Sage Journals: Discover world-class research

Abstract

Background and Aims:

Large unstained cells (LUCs) are peroxidase negative mononuclear cells which include large lymphocytes in healthy individuals, peroxidase negative blasts, plasma cells and atypical lymphoid cells. This study aims to assess the causes for elevated LUC% in ADVIA 2120i by rapid peripheral smear review.

Materials and Methods:

All consecutive anticoagulated blood samples with LUC% >4%, from June 2022 to March 2024, were processed in ADVIA 2120i, and demographic factors, LUC parameters and peripheral smear observations were documented under six diagnostic disease categories. Statistical analysis was performed using SPSS software (IBM, 28.0).

Results:

The median LUC count of 0.5 × 10³ cells/µL with an inter-quartile range of 0.2–2.3 × 10³ cells/µL of blood was observed in haematolymphoid neoplasms. There was a significant overlap of %LUC values among different categories. The majority (69.6%) had % LUC between 4.1% and 9.9%. Marked elevations were observed in AML, CML, ALL and infections.

Conclusion:

Absolute LUC count could be used as a cost-effective and readily available marker to identify a case of clinically suspicious haematolymphoid neoplasm with LUC count more than or equal to 0.5 × 10³ cells/µL of blood in conjunction with other complete blood count parameters. However, reflex peripheral smear testing is mandatory in the current setting for confirmation.

Keywords

Large unstained cells WBC differential counts autoanalyser

Introduction

Large unstained cells (LUCs) are large peroxidase-negative mononuclear cells due to the absence of myeloperoxidase-containing granules in their cytoplasm, and are larger than lymphocytes. In healthy individuals, LUC are mainly large lymphocytes. Still, other cells such as peroxidase-negative blasts, hairy cells, atypical and reactive lymphocytes, atypical lymphoid cells, plasma cells, peroxidase-negative neutrophils and CD34+ve hematopoietic stem cells are also included.^[1] Flowcytometry-based ADVIA 2120i auto-analyser estimates the LUC parameters in anticoagulated blood samples, which are a part of WBC differential in the PEROX channel. LUC parameters have been the subject of decades’ worth of studies evaluating their correlation with other inflammatory biomarkers.^[2] Our study observed the peripheral blood cells causing elevations in %LUC in various disease conditions, and attempted to correlate peripheral blood smear findings with clinical variables.

Materials and Methodology

This prospective observational study was conducted from June 2022 to March 2024. Blood samples with LUC% more than 4% were included in the study. Clotted, haemolysed and lipemic blood samples, samples with LUC% less than or equal to 4% were excluded. We aimed at observing the reasons behind the increase in %LUC and note the cells which fall under this category, and to correlate clinical and other laboratory parameters with peripheral blood smear findings.

Methodology

Clinical details were obtained from the laboratory information system. EDTA-anticoagulated whole blood samples were processed using the ADVIA 2120i autoanalyser, with reflex peripheral blood smear review for cases showing %LUC > 4%, even when other CBC parameters were within normal range. The peripheral smear slides were stained with Leishman stain, examined and both total and differential cell counts were calculated. Any abnormal increases or decreases in cell-counts, and atypical cells identified were documented.

Patients were classified into six diagnostic groups, which included chronic medical disease, infections, haematolymphoid neoplasms, visceral malignancy, multiple co-existing diseases of different groups and no specific clinical disease. Demographic details, LUC parameters and peripheral smear observations were compared under each diagnostic category.

Statistical Analysis Plan

Descriptive statistics were presented with frequency (percentage) & Mean (SD) for the categorical & continuous factors respectively. Median (IQR) was presented while the LUC parameters followed non-normal distribution. Kruskal Wallis test was used to determine the significant difference in LUC% and LUC count in each diagnostic group. Chi-square/Fisher’s exact test was used to determine the association between two independent categorical factors. A P < .05 indicates statistical significance. All the statistical analyses were done by using SPSS software (IBM, 28.0).

Results

A total of 500 consecutive cases which met the inclusion criteria were included. Age range of study population was widely distributed and included four months—89 years with a mean age of 45.6 ± 23.6 years. 95 (19%) were paediatric cases and 405 (81%) were adult cases. 335 (67%) were males and 165 (33%) were females. Most common clinical presentation was pancytopenia (26%), followed by pyrexia of unknown origin (15%). A total of 10% of cases were known haematolymphoid neoplasms under follow-up.

