Reliability of non-contrast magnetic resonance angiography-derived aortic diameters in Marfan patients: comparison of inner vs. outer vessel wall measurements

Figures

• 

  Fig. 1. Para-sagittal ECG-gated non-contrast MRA of the thoracic aorta with in a 28-year-old man with confirmed Marfan syndrome. Indicated measurement levels from proximal to distal: sinuses of Valsalva (1), sinotubular junction (2), ascending aorta (3), aortic arch (4) and descending aorta (5)

• 

  Fig. 2. Para-sagittal non-contrast MRA of the aortic root in a 29-year-old man with Marfan syndrome illustrating the advantage of inner diameter measurements. Note the clear delineation of the inner vessel contour at the level of the sinotubular junction and sinuses of Valsalva (arrowheads) due to stark contrast of the hypointense vessel wall to the hyperintense vessel lumen. Compare the hypointense outer contour of the aorta that is obscured by hypointense perivascular tissue (asterisks). A clear delineation of the outer vessel wall is only possible at the level of the sinuses of Valsalva if adjacent to the hyperintense lumen of the left atrium (arrow). Both readers rated the inner vessel wall delineation for the sinuses of Valsalva and sinotubular junction as excellent (grade 1) whereas the outer vessel wall of the sinuses of Valsalva was rated as well defined (grade 2) and sinotubular junction as poorly defined with low contrast (grade 3). Both observers measured identical inner diameters (40 mm vs. 40 mm; bias 0 mm) and deviating outer diameters (48 mm vs. 44 mm; bias 4 mm) for the sinuses of Valsalva

• 

  Fig. 3. Intra- and interobserver agreement of inner vs. outer aortic diameter measurements using non-contrast MRA at the sinuses of Valsalva. a, b Bland–Altman plots of intraobserver agreement demonstrate a significantly smaller intraobserver variance of a inner measurements compared to b outer measurements (p < 0.001). c, d Analyses of interobserver agreement measurements also demonstrate a significantly smaller interobserver variance of c inner measurements compared to d outer measurements (p = 0.001). Middle solid line indicates mean bias of diameter measurements. Dotted lines indicate limits of agreement

• 

  Fig. 4. Bland–Altman comparison of the measured diameters at the sinuses of Valsalva assessed by inner and outer non-contrast MRA measurements. The plot illustrates a significant difference of − 5.8 ± 2.6 mm for inner measurements compared to outer measurements (p < 0.0001). Middle solid line indicates mean bias of diameter measurements. Dotted lines indicate limits of agreement

• 

  Fig. 5. Bland–Altman comparison of inner and outer non-contrast MRA diameter measurements vs. echocardiographic leading-edge measurements at the sinuses of Valsalva. The plots reveal a difference of a 1.1 ± 2.3 mm for inner MRA measurements (p = 0.014) and a difference of b 6.9 ± 3.1 mm for outer MRA measurements (p < 0.001) when compared to echocardiography. Keep in mind the para-sagittal orientation of MRA-derived measurements vs. parasternal long axis-view orientation of echocardiographic measurements. Middle solid line indicates mean bias of diameter measurements. Dotted lines indicate limits of agreement

• 

  Fig. 6. Para-sagittal non-contrast MRA of three Marfan patients with aortic root aneurysms with diameters larger than 45 mm. a–c All three aneurysms were correctly identified using MRA-derived inner-to-inner measurements. These findings were confirmed by echocardiography and all patients underwent prophylactic aortic root replacement. Both readers rated the inner vessel wall delineation for the sinuses of Valsalva as superior when compared to outer vessel wall delineation. Both observers measured comparable inner diameters: A: 47 vs. 48 mm; B: 51 vs. 51 mm; C: 50 vs. 50 mm which were similar to diameters obtained by echocardiography: A: 46 mm; B: 49 mm; C: 51 mm. Outer diameters showed a higher deviation: A: 50 vs. 52 mm B: 55 vs. 59 mm; C: 53 vs. 57 mm