Osamu Kabutomori

Peripheral large granular lymphocytes in normal pregnant and postpartum women: Decrease in late pregnancy and dynamic change in the puerperium

Large granular lymphocytes (LGLs) have a variety of cytotoxic activities of NK, K and cytotoxic T lymphocytes, suggesting that their morphology is indicative of lytic function. In non-pregnant normal control women (n = 48), the number of LGLs was 0.30 +/- 0.14 x 10(9)/l and the proportion of LGLs in their peripheral lymphocyte fraction was 14.0 +/- 5.4%. The number and proportion of LGLs were significantly decreased in the third trimester of pregnancy (n = 32; 0.19 +/- 0.08 x 10(9)/l, P less than 0.01, and 11.7 +/- 3.8%, P less than 0.05), although an unexpected increase in the proportion of LGLs was observed in the first trimester of pregnancy (n = 24; 17.5 +/- 6.5%, P less than 0.05). After delivery, the number and proportion of LGLs increased rapidly to restore the non-pregnant levels and showed a marked increase in LGL count 4 months postpartum. These data suggest that lymphocyte-mediated cytotoxicity decreases in late pregnancy and increase dynamically after delivery to restore the non-pregnant state.

Increase in platelet-large cell ratio in chronic myeloid leukemia.

Thrombocytosis is defined as a platelet count of more than 400×10/l [1]. Differential diagnosis of thrombocytosis is not always obvious. Recently, automated hematological analysis has been common, and various parameters of blood cells can be used. However, there are many parameters whose significance are still unknown. Platelet–large cell ratio (P–LCR) is one of them, and is defined as the ratio of large platelets to total platelets, and is routinely performed by the automated hematological analyzer Sysmex NE-8000 (Sysmex, Kobe, Japan). In this parameter, a large platelet is defined as a platelet with a volume of 12 fl or more [2]. In this study, we examined the clinical usefulness of P–LCR in the differential diagnosis of thrombocytosis. The subjects studied were 50 patients with reactive thrombocytosis (RT), 30 patients with chronic myeloid leukemia (CML) (chronic phase), 15 patients with essential thrombocythemia (ET), and 5 patients with polycythemia vera (PV). P–LCR was higher in patients with CML (27.7 7.9 [mean S.D.]%; P 0.001), patients with ET (17.7 3.6%; P 0.002), and patients with PV (20.8 2.8%; P 0.005) than in patients with RT (13.4 4.1%) (Fig. 1). These findings are compatible with those of previous reports in which the number of large platelets increased in chronic myeloproliferative disorders, but not in RT [3,4]. Furthermore, the P– LCR was higher in patients with CML than in patients with RT (P 0.001), ET (P 0.001) or PV (P 0.001). When the mean+3 S.D. of the P–LCR in 50 patients with RT was determined as the ‘cut-off’ value for differential diagnosis of thrombocytosis between RT and CML, half of the P–LCRs (15/30) were above the ‘cut-off’ value in patients with CML. All of the P– LCRs were under the ‘cut-off’ in patients with RT, ET, or PV. Therefore, these data indicate the ‘cut-off’ value, which is the mean+3 S.D. of the P–LCR in RT patients is useful for the screening of CML in thrombocytosis.

