Resting Metabolic Rate of the Critically Ill Patient: Measured versus Predicted

Critically ill patients requiring mechanical ventilation are particularly susceptible to malnutrition. A knowledge of the energy requirements of these patients is essential in designing nutritional regimens. This study examines 45 resting energy-expenditure measurements performed in a group (n = 40) of postoperative, critically ill patients who were hemodynamically stable, noncomatose, and receiving mechanical ventilation. It examines in particular to what degree the resting energy expenditure of such patients can be predicted using the Harris-Benedict and Aub-Dubois formulae. Resting energy expenditure was measured using indirect calorimetry. There was only a moderate correlation between measured resting energy expenditure and that predicted using the Harris-Benedict (r = 0.57) and Aub-Dubois (r = 0.59) formulae. There was little correlation between the ratio of the measured to the predicted (Harris-Benedict) resting energy expenditure and age, or the ratio of actual to ideal body weight and body weight. The measured resting energy expenditure differed widely (70-140%) from predicted, reflecting the many complex factors that influence these patients' metabolic rate. The role of standard predictive formulae in such patients is as an arbitrary reference point to be used to define hypermetabolism (measured greater than predicted) and hypometabolism (predicted greater than measured).