Assessment of the relationship between hematologic parameters, (CPD), in screening for COVID-19 severity in women

Abstract

Aims

This study evaluated hematological parameters and cell population data (CPDs) to assess their ability to discriminate between moderate and severe COVID-19 in hospitalized female patients, and to identify potential markers associated with worse outcomes.

Patients & methods

A retrospective study was conducted on 84 adult female COVID-19 patients hospitalized at CHC-UFPR (Brazil) between March 2020 and July 2021. Patients were stratified into moderate (n = 46) and severe (n = 38) disease groups. A control group included 100 healthy female outpatients. Parameters analyzed included D-dimer, WBC count, neutrophil-to-lymphocyte ratio (NLR), and CPDs (LY-X, LY-Y). RT-qPCR was used for SARS-CoV-2 confirmation and variant identification.

Results

Significant differences (p < 0.05) in LY-X, LY-Y, NLR, and WBC were found between moderate and severe groups. D-dimer was elevated in severe cases. Among deceased patients (n = 17), WBC and NLR were markedly increased. ROC curve analysis confirmed the discriminatory power of these markers. No significant association was found between viral genotype and severity (p = 0.9602).

Conclusions

Hematological parameters, particularly CPDs and NLR, are valuable for early stratification of COVID-19 severity. These automated, rapid, and cost-effective measures can support clinical decision-making. However, CPD usage depends on analyzer availability and lacks standardization across platforms.

HIGHLIGHTS

Introduction

• COVID-19 overview: COVID-19 is a highly transmissible and potentially severe infectious disease caused by SARS-CoV-2, associated with a complex systemic inflammatory response.

• Relevance of hematological parameters: Traditional biomarkers have shown limitations in disease stratification. Recent evidence supports the use of hematological indicators such as cell population data (CPD) and D-dimer in evaluating disease severity.

• Study objective: To assess the diagnostic and prognostic potential of CPD parameters, neutrophil-to-lymphocyte ratio (NLR), white blood cell count (WBC), and D-dimer levels in female COVID-19 patients.

Methods

• Design and population: A retrospective analysis was conducted at CHC-UFPR, involving 84 female COVID-19-positive patients (38 with severe and 46 with moderate disease) and 100 female controls without COVID-19.

• Laboratory procedures: Complete blood count (CBC) and CPDs were analyzed using the Sysmex XN-3100. D-dimer measurements were performed on the Sysmex CS Series. RT-qPCR was used for SARS-CoV-2 variant genotyping.

• Statistical analysis: Data were evaluated using R software with GCMR package and ROC curves generated with MedCalc. A 5% significance level was adopted.

Results

• Key biomarkers: LY-X and LY-Y (CPDs), NLR, WBC, and D-dimer levels were significantly associated with disease severity (p < 0.05).

• Prognostic indicators: Elevated NLR and WBC levels were strongly correlated with fatal outcomes.

• Severity vs. outcome: A statistically significant correlation (p < 0.001) was observed between clinical severity and mortality, reinforcing the value of hematological parameters in predicting outcomes.

• Variant analysis: Although the Gamma variant predominated among severe cases and deaths, genotyping results did not statistically influence severity (p = 0.9602).

• ROC performance: LY-X, LY-Y, NLR, and WBC showed statistically significant diagnostic accuracy in distinguishing between moderate and severe disease.

Discussion

• Inflammatory dysregulation: Severe COVID-19 patients exhibited leukocytosis, lymphopenia, and elevated NLR, indicative of systemic immune dysregulation.

• Lymphocyte markers: LY-X and LY-Y effectively represented lymphocyte depletion and inflammatory activity, correlating with disease progression.

• NLR as a predictor: NLR values were markedly elevated in severe cases, reinforcing its utility as an accessible inflammatory marker in COVID-19.

• Role of D-dimer: While elevated in severe cases and useful in early risk assessment, D-dimer showed limited utility in monitoring treatment response.

• Variant considerations: Despite the predominance of the Gamma variant among severe cases, the sample size limited definitive conclusions on variant-specific outcomes.

Final considerations

• Clinical applicability: Hematological parameters, particularly CPDs and NLR, are promising tools for early severity stratification in COVID-19, providing fast, cost-effective, and minimally invasive support for clinical decision-making.

