Outbreak of acute respiratory disease caused by human adenovirus type 7 and human coronavirus-229E in Zhejiang Province, China
Shelan Liu, Jian Cai, and Yumin Li contributed equally to this study.
Abstract
In 2019, an outbreak of pharyngoconjunctival fever (PCF) occurred at a swimming center in Zhejiang Province, China. A total of 97 (13.55%) of the 716 amateur swimmers had illnesses, with 24 patients (24.74%) hospitalized in the pediatric ward. Human adenovirus serotype 7 (HAdV-7) was isolated from one concentrated water from the swimming pool, and 20 of 97 positive cases without liver damage. This outbreak led to a nosocomial outbreak in the pediatric ward, in which 1 nurse had a fever and was confirmed to be adenovirus positive. The hexon, fiber, and penton genes from 20 outbreak cases, 1 water sample, and 1 nurse had 100% homology. Furthermore, 2 cases admitted to the pediatric ward, 2 parents, and 1 doctor were confirmed to be human coronaviruses (HCoV-229E) positive. Finally, all outbreak cases had fully recovered, regardless of a single infection (adenovirus or HCoV-229E) or coinfection of these two viruses simultaneously. Thus, PCF and acute respiratory disease outbreaks in Zhejiang were caused by the completely homologous type 7 adenovirus and HCoV-229E, respectively. The swimming pool water contaminated with HAdV-7 was most likely the source of the PCF outbreak, whereas nosocomial transmission might be the source of HCoV-229E outbreak.
1 INTRODUCTION
Adenoviruses are medium-sized (90−100 nm), non-enveloped icosahedral viruses with double-stranded DNA.1 Human adenoviruses (HAdVs) are classified in the genus Mastadenovirus, which contains seven known HAdV species HAdV-A to HAdV-G2 since the first isolation of adenovirus from children's tonsils in 1953, a total of 113 known genotypes have been identified as defined by the absence of serological cross-neutralization.3-6 HAdV is a common pathogen among children and adults. Syndromes, such as acute respiratory disease in military recruits, pharyngoconjunctival fever (PCF), and epidemic keratoconjunctivitis depending on the HAdV types, have been recognized for many years, although the incidence of clinical adenovirus illness is lower among adults than among children.5-11
PCF, an illness characterized by fever, conjunctivitis, and pharyngitis, is often caused by adenovirus types 3, 4, and 7.4, 7, 8, 12 PCF outbreaks are often found in persons who use common recreational small lakes or swimming pools and bathing facilities without adequate chlorine.4, 7, 8, 12 HAdVs are highly stable and can survive for extended periods in water.4, 13, 14 The transmission of HAdVs in swimming pools can occur through ingestion, inhalation, or by direct eye contact with contaminated water.4
Although the first recorded outbreaks of PCF associated with adenovirus transmission in a swimming pool were reported more than 50 years ago, this type of outbreak has been reported only rarely in the past few decades.4, 7 On July 30, 2019, the Epidemiological Surveillance Network of Zhejiang Province, China was notified of an unusually high number of children with fever, pharyngitis, and/or conjunctivitis who had visited the same swimming center located in the Jinyun District of Lishui City. The present study describes the epidemiological, clinical, environmental, and virological investigations that were performed relative to this outbreak along with the subsequent outbreaks of nosocomial coinfections with adenovirus and human coronavirus 229E (HCoV-229E) in the paediatric ward of the Second People's Hospital of Jinyun county. To our knowledge, this is the largest outbreak of PCF caused by human adenovirus serotype 7 (HAdV-7) and HCoV-229E, and this report is the first communication to describe the isolation of an adenovirus from swimming pool water. This study on an adenovirus outbreak in children before the SARS-CoV-2 pandemic is of great significance to the current investigation of hepatitis of unknown origin in children around the world.15
2 METHODS
2.1 Outbreak recognition
On July 30, 2019, the Jinyun District Center for Disease Control and Prevention was notified by the emergency department of the Second People's Hospital of Jinyun of an unexpected increase in the number of patients with fever, conjunctivitis, and/or pharyngitis along with respiratory symptoms. All of the patients attended different schools and had a history of swimming at the same swimming center located in Huzhen town before illness onset (see Supporting Information: Figure 1). In response, between August 3 and 5, 2019, the Zhejiang Provincial CDC and local CDC initiated an epidemiological investigation to determine the source of origin and the scope of the outbreak and to provide infection control recommendations.
