Oxygen consumption (̇Vo2) measured by indirect calorimetry (Nellcor-Puritan-Bennett 7250; Carlsbad, CA, USA) has been compared with ̇Vo2 calculated by the Fick method in 22 volume-controlled ventilated general surgical patients in the early post-operative period. For 198 pairs of measurements, ̇Vo2 Fick and ̇Vo2 indirect calorimetry correlated significantly (y=1.00x–35.8, P=0.0001, r=0.77). ̇Vo2 indirect calorimetry was 212±32 mL min−1 and ̇Vo2 Fick was 177±41 mL min−1 (P=0.0001). The bias was 35±26 mL min−1. This difference represents 16±13% of the total body ̇Vo2. ̇Vo2 calculated by the Fick method did not accurately predict ̇Vo2 measured by indirect calorimetry, and the two methods were not interchangeable. ̇Vo2 calculated by the Fick method underestimated ̇Vo2 as measured by indirect calorimetry by a systematic quantity that could be attributed, in part, to ̇Vo2 of the lung. Indirect calorimetry should be the preferred method for measuring total body ̇Vo2 in mechanically ventilated surgical patients.