Dual-strain probiotics Bifidobacterium bifidum and Lactobacillus acidophilus reverse gut dysbiosis in preterm neonates: a randomized controlled trial
Article information
Abstract
Background
Preterm neonates exhibit gut dysbiosis, characterized by increased numbers of pathogenic bacteria and decreased Bifidobacterium and Lactobacillus levels. Supplementation with the probiotic Bifidobacterium bifidum/Lactobacillus acidophilus (BB/LA) may reverse gut dysbiosis.
Purpose
To study the effects of BB/LA on the gut microbiota of preterm neonates.
Methods
We enrolled neonates born between July 2022 and September 2023 with a gestational age of <33 weeks or birth weight of <1,500 g. After randomization into probiotic (PG) and control (CG) groups, stool samples were collected at 3 time points: birth (V1), 35 weeks' postmenstrual age (PMA) (V2), and 4 months of age (V3). BB/LA was administered to the PG until PMA 35 weeks. All neonates received a feeding protocol similar to that of predominant breast milk. Stool samples were stored at -80°C, a DNA extraction performed, and 16S rRNA gene sequencing used to define alpha and beta diversities and the relative abundances of the bacteria. Baseline characteristics and clinical outcomes were collected.
Results
We analyzed 68 neonates (33 in the PG, 35 in the CG). The alpha diversities did not differ significantly between the groups at any time point. At V1, beta diversity was not significantly different between the 2 groups. After BB/LA supplementation (V2), beta diversity was significantly greater in the PG versus CG (P=0.004). The relative abundances of Bifidobacterium and Lactobacillus were higher in the PG (both P<0.001), whereas that of Clostridium senso stricto 1 was higher in the CG (P=0.017). Growth parameters, necrotizing enterocolitis, and mortality rate did not differ between groups. No adverse events were observed.
Conclusion
BB/LA led to healthier gut microbiota in preterm neonates as demonstrated by a reversal of gut dysbiosis characterized by increased beta diversity, increased the relative abundances of Bifidobacterium and Lactobacillus, and decreased the relative abundance of Clostridium senso stricto 1.
Key message
Question: Can probiotics BB/LA reverse gut dysbiosis in preterm neonates?
Finding: BB/LA supplementation induced more diverse beta diversity and increased relative abundances of Bifidobacterium, Lactobacillus and decreased relative abundance Clostridium.
Meaning: Early BB/LA supplementation could reverse gut dysbiosis in preterm neonates.
Graphical abstract
Introduction
Preterm neonates often exhibit gut dysbiosis among other gastrointestinal problems. Gut dysbiosis in preterm neonates is caused by many disadvantageous factors such as Cesarean delivery, nonbreast feeding, exposure to broad spectrum antibiotics and prolonged hospital stay with multiple harsh environmental exposures in the neonatal intensive care unit (NICU) [1-3]. Consequently, gut microbiota of preterm neonates typically has aberrant microbial colonization (i.e., reduced diversity) with notable numbers of pathogenic bacteria such as the Enterobacteriaceae which are Gram-negative facultative anaerobes (i.e., Escherichia coli, Klebsiella pneumoniae) and a lower abundance of commensal obligate anaerobes such as Bifidobacterium, Lactobacillus, and Bacteroides [4], the aforementioned microbiota condition can be defined as gut dysbiosis [5]. Intestinal dysbiosis in preterm neonates has been linked to necrotizing enterocolitis (NEC) [6-8]. Thus, preterm neonates are one of the most vulnerable populations to encounter detrimental gastrointestinal problems, most notably NEC [1]. Fecal microbiota analysis of preterm neonates with NEC demonstrated an increased abundance of Enterobacteriaceae and Clostridiaceae and a decrease of Bifidobacteriaceae, Firmicutes, and Bacteroidetes [1,3].. Besides gut dysbiosis, impaired maturation of gastrointestinal tract as well as the underdevelopment of the gut immune system make it difficult for the preterm neonates to tolerate optimal enteral feeding and undergo optimal and sustainable growth [2,9].
