Glenoid rim morphology in recurrent anterior glenohumeral instability.

BACKGROUND: Knowledge regarding the morphology of the glenoid rim is important when patients with recurrent anterior glenohumeral instability are assessed. Ordinary imaging techniques are not always sensitive enough to demonstrate the morphology of the glenoid rim accurately. We developed a method of three-dimensionally reconstructed computed tomography with elimination of the humeral head to evaluate glenoid morphology. The purpose of the present study was to quantify glenoid osseous defects and to define their characteristics in patients with recurrent anterior instability.
METHODS: The morphology of the glenoid rim in 100 consecutive shoulders with recurrent unilateral anterior glenohumeral instability was evaluated on three-dimensionally reconstructed computed tomography images with the humeral head eliminated. The configuration of the glenoid rim was evaluated on both en face and oblique views. Concurrently, we also investigated seventy-five normal glenoids, including both glenoids in ten normal volunteers. Shoulders without an osseous fragment at the anteroinferior portion of the glenoid were compared with the contralateral shoulder in the same patient to determine if the glenoid morphology was normal. In shoulders with an osseous fragment, the fragment was evaluated quantitatively and its size was classified as large (>20% of the glenoid fossa), medium (5% to 20%), or small (<5%). Finally, all 100 shoulders were evaluated arthroscopically to confirm the presence of the lesion at the glenoid rim that had been identified with three-dimensionally reconstructed computed tomography.
RESULTS: Investigation of the normal glenoids revealed no side-to-side differences. Investigation of the affected glenoids revealed an abnormal configuration in ninety shoulders. Fifty glenoids had an osseous fragment. One fragment was large (26.9% of the glenoid fossa), twenty-seven fragments were medium (10.6% of the glenoid fossa, on the average), and twenty-two were small (2.9% of the glenoid fossa, on the average). In the forty shoulders without an osseous fragment, the anteroinferior portion of the glenoid appeared straight on the en face view and it appeared obtuse or slightly rounded, compared with the normally sharp contour of the normal glenoid rim, on the oblique view, suggesting erosion or a mild compression fracture at this site. Arthroscopic investigation revealed a Bankart lesion in ninety-seven of the 100 shoulders and an osseous fragment in forty-five of the fifty shoulders in which an osseous Bankart lesion had been identified with the three-dimensionally reconstructed computed tomography. In the shoulders with distinctly abnormal morphology on three-dimensionally reconstructed computed tomography, the arthroscopic appearance of the anteroinferior portion of the glenoid rim was compatible with the appearance demonstrated by the three-dimensionally reconstructed computed tomography.
CONCLUSIONS: We introduced a method to evaluate the morphology of the glenoid rim and to quantify the osseous defect in a simple and practical manner with three-dimensionally reconstructed computed tomography with elimination of the humeral head. Fifty percent of the shoulders with recurrent anterior glenohumeral instability had an osseous Bankart lesion; 40% did not have an osseous fragment but demonstrated loss of the normal circular configuration on the en face view and an obtuse contour on the oblique view, suggesting erosion or compression of the glenoid rim.