Fat mass by air-displacement plethysmography and impedance in obese/non-obese children and adolescents

Objective. To determine the level of agreement between measurements of body composition by air-displacement plethysmography (ADP) and bioelectrical impedance analysis (BIA) in obese/non-obese children and adolescents. Methods. Fat mass (FM) and fat free mass (FFM) were measured by ADP using the BOD-POD system and foot-to-foot BIA in 187 children and adolescents (75 males and 112 females, aged 5 to 22 years). Obesity was defined as a percentage FM (determined by BOD-POD), as a percentage (%) higher than 25%-35%. Sixty-four subjects were obese and 123 non-obese. Results. Lin's Concordance Coefficient (Rc) between estimates of FM (%) and FFM (kg) was 0.79 (95% CI: 0.73; 0.83) by BIA and 0.96 (95% CI: 0.95; 0.97) by ADP. For the group of patients as a whole, the mean difference (p < 0.001) between methods (the BIA measurement minus the ADP measurement) was -3.39 (95% CI: -4.13; -2.65) for FM (%) and 1.54 (95% CI: 1.10; 1.98) for FFM (kg) (p < 0.001). The limits of agreement were -13.70; 6.90 for FM (%) and 1.40; 7.60 for FFM (kg). In the obese group, the mean difference between methods was -5.01 (95% CI: -6.21; -3.81) for FM (%) and 2.58 (95% CI: 3.45; 1.71) for FFM (kg) (p <0.001). In the non-obese group, these mean differences were 2.49 (95% CI: -3.41; -1.57) and 0.96 (95% CI: 1.43; 0.50), respectively (p < 0.001). Conclusions. Compared with ADP, foot-to-foot BIA overestimates FFM and underestimates FM in obese and non-obese children of either sex. ADP and BIA estimates of FFM and FM are highly correlated for both obese/non-obese children. However, the large limits of agreement suggest that these methods should not be used interchangeably.