Abstract
Background
Granulocyte colony‐stimulating factor (G‐CSF) induces the formation of toxic granulation neutrophils (TGNs), which are found in many inflammatory responses. Cell population data (CPD) may be able to clarify the effect of G‐CSF, and potentially help doctors in discriminating the effect of G‐CSF from other inflammatory situations.
Methods
To achieve this, we performed analyses of leukocyte CPD from normal controls and healthy donors that had received G‐CSF for peripheral blood stem cells (PBSCs) mobilization (G‐CSF group).
Results
Two hundred and seventy‐one subjects were enrolled as normal controls, and 21 subjects were enrolled in the G‐CSF group. Mean volume (MN‐V)‐neutrophils (NE), mean axial light loss (MN‐AL2)‐NE, and all standard deviation (SD) parameters increased significantly, whereas all light scattering parameters, mean median angle light scatter (MN‐MALS)‐NE, mean upper median angle light scatter (MN‐UMALS)‐NE, mean lower median angle light scatter (MN‐LMALS)‐NE, and mean low angle light scatter (MN‐LALS)‐NE reduced significantly in the G‐CSF group. MN‐V‐lymphocytes (LY) from the G‐CSF group showed no significant difference (P = 0.143), whereas MN‐V‐monocytes (MO) were significantly decreased (P < 0.001). Receiver operating characteristic (ROC) curves for the discrimination of the G‐CSF group from normal controls showed excellent sensitivity in SD‐LALS‐NE (at 30.85, sensitivity 95.2%, specificity 76.0%), MN‐AL2‐NE (at 134.5, sensitivity 90.5%, specificity 83.0%), and SD‐AL2‐NE (at 16.4, sensitivity 95.2%, specificity 95.2). Several CPD parameters of lymphocytes and monocytes, as well as neutrophils can be used as markers for determining the effect of G‐CSF.
Conclusion
Our data show that many CPD of leukocytes can be considered to be useful parameters of the effect of G‐CSF. J. Clin. Lab. Anal. 26:194‐199, 2012. © 2012 Wiley Periodicals, Inc.
Keywords: granulocyte colony‐stimulating factor, leukocyte cell population, toxic granulation neutrophils
INTRODUCTION
Granulocyte colony‐stimulating factor (G‐CSF) has sometimes been used after chemotherapy and allogeneic stem‐cell transplantation (SCT) to shorten the neutropenic phase 1, 2, 3. It was reported that G‐CSF induced the formation of toxic granulation neutrophils (TGNs), which have prominent azurophilic cytoplasmic granules in blood smears stained by the Wright or May‐Grϋnwald‐Giemsa technique 4, 5. However, these changes in neutrophils can also be detected in several inflammatory responses including infection, tissue damage, and other inflammatory diseases 6, 7, 8. In addition, the degree of toxic change and neutrophil size was variable; some neutrophils showed toxic changes whereas others did not, even within the same samples. Therefore, microscopic examination by experts can be subjective, and not reproducible 9. In particular, specific features in samples after G‐CSF are more difficult to quantify with the naked eye alone. Recently, the automatic blood cell analyzer DxH800 (Coulter Inc., Miami, FL) has been shown to be effective, revealing significant cell population data (CPD) for each subpopulation of white blood cells (WBCs).
In this study, we analyzed leukocyte CPD of normal controls and healthy donors that had received G‐CSF for peripheral blood stem cells (PBSCs) mobilization to clarify the effect of G‐CSF on the CPD. The purpose of our study was to determine the usefulness of CPD as a marker for determining the effect of G‐CSF.
MATERIALS AND METHODS
Two hundred and seventy‐one subjects without hematological disorders (157 males and 114 females from 14 to 80 years of age) were enrolled as normal controls. Twenty‐one healthy donors that received G‐CSF (G‐CSF group) for PBSCs mobilization (10 males and 11 females from 28 to 71 years of age) were selected to investigate the effects of G‐CSF on CPD. The donor work‐up included medical history, physical examination, ECG, chest x‐ray, blood count, blood chemistry, coagulation screening, and testing for infectious disease markers. Samples with infection were excluded based on medical history, physical examination, and lab findings, and normal controls were selected since an infection caused toxic changes in the neutrophils and the presence of band neutrophils in the blood, which were translated into a change of CPD. The blood samples were analyzed using DxH800, and complete blood count (CBC) and CPD of neutrophils, lymphocytes, and monocytes were obtained.
