Endokrinologie und metabolisches Syndrom
Offener Zugang

Abstrakt

An Optimal Method for Measuring Body Fat in Overweight Individuals in Clinical Practice

Mariana Carvalho de Menezes, Aline Cristine Souza Lopes2, Lorena Pires Cunha, Ann Kristine Jansen and Luana Caroline dos Santos

Background: Excess body fat is associated with increased morbidity and mortality. Accordingly, it is important to choose reliable, simple and straightforward methods for measuring body composition in clinical practice. The aim of this study was to verify the concordance between body composition obtained by Skinfold Thickness (SF) and Bioelectrical Impedance Analysis (BIA) as well as their correlation with anthropometric indices.

Methods: The study was conducted from August 2007 to July 2009 among the adults and elderly patients (n = 85) receiving nutritional counseling at a primary health care unit in a Brazilian city. It was investigated all patients seen in the period. The following indices were used for comparison: Body Mass index (BMI), Waist Circumference (WC), Waist-Hip Ratio (WHR) and body composition evaluated by BIA and the sum of skinfolds (tricipital, bicipital, subscapular and suprailiac). The statistical analyses included the Kolmogorov-Smirnov test, Student’s t-test, Wilcoxon signed-rank test, the Pearson and Spearman correlation tests and the Bland-Altman method. Results: Eighty-five individuals were evaluated, predominantly women (91.8%) and overweight individuals (92.8% adults, 89.7% elderly), with a mean age of 51.8 ± 13.0 years (22;85). The percentage of body fat estimated by SF was significantly higher than that estimated by BIA (42.8%; 95% CI: 12.6-49.2 vs. 40.4%; 95% CI: 21.0-51.2; p<0.001). However, a moderate correlation (r = 0.58; p<0.001) and strong concordance [0.9797 (LC95%= -8.0519; 10.0113)] were observed between the two methods. Significant correlations were found between BMI and WC, but not WHR, with respect to body fat estimated by BIA and SF (r = 0.453 to 0.707; p<0.05). BIA presented stronger correlations with BMI and WC (r = 0.707 and 0.605, respectively) compared with SF (r = 0.493 and 0.453, respectively).

Conclusion: Strong concordance and a significant correlation were observed between BIA and SF, suggesting their validity in measuring body fat among overweight individuals. However, BIA appears to present better results when considering its higher correlation with the anthropometric indicators used.