What's the Take Home?

A 38-Year-Old Woman With Leukocytosis

Ronald N. Rubin, MD1,2Series Editor

1Lewis Katz School of Medicine at Temple University, Philadelphia, Pennsylvania
2Department of Medicine, Temple University Hospital, Philadelphia, Pennsylvania

Rubin RN. A 71-year-old man with pain and stiffness in the hand. Consultant. 2020;60(5):e5. doi:10.25270/con.2020.05.00005

The author reports no relevant financial relationships.

Ronald N. Rubin, MD, Temple University Hospital, 3401 N Broad St, Philadelphia, PA 19140 (blooddocrnr@yahoo.com)

A 38-year-old woman was referred for evaluation of an elevated white blood cell (WBC) count that had been found incidentally on a routine yearly evaluation. She brought her complete blood cell count (CBC) results with her, which had been obtained by her new primary care provider (PCP) 1 month previously (earlier in the year, she had changed PCPs). The CBC revealed a normal hemoglobin level (14.28 g/dL), normal red blood cell indices, and a normal platelet count. However, the total WBC count was 13,400/µL, with essentially all of the elevation accounted for by an increased number of neutrophils, specifically 10,900/µL. Both the Coulter Counter and the automatic “referral” to a commercial pathologist reported rare band forms but no other abnormal WBC forms on examination of the peripheral blood smear. Results of an accompanying basic and comprehensive metabolic and biochemical blood profile were entirely normal.

The patient is a healthy woman whose only major medical diagnosis is asthma, which had been much more severe in childhood but was less so now, except when she experiences a significant upper respiratory tract infection (URI). Her long-term medications include oral contraceptives and acetaminophen and/or ibuprofen as needed for minor orthopedic symptoms that are occasionally associated with her exercise programs.

Physical examination revealed totally normal vital signs. Examination of the head, eyes, ears, nose, and throat revealed normal findings, and there was no thyromegaly. Her chest was essentially clear, save for a rare expiratory wheeze. Abdominal examination findings were normal, with no hepatosplenomegaly. The extremities were negative for rashes, lesions, or edema.

Answer: C, arrange for serial CBCs in the following weeks and months, including attempts to retrieve any prior CBC results.

In the era of the Coulter Counter, that simplest physician order—obtaining a CBC—results in a quick result for not only the hemoglobin and hematocrit values, but also the WBC. And this comes with quite a bit of acceptable preliminary accuracy as to the total number and the breakdown as to which WBCs are present. Then, since we are dealing with programmable machines, there will be asterisks, flags, red alarm messages, or some other form of frantic effort to show the ordering party that an abnormal number—either too high or too low—is present. And hopefully, since most of us are aware that two of the great major differentials that come into play are infection and bone marrow neoplasm, the great chase is under way. The purpose that this article is to put some framework, order, and actuarial logic into the evaluation of such a finding on a CBC.

The initial “classification” involves the level of the number itself and the WBC line involved. There are 2 major WBC lines, the granulocytic or polymorphonuclear (PMN) line and the lymphoid line. The Coulter Counter will register these, but a key part of any and all evaluations is a professionally reviewed evaluation of the sample on smears, be it by a pathologist at the testing site or a hematopathologist or hematologist at the clinic or hospital where the patient is located. Once the specific line can be confirmed, the specific evaluation paradigm pathway will be discussed below.

Also required initially is the number or degree of elevation itself. A key value in the literature, for example, is greater than 50,000/µL. If this is found in the PMN line, this usually will mean a bone marrow malignancy (eg, myeloproliferative disorders, acute myeloid leukemia [AML], chronic myelogenous leukemia [CML]—see below) or a leukemoid reaction usually found in the presence of necrotic tissue such as impending limb loss. It is uncommon for even a severe infection or any drug to raise a PMN line count to these levels. In the lymphoid line, it is also unusual for any reactive or infectious cause to enable the lymphocyte count to attain these levels such that almost always the cause will be a lymphoid malignancy.

