• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
Document
Author
Full name
Natália Tomborelli Bellafronte
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
Ribeirão Preto, 2017
Supervisor
Committee
Chiarello, Paula Garcia (President)
Prada, Patrícia de Oliveira
Machado, Dalmo Roberto Lopes
Suen, Vivian Marques Miguel
Title in Portuguese
Equações de predição de gasto energético de repouso por meio de dados gerados por avaliações de bioimpedância
Keywords in Portuguese
Bioimpedância elétrica
Calorimetria indireta
Composição corporal
Equação de predição
Gasto energético de repouso
Abstract in Portuguese
Avaliar acuradamente o gasto energético de repouso (GER) é de extrema importância no suporte nutricional e a análise de composição corporal influencia seu valor. O estudo teve como objetivos desenvolver equações preditivas de GER por meio de dados de composição corporal obtidos por exame de bioimpedância eléctrica multifrequencial por espectroscopia (BIS); avaliar a adequação das fórmulas mais usuais de predição do GER; medir a correlação dos parâmetros gerados por BIS com o GER, analisar a concordância e a correlação dos dados gerados pelos aparelhos de bioimpedância de frequência simples (BIA) e BIS, além da concordância entre os métodos de classificação do estado nutricional por Índice de Massa Corporal (IMC) e por %MG (Porcentual de Massa Gorda) avaliada por BIA e BIS. Caracterizou-se como um estudo transversal observacional desenvolvido com brasileiros saudáveis, ambos os sexos, entre 20 e 40 anos de idade, estratificados em subgrupos pelo IMC (subnutrido, n=40; eutrófico, n=120; com sobrepeso, n=118 e com obesidade, n=114) e pelo %MG (baixa gordura, n=17; gordura adequada, n=101; excesso de gordura, n=91 e obesidade, n=183). O GER foi medido por calorimetria indireta (CI). Houve emprego do teste de correlação de Spearman e de Pearson e do gráfico de dispersão para avaliar as associações entre as variáveis e de modelos de regressão linear múltipla no desenvolvimento das equações, por método Stepwise. Aplicou-se o teste de BlandAltman, o coeficiente de correlação intraclasse, o teste de Wilcoxon e o coeficiente de correlação kappa para análise de concordância entre medidas e classificações e o teste de Mann-Whitney e Kruskal-Wallis para comparação entre os subgrupos (p<0,05). O GER predito foi considerado adequado quando se encontrou entre 90 e 110% do GER medido por CI. Desenvolveu-se uma equação para a amostra total por sexo e uma para cada categoria do IMC e do %MG e as mesmas apresentaram baixos valores de coeficientes de determinação (R2). As maiores correlações entre as variáveis independentes com o GER ocorreram para o peso, IMC, Massa Gorda7 (MG) e Massa de Tecido Adiposo. Todas as equações usuais avaliadas não foram capazes de predizer corretamente o GER em metade da amostra. As classificações do estado nutricional realizadas por meio do IMC e %MGBIA obtiveram concordância fraca com aquela por %MGBIS. A concordância entre BIA e BIS foi baixa: tecidos corporais de maior hidratação foram superestimados e os menos hidratados subestimados, por BIA frente a BIS, e os vieses entre os dois equipamentos foram maiores com o aumento do IMC. Assim, as equações desenvolvidas apresentaram baixo R2, impossibilitando sua aplicação no cenário clínico. Já as equações de predição do GER avaliadas exibiram baixa adequação, não se recomendando seu uso. A classificação do estado nutricional por meio do IMC subestima as quantidades de MG, sendo mais adequada a utilização de composição corporal para caracterização nutricional. BIA e BIS geram resultados distintos: o tamanho corporal aparece como um fator de confusão na distinção das massas corporais analisadas, mas, a distribuição e a quantidade de água corporal total apresentam-se como fatores limitantes de maior força
Title in English
Resting energy expenditure prediction equation using bioelectrical impedance assessment data
Keywords in English
Indirect calorimetry
Prediction equation, Body composition. Bioelectrical impedance
Resting energy expenditure
Abstract in English
The accurate assessment of resting energy expenditure (REE) is extremely important in nutritional support for energy supply adjustment and body composition analysis plays a significant role in determining its value. The objectives of this thesis was to develop prediction equations of REE using body composition assessment data by bioelectrical impedance; assess the adequacy of the more usual prediction equations of REE against the measured value; measuring the correlation of the parameters generated by multifrequency spectroscopy bioelectrical impedance (BIS) in GER and analyze the agreement and correlation of data generated by bioelectrical impedance devices of simple frequency (BIA) and (BIS), in total sample and between subgroups stratified by body mass index (BMI) and body fat percentage (%BF), in addition to assess the agreement between the classification of nutritional state by BMI and %BF generated by BIA and BIS . This was an observational cross-sectional study with healthy Brazilians, both sexes, between 20 and 40 years old, stratified into subgroups by BMI (malnourished, n=40; eutrophic, n=120; overweight, n=118 and obese, n=114) and by %BF (low fat, n=17; suitable fat, n=101; excess fat, n=91 and obesity, n=183). There was the use of anthropometric and epidemiological parameters and those generated by BIS analysis in the equations's development. REE was measured by indirect calorimetry (IC). Employment Spearman correlation test and scatterplot to assess the associations between the variables and multiple linear regression models in the development of the equations. Application of Bland-Altman analysis, intraclass correlation coefficient, Wilcoxon test and kappa correlation coefficient for agreement analysis between measurements and classifications and the Mann-Whitney and Kruskal-Wallis test to compare the subgroups (p <0,05). Thus, an equation was developed for the total female sample and one for each of the last three categories of BMI and% BF, the same for males. The equations obtain low values of determination coefficient (R2). The highest correlations between the independent variables with REE occurred, for both females and males, with weight, BMI, BF and9 Adipose Tissue Mass. All the usual equations evaluated had low accuracy since none was able to correctly predict GER in 50% or more of the sample, either for the whole sample or stratified by BMI and %BF. The equations with the highest percentages of the sample within the adequacy limits were Owens, Henry-Rees and Livingston-Kohlstadt 2. The worst percentages of the sample within the adequacy limits were those of Ireton-Jones, FAO/WHO/UN 2 and Frankenfield 1. The nutritional status rankings performed through BMI and %BFBIA obtained weak agreement with that by %BFBIS, tending to classify the individual one or two levels below, underestimating the presence of %BF. The agreement between BIA and BIS was low since the equipment presented different results for all the variables, either in the total sample or in the stratified subgroups. The BIA against BIS underestimated the amounts related to the BF and total body water variables and overestimated those concerning the FFM and BCM. The biases between the two equipments were greater with the increase of BMI. Thus, the developed equations have low R2, which makes it impossible to apply them in the clinical setting. The most common and predictive GER prediction equations presented low accuracy, not proving to be adequate for use in clinical practice. The classification of nutritional status through BMI results in errors that compromise the approach of nutritional therapy, underestimating the amounts of BF and its deleterious potencies, so it is more appropriate to evaluate it through body composition. BIA and BIS generate different results, and body size appears as a confounding factor in the body mass distinction analyzed by BIA, but the distribution and amount of total body water is a limiting factor of greater strength for the BIA
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2019-07-29
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2021. All rights reserved.