Coinfections with Respiratory Pathogens among COVID-19 Patients in Korea

The detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in upper and lower respiratory specimens and coinfection with other respiratory pathogens in patients with coronavirus disease 2019 (COVID-19) was investigated. Study subjects (N = 342) were retrospectively enrolled after being confirmed as SARS-CoV-2 positive, and their nasopharyngeal swab (NPS), oropharyngeal swab (OPS), and sputum specimens were restored for SARS-CoV-2 retesting and respiratory pathogen detection. The majority of the subjects (96.5%, N = 330) were confirmed as SARS-CoV-2 positive using NPS/OPS specimens. Among the COVID-19 patients (N = 342), 7.9% (N = 27) and 0.9% (N = 3) were coinfected with respiratory viruses and Mycoplasma pneumoniae, respectively, yielding an 8.8% (N = 30) overall respiratory pathogen coinfection rate. Of the respiratory virus coinfection cases (N = 27), 92.6% (N = 25) were coinfected with a single respiratory virus and 7.4% (N = 2) with two viruses (metapneumovirus/adenovirus and rhinovirus/bocavirus). No triple coinfections of other respiratory viruses or bacteria with SARS-CoV-2 were detected. Respiratory viruses coinfected in the patients with COVID-19 were as follows: rhinovirus (N = 7, 2.1%), respiratory syncytial virus A and B (N = 6, 1.8%), non-SARS-CoV-2 coronaviruses (229E, NL63, and OC43, N = 5, 1.5%), metapneumovirus (N = 4, 1.2%), influenza A (N = 3, 0.9%), adenovirus (N = 3, 0.9%), and bocavirus (N = 1, 0.3%). In conclusion, the diagnostic value of utilizing NPS/OPS specimens is excellent, and, as the first report in Korea, coinfection with respiratory pathogens was detected at a rate of 8.8% in patients with COVID-19.


Introduction
COVID-19, caused by the infection with SARS-CoV-2, was identified from a cluster of pneumonia cases in Wuhan, China, in December 2019, and has spread to other countries since then, resulting in 42 million cases and more than 1.1 million deaths globally as of October 25, 2020 [1].
In Korea, a Chinese female from Wuhan, China, was identified as the first COVID-19 case on January 20, 2020, during a quarantine inspection at Incheon Airport [2]. e Korea government enforced SARS-CoV-2 testing not only for individuals with respiratory symptoms but also for those with histories of either international travel or contact with suspected cases, even when clinical symptoms were absent. Enforcement has caused exponential growth in real-time PCR (RT-PCR) tests to identify COVID-19 with emergency use authorized (EUA) in vitro diagnostic (IVD) assays [3].
For the rapid and accurate diagnosis of SARS-CoV-2 using IVD assays, selecting the appropriate types of specimens to be collected from patients at the right time is important [4]. Lower respiratory specimens such as bronchoalveolar lavage fluid and sputum have been recommended as the best clinical respiratory specimens for detecting SARS-CoV-2 [5,6]. Viral loads of SARS-CoV-2 were higher in nasal swabs than those in throat swabs collected from symptomatic COVID-19 patients [7]. e United States Centers for Disease Control and Prevention has deemed upper respiratory specimens as acceptable for initial diagnostic testing of SARS-CoV-2, and these include nasopharyngeal swabs (NPS), oropharyngeal swabs (OPS), nasal swabs, and saliva [8]. Simultaneous collection and placement of NPS and OPS in the universal transport medium (UTM) using two sets of swabs is recommended to increase sensitivity in real-time polymerase chain reaction (RT-PCR) assays in Korea, as long as the supply of flocked swabs is not limited [9]. We analyzed the number of patients with COVID-19 that were confirmed as positive using NPS/ OPS by measuring the detection rates of SARS-CoV-2 among the patients diagnosed during the screening process in the first outbreak of the disease in Korea. e IVD assays for detecting SARS-CoV-2 were not incorporated for testing coinfections with other respiratory pathogens at the time of screening COVID-19 suspects. Recently, coinfections in COVID- 19 have been reported not to neglect infections by other respiratory pathogens in addition to SARS-CoV-2 [10][11][12][13][14]. is suggests that simultaneous testing for coinfections between SARS-CoV-2 and other respiratory pathogens is required to provide a better patient treatment during the COVID-19 pandemic [15].
is study reports the simultaneous detection of SARS-CoV-2 and the coinfections with frequently reported respiratory viruses and bacteria causing atypical pneumonia (hereinafter atypical bacteria) using commercially available RT-PCR assays in upper (NPS and OPS) and lower (sputum) respiratory tract specimens collected from patients with confirmed cases of COVID-19.

