The Effect of Corticosteroids on Severe Patients of COVID-19: A retrospective study

Qiang Li Huazhong University of Science and Technology Fang Cheng Huazhong University of Science and Technology Yuyong Su Huazhong University of Science and Technology Xuefeng Cai Huazhong University of Science and Technology Fang Zeng (  fancyzeng@126.com ) Huazhong University of Science and Technology Yu Zhang (  zhangwkp@163.com ) Huazhong University of Science and Technology https://orcid.org/0000-0003-3630-9002


Abstract Background
The effect of corticosteroids on COVID-19 remains controversial. This study aims to clarify the potential roles of corticosteroids in severe COVID-19 patients.

Methods
In the current retrospective single-center study, we collected data of 214 severe patients with con rmed COVID-19 in Wuhan Union Hospital from Feb 20th to Mar 1st, 2020. Epidemiological, clinical and treatment were analyzed between patients treated with corticosteroids or not.

Results
Corticosteroids used patients have higher levels of lactate dehydrogenase (LDH), interleukin-6 (IL-6), CD4 + /CD8 + cells, C-reactive protein (CRP) and procalcitonin (PCT). Virus clearance time and hospital length of stay in corticosteroids group were also signi cantly higher. The antiviral treatment and antibiotics treatment in patients given corticosteroids were both signi cantly higher. Antibiotics treatment duration was signi cantly longer in corticosteroids group. And the usage of multiple antibiotics in corticosteroid group was also signi cantly higher. And patients who treated by corticosteroids beyond 5 days showed a signi cantly longer antibiotics duration. Whereas there were no differences on virus clearance time and multiple antibiotics between the patients treated with corticosteroids beyond 5 days and less than 5 days. Multivariate analysis showed that patients with sputum production and higher IL-6 at admission, or treated with corticosteroid therapy were associated with prolonged virus clearance time and lianhua qingwen capsule may contributed to shorten virus clearance time.

Conclusions
The use of corticosteroids could prolong the virus clearance. The bene ts and harms should be carefully weighed in the COVID-19 patients who intend to use corticosteroids. The dosage should be low-tomoderate (≤ 0.5-1 mg/kg per day methylprednisolone or equivalent) and the duration should be short (≤ 5 days) to avoid secondary infections.

Background
Coronavirus disease 2019 (COVID-19) has spread around the world rapidly because of the highly contagious and human-to-human transmissions through direct contact, droplet or fomite [1]. Fever, cough, fatigue, myglgia and dyspnea are the most common symptoms of COVID-19 at onset of illness [2]. In ammatory cytokine storm is typical laboratory abnormalities observed during highly pathogenic coronavirus infections, such as severe acute respiratory syndrome coronavirus (SARS-CoV), the Middle East respiratory syndrome coronavirus (MERS-CoV) and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and is dynamically correlated with the severity of COVID-19 [3][4][5]. Corticosteroids are a good inhibitory of in ammatory factors and often used as an auxiliary treatment for viral pneumonia. In China, corticosteroids are also used for the treatment of COVID-19. However, the effect of corticosteroids on coronavirus pneumonia remains controversial. It was reported that corticosteroids has a bene cial effect on a majority of SARS patients [6]. Whereas, a retrospective study of MERS patients showed that patients treated with corticosteroids were more likely to require mechanical ventilation, vasopressors, and renal replacement therapy [7]. As for COVID-19, whether patients will bene t from corticosteroids still lack of clinical evidence. Therefore, we performed this retrospective single-center study to clarify the potential roles of corticosteroids in severe COVID-19 patients.

Patient selection
This retrospective single-center study included 214 severe COVID-19 patients who were admitted to Wuhan Union hospital from 20 February to 1 March. At the time of this study, all patients were cured and discharged. And patients were divided into two group based on whether they are given corticosteroids. The study was approved by the Medical Ethical Committees of Wuhan Union Hospital. The requirement for written informed consent was waived because there was no intervention for treatment and potential risk to patients.

