3D morphometric evaluation of craniofacial features in adult subjects with Marfan syndrome

Marfan syndrome (MFS) is a hereditable disorder of the connective tissue with an estimated worldwide prevalence of 1:5000, mainly caused by alterations of the extracellular matrix protein fibrillin-1. Cardiovascular manifestations of MFS, including aortic dilation with increased risk of dissection and rupture, are the leading cause of mortality. Therefore, an early and accurate diagnosis of MFS is crucial to prevent the development of complications, even if it may be difficult, due to the variable expression of the disease. It has been suggested that craniofacial abnormalities associated with MFS could predispose to obstructive sleep apnea, which in turn may promote aortic dilation. Since in a previous preliminary study on young subjects with MFS we pointed out some quantitative facial features never described before, we extended the study on adult patients, in order to better characterize the facial phenotype associated with MFS and verify the usefulness of a 3D not invasive quantitative approach for its early recognition. 3D facial images of 49 Italian subjects diagnosed with MFS without previous history of facial injuries or surgery, aged 18-60 years (18 males, mean ± SD age 37 ± 11 years; 31 females, mean ± SD age 40 ± 11 years), and divided in 5 non-overlapping age groups, were obtained by stereophotogrammetry. From the coordinates of 50 soft-tissue facial landmarks, linear distances and angles were measured; z score values were calculated comparing patients with healthy Italian reference subjects (332 males and 329 females), matched for gender and age group. Almost all subjects with MFS (96%) showed a shorter mandibular ramus than controls (mean z score = -1.8) and 100% of subjects showed a greater facial divergence (mean z score = +2.0), thus strengthening the findings of the previous study on young subjects with MFS. Furthermore, 98% of subjects showed a reduced ratio between posterior and anterior facial height (mean z score = -1.8) and 92% of subjects showed a reduced ratio between facial width and facial height (mean z score = -1.5), being both ratios influenced by an expected but overall mild increase of facial height. Remarkable gender differences or age-specific trends were not observed. Facial abnormalities pointed out in the current 3D morphometric study could represent traits of the phenotypic expression of MFS. Furthermore, since they are in accordance with the findings of our previous study on young patients, their detection could facilitate the early diagnosis of the disease, when the phenotype of MFS is not clearly evident or has not yet been clearly expressed. Further studies on more patients are needed to confirm the promising findings.