Incidence, clinical course and risk factor for recurrent PCR positivity in discharged COVID-19 patients in Guangzhou, China: A prospective cohort study.

The phenomenon of COVID-19 patients tested positive for SARS-CoV-2 after discharge (redetectable as positive, RP) emerged globally. The data of incidence rate and risk factors for RP event and the clinical features of RP patients may provide recommendations for virus containment and cases management for COVID-19. We prospectively collected and analyzed the epidemiological, clinical and virological data from 285 adult patients with COVID-19 and acquired their definite clinical outcome (getting PCR positive or not during post-discharge surveillance). By March 10, 27 (9.5%) discharged patients had tested positive for SARS-CoV-2 in their nasopharyngeal swab after a median duration of 7·0 days (IQR 5·0-8·0). Compared to first admission, RP patients generally had milder clinical symptoms, lower viral load, shorter length of stay and improved pulmonary conditions at readmission (p<0.05). Elder RP patients (≥ 60 years old) were more likely to be symptomatic compared to younger patients (7/8, 87.5% vs. 3/19, 18.8%, p = 0.001) at readmission. Age, sex, epidemiological history, clinical symptoms and underlying diseases were similar between RP and non-RP patients (p>0.05). A prolonged duration of viral shedding (>10 days) during the first hospitalization [adjusted odds ratio [aOR]: 5.82, 95% confidence interval [CI]: 2.50-13.57 for N gene; aOR: 9.64, 95% CI: 3.91-23.73 for ORF gene] and higher Ct value (ORF) in the third week of the first hospitalization (aOR: 0.69; 95% CI: 0.50-0.95) were associated with RP events. In conclusion, RP events occurred in nearly 10% of COVID-19 patients shortly after the negative tests, were not associated with worsening symptoms and unlikely reflect reinfection. Patients' lack of efficiency in virus clearance was a risk factor for RP result. It is noteworthy that elder RP patients (≥ 60 years old) were more susceptible to clinical symptoms at readmission.

Introduction

An outbreak caused by a novel human coronavirus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was first detected in Wuhan in December 2019, [1] and has since spread within China and other countries. The WHO has declared the COVID-19 a pandemic on Mar 14, 2020.[2] As of May 19, 2020, more than four million confirmed cases and 315,131 deaths had been reported globally.[3]

So far, over tens of thousands of patients with COVID-19 have been clinically cured and discharged, but multiple COVID-19 cases showed SARS-CoV-2 positive again (redetectable as positive, RP),[4-9] which raises an attention for the discharged patients. Since RT-PCT testing for SARS-CoV-2 is known to have certain range of false negative rate [10], the false negative RT-PCR testing before discharging may be a reason for RP events. On the other hand, patients themselves may have certain characteristics that make them more vulnerable to being RP. Previously, Yao and colleagues conducted postmortem pathologic study in a ready-for-discharge COVID-19 patient (three consecutive PCR tests of nasopharynx swab samples showed negative results) who succumbed to sudden cardiovascular accident, found SARS-CoV-2 remained in lung cells which may account for the RP result in discharged COVID-19 patients.[11] However, more evidence is needed for addressing the following questions include what is the incidence of RP events? What are the clinical characteristics of RP patients before and after discharge? What are the risk factors for patients to get RP? The answers to these questions may lead to recommendations for clinical guideline for virus containment and discharge assessment. Therefore, to facilitate efforts on above questions, we prospectively collected and analysed detailed clinical data from adult patients with laboratory-confirmed COVID-19 and a definite clinical outcome (getting PCR positive or not during post-discharge surveillance) at Guangzhou Eighth People’s Hospital, Guangzhou, China. In this study, we presented the clinical features of RP patients and explored the incidence and risk factors for RP events.

Methods

Study design and participants

This prospective cohort study included a cohort of adult inpatients (≥18 years old) from Guangzhou Eighth People’s Hospital (Guangzhou, Guangdong) with a diagnosis of COVID-19. The diagnosis of COVID-19 was based on the New Coronavirus Pneumonia Prevention and Control Program (7th edition) published by the National Health Commission of China.[12] Overall, 285 patients who were admitted between January 20 and February 18, 2020 were enrolled. March 10, all patients got a definite clinical outcome (becoming RP or haven’t become RP during post-discharge surveillance). The final date of follow-up was March 14, 2020, the day all observed cases were discharged. This study was approved by the institutional ethics board of Guangzhou Eighth People’s Hospital and the requirement for informed consent was waived by the ethics board.

Data collection and processing

Both first and second hospitalization data including demographic information, epidemiological history, clinical signs and symptoms, underlying comorbidities, dynamic laboratory parameters, treatment measures and outcome data, were obtained from the electronic medical record system of Guangzhou Eighth People’s Hospital by a trained team of experienced clinicians, epidemiologists and medical students using a standardized data collection form. According to the COVID-19 management routine of Guangzhou Eighth People’s Hospital, after initial discharging, COVID-19 patients have to undergo a period of isolation (within 15 days after discharge, it is unlikely to have a RP after this time) in the hospital or at home. During the surveillance, nasopharyngeal swab samples of patients were collected by staff of Guangzhou Center for Disease Control and Prevention (Guangzhou CDC) and submitted to Guangzhou Eighth People’s Hospital for Reverse transcription polymerase chain reaction (RT-PCR) test. Patients with positive nucleic acid tests after discharging were diagnosed as being RP and have to be readmitted and receive further medical observation. Two researchers (J.Z.Z. & R.Z.) independently reviewed and analysed the data and a third researcher (F.R.L.) adjudicated any difference in interpretation between the two primary reviewers.

