Objective—To compare the ability of a sidestream
capnograph and a mainstream capnograph to measure
end-tidal CO2 (ETCO2) and provide accurate estimates
of PaCO2 in mechanically ventilated dogs.
Design—Randomized, double Latin square.
Animals—6 healthy adult dogs.
Procedure—Anesthesia was induced and neuromuscular
blockade achieved by IV administration of pancuronium
bromide. Mechanical ventilation was used
to induce conditions of standard ventilation, hyperventilation,
and hypoventilation. While tidal volume was
held constant, changes in minute volume ventilation
and PaCO2 were made by changing the respiratory
rate. Arterial blood gas analysis was performed and
ETCO2 measurements were obtained by use of either
a mainstream or a sidestream capnographic analyzer.
Results—A linear regression model and bias analysis
were used to compare PaCO2 and ETCO2 measurements;
ETCO2 measurements obtained by both
capnographs correlated well with PaCO2. Compared
with PaCO2, mainstream ETCO2 values differed by
3.15 ± 4.89 mm Hg (mean bias ± SD), whereas the
bias observed with the sidestream ETCO2 system
was significantly higher (5.65 ± 5.57 mm Hg).
Regardless of the device used to measure ETCO2,
bias increased as PaCO2 exceeded 60 mm Hg.
Conclusions and Clinical Relevance—Although the
mainstream capnograph was slightly more accurate,
both methods of ETCO2 measurement correlated
well with PaCO2 and reflected changes in the ventilatory
status. However, ETCO2 values > 45 mm Hg may
inaccurately reflect the severity of hypoventilation as
PaCO2 may be underestimated during conditions of
hypercapnia (PaCO2 > 60 mm Hg). (J Am Vet Med Assoc 2002;221:1582–1585)