2.2 Case definition
A suspected case was defined as a person who had swum in either of the two swimming pools at a private, recreational swimming center since July 20, 2019 who presented at the Second People's Hospital of Jinyun or other nearby hospitals with fever, symptoms of conjunctivitis, and/or pharyngitis, and respiratory symptoms or gastrointestinal symptoms without laboratory evidence. A clinically diagnosed case-patient was defined as a suspected case showing epidemiological evidence or positive adenovirus rapid antigen test (rapid colloidal gold test strip). A confirmed case was defined as a suspected case or clinically diagnosed case being confirmed by PCR assay or sequencing for HAdV infection.16
2.3 Field investigation
Questionnaires were designed and used by trained personnel to collect demographic and clinical information from the patients. Affected patients were interviewed to record the following variables: place of residence, age, sex, date of symptom onset, symptoms, presence of complications, swimming pool use, and other potential exposures. The exposed population included 716 swimmers from different schools. The possible occurrence of additional cases of PCF in the surrounding towns and a referral hospital was monitored.
2.4 Surveillance of healthcare workers (HCWs) and bedside caregivers
Surveillance of 16 HCWs (10 nurses and 6 doctors) and 50 bedside caregivers was performed in where the patients were admitted during July 20 and August 31, 2019. A suspected case of HAdV infection in an HCWs and bedside caregivers was defined clinically as acute onset of cough, fever, or conjunctivitis.
2.5 Laboratory investigations
2.5.1 Sample collection and transportation
A total of 20 throat swabs and 1 conjunctival swab were obtained from 19 admitted patients (a total of 24 cases were admitted). One throat swab was from a mild case who presented to the fever clinic and met our case definition of PCF.
Also, 9 of the pharyngeal specimens were collected from 4 bedside caregivers and 5 HCWs, the 3 cases due to non-PCF diseases, who shared with the same ward during August 1 to 6. All samples were transported to the Zhejiang CDC laboratory at 4°C in an ice box and stored at 4°C in a refrigerator until testing for DNA by PCR within 24 h.
Three samples of water were obtained from the swimming pools on July 31, 2019. These samples were concentrated through the adsorption-elution method, and the filter was rinsed with 3 ml of sterile phosphate-buffered saline, which was added to the concentrate.4
2.5.2 Detection for the adenovirus DNA and antigen
At the Zhejiang CDC, adenovirus DNA extraction was performed following the instructions in the kit (catalogue no. 51106; Qiagen Company). All clinical specimens were tested using an HAdV and HCoV-229E viruses real-time PCR (RT-PCR) or PCR assay, as described elsewhere.4 The swimming pool water samples were detected for the hexon gene of the human adenovirus by nested PCR. The S1 gene of HCoV-229E were amplified through nested RT-PCR using primers according to the published paper.17 The primer sequences used for the detection of hexon, fiber, and penton gene of adenovirus type 7 and S1 gene of HCoV-229E by RT-PCR were listed in Supporting Information: Table 1. Additionally, all samples were tested for the 14 most common respiratory pathegen (including influenza, coronavirus, respiratory syncytial virus (RSV), and rhinovirus, etc.) as well as enteroviruses (such as norovirus, enterovirus 71 (EV71), coxsackieviruses A (CoxA), and rotavirus, etc.).4 A rapid colloidal gold test strip was used for detection of adenovirus antigen.
2.5.3 Sequencing and sequences analysis
PCR products were sequenced using the ABI PRISM 3730XL DNA Analyzer (Applied Biosystems) by Sangon Biotech Co., Ltd. In total, 22 of the positive samples of HAdV were typed by sequencing the full genes of the hexon protein gene, the fiber protein gene, and the penton protein gene. Further, the adenovirus type was determined using the three genes and the MegAlign software package.4 Five positive samples of HCoV-229E were typed by sequencing the S1 gene of the spike protein gene and 4 S1 sequences were successfully acquired.
