Persistence of SARS-CoV-2 RNA in Nasopharynx, Blood, Urine and Stool of Patients with COVID 19: A Hospital-based Longitudinal Study

Background: This study was conducted to determine the persistence of SARS-CoV-2 RNA in the nasopharynx, blood, urine and stool of patients with COVID-19. Methods: In this hospital based longitudinal study, 100 confirmed COVID-19 cases were recruited, between March and August 2020 in Guilan province (Northern Iran). Nasopharynx, blood, urine and stool samples were obtained from each patient at the time of hospital admission, discharge, followed by one week after discharge and every 2 weeks until all samples were negative for SARS-CoV-2 RNA by reverse transcription polymerase chain reaction (RT-PCR). Survival analysis was used to identify the duration of virus persistence over time. Results: Positive blood, urine, stool RT-PCR were detected in 24%, 7% and 6% of patient respectively. The median duration of virus persistence in blood, urine and stool were 7 days (95% CI: 6.07–7.93), 6 days (95% CI: 4.16–8.41) and 13 days, 95% CI: 6.96–19.4), respectively. The maximum duration of virus persistent in blood, urine and stool were 17, 11 and 42 days from admission, respectively. Conclusions: : According our results, until obtaining definite evidence of the duration of infective viral shedding, prolonged isolation duration at least 25 days from admission to hospital and strict hygienic measures for about one month were recommended.


Introduction
The Coronavirus disease 2019 (COVID- 19) was declared a global epidemic by the World Health Organization in December 2020 [1]. By November 22, 2020, it has infected almost 57 million people with more than 1,300,000 deaths [2]. Iran was classi ed as the 8th country in terms of COVID-19 morality [3].There were 1,189,203 con rmed cases of COVID-19 with 54,440 deaths in Iran, as of 15 February 2019 until 25 December 2020 [4].
The principle route of COVID-19 transmission was reported to be direct and indirect contact with respiratory droplets [5]. Although, other routes of transmission, such as: mother to fetus, fecal-oral, airborne are currently controversial and subject to future investigations [6,7], however, the epidemiological studies have shown that COVID-19 patients had close contact or proximity in distance with someone who has COVID-19 [8].
Initially, SARS-Cov-2 was isolated and recognized in respiratory samples by real-time RT-PCR [9]. However, in recent studies, viral nucleic acids have subsequently been detected in urine, stool, and gastrointestinal mucosa [10,11]. A study reported, after nasopharyngeal RT-PCR was negative, eight of ten children with COVID-19, had persistently positive RT-PCR on rectal swabs [12].
Base on The Centers for Disease Control and Prevention (CDC) guidelines, all patients with a positive respiratory RT-PCR test must be isolated until at least 24 hours with no fever, and at least 10 days from Page 4/15 initial symptom onset [13]. However, there are several case reports on the persistence of positive RT-PCR in patients with COVID-19, indicating that individuals can remain positive after symptoms have resolved [14,15]. Therefore, the persistence of the virus in body uids can increase the potential risk of virus transmission in asymptomatic or recovered patients [15]. So, it remains uncertain whether the quarantine duration, recommended by CDC is su cient to reduce transmission [16].
Considering that, the frequency with which SARS-CoV-2 RNA can be detected in body uids and the period during which it remains detectable are not well understood; this longitudinal study was conducted to determine the persistence of SARS-CoV-2 RNA in the nasopharynx, blood, urine and stool of patients with coronavirus disease 2019 via collecting sequential samples every 2 weeks.

Study population and design
This hospital based longitudinal study was designed for a period of 6 months, between March 2020, and August 2020. Study participants were selected by a covenant sampling method from patients who were hospitalized with a con rmed coronavirus 2019 diagnosis in the only referral Hospital of Rasht, Guilan (Northern Province of IRAN). Con rmed case of COVID-19 was de ned as positive real-time uorescent quantitative polymerase chain reaction (RT-PCR) of nasopharyngeal specimen [17]. The sample size was estimated to be 100 participants with con dence level of 95% and test power of 80%. If participants refused to give a sample, they were excluded from the study.
The study was approved by the local ethical committee of Guilan University of Medical Sciences, Rasht, Iran (permit code IR.GUMS.REC.1399.013). Written informed consent was obtained from every participant.

