The Influence of Bone Loss on Glenoid Version Measurement: A Computer-Modeled Cadaveric Analysis

To characterize how increasing computed tomography (CT)-quantified glenoid bone loss influences measured version.

Six embalmed cadaveric shoulders were used for this study. Glenoid bone defects were computer modeled in cadaveric shoulders; CT images were obtained and segmented using OsiriX software, creating 3-dimensional en face glenoids. Glenoid defects were made on CT images of intact glenoids superimposed with a glenoid clock face viewed en face to simulate anterior and posterior bone loss. Bony defects in various positions comprising 3%, 9.5%, and 19.5% were created posteriorly. Best-fit circles were superimposed to represent 10% and 25% defects anteriorly. Version was measured using the Friedman method.

The average glenoid version measured 4° of retroversion, 2° after 10% anterior bone loss, and neutral version in the 25% bone loss group. Version was significantly altered when we compared intact glenoids versus 10% and 25% anterior glenoid bone loss (P < .001). Increasing from 10% to 25% bone loss showed a significant difference in measured version (P = .025). Posterior defects from the 6:30 to 8:30 clock-face position averaged 4.6° of retroversion; from the 6:30 to 9:30 clock-face position, 6.2° of retroversion; and from the 6:30 to 10:30 clock-face position, 8.7° of retroversion. When comparing glenoid defects at the 6:30 to 8:30 clock-face position with those involving the 6:30 to 9:30 and 6:30 to 10:30 clock-face positions (P < .001), a 1° correction may be used for every 5% of bone loss to account for version changes seen with bone loss.

In this cadaveric analysis, glenoid version was altered in the setting of increasing posterior and anterior bone loss. A correction factor may be considered to account for this. When comparing glenoid defects at the 6:30 to 8:30 clock-face position with those involving the 6:30 to 9:30 and 6:30 to 10:30 clock-face positions (P < .001), a 1° correction may be used for every 5% of bone loss to account for version changes seen with bone loss.

This cadaveric study shows that glenoid bone loss alters glenoid version, as measured by CT, in a meaningful way. This information is important in managing anterior and posterior shoulder instability, and correction of measured version should be considered in this setting to provide an accurate and comprehensive evaluation.