Acute viral bronchiolitis phenotype in response to glucocorticoid and bronchodilator treatment

Highlights • The frequency of prescription of medications not recommended for bronchiolitis is high.• Atopic phenotype did not interfere with the response to bronchodilators and glucocorticoids.• The lower the respiratory rate, the shorter the oxygen therapy duration for bronchiolitis.


Introduction
Bronchiolitis is one of the most common illnesses of early childhood. 1,22][3] The United States alone accounts for an average of 100,000 hospital admissions per year. 4In Brazil, in 2019, more than 45.000 children were admitted to hospital for bronchiolitis. 5espite medical and technological advances in the last 2 decades, there has not been a substantial change in Acute Viral Bronchiolitis (AVB) treatment and in the mean length of hospital stay for these patients. 6Since 2011, the Brazilian Society of Pediatrics (SBP) in 2011, 7 through the publication of their guidelines, recommended that infants with AVB should only receive clinical management: oxygen therapy, ventilatory support, and hydration according to infants' individual needs.The latest updates of Brazilian and American guidelines, from 2014 accordingly do not recommend using systemic glucocorticoids and bronchodilators as well as carrying out therapeutic tests with bronchodilators, considered in previous years.][11][12] In 2016, Dumas et al. identified different phenotypes of infants admitted to the hospital for AVB: Profile A -Patient treated at the emergency service with wheezing, with a previous history of wheezing and eczema, who showed a greater association with Rhinovirus (RV) infection; Profile B -Similar to Profile A, however without a previous history of bronchospasm or eczema, with a more prevalent Respiratory Syncytial Virus (RSV) infection; Profile C -Moderate-severe respiratory effort, which required a longer hospital stay; Profile D -The group in which patients did not wheeze, who had less severe disease.As a consequence, they suggest a potential difference in the response to the acute treatment and an important implication in the long-term outcome, with future risk for asthma. 1 Rodríguez-Martínez et al., in a review study, describe phenotypes of infants with AVB that are more likely to have a positive response to bronchodilator use: older infants with AVB by RV, especially those with predominant Haemophilus influenzae in their microbiome nasopharyngeal; those affected during the months opposite the months of peak RSV; infants who have wheezing; those with a history of atopic dermatitis or a family history of asthma in a first-degree relative. 3onsidering the emergence of this new knowledge regarding disease heterogeneity and the scarce national production on the frequency with which non-recommended drugs are prescribed in the present environment, this study aimed to analyze: how often infants admitted to hospitals with AVB received bronchodilators, glucocorticoids, and antibiotics; how to assess whether those who received glucocorticoids and bronchodilators and had an atopic phenotype spent less time in hospital and/or less time on oxygen therapy when compared to those who did not have the phenotype.It was chosen to test this response to bronchodilators and glucocorticoids in this specific group of patients due to the clinical similarity of AVB with asthma.Thus, a hypothesis was raised that patients with an atopic phenotype may benefit from using bronchodilators and glucocorticoids, presenting a shorter length of hospital admission and oxygen therapy duration, which corroborates the idea of individualizing the therapeutic approach in AVB guided by anamnesis.To ensure that antibiotic use did not constitute an isolated factor that increased the length of hospital stay, the present study's hypothesis was further tested for these subgroups (those who received and those who did not receive antibiotic therapy).

Study design and place
An epidemiological, retrospective, cross-sectional study was carried out, in which patients under 2 years old admitted with a diagnosis of AVB at the Hospital Universit ario Regional de Maring a (HUM), a tertiary public hospital, were identified through a search of medical records from 2012 to 2019.The study was conducted following the guidelines outlined in the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.

Study population
The numbers of physical records were obtained in two ways.The first corresponded to the period from 2012 to 2014, when the computerized system had not yet been implemented in the hospital, through a search in a book of hospital admission records in the ward or Pediatric Intensive Care Unit (ICU).The search was performed by diagnoses of bronchiolitis, acute respiratory failure, and asthma.In the second, from 2015 to 2019, through the SUS Health Care Management System (GSUS -Sistema de Gestão da Assistência de Sa ude do SUS), the International Classification of Diseases (ICD) was researched: J21, J96 e J45.It was decided to add the other diagnoses in the initial survey in addition to bronchiolitis, considering the possibility of diagnosing bronchiolitis among the alternative ICD.

