Background

Arterial blood gas analysis is the gold standard for monitoring of PaCO2 and PaO2 during mechanical ventilation. However, continuous measurements would be preferred. Transcutaneous sensors continuously measure blood gases diffusing from the locally heated skin. These sensors have been validated in children mostly in intensive care settings. Accuracy in children during general anesthesia is largely unknown.

Aims

We conducted a study in children undergoing general anesthesia to validate the use and to determine the accuracy of continuous transcutaneous measurements of the partial pressures of PCO2 (tcPCO2) and PO2 (tcPO2).

Methods

A prospective observational study in a tertiary care pediatric hospital in the Netherlands, from April to October 2018, in children aged 0-18 years undergoing general anesthesia. Patients were included when endotracheally intubated and provided with an arterial catheter for regular blood sampling. Patients with a gestational age <31 weeks, burn victims and patients with skin disease were excluded. TcPCO2 and tcPO2 measurements were performed with a SenTec OxiVenT™ sensor (SenTec AG, Therwil, Switzerland). Accuracy was determined with an agreement analysis between arterial and transcutaneous PCO2 and PO2 values, and between arterial and end-tidal PCO2 (etCO2) values, according to Bland and Altman, accounting for multiple measurements per subject.

Results

We included 53 patients (median age 4.1 years, IQR 0.7 to 14.4 years) and retrieved 175 samples. TcPCO2-PaCO2 agreement analysis provided a bias of 0.06 kPa (limits of agreement (LOA) -1.18 to 1.31), the etCO2-PaCO2 agreement showed a bias of -0.31 kPa (LOA -1.38 to 0.76). Results of the tcPO2-PaO2 agreement showed a bias of 3.40 to 0.86*(mean tension) kPa.

Conclusions

This study showed good agreement between PaCO2 and tcPCO2 in children of all ages during general anesthesia. Both transcutaneous and end-tidal CO2 measurements showed good accuracy. TcPO2 is only accurate under 6 months of age.