Asymptomatic “Exposers” and “Infectors” are Sources of Infection by SARS-CoV-2: Analysis based on Family Clustering

Background: Previous studies have documented the clinical characteristics of patients with Coronavirus disease 2019(COVID-19) and presented evidence of person-to-person transmission. Limited data are available for patients with asymptomatic infections. Some asymptomatic carriers, whom we characterize as “exposers” or “infectors”, may be responsible for family clustering of COVID-19. Methods: A questionnaire survey and follow-up survey based on media reports were used to assess familial clustering of SARS-CoV-2 infection induced by asymptomatic exposers/infectors. Individual data were collected for all members of each tracked family. A transmission map was then drawn for each family. Results: Our study of 5 families indicated that individuals with no obvious symptoms of COVID-19, regardless of the PCR results, transmitted the virus to other family members who were community contained at home and had no contact with other infected individuals. There was one death case in Family No.3. Conclusion: Asymptomatic exposers/infectors of SARS-CoV-2 were all middle-aged (average age: 44.4 ± years) who had no symptoms but had the ability to disseminate the virus. Medical staff subsequent


Background
Since the rst report of an outbreak of pneumonia with an unknown cause in Wuhan, Hubei Province (China) 1-3 , there has been considerable controversy on the origin of the causative virus (SARS-CoV-2, also referred to as   4 , the epidemiology, and the pathogenesis of this novel coronavirus disease . The COVID-19 outbreak is now characterized as a pandemic because the virus has spread worldwide. Globally, as of 10:00 AM CEST, 6 May 2020, there have been 3,588,773 con rmed cases of COVID-19, including 247,503 deaths, reported to WHO. 5 Why was the outbreak so swift and violent? There was a report of an asymptomatic con rmed German patient who transmitted the virus to other people on January 30, suggesting that asymptomatic carriers were also infectious. 4 Subsequently, an article in the February 2 issue of Science questioned this notion, and concluded that the evidence was still insu cient. 5 Thus, there is controversy regarding the spread of this virus by asymptomatic carriers. As front-line workers in Wuhan, we found some phenomena among our colleagues that could shed light on this notion. In particular, there was evidence of an index case without any signs or symptoms, despite exposed to SARS-CoV-2, whose family members developed symptoms of COVID-19 following contact with the index case. A follow-up of these cases by chest computed tomography (CT) and virus PCR test con rmed infection by SARS-CoV-2.
We classi ed, for the rst time, individuals who had exposed to the con rmed COVID-19 cases, with no symptoms and negative CT and PCR results as "asymptomatic exposers", and those who had no symptoms and negative CT but laboratory-confirmed SARS-CoV-2 infection as "asymptomatic infectors".
There were few case reports of virus transmission by asymptomatic carriers by the news media in China.
We tracked these cases to investigate the possible role of asymptomatic "exposers" or "infectors" as sources of infection, which may help to prevent the transmission of the COVID-19 epidemic.

Data collection
A questionnaire with a two-dimensional code was designed that could be e ciently distributed through the mobile network. This questionnaire rapidly spread among WeChat groups and social media, was completed in the WeChat applet. This also allowed collection of background data and identi ed targeted households by extensive screening. The study subjects also provided informed consent through the media.

Inclusion and exclusion criteria
A family was included if: (a) two or more family members tested positive by PCR; (b) the asymptomatic index case had no signs or symptoms of infection but had close contact with a con rmed patient (highintensity exposure) in the epidemic area, even without any protective measures; (c) family members tested positive, but the index case tested negative (asymptomatic "exposer") or positive (asymptomatic "infector").
A family was excluded if: (a) every member of the familial cluster had signs or symptoms of COVID-19 (i.e., all of them were symptomatic); (b) the asymptomatic family member was not the index case responsible for the familial cluster outbreak.

