Peripheral blood inflammatory markers in predicting prognosis in patients with COVID‐19. Some differences with influenza A

Abstract Background To evaluate the ability of peripheral blood inflammatory markers in predicating the typing of COVID‐19, prognosis, and some differences between COVID‐19 and influenza A patients. Methods Clinical data on 285 cases laboratory‐confirmed as SARS‐CoV‐2 infection were obtained from a Wuhan local hospital's electronic medical records according to previously designed standardized data collection forms. Additional 446 Influenza A outpatients’ hematologic data were enrolled for comparison. Results NLR, SII, RLR, PLR, HsCRP, and IL‐6 were significant higher and LMR was lower in severe COVID‐19 patients than in mild COVID‐19 patients (p < .001). PLR and LMR were lower in the individuals with influenza A than those with COVID‐19 (p < .01). COVID‐19 patients with higher levels of NLR, SII, RLR, PLR, HsCRP, and IL‐6 and lower LMR were significantly associated with the severe type. AUC of NLR (0.76) was larger while the specificity of IL‐6 (86%) and sensitivity of HsCRP (89%) were higher than other inflammatory markers in predicating the typing of COVID‐19. PT had obvious correlation with all the inflammatory markers except RPR. NLR showed positive correlations with AST, TP, BUN, CREA, PT, and D‐dimer. Patients with high IL‐6 levels have a relatively worse prognosis (HR = 2.30). Conclusion Peripheral blood inflammatory markers reflected the intensity of inflammation and associated with severity of COVID‐19.NLR was more useful to predict severity as well as IL‐6 to predict prognosis of COVID‐19. PLR and LMR were initially found to be higher in SARS‐CoV‐2 virus‐infected group than in influenza A.


| INTRODUC TI ON
The epidemic of SARS-CoV-2 infection has spread globally, posing a great threat to public health. 1 SARS-CoV-2 infects the human body through the ACE2 receptor and people who was infected has clinical manifestations such as fever, dry cough, fatigue, and respiratory and digestive systems. 2,3 Patients with mild symptoms account for the majority. 4,5 The mortality of severe patients with SARS-CoV-2 pneumonia is considerable. 6 Severe patients may develop into septic shock, difficult to correct metabolic acidosis, coagulation dysfunction, and multiple organ dysfunction syndrome (MODS), etc rapidly. 7 Past studies have confirmed that cytokine storm/systemic inflammatory response syndrome (SIRS) and subsequent compensatory inflammatory response syndrome (CARS) are involved in the pathophysiology of sepsis, 8 and some studies have found that SARS-COV-2-induced viral sepsis has a large-scale release of inflammatory cytokines and immunosuppression. 7,9,10 It has been reported that some new blood inflammatory indexes [11][12][13] and high-sensitivity C-reactive protein (HsCRP),interleukin 6(IL- 6) are related to a variety of inflammatory reactions including sepsis. Recent researches 14,15 found the disseminated coagulation due to the large production of inflammatory cytokines damaged organs and aggravated the condition. However, further verification is needed in clinical practice. And given the similarity of the symptoms of influenza and COVID-19, the difference of these inflammatory markers between both patients remains to be found. In this study, we retrospectively analyzed the association of these inflammatory markers with clinical typing and prognosis of COVID-19 and some differences between patients of influenza and COVID-19. The laboratory information of outpatients includes complete blood count and HsCRP. The course of disease was defined as the number of days when a patient first noted the onset of symptoms to the day of admission. Comorbidity was determined using the age-adjusted Charlson Comorbidity Index(aCCI) 16,17 and was classified into three categories: no comorbidity (aCCI = 0), mild to moderate comorbidity (aCCI = 1-3), and severe comorbidity (aCCI = 4 or more).

| Diagnostic criteria
Patients were diagnosed with COVID-19 according to the "Guidelines for the Diagnosis and Treatment of New Coronavirus Pneumonia" (5th edition) 18 issued by the National Health Commission of China.

| Statistical analysis
Categorical variables are expressed as frequency or percentage, and significance is tested by chi-square test or Fisher's exact test. The continuous variables of the parameters are expressed as mean ± standard deviation, and significance is tested by t test. Nonparametric variables are expressed as medians and quartiles, and significance was tested by Mann Whitney U test. The diagnostic value of selected parameters used to distinguish between mild and severe COVID-19 patients was evaluated by the receiver operating characteristics (ROC) and the area under the ROC curve (AUC), and the critical value was calculated based on the maximum Youden index.
Binary logistic regression analysis was used to select relevant factors that affect patients with mild and severe COVID-19. Prognostic factors were determined using Cox regression analysis. Analysis was performed using SPSS 24.0 and GraphPad Prism 8 statistical software packages. In all statistical analyses, p < .05 was considered statistically significant.

