Prediction of resting energy expenditure for adolescents with severe obesity: A multi-centre analysis

Background and Objectives: Resting energy expenditure (REE) assessments can help inform clinical treatment decisions in adolescents with elevated body mass index (BMI), but current equations are suboptimal for severe obesity. We developed a predictive REE equation for youth with severe obesity and obesity-related comorbidities and compared results to previously published predictive equations. Methods: Data from indirect calorimetry, clinical measures, and body composition per Dual x-ray absorptiometry (DXA) were collected from five sites. Data were randomly divided into development (N = 438) and validation (N = 118) cohorts. A predictive equation was developed using Elastic Net regression, using sex, race, ethnicity, weight, height, BMI percent of the 95th%ile (BMIp95), waist circumference, hip circumference, waist/hip ratio, age, Tanner stage, fat and fat-free mass. This equation was verified in the validation cohort and compared with 11 prior equations. Results: Data from the total cohort (n = 556, age 15 ± 1.7 years, 77% female, BMIp95 3.3 ± 0.94) were utilized. The best fit equation was REE = −2048 + 18.17 × (Height in cm) – 2.57 × (Weight in kg) + 7.88 × (BMIp95) + 189 × (1 = male, 0 = female), R² = 0.466, and mean bias of 23 kcal/day. Conclusion: This new equation provides an updated REE prediction that accounts for severe obesity and metabolic complications frequently observed in contemporary youth.