Returning to the PMN line, the next important differentiating point is to try to determine whether the WBC elevation is “reactive”—that is to say, the marrow function is normal but the thermostat has been changed upward via cytokines by some stimulus such as infection, stress, or medications. Or, is the WBC elevation occurring autonomously due to abnormal bone marrow anatomy and physiology, extending the metaphor by saying that the bone marrow thermostat is broken. As a group, autonomous marrow indicates malignancy. These situations can be explored and defined using flow cytometry, cytogenetic and molecular testing of bone marrow, or peripheral blood.1 This is the case, for example, in myeloproliferative disorders such as polycythemia vera, wherein an acquired mutation in stem cells (JAK2 mutation) results in both intrinsically abnormal bone marrow proliferation of PMN (and other) forms, as well. Another acquired stem cell chromosomal derangement—the 9/22 translocation or Philadelphia chromosome—results in a gene fusion that amazingly is translated into an oncoprotein, BCR-ABL tyrosine kinase, which similarly causes florid, autonomous stem cell proliferation with extremely high WBC counts as well as unstable mutagenesis of stem cells to even more malignant forms (blast crisis).2

Lymphoid line elevations are far more restricted in differential, since reactive lymphocytosis is very uncommon in adults. A false lymphocytosis may be seen in a variety of viral infections, which in reality are causing temporary neutropenia such that the Coulter Counter shows 60% to 70% lymphocytes in the differential, although the absolute counts remain normal at less than 4000 to 5000/µL. An exception seen in teens and young adults is cytomegalovirus infection and Epstein-Barr virus infection, which do indeed cause mild lymphocytosis, which is a clue to an infectious mononucleosis. The vast majority of true lymphocytosis will thus be due to lymphoid malignancy, the most common being chronic lymphocytic leukemia (CLL), especially in a patient older than 50 years of age. Lymphocytosis is also seen in other lymphomas of the small B cell family such as marginal cell lymphoma and mantle cell lymphoma. And, finally, acute lymphocytic leukemia itself, although less common in adults than in children, will show lymphoid cells in excess, usually larger blasts forms that are easily identified by the reviewer of the peripheral blood smear.

What constitutes a reasonable, useful, and efficient office evaluation (or inpatient evaluation, for that matter)? Initially and easiest, these 5 questions need to be addressed:

1. What is the patient’s accurate and complete history? Is there now or has there been a bona fide infection (eg, pulmonary, dental)? Does the patient have chronic obstructive pulmonary disease or asthma with corticosteroid use involved?

2. If the WBC count is only minimally to moderately elevated, is it real? Are there previous CBC results that can be found and compared? If the patient is otherwise well and without other abnormal findings, can repeated CBCs be done to see whether there is spontaneous resolution?

3. Once the results of multiple WBC studies are available, within weeks to months does the WBC count vary, or is there an inexorable increase?

4. What is the absolute WBC count? Less than 15,000/µL? Greater than 25,000/µL? Greater than 50,000/µL?

5. And of course, what is the specific cell line elevation reported by professional review of the blood smear? Is the dominant form in the PMN line (eg, PMNs, bands) or lymphoid line? Are there immature blast forms?

Once this has been done, the gross diagnosis may become apparent. A modestly elevated WBC count with the PMN line affected that resolves in a month is most likely reactive, perhaps from a minor infection. A modestly elevated WBC count of the PMN line that can be shown to be sporadic often leads to finding corticosteroid medicines being used and causing this to occur. In the absence of such findings, especially if the WBCs progressively increase and work their way above 20,000/µL, more-refined evaluation—often in a hematologist’s office—is indicated. If the predominant forms are PMN line, molecular and genetic markers such as the JAK2 and BCR-ABL proteins will confirm or rule out myeloproliferative or CML causation. Therapeutics have evolved such that excellent treatments exist with good prognosis when these are found. When the lymphoid line is the dominant form, flow cytometry can often define which lymphoproliferative disorder is present, and again, very effective therapeutics are now available for many of these entities.

