Ultrasensitive Exhaled CO2 Detection Provides Accurate, Noninvasive Estimates of PaCO2 During High Frequency Oscillatory Ventilation in Experimental Acute Lung Injury. Tuckosh J, Cox T, Emberger J, et al. A. I. DuPont Hospital For Children, Wilmington, DE.

This study used the COSMO Respiratory Profile Monitor (Novometrics) to continuously measure exhaled CO2 (ExCO2) as an indicator of PaCO2 during HFOV in an animal model of acute lung injury. Five piglets (mean weight 7.8 kg) were studied during transitions from CMV to HFOV at 5, 10, and 15 Hz frequencies before and after acute lung injury was induced by oleic acid. PaCO2 and ExCO2 were measured at the set frequency, during a one second pause in ventilation, and during a 5-second switch to a frequency of 3 Hz. COSMO continuous ExCO2 detection correlated best with PaCO2 when the technique of briefly switching to a 3 Hz frequency during the measurement was employed. The correlation coefficients using this strategy in healthy lungs (p<0.05) were: 5 Hz r=0.91; 10 Hz r=0.87; 15 Hz r=0.97; and for injured lungs (p<0.05): 5 HZ r=0.89; 10Hz r=0.80; 15 Hz r=0.85.

Comment. This device as an ultrasensitive monitor that utilizes a single-beam mainstream sensor to measure CO2 concentration in exhaled gas by infrared absorption. The device appears to be more sensitive than most end-tidal CO2 monitors. Despite this sensitivity, the correlation to PaCO2 during HFOV was only fair unless respiratory pauses or low-frequency (3 Hz) switches were used to improve the measurement. While this seemed to work fairly well in an 8 kg piglet, I can’t help but wonder whether the correlation would be nearly as good in a 1 kg preemie. Also, I fail to see a significant advantage of this technique over transcutaneous pCO2 monitoring during HFOV.