Development of a predictive equation for resting energy expenditure in pediatric patients with oncological diagnosis.

Background and aim: Pediatric cancer is a significant health concern, particularly in low- and middle-income countries with lower cure rates. The nutritional status of these patients is crucial because malnutrition, whether due to a deficiency or excess of energy, can negatively impact treatment response and long-term outcomes. Since resting energy expenditure (REE) is a key parameter for planning appropriate nutritional support, accurate assessment is essential. However, the most precise methods, such as indirect calorimetry (IC), are not always available, leading to predictive equations based on easily accessible variables. These equations may be inaccurate if they are not specifically designed for children with cancer. Therefore, this study presents an equation to estimate REE in pediatric patients with oncological diagnosis and to compare the accuracy of this equation with those of previous equations developed in different pediatric populations to assess its utility in a clinical population. Methodology: A cross-sectional study was conducted in pediatric patients aged 6 to <18 years with a recent oncological diagnosis. After diagnosis, anthropometric measurements were taken, nutritional status was assessed, body composition was determined using bioelectrical impedance, and REE was measured through IC. Results: A total of 226 pediatric participants were evaluated, of whom 203 were included in the final analysis. The majority had solid tumors (68.5%), followed by leukemia (20.2%) and brain tumors (11.3%). Significant differences in anthropometric and biochemical variables were observed among the different diagnoses, with patients with brain tumor having lower REE/kg of body weight. Two new REE prediction equations specific to this population were developed: the INP-simple model, which is based on basic clinical variables, and the INP-Morpho model, which includes body composition. Both new INP equations showed less bias in REE estimation (114.8, 95% CI: -408, 638) than traditional equations, including the Harris-Benedict (-133.6, 95% CI: -671.5, 404.2), FAO (-178.8, 95% CI: -683.9, 326.3), Schofield (-185.4, 95% CI: -697.6, 326.8), IOM (-201, 95% CI: -761.7, 359.7), Oxford (-110.6, 95% CI: -661.4, 440.1), Kaneko (-135.6, 95% CI: -652.5, 381.4) and Müller (-162.6, 95% CI: -715.1, 389.9) equations but not the Molnár equation (-82.3, 95% CI: -741.3, 576.7). Conclusion: Children with cancer often have energy expenditure levels that differ from the recommended values, increasing their risk of malnutrition or obesity. Predictive equations specifically developed for this population may offer improved accuracy for estimating REE in clinical settings, although external validation is still needed.