Study Subjects.
e request forms of those with suspected COVID-19 cases that were submitted to Seegene Medical Foundation, a nonprofit independent clinical reference laboratory, for SARS-CoV-2 testing from February 9 to 23, 2020, were retrospectively reviewed. e request forms were generated and recorded by government healthcare centers, hospitals, or quarantine offices with personal and clinical information, such as name, address, age, birth date, gender, specimen type and collection date, respiratory symptoms, and foreign travel history, based on the guidelines of the Korea Disease Control and Prevention Agency (KDCA). Once request forms with respiratory tract specimens (NPS/OPS and/or sputum) collected from the suspects were submitted, the test results were reported within 24 h of turnaround time. Based on the guidelines for testing COVID-19 in Korea, both upper respiratory tract samples (NPS and OPS) were placed in a UTM to increase test sensitivity [9,18,19].
We reviewed the request forms and enrolled study subjects who were confirmed as COVID-19 positive. e respiratory specimens (UTM containing NPS/OPS as upper respiratory specimens and sputum as the lower respiratory specimen) were restored from storage freezers (−70°C) and applied to extract nucleic acids to reconfirm SARS-CoV-2 and detect respiratory pathogens simultaneously.

Specimen Preparation and Nucleic Acid Extraction.
e collected specimens were subjected to nucleic acid extraction as described below to detect SARS-CoV-2 and respiratory pathogens. A portion (200 µL) of the UTM containing NPS/OPS was directly applied to the nucleic acid extraction. e sputum specimens were inspected for viscosity, homogenized, and diluted with phosphate-buffered saline as recommended by the specimen treatment guidelines of the Korean Society for Laboratory Medicine [9,18]. en, 200 µL of diluted sputum was applied for extracting nucleic acids.
Nucleic acid extraction was performed by mixing the specimen (UTM or diluted sputum) and ready-to-use reagents from a commercial kit (MagNA Pure 96 DNA and Viral NA Small Volume Kit; Roche Applied Diagnostics, Mannheim, Germany) in MagNA Pure 96 instruments (Roche Diagnostics) according to the manufacturer's instructions.
e final elution volume of the nucleic acids extracted from each specimen was approximately 100 µL. e extracted nucleic acids were stored at −70°C and used to detect SARS-CoV-2 and respiratory pathogens.
e nucleic acids (8 µL) extracted from specimens (UTM or diluted sputum) were also mixed in STARlet, as described above for detecting SARS-CoV-2, with the One-step RT-PCR Mastermix (17 µL, consisting of primers, probes, Realtime One-step Enzyme, and buffer) from the commercial kits (Panel 1, 2, and 3 or PneumoBacter). RT-PCR was performed to simultaneously detect respiratory pathogens using CFX96 real-time PCR cyclers (Bio-Rad, Hercules, CA, USA) according to the manufacturer's instructions.
Interpretations for positive and negative detection of SARS-CoV-2 and respiratory pathogens by the multiplex RT-PCR assays were automatically determined using Seegene Viewer software provided by the manufacturer.

Statistical Analysis.
e statistical significance of the differences in the detection rates of respiratory pathogens was analyzed using the chi-square or Fisher's exact tests. e mean value comparisons of age and Ct values among the study subjects were evaluated using the Mann-Whitney test. Statistical analysis was conducted using GraphPad Prism software (version 5.0; GraphPad Software, Inc., La Jolla, CA, USA) and GraphPad InStat software (version 3.0). P values < 0.05 were considered statistically significant.

