Dale, David C.
Department of Washington, University of Washington, Seattle, Washington, USA
Correspondence to David C. Dale, MD, Professor of Medicine, Department of Washington, Box 356422, Health Science Building, Rm AA522, University of Washington, Seattle, WA 98105, USA. Tel: +1 206 543 7215; fax: +1 206 685 4458
White blood cell counts (WBC) and blood neutrophil counts (ANC) are very common tests. Clinicians regularly get an alarm or alert from the laboratory when a patient's ANC is low. How should we interpret a low ANC?
STATISTICAL AND ETHNIC CONSIDERATIONS
The ANC in normal adults of European or Asian origin is usually 1.8–7.2 × 109/l. African-Americans have a wider range of normal values (1.3–7.4 × 109/l), and the population curve is shifted to the left, meaning overall lower ANCs. Although premature babies are often neutropenic [see review ‘Neutropenia in the newborn’ by Maheshwari in this issue (pp. 43–49)], healthy full-term newborn infants have approximately an adult level of blood neutrophils. Neutrophils may decrease slightly during the next 12 months, but then, barring illness, the ANC is at the adult level for the rest of a person's life. Yemeni Jews, African-Americans, Africans living in Africa, Caribbean inhabitants of African descent, Ethiopians, and some Arab populations have lower ANCs than persons of European descent. These population differences may seem unimportant, but they often lead to under-dosing of cancer chemotherapy, interruptions in antiviral therapies for hepatitis, and a greater risk of HIV infection [see ‘Neutrophil counts in persons of African origin’ by Thobakgale and Ndung’u in this issue (pp. 50–57)].
Blood neutrophil counts show diurnal variation and increase with exercise, excitement and any activity increasing cardiac output. For every patient and population, the ANC depends on exactly what the patient's circumstances were when the blood was drawn. The ANC normally does not show cyclic variation, but day-to-day counts vary considerably even with the standardization of the conditions. Patients may also have spuriously high counts on a first ‘anxious’ encounter, and ‘settle down’ to their usual lower counts with subsequent measurements. For patients at the lower end of the normal range, this means that a second or third ANC measurement can fall below the ‘cut off’ of the normal range. If the patient has just begun a new therapy, a clinician may attribute the fall in counts to the drug and not to physiological changes. The best approach for understanding this situation is usually to obtain more measurements of the ANC.
When collected for counting, neutrophils are living cells. Over time, the neutrophil will die in the collection tube, death being via apoptosis. Usually, counts are stable for about 24 h, but if the storage conditions are too hot or too cold, neutrophils will decay more rapidly. This problem may arise with centralized laboratories and shipment irregularities. The simplest approach to this problem is to repeat the count using a local laboratory and control the sample handling. Before electronic counting of blood cells, there was substantial variation in measuring the ANC in neutropenic patients with manual differentials. This kind of variability has largely disappeared with the very large numbers of cells counted with electronic particle counters.
NEUTROPENIA AND THE RISK OF INFECTIONS
Neutropenia is an important determinant of susceptibility to infections. Early in the development of modern cancer chemotherapy, clinical research showed that there was a clear relationship between the severity and duration of neutropenia and the risk for infections, originally reported for patients with acute myeloid leukemia. Researchers, professional organizations and the US Food and Drug Administration adopted standards for categorizing neutropenia as mild, moderate or severe, or grades 1, 2, 3 and 4, severe neutropenia or grade 4 neutropenia being defined as a blood neutrophil level of 0.5 × 109/l. These simple ways of categorizing neutropenia have proven very useful, but it is always very important to interpret these numbers in light of the clinical situation. A low ANC when the counts are falling is far more serious than when the counts are rising. After chemotherapy, and especially after hematopoietic stem cell marrow transplantation, clinicians and patients live expectantly, waiting to see the trajectory of the ANC.
Why is it so important to know whether the count is going up or down? These direction signals are clear indicators of the proliferative activity and production of neutrophils in the marrow and the bone marrow neutrophil supply. In fact, if we could only measure it readily, the post-mitotic neutrophil pool or the ‘marrow neutrophil reserves’ would be a much better index than the ANC for understanding risk of infections after myelosuppressive chemotherapy and in many other clinical situations.
LESS COMMON SITUATIONS
Almost any drug may cause neutropenia, or at least neutropenia is attributed to almost every drug. The risks vary considerably and the mechanisms are difficult to establish. Usually, idiosyncratic drug-induced neutropenia occurs in the first 3 months of a new drug and the setting is often complicated by many concomitant medications. The only appropriate way of dealing with this problem is for clinicians to be vigilant and consider obtaining blood counts and stopping new drugs with even a slight hint of a drug reaction or when patients started in the previous 3 months on a new medication develop a fever.
Neutropenia occurs in a wide variety of autoimmune diseases, but the neutropenia is usually mild or moderate, that is, the counts are greater than 0.5 × 109/l. In these patients, the risk of bacterial infections is often more attributable to the immunosuppressive effects of medications than the disease-related neutropenia.
In severe chronic neutropenia, particularly patients who have ELANE-associated neutropenia, blood neutrophils are selectively decreased, probably because the mutant enzyme expressed in the marrow leads to death via apoptosis of neutrophil progenitors. In contrast to chemotherapy-associated neutropenia, monocyte development is not impaired and these patients usually have increased production of blood monocytes. Although their functions are different, both neutrophils and monocytes are protective of patients from bacterial infections. It is probably the specific host impairment with congenital and cyclic neutropenia that leads to a unique pattern of infections by Clostridia and anaerobic organisms in the deep tissues. This pattern of infections is a reminder that rare diseases, such as the multiple causes of hereditary neutropenia, have some common features and their own special sets of symptoms and complications.
Blood neutrophil counts are an extremely important way to monitor patients for the risk of infection. In fact, for the care of cancer patients, it would be very useful to have much more data on the ANC after various chemotherapy regimens with and without monoclonal antibodies and other biological therapies. It would also be helpful to have more counts for the diagnosis of rare diseases such as congenital and cyclic neutropenia. It would be very helpful if there were a simple way to measure the marrow neutrophil reserves, something simpler than the bone marrow aspirate and biopsy, to know the state of the marrow, perhaps by scanning technologies. On the horizon, there are small devices for monitoring blood counts at home – an important new way to try to understand neutropenia.
Conflicts of interest
Dr Dale is a consultant and has received research support from Amgen, Sanofi-Aventis, Merck and Philips. He has recently reviewed safety data for clinical trials sponsored by Galderma, Sanofi-Aventis, Merck, Esai, Drais, Genkyotex and Boerhinger-Ingelheim.