Radial tear of the medial meniscal root: Reliability and accuracy of MRI for diagnosis

OBJECTIVE. The purpose of this study was to determine the reliability and accuracy of MRI in the diagnosis of radial tears of the medial meniscal root. MATERIALS AND METHODS. The MR images of 192 patients who underwent arthroscopy and MRI of the knee from July 2003 through March 2006 were retrospectively reviewed. MR images were independently scored by two observers for the presence of radial tear of the medial meniscal root. Interobserver agreement in detection of these tears was assessed with kappa values. The differences in areas under the receiver operating characteristic curves were assessed with a univariate z-score test. RESULTS. Arthroscopy revealed that 29 patients had radial tears of the medial meniscal root. The sensitivity, specificity, and accuracy of MRI for one reader were 90% (26/29), 94% (154/163), and 94% (180/192) and for the other reader were 86% (25/29), 95% (155/163), and 94% (180/192). Interobserver agreement for radial tears of the medial meniscal root was very high (kappa = 0.93). The areas under the receiver operating characteristic curves for each reader were 0.97 and 0.96, which were not significantly different. There was no significant difference in detection of medial meniscal root tears on T2-weighted coronal images compared with the overall interpretation for both readers. Other image sequences had significantly different sensitivity or specificity for one or both readers (McNemar statistic). CONCLUSION. MRI of the knee is reliable and accurate for detection of radial tears of the medial meniscal root. Coronal T2-weighted imaging is the most useful MRI sequence.