The CPD of the leukocytes are listed in Table 1. The mean (MN) and standard deviation (SD) values of volume, conductivity, and five‐angle light scattering parameters, corresponding to a total of 14 parameters, were collected. The light scattering parameters were as follows: median angle light scatter (MALS), upper median angle light scatter (UMALS), lower median angle light scatter (LMALS), low angle light scatter (LALS), and axial light loss (AL2). The light scattering parameters reflected cellular complexity, granularity, and nuclear structure, but AL2 reflected cell size on the basis of absorbed light. The protocols of this study were approved by the Catholic Medical Center Institutional Review Board.
Table 1.
CPD of Neutrophils, Lymphocytes, and Monocytes and Abbreviations
| Neutrophils | Lymphocytes | Monocytes | |||
|---|---|---|---|---|---|
| Mean volume | MN‐V‐NE | Mean volume | MN‐V‐LY | Mean volume | MN‐V‐MO |
| SD volume | SD‐V‐NE | SD volume | SD‐V‐LY | SD volume | SD‐V‐MO |
| Mean conductivity | MN‐C‐NE | Mean conductivity | MN‐C‐LY | Mean conductivity | MN‐C‐MO |
| SD conductivity | SD‐C‐NE | SD conductivity | SD‐C‐LY | SD conductivity | SD‐C‐MO |
| Mean median angle light scatter | MN‐MALS‐NE | Mean median angle light scatter | MN‐MALS‐LY | Mean median angle light scatter | MN‐MALS‐MO |
| SD median angle light scatter | SD‐MALS‐NE | SD median angle light scatter | SD‐MALS‐LY | SD median angle light scatter | SD‐MALS‐MO |
| Mean upper median angle light scatter | MN‐UMALS‐NE | Mean upper median angle light scatter | MN‐UMALS‐LY | Mean upper median angle light scatter | MN‐UMALS‐MO |
| SD upper median angle light scatter | SD‐UMALS‐NE | SD upper median angle light scatter | SD‐UMALS‐LY | SD upper median angle light scatter | SD‐UMALS‐MO |
| Mean lower median angle light scatter | MN‐LMALS‐NE | Mean lower median angle light scatter | MN‐LMALS‐LY | Mean lower median angle light scatter | MN‐LMALS‐MO |
| SD lower median angle light scatter | SD‐LMALS‐NE | SD lower median angle light scatter | SD‐LMALS‐LY | SD lower median angle light scatter | SD‐LMALS‐MO |
| Mean low angle light scatter | MN‐LALS‐NE | Mean lower angle light scatter | MN‐LALS‐LY | Mean lower angle light scatter | MN‐LALS‐MO |
| SD low angle light scatter | SD‐LALS‐NE | SD lower angle light scatter | SD‐LALS‐LY | SD lower angle light scatter | SD‐LALS‐MO |
| Mean axial light loss | MN‐AL2‐NE | Mean axial light loss | MN‐AL2‐LY | Mean axial light loss | MN‐AL2‐MO |
| SD axial light loss | SD‐AL2‐NE | SD axial light loss | SD‐AL2‐LY | SD axial light loss | SD‐AL2‐MO |
SD, standard deviation.
PBSCs Mobilization Procedure from Healthy Donor
The 21 healthy donors were all genetically related siblings of the patients who underwent allogeneic SCT for hematologic malignancies at the Catholic Blood and Marrow Transplantation Center between 2009 and 2010. All donors were mobilized with G‐CSF (Filgrastim, Dong‐A Pharmaceutical, Seoul) at doses of 10 μg/kg/day, subcutaneously once a day. G‐CSF was administered for 4 consecutive days and leukapheresis procedures were commenced on Day +4 of the G‐CSF administration. The analyses of CPD were done from a blood sample on Day +4, before leukapheresis.
Statistical Analysis
The average values of CPD parameters are expressed as MN ± SD. To determine differences in effects between G‐CSF and normal controls, statistical analyses were conducted using Mann–Whitney's U‐test. A P‐value of ≤0.05 was regarded as statistically significant. We plotted receiver operating characteristic (ROC) curves to determine the area under the curve (AUC) with cutoff levels for the optimal combination of sensitivity and specificity. The sensitivity and specificity of each parameter for differences in effects between G‐CSF and normal controls were analyzed. Statistical studies were performed using the Statistical Package for the Social Sciences, version 13.0 (SPSS, Inc, Chicago, IL).