Haematolymphoid neoplasms represented 34% of cases followed by infections (25%), chronic medical disease (21%) and visceral malignancy (3%). A total of 8% had co-existence of diseases of various groups, while 9% had no identifiable diagnostic disease [Figure 1].

Figure 1.

Graphical representation of diagnostic categories in study population

LUC Parameters

The average LUC% in the paediatric population (n = 95) was 9.3% ± 6.1% with a median of 7.1% and, in the adult population (n = 405), it was 11.1% ± 12% with a median of 6.4%. Similarly, average LUC count in paediatric population was 0.6 ± 0.5 × 10³ cells/µL of blood with median of 0.4 × 10³ cells/µL of blood and, in adult population, the average was 1.9 ± 6.1 × 10³ cells/µL of blood with a median of 0.4 × 10³ cells/µL of blood. There was no statistical significance between LUC parameters and the age group of the patients.

%LUC was divided into six arbitrary groups: 4.1%–9.9%, 10%–19.9%, 20%–29.9%, 30%–39.9%, 40%–49.9% and greater than or equal to 50%. Approximately 69.5% cases in paediatric population were clustered in subgroup 1 (%LUC = 4.1%–9.9%), and none of the cases had %LUC >/ = 50%. In adults, subgroup 1 (%LUC = 4.1%–9.9%) had 69.8% cases followed by 17.1% in subgroup 2. Eleven cases in the adult population had %LUC >/ = 50%.

Peripheral Blood Smear Results

Peripheral smear revealed following observations: 38% (n = 190) had lymphocytosis, 33.8% (n = 169) had monocytosis, 23.8% (n = 119) had blasts, 7.8% (n = 39) had plasma cells and 6.2% (n = 31) had atypical lymphocytes. Less common findings include, smudge cells (4%, n = 20), eosinophilia (4.2%, n = 21), hypo-granular neutrophils (4.4%, n = 22), basophilia (3%, n = 15), neutrophilia (2.8%, n = 14), neutropenia (0.8%, n = 4), trophozoites (0.4%, n = 2), hyper segmented neutrophils and lymphopenia (0.2%, n = 1, each). Notably, 13.4% (n = 67) had no peripheral smear abnormalities.

Blood smears in the paediatric population showed monocytes (48.4%), lymphocytes (44.2%) and blasts (6.3%) with 7.4% cases having normal blood smears. In adults, lymphocytes were most common (36.5%) followed by monocytes (30.4%), blasts (27.9%), plasma cells (9.6%) and atypical lymphocytes (7.4%). Fourteen point eight percent of blood smears were normal in the adult population.

Frequency of haematolymphoid neoplasms were as follows: 29.1% (n = 44) AML, 16.6% (n = 25) plasma cell dyscrasia, 14.6% (n = 22) lymphomas, 11.9% (n = 18) Chronic lymphocytic leukaemia, 9.9% (n = 15) Acute lymphoblastic leukaemia, 6% (n = 9) Chronic myeloid leukaemia, 6% (n = 9) Myeloproliferative neoplasms, 6% (n = 9) Myelodysplastic neoplasms [Figure 2]. Majority were adults (98.2%) with a male predominance (69.6%). Seventy percent of atypical lymphoid cells, 69.2% of lymphocytes and 67.8% of blasts fell in this disease group. Notably, 13.23% of normal smears belonged to this group.

Figure 2.

Percentage of most observed peripheral blood cells in each morbidity group (X-axis: Diagnostic disease category; Y-axis: percentage of population)

Of 105 chronic medical disease patients, majority were Type 2 diabetes mellitus (66.7%) and %LUC falling between 4.1% and 9.9%. Infections were more common in children (66.6%) with two cases of bacterial sepsis exhibiting highest %LUC of 49% and 41.5%, respectively. Visceral malignancies accounted for the least number of cases (3%) with one case exhibiting %LUC = 50%, which was a case of lung adenocarcinoma undergoing chemotherapy. No specific illness was diagnosed in 9% of cases and was attributed by children with monocytosis (46.3%) and lymphocytosis (29.3%) in peripheral smears. Eight point four percent had multiple co-existing diseases with plasma cells (15.3%), blasts (13.3%), monocytes (8.5%) and lymphocytes (6.8%) in circulation [Figure 3].

Figure 3.

