Three-Dimensional Echocardiographic Left Atrial Appendage Volumetric Analysis

J Am Soc Echocardiogr. 2021 Mar 25;S0894-7317(21)00142-5. doi: 10.1016/j.echo.2021.03.008. Online ahead of print.

Abstract

Background: Left atrial appendage (LAA) echocardiographic assessment is difficult because of its complex shape and relatively small size. Three-dimensional echocardiographic (3DE) imaging can overcome the limitations of two-dimensional (2D) imaging. Pulsed wave (PW) Doppler is the only currently standard LAA functional parameter. The aim of this study was to test a new approach for 3DE volumetric analysis to obtain LAA ejection fraction (EF), its size and shape.

Methods: Transesophageal 2D and 3D LAA images were prospectively obtained in 159 consecutive patients. LAA volumes were measured from 3DE images using available software. PW Doppler was considered the reference value for LAA function, and used for comparison with LAA EF. Comparison to cardiac computed tomography (CT) was performed in a subgroup of 32 patients. Comparisons included linear regression and Bland-Altman analyses. Repeated measurements were performed to assess measurement variability.

Results: Nine patients were excluded because of suboptimal image quality (94%feasibility). 3D LAA calculated EF was in good agreement with LAA PW measurements. 3D morphologic evaluation showed that 43% of the patients had "chicken wing", 33% "cactus", 19% "windsock" and 5% Cauliflower shape. At the time of data acquisition, patients with atrial fibrillation had non-significantly larger LAA end-systolic and diastolic volumes (ESV, EDV), leading to a lower calculated EF. 3DE LAA ESVs were in good agreement with cardiac CT (r = 0.75), with small biases: -2.5±3.9ml. Reproducibility was better for larger LAA volumes.

Conclusions: A novel 3DE approach can determine geometry, size, and function of the LAA. A new parameter, LAA EF, provides functional quantitation.

Keywords: 3-dimensional echocardiography; atrial fibrillation; function; left atrial appendage; shape; size.