INTRODUCTION
Bronchiolitis, a common acute lower respiratory infections (ALRIs) among
infants worldwide1 causes considerable morbidity and
hospitalisations annually (e.g. >3 million
hospitalisations)2. It is a heterogenous,
multi-dimensional disorder from many perspectives, including clinical
phenotypes (during acute disease3 and future
outcomes4-7),
likely
pathophysiology8 and risk
factors9-11.
The importance of improved phenotyping respiratory disorders (e.g.
asthma12 and bronchiectasis13) are
increasingly appreciated as such data will likely improve targeted
interventions. Indeed, to characterise clinical phenotypes (both during
the acute disease and future outcomes) of infants with bronchiolitis
using a multi-dimensional approach, Dumas and colleagues undertook two
important studies4. Using latent class analysis (LCA),
Dumas and colleagues described 4 phenotypes in a large USA cohort, of
which 2 were replicated (with the remaining 2 combined in a single
phenotype) in a Finnish cohort3. Later, using LCA and
they then described a profile (breathing problems/eczema in infancy and
non-respiratory syncytial virus [RSV], mostly human rhinoviruses
[hRV] infection) associated with increased risk of future asthma
phenotype (doctor-diagnosed asthma Hazard
Ratio=2.79;
95% confidence interval [CI] 1.78-4.39 by 3-years of
age)4. These infants also had “higher eosinophil
counts, higher cathelicidin levels, and increased proportions of
Haemophilus-dominant or Moraxella-dominant microbiota
profiles”4.
These longer-term outcomes of infants hospitalised with bronchiolitis
are of considerable interest with associations with
asthma14 and data linking viral agents (particularly
RSV and hRV)15 and/or nasopharyngeal
bacteria5 to future childhood asthma and
allergy16. However, while asthma and allergy are
relevant and important outcomes in mainstream settings, these are
inconsistently reported among and across different populations and
settings2,17. In settings where acute respiratory
infections are prevalent and bronchiectasis is relatively common, such
as among Indigenous children living in high-income
countries10, bronchiolitis is more common and severe
and their non-Indigenous counterparts11. In these
settings, the long-term outcomes are also likely different. Indeed,
Singleton and colleagues found that Alaskan Native children aged
<2-years hospitalized for RSV infection had increased risk for
chronic productive cough at 5 to 8 years of age and recurrent lower
respiratory infections, but not asthma18. Further, we
found that within 13-months post-hospitalisation for bronchiolitis,
Aboriginal and/or Torres Strait Islander (from here referred as
Indigenous) infants 19% (30/157) had bronchiectasis on chest
high-resolution computed tomography (HRCT) scan7.
A better understanding of clinical phenotypes including outcomes in
Indigenous settings are important, in the context of (a) Dumas and
colleagues’3,4 likely paradigm-changing findings for
future intervention studies; (b) the absence of further studies using
LCA to define these bronchiolitis clinical phenotypes; and (c) the
differential long-term outcomes in populations at high-risk of
bronchiectasis. We used LCA to analyse data from Indigenous infants from
Australia who participated in our previous prospective hospitalised
bronchiolitis studies19-21. We aimed to identify
distinct clinical profiles particularly
those
at-risk of future bronchiectasis.