Detection and characterization of respiratory viruses causing acute respiratory illness and asthma exacerbation in children during three different seasons (2011–2014) in Mexico City

Background Viral infections play a significant role in causing acute respiratory infections (ARIs) and exacerbations of chronic diseases. Acute respiratory infections are now the leading cause of mortality in children worldwide, especially in developing countries. Recently, human rhinovirus (HRV) infection has been emerged as an important cause of pneumonia and asthma exacerbation. Objectives To determine the role of several viral agents principally, respiratory syncytial virus, and HRV in children with ARIs and their relationship with asthma exacerbation and pneumonia. Methods Between October 2011 and March 2014, 432 nasopharyngeal samples of children <15 years of age with ARI hospitalized at a referral hospital for respiratory diseases were tested for the presence of respiratory viruses using a multiplex RT-qPCR. Clinical, epidemiological, and demographic data were collected and associated with symptomatology and viral infections. Results Viral infections were detected in at least 59·7% of the enrolled patients, with HRV (26·6%) being the most frequently detected. HRV infections were associated with clinical features of asthma and difficulty in breathing such as wheezing (P = 0·0003), supraesternal (P = 0·046), and xiphoid retraction (P = 0·030). HRV subtype C (HRV-C) infections were associated with asthma (P = 0·02). Conclusions Human rhinovirus was the virus most commonly detected in pediatric patients with ARI. There is also an association of HRV-C infection with asthma exacerbation, emphasizing the relevance of this virus in severe pediatric respiratory disease.


Background
Acute respiratory infections (ARIs) are the leading cause of mortality in children worldwide, particularly in developing countries. It represents an important public health problem in early development, with high mortality and morbidity among children under five years of age. 1 ARIs are classified as upper respiratory tract infections or lower respiratory tract infections (LRTIs) depending on the airways predominately involved. 2 Although ARIs can be caused by bacteria or fungi, viral infections are responsible for most of them. Several viruses have been consistently identified during ARIs: influenza virus, human parainfluenza virus (HPIV), human rhinovirus (HRV), adenovirus (ADV), coronavirus (HCoV), enterovirus, human metapneumovirus (HMPV), and respiratory syncytial virus (RSV). 3 Moreover, viral infections are one of the many risk factors associated with wheezing illnesses and exacerbation of respiratory diseases in children of all ages. 4 HRV has been associated with these exacerbations, including cough, wheezing, shortness of breath, oxygen use, and length of hospital stay. 5,6 In addition, asthma inception and exacerbation had been associated with HRV [7][8][9] and HMPV infection, 10 with some reports estimating that approximately 60% of cases are associated with HRV infection. 11 Human rhinovirus have been classified into two genetic species: HRV-A (including 76 serotypes) and HRV-B (including 25 serotypes). However, recently, HRV-C has been included. HRV-A and HRV-B are associated with the common cold, whereas the role of HRV-C is relatively unknown, but recent reports suggest that HRV-Cs may be more pathogenic than other HRVs. [12][13][14] Virus identification and molecular characterization is fundamental for epidemiological surveillance and control, but also for diagnostic purposes that may lead to specific therapy and an adequate response to treatment because clinical manifestations of virus and bacteria associated with ARI overlap considerably except in epidemic situations. 15 The aim of this study was to determine the association of each type of respiratory viruses with acute hypoxemic respiratory disease mainly asthma acute exacerbation or pneumonia in children admitted to a reference respiratory center in Mexico City during three different seasons.

Study population and clinical data
All samples used in this study were collected during the monitoring of respiratory pathogens in children under 15 years of age who presented clinical signs of acute respiratory infection (ARI) at the Pediatrics' Unit of the National Institute of Respiratory Diseases (INER) in Mexico City. The study had ethical approval from the Research and Ethics Committee of the INER, and informed consent was obtained from all legal guardians. From October 2011 through March 2014, a total of 432 nasopharyngeal swabs (NPS) were collected from hospitalized patients with ARI, all with shortness of breath or hypoxemia (oxygen saturation < 90% on room air). Demographic data and clinical symptoms of the enrolled patients were obtained from clinical charts at the time of the study.
From a clinical point of view, patients were further classified in two groups: (i) pneumonia (all ARI including opacities in the chest roentgenogram) and (ii) those with wheezing and/or asthma and no chest roentgenogram opacities.
Sample collection, nucleic acids extraction, and multiplex RT-qPCR for respiratory virus detection Nasopharyngeal swabs were collected and total nucleic acids were extracted from 200 ll samples using a high-throughput automated extraction system (MagNa Pure; Roche, Indianapolis, IN, USA).
Multiplex reverse transcription-polymerase chain reaction (RT-qPCR) was standardized to detect the main respiratory viruses by high-throughput gene expression analysis using 48.48 dynamic array integrated fluidics chips on the BioMark platform (San Francisco, CA, USA). Details of multiplex RT-qPCR are provided in Supplemental Study design.