RESULTS
Neutrophil CPD Analyses between Normal Control and the G‐CSF Group
Table 2 shows the results of CPD analyses between normal controls and the G‐CSF group. We noted significant differences in all neutrophil CPD. MN‐V‐NE and MN‐AL2‐NE significantly increased in the G‐CSF group relative to the normal controls (P < 0.001). All other light scattering parameters, MN‐MALS‐NE, MN‐UMALS‐NE, MN‐LMALS‐NE, and MN‐LALS‐NE significantly reduced in the G‐CSF group (P < 0.001). In addition, MN‐C‐NE was significantly decreased in the G‐CSF group (P < 0.001). All SD parameters, including SD‐C‐NE increased significantly in the G‐CSF group (P < 0.001). Figure 1 shows ROC of neutrophil CPD for discrimination of the G‐CSF group from normal controls.
Table 2.
The leukocytes CPD of G‐CSF group compared with normal control
| Neutrophil | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Type | MN_V_NE | SD_V_NE | MN_C_NE | SD_C_NE | MN_MALS_NE | SD_MALS_NE | MN_UMALS_NE | SD_UMALS_NE | MN_LMALS_NE | SD_LMALS_NE | MN_LALS_NE | SD_LALS_NE | MN_AL2_NE | SD_AL2_NE |
| Normal (n = 271) | 154.4 ± 8.1 | 19.9 ± 2.7 | 145.0 ± 10.6 | 6.5 ± 1.9 | 138.4 ± 4.8 | 10.3 ± 1.2 | 140.3 ± 9.1 | 11.8 ± 2.2 | 129.9 ± 10.4 | 13.6 ± 2.4 | 156.7 ± 12.1 | 29.5 ± 4.1 | 129.2 ± 5.6 | 13.1 ± 2.6 |
| G‐CSF group (n = 21) | 177.2 ± 9.1 | 32.0 ± 1.6 | 137.6 ± 5.6 | 8.2 ± 1.3 | 123.9 ± 6.7 | 13.9 ± 1.6 | 134.5 ± 6.7 | 18.0 ± 4.7 | 107.9 ± 9.6 | 18.5 ± 2.5 | 123.5 ± 26.8 | 42.5 ± 6.4 | 141.2 ± 5.8 | 19.2 ± 1.5 |
| P‐value a | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 |
| Lymphocyte | ||||||||||||||
| Type | MN_V_LY | SD_V_LY | MN_C_LY | SD_C_LY | MN_MALS_LY | SD_MALS_LY | MN_UMALS_LY | SD_UMALS_LY | MN_LMALS_LY | SD_LMALS_LY | MN_LALS_LY | SD_LALS_LY | MN_AL2_LY | SD_AL2_LY |
| Normal (n = 271) | 89.3 ± 4.0 | 14.4 ± 1.7 | 117.9 ± 8.1 | 9.6 ± 2.4 | 75.3 ± 4.9 | 15.6 ± 1.4 | 78.4 ± 8.6 | 18.2 ± 2.0 | 64.2 ± 5.8 | 18.4 ± 1.9 | 30.6 ± 1.5 | 9.3 ± 1.1 | 55.4 ± 4.2 | 9.6 ± 0.8 |
| G‐CSF group (n = 21) | 90.2 ± 6.1 | 19.7 ± 3.7 | 114.5 ± 3.0 | 11.9 ± 1.6 | 86.5 ± 4.1 | 18.2 ± 2.7 | 91.1 ± 7.5 | 21.2 ± 3.4 | 74.1 ± 2.8 | 20.7 ± 2.9 | 30.2 ± 2.7 | 10.7 ± 2.0 | 61.5 ± 4.9 | 12.0 ± 1.9 |
| P‐value a | 0.143 | <0.001 | 0.009 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | 0.368 | 0.001 | <0.001 | <0.001 |
| Monocyte | ||||||||||||||
| Type | MN_V_MO | SD_V_MO | MN_C_MO | SD_C_MO | MN_MALS_MO | SD_MALS_MO | MN_UMALS_MO | SD_UMALS_MO | MN_LMALS_MO | SD_LMALS_MO | MN_LALS_MO | SD_LALS_MO | MN_AL2_MO | SD_AL2_MO |
| Normal (n = 271) | 171.4 ± 6.4 | 18.9 ± 2.0 | 125.2 ± 8.9 | 6.4 ± 2.3 | 88.5 ± 4.2 | 11.2 ± 1.4 | 99.7 ± 7.5 | 12.3 ± 1.9 | 72.4 ± 6.0 | 14.2 ± 1.8 | 77.3 ± 9.4 | 23.4 ± 4.1 | 104.8 ± 7.1 | 12.9 ± 3.2 |
| G‐CSF group (n = 21) | 169.3 ± 12.5 | 32.8 ± 6.6 | 115.6 ± 8.4 | 15.0 ± 9.7 | 89.1 ± 12.7 | 11.9 ± 2.9 | 100.7 ± 14.6 | 18.2 ± 6.3 | 71.3 ± 11.8 | 15.2 ± 2.9 | 66.6 ± 18.4 | 27.4 ± 4.8 | 110.2 ± 18.3 | 23.1 ± 8.8 |
| P‐value a | <0.001 | <0.001 | <0.001 | <0.001 | 0.040 | 0.297 | 0.025 | <0.001 | 0.411 | 0.072 | 0.004 | <0.001 | 0.003 | <0.001 |
G‐CSF, granulocyte colony‐stimulating factor.
