Rotational malreduction of the syndesmosis: reliability and accuracy of computed tomography measurement methods

Background: Computed tomography (CT)-based indices may be superior to plain radiographs in determining the adequacy of reduction following operative fixation of the syndesmosis in unstable ankle fractures. This study assessed the reliability and accuracy of four CT-based methods for measurement of rotational malreduction of the fibula.

Methods: A simulated Weber C ankle fracture was created by performing an osteotomy in 9 cadaver ankles. The fibula was rotated and fixed in neutral (0 degrees) and 10 to 30 degrees of internal and external rotation. Fifty-two CT images at the level of the syndesmosis were obtained in neutral and rotated positions and presented in random order to 3 independent observers. Measurements were made using commercial imaging software and 4 methods for interpreting CT scans. Interobserver reliability and accuracy were assessed and compared.

Results: Methods 1 and 4 showed high anatomic variability. Methods 1, 2, and 4 had a test-retest repeatability of about 15 degrees. Method 1 varied erratically with direction and degree of malrotation (R (2) = 0.15) and did not permit specification of a neutral range. Method 2 varied consistently and systematically with direction and degree of malrotation (R (2) = 0.88). Receiver operating characteristic curve analysis indicated that method 2 identified malrotation better than did the other methods. Methods 3 and 4 were somewhat more difficult to perform.

Conclusions: Method 2, the angle between the tangent of the anterior tibial surface and the bisection of the vertical midline of the fibula at the level of the incisura, was fairly reliable and accurate and had greater ease of measurement compared with the other methods that were tested.

Clinical relevance: This study demonstrated that assessment of malrotation of fibular fractures by CT scan can be difficult. We believe that of the 4 methods tested in this study, method 2, the angle between the tangent of the anterior tibial surface and the bisection of the vertical midline of the fibula at the level of the incisura, was the most useful.