Introduction: We aimed to investigate whether a modified implantation method facilitating a fully open umbrella can reduce the pericardial effusion/pericardial tamponade (PE/PT) rate after left atrial appendage closure (LAAC) with the LAmbre device compared with the conventional method (CM) in patients with non-valvular atrial fibrillation (NVAF). Methods and results: Patients with NVAF who received either isolated LAAC or combined catheter ablation and LAAC using the LAmbre device at the First Affiliated Hospital of Wenzhou Medical University from January 2018 to December 2019 were enrolled. CM was used for device implantation in the initial 59 patients, while a modified method (MM) was used in the remaining 165 patients. Successful implantation was achieved in 98.3% of patients in the CM group and in 98.8% in the MM group. A higher rate of a fully open umbrella (98.8% vs 69%, P<0.001), less requirement for recapture (46% vs 62.1%, P=0.036), and a lower incidence of delayed PE/PT (1.2% vs 8.6%, P=0.005) were found in the MM group compared with in the CM group. All of the five delayed PT events occurred in patients with combined treatment. An umbrella that was not fully open was the only factor associated with delayed PE/PT events in a multivariable Cox model. Conclusions: LAAC with the LAmbre device using an MM significantly increases the rate of a fully open umbrella and decreases the requirement for recapture and the incidence of delayed PE/PT. This method is more effective in patients with combined treatment.

Introduction: Transesophageal echocardiography (TEE) fails to display optimal views to guide left atrial appendage closure (LAAC) procedure in some patients due to atrio-esophageal anatomical limitation. We aimed to investigate if intracardiac echocardiography (ICE) conducted from left atrium (LA) achieves comparable multi-planar views and clinical outcomes as TEE during LAAC. Methods and Results: This study prospectively enrolled 102 consecutive patients with non-valvular atrial fibrillation receiving LAmbre implants under local anesthesia at the First Affiliated Hospital of Wenzhou Medical University from August 2018 to July 2019. The procedures were guided by either ICE (n=40), TEE (n=50) or combined ICE-TEE (n=12). A novel multi-angled “FLAVOR” approach was used in ICE group for assessment. ICE allowed visualization of implanted device in all patients at all proposed angles with long-axis views, while TEE failed to do so in at least one of the angles in 36% of cases. In the combined ICE-TEE cohort, TEE failed peri-device leak assessment in 2 patients. Rates of procedural success, recapture, resizing and complications were similar between ICE and TEE groups. Fluoroscopy time, radiation dose and volume of contrast use in ICE group were significantly lower than the TEE cohort. At 45-day TEE follow-up, rate and degree of peri-device leaks were similar between the ICE and TEE groups. Conclusions: A systematic approach using ICE to guide LAmbre LAA occlusion was safe and feasible. This method was more reliable in comprehensive, multi-angled imaging assessment, and achieving shorter fluoroscopy time, lower radiation dose and less contrast use than TEE.