| Demographic and clinical characteristics
Demographic and biochemical characteristics of 285 enrolled patients were summarized in Table 1. All of them were local residents of Wuhan. The cases were 211 mild (74%) and 74 severe (26%), 151 females (53%) and 134 males (47%). There was no significant difference in gender (p = .385), the course of the disease was mostly concentrated in 2-3 weeks and has no difference between groups (p = .449). Significant difference between two groups was observed in the median age (p = .001), aCCI score (p < .001), comorbidity categories (p < .001) and complications include hypertension (P = .002), diabetes (p < .001), and heart disease (p = .001). Neutrophils (p < .001), lymphocytes (p < .001), eosinophils (p = .001), and the inflammatory markers NLR, SII, RLR, PLR, HsCRP, and IL-6 were significant higher and LMR was lower in severe patients than in mild (p < .001). Our study also showed significantly difference in the AST, TP, BUN, PT, and D-dimer concentrations (p < .05). Different distribution of patients with abnormal ALT and CREA was observed between two groups (p < .05), and the prognosis of two groups was also different (p < .001) ( Table 1).

| Some differences between patients with COVID-19 and influenza A
The general clinical data were compared between 285 COVID-19 and 446 influenza A patients. There was difference in age, aCCI score, and comorbidity categories between two groups (p < .001).
After preliminary comparison, PLR and LMR were lower in the individuals with influenza A than those with COVID-19 (p < .01) ( Table 2). For the other factors, further expansion of the groups will be necessary.

| Correlation between inflammatory markers and other laboratory parameters
We analyzed the correlation between laboratory indicators includes  (Table 4).

| Inflammatory markers in predicting the prognosis of patients with COVID-19
To investigate the associations between the inflammatory markers

| D ISCUSS I ON
Among the baseline data of these two groups of COVID-19 patients, the median age is higher than other reports 1,6,19 which may be related to the hospital's own patient source. The previous study has noticed the differences of lymphocytes and neutrophils between mild and sever patients. 10,20-22 Also the higher neutrophils count than healthy people were found in COVID-19 patients. 23 In our study, a clear reduction of lymphocytes and increase of neutrophils which was more intense than in mild group were observed in severe patients. The potential reasons of this phenomenon may come from the physiological responses of the innate immune system to systemic inflammation. 24 It has been reported that ACE2 is the receptor of SARS-COV-2 and plays a crucial role in the infection, 25 lymphocytes which express the ACE2 may be a direct target of viruses that vulnerable to be attacked, 26 and SARS-CoV-2-induced NKG2A expression may be correlated with functional exhaustion of cytotoxic lymphocytes at the early stage, which may result in disease progression. 27 Didangelos A 28 used a computational protein-protein interaction network to identify possible SARS-CoV-2 inflammatory mechanisms and bioactive genes, the study found that neutrophils could be recruited by SARS-CoV-2, and lung epithelial cells overexpress neutrophil chemokines after SARS-CoV-2 infection. Complement C3 and tumor necrosis factor (TNF) also have been recently shown to be involved in neutrophil activation and prolong neutrophil survival. 22 Taken together, both lymphopenia and neutrophils increase are the adaptive response of the immune system to SARS-COV-2 invasion. NLR was defined as the ratio of neutrophils and lymphocytes; our study found NLR in the severe group was significantly higher than it in the mild; the AUC of NLR is larger than the other four inflammatory markers; binary logistic regression analysis showed that high NLR levels were more likely to be severe patients; and our analysis also shows that  In contrast to people in mild group, people in sever group were found to have higher IL-6 levels. IL-6 also has the greatest specificity in predicting the sever type of patients with COVID-19 among all the inflammatory markers and is associated with a poor clinical outcome.
In the same vein, much of the literature 7,29-31 on COVID-19 reported the elevated IL-6 levels which might serve as a predictive biomarker for disease severity. 32 More evidence suggests that SARS-CoV-2 has either immune dysregulation or macrophage-activation syndrome, both of which are characterized by pro-inflammatory cytokines, 33 and the immune dysregulation is driven by the Interleukin-6 (IL-6). 34 Individuals with influenza and COVID-19 can present with similar symptoms. 35 Influenza is typical also with the inflammasome in mediating the inflammatory response after infection. 36  29 years), further expansion of the groups is necessary to other factors and in-depth comparisons.
Interestingly, not only are PT and D-dimer observed to be higher in severe patients, they are also significantly associated with inflammatory markers. In accordance with the present results, previous studies 7,14,41 have demonstrated that higher D-dimer concentrations associated with poor prognosis. In an observational study, 14  were not in the early stages of the disease; therefore, the earlier information of inflammatory markers on patient cannot be traced.
So, course of disease was used as a covariate in correlation analysis of the inflammation markers and the classification of COVID-19 patients. Compared with patients with low NLR levels, those with high NLR levels were more likely to be severe patients, which is similar to the findings published by Liu J et al. 43 The results of COX analysis also revealed that the severe comorbidity and the high levels of NLR were associated with a poor prognosis in patients with COVID-19, after adjusted for course of disease.
In summary, our study shows that NLR, SII, RLR, PLR, LMR HsCRP, and IL-6 reflected the intensity of inflammation and associated with severity of patients with COVID-19. NLR was more useful to predict the severity and IL-6 could better predict the prognosis of COVID-19 patients than other inflammatory markers. There was a clear correlation between coagulation indicators and most inflammatory markers. PLR and LMR were initially found to be higher in SARS-CoV-2 virus-infected group than in influenza A.

CO N FLI C T O F I NTE R E S T
The authors disclose no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets used during the current study are available from the corresponding author on reasonable request.