Introduction: In the study, we aimed to evaluate the diagnostic performance of magnetic resonance imaging in comparison with arthroscopy in rotator cuff lesions and pathologies of the long head of the biceps tendon, which are one of the most important causes of shoulder pain and dysfunction.
Material and methods: 180 patients treated with arthroscopic surgery were identified. Sixty-four patients (17 males, 47 females) who had undergone diagnostic magnetic resonance imaging and met the study criteria were enrolled in the study. The magnetic resonance imaging within the last 6 weeks preoperatively and arthroscopic shoulder joint examination findings during the operation were reviewed retrospectively.
Results: Sensitivity was found to be 89.1% and specificity was found to be 94.4% in the full thickness tears, and sensitivity was found to be 93.8% and specificity was found to be 87.5% in the partial tears of the supraspinatus tendon; sensitivity was found to be 33.3% and specificity was found to be 98.4% in the full-thickness tears, and sensitivity was found to be 58.3% and specificity was found to be 80.8% in the partial tears of the long head of the biceps tendon. 24 (37.5%) patients in arthroscopy and 32 (50%) patients in magnetic resonance imaging had more than one tear in the tendon. Furthermore, the combination of the tears of the long head of the biceps tendon was found in the rotator cuff tears arthroscopy of 15 patients (23.4%) and magnetic resonance imaging of 19 (29.6%) patients.
Conclusion: In comparison with arthroscopy, magnetic resonance imaging in rotator cuff tendon pathologies, especially in the supraspinatus tendon tears, demonstrated high accuracy, the kappa value (kappa value: 0.78), high sensitivity and specificity in the strong consistency range. However, sensitivity was low in pathologies of the long head of the biceps tendon. Magnetic resonance imaging remains to be a reliable diagnostic method in shoulder tendon pathologies, especially in rotator cuff tendon pathologies.