The Impact of Probiotics, Prebiotics, and Synbiotics during Pregnancy or Lactation on the Intestinal Microbiota of Children Born by Cesarean Section: A Systematic Review

The gut microbiota is a key factor in the correct development of the gastrointestinal immune system. Studies have found differences between the gut microbiota of newborns delivered by cesarean section compared to those vaginally delivered. Our objective was to evaluate the effect of ingestion of probiotics, prebiotics, or synbiotics during pregnancy and/or lactation on the development of the gut microbiota of the C-section newborns. We selected experimental studies in online databases from their inception to October 2021. Of the 83 records screened, 12 met the inclusion criteria. The probiotics used belonged to the genera Lactobacillus, Bifidobacterium, Propionibacterium, and Streptococcus, or a combination of those, with dosages varying between 2 × 106 and 9 × 1011 CFU per day, and were consumed during pregnancy and/or lactation. Probiotic strains were combined with galacto-oligosaccharides, fructo-oligosaccharides, or bovine milk-derived oligosaccharides in the synbiotic formulas. Probiotic, prebiotic, and synbiotic interventions led to beneficial gut microbiota in cesarean-delivered newborns, closer to that in vaginally delivered newborns, especially regarding Bifidobacterium colonization. This effect was more evident in breastfed infants. The studies indicate that this beneficial effect is achieved when the interventions begin soon after birth, especially the restoration of bifidobacterial population. Changes in the infant microbial ecosystem due to the interventions seem to continue after the end of the intervention in most of the studies. More interventional studies are needed to elucidate the optimal synbiotic combinations and the most effective strains and doses for achieving the optimal gut microbiota colonization of C-section newborns.


Introduction
The gut microbiota (GM) is the set of microorganisms coexisting in the gut of an individual [1]. It plays an active role in the development and maturation of the gastrointestinal mucosal immune system (GALT) and in the defense against intestinal pathogens [1,2]. At birth, the newborn presents an immature immune system, which requires immunogenic stimuli from the developing GM for proper maturation [1,2]. It seems that the development of the GM begins in the womb, contrary to the commonly accepted paradigm of the fetus as a sterile organism [3,4]. In fact, species from the genera Staphylococcus, Lactobacillus, and Bifidobacterium, have been identified in the meconium, placenta, and amniotic fluid of neonates of healthy pregnant women [3,4]. This early GM can be affected by external factors such as the route of birth and diet [2], and is essential for infant and adult health [5,6].
Differences in the degree of gut microbiota development between children born by C-section and those born by vaginal delivery have been found, with the former having a less developed microbiota [7][8][9]. The gut microbiota of C-section newborns contains lower numbers of species of the genera Bifidobacterium, Streptococcus, and Lactobacillus, Nutrients 2022, 14, 341 2 of 12 and higher numbers of potentially pathogenic bacteria, such as Clostridium perfringens or Escherichia coli [10][11][12], compared to vaginally delivered newborns. Although a commonly given explanation is that during passage through the vaginal canal, the newborn acquires different beneficial bacteria that colonize the intestinal tract [7,9], the lack of exposure to vaginal microbiota is unlikely to be the sole contributing factor. Other elements such as intrapartum antibiotic administration, differences in feeding behaviors, maternal obesity, gestational age, limited early skin-to-skin contact after birth, or absence of labor [13][14][15] could also be important drivers of the C-section newborn's GM. Amongst them, neonate feeding seems to play a very important role. Researchers have shown that women who deliver by C-section are less likely to breastfeed, or will delay breastfeeding initiation [16,17]. This is of great importance for GM colonization, since breast milk contains a plethora of beneficial bacteria essential for the optimum immune development and the intestinal colonization in the newborn [18]. The bacterial composition of breast milk is closely related to that of the GM of babies, indicating the existence of the passage of bacteria from mother to the child during the lactation process [2]. Therefore, the lactation process seems to be a key factor in the development and correct establishment of the GM in children born vaginally and by cesarean section [19,20]. In this sense, it has been suggested that external supplementation with probiotics (live microorganisms that, after ingestion in specific numbers, confer health benefits to the host [21]), prebiotics (a substrate that is selectively utilized by host microorganisms, conferring a health benefit [22]), or synbiotics (a mixture comprising live microorganisms and substrate(s) used selectively by host microorganisms that confers health benefits to the host [23]) in the mother could positively contribute to the colonization of the GM of the newborn and therefore contribute to a good immune development in a natural way [24], which is especially important for C-section newborns. We aimed to evaluate, in published analytical experimental studies, the effect of the probiotic, prebiotic, or synbiotic ingestion during pregnancy and/or lactation on the colonization of the GM of C-section newborns.

