Impact of Limosilactobacillus reuteri DSM 17938 on early-life gut microbial diversity

Abstract

BACKGROUND

Gut microbiota dysbiosis plays an important role in functional gastrointestinal disorders. While the effectiveness of Limosilactobacillus reuteri (L. reuteri) DSM 17938 has been evidenced in several pediatric gastrointestinal conditions, its effects on gut microbiota modulation in exclusively breastfed infants remain unclear.

AIM

To evaluate the effect of L. reuteri DSM 17938 supplementation on gut microbiota diversity and gastrointestinal symptoms in healthy, exclusively breastfed infants.

METHODS

Healthy, term infants born at King Chulalongkorn Memorial Hospital (March 2024 to January 2025) were randomized to receive L. reuteri DSM 17938 or a placebo, five drops daily for up to 8 weeks. Fecal samples were collected at 1, 2, and 4 months. The stool microbiota was characterized using 16S rRNA sequencing (Nanopore technology) to assess taxonomic classification, microbial diversity, and relative abundance.

RESULTS

Stool samples from 34 exclusively breastfed infants (17 receiving probiotics, 17 placebo-treated) were analyzed. Baseline characteristics and gastrointestinal symptoms were comparable between groups. Gut microbiota diversity significantly differed with the mode of delivery during the first two months; a higher alpha diversity was observed in neonates born through normal delivery than in babies born through cesarean section (P = 0.028) (Chao 1 index, P = 0.07 and Shannon index, P = 0.01). By 2 months, the probiotic-treated group showed greater alpha diversity than the placebo, particularly in infants born by vaginal delivery (Chao 1 index, P = 0.38 and Shannon index, P = 0.05). In addition, beta diversity differed significantly between the groups (P = 0.04). However, no significant differences in alpha- or beta-diversity were observed at the 4-month time point. Probiotic supplementation was associated with an increased relative abundance of Bifidobacterium, regardless of delivery mode. Higher abundances of Parabacteroides and Hungatella hathewayi were observed in infants born via normal delivery or cesarean section, respectively. No differences in gastrointestinal symptoms were detected.

CONCLUSION

L. reuteri DSM 17938 supplementation during early infancy induced modest and transient changes in gut microbiota composition, particularly increasing Bifidobacterium abundance, without sustained microbiota alterations or significant effects on gastrointestinal symptoms.

Key Words: Gut microbiota; Limosilactobacillus reuteri DSM 17938; Probiotics; Infant; Functional gastrointestinal disorder

Core Tip: While Limosilactobacillus reuteri (L. reuteri) DSM 17938 has demonstrated benefits in several pediatric gastrointestinal conditions, its effects on gut microbiota modulation in exclusively breastfed infants remain unclear. In this prospective sub-study of a randomized controlled trial, L. reuteri DSM 17938 supplementation during the first two months of life was associated with an increased Bifidobacterium abundance and, independently of the delivery modes, higher Parabacteroides and Hungatella hathewayi were the most distinctive species identified in vaginally- and cesarean-delivered infants, respectively. L. reuteri DSM 17938 administration induced only modest and transient microbiota changes, with no evidence of sustained colonization.

INTRODUCTION

The human gut microbiota, composed of over 400 microbial species, contributes to homeostasis, immune development, metabolism, and protection against pathogenic microorganisms[1-3]. Probiotics are live microorganisms that provide health benefits through improving the gut microbiota in children. According to the position paper of the ESPGHAN special interest group on gut microbiota, probiotic strains such as Lacticaseibacillus rhamnosus, Limosilactobacillus reuteri (L. reuteri) DSM 17938, Bifidobacterium infantis BB-02, and Bifidobacterium lactis BB-12 have been recommended for supplementation in the management of various gastrointestinal conditions in children[4].

