VALIDATION OF A SIMPLE METHOD TO COUNT VERY LOW WHITE CELL CONCENTRATIONS IN FILTERED RED-CELLS OR PLATELETS

The increased performance of white cell (WBC) filters makes it difficult to count precisely the number of residual WBCs. Concentrations as low as 0.01 WBC per mu-L cannot be determined with electronic cell counters, conventional hemocytometers, or the flow cytometric techniques currently being used. This article describes a simple, manual method using a Nageotte hemocytometer with a large-volume chamber (50-mu-L) to count the number of WBCs contained in red cell (RBC) suspensions (preparations A, B, and C) and in platelet suspensions (preparation D) diluted 1 in 10 pure, or concentrated two fold. To validate the method, several reference ranges, prepared by successively adding mononuclear cells to a suspension of pure RBCs or platelets, were used. Among the different series, validation ranges varied from 0.2 to 12 to 0.01 to 0.5 WBCs per mu-L and correlation coefficients ranged from 0.929 to 0.996. To determine the limit of accurate detection, accuracy tests (n = 160) were carried out by two experienced operators on samples with WBC concentrations of about 5, 10, and 120 times the concentration at the theoretical limit of detection (1 WBC/chamber). No significant difference was observed in the various types of preparations (A, B, C, D) in the tests performed by the two operators. However, intra-assay coefficients of variation were 18, 9.5, and 2.2 percent, respectively, at WBC concentrations of 5, 1 0, and 120 times that at the theoretical limit of detection. These observations show that a limit of accurate detection (10%) seems to be reached when 10 cells are observed in a Nageotte hemocytometer. Finally, the method is applied to the study of an experimental filter with an efficiency level higher than 5.8 log10.