Comparison of two different measurement methods to determine glenoid bone defects: area or width?

This study compared two different techniques that have been used to measure the glenoids of patients with recurrent anterior shoulder dislocation.

We analyzed 36 patients who had received arthroscopic Bankart repair for anterior shoulder instability. Retrospectively, 3-dimensional computed tomography images of both shoulders were available for these patients. Two measurement methods were compared to determine the glenoid defects. One of these techniques is based on linear measurement, previously defined as the glenoid index. The other method is based on surface area measurement. Subsequently, 3 more diameters and the average values obtained from these diameters were compared with the surface measurement method. Pearson correlation coefficient (r) was assessed to determine the relationship.

There was an almost perfect relationship between measurement methods when the defect area was less than 6% of the inferior glenoid circle (r, 0.915; P < .001). This relation decreased and the difference became more pronounced (r, 0.343; P = .657) when the bone loss exceeded 14% of the inferior glenoid circle. The highest correlations with the actual defects were the average values obtained from 4 different diameters (r, 0.964; P < .001) and the 4-o'clock position of the single diameter measurements (r, 0.860; P = .001). In addition, 11 patients had crescent-like defects, demonstrating a relatively low correlation between the measurement methods (r, 0.679; P = .021).

Although the best correlation was achieved from average values obtained from different diameter positions, in practical use, we advise a linear measurement to estimate the glenoid bone loss at the 4-o'clock position to achieve a high correlation between the measurement techniques.