Age-Related Changes in Whole-Bone Structure and Strength - Skeletal Aging and Osteoporosis: Biomechanics and Mechanobiology - page 7

It is important to note that most studies have been cross-sectional in design, i.e.,
have inferred changes with aging based on differences between groups of donors/
subjects of different ages. This approach has limitations due to secular trends such
as increased stature over the past century. Thus, an older age group is likely to be
shorter than a younger age group, which will introduce bias because of the known
dependence of bone cross-sectional size on bone length [ 5 ]. One way to account
for this is to scale the cross-sectional properties by length, which is done in some
but not all studies.
The ideal study design for assessing bone changes with aging is to follow the
same individuals over time (longitudinal). In a longitudinal study, Lauretani et al.
[ 3 ] performed pQCT scans of the distal tibia in 809 subjects (females and males) at
three timepoints (0, 3, 6 years follow-up). They noted that the age-related changes
in bone parameters were underestimated by cross-sectional analysis compared to
the ''true'' rates of change measured longitudinally. (This is consistent with the
trend that Garn et al. reported for metacarpals [ 11 ]; see Table 2 .) In addition, the
rates of change vary with age and there are age-sex interactions. This is clearly
illustrated in data from their study (Fig. 2 ). For example, both women and men
show an overall increase in subperiosteal area (''total bone area'') with aging, but
they show opposite trends for rates of change. Women have little change early in
life but steep increases later in life ([70 years), while men have relatively steep
increases early (20-40 years) which slow to zero and then actually reverse with
old age ([80 years). While these particular findings may be unique to the tibial site
and the study population (Tuscany, Italy), they highlight the insight that might be
gained from longitudinal studies at sites of greater clinical relevance such as the
proximal femur, distal radius and vertebral body.
2.1.2 Upper Extremity: Metacarpals, Radius and Ulna
Women have smaller bones in the upper extremity than men. Total area of the
metacarpals and radii are approx. 25 % less in women than men [ 12 , 13 ], while
cortical area is approx. 35 % less in women than men [ 9 ].
Diaphyseal expansion with aging also occurs in long bones of the upper
extremity (Table 2 ). Analysis of hand radiographs revealed increased periosteal
and medullary diameters at the midshaft of the metacarpals [ 11 , 12 , 14 ]. Similar to
bones of the lower extremity, increases in periosteal diameter and sub-periosteal
area are modest (1-2 %/decade [ 11 , 12 ]), while medullary diameter increases are
moderate (*15 %/decade in women, 5 %/decade in men [ 14 ]). The radius has
also been examined in several studies, using physical measurement [ 15 ], single
photon absorptiometry [ 16 , 17 ] and, more recently, pQCT [ 9 , 13 , 18 ]. Most of the
cross-sectional studies reveal modest periosteal expansion with relatively greater
medullary expansion leading to a net loss of cortical bone area (*-3 %/decade)
in both women and men [ 9 , 13 , 17 , 18 ]. In partial contrast, Burr et al. reported that
periosteal expansion and increased moment of inertia occurred in men but not in
women [ 15 ].