Magnetic Resonance Imaging of Knee Disorders

Abstract

To prospectively evaluate the clinical value of magnetic resonance imaging of the knee in a referral sports medicine practice, we performed a three-part study. First, we asked 72 consecutive patients a series of clinically relevant questions regarding the ordering of their magnetic resonance imaging scans. Second, we asked the treating physicians at our center if the mag netic resonance imaging findings changed the diagno sis or treatment. Third, we compared the clinical eval uation with the findings on magnetic resonance imaging scans for 37 patients who had arthroscopic confirmation. From the physician's perspective, in only three cases would the results of the scan have changed the diagnosis. Information from the scans was judged to contribute to patient treatment in only 14 of 72 patients. Finally, comparison of clinical evaluation and magnetic resonance imaging findings with findings during arthroscopic procedures showed that clinical evaluation had a sensitivity and specificity of 100% for diagnosis of anterior cruciate ligament injuries, whereas magnetic resonance imaging was 95% sen sitive and 88% specific. For isolated meniscal lesions, the clinical assessment had a sensitivity and specificity of 91 % compared with 82% and 87%, respectively, for magnetic resonance imaging. For evaluation of articu lar surface damage, the predictive value of a positive test was 100% for clinical assessment and 33% for the