Portal de Periódicos da CAPES

Cell population data: Could a routine hematology analyzer aid in the differential diagnosis of COVID‐19?

2020; Wiley; Volume: 43; Issue: 2 Linguagem: Inglês

10.1111/ijlh.13368

1751-553X

Ivana Lapić, Tina Brenčić, Dunja Rogić, Maja Lukić, Iva Lukić, Monika Kovačić, Lorena Honović, Vatroslav Šerić,

Long-Term Effects of COVID-19

The outbreak of coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has lately become a major public health concern, in which laboratory diagnostics plays a pivotal role in terms of timely diagnosis, but also monitoring, prognosis, prediction, and detection of underlying complications. Although the current gold standard for the etiological diagnosis is real-time reverse transcription-polymerase chain reaction (rRT-PCR) from oral or nasopharyngeal swab specimens, laboratory testing in the COVID-19 era goes far beyond virus detection.1 Published data show that many routinely used laboratory tests may assist in distinction of COVID-19 cases from other viral infections or conditions presenting with similar symptomatology as well as in assessment of disease severity.1, 2 So far, the most prominent observed laboratory abnormalities associated with COVID-19 were changes within the complete blood count (CBC) with variable leukocyte count accompanied by persistent lymphopenia, elevations of inflammatory biomarkers, and derangement of the coagulation system.2 Novel automated hematology analyzers provide differentiation of leukocyte subpopulations according to differences in their granularities, size, and nucleic acid content. Cell population data (CPD) are background parameters derived from optical signals during differential leukocyte count that reflect morphological and functional characteristics of neutrophils (NE), lymphocytes (LY), and monocytes (MO) and are readily available with every routinely performed CBC measurement.3 Hereby, we hypothesized that activation of the immune response triggered by SARS-CoV-2 causes cell changes that might be detected and quantified using CPD, and evaluated the potential of CPD parameters in the differential diagnosis of COVID-19. This was a retrospective multicenter study which included 114 COVID-19 patients (median age: 52 years; from 6 to 92, 46% females) admitted to the University Hospital Center (UHC) Zagreb, General Hospital Pula, UHC Osijek, and General Hospital "dr Ivo Pedišić" Sisak, Croatia, during the first peak period of the COVID-19 pandemic in Croatia (from mid-March to April 2020). The control group consisted of 88 suspected but eventually negative COVID-19 patients (median age: 69 years; from 17 to 98, 42% females) who presented with fever and/or respiratory symptoms. Patients whose history included any type of hematologic disorder were excluded. All patients underwent rRT-PCR testing upon admission as well as routine laboratory diagnostics including CBC and biochemistry tests, performed immediately in local laboratories. All data were extracted from results obtained during routine laboratory diagnostics and anonymized so that no additional blood sampling was requested and informed consent was not required. Complete blood count analysis was conducted on Sysmex XN-1000 hematology analyzers (Sysmex Corporation) in laboratories of each respective institution. Sysmex XN-1000 is an automated hematology analyzer that performs CBC, 6-part leukocyte differential, and extended differential count with research CPD parameters by applying a combination of optical fluorescence technique, flow cytometry, and laser-optical recognition. Two channels are used for counting and differentiating leukocyte subpopulations. The WNR channel quantifies basophils and separates them from the total white blood cells (WBC) by means of fluorescence staining and flow cytometry. In the WDF channel, WBC subpopulations are separated by differences in forward- and side-scattered light that reflect size and internal cell complexity (x- and y-axis of the WDF channel scattergram), respectively, as well as intensity of fluorescence that depends on the amount of nucleic acids within the cell nuclei (z-axis of the WDF channel scattergram).3-6 Detailed explanation of the morphological and clinical significance of CPD parameters is summarized in Table 1. The evaluated parameters included total WBC count, its differential in percentage and absolute numbers, CPD parameters of NE, LY, and MO reported in arbitrary units of light scattering (ch), and C-reactive protein (CRP). C-reactive protein was measured using immunoturbidimetric assays provided by the respective analyzer manufacturer: Cobas c501 (Roche Diagnostics) in Zagreb, Architect c4000 (Abbott Laboratories) in Pula, Olympus AU480 (Beckman Coulter) in Osijek, and Dimension EXL with LM (Siemens Healthcare) in Sisak. Data sets were tested for normality using the Shapiro-Wilk test and presented as medians and interquartile ranges. Differences in tested laboratory parameters between two patient groups were assessed using the Mann-Whitney test (Table 2); P < .05 was considered statistically significant. Statistical analysis was performed in MedCalc®, version 19.3.1 (MedCalc). The present analysis of CBC parameters in patients affected with COVID-19 compared to non-COVID-19 patients presenting with overlapping febrile and/or respiratory symptomatology demonstrates that this novel viral infection does not significantly affect the WBC count nor leukocyte subpopulation distribution at the early stage of the disease. Lymphopenia, being predominantly mild, was observed in only 28% of COVID-19 cases, while