Viral etiology of acute respiratory infections in hospitalized children in Novosibirsk City, Russia (2013 - 2017).

BACKGROUND
Acute respiratory infections (ARIs) cause a considerable morbidity and mortality worldwide especially in children. However, there are few studies of the etiological structure of ARIs in Russia. In this work, we analyzed the etiology of ARIs in children (0-15 years old) admitted to Novosibirsk Children's Municipal Clinical Hospital in 2013-2017.


METHODS
We tested nasal and throat swabs of 1560 children with upper or lower respiratory infection for main respiratory viruses (influenza viruses A and B, parainfluenza virus types 1-4, respiratory syncytial virus, metapneumovirus, four human coronaviruses, rhinovirus, adenovirus and bocavirus) using a RT-PCR Kit.


RESULTS
We detected 1128 (72.3%) samples were positive for at least one virus. The most frequently detected pathogens were respiratory syncytial virus (358/1560, 23.0%), influenza virus (344/1560, 22.1%), and rhinovirus (235/1560, 15.1%). Viral co-infections were found in 163 out of the 1128 (14.5%) positive samples. We detected significant decrease of the respiratory syncytial virus-infection incidence in children with increasing age, while the reverse relationship was observed for influenza viruses.


CONCLUSIONS
We evaluated the distribution of respiratory viruses in children with ARIs and showed the prevalence of respiratory syncytial virus and influenza virus in the etiological structure of infections. This study is important for the improvement and optimization of diagnostic tactics, control and prevention of the respiratory viral infections.


Results
We detected 1128 (72.3%) samples were positive for at least one virus. The most frequently detected pathogens were respiratory syncytial virus (358/1560, 23.0%), influenza virus (344/1560, 22.1%), and rhinovirus (235/1560, 15.1%). Viral co-infections were found in 163 out of the 1128 (14.5%) positive samples. We detected significant decrease of the respiratory syncytial virus-infection incidence in children with increasing age, while the reverse relationship was observed for influenza viruses.

Conclusions
We evaluated the distribution of respiratory viruses in children with ARIs and showed the prevalence of respiratory syncytial virus and influenza virus in the etiological structure of PLOS
Although the majority of ARIs are associated with respiratory viruses, antibiotics are often used in the clinical treatment of ARIs. As children with ARIs often have similar clinical symptoms, studying the clinical hallmarks of children with virus-related ARIs and the spectrum of respiratory viruses will help in developing more accurate treatments for ARIs [14]. Rapid diagnosis is important not only for timely treatment starting but also for the detection of a beginning influenza epidemic and the avoidance of unnecessary antibiotic treatment [15,16].
Western Siberia plays a key role in ecology, epizootiology and epidemiology of emerging diseases. This territory was involved in the circulation of A/H5N1 and A/H5N8 avian influenza viruses in 2005-2017 [17,18]. These viruses were spread by wild birds' migration. In Western Siberia migratory flyways of birds' wintering in different regions of the world: South East Asia, Central Asia, Middle East, Hindustan, Europe, and Africa-cross. For this reason, there is high probability of the emergence of humans and animal influenza viruses reassortants, as well as emergence of local outbreaks of human morbidity caused by uncommon variants of influenza viruses. Furthermore, Novosibirsk is the largest transport hub in this part of Russia with numerous international connections, that is important for the spread of ARIs [19,20].
The prevalence of respiratory viruses among children with ARIs differs in different regions and varies over time [21][22][23][24][25]. Thus, to better understand the epidemiology of Acute Respiratory Infections in Novosibirsk region, we investigated etiology of ARIs in children admitted to Novosibirsk Children's Municipal Clinical Hospital in 2013-2017.

Ethics issues
All aspects of the study were approved by the Ethics Committee of the Federal State Budgetary Institution "Research Center of Clinical and Experimental Medicine" (2013-23). Accordingly, written informed consent was obtained from parents prior to sample taking.