Testing process and analysis

Patients’ nasopharyngeal swab specimens were collected for SARS-CoV-2 nucleic acid detection by RT-PCR at admission and once every two or three days during hospitalization and post-discharge surveillance. The detailed protocol of the RT-PCR is described elsewhere.[13] Threshold refers to the critical value of fluorescence signal in exponential growth period. Cycle threshold value (Ct value) refers to the number of cycles when the fluorescence signal reaches the threshold. A Ct-value less than 37 was defined as positive, a Ct-value ≥40 was defined as negative, and a medium load (Ct-value 37–40) was an indication for retesting.[14] Lower Ct value refers to higher viral load. Patients with positive nucleic acid tests in nasopharyngeal swab samples during post-discharge surveillance (within 15 days after discharge) were diagnosed as being RP. COVID-19 Human IgM IgG Assay Kit (ELISA based, produced by Abnova) was used to test the IgG and IgM level in COVID-19 patients.

Discharge criteria for COVID-19

Individuals meeting the following criteria could be discharged: absence of fever for at least three days, substantial improvement in both lungs in chest CT, clinical remission of respiratory symptoms, and two throat-swab samples negative for SARS-CoV-2 RNA obtained at least 24 h apart.[15]

Statistical analysis

Categorical variables are expressed as frequencies and percentages, and continuous variables are expressed as medians and interquartile ranges (IQRs). We compared the differences in epidemiological, clinical, and laboratory findings between patients who had a positive SARS-CoV-2 test after discharge and those who did not. Chi-square or Fisher’s exact tests were used to compare categorical variables between different patient groups, as appropriate, and the Mann-Whitney test was used to compare the continuous variables. When comparing the characteristics of RP patients between the two hospitalizations, the Wilcoxon signed-rank test and McNemar’s test were applied, as appropriate. To evaluate the dynamic changes in laboratory tests, including Ct values, the median value of the first three weeks were compared between the RP and NRP patients. To explore the risk factors associated with being RP, univariate and multivariate-adjusted logistic regression models were used. In the multivariate adjusted model, age, sex, hypertension, diabetes and liver disease were adjusted.

All statistical analyses were performed using Stata SE, version 15 (StataCorp) and graphs were generated and plotted using GraphPad Prism version 8.00 software (GraphPad Software Inc). A P value less than 0.05 (two-tailed) was considered statistically significant.

Results

Clinical data and laboratory findings during first hospitalization (RP vs. NRP patients)

From January 20 to March 4, 2020, 292 patients were admitted to Guangzhou Eighth People’s Hospital. After excluding six patients who were minors (≤18 years) and one death case, we enrolled 285 adult patients with COVID-19 in our final analysis. By March 14, 2020, all patients were discharged. Of these discharged patients, 27 (9.5%) recovered from COVID-19 tested positive for SARS-CoV-2 during post-discharge surveillance. The basic information is shown in Table 1. The median age of the study population was 48.0 years old (IQR 35.0–62.0, range, from 18.0–90.0 years), and 128 patients (44.9%) were men. The median length of stay (LOS) for both RP patients and NRP patients was 18 days. Generally, Demographics, epidemiological history and clinical symptoms did not significantly differ between the two groups.

Table 1

Baseline characteristics of 27 RP patients and 258 non-RP patients.

Characteristics	All patients (n = 285)	RP Patients (n = 27)	NRP patients (n = 258)	p value
Basic information
Age, years	48.0 (35.0–62.0)	44.0 (32.0–62.0)	49.0 (35.0–62.0)	0.450
Men	128 (44.9)	12 (44.4)	116 (44.9)	0.959
Exposed to Wuhan or surrounding cities	179 (62.8)	19 (70.4)	160 (62.0)	0.393
Smoking history	31 (11.0)	1 (3.7)	30 (11.8)	0.332
Severity				0.703
Mild	22 (7.7)	3 (11.1)	19 (7.4)
Moderate	257 (90.2)	24 (88.9)	233 (90.3)
Severe	6 (2.1)	0 (0)	6 (2.3)
Comorbidities
Any Comorbidity	88 (31.4)	8 (29.6)	80 (31.0)	0.832
Hypertension	51 (17.9)	6 (22.2)	45 (17.4)	0.597
Diabetes	24 (8.4)	1 (3.7)	23 (8.9)	0.712
Liver disease	23 (8.1)	2 (7.4)	21 (8.1)	0.343
COPD	19 (6.7)	2 (7.4)	17 (6.6)	0.698
Cardiovascular disease	18 (6.3)	0 (0)	18 (6.9)	0.235
Kidney disease	8 (2.9)	0 (0)	8 (3.1)	..
Cancer	3 (1.1)	0 (0)	3 (1.2)	..
Clinical characteristic
Symptoms
Asymptomatic	34 (11.9)	5 (18.5)	29 (11.2)	0.343
Fever	193 (67.7)	18 (66.7)	175 (67.8)	0.902
Dry cough	159 (55.9)	14 (51.6)	145 (56.4)	0.649
Expectoration	59 (20.7)	6 (22.2)	53 (20.5)	0.838
Chills	58 (20.4)	2 (7.4)	56 (21.7)	0.079
Fatigue	37 (12.9)	4 (14.8)	33 (12.8)	0.764
Myalgia	34 (11.9)	1 (3.7)	33 (12.8)	0.222
Chest CT
Bilateral involvement of chest CT scan	261 (95.6)	27 (100.0)	234 (95.1)	0.241
Small Pulmonary Nodules	13 (4.7)	3 (11.1)	10 (4.1)	0.125

Data are median (IQR) or n (%). p values comparing RP and NRP patients are from χ2, Fisher’s exact test, or Mann-Whitney U test. COPD = Chronic obstructive pulmonary disease. CT = computerized tomography scan. RP = re-detectable as positive. NRP = non-re-detectable as positive.

