# 2009 OPEN FORUM Abstracts

### EVALUATION OF INVERSE FICK CALCULATION FOR SVO2 ESTIMATION

#### Lara Brewer, Matthias Goerges, Joseph Orr; Anesthesiology, University of Utah Health Sciences Center, Salt Lake City, UT

Background: Mixed venous oxygen saturation (SvO2) measurement can be used to assess the global oxygen supply-demand relationship. Direct SvO2 measurement requires either placement of a fiber optic central venous catheter or central venous blood gas measurement. SvO2 can be calculated using the inverse Fick equation if oxygen uptake (VO2), cardiac output (Q) hemoglobin (Hb), and arterial oxygen saturation (SaO2) are known. We used the data provided by a noninvasive cardiac output computer, which utilizes the partial Fick method, as the basis for SvO2 calculation. We compared the estimated SvO2 to the directly measured SvO2 in anesthetized pigs. Method: SvO2 is calculated by the inverse Fick equation as SaO2 less VO2 divided by the product of Q and arterial oxygen content (CaO2). The NM3 monitor (Philips, Wallingford, CT) measures CO2 excretion (VCO2), arterial oxygen saturation via pulse oximetry (SpO2) and Q. Assuming a respiratory quotient of 0.85, the VO2 was calculated from directly measured VCO2. CaO2 was calculated from SpO2 and Hb. Five pigs (30-40 Kg) were intubated and anesthetized with 1 MAC of isoflurane. A pulmonary artery catheter (Edwards Lifesciences, Irvine, CA) was placed to obtain mixed venous blood samples. Arterial blood samples were simultaneously drawn. The NM3 monitor sensor was placed in the breathing circuit between the endotracheal tube and the wye. Q, VCO2, SpO2 and other parameters measured by the NM3 monitor were saved to a computer so that SvO2 could be calculated off line by the inverse Fick method. Continuous infusions of dobutamine or norepinephrine were used to raise Q. Blood gas samples were drawn at baseline levels and during periods of high or low cardiac output. We compared corresponding estimated and direct SvO2 measurements. Results: The average estimated SvO2 was 67.8% and the average directly measured SvO2 was 68% (range of 30.8 to 89.6%). The bias was -0.16% and the standard deviation of the difference was 6.4%. Figure 1 shows a scatter plot of the estimated versus the direct measurement (R2 = 0.92). Conclusions: Direct SvO2 measurement is invasive and expensive. These data show an estimate of SvO2 may be calculated from the parameters provided by a partial rebreathing cardiac output monitor and Hb. This method makes assumptions about the respiratory quotient that are only valid during periods of respiratory stability. The error may potentially be reduced by using SaO2 in place of SpO2 to estimate SvO2. Sponsored Research - Philips Medical