Protection measures for study subjects and researchers
Appropriate protection measures were adopted for all study subjects. The study was deemed safe for all vulnerable subjects (minors, pregnant women, students, persons with no or limited capacities) and would not cause additional harm to these individuals.
Because the subjects contacted by the researchers could not be excluded as asymptomatic "exposers" or "infectors", all researchers adopted level-2 personal protective equipment(PPE) during events that required contact with the subjects (collecting specimens, etc.).

Data analysis and Statistics
After analysis of case data with the above inclusion and exclusion criteria, 5 groups of eligible familial clusters were included. Initially, these families answered a detailed questionnaire about the speci c route and the course of infection and the changes of symptoms. The focus of this study was to track asymptomatic "exposers" or "infectors". Nucleic acid PCR tests and chest CT scans were taken for all the family members. Some subjects who were tested previously were exempted. The con dentiality of all personal data, including medical les, personal and family information, was properly protected.
We collected the diagnostic data of all patients and compare the diagnostic e ciency of CT and nucleic acid tests. Data were collected retrospectively and reported as means ± SDs. P value below 0.05 was considered signi cant. All data were analyzed using SPSS 20.0 software.

Ethical statement
This study was approved by the Ethics Committee of Union Hospital A liated to Tongji Medical College of Huazhong University of Science and Technology.

Epidemiologic characteristics
Two hundred and forty questionnaires were collected and one family was selected for inclusion. Among the patients in the fever clinic, the symptomatic infected cases were tracked and included in the questionnaire for screening and one family met the criteria for inclusion. Through the news media, three families clustering outbreak due to asymptomatic "exposers" or "infectors" were included and back tracking. All ve families were included in the questionnaire for detailed investigation. We analyzed data of the ve familial clusters, each of which had evidence of con rmed infection. Analysis of transmission and the epidemic process According to our de nition, there were 7 asymptomatic exposers or infectors in the 5 families (Tab.1). Most of them were middle-aged (average age: 44.4 ± 14.9 years). The asymptomatic index cases in two families were medical workers who were exposed to subsequently con rmed cases, and did not take necessary protective measures during the very early stage of the epidemic. Thus, these individuals had high-intensity exposure. Although the onset of symptoms in asymptomatic exposers/infectors were generally relatively mild, but there was one death in one of the families.
In family No.1 (Fig.1B), Case 1.1 is classi ed as the asymptomatic exposer de ned by us, the PCR test was always negative despite repeated many times. However, from the analysis of the epidemic transmission process, Case 1.1 was highly suspected, although without any sign or symptoms, and transmitted to the family. Of course, we did not rule out the in uence of false negative nucleic acid test on the results. There was even one case with 8 times of negative nucleic acid test before the con rmation of COVID-19 reported by in China. As to Case 1.1 there was no symptom all the time, no more nucleic acid test was carried out during the follow-up survey. As to Case1.3 and Case1.4, they have typical symptoms and CT imaging of lung, but the nucleic acid test has always been negative, so it can only be classi ed as suspected infectors, or clinical con rmed cases (Fifth edition of diagnosis and Treatment Guide) 6 .
In family No.3 (Fig.1F), Case 3.2 and Case 3.3 were both old people at home, without related contact history. Case 3.1 has been to the epidemic area and had contact with some con rmed patients, but he has no symptoms. Case 3.1 was found in the screening of close contacts after the onset of Case 3.3, and only one positive test was found in ve times, which fully shows the concealment and variability of the virus after infection. However, Case 3.2 died shortly after the onset of the disease.
In family No.5 (Fig.2D.), Case5.2 is a 65 y/o woman who is the aged well living at home. She went to hospital for cough and fever (38.7 ℃) in January 30th. Chest CT examination showed classic groundglass opacities. According to clinical manifestations, she was highly suspected to be infected, and sooner con rmed by the nucleic acid test in January 31st. After the elderly was admitted, her neighbors, their loved ones and others were diagnosed in succession (the nucleic acid test was positive for asymptomatic infection), and the elderly had no history related to epidemic area, nor had they ever contacted with the con rmed or suspected patients. After investigation, her son-in-law Case 5.1 had a contact history with people from the epidemic area on January 17 to 19, and attend family reunion at his mother-in-law's home on January 21. Although Case 5.1 has not been infected, the process of epidemic people and their estimated asymptomatic proportion was 17.9% (95% con dence interval (CI): 15.5-20.2%). Considering the high proportion of the elderly, who is more vulnerable to the coronavirus, on the ship, he suspects that the rate in the general population may be closer to a recently derived estimate of 30.8% (95% con dence interval (CI): 7.7%-53.8%) from data of Japanese citizens evacuated from Wuhan. 13 An epidemiology group in China published a retrospective analysis of 889 cases with asymptomatic infections and found that they accounted for 1.2% of the total. 14 The data of China do not include asymptomatic infections as con rmed cases before April 1, 2020 15 . Wang et al 16 developed a susceptible-exposed-infectious-recovered model to assess the extent of invisible infection. Individual-level data on 25,961 laboratory con rmed COVID-19 cases reported through February 18, 2020 were extracted from the municipal Noti able Disease Report System of Wuhan City and found that at least 59% of infected cases were unascertained in Wuhan, potentially including asymptomatic and mild-symptomatic cases. In addition, the hidden virus carriers have not yet been detected cannot be excluded.

