Preschool Wheeze: Inflammation/Infection Guided Management

This study will determine the efficacy of treating preschool wheeze based on objective biomarkers of inflammation and presence of bacterial infection, and compare this to current clinical guidelines based management. Subjects will be recruited from out-patient clinics and attend for a screening/research visit during which assessments of pathological phenotype (bacterial infection and inflammation) will be made from induced sputum and blood. Their clinical phenotype will be determined by the Consultant. Subsequently, patients will be randomised to one of two arms:

1. current clinical care ( as directed by their Consultant) ii) pathological phenotype based management- presence of eosinophilia in blood (_> 2%) or induced sputum (>2.4%) - will receive twice daily inhaled steroids ( beclometasone 200mcg bd) for 4 months, presence of bacterial infection in induced sputum or oropharyngeal swab- will receive 4 weeks of antibiotics targeted to the isolate, IF BOTH eosinophilic inflammation AND infection are present, then inflammation alone will be treated for 4 months.

All patients will continue on as required bronchodilator therapy, and any other anti-inflammatory therapy ( other than inhaled steroids) e.g. montelukast.

All participants will undergo an initial research/screening visit which will involve the following tests:

Full blood count Total serum immunoglobulin E and radioallergosorbent test to house dust mite, grass pollen, tree pollen, cat, dog, aspergillus, milk, egg, peanut Assessment of lung function by incentive spirometry and lung clearance index (LCI) using the multiple breath washout technique Exhaled nitric oxide- using the offline tidal breathing technique Sputum induction using nebulised hypertonic saline Oro- pharyngeal swab Symptom questionnaire Health related quality of life questionnaire

Management in the pathological phenotype arm:

1. Eosinophilic inflammation, no bacterial infection: Children with sputum eosinophils >2.4% or blood eosinophils equal to and above 2% will be defined as eosinophilic and management will be as follows:
2. Regular inhaled steroids ( beclometasone 200mcg bd) for 4 months in those previously not on any regular therapy. ii) Of children who have already been prescribed inhaled steroid therapy, those with an eosinophilic phenotype will continue for a further 4 months. iii) Children who have already been prescribed a regular leukotriene receptor antagonist will continue this and regular inhaled steroids (beclometasone 200mcg bd) will be added to their maintenance therapy for 4 months. iv) Use of as required bronchodilator therapy will continue
3. Non- eosinophilic, no inflammation, no infection: Children with either sputum eosinophils _<2.4% or blood eosinophils <2% will be defined as non- eosinophilic and management will be as follows:
4. Use of as required bronchodilators for acute symptoms ii) Non- eosinophilic children that have already been prescribed regular inhaled steroids will be asked to stop them for 4 months Children with no bacterial growth from oropharyngeal swab or induced sputum, and without eosinophilic inflammation will continue management according to the non-eosinophilic, no inflammation, no infection profile.
5. Bacterial infection, no eosinophilic inflammation: Children with a positive bacterial culture result from oropharyngeal swab or induced sputum will be treated with a 4 week course of antibiotics. Four weeks of antibiotic therapy will be used as this has been shown to be beneficial in a previous study of severe preschool wheeze. If these children have already been prescribed regular inhaled steroids, they will be stopped for 4 months.
6. Eosinophilic inflammation AND bacterial infection:

Children with raised eosinophils in induced sputum or blood AND positive bacterial culture will be managed according to the eosinophilic inflammation alone guideline. They will only be given inhaled steroids for 4 months, and no antibiotics. This is to assess only one intervention at a time. However, the investigators do not anticipate many children will be in this group as our previous data shows those with positive bacterial cultures are very unlikely to have eosinophilia.

Management in the clinical guidelines arm:

The children will be treated as directed by their Consultant Paediatrician. The prescription for the drugs to be used will be dispensed from the pharmacy department at the Royal Brompton Hospital- as directed by the clinician. Usually only 2 weeks of medication is prescribed by the pharmacy, but for this trial 4 months' supply will be given.

Concomitant treatment

The following medication will be allowed to continue during the intervention period in both study arms:

1. Use of rescue medication with bronchodilators for acute symptoms- salbutamol and/or ipratropium bromide
2. Use of leukotriene receptor antagonist- montelukast- can continue regardless of whether it is being taken regularly or as required for acute symptoms

Monitoring:

The trial will be monitored according to the monitoring plan agreed and written by the Sponsor, based on internal risk assessment procedure. Where appropriate the CI will be asked to complete a copy of the Sponsor's self-monitoring template. It is the responsibility of the CI to ensure this is completed and submitted to the research organisation on request. It is the responsibility of the research organisation to determine the monitoring risk assessment and explain the rationale.

The study may be subject to inspection and audit by Imperial College London under their remit as Sponsor, the Study Coordination Centre and other regulatory bodies to ensure adherence to good clinical practice.

Statistics and Data analysis:

Categorical data will be presented as number and percentage and comparisons between groups done using the chi squared or Fishers exact test. All numerical data will be tested for normality and normally distributed data will be presented as mean standard deviation and comparisons between groups done using the 2 sample independent t test. For the data that are not normally distributed the median (IQR) will be presented and comparisons between groups done using the Mann-Whitney test. Poisson regression will be done to determine the factors that affect the primary endpoint, the number of unplanned hospital visits.

There will be data checks to compare data entered into the registry, against predefined rules for range using the INFORM database.

Type of subjects to be recruited:

All subjects will be aged between 1 and 5 years old. All subjects will have moderate to severe preschool wheeze, defined as _>2 oral steroid bursts for acute wheeze in the last 12 months, with at least one oral steroid burst in the last 6 months.

Number of subjects:

Based on published data for children with moderate preschool wheeze it is apparent that approximately five healthcare attendances occur per child per year. In the current study, the investigators wish to reduce the proportion of healthcare contacts by at least one-third per year. In order to achieve this with 80% power, and accepting statistical significance at the 5% level, the investigators require a minimum of 36 evaluable patients with moderate wheeze per group. Evaluable patients are those from whom at least a blood sample to assess inflammation and an oropharyngeal swab to determine infection can be obtained prior to randomisation. A total of at least 72 children will therefore be recruited. However, this is the minimum number, and the investigators aim to recruit 100 children (50 per group) if possible.

The number of subjects is not based on a precise power calculation as this is a proof of concept study to determine the feasibility of this approach, and to inform a future power calculation for a large multi-centre trial addressing the same question.