Quantitative assessment of experimental fracture repair by peripheral computed tomography

An experimental fracture model was used to assess bone mineral density at the fracture site by peripheral computed tomography and to compare the model with biomechanical, histological, and radiographic methods for the quantification of the fracture repair process. Transverse osteotomies in the mid-diaphysis of 28 tibia of sheep were externally fixed and mineral densities, cross-sectional areas, flexural rigidities, tissue composition, and projected callus area were calculated after 9 weeks of healing time. BMD measured by pQCT was strongly correlated with histologically determined percentages of mineralized tissue in the osteotomy gap (R = 0.71) and in the periosteal callus (R2 = 0.62). The percentage of mineralized tissue in the osteotomy gap was the best predictor of the flexural rigidity of the tibiae (R2 = 0.74). Because of high correlations with the histological findings, the volumetric BMD at the level of the osteotomy gap was also strongly correlated with the biomechanical findings (R2 = 0.70). Neither the cross-sectional area in pQCT nor the projected callus area in plane film radiography were positively correlated to the flexural rigidity of the tibiae. Quantitative computed tomography proved to be a successful estimator for the prediction of the mechanical stability of healing bones. The noninvasive procedure is a reliable tool for the quantification of the fracture repair process in experimental studies and may be useful for treatment decisions in particular clinical situations.