Abstract
Cell counting of bronchoalveolar lavage (BAL) fluid is performed manually in routine practice. This has both methodological and inherent errors; however, the accuracy and suitability of automated counting devices have been questioned. In this study, a Coulter Counter D Industrial model was calibrated and then used to measure the total cell count in unprocessed bronchoalveolar lavage fluid, and compared to a standard manual method. Bronchoalveolar lavage was performed on 34 patients undergoing routine bronchoscopy. An aliquot of unprocessed bronchoalveolar lavage fluid was taken for all counting procedures. Manual counts were performed blind by two experienced independent observers using improved Neubauer chambers. Electronic counting measured 1 ml aliquots suspended in 10 and 20 ml Isoton counting 0.5 and 1 ml duplicates. The correlation coefficients between electronic and manual counts were good. The coefficients of repeatability of electronic counts, for repeat counts, both on the same dilution (Intra-Coulter: 0.1 x 10(5) cells.ml-1) and different dilutions (Interdilution-Coulter: 0.48 x 10(5) cells.ml-1), were superior compared to those for repeat manual counts by the same observer (1.03 x 10(5) cells.ml-1), and counts between observers (1.82 x 10(5) cells.ml-1). This method offers a quick, precise and simple method for counting cells in unprocessed bronchoalveolar lavage fluid, which is both less labour intensive and subjective than manual counting.
