Although bronchiolitis is 1 of the most common conditions in infancy, no medicine has been proven to be consistently beneficial. As a result, the primary objective of frontline providers in primary care and emergency medicine is to triage infants to the appropriate level of care (ie, home, hospital ward, or intensive care) for their hydration and respiratory needs.1
However, prospectively validated risk-stratification tools for infants with bronchiolitis do not exist for frontline providers. Freire et al,2 as part of the Pediatric Emergency Research Networks, addressed this knowledge gap by analyzing data that were collected from 38 emergency departments across the globe to develop a bronchiolitis risk score. In the resulting article,2 the authors report that the usual suspects (ie, being ≤2 months of age; having apnea; oxygen saturation of <90%; signs of increased work of breathing [including nasal flaring, grunting, and retractions]; and dehydration and/or poor feeding) were all predictive of escalated care, which is defined as hospitalization with a high-flow nasal cannula, noninvasive or invasive ventilation, or intensive care admission.
Although many clinicians feel comfortable triaging infants with bronchiolitis at either end of the severity spectrum (ie, well or critically ill), their comfort wanes when triaging moderately ill infants whose outcomes are less certain. For example, using Freire et al’s2 scoring system, a 1-month-old who is not feeding well, is dehydrated, and has a room-air oxygen saturation of <90% in the emergency department triage has 8 risk points and is predicted to have a ∼50% chance of needing escalated care. Clinicians at hospitals without ready access to advanced infant respiratory support measures would need to decide if they should provide hydration and observe the infant for further signs of respiratory distress or recommend immediate transfer to a tertiary- or quaternary-care center, possibly hundreds of miles away. These difficult decisions may become somewhat easier after the authors refine their score during their proposed prospective validation study.
Beyond refining the predictors in the risk score, another critical issue is variability in the outcome (escalated care). The use of different respiratory support methods varies within and between countries,3,4 and there is no accepted measure that can be used to confirm which subset of infants truly needs escalated care. Indeed, without some objective means of understanding the outcome, low oxygen saturations may become a self-fulfilling prophecy of an overly valued measure of severity, leading to overtreatment.5
As researchers develop risk scores and evaluate new respiratory support methods,6,7 there needs to be a simultaneous push to identify bronchiolitis subgroups by improving our understanding of the underlying pathobiology. Without this groundwork, clinicians will continue to treat and researchers will continue to analyze children with bronchiolitis as 1 homogeneous group despite data suggesting that bronchiolitis is heterogeneous.8 Indeed, when the underlying pathobiology of a condition is not accounted for in otherwise rigorous studies, there is an increased chance of obfuscating true associations between exposure and outcome. During the proposed validation study, the authors would ideally create a biorepository to eventually incorporate biologically based identifiers (eg, specific viruses,9 microbiome patterns,10 and host responses11,12) that would be used to improve disease classification and the relevance of the risk score for infants with each identifier.
For example, recent data support clinical, bacterial, and mechanistic differences between respiratory syncytial virus (RSV) and rhinovirus, the 2 viruses that cause 85% of the cases of bronchiolitis requiring hospitalization. Compared with children with RSV bronchiolitis, children with rhinovirus bronchiolitis have a shorter hospital length of stay9,13 and are significantly more likely to be treated with systemic corticosteroids when hospitalized14 and with inhaled corticosteroids the year after hospitalization.15 Furthermore, RSV-only and rhinovirus-only infections are associated with different bacteria in the nasopharyngeal airway at hospitalization.16 And differences in bacterial dominance in the nasopharynx at hospitalization are associated with a higher risk of intensive care use.10 There are also results revealing mechanistic differences between RSV-only and rhinovirus-only infections, including distinct microRNA expression profiles17 and completely separate metabolic pathways.18 Although replication is needed, these differences reveal that the one-size-fits-all approach to caring for and analyzing infants with bronchiolitis may be limited.
Indeed, to make substantial progress in the care of infants with bronchiolitis, clinicians and researchers will first need to understand and embrace the heterogeneity of this common condition. We also need to follow the lead of Pediatric Emergency Research Networks investigators. Global collaboration should be celebrated and replicated given its clear benefits in generalizability. Developing a validated global tool to help clinicians appropriately triage infants with bronchiolitis will not only be helpful, but will also provide a counterweight to our current nihilism and learned helplessness rooted in cycles of promising initial trial results that ultimately lack demonstrable benefit.19 Embracing the heterogeneity of bronchiolitis by incorporating more biologically based identifiers into bronchiolitis studies will help providers to finally overcome the bronchiolitis blues: our collective frustration every winter with having to tell parents “nothing works.”
Opinions expressed in these commentaries are those of the authors and not necessarily those of the American Academy of Pediatrics or its Committees.
FUNDING: Supported by the National Institutes of Health (grant R01AI108588). Funded by the National Institutes of Health (NIH).
COMPANION PAPER: A companion to this article can be found online at www.pediatrics.org/cgi/doi/10.1542/peds.2017-4253.
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.