Infrared monitoring of expired carbon dioxide tension during anesthesia with and without nitrous oxide

Recently, the utilization of infrared analysis of expired gas to monitor end-tidal carbon dioxide partial pressure (PETCO 2 ) during anesthesia has gained in popularity. In contrast to arterial blood gas analysis, PETCO 2 analysis is immediate and continuous. Mass spectrometry has also been used to measure PETCO, and has been found to correlate well with arterial carbon dioxide partial pressure (PaCO,). 1 Other expired gases delivered during anesthesia, particularly nitrous oxide (N20), may affect the accuracy of infrared measurement. Because N2O is used as part of virtually every general anesthetic, it is of clinical importance to determine what effect it has on the value of infrared-determined PETCO2 relative to PaCO,. The study: methods Fifteen adult patients scheduled for routine nonthoracic surgical procedures gave their informed written consent for a study to determine the effects of N0O on infrared CO2 measurement. There were 10 males, aged 21 to 53 years (mean age 35), and 5 females, aged 30 to 47 years (mean age 36). All subjects were ASA physical status 1 or 2 and free of documented pulmonary disease. Following induction of general anesthesia and intubation, ventilation was either controlled or assisted. After isoflurane in 60% N,O and 40% oxygen was breathed from a partial rebreathing circle carbon dioxide absorption system (total flow 5 L/min) for no less than 30 minutes, a 3 ml heparinized blood sample was obtained anaerobically from the radial artery. Simultaneously, PETCO2 was measured with a Cavitron model PM20R infrared carbon dioxide analyzer. A single instrument was used for all studies. Details of calibration were as follows: After a 30 minute warm up period, the sampling flow was adjusted to 150 ml/min. "Zero" was set using ambient air. A calibrating gas consisting of 5.30% carbon dioxide, 12.63% oxygen and 82.07% nitrogen was then sampled and the span adjusted to 40 mmHg. A barometric pressure of 755 mmHg was assumed in all cases. Calibration and waveform signals were recorded in mmHg on a strip-