Marfan syndrome (MFS) is a connective tissue disease with high risk of aortic rupture and dissection. Two-thirds of dissections originate in the ascending aorta (AAo) while one-third occur in the descending aorta (DAo). The identification of changes in aortic dimension such as AAo dilatation plays an important role in risk stratification. However, literature shows that aortic dissections may occur in only moderately dilated aortas as well, predominantly in the descending¹. These findings indicate that aortic dimension alone may not capture the complex changes in aortic geometry that are often encountered in MFS patients, e.g. elongation and changes in overall shape of the aorta. Indeed, recent studies have shown that the aorta in MFS patients can have a different aortic shape (e.g., be more tortuous) and that a more tortuous aortic geometry was a predictor of clinical events, including aortic dissection². However, information on thoracic aortic shape in children with MFS and in age matched healthy pediatric control subjects is lacking. Therefore, the aim of the study was to systematically investigate aortic 3D geometry in a cohort of children and adolescents with MFS compared to an age appropriate control cohort.25 pediatric MFS patients (age 15.6±4.0 yrs, female=11) and 21 healthy subjects (age 16.0±2.6 yrs, female=12) with tricuspid aortic valves were included in this IRB-approved study. MFS was identified according to the 2010 Revised Ghent Criteria³; FBN1 mutation was proven in 21 patients. 4D flow MRI⁴ was performed at 1.5T with full 3D coverage of the thoracic aorta (spatial resolution=2.2-4.1x1.6-2.5x1.9-4.0 mm³; temporal resolution=37.6-40.8 ms) using prospective ECG gating and respiratory navigator gating. All 4D flow MRI data were corrected for velocity aliasing, Maxwell terms and eddy currents⁵. 3D PC-MR angiograms were used to obtain a 3D segmentation of the thoracic aorta (Mimics, Materialise, Leuven, Belgium) (Figure 1A-B). To construct the 3D aortic centerline ten to twelve control points, from the aortic valve (AoV) to the descending aorta at similar transverse level, were manually placed along the entire volume of the thoracic aorta (EnSight, CEI, Apex, NC) (Figure 1C). A spline interpolation was performed to construct a 3D vessel centerline and calculated the following parameters (Figure 1D): 1. Aortic arch width (AAW); 2. Aortic arch height (AAH); 3. Anterior arch width (A); 4. Posterior arch width (P). Accordingly, W/H ratios, A/W ratios and P/W ratios were calculated for each subject.

Basic characteristics:

Age, gender and BSA were comparable between the MFS group and healthy subjects (Table 1). As expected, MFS patients were taller compared to healthy subjects (MFS: 178.6±17.3 cm, Healthy subjects: 161.2±11.2, p<0.001; Table 1). Male subjects were taller compared to female subjects within each patient group; however this difference was not statistically significant (Table 1).

Aortic measures: As summarized in Table 1, MFS patients had greater absolute AAW (56.1±9.0 mm vs. 51.1±6.5 mm, p=0.043) and increased absolute AAH (79.1±10.8 mm vs. 64.1±8.4 mm, p<0.001) compared to the controls. Elevated AAW was mainly the result of a greater anterior portion of the width. This anterior portion was significantly greater in MFS compared to healthy subject (26.9±4.9 mm vs. 21.9±3.1 mm, p<0.001) while the posterior width distance was equal between the groups (29.3±6.3 mm vs. 29.5±5.0 mm, p=0.917). In addition, MFS patients showed increased H/W-ratio compared to healthy controls (1.43±0.24 vs. 1.26±0.14, p=0.006). Body habitus (weight, height and body surface area (BSA)) showed strong relationships with thoracic aortic shape measures. BSA was most prominent associated with aortic width measures (Figure 2A): AAW (r=0.743, p<0.001), anterior width (r=0.579, p<0.001), and posterior width (r=0.591, p<0.001). Body height showed a strong relationship with aortic arch height (r=0.712, p<0.001; Figure 2B).

The findings of this study demonstrate that pediatric MFS patient already have an altered thoracic aorta geometry compared to a group of age appropriate healthy subjects. Both aortic arch width and height were significantly altered and increased. Elevated AAW was mainly the result of a greater anterior portion. Considering that the distal descending aorta is more in an anatomically fixed position, this might suggest that this elongation of the ascending aorta may occur simultaneously with expansion of the aortic root and ascending diameter in Marfan patients. In accordance with the literature, height was found as a contributing factor of the differences between MFS and healthy control group. However, we did not investigate the relationship with aortic diameters and further longitudinal alterations in aortic shape in Marfan disease and the relation/influence of this aortic shape differences on aortic hemodynamics need to be investigated.