Objective: To examine the interrelationships of body composition variables derived from simple anthropometry, bioelectrical impedance (BIA) and dualenergy X-ray absorptiometry (DXA).
Population, material and methods: A total of 230 healthy preschool children (118 males and 112 females) age 4-6 years, attending a day care center, were examined to determine body mass index (BMI) and waist circumference (WC). Percentage of body fat content (%BF) was predicted by BIA (Maltron BF- 900) and DXA (Lunar DPX-L, pediatric software).
Results: BMI and WC did not show significant differences among sex, while %BF by BIA or DXA did. BIA measures were lower than DXA s (p <0.0001). Correlation between BIA vs. anthropometric methods and WC vs. DXA were moderate (Pearson r= 0.43 to 0.53), whereas the other correlations were strong (r= 0.71 to 0.83). However Bland Altman comparison showed wide limits of agreement between BIA and DXA; BIA significantly underestimated %BF as determined by DXA (p <0.0001).
Conclusion: The methods used to estimate body composition in healthy preschool children highly correlated. However, the Bland Altman procedure suggests that BIA and DXA should not be used interchangeably.