[Reliability of three dimensional CT measurement of glenoid track in patients with recurrent anterior dislocation of shoulder].

OBJECTIVE

To evaluate the reliability and repeatability of three-dimensional (3D)-CT measurement of glenoid track.

METHODS

Glenoid and Hill-Sachs bone defects in 60 patients with recurrent anterior dislocation of the unilateral shoulder were evaluated by 3D-CT and analyzed by three independent observers(shoulder orthopedists) according to the Di Giacomo method. These injuries were classified to formulate a surgical protocol. All the patients were repeatedly measured 1 week later. Intraclass correlation coefficient (ICC) and coefficient of variation (COV) were used for reliability analysis.

RESULTS

Interobserver reliability was "very good" for both measurements of glenoid diameter and glenoid bone defects, and "good" for measurements of Hill-Sachs interval. Interobserver agreement was high on the assessment of the extent of the glenoid defect, and poor on track/off track assessment of the Hill-Sachs lesion. Intraobserver reliability for measurements of glenoid diameter, glenoid defect, and Hill-Sachs interval was "very good". The COV was 4.1% for glenoid diameter, 4.4% for glenoid defect, and 21.1% for Hill-Sachs interval.

CONCLUSION

The reliability and reproducibility of 3D-CT measurements of glenoid bone defects are good, but the assessment of humeral head bone defects shows large variability with poor reliability.