While weightbearing computed tomography (WBCT) allows three-dimensional (3D) visualization of the distal syndesmosis, image interpretation has largely relied on one-dimensional (1D) distance and, more recently, two-dimensional (2D) area measurements. This study aimed to (1) determine the sensitivity and specificity of 2D area and 3D volume WBCT measurements towards detecting subtle syndesmotic instability, (2) evaluate whether the patterns of changes in the 3D shape of the syndesmosis can be attributed to the type of ligament injury. A total of 24 patients with unilateral subtle syndesmotic instability and 24 individuals with uninjured ankles (controls) with bilateral ankle WBCT were assessed retrospectively. First, 2D areas at 0, 1, 3, 5, and 10 cm, and 3D volumes at 1, 3, 5, and 10 cm above the tibial plafond were measured bilaterally. Secondly, the 3D model of the distal tibiofibular space was created based on WBCT in a subset of 8 out of 24 patients in whom the type of ligament injury was recognized via magnetic resonance imaging. The 3D model of the injured side was superimposed on the uninjured contralateral side to visualize the pattern of changes in different planes. Volume measurement up to 5 cm above the tibial plafond showed the lowest p-value (<0.001 vs. other methods), higher sensitivity (95.8%, 95% confidence interval [CI]: 87.8-100), and specificity (83.3%, 95% CI: 68.4-98.2) for detection of syndesmotic instability. No specific pattern of changes in the 3D shape could be attributed to a type of ligament rupture. We suggest 3D volume measurements, best measured up to 5 cm proximal to the plafond, as a promising means of diagnosing syndesmotic instability, particularly for subtle cases that are hard to detect. Clinical significance: The ability to compare the ankle joints bilaterally in a 3D manner under physiologic weight provided by weightbearing CT has led to a more accurate diagnostic method. Using volumetric measurement up to 5 cm above the tibial plafond showed higher sensitivity and specificity for recognizing an unstable syndesmosis, especially in subtle cases. However, our preliminary investigations showed that the pattern of 3D alterations in the distal tibiofibular joint space based on WBCT images does not indicate the type of syndesmotic ligamentous injury. Our results can also help image viewing programs to improve their measurement tools to facilitate 3D measurement for the syndesmosis as well as other conditions that may benefit from 3D evaluation of the clinical images.
Keywords: accuracy; image interpretation; sensitivity; specificity; stress imaging; syndesmosis; weightbearing CT.
© 2021 Orthopaedic Research Society. Published by Wiley Periodicals LLC.