Prediction of Sustentaculum Tali Sustentacular Screw Length based on Linear Regression Model

Abstract

Calcaneal fractures present challenges due to the complex anatomy and difficulty in achieving precise screw placement. Traditional methods often rely on empirical screw insertion, leading to complications like joint penetration or nerve damage. We selected 66 adult calcaneus specimens and conducted analyses using micro-CT scanning and anatomical measurement techniques, dividing the calcaneus into the anterior, middle, posterior, and ST regions. The calcaneus was divided into anterior, middle, posterior, and ST regions. Linear regression was used to analyze the relationship between anatomical parameters and the lengths of screws for the posterior facet (PF) and articulatio calcaneocuboidea (AC). The derived models for AC/PF screw length based on CT data are: G1 = -1.96 + 0.71F1 + 0.48F8 + 0.39F9 (AC screw) and G2 = 3.95 + 0.28F1 + 0.59F8 + 0.31F9 (PF screw), with similar results for anatomical data. Predicted screw lengths were validated through Micro-CT imaging, confirming accurate insertion without perforating the medial calcaneal cortex. In conclusion, linear regression models based on Micro-CT and anatomical data can accurately predict AC/PF screw lengths, improving surgical precision and outcomes. Meanwhile, we’ll keep collecting more data to validate and improve the models. Additionally, we’ll explore new methods like machine learning to enhance prediction accuracy in the future.

Study design: Experimental cadaveric study with anatomical and Micro-CT-based measurements.