Background/objectives: Using dual X-ray absorptiometry (DXA) to assess body composition in children has limitations including expense, lack of portability, and exposure to radiation. The aims of this study were to examine: (1) validity of quantitative ultrasound (QUS) against DXA for measuring bone density and (2) the validity of in-built algorithm of bioelectrical impedance analysis (BIA) for measuring body composition in children (8-13 years) living in New Zealand.
Subjects/methods: Total body less head (TBLH), bone mineral content (BMC), bone mineral density (BMD), and body composition were measured with DXA (QDR Discovery A, Hologic, USA); calcaneal BMD and stiffness index (SI) with QUS (Sahara QUS, Hologic, USA), and BIA on the InBody 230 (Biospace Ltd., Seoul, Korea). Relative validity was assessed using Pearson's and Lin's concordance correlation coefficients (CCC), and Bland-Altman plots.
Results: In 124 healthy children, positive correlations between QUS SI and DXA (BMC and BMD) were observed (range = 0.30-0.45, P < 0.01). Results from Lin's CCC test showed almost perfect correlations between BIA and DXA fat free mass (0.96), fat mass (0.92), and substantial correlation for percentage of fat mass (0.75) (P < 0.05).
Conclusion: Although BIA results were not as accurate as DXA and DXA remains the gold standard method for clinical assessment, BIA can be an alternative method for investigating body composition among children in large cohort field studies. Calcaneal QUS and DXA are not interchangeable methods for measuring bone density in children similar to our study population.