Objective: To quantify resting and total energy expenditure in patients who have suffered severe trauma and sepsis. Design: Prospective, unblinded, observational, nonrandomized study. Setting: Critical care unit of a Level I adult trauma center. Patients: Immediate posttrauma patients or trauma patients exhibiting signs of sepsis with multiple organ dysfunction. Interventions: An indirect calorimeter was used to measure energy expenditure at rest (resting energy expenditure) at 0700 and 1900 hrs. The energy expenditure measurement was then continued for up to 12 hrs (total energy expenditure). Clinical data were collected for computation of an illness severity score. Results: Thirteen trauma and 20 septic patients were studied 240 times. All patients were mechanically ventilated. Morphine or fentanyl was infused during 99% of studies. Neuromuscular blocking agents were used in 42% of septic studies. Both the trauma and septic groups were hypermetabolic (mean trauma resting energy expenditure, 36 ± 6 kcal/kg; mean septic resting energy expenditure, 44 ± 8 kcal/kg; p < .05). Total energy expenditure was similar to resting energy expenditure (trauma total energy expenditure = resting energy expenditure × 1.035 ± 0.078, septic total energy expenditure = resting energy expenditure × 1.039 ± 0.071). Total energy expenditure and resting energy expenditure were linearly related (r2 = .89, p < .0001). Conclusions: Trauma and septic patients are hypermetabolic, even when heavily sedated or medically paralyzed. A measurement of resting energy expenditure is a close approximation of total energy expenditure in most patients. (Crit Care Med 1994; 22:1796–1804)