Decrease in the Density of IgG-Fc Receptor III (CD16) on 'Toxic' Neutrophils

Osamu KabutomorI, Central Laboratory for Clinical Investigation, Osaka University Hospital, 1-1-50 Fukushima, Fukushima-ku Osaka 553 (Japan) Changes in human peripheral blood neutrophils frequently occur during serious bacterial infections and certain other inflammatory states [1]. Best known is the ‘shift to the left’ defined by the presence of band forms of neutrophils, metamyelocytes, and sometimes myelocytes [2]. ‘Toxic’ neutrophils, which have azurophilic cytoplasmic granules in blood smears stained by the Wright or May-Grünwald-Giemsa stain [3], are another sign of bacterial infection in the blood. The migration, chemotaxis, phagocytosis, and bactericidal activity (measured by counting of the number of colonies formed by surviving bacteria) of toxic neutrophils are less than those of normal neutrophils [4], but myelo-peroxidase activity, a major bactericidal factor in phagocytes, is higher in toxic neutrophils [5, 6]. Therefore, the decrease in bactericidal activity observed in such neutrophils may arise from the decrease in phagocytosis. Much has been reported about the relationship between cell function and cell surface antigens, and measurement of the density of cell surface antigens is useful for the diagnosis of some pathological conditions [7, 8]. The cell surface antigen(s) related to phagocytosis is unknown, but it seems that the CD16 antigen, which is an IgG-Fc receptor [9], is important for immunophagocytosis by neutrophils [10]. In this study, we examined the relationship between the density of CD16 antigen on neutrophils and the number of toxic neutrophils in patients with bacterial infections. In a sample, the mean fluorescence intensity of CD16+ neutrophils stained with an anti-CDló monoclonal antibody conjugated with fluorescein isothiocyanate reflects the density of CD16 antigen on neutrophils. This intensity was exu 500S. o .¤ > c 3⁄4 φ is ¢ø Φ

Induction of toxic granulation in neutrophils by granulocyte colony‐stimulating factor

To the Editor: Toxic granulation neutrophils (TGNs), which have prominent azurophilic cytoplasmic granules in blood smears stained by the Wright or May– Grünwald–Giemsa technique, appear in blood when inflammation occurs. Toxic granulation in neutrophils is an increase in acid mucosubstance in azurophilic granules (1). There is a positive correlation between the levels of C-reactive protein and the percentages of TGNs in blood samples from patients with inflammatory diseases (2), and so TGNs are one of the important inflammation makers (3). Inflammatory response is the mechanism by which the body defends against infection and repairs tissue damage, and the serum concentration of granulocyte colony-stimulating factor (G-CSF) and interleukin-6 increases in inflammation. It was reported that G-CSF induced TGN formation, but the concentration of G-CSF used in the study (10–100 lg mL) (4) was markedly high, as compared with that in blood from patients with infection (30–3199 pg mL) (5). In this study therefore we examined the effect of physiological levels of G-CSF on TGN formation. Neutrophils changed to TGN-like cells in response to 1000 pg mL of G-CSF added in peripheral blood from a healthy subject. Both the size and the number of azurophilic granules in those TGN-like cells were smaller [Fig. 1A(a)] than those in TGNs in peripheral blood from a patient with inflammatory disease [Fig. 1A(b)]. The formation of TGNlike cells was examined 30, 60, 90, and 120 min after the addition of 100, 500, and 1000 pg mL of G-CSF to peripheral blood samples from six healthy subjects. TGNs or TGN-like cells were counted on blood films stained by the May– Grünwald–Giemsa technique. Two hundred mature neutrophils were examined (Magnification ·1000), and the results were expressed as the percentage of TGN-like cells in mature neutrophils. The percentages of TGN-like cells increased in response to G-CSF in both a doseand timedependent manner (Fig. 1B). We observed increases of TGNs in peripheral blood from three patients during treatment with recombinant human G-CSF to support myelopoiesis under cancer chemotherapy (data not shown). These data indicate that G-CSF induces toxic granules in neutrophils. The significance of toxic granules induced by G-CSF in inflammation is still

Characteristic Changes in Platelet-Large Cell Ratio, Lactate Dehydrogenase and C-Reactive Protein in Thrombocytosis-Related Diseases

We examined the clinical usefulness of 3 parameters of routine laboratory tests [platelet-large cell ratio (P-LCR), lactate dehydrogenase (LDH) and C-reactive protein (CRP)] in 84 patients with thrombocytosis-related diseases (reactive thrombocytosis, chronic myeloid leukemia, essential thrombocythemia and polycythemia vera). These thrombocytosis-related diseases were characterized using the 3 parameters P-LCR, LDH and CRP as follows: high P-LCR and high LDH in chronic myeloid leukemia; high CRP in reactive thrombocytosis; slightly high P-LCR and high LDH in essential thrombocythemia and polycythemia vera. For essential thrombocythemia and polycythemia vera, levels of P-LCR and CRP were nearly identical, but the LDH level in essential thrombocythemia was significantly higher than in polycythemia vera. These characteristics of P-LCR, LDH and CRP may be useful for simple and very rough differentiation of the thrombocytosis-related disease mentioned above.

