Comparison of pediatric end-tidal CO2 measured with nasal/oral cannula circuit and capillary PCO2

This study was designed to determine whether end-tidal carbon dioxide (ETCO2) values obtained by noninvasive oral/nasal cannula circuit with side-stream capnometry correlate reliably with capillary PCO2 (CapCO2) in a pediatric population without cardiopulmonary problems. Each patient was monitored until a reliable 5-minute ETCO2 waveform was obtained. A capillary blood gas sample was drawn while, simultaneously, ETCO2 was recorded. The difference between CapCO2 and ETCO2 levels was tested with a paired t-test at P < .001. The limits of agreement were established with a 95% confidence level. The stability of the measured difference across the range of mean scores (CapCO2 + ETCO2/2), age, and respiratory rate was tested using simple linear regression. Fifty-eight children (23 girls and 35 boys) had mean ETCO2 readings of 33.96 mm Hg (SD 4.26), and mean CapCO2 readings of 35.93 (SD 4.04). A relative average bias of 1.96 with ETCO2 lower than CapCO2 was established with 95% limits of agreement of +/- 5.2 mm Hg (t = 5.71). Variability of difference scores was not related to range of mean scores (r = .08), age (r = .09), or respiratory rate (r = .25). End-tidal CO2 measured by an oral/nasal cannula capnometry circuit is a noninvasive method of assessing indirect measurements of PCO2 in a normal pediatric population.