The medians of RP and NRP patients’ parameters during whole hospitalization and each week after admission are shown in Table 2. Compared with NRP patients, RP patients had a significantly lower median LDH level during hospitalization (159.5 vs. 186.0, p = 0.034). RP patients showed lower LDH (159.0 vs. 192.0, p = 0.034) than NRP patients at first week after admission, whereas eosinophil count was higher (0.05 vs. 0.02, p = 0.018). Concerning Ct values of N and ORF gene, there were no significant differences between the two groups within two weeks after admission. Eventually, RP patients’ median Ct values of ORF gene were significantly lower than NRP group (35.5 vs. 39.0, p = 0.031) at third week. Similar results also observed in Ct values of N gene (Fig 1). The details of other markers between the two groups are described in S1 Table.

Table 2

Laboratory indicators of RP and NRP patients for the first three weeks of hospitalization.

Parameter, week	Normal Range	All patients	RP Patients	NRP patients	p value
Positive Ct value (N)	≥40	37.0 (34.5–38.0)	35.5 (34.0–37.0)	37.0 (35.0–38.0)	0.044
1		36.0 (33.0–38.0)	35.0 (33.0–36.0)	36.0 (33.0–38.0)	0.197
2		37.0 (34.0–38.8)	35.8 (31.8–38.3)	37.0 (34.5–39.0)	0.275
3		38.0 (33.0–39.0)	36.0 (33.0–38.0)	38.0 (37.0–39.0)	0.045
Positive Ct value (ORF1ab)	≥40	38.0 (35.0–39.0)	37.0 (34.0–38.0)	38.0 (35.5–39.0)	0.061
1		37.0 (34.0–38.0)	35.0 (33.0–37.5)	37.0 (34.0–39.0)	0.088
2		38.0 (34.0–39.0)	36.3 (33.5–39.0)	38.0 (35.0–39.0)	0.224
3		38.0 (36.0–40.0)	35.5 (33.0–39.0)	39.0 (38.0–41.0)	0.031
White blood cell count, ×109 /L	3.5–9.5	5.3 (4.4–6.3)	5.2 (4.2–5.6)	5.3 (4.4–6.4)	0.301
1		4.9 (3.9–6.3)	4.8 (4.2–5.5)	5.1 (3.9–6.3)	0.297
2		5.4 (4.5–6.8)	5.1 (4.6–5.7)	5.6 (4.5–6.8)	0.246
3		5.5 (4.6–6.5)	5.4 (4.4–6.3)	5.5 (4.6–6.5)	0.489
Eosinophil, ×109 /L	0.02–0.52	0.08 (0.04–0.12)	0.08 (0.04–0.12)	0.08 (0.04–0.12)	0.767
1		0.03 (0.0–0.07)	0.05 (0.02–0.1)	0.02 (0.00–0.07)	0.018
2		0.09 (0.05–0.14)	0.08 (0.04–0.10)	0.1 (0.05–0.1)	0.208
3		0.12 (0.08–0.20)	0.11 (0.08–0.17)	0.12 (0.08–0.20)	0.729
Lactate dehydrogenase, U/L	125–243	185.0 (154.5–225.0)	159.5 (139.0–197.0)	186.0 (155.5–229.0)	0.034
1		188.0 (151.0–238.0)	159.0 (140.0–196.0)	192.0 (152.0–243.0)	0.022
2		183.0 (149.0–238.0)	159.5 (147.0–213.0)	184.0 (151.0–238.0)	0.493
3		179.0 (150.0–213.5)	166.0 (129.0–192.0)	180.0 (151.0–218.0)	0.069
C-reactive protein, mg/L	<10	5.0 (5.0–12.7)	5.0 (5.0–5.0)	5.0 (5.0–13.4)	0.094
1		5.0 (5.0–23.5)	5.0 (5.0–10.2)	5.0 (5.0–26.2)	0.034
2		5.0 (5.0–11.5)	5.0 (5.0–5.0)	5.0 (5.0–12.3)	0.301
3		5.0 (5.0–5.0)	5.0 (5.0–5.0)	5.0 (5.0–5.0)	0.537

Data are median (IQR) value of first three weeks after admission., the number of available test result of RP patients for the first, second and third weeks were 27, 27, 19, in contrast, 258, 249, 184 in NRP patients. P values comparing RP and NRP patients are from Mann-Whitney U test. RP = redetectable as positive. NRP = non-redetectable as positive.

Fig 1

Comparison of viral dynamics between RP and NRP patients.

Figure shows temporal changes in median Ct value of N gene (A) and ORF gene (B) in different time period. Since we have only collected and analyzed the data during patients’ hospitalization, Ct value at admission and readmission were unavailable. The dotted line in red shows the lower normal limit of Ct values. Ct = cycle threshold. RP = redetectable as positive. NRP = non-redetectable as positive.

As for clinical course, RP and NRP patients’ length of hospital stay (LOS) were both 18 days. For RP patients, the median duration of viral shedding (N gene) after admission was 14.0 days (IQR 8.0–20.0 days) (Fig 2), which was significantly longer than those in NRP patients (7.0 days [IQR 7.0–10.0]) (p<0.001). 62.9% RP patients and 23.6% NRP patients presented positive RNA detection tests (N gene) for more than 10 days since hospital admission. The results were similar in ORF gene (Table 3).

Fig 2

Clinical course, complications and duration of viral shedding from illness onset in patients hospitalized with COVID-19.

Figure shows median duration of hospitalization and positive nucleic acid Ct value and onset of several complications. RP = redetectable as positive. NRP = non-redetectable as positive.

Table 3

Clinical course and RNA test result of 27 RP patients and 258 non-RP patients.