2.5.4 Phylogenetic analyses
A total of 64 sequences of the hexon protein gene, the fiber protein gene, and the penton gene segments from 20 pediatric cases, 1 nurse, and 1 water sample from the large swimming pool) and 4 S1 protein gene of the HCoV-229E viruses from admitted case 13, admitted case 7, and her parents were uploaded to the NIH genetic sequence database (GenBank, https://www.ncbi.nlm.nih.gov/genbank/). Details of the sequences and the sequence GenBank number are listed in Supporting Information: Tables 2 and 3. The hexon of the HAdV and S1 genes of HCoV-229E were phylogenetically analyzed separately, together with referenced sequences from GenBank. The nucleotide sequence-based phylogenetic trees were constructed via maximum likelihood method with bootstrap analysis (n = 1000) using the MEGA5.0 program.18
2.5.5 Water testing
Samples of concentrated swimming pool water were tested for adenovirus and most common enteroviruses (norovirus and rotavirus).4 Meanwhile, water was tested for levels of residual chlorine as well as total chlorine and pH with diethylphenylenediamine using the Palin method after the voluntary closing of the pool on August 2. Water was tested for aerobic bacterial count and coliform bacteria, in accordance with the national guidelines of China for the microbiological examination of swimming pool water (PRC Ministry of Health).4
2.6 Statistical analysis
All statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS; version 26.0). Categorical variables were compared using the χ2 test and Fisher's exact test. The Mann−Whitney test was used to compare continuous variables. All statistical tests were two-sided, and a p value of 0.05 was considered statistically significant. All statistical analyses in this study were referenced to the medical research of the statistical guideline written by Lee.19
2.7 Role of the funding source
This study was supported by the Zhejiang Provincial Program for the Cultivation of High-Level Innovative Health Talents, which had no role in study design, data collection, data analysis, data interpretation, writing of the report, or the decision to submit.
3 RESULTS
3.1 Epidemic of the PCF outbreak
The outbreak occurred in the summer of 2019, between July 23 and August 4, and spanned 2 weeks. During the outbreak, 97 children reported sick, and 22 cases were confirmed as adenovirus infections. A total of 75 met the clinically diagnosed definition. 24.75% (24/97) of cases were admitted to hospitals, but there were no severe or fatal cases. In total, 87 of the children (89.70%) had recently used the affected swimming pools, which was considered the source of infection. Ten (10.30%) other cases amongst the families of the swimmers were noted as secondary cases due to having had close contact with a primary case.
Index case, a 10-year-old female student, developed a high fever on July 23 and visited the swimming center on July 24. On July 25, she was examined at the Second People's Hospital of Jinyun due to fever, pharyngeal redness, tonsil enlargement, and diarrhoea. She was admitted to the hospital on the same day. Her white blood cell (WBC) count was found to be 11.8000 per mm3 (normal range: 4500−10 000 WBC per mm3). Neutrophils made up 80.04% of all the WBC (normal range: 40%−70%). The rapid influenza direct antigen testing was negative for the influenza A and B viruses. Through antivirus treatment and symptomatic treatment, this patient was discharged on August 5. The last case was onset fever on August 4. The timeline distribution of these cases is shown in Figure 1, where it can be seen that the single peak of the outbreak (a total of 45 cases, accounting for 46.39% of all outbreak cases) occurred during July 30 to August 1. This curve was consistent with the hypothesis of a persistent common source. In addition, 10 isolated secondary cases resulted from person-to-person transmission, mainly within families.
Amongst the 97 outbreak cases, children were most affected (56/97, 57.73%) living in Huzhen town where the swimming center is situated (Supporting Information: Figure 1). A total of 34.02% (33/97) came from Dongfang town, whilst 3.09% (3/97) were from the Sanxi countryside. A further 2.06% (2/97) of cases lived on Wuyun Street and Xiandu Street, and 1.03% (1/97) lived in the Shuhong countryside (see Supporting Information: Figure 1).
Of the 97 outbreak cases, there were 53 (54.64%) boys and 44 (45.36%) girls. The male-to-female ratio was 1.20:1. All cases were between 2 and 15 years old. A total of 80 cases (82.47%, 80/97) were between 9 and 14 years old (see Supporting Information: Figure 2). Three of the cases were preschool children, 6 were in kindergarten, and 88 cases were in primary or middle school.
