Non-invasive measurement of cardiac output in patients with acute lung injury using the carbon dioxide rebreathing method.

STUDY OBJECTIVE: To compare measurement of cardiac output by the CO2 rebreathing method with the thermodilution cardiac output technique in mechanically ventilated patients with acute lung injury. DESIGN: Prospective study comparing two methods of cardiac output measurement in 22 consecutive patients with acute lung injury. SETTING: Intensive care unit of a university hospital. PATIENTS: Twenty-two mechanically ventilated patients with acute lung injury monitored with systemic and pulmonary artery catheters. MEASUREMENTS AND RESULTS: Cardiac output was determined using both the thermodilution technique and an indirect CO2 Fick method. Veno-arterial CO2 content difference was calculated from an estimated mixed venous CO2 tension obtained by an equilibrium CO2 rebreathing method and measured arterial CO2 tension. Carbon dioxide pressure was converted to content using the equation of the CO2 dissociation curve described by McHardy. A wide range of cardiac index was studied from 2.7-5.7 l/min/m2. There was a significant correlation between thermodilution and CO2 rebreathing methods (r2 = 0.82, p < 0.01). The mean difference between the CO2 rebreathing method and thermodilution was 0.05 l/min/m2, with a standard deviation for the bias of 0.38 l/min/m2. CONCLUSION: Our results suggest that the CO2 rebreathing method may be a reliable non-invasive technique to determine cardiac output in mechanically ventilated patients with acute lung injury.