Characterization of rhinovirus/enterovirus complex infection in children living at high altitudes: A cross sectional study

Background: Respiratory tract infections caused by the rhinovirus/enterovirus (RV/EV) complex have traditionally been considered to be minor, self-limited infections in children, with few complications. There are no previous studies of patients living at high altitudes that characterize severe cases of this infection. Methods: This was a cross-sectional study including patients from 1 month to 18 years old who had been hospitalized for acute respiratory tract infections between October 2015 and December 2019, and had had a viral panel with RT-PCR during their hospitalization. Results: During the study period, 645 RT-PCRs were performed, with the two main etiological agents identied being RV/EV (224) and respiratory syncytial virus (68). The median age of patients with RV/EV complex was 27 months (IQR: 8-70), 55.8% were boys and the average length of hospital stay was 12 days (IQR: 6-24). Severe RV/EV complex infections required more transfers to intensive care (11% vs 47%), showed more viral coinfection (OR: 2.13,95%, 95%CI: 1.42-4.64) and had less bacterial coinfection (OR: 0.55, 95%CI: 0.31-0.98) than RSV infections, with no difference in mortality ( 2.4% vs. 2.1%, P:0.09). Post-transplant patients (OR: 3.35, 95%CI: 1.10-11.34) and those with comorbidities (OR: 3.97, 95%CI: 2.23-7.08) had the highest risk of RV/EV infection. The RV/EV group had a higher risk of presenting acute respiratory distress syndrome (ARDS) (OR: 3.6, 95%CI: 1.07-12:18), especially in premature infants (p: 0.05; exp (B), 2.99; 95%CI= 1.01-8.82), those with heart disease (p: 0.047; exp(B), 2.99; 95%CI = 1.01-8.82) and those with inborn errors of metabolism (p: 0.032; exp


Background
Acute respiratory infection (ARI) causes about four million deaths per year in children under ve years of age. It is one of the main causes of hospitalization and admission to intensive care in pediatrics.
According to the World Health Organization (WHO), it is estimated that between 2-3% of children under two years of age in low-and middle-income countries have had severe pneumonia requiring hospitalization, and ARI mortality rates n these countries range from 60 to 100 cases per 1,000 children under the age of ve [1].
In children under two years of age, viruses are the main etiological agents of ARI. Respiratory syncytial virus (RSV) has been considered to be the most common germ and it usually affects children who have risk factors such as prematurity, heart disease or bronchopulmonary dysplasia [2]. Other viruses such as the rhinovirus / enterovirus complex (RV / EV) are thought to commonly cause uncomplicated acute upper respiratory infections that are usually self-limiting and have a benign course. It has recently been noted that this virus, from the genus Enterovirus and family Picornaviridae, can invade the lower respiratory tract and cause complicated disease in children [3]. Additionally, this virus has been associated with a higher frequency of bacterial coinfection and the presence of complicated pneumonia, and it accounts for up to 14% of acute respiratory distress syndrome (ARDS) in children living in highincome countries.
In low and middle-income countries, there is no registry for RV / EV complex infection, its natural course or associated risk factors. In these countries, the di culty in accessing health services, chronic noncommunicable diseases, and social inequality, frequently make ARI one of the main causes of morbidity and mortality [1]. Moreover, many of these patients live at moderate or high altitudes, which leads to the development of respiratory compensatory mechanisms to tolerate relative hypoxemia [4].
These children tend to be more prone to frequent upper and lower respiratory tract problems. Therefore, in this study we intend to describe the main characteristics of hospitalized patients with ARI due to the RV / EV complex, the risk factors associated with severe infection and their clinical course compared to RSV infections, in children who live at moderate altitudes.

