DXA performance in a pediatric population: precision of body composition measurements in healthy term-born infants using dual-energy X-ray absorptiometry

Abstract

Dual-energy X-ray absorptiometry (DXA) has been hailed as a golden standard for measuring body composition in adults but remains to be fully assessed for the infant population. A total of 64 newborn infants were allocated to 1 of 3 groups. All underwent 2 Hologic Discovery A DXA scans. Suboptimal scans were reconstructed, and an investigation into the success of adjustment was carried out. Depending on group, the factors of weight change and repositioning were investigated. Test-retest variation and coefficients of variation for DXA body composition estimates were calculated. Furthermore, the effects of flannel sheets and breast milk were investigated using a pediatric phantom. Reconstruction of suboptimal scans resulted in more accurate body weight estimates. Moderate weight change and repositioning had no significant effect on the variation between scans. No significant body composition changes occurred between scans. The test-retest variation varied between 6.3% and 11.8%. Flannel sheets and breast milk affected DXA results significantly. High precision of DXA measurements was obtained in our newborn population. Reconstructing scans is a viable way of correcting minor movement artifacts. Moderate weight changes and repositioning have no significant effect on DXA results, whereas flannel sheets and milk do.

Keywords: Body composition; dual-energy X-ray absorptiometry; neonatal; term born.