Clinical characteristics of asymptomatic carriers
Asymptomatic exposers/infectors have no signs or symptoms, and may even have no positive chest CT results. Hu et al 17 investigated the clinical characteristics of 24 cases with asymptomatic infection screened from close contacts of COVID-19 patients (or suspected patients) in Nanjing, none of the 24 asymptomatic cases presented any obvious symptoms while nucleic acid screening. 5 cases (20.8%) developed symptoms (fever, cough, fatigue, etc.) during hospitalization. 20 (50.0%) cases showed typical CT images of ground-glass chest and 5 (20.8%) presented stripe shadowing in the lungs. The remaining 7 (29.2%) cases showed normal CT image and had no symptoms during hospitalization. These 7 cases were younger (median age: 14.0 years; P=0.012) than the rest.

Causes of viral dissemination by asymptomatic carriers
A German research group published their research in medRxiv, on March 8th, 18 they found that SARS-CoV-2 could actively replicate in the upper respiratory tract. In other words, when the symptoms were mild, the virus was released easily by coughing or sneezing. This process was called "virus abscission" and spread to others. Of course, the chance of transmission caused by pathogen discharge in vitro is relatively less than that of con rmed cases due to the absence of clinical symptoms such as cough and sneeze.
After close contact with con rmed cases or tourism history of epidemic area, some patients with symptoms enter the diagnostic process, most of them do not have symptoms. Some of which were diagnosed as asymptomatic infectors by positive Nucleic acid PCR result, while others with negative result should be classi ed as asymptomatic exposers. In view of this situation, we propose a treatment process for asymptomatic infectors and asymptomatic exposers. (Fig4) How can an individual carrying negative PCR results cause infections in close contacts? There are three possibilities. Firstly, the results may be "false negative". Secondly, the person may have transmitted the virus before developing immunity, so the viral load at the time of PCR testing was below the detection limit. Thirdly, viable viruses may have been on their skin or clothing and spread among close contacts. It is suggested that their self-segregation will play an important role in the epidemic prevention and control (Fig.4) .

Family clusters of infection caused by medical staff
Healthcare workers and elderly people had higher attack rates and severity risk increased with age and special efforts are needed to protect vulnerable populations, including healthcare workers, elderly and children 16 . In our study, two families were exposed to index cases who worked in a hospital, which was not an isolated phenomena that appeared by accident. During the very earlier stages of the epidemic when the cause was unknown, some medical care providers did not have necessary protective measures. Therefore, the staff participating in the care of COVID-19 patients in the hospital should be considered as a high-risk group, and they should be quarantined so as to protect their families (elderly and children) even if they have no obvious symptoms.
Interpretation and response to false negative results Xie X et al 19 reported some patients with positive chest CT ndings may present with negative results of RT-PCR. How did this happen? Firstly, in principle, the speci city is high while the sensitivity is relatively low, so there is the possibility of false negative results. Secondly, the PCR kit has a detection limit and the result is positive only when the number of viruses in the sample exceeds that limit. Thirdly, apart from sampling technique, samples from the lower respiratory tract are best while most samples were from the upper respiratory tract considering inconvenient and invasive. Fourthly, other factors such as preservation, transportation of the sample and so on. The presence of some false negative results is inevitable. However, it is necessary to improve our understanding and correctly interpret the results. So that the PCR detection can play the greatest role in diagnostic virology, rather than misleading to the opposite direction.

