Abstract
Purpose: To evaluate the diagnostic accuracy of arthrography, magnetic resonance (MR) imaging, and MR arthrography in the detection and localization of defects of the triangular fibrocartilage (TFC) in cadaveric wrists, using arthroscopy as a reference standard.
Material and Methods: Twenty-four specimen wrists were evaluated. The different imaging modalities were blinded to reviewers and were interpreted independently. A classification of TFC defects was used for the evaluation of images in the different imaging modalities, thus permitting a more uniform correlation. Two cases were excluded from the MR imaging study because of poor image quality. Contingency tables with the chi-square test and Fisher's exact test were used for statistical analysis.
Results: Defects of the TFC were identified in 17 of the 24 specimen wrists by means of arthroscopy, and 16 defects were observed when arthrography was carried out. With MR imaging 14 defects of the TFC were detected in the 22 specimen wrists evaluated, and with MR arthrography 16 defects were observed. Most defects were central or combined (two or more defects). In comparison to arthroscopy, the accepted diagnostic gold standard, the following results were found for arthrography in the detection of TFC defects: sensitivity 95%, specificity 100%, and accuracy 95% (P<0.0005); for MR imaging: sensitivity 86%, specificity 85%, and accuracy 70% (P<0.002); and for MR arthrography: sensitivity 100%, specificity 85%, and accuracy 95% (P<0.0005).
Conclusion: The results of the study seem to indicate that both arthrography and MR arthrography have high accuracy, and either would be useful for evaluation of the TFC. The combined approach using both techniques would have a very high accuracy equivalent to that resulting from arthroscopy.