	RP (n = 27)	NRP (n = 258)	p value
First-hospitalization
Time from illness onset to admission, days	3.0 (1.0–5.0)	3.0 (1.0–7.0)	0.923
Length of stay, days	18.0 (13.0–24.0)	18.0 (13.0–25.0)	0.822
Duration of viral shedding (N) after admission, days	14.0 (8.0–20.0)	7.0 (7.0–10.0)	<0.001
Distribution, no (%)
≤10 days	10 (37.0)	197 (76.4)	<0.001*
>10 days	17 (62.9)	61 (23.6)
Duration of viral shedding (ORF) after admission, days	16.0 (8.0–21.0)	7.0 (7.0–10.0)	<0.001
Distribution, no (%)
≤10 days	8 (29.6)	203 (78.7)	<0.001*
>10 days	19 (70.4)	55 (21.3)
Rehospitalization
Quarantine site before rehospitalization
Hospital	9 (33.3)	-	-
Home	18 (66.7)	-	-
Time from discharge to retest positive, days	7.0 (5.0–8.0)	-	-
Length of stay, days	7.0 (5.0–11.0)	-	-
Duration of viral shedding after being RP (N gene), days	3.0 (3.0–10.0)	-	-
Duration of viral shedding after being RP (ORF gene), days	7.0 (6.0–10.0)	-	-.
Lung inflammation compared with first hospitalization
Normal	1 (3.7)	-	-
Improved	21 (77.8)	-	-
Stable	5 (18.5)	-	-
Aggravated	0	-	-

Data are median (IQR) or n (%). P values comparing RP and NRP patients are from χ2, Fisher’s exact test, or Mann-Whitney U test. ICU = intensive care unit. RP = redetectable as positive. NRP = non-redetectable as positive.

*χ2 test comparing all subcategories.

Clinical data and laboratory findings of RP patients (first vs. second hospitalization)

After discharged, RP patients readmitted to hospital after a median of 7.0 days (IQR 5.0–8.0 days) of surveillance. Compared with the first hospitalization, more asymptomatic persons (17 [62.9%] vs. 5 [18.5%], p = 0.013), shorter length of hospitalization (7.0 days [5.0–11.0] vs. 18.0 [13.0–24.0], p<0.001) and higher Ct value of N gene (37.5 [36.0–38.5] vs. 35.0 [33.0–37.0], p = 0.042) were presented in RP patients’ rehospitalization (S2 Table). Elder RP patients (≥ 60 years old) were more likely to be symptomatic compared to younger RP patients (7/8, 87.5% vs. 3/19, 18.8%, p = 0.001) at readmission (Fig 3). Of those who underwent detection of the specific binding antibody to SARS-COV-2 in the plasma, twenty (100.0%) and sixteen (80.0%) showed positivity of IgG and IgM. During rehospitalization, duration of viral shedding from first positive tests (N gene) was 3.0 days (IQR 3.0–10.0 days) and 7.0 days (IQR 6.0–10.0 days) for ORF gene. 21 (77.8%) patients showed improved lung condition at readmission. (Table 3). The monitoring results of RNA for 27 RP patients during rehospitalization were shown in Fig 3 and S1 Fig.

Fig 3

Comparison of the two hospitalization courses of 27 RP patients and result of series SARS-CoV-2 RNA test in nasopharyngeal swab specimens during the second hospitalization.

Comparisons of clinical condition between first and second hospitalization are shown for each RP patient (upper panels). Timeline of series SARS-CoV-2 RNA test (lower panels) during rehospitalization are shown. *Ct value <35 refers to whether the lowest Ct value during hospitalization is lower than 35. **Discharge indicates two throat-swab samples negative for SARS-CoV-2 RNA obtained at least 24 h apart. This figure showed that elder RP patients (≥ 60 years old) were more likely to be symptomatic compared to younger RP patients (7/8, 87.5% vs. 3/19, 18.8%, p = 0.001) at readmission. RP = redetectable as positive. NRP = non-redetectable as positive.

Chest CT images during two hospitalizations of three RP patients and CT images taken at two different time during first hospitalization of one NRP patient were shown in Fig 4. The typical findings of chest CT images of RP patients at readmission were the improved bilateral pulmonary inflammation.

Fig 4

Chest CT images.

(A) Transverse chest CT images from a 62-year-old woman who got RP 14 days after discharge, showing multiple inflammation in bilateral lungs at readmission (lower panel), which has partly absorbed compared to the condition at first discharged (upper panel). (B) Transverse chest CT images from a 30-year-old man who got RP 8 days after discharge, showing improved multiple inflammation and decreased shadows of fibrotic streaks at readmission (lower panel) compared to the condition at first discharged (upper panel). (C) Transverse chest CT images from a 32-year-old woman who got RP 6 days after discharge, showing inflammation on bilateral lower lobe at readmission (lower panel), which has partly absorbed compared to the condition at first discharged (upper panel). (D) Transverse chest CT images from a 68-year-old male NRP patient, showing multiple inflammation in bilateral lungs 14 days after admission (lower panel), with no obvious change compared with the condition at admission (upper panel).

Risk factors for RP events

In the univariate logistic regression model, decreased median Ct values of ORF gene at week three after admission (OR 0.76, 95% CI 0.60–0.97) and duration of viral shedding from admission greater than 10 days (OR 5.49, 95% CI 2.39–12.62 and OR 8.77, 95% CI 3.64–21.09, for N gene and ORF gene, respectively) were associated with increased risk of being RP. When adjusting for age, sex, hypertension, CVD and liver disease, our regression model showed similar results (Table 4).

Table 4

Univariable and multivariable analysis of risk factors associated with RP events.