In contrast to gut dysbiosis, healthy gut microbial colonization in neonates also known as gut eubiosis, has been shown as a pivotal role in maintaining balanced intestinal immunity, nutrient digestion and absorption, intestinal motility, and energy utilization [1]. Intestinal eubiosis is typically observed in term, vaginally delivered, exclusively breastfed infants [3,5,10]. The proposed healthy gut microbiota is characterized by an abundance in Gram-positive Firmicutes such as Staphylococcus, Bifidobacterium, and Lactobacillus with the ability to produce glycosidases necessary to digest human milk oligosaccharides (HMOs) [3]. Probiotics supplementation in the early life of preterm neonates might reverse or modify the gut dysbiosis to resemble those of term neonates with gut eubiosis.
Probiotics are live microorganisms that, when administered in adequate amounts, confer a healthy benefit to the host [11]. Early supplementation of probiotics, either multistrain or single-strain, to preterm neonates aims to alter gut microbiota environment by colonizing the immature gut with commensal bacteria which are naturally present in the presumably eubiotic environment of exclusively breastfed term newborns [9,12]. When the gastrointestinal tract is colonized by these bacteria derived from the probiotics, studies have shown that the preterm neonates are less likely to develop late-onset sepsis and NEC [1,3,13]. However, the beneficial mechanisms of probiotics remain largely unknown but are thought to involve changes in intestinal permeability, enhanced mucosal IgA responses to foreign antigens and increased production of anti-inflammatory cytokines [14]. Promising beneficial effects of probiotics have been shown in various clinical trials in the prevention of NEC [14-19], late-onset sepsis [15,16], and reduction in mortality [14,15,17,19]. However, many other previous studies in this area demonstrated inconsistent results, in which some studies did not show clinical benefits of probiotics supplementation in preterm neonates. Two meta-analyses concluded that only specific strains of probiotics such as various strains of Bifidobacterium spp. yielded promising clinical benefits while others did not [16,18]. The 2020 European guideline for using probiotics in preterm neonates only provides a conditional recommendation on administering specific strains of probiotics in the prevention of NEC [20].
A dual-strain probiotics consisting of Bifidobacterium bifidum/Lactobacillus acidophillus (BB/LA) has been shown in neonates to increase the abundance of Bifidobacterium spp. and Lactobacillus spp. with a reduction of Enterobacteriaceae, compared with placebo/the control group [12,21]. The use of probiotics BB/LA has increased among preterm newborns in many NICUs across Europe with good safety profiles [22,23] and has been shown to significantly reduce the risk of NEC, nosocomial bacteremia, sepsis, and death [24-26]. It is well known that breast milk feeding reduces NEC risk in preterm neonates [27,28]. Probiotics has a synergistic effect among the breastfed neonates in altering the gut microbiota [21,29] and preventing NEC [18,23]. A reasonable explanation would be that breast milk contains prebiotic HMOs which act as an important metabolic source of growth for the Bifidobacterium [11,25]. Nowadays many NICUs encourage breast milk as a primary source of enteral feeding in the preterm neonates, we believe that BB/LA would yield advantageous gut microbiota profile for the predominantly breastfed preterm neonates.
Probiotics BB/LA administration in the neonatal period has been shown to maintain the colonization of Bifidobacterium after discontinuing it during the infancy period [21], indicating the sustainable beneficial effects. According to previously mentioned studies, we hypothesized that preterm neonates generally exhibit gut dysbiosis earlier in life, which is characterized by high levels of Enterobacteriaceae and Clostridiacieae and relatively low levels of Bifidobacterium and Lactobacillus. With the aid of probiotics BB/LA, the gut dysbiosis might be reversed or attenuated. Since there are only a handful of randomized controlled studies on the prevention of gut dysbiosis by BB/LA, therefore we aim to compare gut microbial profiling of the preterm infants who receive in-hospital probiotics BB/LA and those who did not.
