Origin of the neonatal gut microbiota and probiotic intervention in the third trimester: a randomized controlled trial
In 30 mother-neonate dyads, maternal gut microbiota was the primary source of neonatal gut microbiota on days 3 and 14, but third-trimester maternal supplementation with Bifidobacterium and Lactobacillus did not significantly alter the abundance of these bacteria in maternal gut, vaginal, or placental microbiota at term.
| Population | 30 pregnant women in the third trimester and their neonates (mother-neonate dyads) |
|---|---|
| Intervention | Live combined Bifidobacterium and Lactobacillus tablets administered to the mother in the third trimester |
| Comparator | No intervention (control group) |
| Outcome | Source of neonatal gut microbiota (meconium); Source of neonatal gut microbiota (days 3 and 14); Bifidobacterium abundance in maternal microbiota at term (probiotic vs control); Lactobacillus abundance in maternal microbiota at term (probiotic vs control); Neonatal gut microbiota composition at day 14 (probiotic vs control); Neonatal microbiota beta diversity (NMDS) |
Summary of findings
| Outcome | Effect | 95% CI | Certainty | Clinical relevance | Notes |
|---|---|---|---|---|---|
| Source of neonatal gut microbiota (meconium) | SourceTracker proportional contribution: placenta > maternal gut (exact values not reported) | — | Low | — | 1 studies |
| Source of neonatal gut microbiota (days 3 and 14) | SourceTracker proportional contribution: maternal gut > placenta (exact values not reported) | — | Low | — | 1 studies |
| Bifidobacterium abundance in maternal microbiota at term (probiotic vs control) | No significant difference reported; p-value and effect size not provided | — | Low | — | 1 studies |
| Lactobacillus abundance in maternal microbiota at term (probiotic vs control) | No significant difference reported; p-value and effect size not provided | — | Low | — | 1 studies |
| Neonatal gut microbiota composition at day 14 (probiotic vs control) | Some taxa differed between groups; in the quantitative effect sizes or CI reported | — | Low | — | 1 studies |
| Neonatal microbiota beta diversity (NMDS) | Correlation between maternal and neonatal microbiota reported; in the numeric effect size or CI provided | — | Low | — | 1 studies |
Context
The origin of neonatal gut microbiota remains debated; identifying maternal sources and the impact of maternal probiotic interventions has implications for neonatal immune and metabolic health. Animal studies suggest vertical transmission of maternal bacteria to the fetal gastrointestinal tract, but human data are scarce. This study attempts to map these routes using 16S rRNA sequencing and source-tracking modeling.
What the study showed
SourceTracker indicated that the placenta contributed more to meconium microbiota, while maternal gut microbiota was the predominant source on neonatal days 3 and 14. In the probiotic group, there was no significant difference in Bifidobacterium, Lactobacillus, or Streptococcus abundance in maternal gut, vaginal, or placental microbiota at term compared to controls (specific p-values and effect sizes not reported). Some other bacteria differed between groups in maternal and neonatal microbiota, but taxa and magnitudes were not quantified in a standardized manner in the available summary.
How it was done
RCT with 30 mother-neonate dyads (1 excluded from 31 recruited), divided into probiotic and control groups; samples collected from maternal gut, vaginal, and placental microbiota at term, and neonatal microbiota in meconium, day 3 and day 14; analysis by 16S rRNA V4 region sequencing, NMDS, and SourceTracker. Intervention duration corresponded to the third trimester.
Effect magnitude
The study did not report quantitative effect sizes (RR, OR, SMD, MD) or 95% confidence intervals for primary outcomes; findings are descriptive and based on microbial composition analyses without standardized inferential statistics reported.
Risk of bias
Extremely small sample (n=30 dyads) severely limits statistical power and generalizability; absence of effect sizes, 95% CIs, and multiple comparison correction limits causal inference. Risk of bias tool (RoB 2) was not applied or reported; blinding of participants and assessors is not clearly described. 16S rRNA V4 analysis has limited taxonomic resolution at genus level, preventing strain-level identification of transmitted bacteria.
What this study does NOT prove
This study does not prove causality in vertical transmission of maternal bacteria to neonates, nor that maternal probiotics modify neonatal microbiota in a clinically relevant way. Results are not generalizable beyond the small, selected Chinese sample.
In clinical practice
The data do not support recommending maternal Bifidobacterium/Lactobacillus supplementation in the third trimester to modify gut, vaginal, or placental microbiota at term. Clinicians should not extrapolate these findings to clinical decisions; the study serves only as a hypothesis generator.
Limitations
Extremely small sample (n=30 dyads) severely limits statistical power and generalizability; absence of effect sizes, 95% CIs, and multiple comparison correction limits causal inference. Risk of bias tool (RoB 2) was not applied or reported; blinding of participants and assessors is not clearly described. 16S rRNA V4 analysis has limited taxonomic resolution at genus level, preventing strain-level identification of transmitted bacteria.
What is still missing
RCTs with adequate sample sizes (≥100 dyads per group), clinical outcomes (allergy, neonatal infection), and shotgun metagenomic sequencing to confirm strain-specific vertical transmission and causality.
Technical appendix
Version history
- 1.0 · 2026-06-24 — Auto-generated under Evidence Standard v1.0