Mixed venous oxygen saturation (SvO2) is important when evaluating the balance between oxygen delivery and whole-body oxygen consumption. Monitoring SvO2 has so far required blood samples from a pulmonary artery catheter. By combining volumetric capnography, for measurement of effective pulmonary blood flow, with the Fick principle for oxygen consumption, we have developed a continuous noninvasive method, capnodynamic SvO2, for assessment of SvO2. The objective of this study was to validate this new technique against the gold standard cardiac output (CO)-oximetry SvO2 measurement of blood samples obtained from a pulmonary artery catheter and to assess the potential influence of intrapulmonary shunting.
Eight anesthetized mechanically ventilated domestic-breed piglets of both sexes (median weight 23.9 kg) were exposed to a series of interventions intended to reduce as well as increase SvO2. Simultaneous recordings of capnodynamic and CO-oximetry SvO2 as well as shunt fraction, using the Berggren formula, were performed throughout the protocol. Agreement of absolute values for capnodynamic and CO-oximetry SvO2 and the ability for capnodynamic SvO2 to detect change were assessed using Bland-Altman plot and concordance analysis.
Overall bias for capnodynamic versus CO-oximetry SvO2 was −1 percentage point (limits of agreement −13 to +11 percentage points), a mean percentage error of 22%, and a concordance rate of 100%. Shunt fraction varied between 13% at baseline and 22% at the end of the study and was associated with only minor alterations in agreement between the tested methods.
In the current experimental setting, capnodynamic assessment of SvO2 generates absolute values very close to the reference method CO-oximetry and is associated with 100% trending ability.