Background: Accurate knowledge of the scapular anatomy is fundamental for the preoperative evaluation but some bony landmarks are difficult to identify. Statistical approaches based on subject-specific parametric models could be used to overcome this difficulty. The aim of this study was to propose a quantitative parametric model of the scapula and to analyze correlations between descriptive morphologic parameters.
Materials and methods: Forty-three scapulae were scanned and reconstructed. Each 3D scapula was regionalized and a simple geometric element was best fitted on each region using least square method. Descriptive parameters of each region were obtained. Correlation and linear regression analyses were performed between all measurements in order to assess parameters that can be used as predictors of the other descriptive parameters.
Results: Morphometric scapular measurements from 3D reconstructions were obtained. Correlation and linear regression analyses assessed correlations between the glenoid width and both the glenoid height and the acromial width. Also, we obtained correlation between the orientation of the inferior part of the acromion on the A-P view and on the axillary view.
Discussion: Parametric models are widely used in biomechanics for identifying anatomical landmarks or rotations centers of these structures. For the scapula, no such model is available. We elaborated a first parametric model of scapula based on a large database of 43 scapulae. Our morphometric measurements are very close to others founded in literature. Correlations obtained should help to progress toward relevant subject-specific models of the scapula based on reduced information.