Methods
We enrolled preterm newborns who were born between July 2022 and September 2023. This clinical trial was reviewed and approved by the Institutional Review Board of Faculty of Medicine Ramathibodi Hospital (MURA 2022/153) and registered in the Thai Clinical Trials Registry (TCTR20220829005). This study was carried out with strict adherence to the World Medical Association’s Declaration of Helsinki. Written informed consent was obtained from the parents or guardians of all participants. The inclusion criteria were as follows: preterm newborns with a gestational age (GA) <33 weeks or birth weight <1,500 grams (i.e., very low birth weight). The exclusion criteria were cyanotic congenital heart disease, chromosomal or gastrointestinal tract anomalies, or enteral feeding was withheld for more than 14 consecutive days. The enrolled infants were randomized into the probiotics group (PG) and control group (CG).
Preterm newborns in PG received BB/LA (Infloran, Laboratorio Farmaceutico SIT Mede, Italy). One capsule of BB/LA (250 mg) contains 1 billion colony forming unit (CFU) lyophilized Bifidobacterium bifidum and 1 billion CFU lyophilized Lactobacillus acidophilus. After having been fed breast milk for one day without complications, one capsule of BB/LA was administered once daily via dissolving in 1–2 mL of sterile water prepared by neonatal registered nurses, then tube-fed before feeding milk. BB/LA preparations were freshly prepared right before administering to the infants. Throughout the trial, BB/LA was kept in a 4°C temperature-controlled refrigerator, separated from the milk preparation station to avoid cross-contamination with CG. BB/LA was administered until postmenstrual age (PMA) of 35 weeks and continued until the infants were discharged. In CG, besides a standard neonatal care similar to the PG, preterm newborns did not receive BB/LA during the study. The feeding protocol strongly supported the use of breast milk as the first choice in all studied infants. However, when there was insufficient breast milk in some infants, pasteurized donor human milk (DHM) would be given after written informed consent was obtained from the parents. Preterm newborns in both groups received similar standardized feeding protocol and medical care without interference from the researchers.
Perinatal and maternal data was obtained by interviewing the mother according to the case record form and from electronic medical records. All preterm newborns would receive empirical antibiotics if there was clinical suspicion of neonatal sepsis. Then, if blood cultures were negative at 48 to 72 hours and the newborns had no signs of sepsis, the antibiotics were discontinued. Enteral feeding was initiated for studied newborns within the first few days of life with maternal breast milk (MBM) or DHM in case of unavailable mother’s breast milk. MBM or DHM was continuously given at least 2 weeks after birth but not more than 4 weeks. After 2 weeks, if there was no MBM, premature formula milk was gradually introduced, alternating with MBM/DHM. At any point, if MBM became available, the infants would receive MBM as the first choice. Full enteral feeding was defined as a total volume of 150 mL/kg/day.
The focused clinical outcomes associated with gut dysbiosis were NEC with modified Bell’s stage ≥2A, late-onset neonatal sepsis, and death. Time to full enteral feeding, duration of hospitalization, body weight at discharge, 2 months and 4 months, respiratory distress syndrome and bronchopulmonary dysplasia were also recorded. Factors altering gut microbiota of preterm newborns such as duration of antibiotics administration, route of delivery and types of feeding were also recorded.
1. Collection of stool samples
Nurses in the NICU collected fresh stool samples from the diapers of the infants into a sterile container at 3 different time points. Stool samples were collected within 72 hours after birth as meconium (V1) and after PMA 35 weeks (V2). After being discharged from the hospital, infants in both groups did not further receive BB/LA. At chronological age of 4 months, the infants were scheduled for follow up and vaccination at our outpatient clinic. At the clinic, stool samples were collected as the third sample (V3) for postdischarge analysis.
