Microstructure and homogeneity of distribution of mineralised struts determine callus strength

Non-invasive assessment of fracture healing, both in clinical and animal studies, has gained favour as surrogate measure to estimate regain of mechanical function. Micro-computed tomography (µCT) parameters such as fracture callus volume and mineralisation have been used to estimate callus mechanical competence. However, no in-depth information has been reported on microstructural parameters in estimating callus mechanical competence. The goal of this study is to use differently conditioned mice exhibiting good and impaired fracture healing outcomes and investigate the relationship between µCT imaging parameters (volume, mineralisation, and microstructure) that best estimate the callus strength and stiffness as it develops over time. A total of 99 mice with femoral fracture and intramedullary stabilisation were divided into four groups according to conditioning: wild type, NF1 knock-out, RAG1 knock-out and macrophage depleted. Animals were sacrificed at 14, 21, 28 or 35 days and µCT parameters and torsional stiffness and strength were assessed post-sacrifice. Using linear regression for all groups and time points together, torsional stiffness could be estimated with strut thickness, strut number and strut homogeneity (R² = 0.546, p < 0.0001); torsional strength could be estimated using bone mineral density, strut thickness and strut homogeneity (R² = 0.568, p < 0.0001). Differently conditioned mice that result in different fracture healing outcomes have been shown to result in varying structural, material and volumetric µCT parameters which can be used to estimate regain of bone strength. This study is the first to demonstrate that microstructure and strut homogeneity influence callus stiffness and strength.