Data collection
On admission, severe illness was de ned according to the criteria of Chinese management guideline for COVID-19 (version 7.0) [8]. The following information was extracted from electronic medical records by a standardised case report form: epidemiological, clinical characteristics, treatments, laboratory parameters and outcome data of con rmed severe cases of COVID-19. The date of disease onset was de ned as the day when the symptom was noticed.

Statistical analysis
Categorical variables were described as frequency rates and percentages. Proportions for categorical variables were compared using the χ 2 test, although the Fisher exact test was used when the data were limited. Means for continuous variables were compared using the Mann-Whitney test, and were described using median and interquartile range (IQR) values. When analyzing the in uencing factors of virus clearance time, pearson correlation was used for univariate analysis, and then multiple linear regression was used to select independent risk factors. All statistical analyses were performed using SPSS version 20.0 software. P<0.05 was considered statistically signi cant.

Results
Baseline characteristics of 214 COVID-19 patients 214 patients were included in this study, and 34 patients were treated with corticosteroid. The comparison of baseline characteristics from the two group were shown in Table 1. The median age of the two groups (non-corticosteroid and corticosteroid) was 60 (IQR 52-67) and 65 (IQR 55-73) years old, respectively. The median time from illness onset to admission was 12.0 days (IQR 8.0-15.0 There was no signi cant difference of age, sex, comorbidities and clinical symptoms between the two groups. However, the median time from illness onset to admission in corticosteroid groups was signi cantly shorter than that of non-corticosteroid groups (P=0.007), which may be the result of the faster progress of COVID-19.

Laboratory parameters
Major laboratory parameters were tracked from illness onset (Table 2). Creatinine, blood urea nitrogen, AST, ALT, total bilirubin, ESR, white blood cell, neutrophil, lymphocyte, hemoglobin, platelet, eosinophils count, FIB, D-Dimer, total lymphocytes, CD4, CD8, B cells and NK cells did not differ between patients who received corticosteroid treatment and patients who did not receive corticosteroid treatment.
Lactate dehydrogenase (LDH) was signi cantly higher in corticosteroid group than non-corticosteroid group (P=0.036), as well as higher levels of CD4 + /CD8 + cells (P=0.046). Virus clearance time (P=0.007) and hospital length of stay (P<0.001) were signi cantly prolonged in corticosteroid compared with noncorticosteroid group throughout the clinical course. And patients who required corticosteroid treatment were more likely to have higher levels of in ammatory indicators, including CRP (P=0.017), PCT (P=0.043), and IL-6 (P=0.029).

Treatment
According to the medication use of 214 patients, the patients generally received antiviral, antibiotic, traditional Chinese medicine (including Lianhua qingwen capsule and traditional Chinese medicine decoction), immune enhancer, intestinal microecological regulator and sedative hypnotic drugs ( Table 3). Compared with patients who did not receive corticosteroid treatment, the usage rate of chloroquine phosphate patients in corticosteroid group were signi cantly higher in patients given corticosteroid (P=0.002). Moreover, the usage of antibiotic was also signi cantly higher in corticosteroid group (P<0.001). However, patients in non-corticosteroid group were more likely to received traditional Chinese medicine therapy, including Lianhua qingwen capsule (P=0.035) and traditional Chinese medicine decoction (P=0.017).
To detect the effect of corticosteroid on secondary infections, we analyzed the antibiotic treatment of these patients. As shown in Table 4, the antibiotics duration in patients given corticosteroid was signi cantly longer than those of non-corticosteroid. And the usage of multiple antibiotics in corticosteroid group was also signi cantly higher. These indicated that patients treated by corticosteroid were more likely to get secondary infections.
As corticosteroid was recommended to use in COVID-19 patients in low dose and short term (3-5 days), we analyzed the patients given corticosteroid beyond 5 days and those less than 5 days. As Table 5 showed, patients who treated by corticosteroid beyond 5 days showed a signi cantly longer antibiotics duration.
However, there were no statistic difference in the virus clearance time and multiple antibiotics between the two groups. These results suggested that short-term use of corticosteroid may not increase the risk of secondary infections, but do prolong the virus clearance time.