Value of CT imaging
Generally, CT provided more rapid detection of COVID-19 than PCR Fig.3 , during the early stage of the epidemic, there were few false negative PCR results because of a shortage of PCR kits. It was reported that positive cases could appear even after 2 or 3 negative PCR results 20-21. Particularly, in the fth edition of the NHC guidelines for diagnosis and treatment 6 , in high incidence areas, CT alone was regarded as the standard for clinical diagnosis, and these cases were classi ed into clinical diagnosis cases, such as Case1.3 and Case1.4. (Fig.1) Moreover, the correlation between PCR positivity and symptoms of COVID-19 is not signi cant. A CT examination can provide early identi cation of pulmonary parenchymal lesions, which is helpful for guiding treatment decisions. Lung improvements (based on CT) can allow caregivers to adjust the treatment process and even assess prognosis. The results from CT imaging are more important than those from PCR testing for determining the seriousness of an infection [20][21] . However, PCR detection is necessary to establish etiology and con rmation of the diagnosis because CT imaging cannot distinguish among different causes of pneumonia.

Conclusion
We can draw some conclusions from this study. Firstly, asymptomatic exposers or infectors, are mostly middle-aged people with strong immunity system, but are nonetheless likely to disseminate the virus. Secondly, asymptomatic "exposers" or "infectors" in two of the analyzed families were medical workers, indicating that health care workers participating in the treatment of COVID-19 cases are under high risk.
Even if these individuals have no any symptoms, they should be quarantined from their families to prevent family outbreaks. Thirdly, it is necessary to strictly quarantine and observe all people from an epidemic area, rather than simply screening body temperature. We recommend routine performance of PCR testing and chest CT scans. Fourthly, the onset of symptoms caused by asymptomatic "exposers" or "infectors" were generally relatively mild, but there was one death in one of the families. This require further study, but the harm of asymptomatic infection should not be ignored. Finally, urgent measures are needed to contain mild and asymptomatic cases that are exacerbating the pandemic, strong social distancing is the only way to stop the spread of the virus, such as closure of schools, the cancellation of public gatherings, and the keeping of people at home and away from public places.
In general, the focus of this study is on the family cluster outbreaks caused by asymptomatic "exposers" or "infectors", and only 5 families are traced, which is the inadequacy of this study. On March 30, 2020, the o cial website of the National Health Commission of China began to release the number of asymptomatic infections. As of April 14, 6764 cases of asymptomatic cases have been reported in China, including 588 imported cases from abroad. There were 1297 con rmed cases, 251 of which were imported from abroad, 4444 were removed from medical observation, 109 of which were imported from abroad. 1023 asymptomatic cases still under medical observation, 228 of which were imported from abroad. 22 Compared with 82601 con rmed cases in China, the proportion of asymptomatic cases is 7.6%, which is different from 18-59% estimated in other literatures. 12 Then we submitted an application to the ethics committee for exemption from signing an informed consent and the application has been approved.

Consent for publication
Multiple patients involved were informally asked if they felt the results reported herein were re ective of their illness experience to check the validity of the ndings.

Availability of data and materials
All data generated or analysed during this study are included in this published article.

Competing interests
All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_ disclosure.pdf and declare: no nancial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have in uenced the submitted work  Time from the onset of symptoms to positive CT results and positive PCR results.