Jonathan Ck Wells1, Jane E Williams2, Leigh C Ward3, Mary S Fewtrell2. 1. Childhood Nutrition Research Centre, Population, Policy and Practice Reseach and Teaching Department, UCL Great Ormond Street Institute of Child Health, 30 Guilford Street, London, WC1N 1EH, UK. Electronic address: Jonathan.Wells@ucl.ac.uk. 2. Childhood Nutrition Research Centre, Population, Policy and Practice Reseach and Teaching Department, UCL Great Ormond Street Institute of Child Health, 30 Guilford Street, London, WC1N 1EH, UK. 3. School of Chemistry and Molecular Biosciences, The University of Queensland, Brisbane, Australia.
Abstract
BACKGROUND & AIMS: Bioelectrical impedance analysis (BIA) is widely considered a body composition technique suitable for routine application. However, its utility in sick or malnourished children is complicated by variability in hydration. A BIA variant termed vector analysis (BIVA) aims to resolve this, by differentiating hydration from cell mass. However, the model was only partially supported by children's data. To improve accuracy, further adjustment for body shape variability has been proposed, known as specific BIVA (BIVAspecific). METHODS: We re-analysed body composition data from 281 children and adolescents (46% male) aged 4-20 years of European ancestry. Measurements included anthropometry, conventional BIA, BIVA outcomes adjusted either for height (BIVAconventional), or for height and body cross-sectional area (BIVAspecific), and fat-free mass (FFM) and fat mass (FM) by the criterion 4-component model. Graphic analysis and regression analysis were used to evaluate different BIA models for predicting FFM and FM. RESULTS: Age was strongly correlated with BIVAconventional parameters, but weakly with BIVAspecific parameters. FFM correlated more strongly with BIVAconventional than with BIVAspecific parameters, whereas the opposite pattern was found for FM. In multiple regression analyses, the best prediction models combined conventional BIA with BIVAspecific parameters, explaining 97.0% and 89.8% of the variance in FFM and FM respectively. These models could be further improved by incorporating body weight. CONCLUSIONS: The prediction of body composition can be improved by combining two different theoretical models, each of which appears to provide different information about the two components FFM and FM. Further work should test the utility of this approach in pediatric patients.
BACKGROUND & AIMS: Bioelectrical impedance analysis (BIA) is widely considered a body composition technique suitable for routine application. However, its utility in sick or malnourished children is complicated by variability in hydration. A BIA variant termed vector analysis (BIVA) aims to resolve this, by differentiating hydration from cell mass. However, the model was only partially supported by children's data. To improve accuracy, further adjustment for body shape variability has been proposed, known as specific BIVA (BIVAspecific). METHODS: We re-analysed body composition data from 281 children and adolescents (46% male) aged 4-20 years of European ancestry. Measurements included anthropometry, conventional BIA, BIVA outcomes adjusted either for height (BIVAconventional), or for height and body cross-sectional area (BIVAspecific), and fat-free mass (FFM) and fat mass (FM) by the criterion 4-component model. Graphic analysis and regression analysis were used to evaluate different BIA models for predicting FFM and FM. RESULTS: Age was strongly correlated with BIVAconventional parameters, but weakly with BIVAspecific parameters. FFM correlated more strongly with BIVAconventional than with BIVAspecific parameters, whereas the opposite pattern was found for FM. In multiple regression analyses, the best prediction models combined conventional BIA with BIVAspecific parameters, explaining 97.0% and 89.8% of the variance in FFM and FM respectively. These models could be further improved by incorporating body weight. CONCLUSIONS: The prediction of body composition can be improved by combining two different theoretical models, each of which appears to provide different information about the two components FFM and FM. Further work should test the utility of this approach in pediatric patients.
Authors: Mary Horlick; Stephen M Arpadi; James Bethel; Jack Wang; Jack Moye; Patricia Cuff; Richard N Pierson; Donald Kotler Journal: Am J Clin Nutr Date: 2002-11 Impact factor: 7.045
Authors: Claire M Nightingale; Alicja R Rudnicka; Chris G Owen; Derek G Cook; Peter H Whincup Journal: Int J Epidemiol Date: 2010-11-01 Impact factor: 7.196
Authors: H Croker; R M Viner; D Nicholls; D Haroun; P Chadwick; C Edwards; J C K Wells; J Wardle Journal: Int J Obes (Lond) Date: 2011-09-20 Impact factor: 5.095
Authors: Camila E Orsso; Maria Cristina Gonzalez; Michael Johannes Maisch; Andrea M Haqq; Carla M Prado Journal: Eur J Clin Nutr Date: 2021-10-07 Impact factor: 4.884
Authors: Kaspar Staub; Nicole Bender; Turgut Ekingen; Cynthia Sob; Christina Hartmann; Frank J Rühli; Katarina L Matthes Journal: BMC Public Health Date: 2022-05-05 Impact factor: 4.135