Methods

Multicentre (8 CF centres in Australia and New Zealand), randomized controlled study.

Participants

170 Infants younger than 6 months age, with confirmed diagnosis of CF, diagnosed through newborn screening programs.

84 infants were randomized to receive BAL‐directed therapy (80 completed study) and 86 randomized to receive standard therapy (77 completed study).

BAL‐directed group

Mean age (SD) 3.8 (1.6) years.

Gender split: 44 male/40 female.

Mean (SD) weight at enrolment: 5.7 kg (1.40).

Number of participants with homozygous ΔF508 mutation: 57 (68%).

Number of participants with pancreatic insufficiency: 73 (87%).

Number of participants with meconium ileus: 17 (20%).

Number of participants born pre‐term (under 37 week gestation): 8 (10%).

History of exposure to tobacco smoke during pregnancy present in: 22 (26%).

History of concurrent smoking in the household present in: 30 (36%).

Standard therapy group

Mean age (SD) 3.7 (1.7) years.

Gender split: 44 male/42 female.

Mean weight (SD) at enrolment: 5.6 kg (1.5).

Number of participants with homozygous ΔF508 mutation: 54 (64%).

Number of participants with pancreatic insufficiency: 71 (85%).

Number of participants with meconium ileus: 16 (19%).

Number of participants born pre‐term (under 37‐week gestation): 9 (11%).

History of exposure to tobacco smoke during pregnancy present in: 13 (15%).

History of concurrent smoking in the household present in: 23 (28%).

Interventions

Intervention: BAL‐directed therapy for pulmonary exacerbations until age 5 years.

The participants in the BAL‐directed therapy groups underwent BAL at following times:

1. before 6 months age when well;

2. when hospitalized for a pulmonary exacerbation (unwell with change in respiratory symptoms from baseline);

3. if P. aeruginosa was cultured from oropharyngeal specimens;

4. following P. aeruginosa eradication therapy.

Control: Standard therapy (directed by clinical features and oropharyngeal swab cultures) for pulmonary exacerbations until age 5 years.

The standard therapy included oropharyngeal swab at following time points:

1. when a child was unwell with a change in respiratory symptoms from baseline (pulmonary exacerbation);

2. at the end of the antibiotic eradication treatment for P. aeruginosa

Outcomes

Reported at at 5 years age.

Primary outcome measures

• Prevalence of P. aeruginosa on BAL cultures (defined as ≥ 10³ CFU/ml)

• Total CF‐CT score (as percentage of the maximum score) on high resolution chest CT scan

Secondary outcome measures

• Weight z score

• Height z score

• BMI z score

• Lung function parameters: standard spirometry measures (FEV₁, FVC)

• CF‐CT components: bronchiectasis score, parenchymal disease score, mucus plugging score, airway wall thickening score and air trapping score

• Respiratory exacerbation rate

• Number and duration of hospitalizations for respiratory exacerbations not associated with P. aeruginosa infection

• Number of episodes of P. aeruginosa infection per child per year

• Final BAL microbiology and inflammatory indices

Notes

The BAL‐directed therapy group had BAL before age 6 months when well, when hospitalized for pulmonary exacerbations, when P. aeruginosa was cultured from their oropharyngeal specimens and following P. aeruginosa therapy.

The standard therapy included taking oropharyngeal swabs when having pulmonary exacerbation and at the end of antibiotic therapy. Children in both groups had BAL and HRCT scan of chest at 5 years age.

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

After consent, the participants were randomly assigned in 1:1 ratio to 2 groups by a centralized computer‐generated schedule with stratification by site and sex.

Allocation concealment (selection bias)

Low risk

The allocation was done by a centralized computer‐generated schedule; the randomization key was concealed and held remotely. Allocation was revealed by telephone after confirmed recruitment.

Incomplete outcome data (attrition bias)
All outcomes

Low risk

Although the study was set up to be analysed on intention‐to‐treat basis, the participants with missing outcomes were not included in the primary analysis. The risk of bias is considered moderate to low as less than 10% of the data were missing and the reasons of exclusions were balanced across both groups.

Selective reporting (reporting bias)

Low risk

The economic analysis was planned but was not reported initially but now has been published. It has been included in the updated version of the review. All other outcomes planned to be assessed in the protocol were reported.

Other bias

Low risk

No other potential source of bias was identified.

Blinding of participants and personnel (performance bias)
All outcomes

Low risk

The participants and the personnel were not blinded to the randomization (which might not have been possible in this study setting). However, the risk of bias is low as the primary outcome measures were unlikely to be influenced by the lack of blinding.

Blinding of outcome assessment (detection bias)
All outcomes

Low risk

Risk of bias is low as the outcome assessors were blinded for both the primary outcome measures.