K. Brookewavell et al., ULTRASOUND AND DUAL X-RAY ABSORPTIOMETRY MEASUREMENT OF THE CALCANEUS- INFLUENCE OF REGION OF INTEREST LOCATION, Calcified tissue international, 57(1), 1995, pp. 20-24
Ultrasound measurements of the calcaneus are related to incidence of o
steoporotic fracture. Such measurements are generally made at fixed co
ordinates relative to a footplate. This study compares measurements at
an anatomically located region of interest (ROI(anat)) and at fixed c
oordinates (ROI(fixed)), with bone mineral density measurements, in 84
postmenopausal women. Bone mineral density (BMD) was assessed using d
ual energy X-ray absorptiometry at both ROIs as well as at lumbar spin
e and femoral neck. Broadband ultrasound attenuation and velocity of s
ound were measured using a CUBA system at ROI(anat) and ROI(fixed). Ad
ditionally, broadband ultrasound attenuation at ROI(fixed) was measure
d using a Walker Sonix instrument. Mean bone mineral density, broadban
d ultrasound attenuation and velocity of sound did not differ signific
antly between ROI(fixed) and ROI(anat), although broadband ultrasound
attenuation by Walker Sonix (81.4 +/- 14.6 dBMHz(-1)) was significantl
y (P < 0.001) greater than that by CUBA (63.7 +/- 14.2 dBMHz(-1)). The
relationship between broadband ultrasound attenuation and BMD differe
d significantly between the 2 ROIs and the correlation of this relatio
nship was significantly greater at ROI(fixed) than at ROI(anat) (r = 0
.74 versus 0.46, P < 0.01). The differing relationship may reflect str
uctural variation at different regions. ROI selection may thus be a po
ssible confounding factor in ultrasound measurement.