Comparação entre percentuais de gordura corporal estimados por bioimpedância bipolar e tetrapolar / Comparison between body fat percentages estimated by bipolar and tetrapolar bioimpedance

RESUMO

O objetivo deste estudo foi avaliar se os resultados relativos ao percentual de gordura corporal obtidos pelos métodos de bioimpedância bipolar são semelhantes ao método tetrapolar. Foi realizado um estudo transversal, com 88 indivíduos (≥ 18 anos), de ambos os sexos, de Taubaté - SP, definidos por amostragem probabilística não aleatória, os quais foram submetidos aos testes de bioimpedância tetrapolar de corpo inteiro, bipolar para membros superiores (BMS) e bipolar para membros inferiores (BMI), após seguirem os pré-requisitos para a sua realização. Foram obtidos os valores estimados de %GC e classificada sua adequação, por cada aparelho. Calculou-se a sensibilidade, a especificidade e os valores preditivos e traçou-se a curva ROC dos aparelhos de BMI e BMS, tendo como padrão os valores obtidos pela BIA tetrapolar. A idade média dos indivíduos foi 23,36±5,75 anos, sendo 88,64% do sexo feminino. As médias de %GC estimadas pelos três aparelhos (BIA tetrapolar=27,22%; BMS=24,96% e BMI=27,82%) para ambos os sexos foram classificadas como adequadas (<25 % para homens e <35% para mulheres), de acordo com a Organização Mundial de Saúde. Em relação à BIA tetrapolar, a BMS e a BMI apresentaram sensibilidades semelhantes, porém, a BMS foi mais específica do que a BMI. Pela análise das curvas ROC, observou-se maior área sob a curva para BMS. Conclui-se que a bioimpedância para membros superiores apresentou maior eficiência em predizer o diagnóstico de gordura corporal adequada e excessiva quando comparada à bioimpedância tetrapolar.

ABSTRACT

The objective of the study was to analyze and to compare if the body fat percentage (BF%) estimated by bipolar bioelectrical impedance analyzer (BIA) is similar to the tetrapolar method. This was a cross-sectional study, with 88 subjects (≥18 years old) from Taubaté – SP chosen by non-SP chosen by non-random probability sample, which was submitted to tests of tetrapolar, hand-to-hand and foot-to-foot BIA, after taking the pre-requisites for their performance. Estimated values of %BF were obtained and their adjustment for each instrument was classified. Sensibility, specificity and predictive values were calculated and the ROC curve for the foot-to-foot BIA, after taking the pre-requisites for their performance. Estimated values of %BF were obtained and their adjustment for each instrument was classified. Sensibility, specificity and predictive values were calculated and the ROC curve for the foot-to-foot and hand-to-hand analyzers was done taking the standard values obtained by tetrapolar BIA. The average age of subjects was 23,36±5,75 years old, 88,64% were female. The average %BF estimated by instruments (tetrapolar BIA=27,22%; hand-to-hand BIA=24,96% e foot-to-foot BIA=27,82%), for both sex, were classified as appropriate (<25% for male and 35% for female), according to World Health Organization. In relation to tetrapolar BIA, hand-to-hand and foot-to-foot pretend a similar sensibility. However, hand-to-hand was more specific than foot-to-foot analyzer. Analyzing ROC curves, a bigger area under the curve obtained by hand-to-hand was observed showing greater efficiency to predict body fat when compared with tetrapolar than foot-to-foot analyzer. In conclusion, hand-to-hand bioelectrical impedance had greater efficiency in predicting a diagnosis of ideal and excessive body fat compared to the tetrapolar method.