3.2 Clinical manifestations of PCF outbreak
Symptoms usually associated with PCF and their frequency of occurrence amongst the 97 patients are shown in Table 1. All of the patients (100%, 97/97) had fever, with 38.14% (37/97) having a high fever over 40°C. In addition to fever, pharyngitis (92.78%, 90/97) and fatigue (42.2%, 41/97) were the most commonly reported symptoms. There was no difference in symptoms noted according to age, sex, or direct exposure to the pools. The clinical course usually lasted approximately 7 days.
| Signs and symptoms | No of cases | Percentage（%） |
|---|---|---|
| Fever | 97 | 100 |
| Sore throat | 90 | 92.78 |
| Fatigue | 41 | 42.27 |
| Dizziness | 40 | 41.24 |
| Headache | 30 | 30.93 |
| Vomiting | 22 | 22.68 |
| Cough | 18 | 18.56 |
| Conjunctivitis | 17 | 17.53 |
| Diarrhoea | 16 | 16.49 |
| Myalgia | 11 | 11.34 |
A total of 24 patients (24.74%, 24/97) were hospitalized for persistent fever, 1 of whom developed pneumonia. The median age of the hospitalized patients was 9 years old, ranging from 2 to 14 years, and they comprised 16 boys (66.67%) and 8 girls (33.33%). All cases received symptomatic treatment in the hospital and were discharged fully. The median incubation was 4.0 days among the 24 admitted cases (3.5 days in the other outpatients, p = 0.065). The median time from illness onset to hospital admission was 3 days (0−6 days). The average hospital stay was 7 days (3−11 days). The average time from illness onset to discharge was 10 days (3−17 days) see Figure 2.
3.3 Swimming pools related with PCF outbreak
The Jinshatan swimming center is located in the suburb of Huzhen in the Jinyun district, Lishui city (see Supporting Information: Figure 1). The center has 14 workers, including 8 professional swimming coaches, who work in 2 outdoor swimming pools. The large pool is 50 × 25 m in size and is divided into a deep-water zone with a depth of 1.2 m and a shallow water zone with a depth of 0.8 m. The small a wading pool, located 20 m away from the large pool, is 10 × 10 m in size and includes only a shallow water zone with a depth of 10 cm (see Supporting Information: Figure 3). The two pools have separate water inlets and drainage systems, and their water is not mixed.
The large pool is used by adults and children over 3 years of age. A total of 716 students from five schools along with other swimmers had visited the large swimming pool between July 3 and 30, 2019. The swimming training schedules included five sessions: 9:00−10:30 a.m., 10:30 a.m.−12:00 p.m., 13:30−15:00 p.m., 15:00−16:30 p.m., and 16:30−18:00 p.m. The average number of swimmers was approximately 450 persons per day, with an average of 100−200 per session.
The small pool is intended for children between the ages of 2 and 6 years. Approximately 30 children and their parents visit the small pool each day. Of the 97 patients who were positive for HAdV type 7, 75 cases had used the large outdoor swimming pool, 2 cases used the small pool, and 18 cases were exposed to both pools.
3.4 Laboratory results
Adenovirus antigen rapid testing from pharyngeal specimens was performed in 65 of 716 swimmers who developed acute respiratory symptoms from July to August 2019. The results showed that 25 cases were adenovirus antigen-positive see Supporting Information: Table 4. A total of 20 pharyngeal specimens and 1 conjunctival swab were collected from the 19 of 24 admitted cases along with 1 pharyngeal specimen from 1 outpatient. A total of 22 samples were tested for the presence of adenovirus DNA by PCR. The results showed that 20 samples were adenovirus positive. The mean cycle threshold (Ct) value was 24, ranging from 19.1 to 30.3 (see Table 2). The 22 complete hexon, penton, and fiber gene sequences were searched for in GenBank with blast analysis and identified as HAdV-7. The nucleotide identity of all sequences was 100%, indicating that this outbreak was likely caused by the same source of infection, see Figure 3.
