Clinical Characteristics of COVID-19 Patients with Recurrent PCR Positivity After Hospital Discharge

Objective: This study aimed to investigate the clinical characteristics of COVID-19 patients with recurrent SARS-CoV-2 positivity after hospital discharge. Methods: This retrospective study included COVID-19 patients who were readmitted for recurrence of positive SARS-CoV-2 RNA. Univariate and multivariate analyses were performed to assess the risk factors associated to the duration of recurrent RNA positivity. Results: Among the 287 discharged COVID-19 patients, 33 (11.5%) patients with recurrent PCR positivity were included. Among these patients, 21 (63.7%) patients were female, their mean age was 48.7 (±19.7) years old. 22 (66.7%) patients were asymptomatic. The following clinical features were presented in other patients: cough, fatigue, sore throat, fever and expectoration. The chest CT �ndings revealed that 8 (24.2%) patients were characterized by deterioration compared to the previous results. The median duration of recurrent RNA positivity was 9.0 days (IQR, 6.0, 15.0). We found that increased serum SARS-CoV-2-specic IgG antibody titer, elevated serum creatinine level, and female gender were the risk factors for the prolonged duration of recurrent RNA positivity. Conclusion: SARS-CoV-2 turned positive in a minority of discharged patients with COVID-19. Most patients experienced mild clinical course. Increased IgG antibody titer, creatinine and female gender were correlated to the prolonged RNA clearance time.


Introduction
] According to the World Health Organization report, as of May 14, cases were reported in more than 200 countries, and the total number of con rmed cases and deaths has reached 424,8389 and 29,2046, respectively. [4]number of studies have described the clinical characteristics of hospitalized patients with COVID-19 infection. [5,6] ome studies have reported that few of these patients were still virus carriers after recovering from COVID-19 infection. [7,8]Approximately 14.5% of discharged COVID-19 patients were sent to the hospital again due to positive RT-PCR results on virus again, and the median time interval from negative to recurrent positive for these patients was 7.32±3.86days. [8]There was a negative correlation between the serum D-Dimer level before discharge and the duration of treatment.Lymphocyte concentrations before discharge were positively correlated with the time interval for virus reappearance. [8]wever, to date, the clinical and imaging features are not entirely clear.In addition, no study has revealed the viral shedding duration after recurrence and its in uencing factors.The present study reports 33 patients with polymerase chain reaction (PCR) assays of SARS-CoV-2 recurrent positive after discharge, and analyzes the clinical characteristics of this population.

Study Design
East District of Renmin Hospital, Wuhan University, is a designated hospital for the treatment of COVID-19 patients in Wuhan, where all patients are diagnosed with COVID-19 according to the WHO Interim Guidance.When patients were discharged for the rst time, these patients meet the following criteria proposed by the National Health Commission of China: (1) normal temperature that lasts longer than three days, (2) signi cant improvement in respiratory symptoms, (3) substantially improved acute exudative lesions on chest computed tomography (CT) images, and (4) the respiratory nucleic acid was negative for two consecutive times (with at least a 24-hour sampling time interval). [9]After discharge, all patients were quarantined for two weeks, and nasopharyngeal swab samples were collected every three or four days for the RT-PCR detection of COVID-19, in the same manner as that performed in a hospital.
From February 1, 2020 to April 1, 2020, a total 287 COVID-19 infected patients were discharged from 23 rd and 24 th ward.Among these patients, 33 patients with recurrent PCR positivity after hospital discharge were readmitted to the hospital.The positive recurrence of PCR assays was de ned as at least two RNA tests of SARS-CoV-2 was positive after discharge.The ethics approval was granted by the Ethics Board of the Institute of People's Hospital of Wuhan University (WDRY2020-K068).The informed consent was waived due to the retrospective nature of the study.

Data collection
The data of demographic information, clinical characteristics, laboratory results and treatment process were extracted from the electronic medical records of these patients.All data were reviewed by two doctors (K.W. and H.Z.), and a third researcher (P.T.) determined whether there was a difference in interpretation between the two reviewers.

