Effect of Commonly Used Pediatric Antibiotics on Gut Microbial Diversity in Preschool Children in Burkina Faso: A Randomized Clinical Trial

Abstract

Background: Exposure to antibiotics may result in alterations to the composition of intestinal microbiota. However, few trials have been conducted, and observational studies are subject to confounding by indication. We conducted a randomized controlled trial to determine the effect of 3 commonly used pediatric antibiotics on the intestinal microbiome in healthy preschool children.

Methods: Children aged 6-59 months were randomized (1:1:1:1) to a 5-day course of 1 of 3 antibiotics, including amoxicillin (25 mg/kg/d twice-daily doses), azithromycin (10 mg/kg dose on day 1 and then 5 mg/kg once daily for 4 days), cotrimoxazole (240 mg once daily), or placebo. Rectal swabs were obtained at baseline and 5 days after the last dose and were processed using 16S rRNA gene sequencing. The prespecified primary outcome was inverse Simpson's α-diversity index.

Results: Post-treatment Simpson's diversity was significantly different across the 4 arms (P = .003). The mean Simpson's α-diversity among azithromycin-treated children was significantly lower than in placebo-treated children (6.6; 95% confidence interval [CI], 5.5-7.8; vs 9.8; 95% CI, 8.7-10.9; P = .0001). Diversity in children treated with amoxicillin (8.3; 95% CI, 7.0-9.6; P = .09) or cotrimoxazole (8.3; 95% CI, 8.2-9.7; P = .08) was not significantly different than placebo.

Conclusions: Azithromycin affects the composition of the pediatric intestinal microbiome. The effect of amoxicillin and cotrimoxazole on microbiome composition was less clear.

Clinical trials registration: clinicaltrials.gov NCT03187834.

Keywords: antibiotics; microbiome; randomized controlled trial; sub-Saharan Africa.

Figure 1.

Consolidated Standards for Reporting of Trials study flow diagram.

Figure 2.

Distributions of Simpson’s (A) and Shannon’s (B) alpha diversity indices at baseline and 5 days after the final study treatment for children treated with placebo (black lines), azithromycin (red lines), amoxicillin (blue lines), and cotrimoxazole (green lines).

Figure 3.

Principal coordinates analysis (PCoA) plots comparing the Euclidean distance between placebo and azithromycin (A), placebo and amoxicillin (B), and placebo and cotrimoxazole (C). Centroids are depicted with square points.