Materials and Methods
This systematic review was carried out following prospective registration (PROSPERO ID: CRD42021241641) and reported according to the PRISMA statement [25].

Literature Search
Searches were conducted in the PubMed, Web of Science, and Scopus databases from their inception to October 2021. ScienceDirect was used as additional source. Keywords used were probiotic, synbiotic, prebiotic, Lactobacillus, Bifidobacterium, oligosaccharides, pregnancy, lactation, breastfeeding, and cesarean. Boolean descriptors AND, OR, and NOT were used, as well as performing truncations (*) of the different terms. To further define the results, additional filters were used. The searching strings used in each of the databases are presented in Table S1.

Selection Criteria
We included experimental studies published from the time of inception, until October 2021, written in English with full text available, and conducted in humans. Studies that did not refer to the efficacy of the use of probiotics, prebiotics, or synbiotics during pregnancy or lactation on the gut microbiota of C-section newborns, and studies with unclear data information about the interventions were excluded. The selection process is further described in Section 3.1.

Data Extraction and Analysis
Data were extracted by CA-P and NC-I, and contrasted by SM-P. From each selected publication, information on authorship, year of publication, place of completion of the study, study population, microorganisms and/or oligosaccharides and dosage used, ad-Nutrients 2022, 14, 341 3 of 12 ministration vehicle, intervention duration, and outcomes regarding gut microbiota of neonates was obtained.

Quality Assessment
Selected studies were evaluated by C.A.-P. and N.C.-I., and disagreements were contrasted by S.M.-P. using the 2020 update of the Cochrane Risk of Bias (RoB2) assessment tool [26]. Five areas were evaluated in risk of bias: selection, performance, detection, attrition, and reporting. Every single item was evaluated to have a high, low, or unclear risk of bias and an overall estimation was obtained for each study, which was classified as a low, medium, or high risk of bias.

Selection Process
The selection process is presented in Figure 1, according to the PRISMA flow diagram [27].

Data Extraction and Analysis
Data were extracted by CA-P and NC-I, and contrasted by SM-P. From each selecte publication, information on authorship, year of publication, place of completion of th study, study population, microorganisms and/or oligosaccharides and dosage used, ad ministration vehicle, intervention duration, and outcomes regarding gut microbiota of ne onates was obtained.

Quality Assessment
Selected studies were evaluated by C.A.-P. and N.C.-I., and disagreements were con trasted by S.M.-P. using the 2020 update of the Cochrane Risk of Bias (RoB2) assessmen tool [26]. Five areas were evaluated in risk of bias: selection, performance, detection, attr tion, and reporting. Every single item was evaluated to have a high, low, or unclear ris of bias and an overall estimation was obtained for each study, which was classified as low, medium, or high risk of bias.

Selection Process
The selection process is presented in Figure 1, according to the PRISMA flow diagram [27]. The electronic search using the strategy previously described yielded a total of 8 records (19 PubMed, 38 Scopus, 18 Web of Science, 8 ScienceDirect). After duplicates wer deleted, 54 titles and abstracts were screened. Of these, 34 articles not meeting the elig bility criteria (1 study on animals, 11 not complying with the design, 22 not assessing th The electronic search using the strategy previously described yielded a total of 83 records (19 PubMed, 38 Scopus, 18 Web of Science, 8 ScienceDirect). After duplicates were deleted, 54 titles and abstracts were screened. Of these, 34 articles not meeting the eligibility criteria (1 study on animals, 11 not complying with the design, 22 not assessing the association of the review) were dismissed. From the 20 articles that were fully read, 9 did not meet the eligibility criteria (2 not complying with the design, 7 not assessing the association of the Nutrients 2022, 14, 341 4 of 12 review) and were also dismissed. The reference lists of the remaining 11 articles revealed 1 further citation. Finally, 12 records were included in the systematic review.

