Calculated versus measured oxygen consumption during and after cardiac surgery. Is it possible to estimate lung oxygen consumption?

Background: Lung tissue is metabolically active and consumes oxygen. The oxygen content difference between arterial and mixed venous blood does not include the effect of pulmonary tissue oxygen uptake. Thus, oxygen consumption (VO2) of the lung should be reflected as a difference between VO2 measured by gas exchange and VO2 derived by the Fick principle. The purpose of this study was to measure in clinical conditions this difference (taken to represent the VO2 of the lung), and to evaluate the sources of error in lung VO2 estimation. Methods: Nine patients undergoing coronary artery bypass grafting were studied. VO2 was measured by indirect calorimetry (VO(2)gasex) and compared to Fick-derived VO2 (VO(2)Fick) after induction of anaesthesia, after closure of the chest, at admission to intensive care, after stabilization of haemodynamics and during weaning from mechanical ventilation. The Fick-derived VO2 was calculated from blood samples taken at the beginning and at the end of each 20 min measurement period, and the mean of 12 consecutive thermodilution cardiac output measurements taken during each 20 min measurement period. Results: VO(2)gasex was higher than VO(2)Fick (P<0.01; in all except 4 of 45 measurements). The difference between the measured and the calculated VO2 was 33+/-25 ml/min (mean+/-SD, range -16-100 ml/min). This difference represented 14+/-3% (range 11-18%) of the whole body VO2. The VO2-difference was highest after the induction of anaesthesia (50+/-19 ml/min; range 20-81 ml/min, P<0.03) and lowest on arrival at the intensive care unit (10+/-16 ml/min; range -16-39 ml/min). Core temperature did not correlate with the oxygen consumption difference. Conclusions: A constant difference between measured and calculated VO2 can be detected in carefully controlled clinical conditions. The difference between the two methods is due to both lung oxygen consumption and errors in the measurement of VO2, thermodilution cardiac output, haemoglobin and blood oxygen contents. We suggest that the perioperative changes of the VO2-difference are due not only to variation of the measurements but also to changes in lung metabolic activity.