Systemic In ammatory Cytokines Associate With SARS-COV-2 Viral Shedding Time in Covid-19 Inpatients


 Background: Since December 2019, coronavirus disease 2019 (COVID-19), as an infectious disease with cytokine storm, has become an emerging global challenge. To assess the duration of SARS-COV-2 viral shedding and associated risk factors in COVID-19 patients.Methods: COVID-19 patients with interleukin (IL)-1b, soluble interleukin-2 receptor (sIL-2R), IL-6, IL-8, IL-10 and tumor necrosis factor (TNF)-α cytokines data consecutively admitted to Tongji Hospital from January 27, 2020 through February 5, 2020 were enrolled and been followed up until March 24, 2020. We utilized Kaplan-Meier method and Cox proportional hazards regression analysis to assess the duration of viral shedding and risk factors affecting virus clearance.Results: 246 inpatients with laboratory confirmed COVID-19 were enrolled. The median duration of viral shedding was 24 days, ranging from 6 to 63 days. Age, severity of COVID-19, albumin, lactate dehydrogenase (LDH), D-dimer, ferritin and sIL-2R were associated with duration of viral shedding. Administration of lopinavir-ritonavir, arbidol, oseltamivir and intravenous immunoglobulin did not shorten viral shedding time. Multivariate cox regression analysis revealed that sIL-2R, LDH and severity of COVID-19 were independent factors associated with duration of viral shedding. At stratified analysis, the viral shedding time was positively correlated with age, sIL-2R and LDH in non-corticosteroid subgroup, while negatively correlated with lymphocyte count in corticosteroid group. Conclusions: The present study demonstrated that elevated sIL-2R, increased LDH and severe status were related to prolongation of viral shedding in COVID-19 inpatients. Further research is urgent to investigate the mechanism of immune reaction involved in the virus clearance process and aim to the optimal antiviral therapy.

, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), presents a global challenge. Up to May 4, 2020, more than 3 million con rmed cases of COVID-19 have been reported, of which more than 239,000 have died. 1 As SARS-CoV-2 is an emerging virus, there is no effective anti-viral drug and vaccine to treat this disease. Supportive therapy is the main pattern. Whether corticosteroids are bene cial or lead to delayed virus shedding remains controversial.
Previous studies found that the SARS-CoV-2 persist longer with higher load in severe patients. 2 Our previous studies found that high cytokine levels, such as soluble interleukin-2 receptor (sIL-2R), interleukin (IL)-6, IL-10 and tumor necrosis factor (TNF)-α were signi cantly associated with severe COVID-19. 3 The in ammatory storm caused by SARS-CoV-2 plays an important role in the deterioration of the disease. 4 However, the association between duration of viral shedding and in ammatory factors was limited.
The assessment of risk factors for delayed virus shedding will bene t the treatment and management of COVID-19. The goal of the present study was to conduct a large population-based study to describe the data for duration of viral shedding, and assess the association between duration of viral shedding and relevant characteristics, including age, gender, smoking history, severity of the disease, comorbidities, in ammatory factors and treatment method.

Study design and participants
This study is an observative study of hospitalized patients with COVID-19 enrolled at Sino-French New City Branch of Tongji Hospital, Huazhong University of Science and Technology in Wuhan January 27, 2020 through February 5, 2020. Tongji hospital, HUST is one of the major nationally designated hospitals for COVID-19 in Wuhan. COVID-19 was diagnosed according to WHO interim guidance. 5 Only the cases with complete cytokines data, including IL-1b, sIL-2R, IL-6, IL-8, IL-10 and TNF-α, were included in the analysis. Patients were divided into severe COVID-19 group and non-severe COVID-19 group according to 2019 clinical practice guideline from IDSA and ATS for diagnosis and treatment of adults with community-acquired pneumonia. 6 Laboratory con rmed case of COVID-19 was de ned as a positive result to next-generation sequencing or real-time reverse-transcriptase polymerase chain reaction assay for nasopharyngeal or pharyngeal swab specimens. 7 Patients were followed up from the date of admission to March 24, 2020. Viral shedding was de ned as SARS-CoV-2 nucleic acid test turned negative for two consecutive times (minimum 24h sampling interval). Duration of viral shedding was de ned as the duration from illness onset to the date of rst negative result of viral shedding. This study was approved by Institutional Review Board of Tongji Hospital, Huazhong University of Science and Technology. Written informed consent was waived owing to the rapid emergence of this infectious disease.