Characteristics of Studies Selected
Information on authorship and year of publication, study population, type and duration of the intervention, and outcomes regarding gut microbiota are shown in Table 1.
From the twelve studies included, two investigated the effect of the interventions in the mother and the offspring [29,36], whereas the remaining ten studies investigated the effect of the interventions only in the infants. Three studies used only C-section newborns as the study population [30,37,39]; the remaining studies investigated the effect of the interventions in infants born by both delivery types (vaginally and C-section). The sample size varied from 40 [32] to 422 [36].
When the intervention included the pregnant female, this took place from week 36 of gestation until delivery [29,36]. In infants, most of the interventions started at birth [28][29][30][31]33,36,37,39] or within 3 days after delivery [32,34]. The interventions in the remaining studies started later: one week after birth [35] or at almost one month of life [38].
All the studies found beneficial effects of the interventions on the gut microbiota of CD infants.

Interventions with Probiotics
All the selected studies reported information about the specific strains used. Studies investigating probiotics used either a single strain, belonging either to the genera Bifidobacterium [28,30,35] or Lactobacillus [32], or multi-strain combinations [29,33,36,37]. Some of the multi-strain combinations included, in addition to Bifidobacterium or Lactobacillus strains, others belonging to the genera Streptococcus [29] or Propionibacterium [36].

Interventions with Prebiotics
From the twelve included studies, only one used a prebiotic as unique intervention [38]. In another study, the intervention with prebiotics was compared with an intervention with synbiotics [34]. The prebiotics used were milk-derived oligosaccharides [38] and a combination of short-chain galacto-oligosaccharides and long-chain fructo-oligosaccharides [34].

Interventions with Synbiotics
Four studies used synbiotics as intervention [31,34,36,39], with newborns as the population under study. Strains of Bifidobacterium breve were the most used in the synbiotic combinations [34,36,39]; in one study together with other strains from the genus Lactobacillus in a multi-strain mixture [36]. The prebiotic components were either galacto-oligosaccharides alone [36] or in combination with fructo-oligosaccharides [34,39], and bovine milk-derived oligosaccharides [31].

Stool Sample Collection and Microbial Analysis Methods
Fecal sample collection was performed at home in all the studies. In all the studies where this process was described in detail, the sample was stored either in domestic freezers [35,37] or in fridges [32][33][34]39] at home prior to delivery to the place of analysis, either as collected [32,34,35,39], mixed with storage media [30,37], or maintained in anaerobic conditions [33]. At the place of analysis, samples were stored at −40 • C [32], −70 • C [37], or −80 • C [29,30,[33][34][35]39] until subsequent analysis. None of the studies specified a maximum storage time required for analysis.
All the studies used a variety of 16S RNA-based methods for the analysis of the IM; some of them were combined with traditional culture methods [30,31,37]. Table S2 shows the evaluation of the methodological quality of the 12 studies included in this systematic review. Two studies were considered to have high overall risk of bias [28,35], three had an unclear overall risk of bias [31,34,39], and seven had a low overall risk of bias [29,30,32,[36][37][38].

