Intermodality variation of aortic dimensions: How, where and when to measure the ascending aorta

Background: No established reference-standard technique is available for ascending aortic diameter measurements. The aim of this study was to determine agreement between modalities and techniques. Methods: In patients with aortic pathology transthoracic echocardiography, computed tomography angiography (CTA) and magnetic resonance angiography (MRA) were performed. Aortic diameters were measured at the sinus of Valsalva (SoV), sinotubular junction (STJ) and tubular ascending aorta (TAA) during mid-systole and end-diastole. In echocardiography both the inner edge-to-inner edge (I-I edge) and leading edge-to‑leading edge (L-L edge) methods were applied, and the length of the aortic annulus to the most cranial visible part of the ascending aorta was measured. In CTA and MRA the I-I method was used. Results: Fifty patients with bicuspid aortic valve (36 ± 13 years, 26% female) and 50 Turner patients (35 ± 13 years) were included. Comparison of all aortic measurements showed a mean difference of 5.4 ± 2.7 mm for the SoV, 5.1 ± 2.0 mm for the STJ and 4.8 ± 2.1 mm for the TAA. The maximum difference was 18 mm. The best agreement was found between echocardiography L-L edge and CTA during mid-systole. CTA and MRA showed good agreement. A mean difference of 1.5 ± 1.3 mm and 1.8 ± 1.5 mm was demonstrated at the level of the STJ and TAA comparing mid-systolic with end-diastolic diameters. The visible length of the aorta increased on average 5.3 ± 5.1 mmW during mid-systole. Conclusions: MRA and CTA showed best agreement with L-L edge method by echocardiography. In individual patients large differences in ascending aortic diameter were demonstrated, warranting measurement standardization. The use of CTA or MRA is advised at least once.