S1 Lower airway pathological phenotypes do not relate to clinical phenotypes in preschool children with severe, recurrent wheeze

Abstract

Background Recurrent wheeze attacks are a leading cause of hospital admissions in preschool children. Management to prevent attacks is predominantly determined by clinical phenotypes (episodic viral wheeze and multiple trigger wheeze), which rely on accurate parental recall of symptom pattern. Management strategies based on more objective measures are needed to reduce healthcare burden. Atopy and bacterial airway infection are two objective measures which may be useful to guide the management of preschool wheeze. We hypothesised that during stable disease, atopic children with severe, recurrent wheeze would have blood and airway eosinophilia, while those with bacterial airway infection would be neutrophilic. We investigated peripheral blood and airway granulocyte inflammation, and airway remodelling, in severe, recurrent preschool wheeze, and related pathology to clinical phenotype, atopy and bacterial airway infection. Methods Preschool children with severe, recurrent wheeze (n=24; median age 30.9 months) and non-wheezing controls (n=30; median age 61.9 months) undergoing clinically indicated bronchoscopy were included. Granulocyte inflammation (neutrophils and eosinophils) was analysed in peripheral blood, bronchoalveolar lavage (BAL) and endobronchial biopsies. Airway remodelling (airway smooth muscle (ASM) volume fraction and reticular basement membrane (RBM) thickness) was quantified in endobronchial biopsies. Inflammation and remodelling were related to atopic status (elevated specific IgE to aeroallergens), bacterial airway infection and clinical phenotype. Results Neither peripheral nor airway granulocyte inflammation were different between the clinical phenotypes. Atopic preschool wheezers had increased BAL eosinophils (median, 6.7% [2.9–11.7]) compared to non-atopic wheezers (median, 0.7% [0.4–3.2]; p=0.03) and controls (median, 1.0% [0.2–2.7]; p=0.02). Wheezers with positive BAL bacterial culture had increased BAL neutrophils (median, 22.7% [6.3–47.2]) compared to culture negative wheezers (median, 3.7% [1.3–7.1]; p=0.01). Blood inflammation did not correlate with BAL or biopsy inflammation. ASM volume fraction was increased in atopic wheezers (median, 12.9% [7.3–15.9]) compared to controls (median, 6.6% [4.3–9.5]; p=0.04). Conclusions In contrast to clinical phenotypes, stratifying preschool children with severe, recurrent wheeze by atopy and bacterial airway infection distinguished pathological phenotypes. Blood inflammation did not reflect airway inflammation in severe preschool wheeze. Prevention of preschool wheeze attacks may be more effective with management of pathological phenotype, rather than clinical phenotype.