Introduction

Lower respiratory tract illnesses are a frequent occurrence for preschoolers, often accompanied by wheezing, and associated with increased risks of recurrent wheezing and subsequent asthma diagnosis. Approximately one third of children will experience at least one episode of wheezing during the first 3 years of life, and approximately half of these children will continue to experience wheezing episodes until at least 6 years of age [1]. These episodes are often severe and exert substantial morbidity, including unscheduled visits to physician offices, urgent care centers, and emergency departments. Many of these young children have been diagnosed with asthma, and among them, 20.9% seek emergency department care and 6.5% are hospitalized each year [2]. There is mounting evidence that asthma pathogenesis is linked to events during early childhood, particularly viral respiratory tract infections and their interactions with the recently identified and characterized respiratory tract microbiome [3, 4]. Since asthma is a disease with substantial health care costs and morbidity, advancing our knowledge of potential treatments that reduce the frequency, severity, and sequelae of these early-in-life lower respiratory tract illnesses could make a significant impact. In this review, we will examine the role of antibiotics in the management of these episodes, with a focus on recent developments.

Impact of Lower Respiratory Tract Infections on Recurrent or Persistent Wheezing

Bronchiolitis is an important cause of morbidity in young children, particularly infants. A study of U.S. hospital discharges between 1988 and 1996 found that 47% of lower respiratory tract discharges and 16% of overall discharges were due to bronchiolitis [5]. A subsequent study found that on average each year, respiratory syncytial virus (RSV) bronchiolitis in particular accounts for an estimated two million outpatient visits and 57,527 hospitalizations among children 5 years old and younger [6]. The occurrence of viral respiratory tract infections with wheezing increases the risk of subsequent asthma development, particularly when underlying atopy is present [7, 8], and the wheezing illnesses are triggered by human rhinovirus. Preschoolers with wheezing during rhinovirus respiratory tract infections are at increased risk of asthma [9] and have lower lung function at age 8 years compared with those who do not have rhinovirus wheezing illnesses during the first 3 years of life [10]. The severity of the respiratory tract infection and the specific viral pathogen (e.g., rhinovirus C) causing the illness also influence the risk of subsequent asthma [11,12,13].

Assessing the role of RSV infection with subsequent asthma risk is particularly important, as RSV infection is almost universally acquired during the first 2 years of life [14]. The Tucson Children’s Respiratory Study reported that children under 3 years of age with predominantly mild RSV lower respiratory tract infections were at increased wheezing risk at age 11 years but not at age 13 years, suggesting that the risk after mild RSV infection may diminish with time [15]. Recently, additional studies strengthened the evidence supporting the association between severe RSV infection during infancy and subsequent wheezing risk. A large retrospective cohort study of over 70,000 children in Northern California showed that mild and severe RSV infections during infancy, defined as requiring outpatient evaluation or prolonged hospitalization, respectively, were associated with increased risk of wheezing at both ages 3 and 5 years, with an escalating risk linked to increasing RSV illness severity [16]. Infants with a severe RSV infection were at higher risk for future recurrent wheezing even when compared with premature infants born at 32–33 weeks of gestation. In addition to being associated with greater risk of subsequent recurrent wheezing, severe RSV infection is also linked to greater asthma risk. Several studies have found positive associations between severe RSV infection during infancy and subsequent development of asthma [17, 18], with the RSV Bronchiolitis in Early Life (RBEL) cohort reporting that severe RSV bronchiolitis requiring hospitalization during infancy is followed by an approximately 50% likelihood of developing physician-diagnosed asthma by 7 years of age [19].

The exact mechanisms responsible for the association between bronchiolitis during infancy and subsequent increased recurrent wheezing or asthma risk remain uncertain, but recent studies have identified potential associations and pathways. Increased nasal epithelial cell expression of the chemokine CCL5, a chemoattractant for inflammatory cells such as T-lymphocytes, during RSV bronchiolitis was found to be a predictor of physician-diagnosed asthma [19]. Higher levels of IL-3 and IL-12p40 in bronchoalveolar lavage fluid during hospitalization for RSV bronchiolitis were associated with increased risk of recurrent wheezing [20]. Rare non-synonymous genetic variants, notably in ADRB2, have been found to contribute to asthma development following severe RSV bronchiolitis [21]. Another potential association has been reported between the TLR9 rs1870084 gene polymorphism and wheezing after bronchiolitis [22].

