Sysmex SE-9000 hematology analyzer: performance evaluation on leukocyte differential counts using an NCCLS H20-A...

We evaluated the performance (ie, imprecision, inaccuracy, and analytic sensitivity) of the Sysmex SE-9000 commercial hematology analyzer (TOA Medical Electronics, Kobe, Japan) on differential leukocyte counts according to the National Committee for Clinical Laboratory Standards H20-A protocol. The results obtained were compared with those from the Bayer H6000 and H3 (Bayer Diagnostic Division, Tarrytown, NY), the Coulter MAXM (Miami, Fla), and the microscopic method. Altogether, samples from 462 subjects were analyzed. The results show a substantial superimposition of reference intervals between the methods. The imprecision of the SE-9000 is low for all the leukocyte subpopulations, with the exception of basophils (coefficient of variation: neutrophils, 3.35%; lymphocytes, 4.25%; monocytes, 7.9%; eosinophils, 9.5%; and basophils, 44.2%) and is consistently lower than that of manual counts. The correlation with other methods is high, with the exception of basophils (r2: neutrophils, 0.94-0.95; lymphocytes, 0.93-0.97; monocytes, 0.76-0.85; eosinophils, 0.96-0.99; and basophils, 0.02-0.56). When compared with the microscopic method, an overestimation of neutrophils is seen mostly at low concentrations (mean difference, 2.63), and an underestimation of lymphocytes is seen at high concentrations (mean difference, -3.1). The clinical sensitivity was good, with an agreement of 75.7% on morphologic and 89.6% on distributional abnormalities. With a new analytical channel for immature cells (IMI), the analyzer shows high sensitivity in detecting immature cells of the granulocytic lineage (from 94.4% for immature granulocytes to 96% for myeloblasts).