Purpose
Imaging for rotator cuff injury may provide false-positive and false-negative results. Moreover, many studies have questioned both the accuracy and reliability of orthopedic special tests of the shoulder. Herein, this study evaluates the sensitivity, specificity, and accuracy of specific clinical tests and MRI of rotator cuff tears in relation to arthroscopic findings.
Patients and methods
This study was conducted on 45 patients (27 males and 18 females), who were diagnosed clinically to have rotator cuff tears and/or biceps tendon pathology. Mean patients’ age was 26.18 years old. The special clinical tests for rotator cuff tears included in this study were Jobe’s test for supraspinatus tear, drop arm test for infraspinatus tear, lift-off test for subscapularis tear, and speed test for biceps tendinitis. Thereafter, all patients underwent radiographs, MRI, and shoulder arthroscopy.
Results
Jobe’s test showed sensitivity, specificity, and accuracy of 85.7, 75, and 80.4%, respectively, whereas those of MRI were 85.7, 87.5, and 86.6%, respectively. Drop arm test showed sensitivity, specificity, and accuracy of 100, 69.2, and 84.6%, respectively, whereas those of MRI were 100% for each. Lift-off test showed sensitivity, specificity, and accuracy of 50, 76.9, and 63.5%, respectively, whereas those of MRI were 100, 92.3, and 96.2%, respectively. Speed test showed 80% for each of sensitivity, specificity, and accuracy, whereas those of MRI were 80, 100, and 90%, respectively.
Conclusion
We could conclude that Jobe’s test and MRI had good correlation with arthroscopy in the diagnosis of supraspinatus tear. Drop arm test had weak correlation whereas MRI had good correlation with arthroscopy in the diagnosis of infraspinatus tear. Lift-off test slightly overestimated the diagnosis of subscapularis tear, whereas MRI findings were close to those of arthroscopy. Speed test and MRI correlated fairly well with arthroscopy in the diagnosis of biceps tendon pathology.