Age-Related Changes in Whole-Bone Structure and Strength - Skeletal Aging and Osteoporosis: Biomechanics and Mechanobiology - page 21

of model predictions to mechanical properties in cadaveric samples, these methods
open the door to analyze samples from many hundreds of subjects from popula-
tion-based and clinical trials. A variant on this approach is to use microCT to
generate a voxel-based model which accurately renders the micro- as well as the
macro-structure. These microCT-based FEA models have been used most widely
to generate models of small (*1 cm) trabecular samples, but in recent years have
been used to generate whole-bone models, most often using HR-pQCT images of
the distal radius. Either of these FEA approaches produces an estimate of bone
strength.

We review the available data on age and estimated whole-bone

mechanical properties from CT-based FEA in this section.

3.2.1 Lower Extremity: Femoral Neck and Distal Tibia

Loading the proximal end of the femur in a configuration replicating a fall to the
side is the loading mode most relevant to hip fractures, which are fractures of the
proximal femur and nearly always occur as a result of a fall to the side [ 56 , 57 ].
There are limited experimental data on age and mechanical properties of the
proximal femur/femoral neck. Courtney et al. [ 58 ] compared femora from young
donors (17-51 yrs, mean 33 yrs; n = 9) to femora from old donors (59-83 yrs,
mean 74 yrs; n = 8) using fall-configuration loading. The old femurs had 30 %
lower stiffness, 50 % lower maximum force (''strength'') and required 70 % less
energy to fracture compared to the young femurs. This corresponds to a decrease in
strength of -12 %/decade. Limitations of this study include the grouping of female
(n = 7) and male (n = 10) donors, and the small sample size which did not allow
for linear regression analysis. In another experimental study using a falls config-
uration, Roberts et al. [ 59 ] tested 73 elderly cadaveric femora (range 55-98 yrs,
mean 74 yrs; 48 female, 25 male). They did not focus on the effects of age, but did
report a weak, non-significant correlation between failure load and age (r =-0.14;
p = 0.08; sexes pooled); separate regression analyses for female and male were not
reported. This result suggests that there is only a weak association of age and
proximal femur failure force from middle- to old-age. In a recent study, Epelboym
et al. [ 31 ] subjected 49 female cadaveric proximal femora (29-93 years of age) to a
failure load in a simulated fall-to-the side, and reported that maximum load
(r 2 = 0.19, p \ 0.004) but not stiffness (r 2 = 0.05, p \ 0.18) decreased with age.
While the load-age regression was only moderately strong, the effects are more
dramatic when comparing subsets of young versus old subjects. A comparison of
properties between women less than 50 years of age (n = 6) to women older than
80 years of age (n = 13) revealed a 35 % reduction in stiffness, a 46 % reduction in
maximum load, and a 50 % reduction in work-to-fracture. A multiple regression
analysis indicated that cortical area and trabecular BMD were the most significant
contributors to the variation in maximum load.
One limitation in accurately determining age-related changes from mechanical
test data is that they are based on relatively small samples. By contrast, studies that
use QCT-based FEA have access to many more samples. Keaveny et al. [ 60 ]