The Role of the Microbes and the Microbiome on Recurrent or Persistent Wheezing

Recent work highlights the frequent co-presence of bacteria during viral respiratory tract infections, and the presence of certain bacteria impacts the outcomes of such illnesses. Suarez-Arrabal et al. assessed RSV-infected infants requiring hospitalization compared with healthy aged-matched controls who were enrolled either at primary care offices or prior to scheduled non-respiratory tract surgical procedures such as hernia repairs [23]. Compared with controls, infants with RSV infection were more likely to be culture positive for the potentially pathogenic organisms Haemophilus influenzae, Streptococcus pneumoniae, and Moraxella catarrhalis (81% in RSV-positive infants vs 65% in controls) in samples obtained via nasopharyngeal swab and nasal wash. Thirty-two percent of RSV-positive infants versus 4% of controls had growth of more than one potentially pathogenic bacterial species, and RSV-positive infants had increased numbers of white blood cells in nasopharyngeal fluid. Their analysis was limited by high rates of antibiotic use in the prior 2 weeks, which resulted in a significant decrease in bacterial recovery. A study from the Copenhagen Prospective Study on Asthma in Childhood (COPSAC) cohort found that the presence of pathogenic bacteria or viruses in the hypopharynx during a wheezy episode was not associated with the duration of the episode [24•].

There is evidence that children who develop recurrent or persistent wheezing have differences in their upper airway microbial composition compared with those who do not. In the COPSAC study, Bisgaard and colleagues found that one-month old infants with hypopharyngeal colonization with S. pneumoniae, M. catarrhalis, and/or H. influenzae were at increased risk of recurrent wheezing and later diagnosis of asthma at 5 years of age [25]. A retrospective study assessed bronchoalveolar lavage samples from children aged 4 to 38 months with persistent wheezing despite treatment with inhaled corticosteroids and found that 48.5% of the children had positive lower airway cultures for potentially pathogenic bacteria, with non-typeable H. influenzae, S. pneumoniae, and M. catarrhalis being the most common [26].

Environmental microbe exposure has also been linked to an altered risk of recurrent wheezing, with more diverse bacterial exposures being associated with a decreased risk for recurrent wheezing. In an urban population at high risk for asthma, children who were exposed to the highest diversity of bacterial content in house dust were at lower risk of recurrent wheezing at age 3 years. Additionally, lower exposure to Firmicutes and Bacteriodetes in house dust during the first year of life was associated with atopic wheeze [27]. Similarly, environmental exposure to a wide range of microbes [28] and higher endotoxin levels [29] in house dust found in houses from certain farming communities has been associated with lower asthma rates.

Increasing evidence has highlighted a critical time window for establishing protective or detrimental microbiota compositions. Alterations in the infant microbiome, specifically reflected by the presence and/or abundance of specific organisms and the overall degree of microbiome diversity, may be uniquely protective. Cardenas et al. assessed microbiota composition in oropharyngeal swabs from infants who were 10.2 months old on average [30]. This population from rural Ecuador was unique in that they had minimal antibiotic exposure and no inhaled corticosteroid exposure. This case control study found that compared with controls, oropharyngeal samples from infants who wheezed had increased Streptococcus spp. and Haemophilus spp., similar to the findings of Bisgaard et al. [25], as well as increased Staphylococcus spp. The Canadian Healthy Infant Longitudinal Development (CHILD) Study found that the abundance of four bacterial species, specifically Lachnospira, Veillonella, Faecalibacterium, and Rothia, was transiently decreased in stool samples of children with early onset atopic wheezing when compared with controls at 3 months of age [31••]. By 1 year of age, these differences were much less apparent. A study by Abrahamsson et al. found that a lack of microbial diversity during early infancy appeared to be detrimental, as lower microbial diversity in stool samples from the first week and month of life was associated with increased asthma rates at 7 years of age [32].

Role of Antibiotics in the Treatment of Bronchiolitis and Wheezing in Young Children

Despite the unequivocal role of viral infections in the pathogenesis of acute bronchiolitis, antibiotics are often used clinically in this condition, being prescribed during nearly one third of emergency department visits for bronchiolitis [33] despite limited evidence for this practice. A 2014 Cochrane Report assessing the use of antibiotics for bronchiolitis in children under 2 years of age included seven studies with 824 participants [34]. The primary outcome was duration of symptoms, including wheezing in some studies. They found insufficient evidence to support the use of antibiotics, although they comment that perhaps a subgroup would benefit from antibiotic treatment. The 2014 American Academy of Pediatrics bronchiolitis guidelines recommend that antibiotics should not be used for routine treatment of bronchiolitis in infants and children, and should only be prescribed if there is a concomitant bacterial infection or a strong suspicion of one [35•]. Another 2014 Cochrane Report comparing antibiotics with no antibiotics for children aged 2–59 months with non-severe pneumonia and wheeze found no studies that completely fulfilled their inclusion criteria [36]. A 2012 Cochrane Report assessed the use of antibiotics for persistent cough or wheezing following acute bronchiolitis in children. Only one study met their inclusion criteria, and it did not find a therapeutic effect [37].