2. DNA extraction
All stabilized samples were then transferred to the Center for Medical Genomics, Faculty of Medicine Ramathibodi Hospital, where they were stored at -80°C until processing. Metagenomic DNA was extracted using QIAamp PowerFecal DNA Kit (QIAGEN, Germany) according to the manufacturer's protocol. DNA quality and quantity were measured using NanoDrop One spectrophotometer and Qubit 3.0 fluorometer (Thermo Fisher Scientific, USA), with acceptable quality thresholds of A260/280 ratio between 1.8-2.0 and A260/230 ratio above 1.4.
3. Polymerase chain reaction amplification and DNA sequencing
The V3–V4 hypervariable regions of bacterial 16s ribosomal RNA genes were amplified using a 2-step polymerase chain reaction (PCR) approach. The first PCR amplification was performed using primers 341F (5′-CCTAYGGGRBGCASCAG-3′) and 806R (5′-GGACTACHVGGGTWTCT AAT-3′). Each PCR reaction mixture (total volume 15 μL) contained Phusion High-Fidelity PCR Master Mix (New England Biolabs, USA), 0.2 μM of each primer, and approximately 10 ng of template DNA. The PCR conditions were as follows: Initial denaturation at 98°C for 1 minute; 30 cycles of denaturation at 98°C for 10 seconds, annealing at 50°C for 30 seconds, and extension at 72°C for 30 seconds; followed by a final extension at 72°C for 5 minutes. Index sequences were subsequently added in the second PCR step using NEBNext Ultra II FS DNA PCR-free Library Prep Kit (New England Biolabs). The prepared libraries underwent paired-end sequencing (2×250 bp) using NovaSeq 6000 SP Reagent Kit v1.5 (500 cycles) on the Illumina NovaSeq 6000 platform.
4. Bioinformatics analysis of gut microbiota composition
Raw sequencing data underwent quality control and preprocessing using FLASH software before being analyzed with Quantitative Insights into Microbial Ecology 2 (QIIME 2). The sequences were denoised using DADA2 to generate amplicon sequence variants (ASVs), which were then taxonomically classified using the SILVA 138 ribosomal RNA database. ASV count tables were generated and grouped at various taxonomic levels.
Microbial community diversity was assessed through both alpha and beta diversity analyses. Alpha diversity, which reflects the complexity of species diversity within samples, was evaluated using 4 indices: abundance-based coverage estimator, Chao1 (species richness estimation), Shannon index (diversity considering both richness and evenness), and Simpson index (community dominance). For beta diversity analysis, principal coordinate analysis (PCoA) with Bray-Curtis dissimilarity matrices was performed to compare microbial community compositions between the 2 study groups, as well as across different visiting timepoints. Differential abundance analysis was conducted using linear discriminant analysis effect size (LEfSe) when differences in taxonomic abundance were statistically different and then visualized through taxonomic bar charts. All diversity metrics and LEfSe analyses were performed using MicrobiomeAnalyst 2.0 (https://www.microbiomeanalyst.ca/). The data were then subjected to comprehensive statistical analyses to identify significant associations and differences between groups.
QIIME generates operational taxonomic unit tables containing information regarding the microbiome composition and relative abundances of each bacterial taxon (from phylum to genus) in each sample. QIIME was used to assign taxonomy and determine alpha and beta diversities. To measure alpha diversity, 4 parameters were calculated; Chao 1, Shannon, and Simpson indices. To assess beta diversity, PCoA of Bray-Curtis dissimilarity was performed to depict the dissimilarity of bacterial populations between the PG and CG. If differences in taxonomic abundance were statistically different between the groups, LEfSe was performed.
5. Biostatistical analyses
We conducted statistical analysis using STATA/SE program ver. 18.0 (StataCorp, USA). For comparison of categorical data, we used Fisher exact test and chi-square test. The normal distribution of continuous data was evaluated using the Shapiro-Wilk test. For comparison of continuous data with normal distribution, we used Student t test. For continuous data with nonnormal distribution, we used quantile regression.
