Determining the accuracy of predictive energy expenditure (PREE) equations in severely obese adolescents

Background & aims: Severe obesity in children and adolescents is now a serious global health concern. Accurate measurements of resting energy expenditure (REE) is a key foundation for successful obesity treatment. Clinical dietitians rely heavily on measured or calculated REE to tailor dietary interventions. Indirect calorimetry (IC) is the gold standard for measuring REE. However, predictive resting energy expenditure (PREE) equations are commonly used when IC is unavailable due to cost or practicality. PREE equations differ based on variables such as age, gender, weight, and height and selecting the most accurate PREE for an individual is crucial to avoid over or underestimation of energy requirements. Published studies investigating the accuracy of PREE equations in obese children and adolescents have reported inconsistent findings, which likely result from heterogeneity in the patient populations studied. Accordingly, this study aimed to (a) assess the accuracy of the published PREE equations in a group of severely obese (SO) adolescents using IC measurement, and (b) determine if there is a BMI threshold at which the PREE equations become less accurate. Methods: SO adolescents were studied using IC. REE was calculated using nine commonly used PREE equations. Generalized linear regression equations were used to compare absolute and relative differences between calculated and measured REE (MREE) for each PREE equation. Accuracy was calculated as the percentage of subjects with PREE values within 10 percent of MREE. Results: 226 SO adolescents (mean +/- SD age: 15.9 +/- 1.9 years; weight: 126.9 +/- 24.5 kg; BMI: 44.9 +/- 8.1 kg/m(2)) participated. Mean MREE was 2163 +/- 443 kcal/d. PREE calculated by the Mifflin equation was the only equation without a statistically significant bias compared to MREE (mean bias of -23 +/- 307 kcal/d; p = 0.26). Mifflin was also the most accurate with 61% of individuals within +/- 10% of MREE. PREE equations accuracy was not associated with degree of BMI elevation (31-69 kg/m(2)). Conclusions: In adolescents with severe obesity, the Mifflin equation best predicts REE. This should be the equation applied when using PREE to optimize nutritional care in this population. (C) 2016 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.