Objective: The purpose of this study was to evaluate current methods of predicting energy requirements and to develop and validate new equations derived from energy expenditure measurements of ventilated, critically ill children.
Design: Prospective, observational, sequential study.
Setting: Pediatric intensive care unit.
Patients: A total of 100 ventilated, critically ill children who fit the criteria of energy expenditure measurement. Additional patients (n = 25) were included in the validation study.
Intervention: An indirect calorimeter was used to measure energy expenditure for a period of 30 mins.
Measurements and main results: The mean measured energy expenditure was 185+/-51 kJ/kg per day. Predicted energy expenditure from standard equations was compared with measured energy expenditure by using the Bland and Altman "methods comparison procedure," and poor precision and accuracy were observed. Patient variables were collected at the time of measurement, and multiple regression analysis was performed to determine the independent contribution of each variable to measure energy expenditure. New predictive equations were formulated and validated with additional energy expenditure measurements. Patient variables that did not correlate significantly with energy expenditure were gender, Pediatric Risk of Mortality score, and commencement of nutritional support. An equation was derived from patient variables (age, weight, weight for age Z score, body temperature, number of days after intensive care admission, and primary reason for admission) that correlated significantly (r2 = .898) with measured energy expenditure. A second, simplified equation (energy expenditure kJ/day = ¿17 x age [months]¿ + ¿48 x weight [kg]¿ + ¿292 x body temperature degrees C¿ - 9,677) was produced (r2 = .867). Validation found no significant difference between measured and predicted energy expenditure by the new equations; however, the equations did not predict accurately for patients <2 months of age.
Conclusion: The new equations provide a more accurate alternative to current predictive methods in assessing energy requirements of ventilated, critically ill children.