Gasto energético de repouso em idosos longevos saudáveis: fatores associados e uso de fórmulas preditivas

Abstract

Introduction: The aging process is associated with a progressive reduction in resting energy expenditure (REE). Although indirect calorimetry (IC) is considered the gold standard for assessing REE, predictive mathematical equations are more commonly used in clinical practice. Objective: To evaluate the accuracy and agreement between measured resting energy expenditure (mREE) and estimated resting energy expenditure (pREE), as well as their associations with body composition, strength, and engagement in physical activity or exercise in healthy long-lived elderly individuals. Methods: Cross-sectional study with 74 elderly individuals (45 women and 29 men) aged ≥ 80 years, who were healthy. REE was measured by IC (after a 12-hour overnight fast) and estimated by 11 prediction equations. Body composition was assessed by bioelectrical impedance analysis. The normality was assessed by the Shapiro-Wilk test. Student's t-tests and Mann-Whitney tests were used for comparing means and medians, respectively, between sexes. Proportion comparisons were made using the chi-square test. Comparison between mREE and pREE values was performed using the Wilcoxon test. Spearman and Pearson correlation was conducted to compare associations. Variations of 10% from mREE were used as an accuracy measure. Individual REE agreement was examined by tertiles and Bland-Altman analysis. Results: The median age was 85 years (82.00 – 85.25). The mREE showed moderate correlation with all pREE (0.30 ≤ | r | < 0.60). When considering genders, correlations were significant only among men. All equations overestimated REE values. The smallest total difference between mREE and pREE was achieved by the equation by Mifflin et al. (1990) (237.16 kcal/d), for women by Porter et al. (2023) (247.43 kcal/d), and for men by Anjos et al. (2014) (326.59 kcal/d). The highest accuracy among women was identified by the equation Porter et al. (2023) (26.7%), while Anjos et al. (2014) provided higher accuracy in the total group and men (23.0% and 20.7%, respectively). This equation also showed better individual agreement in tertile classification (40.5%). Low agreement was identified in all formulas by Bland-Altman plots. There was a strong correlation between mREE and lean body mass (r=0.329, p=0.008), fat mass in men (r=0.607, p=<0.001), and calf circumference for the total (r=0.322, p=0.001) and for men (r=0.419, p=0.009). Conclusion: We identified the need for caution when using REE estimation equations in long-lived elderly individuals, with the Anjos et al. (2014) equation being the least inaccurate, although overall accuracy and individual agreement are limited. Body composition, particularly fat-free mass, lean body mass, and fat mass, influenced REE in elderly individuals.