Factors influencing the estimation of extravascular lung water by transpulmonary thermodilution in critically ill patients

Objective: To investigate factors that may influence the estimation of extravascular lung water (EVLW) with a single (cold) indicator compared with assessment using two indicators (thermo-dye dilution). Design: Post hoc analysis of an electronic hemodynamic database. Setting: Surgical intensive care unit of a university hospital. Patients: Forty-eight critically ill patients monitored by the thermo-dye dilution technique in the postoperative period. Interventions: None. Measurements and Main Results: The EVLW was simultaneously assessed by the thermo-dye dilution technique (EVLWref) and estimated by transpulmonary thermodilution (EVLWest). EVLWref index ranged between 1 and 40 mL/kg (mean 10 +/- 7 mL/kg) and EVLWest between 2 and 39 mL/kg (mean 9 +/- 6 mL/kg). EVLWref was closely correlated (r = .96) with EVLWest. The mean difference (bias) between EVLWref and EVLWest was -0.5 +/- 1.9 mL/kg. The bias was not influenced by the weight, height, body surface area, body mass index, Pao(2), intrathoracic blood volume, cardiac output, or dosage of vasoactive agents. In contrast, the bias was slightly but significantly influenced by EVLWref, Pao(2)/Fio(2) ratio, tidal volume, and level of positive end-expiratory pressure. Conclusions. In our surgical intensive care unit population, the estimation of EVLW by transpulmonary thermodilution was influenced by the amount of EVLW, the Pao(2)/Fio(2) ratio, the tidal volume, and the level of positive end-expiratory pressure. However, compared with the double indicator method, transpulmonary thermodilution estimation remained clinically acceptable even in patients with severe lung disease.