A case of viral infection. (A) PEROX scattergram showing an absolute LUC of 0.21 × 10³ cells/µL of blood (B) and (C) (40×) Reactive lymphocytes in peripheral smear

The highest mean and median %LUC were noted in haematolymphoid neoplasm group (15.2% ± 16.1% and 8.7%, respectively), while the lowest were in chronic medical disease group (6.6% ± 5.1% and 4.9% (4.3%–6.4%), respectively). However, the differences in mean and median values among the diagnostic disease groups were not statistically significant [Table 1].

Table 1.

LUC% among each diagnostic group

Disease Category	LUC in Percentage				P Value*
Disease Category	n	Mean ± SD	Median (IQR)	Range	P Value*
Chronic medical illness	105	6.6 ± 5.1	4.9 (4.3–6.4)	4.1–35.4	.08^*
Infections	126	9.4 ± 6.9	6.9 (5–12.1)	4.1–49
Haematolymphoid malignancy	168	15.2 ± 16.1	8.7 (5.2–16.4)	4–88.1
Multiple co-morbidities	42	9.9 ± 5.9	7.8 (5.3–14.2)	4.1–24.9
Organ malignancy	14	12.8 ± 12.3	8.2 (5.2–18.5)	4.1–50
No specific clinical findings	45	7.5 ± 3.6	6 (4.5–9.7)	4.1–15.9

Note: *Kruskal Wallis test.

The highest mean and median absolute LUC count was noted in the haematolymphoid neoplasm group [e.g., Figures 3–5] (mean = 3.4 ± 8.5 × 10³ cells/µL of blood and median = 0.5 × 10³ cells/µL of blood (0.2–2.3)), while the lowest were in visceral malignancy group (mean = 0.2 ± 0.2 × 10³ cells/µL of blood and median = 0.2 × 10³ cells/µL of blood (0.1–0.3)). The difference in median between haematolymphoid neoplasm group and other disease groups was proven to be statistically significant [Table 2].

Figure 4.

A case of Acute lymphoblastic leukaemia. (A) PEROX scattergram showing an absolute LUC count of 0.57 × 10³ cells/µL of blood. (B) Circulating lymphoblasts (20×). (C) Lymphoblasts with round to oval, medium to large sized nucleus with fine chromatin, occasional prominent nucleoli and scant basophilic cytoplasm (40×)

Figure 5.

A case of myelodysplastic neoplasm. (A) PEROX scattergram showing an absolute LUC count of 0.89 × 10³ cells/µL of blood. (B and C) (40×): Hypo-granular neutrophils in peripheral smear

Table 2.

LUC count representation in each diagnostic group

Diagnostic Subgroup	Mean ± SD	Median (IQR)	Range	P Value
Medical disease	0.6 ± 1.4	0.3 (0.2–0.5)	0–10	<.001
Infections	1.1 ± 4	0.4 (0.2–0.7)	0–41
Haematolymphoid malignancy	3.4 ± 8.5	0.5 (0.2–2.3)	0–64
Multiple disease	0.6 ± 0.9	0.3 (0.2–0.6)	0–5
Organ malignancy	0.2 ± 0.2	0.2 (0.1–0.3)	0–1
No specific clinical findings	0.5 ± 0.5	0.4 (0.3–0.7)	0–3

Note: *Kruskal Wallis test; Boldface indicates statistical significance.

In LUC% subgroup 1 (4.1%–9.9%), monocytes, lymphocytes, blasts, plasma cells and hypo-granular neutrophils were most observed. Notably, blasts were most common when %LUC was greater than 30%, which was statistically significant [Table 3].

Table 3.

Percentage of most commonly observed cells in each %LUC subgroup

Blood Cells	LUC Percentage, n (%)						P Value^*
	4.1–9.9	10–19.9	20–29.9	30–39.9	40–49.9	≥50
	(n = 348)	(n = 93)	(n = 29)	(n = 11)	(n = 7)	(n = 11)
Lymphocytosis	121 (34.8)	38 (40.9)	19 (65.5)	5 (45.5)	3 (42.9)	4 (36.4)	.042
Monocytosis	127 (36.5)	31 (33.3)	3 (10.3)	1 (9.1)	3 (42.9)	3 (27.3)	.038
Within normal limits	58 (16.7)	7 (7.5)	2 (6.9)	−	–	–	.043
Hypo-granular neutrophils	14 (4)	2 (2.2)	4 (13.8)	1 (9.1)	–	1 (9.1)	.118
Blasts	51 (14.7)	32 (34.4)	14 (48.3)	9 (81.8)	5 (71.4)	8 (72.7)	<.001
Plasma cells	32 (9.2)	4 (4.3)	1 (3.4)	1 (9.1)	1 (14.3)	–	.461
Atypical cells	14 (4)	9 (9.7)	6 (20.7)	1 (9.1)	–	1 (9.1)	.007

Note: ^*Chi Square test; Boldface indicates statistical significance.

