RESUMO
OBJECTIVE: To compare body composition as assessed by conventional and vector bioelectrical impedance analysis according to the nutritional cataloging using body mass index (BMI) in a group of institutionalized elderly. METHODS: Cross-sectional study in 38 institutionalized elderly. Body composition was estimated by bioimpedance analysis. Differences in body composition were analyzed using t-test and ANOVA, or their corresponding nonparametric tests. Statistical significance was set at p<0.05. RESULTS: Based on BMI, the sample showed overweight (average BMI: 26.4kg/m2), and women had higher BMI values than men (28.9 vs. 25.5kg/m2). Based on waist circumference, abdominal obesity was detected in 60.7% of men and 80% of women. Conventional bioimpedance analysis (BIA) yielded high fat mass values and slightly depleted skeletal muscle mass, compatible with sarcopenic obesity. All individual impedance vectors were located on the right of the major axis of the tolerance ellipses, reflecting body-cell-mass depletion in all subjects, regardless of BMI cataloging. CONCLUSIONS: Bioelectrical impedance vector analysis (BIVA) detects body compartment changes in institutionalized elderly that are not identified by the most widely used clinical practice nutritional indicators, such as BMI, waist circumference, and BIA-estimated body composition.
Assuntos
Antropometria/métodos , Composição Corporal , Impedância Elétrica , Obesidade/diagnóstico , Idoso , Idoso de 80 Anos ou mais , Índice de Massa Corporal , Estudos Transversais , Feminino , Avaliação Geriátrica , Humanos , Institucionalização , Masculino , Avaliação Nutricional , Estado Nutricional , Projetos Piloto , Instituições Residenciais , EspanhaRESUMO
INTRODUCTION: although there are precise and accurate techniques for estimating resting energy expenditure, like the indirect calorimetry (IC), daily practice needs faster, easier and cheaper methods as the predictive equations. OBJECTIVE: the aim of the study was to develop a new predictive equation for estimating resting energy expenditure (REE) for healthy Spanish population. Méthods: the REE of 95 healthy normal weighted volunteers was determined by indirect calorimetry (IC). The new equation was obtained by multiple lineal regression by using the analytical criteria of the Cp of Mallows and the adjusted R2. Then, the behavior of the new formula was studied in a group of overweight volunteers through the intraclass correlation coefficient (ICC) and Bland-Almand plots. The level of signification was reached at p < 0,05. RESULTS: the average age was 42 years (range: 2.0-63.2). Mean REE determined by IC was 1 589.1 kcal/d (312.0). The selected equation was: [y = 1 376.4 - 308 Sex (M = 0; W = 1) + 11.1 Weigh (kg) - 8 Age (years)] (R2: 0.68; EE: 175.95). The ICC between the new equation and the IC in normal weighted subjects was 0.901 (95%CI: 0.851 - 0.934). The new formula showed a good level of agreement in the overweight group (ICC: 0.880; 95%IC: 0.772 - 0.937). CONCLUSIONS: we propose a new predictive equation for estimating the REE for healthy Spanish population which has an easy application and includes sex, age and weigh. The selected equation shows an adequate behavior in overweight subjects too.
Introducción: aunque se dispone de técnicas precisas y exactas para la estimación del GER, como la calorimetría indirecta (CI), en la práctica diaria se precisan métodos rápidos, fáciles de aplicar y económicos, como los modelos predictivos. Objetivo: desarrollar una nueva ecuación predictiva del gasto energético en reposo (GER) para población española adulta sana. Métodos: se determinó el GER en 95 sujetos sanos con normopeso mediante CI. Se utilizó la regresión lineal múltiple para la obtención del modelo, empleando como criterios analíticos la Cp de Mallows y el R2 ajustado. Se estudió el comportamiento del modelo generado en una muestra de 39 sujetos con IMC ≥ 25 kg/m2 mediante el coeficiente de correlación intraclase (ICC) y la prueba de Bland-Altman. La significación se alcanzó con p < 0,05. Resultados: la edad media fue de 42 años (rango: 23,0 63,2). El valor medio del GER estimado fue de 1.589,1 kcal/d (312,0). La ecuación seleccionada fue: [GER (kcal/d) = 1.376,4 308 Sexo (V = 0; M = 1) + 11,1 Peso (kg) 8 Edad (años)] (R2: 0,68; EE: 175,95). El CCI entre el nuevo modelo predictivo y la CI fue de 0,901 (IC del 95%: 0,851 0,934) en la muestra con normopeso. La aplicación del modelo en la muestra de sujetos con IMC ≥ 25 kg/m2 alcanzó una buena concordancia (CCI de 0,880; IC del 95%: 0,772 0,937). Conclusiones: se presenta un nuevo modelo de estimación del GER para población sana española fácilmente aplicable en la práctica diaria, que incluye las variables sexo, edad y peso. La ecuación propuesta presenta un comportamiento adecuado en sujetos con IMC ≥ 25 kg/m2.