Metopic "ridge" vs. "craniosynostosis": Quantifying severity with 3D curvature analysis

Abstract

Background: There has been an increased incidence of surgical treatment for metopic craniosynostosis (MCS) over the past decade. MCS presents as a wide spectrum, ranging from severe trigonocephaly on one end to an isolated metopic ridge on the other. Current surgical diagnosis relies upon subjective clinical assessment of patients' cranial shape, which is often combined with impressions from radiologic imaging. In our study, we have developed a semi-automated methodology using three-dimensional curvature analysis to rigorously separate the phenotypes along the spectrum.

Methods: Three clinically distinct groups of patients with CT images were obtained: 1) Normal subjects without any deformity; 2) "Benign" metopic ridge (BMR) without classic trigonocephaly; 3) Severe trigonocephaly with MCS. CT scans were converted into three-dimensional skull surface images, and curvature analysis was performed using two user-defined regions of interest (mid-forehead and lateral orbit).

Results: Three-dimensional curvature analysis of mid-forehead strip, and right/left lateral orbital areas was performed in thirty patients. The difference in average mean curvature between true MCS and BMR was 10.5 m(-1) and -18.6 m(-1) for mid-forehead strip and right/left lateral orbital rim, respectively. The average mean curvatures of mid-forehead strip and right/left lateral orbit areas among the three groups were significantly different (p < 0.0001). K-means clustering reliably classified patients into different severity groups based on average mean curvature of the two regions of interest.

Conclusions: The described methods are effective in classifying severity of orbitofrontal deformity in the spectrum of MCS, which can aid providers in identifying an appropriate threshold for surgical treatments of MCS.

Keywords: Computed tomography; Curvature analysis; Metopic craniosynostosis; Metopic ridge; Trigonocephaly.