Diagnostic utility of MRI and MR arthrography for detection of ligamentum teres tears: a retrospective analysis of 187 patients with hip pain.

Abstract

The purpose of this study was to compare the diagnostic utility of MRI and MR arthrography for ligamentum teres tears in patients with hip pain. This is a retrospective study involving 187 patients who underwent MRI or MR arthrography for hip pain. This study included 103 male and 84 female patients with average age of 39 years. Three experienced musculoskeletal radiologists reviewed the MRI and MR arthrography studies to assess the ligamentum teres tears. The criteria for diagnosing normal or abnormal (i.e., ligament), partial or degenerative or complete tear of ligamentum teres were defined on the basis of several imaging characteristics. The MRI and MR arthrography results were correlated with arthroscopy, which served as the reference standard. Statistical analysis was performed to calculate the diagnostic yield, diagnostic accuracy, and diagnostic performance of MRI and MR arthrography in detecting partial or degenerative and complete ligamentum teres tears. Overall comparative performance of MRI and MR arthrography was assessed using Kruskal-Wallis test. For partial ligamentum teres tears, MRI showed lower sensitivity, specificity, and positive predictive value (0.41, 0.75, and 0.32, respectively) as compared to MR arthrography (0.83, 0.93, and 0.76, respectively), whereas the negative predictive value of MRI (0.82) was comparable to that of MR arthrography (0.95). No statistically significant difference (p < 0.05) could be identified between MRI and MR arthrography for diagnosing complete ligamentum teres tears. Hip MRI is equally suited for diagnosis of complete ligamentum teres tears when compared with MR arthrography. By contrast, for partial or degenerative ligamentum teres tears, MR arthrography offers the advantage of better arthroscopic correlation.