From the Neonatal Intensive Care Unit, Royal Alexandra Hospital; and the Department of Pediatrics, University of Alberta, Edmonton.
Bacterial and fungal sepsis are major causes of morbidity and mortality in the newborn. Multiple factors contribute to this increased susceptibility to infection, including quantitative and qualitative neutrophil defects, with a reduction in neutrophil number and function. Neutropenia in the newborn may occur in association with sepsis and has a poor prognosis. In addition to antibiotic therapy and supportive care, granulocyte colony-stimulating factor (G-CSF) and granulocyte-macrophage colony-stimulating factor (GM-CSF) have been used to reduce morbidity and mortality. Granulocyte CSF is the physiological regulator of neutrophil production and function. Administration of G-CSF results in increased neutrophil production and counts and improved neutrophil function. Several studies of animal and human newborns having neutropenia or suspected sepsis investigated the use of G-CSF and GM-CSF to elevate neutrophil counts and reduce morbidity and mortality in this population. Results of small clinical trials using G-CSF and GM-CSF in very low-birth-weight infants having neutropenia show increased neutrophil counts and a reduced incidence of sepsis during the neonatal period. Despite these promising early results, further studies of the safety and efficacy of G-CSF and GM-CSF administration in neonates are required before their routine use can be recommended as either prophylaxis or treatment for neonatal sepsis.
Bacterial and fungal sepsis are major causes of morbidity and mortality in the newborn. Despite advances in neonatal intensive care in recent decades, sepsis-associated mortality rates in very low-birth-weight infants have remained constant at nearly 15%.1 This plateau in mortality likely reflects an increased susceptibility of the neonate to infection; multiple factors contributing to this include quantitative and qualitative neutrophil defects with reduced numbers of stored and precursor neutrophils per kilogram body weight, delayed upregulation of neutrophil production in response to infection, and reduced neutrophil function.2
Neutropenia in the newborn may occur in association with sepsis or in other situations such as maternal hypertension.3,4 Neutropenia in neonates increases the risk of sepsis and is associated with a poor prognosis.3- 8 As a result, in addition to the standard therapy for neonatal sepsis with antibiotic medications and supportive care, several novel forms of immunotherapy such as granulocyte transfusions and intravenous immunoglobulin administration have been used to reduce mortality—without any proven benefit.9,10 Granulocyte colony-stimulating factor (G-CSF) is the physiological regulator of neutrophil production and function. Its actions include increased neutrophil and neutrophil superoxide production and bactericidal activity.11,12 Several studies of animal and human newborns having neutropenia or suspected sepsis investigated the use of G-CSF and granulocyte-macrophage colony-stimulating factor (GM-CSF) to increase neutrophil counts and reduce morbidity and mortality in this population.
Neutrophils develop from myeloid progenitor cells, passing through a number of stages. Myeloblasts, promyelocytes, and myelocytes all have the capacity for cell division and are referred to as the neutrophil proliferative pool. As neutrophils mature, they lose the capacity for cell division. Metamyelocytes, band neutrophils, and segmented neutrophils are termed the neutrophil storage pool (NSP).13 Results of studies14 in adult humans show that the neutrophil proliferative pool and NSP contain 2×109 and 6×109 cells per kilogram body weight, respectively. Although the size of the NSP and the neutrophil proliferative pool have not been quantified as accurately in neonates as in adults, both pools are smaller in newborns than in adults. Several factors enhance the release of cells from the NSP into the blood, including treatment with G-CSF and corticosteroids.14,15 The equilibrium between circulating and marginated neutrophil pools has been shown experimentally in animals and adult humans to be affected by several factors, such as administration of epinephrine or endotoxin, and in neonates by strenuous crying.14,16 The half-life of neutrophils in the blood is approximately 6 hours because they quickly leave the blood to enter the tissues.17 The length of time that neutrophils spend in the tissues is not completely clear but is affected by the presence of tissue damage or infection.18
Results of studies19 on human abortuses show that virtually no neutrophils are present before 14 weeks' gestation. Subsequently, the neutrophil count rises progressively to term, with a doubling of the absolute neutrophil count from 14 to 18 weeks' gestation and a 4-fold increase by 24 to 32 weeks' gestation.20 At birth, neonates delivered vaginally have higher neutrophil counts than those delivered by cesarean section.21 After birth, neutrophil counts peak between 12 and 24 hours and subsequently decrease, achieving a stable lower value of approximately 1800×109/L by 72 hours of life.22 Neutrophil counts in very low-birth-weight infants have a wider variation and are, on average, lower than in full-term infants, with neutropenia being defined by some authors23 as a concentration less than 1800×109/L at 12 hours and less than 1000×109/L by 48 hours, since this is the lower limit of the 90% reference range.
