Article Text

Download PDFPDF
From bronchiolitis endotyping to asthma risk assessment
  1. Silvia Carraro,
  2. Valentina Agnese Ferraro,
  3. Stefania Zanconato
  1. Women’s and Children’s Health Department, University of Padua, Padova, Veneto, Italy
  1. Correspondence to Professor Silvia Carraro, Women’s and Children’s Health Department, University of Padua, Padova 35128, Veneto, Italy; silvia.carraro{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Bronchiolitis is the most common lower respiratory tract infection in young children, and it is associated with significant morbidity and high rate of hospitalisation.1 In addition, epidemiological data show that bronchiolitis during early life may be associated with long-term respiratory morbidity.2 The identification of children who will present with recurrent respiratory symptoms after acute bronchiolitis represents a research challenge with possible significant impact on clinical practice. A promising approach to answer such research question derives from the application of multivariate statistical methods that can manage simultaneously a multiplicity of data obtained from different sources.

In their multicentre study, Fujiogi et al3 analysed data collected in 917 infants (<1 year of age) hospitalised for acute bronchiolitis in 17 US centres participating to the 35th Multicenter Airway Research Collaboration study. In enrolled children, nasopharyngel airway specimens were collected and analysed through both lipidomic and transcriptomic approaches.

Applying a consensus clustering analysis, the authors could classify these children according to four endotypes, each defined by a variable combination of clinical, virological and lipidomic data. Thanks to the lipidomic approach the authors could simultaneously study a great number of lipids, enabling a deep characterisation of recruited infants from a metabolic standpoint. Nonetheless, the most interesting aspect of the study is the integrated use of lipidomic and clinical data to endotype the patients. In fact, biochemical–metabolic data on one side and clinical data on the other side provide complementary information and, when considered together, they can offer a more accurate classification of the patients. Likewise, a recent study conducted in patients with severe allergic asthma demonstrated that building models that include both clinical and omic data improve patients’ classification, generating a more complete picture of patients’ phenotype.4 When dealing with complex multifactorial conditions like asthma, the accuracy and thoroughness of patients’ characterisation is of utmost importance inasmuch it can lay the foundation for a personalised treatment.

Among the four endotypes described in infants with bronchiolitis by Fujiogi et al,3 endotype A presents, from a clinical standpoint, many of the typical features of classic bronchiolitis (younger age of children, higher proportion of males, higher prevalence of RVS infection, lower prevalence of respiratory problems and atopy in personal history, and lower prevalence of asthma and eczema in the family). Interestingly, this endotype was characterised from a lipidomic standpoint by a mixed lipid profile, with no predominance of a specific group of lipids. Endotype A was, therefore, used as the reference when evaluating the association of each of the remaining three endotypes with the clinical outcomes of interest (recurrent wheeze and asthma). Through such analysis, the authors found that only endotype D was significantly associated with recurrent wheeze by the age of 3 and with asthma at the age of 6. From a clinical standpoint, infants with endotype D were characterised by a high proportion of breathing problem history, eczema and IgE sensitisation, high proportion of rhinovirus infection, and parental history of asthma and eczema. Form a lipidomic standpoint, compared with endotype A, endotype D demonstrated lower levels of sphingolipids and polyunsaturated fatty acids. A recent study analysing through a metabolomic approach serum samples collected in infants belonging to two mother–child cohorts led to similar results, showing that reduced levels of ceramides and sphingomyelins were associated with increased risk of developing wheezing/asthma by the age of 3.5 Taken together, these data suggest that a perturbed sphingolipid metabolism may have an early pathogenetic role in asthma. Moreover, in general terms, these studies highlight the potential role of omic approaches in the understanding of the metabolic pathways involved in a complex multifactorial disease like asthma.

The relationship between acute viral bronchiolitis in infancy, recurrent wheeze during preschool years and, eventually, asthma development has long been studied, but the pathogenetic connections across these conditions have not been fully clarified yet.

In particular, the linkage between bronchiolitis and recurrent wheezing has long been debated and two main hypotheses have been suggested: either acute viral bronchiolitis has a causal role and induces airway modifications which favour recurrent wheeze, or both conditions are the clinical expression of a common biological (genetic or immunological or related to airway function) background.6 7

In the context of this debate, a recent study from our group demonstrated that the metabolomic urinary profile at birth shows no association with the following development of acute bronchiolitis and postbronchiolitis recurrent wheeze, arguing against the existence of an underlying biochemical–metabolic substrate predisposing to both conditions.8 On the other hand, a previous study that applied the same approach described a discriminant metabolomic profile in infants with acute viral bronchiolitis who later developed recurrent wheezing.9 Taken together, these studies suggest that metabolomic analysis, although unable to identify at birth a profile predictive of postbronchiolitis wheezing, can capture the metabolic disruption caused by bronchiolitis and associated with the following wheezing illness.

The paper by Fujiogi et al,3 classifying 917 infants in four endotypes according to clinical, virological and lipidomic features, and showing how the described endotypes have a different risk of developing recurrent wheeze and asthma, strengthen the hypothesis that the metabolic arrangement during acute infection could be considered together with clinical features to stratify children according to the risk of developing wheeze and asthma. In the next future, the endotyping of children with bronchiolitis could have a significant impact on clinical practice, enabling the early identification of patients who deserve a closer follow-up because of a higher risk of wheeze and asthma, and who might be treated with targeted preventive or therapeutic strategies.

Further studies are needed to test the applicability of the proposed endotypes in other independent populations and also to follow the lipidomic trajectory over time after the acute infection to further investigate the pathogenetic connections between bronchiolitis and the subsequent development of recurrent respiratory symptoms.

Ethics statements

Patient consent for publication

Ethics approval

Not applicable.



  • Contributors SC wrote the manuscript; VAF and SZ critically reviewed the manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests SC and VAF have no conflict of interest to declare; SZ reports financial support from Sanofi outside this work.

  • Provenance and peer review Commissioned; externally peer reviewed.

Linked Articles