Comparison of Glenoid Dimensions Between 3D Computed Tomography and 3D Printing.

Abstract

Glenoid dimensions can be measured in vivo with various imaging methods including two-dimensional (2D) and three-dimensional computed tomography (CT) and magnetic resonance imaging scans. Printing of three-dimensional (3D) models of the glenoid using imaging data is feasible and can be used to better understand skeletal trauma and complex skeletal deformations such as glenoid bone loss in patients with shoulder instability. The purpose of this study was to compare measurements of glenoid dimensions on 3D CT scan reconstructed models and 3D printed models of the glenoid. CT scans from 62 young, male adults acquired for non-trauma-related causes were evaluated. Following volume rendering, a stereolithography model of each scapula was constructed and a 3D model was printed. Additionally, 3D CT models of each glenoid were reconstructed using dedicated software. Measurements of the maximum glenoid height and width were performed on both the 3D printed and the 3D reconstructed models. To assess intra- and interrater reliability, measurements of 15 glenoids were repeated by two observers after three weeks. The measurements of the 3D printed and 3D reconstructed models were compared. Inter- and intra-rater reliability was excellent or perfect. Analysis of height and width values demonstrated a strong correlation of 0.91 and 0.89 respectively (p<0.001) for both the 3D printed models and the 3D reconstructed models. There was a strong correlation between the height and width, but no significant difference between the glenoid width and height in both models. There was no statistical significance between height and width when measurements on the two models were examined (p=0.12 and 0.23 respectively). 3D printed glenoid models can be used to evaluate the glenoid dimensions, width, and height, as they provide similar accuracy with 3D reconstructed models as provided from CT scan data.