Noninvasive determination of cardiac output in a model of acute lung injury.

OBJECTIVE: To examine the utility of single breath CO2 analysis as a noninvasive measure of cardiac output in a model of acute lung injury.

SETTING: An animal laboratory in a university-affiliated medical center.

DESIGN: A prospective, animal cohort study comparing 21 parameters derived from single breath CO2 analysis with cardiac output determined by an ultrasonic flow probe.

SUBJECTS: Six adult sheep with saline lavage-induced acute lung injury.

INTERVENTIONS: Animals were treated with repetitive saline lavage to achieve a uniform degree of acute lung injury (PaO2 of < 100 torr [< 13.32 kPa] on an FIO2 of 1.0). Cardiac output was manipulated by successive injections of an hydraulic constrictor placed around the inferior vena cava and measured using an ultrasonic flow probe. Twenty-one derived components of the CO2 expirogram were evaluated as predictors of cardiac output.

MEASUREMENTS AND MAIN RESULTS: Thirty-eight measurements of cardiac output were available for comparison with derived variables from the CO2 expirogam. Stepwise linear regression identified four variables for the equation predicting cardiac output: a) PaO2/FIO2 ratio; b) the angle between the slope lines for phases II and III divided by the tidal volume; c) mixed expired CO2 tension; and d) physiologic deadspace to tidal volume ratio. The multivariate equation was highly statistically significant and explained 80% of the variance (adjusted R2 = .80, p < .0001). The blas and precision of the calculated cardiac output were .00 and .38, respectively. The mean percent difference for the cardiac output estimates derived from the single breath CO2 analysis station was -0.01%.

CONCLUSIONS: Our results indicate that changes in cardiac output can be determined using components of the CO2 expirogram with a high degree of reliability in animals with induced acute lung injury. Specifically, the use of four parameters derived from a plot of expired CO2 concentration vs. expired volume predict changes in cardiac output in adult sheep with induced lung injury with an adjusted coefficient of determination of .80. Prospective application of this technology in the clinical setting with the rapidly changing physiology that is characteristic of the acutely ill patient will be essential in determining the clinical usefulness of single breath CO2 analysis as a noninvasive measure of cardiac output.