	Univariate OR	p value	Adjusted OR*	p value
Basic information
Age, years	0.99 (0.97–1.02)	0.458	..	..
Male (vs. female)	0.98 (0.44–2.17)	0.959	..	..
Clinical severity
Mild	1 (reference)		1 (reference)
Moderate	0.65 (0.18–2.37)	0.522	0.72 (0.19–2.69)	0.632
Comorbidity
Any comorbidity	0.91 (0.38–2.17)	0.832	0.89 (0.12–5.02)	0.711
Diabetes	0.39 (0.05–3.03)	0.370	..	..
Hypertension	1.35 (0.52–3.54)	0.539	..	..
Liver diseases	0.90 (0.20–4.08)	0.894	..	..
Laboratory findings
Median Ct value (N gene)	0.96 (0.87–1.07)	0.496	0.96 (0.86–1.07)	0.487
Week1	0.96 (0.87–1.05)	0.367	0.96 (0.87–1.06)	0.385
Week2	0.95 (0.84–1.07)	0.380	0.94 (0.82–1.08)	0.368
Week3	0.85 (0.71–1.03)	0.103	0.88 (0.70–1.10)	0.256
Median Ct value (ORF gene)	0.91 (0.82–1.01)	0.071	0.89 (0.80–0.99)	0.042
Week1	0.93 (0.84–1.04)	0.193	0.93 (0.83–1.03)	0.167
Week2	0.90 (0.79–1.04)	0.144	0.87 (0.75–1.02)	0.078
Week3	0.76 (0.60–0.97)	0.030	0.69 (0.50–0.95)	0.022
Eosinophil,×109 /L	0.84 (0.63–1.11)	0.213	1.59 (0.24–10.75)	0.633
Week1	9.42 (0.07–54.09)	0.374	9.30 (0.06–49.06)	0.390
Lactate dehydrogenase, U/L	0.99 (0.99–1.00)	0.133	0.99 (0.99–1.00)	0.165
Week1	0.99 (0.99–1.00)	0.065	0.99 (0.99–1.00)	0.072
Week3	0.99 (0.98–1.00)	0.122	0.99 (0.98–1.00)	0.193
C-reactive protein, mg/L	0.98 (0.93–1.02)	0.303	0.98 (0.93–1.03)	0.392
Week1	0.97 (0.94–1.01)	0.102	0.97 (0.94–1.01)	0.108
Clinical course
Duration of viral shedding from admission, days
N gene
≤10	1 (reference)		1 (reference)
>10	5.49 (2.39–12.62)	<0.001	5.82 (2.50–13.57)	<0.001
ORF gene
≤10	1 (reference)		1 (reference)
>10	8.77 (3.64–21.09)	<0.001	9.64 (3.91–23.73)	<0.001

OR = odds ratio.

*Adjusted for age, sex, hypertension, diabetes and liver disease. OR value in continuous variables is the risk related to per 1 unit increase.

Discussion

This study reported the incidence rate of and risk factors for RP events in adult patients with COVID-19 in Guangzhou. Additionally, the epidemiological, clinical and virological features of RP and NRP patients were compared. The viral load in both RP patients and NRP patients’ nasopharyngeal swab samples were monitored with sustained viral detection by RT-PCR. As of March 14, 2020, the end of the follow-up, 27 (9.5%) patients had become RP during their post-discharge surveillance after a median duration of 7.0 days. We revealed that a longer duration of viral shedding and higher viral load in the later stage of hospitalization were risk factors for RP events in patients with COVID-19. Furthermore, our study found that elder RP patients (≥ 60 years old) were more susceptible to clinical symptom at readmission.

27 of 285 (9.5%) individuals had tested positive for SARS-CoV-2 in nasopharyngeal swab after discharged. Previously, Zhongnan Hospital has reported two (3·23%) medical staff had tested positive after discharged.[7] In a study on 209 discharged patients, conducted by Tang et al [16], 9 (4.3%) re-tested positive in throat swabs only, 13 patients (6.2%) re-tested positive in anal swabs only, and 22 (10.5%) re-tested positive in either. The study by Zhongnan Hospital merely included medical staff and the Third People’s Hospital of Shenzhen had only enrolled 49.4% (209/423) COVID-19 patients diagnosed in Shenzhen. The incidence rate of our study was more representative since we have included the majority (75.6%, 285/377) of COVID-19 patients in the international city: Guangzhou, with no restrictions other than age. Since the outbreak of COVID-19 occurred earlier in China than in other countries and the outbreak is still increasing or plateauing in many other countries,[3] the incidence rate reported in this study may provide a reference for the global disease management, especially in the populous developing countries.

Retested positivity unlikely reflect reinfection, since most RP patients in current study showed no obvious clinical symptoms or disease progression indicated by laboratory and CT findings and did not contact with other infectious patients. When we focus on chest CT rather than viral load, most of the RP patients seem to be normal convalescent patients with favorable inflammatory absorption. For these patients, unless there is a clinical symptom worsening, excess clinical intervention may not be necessary. It has been shown that the plasma levels of the IgM and IgG antibodies specific to SARS-CoV-2 and the series of immune cells produced during recovery play important roles in virus neutralization and prevention against further infection [17,18]. In a study of rhesus macaques infected with SARS-CoV-2, the animals did not develop reinfection following recovery and re-challenge [19]. Thus, we concluded that RP events are more likely caused by false negative RT-PCR tests before discharging.

Concerning clinical symptom were more commonly seen in elder RP patients. Previous studies found that age-dependent defects in T-cell and B-cell function and the excess production of type 2 cytokines could lead to more prolonged proinflammatory responses and constant clinical symptom.[20]. On this basis, we speculate that the remained clinical symptoms may be related to the poor recovery ability and prolonged body responses of the elderly just recovered from COVID-19. Thus, enhanced follow up medical examination and treatment should be carried out in time for discharged elderly patients.

