Validação da impedância bioelétrica para determinação da composição corporal em mulheres no período pós parto

Abstract

Postpartum women are at risk of body weight retention and consequently becoming obese. Usually, during postpartum period there is a greater concern in weight loss instead of measuring which body compartment is being mobilized. Thus, there is a growing interesting to validate field and affordable techniques, such as the bioelectrical impedance (BIA) to determine body composition in this segment of the population. The purpose of this study was to validate different BIA devices using dual-energy x-ray absorptiomety (DXA) as a goldstandard in postpartum women. Forty-six postpartum women aged 18 years and over were enrolled in the study. They were recruited soon after delivery from two health units (Hospital São José of Mesquita District and Policlínica Piquet Carneiro, located in Rio de Janeiro city).All measurements were performed at the Laboratório Interdisciplinar de Avaliação Nutricional da Universidade do Estado do Rio de Janeiro. Participants were instructed to follow a standardized protocol for body composition measurements. Anthropometric variables(body mass - BM, stature, waist and hip circumferences) were collected. Body composition was determined by DXA and also obtained using the following BIA devices :model RJL 101(RJL Systems®, Inc. Clinton, USA), model BIOD 450 (Biodynamics Corporation®, USA)and I model BC 533 (Tanita®, USA). Descriptive statistical analysis was done for continuous variables. ANOVA was used to verify differences between mean values of fat free mass(FFM), fat mass (FM) and percentage body fat (%BF) within BIA devices and DXA. Concordance analysis between each body component obtained by DXA and all BIA devices were performed using Bland & Altman plots and the Concordance Correlation Coefficient(CCC). ANOVA was used to compare mean values within BIA and DXA, and Tukey post-hoc was adopted. SPSS and MedCalc software were used in the analysis. Mean values (±standard deviation) of age, BM, body mass index, FM by DXA, FFM by DXA, %BF by DXA were 25.5 ± 4.6 years, 73.6 ± 12.2kg, 28.7 ± 4.3kg/m², 29.8±7.4kg, 43.2 ± 5.7kg, 41.8 ± 4.3%,respectively. When comparing the results within different equipments (DXA, RJL, BIOD and TAN), it was verified that there was significant difference between DXA and BIA equipments for all body components, except for MG from RJL (27.1Kg) and TAN (27.1Kg) in relation toD XA (29.8Kg). The results showed good reproducibility in relation to BIA TAN (FFM, r=0.74; FM, r=0.90) and RJL (FFM, r=0.78; FM, r=0.89). It was verified that BIA TAN and RJL yielded similar estimates for FFM, FM and % BF. When all women were considered together no significant difference was observed for FM obtained by TAN, RJL and DXA. When considering BMI categories the estimates were better for obese women. However, the agreement was not satisfactory for body components supplied by BIA in relation to DXA. All BIA devices tended to underestimated % BF and overestimated FFM for this sample.