key: cord-0999596-8920vemr
authors: Yu, Ting; Tian, Chunxia; Chu, Si; Zhou, Haifeng; Zhang, Zili; Luo, Shanshan; Hu, Desheng; Fan, Heng
title: COVID‐19 patients benefit from early antiviral treatment: a comparative, retrospective study
date: 2020-06-03
journal: J Med Virol
DOI: 10.1002/jmv.26129
sha: 48de69c7078bfc8b18f21bc6320137d6cb79e623
doc_id: 999596
cord_uid: 8920vemr

BACKGROUND: The outbreak of COVID‐19, caused by SARS‐CoV‐2, started in December 2019, Wuhan, China. We aimed to figure out the time‐point and duration of using antiviral drugs for receiving the maximal effects in patients with COVID‐19. METHODS: In this study, we enrolled 129 confirmed COVID‐19 mild to moderate patients who had been treated with antiviral drugs during their hospitalization in Wuhan Union Hospital China. The patients were divided into early antiviral treatment group and late antiviral treatment group. The demographic data, laboratory tests, the virus clearance time, chest computed tomography (CT) scans, etc. were extracted, calculated and compared between two groups. RESULTS: Our data showed that the median time from illness onset to initiation of antiviral treatment was 6 days in all patients. The group with early antiviral treatment demonstrated 7 days shorter in the virus clearance time when compared to the group with late antiviral treatment. After virus clearance, the group with early antiviral treatment showed milder illness than the group with late antiviral treatment. CONCLUSIONS: Early antiviral treatment could effectively shorten the virus clearance time, and prevent the rapid progression of COVID‐19. Therefore, the COVID‐19 patients should receive combined therapies with antiviral treatment at early stage. This article is protected by copyright. All rights reserved.

(34.88%), followed by fatigue (30.23%), expectoration (19.38%) and myalgia (19.38%). For patients with fever, their axillary temperature was mostly between 37.3-38°C (Table 1) . Nearly half of patients had comorbidities, the most comorbidities were hypertension (35.66%), diabetes (11.63%) and heart disease (10.85%) ( Table 1) . Except for the time from illness onset to hospital admission, there were no significant difference in age, sex, exposure history, comorbidities, signs and symptoms between the two groups ( Table 1) .

In terms of treatment, all the (100%) patients received antiviral treatment. The main antiviral drugs included arbidol (97.67%), interferon (24.03%), ribavirin (13.95%), and oseltamivir (8.53%); 105 (81.40%) patients received antibiotic treatment; 3 (2.33%) patients received antifungal treatment; 28 (21.71%) patients received glucocorticoids; 95 (73.64%) patients received oxygen therapy; 8 (6.20%) patients received immunotherapy (Table 2) . Significant difference was only observed in glucocorticoids (28.79% vs. 14.29%) between patients with early antiviral treatment and late (Table 2) .

This article is protected by copyright. All rights reserved.

Among all patients, the median time from illness onset to positive SARS-CoV-2 RNA detection was 10 days (IQR 3-16). As some inpatients were transferred from other hospitals, viral RNA detection has been accomplished before admission. The median time from illness onset to initiation of antiviral treatment was 6 days (IQR 3-12). In addition, the median duration of antiviral medication during illness was 

25 days (IQR 20-31) in those who had taken antivirals more than 19 days.

Likewise, the median time from illness onset to virus clearance was 26 days (IQR 19.5-32.5) in the late treated patients who had taken antivirals for 19 days or less, whereas the median time was 34.5 days (IQR 28.5-39) in those who had taken antivirals more than 19 days (Figure 3) . In order to identify the optimal duration of antiviral medication, patients were divided into 6 groups by duration: 0-7d, 8-14d, 15-21d, 22-28d, 29-35d, 36-42d. In the "0-7d" group, the virus clearance time was significantly shortened compared to other group. Besides, virus clearance time gradually increased in a time-dependent manner ( Figure 3) .

