How Should We Evaluate Syndesmosis Reduction on CT Scan?

CATEGORY: Ankle, Sports, Trauma, Syndesmosis. INTRODUCTION/PURPOSE: Accurate reduction of the syndesmosis is essential to the restoration of normal ankle mechanics, but radiographic evaluation of reduction quality has been demonstrably unreliable. CT imaging has traditionally been considered the gold standard, but there is no uniformly accepted method by which to assess syndesmotic reduction on CT. The aim of this study was to evaluate intra- and interobserver reliability of published measurements for evaluation of syndesmosis reduction using weight-bearing CT scan. METHODS: Medical records were reviewed to identify patients who underwent surgical stabilization of a unilateral syndesmotic injury. Exclusion criteria included patients less than 18 years old, ipsilateral pilon fracture, history of contralateral ankle or pilon fracture or syndesmosis injury, BMI >40, and any neurological impairment. Twenty eligible patients were recruited to participate in the study and underwent weight-bearing CT evaluation for both ankles simultaneously. The quality of syndesmotic reduction was evaluated using axial CT images at a level 1 cm proximal to the tibial plafond using nine published radiological measurements (Figure 1). Inter- and intraobserver reliability were assessed for each evaluation method. Side by side method (qualitatively observing the repaired laterality side by side with the contralateral control) was assessed using Kappa interrater agreement, while the reliabilities for all other continuous methods were assessed with interclass correlation coefficients (ICC). RESULTS: The study cohort consisted of ten males and ten females with mean age of 43 years (range, 23-67 years). The intraobserver reliability of the side by side method was 0.49 and 0.24 for the first and second observers respectively, and the interobserver reliability was 0.24. The syndesmotic area showed the highest interobserver reliability (0.96), the highest intraobserver reliability for observer 2 (0.97), and the second highest intraobserver reliability for observer 1 (0.92). Fibular rotation had the second highest interobserver reliability in our results (0.84), with high intraobserver reliability of 0.91 and 0.8 for first and second observers, respectively. The other six radiological measurements had interobserver reliability that ranged between 0.62 and 0.83. CONCLUSION: Qualitatively assessing syndesmotic reduction via side by side comparison with the uninjured ankle had the least intra- and interobserver reliability and should not be relied upon to determine syndesmotic reduction quality. In contradistinction, syndesmotic area calculation showed the highest reliability for evaluating syndesmosis reduction, followed by fibular rotation. Additional efforts should be considered to standardize such measurements within CT image viewers, so they become seamlessly accessible in the clinical setting.