Capnography in newborns under mechanical ventilation and its relationship with the measurement of CO2 in blood samples

Abstract

IntroductionMonitoring the partial pressure of CO2 (PCO2) in newborns who require ventilation would allow avoiding hypocapnia and hypercapnia. The measurement of end-tidal carbon dioxide (ETCO2) is an alternative rarely implemented in this population. ObjectiveTo evaluate the relationship between ETCO2 and PCO2 in newborns. MethodsCross-sectional study comparing two PCO2 measurement methods, the conventional one by analysis of blood samples and the one estimated by ETCO2. The study included hospitalized newborns that required conventional mechanical ventilation. The ETCO2 was measured with a Tecme GraphNet® neo, a neonatal ventilator with an integrated capnograph, and we obtained the ETCO2–PCO2 gradient. We conducted correlation and Bland-Altman plot analyses to estimate the agreement. ResultsA total of 277 samples (ETCO2 / PCO2) from 83 newborns were analyzed. The mean values ​​of ETCO2 and PCO2 were 41.36mmHg and 42.04mmHg. There was a positive and significant correlation between ETCO2 and PCO2 in the overall analysis (r=0.5402; P<.001) and in the analysis of each unit (P<.001). The mean difference was 0.68 mmHg (95% CI, –0.68 to 1.95) and was not significant. We observed a positive systematic error (PCO2 > ETCO2) in 2 of the units, and a negative difference in the third (PCO2 < ETCO2). DiscussionThe correlation between ETCO and PCO2 was significant, although the obtained values ​​were not equivalent, with differences ranging from 0.1mmHg and 20mmHg. Likewise, we found systematic errors that differed in sign (positive or negative) between institutions.