Factor of Risk for Fracture - Skeletal Aging and Osteoporosis: Biomechanics and Mechanobiology - page 134

more non-invasive imaging modalities (e.g., quantitative computed tomography
(QCT), or DXA) and then mechanically tested to failure in loading conditions
designed to mimic those that cause fractures. Then, using those data, whole bone
strength is estimated in vivo using regression equations between whole bone
strength and aBMD, volumetric bone mineral density (vBMD) and/or other mass
or geometric features derived from the imaging modalities. Whole bone strength
may also be estimated from finite element models based on QCT or high-reso-
lution peripheral QCT (HR-pQCT). These approaches are described in more detail
elsewhere, but are generally better predictors of bone strength than bone density
measurements alone [ 10 , 12 , 33 , 60 , 75 ].
It is important to remember that the strength of a bone depends greatly on the
loading conditions, that is the strength for one activity may not be the same as for
another. For example, a proximal femur withstands much higher force magnitudes
when tested in a single-leg stance configuration than it does when tested in a
configuration designed to simulate a sideways fall [ 35 , 38 , 39 ]. Even different
directions of sideways falls can significantly impact femoral strength [ 59 ]. In
general, bone is weaker in shear or tension conditions than in compression
[ 61 , 72 ]. Furthermore, yield strength of bone decreases with higher strain rates
[ 25 ], which may be of importance for fracture under traumatic conditions such as
fall impact loading. Thus, characterizing the activities that cause fractures and
reproducing those loading conditions for the estimates of bone strength is critical,
though challenging.
2 Examinations of Factor of Risk
Once estimates of both loading and bone strength are made, the factor of risk can
be calculated. In recent years, a number of retrospective studies and a few pro-
spective studies have examined the factor of risk for fractures of the hip, vertebral
body and wrist. These are all common fractures in osteoporotic adults, together
representing an estimated 60% of osteoporotic fractures [ 7 ]. This section critically
reviews this literature, which represents early attempts to incorporate both loading
and strength when examining the risk of fractures. We pay particular attention to
(1) the study design, as prospective studies are the gold standard for examining
predictors of fracture risk, and (2) whether current implementations of the factor of
risk perform better than BMD in predicting fracture risk.
2.1 Studies of the Factor of Risk for Hip Fracture
Hip fractures are common injuries in older adults and their incidence is increasing;
they are costly and associated with high rates of morbidity and mortality [ 7 , 41 , 57 ].
More than 90% of hip fractures are associated with a fall [ 41 ], but only about 1%