A Validation of Equations for Predicting the Energy Expenditures of Hospitalized Patients

Abstract

Indirect calorimetry most accurately determines energy expenditures (EE) for the purpose of establishing nutrient requirements in hospitalized patients. However, the high costs associated with clinically using indirect calorimetry have renewed interest in predictive energy expenditure equations. We compared measured energy expenditures (MEEs) to those calculated by two commonly used equations: the Harris Benedict (HB) with a disease related adjustment factor of 1.2 (HB1.2) and the Ireton-Jones (IJ) in 120 patients (53 female/67 male). Patients were categorized according to the following diagnoses: burn (18%), non-burn trauma (30%), and non-trauma/non-burn (52%). Forty-one percent of the patients were ventilator dependent (VD). Overall there was no significant difference between the HB1.2 and the MEE. The HB1.2 correlated significantly with the MEE (r = 0.643, p= < 0.001). There was no significant difference between the LI and the MEE and the two correlated significantly (r = 0.735, p= < 0.001). When evaluating EE by subgroups, the HB1.2 significantly underestimated the MEE of both burn and VD patients (p = 0.01), while the IJ was not significantly different from the MEE. The correlation of the IJ with the MEE was higher than that found with the HB1.2. The IJ accurately predicts the EE of hospitalized patients with various diagnoses.