Diagnosis of glenoid labral tears. A comparison between magnetic resonance imaging and clinical examinations

We studied 54 patients with shoulder pain secondary to anterior instability or glenoid labral tears refractory to 6 months of conservative management with no evidence of rotator cuff lesions. All patients had sufficient preoperative clinical data, magnetic resonance imaging, and shoulder arthroscopy results for analysis. The ability to predict the presence of a glenoid labral tear by physical examination was compared with that of magnetic resonance imaging (conventional and arthrogram) and confirmed with arthroscopy. There were 37 men and 17 women (average age, 34 years) in the study group. Of this group, 64% were throwing athletes and 61% recalled specific traumatic events. Clinical assessment included history with specific attention to pain with overhead activities, clicking, and instances of shoulder instability. Physical examination included the apprehension, relocation, load and shift, inferior sulcus sign, and crank tests. Shoulder arthroscopy confirmed labral tears in 41 patients (76%). Magnetic resonance imaging produced a sensitivity of 59% and a specificity of 85%. Physical examination yielded a sensitivity of 90% and a specificity of 85%. Physical examination is more accurate in predicting glenoid labral tears than magnetic resonance imaging. In this era of cost containment, completing the diagnostic workup in the clinic without expensive ancillary studies allows the patient's care to proceed in the most timely and economic fashion. Glenoid labral tears have been associated with overhead throwing activities, trauma, and shoulder instability. Assessment of an athlete with shoulder pain should take into account a careful history of clicking sounds or catching, symptoms with overhead activities, reports of instability, or previous trauma. On physical examination, patients with labral tears often demonstrate objective instability with or without clicking or catching during glenohumeral rotation. Plain radiographs have not been helpful, and radiologists have relied on techniques from arthrogram to arthrotomogram, CT arthrogram, magnetic resonance imaging (MRI), and MR arthrogram to assist in the diagnosis. Various sensitivities and specificities have been reported for these tests. However, a large degree of intra- and interobserver variability has been demonstrated, and the degree to which these studies are helpful in preoperative planning has been questioned. No previous study to our knowledge has involved MRI in a direct comparison of other diagnostic modalities. Therefore, the purpose of this study is to investigate the accuracy of MRI and physical examination in the diagnosis of glenoid labral tears.