Validation of transcutaneous carbon dioxide monitoring using an artificial lung during adult pulsatile cardiopulmonary bypass

Abstract

Measurements of transcutaneous carbon dioxide (tcCO₂) have been used in multiple venues, such as during procedures utilizing jet ventilation, hyperbaric oxygen therapy, as well as both the adult and neo-natal ICUs. However, tcCO₂ measurements have not been validated under conditions which utilize an artificial lung, such cardiopulmonary bypass (CPB). The purpose of this study was to (1) validate the use of tcCO₂ using an artificial lung during CPB and (2) identify a location for the sensor that would optimize estimation of PaCO₂ when compared to the gold standard of blood gas analysis.

tcCO₂ measurements (N = 185) were collected every 30 min during 54 pulsatile CPB procedures. The agreement/differences between the tcCO₂ and the PaCO₂ were compared by three sensor locations. Compared to the earlobe or the forehead, the submandibular PtcCO₂ values agreed best with the PaCO₂ and with a median difference of –.03 mmHg (IQR = 5.4, p < 0.001). The small median difference and acceptable IQR support the validity of the tcCO₂ measurement. The multiple linear regression model for predicting the agreement between the submandibular tcCO₂ and PaCO₂ included the SvO₂, the oxygenator gas to blood flow ratio, and the native perfusion index (R² = 0.699, df = 1, 60; F = 19.1, p < 0.001).

Our experience in utilizing tcCO₂ during CPB has demonstrated accuracy in estimating PaCO₂ when compared to the gold standard arterial blood gas analysis, even during CO₂ flooding of the surgical field.