Methods
A cross-sectional study was carried out in children from 1 month to 18 years old hospitalized at the Fundación Cardioinfantil-Instituto de Cardiología, a tertiary university hospital located in Bogotá, Colombia. Patients who underwent multiple RT-PCR tests (FilmArray® BioMériux) and who were hospitalized for ARI on the regular hospital oor or in pediatric intensive care between October 2015 and December 2019 were included. According to institutional protocol, children with ARI who are going to be hospitalized should have an RT-PCR taken.
Post-transplant surgery patients who underwent multiple respiratory RT-PCR tests without having symptoms or a diagnosis of ARI were excluded. This study was done in accordance with the Helsinki declaration and approved by the institutional ethics committee and was registered with number PM-35-2020. The informed consent was signed by the parents or legal guardian in all the participants.
The information was taken from the electronic charts of the hospitalized children who had undergone multiple RT-PCRs. The analyte used in this technique for virus detection was Type IA10-2003 Zeptometrix 0810161CF. All the information was entered in a database created for this study, to which only the principal investigators had access.
The multiplex RT-PCR respiratory assay was ordered for patients with upper and lower respiratory symptoms at the discretion of the attending physician. A respiratory therapy specialist took samples from nasopharyngeal aspirates, and they were processed within 30 minutes of collection, according to the institutional standard for sample handling. The RT-PCR detects 17 viruses and 3 bacteria, speci cally: adenovirus, coronavirus OC43, coronavirus NL63, metapneumovirus, HRV / HEV, in uenza A, in uenza A / H1, in uenza A / H1-2009, in uenza A / H3, in uenza B, parain uenza 1, parain uenza 2, parain uenza 3, parain uenza 4, RSV, Bordetella pertussis, chlamydia and mycoplasma; with a sensitivity and speci city of 95% and 99%, respectively [5], and is currently the gold standard for detecting these microorganisms [6].
Clinical behavior was evaluated in terms of the main viral respiratory infection complications, such as transfer to the intensive care unit, need for ventilatory support and mechanical ventilation, and length of hospital stay. The clinical manifestations were classi ed into different diagnoses such as rhinopharyngitis, laryngitis, croup, bronchiolitis, tracheitis, pneumonia or ARDS, according to the criteria of the attending physician. The data on RSV was recorded during the same time period.
The outcome of unsatisfactory clinical behavior during RV / EV infection was de ned as the requirement for an intensive care unit and/or mechanical ventilation, and the length of ventilatory support and hospital stay in pediatric patients hospitalized for ARI. Inpatient deaths from any cause in children with a positive RV / EV test were considered.
Severe ARI was de ned as the use of oxygen with an inspired oxygen fraction (FiO2) greater than 40% and the need for non-invasive mechanical ventilation (e.g., high-ow nasal cannula), or invasive mechanical ventilation, including hemodynamic instability requiring vasoactive support. Patients in whom a typi ed viral infection was identi ed 48 hours after admission, and who did not have respiratory symptoms on admission, were considered to have a nosocomial infection. Viral coinfection was established if two or more respiratory viruses were isolated in the RT-PCR assay, and bacterial coinfection was determined if blood cultures and / or orotracheal secretion cultures were positive and / or procalcitonin was > 0.5 μg / L.

Statistical analysis
Descriptive statistics were used for demographic variables and a measure of central tendency and dispersion for continuous variables, according to the distribution of the variable de ned by the Shapiro Wilk normality test. Absolute and relative frequencies were described for the qualitative variables. For the factors related to disease severity, a bivariate analysis was performed using Student's t test for independent samples when the variable was quantitative and parametric with a normal distribution. A Mann-Whitney U test was used for non-normal distribution.
An exploratory bivariate analysis was performed to compare the two agents, RV /EV and RSV.
Hospitalized and intensive care patients were analyzed. For the multivariate analysis, binary logistic regression was performed in order to control for confounding factors, mainly the severity of the disease, evaluated according to the PIM2 scale [7]. The aim of this analysis was to establish the risk factors associated with severe RV / EV infection and determine their association with the outcomes of interest. Statistical analysis was performed using SPSS 25 (IBM 16), and a P = <0.05 was considered statistically signi cant.
We tried to control biases with various strategies. To control for information bias, an exhaustive search of the electronic chart was carried out; if information was lacking, other sources of information were searched, such as the data recorded in the clinical laboratory history or the nurses' notes. The central laboratory's RT-PCR sample collection and processing procedures were standardized. This research was carried out in a single university center that cares for all kinds of pathologies in the emergency room, due to its complexity and size.