Ethics Statement.
We retrospectively analyzed the request forms and RT-PCR assay results of the study subjects, which were exempted from informed consent. is study was approved by the institutional review boards of Seegene Medical Foundation (SMF-IRB-2020-003) and Kyungpook National University Hospital (DGIRB202003001-HE002).

Study Subject Enrollment.
A total of 342 patients were enrolled in the study. After retrospectively reviewing the request forms and test results of suspects (N � 20,054) conducted for screening COVID-19 from February 7 to 23, 2020, we found that 404 patients (2.0%) were confirmed as SARS-CoV-2 positive based on the Allplex 2019-nCoV assay. Among those with positive test results, patients without personal information (N � 62; 15.3%), such as age, sex, and region of residence, were excluded. Finally, 342 subjects (84.7%) who were SARS-CoV-2 positive were enrolled. e mean age ± SEM (standard error of the mean) of the study subjects was 47.8 ± 0.9 years (standard deviation: 15.7 years; 95% CI: 46.1-59.4 years) with a median age of 51.0 years. e detection rates of SARS-CoV-2 by the age were 0.9% (N � 3) for those below 20 years of age, 29.5% (N � 101) for those aged 20-39 years, 44.4% (N � 152) for those aged 40-59 years, and 25.1% (N � 86) for those aged > 60 years.
is indicates that SARS-CoV-2 was mostly detected in those who were 40-59 years old. e ratio of males to females was 37.1% (N � 127) and 52.9% (N � 215), respectively. e residential areas of the study subjects were predominantly Kyungpook Province (N � 320, 93.6%) including Daegu city, where the first outbreak of COVID-19 occurred in February, followed by
Of the respiratory virus coinfection cases (N � 27), 25 of SARS-CoV-2-positive cases (92.6%) were coinfected with a single respiratory virus, and two cases (7.4%) (MPV/Adv  All cases of respiratory bacterial coinfections (N � 3; male : female � 2 : 1) were M. pneumoniae detected in both NPS/OPS and sputum specimens without any other atypical bacteria detected (Table 1). Additionally, no triple coinfections of other respiratory viruses or bacteria with SARS-CoV-2 were detected. e ratio of males to females in the coinfection cases was 4 : 6 ( Table 1), but no difference in coinfection by gender was observed (P � 0.3507), suggesting that both male and female confirmed as COVID-19 have the same level of susceptibility to coinfections with respiratory pathogens. e mean age of the coinfection cases (N � 30, 8.8%) with respiratory pathogens (both viruses and bacteria) was 36.4 ± 2.9 years, which was significantly less than the mean age (48.6 ± 0.9) of patients without coinfections (P < 0.0001). e mean ages for the viral (N � 27) and bacteria (N � 3) coinfection cases were 36.9 ± 3.1 years and 32.0 ± 6.0 years, respectively, which were not statistically significant (P � 0.6162) ( Table 1).

