Because periprosthetic bone loss impacts revision total hip arthroplasties and subsequent patient recovery, it is important to understand the consequences of stress shielding. We characterized bone loss attributed to stress shielding by investigating the influence of patient activity and demographics on reductions in cortical bone cross-sectional area, bone mineral density, and bone mineral content. Cortical bone shape, bone mineral content, and bone mineral density were measured in implanted and contralateral nonimplanted cadaveric femurs. Geometric measurements of transverse cross sections were measured with imaging software. Dual energy x-ray absorptiometry measured bone mineral content and bone mineral density. Patient activity was estimated by the mechanical usage score. The mechanical usage score was comprised of the ambulatory components of the Harris hip score. Regression analysis revealed a lower mechanical usage score and patient weight correlated with greater bone loss. Age, implant size, and time in situ did not correlate to bone loss. The implanted femurs were reduced in cross-sectional area and rigidity, and expanded endosteally and periosteally. Principal axis location did not change. Bone loss was more likely to occur uniformly than to change cortical bone distribution. Bone loss was minimized in patients with higher activity levels.

No abstract

When testing the effects of a femoral component on cortical bone following total hip arthroplasty, the patient's implanted femur is often compared with his/her contralateral nonimplanted femur, with differences attributed to the femoral component. However, if normal anatomical differences exist between bilateral femurs, they need to be quantified in order to validate whether the differences between implanted and nonimplanted bilateral femurs are due to the implant or possibility due to intrinsic differences before implantation. This study quantified the geometric properties of cortical bone shape between seven pairs of bilateral, cadaveric, human femurs. The null hypothesis tested stated that the bilateral femurs would not be significantly different in cortical bone geometry. Digitized images of cortical bone cross-sections taken at percent biomechanical lengths (levels 1-8) were used to calculate bone geometry measurements. The paired t-test showed that the only significant difference was in the location of principal axes at the most proximal location, level 1 (p = 0.015). All other measurements and levels were not significant with percent differences less than 6.6%. In conclusion, the data supports attributing cortical bone shape differences between implanted and contralateral nonimplanted femurs in levels 2-8 to the presence of the implant when the significant differences are greater than 6.6%.