For microbiome statistical analyses, alpha diversity comparisons between groups were performed using univariate and multivariate linear regression analyses. For nonparametric comparisons, Wilcoxon rank-sum tests were applied with false discovery rate control using the Benjamini-Hochberg method. P values for beta diversity were calculated using permutational multivariate analysis of variance. Statistical significance was set at P value <0.05.
Results
1. Perinatal and maternal characteristics
In our study, 93 preterm newborns were initially enrolled and 25 of them were excluded; 2 had cyanotic congenital heart disease, 2 had gastrointestinal tract anomalies, 9 died before randomization, 2 referred out, and 10 of their parents refused to participate. We had 68 patients remained in the trial (33 in PG and 35 in CG). Upon the process of fecal DNA extraction and PCR amplification, some samples were unqualified for sequencing, thus were excluded from analysis (Fig. 1). Baseline data including maternal age, mode of delivery, sex, GA, birth weight, Apgar scores and perinatal antibiotics exposure were comparable in both groups (Table 1). All infants were given mixed feeding of 50% fortified breast milk and 50% premature formula when the second stool samples (after PMA 35 weeks) were collected.
Flow diagram of the study. GA, gestational age; BW, body weight; PMA, postmenstrual age.
Maternal and perinatal characteristics of the preterm newborns
2. Gut microbiota profile
The alpha diversity of meconium samples (V1) between PG and CG were not statistically different (Supplementary Fig. 1). Also, the beta diversities of meconium samples represented by Bray-Curtis index were comparable (P=0.323) (Supplementary Fig. 2). Taxonomy abundance of meconium samples in both groups revealed predominance of Staphylococcus as shown in dark green color in Fig. 2A. The relative abundance of Staphylococcus in infants delivered via Cesarean section was significantly higher than those who were vaginally delivered (P=0.04) (Supplementary Fig. 3).
Taxonomy abundances of stool samples across 3 time points (V1–V3). (A) At V1 (birth), an overall predominance of Staphylococcus was seen in both groups. (B) At V2 (35 weeks' postmenstrual age), several peaks of Bifidobacterium were seen in the probiotics group along with a predominance of Klebsiella in the control group. (C) At V3 (4 months of age), an overall predominance of Bifidobacterium was noted in both groups.
The alpha diversity of V2 was also not statistically different between the 2 groups (Supplementary Fig. 4). Whereas, the beta diversity of V2 in the PG was significantly diverse when compared with the CG group (P=0.004) (Fig. 3). The taxonomy abundance of V2 showed a more predominance of Bifidobacterium in the PG as shown in dark pink color in Fig. 2B. LefSe analysis depicted several differences in the relative abundance of certain bacteria including higher abundances of Bifidobacterium spp. (P<0.001) (Fig. 4A) and Lactobacillus spp. in the PG (P<0.001) (Fig. 4B), and a higher abundance of Clostridium senso stricto 1 in the CG (P=0.017) (Fig. 4C). The relative abundances of Klebsiella, Escherichia-Shigella, Enterococcus, Enterobacter, Staphylococcus, Streptococcus, and Serratia were not statistically different between the 2 groups.
Beta diversity at time point V2 (35 weeks' postmenstrual age). PERMANOVA, permutational multivariate analysis of variance; CG, control group; PG, probiotics group.
Relative abundances of Bifidobacterium (A), Lactobacillus (B), and Clostridium senso stricto 1 (C) at V2 (35 weeks' postmenstrual age).
The alpha and beta diversities at V3 were not statistically different between the PG and CG (Supplementary Figs. 5 and 6, respectively). Taxonomy abundance of V3 showed a predominance of Bifidobacterium spp. in both groups as shown in dark green color (Fig. 2C).
Supplementary Fig. 7 demonstrates a higher Simpson index at V2 in both groups but the alpha diversity defined by this index seemed to decrease more in the PG vs. CG group in V3 without statistical significance (P=0.12).