Macrolide antibiotics, in particular, have been the focus of recent clinical studies, as they have anti-inflammatory properties and reduce neutrophil numbers [38, 39]. Macrolides have been demonstrated to be beneficial in inflammatory airway diseases such as cystic fibrosis, diffuse panbronchiolitis, COPD, and in some cases, asthma, although studies have yielded mixed results [39]. The mechanism of action, anti-inflammatory and/or antimicrobial, in these conditions remains uncertain. Several studies assessing the impact of antibiotic use on the duration of hospitalization for bronchiolitis have been consistently negative (Table 1). A Dutch study assessed the impact of azithromycin use by infants 2 years old and younger hospitalized with RSV bronchiolitis, and found no differences in mean hospitalization duration or RSV symptom score improvement when compared with placebo [40]. A randomized controlled trial in Brazil assessed the use of a 7-day course of azithromycin in infants less than 12 months of age who were hospitalized with acute bronchiolitis [41]. A virus was detected in 63–64% of patients, of which 90–95% were positive for RSV. There was no difference in duration of either hospitalization or supplemental oxygen use. McCallum et al. performed two studies of azithromycin use in the setting of acute bronchiolitis, neither of which found an association between macrolide use and bronchiolitis symptom improvement. The use of one dose of azithromycin in children 18 months and younger hospitalized for bronchiolitis was not associated with significant differences in supplemental oxygen requirement or rehospitalization [42]. Nasopharyngeal bacterial carriage was found to be reduced among the azithromycin-treated children. In another study, 219 infants who were 24 months of age and younger and hospitalized for bronchiolitis were randomized to receive three once-weekly doses of azithromycin or placebo, and no differences in time receiving supplemental oxygen, symptoms, or need for rehospitalization within 6 months were found between the treatment groups [43•].

Table 1 Summary of referenced studies of the role of azithromycin in children with bronchiolitis
Full size table

In contrast, two recent trials have shown the efficacy of azithromycin in preschool children with recurrent lower respiratory tract illnesses (Table 2). A large, multicenter trial by the NHLBI’s AsthmaNet explored the preemptive use of azithromycin in an effort to prevent severe episodes. This trial assessed if azithromycin, when started at the earliest signs of a respiratory tract illness and before the development of lower respiratory tract symptoms, in high-risk children aged 12–71 months, could prevent progression to severe episodes [44••]. The duration of the study was 12–18 months for each participant, with a maximum of four treated illnesses per participant. Children had histories of recurrent severe lower respiratory tract illnesses but were not receiving daily asthma controller therapy. Four hundred forty-three children were included in the primary outcome analysis, and these children experienced a total of 937 treated respiratory tract illnesses, 80–83% of which had a virus detected in nasal wash samples. Compared with placebo, a 5-day course of azithromycin resulted in a significant reduction in the risk of progression to a severe episode, with a hazard ratio of 0.64 (95% CI, 0.41–0.98, p = 0.04) and less rescue albuterol use during severe illnesses. Azithromycin-resistant organisms in oropharyngeal samples were infrequently acquired, occurring in six of 36 participants treated with azithromycin and four of 37 treated with placebo.

Table 2 Summary of referenced studies of the role of azithromycin in children with recurrent wheezing
Full size table

Another randomized controlled trial found that azithromycin use during an episode of lower respiratory tract symptoms resulted in a shorter duration of asthma-like episodes. This trial, from the COPSAC cohort, assessed 72 children aged 1–3 years with a history of recurrent asthma-like symptoms who were randomly assigned to receive either a 3-day course of azithromycin or placebo during asthma-like episodes, which were physician-evaluated and defined as having at least 3 consecutive days of troublesome lung symptoms prior to starting study medication [45••]. Hypopharyngeal aspirates were cultured for selected bacteria (H. influenzae, S. pneumoniae, and M. catarrhalis). Nasopharyngeal aspirates were assessed for rhinoviruses, RSV, and enteroviruses via PCR. 60% of samples collected were positive for at least one of the bacterial species, with M. catarrhalis being the most common (47%). Forty-three percent of the samples collected were PCR-positive for a virus, with an incidence of 16–20% for each type of virus. One hundred fifty-eight asthma-like episodes occurred in the two groups, and the average duration of each episode after treatment initiation was significantly shorter in the azithromycin group compared with placebo (3.4 vs 7.7 days). They found that early treatment with azithromycin, prior to day 6 of symptoms, resulted in the greatest reduction in the episode duration (83%).

