Validation of surrogate limb analysis for body composition in children by dual energy X-ray absorptiometry (DXA)

Dual energy X-ray absorptiometry (DXA) is a recognized tool for measurement of body composition and provides benefits of low dose radiation, quick scan time and multiple measurement options. Challenges arise in scanning children, particularly with limb movement. We aimed to validate the use of surrogate limb substitutions compared with whole-body scans by DXA for measuring fat, lean and bone mass in children.

DXA scans were obtained from 3-year-old children who had normal positioning and no limb movement (n=246) or movement of a single limb (n=55). By replacing the measurements of one scanned limb with those of the opposite limb, we obtained an estimate value that was compared with the original whole-body scan measures for fat, lean and bone mass, percent whole-body fat and total mass for scans without or with movement.

Original normal scan analyses were highly correlated with estimates using substitution of the surrogate limb for all body compartments (R(2)=0.986-0.999, P<0.005). The Bland-Altman analysis demonstrated high levels of agreement between the estimates using surrogate limb and original whole-body measurements. Differences in estimate values and variance were greater with limb substitution in scans with movement compared to without movement.

Use of limb surrogate methodology for fat mass, lean mass, bone mineral content, percent fat and total mass, is a valid alternative that minimizes variation in estimates of body composition from DXA scans in young children in which a single limb is affected by movement. This will allow inclusion of scans with movement artifact in a single limb for data analysis.