Relationship between tibio-fibular overlap ratio and lower limb torsion in an Asian population

Lower limb torsion affects the accuracy of knee coronal alignment measurements, which has an impact on the planning for surgical management of knee osteoarthritis (KOA). Current computed tomography (CT) scans for evaluating torsion are costly and cannot be performed for every case. This study aimed to investigate the relationship between distal and proximal tibio-fibular overlap ratio measured on long-limb radiographs with femoral and tibial torsion quantified by CT scan in an Asian population with KOA.

A retrospective analysis was performed on patients who were planned for robotic total knee arthroplasty (TKA) or unicompartmental knee arthroplasty (UKA) between September 2018 and April 2024 in a single tertiary institution. Proximal and distal tibio-fibular overlap was defined as the longest overlapping length between the medial cortex of the fibula and the lateral cortex of the tibia, measured on weight-bearing long-leg radiographs. Femoral and tibial torsion were assessed on CT images using the Lee and bimalleolar method, respectively. Spearman's rank correlation coefficient was used to examine associations between the tibio-fibular overlap ratio and tibial and femoral torsion. Receiver operating characteristic (ROC) curve analysis was used to assess the ability of the overlap ratio to discriminate between increased and decreased torsion, with area under the curve (AUC), sensitivity, and specificity reported. Inter- and intra-observer reliability were assessed using intraclass correlation coefficients (ICCs) in a random sample of 30 patients.

A total of 217 patients were included (74 males and 143 females; 66.3 ± 8.8 years old). The median tibio-fibular overlap ratio and femoral torsion was 0.9 (0.7-0.8) and 26.9° (-10.7° to 7°), respectively. The mean tibial torsion was 20.7° + 8.55°. Overall, a moderate positive Spearman correlation of tibio-fibular overlap ratio with tibial torsion (ρ = 0.54, P < 0.001) was observed. ROC analysis for tibio-fibular overlap ratio in predicting both decreased (AUC 0.91 ± 0.05; 95%, CI 0.81-0.99) and increased tibial torsion (AUC 0.77 ± 0.07; 95%, CI 0.64-0.90) was excellent and acceptable, respectively. For predicting low tibial torsion, the optimal cut-off was <0.5 (sensitivity 99%, specificity 82%). For predicting high tibial torsion, the optimal cutoff was >0.8 (sensitivity 76%, specificity 71%). Reliability analysis revealed excellent intra-observer and interobserver reliability.

This study suggests the utility of the tibio-fibular overlap ratio as a potential screening tool for underlying tibia torsional deformity. Further evaluation of tibial torsion is recommended for knees with tibio-fibular overlap ratio of <0.5 or >0.8.