Discussion
The differences found in the gut microbiota of vaginally delivered newborns and C-section newborns show a more immature and less effective GM in the C-section newborns. This causes susceptibility to develop certain metabolic or immune disorders [7,9]. Probiotic, prebiotic, and synbiotic interventions led to a beneficial gut microbiota in C-section newborns, closer to that of vaginally delivered newborns, especially regarding Bifidobacterium colonization.
Regarding probiotic interventions, the effects observed on GM were more evident when multi-strain combinations were used [32,33]. In synbiotic formulas, probiotic strains (all from Bifidobacterium genus) were combined with galacto-oligosaccharides, fructooligosaccharides, or bovine milk-derived oligosaccharides. This could represent an optimal strategy to achieve the restoration of GM in CD-delivered infants. It is thought that interventions with Bifidobacterium strains alone could be insufficient to promote an effect on GM [40]. This could be due to the difficulty of achieving permanent colonization of the infant gut due to competition with autochthonous microbiota. In this sense, it has been suggested that the combination of Bifidobacterium with a prebiotic or with breastfeeding, which provides milk oligosaccharides, would be more successful regarding colonization [41]. In fact, Chua et al. [34] found that a synbiotic intervention combining B. breve M16V with galactoand fructo-oligosaccharides increased infant gut colonization by the probiotic strain and by other members of the Bifidobacterium genus compared to formula-fed infants [34]. In C-section infants, this colonization was similar to that one of vaginally delivered infants. In addition, Cooper et al. [31] found a strong bifidogenic effect of a synbiotic preparation containing B. lactis CNCM I-3446 and bovine milk-derived oligosaccharides, which was more evident in C-section newborns. This is of a great importance due to the relevance of Bifidobacterium colonization in early life immune programming [42,43].
The studies indicate that the sooner the intervention begins, the more successful the effect achieved [41], since the first three months of life are a key window for GM recovery in C-section infants, especially regarding Bifidobacterium. In fact, most of the studies included in this review started the interventions immediately after birth, although the rest of the included studies found beneficial modifications of the GM of newborns even when begin-ning intervention later [35,38]. These early interventions can restore the low Bifidobacterium presence in C-section infants within a week [31]. The intervention period is also important. In the present review, although some authors state that a minimum of 3 months would be needed to restore the GM of C-section infants by using probiotic strains [36], we have found that shorter intervention durations are also effective [28,35,37,39].
Regarding the administration of probiotics during pregnancy, the two studies included started the intervention in the last month of pregnancy. Whether these prenatal interventions exert an effect on C-section infants remains unclear. Mastromarino et al. [29] found that a multi-strain probiotic combination positively influenced the beneficial microbiota of breast milk, by means of as systemic effect exerted by the probiotics, but this effect was less evident in C-section newborns. On the contrary, Korpela et al. [36] observed that most of the cesarean-associated changes in the fecal microbiota of infants were corrected or reduced by a probiotic supplementation to mother and infant, indicating that that breastfeeding together with probiotic supplementation offered optimal results in terms of supporting the microbiota development in these infants. However, since both the mother and the infant received the same probiotic supplement, it was not possible to elucidate the role of the maternally ingested probiotic on the infant GM.
Another important point is the persistence of probiotic colonizers over time. In most of the studies in our review, the probiotic strain was found for either a short period of time or not at all in feces. Despite this, a beneficial effect of probiotic strains on the GM has been observed [33], demonstrating persistent changes in the infant microbial ecosystem after the end of the intervention in most of the studies. One explanation could be that probiotic strains help to create a microbial ecosystem that facilitates the growth of autochthonous beneficial bacteria, which would, in turn, be responsible for the health benefits observed. In this sense, it has been observed that intervention with prebiotics is able to increase Bifidobacterium populations by increasing the endogenous population of Bifidobacterium in healthy, term infants [44].
Regarding the use of these products by the general population, consumers and medical providers must bear in mind there are very important details that must always be available. These include the specific strain/s, the number of microorganisms, the treatment duration, the route of administration, the formulation, the shelf-life, and the storage conditions, which unfortunately are often missing. In this sense, it is worthy to mention that Lactobacillus nomenclature has recently been changed [45] based on several genetic approaches and markers, providing a better ecological and functional vision. As an example, following this new classification, Lactobacillus reuteri is now named Lacticaseibacillus reuteri. Consequently, the labels of probiotic products will need to be updated and scientists will need to take the new names into account for future publications and new patents. In addition, this will be critical when performing literature searches.

Strengths
With the exception of one study, all included studies were randomized controlled trials, giving the highest degree of evidence. The review includes recent studies and provides specific strains, doses, and intervention times. We have conducted the review based on PRISMA guidelines, selecting studies published since database inception. Moreover, most of the studies have low risk of bias.

Limitations
We found high variability regarding the strains and dosages of probiotic microorganisms alone or in combination with prebiotics, which makes it difficult to suggest a specific strain or dosage. In addition, the methodology regarding fecal collection and microbial analyses was heterogenous, which makes it difficult to compare results. In some studies, the influence of important factors such as feeding (breastfed or formula) and antibiotic intake were not taken into account.

Conclusions
The intake of probiotics, prebiotics, and synbiotics, especially during lactation, results in beneficial effects on the gut microbiota of newborns, especially C-section newborns. These interventions are more effective when ingestion begins soon after birth, especially for restoring the population of bifidobacteria. More interventional studies are needed to elucidate the optimal synbiotic combinations and the most effective strains and doses to achieve the optimal gut microbiota colonization of C-section newborns.
Supplementary Materials: The following supporting information can be downloaded at: https://www. mdpi.com/article/10.3390/nu14020341/s1. Table S1. Search strategies used for this review, Table S2. Summary of assessing risk of bias according to ROB2 checklist. Institutional Review Board Statement: Ethical review and approval were waived for this study, because the Systematic Review relies on retrieval and synthesis of data from existing approved human trials.
Informed Consent Statement: Patient consent was waived because the Systematic Review relies on retrieval and synthesis of data from existing approved human trials.

Data Availability Statement:
The publications analyzed for this systematic study can be accessed from their respective journals, whereby access restrictions may apply.