Estimating Energy Expenditure in Critically Ill Children: Still Shooting in the Dark?

Artificial Neural Network Algorithms to Predict Resting Energy Expenditure in Critically Ill Children

INTRODUCTION

Accurate assessment of resting energy expenditure (REE) can guide optimal nutritional prescription in critically ill children. Indirect calorimetry (IC) is the gold standard for REE measurement, but its use is limited. Alternatively, REE estimates by predictive equations/formulae are often inaccurate. Recently, predicting REE with artificial neural networks (ANN) was found to be accurate in healthy children. We aimed to investigate the role of ANN in predicting REE in critically ill children and to compare the accuracy with common equations/formulae.

STUDY METHODS

We enrolled 257 critically ill children. Nutritional status/vital signs/biochemical values were recorded. We used IC to measure REE. Commonly employed equations/formulae and the VCO₂-based Mehta equation were estimated. ANN analysis to predict REE was conducted, employing the TWIST system.

RESULTS

ANN considered demographic/anthropometric data to model REE. The predictive model was good (accuracy 75.6%; R² = 0.71) but not better than Talbot tables for weight. After adding vital signs/biochemical values, the model became superior to all equations/formulae (accuracy 82.3%, R² = 0.80) and comparable to the Mehta equation. Including IC-measured VCO₂ increased the accuracy to 89.6%, superior to the Mehta equation.

CONCLUSIONS

We described the accuracy of REE prediction using models that include demographic/anthropometric/clinical/metabolic variables. ANN may represent a reliable option for REE estimation, overcoming the inaccuracies of traditional predictive equations/formulae.

Personalized nutrition approach in pediatrics: a narrative review

Dietary habits represent the main determinant of health. Although extensive research has been conducted to modify unhealthy dietary behaviors across the lifespan, obesity and obesity-associated comorbidities are increasingly observed worldwide. Individually tailored interventions are nowadays considered a promising frontier for nutritional research. In this narrative review, the technologies of importance in a pediatric clinical setting are discussed. The first determinant of the dietary balance is represented by energy intakes matching individual needs. Most emerging studies highlight the opportunity to reconsider the widely used prediction equations of resting energy expenditure. Artificial Neural Network approaches may help to disentangle the role of single contributors to energy expenditure. Artificial intelligence is also useful in the prediction of the glycemic response, based on the individual microbiome. Other factors further concurring to define individually tailored nutritional needs are metabolomics and nutrigenomic. Since most available data come from studies in adult groups, new efforts should now be addressed to integrate all these aspects to develop comprehensive and-above all-effective interventions for children. IMPACT: Personalized dietary advice, specific to individuals, should be more effective in the prevention of chronic diseases than general recommendations about diet. Artificial Neural Networks algorithms are technologies of importance in a pediatric setting that may help practitioners to provide personalized nutrition. Other approaches to personalized nutrition, while promising in adults and for basic research, are still far from practical application in pediatrics.

Estimation of Resting Energy Expenditure Using Predictive Equations in Critically Ill Children: Results of a Systematic Review

Provision of adequate energy intake to critically ill children is associated with improved prognosis, but resting energy expenditure (REE) is rarely determined by indirect calorimetry (IC) due to practical constraints. Some studies have tested the validity of various predictive equations that are routinely used for this purpose, but no systematic evaluation has been made. Therefore, we performed a systematic review of the literature to assess predictive equations of REE in critically ill children. We systematically searched the literature for eligible studies, and then we extracted data and assigned a quality grade to each article according to guidelines of the Academy of Nutrition and Dietetics. Accuracy was defined as the percentage of predicted REE values to fall within ±10% or ±15% of the measured energy expenditure (MEE) values, computed based on individual participant data. Of the 993 identified studies, 22 studies testing 21 equations using 2326 IC measurements in 1102 children were included in this review. Only 6 equations were evaluated by at least 3 studies in critically ill children. No equation predicted REE within ±10% of MEE in >50% of observations. The Harris-Benedict equation overestimated REE in two-thirds of patients, whereas the Schofield equations and Talbot tables predicted REE within ±15% of MEE in approximately 50% of observations. In summary, the Schofield equations and Talbot tables were the least inaccurate of the predictive equations. We conclude that a new validated indirect calorimeter is urgently needed in the critically ill pediatric population.).