| Case number | Case type | Gender | Age (years old) | Onset date | Clinical presentation | Sampling date | Type of samples | Human adenovirus（Ct value） | Other pathogens （Ct value） |
|---|---|---|---|---|---|---|---|---|---|
| 1 | Admitted case | Female | 4 | 28-7-2019 | Acute purulent tonsillitis | 2-8-2019 | Pharyngeal swab | Negative | Echo11 （29.9） |
| 2 | Admitted case | Female | 10 | 27-7-2019 | Acute purulent tonsillitis | 2-8-2019 | Pharyngeal swab | 22.4 | Negative |
| 3 | Admitted case | Male | 8 | 28-7-2019 | Acute purulent tonsillitis | 2-8-2019 | Pharyngeal swab | 25.5 | Negative |
| 4 | Admitted case | Male | 12 | 28-7-2019 | Acute tonsillitis | 2-8-2019 | Pharyngeal swab | 24.4 | Negative |
| 5 | Admitted case | Male | 6 | 30-7-2019 | Acute tonsillitis | 2-8-2019 | Pharyngeal swab | 21.5 | Negative |
| 6 | Admitted case | Female | 8 | 27-7-2019 | Acute upper respiratory tract infection | 2-8-2019 | Pharyngeal swab | 23.1 | Negative |
| 7 | Admitted case | Female | 13 | 26-7-2019 | Acute bronchitis | 2-8-2019 | Pharyngeal swab | 28.8 | HCoV-229E (32.5) |
| 8 | Admitted case | Male | 14 | 25-7-2019 | Acute purulent tonsillitis and conjunctivitis | 2-8-2019 | Pharyngeal swab | 30.3 | Negative |
| 9 | Admitted case | Male | 9 | 28-7-2019 | Acute purulent tonsillitis | 2-8-2019 | Pharyngeal swab | 24.0 | Negative |
| 10 | Admitted case | Male | 3 | 30-7-2019 | Acute upper respiratory tract infection | 2-8-2019 | Pharyngeal swab | 21.1 | Negative |
| 11 | Admitted case | Male | 8 | 28-7-2019 | Acute pharyngitis | 2-8-2019 | Pharyngeal swab | 25.1 | Negative |
| 12 | Admitted case | Female | 10 | 22-7-2019 | Acute tonsillitis | 2-8-2019 | Pharyngeal swab | Negative | Negative |
| 13 | Admitted case | Female | 9 | 31-7-2019 | Acute upper respiratory tract infection | 2-8-2019 | Pharyngeal swab | 22.0 | HCoV-229E (33.3) |
| 14 | Admitted case | Male | 10 | 28-7-2019 | Acute tonsillitis | 2-8-2019 | Pharyngeal swab | 25.3 | Negative |
| 15 | Admitted case | Female | 9 | 26-7-2019 | PCF | 2-8-2019 | Pharyngeal swab | 28.4 | Negative |
| 16 | Admitted case | Female | 8 | 28-7-2019 | Acute tonsillitis | 2-8-2019 | Pharyngeal swab | 22.4 | Negative |
| 17 | Admitted case | Male | 9 | 29-7-2019 | Acute tonsillitis and gastroenteritis | 2-8-2019 | Pharyngeal swab | 21.1 | Negative |
| 18 | Admitted case | Male | 5 | 31-7-2019 | Acute tonsillitis | 2-8-2019 | Pharyngeal swab | 21.1 | Negative |
| 19 | Admitted case | Male | 13 | 29-7-2019 | Acute upper respiratory tract infection | 2-8-2019 | Pharyngeal swab | 21.7 | Negative |
| 20 | Admitted case | Male | 14 | 25-7-2019 | Acute purulent tonsillitis and conjunctivitis | 2-8-2019 | Conjunctival swab | 30.2 | Negative |
| 21 | Admitted case | Male | 3 | 31-7-2019 | Pneumonia | 3-8-2019 | Pharyngeal swab | 21.4 | Echo11 (33.1) |
| 22 | Outpatient | Male | 7 | 1-8-2019 | Acute tonsillitis | 3-8-2019 | Pharyngeal swab | 19.1 | Negative |
- Abbreviations: Ct, cycle threshold value; Echo11, echovirus 11; HCoV-229E, human coronaviruses 229E; PCF, pharyngoconjunctival fever.
In addition, the 22 samples from this PCF outbreak were also tested by PCR for 14 other pathogens known to cause acute respiratory diseases and intestinal infections. Two admitted patients (case 7: 13-year-old girl and case 13: 9-year-old girl) were confirmed to have coinfection with HCoV-229E and HAdV-7 (see Table 2). Two other patients (case 1: 4-year-old girl and case 21: 3-year-old boy) were found to have coinfection with human echovirus 11 and HAdV-7 (see Table 2).
The water in the two swimming pools is exchanged via a water circulation system every 2 days, disinfected with chlorine once every night, disinfected 30 min before opening of business, and self-measured for residual chlorine after disinfection. However, the disinfectant concentrations registered on July 22 and 23 (before the day of the outbreak) were insufficient in both the large and small pools (levels of free chlorine were below the recommended level of 0.3−0.5 mg/L according to the national guideline).
