Background. Left atrial appendage occlusion (LAAO) is an established alternative to oral anticoagulation in patients with atrial fibrillation. Data on periprocedural outcomes and long-term safety of LAAO in patients with preprocedural thrombus or sludge within the left atrial appendage (LAA) are limited.
Methods. In this single-center study, all consecutive patients undergoing LAAO between January 2019 and March 2024 were included. All patients underwent transesophageal echocardiography before the intervention and at 45-day follow-up. The incidence of periprocedural complications, peri-device leakage (PDL) at 45 days, and a composite clinical endpoint including all-cause mortality, stroke, systemic embolism, device-related thrombosis, and major bleeding were analyzed.
Results. A total of 283 patients were included, of whom 34 (12.0%) presented with LAA sludge or thrombus before LAAO. Patients with LAA sludge or thrombus were older (74.5 [70.8 – 79.0] years versus 72.0 [67.0 – 77.0] years, p=0.05), while other baseline characteristics including sex, cardiovascular risk factors, left ventricular ejection fraction, and prior thromboembolic or bleeding events, did not differ (all p>0.05). Periprocedural stroke occurred more frequently in patients with LAA sludge or thrombus compared to those without (5.9% [n=2/34] versus 0.4% [n=1/249], p=0.04). Other periprocedural complications including intra-hospital death (n=0), pericardial tamponade (0% versus 2.0%, p=0.52), device embolization (0% versus 0,4% [n=1/249], p=0.71), and major bleeding (2.9% versus 0.8%, p=0.32) did not differ between groups.
At 45-day transesophageal echocardiographic follow-up, the incidence of PDL was comparable between groups (PDL <3 mm: 14.7% versus 22.1%, p=0.21; PDL 3-5 mm: 5.9% versus 8.4%, p=0.50; PDL >5 mm: 2.9% versus 4.5%, p=0.59).
During a median follow-up of 394 days (IQR 314 – 697), the composite clinical endpoint occurred in 20 patients (7.1%). Patients with preprocedural LAA sludge or thrombus were at higher risk for composite endpoint (14.7% [5/34] versus 6.0% [15/249], hazard ratio [HR] 3.34, 95% confidence interval [CI] 1.18 – 9.45; p=0.02). This association remained significant after adjustment for age, sex, diabetes mellitus, left ventricular ejection fraction, and renal function (HR 4.17, 95% CI 1.41 – 12.33, p=0.01).
Conclusions. The presence of LAA sludge or thrombus prior to LAAO was associated with a higher risk of periprocedural stroke and adverse clinical outcomes during follow-up, despite similar device sealing. Although LAAO appears technically feasible in carefully selected patients with LAA sludge or thrombus, these findings underscore the need for meticulous procedural planning and close post-procedural surveillance. Larger multicenter studies are warranted to better define optimal management and outcomes in this high-risk subgroup.
