Extravascular lung water and intrathoracic blood volume: double versus single indicator dilution technique.

Abstract

The accuracy of single thermodilution was assessed in measuring extravascular lung water (EVLW) and intrathoracic blood volume (ITBV). Single thermodilution (ST) was prospectively compared with thermo-dye dilution (TD) in 13 mechanically ventilated pigs using the Pulsion Cold Z-021 monitor. Lung injury was induced with oleic acid injection. EVLWI(TD) increased from 4.3 +/- 1.4 to 11.3 +/- 4.5 ml/kg (p = 0.0014) and ITBV(TD) decreased from 581 +/- 66 to 540 +/- 85 ml (p = 0.039) after induction of lung injury. EVLW(ST) was systematically overestimated (0.5-1 ml/kg) and ITBV(ST) was slightly underestimated (15-20 ml) when calculated automatically by the Pulsion Cold Z-021 monitor. This bias could be removed by adjusting two coefficients needed for the computation of ITBV(ST) and EVLW(ST) so that the following regression equations were obtained: EVLWI(ST) = 0.98 EVLWI(TD) + 0.27 (r = 0.94, p < 10(-6)) and ITBV(ST) = 1.0 x ITBV(TD) + 0 (r = 0.87, p < 10(-6)). Single thermodilution allows estimation of ITBV and EVLW values with reasonable accuracy and therefore provides useful information about the cardiac preload and the severity of lung injury. However, automatic calculation of EVLWI(ST) and ITBV(ST) using the Pulsion Cold Z-021 might be biased. Therefore, researchers and clinicians should validate single thermodilution for the given settings, before employing this method to estimate EVLW and ITBV.