Estimation of lumbar spine bone mineral density by dual-energy X-ray absorptiometry:: standard anteroposterior scans vs sub-regional analyses of whole-body scans

Dual-energy X-ray absorptiometry of the lumbar spine (anteroposterior (AP) L(1)-L(4) DXA) is one of the standards for testing bone density (BMD). Whole-body (WB) DXA scanning allows analysis of the L-spine (WB-LS). The objective of this study was (i) to predict the AP L1-L4 BMD from the WB-LS BMD, and (ii) to compare WB and AP L1-L4 T-scores. 132 women participated in the study (age, 50.84 +/- 9.69 years; range, 21-74 years). AP L(1)-L(4) BMD was measured on a Hologic QDR 4500-A densitometer, and was followed by a WB scan. Linear regression analyses and curve estimations were performed for area, bone mineral content (BMC), BMD and T-scores. Precision errors for BMD from AP scans and WB-LS scans were 1.37% and 1.65%, respectively. The best linear correlations were found for BMD, followed by the T-scores (R = 0.882 and 0.871; p < 0.001). A power model for BMC (R(2) = 0.630) and a model S for BMD and the scanned area (R(2) = 0.841 and 0.539) were built. Discordant classification between AP L1-L4 and WB T-scores was found in 12.73% and 19.09% cases, respectively. WB T-score cut-off values of -2.35 and -0.90 produced the best sensitivity-to-specificity ratio in the diagnosis of lumbar spine osteoporosis and low bone mass. In conclusion, WB-LS analysis may provide data for the indirect estimation of spinal BMD. (C) 2008 The British Institute of Radiology.