AIR.

Methods	Design: randomised controlled trial (NCT00214526) Number of participating centres: 11 from Canada, United Kingdom, Brazil and Denmark
Participants	Inclusion criteria: Airflow obstruction (pre‐bronchodilator forced expiratory volume in one second (FEV1) of 60% to 85% of predicted value) Airway hyperresponsiveness (provocative concentration of methacholine required to lower the FEV1 by 20% (PC20) of less than 8 mg/mL) Stable asthma during the six weeks before enrolment (absence of unscheduled physician visits for asthma care, unchanged use of maintenance asthma medication and stable use of rescue medication) Worsening asthma control after abstention from LABA for two weeks (increase in ACQ score ≥ 0.5, or a decline of 5% in PEF during second week) Exclusion criteria: ≥ Three lower respiratory tract infections requiring antibiotics during previous 12 months, or Respiratory tract infection within previous six weeks No. of randomly assigned participants: 112 (56 BT vs 56 control group) Age (mean (SD)): BT group: 39.36 (11.18) versus control group: 41.65 (11.35) Sex (% male): BT group: 44 versus control group: 43 PC20 mg/mL (geometric mean (95% CI)): BT group: 0.25 (0.16 to 0.40) versus control group: 0.35 (0.23 to 0.52) Pre‐bronchodilator FEV1 (% predicted; mean (SD)): BT group: 72.65 (10.41) versus control group: 76.12 (9.28) Dose ICS (µg; mean (SD)): BT group: 1351 (963) versus control group: 1264 (916) Dose LABA (µg; mean (SD)): BT group: 111.3 (35.9) versus control group: 105.8 (30.8) Asthma severity (number of participants with moderate/severe asthma, according to the guidelines of the Global Initiative for Asthma): BT group: 21/26, respectively, versus control group: 34/28, respectively AQLQ score: BT group: 5.58 (1.05) versus control group: 5.72 (0.94)
Interventions	Intervention: Bronchial thermoplasty plus medical management. Three bronchial thermoplasty procedures performed three weeks apart Control: medical management
Outcomes	Main outcome: Mild exacerbation rate (change from baseline) Secondary outcomes: Severe exacerbation rate Pre‐bronchodilator FEV1 (percentage predicted) Post‐bronchodilator FEV1 (percentage predicted) Peak expiratory flow (morning and evening) (change from baseline) Asthma Control Questionnaire (ACQ) score (change from baseline) Use of rescue medications Use of maintenance medications Asthma Quality of Life Questionnaire (AQLQ) score (change from baseline) Total symptom score (change from baseline) Percent of symptom‐free days (change from baseline) Respiratory adverse events during treatment period Hospitalisations during treatment period for respiratory adverse events (percentage of participants) Hospitalisations during post‐treatment period (number of participants)
Notes	Funded by Asthmatx Inc
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Centrally generated computer randomisation sequence in blocks of four participants
Allocation concealment (selection bias)	Low risk	Randomisations provided in sealed envelopes to each participating centre
Blinding of participants and personnel (performance bias) All outcomes	High risk	Open trial
Blinding of outcome assessment (detection bias) All outcomes	High risk	Outcomes collected by participants in diaries monitored by the research staff at medical meetings or in telephone interviews
Incomplete outcome data (attrition bias) All outcomes	Low risk	Intention‐to‐treat analysis planned with no imputation of missing data 112 participants randomly assigned (56 to each group, but baseline data reported for 109 participants) Participants lost to follow‐up (three‐month visit): four participants in the bronchial thermoplasty group versus eight in the control group (three and four participants, respectively, withdrew consent; one and four participants, respectively, were lost to follow‐up) Participants lost to follow‐up (12‐month visit): complete data for 52 participants in the bronchial thermoplasty group versus 49 in the medical management group
Selective reporting (reporting bias)	Low risk	All outcomes described in the study protocol (NCT00214526) reported in the main publication (Cox 2007) Most trial results reported in graphics, but complementary data on electronic appendices allow a more comprehensive analysis of the results