In-hospital complications after left atrial appendage occlusion and their impact on long-term mortality: a multicenter study

Background

Left atrial appendage occlusion (LAAO) is increasingly performed in patients with atrial fibrillation at high thromboembolic risk and contraindicated to long-term oral anticoagulation due to major bleeding risk. However, the incidence, determinants, and prognostic impact of in-hospital complications remain incompletely characterized in this high bleeding risk population.

Purpose

To assess the incidence and predictors of in-hospital major adverse events (IHMAE) following LAAO and their impact on long-term mortality.

Methods

We conducted a multicenter observational study including 1085 consecutive patients undergoing LAAO between 2012 and 2025 in three tertiary French centers. Baseline, procedural, and follow-up data were collected. IHMAE were defined as a composite of death, stroke/systemic embolism, pericardial effusion/tamponade, device embolization, major bleeding, or vascular complications. Logistic regression analysis was performed to identify independent predictors of IHMAE. The association between IHMAE and all-cause mortality was assessed using Kaplan–Meier analysis and Cox regression, restricted to patients discharged alive.

Results

Patients were mostly males (63.3%) with a mean age of 77.2±10 years and had high thromboembolic and bleeding risks (CHA₂DS₂-VASc 4.7±1.5; HAS-BLED 3.6±1.0). Procedural success was high (99.5%). Median length of stay was 3 days [IQR 2–3].

IHMAE occurred in 6.3% of patients. The most frequent complications were pericardial effusion/tamponade (2.9%) and major bleeding (2.6%), followed by vascular complications (1.2%). Periprocedural or in-hospital death occurred in 0.6%, stroke/systemic embolism in 0.4%, and device embolization in 0.1%. No baseline characteristic independently predicted IHMAE in multivariable analysis.

During a median follow-up of 702 days [IQR 211–1437], all-cause mortality occurred in 37.8%, recurrent thromboembolic events in 8.8%, recurrent bleeding events in 18.5%, and rehospitalization in 62.3%. IHMAE were associated with higher post-discharge mortality in the unadjusted analysis (9.1% vs 4.6%, p=0.004). However, after exclusion of in-hospital deaths, this association was not significant in time-to-event analysis (log-rank p=0.704) or after adjustment for age and sex (HR 0.94, 95% CI 0.64–1.36, p=0.728).

Conclusion

In this large multicenter cohort, LAAO was associated with a low rate of in-hospital complications, mainly driven by manageable bleeding and pericardial events. IHMAE did not independently predict long-term mortality after time adjustment. Our findings suggest that most complications are managed appropriately during hospitalization and have limited impact on long-term prognosis. Mortality of these high-risk patients remains high at relatively short term follow-up.Abstract FigureFor image description, please refer to the figure legend and surrounding text.Abstract TableFor image description, please refer to the figure legend and surrounding text.