHADS-VASc (4.3% vs 4.7%, p=0.005) scores. After 1:1 PSM, 192 patients were enrolled in the final analysis (LAAO n=96; NOACs n=96). At two-year follow-up, no significant differences in thromboembolic (7.3% vs 6.3%, p=0.966) and ISTH major bleeding event rates (6.7% vs 4.8% p=0.503) were found between the two unmatched groups. All-cause death was significantly higher in the LAAO group (18.7% vs 10.6%; p=0.049). After PSM, all-cause death, thromboembolic and ISTH major bleeding event rates were similar between the groups. Significant independent predictors of all-cause death were dialysis (HR 5.65, 95% CI: 2.16-14.85, p<0.001) and age (HR 1.08, 95% CI: 1.05-1.13, p<0.001). Conclusions: In NVAF patients at HBR, LAAO and NOACs performed similarly in terms of thromboembolic and major bleeding events up to two-year follow-up. Our findings warrant further investigation in randomised trials and therefore can be considered as hypothesis-generating.