Evaluation of the Coulter STKS five-part differential

The authors evaluated the Coulter STKS (Coulter Corp., Hialeah, FL) five-part differential in a tertiary-care hospital using samples with a broad range of distributional and morphologic abnormalities. Particular attention was given to the performance of the instrument-generated suspect flags that occur as an aid to identify samples with abnormal leukocytes. A morphologically abnormal, or positive, blood smear was defined by the presence of any blasts, malignant lymphoid cells, grossly dysplastic neutrophils, nucleated red blood cells (nRBC), platelet clumps, or reactive lymphocytes of more than 5%. The presence of any white blood cell-related suspect flag, except for Immature Granulocyte/Bands (i.e., Blasts, Variant Lymph, NRBC, Platelet Clumps, Review Slide, or WBC*R), was considered to be a positive instrument result. The STKS showed excellent quantitative results for the WBC differential compared with the manual differential when these "morphologic abnormalities" were absent in a 400-cell manual differential or low in numbers (< or = 5%). Specificity of these non-immature granulocyte/band suspect flags was good, with a false-positive rate of only 11.7%. Overall sensitivity in 113 samples with morphologic abnormalities was 67.3%. Sensitivity to detection of > or = 1% abnormal WBCs or > or = 1 nRBC/100 WBCs (a subset of 78 samples) was 80.8%. Sensitivity to detection of more than 5% abnormal WBCs or more than 5 nRBC/100 WBCs (a subset of 53 samples) was 84.9%. The primary deficiency was the inability of the STKS to flag samples with lymphoma cells, lymphoid blasts, or more than 5% reactive lymphocytes.