Variables of interest
The following were investigated in the medical records: age (days); sex; weight; gestational age at birth; comorbidities; final diagnosis (in discharge summary); the presence of wheezing; history of fever measured or reported by a caregiver; family history of atopy (father, mother or siblings with a history of allergic rhinitis, asthma or atopic eczema); history of atopy (previous wheezing, atopic eczema, allergic rhinitis); Exclusive Breastfeeding (EBF) duration; data from the first blood count; length of stay; and oxygen therapy.It also verified the prescription of antibiotics, systemic glucocorticoids and inhaled bronchodilators during the hospital admission period.Regarding vital signs (Heart Rate [HR], Respiratory Rate [RR] and Oxygen Saturation [SpO 2 ]) in order to mitigate the effect of body temperature on other signs, the first set with a temperature below 37.5 °C was considered from admission.Afterward, the score of vital signs in pediatrics (Vipe) 13 was used, which attributed a score to each sign according to infants' age, and the sum of this score was categorized into one of the 5 groups of score severity (blue, green, yellow, orange and red -from least to most severe).

Exclusion criteria
Those patients who did not present a diagnosis of AVB in the discharge summary (which is the definitive admission diagnosis), premature individuals (less than 37 weeks), and those with other associated diagnoses, such as pneumonia, genetic syndromes, heart disease, pneumopathy, neuropathies, pertussis syndrome and patients who died (Fig. 1).Infants who did not present with wheezing at any point during hospital admission were also excluded.Among infants who received antibiotic therapy, those who presented leukocytes above 15,000 with a number of neutrophils above 50 % of the total leukocytes were excluded.
The choice of this criterion was based on laboratory criteria present in protocols for fever without localizing signs in infants, which searches for bacterial disease in children under 36 months of age. 14fter calculating the frequency of use of bronchodilators and systemic glucocorticoids, one infant who had not received glucocorticoids was also excluded, with no need to be excluded due to non-use of bronchodilators (100 % received them) (Fig. 1).

Statistical analysis
Wilcoxon's non-parametric hypothesis test was used with a significance level of 5 % to test the association of length of stay and oxygen therapy duration separately with history of atopy, and family history of atopy.This analysis was also performed in the subgroup that received antibiotic therapy, comparing it with the group that did not receive it, both in the atopic group and in the group without atopy.Afterward, the association test was applied to compare the length of stay/oxygen therapy duration in the group that received antibiotics with the group that did not, regardless of personal or family history of atopy.The association between length of stay and/or oxygen therapy duration with sex, fever, Vipe score, HR, SatO 2 , RR, and Exclusive Breastfeeding (EBF) duration was also tested.Data

Results
Thus, 207 medical records of individuals younger than two years were selected, but 149 were excluded (Fig. 1).Of the 58 infants included in the study, 32 (55.2 %) were male.The median age was 109 days, with 18 infants over 180 days old and only 5 over one year old.According to the parameters adopted by the Ministry of Health, 15 52 (89.4 %) infants were classified as having adequate weight for their age (Table 1).
The mean length of stay was 5.4 days (±3.2).This calculation was only possible with 50 infants, since, after discharge from the ICU, 8 patients returned to the hospital of origin, where they completed the length of stay.Twenty-one infants (36.2 %) were admitted to the ICU with a mean of 2.6 days (standard deviation of ±1.9) of stay in this sector.No patient required mechanical ventilation, and 43 (74.1 %) infants received oxygen therapy at some point.Of these, 4 received oxygen for less than 24 h.The mean time of those who received oxygen for more than 24 h was 2 days (variance 5.4 ± 2.3 days) (Table 1).
All patients received systemic glucocorticoids and inhaled bronchodilators within the first 24 h of admission, and 62.1 % (36/58) received some antibiotic for at least 2 days, starting within 24 h of admission: 21 received azithromycin (58 % of those who used antibiotics); 3 received ceftriaxone; 3 received crystalline penicillin; 3 received ampicillin; and 1 used amoxicillin as sole antibiotic therapy.Among the infants, 5 received 2 antibiotics (4 azithromycin and ampicillin or ceftriaxone or crystalline penicillin), and 1 received ampicillin and gentamicin.
Of the infants older than 180 days, 12 had information on EBF duration, and 66.7 % (8) of these were breastfed from 30 to 119 days.There was no influence of EBF duration on length of hospital stay or oxygen therapy duration (p = 0.60; 0.23) (Table 2).
Vital signs were described for 52 (89.7 %) infants, and 58.6 % (34) reported fever at some point during the illness.Of these, 26 (50.0 %) received orange stratification by the Vipe score (Table 1).No change in vital signs, tested for different cut-off points, was associated with length of stay, and only RR was associated with oxygen therapy duration (p = 0.03), in which infants with RR with a score ≤1 received oxygen therapy for less time (Table 2).
As for the history of atopy, 5 (8.6 %) patients had a history of eczema or allergic rhinitis or a previous episode of wheezing, 9 (15.5 %) with undetermined data (no information in the medical record).Twenty-two (47.8 %) had first-degree relatives with a history of atopy (Table 1).
There was no statistically significant difference in the length of hospital stay or oxygen therapy duration when considering the atopic phenotype personal or family (Table 2).When infants without a family history of atopy were compared to each other, those who had received antibiotics had a longer hospital stay than those who did not (p = 0.01) (Table 3).In general, those who received antibiotics tended to remain admitted to the hospital longer than those who did not (p = 0.07) (Table 2).