Results of studies24 in adult animals show that approximately 60 to 90 minutes after inoculation with bacteria, there is neutrophilia with an accompanying increase in band forms. As neutrophils are released from the marrow to the blood, bone marrow NSP decreases. In sepsis, neutrophil production is increased, with an acceleration of cycling of neutrophil precursors. If tissue uptake occurs at a rate greater than neutrophil production and release from the NSP, neutropenia develops. Neutrophil proliferation in neonates is at near-maximum capacity in the normal state, resulting in inadequate reserve production capacity. In human neonates, the latency period for neutrophilia to develop after inoculation is prolonged by an uncertain duration, and a marked increase in neutrophil production is not evident.25
Granulocyte CSF, the primary physiological regulator of neutrophil production, is an 18-kd glycosylated polypeptide consisting of 27 amino acids11 produced primarily by monocytes, macrophages, fibroblasts, and endothelial cells.26 Actions of G-CSF include maturation of committed myeloid progenitors; release of neutrophils from the bone marrow NSP to the blood; and activation of neutrophil functions, including chemotaxis, superoxide generation, phagocytosis, and microbial killing.15 Administration of G-CSF to primates with normal white blood cell counts results in dose-dependent neutrophilia.27
After isolation of murine G-CSF in 1983,28 human G-CSF was isolated in 1985.29 Subsequently, the gene for G-CSF was located on the long arm of chromosome 17.30
Granulocyte CSF has been used in adults having chemotherapy-induced neutropenia since the late 1980s and has been shown to cause a dose-dependent increase in blood neutrophil concentration and to reduce the period of neutropenia.12 Subsequently, G-CSF has been used in children having chemotherapy-induced neutropenia and congenital or acquired neutropenia with similar success.31- 34 Use of G-CSF also hastens myeloid recovery after myeloablative therapy for bone marrow transplantation.35
Granulocyte CSF can be measured in cord blood from full-term and preterm neonates.36 In some studies,36,37 the concentration of G-CSF measured in cord blood or in the immediate postnatal period correlated with gestational age and blood neutrophil concentration. However, other studies38,39 did not find any such correlation. Compared with healthy adults, preterm and term infants without neutropenia have higher G-CSF levels.40,41 Levels are higher in preterm than in term infants.41,42 However, despite markedly elevated G-CSF levels in adults having neutropenia, levels in neonates having neutropenia are not elevated, particularly in preterm infants. Elevated G-CSF levels are associated with infection, fetal distress, premature rupture of membranes, birth asphyxia, vaginal delivery compared with cesarean section, and in the second twin delivered compared with the first.42 Neonates with signs of infection have higher G-CSF levels than healthy neonates but not as high as infected adults.37 These effects are explained by the fact that, although neonatal myeloid progenitor cells are equally responsive as adult cells to the actions of G-CSF, monocytes from neonates generate less G-CSF in response to stimulation than adult monocytes because of posttranscriptional instability rather than diminished G-CSF transcription.38,43,44
In the first study of the use of G-CSF in a human neonate, Roberts et al45 used G-CSF in a 654-g infant—the result of a pregnancy complicated by severe maternal hypertension—who was persistently neutropenic and had 5 episodes of sepsis. Administration of G-CSF (10 µg/d) resulted in neutrophilia without any further episodes of infection. The drug was administered for 7 months without any adverse effects. Gillan et al46 studied 42 neonates suspected of having sepsis in the first 72 hours of life. The infants were randomized to receive either placebo or G-CSF (1, 5, 10, or 20 µg/kg per day), and a dose-dependent increase in blood and bone marrow neutrophils was demonstrated without any adverse effects.46 Bedford Russell et al47 administered G-CSF (5 µg/kg per day for 5 days, increased to 10 µg/kg per day if no response) to 12 preterm infants having neutropenia and suspected of having sepsis, and showed a significant increase in neutrophil counts after a median of 4 days.
