Selection of Macrolide and Non-Macrolide Resistance with Mass Azithromycin Distribution: A Community-Randomized Trial

BACKGROUND: Biannual mass azithromycin distributions to preschool children for 2 years have been shown to reduce childhood mortality in sub-Saharan Africa, but at the cost of amplifying macrolide resistance. Here we investigated the gut resistome of children after 4 additional biannual distributions were given. METHODS: In the Niger site of the MORDOR (Macrolides Oraux pour Réduire les Décès avec un Oeil sur la Résistance) trial, 30 villages were enrolled in a sister trial in which they were randomized to mass distribution of either azithromycin or placebo every 6 months for 4 years, with treatments offered to all children 1 to 59 months of age. Rectal samples were collected at baseline, 36 months, and 48 months for gut resistome analysis. All field and laboratory personnel were masked to the participants’ original assignments. The primary outcome was the ratio in macrolide resistance determinants between treatment arms at 48 months. RESULTS: Over the entire 48-month period, mean (±SD) drug coverage was 86.6±12% in the placebo villages and 83.2±16.4% in the azithromycin villages. Macrolide resistance determinants were more common in the azithromycin arm compared to the placebo arm at 36 months (7.4-fold difference, 95% confidence interval 4.0 to 17.9-fold) and at 48 months (7.5-fold difference, 95% CI: 4.0 to 21.7-fold). Continued mass azithromycin distributions also selected for non-macrolide resistance determinants, including beta-lactams, the antibiotic class prescribed most frequently in this region. CONCLUSIONS: The study revealed that repeated mass azithromycin distributions may propagate antibiotic resistance. (ClinicalTrials.gov, NCT02047981)