Clinical implications of continuous measurement of energy expenditure in mechanically ventilated patients.

Energy expenditure was monitored in 20 critically-ill mechanically ventilated patients using the Siemens-Elema Oxygen Consumption Calculator (OCC 980). Energy expenditure was measured continuously over the 24-h period in all patients (altogether, over 2500 patient hours; range 48-288 h). A predicted energy expenditure was calculated for each patient from standard tables for basal metabolic rates modified according to previously published reports on the influence of trauma, infection and elevated body temperature. For all patients combined, the agreement between the predicted and the measured energy expenditure was good. However, in individual patients the measured energy expenditure varied between 48 and 148% of the predicted value. The measured energy expenditure in surviving traumatized and/or septic patients correlated well (95-100%) with the predicted value at the time when weaning off the ventilator could be initiated. On the first day of measurements, the energy expenditure (in % of the predicted value) in the six patients who later died was significantly lower than in surviving patients (84 +/- 6 vs 107 +/- 2%; p < 0.01). Over a 24-h period, energy expenditure, defined as the value noted during a stable 30-40-min period of measurement, varied between 12 and 50% in the individual patients. This study shows that energy expenditure cannot be accurately predicted in the individual patient, that an energy expenditure below predicted values appears to be indicative of a poor prognosis and that short periods of energy expenditure monitoring may fail to reflect 24-h conditions.