Predictive equations are inaccurate to assess caloric needs in non-white adults from Chile

Objectives: Predictive equations are frequently used to estimate resting energy expenditure (REE) because indirect calorimetry (IC) is not always available and is expensive. The aim of this study was to determine the concordance between the estimation of REE using predictive equations and its measurement by IC. Methods: This was an analysis of the registry of indirect calorimetry performed in non-hospitalized participants. Harris-Benedict, FAO/WHO/UNU, Mifflin St. Jeor, and European Society for Clinical Nutrition and Metabolism (ESPEN) equations were used to estimate REE in these individuals. The concordance between measured and estimated REE using real, ideal, and adjusted weight was calculated using the concordance coefficient analysis of Lin and Bland- Altman plots in all participants and in subgroups separated according to their body mass index. Results: We retrieved 680 measurements and discarded 247 that did not comply with the inclusion criteria. Thus, we studied 433 participants ages 36 y (29-48 y). Of the participants, 341 were women (79%) and the participants had a body mass index (BMI) of 30 kg/m(2) (26.7-33.1 kg/m(2)). All predictive equations had concordance values <0.90. The proportion of participants in which the difference was >10% ranged from 36% to 87%. The ESPEN equation had the greater proportion of erroneous estimations of REE in all participants and BMI subgroups when real weight was used. Conclusions: We observed a low level of concordance between REE estimated using predictive equations and measured by IC. These results should alert clinicians about the inaccuracy of predictive equations. (C) 2020 Elsevier Inc. All rights reserved.