When bone is loaded beyond its failure point, it develops damage in the form of microcracks. Normally, microcracks are repaired by the remodeling process, limiting the number of in vivo microcracks. However, if the rate of microdamage accumulation increases or the rate of remodeling slows, microdamage can accumulate, reducing bone stiffness and strength and may lead to stress fractures or fragility fractures.

A new technique for visualizing microdamage in vitro has been developed that uses chelating fluorochromes to label microcracks. Sequential staining is used to distinguish between microdamage that occurred before testing and damage created during testing. Microdamage parameters quantified include the total number of microcracks, total length of microcracks, damaged area, the number of trabeculae containing microcracks, the pattern of microcracking, the extent of microdamage across the thickness of the trabeculae, and the size of the damage-containing region in the specimen.

The chelating fluorochrome marker technique was used to label and quantify microdamage in specimens of bovine trabecular bone damaged in uniaxial compression and compressive fatigue, and relationships between microdamage parameters and changes in mechanical properties (maximum compressive strain, modulus reduction) were quantified. The progressions of damage accumulation during a single compression cycle and during fatigue to failure were determined. Comparisons were made between specimens tested in different loading modes, including uniaxial compression, compressive fatigue, and compressive creep (Pierce, 1999). Microdamage accumulation increases with increasing specimen strain and with increasing stiffness loss.

A model was developed to predict modulus reductions based on observed microdamage using cellular solid principles. The predictions were compared to experimentally measured changes in mechanical properties during uniaxial compressive loading and compressive fatigue. There was good agreement between model predictions and experimental results for specimens tested in uniaxial compression. The model predictions were less accurate when applied to specimens tested in compressive fatigue.