Identification of biomechanical alterations associated with nonunion after tibial shaft fractures using musculoskeletal simulation based on motion capture and instrumented insole data.

INTRODUCTION

Musculoskeletal simulations based on gait analysis data provide access to parameters such as joint reaction forces, muscle forces, and push-off moments, which cannot directly be measured in clinical practice. Applying these methods may help evaluate weight-bearing patterns and identify mechanical risk factors associated with the development of nonunion. This study aimed to investigate differences in the longitudinal development of forces in the musculoskeletal system between cases with and without union in tibial shaft fractures.

MATERIALS AND

METHODS

Longitudinal 3D marker-based motion capture and instrumented insole data of thirteen patients (8 union; 5 nonunion) with tibial shaft fractures were used to generate individualized musculoskeletal models using the AnyBody™ software.

RESULTS

In a linear mixed-effects model, the fractured limb showed impaired biomechanical parameters compared with the contralateral limb, including knee joint reaction forces (KJRFs), muscle forces, and push-off moments (all p < 0.001). During follow-up, parameters of the fractured leg increased over time, including KJRFs (p = 0.011), muscle forces (p ≤ 0.030), and ground reaction force (p = 0.008). The union group exhibited greater muscle forces, particularly in the soleus (p = 0.014) and gastrocnemius muscles (p = 0.049). The nonunion group exhibited significantly greater asymmetry between healthy and fractured leg than the union group in ground reaction force (β = 0.368, p = 0.011), KJRFs (β = 0.364, p = 0.028), and push-off moment (β = 0.457, p = 0.028). In additional time point-specific analyses no significant differences were found at 6 and 12 weeks.

CONCLUSION

Biomechanical loading patterns differed between patients with union and nonunion throughout fracture healing, with greater between-limb asymmetries in push-off moment, joint reaction forces, and plantar flexor muscle forces in the nonunion group. Larger studies are required to validate these findings using the proposed workflow.