In multiple respiratory viruses, viral load could be a predictor of disease stage and progression.[20-23] In this study, Ct values of respiratory tract samples from both RP and NRP patients with COVID-19 peaked in the first week after admission which was similar to the results reported in Beijing,[24] but distinct from those observed in patients with SARS, which normally peaked at approximately ten days after onset.[25] Furthermore, we found that a higher level of viral load during the later stage of hospitalization and a longer duration of viral shedding were risk factors for RP events. Sustained viral shedding has been found to be associated with antiviral resistance in patients infected with the influenza H7N9 virus.[26,27] On this basis, we speculate that the higher viral load and longer duration of positive test results in RP patients may be the consequence of their deficiency in control of viral replication and antiviral resistance. This makes them more susceptible to positive result after discharged. Since many RP patients shifted from the negative to positive again during their second hospitalization, as they performed at their first discharge. This phenomenon may be caused by the lack of efficiency in virus clearance in RP patients. It was observed these patients had returned to true negative soon but their potential infectiousness remained unknown; thus, prolonged hospitalization for patients with lower efficiency in virus clearance might be a safer measure.

This study has several limitations. First, it’s a single-centre study. However, by including adult patients with diverse characteristics, we believe our study population is representative. Second, the estimated duration of viral shedding is limited by the frequency of nasopharyngeal swab samples collection. Third, due to the limited duration of observation, the evaluation for further clinical progression in RP patients could not be carried out, which need a long-term follow-up. Fourth, despite no individual was infected by RP patients in this study, as post-discharge patients were under isolation, we were unable to effectively assess the infectiousness of RP patients.

In conclusion, a nearly 10% incidence of RP events observed in this study suggests numerous COVID-19 patients in the world may get RP. It is expected these patients would return to true negative soon, and unlikely they would get reinfection and remain infectious. We found that a prolonged duration of viral shedding during first hospitalization was a risk factor for RP events which may provide implication on further virological research. However, the clinical symptoms shown in elder RP patients at readmission should not be ignored, suggesting more post-discharge clinical attention on elder COVID-19 patients.

Ethics approval and consent to participate

This study was approved by the institutional ethics board of Guangzhou Eighth People’s Hospital and the requirement for informed consent was waived by the ethics board.

Laboratory findings for RP and non-RP patients.

Data of the indicators that have been measured many times are median (IQR) value of first three weeks after admission. Others are data at admission. In the first, second and third weeks, the number of available test results of RP patients were 27, 27, 19, in contrast, 258, 249, 184 in NRP patients. P values comparing RP and NRP patients are from Mann-Whitney U test. GFR = glomerular filtration rate. ALT = Alanine aminotransferase. AST = Aspartate aminotransferase. RP = redetectable as positive. NRP = non-redetectable as positive.

(DOCX)

Click here for additional data file.

Clinical characteristics, treatment and laboratory findings of 27 RP patients at first admission and readmission.

Data are median (IQR) or n (%). *Results of CT scan for first admission were shown in Table 1. **Only 20 RP patients have tested for Antibody. Results of Ct values and CD cell were median value during hospitalization. RP = redetectable as positive. NRP = non-redetectable as positive.

(DOCX)

Click here for additional data file.

Entire distribution of times of RP patients.

(TIF)

Click here for additional data file.

10 Jul 2020

Dear Professor Wu,

Thank you very much for submitting your manuscript "Incidence, clinical course and risk factor for recurrent PCR positivity in discharged COVID-19 patients in Guangzhou, China: a prospective cohort study" for consideration at PLOS Neglected Tropical Diseases. As with all papers reviewed by the journal, your manuscript was reviewed by members of the editorial board and by several independent reviewers. The reviewers appreciated the attention to an important topic. Based on the reviews, we are likely to accept this manuscript for publication, providing that you modify the manuscript according to the review recommendations.

Please prepare and submit your revised manuscript within 30 days. If you anticipate any delay, please let us know the expected resubmission date by replying to this email.

When you are ready to resubmit, please upload the following:

[1] A letter containing a detailed list of your responses to all review comments, and a description of the changes you have made in the manuscript.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out

[2] Two versions of the revised manuscript: one with either highlights or tracked changes denoting where the text has been changed; the other a clean version (uploaded as the manuscript file).

Important additional instructions are given below your reviewer comments.

Thank you again for your submission to our journal. We hope that our editorial process has been constructive so far, and we welcome your feedback at any time. Please don't hesitate to contact us if you have any questions or comments.

Sincerely,

Andrés Felipe Henao-Martínez, M.D.

Deputy Editor

PLOS Neglected Tropical Diseases

Andrés Henao-Martínez

Deputy Editor

PLOS Neglected Tropical Diseases

***********************

Reviewer's Responses to Questions

Key Review Criteria Required for Acceptance?

As you describe the new analyses required for acceptance, please consider the following:

Methods

-Are the objectives of the study clearly articulated with a clear testable hypothesis stated?

-Is the study design appropriate to address the stated objectives?

-Is the population clearly described and appropriate for the hypothesis being tested?

-Is the sample size sufficient to ensure adequate power to address the hypothesis being tested?

-Were correct statistical analysis used to support conclusions?

-Are there concerns about ethical or regulatory requirements being met?

Reviewer #1: Objectives are well documented as the population subgroup study was looking at is positive RP after discharge.

Study design is aptly selected as patient population is prospectively followed and a specific cohort was selected.

Yes, population was clearly described as only who were positive RP after discharge.

Sample size is small, 27, this is not an adequate power for the hypothesis tested. Having said that, the incidence of RP positive have been around 7 to 10% in previous studies too

Yes correct statistical analysis was used for both continuous and categorical variables.