Patients and specimens
Children 0-15years of age were enrolled in the study within 3 days of illness onset and had at least two of the following symptoms: fever, sore throat, cough, rhinorrhea, nasal congestion, sputum, shortness of breath, lung auscultation abnormalities, tachypnea, and chest pain. Paired nasal and throat swabs were collected from each patient admitted to Novosibirsk Children's Municipal Clinical Hospital by hospital nurses. A total of 1560 samples collected during four epidemic seasons of 2013-2017 (October-April) were enrolled to the study. The epidemiological and clinical information including case history, symptoms, physical signs, and examination were included in a standardized questionnaire. The samples were placed immediately in viral transport medium (Eagle MEM, BSA and antibiotics) and stored at 4-8˚C prior transportation to the laboratory (not more than 24 hours). Detection of respiratory viruses was performed immediately after delivery to the laboratory. All specimens were tested for 15 common respiratory viruses, including influenza virus types A, B (IFVA and IFVB), human parainfluenza virus (HPIV) types 1-4, human respiratory syncytial virus (HRSV), human metapneumovirus (HMPV), four human coronaviruses (HCo), human rhinovirus (HRV), human adenovirus (HAdV) and human bocavirus (HBoV), using a real-time PCR assay-kit.

Nucleic acid extraction and reverse transcription
Viral nucleic acids were extracted from nasal and throat swabs using RNA/DNA extraction kit «RIBO-sorb» (Interlabservice, Russia) according to the manufacturer's instructions. The extracted viral nucleic acid was immediately used to perform the reaction of reverse transcription using commercial kit "REVERTA-L" (Interlabservice, Russia).

Statistical analysis
Two-tailed chi-square test (two by two table) was performed to compare the infection rates for respiratory viruses among different age groups. P-value <0.05 was considered to be statistically significant.

Patient characteristics
Totally, 1560 samples collected from patients with ARIs during the period from December 2013 to April 2017 were enrolled in the investigation. There were 824 males (52.8%) and 736 females (47.2%), and the patient's ages ranged from 3 months to 15 years. The most numerous age group (43.2%) was between 1 and 3 years old. The age distribution is shown in Table 1.
the incidence of respiratory viral infection between boys (601/824; 72.9%) and girls (527/736; 71.6%) (χ 2 = 0.345, p>0.05). The respiratory virus positive rate appeared to decrease with age. The lowest positive rate was observed in the age group of more than 6 years old (167/302; 55.3%), while the positive rates in age groups less than 6 years old were more than 70% (Table 1). Statistically significant difference was observed between the age group of more than 6 years old and other age groups (χ 2 = 54.113, p<0.01). No statistically significant difference was observed among the age groups less than 1 year old, 1-3 years, and 4-6 years.

Age and gender distribution
The data were analyzed with regard to the age and gender distribution of virus infection. No difference in the etiological distribution of viral pathogens was observed between males and females. Among detected respiratory viruses, HRV, HPIV, HCoV, and HAdV did not have statistically significant difference in the distribution among the different age groups. HMPV was detected in age group less than 1 year old much more frequently than in children 1-3 years old (χ 2 = 4.986, p<0.05) and 7-15 years old (χ 2 = 8.174, p<0.05). HBoV was significantly more frequently observed in children younger than 3 years old compare with children of 4-15 years old (χ 2 = 28.523, p<0.005). The incidence of HRSV decreased significantly with increasing age (p < 0.05) dropping from 35.1% in children younger than 1 year old to 5.3% in the school-age children (7-15 years old group), while the reverse relationship was observed for IFV (Fig 3).

Seasonal distribution
The data were analyzed with regard to the seasonality. Overall, virus detection positive rate did not differ significantly between seasons, ranging from 71.3% to 73.3%.

Multiple infections
Co-infections with two or more viruses were detected in 163 out of the 1128 (14.5%) positive samples (Table 1). Dual infections accounted for 11.4% (129/1128) of all positive samples and

