Fundamental evaluation of dual x-ray absorptiometry (DXA) for measurement of bone mineral density in rats

Abstract

This study was designed to assess the precision and accuracy of newly developed ultra high resolution mode (rat mode) in DXA (Hologic, QDR-1000), determine how body thickness affects measured BMD values and to derive a formula by which BMDs in animals with varying degrees of body thickness can be compared. The long term reproducibility on two phantoms (BMD: 170 and 300 mg/cm²) were under CV 1.0%. The repeated precision in vivo and in vitro lumbar spines and phantoms were within CV 1.5%. Accuracy was evaluated by determining the correlation coefficient between ash weight and tibial BMC. The correlation was excellent (r=0.999, p<0.001) over the ash range of 250–600 mg. Using single regression equations, QDR-1000 BMC values were compared with those obtained by conventional methods. There was a close linear correlation with both SPA (Norland Co., r=0.997) and DCS-600 (Aloka Co., r=0.996) measurements. The effects of body thickness were assessed by immersing phantoms at various water depths. There was a significant linear decrease in BMD, as measured by QDR-1000, with increasing water depth. BMDs in vivo with varying body thickness can be compared with each other by using the following correcting formula: BMD₁=(BMD₂+5.5w+7.6) × 10³/(977-4.8w), where BMD₁=expected BMD of extracted bone (mg/cm²), BMD₂=BMD in vivo (mg/cm²), w=body thickness (cm). There was a significant positive correlation (r=0.996, p<0.001) between calculated BMDs from this equation and the values actually obtained. These results confirm that QDR-1000 rat mode yields data useful for assessing BMD and BMC in small animal bones.