Processing of lung lavage fluid causes variability in bronchoalveolar cell count.

To study the effects on cell counts of different ways of processing bronchoalveolar lavage (BAL) fluid, we performed 77 BAL in 19 healthy control subjects and in 58 patients with sarcoidosis. We investigated the role of readers, hemocytometers, mode of identification of macrophages, cell washing, and speed of cytocentrifugation. No significant effect of reader or hemocytometer was observed. The percentage of macrophages, determined as large cells stained with neutral red in a Malassez hemocytometer (63.9 +/- 24.6%, mean +/- SD) was lower than the percentage of macrophages determined by May Grunwald Giemsa staining (76.3 +/- 19.2%, p less than 10(-9)). Cell counts decreased 34% after 2 washings (p less than 0.001), and more lymphocytes were counted after cytocentrifugation at 90 g (33.2 +/- 25.3%) than at 23 g (27.7 +/- 22.1%). We conclude that bronchoalveolar cell counts vary with changes in processing lung lavage fluid and that this variability should be considered when using BAL cell counts.