H. Kroger et al., COMPARISON OF DIFFERENT MODELS FOR INTERPRETING BONE-MINERAL DENSITY-MEASUREMENTS USING DXA AND MRI TECHNOLOGY, Bone, 17(2), 1995, pp. 157-159
Bone mineral density measurements using dual X-ray absorptiometry (DXA
) are commonly expressed as areal density (g/cm(2)). However, areal BM
D (BMD(areal)) is dependent on bone size and this can lead to erroneou
s interpretations of BMD values. We have previously presented a simple
method for calculating apparent volumetric bone mineral density (BMD(
vol)) using ancillary DXA-derived data. In the present study we tested
the validity of our model using in vivo volumetric data obtained from
magnetic resonance imaging (MRI) of lumbar vertebrae. BMD(areal) and
BMD(vol) of L(3) were measured from sixteen pairs of identical twins (
24 men, 8 women), aged 25-69 years. The dimensions of the lumbar verte
bra L(3) were measured from MR images and BMD values were corrected fo
r these dimensions. The DXA-derived apparent volumetric bone mineral d
ensity (BMD(vol)) correlated moderately with MRI-derived BMDs (r value
s from 0.665 to 0.822). In contrast to BMD(areal), BMD(vol) and MRI-de
rived BMDs were not related to body size variables. All these volume-c
orrected BMDs diminished the erroneous effect of vertebral size on are
al BMD. We conclude that the simple DXA-derived BMD(vol) can be used f
or normalization of bone mineral density values in subjects of differe
nt body sizes, and especially in growing children.