Measurements
The samples of nasopharynx, blood, urine and stool were obtained from each participant through following schedule: at the time of admission -at the time of discharge -one week after discharge and every 2 weeks until all samples were negative for SARS-CoV-2 RNA on RT-PCR. At the time of admission, all four samples (nasopharynx, blood, urine and stool) were collected from each participant and were analyzed for SARS-CoV-2 RNA by PCR. At the time of discharge nasopharynx and stool samples were collected from each participant, but blood and urine specimen were collected if they were positive for SARS-CoV-2 RNA, at the time of admission. At each follow up visit, samples of nasopharynx, blood, urine and stool were collected if they were positive for SARS-CoV-2 RNA, at the previous visit.
Nasopharynx and stool samples were obtained by sterile, Dacron swabs. Also, ve ml whole blood and urine were collected for SARS-CoV-2 virus-speci c real-time RT-PCR. The samples were processed immediately after sampling by a trained laboratory technician. The samples were tasted for SARS-CoV-2 virus using the Ribo Virus RT-PCR Kit (Sacace Biotechnologies, Como, Italy) according to the manufacturer's instructions.
In addition, relevant information on clinical and demographic characteristics of participants was collected. The collected data on sociodemographic characteristics of patient was included age, gender, marriage, job, education, Residency, socioeconomic status. The collected data on clinical manifestation of COVID-19 were included fever, cough, sore throat, dyspnea, weakness, muscular pain, headache, diarrhea, nausea and vomiting and chills. Information on underlying disease (including diabetes, hypertension, cardiovascular disease, immunode ciency, cancer and respiratory disease) and in ammatory markers (including WBC, ESR and CRP) were also collected. Hospital treatment plans were categorized in 4 groups, including hydroxychloroquine, anti-viral (such as lopinavir, Sovodak (Sofosbuvir at 400 mg and daclatasvir at 60 mg and Remdesivir), interferon β1 ( ve doses (44µg each dose) /daily for 3 days a week / subcutaneous) and local treatment (including diphenhydramine, acetaminophen, zinc, vitamin C, famotidine). All the patients received corticosteroids (dexamethasone 8 mg/daily).

Data analyses
Comparisons of qualitative variables (clinical and demographic factors) between three groups (categorized based on duration of SARS-CoV-2 RNA persistence in the nasopharynx) were performed with the Chi-square or Fisher exact tests. Survival analysis was performed to identify the median and 95 percentile durations of SARS-CoV-2 persistence. To nd clinical and demographic factors that might be associated with persistence of SARS-CoV-2 RNA, Cox regression analysis was used. Data analyses were performed using SPSS version 17.0 software (SPSS Inc., Chicago, IL, USA).

Results
During the study period, 106 hospitalized patients with positive nasopharyngeal RT-PCR of SARS-CoV-2 RNA were enrolled to the study, but 6 of these were died (5.66%). Mean age of the participants was 53.30 ± 13.03 years (range 29-86 years) and the majority of the patients were male (60%), were married (84%), had educated, less than diploma (63%), were urban residents (85%), were employed (68%) and had low socioeconomic status(82%). The result of SARS-CoV-2 RNA RT-PCR on different sample, over time is shown in Fig. 1. Only 24% of RT-PCR on blood sample was positive at admission and one sample remained positive at discharge and was cleared after 1 week. RT-PCR on urine sample was positive in 6 % of participants at admission and all were cleared at discharge. RT-PCR was positive in stool samples at admission in 4 participants and 2 remained positive at discharge, in addition, 2 new stool samples were positive at discharge. Three of them were cleared after one week and one was cleared 5 weeks after discharge. At discharge, 30% of nasopharyngeal RT-PCR was positive and all were cleared after one weak.
Participants' clinical and demographic characteristics based on duration of virus persistence in the nasopharynx are shown in table2. Nearly 11% of patients had virus persistence in nasopharynx longer than 14 days from the rst positive test. Demographic characteristics of patient were not signi cantly different among the 3 groups. Similarly, there was no signi cant association between COVID-19 The median duration of SARS-CoV-2 persistence in nasopharynx from the rst positive RT-PCR test at admission time was 8 days (95% CI: 6.91-9.09) and maximum duration of virus persistence among them was 25 days from admission (Fig. 2). The median duration of SARS-CoV-2 persistence in blood from the rst positive RT-PCR test at admission time was 7 days (95% CI: 6.07-7.93) and maximum duration of virus persistence among them was 17 days from admission ( gur2). The median duration of SARS-CoV-2 persistence in stool from the rst positive RT-PCR test was 13 days (95% CI: 6.96-19.4) and maximum duration of virus persistence among them was 42 days from admission (Fig. 2). The median duration of SARS-CoV-2 persistence in urine from the rst positive RT-PCR test at admission time was 6 days (95% CI: 4. 16-8.41) and maximum duration of virus persistence among them was 11 days from admission ( Fig. 2).
In our analysis of factors that might be associated with durations of SARS-CoV-2 persistence by Cox regression, we did not nd any statistically signi cant associations (data not presented).
Frequency of some COVID 19 gastrointestinal (GI) symptoms according to result of stool RT-PCR are presented in Fig. 3. Abdominal pain and diarrhea were signi cantly more common in stool RT-PCR positive patient than stool negative ones.
In analysis of clinical and demographic factors that might be associated with positive blood and urine RT-PCR, we did not nd any statistically signi cant associations (data not presented).

