Abstract
Objectives:
Current knowledge of the appropriate site of osteochondral allograft harvest to match glenoid morphology for the purposes of glenoid resurfacing is lacking. This has led to difficulty with adequately restoring the geometry of the glenoid using current available techniques. The purpose of this study was to quantify the articular surface topography of the glenoid and medial tibial plateau via 3-dimensional (3D) modeling to determine if the medial tibial articular surface provides an anatomic topographic match to the articular surface of the glenoid. We hypothesized that the medial tibial plateau will provide a suitable osteochondral harvest site due to its concavity and anatomic similarity to the glenoid.
Methods:
Computed tomography (CT) was performed on four cadaveric proximal tibias and four scapulae, allowing for 16 glenoid-tibial comparative combinations. 3D CT models were created and exported into point cloud models. A local coordinate map of the glenoid and medial tibial plateau articular surfaces was created. Two zones of the medial tibial articular surface (anterior and posterior) were quantified. The glenoid articular surface was defined as a best-fit circle of the glenoid articular surface maintaining a 2mm bony rim (Figure 1). This surface was virtually placed on a point on the tibial articular surface in 3D space. The tibial surface was segmented and its 3D surface orientation was determined by an eigenvector in the direction of its surface. 3D orientation of the glenoid surface was reoriented so that an eigenvector in the direction of the glenoid surface matched that of the tibial surface (Figure 2). The least distances between the point-clouds on the glenoid and tibial surfaces were calculated. The glenoid surface was rotated 360 degrees around the eigenvector with one degree increments and the mean least distance was determined at each rotating angle. A non-parametric wilcoxon signed rank statistical analysis was performed to compare the findings between the anterior and posterior aspects of the medial tibial articular surface with respect to the glenoid.
Results:
When the centroid of the glenoid surface was placed on the medial tibial articular surface, it covered approximately two-thirds of the anterior or posterior tibial surfaces. Overall, the mean least distance difference in articular congruity of all 16 glenoid-medial tibial surface combinations was 0.74mm (Std. Deviation ± 0.13). The mean least distance difference of the anterior and posterior two-thirds of the medial tibial articular surface was 0.72mm (± 0.13) and 0.76mm (± 0.16), respectively. There was no significant difference between and the anterior and posterior two-thirds of the tibia with regard to topographic match of the glenoid (p=0.187).
Conclusion:
We describe a novel methodology to quantify the topography of the tibial and glenoid articular surfaces. The findings suggest that the medial tibial articular surface provides an appropriate anatomic match to the glenoid articular surface. To the authors knowledge, this is the first time this relationship has been quantified. Both the anterior and posterior two-thirds of the medial tibial articular surface can serve as potential sites for osteochondral graft harvest. In addition, this methodology can be applied to future studies evaluating the ideal sites of graft harvest to treat zonal glenoid bone wear and/or loss.