3. Clinical outcomes
Most clinical outcomes including early-onset and late-onset neonatal sepsis, NEC stage ≥2A and lung complications were comparable as shown in Table 2. The number of days to full enteral feeding was approximately 10 days in both groups with comparable length of hospitalization of 44 days. There were 2 deaths in the PG; one died without known cause at home and the other died from severe congenital brain anomaly, which was diagnosed after randomization. One infant in the CG died due to spontaneous intestinal perforation. Also, the body weight at 2 months and 4 months between the PG and CG were not statistically different.
Clinical outcomes
Discussion
The gut microbiota profiles of meconium samples in preterm newborns during the first few days of life in our study were predominantly colonized with Staphylococcus, which is consistent with previous studies [30-32]. This can be explained by the fact that most of the preterm infants in this study were delivered via Cesarean delivery, thus making the gut microbiome of their early life resembled those of the skin surface of their mothers [3,33,34]. Therefore, the overall profile of gut microbiome in the first days of life was therefore not diverse. In accordance with our hypothesis, we found significant difference in the gut microbiota profile after we administered probiotics BB/LA to the newborns. The gut microbiota demonstrated a more eubiotic environment [3,5,10], with significantly higher abundances of Bifidobacterium and Lactobacillus [31]. Whereas in the CG, the relative abundance of Clostridium senso stricto 1 was significantly higher. In previous studies, Clostridium senso stricto 1 was more abundant preceding the development of NEC in the preterm newborns [35]. Another signal of gut dysbiosis was that Bifidobacterium and Lactobacillus were virtually non-existent in CG at PMA 35 weeks. Lower numbers of Bifidobacteriaceae and higher numbers of Clostridiaceae in the preterm newborns could be a signal of gut dysbiosis [3], which was also associated with NEC [36]. Although not statistically different, probably due to a relatively small study population, preterm newborns in the CG had a slightly higher incidence of NEC stage ≥2A (12% vs. 3% in the PG, Table 2). Our research findings are consistent with the previous studies that showed the ability of probiotics to modulate gut microbiota of preterm newborns into a healthier environment [12,21,29,30,37].
We did not find any differences in alpha diversities between the 2 groups at all 3 time points, which are consistent with a previous study by Chang et al. [30]. Alpha diversity measures the richness of bacteria in each sample, hence at birth, the considerably low alpha diversities in both groups were not unexpected because of similar perinatal characteristics including route of delivery (mostly Cesarean delivery), one of the crucial factors contributing to gut dysbiosis early after birth [1,2]. Despite not statistically different after PMA 35 weeks, the alpha diversity in both groups increased overtime which is consistent with the study by Korpela et al. [32], where the diversity of gut microbiota in preterm newborns increased along with postnatal age. The cause of increased alpha diversity might be different between the neonates in both groups. In the PG, their gut microbiota profile was largely modulated by the BB/LA resulting in the higher abundance of Bifidobacterium and Lactobacillus. Infants in the CG were also exposed to the harsh environment such as NICU setting, antibiotics administration, and formula feeding, thus potentially resulting in a higher number of Clostridium senso stricto 1 [35]. Beta diversity at PMA 35 weeks demonstrated a clear dissimilarity of gut microbiota between the 2 groups, in which PG had a more diverse profile consistent with the study by Chang et al. [30].
From discharged until 4 months of age, infants in both groups relocated to their home environment for several weeks, which likely altered their gut microbiota profiles. We found no difference of alpha and beta diversities between the 2 groups. This might be explained by the fact that they were not exposed to hospital environment for several weeks and most of them received similar type of feeding. Probiotics BB/LA could not maintain gut microbiota diversity after its discontinuation. Surprisingly, gut microbiota in both groups showed a predominance of Bifidobacterium despite discontinuing BB/LA after discharged. In addition, probiotics BB/LA might pertain to the colonization of Bifidobacterium in the gut of former preterm infants [21] until 4 months of age after discontinuation. In our research, most patients after discharged receive half of their intake as breast milk and the other half as formula milk. This might explain that the predominance of Bifidobacterium was enhanced by breast milk that contained HMOs which serve as a prebiotic for Bifidobacterium [11,25]. Unfortunately, we were not able to perform the LefSe analysis to depict a statistical difference of Bifidobacterium between the 2 groups because of the limited number of stool samples.
