Mechanical property degradation is one manifestation of fatigue damage accumulation in bone and o th er com posite m aterials. Fatigue dam age accum ulation as a result of increased loading levels has been demonstrated in living bone, and postulated as a stimulus to the rem odeling response of living bone. Remodeling is the combined processes of destruction o f old bone and successive laying down of new bone whereby older bone tissue is replaced with new bone. An understanding o f fatigue dam age accum ulation in devitalized bone may aid in the understanding of the specific mechanisms which activate the rem odeling process. This study concentrates on changes in secant modulus and cyclic energy dissipation behavior during axial load-controlled fatigue loading of cortical bone specim ens. Mathem atical expressions for modulus degradation were determ ined in tensile and com pressive fatigue as functions of life fraction and loading level. Cyclic and total lifetim e energy dissipation w ere related to the magnitude of applied tensile, compressive, and fully reversed fatigue loading. Finally, cyclic mechanical properties at low loading levels were determ ined before and after loading at higher loads, in order to determ ine the effectiveness of nondestructive testing techniques in bone.
Tensile fatigue specim ens exhibited greater property degradation during the first quarter of fatigue life than they did during the remaining three quarters of life. Secant modulus degradation in tensile fatigue was found to be a linear function of both the logarithm of the cycle num ber and the effective cyclic strain range above a threshold of 2600 to 2900 microstrain (με). Secant modulus reduction during com pressive fatigue was shown to be a linear function of both the logarithm of the remaining life fraction and of the applied effective strain range exceeding 3900 to 4000 pE. These equations suggest a dramatic increase in tensile and compressive fatigue damage accumulation behavior at approximately 2750 and 3950 με, respectively. The greater fatigue resistance of bone in compression is thus reflected in differing tensile and com pressive property degradation during cyclic loading. Despite extensive secant modulus degradation, little degradation in tangent modulus at low loading levels was observed during fatigue tests conducted in this study.
Cyclic energy dissipation was shown to be proportional to the 2.1 power of the applied effective strain range in tensile, com pressive, and fully reversed fatigue loadings below 2500 με. Above 2500 με, tensile fatigue loading caused cyclic en erg y dissipation proportional to the 5.8 power of the applied effective strain range. Com pressive fatigue loading dissipated cyclic energy proportional to the 4.9 power of the applied effective strain range over 4000 pe. Tensile and com pressive cyclic energy dissipation behavior in cortical bone demonstrated sharp transitions at effective strain ranges of 2500 and 4000 με in tensile and compressive fatigue, respectively. This finding appears to be consistent with the tension and com pression critical dam age strain thresholds observed in the secant modulus degradation analyses
Fatigue dam age accum ulation was found to have little effect on cyclic mechanical properties (e.g. secant modulus or cyclic energy dissipation) at effective strain ranges below 2500 με. A bove 2500 με, property degradation was clearly evident after prior fatigue loadings at higher levels in tensile fatigue, and was less evident in com pressive and fully reversed fatigue tests.
These findings suggest that secant modulus degradation and cyclic energy dissipation are greatly increased after critical dam age strain thresholds of 2500 and 4000 με in tensile and com pressive fatigue, respectively. Tensile and com pressive fatigue loading also caused different forms of modulus degradation over fatigue life. Prior fatigue dam age accum ulation causes property degradation at loadings above 1500 με, primarily in tensile fatigue loading. These loading levels are within the ranges observed in living anim als, and thus these phenom ena may play a role in initiating the rem odeling response in live bone tissue.