Associated Biochemical and Hematological Markers in COVID-19 Severity Prediction

Background The global threat of COVID-19 has created the need for researchers to investigate the disease's progression, especially through the use of biomarkers to inform interventions. This study aims to assess the correlations of laboratory parameters to determine the severity of COVID-19 infection. Methods This study was conducted among 191 COVID-19 patients in Sumeru Hospital, Lalitpur, Nepal. According to their clinical outcomes, these patients were divided into severe and nonsevere groups. Inflammatory markers such as LDH, D-dimer, CRP, ferritin, complete blood cell count, liver function tests, and renal function tests were performed. Binary logistic regression analysis determined relative risk factors associated with severe COVID-19. The area under the curve (AUC) was calculated with ROC curves to assess the potential predictive value of risk factors. Results Out of 191 patients, 38 (19.8%) subjects died due to COVID-19 complications, while 156 (81.7%) survived and were discharged from hospital. The COVID-19 severity was found in patients with older age and comorbidities such as CKD, HTN, DM, COPD, and pneumonia. Parameters such as d-dimer, CRP, LDH, SGPT, neutrophil, lymphocyte count, and LMR were significant independent risk factors for the severity of the disease. The AUC was highest for d-dimer (AUC = 0.874) with a sensitivity of 82.2% and specificity of 81.2%. Similarly, the cut-off values for other factors were age >54.5 years, D-dimer >0.91 ng/ml, CRP >82.4 mg/dl, neutrophil >78.5%, LDH >600 U/L, and SGPT >35.5 U/L, respectively. Conclusion Endorsement of biochemical and hematological parameters with their cut-off values also aids in predicting COVID-19 severity. The biomarkers such as D-dimer, CRP levels, LDH, ALT, and neutrophil count could be used to predict disease severity. So, timely analysis of these markers might allow early prediction of disease progression.


Introduction
Coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus-2 has become a major global health problem leading to pandemic outbreaks [1].Most studies illustrated that elevated ferritin, CRP, and d-dimer were associated with mortality risk [2].Early severity prediction can help to provide quality treatment, which might enhance the survival rate by lowering mortality [3].For the prognosis of the infection, laboratory parameters such as the hematological profle and the infammatory index have a signifcant predictive value.Laboratory features such as leukocytosis, lymphopenia, neutrophilia, NLR, PLR, CRP, LDH, D-dimer, ferritin level, and liver enzymes are signifcantly associated with the severity of the disease [4].With an increase in the severity of the disease, lymphopenia is encountered at a higher rate in the nonsurvival patients compared to the survivors [5].In COVID-19, due to viral activity, hematological and infammatory marker changes occur that cause damage to liver tissue leading to a rise in ALT, AST, and LDH activity [5].
A study by Shang et al. states that hemodialysis patients possess a greater risk of COVID-19 infection [6].Conversely, comorbidities such as diabetes, hypertension, immunodefciency, and cardiovascular disease are considered risk factors for COVID-19 infection whereas acute respiratory distress syndrome (ARDS), septic shock, acute kidney injury, cardiac injury, and multiorgan failure are identifed as complications of COVID-19 disease [7].
A better understanding of early prognostic clinical laboratory parameters helped to save many lives by enabling timely intervention and better resource allocation since ICU capacity is limited in most of the countries [8].Hence, this study aims to fnd the clinical and laboratory biomarkers associated with COVID-19 severity.

Materials and Methods
Patients diagnosed with COVID-19 by quantitative reverse transcription PCR (qRT-PCR) and admitted to Sumeru Hospital Pvt. Ltd., Lalitpur, Nepal, from March 2021 to October 2021, were included in this study.

Exclusion and Inclusion Criteria.
Patients with COVID-19 infection were further verifed by positive reverse transcription polymerase chain reaction (RT-PCR), and clinical symptoms were included in the study.Patients with asymptomatic conditions and duplicate samples were excluded from the study.
All patients above 18 years, fulflling the criteria mentioned previously, were given informed and written consent and recorded their demographic data, comorbidities, and laboratory parameters using standard performa.