Discussion
In Korea, both NPS and OPS were placed in a UTM after being collected from suspects of COVID-19 for screening [9]. e UTM was transported to the laboratory for COVID-19 testing immediately after collection, with or without sputum samples. No other upper or lower respiratory tract samples were collected and tested. Among the suspects (N � 20,054) tested from February 9 to 23, 2020, during the first outbreak in Korea, the positive rate of SARS-CoV-2 was 2.0% (N � 404) using an RT-PCR assay, which was lower than that found (9.5%; N � 116) in a previous study among patients tested in the USA (N � 1,217) between March 3 and 25, 2020 [22]. In this study, a total of 342 patients were enrolled among those who tested positive (N � 404) for SARS-CoV-2.
Among the study subjects (N � 342), 96.5% (N � 330) were positive for SARS-CoV-2 through the RT-PCR assay on NPS/OPS specimens, regardless of sputum collection (Figure 1). Additionally, based on a comparison of the mean Ct values for the genes specific to SARS-CoV-2, higher viral loads of SARS-CoV-2 were observed in NPS/OPS than in sputum (Figure 2), which supports that rates of SARS-CoV-2 detection in NPS/OPS are excellent (Figure 1). is suggests that NPS/OPS is better than sputum for detecting SARS-CoV-2 in molecular diagnostic assays and that it is unnecessary to enforce sputum sampling, especially from patients who are unable to expectorate. However, in previous studies, sputum specimens were considered superior to nasal samples for the molecular detection of SARS-CoV-2. Pan et al. [6] measured the viral loads of SARS-CoV-2 in the throats, sputum, urine, and stool of two COVID-19 patients at the early stage of symptom onset. ey found that the viral loads were higher in sputum samples than in throat samples, especially in the early days after onset. Wang et al. [5] compared the detection rates of SARS-CoV-2 across multiple types of specimens collected from patients with COVID-19 and found that bronchoalveolar lavage fluid had the highest positive rate (93%), followed by sputum (72%), nasal swab (63%), and pharyngeal swabs (32%). Based on the literature published, bronchoalveolar lavage fluid was recommended as the best clinical specimen for detecting SARS-CoV-2 in patients, and sputum was the second most strongly recommended specimen, whereas pharyngeal swabs were given a moderate recommendation [23,24]. In contrast, our data support that pharyngeal swab samples were superior to sputum for detecting SARS-CoV-2 among the study subjects who underwent COVID-19 screening. A finding consistent with that of the current study was also observed in a case study of two Korean patients, which reported that viral loads of upper respiratory specimens (placing both NPS and OPS in the same UTM) were similar to or sometimes higher than those in sputum using RT-PCR detection [25]. e rate of coinfection in patients with COVID-19 (N � 342) with respiratory pathogens was 8.8% (N � 30) ( Table 1) and 7.9% (N � 27) and 0.9% (N � 3) having viral and atypical bacterial coinfections, respectively. e rate of viral coinfection has been previously reported to range from 2.0% to 19.8% in different countries [22,[26][27][28]. e coinfection rate with respiratory viruses was 2.0% (N � 39) among 1,996 COVID-19 patients hospitalized in New York City [26]. At a hospital in Shenzhen, China, six (3.2%) of 92 COVID-19 patients hospitalized were cases with viral coinfections [27]. Another study conducted in Wuhan, China, reported that, among patients with COVID-19 (N � 104), 5.8% (N � 6) had coinfections with non-SARS-CoV-2 coronaviruses (2.9%, N � 3), influenza A (2.9%, N � 3), and rhinovirus (1.9%, N � 2) [28]. In the study by Kim et al. [22] with patients in California, USA, 19.8% (N � 23) of SARS-CoV-2-positive patients (N � 116) were positive for other respiratory pathogens. As the first report on coinfections with respiratory pathogens among COVID-19 patients in Korea, especially in those from the first outbreak in February 2020, 7.9% were found to have viral coinfections, which is a rate comparable with those of previous reports.
is is consistent with the conclusions of a previous report by Zhu et al. [14], who suggested that both upper and lower respiratory tract specimens should be collected for detecting respiratory pathogen coinfections in COVID-19 patients, and these specimens should be considered for testing during the diagnosis and treatment of COVID-19.
