Regional lung hematocrit variation and assessment of acute lung injury.

Estimating blood content in the lung remains a key step in calculating lung water volume and microvascular permeability. We studied the effect of regional lung hematocrit (Hct) variation on assessment of acute lung injury. Escherichia coli endotoxin was administered in guinea pigs intravenously. Lung injury was evaluated by measuring the wet-to-dry weight ratio (W/D) and transvascular 125I-labeled albumin leakage for 3 h [tissue-to-plasma 125I-albumin ratio (T/P)] in five tissue samples from each animal. Residual blood content was corrected using either 51Cr-red blood cells as a blood cell marker, 99mTc-albumin as a plasma marker, or both, injected 10 min before the guinea pigs were killed. Lung Hct, estimated from the marker counts of lung and peripheral blood samples, was lower than peripheral blood Hct; intraindividual variation, represented by the standard deviation in each subject, was 0.024 +/- 0.015 for the control group (coefficient of variation 8.0 +/- 5.1%) and 0.026 +/- 0.013 for the endotoxin group (coefficient of variation 8.5 +/- 4.1%). Uncorrected W/D for residual blood content was greater than the corrected W/D. 99mTc-albumin correction gave values closer to the W/D corrected by both markers. T/P corrected by 99mTc-albumin showed smaller data variations than the values obtained with 51Cr-red blood cell correction, which was affected by variations in lung Hct. We recommend using a plasma marker to correct for blood content in assessing acute lung injury by W/D and T/P.