Abstract
BACKGROUND: Three-dimensional (3D) transesophageal echocardiography (TEE) has been successfully used in the sizing of left atrial appendage (LAA) occlusion devices, but its use has not yet been studied in LAA clip devices. We sought to develop and validate the novel use of 3D-TEE sizing in a novel LAA clip device for atrial fibrillation (AF) patients undergoing video-assisted thoracic surgery (VATS) ablation. METHODS: Consecutive patients with isolated AF undergoing LAA clipping or excision during VATS ablation were included in the study between June 2021 and September 2022 at Fuwai Hospital. The patients underwent 3D-TEE examinations preoperatively and postoperatively. The VATS length, LAA clip effective length, and LAA excision margin length were recorded. A correlation analysis, intraclass correlation coefficient (ICC) analysis, and Bland-Altman plot analysis were conducted to examine the TEE parameters, VATS length, LAA clip effective length, and LAA excision margin length. RESULTS: In total, 26 AF patients undergoing LAA clipping and 15 undergoing LAA excision were included in the study. In the LAA clipping group, in which the Atriclip size served as the control, the 3D-TEE with volumetric measurement (the perimeter-derived maximum orifice diameter) (R=0.938; ICC =0.934; Bland-Altman plot variability, 3.85%) showed the best sizing efficacy for the LAA clip device among the 3D-TEE with multiplanar reformatting sizing (the perimeter-derived maximum orifice diameter) (R=0.808; ICC =0.772; Bland-Altman plot variability, 3.85%), VATS sizing (R=0.851; ICC =0.756; Bland-Altman plot variability, 11.54%), and VATS plus 0.5-cm sizing (R=0.851; ICC =0.775; Bland-Altman plot variability, 11.54%) measurements (all P<0.001). In addition, for the distribution of matched sizing in the LAA clip group, 3D-TEE with volumetric measurement sizing (20/26) had a higher proportion than 3D-TEE with multiplanar reformatting sizing (11/26, P=0.011), VATS sizing (9/26, P=0.002), and VATS plus 0.5-cm sizing (14/26, P=0.08). Using the LAA excision margin length as the control, the mean difference in the LAA diameter was 1.17 cm [95% confidence interval (CI): 0.71-1.62 cm , P<0.001] in the maximum orifice diameter of two-dimensional-TEE, 0.15 cm (95% CI: -0.32 to 0.61 cm , P=0.523) in the perimeter-derived 3D multiplanar reformatting (the maximum orifice diameter), and 0.03 cm (95% CI: -0.47 to 0.53, P=0.901) in the perimeter-derived 3D volumetric (3DV) measurement (the maximum orifice diameter), and the related Pearson correlation coefficients for these modalities were 0.760 (P=0.001), 0.843 (P<0.001), and 0.963 (P<0.001), respectively. CONCLUSIONS: Our study showed that 3D-TEE might be employed in the sizing of a novel LAA clip device using the VATS approach in patients with AF. The 3DV measurement (the perimeter-derived maximum orifice diameter) was superior to the VATS measurement. These findings might also apply to LAA VATS excision patients with AF.