Accuracy of Magnetic Resonance Imaging of the Knee in the Community Setting

• Published in: Sports health Vol. 4; no. 3; pp. 222 - 231
• Main Authors: Hardy, Jolene C., Evangelista, Gregory T., Grana, William A., Hunter, Robert E.
• Format: Journal Article
• Language: English
• Published: Los Angeles, CA SAGE Publications 01.05.2012
• Subjects: Accuracy; Associations; Evaluation; Facilities; Knees; Primary Care; Scanning; Specificity; Statistics; Surgery; knee; magnetic resonance imaging; clinical evaluation; meniscus
• ISSN: 1941-7381; 1941-0921; 1941-0921
• DOI: 10.1177/1941738111434396

Background: Magnetic resonance imaging (MRI) is routinely used in the diagnosis of sports-related knee injuries. Purpose: To determine the accuracy, sensitivity, and specificity of MRI compared with clinical evaluation in the diagnosis of meniscal pathology when the MRI facility and the radiologist are not preselected. Methods: A total of 288 knee arthroscopies were retrospectively compared. Patients were divided into 3 groups: those who had MRI performed and interpreted at a single institution, MRI performed and interpreted at community facilities, or a clinical evaluation by a senior orthopaedic surgeon. Results: The sensitivity, specificity, and accuracy of the diagnosis of medial meniscal pathology at a single institution were 90%, 59%, 76%; in community facilities, 73%, 68%, 70%; and by a clinical evaluation, 93%, 55%, 73%, respectively. For lateral meniscal pathology, the results were as follows: single institution, 75%, 76%, 81%; community facilities, 60%, 88%, 79%; and clinical evaluation, 45%, 90%, 79%, respectively. Sensitivity for medial meniscus was greater than for lateral meniscus, but specificity of diagnosis was better for lateral meniscus by MRI and clinical evaluation. While not statistically significant, there was increased sensitivity in the diagnosis of medial meniscus and lateral meniscus at SIs, but they have less specificity than at community facilities. The number of false-positive diagnoses (ie, no intra-articular pathology) that resulted in surgery was 4 of 288 (1.39%). The overall accuracy for medial meniscus by MRI was 73% vs 73% for clinical evaluation. The overall accuracy for MRI for lateral meniscus was 78% vs 79% for clinical evaluation. Conclusion: Routine MRI may not be more beneficial than clinical evaluation when there is no preselection of MRI facility and interpreting radiologist. Clinical Relevance: The use of MRI for diagnosing meniscal pathology should be reserved for those cases where the orthopaedic clinical examination is ambiguous.