Is a single bioelectrical impedance equation valid for children of wide ranges of age, pubertal status and nutritional status? Evidence from the 4-component model

BACKGROUND/OBJECTIVES: Bioelectrical impedance analysis (BIA) is widely used to predict body composition in paediatric research and clinical practice. Many equations have been published, but provide inconsistent predictions. AIMS: To test whether a single equation for lean mass (LM) estimation from BIA is appropriate across wide ranges of age, pubertal status and nutritional status, by testing whether specific groups differ in the slope or intercept of the equation. SUBJECTS/METHODS: In 547 healthy individuals aged 4-24 years (240 males), we collected data on body mass (BM) and height (HT), and lean mass (LM) using the 4-component model. Impedance (Z) was measured using TANITA BC418MA instrumentation. LM was regressed on HT2/Z. Multiple regression analysis was conducted to investigate whether groups based on gender, age, pubertal status or nutritional status differed in the association of LM with HT2/Z. RESULTS: BM ranged from 5 to 128 kg. HT2/Z was a strong predictor of LM (r(2) = 0.953, s.e.e. = 2.9 kg). There was little evidence of a sex difference in this relationship, however, children aged 4-7 years and 16-19 years differed significantly from other age groups in regression slopes and intercepts. Similar variability was encountered for pubertal stage, but not for nutritional status. CONCLUSIONS: No single BIA equation applies across the age range 4-24 years. At certain ages or pubertal stages, the slope and intercept of the equation relating LM to HT2/Z alters. Failure to address such age effects is likely to result in poor accuracy of BIA (errors of several kg) for longitudinal studies of change in body composition.