Based on a meta-analysis by Lansbury et al. [12] conducted with the previously published studies on the types of respiratory viruses detected as coinfection agents among patients with SARS-CoV-2, the most common coinfecting virus was RSV, followed by Flu A, HRV, PIV, and other coronaviruses. A study in China, which was not included in the meta-analysis, consistently reported that RSV was the most commonly detected among coinfecting viruses in COVID-19 patients diagnosed between January 19 and February 26, 2020 [15]. However, in the USA, Kim et al. [22] and Richardson et al. [26] detected rhinovirus/enterovirus as the most common coinfection agents, followed by RSV, among the study subjects they enrolled in March 2020. In this study with Korean study subjects enrolled in February 2020, the most common respiratory virus detected as a coinfection agent was HRV (2.1%, N � 7; one double viral coinfection with HBoV (HRV/HBoV) and 6 single viral coinfection cases in SARS-CoV-2-positive patients) (Table 1). Six cases (N � 1.8%) of RSV and 5 cases (N � 1.5%) of single viral coinfections by non-SARS-CoV-2 coronaviruses (three cases of 229E, one case of NL63, and one case of OC43) were also observed ( Table 1).
Among the atypical bacteria tested, only M. pneumoniae was detected at the level of 0.9% (N � 3) of the study subjects (N � 342) ( Table 1). In some other studies, however, C. pneumoniae, another atypical bacterium, was identified as another coinfection agent in patients with COVID-19 in addition to M. pneumoniae [14,15,26]. e coinfection rates in the above studies for M. pneumoniae and C. pneumoniae ranged from 1.6% to 4.8% and 2.5%-5.2%, respectively. Lansbury et al. [12] emphasized, in their meta-analysis study, that M. pneumoniae was the most common bacteria detected in patients with COVID-19 who had respiratory bacterial coinfections, followed by P. aeruginosa, H. influenza, Klebsiella pneumoniae, and Chlamydophila spp., suggesting that in addition to coinfection by atypical bacteria, other respiratory bacteria were also the candidates of coinfections in COVID-19 patients.
Unlike Kim et al. [22] who reported no difference in age between coinfection and SARS-CoV-2 only (non-coinfection) groups, the mean ages of these groups were significantly different in this study (36.4 ± 2.9 vs. 48.6 ± 0.9 years old, respectively; P < 0.0001), where most coinfections with respiratory pathogens were detected at ages between 20 and 39 (N � 16, 4.7%) (Figure 3). is suggests that young individuals with COVID-19 are more susceptible to respiratory pathogen coinfections. e viral loads of SARS-CoV-2 in COVID-19 patients with coinfection and patients without coinfections were not significantly different as Ct values for the target genes (E, RdRP, and N) were statistically the same between these patient groups when the Allplex 2019-nCoV assay was applied (P ≥ 0.5096).
is suggests that the viral load of SARS-CoV-2 is not responsible for causing respiratory pathogen coinfections. e current study had several limitations. First, the effects of coinfections on the treatment outcomes of COVID-19 patients are unknown. e study subjects were retrospectively enrolled during the COVID-19 screening process in the region of the first outbreak in Korea. us, the scope of this study was to investigate the coinfection rates of respiratory pathogens among patients with COVID-19 who were not hospitalized, but diagnosed as positive for SARS-CoV-2 during the screening process. Second, the results of the study cannot represent the coinfection rates in winter when respiratory pathogens such as influenza are dominant. e study subjects were enrolled only in February; therefore, it would be interesting to investigate the rates of coinfection with respiratory pathogens across the entire winter season, especially from November to March. ird, the commercial multiplex RT-PCR assays (Allplex TM Respiratory Panel 1, 2, and 3) that we employed were able to detect the subtypes found in Flu, RSV, PIV, and non-SARS-CoV-2 coronaviruses as described in Materials and Methods but unable to distinguish the HRV subtypes (A, B, and C), which were previously included in a report by Kuypers et al. [30]. Finally, this study did not investigate coinfections in COVID-19 patients with either opportunistic pathogens such as Streptococcus and Actinomyces or fungi such as Aspergillus and Candida, as Chen et al. [10,31] described. e broad range of coinfecting agents in COVID-19 patients should be investigated.

Data Availability
e data used to support the findings of this study are available from the corresponding author (Dr. Nackmoon Sung (paratb@gmail.com)) upon request. Requests will be reviewed by coauthors and decided to provide.

Disclosure
Present address of Kyoung Ho Roh: Department of Laboratory Medicine, National Health Insurance Service Ilsan Hospital, Goyang 10444, Gyeonggi, Rep. of Korea.