Analisi metabolimetrica di un campione di donne obese soggette a Weight Cycling

Background: Obesity, a major health problem, is usually ascribed to energy imbalance. In obese people, this might be due to hypothalamic alterations in the control of body composition, thermogenesis and energy expenditure (EE), especially in “weight cyclers” (WC). Aim of the study: To verify energy balance and accuracy of the procedure. Materials and methods: In a group of 8 obese WC women (BMI=32.8) energy balance was assessed throughout a period of 9 months. Energy expenditure was estimated through a SenseWear Armband © (AB) alternate to a physical activity recording, whereas caloric intake through food consumption diaries compiled by subjects. A control group of 15 normal weight women (BMI=21.5) underwent the same procedures and was monitored for 1 month. We confirm validity of the procedure through a comparison by an indirect calorimeter in a group of 12 WC obese women (BMI 33.9). A systematic revision of the comparison between AB and indirect calorimetry (IC) was performed to verify the objectivity of the method. Results: In the analysis of data, energy balance was 869kcal negative than predicted contemplating body composition shifting. This gap was only 350kcal in the control group. Resting metabolic rate (RMR) measured via IC was 20% less than estimated with Armband, the difference was statistically significant with T-test (p < 0.05). Bland-Altman test showed a weak compatibility between methods. Systematic revision confirmed an overestimation trend, particularly in RMR. Discussion: AB overestimates EE in our sample and this trend was confirmed by systematic revision of the data literature. Methods are not interchangeable because Bland-Altman test has a too large bias. Conclusions: Although it's likely present an underestimation in food recording, AB overestimates EE in this small group of WC women. Estimating energy balance in dynamic conditions is a very hard task. Limitations of the study: Measures were not taken in the same experimental conditions. Sample is small