Page 27 - Read Online

Horwell et al. Microbiome Res Rep 2025;4:1 https://dx.doi.org/10.20517/mrr.2024.32 Page 5 of 20

Bifidobacterium has been shown in both human and animal studies to reduce the risk of necrotising
[71]
enterocolitis in newborns .
This microbiomic theory of disease has a large and convincing body of evidence to support it. However,
when looking at the literature, there are two important deficits that require attention. Firstly, the majority of
studies showing metagenomic differences in VD and CS children, while replicated many times, were not
designed to determine why these differences occur. Are the differences related to the method of delivery
itself - i.e., is the vaginal microbiome the only variable, or are there other unrecognised environmental
factors at play? Studies traditionally have used genetic operational taxonomic units (OTUs) to show
significant similarity in intra-mother-child dyads vs. inter-dyads ; thereby, evidence of VD vertical
[72]
transmission is only strongly inferred. To the best of our knowledge, there have been no mechanistic studies
(e.g., using tagged bacterial species) to definitively show colonisation from the vaginal microbiome to the
neonate’s GIT. Indeed, Bacteroides fragilis, the species that correlates strongly with VD, is only found in

6.6% of the vaginal microbiome of pregnant women, suggesting inoculation from an environment
other than the vagina . Interestingly, studies using cultured vaginal bacteria from the mother,
[73]
administered orally to the neonate shortly after CS, have reported both positive and negative effects
on the infant’s microbiome compared to placebo, raising questions over the extent played by
VD in forming the newborn’s microbiome [74,75] . In a similar vein, there is an increasing trend of the lay
press in the advocation of “vaginal seeding”, inoculating a swab or gauze with vaginal fluid to transfer the
flora to a newborn. More research is urgently needed to assess not only the efficacy of this but also to
assess for any safety concerns, a view reflected by the American College of Obstetricians and
Gynaecologists . Another variable that may influence the paradigm is the routine use of preoperative
[76]
[77]
antibiotics during CS, which has significant effects on the neonatal microbiome . The evidence suggests
that the routine use of pre-cord clamping antibiotics significantly reduces the abundance of bacteria,
[77]
predominantly from reduced Clostridium spp., an effect that can still be observed after one year of age .
Interest in antibiotic usage after cord clamping is increasing internationally, where the antibiotic
compound does not reach the foetus. Randomised control trials of this method have shown that it
successfully mitigates the effects on the neonatal microbiome . Furthermore, a Swiss lead non-inferiority
[77]
trial found that the technique had no negative effect on the mother (e.g., surgical site infection), and its
routine adoption into clinical practice should be encouraged . However, it should be noted that there is
[78]
conflicting evidence on this; a Dutch group has recently reported no significant difference in
microbiomic composition between infants born to CS with antibiotics given pre- vs. post-cord clamping .
[79]
Finally, maternal factors seem to influence microbiome dynamics, with one study showing a significant
difference in children born by emergency CS compared to elective CS, suggesting that antenatal health
regulates the microbiomic outcome . Recent papers have shown that the differences in the
[80]
microbiome of CS vs. VD are very short-lived. In one large longitudinal study using whole genome shotgun
(WGS) metagenomic analysis, a more sensitive method compared to the commonly used 16S rRNA
sequencing, the microbiome between CS and VD becomes indistinguishable at six weeks . This provides a
[81]
very limited window of opportunity for such a dynamic microbiome to determine long-term health
outcomes for the child.

The second aspect to mention is the absence of prospective trials showing a pathogenic mechanism linking
CS and negative long-term health outcomes. The evidence that does exist, exclusively from epidemiological
studies, has been inconsistent in controlling for variables that are known to impact the health of a newborn.
These include: (i) maternal and neonatal co-morbidity during pregnancy and after birth; (ii) rates of
breastfeeding (known to be lower after CS); (iii) the impact of a hospital microbiome compared to a
maternity unit / home birth; (iv) increased maternal age, all of which are known to be independently
associated with paediatric health . This list is not exhaustive and, importantly, neglects the unknown
[82]