INFRARED END-TIDAL CO2 MEASUREMENT DOES NOT ACCURATELY PREDICT ARTERIAL CO2 VALUES OR END-TIDAL TO ARTERIAL P-CO2 GRADIENTS IN RABBITS WITHLUNG INJURY

End-tidal P-CO2 (P-ETCO2) measurements from two commercially available neonatal infrared capnometers with different sampling systems and a m ass spectrometer were compared with arterial P-CO2 (P-aCO2) to determi ne whether the former could predict the latter in mechanically ventila ted rabbits with and without lung injury. The effects of tidal volume, ventilator frequency and type of lung injury on the gradient between P-ETCO2 and P-aCO2 (Delta P((a-ET)CO2)) were evaluated. Twenty rabbits were studied: 10 without lung injury, 5 with saline lavage and 5 with lung injury by meconium instillation. Paired measurements of P-ETCO2 by two infrared capnometers and a mass spectrometer were compared to P -aCO2. In the rabbits without lung injury, the values from the infrare d capnometers and mass spectrometer correlated strongly with P-aCO2 (r greater than or equal to 0.91) despite differences in the slopes of t he linear regression between P-ETCO2 and P-aCO2 and in Delta P((a-ET)C O2) (P < 0.05). Values from the mainstream IR-capnometer more closely approximated the line of identity than the regression between the side stream IR-capnometer values or the mass spectrometer and P-aCO2, but t ended to overestimate, P-aCO2. The Delta P((a-ET)CO2) was similar at a ll tidal volumes and ventilator frequencies, regardless of capnometer type. In the rabbits with induced lung injury, while there was a posit ive correlation between the slopes of the regression between P-ETCO2 a nd P-aCO2 for both capnometers (r greater than or equal to 0.70), none of the regression slopes approximated the line of identity. The Delta P((a-ET)CO2) was greater in (a-ET)co, was greater in rabbits with inj ured than noninjured lungs (P < 0.05). The Delta P((a-ET)CO2) was simi lar among capnometers regardless of tidal volume, ventilator frequency , or type of lung injury. The 95% confidence interval of plots P-aCO2 against P-ETCO2 was large for rabbits with injured and noninjured lung s. The Delta P((a-ET)CO2) changed In an erratic manner with changing P -aCO2. We conclude that P-ETCO2 measurements by infrared capnometer or mass spectrometer does not accurately predict P-aCO2 or Delta P((a-ET )CO2) in the presence of lung injury, and should be used with caution in the management of critically ill infants. (C) 1994 Wiley-Liss, Inc.