College of American Pathologists
Printable Version

  Q & A





cap today

January 2004

Q. Our laboratory makes buffy coats and performs the differential on white blood cell counts of less than 1,000. Are buffy coat preps a useful procedure, or should differentials not be performed below a certain level of WBC count?

A. It sounds as though your procedure might mandate the cancellation of automated differential counts on patients with leukocyte counts below 1,000/μL (1 × 109/L). However, it is important to note that if the automated differential count yields an unflagged result, this is the most precise, reproducible, and accurate differential count the laboratory can provide on low-count samples. Even if a flagged automated result is issued, it may be possible to release the automated differential count if a manual scan does not reveal cell types not quantifiable by the instrument, such as blasts or lymphoma cells.1

Some laboratories perform only manual differential counts on low-count samples when there is no clinical justification for that practice. The automated result is no less likely to be accurate in a patient with 1,000 leukocytes per microliter than in a patient with 10,000 leukocytes per microliter. Indeed, low-count samples play to the strengths of high-throughput automated technology. Therefore, this is the preferred method for differential leukocyte counting, regardless of the total leukocyte count, if other criteria for the release of an automated differential count are met.

If, however, the automated differential count fails-for instance, the instrument does not issue differential count values due to marked interference or a manual scan reveals cell types not quantifiable by the instrument-the laboratory has several choices.

Some laboratories choose not to offer differential counts for patients with leukocyte counts of less than 1,000/μL. This could be considered a valid option since, by definition, no single cell type in such a patient can be present at a level greater than 1,000/μL. In other words, the patient in that case is known to be absolutely neutropenic and lymphopenic without performance of a differential count. This option, however, does not allow for the identification of aberrant circulating cell types, such as blasts. To avoid this weakness, the laboratory can offer a microscopic interpretation (without quantitative differential count) of the blood smear by a pathologist or other qualified hematomorphologist to look for abnormal circulating cell types.

Some laboratories, like yours, perform manual differential counts on buffy coat preparations. This method provides a greater number of cells per unit area on the slide, which leads to a timely differential count. I have concerns, however, about performing differential counts on buffy coat peripheral blood preparations. A buffy coat sample is a manipulated sample for which no control or reference ranges are typically available. Therefore, it is difficult to establish the clinical relevance of rare blasts or other abnormal cell types in such samples. There are no data establishing the extent to which such abnormal cells are seen in normal individuals whose blood samples have been purposefully concentrated by differential centrifugation.

Some laboratories simply perform truncated differential counts on low-count specimens-for example, using 25 or 50 cells instead of the usual 100 or 200 cells. An advantage of this method is that the cells are being viewed in the usual setting in which we derive reference ranges and criteria regarding what is normal versus abnormal. The obvious disadvantage is the statistical imprecision that comes from counting such a limited number of cells.

Other options can be considered on a case-by-case basis. For instance, some laboratories offer automated absolute granulocyte counts as a stand-alone test order, instead of a full differential count, when a differential count is ordered simply to monitor a patient's absolute neutrophil count.2 Alternatively, the clinical service's need for differential counts on leukopenic patients may depend on the circumstances surrounding the leukopenia-for example, induced by chemotherapy or randomly discovered on routine physical exam.

It is highly likely that automated differential counts are underutilized to assess blood samples with low leukocyte counts and that optimizing the use of automated methods should solve many problems. When automated analysis is not possible or not indicated, the laboratory must choose, based on careful consultation with clinical services, the best approach to manual counting.


  1. Lantis KL, Harris RJ, Davis G, et al. Elimination of instrument-driven reflex manual differential leukocyte counts. Optimization of manual blood smear review criteria in a high-volume automated hematology laboratory. Am J Clin Pathol. 2003; 119: 656-662.
  2. Jatoi A, Jaromin R, Jennings L, et al. Using the absolute neutrophil count as a stand-alone test in a hematology/oncology clinic: an abbreviated test can be preferable. Clin Lab Manage Rev. 1998; 12: 256-260.

William G. Finn, MD
Clinical Associate
Professor of Pathology
Director of Hematopathology
University of Michigan
Ann Arbor
Member, CAP Hematology/Clinical Microscopy Resource Committee

Q. Do the chromosomes involved in the translocation t(9;22) giving rise to the Philadelphia chromosome (bearing the fusion gene BCR-ABL) in human chronic myeloid leukemia generally belong to one of the two parental genomes?

A. This question has been studied extensively by a number of investigators and is nicely summarized in a recent report.1 In 1992, Haas et al studied 11 patients with chronic myeloid leukemia and used chromosome polymorphisms to study the parental origin of the Philadelphia chromosome.2 In each of the 11 cases, the Philadelphia chromosome appeared to involve the maternal chromosome 22. This observation led the investigators to suggest that the Philadelphia chromosome may be maternally derived and is associated with gene imprinting. Several subsequent publications reported that ABL and BCR genes are not imprinted, based on molecular methods.3-5

In 1998, Nakamura et al1 attempted to repeat the experiments of Haas et al2 by looking at chromosome polymorphisms in five patients with t(9;22) and CML. These investigators found that some patients had a maternally derived abnormal chromosome 9 and paternally derived Philadelphia chromosome. Thus, the Philadelphia chromosome does not appear to have a specific parental origin.


  1. Nakamura H, Itoyama T, Niikawa N, et al. No parental origin bias for the rearranged chromosomes in myeloid leukemias associated with t(9;22), t(8;21), and t(15;17). Leuk Res. 1998;22:793-796.
  2. Haas OA, Argyriou-Tirita A, Lion T. Parental origin of chromosomes involved in the translocation t(9;22). Nature. 1992;359:414-416.
  3. Fioretos T, Heisterkamp N, Groffen J. No evidence for genomic imprinting of the human BCR gene. Blood. 1994; 83: 3441-3444.
  4. Litz CE, Copenhaver CM. Paternal origin of the rearranged major breakpoint cluster region in chronic myeloid leukemia. Blood. 1994;83:3445-3448.
  5. Melo JV, Yan XH, Diamond J, et al. Balanced parental contribution to the ABL component of the BCR-ABL gene in chronic myeloid leukemia. Leukemia. 1995; 9: 734-739.

Gordon Dewald, PhD
Cytogenetics Laboratory
Mayo Clinic
Rochester, Minn.
Consultant, CAP/ACMG
Cytogenetics Resource Committee