A novel approach to assess body composition in children with obesity from density of the fat-free mass

Background & aims: Assessment of Fat Mass (FM) and fat-free mass (FFM) using Air-displacement plethysmography (ADP) technique assumes constant density of FFM (D_FFM) by age and sex. It has been recently shown that D_FFM further varies according to body mass index (BMI), meaning that ADP body composition assessments of children with obesity could be biased if D_FFM is assumed to be constant. The aim of this study was to validate the use of the calculations of D_FFM (rather than constant density of the FFM) to improve accuracy of body composition assessment in children with obesity.

Methods: cross-sectional validation study in 66 children with obesity (aged 8-14 years) where ADP assessments of body composition assuming constant density (FFM_BODPOD and FM_BODPOD) were compared to those where D_FFM was adjusted in relation to BMI (FFM_adjusted and FM_adjusted), and both compared to the gold standard reference, the 4-component model (FFM_4C and FM_4C).

Results: FFM_BODPOD was overestimated by 1.50 kg (95%CI -0.68 kg, 3.63 kg) while FFM_adjusted was 0.71 kg (-1.08 kg, 2.51 kg) (percentage differences compared to FFM_4C were 4.9% (±2.9%) and 2.8% (±2.1%), respectively (p < 0.001)). Consistently, FM was underestimated by both methods, representing a mean difference between methods of 4.0% (±2.9%) and 6.8% (±3.8%), respectively, when compared to the reference method. The agreement and reliability of body composition assessments were improved when adjusted using calculations (adjusted models) rather than assuming constant D_FFM.

Conclusions: The use of constant values for fat-free mass properties may increase bias when assessing body composition (FM and FFM) in children with obesity by two-component techniques such as ADP. Using adjusted corrections as proposed in the present work may reduce the bias by half.