BACKGROUND: The total and differential cell count of bronchoalveolar lavage(BAL) fluid are useful assessing activity, prognosis and response to therapy in diffuse interstitial lung disease. But controversy exist as to the appropriate method in processing BAL fluid. Therefore we investigated the effect of gauze filtration, centrifugation and different storage time of BAL fluid on the total and differential cell count. METHOD: We obtained BAL fluid from 6 persons with no active lung lesion and divided pooled BAL fluid into several siliconized glass tubes and filtered through 0,1, 2, 4 folds of cotton guaze(pore size:lmm), and compared total cell count using hemocytometer after trypan blue staining and differential cell count after Wright-Giemsa staining of cytocentrifuged preparations. And we also counted total and differential cell count after centrifugation(400g for 30 mm) and various storage time(2hr, 24hr, and 48hr). RESULTS: There was no difference in total and differential cell count according to folds of gauze filtraion. But without gauze filtration, mucus threads that hampered total and differential cell count were found in 2 cases (33%). Centrifugation resulted in loss of total cell count(24+/-18%) without change in differential cell count. There was no change in total cell count after 21w storage but significant cell loss was found after 241w storage time(24hr : 28+/-21%, 48hr: 41+/-24%). However there was no change in differential cell count with 48hr storage time. CONCLUSION: Total and differential cell count of BAL fluid may be best performed after cotton gauze filtration without centrifugation and within 2 hours.