Significance of QCT bone mineral density and its standard deviation as parameters to evaluate osteoporosis

A study using quantitative CT (QCT) of the spine was carried out to determine whether the standard deviation (SD) of the bone mineral density (BMD) within a given region of interest (ROI) could be used as a parameter to evaluate osteoporosis.

A low dose single energy protocol was used. The elliptical ROIs inside the trabecular bone of the vertebral bodies T12-L3 were analyzed for four patient groups: Group 1, 52 healthy premenopausal women (age 41 +/- 2 years); group 2, 119 healthy early postmenopausal women (53 +/- 4 years); group 3, 45 postmenopausal relatively healthy women (age 65 +/- 5 years); group 4, 26 osteoporotic women (age 67 +/- 5 years). Average group mean BMD values and their coefficients of variation (CV = SD/BMD) were calculated. The t values, percent decrements, z-scores, and analysis of variance (ANOVA) served to compare capabilities of the BMD and the CV to discriminate groups pairwise using all possible group combinations.

The use of z-scores and percent decrements gave ambivalent and mostly insignificant results. The CV performed better than BMD in separating group pairs (1,3), (1,4), (2,3), and (2,4), but BMD was superior for group pairs (1,2) and (3,4). Using SD as an independent variable in addition to age and BMD in the ANOVA did not significantly change r2 or the standard error of the estimate. The t test showed highly significant better discriminatory capabilities for BMD compared to CV>

The results of our study did not indicate a significant potential of the BMD SD as measured in trabecular single energy low dose spinal QCT to improve the discriminatory capabilities of BMD for a separation of osteoporotic from nonosteoporotic subjects.