Data collection
The demographic data were obtained by face-to-face or telephone interview. Laboratory examination at admission, treatment during hospitalization and outcome data were extracted from electronic medical records. Patient data were cross-checked for consistency before nal data entry and then entered into a computerized database.

Statistical analysis
Continuous variables were expressed as median and interquartile range (IQR). Categorical variables were expressed as number (%). Survival curves were performed by Kaplan-Meier method and validated by Logrank, Breslow and Tarone-Ware tests. The time-dependent ROC curve analysis was conducted and the area under the curve (AUC) value was used to evaluate the predictive effect of laboratory ndings. Adjusted hazard ratios (HRs) along with 95% con dence intervals (CIs) were calculated by using the Cox proportional hazards regression model. Age, diabetes, lymphocyte count, albumin, sIL-2R and usage of systemic corticosteroids were chosen for multivariate cox regression analysis on the basis of univariate cox regression analysis and clinical signi cance. The correlation between duration of viral shedding and age, sIL-2R, LDH and lymphocyte count was analyzed by Spearman's rank correlation. When the P value was <0.05, the difference was regarded as statistically signi cant. All statistical tests were two tailed. Overall statistical analysis was performed by the software statistical package for social science version 23.0 (SPSS Inc., Chicago, IL, USA) and R software for Windows, version 3.6.1.

Patient Characteristics
A total of 246 patients with COVID-19 were enrolled January 27, 2020 through February 5, 2020. Median age of the patients was 58 years (IQR 47-67). 126 (51.2%) patients were male. 22 (9.0%) patients were current smokers. 72 (29.3%) patients had hypertension, 45 (18.3%) patients had diabetes. 16 (6.5%) patients had coronary heart disease. 16 (6.5%) patients had tumor or impaired immune system disease.  Table 1 in the Online Repository. The overall median duration of viral shedding was 24 days. The shortest duration of viral shedding was 6 days from illness onset, whereas the longest one was 63 days. One patient was persistent positive until the end of follow-up date.
By the analysis of Kaplan-Meier method, the severity of disease, old age, low albumin level, high LDH, high D-dimer, high ferritin, high sIL-2R and low lymphocyte count were associated with longer duration of viral shedding time (Table 2, Figure 1). Gender, smoking history, comorbidity, and administration of lopinavir-ritonavir, arbidol, oseltamivir and intravenous immunoglobulin were not associated with duration of viral shedding ( Table 2).

Time-dependent ROC curve analysis
The time-dependent ROC curve analysis was conducted and the AUC value was used to evaluate the predictive effect of albumin, LDH, D-dimer, ferritin, sIL-2R. As shown in Figure 2, the AUC for albumin was 0.646, 0.708 and 0.665 at 20, 40, and 60 days, respectively; the AUC for LDH was 0.627, 0.747 and 0.762 at 20, 40, and 60 days, respectively; the AUC for D-dimer was 0.648, 0.714 and 0.710 at 20, 40, and 60 days, respectively; the AUC for ferritin was 0.628, 0.652 and 0.683 at 20, 40, and 60 days, respectively; the AUC for sIL-2R was 0.658, 0.660 and 0.710 at 20, 40, and 60 days, respectively.
Univariate and multivariate cox regression analysis As shown in Table 3 (Table 3).

Strati ed analysis
According to whether systemic corticosteroid was used, the correlation between duration of viral shedding and age, sIL-2R, LDH and lymphocyte count was strati ed analyzed. As shown in Figure 3

