It has been suggested that bone remodeling targeted to microscopic tissue damage can impair trabecular bone biomechanical properties, potentially modifying overall bone strength. In this study, we evaluate microscopic tissue damage and remodeling cavities using experimental and computational methods.

Cyclic loading experiments were performed on isolated rat caudal vertebrae (n=24) to evaluate the progression of microscopic tissue damage in trabecular bone in-vitro. Vertebrae were potted in bone cement and subjected to cyclic loading between 0 – 260N. Cyclic loading was terminated at secondary and tertiary phases of the creep-fatigue curve. Trabecular microfracture was the primary form of damage in trabecular bone and the number of microfractures increased with the amount of cyclic loading. Only small amounts of microscopic tissue damage were observed in the cortical shell, demonstrating that the damage occurs in trabecular bone prior to complete fracture of vertebrae.

Modifications to the rat tail loading model developed by Chambers and colleagues were considered to evaluate the feasibility of using the model to generate microscopic tissue damage in trabecular bone without fracturing the vertebrae in-vivo. Protocols were developed to apply cyclic loading to caudal 8th vertebrae (C8) in-vivo (n=20) or in-situ (n=15). Two pin types:​ smooth and threaded, two pin sizes:​ 1.6mm and 2.0mm dia. and four time points after the surgery:​ 0, 1, 2 and 4 weeks were considered. Our results indicated that the rat tail loading model may not be used for generating microscopic tissue damage in trabecular bone in-vivo.

Finite elements models of idealized trabeculae were generated to determine a potential range for stress concentrations factors of remodeling cavities. Two types of trabecula:​ rod-like and plate-like and two types of loading conditions:​ pure tension and pure bending were considered. Finite element models of two rod-like and three plate-like rat trabeculae were generated to confirm the results from idealized models. Our results suggests that stress concentration factors could be as high as 7 for pure tension and could be very high (>20) for pure bending. Results from rat trabeculae simulations were similar to that of idealized models. Our results suggest that both cavity morphology and local thinning considerably increase local tissue stresses and hence may affect failure of an individual trabecula and trabecular bone at continuum level.