• Advantages of CPDs: These parameters offer real-time insights during routine blood analysis, without requiring additional samples or testing time.

• Limitations and future directions: The need for specialized equipment and the lack of standardization may restrict widespread CPD implementation. Further studies with larger, more diverse populations are warranted.

1. Introduction

COVID-19 is an acute respiratory infection caused by SARS-CoV-2, characterized by its potential severity, high transmissibility, and global distribution [1]. SARS-CoV-2, a betacoronavirus (subgenus Sarbecovirus, family Coronaviridae), was initially identified in bronchoalveolar lavage samples from patients with pneumonia of unknown etiology in Wuhan, Hubei Province, China, in December 2019 [2].

Clinical manifestations of coronavirus disease include symptoms such as fever, dyspnea, cough, and anosmia. The incubation period is typically 14 days after exposure and symptomatic infections range from mild to critical, with the majority being non-sever [3–12].

Severe disease can occur in otherwise healthy individuals of any age but is predominantly observed in older adults with specific underlying medical comorbidities. COVID-19 vaccination significantly reduces the risk of severe disease even in susceptible populations [13].

Specific demographic characteristics and laboratory abnormalities have also been associated with severe disease among hospitalized COVID-19 patients such as lymphopenia, elevated transaminases, lactate dehydrogenase, and elevated inflammatory markers (ferritin, C-reactive protein), along with coagulation test abnormalities [8,14,15].

The gold standard for biomolecular diagnosis of COVID-19 is the RT-qPCR, complemented by readily available rapid tests, ensuring widespread access to the general population at an affordable cost. Given the limitations of many laboratory biomarkers, recent years have seen the proposal of numerous new markers related to hematological parameters for infectious diseases [16]. However, none of these markers can be recommended as a sole criterion for COVID-19 or any other infection, independent of the etiological agent. It is advocated the use of a combination of biological parameters alongside clinical and imaging assessments to compose a comprehensive severity score for the disease.

A complete blood count (CBC) is widely employed for diagnosing acute infections and the modern hematology analyzers provide precise and cost-effective leukocyte differential counts swiftly. These enhanced technologies facilitate the inclusion of cell population data parameters (CPD) along with the basic CBC. CPDs offer quantitative and qualitative insights into the morphological and functional characteristics of leukocytes, enabling a more detailed study of cells in response to stimuli such as infections, providing valuable information on cellular activation status [17].

Thus, the central premise of our investigation contends that the measurement of CPD, and hematological parameters and D-dimer assays, could represent a robust approach, enhancing our understanding of the multifaceted implications of the complex pathophysiology induced by COVID-19 on the hematological dynamics. Sex-based differences in physiology, hormones, and behavior can significantly influence study outcomes. Focusing on a female-only cohort enhances analytical specificity and minimizes biological variability. Additionally, women have historically been underrepresented in biomedical research. By examining a homogeneous female sample, this study aims to generate findings that are both biologically meaningful and clinically applicable to women.

The aim of this study is to analyze the correlation between hematological markers and the severity of SARS-CoV-2 infection, proposing their application in the early stratification of the disease.

2. Methods

2.1. Study site

This retrospective study was conducted at the Hematology Laboratory of the Clinical Hospital Complex of University Federal do Paraná (CHC-UFPR), following thorough review and approval by the Research Ethics Committee, under protocol number CAAE 36697020.7.0000.0096. The research was carried out in close collaboration with the Molecular Virology Laboratory at UFPR, Curitiba, Paraná, Brazil. As this was a retrospective study, the ethics committee released the need for a free and informed consent form. This study adheres to the ethical principles contained in the Declaration of Helsinki, ensuring that all research procedures and methods are carried out in accordance with the rights of the participants and with respect for their dignity.

2.2. Specimens and patients

The subjects enrolled in the study were attended to in the screening, emergency, and/or hospitalized departments from March 2020 to July 2021 at the CHC-UFPR. The selection of study participants adhered to the following inclusion criteria: female patients aged >18 years with a positive RT-qPCR test for SARS-CoV-2, attended to and hospitalized in the wards or Intensive Care Unit (ICU) of CHC. Exclusion criteria encompassed female patients with a positive test for SARS-CoV-2 who concurrently presented hospital-acquired infection with COVID-19.

