Underestimation of thermal lung water volume in patients with high cardiac output

When utilizing the intravascular double-indicator dilution technique to measure extravascular lung water, blood flow may be so high that diffusion equilibrium of the diffusible indicator fails to occur and the water distribution space is underestimated during the first 7 days after thermal injury. We serially measured cardiac index and lung water in five severely burned patients (mean age 24 years, range 18 to 33 years; mean burn size 56% total body surface, range 43% to 80%) by a rebreathing method utilizing two gases of differing solubility and by the thermal-indocyanine green dye (ICG) double-indicator dilution technique. Rebreathing lung water, determined by a time- and blood flow-insensitive method, increased significantly over the study period, from 6.6 ml/kg on admission to the hospital 11.3 ml/kg on postburn day 6 (+70%, P less than 0.01). Thermal-ICG lung water decreased slightly as blood flow rose. Rebreathing lung water correlated with clinical data in a patient with pulmonary edema, while thermal-ICG lung water changed in the opposite direction. Our data suggest that the thermal-ICG technique may be diffusion limited by short transit times at the high flows characteristic of burned and other critically ill patients with hyperdynamic circulations. Additionally, segmented redistribution of pulmonary blood flow known to occur in burn patients may contribute to underestimation of lung water.