Relationship between inflammatory cells in bronchoalveolar lavage fluid and pathologic changes in the lung interstitium.

Bronchoalveolar lavage (BAL) is recognized as an important research tool for various lung diseases, but it is still uncertain whether inflammatory cells in BAL fluid (BALF) accurately reflect pathologic changes in the lung interstitium. We used a morphometric method to quantify the density of inflammatory cells in the lung interstitium by utilizing a computer-aided graphic analyzer and compared those findings with BALF results. Two types of animal models were studied, i.e., endotoxemia (Escherichia coli endotoxin) and hypersensitivity pneumonitis (inhaled ovalbumin). Male Wistar rats were used; the right lungs were lavaged and the left lungs were prepared for morphometric study. In the endotoxemia model, the neutrophil fraction in BALF and the neutrophil density in the lung interstitium correlated significantly at 18 h (r = 0.81, p < 0.05) and 24 h (r = 0.81, p < 0.05) but not at any other time points after injection. In the hypersensitivity pneumonitis model, the neutrophil fraction in BALF and the neutrophil density in the lung interstitium correlated significantly (r = 0.80, p < 0.05) only at 3 h after inhalation. The lymphocyte fraction in BALF and the lymphocyte density in the lung interstitium were correlated positively at 3 h (r = 0.83, p < 0.05), 1 day (r = 0.82, p < 0.05), 2 days (r = 0.67, p = NS), and 4 days (r = 0.87, p < 0.05), but not at 6 days after inhalation. Our data suggest that neutrophil fraction in BALF does not reflect neutrophil populations in the lung interstitium except at the time of maximal neutrophil count in lung lavage. For lymphocytes in the hypersensitivity pneumonitis model, those in BALF and in the lung interstitium roughly correlate in the majority of measurements.