Publications of Faculty of Medicine:FROZEN SHOULDER: RELIABILITY OF SHOULDER MR ARTHROGRAPHY: Abstract

Title:
FROZEN SHOULDER: RELIABILITY OF SHOULDER MR ARTHROGRAPHY
Full paper Not Available
Abstract:

PURPOSE: The purpose of this study is to evaluate the MR arthrographic findings in patients with frozen shoulder and to test its reliability using shoulder arthroscopic findings as the gold standard. MATERIALS AND METHODS: During a period of 13 months, 19 patients clinically diagnosed as frozen shoulder syndrome were evaluated using MR arthrography and results were confirmed by shoulder arthroscopy within 3 months of the MR arthrography. Nineteen other subjects free of frozen shoulder as proved clinically and arthroscopically were used as control subjects matching with the patient group in number, sex, and age, those subjects underwent MR arthrography for other shoulder pathologies within 3 months before surgery. Evaluation criteria of the MR arthrography were; thickness of the coracohumeral ligament (CHL), thickness of the joint capsule, volume of the axillary recess, abnormalities of the CHL, subcoracoid fat triangle, superior glenohumeral ligament, subscapularis tendon, long head of biceps tendon, and subscapularis recess. RESULTS: Thickened coracohumeral ligament (Cl-IL) was significantly more common in patients with frozen shoulder than in control subjects (12 patients vs I control subject) with a threshold value of 4 mm, it has 63% sensitivity and 95% specificity for diagnosing frozen shoulder, while thickened capsule in the rotator cuff interval with a threshold value of 7 mrn was having 68% sensitivity and 84% specificity and was more prevalent in patients than control subjects, in contrary to thickened capsule in the axillary recess which was not significant. Small axillary recess volume with a threshold value of 0.53 mL was more common in patients with frozen shoulder, having sensitivity of 84% and specificity of 74%. Complete obliteration of the subcoracoid fat triangle has a specificity of 100% for diagnosing frozen shoulder, yet, not sensitive (42%). Abnormality of the CHL and synovitis-like abnormality at the superior border of the subscapularis tendon and in subscapularis recess were more common in patients than control subjects. CONCLUSION: MR arthrography is a reliable imaging modality for diagnosing frozen shoulder, with characteristic findings including thickened CHL, thickened capsule at the rotator cuff interval and complete obliteration of the subcoracoid fat triangle.