Comparison of adult and paediatric spine and whole body software for the Lunar dual energy X-ray absorptiometer.

Abstract

Simple phantoms were devised to compare the performance of adult (software 3.64) and paediatric (software 3.8 g) spine and whole body software developed for the Lunar dual energy X-ray absorptiometer. Rectangular slabs of aluminium with high (1.18 g cm-2) and low (0.57 g cm-2) density were used to represent bone mineral. For spine measurements, the phantoms were scanned in water at depths of 5-20 cm. For whole body measurements, the phantoms were scanned with known amounts of oil and water to represent fat and lean tissue. This simulated tissue depths of 5.5-19.7 cm and body composition ranging from 14-29% fat. There were systematic differences in spine and whole body bone mineral content (BMC), bone area (BA) and bone mineral density (BMD) measurements and also between adult and paediatric software versions. The magnitude and direction of these differences were dependent on BMD of the phantom and tissue depth. Similar systematic differences were observed in vivo when volunteers were scanned using adult and paediatric software. Paediatric software enabled measurements to be made at low tissue depths. The weights of fat, lean and total soft tissue measured by the adult and paediatric whole body software were similar to the values calculated from the known composition of the phantom. Precision estimates for all softwares were excellent. In conclusion, paediatric software should improve bone mineral measurements of children but the discrepancies between adult and paediatric softwares may cause problems in longitudinal studies of skeletal growth and when compiling reference data from infancy through to adulthood.