2.3. Study design

The study included two distinct groups (Figure 1), the first consisting of 84 female patients diagnosed with COVID-19, of whom 38 had severe cases requiring hospitalization in intensive care units (ICUs) and 46 had moderate cases requiring hospitalization in general wards. The second group consisted of 100 female patients without a diagnosis of COVID-19, a control group whose objective was to validate the hematological laboratory parameters proposed in the study, establishing a standard normal range for the population under investigation and analyzing the Receiver Operating Characteristic (ROC) curves to obtain the diagnostic accuracy of the relevant parameters in the study. It is important to note that the patients in the control group were outpatients with CBC and D-dimer results within the normal and reference criteria established for these tests. Both groups included individuals between the ages of 20 and 90 years, with a median age of 54 years for the control group and 56 years for the COVID-19 group. This study evaluated five continuous response variables, namely D-dimer, white blood cell (WBC) count, neutrophil-to-lymphocyte ratio (NLR), and cell population data (CPDs) related to WBC count and lymphocyte subpopulations (LY-X, LY-Y). Categorical covariates were hospital admission and outcome. Samples were collected in tubes containing K2-EDTA and processed on the Sysmex XN-3100 hematology analyzer. For D-dimer dosing, samples were collected in 3.2% sodium citrate and processed on Sysmex CS Series automated coagulometric analyzers, models 2100 and 2500.

Figure 1.

Flowchart. CPD: cell population data; D-dimer: mg/FEU: milligrams/fibrinogen equivalent units; WBC: white blood cell count expressed as × 1000/mm³; NLR: neutrophil to lymphocyte ratio; LY-X: lymphocyte complexity; LY-Y: lymphocyte fluorescence intensity.

2.4. Automated hematology analyzer

We used the Sysmex XN-3100 hematology analyzer (Sysmex Corporation, Kobe, Japan), which is an automated analyzer that uses different principles based on fluorescence flow cytometry to provide information on the shape, size and internal granularity of cells, thus generating total and differential leukocyte counts, as well as CPDs. It has a leukocyte differentiation channel that discriminates leukocytes by virtue of signals that are plotted on a scatter graph. The optical signals along the X-axis (lateral dispersion) are proportional to internal complexity; fluorescence along the Y-axis correlates with nucleic acid content (cell activation), while forward dispersion (Z-axis) is related to cell size. These parameters represent the distribution ranges of cell population clouds within the analysis channels (WDF: WBC differential and WNR: White cell nucleated). Among the various parameters of CPDs reported along the X, Y axes for all normal leukocyte populations, the parameters related to lymphocytes stand out in this study: X-axis the internal complexity of lymphocytes (LY-X), Y-axis the fluorescence intensity of lymphocytes (LY-Y) [17,18].

2.5. Automated coagulometric analyzer

This study used Sysmex CS series coagulometric analyzers, models 2100 and 2500 (Sysmex Corporation, Japan), which have more than one analytical module available, including coagulometry, immunoturbidimetry and colorimetry. D-dimer dosing was carried out using immunoturbidimetry, which is an analytical test that uses immunoprecipitation and light scattering to quantify the D-dimers present in plasma (polystyrene particles covered with monoclonal antibody undergo aggregation when in contact with samples containing D-dimer and the turbidity generated is proportional to the concentration).

2.6. RT-qPCR genotyping

Two probe-based genotyping protocols were used in sequence to identify variant of concern (VoC) [19]. The first used two deletion regions as targets – 69–70del in the spike and 3675–3677 del and in ORF1-a/b to identify Alpha, Beta/Gamma, without discriminating between the latter two [20]. The other protocol used primers 73 from the Thermo Fischer Scientific panel (Thermo Fisher Scientific, Applied Biosystems, US), which targets mutations K417T, E484K, N501Y, P681R, to identify Delta and Zeta variants, as well as differentiating Beta from Gamma.

2.7. Statistical analysis

Statistical analysis of the data was conducted using R Core Team software version (2021) and the GCMR (Gaussian Copula Marginal Regression) package. The significance level of 5% was adopted. To assess the normality of the data, the Shapiro-Wilk test was used. A normal (Gaussian) distribution was assumed for the response variable when p > 0.05. Otherwise, when p < 0.05, a Gamma distribution was assumed for the response variable. The null hypothesis H0 assumed normality for the proposed data. To evaluate the response performance of hematological laboratory parameters, statistical analyses were conducted using MedCalc software version 20.218 (MedCalc Software, Mariakerke, Belgium). The results were presented as ROC curves, indicating the sensitivity, specificity, and accuracy percentages of the parameters.