Statistical method
The data were analyzed by the means of SPSS version 19.0.Continuous variables were presented as mean (SD) or median (IQR), and compared by Mann-Whitey U-test.Categorical variables were expressed in frequency (percentage), and compared by Chi-square or Fisher's exact test, when appropriate.A P-value of <0.05 was considered statistically signi cant.Simple linear regression analysis and multiple linear regression analysis were used to evaluate the correlation between these two variables.

Imaging features and treatment
Compared with the chest CT images before leaving the hospital, the CT ndings at readmission to the hospital revealed that eight (24.2%)patients were characterized by deterioration of original lesions, four patients presented with stable lesions, nine patients presented with improved lesions, and 12 patients presented with disappearance of original lesions.When these patients were readmitted, 12 of these patients received antiviral therapy, including abidol, hydroxychloroquine and ribavirin.All patients recovered, and were discharged.The median duration of recurrent RNA positivity was 9.0 days (IQR, 6.0, 15.0; Figure 1).No new COVID-19 was found among the close contacts of these patients during the study period.

Risk factors of the duration of recurrent RNA positivity
According to the duration of recurrent RNA positivity, these patients were divided into two groups: viral shedding after recurrence in the short-term group, in which the duration of viral shedding after recurrence was within 10 days (n=19, 57.58%); viral shedding after recurrence in the long-term group, in which the duration of viral shedding after recurrence was beyond 10 days (n=14, 42.42%).There was no distinguished difference in terms of age, gender, healthcare worker, signs and symptoms, or comorbidities between these two groups (Table 1).Compared with patients in the viral shedding after recurrence in the short-term group, the serum creatinine level, serum SARS-CoV-2-speci c IgG antibody titer and serum SARS-CoV-2-speci c IgM antibody titer of patients signi cantly increased in the viral shedding after recurrence in the long-term group (Table 2).The correlation analysis indicated that the duration of recurrent RNA positivity was positively correlated with both serum creatinine level (r=429, P=0.016) and serum SARS-CoV-2-speci c IgG antibody titer (r=0.507,P=0.005).A multiple linear regression analysis was carried out, with age, gender, serum IgG titer, serum IgM titer and serum creatinine level as independent variables.The results revealed that serum IgG antibody titer (r=0.016,P=0.016) and serum creatinine level (r=0.318,P=0.002) were risk factors for the prolonged duration of recurrent RNA positivity (Table 3).

