College of American Pathologists
Printable Version

  Q and A





July 2009

Fredrick L. Kiechle, MD, PhD

Question Q. Our hematology laboratory would like to institute guide-lines for repeat manual differentials on oncology patients with low WBC counts that typically run <0.1 or 0.1 x 109/L. We have been getting orders for differentials on patients who show no variation in the low WBC count. Can you suggest utilization guidelines for repeat diffs on hematology oncology patients?

A. Katherine Galagan, MD, chief of pathology and director of clinical laboratories at Virginia Mason Medical Center, Seattle, answered a related question in the August 2008 issue of CAP TODAY, in which a reader asked whether there are accepted policies regarding the performance of repeated manual differential white blood cell counts for patients with persistent low total leukocyte counts. In her answer, Dr. Galagan referred to an article put forth by the International Society for Laboratory Hematology, in which an international group of 20 laboratorians suggested criteria for action following automated CBC and WBC differential analysis.1 For patients with repeated low leukocyte counts of <4.0 x 109/L, the group recommends a slide review (or manual differential) with the first low result and then a repeat slide review if the delta check fails within three days.

The literature is rather sparse with respect to repeat manual differentials on oncology patients with very low WBC counts (<0.5 x 109/L). Utilization guidelines are not readily found. It is standard in most hospitals to admit oncology patients with fever and an absolute neutrophil count of <0.5 x 109/L for broad-spectrum intravenous antibiotics.2 In one article addressing pediatric patients, the authors found that pediatric oncology outpatients with fever and neutropenia who present with an initial monocyte count of ≥0.1 x 109/L, without comorbidity or abnormal chest radiograph, are at a lower risk for significant bacterial infection and can be considered for less aggressive initial therapy.3 It is standard for an oncology outpatient at my hospital to have a manual differential performed for a WBC count of ≤0.5 x 109/L. However, the practice of repeated manual differentials on oncology inpatients with very low WBC counts is not addressed well in the literature.

I informally surveyed the members of the CAP Hematology/Clinical Microscopy Resource Committee for institutional guidelines for repeat manual differentials on oncology inpatients with very low WBC counts. The results showed a wide variation in practice. Some hospital laboratories per-form manual differentials on all low WBC counts, while in my own hospital laboratory manual differentials are not routinely performed on WBC counts of ≤0.5 x 109/L in adult or pediatric oncology inpatients. In one children’s hospital laboratory, blood films with WBC counts of <0.5 x 109/L are scanned for evidence of a blast population, with results reported to reflect the number of cells scanned and the blast percentage found. In an-other large laboratory, blood films are scanned when the WBC count is <0.5 x 109/L, but a manual differential is not always performed. In those cases where manual differentials are performed (often by physician request), the laboratory will perform a manual differential once every 24 hours, unless there has been a significant change in the WBC, RBC, or platelet counts. This is an area where utilization guidelines would be very helpful.


  1. Barnes PW, McFadden SL, Machin SJ, et al. The international consensus group for hematology review: suggested criteria for action following automated CBC and WBC differential analysis. Lab Hematol. 2005;11:83–90.
  2. Pizzo PA, Robichaud KJ, Gill FA, et al. Duration of empiric antibiotic therapy in granulocytopenic patients with cancer. Am J Med. 1979;67:194–200.
  3. Klaassen RJ, Goodman R, Pham B, et al. “Low-risk” prediction rule for pediatric oncology patients presenting with fever and neutropenia. J Clin Oncol. 2000;18: 1012–1019.

Tracy I. George, MD
Assistant Professor of Pathology
Associate Director,
Clinical Hematology Laboratory
Associate Director,
RBC Special Studies Laboratory
Stanford Hospital and Clinics
Lucile Packard Children’s Hospital
Stanford University School of Medicine
Stanford, Calif.

Chair, CAP Hematology/
Clinical Microscopy Resource Committee

Question Q. I have recently seen mean corpuscular volumes drop drastically in newborns—from 115 fL at birth to 102 fL at three days of age. Why is it not stable, and is this normal?

A. The MCV in newborns is higher than that observed in adults and decreases with increasing gestational and postnatal age. The mean MCV is 119 fL in premature infants at 25 weeks of gestation and 106 fL in term infants at 40 weeks of gestation. There is a significant decrease in mean MCV after birth also, with term infants demonstrating a mean MCV of 106 fL at less than 24 hours, 98 fL at 96 hours, and 94 fL at one week of age. The change in MCV values probably results from a combination of factors, such as the shift from hepatosplenic to marrow erythropoiesis and a transient decrease in reticulocytes that occurs due to oxygen-delivery changes after birth. In ill newborns, transfusion therapy can cause a large change in MCV because transfused adult cells have a lower MCV. Tables of hematologic values in the first week of life are available in texts of pediatric hematology, and normal values used in nurseries should be based on validated literature or local studies.

