Background: Anterior glenohumeral instability is associated with anteroinferior rim defect of glenoid. Evaluation of such bony lesions may play a role in determining proper treatment and predicting recurrence of instability. The main purpose of the current study was to investigate the relation between number of dislocations and extent of bony lesion of the glenoid and also determine the sensitivity and specificity of 3-D CT for detection of patients requiring bone-grafting surgery. Methods: Thirty patients with recurrent anterior shoulder dislocation contributed to this study. The patients underwent bilateral 3-D CT imaging to determine the glenoid index (GI). Then, the correlation between the number of dislocations and GI, and also sensitivity and specificity of CT were determined. Results: The bony lesion was present in 28(94%) of patients. There was no meaningful relationship between the number of dislocations and GI (p=.05). Based on the CT, 2 patients required open surgery and bone grafting and 28 patients only Bankart procedure. This was proved to have been wrong in 2 cases at surgery. So, the sensitivity and specificity of 3-D CT was 50% and 96% respectively. Conclusion: The number of dislocations cannot determine the extent of the bony glenoid defect in recurrent shoulder dislocation lesion. 3-D CT can help in knowing the expansion and location of the lesion, but is not a reliable tool for surgical technique.