Discussion

Combination of CBC components such as neutrophil-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR) and monocyte-to-lymphocyte ratio (MLR) have been investigated as markers of inflammation.^[3-6] In addition to these, LUC parameters, which include %LUC and LUC count, have also been investigated for their potential role as inflammatory biomarkers, markers to assess successful stem cell transplants and tumour biomarkers in suspicious leukaemia and lymphomas.^[7-9] Although high throughput technologies are available for the correct identification of the clinical disease, a complete blood count serves as a primary tool to arrive at a differential diagnosis cost-effectively.

In our study, there were 405 adults and 95 paediatric cases, with a male predominance in both groups. Infections were more common in children, and haematolymphoid malignancies were more common in adults. No specific clinically detectable disease was seen in 9% of the study population, which was majorly attributed to children (n = 48). The clinical presentation was non-specific, with pancytopenia and pyrexia of unknown origin being more common.

From our observation, the median LUC count of 0.5 × 10³ cells/µL with an inter-quartile range of 0.2–2.3 × 10³ cells/µL of blood was observed in haematolymphoid neoplasms, which were found to be higher than the median LUC count of other diagnostic categories, and this difference was found to be statistically significant (P < .001). There was a significant overlap of %LUC values among different categories, and no statistical correlation was obtained. The majority (69.6%) had %LUC between 4.1% and 9.9%. Marked elevations were observed in AML, CML, ALL and infections.

Reactive mature lymphocytes, monocytes, lymphoblasts, myeloblasts, monoblasts, atypical lymphocytes, hypogranular neutrophils, basophils and plasma cells were the most commonly encountered peripheral blood cells in the study population. Each of these cells has variable MPO staining characteristics.

Historically, MPO activity has been considered to be the only unambiguous differentiating marker of myeloblasts that is exclusive to the myeloid lineage. Due to its simplicity of use and interpretation, MPO cytochemistry has shown to be the most effective method for identifying the myeloid lineage of blasts in terms of both specificity and general acceptability. More than or equal to 3% of blast cells exhibiting positive staining in a peripheral blood smear is considered a positive MPO reaction. Less than 3% MPO staining is considered a negative MPO reaction.^[10]

Most of the myeloblasts are MPO-positive. Previous studies have shown that AML with minimal differentiation has strongly MPO-negative myeloblasts, which mimic lymphoblasts.^[11] Hence, MPO-negative myeloid blasts in AML without differentiation can be scattered in the LUC area, attributing to LUC% elevation.

In a study by Vainchenkar W et al., strong MPO-positive myeloblasts were noted in AML with maturation and Acute promyelocytic leukaemia.^[12] In our study, six cases of AML with maturation showed predominantly circulating myeloblasts with LUC% >4% that had MPO positivity in flow-cytometric analysis. Rosniak et al. have described that those myeloid cells with partial or complete MPO deficiency were clustered to monocytic or LUC area in the scattergram.^[13] These findings suggest that a small percentage of myeloblasts with scant or absent myeloperoxidase in their cytoplasm in these cases could have been clustered onto the LUC area, attributing to elevations in LUC%.

Similar to a study by Roh et al., in a few AML cases of our study, hypogranular neutrophils were seen in peripheral circulation, which were likely to be counted as monocytes or LUCs due to decreased myeloperoxidase index.^[14]

In AML with myelomonocytic and monocytic differentiation, monocytes are part of the neoplastic process. These monocytes and myeloblasts with scant granularity are counted as LUCs. Our study implies that, if the %LUC is more than 50% in a case of leukemic presentation, possibility of AML with monocytic, myelomonocytic differentiation and AML with dysplasia is to be considered in the differential diagnosis.

Lymphocytes, atypical lymphoid cells, plasma cells, lymphoblasts and basophils are invariably MPO-negative cells. Similarly, elevated mature lymphocytes were noted predominantly in haematolymphoid neoplasms (40.74%) and infections (22.75%) in this study.

