Background: Previous studies have shown cross-sectional 3-dimensional (3D) transesophageal echocardiographic (TEE) measurements to severely underestimate multidetector row computed tomographic (MDCT) measurements for the assessment of aortic annulus before transcatheter aortic valve replacement. This study compares annulus measurements from 3D-TEE using off-label use of commercially available software with MDCT measurements and assesses their ability to predict paravalvular regurgitation.
Methods and results: One hundred patients with severe, symptomatic aortic stenosis who had both contrast MDCT and 3D-TEE for annulus assessment before balloon-expandable transcatheter aortic valve replacement were analyzed. Annulus area, perimeter, and orthogonal maximum and minimum diameters were measured. Receiver operating characteristic analysis was performed with mild or greater paravalvular regurgitation as the classification variable. Three-dimensional TEE and MDCT cross-sectional perimeter and area measurements were strongly correlated (r=0.93-0.94; P<0.0001); however, the small differences (≤1%) were statistically significant (P=0.0002 and 0.0074, respectively). Discriminatory ability for ≥ mild paravalvular regurgitation was good for both MDCT (area under the curve for perimeter and area cover index=0.715 and 0.709, respectively) and 3D-TEE (area under the curve for perimeter and area cover index=0.709 and 0.694, respectively). Differences in receiver operating characteristic analysis between MDCT and 3D-TEE perimeter and area cover indexes were not statistically significant (P=0.15 and 0.35, respectively).
Conclusions: Annulus measurements using a new method for analyzing 3D-TEE images closely approximate those of MDCT. Annulus measurements from both modalities predict mild or greater paravalvular regurgitation with equivalent accuracy.
Keywords: aortic valve insufficiency; aortic valve stenosis; echocardiography; tomography.