Results
RT-PCR was performed on 645 patients in the study period, of whom 224 were positive for RV / EV and 68 were positive for RSV. The demographic characteristics of the patients are detailed in Table 1. The patients with RV / EV complex infection had a median age of 27 months (IQR: 8-70) and were predominantly male (54.8%). This infection was seen more frequently in patients with solid organ transplantation (10.3%) and prematurity (15.1%) (Supplement 1). Hypoxemic respiratory failure 1 (7.7) Severe sequelae -DNR 1 (7.7) Bradycardia and asystole 1 (7.7) Refractory shock 5 (38.5) Most of the patients with RV / EV complex infection had a normal weight (58.6%); nevertheless, 19.2% of the cases had severe malnutrition, and obesity was observed in 11 patients (3.8%). The predominant symptoms were cough (46.2%) and dyspnea (37.7%), with some gastrointestinal (15.4%) and neurological symptoms (4.1%) ( Table 1).
The nal diagnoses of RV / EV and RSV infection patients were pneumonia in 114 cases (39%) and bronchiolitis in 59 cases (20.2%), 35 (12%) of whom developed ARDS. Rhinopharyngitis, asthmatic crisis, and recurrent wheezing occurred in 30 children (10.3%); croup and tracheitis in 10 patients (12%), and severe respiratory infection criteria were identi ed in more than 50% of the study population, for a total of 162 cases (55.5%). The main viruses found to cause viral coinfection with the RV / EV complex are described in Table 2 and were found in 64 cases (Table 2). The patients with RV / who had to be transferred to the PICU were younger than those who did not need to be transferred [13 months (IQR: 4-49) vs 30 months (IQR: 10-65) (P = 0.004)]. (Table 3). The predominant symptoms of the patients transferred to the PICU were more severe from the beginning.
More dyspnea was observed at the time of emergency admission among those who were transferred to the PICU vs those who continued in general hospitalization (64.5% vs 24.4%; P < 0.001). The most common comorbidities in the population admitted to the PICU were heart disease (24.4%), liver disease (15.1%), prematurity (15.1%), and bronchopulmonary dysplasia (13.4%). Other comorbidities included primary immunode ciency (12.8%), transplantation (9.9%), metabolic disease (8.7%), and kidney disease (8.1%). Some type of neoplasm was present in 12 patients (7.0%) in the PICU vs 30 patients (25%) not in the PICU (P = < 0.001). Pneumonia was the predominant cause of admission to the PICU, with 72 patients (41.9%), followed by bronchiolitis with 18%, and asthmatic crisis with 11%. Severe ARI was identi ed in 86.6% of the cases that were transferred to the PICU, but only in 10.8% of those who were not transferred (P = < 0.001). In patients with a solid organ transplant, RV / EV infection was more frequently observed than RSV, 10.3% vs 1.0% (OR: 3.35 95%CI: 1-11.34) (P = 0.04). Viral coinfection was identi ed in 55 patients (18.8%) with RV / EV infection vs 9 patients (3.1%) with RSV (OR: 2.13 95%CI: 1-4.58) (P = 0.04). Bacterial coinfection was observed in 19.5% of patients with RV / EV infection vs 8.9% with RSV (OR: 0.55 95%CI: 0.31-0.98) (P = 0.04), as illustrated in Table 4. The need for PICU was more frequently seen in the group with RV / EV infection, with 47.3% of these patients vs 11.6% in RSV (P = 0.09). Acute respiratory distress syndrome developed more frequently in the group with RV / EV infection than in children with RSV infection (11% vs 1%; P = 0.003). When different confounding factors (mainly the severity of the disease measured by the PIM2 scale) were controlled for, ARDS in patients infected by RV / EV was observed more frequently in those with heart disease, premature infants and those with inborn errors of metabolism (Table 5).