Experimental
Protocol.Blood sample was collected in diferent vacutainer system.Whole blood was collected in an EDTA vacutainer and analyzed hematological parameters: complete blood cell count (CBC) including hemoglobin, red blood cell (RBC), packed cell volume (PCV), platelet count, white blood cell (WBC), and diferential count using a fully automated blood cell counter (Sysmex XN-350).Trisodium citrate plasma was collected to analyze the D-dimer test using an immunofuorescent analyzer (iChroma II).Serum was separated to analyze diferent biochemical tests such as random blood sugar (RBS), lactate dehydrogenase (LDH), renal function tests including urea, creatinine, sodium, and potassium, and liver function tests including total bilirubin (TB), direct bilirubin (DB), alanine transferase (ALT), aspartate transferase (AST), and alkaline phosphatase (ALP) using fully automated biochemistry analyzer (Erba XL 200).Ferritin was analyzed from fully automated CLIA (Roche Cobas e411), and C-reactive protein (CRP) was analyzed by an immunofuorescent analyzer (iChroma II).Daily internal quality control and quarterly external quality control were conducted to validate tests.
COVID-19 diagnosis and clinical classifcation were made from the new coronavirus pneumonia diagnosis and treatment plan (Trial Version 7) established by the People's Republic of China's National Health Commission as follows: (1) Mild, with few symptoms and no evidence of pneumonia on imaging (2) Moderate, with fever and signs of pneumonia on imaging (3) Severe, with any of the following: the following symptoms of respiratory distress were observed: (a) respiratory rate 30 beats/min; (b) oxygen saturation 93%; (c) arterial blood oxygen partial pressure 300 mmHg; and (d) pulmonary imaging revealed a lesion that had advanced by more than 50% in 24-48 hours (4) A critical condition, one of the following: a respiratory failure that necessitates mechanical ventilation, shock, and a need for ICU admission due to multiple organ failure [9] Furthermore, we categorized the COVID-19-infected patients into two categories as mild and moderate or severe according to the above-mentioned criteria.

Statistical Analysis.
Te data were analyzed in IBM SPSS (IBM Corp. Released 2019.IBM SPSS Statistics for Windows, Version 26.0.Armonk, NY: IBM Corp).Normally distributed data were expressed as mean and standard deviation and compared by independent sample t-test.Nonnormally distributed data were presented as median and interquartile range and compared by the Mann-Whitney U test.Te Chi-square test compared categorical variables.Pearson's correlation was used to determine the correlation between severity and variables.All variables were subjected to univariate logistic regression, and odds ratios were calculated between nonsevere and severe groups, with a 95% confdence interval.Variables were included in binary logistic regression if the corresponding P value was less than 0.05.Binary logistic regression analysis was used to determine relative risk factors associated with severe COVID-19.Te area under the curve (AUC) was calculated with ROC curves to assess the cut-of values of potential predictive factors.

Results
Out of 191 patients, 101 patients belong to the mild symptomatic group and 90 to the severe group.Te average age of patients was found to be 54.9 ± 17.9 years.Te mean age of the severe group (60.31 ± 17.35) was signifcantly higher (P < 0.001) than that of the mild group (50.08 ± 17.10) (Table 1).
Based on the clinical history, the prevalence of comorbidities such as CKD, HTN, DM, COPD, anemia, and pneumonia was signifcantly higher in the severe population (P < 0.001, P � 0.034, P � 0.049, P < 0.001, P � 0.013, and P � 0.005, respectively).Te demographic, clinical, and laboratory parameters of the patients are presented in Table 1.
In the univariate analysis, risk factors associated with disease severity were age, NLR, LMR, PLR, LCR, ferritin, LDH, CRP, D-dimer, RBC count, Hb, WBC count, neutrophilia, lymphopenia, PCV, urea, creatinine, and aminotransferases.Variables that signifcantly afect the disease progression revealed by univariate analysis were entered into multivariate logistic regression analysis.Variables such as age, LMR, LDH, CRP, D-dimer, neutrophil, lymphocyte count, and AST remained the independent risk factors for the severity of the disease (Table 2).
Te AUROC, standard error, and 95% CI of each parameter are shown in Table 3, and the cut-of value was estimated from Figure 1.Te area under the curve was highest for D-dimer (AUC � 0.874, 95% CI � 0.824-0.924,P < 0.001), and the best cut-of point was 0.91 ng/ml, with a sensitivity of 82.2% and specifcity of 81.2%.Similarly, the ROC curve of CRP (AUC � 0.857, 95% CI � 0.803-0.910,P < 0.001) suggested the best cut-of point of 82.4 mg/dl with a sensitivity of 82% and specifcity of 75.2%.Similarly, the Advances in Medicine cut-of values for age, neutrophil, LDH, ALT, and ferritin were 54.5 years, 78.5%, 600 IU/L, 35.5 IU/L, and 43.5 ng/ml, respectively, for the severity prediction (Figure 1).