All the CML cases in our study presented in the chronic phase with a median LUC count of 0.4 × 10³ cells/µL of blood. The majority had a %LUC between 10% and 19.9%. Basophils and myeloid blasts were the most common peripheral blood cells observed in circulation. In a previous study by Peterson et al. and others, it was observed that blasts in a substantial proportion of CML blast crisis (BC) cases were negative for MPO, even though roughly 70% of CML BC cases were myeloid, and BC follows a previous phase that exhibits florid myeloproliferation.^[15,16] Similarly, in a study by Mona et al., 154 out of 161 CML cases had circulating MPO negative cells in circulation, and seven cases had weak MPO staining (4%–20%), which suggests that circulating CML blasts in all phases have an intrinsic MPO negativity or weak positivity.^[17-21] Similarly, basophils have metachromatic granules in their cytoplasm, which are negative for MPO staining. %LUC >4% in CML cases of our study can be explained by clustering of circulating MPO negative myeloblasts and basophils in the LUC area in PEROX scattergram.

Myelocytes produce MPO; Neutrophils do not produce MPO; instead, they store it. Neutrophil granules in acute bacterial sepsis release MPO. Hence, hypogranular neutrophils are seen in sepsis patients’ blood. Also, neutrophil dysfunction is brought on by cytokine storm. As a result, these hypogranular neutrophils appear as large, unstained cells or monocytes distributed to the left of the PEROX cytogram.^[22]

In viral infections and chronic inflammatory states, such as diabetes, hypertension, autoimmune diseases and chronic kidney disease, monocytes and reactive lymphocytes were markedly increased in circulation, as they are mediators of inflammation. In these systemic illnesses, immune infiltration by T cells and B cells into target tissues and release of cytokines and chemokines sets a chronic low-grade localised and systemic inflammation. In contrast, short-chain fatty acids modulate gene expression of monocytes and reduce inflammation. It also regulates the release of regulatory T cells and thereby suppresses the function of inflammatory T cells.^[23] This forms the basis for chronic active inflammation in the above-mentioned metabolic diseases.

In autoimmune diseases, the predominance of leukopenia, along with peripheral blood monocytosis and lymphocytosis, was observed. Neutrophilia is the characteristic feature of RA, indicating active joint disease,^[24] but the current study showed no such findings.

Limitations

For assessment, we have not included a completely healthy control cohort or cohort with %LUC</ = 4%, for definitive comparison of pathological versus non-pathologic samples. Also, treatment history was not included as a part of the study.

Conclusion

Our study is a large-scale prospective study, which includes a spectrum of clinical disease. Absolute LUC count could be used as a cost-effective and readily available marker to identify a case of clinically suspicious haematolymphoid neoplasm with LUC count more than or equal to 0.5 × 10³ cells/µL of blood in conjunction with other complete blood count parameters. However, reflex peripheral smear testing is mandatory in the current setting for confirmation. The characteristic LUCs noted in circulation were reactive lymphocytes, monocytes, blasts including lymphoblasts, myeloblasts, monoblasts, plasma cells, hypogranular neutrophils, basophils and atypical lymphoid cells. Myeloblasts are myeloperoxidase reactive cells but depending upon the maturation, these cells are counted as LUCs in the PEROX scattergram. But the questionable nature of cells in AML with maturation displaying elevated LUC% warrants further investigation.

Footnotes

Declaration of conflicting interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Author Mamta Soni is a member of the Editorial Board of Apollo Medicine. The author did not take part in the peer review or decision-making process for this submission and has no further conflicts to declare.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

Institutional ethical committee approval number

The study was approved by the Institutional Ethical Committee under approval number AMH-DNB-013/05-22.

Informed consent

Not applicable.

Credit author statement

Study designed, conducted and manuscript written by Dr Dhivya.

Study conceptualized, guided and manuscript reviewed and revised by Coauthor Dr Mamta Soni.

Data availability

Available on request.

Use of artificial intelligence

Nil.

References

Bain

. Blood Cells: A Practical Guide. John Wiley & Sons, Chichester; 2021.

Cakir

, Cakir

, Atalay

, Koc

. Large unstained cells are correlated with inflammatory biomarkers in patients with invasive aspergillosis. Turk J Biochem. 2018 May 1;43(3):306–11.

Prabawa

, Bhargah

, Liwang

, . Pretreatment neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) as a predictive value of hematological markers in cervical cancer. Asian Pac J Cancer Prev. 2019;20(3):863.

Russu

, Mureșan

, Arbănași

, . The predictive role of NLR and PLR in outcome and patency of lower limb revascularization in patients with femoropopliteal disease. J Clin Med. 2022 May 6;11(9):2620.

