Femoral strength after cephalomedullary nail removal can be predicted preoperatively using CT based FE models

Abstract

Removals of cephalomedullary nails (CMNs) after healed pertrochanteric femur fractures are sometimes requested by patients or medically indicated due to pain or screw cut-out. However, CMN removal carries a high risk of secondary femoral neck fracture, even in the absence of trauma. Consequently, decisions on nail removal and establishing a safe post-operative loading regimen can be challenging. This study investigated if finite element (FE) models can pre-operatively predict femoral strength after CMN removal to support these clinical decisions. Nine proximal femora of body donors who were treated with a CMN during their lifetime were included. Computed tomography (CT) scans were acquired with the CMN still in place, followed by virtual implant removal using image processing. Based on this scan, non-linear voxel-based FE models were created and femoral strength was predicted for a one-legged stance configuration. For validation, the CMNs were physically removed and femoral strength was assessed in a material testing machine. The FE models predicted the femoral strength accurately relative to the experiments (R² = 0.94, CCC = 0.97). In conclusion, CT-based FE models demonstrate potential to predict femoral strength after CMN removal pre-operatively. This could help patients and clinicians to make an informed decision on implant removal and permissible post-operative weight-bearing.