Assessment of the left atrial appendage structure and morphology: comparison of real-time three-dimensional transesophageal echocardiography and computed tomography

• DOI: 10.1007/s10554-016-1044-4
• Publication Date: 2016-12-24T15:22:48Z
• Authors: Wenjuan Bai; Zhongxiu Chen; Hong Tang; Hui Wang; Wei Cheng; Li Rao
• Keywords: Adult; Echocardiography, Doppler, Pulsed; Male; Observer Variation; China; Chi-Square Distribution; Echocardiography, Three-Dimensional; Middle Aged; 03 medical and health sciences; 0302 clinical medicine; Predictive Value of Tests; Case-Control Studies; Atrial Fibrillation; Catheter Ablation; Linear Models; Humans; Atrial Appendage; Atrial Function, Left; Female; Prospective Studies; Echocardiography, Transesophageal; Aged

In this study, the left atrial appendage (LAA) structure and morphology were assessed using real-time three-dimensional transesophageal echocardiography (RT3D-TEE) and enhanced cardiac computed tomography (CT) scanning, and the two measurements obtained using these different techniques were compared. Two hundred nonvalvular atrial fibrillation (NVAF) patients who were prepared to undergo radiofrequency ablation were enrolled, and 62 controls were enrolled prospectively. RT3D-TEE and CT were performed, and the following parameters were measured: LAA orifice diameter, area, depth, maximum volume, and emptying velocity. The differences between the NVAF patients and controls were compared, and the differences and correlations in the LAA parameters between the two imaging techniques were assessed. The NVAF patients had significantly bigger orifice sizes, orifice areas, and maximum volumes, and lower emptying rates. The AF cohort had roughly equal proportions of the four morphological types, whereas the controls predominantly had the windsock type. A Bland-Altman analysis demonstrated that the LAA measurements obtained using RT3D-TEE were lower than those obtained with the CT. However, linear regression analysis showed the good correlations between the RT3D-TEE and CT measurements of the LAA orifice area, depth and volume (r = 0.86, 0.63, and 0.75, respectively). The use of RT3D-TEE for the visualization and quantitative analysis of LAA parameters is feasible, and the obtained measurements correlate well with those obtained with cardiac CT. This technique may be an ideal tool for guiding LAA occlusion procedures, and combining these two methods may enhance the accuracy of LAA measurements.