Discussion
In this study of children with i RV / EV complex infection living at high altitudes, 47.3% were found to have severe disease that required transfer to pediatric intensive care. Children with comorbidities such as prematurity, heart disease and inborn errors of metabolism were especially prone to severe disease. The clinical course frequently includes ARDS and a mortality similar to RSV, which has traditionally been responsible for a signi cant burden of disease [8].
The study population's median age was much lower than that reported in the literature in high-income countries. Sapeder et al., in a retrospective cohort study carried out in Baltimore with 519 patients, showed a median age of 2.7 years (very similar to what we found in our group) for those hospitalized for RV / EV with severe infection [9]. In a study in New York with 155 children, Smith et al. found a median age of 4 years [5].
Vásconez-García et al. have linked nutritional risk and the presence of worse outcomes and high mortality during severe viral respiratory infection in Latin American individuals with comorbidities [10][11]. However, in our population, malnutrition was observed in only 41%, a nding consistent with the dietary problems of low-and middle-income countries where the prevalence of chronic noncommunicable diseases such as overweight is progressively increasing [12][13][14][15] Traditionally, severe ARI has been described in patients with comorbidities. Tijerina et al., in a crosssectional study in Mexico with 295 patients, described a higher rate of hospitalization in patients with prematurity [16] and bronchopulmonary dysplasia, relating it to the lack of coverage in vaccination, abandonment of breastfeeding, poverty, social commitment and the socioeconomic level [11] [17]. In our population, we found that children hospitalized for RV / EV presented signi cant comorbidities and frequently needed to be transferred to intensive care. As mentioned previously, this virus has generally been thought to cause self-limited upper respiratory infection in most cases [18], but studies like ours show that, in risk groups, this virus can have a similar behavior in terms of severity to others previously described, such as RSV.
We found that younger patients infected with RV / EV required transfer to intensive care more often and had a longer hospital stay. Roeleveld et al., in high-income countries, found that RV / EV infection, regardless of comorbidities, is not associated with a longer length of hospital stay [19]. The difference in our population can be explained by the greater di culty in accessing health services in low-and middleincome countries, which often leads to late consultations. This is related to the signi cant frequency of ARDS seen in our series in children with RV / VE, which is higher than that observed with RSV.
Traditionally, RSV causes signi cant morbidity / mortality in children under two years of age. The study by Bianchini et al. shows that this virus is the second leading cause of infant mortality and the leading cause of lower respiratory tract infections in children worldwide, with clinical manifestations and severe complications including ARDS [20]. Having seen these results with RV / EV infection, it is striking that this germ had a more aggressive behavior, with a higher frequency of ARDS and viral coinfection and a greater need for respiratory support. This has not been described before in middle-income countries. We consider that it is important to look further into the pathophysiological mechanisms that explain this more severe behavior of RV / EV.
Messacar et al. found that pediatric patients admitted with the RV / EV complex had the same probability of requiring admission to the PICU or mechanical ventilation as children admitted with other respiratory viruses, historically considered to be more severe, such as in uenza [21] [22]. In a previous publication by our group, the RV / EV complex was described as the most frequent etiological agent (30%), followed by RSV (19%), parain uenza (7.4%) and adenovirus (5.7%). In uenza only accounted for 1.2% of the total RT-PCRs taken [23]. In his series, Messacar found that only 18% of the patients with RV / EV required transfer to intensive care, while 60% of our patients were transferred [21]. Socioeconomic factors, comorbidities and limitation of access to health services can explain this situation.
Despite the fact that it has been considered to be a minor infection, the risk of developing ARDS was more frequent with the RV / EV complex than with RSV infection in our population. This is an important fact, and we must consider that children admitted to intensive care with RV / EV infection may have an unsatisfactory course and require more frequent and intense support than those with other viruses such as RSV [5]. Comorbidities such as prematurity, heart disease and inborn errors of metabolism may partially explain this evolution, but this implies that, in these risk groups, RV / EV infection cannot be considered to be a mild infection, and its possible complications must be taken into account [5].
We consider that our study has several limitations. In the rst place, it is the experience of a single university center that only evaluated children who, due to their severity, required hospitalization. This may be an explanation of why children with some comorbidities may require more transfers to intensive care. However, RV / EV infection was more severe and frequent in patients with comorbidities than RSV was in the same group. Another limitation of our study is that we did not have a control group with a negative RT-PCR and ARIthat required a transfer to intensive care. However, when compared with RSV, we observed that the RV / EV behavior is more severe and has a more aggressive natural course. Additionally, the commercial brand of RT/PCR did not allow us to differentially evaluate the enterovirus and rhinovirus to understand the differences between these viruses from the same family [24]. In general, they are analyzed together in most of the tests available commercially today.
The ethics committee of the Fundación Cardioinfantil-Instituto de Cardiología approved this study with committee's reference number PM-35-2020 and the parents of the children approved to participate in this study. This study was done in accordance with the Helsinki declaration. The informed consent was signed by the parents or legal guardian in all the participants.

Consent for publication
Not applicable.