[The value of capnography and exhaled CO2 in neonatal intensive care units].

OBJECTIVE: The objective of this study was to investigate the reliability of end-tidal CO2 (PetCO2) as a non-invasive guide of PaCO2 in the newborn and to analyze the influence of the relationship between ventilation-perfusion in the correlation between both determinations.

PATIENTS AND METHODS: End-tidal CO2 (PetCO2) was monitored by capnography in 9 ventilated newborns: 146 arterial blood gas samples were drawn and the results were compared with the PetCO2 values obtained. The gradient or difference between PaCO2 and PetCO2 was calculated to determine the correlation. The ratio a/AO2 was used as an indirect indicator of the ventilation/perfusion relationship (V/Q ratio).

RESULTS: The mean gestational age was 30.9 +/- 2.8 weeks and birth weight 1,648 +/- 596 g. The age at the beginning of the study was 2 +/- 1.5 days. The diagnoses corresponded to 5 cases of RDS (56%), 2 cases of wet lung syndrome (22%), 1 case of pneumonia (11%) and 1 pneumothorax (11%). The results of this monitoring were classified in function of the a/AO2 ratio obtained: Group A, a/AO2 < 0.2 and PaCO2-PetCO2 gradient = 13.3 +/- 5; Group B, a/AO2 = 0.2-0.29 and PaCO2-PetCO2 gradient = 8 +/- 2.7; and Group C, a/AO2 > 0.29 and PaCO2-PetCO2 gradient = 2 +/- 1.7. The results show a very good correlation from a a/AO2 ratio > or = 0.3 onwards. The a/AO2 ratio is the major determinant of PaCO2-PetCO2 differences and respiratory frequency has less influence.

CONCLUSIONS: 1) Monitoring of end-tidal CO2 does not maintain a good correlation with PaCO2 in serious lung illness. 2) End-tidal CO2 measurement is an effective and accurate technique for the monitoring of newborns when the a/AO2 ratio > or = 0.3 and it can be useful for weaning of mechanical ventilation. 3) PaCO2-PetCO2 differences accurately show the changes in the ventilation-perfusion relationship.