Univariate analysis of virus clearance time
In univariable analysis, sex, longer time from illness onset to hospital admission, sputum production symptom, chloroquine phosphate therapy, non-Lianhua qingwen capsule therapy, immune enhancer, intestinal microecological regulator, corticosteroid, and sedative hypnotic therapy, higher levers of LDH, CRP, ESR, IL-6 and FIB, as well as lower lymphocyte count were associated with longer virus clearance time (Table 4).

Prognostic factors of virus clearance time
We included 15 signi cant variables in univariable analysis for multiple linear regression. The results showed that patients with sputum production and higher IL-6 at admission, or treated with corticosteroid therapy were associated with prolonged virus clearance time. Whereas, patients treated with lianhua qingwen capsule were more likely to shorten virus clearance time.

Discussion
There is no effective antiviral treatment for the novel virus, SARS-CoV-2, at present. And patients con rmed with COVID-19 were mainly treated by symptomatic therapy. In ammatory cytokine storm is a risk factor in severe COVID-19 patients.
Corticosteroids has been used for the treatment of severe pneumonia in clinical. However, it was controversial that whether corticosteroids should be used in COVID-19 patients. Some scholars claimed that clinical evidence does not support corticosteroid treatment for COVID-19 [9]. Whereas, Chinese management guideline for COVID-19 (version 7.0) [8]and a team of front-line physicians from China suggested that short term of corticosteroids at low dose could be used prudently in critical COVID-19 patients [10]. Therefore, it is important to provide evidence for corticosteroid used in COVID-19 patients.
In the current retrospective study, we analyzed the clinical features and medication of severe COVID-19 patients who treated with corticosteroids. There are no signi cant differences in demographic and epidemiologic characteristics except for the time from illness onset to hospital admission, which indicated that the COVID-19 patients who used corticosteroids were more likely to have a faster progression of COVID-19 at illness onset.
Similar with the premise of the use of corticosteroids [8], the severe COVID-19 patients who used corticosteroids were more likely to be an in ammatory condition at admission. Interestingly, the LDH level in patients given corticosteroids was signi cantly higher, which suggested that the injury liver function may associated with in ammatory storm. Consistent with the statement in Chinese management guideline for COVID-19 (version 7.0), we found that patients given corticosteroids have the signi cantly longer virus clearance time and hospital length of stay.
As to the treatment, the use of chloroquine phosphate and ribavirin were signi cantly higher in patients treated with corticosteroids, indicating that COVID-19 patients given corticosteroids need more antiviral treatment. In according with previous study [11], the antibiotic treatment in corticosteroids used patients was also signi cantly higher. Moreover, the antibiotics treatment duration and multiple use of antibiotics were both signi cantly higher in patients given corticosteroids. These results suggested that corticosteroids used patients have higher secondary infections. Further study showed that short term use of corticosteroid signi cantly reduced the antibiotics duration compared to long term use of corticosteroid (> 5 days). Therefore, these results demonstrated that the use of corticosteroid do prolong the virus clearance in severe COVID-19 patients and the use of corticosteroid should be less than 5 days to reduce secondary infections.
At present, many risk factors of COVID-19 have been identi ed, such as IL- 6[12]. In this study, we found that sputum production and higher IL-6 at admission, or treated with corticosteroid therapy were associated with prolonged virus clearance. Interestingly, lianhua qingwen capsule may contribute to shorten virus clearance time.
This study has some limitations. Firstly, this is a single center retrospective study, and a large-scale research was needed to provide high quality evidence. Secondly, fatal cases of COVID-19 were excluded and selection bias might occur. Therefore, additional studies are needed to investigate the effect of corticosteroid on patients with 2019-nCoV pneumonia.

Conclusion
In summary, we reported the clinical features and medication and analyzed the corticosteroid treatment in severe COVID-19 patients. The bene ts and harms should be carefully weighed in the COVID-19 patients who intend to use corticosteroids and the dosage should be low-to-moderate (≤ 0.5-1 mg/kg per day methylprednisolone or equivalent ) and the duration should be short (≤ 5 days) to avoid secondary infections.