There were several advantages in this clinical trial. We compared the gut microbiota profiles of preterm newborns in 3 different time points; at birth, at PMA 35 weeks and at age 4 months between the infants who received probiotics BB/LA and infants who did not. This comparison elucidated the characteristics of gut microbiota of preterm newborns which exhibited gut dysbiosis. Our NICU has a unified feeding protocol where all preterm newborns mostly receive breast milk or DHM in early weeks after birth. This research also showed that continuing breast milk after discharge from the hospital helped sustaining the abundance of Bifidobacterium. This result again stresses the importance of breast milk in preterm newborns in improving the chance of maintaining gut eubiosis. However, there were some limitations in this study. The number of infants and stool samples for gut microbiota analysis were relatively small. Therefore, we were not able to demonstrate significant differences of gut microbiome at 4 months of follow-up and did not find clinical significances on the use of probiotics BB/LA such as the prevention of NEC and late-onset neonatal sepsis. Our technique of 16s rRNA sequencing only detected the bacteria at the genus level, so we could not tell whether the Bifidobacterium and Lactobacillus found in the stool samples belonged to the probiotic’s species (BB/LA) or not.
In conclusion, this research showed that probiotics BB/LA modulated gut microbiota of preterm newborns into a more eubiotic environment with diverse commensal bacteria such as Bifidobacterium and Lactobacillus and low pathogenic bacteria such as Clostridium senso stricto 1. Studying the metabolomic profile of preterm gut microbiota affected by probiotics may shed the lights on mechanism of action of probiotics and further explain their benefits on clinical outcome.
Supplementary materials
Supplementary Figs. 1-7 are available at https://doi.org/10.3345/cep.2025.00374.
Alpha diversity at V1 (birth). (A) Abundance-based coverage estimator (ACE) index. (B) Chao1 index. (C) Shannon index. (D) Simpson index. No significant intergroup differences were noted in any parameters.
Beta diversity at V1 (birth). PERMANOVA, permutational multivariate analysis of variance; CG, control group; PG, probiotics group.
Relative abundance of Staphylococcus after vaginal versus cesarean delivery. At V1 (birth), cesarean-delivered infants showed a significantly higher relative abundance of Staphylococcus than those who were vaginally delivered.
Alpha diversity at V2 (35 weeks' postmenstrual age). (A) Abundancebased coverage estimator (ACE) index. (B) Chao1 index. (C) Shannon index. (D) Simpson index. No significant intergroup differences were noted in any parameters.
Alpha diversity at V3 (4 months of age). (A) Abundance-based coverage estimator (ACE) index. (B) Chao1 index. (C) Shannon index. (D) Simpson index. No significant intergroup differences were noted in any parameters.
Beta diversity at V3 (4 months of age). PERMANOVA, permutational multivariate analysis of variance; CG, control group; PG, probiotics group.
Simpson index alpha diversity at V1–V3 by study group. The Simpson index was low at birth (V1) but increased after 35 weeks' postmenstrual age (V2) in both groups. At 4 months of age (V3), the Simpson index was decreased more in the probiotics versus control group, but the difference was insignificant. The dashed line represents the mean value.
Notes
Conflicts of interest
No potential conflict of interest relevant to this article was reported.
Funding
This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Acknowledgments
This study was financially supported by a grant from MU’s Strategic Research Fund from Mahidol University. PT received a mid-career Career Development Grant from Faculty of Medicine Ramathibodi Hospital.
Author Contribution
Conceptualization: SS, PT, PP, PN; Data curation: SS; Formal analysis: SS; Funding acquisition: PT ; Methodology: SS, PT, PP, PN; Project administration: SS; Visualization: SS; Writing-original draft: SS; Writing-review & editing: PT, PP, PN