Statistical significances were tested by Mann–Whitney's U‐test.
Figure 1.
ROC curves of neutrophil CPD for discrimination of the G‐CSF group from normal controls. SD‐LALS‐NE at 30.85, sensitivity 95.2%, and specificity 76.0%; MN‐AL2‐NE at 134.5, sensitivity 90.5%, and specificity 83.0%; SD‐AL2‐NE at 16.4, sensitivity 95.2%, and specificity 95.2%.
Lymphocyte CPD Analyses between Normal Controls and the G‐CSF Group
For the analyses of lymphocyte, all lymphocyte CPD except MN‐V‐LY, MN‐LALS‐LY revealed significant differences between the G‐CSF group and normal controls. There was no significant difference in MN‐V‐LY of the G‐CSF group relative to the normal controls (P = 0.143). However, MN‐AL2‐LY significantly increased in the G‐CSF group relative to the normal controls (P < 0.001). All light scattering parameters except MN‐LALS‐LY were significantly higher in the G‐CSF group than in the normal control. MN‐C‐NE significantly increased in the G‐CSF group (P = 0.009). All SD parameters including SD‐C‐NE increased significantly in the G‐CSF group (P < 0.001). Figure 2 shows ROC of lymphocyte CPD for discrimination of the G‐CSF group from normal control.
Figure 2.
ROC curves of lymphocyte CPD for discrimination of the G‐CSF group from normal controls. MN‐MALS‐LY at 78.5, sensitivity 95.2%, and specificity 76.4%; MN‐UMALS‐LY at 87.5, sensitivity 85.7%, and specificity 92.3%; MN‐LMALS‐LY at 69.5, sensitivity 95.2%, and specificity 77.1%.
Monocyte CPD Analyses between Normal Controls and the G‐CSF Group
For the analyses of monocytes, MN‐V‐MO decreased significantly in the G‐CSF group relative to the normal controls (P < 0.001). However, MN‐AL2‐MO significantly increased in the G‐CSF group relative to the normal controls (P = 0.003), whereas MN‐LALS‐MO reduced significantly in the G‐CSF group (P = 0.004). However, other light scattering parameters, MN‐MALS‐MO, MN‐UMALS‐MO were significantly higher in the G‐CSF group. Among SD parameters, SD‐V‐MO, SD‐C‐MO, SD‐UMALS‐MO, SD‐LALS‐MO, SD‐AL2‐MO significantly increased in the G‐CSF group. Figure 3 shows the ROC of monocyte CPD for discrimination of the G‐CSF group from normal controls.
Figure 3.
ROC curves of monocyte CPD for discrimination of the G‐CSF group from normal controls. SD‐V‐MO at 26.8, sensitivity 95.2%, and specificity 100%; SD‐C‐MO at 6.90, sensitivity 76.2%, and specificity 72.7%; SD‐AL2‐MO at 15.1, sensitivity 81.0%, and specificity 93.7%.