3. Results

The statistical analysis focused on three main parameters: D-dimer, WBC and NLR. Furthermore, the analysis of disease severity highlighted the relevance of CPDs of LY-X and LY-Y lymphocyte subsets in differentiating between moderate and severe disease. Table 1 summarizes the analysis of hematologic parameters in relation to COVID-19 severity.

Table 1.

Statistical analysis of hematological parameters in COVID-19 severity, comparing moderate and severe disease.

*Hospital admission		Patient COVID-19 (n = 84)
	COVID-19 moderate (n = 46) IIQ			COVID-19 severe (n = 38) IIQ
Parameters	Med	P25	P75	Med	P25	P75	p-Value	Reference values
D-dimer	1.31	0.63	3.32	2.19	1.0	4.24	<0.001	<0.55 mg/FEU
WBC	7.1	5.5	10.6	10.7	7.2	12.4	<0.001	3.8–11.0 × 10³/mm³
NLR	6.05	3.23	11.1	10.15	6.13	17.9	<0.001	<2.1 (1.5–2.0)
LY-X	78.1	76.4	81.1	77.3	75.3	78.8	0.023	80.1 (79.6–80.4)
LY-Y	71.9	70.7	75.3	70.2	67.7	72.7	0.002	72.8 (72.3–73.5)
**Hospital outcomes
D-dimer	1.44	0.62	5.91	1.09	0.63	4.62	0.128	<0.55 mg/FEU
WBC	7.9	5.9	11.2	10.9	7.7	15.6	<0.001	3.8–11.0 × 10³/mm³
NLR	3.95	2.3	7.4	4.70	2.4	14.5	<0.001	<2.1 (1.5–2.0)
LY-X	79.6	77.5	81.2	78.0	76.5	80.4	0.026	80.1 (79.6–80.4)
LY-Y	74.9	72.1	77.0	72.9	67.5	75.7	0.024	72.8 (72.3–73.5)

Legend Tables 1 and 2: Med.: median; (IIQ): interquartile range; P25: 25th percentile; P75: 75th percentile; D-dimer: mg/FEU: milligrams/fibrinogen equivalent units; WBC: blood cell count expressed as × 1000/mm³; NLR: neutrophil to lymphocyte ratio; LY-X: lymphocyte complexity; LY-Y: lymphocyte fluorescence intensity; *hospital admission: tests performed in the first 24 hours of hospitalization; **hospital outcome: last test performed before hospital discharge or death.

Table 2 presents the analysis of hematologic parameters at the time of outcome for the group of 84 will patients, with 20% of cases resulting in death and 80% of cases in recovery. The parameters that stood out were WBC and NLR in cases that resulted in death.

Table 2.

Statistical analysis of hematological parameters in the clinical outcome of COVID-19.

*Hospital admission	Patient COVID-19 (n = 84)
	Clinical outcome at discharge (n = 67) IIQ			Clinical outcome: mortality (n = 17) IIQ
Parameters	Med	P25	P75	Med	P25	P75	p-Value	Reference values
D-dimer	1.20	0.74	3.24	3.70	2.31	11.0	<0.01	<0.55 mg/FEU
WBC	8.6	6.0	10.9	11.5	7.8	13.0	<0.001	3.8–11.0 × 10³/mm³
NLR	7.0	4.0	12.6	15.2	4.5	22.0	0.022	<2.1 (1.5–2.0)
**Hospital outcomes
D-dimer	1.09	0.58	3.34	2.33	1.01	5.9	<0.001	<0.55 mg/FEU
WBC	7.9	6.3	10.9	16.1	12.5	20.8	<0.001	3.8–11.0 × 10³/mm³
NLR	3.3	2.3	5.5	16.6	6.8	42.5	0.002	<2.1 (1.5–2.0)

Figure 2 displays the correlation between severity and outcome of the 84 patients with COVID-19 was established using Fisher’s exact test, which indicated a statistically significant difference (p < 0.001) in cases that progressed to the worst outcome, with (n = 15) severe disease and (n = 2) moderate disease in cases that died. In addition, the analysis of the outcome shows that 67 patients recovered from the disease, including 23 patients with severe COVID-19.

