Prediction and evaluation of resting energy expenditure in a large group of obese outpatients

    loading  Checking for direct PDF access through Ovid

Abstract

BACKGROUND/OBJECTIVES:

The aim of this study was to compare resting energy expenditure (REE) measured (MREE) by indirect calorimetry (IC) and REE predicted (PREE) from established predictive equations in a large sample of obese Caucasian adults.

SUBJECTS/METHODS:

We evaluated 1851 obese patients (body mass index (BMI)>30 kg m-2) aged between 18a and 65 years. Data were obtained by comparing MREE with PREE, derived from different equations, within and between normal weight and obese groups. The mean differences between PREE and MREE as well as the accuracy prediction within ±10% level were investigated in the whole sample and in three subgroups, classified by BMI (Group 1 = 30–39.9 kg m-2; Group 2 = 40–49.9 kg m-2; Group 3>50 kg m-2).

RESULTS:

We observed that FAO, Henry and Muller3 (body composition (BC)) equations provided good mean PREE-MREE (bias -0.7, - 0.3 and 0.9%; root mean standard error (RMSE) 273, 263 and 269 kcal per day, respectively); HB and Henry equations were more accurate individually (57 and 56.9%). Only the Muller1 (BC) equation gave the lowest PREE-MREE difference (bias -1.7%; RMSE 228 kcal per day) in females, while Johnstone and De Lorenzo equations were the most accurate (55.1 and 54.8%). When the sample was split into three subgroups according to BMI, no differences were found in males; however, the majority of the equations included in this study failed to estimate REE in severely obese females (BMI>40 kg m-2). Overall, prediction accuracy was low (˜55%) for all predictive equations, regardless of BMI.

CONCLUSIONS:

Different established equations can be used for estimating REE at the population level in both sexes. However, the accuracy was very low for all predictive equations used, particularly among females and when BMI was high, limiting their use in clinical practice. Our findings suggest that the validation of new predictive equations would improve the accuracy of REE prediction, especially for severely obese subjects (BMI>40 kg m-2).

Related Topics

    loading  Loading Related Articles