Comparison of bone density measurement techniques: DXA and Archimedes' principle

The standard method for determination of density (g/cm3) of bones from small animals has been the application of Archimedes' principle. A recent development has been software for the determination of "density" (g/cm2) of small animal bones with dual-energy X-ray absorptiometry (DXA). We compared Archimedes' principle and DXA (Hologic QDR-2000) in the measurement of the densities of whole and hollowed femurs of 5- to 6-month-old retired female breeder rats. In an attempt to ensure detectable treatment differences, rats were used from a low-vitamin D Holtzman and a supplemental-vitamin D Sprague-Dawley colony. Whole femur densities were higher for supplemental-vitamin D colony rats than for low vitamin D rats using both techniques (Archimedes' principle, p < 0.002; DXA, p < 0.005), and the densities from the two techniques were highly correlated (r = 0.82, p < 0.0001). Actual density values were higher for Archimedes' principle than for DXA. Other variables such as femur ash weight and calcium content were also highly correlated to densities with both techniques. Hollowed femur density values were higher than whole femur values with Archimedes' principle but lower with DXA. Colony effects for hollowed femur densities were diminished with Archimedes' principle (p < 0.03) and eliminated with DXA (p < 0.53). Investigation of whole bones is more biologically relevant, and both techniques were effective in detecting differences between whole femurs from low-vitamin D and supplemental-vitamin D colony rats.