Major concern is this being a prospective study, informed consent is required as patient population was tested post discharge and the main criteria for inclusion in the study was RP positive. Manuscript mentions the requirement was waived by ethics board. PLEASE EXPLAIN WHY

Reviewer #2: (No Response)

Reviewer #3: The objectives of the study are clear: analyse epidemiological, virological and clinical data from patients admitted in the hospital and by comparing them with the definitive outcome of retesting positive or not after discharge, identify potential risk factors for restesting positive. It is a perspective study, which is good for its objectives, because it gives to this analysis more power to discover relevant risk factors.

The characteristic of the populations might be described a little more in depth. Still the authors discarded very few cases and for very clear reasons, so I would deem the population apropriate.

The sample size is large enough to study with enough confidence the problem under consideration.

The statistical analysis are appropriate to test the hypotheses considered.

No ethical concerns (it has all been approved by the responsible ethical committee).

--------------------

Results

-Does the analysis presented match the analysis plan?

-Are the results clearly and completely presented?

-Are the figures (Tables, Images) of sufficient quality for clarity?

Reviewer #1: Yes analysis was appropriate, and results were clearly mentioned.

Tables did have all the necessary variables.

I can understand the CT images included for three RP patients and one NRP patient. But the relevance and clinical significance need to be explained. Review of literature shows very little correlation between imaging and patient clinical improvement. Even though the findings of eosinophil count LDH appear to have more clinical relevance in follow up too.

Reviewer #2: (No Response)

Reviewer #3: The analysis match the plan and the results are presented very clearly.

The tables and Images are very clear and interesting. In particular figure 3 is very good, gives a very clear visual impact on the point the authors are making.

In figure 2 maybe showing the entire distribution of the times instead of the median might be more informative. This is just a suggestion, it might also be more confusing, still it could be provided as a supplementary figure.

--------------------

Conclusions

-Are the conclusions supported by the data presented?

-Are the limitations of analysis clearly described?

-Do the authors discuss how these data can be helpful to advance our understanding of the topic under study?

-Is public health relevance addressed?

Reviewer #1: Conclusions are partially supported by the data. I agree with conclusions about patient population >60 who are RP positive, viral load predicting disease progression and risk factors identified for being RP positive.

My concern is with the conclusion that these RP patients unlikely are infectious. Even though a considerable conclusion can be made that they are unlikely reinfected based on provided biochemical data and symptoms, we have no evidence that they are not infectious to other population or health care personnel. Please explain

Limitations have been documented well. Please explain the diverse characteristics of the adult population, which was described in the manuscript that makes the study representative.

Data or study findings definitely help in better understanding of the pandemic but conclusions are not completely accurate.

We can only address public health relevance, if the diverse characteristics are explained. Also such a small population with not adequate power is a concern. But in setting of a pandemic, can be considered for further revisions

Reviewer #2: (No Response)

Reviewer #3: The conclusions are supported by the data presented.

I find that overall the discussion is broad and exaustive and very relevant to understand the impact and relevance of this study on the topic.

the public health relevance of the study is clearly addressed.

I have some issues on the limitations they describe:

- the characteristic of the populations are identified as the cause of discrepancies with previous studies on the same subject, but the authors do not detail which characteristic my affect the difference? Is the average age different? Is the ethnicity different? (line 203-204)

-I do not find 285 and 209 (lines 202-203) to be so different as sample sizes and the authors did not detail which population characteristics differ among the three studies. Can the authors comment on this with more detail?

- How does this number of redetectable positives compare with the specificity of the rt-pcr test they used?

- Since some of the authors are directly responsible for performing the nasal-throat swab on patients, could they comment on how the way the nasal-throat swab is performed might affect (if at all) their results? Would a sputum-based test impact in any way on their results?

--------------------

Editorial and Data Presentation Modifications?

Use this section for editorial suggestions as well as relatively minor modifications of existing data that would enhance clarity. If the only modifications needed are minor and/or editorial, you may wish to recommend “Minor Revision” or “Accept”.

Reviewer #1: 1) Explain conclusion that these RP patients unlikely are infectious.

2) diverse characteristics of the adult population

3)explain informed consent process

4)

Reviewer #2: (No Response)

Reviewer #3: I have a couple of general issues that are unclear for researchers not completely familiar with the local response to CoVid-19. I would suggest the authors comment more in their manuscript:

1. Why is there patients follow up after discharge? Is it a feature added from this study or is it common practice? If it is a feature of this study they should detail how the frequency of re-testing was decided.

2. Why if the discharged patients re-test positive they are re-admitted into the hospital? Is it common practice in their region? Is it a feature added by this specific study?

3. How does this number of redetectable positives compare with the specificity of the rt-pcr test they used?

4. Is there any possibility that any of the included patients has had a RP after the end of the study? They should clearly explain why this is not possible (i.e. all the patients have been discharged and checked for more than x weeks and it is unlikely to have a RP after this time).

In figure 2 maybe showing the entire distribution of the times instead of the median might be more informative. This is just a suggestion, it might also be more confusing, still it could be provided as a supplementary figure.

--------------------

Summary and General Comments

Use this section to provide overall comments, discuss strengths/weaknesses of the study, novelty, significance, general execution and scholarship. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. If requesting major revision, please articulate the new experiments that are needed.

Reviewer #1: Study strengths

1) design of study and data provided

2) able to define incidence, risk factors for RP positive, and clinical characteristics

3) population selection and manuscript writing

Weakness have been defined above. Sample size, power and generalizability to society are concerns

One other major concern is the availability of Ct values described and the test availability in most nations.

Study has significance in the present pandemic situation. Study has novelty in the number of tests done and follow up period.

Ethical concerns as above with informed consent

Reviewer #2: Results:

1. Table 2 Normal range for CT value ≥40 needs to be defined, as per my understanding it should be ≤40, if the authors mean something else they should explain it in the footnotes.

