Validation of a CT-based determination of the glenohumeral index

The amount of bony support by the glenoid can be determined using the glenohumeral index, i.e. the maximum anteroposterior (AP) diameter of the humeral head divided by the maximum AP dimension of the glenoid. This index has been described theoretically, but has never been validated in practice. In this study we used 20 cadaver shoulders to determine the glenohumeral index in two different ways. One method evaluated the glenohumeral index on a CT scan of the shoulders. The second method determined the anatomical glenohumeral index of the same shoulders by direct measurement of anatomical specimens using a digital caliper. All CT and caliper measurements were repeated by three different investigators. We used the Wilcoxon Signed Rank Test, to calculate the statistical significance of intra-observer and inter-observer differences in measurements on CT and with the caliper. Statistical analysis showed no significant differences between CT scan and caliper measurements for each investigator separately, but we found a statistically significant inter-observer variability concerning the caliper measurements obtained by two different investigators. This study demonstrates that a two-dimensional CT scan of the shoulder is a reliable and very accurate tool to calculate the glenohumeral index, as the values measured for the AP diameter of the humeral head and the AP dimension of the glenoid compare well with those measured in vitro on anatomical specimens.