The effect of antibiotics on the intestinal microbiota in children - a systematic review

Abstract

Background: Children are the age group with the highest exposure to antibiotics (ABX). ABX treatment changes the composition of the intestinal microbiota. The first few years of life are crucial for the establishment of a healthy microbiota and consequently, disturbance of the microbiota during this critical period may have far-reaching consequences. In this review, we summarise studies that have investigated the effect of ABX on the composition of the intestinal microbiota in children.

Methods: According to the PRISMA guidelines, a systematic search was done using MEDLINE and Embase to identify original studies that have investigated the effect of systemic ABX on the composition of the intestinal microbiota in children.

Results: We identified 89 studies investigating a total of 9,712 children (including 4,574 controls) and 14,845 samples. All ABX investigated resulted in a reduction in alpha diversity, either when comparing samples before and after ABX or children with ABX and controls. Following treatment with penicillins, the decrease in alpha diversity persisted for up to 6-12 months and with macrolides, up to the latest follow-up at 12-24 months. After ABX in the neonatal period, a decrease in alpha diversity was still found at 36 months. Treatment with penicillins, penicillins plus gentamicin, cephalosporins, carbapenems, macrolides, and aminoglycosides, but not trimethoprim/sulfamethoxazole, was associated with decreased abundances of beneficial bacteria including Actinobacteria, Bifidobacteriales, Bifidobacteriaceae, and/or Bifidobacterium, and Lactobacillus. The direction of change in the abundance of Enterobacteriaceae varied with ABX classes, but an increase in Enterobacteriaceae other than Escherichia coli was frequently observed.

Conclusion: ABX have profound effects on the intestinal microbiota of children, with notable differences between ABX classes. Macrolides have the most substantial impact while trimethoprim/sulfamethoxazole has the least pronounced effect.

Keywords: 16S rRNA; cephalosporin; intestine; macrolide; microbiome; penicillin; sequencing; stool.

Figure 1

Selection of studies.

Figure 2

Duration of decreased alpha diversity after ABX reported in different studies. Red bars depict duration of decreased alpha diversity in ABX group compared to controls (red bars); black bars depict the duration of follow-up after ABX. Studies specifying Shannon index, inverse Simpson index or richness for the ABX and control group were included. Studies which did not specify the time points when the alpha diversity was investigated were excluded.

Figure 3

Differences in bacterial abundance between children treated with penicillins and controls. Included studies investigated amoxicillin (71, 89) and amoxicillin with or without clavulanate and penicillin V (70). Studies which did not compare ABX group to controls and studies not providing p-values were excluded.

Figure 4

Differences in bacterial abundance or colonization rate between children treated with penicillins plus aminoglycosides and controls. Studies included investigated ampicillin plus gentamicin (52, 58, 69, 81, 93, 94), penicillin plus gentamicin (81, 83), and penicillin plus tobramycin (26). Studies which did not compare ABX group to controls and studies not providing p-values were excluded.

Figure 5

Differences in bacterial abundance or total bacterial count between children treated with macrolides and controls. Included studies investigated azithromycin (80, 95, 103), azithromycin and clarithromycin (70) or did not separately analyse different macrolides (71). Studies which did not compare ABX group to controls and studies not providing p-values were excluded.

Figure 6

Differences in bacterial abundance or colonisation rate between ABX groups and controls of studies investigating various ABX without separate analysis of individual ABX. Included studies investigated: ampicillin, nafcillin, gentamicin, tobramycin, others (97), penicillin (plus aminoglycoside), others (92), ampicillin/sulbactam, cefotaxime (72), penicillin plus gentamicin, amoxicillin plus gentamicin, amoxicillin plus ceftazidime, others (48), ampicillin and cefotaxime, amikacin, vancomycin, others (65), ampicillin, cefotaxime, gentamicin, others (54), cephalosporin, penicillin, others (113), penicillins, macrolides, cephalosporins, others (115), beta-lactams, aminoglycosides, vancomycin, others (73), benzylpenicillin, cloxacillin, flucloxacillin, others (32), benzylpenicillin, cloxacillin, flucloxacillin, others (31), penicillin, penicillin plus gentamicin, others (33), or did not specify which ABX they investigated (28, 29, 42, 56, 61, 82, 98). Studies which did not compare ABX group to controls and studies not providing p-values were excluded.