Variability in the Contour of Cadaveric Anterior and Posterior Glenoids Based on Ipsilateral 3-Dimensional Computed Tomography Reconstructions: Implications for Clinical Estimation of Bone Loss

Purpose: To compare differences in the contour of the anterior and posterior glenoid in 3-dimensional computed tomography reconstructions as a reason for variability in estimating glenoid bone loss.

Methods: Three reviewers digitally placed 11 points (1 at 6 o'clock, 5 from 3 to 6 o'clock, and 5 from 6 to 9 o'clock) along the inferior half of glenoid 3-dimensional computed tomography reconstructions of 32 cadaveric specimens. A best-fit circle for the whole glenoid was determined from all 11 points. The anterior 6 points determined an anterior-based circle, as for estimation of posterior bone loss. The posterior 6 points determined a posterior-based circle, as for estimation of anterior bone loss. The area and radii were determined for all circles. Individual intraclass correlations were used to evaluate reliability. Paired t-tests and Wilcoxon signed rank tests were used to compare areas and radii of the anterior and posterior circles relative to the best-fit glenoid circle. Significance was defined as P < .05.

Results: There was good-to-excellent inter-rater reliability for the overall best-fit circle (intraclass correlation coefficient [ICC] = 0.97 [95% confidence interval (CI), 0.95-0.99]), anterior circle (ICC = 0.82 [95% CI, 0.71-0.90]), and posterior circle (ICC = 0.78 [95% CI, 0.65-0.88]). The average area of the posterior-based circle was significantly greater than the area of the best-fit circle for the intact glenoid (111% [95% CI, 104.3%-117.7%]; P = .0016). The average radius of the posterior-based circle was significantly larger than the radius of the intact glenoid (105% [95% CI, 101.7%-108.3%]; P = .0042). There were no significant differences between the average area (96% [95% CI, 89.0%-103.0%]; P = .16) and average radius (97% [95% CI, 93.5%-100.5%]; P = .19) for the anterior-based circle relative to the actual whole glenoid.

Conclusions: Estimation of anterior glenoid bone loss based on the normal posterior glenoid rim may overestimate glenoid bone loss because of differences in the contour of the anterior and posterior glenoid, whereas estimations of posterior glenoid bone loss based on the anterior rim did not differ significantly from the intact glenoid.

Clinical relevance: A circle based on the remaining intact portion of the injured glenoid may lead to an inaccurate estimate of glenoid bone loss.