2. There is a difference in the viral shedding days between the N gene and ORF gene, can the authors postulate a probable cause for this phenomenon.

3. Line 168 results section mentions about the immunological assay (IgG and IgM) theses need to be defined in the methodology (test kit used, type of assay I,e, ELISA based or ICT).

4. Table 2 shows a p value of 0.045 for difference in Ct value at 3rd week however the results in table 4 are completely different. If both are comparison between RP and NRP then why this discrepancy. If these are two different analysis the title of the table should define it accordingly.

Discussion:

1. Line 198-202 the language needs to be modified it sounds a bit confusing

2. Line 203- “The incidence rate of our study is representative due to the diverse characteristics of the study population”. What characteristics are the authors talking about?

3. The author’s need to provide a justification or their view on why the patients who were negative at time of discharge after first hospitalization became positive? Is there any limitation of the test (in view of the Limit of detection) or is there a possibility that these might be case of reinfection?.

4. The justification produced to rule out reinfection completely is not reasonable, understanding that many cases now come as asymptomatic and knowing that the antibodies might have some effect on the virus the re-infected patients might present with mild symptoms, the authors can presume that these were not re-infection cases but the study methodology is not sufficient to rule out the same.

Reviewer #3: I think this manuscript is good enough to deserve pubblication with minor changes that I detailed above.

One of its major strength is that it is a perspective study and very few patients have been discarded and for very limited reasons (death and <18 years of age), so the study is significative, in that adds new knowledge to the field. The study evaluates a range of different epidemiological, clinical and virological variables and how they influence the re-positive testing outcome using appropriate statistical tests.

Some parts should be clearer, like what is part of the study design and what is rutine in Guangzhou Eighth People's Hospital (Guangzhou, Guangdong). For example is the follow-up testing after discharge rutin or part of the study? Is readmission to hospital upon positive re-testing rutine or part of the study design?

Every study of this kind is novel, in the sense that explores a new population and a slighltly different set of variables. Also this study has a slightly bigger sample size compared to the previous ones.

It is significant in that it gives risk factors to predict patients that might re test positive after testing negative. This is importat for example to exclude re-infections. But also to help any national healthcare system to be prepared to such cases (and how many to expect).

As a general execution it is well conceived, executed and written. They also provide all the data in tables that are very easy to read.

--------------------

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Pradeep Yarra

Reviewer #2: No

Reviewer #3: Yes: Alice Ledda

Figure Files:

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org.

Data Requirements:

Please note that, as a condition of publication, PLOS' data policy requires that you make available all data used to draw the conclusions outlined in your manuscript. Data must be deposited in an appropriate repository, included within the body of the manuscript, or uploaded as supporting information. This includes all numerical values that were used to generate graphs, histograms etc.. For an example see here: http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001908#s5.

Reproducibility:

To enhance the reproducibility of your results, PLOS recommends that you deposit laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see http://journals.plos.org/plosntds/s/submission-guidelines#loc-materials-and-methods

16 Jul 2020

Submitted filename: detailed list of responses.docx

Click here for additional data file.

27 Jul 2020

Dear Professor Wu,

We are pleased to inform you that your manuscript 'Incidence, clinical course and risk factor for recurrent PCR positivity in discharged COVID-19 patients in Guangzhou, China: a prospective cohort study' has been provisionally accepted for publication in PLOS Neglected Tropical Diseases.

Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow up email. A member of our team will be in touch with a set of requests.

Please note that your manuscript will not be scheduled for publication until you have made the required changes, so a swift response is appreciated.

IMPORTANT: The editorial review process is now complete. PLOS will only permit corrections to spelling, formatting or significant scientific errors from this point onwards. Requests for major changes, or any which affect the scientific understanding of your work, will cause delays to the publication date of your manuscript.

Should you, your institution's press office or the journal office choose to press release your paper, you will automatically be opted out of early publication. We ask that you notify us now if you or your institution is planning to press release the article. All press must be co-ordinated with PLOS.

Thank you again for supporting Open Access publishing; we are looking forward to publishing your work in PLOS Neglected Tropical Diseases.

Best regards,

Andrés Felipe Henao-Martínez, M.D.

Deputy Editor

PLOS Neglected Tropical Diseases

***********************************************************

25 Aug 2020

Dear Professor Wu,

We are delighted to inform you that your manuscript, "Incidence, clinical course and risk factor for recurrent PCR positivity in discharged COVID-19 patients in Guangzhou, China: a prospective cohort study," has been formally accepted for publication in PLOS Neglected Tropical Diseases.

We have now passed your article onto the PLOS Production Department who will complete the rest of the publication process. All authors will receive a confirmation email upon publication.

The corresponding author will soon be receiving a typeset proof for review, to ensure errors have not been introduced during production. Please review the PDF proof of your manuscript carefully, as this is the last chance to correct any scientific or type-setting errors. Please note that major changes, or those which affect the scientific understanding of the work, will likely cause delays to the publication date of your manuscript. Note: Proofs for Front Matter articles (Editorial, Viewpoint, Symposium, Review, etc...) are generated on a different schedule and may not be made available as quickly.

Soon after your final files are uploaded, the early version of your manuscript will be published online unless you opted out of this process. The date of the early version will be your article's publication date. The final article will be published to the same URL, and all versions of the paper will be accessible to readers.

Thank you again for supporting open-access publishing; we are looking forward to publishing your work in PLOS Neglected Tropical Diseases.

Best regards,

Shaden Kamhawi

co-Editor-in-Chief

PLOS Neglected Tropical Diseases

Paul Brindley

co-Editor-in-Chief

PLOS Neglected Tropical Diseases