Development of an instrument to indirectly monitor arterial pCO2 during cardiopulmonary bypass

Abstract

Arterial blood carbon dioxide tension (P_aCO₂) during cardiopulmonary bypass (CBP) is important to the conduct of perfusion with alpha-stat or pH-stat strategy. Temperature changes during CBP complicate any attempts to monitor carbon dioxide tension in the exhaust outlet of an oxygenator (P_exCO₂) because CO₂ becomes more soluble with decreasing temperatures. Normally, this would have been the obvious and easy choice of method to indirectly measure the patient’s P_aCO₂.

Several tests have been performed with ordinary capnographs modified to measure pCO₂ at the oxygenator exhaust gas port. These tests have shown varying degrees of precision (Br J Anaesth 1999; 82(6): 843-46; J Extra-Corpor Technol 2003; 35(3): 218-23; Br J Anaesth 2000; 84: 536; J Extra-Corpor Technol 1994; 26: 64-67). Some of the best results have been achieved by Potger et al. (J Extra-Corpor Technol 2003; 35(3): 218-23), who found a strong correlation between the arterial temperature-corrected P_exCO₂ when using a standard capnograph monitoring the P_aCO₂ measured from a blood gas analyser (P_bCO₂).

Our group has developed a new instrument, especially designed for oxygenator gas exhaust monitoring. The new instrument has automatic temperature correction, enabling it to show both original and corrected pCO₂ values, simultaneously. Ordinary capnograph functions, such as zeroing, flow control and calibration routines, are included. The solution consists of a pCO₂ sensor module, a temperature sensor, a water trap and a dedicated PC mounted on a heart-lung machine. Since the heart-lung machine was already equipped with a computer for data logging and a temperature sensor, only a box containing the pCO₂ sensor module and the water trap had to be added.

The PC uses a specially written program designed to collect data, make the necessary calculations and display the results on the computer screen. A temperature correction was developed based on a linear regression analysis for a data-set of 15 patients, assuming that the deviation between the measured P_exCO₂ from the oxygenator exhaust outlet and the P_bCO₂ from the blood gas analyser was linearly dependent on arterial temperature alone.

Eighty-six blood gas samples were compared to the corrected P_exCO₂ values. The final product displayed good qualities of stability and was accurate when temperature fluctuated from 32 to 388C, even during rewarming, which has been reported to be a problem for other P_exCO₂ investigations (J Extra-Corpor Technol 2003; 35(3): 218-23).