64-slice MDCT arthrography in shoulder instability: our experience

PURPOSE: This study was performed to assess the diagnostic accuracy of air-contrast 64-slice multidetector computed tomography (MDCT) arthrography in the evaluation of glenohumeral joint instability by comparison with conventional arthroscopy. MATERIALS AND METHODS: Fifty patients with a history of shoulder instability underwent MDCT arthrography with thin collimation scans. The raw data were transferred to a workstation and processed using multiplanar reformation (MPR) and volume rendering (VR) algorithms. All patients subsequently underwent conventional arthroscopy. The results of the two techniques were compared and their sensitivity and specificity calculated. RESULTS: We diagnosed eight anterosuperior labrum lesions (group 1), 32 anteroinferior labrum lesions (group 2) and two posterior labrum lesions (group 3). Overall sensitivity and specificity (groups 1, 2, 3) were 88% and 100%, respectively. In group 1, sensitivity was only 66% (four false negatives), whereas in groups 2 and 3, it was 94% (two false negatives) and 100%, respectively. The labrum lesions were also found to be associated, with 100% sensitivity and specificity, with 20 lax capsules, 17 Hill-Sachs lesions, five Bankart lesions, two Perthes lesions and three complete rotator-cuff tears. CONCLUSIONS: Air-contrast MDCT arthrography is fast, reproducible, well tolerated and very accurate in the evaluation of lesions causing shoulder instability.