Numerous protocols are used for indirect calorimetry in research and clinical settings. The objective of the current study was to validate in critically ill patients an abbreviated protocol that uses five consecutive stable 1-minute readings of oxygen consumption (VO2), carbon dioxide production (VCO2), and minute ventilation (VE) in a range of +/− 5%, versus a more standard protocol that uses 30 consecutive stable one minute readings of VO2, VCO2, and VE in a range of +/− 10%. Indirect calorimetry was performed on resting, mechanically ventilated, critically ill patients. The first 5-minute period in which coefficients of variation for VO2, VCO2, and VE were < or = 5% was compared to the first 30-minute period in which coefficients of variation for these variables were < or = 10%. Thirty-four critically ill patients were studied. Twenty four patients (70%) successfully completed both protocols (Success Group). Eighteen percent of subjects completed neither the abbreviated nor the 30 minute protocol, and 12% completed only one protocol (Fail Group). The Success Group was marked by a higher incidence of sedation and/or medical paralysis. There were no significant differences in VE, VO2, VCO2, respiratory quotient, or energy expenditure between the protocols in the Success Group or the Fail Group. Coefficients of determination (R2) for VO2 and VCO2 between the two methods in the Success Group were 0.99. In the Fail Group, R2 values ranged from 0.75 for VCO2 to 0.91 for VO2. In sedated, mechanically ventilated patients, an indirect calorimetry test of five consecutive 1-minute periods with coefficient of variation < or = 5% is equivalent to a longer test consisting of 30 consecutive 1-minute periods with coefficient of variation < or = 10%.