The Multifactorial Relationship Between Bone Tissue Water and Stiffness at the Proximal Femur

Bone mechanical function is determined by multiple factors, some of which are still being elucidated. Here, we present a multivariate analysis of the role of bone tissue composition in the proximal femur stiffness of cadaver bones (n = 12, age 44-93). Stiffness was assessed by testing under loading conditions simulating a sideways fall onto the hip. Compositional properties of cortical and trabecular tissues were quantified in femoral neck cross sections by Fourier transform infrared (FTIR) spectroscopy and near infrared (NIR) spectroscopy. In addition, cross-sectional areas and cortical thickness and tissue mineral density (TMD) were measured at the femoral neck. Pearson correlation analysis showed a significant (p < 0.05) negative relationship between bone stiffness and cortical and trabecular water content, both total (r = -0.63) and tightly bound to matrix and mineral (r = -55). Additionally, significant (p < 0.05) positive correlations were found between stiffness and bone area, both total (r = 0.67) and trabecular (r = 0.58). However, linear regression using each of these properties to predict bone stiffness resulted in weak models (R² = 0.36-0.48). Interestingly, we found markedly stronger models (cross-validated R² = 0.80-0.92) by using partial least squares (PLS) regression to predict stiffness based on combinations of bone properties. The models with highest R² values were found when including bone water parameters as explanatory variables, both total and tightly bound, in cortical and trabecular. This study provides new insights by revealing a multifactorial relationship in which higher bone water content across different tissue compartments contributes to lower bone stiffness, highlighting bone water as a potential biomarker of bone quality and proximal femur mechanical function.