Discussion
Acute respiratory infections are a serious health and economic problem, causing high morbidity and significant economic losses due to temporary disability payments and medical costs. Children are the most vulnerable group to the development of the disease. As several authors mentioned previously [26], ARIs can lead to serious diseases such as bronchiolitis and pneumonia and sometimes even cause death in infants and children worldwide. Nevertheless, most of the data on the epidemiological features and etiological structure of ARIs were from moredeveloped countries, and less is known about the etiology of ARIs in Russia. In the present study, we examined the viral etiology of ARIs in hospitalized children in Novosibirsk by Real-Time RT-PCR assay.
We detected at least one of the tested viruses in 72.3% (1128/1560) of the samples. In similar studies conducted in different regions of the world, the virus detection rate ranged from less than 50% to 75% [27][28][29]. For example, in studies performed in China, the proportion of positive samples in children with ARIs ranged from 37.6% to 78.7% [15,17,30,31]. The previous study of respiratory infections among children in European part of Russia revealed the 71.5% detection rate of respiratory viruses [32].
The percentage of the respiratory viruses' detection varies in different years in different regions, which may be associated with climatic and environmental factors, population distribution, economic status and diagnostic methods used [1]. In addition, seasonality of sampling can also lead to differences in the level of viruses' detection in various studies. Thus, Ju X. et al. carried out a study continuously from July 2011 through July 2013 and found 48.66% of samples to be positive for at least one respiratory virus [33]. In contrast, we collected samples only during epidemic seasons of ARIs, so the positivity rate in our study was considerably higher. Furthermore, acute respiratory infections can be caused by viruses that are not yet known, as well as bacteria that have not been included in these studies [14].
We found that the prevalence of respiratory viruses did not differ between boys and girls (72.9% and 71.6% respectively), which confirms the absence of a gender-based susceptibility to respiratory viral infections [14]. However, we observed a decrease in the respiratory viruses' detection rate with age, with the lowest detection rate in school-age children compared to children under 7 years of age (55.3% versus 76.4%, respectively). These data are consistent with findings obtained in other regions of Russia [32] and it may be due to decreased sensitivity to respiratory virus infections in older children.
Etiology of ARIs and the respiratory virus prevalence varies in different studies. In the United States, IFV, HRSV and HPIV were the most frequently detected [34]. Studies conducted in France have shown that metapneumovirus and respiratory syncytial virus are the most common [35]. IFV, HRSV and HRV were the most commonly detected respiratory viruses among children with ARIs in most regions of China [14]. The study of ARIs etiologic structure showed that HRSV, HRV, HPIV and IFV were registered significantly often among children in western part of Russia [32].
In our study the most common viruses detected were HRSV and IFV, followed by HRV. The age distribution of ARIs indicated that children under 3 years old were more likely to be infected by HRSV confirmed the importance of RSV in children with ARIs, especially in children < 4 years of age [36][37][38][39]. We observed a high rate of HRSV-detections in 2013-2014 (44.4%), while in 2016-2017 it was less than 10% which could be due to the annual variation in the circulation pattern of HRSV. Such year-to-year variation in the epidemiological patterns of viral infections confirms importance of the long-term study of the ARIs epidemiology [40].
Influenza virus is one of the major causative agents of respiratory disease in humans and may lead to serious illness [41]. In temperate countries influenza outbreaks usually occur during the winter season. Finally, in our study, IFV (344/1560, 22.1%) was the second most frequent detected pathogen with markable seasonality in winter months. During the 2013-2014 epidemic season, influenza virus detection rate was low-6.9% of all respiratory viruses, which was in accordance with the official influenza surveillance results of Ministry of Health, Russia [42]. In our study in 2014-2015, influenza viruses were detected in 17.2% among all respiratory viruses. Among those, the percentage of influenza A virus accounted for 73.3% and influenza B virus made up 26.7% of all detected influenza viruses. At the same time, in Russia influenza B was the main etiological agent, accounting for 50.6% of all detected influenza viruses. In 2015-2016, influenza A(H1N1)pdm09 virus was predominant in Russia, accounting for 79% of all influenza-positive samples consistent with our results (82%). In 2016-2017 influenza A(H3N2) virus was dominant in Russia detected in 61.3% of all influenza virus cases [42]. In our study influenza A and B viruses were detected with approximately the same frequency (48% and 52% respectively).
With the introduction of molecular techniques, the detection of multiple co-infecting viruses has become common, though the prevalence of each virus varies between studies [43]. In our study detection rate of viral co-infection was 14.5% among the positive samples. According to the previous reports, the incidence of viral co-infection in children can reach 30% [44]. Co-infection is most often found in children under the age of 5, due to the immaturity of the immune system and, thus, greater susceptibility to infection [45]. In our study, we significantly more frequently detected cases of simultaneous infection with two or more viruses in children under 7 years of age compared with children of school age (12.2% versus 3.4%), while there was no significant difference in the incidence of viral co-infection between the age groups of 0-1 year, 1-3 years and 4-6 years.

Conclusion
In conclusion, in our study we investigated the etiological structure of acute respiratory viral infections in hospitalized children in Novosibirsk, Russia, and evaluated age and seasonal distribution of the various respiratory viruses. Systematic monitoring of respiratory viruses is necessary to better understand the structure of respiratory infections. Such studies are important for the improvement and optimization of diagnostic tactics, as well as measures for the control and prevention of the respiratory viral infections.