Discussion
Te pandemic outbreak of COVID-19 infection worldwide has created a heavy burden on healthcare services.Mortality is the major issue in dealing with this pandemic, and early identifcation of severe and critical cases is essential to minimize the mortality and improve the recovery rate [10].Te fatality rate was 4.3-15% among hospitalized COVID-19 patients [6].Terefore, to determine the severity of COVID-19 infection, it is necessary to ascertain the early predictors that could help clinicians to identify the early stage of infection for timely diagnosis and treatment [11].We examined the demographic and laboratory characteristics of 191 COVID-19-infected individuals in this study, and 38 (19.8%) died afterward.In our study, the population of the moderate and severe groups was signifcantly higher in age as compared to that of the mild group, with the best cutof point by ROC being 54.4 years, which is similar to the study of Wang et al. [12].A multicenter study showed that in the older age group, the risk of mortality increased by 18% [13].COVID-19 patients sufering from comorbidities were at high risk for mortality [13].Several studies suggested that proper consideration must be provided to comorbidity [14].In our study, the patients sufering from hypertension, CKD, DM, COPD, pneumonia, and anemia were highly severe, similar to a study by Yilmaz et al. [7,13].Similarly, a higher mortality was observed in the COVID-19 patients sufering from CKD, which is in accordance with the study of Posso et al.Furthermore, studies suggested that COVID-19 infection and human immunological response could lead to kidney illness [15].
Among the laboratory fndings, Hb, total WBC count, neutrophil count, lymphocyte count, urea, creatinine, potassium, ALT, AST, Ferritin, LDH, CRP, D-dimer, NLR, and PLR were found to be signifcantly higher; however, RBC count, LMR, and LCR were lower among the moderate or  4 Advances in Medicine severe group of the population as compared to the mild group, which is similar with several studies [3,14,16].In severe cases, neutrophilia might be due to the activation of neutrophils as an immunological response to the virus [14,16].Tis result is in support of the study of Yang et al. reported higher NLR and PLR and lower LMR and LCR in the severe group [17].NLR and PLR were positively correlated with critical illness, whereas LMR and LCR were negatively correlated.In this study, D-dimer is one of the major risk factors associated with fatal outcomes in COVID-19 patients.Ddimer increases in the disseminated intravascular coagulation are considered an early stage pulmonary intravascular coagulopathy.A study suggested severe thrombosis in tiny capillaries and microvasculature in the lung tissue of COVID-19 patients, indicating that D-dimer is a signifcant prognostic indicator in individuals with probable infection and sepsis [18].
CRP, an acute-phase protein, is independently associated with critical illness in COVID-19.Our study showed higher CRP levels in the severe group of patients with an odds ratio of 1.021 in multivariate logistic regression, which is in accordance with the study of Sharifpour et al. [19].CRP being an independent prognostic factor may facilitate risk stratifcation and prognostication.In this study, CRP and d-dimer were found to be independent risk factors associated with fatal outcomes, which is similar to the study of Liu et al. and Wang et al. [12,20].
Similarly, based on ROC analysis among severe/moderate and mild populations, the cut-of values of diferent variables such as age is >54.5 years, D-dimer is >0.91 ng/ml, CRP is >82.4 mg/dl, neutrophil is >78.5%,LDH is >600 U/L, and SGPT is >35.5 U/L for severity prediction.It indicates the progression of the disease to a critical case which must be observed for the prevention of COVID-19 complications [12].Tese fndings suggest that routine monitoring of these parameters is useful for improving the early diagnosis of critical COVID-19 and establishing an accurate therapeutic strategy [21].
However, the study was limited to a single center, and the sample size was smaller.In addition, the patient's disease state, clinical symptoms, onset, and exposure history were not considered.Furthermore, radiological diagnoses such as CT scan and other laboratory indicators associated with severity, such as cytokines and procalcitonin, were not measured.

Conclusion
Tis study revealed that elder age, D-dimer, CRP levels, LDH, ALT, and neutrophil count were independent severity risk factors as compared to the case of mild patients.Evaluation of these factors could aid in detecting disease progression.Early medical care and assistance for these high-risk patients may help to minimize the disease's fatality rate.Advances in Medicine

5. 1 .
Clinical Signifcance.Early diagnosis and monitoring of the disease progression play crucial roles in the management of COVID-19 complications.[22]Screening of biochemical and hematological biomarkers, primarily D-dimer, CRP, LDH, ALT, and neutrophil count helps triaging COVID-19infected elderly individuals early in the disease's course.
segments are produced by ties.

Figure 1 :
Figure 1: Receiving operating characteristics for generating cut-of values of d-dimer, LDH, ferritin, age, CRP, SGPT, and neutrophil for the severity of COVID-19 infection.

Table 1 :
Sociodemographic and other laboratory investigations with the prognosis of COVID-19.

Table 2 :
Univariate and multivariate logistic regression analyses of risk factors associated with severe COVID-19.

Table 3 :
Predictive efcacy of the severe COVID-19 risk model and early predictors.