Therefore, our study suggested that the applied antiviral medication within 7 days was the optimal period to get a best virus clearance time. In addition, linear regression analysis showed that time of initiation of antiviral treatment and duration of medication correlated with virus clearance time (both P＜0.001) ( Figure 4) . Furthermore, we also compared age and sex factors, and found that the effect of age and sex on the time of virus clearance was not obvious (Table S1, Figure S1 ). Due to the significant difference in glucocorticoids treatment between the two groups, we further analyzed the correlation of glucocorticoids treatment and virus clearance time. Univariate analysis demonstrated that glucocorticoids treatment was slightly negatively correlated with virus clearance time ( Figure   S1 ). With the elimination of confounding factors such as glucocorticoids treatment, sex, and age, multivariate regression analysis further revealed that time This article is protected by copyright. All rights reserved.

of initiation of antiviral treatment and duration of antivirals medication were independent impact factors for virus clearance time. Especially, time of initiation of antiviral treatment was highly correlated with the time of virus clearance (Table S2) . Our data indicated early use of antiviral drugs and medication duration within 7 days could effectively shorten the virus clearance time.

All laboratory tests were traced from patients on admission. After virus clearance, lymphocytes absolute counts in all patients increased. The level of monocytes decreased in patients with early antiviral treatment, but it still increased in patients treated late (Figure 1 ; Table S4 ). In terms of blood coagulation function, the median level of fibrinogen (FIB) of the two groups on admission was above the normal range, whereas the value returned to within normal range after the virus clearance ( Figure 1 ; Table S3; Table S4) . After virus clearance, we observed a clear distinction of the levels of serum inflammation markers between early and late antiviral treatment groups. More specifically, the level of C-reactive protein (CRP), erythrocyte sedimentation rate (ESR) and interleukin-6 (IL-6) decreased significantly after virus clearance in patients with the early antiviral treatment.

However, IL-6 and ESR in patients treated late were clearly higher than baseline ( Figure 1 ; Table S3; Table S4 ). In addition, we also noted the abnormal This article is protected by copyright. All rights reserved.

frequency of white blood cells and neutrophils in patients with late treatment was higher than that of patients with early treatment (Table S3; Table S4 ). Moreover, for blood biochemistry markers, the heart, liver and kidney function markers fluctuated within the normal range after virus clearance (Figure 1; Table S4 ).

And the median value of lactic dehydrogenase (LDH) was lower than baseline in all patients (Figure 1 ). However, compared with patients treated early, the alanine aminotransferase (ALT), aspartate aminotransferase (AST) and blood urea nitrogen (BUN) in patients treated late were higher after virus clearance ( Figure   1 ; Table S4 ). These may be due to the potential injury in liver and kidney in patients treated late. On chest CT images, lung infections in both groups were effectively relieved after virus clearance. CT images showed bilateral distribution of patchy shadows and ground glass opacity were occurred in most patients on admission while there was significant lesions absorption after virus clearance ( Figure 2 ). It can be seen that after virus clearance, the patients did not have obvious multiple organs damage, and the coagulation disorder and inflammatory response were gradually controlled. Besides, patients treated early had a milder inflammatory response than those treated late after virus clearance. This article is protected by copyright. All rights reserved.

The level and duration of virus replication determine the capacity of transmission and provides instructive information on the isolation of patients. Since coronavirus RNA detection is more sensitive, most guidelines involve qualitative or quantitative viral RNA detection as an informative test in standards of discharge and terminating isolation. 11 In previous studies, it was found that for survivors, the median duration of virus shedding from the onset of disease was 20 days, but the virus continued to be detectable until death in non-survivors. 12 The long-term presence of the virus in the body was not conducive to the prognosis of the disease. In severe influenza virus infection, late antiviral treatment was associated with prolonged virus detection, and prolonged viral shedding was an independent risk factor for disease progression. 13 Effective antiviral treatment might improve outcome in COVID-19. But no shortening in the duration of viral shedding after lopinavir/ritonavir treatment was observed in previous studies. 14 In addition, optimized regimen of using antiviral drugs to curtail the virus clearance time had not been well described.

