One hundred forty patients with severe aortic valvular disease or aortic dilatation were prospectively evaluated using 2D TTE and MDCT (n = 70) or MRI (n = 70). The aorta was measured at three different levels: sinuses of Valsalva, sinotubular junction, and ascending aorta. Three different measurements were made by 2D TTE—inner edge to inner edge, leading edge to leading edge (L-L), and outer edge to outer edge—and then compared with the inner edge–to–inner edge and outer edge–to–outer edge measurements of cusp-to-cusp and cusp-to-commissure diameters by MDCT or MRI. Inter- and intraobserver variability was analyzed.
Aortic diameters by 2D TTE, MDCT, and MRI showed excellent inter- and intraobserver variability using all conventions. Significant underestimation was observed of all aortic diameters assessed by 2D TTE using the inner edge–to–inner edge convention compared with those obtained by MDCT or MRI (P < .0001). However, excellent accuracy was observed by 2D TTE when the L-L convention was used and compared with the internal diameter by MDCT and MRI (mean differences, 0.6 ± 2.6 mm [P = .158] for MDCT and 0.4 ± 3.5 mm [P = .852] for MRI). Cusp-to-cusp diameters were slightly larger than cusp-to-commissure diameters. The diameter by 2D TTE using the L-L convention correlated best with the noncoronary cusp–to–right coronary cusp diameter determined by both MDCT and MRI.
Aortic root and ascending aortic diameters measured by 2D TTE using the L-L convention showed accurate and reproducible values compared with internal diameters assessed by MDCT or MRI. This approach permits a multimodality follow-up of patients with aortic diseases and avoids disparities in measurements obtained by different conventions.