Oxygen consumption was measured in infants, children, and adolescents during diagnostic heart catheterizations. A total of 825 measurements of oxygen consumption (VO2) was performed in 504 subjects using a semiopen hood system and a paramagnetic oxygen analyzer. In 256 subjects under 3 years of age, body dimensions and heart rate were found to be significant factors for oxygen consumption. The regression equation for both sexes was: VO2/BSA (ml/min.m2) = 3.42.height (cm) - 7.83.weight (kg) + 0.38.HR - 54.1 (r2 = 0.39, SD = 38.7), where BSA is body surface area and HR is heart rate. VO2/BSA was significantly lower in infants less than 3 months of age (133 +/- 33 ml/min.m2) compared with infants of 3-12 months (171 +/- 37 ml/ min.m2; p < 0.01). In 272 children aged 3 years and older and adolescents, gender was a significant factor in oxygen consumption together with BSA and HR. The regression line equation for males was VO2/BSA (ml/ min.m2) = 0.79.HR - 7.4.BSA(m2) + 108.1 (r2 = 0.45, SD = 34.2). The regression line equation for females is VO2/BSA (ml/min.m2) = 0.77.HR - 5.2.BSA(m2) + 106.8 (r2 = 0.43, SD = 34.4). Hematocrit, systemic oxygen saturation, and blood pressure were not significant factors. The predictive value of nomograms for oxygen consumption is limited because of the large interindividual variations not explained by differences in gender, body size, or simple hemodynamic variables. Preferably, oxygen consumption is measured; but if nomograms for oxygen consumption are used for hemodynamic assessment, the wide confidence intervals should be considered.