Assessment of the aortic annulus by multislice computed tomography, contrast aortography, and trans-thoracic echocardiography in patients referred for transcatheter aortic valve implantation

Objective: We sought to determine the level of agreement and the reproducibility of trans-thoracic echocardiography (TTE), contrast aortography (CA) and multislice computed tomography (MSCT) for the assessment of the aortic annulus, in patients referred for Transcatheter Aortic Valve Implantation (TAVI). Background: Correct measurement of the aortic annulus is important for TAVI. Methods: The dimensions of the aortic annulus were measured using TTE, CA and MSCT in 70 patients with severe aortic stenosis, referred for TAVI. Agreement between imaging techniques and interobserver variability was assessed using the Bland - Altman method and a linear regression model. Results: The MSCT Coronal view provided the largest mean annulus diameter (26.3 mm) followed by CA (24.4 mm), MSCT Mean (23.7 mm), TTE (22.6 mm), and MSCT Sagittal (21.8 mm) view. Differences in the annulus measurements were significant: MSCT Coronal view versus CA (mean, 95% confidence interval, Pearson's correlation) 2.0 mm, -1.9 to 6.0 mm, r = 0.72, CA versus MSCT Mean 0.2 mm, -3.3 to 3.7 mm, r = 0.76, MSCT Mean versus TTE 1.3 mm, -2.9 to 5.5 mm, r = 0.61, TTE versus MSCT Sagittal view 0.9 mm, -3.6 to 5.4 mm, r = 0.59, CA versus TTE 1.5 mm, -3.0 to 5.9 mm, r = 0.57. Interobserver variability was: TTE (mean, 95% confidence interval, Pearson's correlation) 0.29 mm, -4.2 to 4.8 mm, r = 0.57, CA 0.14 mm, -3.5 to 3.8 mm, r = 0.77, MSCT Mean 0.20 mm, -1.4 to 1.8 mm, r = 0.95. Conclusions: We found significant differences in the dimensions of the aortic annulus measured by MSCT, CA, and TTE. Interobserver variability for TTE and CA was substantially higher compared with MSCT.