To determine whether the pools were a possible source of the outbreak, 3 water samples (2 from the large pool and 1 from the small pool) were collected from points 30 or 10 cm deep at different locations in the pools on July 31. The water samples were tested for adenovirus, aerobic bacterial count, and coliform bacteria, in accordance with the national guidelines of China for the microbiological examination of swimming pool water (PRC Ministry of Health). One of the water samples collected in the large swimming pool was positive for HAdV. Additionally, the bacteria count from the 3 water samples increased compared to the normal range. Partial hexon, penton, and fiber gene sequences obtained from the water samples were 100% identical to the sequences obtained from the swab samples of the outbreak patients, which were clustered with HAdV-7 within species B.
3.5 Nosocomial outbreak of HAdV-7 and HCoV-229E
A total of 24 PCF outbreak cases required admission to the paediatric ward from the Second People's Hospital of Jinyun county, which had a total of 6 doctors, 10 nurses, and 35 beds (see Supporting Information: Figure 4). On August 5, 1 nurse reported the onset of fever. In response this event, we conducted a nosocomial infection investigation. A total of 56 throat swabs were collected from 48 bedside caregivers, 2 nurses, 3 doctors, and 3 patients who were admitted on the same floor for other respiratory diseases. Amongst these samples, we identified 3 different nosocomial infections outbreaks in this paediatric ward.
First, 1 nurse who had cared for the PCF outbreak cases was positive for adenovirus genotype 7. Her full gene hexon, fiber, and penton of adenovirus was 100% identical to the PCF outbreak cases, see Figure 3. This suggested that her nosocomial adenovirus infection originated from the swimming pool-related HAdV-7 outbreak, see Figure 4.
Second, 5 nosocomial HCoV-229E infections were recorded, including 2 of the admitted cases (case 7 and case 13), 2 caregivers for case 7, and 1 doctor. Epidemiological investigation showed that case 7 and case 13 were admitted in different rooms on July 30 and August 1, respectively; neither they nor their caregivers had contact with each other. The four sequences for all 5 confirmed HCoV-229E cases were very close and generated independent clusters that shared 97.5%−99.9% identity, see Figure 5. Current epidemiological and microbiological results supported that HCoV-229E was most likely transmitted from the HCWs to the patients and their caregivers, see Figure 4.
Third, 1 nurse and 1 non-outbreak case were confirmed rhino positive, and the two sequences were identified to be the same. The potential transmission was determined to be via the patient to the HCWs. The nosocomial outbreak transmission route can be seen in Figure 4.
4 DISCUSSION
Swimming pool-related outbreaks of viral infection most often involve adenovirus, norovirus, hepatitis A virus, and echovirus.7, 20, 21 This article describes a swimming pool-associated outbreak of PCF affecting children in Lishui city of Zhejiang Province, China before the SARS-CoV-2 pandemic. Laboratory investigations demonstrated that HAdV-7-contaminated swimming pool water was the most likely source of infection. The admitted outbreak cases further led to nosocomial outbreak, caused by HAdV-7, HCoV-229E, and rhinovirus amongst the HCWs, bedside caregivers, and other non-outbreak cases in the pediatric ward.
Adenoviruses have been observed to be prevalent in rivers, coastal waters, swimming pools, and water supplies worldwide.22 Adenoviruses have also been detected in swimming pool water in the context of outbreaks of PCF.4 Adenovirus type 3, type 4, and type 7 have been most frequently found in these outbreaks.4, 7, 12, 23-25 For example, a swimming pool-associated outbreak of PCF in China was caused by human adenovirus type 4 in 2018 in Beijing. Another was caused by Ad3a2 in Shenzhen in 1997, and a third was caused by adenovirus serotype 7 in Hangzhou in 1991.4, 26 In this study, a total of 97 cases of swimming-related pharyngeal conjunctival fever outbreak were found amongst more than 700 swimmers. The outbreak was found in children ranging from 2 years to 15 years old; no infections were found in adult swimmers. The epidemic characteristics were apparent aggregation and fast spread, and the number of cases increased quickly within a short period of time.
