Main content area

Concordance between whole- and half-body scans to evaluate body composition in dual-energy X-ray absorptiometry in children and adolescents with different nutritional and pubertal conditions

Ferreira, Mariana Simões, Marson, Fernando Augusto Lima, Wolf, Vaneza Lira Waldow, da Silva, Marcos Tadeu Nolasco, Zambon, Mariana Porto, Antonio, Maria Ângela Reis de Góes Monteiro, Ribeiro, José Dirceu, Mendes, Roberto Teixeira
Nutrition 2019 v.66 pp. 78-86
adolescents, body composition, children, densitometers, dual-energy X-ray absorptiometry, gender, nutritional status, regression analysis
Evaluation of body composition is a relevant clinical instrument for the follow-up assessments of children and adolescents, and dual-energy X-ray absorptiometry (DXA) is an accurate method for the pediatric population. However, DXA has limited scan area for the obese population. Thus, half-body scans emerged as an alternative to evaluate individuals with obesity. The aim of this study was to compare the body composition of children and adolescents with whole- and half-body DXA scans, considering nutritional status, pubertal development, sex, and age.This was a cross-sectional, analytical, and diagnostic intervention study with a sample of 82 participants of both sexes between 4 and 20 y of age. Body composition was evaluated by DXA using an iDXA bone densitometer (GE Healthcare Lunar, Madison, WI, USA). Two evaluations were performed: whole-body and half-body scans. The Bland–Altman correlation and linear regression tests were applied to identify the presence of association bias between the techniques. α = 0.05 was set.Of the 82 participants, 20 were excluded. A high correlation was observed between the data (correlation coefficient ∼0.999). Bland–Altman plots and regression analyses demonstrated correlation and randomness bias between whole- and half-body scan techniques in obese or normal weight participants for all DXA markers.The use of half-body scans was feasible and accurate to evaluate whole-body composition. The difference bias between techniques occurred randomly and was clinically irrelevant. A high correlation was observed between half- and whole-body analysis techniques.