Discussion
Little do we know about the duration of viral shedding of COVID-19, a novel infectious disease. As far as we know, this study proposed the longest follow-up duration of the viral shedding in COVID-19. Our result revealed that the viral shedding time ranged from 6 to 63 days, which was wider than previous study in Wuhan with a range of 8 to 37 days 8 and another study performed in Shanghai with a range of 2 to 22 days. 9 The longer duration of viral shedding time in this study may result from a higher severity of the patients recruited in Tongji hospital and a longer duration of follow-up time.
In univariate cox regression analysis, age was signi cantly associated with duration of viral shedding. Elderly patients experience a marked decline of immune function. The age-dependent defects in T-cell and B-cell function and the excess production of type 2 cytokines could lead to a de ciency in control of viral replication. 8,10 Diabetes was the most common comorbidity in elderly patients with COVID-19. Previous study found that COVID-19 patients with diabetes were at higher risk of severe pneumonia, release of tissue injury-related enzymes, excessive uncontrolled in ammation response, hypercoagulable state and rapid deterioration. 11 However, there was no difference in viral shedding time between patients with or without diabetes in our study. Larger sample studies are needed to explore virus clearance ability of patients with diabetes. In the present study, although the P value of Log-rank test is greater than 0.05, but the P values of Breslow and Tarone-Ware tests were less than 0.05, lymphocyte count < 0.5×10 9 /L may be associated with delayed virus shedding. Lymphocytes play a crucial role in immune system.
Lymphopenia was commonly observed in severe COVID-19 illness. 8 Analyze of lymphocyte subsets demonstrated that absolute numbers of T lymphocytes, CD4+ and CD8+ T cells decrease in nearly all the patients, and were markedly lower in severe cases. 12 Corticosteroids could suppress cytokine storm of COVID-19, but also inhibit immune response and holds the possibility to delay viral clearance. In severe acute respiratory syndrome and Middle East respiratory syndrome, corticosteroids were not routinely recommended. 13 Whether corticosteroids are bene cial or lead to poor clinical outcomes in COVID-19 remains controversial. Our study revealed that for the comparison between high dose systemic corticosteroids and no use of systemic corticosteroids, although the P value of Log-rank test is greater than 0.017, the P values of Breslow and Tarone-Ware tests were less than 0.017. High dose systemic corticosteroids may associate with prolonged viral shedding compared with no use of systemic corticosteroid. This may due to the impaired immune response aggravated by high-dose systemic corticosteroids, but also could be an indicator of severe condition when high dose corticosteroid would be used. Further research is needed to clarify the role of systemic corticosteroid in viral shedding of COVID-19. The present study also identi ed that administration of lopinavir-ritonavir did not shorten viral shedding time (Log Rank, P>0.05; Breslow, P>0.05; Tarone-Ware, P>0.05). Our result was consistent with previous clinical trial that the percentage of patients with detectable viral RNA for SARS-CoV-2 at various time points were similar between the lopinavir-ritonavir group and the standard-care group. 14 Oseltamivir is an antiviral used to treat in uenza virus. December and January were the high incidence period of in uenza in Wuhan previously. Co-infected with in uenza virus was likely existed in COVID-19 patients. Although oseltamivir was prescribed to 39% patients, our result showed that oseltamivir did not affect duration of viral shedding. Thus, we speculate that the seasonal in uenza may not affect the viral shedding of SARS-Cov-2. Arbidol is a broad-spectrum antiviral, which can inhibit numerous enveloped or non-enveloped RNA or DNA viruses. Previous study showed arbidol treatment potentially improved discharging rate and decrease the mortality rate of COVID- 19. 15 While, this study indicated that arbidol did not shorten duration of viral shedding. High quality clinical research is needed to verify the clinical e cacy of arbidol on COVID-19 patients.
In our study, virus shedding was delay in severe patients. Previous study suggests high viral load is associated with COVID-19 severity. 16 The increased viral load may contribute to the longer viral shedding in severe cases of COVID-19. The other reason may due to the impaired immune since the severe COVID patients presented with signi cantly decreased lymphocyte. In previous studies, continuously decrease of serum albumin is observed in critical cases, which was associated with poor clinical outcome in hospitalized patients. 17 Albumin can re ect the basic nutrition status. Over consumption of COVID-19 and inadequate nutrition intake may lead to hypoalbuminemia. This study demonstrated low level albumin associated with delayed viral shedding in consistent with previous study nding that albumin level was signi cantly correlated with SARS-CoV-2 viral load. 16 Therefore, nutrition support is very important in the healthcare of COVID-19. LDH was a glycolytic enzyme, which widely exist in many organs and tissues, especially in fast growing tumors. LDH was often regarded as an indicator of organ injury and tumor burden. 18 Previous studies found that high level LDH were common in severe cases and correlated with severe acute lung injury. 12,16 The present study revealed that high level LDH was associated with prolongation of viral shedding time, which may be due to the severe lung injury. Our study also found that elevated D-dimer level was associated with delayed viral shedding. Serum D-dimer can re ect brinolytic activities and was also an in ammatory biomarker. 19 Previous studies found that severe cases were commonly complicated with coagulopathy, markedly elevated D-dimer was related to poor prognosis of severe COVID- 19. 20 In ammatory cytokines would be highly associated with severity and immune status of COVID-19. The present study indicated that increased levels of ferritin and sIL-2R were related to prolongation of viral shedding time. Hyperferritin, a marker of macrophage activation, re ect the activation of reticuloendothelial system. 21 As an acute-phase reactant, ferritin level can be used as an indicator to re ect the degree of in ammation of inpatients with COVID-19. IL-2R is composed of three subunits: α, b, and g c . While b and g c subunits are constitutively active, the α subunit is induced and expressed only after mononuclear cell activation. 22,23 The sIL-2R is generated in "activation-dependent" manner, it is cleaved from IL-2Rα protein once T cells are activated. 22,23 Therefore, sIL-2R is used as a classical hallmark of T cell activation. 23 It can also be released from activated B cells, monocytes and dendritic cells. 23 Signi cant increase of sIL-2R are seen in autoimmune disease, organ transplant rejection and in many T cell and B cell neoplasms. 24 In autoimmune disease, sIL-2R level associated with disease activity and re ects immune system activation. 25 In clinic, sIL-2R is also an important diagnostic and disease marker in hemophagocytic syndromes/hemophagocytic (HPS/HLH). 22,23 High levels of sIL-2R are associated with poor prognosis, and falling sIL-2R corresponds to disease remission or treatment response in HPS/HLH. 23 A cytokine pro le resembling secondary hemophagocytic lymphohistiocytosis (sHLH) was seem in COVID-19. 26 sIL-2R can be used as a biomarker of immune system activation in patients with COVID-19 and predict the prognosis of COVID-19.
The duration of viral shedding was correlated with viral load, host immune response and treatment. Administration of systemic corticosteroid may impair immune response and delay viral clearance. When systemic corticosteroid was not used, the duration of viral shedding was correlated with age. The viral shedding time was longer in elder patients. However, when systemic corticosteroid was used, the duration of viral shedding was not correlated with age. Similarly, the duration of viral shedding was correlated with LDH when systemic corticosteroid was not used. The viral shedding time was longer in patients with elevated LDH. But when systemic corticosteroid was uses, the predict effect of LDH disappeared due to the impairment of systemic corticosteroid. When systemic corticosteroid was not used, the duration of viral shedding was not correlated with lymphocyte count. In cases administrated with systemic corticosteroid, the duration of viral shedding was negatively correlated with lymphocyte count. This may because high dose corticosteroid would be used in severe cases, in whom lymphopenia was very common, and viral shedding time was longer. sIL-2R was positively associated with duration of viral shedding no matter whether systemic corticosteroid was used or not. The result suggests that sIL-2R was an independent factor for predicting viral shedding time. It is a key factor for the virus inducing immune reaction in COVID-19 patients. The delayed viral shedding may associate with high virus load which stimulates the IL-2R related immune activation.
To our knowledge, this is the most detailed study of duration of viral shedding and its association with cytokines in COVID-19 patients. However, there are some limitations to this study. Firstly, there was potential false negative in SARS-CoV-2 nucleic acid test. Secondly, due to limited medical resources during the outbreak, only patients with severe symptoms could be admitted to hospital and most patients were severe cases. The durations of viral shedding from illness onset in this study tended to be very long and may differ from those in all population of COVID-19 patients. Duration of viral shedding in asymptomatic patients need to be further investigated. Finally, this report is a descriptive study, further research is urgent to investigate the mechanism of immune reaction involved in the virus clearance process.