Shi

, Qin

, Wang

, . Elevated neutrophil-to-lymphocyte ratio and monocyte-to-lymphocyte ratio and decreased platelet-to-lymphocyte ratio are associated with poor prognosis in multiple myeloma. Oncotarget. 2017 Mar 3;8(12):18792.

Kasimir-Bauer

, Karaaslan

, Hars

, Hoffmann

, Kimmig

. In early breast cancer, the ratios of neutrophils, platelets and monocytes to lymphocytes significantly correlate with the presence of subsets of circulating tumor cells but not with disseminated tumor cells. Cancers. 2022 Jul 6;14(14):3299.

Kanduła

, Janowski

, Więckowska

, . High molecular risk variants, severe thrombocytopenia and large unstained cells count affect the outcome in primary myelofibrosis. J Appl Genet. 2023 Sep;64(3):479–91.

Rabizadeh

, Pickholtz

, Barak

, Isakov

, Zimra

, Froom

. Acute leukemia detection rate by automated blood count parameters and peripheral smear review. Int J Lab Hematol. 2015 Feb;37(1):44–9.

Vanker

, Ipp

. The use of the full blood count and differential parameters to assess immune activation levels in asymptomatic, untreated HIV infection. S Afr Med J. 2014 Jan 20;104(1):45–8.

10.

Childs

, Stass

. Introduction. Characterization and diagnosis of the acute leukemias. In: Stass

, ed. The Acute Leukemias: Biologic, Diagnostic and Therapeutic Determinants. Marcel Dekker, New York; 1987:1–26.

11.

Anand

, Ghara

, Kumar

, . Myeloperoxidase cytochemical negativity: an unexpected but intrinsic property of blasts of all phases of chronic myeloid leukemia. Ann Hematol. 2005 Dec;84:767–70.

12.

Vainchenker

, Villeval

, Tabilio

, . Immunophenotype of leukemic blasts with small peroxidase-positive granules detected by electron microscopy. Leukemia. 1988 May 1;2(5):274–81.

13.

Rośniak-Bąk

, Jeleń

, Bąk

. Analysis of separation of white blood cells in peripheral blood as an indicator of MPO deficiency. Ann Clin Lab Sci. 2017 Aug 1;47(4):422–31.

14.

Roh

, Ham

, Song

, Hwang

, Lee

. Myeloperoxidase deficiency manifesting as pseudoneutropenia with low mean peroxidase index and high monocyte count in 4 adult patients. Lab Med. 2020 Mar 10;51(2):e16–9.

15.

Vardiman

, Pierre

, Thiele

, Imbert

, Brunning

, Flandrin

. Chronic myelogenous leukemia. In: Jaffe

, Harris

, Stein

, Vardiman

, eds. Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. IARC, Lyon; 2001:20–6.

16.

Anastasi

, Vardiman

. Chronic myelogenous leukemia and the chronic myeloproliferative diseases. In: Knowles

, ed. Neoplastic Hematopathology. Lippincott Williams and Wilkins, Philadelphia, PA; 2000:1745–90.

17.

Peterson

, Bloomfield

, Brunning

. Blast crisis as an initial or terminal manifestation of chronic myeloid leukemia. A study of 28 patients. Am J Med. 1976;60:209–20.

18.

Catovsky

, O’Brien

, Cherchi

, Benavides

. Ultrastructural, cytochemical, and surface marker analysis of cells during blast crisis of chronic granulocytic leukaemia. Boll Ist Sieroter Milan. 1978;57:344–54.

19.

Vardiman

, Coelho

, Golomb

, Rowley

. Morphologic and cytochemical observations on the overt leukemic phase of therapy-related leukemia. Am J Clin Pathol. 1983;79:525–30.

20.

Saikia

, Advani

, Dasgupta

, . Characterization of blast cells during blastic phase of chronic myeloid leukemia by immunophenotyping—experience in 60 patients. Leuk Res. 1988;12:499–506.

21.

Karnik

, Nair

, Zingde

, . Use of polyclonal anti-myeloperoxidase antibody in myeloid lineage determination. Indian J Med Res. 1994;100:272–6.

22.

Yonezawa

, Horie

, Yoshioka

, . Association between the neutrophil myeloperoxidase index and subsets of bacterial infections. Int J Lab Hematol. 2010;32(6p2):598–605.

23.

Salamandris

, Antonopoulos

, Oikonomou

, . The role of inflammation in diabetes: current concepts and future perspectives. Eur Cardiol Rev. 2019;14(1):50.

24.

Vaitinadin