Improving the estimate of the effective elastic modulus derived from three-point bending tests of long bones

Three-point bending tests are often used to determine the apparent or effective elastic modulus of long bones. The use of beam theory equations to interpret such tests can result in a substantial underestimation of the true effective modulus. In this study three-dimensional, nonlinear finite element analysis is used to quantify the errors inherent in beam theory and to create plots that can be used to correct the elastic modulus calculated from beam theory. Correction plots are generated for long bones representative of a variety of species commonly used in research studies. For a long bone with dimensions comparable to the mouse femur, the majority of the error in the effective elastic modulus results from deformations to the bone cross section that are not accounted for in the equations from beam theory. In some cases, the effective modulus calculated from beam theory can be less than one-third of the true effective modulus. Errors are larger: (1) for bones having short spans relative to bone length; (2) for bones with thin vs. thick cortices relative to periosteal diameter; and (3) when using a small radius or "knife-edge" geometry for the center loading ram and the outer supports in the three-point testing system. The use of these correction plots will enable researchers to compare results for long bones from different animal strains and to compare results obtained using testing systems that differ with regard to length between the outer supports and the radius used for the loading ram and outer supports.