DOACs and LAAO Safe and Effective for Patients With NVAF and High Bleeding Risk

Left atrial appendage occlusion (LAAO) is as effective as direct oral anticoagulants (DOACs) in preventing ischemic events at 5 years in patients with nonvalvular atrial fibrillation (NVAF) with a high bleeding risk, researchers reported in the International Journal of Cardiology.

The observational, prospective, nonrandomized study was conducted at a hospital in Italy and included patients with NVAF at high bleeding risk who initiated a DOAC or had successful LAAO from July 2009 to December 2016.

High bleeding risk was defined based on a HAS-BLED score 3 or higher. The high bleeding risk patients were prospectively followed up to 2021.

The primary effectiveness endpoint included thromboembolic events, and the primary safety endpoint was major bleeding per the International Society of Thrombosis and Hemostasis (ISTH) classification.

The cohort included 940 participants with NVAF, of whom 382 patients had a HAS-BLED risk score of 3 or higher and represented the high bleeding risk study population (193 patients in the LAAO group [mean age, 74.2±7.7 years; 32.6% women] and 189 patients in the DOAC group [mean age, 77.7±6.9 years; 30.7% women]). Participants who had percutaneous LAAO had more comorbidities such as diabetes and an increased HAS-BLED score (4.2 for the LAAO group vs 3.3 for the DOAC group, P <.001). Patients who used DOACs had a greater ischemic risk according to CHA₂DS₂-VASc score (4.8 vs 4.3, P =.005).

The median follow-up in the 2 unmatched high bleeding risk groups was 5.1 years (IQR, 2.6-6.7). The combined safety and effectiveness endpoint was significantly increased in the LAAO group (P =.042), primarily due to a significantly greater number of thromboembolic events (P =.047). The 2 groups had a comparable rate of ISTH-major bleeding events (P =.221). The all-cause death rate was significantly greater in the LAAO group (35.8% vs 26.6%; P =.028) compared with the DOAC group. In addition, the LAAO group had a higher cardiac death rate (13.3% vs 6.8%; P =.016).

After 1:1 propensity-score matching, data from 192 patients (96 in the DOAC group and 96 in the LAAO group) were adjusted for variables in the CHADs-VASc and HAS-BLED scores. No significant difference in thromboembolic events was found between the 2 matched groups, although an absolute higher number of thromboembolic events occurred in the LAAO group vs the DOAC group (13.3% vs 9.5%, P =.357). Both groups had a major bleeding rate of 7.5 (P =.918).

The DOAC group had a significantly increased overall bleeding rate (25.0% vs 13.7%, P =.048). All-cause and cardiac mortality were significantly increased for the LAAO group vs the DOAC group (P =.020 and .021, respectively).

Multivariable Cox regression analysis showed that significant independent predictors of all-cause death were age (hazard ratio [HR], 1.1; 95% CI, 1.05-1.15; P <.001), heart failure (HR, 2.3; 95% CI, 1.3-4.0; P =.003), and LAAO indication (HR, 1.9; 95% CI, 0.9-10.4; P =.033).

Limitations of the study include the observational design, and some potential confounders are not directly measured or included in the database. In addition, the mortality rate, which is particularly high in the LAAO group, could have affected the rates of thromboembolic and bleeding events. Also, the sample size is relatively small, and discharge antithrombotic therapies in the LAAO group are not standardized.

“This single-center, real-world, 1:1 propensity-matched study confirmed that LAAO was as effective as DOACs in preventing ischemic events at 5-year follow-up,” wrote the investigators. “On the other hand, DOACs were extremely safe with no excess of major bleeding. Thus, both treatments can be considered valuable at-long term follow-up in patients at high bleeding risk.”