Truong Quynh A, Legasto Alan C, Deaño Roderick C, Bachman Daniel P, Bhatt Deep, Al'Aref Subhi J, Devereux Richard B, Wong S Chiu, Salemi Arash, Szymonifka Jackie
Department of Radiology, Weill Cornell Medicine, New York, NY.
Division of Cardiology, Department of Medicine, Weill Cornell Medicine, New York, NY.
Radiol Cardiothorac Imaging. 2019 Aug;1(3). doi: 10.1148/ryct.2019180025. Epub 2019 Aug 29.
We sought to determine the performance of an automated computed tomography (CT) software that provides dynamic annular measurements of all available cardiac phases for transcatheter aortic valve replacement (TAVR) sizing as compared to the standard single manual measurement.
In 60 TAVR patients (84±7 years, 60% male) who underwent pre-procedural CT scans, we measured the aortic annular diameters, perimeter, and area using (1) the dynamic automated CT measurements and (2) standard single manual measurement from the cardiac phase of maximum systolic opening by visual estimate.
The automated software was successful in providing annular measurements in 43/60 (72%) of cases, with the remainder requiring semi-automated contours. The maximum dynamic automated values were predominantly in systole (46/60[77%] for diameter, 44/60[73%] for perimeter, 48/60[80%] for area), and was a different phase from the standard manual phase in 46/60 (77%) cases. The maximum dynamic automated annular values were larger than the standard manual values measured (Δdiameter 0.35 mm, p=0.04; Δperimeter 1.71 mm, p<0.001; Δarea 15.6 mm, p<0.001). When comparing standard manual to the same phase by automated measurements, while there was no difference in annular mean diameter (p=0.80), perimeter and area were larger with the automated measurements (Δperimeter 0.95 mm, p=0.002; Δarea 10.8 mm, p=0.03). However, the maximum automated measurements were consistently larger than the same phase automated measurements (Δdiameter 0.13 mm, p<0.001; Δperimeter 0.42 mm, p<0.001; Δarea 4.4 mm, p<0.001).
Automated maximum dynamic CT annular measurements provide larger values than standard manual and same phase automated measurements.
我们旨在确定一种自动计算机断层扫描（CT）软件的性能，该软件可对所有可用心脏相位进行动态主动脉环测量，用于经导管主动脉瓣置换术（TAVR）的尺寸确定，并与标准单手动测量进行比较。
在60例接受术前CT扫描的TAVR患者（84±7岁，60%为男性）中，我们使用（1）动态自动CT测量和（2）通过视觉估计从最大收缩期开放心脏相位进行的标准单手动测量来测量主动脉环直径、周长和面积。
自动软件在43/60（72%）的病例中成功提供了主动脉环测量，其余病例需要半自动轮廓。最大动态自动测量值主要出现在收缩期（直径46/60[77%]、周长44/60[73%]、面积48/60[80%]），并且在46/60（77%）的病例中与标准手动测量的相位不同。最大动态自动主动脉环测量值大于测量的标准手动值（直径差值0.35mm，p = 0.04；周长差值1.71mm，p < 0.001；面积差值15.6mm，p < 0.001）。当将标准手动测量与自动测量的同一相位进行比较时，虽然主动脉环平均直径没有差异（p = 0.80），但自动测量的周长和面积更大（周长差值0.95mm，p = 0.002；面积差值10.8mm，p = 0.03）。然而，最大自动测量值始终大于同一相位的自动测量值（直径差值0.13mm，p < 0.001；周长差值0.42mm，p < 0.001；面积差值4.4mm，p < 0.001）。
自动最大动态CT主动脉环测量提供的值大于标准手动测量和同一相位自动测量的值。