Figure 2.

COVI-19 severity vs. outcome clinical.

Regarding genotyping, no significant results were observed when comparing the severity of the cases studied with the respective genotypes tested (p = 0.9602, Fisher’s test). The genotypes detected in the samples were Wild (13%), Alpha (1%), Gamma (34%) and Zeta (1%), totaling approximately 50% of the 84 patients. Of the 84 samples, 60 were tested and 11% were inconclusive, where the genotype could not be defined. Of the 17 patients who died, 82% were affected by the Gamma variant. This includes 12 severe and 2 moderate cases of COVID-19.

Tables 3 and 4 and Figure 3 show the results of the ROC curves, which confirm some of the statistical findings described above. In the comparison between moderate and severe COVID-19 in the ROC curves, the following hematological parameters: LY-X, LY-Y, NLR and WBC presented p-values <0.05, which corroborates the statistical results reported in Table 1 and demonstrates a discriminatory character for the severity of the disease.

Table 3.

ROC curves comparing COVID-19 groups.

Parameters	Patient (n = 84) versus moderate (n = 46)	Patient (n = 84) versus severe (n = 38)	Moderate (n = 46) versus severe (n = 38)
	p-Value
LY–X	0.0243	<0.0001	0.0074
LY–Y	0.9223	0.0176	0.0006
NLR	<0.0001	<0.0001	0.0284
WBC	0.6415	<0.0001	<0.0001

Note: Bolded values indicate a statistically significant difference between severe and moderate COVID-19 cases p < 0.05 for LY-X, LY-Y, NRL and p < 0,001 for WBC.

Legend table 3,4. D-dimer: mg/FEU: milligrams/fibrinogen equivalent units; WBC: white blood cell count expressed as × 1000/mm³; NLR: neutrophil to lymphocyte ratio; LY-X: lymphocyte complexity; LY-Y: lymphocyte fluorescence intensity.

Table 4.

Comparative statistical analysis of ROC curves in relation to parameters: WBC, NRL, D-dimer and LY-X, LY-Y.

Parameters	COVID-19	Accuracy	Sensitivity (%)	Specificity (%)	Youden index	Cut off
WBC	Patients (84)	0.643	38.1	97	0.35	>10420
	Moderate (46)	0.669	57.5	72	0.29	>8480
NLR	Severe (38)	0.876	77.3	91.7	0.69	>10420
	Patients (84)	0.913	89.29	90	0.79	>2.7
	Moderate (46)	0.504	27.6	68.5	0.04	>2.7
D-DIMER	Severe (38)	0.584	35.8	77	0.13	>3.5
	Patients (84)	0.928	92	93.5	0.85	>0.55
	Moderate (46)	0.512	34.9	64.9	0	>0.6
LY-X	Severe (38)	0.570	41.5	63.35	0.04	>1.0
	Patients (84)	0.702	68.9	71.4	0.40	≤77.9
	Moderate (46)	0.808	57.5	99	0.56	≤77.9
LY-Y	Severe (38)	0.894	76	97.6	0.74	≤78.2
	Patients (84)	0.617	63.1	62	0.25	>70.6
	Moderate (46)	0.610	54.8	66.4	0.21	>70.6
	Severe (38)	0.765	48.7	98	0.47	>65.8

Note: This table presents the diagnostic accuracy of the analyzed parameters. Not all parameters demonstrated good accuracy, as reflected in the tabulated results. These findings are further explored in the Discussion and Conclusion, where the most reliable indicators for screening COVID-19 severity are highlighted.

Figure 3.

Shows the ROC curves for WBC (leukocytes), lymphocyte subpopulation (LY-X and LY-Y), NLR (neutrophil/lymphocyte ratio), and D-dimer.

Hematological parameters, including D-dimer, suggest differentiation between moderate and severe disease, particularly during the assessment of patients at hospital admission. This indicates changes in the body’s innate cellular response to SARS-CoV-2 and disease progression in severe cases.