Kocherlakota and La Gamma48 evaluated the administration of G-CSF to neonates suspected of having sepsis complicated by neutropenia. They documented increased circulating neutrophil counts. In a subsequent nonrandomized, nonmasked, preliminary study,49 they demonstrated that administration of G-CSF in neonates having prolonged preeclampsia-associated neutropenia significantly reduced the incidence of sepsis in the first 28 days of life, from 54% to 13%. Several other studies50,51 involving small numbers of neonates having neutropenia due to various causes found increased neutrophil counts associated with use of G-CSF at dosages averaging 5 µg/kg per day.
In a recent study, Schibler et al52 randomized 20 neonates having neutropenia and suspected of having sepsis to receive either G-CSF (10 µg/kg/d) or placebo for the first 3 days of life. Bone marrow aspirates were performed, either at study entry or on day 2. Neutrophil counts increased significantly in both groups over time, with no significant differences in neutrophil count, bone marrow NSP, neutrophil proliferative pool, severity of illness, morbidity, or mortality. The lack of effect for G-CSF use may be explained by the fact that endogenous G-CSF levels were elevated in both groups at study entry, and, during the study, there were no significant differences in their G-CSF concentrations.
Results of initial studies53 suggested that administration of G-CSF to a pregnant woman at risk for imminent preterm delivery might result in transplacental passage of G-CSF and an increased neutrophil count in the neonate. The effect was affected by the time between administration of G-CSF and delivery, and may have been confounded by differences in the degree of prematurity between patients and controls.53 Although results of the above-mentioned studies showed that G-CSF increased neutrophil counts in neonates having neutropenia or suspected of having sepsis, there are no randomized placebo-controlled trials on whether G-CSF improves survival for neonates having early-onset bacterial sepsis, or whether G-CSF improves the outcome for neonates having neutropenia.
The physiological features of GM-CSF are similiar to those of G-CSF.54 However, GM-CSF is an earlier and more widely acting cytokine than G-CSF, which induces proliferation and differentiation of multilineage hematopoietic progenitors.54 This increases monocyte, macrophage, and eosinophil counts more than G-CSF but has less effect on neutrophil counts.55 In addition, GM-CSF has a greater effect on chemotaxis and the bactericidal function of neutrophils than does G-CSF.55
Studies on the use of GM-CSF in human newborns are limited. Cairo et al56 administered placebo or GM-CSF (5 or 10 µg/kg per day for 7 days) to 20 very low-birth-weight infants starting in the first 72 hours of life. A significant dose-dependent increase in blood and bone marrow neutrophil counts lasted approximately 5 days after administration of the last dose. Subsequently, this group randomized 61 very low-birth-weight infants to receive either placebo or GM-CSF (8 µg/kg per day for 28 days) starting within 3 days of birth.57 A significant (61%) reduction in the incidence of nosocomial infection occurred in the treatment group.
