S64 Use of pathological phenotype to determine optimal management for moderate to severe preschool wheeze

Introduction Persistent episodes of wheezing are common in preschool children, however we have few effective therapies. We hypothesised that objective biomarker based management of preschool wheeze would be superior to current clinical guidelines. Methods A single-centre randomised, controlled trial in children aged 1–5 years with moderate to severe recurrent wheeze, requiring at least 2 admissions ± short courses of oral steroids in the last 12 months, with at least one in the last 6 months. Children were recruited from September-April over a 3 year period. Clinical (episodic viral wheeze [EVW], or multiple trigger wheeze [MTW]1) and pathological phenotypes based on blood eosinophilia >3%, or bacterial infection in sputum or cough swab were determined at recruitment. Children were randomised to pathological phenotype based management (beclomethasone 400 mcg/day if blood eosinophils >3%, or targeted antibiotics if positive culture on sputum/cough swab) or clinician directed care (control arm) for 4 months. Primary outcome was number of unscheduled healthcare visits (UHCVs). Daily symptoms were reported via a text message system. Patients treated with inhaled corticosteroids (ICS) had adherence assessed using an electronic monitoring device. Results 60 children were randomised, 30 in each group. Baseline blood eosinophils were similar in the two groups (5.18% control, 5.15% intervention). 6 children had positive sputum cultures. 38/60 had EVW and 22/60 had MTW. Prevalence of clinical phenotypes was similar in both groups (control-EVW 18/30, MTW 12/30; intervention- EVW 20/30, MTW 10/30). In both groups 20/30 (67%) were prescribed ICS, with median adherence 67% (range 0–91%). There was no significant difference in the rate of UHCVs or symptoms between the two groups (p=0.46). Conclusions Phenotype based management of children with moderate to severe preschool wheeze did not result in a significant reduction in UHCVs compared to clinical guideline based management. However, 56% of children in the control group with EVW were prescribed ICS by their clinician even though this is not recommended in clinical guidelines and 80% of those with EVW in the pathological phenotype had blood eosinophilia, suggesting little relationship between clinical phenotype and objective biomarkers to guide ICS prescription. Reference Brand PJ, et al. Eur Respir J 2008