Unusual eosinophilia not detected by an automated haematological analyser in a patient with liver cirrhosis.

The use of automated haematological analysers to differentiate leucocytes has become more widespread. Unusual eosinophilia in a 57 year old man with liver cirrhosis, caused by hepatitis C infection, and abnormal blood counts detected using a manual method (eosinophils, 50%) was not detected by an automated analyser using the electrical impedance method (0.3%) or the optical method (14.1%). It is important to check blood films when cell counts are apparently abnormal, even for automated haematological examination.

Incorrect measurement of leukocyte counts in post-bone marrow transplantation (P-BMT) patients

Recent hematological analyzers [1] can measure a very small number of leukocytes (10/ml), but some types of analyzers do not wash the detection chamber between each measurement. Leukocyte carryover is negligible in hematological analysis for patients with large numbers of leukocytes (>1000/ml) [2]. However, there is no study investigating leukocyte carryover in patients with small numbers of leukocytes (<500/ml). Recently, the number of patients with leukocytopenia has been increasing as intensive therapies such as post-bone marrow transplantation (post-BMT) and chemotherapy prevail. Therefore, we examined the effect of leukocyte carryover on the measurement of leukocyte number in patients with postBMT. Ethylenediaminetetraacetate (EDTA)-2K-anticoagulated venous blood samples were obtained from specimens sent to the hematology laboratory for routine testing. Using the hematological analyzer Sysmex NE8000 (Kobe, Japan) with the impedance method, we examined the reproducibilities of measurements of small numbers of leukocytes from seven post-BMT patients by washing the detection chamber between each measurement. Furthermore, we examined the effect of leukocyte carryover on the measurement of a small number of leukocytes from three post-BMT patients by not washing the detection chamber after measurement of each routine sample. Standard deviations (SDs) of the measurements of a small number of leukocytes (means: 20/ml–400/ml) from seven post-BMT patients were very small (10/ml–20/ ml) when we washed the detection chamber between each measurement. Therefore, the washing procedure facilitates correct and stable measurement of small numbers of leukocytes in the hematological analyzer. Without the washing procedure between each measurement, however, the numbers of leukocytes from three post-BMT patients increased after measurement of routine samples (Table 1). The effect of leukocyte carryover was larger in routine samples with larger numbers of leukocytes (Table 1). An ascending trend of leukocyte numbers means the recovery of leukopoiesis in post-BMT patients, even if the number of leukocytes is small [3]. Therefore, it is important for the correct and stable measurement of leukocyte numbers in patients with leukocytopenia such as post-BMT patients to remove the effect of leukocyte carryover from routine samples by washing the detection chamber before each measurement.

Phagocytosis decreases the density of IgG-Fc receptor III on neutrophils.

ly occur during serious bacterial infections and in inflammatory states [1]. The best-known change is the “shift to the left,” defined by the presence of a band of neutrophils, metamyelocytes, and sometimes myelocytes in the blood [2]. Cytoplasmic vacuolation is another sign of bacterial infection in the blood, and it is one of the characteristics of toxic neutrophils in blood smears stained by the Wright or May–Giemsa stain [3, 4]. Phagocytosis is weaker in toxic neutrophils than in normal neutrophils [5]. This may be explained by our previous finding that the density of the CD16 antigen, which is an IgG-Fc receptor III, is decreased on ‘toxic’ neutrophils [6]. However, it has been reported that the density of the CD16 surface antigen on neutrophils is increased by stimulation with formylnorleucylleucylphenylalanine (fNLLP) [7]. In this study, we examined whether or not phagocytosis decreases the density of the CD16 antigen on neutrophils. The density of the CD16 antigen on phagocytic neutrophils was measured in ten normal subjects before and after incubation of 1 ml of heparinized whole blood with latex particles (Ø 0.81 μm, Bacto, USA) for 2 h at 37 °C. The mean fluorescence intensity of CD16 positive neutrophils stained with an anti-CD16 monoclonal antibody reflects the density of the CD16 antigen on neutrophils [6]. This density was examined with flow cytometry (FACScan, Becton Dickinson, USA). The density of the CD16 antigen on neutrophils after phagocytosis (mean±SD; 529±209, P<0.005) was significantly lower than that before phagocytosis (755±230) (Fig. 1).