Estimated vs measured energy expenditure in ventilated surgical-trauma critically ill patients

Background The American and European guidelines recommend measuring resting energy expenditure (REE) using indirect calorimetry (IC). Predictive equations (PEs) are used to estimate REE, but there is limited evidence for their use in critically ill patients. The aim of this study is to evaluate the degree of agreement and accuracy between IC-measured REE (REE-IC) and 10 different PEs in mechanically ventilated critically ill patients with surgical trauma who met their estimated energy requirement. Methods REE-IC was retrospectively compared with REE-PE by 10 PEs. The degree of agreement between REE-PE and REE-IC was analyzed by the Bland-Altman test (BAt) and the concordance correlation coefficient (CCC). The accuracy was calculated by the percentage of patients whose REE-PE values differ by up to +/- 10% in relation to REE-IC. All analyses were stratified by gender and body mass index (BMI; <25 vs >= 25). Results We analyzed 104 patients and the closest estimate to REE-IC was the modified Harris-Benedict equation (mHB) by the BAt with a mean difference of 49.2 overall (61.6 for males, 28.5 for females, 67.5 for BMI <25, and 42.5 for BMI >= 25). The overall CCC between the REE-IC and mHB was 0.652 (0.560 for males, 0.496 for females, 0.570 for BMI <25, and 0.598 for BMI >= 25). The mHB equation was the most accurate with an overall accuracy of 44.2%. Conclusion The effectiveness of PEs for estimating the REE of mechanically ventilated surgical-trauma critically ill patients with is limited. Nonetheless, of the 10 equations examined, the closest to REE-IC was the mHB equation.

Automated summary

This study evaluated the agreement and accuracy between indirect calorimetry and ten different predictive equations in mechanically ventilated surgical trauma critically ill patients. The results showed that the accuracy of the predictive equations was limited, but among the ten equations, the modified Harris-Benedict equation was the closest to indirect calorimetry.