Micro-CT-derived nonlinear micro-finite element estimation of compressive fracture load in murine vertebrae: a pilot...

Fragility fractures caused by osteoporosis remain a growing health burden, highlighting the need for reliable, non-destructive methods to assess bone strength. We aimed to develop and assess the feasibility of a micro-computed tomography (µCT)-based nonlinear micro-finite element analysis (µFEA) workflow for estimating compressive fracture load in murine vertebrae. Fourth lumbar vertebral bodies from untreated and sclerostin antibody-treated (Scl-Ab) female C57BL/6 mice were imaged ex vivo using µCT and subjected to destructive axial compression testing. µFEA models incorporated heterogeneous isotropic material properties derived from voxel grey values and density-dependent local yield and post-yield softening definitions. µFEA-estimated vertebral fracture load strongly correlated with experimental fracture load (R² = 0.929), while µFEA-estimated axial stiffness showed a moderate-to-high association with experimental values (R2 = 0.70). Across the entire dataset, bone volume fraction (BV/TV) was also strongly associated with experimental fracture load (R2 = 0.928). After treatment stratification, associations were substantially weaker, with µFEA showing a numerically stronger association than BV/TV in the control group (R2 = 0.323 vs. 0.049 but not in the Scl-Ab-treated group (R2 = 0.231 vs. 0.035). This pilot study demonstrates the technical feasibility of nonlinear heterogeneous µFEA and provides preliminary evidence that approach reflects broad treatment-related differences in murine vertebral strength. However, larger cohorts and murine-specific material calibration are required before predictive validity can be established.