Efficacy of diagnostic magnetic resonance imaging for articular cartilage lesions of the glenohumeral joint in patients with instability

The purpose of this study was primarily to assess the diagnostic performance of magnetic resonance imaging (MRI) in detecting articular cartilage injuries in patients with glenohumeral instability. A secondary purpose was to assess the diagnostic performance of MRI for detection of Hill-Sachs and Bankart lesions. A cohort of 87 consecutive patients who underwent diagnostic MRI and shoulder arthroscopy for instability from 1997 to 2006 were identified. Fifty-five patients (63.2%) underwent MRI with intra-articular contrast medium and 32 patients (36.8%) underwent MRI without contrast medium. MR images were reviewed by two radiologists and interpreted by consensus for the presence of articular cartilage lesions (including Hill-Sachs and Bankart lesions), which were then confirmed by reviewing the operative report and images recorded at arthroscopy. Mean patient age was 27.0 +/- 10.2 years with a mean clinical and radiographic follow-up of 29 (range 3-72) months. Cartilage injuries were detected arthroscopically in 55 patients (63%). Bankart and Hill-Sachs lesions were identified arthroscopically in 66 patients (75.9%) and 55 patients (63.2%) respectively. The overall sensitivity and specificity for detection of glenohumeral articular cartilage lesions by MRI were 87.2% and 80.6% respectively. The sensitivity and specificity of MRI in detecting Bankart lesions was 98.4% (95% CI 91.9, 99.7) and 95.2% (95% CI 77.3, 99.2) respectively. The sensitivity and specificity of MRI in detecting Hill-Sachs lesions was 96.3% (95% CI 87.6, 98.9%) and 90.6% (95% CI 75.7, 96.9) respectively. No statistically significant difference was found between MRI examinations with and without intra-articular gadolinium (p = 0.89). Magnetic resonance imaging demonstrates high sensitivity and specificity for the diagnosis of articular cartilage injuries in patients with glenohumeral instability. MRI with or without intra-articular contrast medium in this study were equally reliable as a non-invasive method for assessment of articular cartilage damage of the glenohumeral joint prior to diagnostic arthroscopy.