Dual-energy X-ray absorptiometry (DXA) is used to measure body composition in newborns; however, data on DXA accuracy are limited. We investigated the reliability of body composition measurements by DXA. The present study included 207 normal-term newborn babies, recruited from a larger study on the determinants of birth weight in healthy pregnancies (STORK) between 2005 and 2008. Reliability analysis of total fat mass (FM(DxA)), fat-free mass, lean mass (LM(DxA)), bone mineral content (BMC), and bone mineral density (BMD) were based on 2 DXA scans of 50 neonates. We also performed a comparison analysis for DXA (FM(DxA)) measurements and caliper (CLP) or circumference (CF) measurements of trunk and extremities (performed on all neonates, n=207). Reliability: All intraclass correlation coefficients (ICC) were satisfactory to excellent for total body and the extremity-compartment FM(DxA), LM(DxA), BMD, and BMC; ICC ranged from 0.86 to 0.96 but with a lower ICC for trunk FM(DxA). For comparison analysis, the Pearson correlation coefficients for CLP vs DXA and CF vs DXA ranged from 0.48 to 0.79 and 0.41 to 0.77, respectively. Quadriceps CLP and CF measurements correlated best with the most reliable DXA results, whereas more modest correlations were found for the trunk region. DXA measurements of body composition demonstrated good reliability and can be used as a reference method in neonates. CLP and CF measurements are appropriate for larger cohorts or when DXA is unavailable, and they provide fair rough estimations of fat mass.
The International Society for Clinical Densitometry. Published by Elsevier Inc. All rights reserved.