Discussion
The present study did not show any difference in length of hospital stay or oxygen therapy duration of infants with AVB who received systemic glucocorticoids and bronchodilators, when considering the phenotype as follows: presence or absence of a personal history of atopy or a family history of atopy.Another highlight was the high frequency of prescription drugs not recommended by national guidelines. 9irolos et al. (2019), in a systematic review, compared the recommendations of 32 guidelines on AVB published until 2017.Of these, 22 recommend against using bronchodilators, 14 consider the findings controversial or consider the therapeutic test in specific situations, and only 3 recommend its use.Glucocorticoids as well as antibiotics are not recommended by any, and the non-recommendation of using glucocorticoids is present in 29 (90.6 %) of the 32 guidelines. 180][21][22] Thus, a group of patients could benefit from using medications that, until now, have not been recommended. 9,11,18,23 Megalaa et al. (2018), for instance, suggest that infants with asthma, mistakenly diagnosed with AVB, could benefit from using glucocorticoids and bronchodilators. 20n this line, a Pakistani multicenter study with 212 infants compared the clinical response between patients with a family history of atopy and patients without a family history to prednisolone use in the face of AVB and observed a shorter hospital stay and a significant improvement in symptoms among patients with a family history of atopy, 19 which is in contrast to the present findings.This difference can be explained by the difference in the predominant age group in the studiesthe infants had a lower mean age, with the majority (68.9 %) being younger than 6 months (mean 4.8 ± 4), while in the Pakistani study, the majority was older than 6 months (mean 5.39±3.11),with a lower age of inclusion in the study of 2 months.Rodríguez-Martínez (2021) suggests, in his review, that infants with AVB older than 6 months and with a parental history of asthma may be potential responders to bronchodilators. 3 randomized clinical trial conducted in Qatar found shorter hospital stays for infants with AVB who received oral dexamethasone and salbutamol and had risk factors for asthma such as a history of eczema or a family history of eczema or asthma (OR = 0.69, 95 % CI 0.51 to 0.93) when compared to infants without risk factors.24 This study, unlike ours, considered preterm infants older than 34 weeks, who may have an immune response and clinical behavior different from infants exclusively older than 37 weeks.25 Different tools for severity categorization were used in the studies from Pakistan, Qatar, and the present study: Modified Respiratory Distress Assessment Instrument (mRDAI) Score, Wang Bronchiolitis Severity Score and Vipe score, respectively.13,26,27 Although the Vipe score only considers vital signs and does not compare respiratory effort data, the choice of this severity categorization tool is justified as it considers objective data routinely documented in medical records and allows comparing RR between different age groups, which is not included in other scores.26,27 Furthermore, the Vipe score uses age segmentation for reference values of vital signs, as recommended by the World Health Organization: 0-2, 3-11, 12-24 months.28 Additionally, the present analysis resulted in the association of the Vipe score for RR ≤ 1 for age (Table 2) with shorter oxygen therapy time (p = 0.03), this variable being data already related to the clinical evolution of pediatric patients.29 2016), in a multicenter study, showed that infants admitted to the hospital due to AVB caused by RV have phenotypic characteristics of asthmatic infants (older infants, with a history of wheezing or a history of eczema) when compared to AVB caused by RSV. 12 Sena et al. (2021) identified a higher risk of asthma in infants, and children of an asthmatic mother, with AVB via RV.30  Lukkarinen et al. (2013), in a randomized, double-blind study, found benefit in using prednisolone by patients admitted to the hospital for AVB caused by RV and/or with a history of eczema, in reducing recurrent wheezing.31 Martínez et al. (2022) associated the decrease in using bronchodilators with earlier hospital discharge, which could save material and human resources.32 Using intravenous antibiotics has already been described in other works as an isolated risk factor for increasing children's length of stay.33,34 In the present study, there was a tendency towards an increase in the length of hospital stay related to using antibiotics, regardless of history of atopy (p = 0.07).When considering a family history of atopy, infants without a history who received antibiotics remained admitted to the hospital longer than those who did not (p = 0.01).This result probably did not extend to the other analyses due to the small sample values of the assessed subgroups, which is one of the limitations of the present study.Another finding that corroborates a possible isolated influence of using antibiotics on length of stay was that it was not found a longer period of oxygen therapy in the group that had a longer length of stay associated with antibiotic use, i.e., this infant's clinical evolution did not differ between the groups (with and without antibiotics).This influence on length of stay can be explained by the fact that the attending physician wants to guarantee complete treatment of these infants in a hospital environment, either due to insufficient trust in caregivers or due to the difficulty of accepting oral medications presented by some babies, requiring a complete intravenous treatment.
