Background Patients with severe asthma are at increased risk of exacerbations and lower respiratory tract infections (LRTI). Severe asthma is heterogeneous, encompassing eosinophilic and non-eosinophilic (mainly neutrophilic) phenotypes. Patients with neutropilic airway diseases may benefit from macrolides. Methods We performed a randomised double-blind placebo-controlled trial in subjects with exacerbationprone severe asthma. Subjects received low-dose azithromycin (n=55) or placebo (n=54) as add-on treatment to combination therapy of inhaled corticosteroids and long-acting β 2 agonists for 6 months. The primary outcome was the rate of severe exacerbations and LRTI requiring treatment with antibiotics during the 26-week treatment phase. Secondary efficacy outcomes included lung function and scores on the Asthma Control Questionnaire (ACQ) and Asthma Quality of Life Questionnaire (AQLQ). Results The rate of primary endpoints (PEPs) during 6 months was not significantly different between the two treatment groups: 0.75 PEPs (95% CI 0.55 to 1.01) per subject in the azithromycin group versus 0.81 PEPs (95% CI 0.61 to 1.09) in the placebo group (p=0.682). In a predefined subgroup analysis according to the inflammatory phenotype, azithromycin was associated with a significantly lower PEP rate than placebo in subjects with noneosinophilic severe asthma (blood eosinophilia ≤200/ml): 0.44 PEPs (95% CI 0.25 to 0.78) versus 1.03 PEPs (95% CI 0.72 to 1.48) (p=0.013). Azithromycin significantly improved the AQLQ score but there were no significant between-group differences in the ACQ score or lung function. Azithromycin was well tolerated, but was associated with increased oropharyngeal carriage of macrolide-resistant streptococci. Conclusions Azithromycin did not reduce the rate of severe exacerbations and LRTI in patients with severe asthma. However, the significant reduction in the PEP rate in azithromycin-treated patients with non-eosinophilic severe asthma warrants further study. ClinicalTrials.gov number NCT00760838.