In this study, we depicted the intervention timing and total duration of antiviral medication, and evaluated its impact on virus clearance time and disease prognosis. Based on the median time from illness onset to initiation of antiviral treatment (6 days), we divided 129 mild to moderate COVID-19 patients into This article is protected by copyright. All rights reserved.

early antiviral treatment group and late antiviral treatment group in order to investigate the impact of timing of antiviral treatment on virus clearance time. The characteristics of the patients at baseline were generally consistent across the two groups. At the illness onset, patients mainly exhibited fever, cough, chest congestion and other symptoms. These symptoms gradually subsided as the virus was cleared away. Although we observed that there were slightly more obvious initial sympotoms in patients with early treatment of antivirals, there was no significant difference between the two groups. It might be because patients with obvious early symptoms were more likely to pay attention to their disease. Thus, they took antiviral drugs earlier. Meanwhile, we found that the frequency of use of glucocorticoids had differences between two groups. As we known the use of glucocorticoids is controversial, this might related with the heterogeneity of treatment regimens and also was a confounder. Univariate regression analysis demonstrated that glucocorticoids treatment was slightly negatively correlated with virus clearance time. Thus, this confounder was excluded due to the weak correlation. With the virus clearance, the blood biochemistry, coagulation function, and inflammation markers of most early treated patients gradually returned to normal range, and chest CT showed that lung infections gradually were controlled. Also, there was no clearly multiple organs damage. However, the improvement of abnormal markers in patients with late treatment was not as good as that in patients treated early, and the inflammatory storm still existed. It might Accepted Article be related to the fact that the virus in patients with late antiviral treatment were not well contained. Thus, the virus might replicate in the body for a longer time, attack the body further, and cause a relatively durable inflammatory response.

Taken together, our results showed that patients with early treatment was more likely to recover after virus clearance.

In this cohort, arbidol was the most widely used antiviral drug. Its usage rate is as high as 97.6%. Arbidol has been demonstrated to be a broad spectrum antiviral drug for prophylaxis and treatment of influenza. 15 It can not only interfere with virus-induced membrane fusion, but also degrade viral RNA (mRNA) to inhibit protein synthesis and thereby block the early replication of the virus. 16 This mode of action is mainly due to the disruption of key steps in virus-cell interactions. 17 In previous studies, its inhibitory activity has been extended to other human viruses, including respiratory syncytial virus (RSV), severe acute respiratory syndrome (SARS) and herpes simplex virus type I (HSV-1). 18, 19 Up to now, there were no licensed vaccines or antiviral medicines for SARS-CoV-2 infection. Broadspectrum antiviral drugs, including arbidol, could be a useful alternative therapy.

It might be beneficial for patients with COVID-19.

As showed in our study, the median virus clearance time in patients with early antiviral treatment was significantly lower by 7 days than that in patients who were treated late, and the time was further shortened after optimizing the duration This article is protected by copyright. All rights reserved.

of antivirals medication. Hence, we suggest that antiviral drugs should be administered to patients with COVID-19 as early as possible, because late antiviral drugs application could delay the clearance of virus and increase severe inflammatory response risk. Based on these findings, we further subdivided the duration of medication into 6 time intervals to ascertain the optimal duration of antiviral medication. We found that the virus clearance time was the shortest in patients taking antiviral medication within 7 days, which was consistent with the medication duration recommended in the guidelines. The virus clearance time P﹤0.05 was considered statistically significant. 

Clinical features of patients infected with 2019 novel coronavirus in Wuhan

Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study

Outcomes, median (IQR) Time from illness onset to be confirmed by SARS-Cov-2 RNA detection, days

This article is protected by copyright. All rights reserved.

We would like to thank the authors of the primary studies for their timely and helpful responses to our information requests. Moreover, we thank all members of Union hospital for helpful suggestions and discussions.

DH and HF designed this study. TY, CT, SC, HZ, and ZZ collected the epidemiological and clinical data. TY and CT processed statistical data. TY, and SL drafted the manuscript.

The authors do not have any professional and financial affiliations that may be perceived to have biased the presentation. This article is protected by copyright. All rights reserved.

Interferon