DISCUSSION
G‐CSF has been identified as a glycoprotein that stimulates the survival, proliferation, differentiation, and functional activation of neutrophilic granulocytes both in vivo and in vitro 10, 11. When the serum concentration of G‐CSF is increased, it is known that TGNs increase in the blood. Takeshita at al. reported that by using a high concentration of G‐CSF (10–100 μg/mL), G‐CSF‐induced TGN formation 4. Another study using physiological levels of G‐CSF showed that neutrophils changed to TGN‐like cells in response to 1000 pg/mL of G‐CSF added in peripheral blood from a healthy subject 5.
However, since all previous studies were performed with microscopic examinations using the naked eye alone, it was difficult to accurately quantify changes. In particular, it would be impossible to detect changes in lymphocytes and monocytes. In this context, we performed analyses of leukocyte CPD to determine the usefulness of CPD as a marker for investigating the effect of G‐CSF.
In this study, we noted significant differences between samples from healthy donors that had received G‐CSF (G‐CSF group) and normal controls with regards to MN‐V‐NE and SD‐V‐NE. This showed that the neutrophils of subjects in the G‐CSF group were larger and their size variations were greater than in the normal controls. Meanwhile, the monocyte volume parameter (MN‐V‐MO) of the G‐CSF group decreased and there was no difference in the lymphocyte volume parameter (MN‐V‐LY), even though large variations in size (SD‐V‐LY, SD‐V‐MO) were observed. These CPD of monocytes and lymphocytes make it possible to compare the change of morphology by administration of G‐CSF, which is impossible by microscopic examinations. All lymphocyte SD parameters increased in the G‐CSF group compared with normal controls. This showed that certain lymphocytes were altered as a result of G‐CSF administration. In this study, we used samples after applying extrinsic G‐CSF with a dose of 10 μg/kg/day for 4 consecutive days. However, the serum concentration of G‐CSF can also increase during inflammation responses. Kawakami et al. examined serum G‐CSF levels in patients with infections and reported, the serum G‐CSF level in the acute stage of infection was significantly higher than after the recovery phase, which was the same as the level in normal controls 12. Although we excluded samples that had infection, the possibility of including undetected infection remains. Thus, further analyses of leukocyte CPD according to serum G‐CSF levels including physiological levels are needed.
Interestingly, in blood samples of G‐CSF group, many neutrophil CPD were similar to the CPD of patients with acute bacterial infection. One possible hypothesis to explain this similarity is that TGNs also increase after an inflammatory response, which is the defense mechanism against infection and tissue damage. Chaves et al. showed a significant increase of neutrophil volume and significant decrease of neutrophil light scatter by the Coulter‐automated hematology analyzer in bacteremic cases 13. Another study from Chaves at al. in 70 patients with positive blood cultures for bacteria reported that the neutrophil volume distribution width significantly increased in the bacteremic patients (P < 0.001) 14. Our previous analyses in 117 patients with bacterial sepsis and MN‐V‐NE and MN‐AL2‐NE were significantly increased in the sepsis patients relative to the normal controls (P < 0.05) (data not shown). However, several lymphocyte and monocyte CPD of the G‐CSF group were different from those of previously analyzed sepsis samples (unpublished data). Particularly with sepsis samples, MN‐V‐LY and MN‐V‐MO were significantly increased, whereas MN‐V‐LY of the G‐CSF group was not significantly different compared with normal controls, and MN‐V‐MO was significantly decreased. These results may be due to certain lymphocytes and monocytes in the circulating blood after bacterial infection being different from those influenced by G‐CSF.
Although toxic granulation changes of neutrophils in patients with infection could be caused by increasing the intrinsic G‐CSF, differences in the volumetric parameters in lymphocytes and monocytes of patients with infection is distinct from the effect of increasing G‐CSF. The clinical importance of these results is that G‐CSF was used in patients that underwent chemotherapy or allogeneic SCT and may have a high risk of infection. This suggests that CPD parameters could be used as an indicator to discriminate the effects of G‐CSF from acute bacterial infection, particularly when the patients received G‐CSF.
