Predictive energy equations are inaccurate for determining energy expenditure in adult burn injury: a retrospective observational study

Background: Severe burn injuries are associated with hypermetabolism. This study aimed to compare the measured energy expenditure (mEE) with predicted energy requirements (pERs), and to correlate energy expenditure (EE) with clinical parameters in adults with severe burn injury. Methods: Data were retrospectively analysed on 29 burn patients (median (interquartile range) age: 46 (28–61) years, % total body surface area burn: 37% (18–46%)) admitted to an intensive care unit. Indirect calorimetry was performed on 1–4 occasions per patient to measure EE. mEE was compared with pER calculated using four prediction equations. Bland–Altman and correlation analyses were performed. Results: Mean ± SD mEE was 9752 ± 2089 kJ/day (143 ± 32% of predicted basal metabolic rate). Bland–Altman analysis demonstrated clinically important overestimation for three of the four prediction equations and wide 95% limits of agreement for all equations. Overestimation of EE was more marked early post-burn. mEE correlated with day post-burn (r = 0.42, P = 0.004) and number of operations prior to first EE measurement (r = 0.34, P = 0.016), but not with % total body surface area (r = 0.02, P = 0.9). Conclusions: Patients with severe burn injury exhibit hypermetabolism. The observed poor agreement between pER and mEE at an individual level indicates the value of indirect calorimetry in determining EE in burn injury.