I read with interest the reply of the AIR2 trial authors entitled “Bronchial thermoplasty: ready for prime time—evidence is there!” One of the major findings in the AIR2 trial reported by Castro et al was the statistically significant difference in the rate of severe asthma exacerbations per subject per year between the bronchial thermoplasty arm and the sham arm. It is important to observe that the two groups were not matched in their baseline exacerbation rate per subject per year (see Table 1 in the AIR2 trial). Therefore, the difference in the asthma exacerbation rate could simply be a result of the fact that more patients with frequent baseline exacerbations (ie, frequent exacerbators) were included in the sham arm than in the bronchial thermoplasty arm. It remains unclear why important baseline characteristics such as number of exacerbations per subject per year, ED visits in the previous year, and work absenteeism were not considered in the baseline patient characteristics, while comparison between the two arms with regard to these outcomes was planned for and conducted. If there was an intention to measure these outcomes and compare them between the two groups, the two arms should have been matched in these characteristics as well.