Discussion
In this longitudinal study, we demonstrate that the maximum duration of nasopharyngeal (25 days) and fecal (42 days) SARS-CoV-2 RNA shedding is longer than the recommended (at least 24 hours with no fever, and at least 10 days from initial symptom onset) quarantine duration [13].
Nearly, one third of our hospitalized COVID-19 patients had positive nasopharyngeal RT-PCR at the time of discharge and about one tenth of them had virus persistence in nasopharynx longer than 14 days from the rst positive test. Maximum duration of nasopharyngeal shedding was 25 days from admission. These ndings are compatible with some studies that revealed, some patients continued to be upper respiratory tract RT-PCR positive after discharge from hospital, for the next few days [18,19]. We did not identify any determinants of viral persistence, but a study in china demonstrated that the prolonged presence of the virus in upper respiratory tract was associated with disease severity [20].While, another study in Portugal revealed the viral RNA persistence not associated with disease severity and stronger immune response are the determinants of virus RNA clearance [21].
Positive blood SARS-CoV-2 RNA RT-PCR was detected in about one third of our patient, but most of them turned to negative at the time of discharge. The studies on detecting SARS-CoV-2 RNA in blood were limited, but a study in china revealed that the SARS-CoV-2 RNA detected in the blood of 6 out of 57 Chinese patients and all 6 positive bloods RT-PCR had a severe clinical picture [22]. The higher positive rate of blood RT-PCR in our study was probably due to the disease severity in our patients and hospitalbased design.
Our ndings revealed that the positive urine SARS-CoV-2 RNA RT-PCR was less common (7 out of 100 patients) in our study population and all of them turned to negative at the time of discharge. Our result was in accordance with studies in Turkey and China that demonstrated nearly 5-7% of COVID-19 patient had positive urine RT-PCR [23,24].
Positive stool SARS-CoV-2 RNA RT-PCR was detected in only 6 out of 100 patients in our study, but the duration of time for a positive stool RT-PCR test to turn negative was longer than nasopharynx, urine and blood RT-PCR tests, with a maximum duration of fecal shedding of 42 days. Therefore, fecal positive hospitalized patients with positive stool RT-PCR, represent the need for precautions and protective equipment for interventional procedures involving the gastrointestinal tract in a hospital environment., Similarly, a study in china demonstrates that an asymptomatic case was positive in the stool RT-PCR for a period as long as 42 days and also, reported that in almost two third of patients, the clearance of fecal shedding takes a longer time than the nasopharyngeal sample [19].
Although, our ndings demonstrate that the maximum duration of nasopharyngeal and fecal SARS-CoV-2 RNA shedding is longer than the recommended quarantine duration by the CDC 13 , however, it is unclear whether individuals with persistent positive RNA PCR represent an infectious risk. Future studies are needed to determine whether PCR positivity is due to infective virus or non-infective nucleic acid fragments. Therefore, until such studies are concluded, prolonged isolation duration at least 25 days may be recommended, and aggressive contact tracing might also be considered. Also, patients are advised to take strict hygiene measures, especially those with gastrointestinal symptoms for about one month, in order to prevent potential fecal-oral transmission.
The strengths of the current study are the long term follow up and detecting virus RNA in respiratory and extra-respiratory sites. However, the hospital-based design is the limitation of our study, leading to more severe patient enrollment that limits the generalizability of ndings. Another limitation of our study was that the day of symptom onset was not available, and all calculations and analyses were based on rst nasopharyngeal RT-PCR test at admission time thus, the duration of viral shedding was underestimated. Another limitation of current study was that the, virus isolation and tests of specimens' infectivity were not conducted and CT value of RT-PCR was not available.

Conclusion
Our ndings demonstrate that the maximum duration of nasopharyngeal and fecal SARS-CoV-2 RNA shedding in hospitalized COVID-19 patients is longer than the recommended quarantine duration by CDC. Therefore, until obtaining de nite evidence of the duration of infective viral shedding, prolonged isolation duration for at least 25 days from admission to hospital may be recommended and aggressive contact tracing might also be considered. Also, patients are advised to take public health measures such as strict personal hygiene, especially those with gastrointestinal symptoms for about one month, in order to prevent potential fecal-oral transmission.

Abbreviations CDC: The Centers for Disease Control and Prevention
Declarations