Background: Accurate and precise methods to measure of body composition in infancy and childhood are needed.Objectives: This study evaluated differences and precision of three methods when compared with the four-compartment (4C) model for estimating fat mass (FM).Methods: FM of children (age 14 days to 6 years of age, N = 346) was obtained using quantitative nuclear magnetic resonance (QMR, EchoMRI-AH), air-displacement plethysmography (ADP, PeaPod, less than or equal to 8 kg, BodPod age 6 years or older), and dual-energy X-ray absorptiometry (DXA, Hologic QDR). The 4C model was computed. Correlation, concordance, and Bland-Altman analyses were performed.Results: In infants, PeaPod had high individual FM accuracy, whereas DXA had high group FM accuracy compared with 4C. In children, DXA had high group and individual FM accuracies compared with 4C. QMR underestimated group FM in infants and children (300 and 510 g, respectively). The instrument FM precision was best for QMR (10 g) followed by BodPod (34 g), PeaPod (38 g), and DXA (45 g).
Conclusions:In infants, PeaPod was the best method to estimate individual FM whereas DXA was best to estimate group FM. In children, DXA was best to estimate individual and group FM. QMR had the highest instrument precision. K E Y W O R D S air displacement plethysmography, body composition, dual-energy X-ray absorptiometry, fat mass, quantitative nuclear magnetic resonance 1 | INTRODUCTION Recent projections estimate that approximately 57% of children will be obese by 35 years of age, with half of those children developing obesity much earlier in childhood. 1 Previous analyses have shown that body mass index (BMI) >85th percentile at 2 weeks of age is associated with a two-fold risk of being overweight by the age of five. 2Further, increased adiposity (%fat mass [FM]) at birth persists through 5 months of age independent of early infant feeding and rapid infant growth. 3 BMI is commonly used in infants and children to assess growth, 4 effectiveness of intervention to reduce paediatric obesity, 5 and for prediction of obesity 6 and chronic diseases development. 7,8 While BMI in adults is related to clinical outcomes, its predictive value and usefulness in infants, children, and adolescents are less clear. 9 Further, complicating the use of BMI in infants and children is the interindividual variability for periods of rapid growth, for which BMI cannot Abbreviations: 4-C, four-compartment; ADP, air displacement plethysmography; BMI, body mass index; BV, body volume; BW, body weight; BMC, bone mineral content; DXA, dualenergy X-ray absorptiometry; FM, fat mass; MRI, magnetic resonance imaging; QMR, quantitative magnetic resonance; TBW, total body water.