4. Discussion

COVID-19 presents clinical and laboratory characteristics that result from local and systemic inflammatory processes. These processes are potentiated by preexisting comorbidities, such as diabetes, obesity, systemic arterial hypertension, cardiovascular, renal, and pulmonary diseases, as well as biological factors such as gender and age [21]. However, investigations into the critical factors that profoundly influence the severity of the disease and death were prompted by the discrepancies in the clinical signs of COVID-19 observed among these patients [22].

The studies conducted in China reported that changes in peripheral blood inflammatory cells can indicate the severity and spread of COVID-19 [23,24]. They found that the majority of patients with severe COVID-19 symptoms had increased serum levels of cytokines and chemokines, as well as high neutrophil production and low lymphocyte production. This indicates that the hyperinflammatory response was responsible for the severity of the disease [25].

A meta-analysis revealed that 35–75% of patients developed lymphopenia [26]. The study found that 45% of severe cases had marked lymphopenia at the onset of SARS-COV-2 infection, which supports previous research. The LY-X and LY-Y parameters, which represent lymphopenia in this study, were found to be discriminatory between severe and moderate COVID-19, as shown by both logistic statistical analysis (p < 0.05).

The neutrophil-to-lymphocyte ratio (NLR) is a widely used marker for predicting bacterial infections, including respiratory syndromes and pneumonia. The study found that patients with a more severe condition had an increase in leukocytes and NLR compared to the mild-moderate group [27]. The NLR median value on admission was approximately 5 times higher than the reference value. Furthermore, a statistically significant difference between moderate and severe disease was observed (p = 0.0284).

It was observed that several COVID-19 patients experienced an increase in neutrophil count and a decrease in lymphocyte count during the severe phase, indicating significant dysregulation of the immune system and a critical condition in the most severe cases of the infection [8].

A retrospective multicenter study conducted in China during the first two months of the pandemic found that 46.4% of patients with SARS-COV-2 infection had elevated D-dimer values (>0.5 mg/L), with the highest results observed in the most severe cases [14]. The study results indicate an average 2.4-fold increase in the D-dimer values of patients with severe COVID-19. In the multivariate analysis, D-dimer was found to be an effective parameter for screening disease severity. However, although it has been identified that patients with higher initial levels of inflammatory markers such as ferritin, CRP, procalcitonin, and D-dimer are more likely to progress to critical illness and have a worse outcome, monitoring the treatment of patients using these markers did not prove to be a reliable tool for clinical decision-making [6,28].

Additionally, genotypes did not have any influence on the outcome of the most severe cases. However, it is important to note that the sample size for this analysis was small. Out of the 84 samples included, only 60 were tested and 11 were inconclusive. The study revealed that 34 (57%) patients had the gamma variant, with 70% of them being between the ages of 20 and 60. This indicates that a younger population was infected at a time when vaccination was beginning in the country for the population most at risk, which until then had been represented by the elderly (>65 years) and health professionals.

The study suggests that combined use of CPD, D-dimers, and NLR offers complementary, accessible, and cost-effective tools for early risk stratification in COVID-19. These markers can support clinical decision-making, optimize triage, and improve resource allocation, particularly in settings with limited capacity.

Although initial reports suggested higher mortality and severity of COVID-19 among men, subsequent epidemiological data have shown that women constitute a significant proportion of global COVID-19 cases. Nevertheless, studies focusing specifically on the female population particularly with regard to long-term outcomes, symptomatology, and comorbidities remain limited in the current literature. The present study aims to address this gap by providing sex-specific data that may support more equitable and targeted public health strategies.

The importance of sex-balanced research is recognized, and comparative studies involving both sexes are essential for a comprehensive understanding of COVID-19. However, research specifically focused on women is also considered necessary to identify potential differences in disease presentation, treatment response, and recovery patterns.

5. Conclusions

This study suggests that hematological laboratory parameters are effective screening tools for COVID-19 cases. Data collected at patient admission can aid in the classification of disease severity. Genotyping did not show any discernible influence on disease severity or outcomes. The significance of CPDs in this study cannot be overstated. They are generated during automated differential counting without the need for additional samples and provide a quantitative, objective, and highly accurate alternative to manual differential counts. Furthermore, CPDs have fast turnaround times and are cost-effective. However, their widespread use may be limited by the dependence on specific instruments. Despite their advantages, the lack of standardization is a limitation that requires cautious consideration of CPDs in clinical decisions. It is important to note that early diagnosis is crucial, as rapid intervention has the potential to reduce mortality.

