Assumed oxygen consumption frequently results in large errors in the determination of cardiac output.

Abstract

We sought to investigate the differences in assumed and measured oxygen consumption values for the determination of cardiac output by using the Fick principle in a pediatric population with congenital heart disease. The patient population consisted of 143 patients with a mean age of 11.3 years (age range, 2 days to 23.8 years) undergoing cardiac catheterization during general anesthesia and with mechanical ventilation. Oxygen consumption was measured with a standard commercial analyzing system (Deltatrac II; Datex, Engström, Helsinki, Finland). Assumed oxygen consumption values were calculated according to the formulas of Krovetz and Goldbloom and LaFarge and Miettinen. Comparisons between measurements and assumptions were performed by Bland-Altman plots. Two-sided paired t tests were used to assess a difference of the assumed and measured values. The range of measured oxygen consumption values was between 55.2 and 249 mL . min -1 . m -2 . The Krovetz-Goldbloom formula led to systematically larger values compared with the measured values (P = .0001; mean difference of -53.3 mL . min -1 . m -2 ; 95% confidence interval, -56.7 to -49.8 mL . min -1 . m -2 ). The use of the LaFarge-Miettinen formula tends to overestimate oxygen consumption (P = .0037; mean difference of -15.9 mL . min -1 . m -2 ; 95% confidence interval, -26.5 to -5.4 mL . min -1 . m -2 ). A similarly poor agreement was found when analyzing a subgroup of 25 patients with Fontan-type circulation. The use of assumed instead of measured oxygen consumption values introduces large errors in the determination of cardiac output.