Although only occasional adverse effects of G-CSF and GM-CSF treatment in neonates have been described, some concerns have been raised.50,51,58,59 Significant reductions in platelet counts have been reported after treatment with G-CSF in several studies in neonates and children. It is not clear whether this thrombocytopenia is an adverse effect of G-CSF administration or whether it may be attributable to associated sepsis and the critically ill state of the population involved.47,60 However, in a study by Donadieu et al,34 thrombocytopenia occurred after G-CSF administration to children having chronic neutropenia in the absence of sepsis. In contrast, GM-CSF treatment seems to cause an elevation in platelet counts.56 In adults, adverse effects include fever and bone pain and are more common with GM-CSF than with G-CSF.61 In addition, a "first-dose" effect of pulmonary sequestration of neutrophils has been described after high-dose intravenous administration of G-CSF and GM-CSF.61 This effect is avoided by slow intravenous infusion or subcutaneous injection.62 Concerns have been expressed that pulmonary sequestration of neutrophils with activation of cytokine cascades could exacerbate short- and long-term lung injury.58 This effect is a particular concern in extremely premature infants having acute respiratory distress syndrome who are already at considerable risk of bronchopulmonary dysplasia. Indeed, a study52 demonstrated a nonsignificant trend toward a slightly higher incidence of chronic lung disease among G-CSF recipients.
A theoretical concern exists that exposure to G-CSF may predispose neonates to leukemia in later life, although no cases directly attributable to G-CSF treatment have been documented to date, to our knowledge. However, a few children have developed leukemia on treatment with G-CSF in Kostmann syndrome where there is primary congenital neutropenia.63 Several factors other than G-CSF administration may contribute to the increased risk of leukemia in patients having Kostmann syndrome, which in itself may be a preleukemic condition. These factors include structural abnormalities in neutrophils and the G-CSF receptor, and an increase in survival by preventing early neutropenia-related septic deaths using long-term administration of G-CSF.64 Some evidence for the safety of G-CSF treatment comes from a 2-year follow-up65 of 21 neonates who received G-CSF for 3 days and subsequently had normal hematologic, immunologic, and neurologic development.
Despite evidence of benefit from animal studies and some data in the human newborn that increased neutrophil counts occur with apparent short- and long-term safety, there is insufficient evidence to date to support the use of G-CSF or GM-CSF in routine clinical practice as prophylaxis against or treatment for sepsis in the newborn, with or without neutropenia. Further randomized controlled trials are required to demonstrate the minimum effective dose with an acceptable safety profile. A role may well emerge for a particular CSF (G-CSF or GM-CSF) in certain subgroups such as extremely low-birth-weight infants having neutropenia. The exact role for G-CSF and GM-CSF treatment in the neonatal intensive care unit awaits further clarification.
Accepted for publication March 11, 1999.
Neither author has any affiliation with or financial interest in any organization or entity that conflicts with or has financial interest in the subject matter discussed in the article.
We thank Paul Byrne, MD, FRCPC, University of Alberta Hospital and University of Alberta, Edmonton, for his critical review of the manuscript.
Reprints: Horacio Osiovich, MD, FRCPC, Neonatal Intensive Care Unit, Royal Alexandra Hospital, 10240 Kingsway Ave, Edmonton, Alberta, Canada T5H 3V9 (e-mail: firstname.lastname@example.org).
Thank you for submitting a comment on this article. It will be reviewed by JAMA Pediatrics editors. You will be notified when your comment has been published. Comments should not exceed 500 words of text and 10 references.
Do not submit personal medical questions or information that could identify a specific patient, questions about a particular case, or general inquiries to an author. Only content that has not been published, posted, or submitted elsewhere should be submitted. By submitting this Comment, you and any coauthors transfer copyright to the journal if your Comment is posted.
* = Required Field
Disclosure of Any Conflicts of Interest*
Indicate all relevant conflicts of interest of each author below, including all relevant financial interests, activities, and relationships within the past 3 years including, but not limited to, employment, affiliation, grants or funding, consultancies, honoraria or payment, speakers’ bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued. If all authors have none, check "No potential conflicts or relevant financial interests" in the box below. Please also indicate any funding received in support of this work. The information will be posted with your response.
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited: 8
Customize your page view by dragging & repositioning the boxes below.
Enter your username and email address. We'll send you a link to reset your password.
Enter your username and email address. We'll send instructions on how to reset your password to the email address we have on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.