The frequency with which drugs not recommended by Brazilian guidelines were prescribed 8 is much higher than that found in the literature as a practice in other countries (Table 4).The most prescribed antibiotic was azithromycin, which, despite the already described immunomodulatory effect of macrolides, 35 does not present data that justify its use in AVB. 32This high frequency of antibiotic prescription is probably due to clinical concerns about co-infections and complications in cases of severe AVB.
A study published in 2017 with 38 hospitals from 8 countries showed that, on average, 14 % of infants with AVB received a bronchodilator and 2.3 % a glucocorticoid. 36In Brazil, there are few studies on the frequency with which these drugs are prescribed for AVB.Despite the result portraying the medical conduct in a single service, it is a tertiary public hospital, with a sentinel profile for surveillance of respiratory viruses.It is suggested, however, that multicenter studies be carried out involving hospitals in the state of Paran a, for greater robustness and external evidence validity.Wilcoxon test: Group 1B (received antibiotics and have a personal and family history of atopy); Group 2B (did not receive antibiotics and have a personal and family history of atopy); Group 3B (received antibiotics and no personal or family history of atopy); Group 4B (did not receive antibiotics and has no personal or family history of atopy).Other limitations that can be mentioned in this study are the nature of the data and the incomplete filling of medical records and the nonidentification of viruses for analysis.Moreover, the sample was reduced as a result of strict exclusion criteria in an attempt to avoid any bias, such as concomitant bacterial infection, which could influence the present results.Despite the limitations, the study provides useful data to corroborate results that guide the main guidelines for AVB treatment in the world, when considering a family history of atopy or a history of atopy in infants.
It was concluded that the presence of an atopic phenotype did not interfere with the length of stay and/or oxygen therapy duration of infants with AVB who received inhaled bronchodilators and systemic glucocorticoids in the same way as breastfeeding duration.Among the vital signs, only RR was related to oxygen therapy duration.It was highlighted here an increased length of stay for infants without a family history of atopy who used antibiotics when compared to those who did not.Until new studies are carried out considering the phenotype for AVB treatment, guidelines based on clinical evidence proposed by current guidelines should guide the clinical practice of professionals who assist this population.Better strategies are needed to promote stricter protocol implementation and professional training in AVB diagnosis and care.This will be reflected in the reduction of expenses with drugs that do not offer a proven benefit, also reducing the frequency of their side effects and patient's length of stay.
were stored in Microsoft Excel® version 2211 and analyzed in R® version 4.1.2(2021−11−01).This study was submitted to the Standing Research Ethics Committee Involving Human Beings (COPEP) of the Universidade Estadual de Maring a, and was approved on December 23, 2021 under Certificado de Apresentação para Apreciação Etica (CAAE -Certificate of Presentation for Ethical Consideration) n°54,236,221.2.0000.0104(Opinion 5.186.320).Since it is secondary data, there was a waiver of the Informed Consent Form (ICF).

Table 1
Clinical and epidemiological characteristics of infants treated at the Hospital Regional Universit ario de Maring a from 2012 to 2019.

Table 2
Comparison of variables regarding length of stay and oxygen therapy duration in infants admitted to the Hospital Regional Universit ario de Maring a from 2012 to 2019.

Table 3
Comparison of length of stay of infants admitted to the Hospital Universit ario Regional de Maring a with a history of atopy and a family history of atopy with infants without a history in the subgroups that received antibiotics or not from 2012 to 2019.

Table 4
Comparative values of the frequency of prescription of bronchodilator, glucocorticoid and antibiotic for patients with AVB between different studies of different nationalities.