The aim was to define the operative and MRI arthrogram findings in recurrent post stabilization instability, to establish the accuracy of MRI findings compared to surgical findings, and to define the role of MRI in evaluation and planning for these patients.

The operative findings in 25 consecutive patients undergoing revision shoulder stabilization procedures were reviewed. 18 of these patients had MRI arthrograms prior to their revision procedure. All revision procedures were performed by the same surgeon, and all MRI scans reviewed by 2 musculoskeletal radiologists.

Primary and contributing factors for instability were identified for each patient. Primary factors for failure included; new injury at a different site to the index repair (6 patients), laxity of the inferior glenohumeral ligament (5 patients), failure of the index repair (7 patients) and failure to address the pathology at the index procedure (7 patients).

MRI arthrography had 85% sensitivity, 100% specificity and 89% accuracy.

MRI arthrography is accurate in assessment of the labrum in recurrent post stabilization instability. Recurrent instability is complex and multifactorial, but a primary factor can often be identified. In some cases new trauma results in injury at a different site to the initial repair. MRI arthrogram may assist in surgical planning. If a labral injury is present without laxity or a glenoid bony defect, arthroscopic revision may be undertaken. If significant capsular laxity or bone deficiency is present, then an open procedure with capsular shift and rotator interval closure may be appropriate.