Precision and accuracy of dual-energy X-ray absorptiometry for determining in vivo body composition of mice

Abstract

Objective: To evaluate the precision and accuracy of dual-energy X-ray absorptiometry (DXA) for the measurement of total-bone mineral density (TBMD), total-body bone mineral (TBBM), fat mass (FM), and bone-free lean tissue mass (LTM) in mice.

Research methods and procedures: Twenty-five male C57BL/6J mice (6 to 11 weeks old; 19 to 29 g) were anesthetized and scanned three times (with repositioning between scans) using a peripheral densitometer (Lunar PIXImus). Gravimetric and chemical extraction techniques (Soxhlet) were used as the criterion method for the determination of body composition; ash content was determined by burning at 600 degrees C for 8 hours.

Results: The mean intraindividual coefficients of variation (CV) for the repeated DXA analyses were: TBMD, 0.84%; TBBM, 1.60%; FM, 2.20%; and LTM, 0.86%. Accuracy was determined by comparing the DXA-derived data from the first scan with the chemical carcass analysis data. DXA accurately measured bone ash content (p = 0.942), underestimated LTM (0.59 +/- 0.05g, p < 0.001), and overestimated FM (2.19 +/- 0.06g, p < 0.001 ). Thus, DXA estimated 100% of bone ash content, 97% of carcass LTM, and 209% of carcass FM. DXA-derived values were then used to predict chemical values of FM and LTM. Chemically extracted FM was best predicted by DXA FM and DXA LTM [FM = -0.50 + 1.09(DXA FM) - 0.11(DXA LTM), model r2 = 0.86, root mean square error (RMSE) = 0.233 g] and chemically determined LTM by DXA LTM [LTM = -0.14 + 1.04(DXA LTM), r2 = 0.99, RMSE = 0.238 g].

Discussion: These data show that the precision of DXA for measuring TBMD, TBBM, FM, and LTM in mice ranges from a low of 0.84% to a high of 2.20% (CV). DXA accurately measured bone ash content but overestimated carcass FM and underestimated LTM. However, because of the close relationship between DXA-derived data and chemical carcass analysis for FM and LTM, prediction equations can be derived to more accurately predict body composition.