Role of Magnetic Resonance Imaging (MRI) in the Diagnosis of Rotator Cuff Injuries and Correlation With Arthroscopy Findings

Abstract

Background and aim: The prevalent cause of shoulder pain is rotator cuff tears (RCT), which induce profound discomfort and morbidity. Hence, their detection and appropriate management become important to alleviate morbidity and enhance quality of life. Imaging has an important role in the diagnosis of such patients to guide for further management. A wide array of radiological approaches has been explored for the detection of RCT. The study aimed to assess the sensitivity (SN) and specificity (SP) of regular magnetic resonance imaging (MRI) in the diagnosis and characterization of RCT and furthermore to link the results of MRI with the arthroscopy findings. The limitations and pitfalls if any with MRI patients presenting were also addressed.

Materials and methods: The study was a diagnostic evaluation study conducted at the Department of Radiodiagnosis, Muslim Educational Society (MES) Medical College, Perinthalmanna, India, among individuals with RCT between December 1, 2015, and May 31, 2017. Patients who were referred for MRI and then taken up for arthroscopy were included. The findings for each tendon of rotator cuff on MRI were contrasted with that of arthroscopy, and the percentage of agreement was calculated. Additional findings relevant to the study were also noted. SN, SP, positive predictive value (PPV), and negative predictive value (NPV) of MRI were determined by taking arthroscopic findings as the gold standard.

Results: The study was comprised of 36 patients assessed using MRI and arthroscopy for RCT. The study participants ranged in age from 31 to 70 years, with a mean of 52.69±8.86 years. The majority of the patients (69.4%) were between the ages of 41 and 60 years. MRI had 100% SN and SP for full-thickness supraspinatus (SS) tear, 50% SN and 100% SP for partial-thickness SS tear, 100% SN and 80% SP for full-thickness infraspinatus (IS) tear, 75% SN and 80% SP for partial-thickness IS tear, and 95% SN and 86.6% SP for subscapularis (SC) tear.

Conclusion: The MRI RCT investigation had a high SN, SP, and PPV compared to arthroscopy. The most frequently involved tendon reported in the present study was SS (n=35; 97.22%) followed by IS (n=32; 88.88%) and SC (n=22; 61.11%). The teres minor tendon was least commonly affected (n=0). Moreover, 61.11% (n=22) of the patients had joint effusion, 41.66% (n=15) had subacromial-subdeltoid bursal effusion, and 27.77% (n=10) had subcoracoid effusion, suggesting that RCT include joint effusion or bursal fluid. Acromioclavicular (AC) joint hypertrophy was found in 53% (n=19) of the patients, and 90% (n=17) were over 45 years old, indicating an association between age, AC joint hypertrophy, and RCT. Therefore, MRI has a good SN and SP for detecting various RCT. Therefore, it could be used to investigate a suspected RCT and should be considered a near-reference standard to arthroscopy for RCT diagnosis.