Magnetic resonance imaging (MRI) has been widely used for the diagnosis of meniscal tears, but its diagnostic accuracy, depending on the type and location, has not been well investigated. We aimed to evaluate the diagnostic accuracy of MRI by comparing MRI and arthroscopic findings. Preoperative 3.0-T MRI and arthroscopic findings from 2005 to 2018 were reviewed to determine the presence, type, and location of meniscus tears. In addition, subgroup analysis was performed according to anterior cruciate ligament (ACL) injury. The exclusion criteria were as follows: (1) Inflammatory arthritis, (2) other ligament injuries, (3) inability to classify meniscal tears due to degenerative arthritis, (4) over 90 days from MRI to surgery, and (5) incomplete data. Of the 2998 eligible patients, 544 were finally included. The sensitivity and specificity of MRI in determining medial and lateral meniscus tears were 91.8% and 79.9%, and 80.8% and 85.4%, respectively. The accuracy of MRI in the ACL-injured group was lower than that in the ACL-intact group (medial meniscus: 81.7% vs. 88.1%, p = 0.041; 72.9% vs. lateral meniscus: 88.0%, p < 0.001). MRI accuracy was low for the longitudinal tears of the posterior horn of the medial meniscus in the ACL-injured group. MRI could be a diagnostic tool for meniscus tears, but has limited accuracy in their classification of the type and location. Hence, care should be taken during arthroscopic assessment of ACL-injured patients due to low diagnostic accuracy of preoperative MRI.
Comparing shoulder maneuvers to magnetic resonance imaging and arthroscopic findings in patients with supraspinatus tears
