The use of uncuffed endotracheal tubes (ETT) in pediatric patients raises concern over the accuracy of indirect calorimetry measurement in the presence of a gas leak around an ETT. We examined the effects of ETT gas leak on respiratory gas measurements in a dog model. Mongrel dogs (n=12) were sedated, paralyzed, intubated, and placed on mechanical ventilation. Leak was achieved by adjusting cuff volume. Oxygen consumption (VO2), CO2 production (VCO2), respiratory exchange ratio (RER), and resting energy expenditure (REE) were measured at each leak pressure (Pleak). Peak inspiratory pressure (PIP), Pleak, inspiratory and expiratory tidal volume (VTinsp, VTexp), VE, end tidal CO2 (ETCO2), and blood gases were recorded at each leak pressure. VO2, VCO2, and REE decreased significantly with increasing gas leak. There was a linear relationship between VO2, VCO2, and REE with both TVratio (VTexp/VTinsp) and Pdiff (PIP-Pleak). Multiple regression equations based on TVratio and Pdiff were obtained which allowed correction of the measurement error in VO2, VCO2, and REE, with correlation coefficients (R 2) of 0.71, 0.75, and 0.73, respectively. ETT gas leak affects measurements of VO2, VCO2, and REE, but not RER. Measurements made with TVratio > 0.80 required no correction. Measurements made with TVratio > 0.45 could be corrected to actual values in our model with regression equations based on TVratio and Pdiff. We conclude that indirect calorimetry measurements can be useful in the presence of tracheal gas leak around an ETT. © 1990, Sage Publications. All rights reserved.
