The interaction of microstructure and volume fraction in predicting failure in cancellous bone

Inroads have been made in the diagnosis and treatment of osteoporosis, yet dual-energy X-ray absorptiometry is still the primary diagnostic modality. This method provides 2D projections of an irregular 3D construct. However, human cancellous bone is highly heterogeneous with varying material properties. Therefore, to properly assess fracture risk, it is imperative to take into consideration microstructural indices besides subregional bone volume fraction (BV/TV). A power law model with average BV/TV as the independent variable describes 38% of the variation in yield strength; however, this predictive power is increased to 56% when BV/TV of the weakest subregion is considered. Of twenty-five specimens studied, 76% had minimum BV/TV, maximum principal Eigen value of the fabric tensor (H-1) and minimum connectivity density (Conn.D) values within the visually determined failure regions. These three independent rnorphometric indices yielded significant differences between the failure and non-failure regions of each specimen. From the results, we conclude that subregions with minimal BV/TV values are better predictors of mechanical failure in cancellous bone than average specimen BV/TV Addition of microstructural indices augments this predictive power to generate a trabecular failure prediction model based on volume fraction and cancellous bone microstructure specifically in areas where trabecular failure is most likely to occur. (c) 2006 Elsevier Inc. All rights reserved.