Conclusions
The median duration of viral shedding was 24 days. Low dose of systemic corticosteroids did not prolong viral shedding compared with no use of systemic corticosteroids. The current antiviral treatments including lopinavir-ritonavir, arbidol, oseltamivir and intravenous immunoglobulin did not shorten viral shedding time. sIL-2R >710 U/mL, LDH>250 U/L and severe COVID-19 were related to prolongation of viral shedding. Optimal antiviral treatment strategy is urgently needed in the future.

Funding
This research did not receive any speci c grant from funding agencies in the public, commercial, or notfor-pro t sectors.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval and consent to participate
This study was approved by the Ethics Commission of Tongji Hospital (TJ-C20200122).

Consent for publication
Not applicable.
Competing interests HR is member of and receives funding from the German Lung Centre (DZL) and of the Universities Giessen and Marburg Lung Centre (UGMLC). The other authors declare that they have no competing interests.    The correlation between duration of viral shedding and age, lactate dehydrogenase, soluble interleukin-2 receptor and lymphocyte count strati ed by the usage of systemic corticosteroid. LDH lactate dehydrogenase; sIL-2R soluble interleukin-2 receptor.

Figure 3
The correlation between duration of viral shedding and age, lactate dehydrogenase, soluble interleukin-2 receptor and lymphocyte count strati ed by the usage of systemic corticosteroid. LDH lactate dehydrogenase; sIL-2R soluble interleukin-2 receptor.