In summary, our data show that several CPD of lymphocyte, monocytes, and neutrophils can be considered useful parameters for determining the effect of G‐CSF. If these parameters are incorporated into a decision rule, they should help doctors in discriminating the effect of G‐CSF from other inflammatory situations. Further studies aimed at verifying the CPD according to the concentration of G‐CSF, and defining potential factors affecting leukocyte CPD are warranted.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
REFERENCES
- 1. Bishop MR, Tarantolo SR, Geller RB, Lynch JC, Bierman PJ, Pavletic ZS, Vose JM, Kruse S, Dix SP, Morris ME, Armitage JO, Kessinger A. A randomized, double‐blind trial of filgrastim (granulocyte colony‐stimulating factor) versus placebo following allogeneic blood stem cell transplantation. Blood 2000;96(1):80–85. [PubMed] [Google Scholar]
- 2. Dallorso S, Rondelli R, Messina C, Pession A, Giorgiani G, Fagioli F, Locatelli F, Manzitti C, Balduzzi A, Prete A, Cesaro S, Lanino E, Dini G. Clinical benefits of granulocyte colony‐stimulating factor therapy after hematopoietic stem cell transplant in children: Results of a prospective randomized trial. Haematologica 2002;87(12):1274–1280. [PubMed] [Google Scholar]
- 3. Ozcan M, Ustun C, Akcaglayan E, Akan H, Arslan O, Ilhan O, Beksac M, Gurman G, Demirer T, Arat M, Celebi H, Konuk N, Uysal A, Koc H. Recombinant human granulocyte colony‐stimulating factor (rh‐G‐CSF) may accelerate hematopoietic recovery after HLA‐identical sibling allogeneic peripheral blood stem cell transplantation. Bone Marrow Transplant 2001;27(5):499–505. [DOI] [PubMed] [Google Scholar]
- 4. Takeshita S, Sekine I, Fujisawa T, Yoshioka S. Studies of peripheral blood toxic neutrophils as a predictor of coronary risk in Kawasaki disease—The pathogenetic role of hematopoietic colony‐stimulating factors (GM‐CSF, G‐CSF). Acta Paediatr Jpn 1990;32(5):508–514. [DOI] [PubMed] [Google Scholar]
- 5. Kabutomori O, Kanakura Y, Watani YI. Induction of toxic granulation in neutrophils by granulocyte colony‐stimulating factor. Eur J Haematol 2002;69(3):187–188. [DOI] [PubMed] [Google Scholar]
- 6. Kabutomori O, Kanakura Y, Iwatani Y. 2002. Movement of toxic granulation neutrophils and C‐reactive protein in inflammatory processes. Am J Med 2002;112(7):595–596. [DOI] [PubMed] [Google Scholar]
- 7. Shapiro MF, Greenfield S. The complete blood count and leukocyte differential count. An approach to their rational application. Ann Intern Med 1987;106(1):65–74. [DOI] [PubMed] [Google Scholar]
- 8. Cornbleet PJ. Clinical utility of the band count. Clin Lab Med 2002;22(1):101–136. [DOI] [PubMed] [Google Scholar]
- 9. Pierre RV. Peripheral blood film review. The demise of the eyecount leukocyte differential. Clin Lab Med 2002;22(1):279–297. [DOI] [PubMed] [Google Scholar]
- 10. Platzer E, Welte K, Gabrilove JL, Lu L, Harris P, Mertelsmann R, Moore MA. Biological activities of a human pluripotent hemopoietic colony stimulating factor on normal and leukemic cells. J Exp Med 1985;162(6):1788–1801. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Souza LM, Boone TC, Gabrilove J, Lai PH, Zsebo KM, Murdock DC, Chazin VR, Bruszewski J, Lu H, Chen KK, Barendt J, Platzer E, Moore MAS, Mertelsmann R, Welte K. Recombinant human granulocyte colony‐stimulating factor: effects on normal and leukemic myeloid cells. Science 1986;232(4746):61–65. [DOI] [PubMed] [Google Scholar]
- 12. Kawakami M, Tsutsumi H, Kumakawa T, Abe H, Hirai M, Kurosawa S, Mori M, Fukushima M. Levels of serum granulocyte colony‐stimulating factor in patients with infections. Blood 1990;76(10):1962–1964. [PubMed] [Google Scholar]
- 13. Chaves F, Tierno B, Xu D. Quantitative determination of neutrophil VCS parameters by the Coulter automated hematology analyzer: New and reliable indicators for acute bacterial infection. Am J Clin Pathol 2005;124(3):440–444. [DOI] [PubMed] [Google Scholar]
- 14. Chaves F, Tierno B, Xu D. Neutrophil volume distribution width: A new automated hematologic parameter for acute infection. Arch Pathol Lab Med 2006;130(3):378–380. [DOI] [PubMed] [Google Scholar]