Appendix A.

Table A1.

Table of additional data: controls samples.

No Covid-19	Control samples (100) iIQ		Reference values
Parameters	Median	P25–P75
Age	54	44–67
D-dimer	0.36	0.27–0.45	**<0.55 mg/FEU
WBC	6.885	5.962–8.040	4.0–12.0 × 10³/mm³
NLR	1.69	1.20–2.30	<2.0%
WNR-X	149.5	148.2–152	141.40 − 159.20 ch
WNR-Y	102.6	99.9–109.2	95.0 − 113.80 ch
WNR-WX	300	276–407	227–491 ch
WNR-WY	670	635–699	585–782 ch
NE-SSC	151.4	146.8–153.6	150.6 − 52.0 ch
NE-SFL	47.85	46.3–49.6	49.4–50.3 ch
WDF-WX	456	441–533	297–772 ch
WDF-WY	1.580	1410–1753	1.070–3.109 ch
LY-X	79.95	78.4–81.7	79.6–80.4 ch
LY-Y	69.80	68.2–72	72.3–73.5 ch
RBC	4.67	4.36–4.87	4.05–5.25 × 106
HB	13.80	13–14.5	12.5–15.7 g/dl
HCT	40.90	38.9–43	36.7–46.3%
RDW	12.7	12.3–13.2	<14.8%
PLQ	265	222–291	140–400/mm³

^**Reference value established by the test manufacturer.

Note: The other reference values represent the average female outpatient population attended at CHC-UFPR (Curitiba, PR, Brazil), established according to the recommendations of CLSI document EP28-A3c.¹⁰Control group consisted of 100 asymptomatic women undergoing elective exams (age range 18–98).

These are results of the laboratory markers listed in patients who do not have COVID-19. These values are consistent with normal reference ranges for a healthy population matched by age to the studied COVID-19 cohort.

Med.: Median (IIQ): interquartile range; P25: 25th percentile; P75: 75th percentile; D-dimer: in milligrams/fibrinogen equivalent units; WBC: white blood cell count expressed as × 10³/mm³; NLR: neutrophil to lymphocyte ratio; WNR-X: leukocyte complexity reported in arbitrary units of light scatter(ch); WNR-Y: leukocyte fluorescence intensity (ch); WNR-WX: width of dispersion of leukocyte complexity(ch); WNR-WY: width of dispersion of leukocyte fluorescence intensity(ch); NE-SSC: neutrophils complexity (ch); NE-SFL: neutrophils fluorescence intensity(ch); WDF-WX: differential of leukocyte channel complexity and width of dispersion(ch); WDF-WY: differential of leukocyte channel fluorescence intensity and width of dispersion(ch); LY-X: lymphocyte complexity(ch); LY-Y: lymphocyte fluorescence intensity (ch); RBC: red blood cells count expressed as × 10⁶/mm³; HB: hemoglobin g/dl; HCT: hematocrit in percentile; RDW: red cell distribution width or anisocytosis coefficient; PLQ: Platelets count expressed as × 1000/mm³.

Author contributions

Conceptualization, S.R.C; S.C.C; M.B.N; methodology, S.C.C; S.M.L.; L.L.M.A.; L.BBC; and B.A.L software, S.C.C; S.R.C.; R.R.P; validation; S.C.C; S.R.C; M.B.N; formal analysis, S.C.C; S.R.C M.B.N.; investigation, S.C.C; S.M.L.; L.L.M.A.; L.BBC; and B.A.L; writing—original draft preparation, S.C.C; writing—review and editing, S.C.C; M.B.N.; S.R.C.; S.M.R; visualization, S.C.C; S.R.C; M.B.N; S.M.R; R.R.P; S.M.L.; L.L.M.A.; L.BBC; and B.A.L; supervision, S.R.C and M.B.N.; project administration, M.B.N.; resources and funding acquisition, S.R.C. All authors have read and agreed to the published version of the manuscript.

Disclosure statement

The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Data availability statement

The authors confirm that the data supporting the findings of this study are available within the article and additional data can be accessed at the URL:<https://hdl.handle.net/1884/87527> or the data will be shared by the corresponding author translation.