Objectives The estimation of energy requirements in critically ill children (CIC) currently relies on predictive formulae, which have been shown to be inaccurate in this population. Routine indirect calorimetry is a suitable alternative, but is technically challenging and costly. The aim of this study was to develop a new PICU-specific predictive equation.
Methods This study included all CIC with an endotracheal tube leak of <10% and fractional inspired oxygen of <60%. A 30 minute steady-state energy expenditure (EE) measurement was performed with the Deltatrac II. Polynomial regression analysis was used to establish the impact of diagnosis, day of admission, temperature, severity of disease and medication on EE and for the development of the new equation. This was followed by a validation study.
Results One hundred patients were enrolled. Day of admission (p = 0.976), temperature (p = 0.212) and severity of disease (p = 0.794) did not impact significantly on EE, however diagnosis did (p = 0.055). The new equation included weight, age and diagnosis and accounted for 83.3% of variation in energy expenditure (R2 = 0.833). The validation study (n = 25) indicated a mean difference of 16.7% between measured and predicted EE with no statistical difference (p = 0.433). Accuracy increased in patients <3 years (<10%).
Conclusion This new predictive equation estimates EE more accurately than previously published equations. However further research is required in older patients and those with cardiac and liver disease due to limited numbers recruited in this study.