On occasion, these studies will eventuate into an indication for bone marrow biopsy for diagnosis, and an acute leukemia will be found, which is far more ominous therapeutically and prognostically.

Taking this discussion into context with the presented patient, she was totally asymptomatic and without findings on examination such that initiating antibiotics for a phantom infection (Answer D) would be a blind stab and is the incorrect approach here. The reviewed smear showed mainly PMN forms with rare band forms (also in the PMN line); thus, the count was only minimally elevated such that an immediate bone marrow biopsy is far too early (Answer B). There was asthma in the patient’s history, and so attempts to find past CBC data and episodic corticosteroid exposures is worth the effort while repeated studies over the next 6 weeks or so are performed. The counts seen are not inherently dangerous, so this conservative course (Answer C) is quite reasonable and is the best choice here. This physician would not proceed to Answer A—the detailed evaluation for an autonomous malignant cause for the WBC elevation—with the limited amount of nonominous data available right now. Having said this, in the current era of performing every test available in every patient in every situation immediately and all the time, I could not say that Answer A is absolutely incorrect and does not occur. Again, this author would not do so with the data presented.


The patient was given a schedule of follow-up CBC testing, including at the present visit and then twice monthly thereafter. The WBC count on the day of visit was 11,000/µL with normal differential, and the following two WBC counts were entirely normal. The slides were evaluated by a staff hematopathologist and were found to be normal.

At the patient’s 10-week follow-up appointment, she brought her medical records from the past 4 years, several of which were at different primary care practices using different reference laboratories. There were 5 different CBC results, including the most recent described in the case presentation. The results of 3 were entirely normal. A fourth, from 3 years prior, did show a similarly elevated WBC/PMN value. There were 2 normal results between that study and the one that prompted the consultation.

On further questioning, she recalled her having episodes of “reactive airways” due to her asthma history and a URI, and on occasion receiving an azithromycin dose pack and a methylprednisolone dose pack for this, including in the month prior to her visit to her new PCP. For now, therefore, her episodic leukocytosis will be assumed to be drug-related (corticosteroids). She will have 6-month evaluations, and if her WBC count is shown to to be steadily increasing over time, or manifests abnormal morphology, a more detailed evaluation as discussed in Answer A will be performed.


1. Riley LK, Rupert J. Evaluation of patients with leukocytosis. Am Fam Physician. 2015;92(11):1004-1011.

2. Cerny J, Rosmarin AG. Why does my patient have leukocytosis? Hematol Oncol Clin North Am. 2012;26(2):303-319. doi:10.1016/j.hoc.2012.01.001

3. Gibson C, Berliner N. How we evaluate and treat neutropenia in adults. Blood. 2014;124(8):1251-1378. doi:10.1182/blood-2014-02-482612


The major categories of human leukocytes (WBCs) consist of the neutrophil (PMN) line and the lymphoid line. Leukocytosis refers to abnormal elevations in these cells. In the PMN line, leukocytosis can be reactive as with infections, drug-related (especially corticosteroids) or neoplastic/clonal, as with myeloproliferative disorders and AML. In the lymphoid line, leukocytosis is most frequently neoplastic/clonal, namely CLL and lymphomas. Office evaluation should include serial CBCs to confirm that the finding is real and whether episodic and medication-related or steadily progressive, as well as the degree of leukocytosis. Also, a professionally evaluated blood smear examination is mandatory. Thereafter, frequently in the hands of a hematologist, flow cytometry, cytogenetics, and molecular markers (eg, JAK2, BCR-ABL) and possible bone marrow biopsy are usually able to confirm a precise diagnosis.2 Abnormally low leukocyte counts are most commonly the PMN line—neutropenia. Again, serial measurements and past CBC results are helpful in establishing cause. Ethnic neutropenia is most common in patients of African/Mediterranean descent. Common acquired causes are viral infection and drugs.3