CRP was only mildly to moderately increased in the majority of our COVID-19 patients. However, an increase in WBC count driven by a growing NE population, with diminishing numbers of LY, MO, and EO, and accompanied by CRP elevations could be expected at the later stage of the disease, indicating disease progression and poor outcome.2, 7-9 The underlying mechanism for LY reduction is not fully understood, yet it is probable that increased LY apoptosis happens as a consequence of the pro-inflammatory cytokine storm, which causes direct cytopathic effect as well as induces the expression of ACE2 receptor on LY surface that in turn serves as the target for SARS-CoV-2.7, 10 Most CPD parameters in COVID-19 patients displayed lower and rather different values compared to non-COVID-19 patients with respiratory flu-like symptoms. Typically, parameters related to NE size (NE-FSC), nucleic acid content (NE-SFL), and internal complexity (NE-SSC) coupled with increased NE count reflect NE activation in fever, usually caused by bacterial infections.3 Obviously, such pattern was not observed in our COVID-19 group. However, their substantial increase could be expected in patients developing COVID-19-associated complications, in the first place bacterial coinfections.8 Similarly, COVID-19 patients were not characterized by notable MO activation, with MO-X, MO-Y, and MO-Z being significantly lower than in non-COVID-19 cases. Conversely, studies performed on a different hematology analyzer that applies direct impedance, radio frequency current, and light scattering for WBC counting and differential showed that MO-related CPD parameters were significantly increased in COVID-19 and in correlation with disease progression, thus potentially providing satisfactory discriminating capacity at the early stage of COVID-19 as well as disease monitoring.11, 12 These contradictory results indicate that CPD largely depend on the technology used within the hematology analyzer and cannot be interpreted nor used interchangeably. Notably, LY-WZ was prominently represented in the COVID-19 group displaying higher and nonoverlapping values compared to the control group, thereby reflecting larger LY size and subsequent wider light dispersion. On the contrary, a significantly lower trend of values was observed for LY-X, LY-Y, and LY-Z in COVID-19 patients. This interesting pattern of LY-associated CPD parameters might suggest that infection caused by SARS-CoV-2 at the early stage induces LY volumetric expansion, with nonevident or only subtle morphological changes when compared with other respiratory infections. LY activation reflected by LY-WZ, coupled with lymphopenia and predominantly normal values of both total WBC and absolute NE count, could be a combined result of the antiviral immune response and inflammatory reaction that in conjunction with clinical data might suggest COVID-19 etiology. Other CPD parameters of NE, MO, and LY related to internal cell complexity mostly presented with overlapping values; thus, these CPD may not be useful in separating COVID-19 from non-COVID-19 cases at the early stage of the disease. However, it can be justifiably presumed that disease progression with associated inflammatory complications can lead to additional significant changes in cell complexity, morphology, and size, and could be therefore reflected by changes of CPD parameters. Thus, additional investigations, including monitoring of CPD changes at different time points, especially 7-14 days from symptom onset which is known to be the peak cytokine production period, as well as in relation to disease progression, are needed. Therefore, studying our patients at a single time point can be considered as a limitation of our study. On the other hand, our data taken at presentation may be of help in early patient stratification. Our results are also limited by the lack of clinical data and of the definite cause for respiratory and/or febrile symptoms in COVID-19-negative patients. Also, CPD data are dependent on the type and technology used within the hematology analyzer, so these findings are not transferable to other analytical platforms. In conclusion, the data obtained herein demonstrate that at the early stage of COVID-19, the usually reported WBC parameters are only mildly altered. Most of the CPD obtained by the technology applied on Sysmex XN-1000 hematology analyzer were significantly different compared to other conditions presenting with similar symptomatology, with LY-WZ elevation being the most pronounced finding indicating that increase in LY size is associated with response to SARS-CoV-2 infection. Further studies, especially longitudinal monitoring of COVID-19 patients, are needed to better elucidate the relationship between changes of hematological parameters and disease progression, as well as to understand the underlying pathophysiological mechanisms and the true effect of SARS-CoV-2 stimuli on WBC morphology changes that are eventually reflected by CPD. Additional studies should also focus on other viral diseases with similar clinical presentation as these CPD changes may not be unique for COVID-19. None to declare. IL conceived and designed the study. IL, TB, ML, ILu, and MK performed laboratory measurements. IL, TB, DR, ML, ILu, MK, LH, and VŠ analyzed and interpreted the data. IL, TB, and DR wrote the manuscript. ML, ILu, MK, LH, and VŠ revised the manuscript critically. All authors approved the final version of the article for submission. All data from this research are available from corresponding author upon reasonable request.