Robert Novak, MD
Department of Pathology
Children’s Hospital
Medical Center of Akron
Akron, Ohio

Immediate past chair,
CAP Hematology/Clinical
Microscopy Resource Committee

Question Q. We’ve had questions from clinicians about the rate and extent of change of thyroidstimulating hormone, or TSH, values in patients tested and then retested within a week. These clinicians claim their patients have not changed medications or life patterns. I realize there can be variations in TSH values owing to the pulsatile nature of the hormone. However, to what extent can I see these variations in the values?

A. Biological and analytical variation can be considerable in the measurement of TSH. In a 2003 paper by Andersen, et al.,1 15 healthy subjects had 12 monthly measurements of thyroid function. The TSH coefficient of variation for individuals varied between approximately 15 percent and approximately 35 percent. As the CV represents ±1 standard deviation, these variations represent only 68 percent of the total variation, suggesting that much wider intraindividual variations in TSH are possible. The 2003 National Academy of Clinical Biochemistry Laboratory Practice Management Guidelines on thyroid function testing2 reported intraindividual CVs for TSH of 19.3 percent for one week, 20.6 percent for six weeks, and 22.4 percent for one year. These guidelines also said that diurnal variation introduces significant variation in TSH measurements, with a nocturnal peak TSH value that is twice the nadir TSH value.

One of the chief sources of biological variation in the measurement of TSH (and some of the other pituitary hormones) is the pulsatile pattern of secretion and fairly short half-lives of approximately one to two hours.3 In this study TSH was released in approximately 18 pulses over the course of a day, and the TSH half-life was approximately 60 to 90 minutes. The measured TSH concentration then depends, in part, on how much time has passed since the last TSH pulse when the patient was sampled. The half-life of TSH may also vary between individuals, depending upon interindividual differences in the degree of sialylation and sulfation of TSH (for example, variation in oligosaccharide structure).4,5

To reduce intraindividual variations, patients should undergo venipuncture at a consistent time of day and the same laboratory and testing platform should be used. If the TSH result varies within the reference interval, it is likely that such variations represent expected biological and analytical variations.

One question to ask the clinician is why TSH is being measured weekly in somebody who is said to be clinically stable. An excellent review on TSH variation was published in 1996 by D.A. Fisher, MD.6


  1. Andersen S, Bruun NH, Pedersen KM, et al. Biologic variation is important for interpretation of thyroid function tests. Thyroid. 2003;13:1069–1078.
  3. Keenan DM, Roelfsema F, Biermasz N, et al. Physiological control of pituitary hormone secretory-burst mass, frequency, and waveform: a statistical formulation and analysis. Am J Physiol Regul Comp Physiol. 2003;285:R664–R673.
  4. Fares F. The role of O-linked and N-linked oligosaccharides on the structure-function of glycoprotein hormones: development of agonists and antagonists. Biochim Biophys Acta. 2006;1760:560–567.
  5. Persani L, Borgato S, Romoli R, et al. Changes in the degree of sialylation of carbohydrate chains modify the biological properties of circulating thyrotropin isoforms in various physiological and pathological states. J Clin Endocrinol Metab. 1998;83:2486–2492.
  6. Fisher DA. Physiological variations in thyroid hormones: physiological and pathophysiological considerations. Clin Chem. 1996;42:135–139.

William E. Winter, MD
Glen Hortin, MD
Neil S. Harris, MD
Department of Pathology, Immunology, and Laboratory Medicine
University of Florida
College of Medicine

Question Q. What information can be obtained from performing linearities on point-of-care precalibrated instruments that cannot be obtained from running two levels of controls?

A. To establish linearity, one needs at least three points for a line. Routine QC usually consists of only two points, generally high and low values for an analyte. The CAP’s linearity Surveys include multiple points to adequately challenge and confirm linearity across the range of an assay. The Survey also allows users to verify their Analytical Measurement Range, or AMR. The AMR is the range of an assay that can be obtained without any specimen manipulation, either by dilution or concentration. This is addressed in chemistry and toxicology checklist item CHM.12950, which says the AMR “is the range of analyte values that a method can directly measure on the specimen without any dilution, concentration, or other pretreatment not part of the usual assay process.” AMR verification requires users to prove that an assay can accurately measure an analyte as low and as high as a manufacturer claims or as was originally validated when the assay was introduced. To do that, one must use at least one low, mid, and high point spanning the AMR at periodic intervals to confirm that the AMR remains the same as reagent lots change. Two QC points may not adequately cover the AMR.

Cynthia Bowman, MD
Vice Chair, Department of
Laboratory Medicine
Long Island Jewish Medical Center
New Hyde Park, NY

Chair, CAP Point of Care
Testing Committee

Dr. Kiechle is medical director of clinical pathology, Memorial Healthcare, Hollywood, Fla.