Discussion
The present study describes the clinical characteristics of 33 COVID-19 patients with recurrent PCR positivity after discharge.Most of these patients were asymptomatic during the duration of the recurrent RNA positivity.These asymptomatic carriers brought more challenges to the identi cation and control of the COVID-19 epidemic worldwide.
In the present study, some patients presented with mild nonspeci c symptoms, including cough, fatigue, sore throat, fever and expectoration.In other reports, merely 32% of patients had mild cough during the duration of the recurrent RNA positivity. [8]Furthermore, many cases reported that the clinical symptoms of COVID-19 patients with recurrent PCR positivity did not aggravate. [7,10] owever, there has been no reports on the infectious capacity of COVID-19 patients with recurrent PCR positivity.Therefore, it remains di cult to identify and control such patients through clinical symptoms.In order to avoid such patients from becoming a potential source of infection again, all discharged patients should be quarantined and regularly tested for infectivity assessment by rechecking the PCR.
The median duration of recurrent RNA positivity was 9.0 days.Furthermore, it was found that serum SARS-CoV-2-speci c IgG antibody titer, serum creatinine level and female gender were risk factors for the prolonged duration of recurrent RNA positivity.The duration of positive RT-PCR persistence was associated with antibody response and clinical manifestations.Patients with symptoms and the development of anti-SARS-CoV-2 IgM antibodies had a shorter duration of positive RT-PCR results, and had no worsening clinical conditions, when compared to patients without the presence of anti-SARS-CoV-2 IgM antibodies. [11]Speci c IgG antibodies are important for protecting the host from infection by blocking viral entry into host cells after viral infection. [12]The median duration of IgG was detected at 14 days (IQR, 10-18) after symptom onset, with a positive rate of 77.9%. [13]There was no association between plateau IgG levels and the clinical characteristics of patients. [14]According to the report of Kaijin Xu, male gender was an independent risk factor for prolonged viral RNA shedding in COVID-19 patients. [15]In the present study, female patients presented with a longer duration of recurrent RNA positivity.
Histopathological studies have provided direct evidence of the invasion of SARS-CoV-2 into kidney tissues.
[16] Cheng Y et al. reported that elevated baseline serum creatinine was an independent risk factors for in-hospital death. [17]However, there has been no report on the relationship between serum creatinine level and the duration of recurrent RNA positivity.Hence, the in uence factor on the duration of recurrent RNA positivity needs further studies.
The cause for the recurrent positive viral RNA test after discharge in COVID-19 patients remains unclear.The de ciency of cellular immune function may lead to the incomplete shedding of the virus, resulting to the occurrence of "recurrence".Yao X et al. reported a case with negative detection of SARS-CoV-2 virus nucleic acid from nasopharyngeal swabs.However, the autopsy revealed residual SARS-CoV-2 in the lungs.These results highlight the remaining of SARS-CoV-2 in the lungs of discharged COVID-19 patients. [18]Comorbidity and therapeutic drugs, such as antiviral drugs and glucocorticoids, may also affect virus shedding. [19]After becoming infected with the virus, the human body responds with speci c antibodies production to avoid reinfection of the virus. [14]If SARS-CoV-2 mutates within a short period of time, similar to hepatitis C virus, the patient may be infected again. [20]The possibility of reinfection is unknown.
Two consecutive negative RNA tests is recognized as one of the discharge criteria. [9]According to the report of Wu et al., the positive rate of the SARS-CoV-2 RT-PCT test of nasopharyngeal swab, sputum, blood, feces and anal swabs is 38.13%, 48.68%, 3.03%, 9.83% and 10.00%, respectively. [21]Xiao et al. reported that 21.4% of patients presented with a positive virus RNA test again after two consecutive negative tests.It has been speculated that these patients may have experienced a false negative RT-PCR result or prolonged viral clearance, rather than a "turn positive" or "recurrence".However, the clinical characteristics and subgroup analysis was not performed for these patients. [22]Hence, the false negative RT-PCR result or prolonged viral clearance may be the cause for these patients to turn from negative to positive after discharge.It has been suggested that performing more viral RNA tests, combined with clinical improvements, could avoid false negative and reduce the recurrence.
In order to prevent the missed diagnosis of discharged patients, whose PCR assays turned positive, the following suggestions are given: (1) Both nasopharyngeal and oropharyngeal swabs test for SARS-CoV-2 RNA should be performed to reduce the false-negative rate.Furthermore, more tests, more specimens, and more methods should be considered.( 2) Patients in convalescence should also be regularly tested for infectivity assessment, and all discharged patients should be home quarantined for at least 14 days.(3) Some laboratory examinations, such as serum creatinine level, SARS-CoV-2-speci c antibody titer, Ddimer, lymphocyte count and platelet count, should be combined with the RT-PCR negative test as an additional measure, in order to ensure that the infected patient has completely recovered, and can be released from quarantine. [10,23]  present study had several limitations.First, merely 33 discharged patients were included.Collecting data from a larger cohort would provide a more comprehensive understanding of these discharged patients.Second, the present study lacks a control group of COVID-19 patients discharged at the same time.Furthermore, even if anti-virus medicine was used for these readmitted patients, it was di cult to assess its effect on virus clearance.The etiological mechanism of COVID-19 patients with recurrent PCR positivity after hospital discharge should also be investigated.

Table 1 .
Clinical characteristics of COVID-19 patients with recurrent PCR positivity

Table 2 .
Laboratory results, treatment, and clinical outcomes of patients with recurrent PCR positivity after readmission

Table 3 .
Multiple linear regression analysis of the correlation between different variables and duration of recurrent