The role of the pools and, more specifically, the pool water as a potential common source for these 97 cases of PCF was supported by epidemiologic and laboratory findings. The transmission of this virus can occur both through intake of swimming pool water or through direct contact between the water and the conjunctival mucosa or upper respiratory tract.21 There are several reasons to attribute this outbreak to adenovirus. First, adenoviruses are non-enveloped viruses that are unusually resistant to physical and chemical agents, which gives them prolonged survival capacity in the water, cases and environments.2, 27 Second, as a result of the training class being conducted in the outdoor swimming pool, the environmental temperature was especially high; the residual chlorine in the water became volatile much faster than usual. Likewise, the consistently inadequate amount of chlorine and the poor mixing of the chemical in this large pool were responsible for the survival of the virus in the water.12, 26 Third, the swimming classes comprised large numbers of students; each class had more than 100−200 people in one swimming pool, which resulted in frequent contact between the swimmers. More importantly, overcrowding of too many swimmers in the water and bathroom could have contributed to the infections and their spread. These potential reasons are consistent with other reports.4, 26
The incubation period of adenovirus is usually 2−14 days. Here we found the median incubation was 3.5−4 days for the PCF outbreak cases. HAdV-7 has been associated with severe or fatal disease in immunocompetent patients and tends to be more virulent than other serotypes.28-30 In this outbreak, all cases had a good outcome after 7-day hospital stays. The 97 cases described here demonstrated a remarkable heterogeneity of symptoms, with a high percentage of fever, sort throat, and conjunctivitis. These clinical features were as reported in other swimming pool-related outbreaks of adenovirus infection.4, 7 Notably, the liver function of all of these outbreak cases, which occurred before the SARS-CoV-2 pandemic, was normal, which was very different from the outbreak of hepatitis related to adenovirus in younger children in 2021.31 Moreover, the adenovirus sequences identified from 20 cases in this outbreak shared the same identity, which helps confirm that the common epidemiology of these cases described one disease entity.
To date, seven HCoVs have been identified, of which four (NL63, HKU1, OC43, and 229E) are known to cause mild to moderate upper-respiratory tract illness, presumably contributing to 15%−30% of cases of common colds in humans.32 According to the report by Feng et al.32 from China, of 28 369 acute lower respiratory infections, 10 387 (36.6%) were positive for at least one etiology. The most frequently detected virus was RSV (9.9%), followed by influenza (6.6%), parainfluenza virus (4.8%), adenovirus (3.4%), bocavirus (1.9%), human metapneumovirus (1.5%), and HCoV (1.4%). Codetections were found in 7.2% of patients. At the same time, HCoV-229E has been recently recognized as a potential agent of nosocomial viral respiratory infections.33
Previous studies conducted in neonatal and paediatric intensive care units demonstrated coronavirus nosocomial outbreaks.34, 35 The annual incidence of HCoV nosocomial infections in 1998 was 28.3%, which compared with 25% in 1995 and 11% in 1997, of which 52.4% of medical staff members were infected.35 In this outbreak, we found 1 medical staff member, 2 admitted cases related to the adenovirus outbreak as well as their bedside caregivers to be infected with HCoV-229E. Moreover, all of the HCoV cases were symptomatic at the time of infection. This field investigation suggested that the nosocomial infections with HCoV-229E occurred through medical staff-to-patient-to-bedside caregiver transmission.36 Some research underscored the importance of adenovirus as a cause of nosocomial infection in the immunosuppressed populations.37 Here, we confirmed that the adenovirus hospital acquired infections occurred in the normal immune population, where a nurse was confirmed to be adenovirus positive via outbreak cases to medical staff transmission. Both respiratory viruses are spread from person to person by aerosol transmission and via contaminated hands and fomites. This nosocomial outbreak occurred for several reasons. First, and most importantly, the cross-infection within the hospital occurred due to the fact that medical care workers did not perform the strict universal precautions related to hand washing. Second, HCoV can survive for as long as 6 days in suspension and 3 h after drying on surfaces. This relatively high stability and long-term survival of adenoviruses in the environment suggests that they could be a source of hospital-acquired infections if the surface disinfection and ventilation in the ward are inadequate.36 We also noted that the relatively long incubation period of PCF and the continued shedding of adenoviruses by children contribute to the probability that the virus might cross-transmit amongst nurses, who are most often in contact with these patients.12 Third, HCoV-229E presents mild symptoms and is not easily detected early, which induced the human-to-human transmission during the 2-week hospitalizations.38