Statistical analysis
Data were analyzed in R version 2.15.3 (Vienna, Austria) and SPSS version 20 (Armonk, NY, USA) and STATA 11 (College Station, TX, USA). Differences with P value <0Á05 were considered significant. For univariate comparisons between groups in continuous variables, we used the Mann-Whitney U-test for independent samples and the chi-squared test for categorical outcomes. To estimate the significance and magnitude of association between risk factors and viral infection, odds ratios (ORs) and 95% confidence intervals (95% CIs) were calculated using logistic regression. Multivariate regression was adjusted by age, comorbidities such as gastroesophageal reflux, and bacterial and viral coinfection.

Demographic characteristics
Between October 2011 and March 2014, a total of 432 hospitalized children with ARI, most of them with community-acquired pneumonia, influenza-like illness, or asthma exacerbation, were enrolled in this study ( Table 1). The median age of study participants was 36 months, and 51Á4% participants were male.

Seasonal distribution
Viral infections were highly active during winter seasons and almost absent in the middle months of the year. HRV was detected almost throughout the whole year with a diminishing in the number of cases during the middle months of the year and a maximum peak during winter season ( Figure 1A). RSV-A and RSV-B showed similar patterns as HRV, with some cases during spring season ( Figure 1B), while other viruses such as influenza virus A (IFV-A), ADV, HMPV, and HPIV were less frequent; however, while ADV, HMPV, and HPIV infections were detected during spring and autumn seasons, IFV-A showed a clear occurrence during winter, with an increase in the number of cases in 2011 and 2013 winter seasons, but almost absent in winter 2012 ( Figure 1C).
The 115 samples positive to HRV infection were typified, and 49Á4% of the samples were classified as HRV-C. To determine the influence of comorbidities and other factors such as age, bacterial, and viral coinfections, multivariate logistic regression was realized. Remarkably, the relationship between HRV-C and asthma is maintained (P = 0Á02; OR=2Á53 [95% CI: 1Á14-5Á59]). The rest of types and subtypes of respiratory viruses and comorbidities such as gastroesophageal reflux were not associated with asthma either in the univariate analysis or in the adjusted analysis (data not shown).

Molecular detection and characterization of viral pathogens present in respiratory diseases is one of the most important
In the present work, we were able to evaluate the role of viral agents in ARIs among a hospitalized pediatric population, finding that 59Á7% of the patients tested had at least one viral infection. This result is similar to a previous etiological report made in Mexico City, where 63% of the hospitalized children were positive to viral mono-infection. 16 On average, viruses are usually detected in approximately 30-50% of hospitalized children with severe respiratory infections in developing countries. [17][18][19] Contrary to previous reports, we did not find a relationship between RSV infection and wheezing or asthma diseases as previously has been stated 20 ; however, we found a significant relationship between RSV infection and severity symptoms such as diarrhea, hyporexia, and crackles, and also an association with pneumonia diagnosis. Our study provides additional evidences against the belief that RSV is the most important virus associated with asthma exacerbation in children.
We found a low proportion (3Á5%) of HMPV infection in our study, consistent with other result found in Mexico City. 16 However, unlike other reports where HMPV infections are characterized by more severe disease or symptoms indistinguishable from those of an RSV infection, 21 we could not find a relationship between HMPV infection and symptoms or laboratory parameters with exception of fever; however, we found a higher hospitalization length in children with HMPV infection, suggesting an important role of the virus in disease severity that needs to be confirmed.
HRV was the most frequent viral pathogen found in NPS of hospitalized children with respiratory symptoms (27Á3%). This is a higher incidence compared to what was recently shown in a study in Mexico City where a range from 14% to 17% of the hospitalized children tested positive for HRV infection, 16,22 . Recently has been reported an increase in ARIs caused by HRV 23 suggesting an emergent clinical importance in respiratory diseases principally in children.
We found a relationship between HRV infections with respiratory distress and asthma diagnosis features, differing from a previous report, 22 which suggests that HRV is associated only with very mild and mild illnesses. However, in our work and in accordance with previous studies, 8,24,25 we reported a high proportion (75Á7%) of children who presented wheeze and HRV infection. Moreover, HRV subtype C showed relationship between asthma diagnosis but no HRV-A or HRV-B. These data support previous reports that suggest an important role of HRV-C in LRTIs and the relevance of the HRV genotype as determinant of disease severity. 26 This is the most extensive analysis made that was able to detect the respiratory viruses present in the upper respiratory tract of hospitalized children with ARIs in Mexico City, particularly assessing the role of rhinovirus infection in asthma illnesses. As HRVs are often considered to be of little health impact and clinical significance, this type of studies is essential to know the genomics, epidemiology, and clinical impact of the HRV-C strains in order to be capable of confronting and even predicting future outbreaks and be able to respond in a fast and effective way, especially in developing countries.

Funding
This work was supported by Instituto de Ciencia y Tecnolog ıa DF (ICYT DF) grant PICSA12-64. VAHH was supported by a scholarship from CONACYT, Mexico.

Conflict of interests
None.

Ethical approval
The Science and Bioethics Committee of the INER revised and approved the protocol and the consent procedure (B2613). For all pediatric patients, the corresponding legal guardians provided written informed consent.