Comparison of shoulder ultrasound and MR imaging in diagnosing full-thickness rotator cuff tears
Introduction
Imaging diagnosis is important for the management of patients whose symptoms are related to the rotator cuff [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], particularly for the surgeons to initiate an operation plan. Due to technical improvements and advances in the understanding of anatomic and pathologic characteristics of the rotator cuff, ultrasound (US) and magnetic resonance imaging (MRI) have gradually replaced the role of arthrography for diagnosing rotator cuff tear during the past decade [1], [2], [3], [4], [5], [6], [7], [8], [9], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21]. Both methods are now accepted as routine diagnostic regimens [2], [3], [4], [5], [6], [7], [8], [9]. However, the decision as to which one is better remains undetermined, and only limited English radiological literatures have compared the two modalities in interpretation of a full-thickness rotator cuff tear (FTRCT). The purpose of the present study is twofold: one is to evaluate the concordance of the two modalities in interpretation of FTRCTs and another is to evaluate their diagnostic performance using surgical findings as gold standard.
Section snippets
Patients
A retrospective search of reports in computer system of our department from July 1996 to October 2000 yielded 506 consecutive patients who underwent both shoulder US and MRI examinations in our hospital. Excluding cases examined for the evaluation of shoulder joint infectious or neoplastic disease and those who had previous shoulder operation, 422 patients remain. Using shoulder surgery after imaging study as an additional criterion, 75 patients were collected. Because a 5-year experienced
Results
The κ value of percentage concordance between MRI and US for interpretation of FTRCTs in Group 1 (Fig. 1) was .35 (P=.065), indicating a fair agreement; that of the Group 2 (Fig. 2) was .97 (P=.012), indicating an almost perfect agreement.
Table 1, Table 2, Table 3, Table 4 list the results of US, MRI, and the corresponding surgical findings of surgically proven patients. There were three FTRCTs undetected by both modalities in Group 1. All of them received surgery at least 80 days after imaging
Discussion
After the first article about the use of US for assessment of rotator cuff in 1979 [11] and that for MRI in 1986 [12], both modalities have gradually replaced the role of arthrography in diagnosing rotator cuff tear [1], [2], [3], [4], [5], [6], [7], [8], [9], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21]. Both US and MRI carry certain advantages: nonradiative, noninvasive, and as well as offering a multiplanar imaging. However, US examination is economic, time saving, and
Acknowledgements
The authors wish to thank Ms. Chang CH for assistance with manuscript preparation. The study is supported by a grant from Taipei Veterans General Hospital.
References (23)
Magnetic resonance imaging in musculoskeletal assessment: a surgeon's perspective
Top MRI
(1996)- et al.
Rotator cuff tears: diagnostic performance of MR imaging
Radiology
(1989) - et al.
Rotator cuff tears: prospective comparison of MR imaging with arthroscopy, sonography and surgery
Am J Roentgenol
(1989) - et al.
Rotator cuff sonography: a reassessment
Radiology
(1989) - et al.
Rotator cuff lesions: signal patterns at MR imaging
Radiology
(1990) - et al.
Magnetic resonance imaging of the shoulder: sensitivity, specificity, and predictive value
J Bone Jt Surg
(1991) - et al.
Full-thickness tears of the rotator cuff of the shoulder: diagnosis with MR imaging
Am J Roentgenol
(1992) - et al.
Sonography of the shoulder in patients with tears of the rotator cuff. Accuracy and value for selecting surgical options
Am J Roentgenol
(1993) - et al.
Die Wertigkeit der Sonographie bei Rupturen der Rotatoren-manschette
Unfallchirurg
(1993) - et al.
The clinical role of shoulder joints
Ann Rheum Dis
(1998)