Objective:To explore the accuracy of a continuous transcutaneous CO2(TcCO2) monitor, used in an intermittent rather than a continuous fashion, to obtain quick (<5 mins) CO2readings.Design:Prospective study.Setting:An urban pediatric intensive care unit in a university teaching hospital.Patients:A convenience sample of pediatric patients with indwelling arterial catheters.Intervention:Transcutaneous monitoring was done simultaneous with arterial blood gas monitoring.Measurements and Main Results:There were 49 simultaneous-readings on 19 patients, age 5 days to 16 years, with 13 different diagnoses. The TcCO2was related to the PCO2by a Pearson product coefficient of 0.79 (p< .0005), with a mean difference of 1.94 (TcCO2> PCO2) and 95% confidence interval of −0.12 to 4.07. The scatterplot produces a regression line characterized by the following equation: PCO2= (TcCO2× 1.05) − 4.08.Conclusions:Further study to evaluate intermittent TcCO2as a practical clinical variable is warranted. This study should encourage refinement of the technology to be more accurate for this use.