Factors associated with "Frequent Exacerbator" phenotype in children with bronchiectasis: The first report on children from the Australian Bronchiectasis Registry.

Introduction: In adults with bronchiectasis, multicentre data advanced the field including disease characterisation and derivation of phenotypes such as 'frequent exacerbator (FE)' (≥3 exacerbations/year). However, paediatric cohorts are largely limited to single centres and no scientifically derived phenotypes of paediatric bronchiectasis yet exists. Using paediatric data from the Australian Bronchiectasis Registry (ABR), we aimed to: (a) describe the clinical characteristics and compare Indigenous with non-Indigenous children, and (b) determine if a FE phenotype can be identified and if so, its associated factors.

Methods: We retrieved data of children (aged <18-years) with radiologically confirmed bronchiectasis, enrolled between March 2016-March 2020.

Results: Across five sites, 540 children [288 Indigenous; median age = 8-years (IQR 6-11)] were included. Baseline characteristics revealed past infection/idiopathic was the commonest (70%) underlying aetiology, most had cylindrical bronchiectasis and normal spirometry. Indigenous children (vs. non-Indigenous) had significantly more environmental tobacco smoke exposure (84% vs 32%, p < 0.0001) and lower birth weight (2797 g vs 3260 g, p < 0.0001). FE phenotype present in 162 (30%) children, was associated with being younger (OR = 0.85, 95%CI 0.81-0.90), more recent diagnosis of bronchiectasis (OR = 0.67; 95%CI 0.60-0.75), recent hospitalization (OR = 4.51; 95%CI 2.45-8.54) and Pseudomonas aeruginosa (PsA) infection (OR = 2.43; 95%CI 1.01-5.78). The FE phenotype were less likely to be Indigenous (OR = 0.14; 95%CI 0.03-0.65).

Conclusion: Even within a single country, the characteristics of children with bronchiectasis differ among cohorts. A paediatric FE phenotype exists and is characterised by being younger with a more recent diagnosis, PsA infection and previous hospitalization. Prospective data to consolidate our findings characterising childhood bronchiectasis phenotypes are required.