Comparison Of Methods T Assess Percent Fat And Lean Mass With Bioelectrical Inpedance Analysis (bia) In Zambian Children

Introduction
Zambian mothers and pregnant women face problems of rapid population growths and this puts them under increasing
pressure. Gender and race specific calculation are necessary in order to have a more valid assessment on children body
composition.

Methods
To assess reproducibility of results of two equations for percent fat and one for lean mass in Zambian children and
compare the results with those of Bioelectrical Impedance Analysis (BIA). 139 children with mean age of 4.9 years
were evaluated. BIA (TANITA®) TBF 300A was used to assess percent fat, and lean mass for those children.
Slaughter’s and Houtkooper’s equations were used to calculate fat and lean mass. Lohman’s equation was used to
predict percent fat. The results of these equations were compared to those of BIA. Paired t-test was used to identify a
difference between means and Bland-Altman plots to evaluate precision of measurements. Pearson correlation
coefficient was used to assess the reliability of reproducibility.

Results
Regardless of which equation was used the mean differences between the equations and TANITA were significant for
percent fat (p<0.001), lean mass (p<0.001). There was a correlation between the difference of TANITA-Slaughter’s,
TANITA- Houtkooper’s equation and those of the Slaughter’s and Houtkooper’s equation mean respectively (r=0.553,
p<0.001), (r=-0.289,p<0.05). No correlation was found for the difference between TANITA-Lohman and Lohman’s
equation mean (r=0.161,p=0.263). There was a difference (p<0.05) between Lohman’s and Slaughter’s equation for
percent fat. However, reproducibility of results was comparable when using Bland-Altman plots for all methods.

Discusion/Conclusions
Lohman’s equation for percent fat was not correlated with the results from TANITA. A significant correlation was
found between Slaughter’s equation for percent fat and Houtkooper’s equation for lean mass with TANITA.
Given that we had a good correlation between the compared equations, this does not mean that we can use either one of
them interchangeably to assess Fat percent or Fat Free Mass in children.

References
[1]. Story M, et al. (2003) J. Ethn Dis.13,S54-64.
[2]. Tershakovec AM,et al. (2003) Int J Obes Relat Metab Disord, 1,19-24.
[3]. Wong WW, et al. (2002). Am J Clin Nut, :384-9.

NOTA: O texto com a iconografia está no anexo.

Arquivo