Effect of Biannual Mass Azithromycin Distributions to Preschool-Aged Children on Trachoma Prevalence in Niger: A Cluster Randomized Clinical Trial

Abstract

Importance: Because transmission of ocular strains of Chlamydia trachomatis is greatest among preschool-aged children, limiting azithromycin distributions to this age group may conserve resources and result in less antimicrobial resistance, which is a potential advantage in areas with hypoendemic trachoma and limited resources.

Objective: To determine the efficacy of mass azithromycin distributions to preschool-aged children as a strategy for trachoma elimination in areas with hypoendemic disease.

Design, setting, and participants: In this cluster randomized clinical trial performed from November 23, 2014, until July 31, 2017, thirty rural communities in Niger were randomized at a 1:1 ratio to biannual mass distributions of either azithromycin or placebo to children aged 1 to 59 months. Participants and study personnel were masked to treatment allocation. Data analyses for trachoma outcomes were performed from October 19, 2021, through June 10, 2022.

Interventions: Every 6 months, a single dose of either oral azithromycin (20 mg/kg using height-based approximation for children who could stand or weight calculation for small children) or oral placebo was provided to all children aged 1 to 59 months.

Main outcomes and measures: Trachoma was a prespecified outcome of the trial, assessed as the community-level prevalence of trachomatous inflammation-follicular and trachomatous inflammation-intense through masked grading of conjunctival photographs from a random sample of 40 children per community each year during the 2-year study period. A secondary outcome was the seroprevalence of antibodies to C trachomatis antigens.

Results: At baseline, 4726 children in 30 communities were included; 1695 children were enrolled in 15 azithromycin communities and 3031 children were enrolled in 15 placebo communities (mean [SD] proportions of boys, 51.8% [4.7%] vs 52.0% [4.2%]; mean [SD] age, 30.8 [2.8] vs 30.6 [2.6] months). The mean coverage of study drug for the 4 treatments was 79% (95% CI, 75%-83%) in the azithromycin group and 82% (95% CI, 79%-85%) in the placebo group. The mean prevalence of trachomatous inflammation-follicular at baseline was 1.9% (95% CI, 0.5%-3.5%) in the azithromycin group and 0.9% (95% CI, 0-1.9%) in the placebo group. At 24 months, trachomatous inflammation-follicular prevalence was 0.2% (95% CI, 0-0.5%) in the azithromycin group and 0.8% (95% CI, 0.2%-1.6%) in the placebo group (incidence rate ratio adjusted for baseline: 0.18 [95% CI, 0.01-1.20]; permutation P = .07).

Conclusions and relevance: The findings of this trial do not show that biannual mass azithromycin distributions to preschool-aged children were more effective than placebo, although the underlying prevalence of trachoma was low. The sustained absence of trachoma even in the placebo group suggests that trachoma may have been eliminated as a public health problem in this part of Niger.

Trial registration: ClinicalTrials.gov Identifier: NCT02048007.

Conflict of interest statement

Conflict of Interest Disclosures: Dr West reported receiving grants from the Bill & Melinda Gates Foundation during the conduct of the study. Dr Arnold reported receiving grants from the Bill & Melinda Gates Foundation and the National Institutes of Health (NIH) during the conduct of the study. Dr Porco reported receiving grants from the NIH during the conduct of the study. Dr Lietman reported receiving grants from the Bill & Melinda Gates Foundation during the conduct of the study. No other disclosures were reported.

Figures

Figure 1.. Study Flow Diagram

All communities in the Loga and Boboye departments of Niger were screened for eligibility, with 86 excluded because the population was not in the desired range (ie, 200-2000 people). Of the remaining eligible clusters, 1 was allocated for an ancillary trial on antibiotic resistance, 615 were randomly chosen for the main child mortality trial, and 30 were randomly selected for the present trial. The 30 communities were randomized to biannual mass administration with either azithromycin or placebo for 2 years. Repeated cross-sectional random samples of children were selected from the preceding study census at months 0, 12, and 24 for conjunctival photography.

Figure 2.. Longitudinal Prevalence of Trachoma

Each thin line represents the prevalence of a study cluster over time, and the thick line represents the mean prevalence across all communities in the treatment group. A and B, Prevalence of trachomatous inflammation–follicular and trachomatous inflammation–intense among children aged 1 to 59 months; C and D, seroprevalence of pgp3 and CT694 antibodies among children aged 12 to 59 months.

Figure 3.. Chlamydia Antibody Responses Across Assessments

Immunoglobulin G responses to CT694 and pgp3 are shown separately as smoothed curves of the continuous antibody response across ages, with the antibody response expressed as the log10 median fluorescence intensity minus background. Shaded bars represent approximate simultaneous CIs. Children in the azithromycin group are shown in blue and children in the placebo group are shown in orange.