Abstract
PURPOSE OF THE STUDY To evaluate the efficacy of magnetic resonance (MR) examination in intra- and extra- articular shoulder disorders, in comparison with arthroscopic findings, in patients with both acute injuries and chronic problems of the shoulder. MATERIAL AND METHODS Conventional MR images and MR arthrograms of the shoulder were obtained in 35 patients treated between January 2004 and January 2006.
Each MR image was evaluated by five radiologists experienced in assessing findings on the musculoskeletal system. Subsequently, the patients underwent shoulder arthroscopy performed by shoulder arthroscopy surgeons.
Each detailed arthroscopic report was compared with the pre-operative MR findings, and the sensitivity and specificity of MR examination were calculated. The arthroscopic findings served as standards of reference for comparison.
RESULTS SLAP lesions were found intra-operatively in seven out of the 35 patients (20 %). MR sensitivity was 43 % (3 to 7 patients) and specificity was 96 % (27 to 28).
The accuracy of MR for SLAP lesion diagnosis was 86 % (30 of 35 patients). Tears in the anterior labrum were diagnosed by arthroscopy in 16 of 35 patients (46 %); MR sensitivity was 44 % (7 of 16 patients) and specificity was 89 % (17 of 19).
The accuracy of MR for diagnosing labral tears was 68 % (24 of 35 patients). DISCUSSION When designing our study, we tried to avoid limitations of similar investigations.
We employed a nuclear magnetic resonance system with high resolution and standardized both the arthroscopic technique and the evaluation of MR images. Our results of diagnosing rotator cuff disorders and SLAP lesions are in agreement with those of other relevant studies.
The limitation of our study was a small size of the group. CONCLUSIONS Although MR imaging is a sensitive method suitable for diagnosing some shoulder disorders, the MR imaging of SLAP lesions and labral tears does not give results accurate enough to be used for pre-operative planning.