The following preventive and control measures were proposed as a result of this PCF outbreak. First, the swimming pool was closed starting on August 1 and was thoroughly cleaned. The pool was disinfected to ensure adequate chlorination and filled with new water.12 Second, we conducted symptom surveillance of the swimmers and their contacts; throat swabs were collected and tested for the DNA of adenovirus.26 Third, physical examinations of healthy swimmers were performed before they were allowed back into the swimming water. We excluded all ill persons from the pool for at least 2 weeks after resolution of their symptoms.12 Fourth, health education and health promotion work were performed through communication with the parents, customers and the media to eliminate a public pandemic.26 At the same time, we developed a surveillance system for detecting respiratory tract infection cases in all hospitals in the local area. Finally, we monitored the number of swimming pool visitors in real time and conducted regular measurements of the residual chlorine concentrations at all swimming pools to help prevent transmission of adenoviruses and other microorganisms.4
The investigation of this PCF outbreak also allowed us to define measurements for prevention of nosocomial infections. Prevention of HCoV-229E transmission takes the same form as preventing human adenovirus in this ward. First amongst these precautions is the isolation of infected patients as soon as possible. Second, due to the viruses' possible prolonged survival on different surfaces and the decreased effectiveness of the usual antiseptic agents, surface disinfection as well as ventilation are key techniques to control nosocomial outbreaks. Third, the HCWs are expected to adhere to appropriate hand hygiene guidelines. Fourth, we implemented screening for carrier detection and acute respiratory symptom surveillance, which are generally recommended for infection control. Of all the measures that were implemented, those that seemed to be most effective were the observation of hand hygiene practices and the supervision of isolation precautions. After the final case was discharged from the ward at the end of August 2019, no new cases were reported.
In conclusion, this article described a swimming pool-associated outbreak of PCF that affected the children in Zhejiang Province, China. Laboratory investigations demonstrated that HAdV-7-contaminated swimming pool water was the outbreak source. The consistently inadequate amount of chlorine in the pools and the overcrowding of swimmers in the pools were responsible for this outbreak. This PCF outbreak further led to a nosocomial outbreak of adenovirus and HCoV-229E amongst the outbreak cases, HCWs and bedside caregivers in the paediatric ward. This hospital outbreak was associated with cross-infections during hospital stays of long duration and long virus survival time. These factors were exacerbated by flaws in personal protective equipment, especially the hand hygiene of the healthcare staff. Finally, all outbreak cases had fully recovered, regardless of a single infection (adenovirus or HCoV-229E) or coinfection of these two viruses simultaneously.
This study has several limitations. First, the number of cases was most likely under-reported, given that people with mild or no symptoms may not have sought medical care. Second, no environmental samples in the hospital were tested during this outbreak, which would have provided direct evidence of nosocomial infections. Third, adenovirus and HCoV-229E in China are not statutorily reportable infectious diseases, and these viruses have not been incorporated into the monitoring system. Therefore, the baseline levels of these two viruses were not known in this area. Fourth, we did not collect the blood or test the antibody against human adenovirus 7 in these monitored patients, which the mild and asymptomatic infections might not be detected during this adenovirus outbreak. Last, this is a case series of PCF and acute respiratory disease outbreaks, which lacks a comparison (control) group. We did not obtain the baseline of human adenovirus 7 and HCoV-229E in the healthy populations of this area.
AUTHOR CONTRIBUTIONS
All coauthors conceptualized the study. Shelan Liu drafted the first version of the manuscript and obtained funding for the study. Lihong Ying, Huan Li, An Zhu, Lin Li, and Haiying Zhu collected outbreak cases for the study. Jian Cai and Yumin Li verified and analyzed data for the study. Shengcao Dong, Ruyao Ying, Ta-Chien Chan, and Jinren Panc conducted the field investigation. Yin Chen made the detection of all samples and sequences analysis. All authors contributed data to the study and to the data interpretation, critically reviewed the manuscript, and approved the final manuscript for submission.
ACKNOWLEDGMENTS
We thank the local CDC and the local hospitals workers who has helped us to conduct the field investigation. We also thank Dr. King Wun Lau, from Shanghai High School International Division of China who helped us to search the published articles and to input the raw data.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
Open Research
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
Informed consent was obtained from all participants according to the study protocol. The protocol was reviewed and approved by the institutional review board of the Zhejiang Provincial Center for Disease Control and Prevention, China (Zhejiang Provincial CDC, Ethics approval number: 2019-012).