There is considerable evidence that antibiotic use contributes to the development of antimicrobial resistance and may also negatively alter the microbiome. A study by Korpela et al. of antibiotic use in 2–7-year old Finnish children found that early antibiotic use was associated with altered intestinal microbiota composition and increased risk of asthma [46•]. The altered microbiota composition was notable for decreased Actinobacteria, increased Bacteroidetes and Proteobacteria, and increased rates of macrolide resistance. A retrospective population-based cohort study found that antibiotic use during the first year of life was associated with an increased risk of asthma development by 3 years of age, with a positive relationship between number of antibiotic courses received and risk of asthma [47]. A cross-sectional study showed that bronchiolitis during the first 2 years of life and antibiotic exposure during infancy were associated with asthma development during adolescence [48]. While these studies suggest that antibiotic use and subsequent asthma development are related, the cause of this relationship is still unknown. Furthermore, this association is likely to be complicated by confounding by indication. Children with significant lower respiratory tract illnesses are more likely to receive antibiotics for treatment of these illnesses, and thus it is possible that while antibiotics could be causative of subsequent asthma development, antibiotic use could also be a marker of an underlying predisposition for, or a comorbidity associated with, asthma.

Role of Antibiotics for the Prevention of Asthma

While macrolide use does not significantly alter the course of acute bronchiolitis, recent studies suggest that azithromycin may have a role in the prevention of recurrent wheezing following bronchiolitis. The first was a randomized trial of 21 infants with RSV bronchiolitis which compared clarithromycin to placebo for 3 weeks, and reported that clarithromycin use was associated with a shorter length of hospital stay, shorter duration of supplemental oxygen requirement, lower amount of β2 agonist use, and lower rates of readmission to hospital over the ensuing 6 months [49]. However, subsequent communications have questioned the validity of these findings [50, 51]. The second study was a proof-of-concept, randomized, placebo-controlled trial involving 40 children 1–18 months of age without a prior wheezing history who were hospitalized with RSV bronchiolitis [52••]. A unique aspect of this study was inclusion of a homogeneous population of otherwise healthy full-term infants. When added to routine bronchiolitis care, azithromycin treatment for 14 days resulted in a lower risk of recurrent wheezing in the subsequent year when compared to placebo. It is interesting to note that wheezing outcomes between the treatment and placebo group were similar at 6 months, suggesting that the effect of azithromycin therapy may be predominantly on illnesses occurring more than 6 months after treatment [53]. Additionally, nasal lavage fluid IL-8 levels, a marker of neutrophilic airway inflammation, were reduced, although serum IL-8 levels were not [52••]. The upper airway microbiome was also assessed, and treatment with azithromycin for acute RSV bronchiolitis was found to result in changes in airway microbiome composition, particularly a reduction in Moraxella [54]. Azithromycin was well-tolerated, with mild gastrointestinal adverse events including diarrhea, vomiting, and abdominal pain reported at similar rates in the placebo group compared with the treatment group [52••]. These studies were limited by their small sample size. Larger studies are required to determine if this therapy provides clinically important and durable effects on recurrent wheezing and asthma as well as the consequences of this therapy on the microbiome.

Conclusion

In summary, early-in-life lower respiratory tract infections are strongly associated with augmented risks of persistent wheezing and asthma development. Most research on this topic has focused on viral causes of lower respiratory tract disease, particularly RSV and rhinovirus. However, increasing knowledge about the infant microbiome and effects of environmental exposure on expression of respiratory disease suggests that subsequent asthma development is also associated with alterations in microbiome composition. Current evidence indicates that while macrolides do not impact the immediate course of illness for infants hospitalized with bronchiolitis, azithromycin may have a role in the prevention of future post-bronchiolitis wheezing episodes in high-risk children with severe RSV bronchiolitis. Furthermore, recent evidence suggests that azithromycin use early during respiratory episodes in high-risk preschoolers with recurrent wheeze can shorten episode duration and lower the risk of progression to severe lower respiratory tract illnesses. More research is needed to confirm these beneficial effects, to better characterize azithromycin’s mechanism of action in these disorders, to further define the risk of antibiotic resistance, as well as to explore the potential roles for other antibiotics in the management of early life recurrent wheezing.