Among these, L. reuteri DSM 17938 has been shown to have an effective role in gastrointestinal conditions, including acute gastroenteritis[5], antibiotic-associated diarrhea[6], and infantile colic[7]. Although several randomized controlled trials have shown the administration of L. reuteri in reducing crying time in infantile colic, especially in exclusively breastfed infants[8-10], modulation of gut microbiota has been reported controversially. In the study reported by Savino et al[8], a significant increase in Lactobacilli and a decrease in Escherichia coli were observed in the stool of colicky infants supplemented with L. reuteri, while other studies showed no significant difference in E. coli density between the two groups[9,10].

Apart from its therapeutic role, L. reuteri has been shown to be effective in the prevention of gastrointestinal disorders in infants, such as regurgitation, colic, and constipation[11]. In addition, the relative abundance of the genus Lactobacillus and the phylum Bacillota in the first stool sample is lower in infants diagnosed with colic than in healthy infants[12]. However, evidence regarding changes in the longitudinal gut microbiota in infants with L. reuteri supplementation remains limited.

Therefore, this study aimed to evaluate the effects of daily L. reuteri DSM 17938 supplementation for up to 8 weeks on gut microbiota composition and gastrointestinal symptoms in healthy and exclusively breastfed term infants during the first four months of life.

MATERIALS AND METHODS

Study design and participants

This study selected 36 exclusively breastfed infants enrolled in the PROOF study (ClinicalTrials.gov: No. NCT06309199), a prospective, double-blind, randomized controlled trial conducted at King Chulalongkorn Memorial Hospital. The PROOF study aims to investigate the incidence of functional gastrointestinal disorders in Thai infants and the effects of L. reuteri DSM 17938 on the prevention of functional gastrointestinal disorders during early life. Eligible participants were healthy, born at term infants (gestational age 37 weeks to 41 weeks) who were appropriate for gestational age and whose mothers did not receive probiotics during the third trimester of pregnancy. All infants were exclusively breastfed, receiving only breast milk, with minimal supplementation allowed for illness-related medications and any necessary interventions for the study. Exclusion criteria included parental refusal of permission for infant participation, inability to attend follow-up visits, and infants having specific diseases at the time of enrollment. The study adhered to the CONSORT guidelines and was approved by the Institutional Review Board (IRB) of the Faculty of Medicine, Chulalongkorn University (No. 0855/66). Written informed consent was obtained from both parents or guardians.

Intervention

Using computer-generated block randomization (block size of 4), participants were allocated into two groups. Infants in the probiotic group orally received L. reuteri DSM 17938 in sunflower oil (i.e., 5 drops) at a daily dose of at least 10⁸ colony-forming unit within 7 days of life and for up to 8 weeks. The placebo group received a visually identical placebo containing sunflower oil and medium-chain triglycerides.

The sample size was calculated based on the microbial diversity score for this substudy as a part of the PROOF study. Based on these previous data, we assumed mean scores of 32.8 and 30.4, with standard deviations of 1.4 and 1.2 for L. reuteri DSM 17938 and the placebo group, respectively. With a 5% margin of error, a significance level (α) of 0.05, and 90% power (1-β), the required sample size was calculated to be at least 6 participants per group, ensuring sufficient power to detect meaningful differences.

Data collection and management

Baseline data, including demographic, socioeconomic, and family medical history, were collected through maternal interviews. Anthropometric data were recorded by nurses. Parents received a structured diary to track gastrointestinal symptoms, bowel movements, and probiotic/placebo administration over the first two months. Follow-ups were conducted at 1, 2, and 4 months (± 14 days) for interviews, physical examinations, and stool collection. The Thai version of the Rome IV diagnostic questionnaire[13] was electronically sent for completion or direct interviewing at 3 months and 4 months. Compliance was evaluated by weighing the intervention bottles before dispensing and upon return at the 1- and 2-month follow-ups. The weight difference was used to calculate the number of intervention days, providing an objective measure of adherence.

Fecal sample collection

Fecal samples were collected at 1, 2, and 4 months by research nurses and doctors. Samples were obtained either from diapers uncontaminated with urine or by gentle rectal stimulation using a sterile cotton bud soaked in sterile water. Stool samples were stored in sterile tubes at -80 °C until analysis.

DNA extraction, 16S polymerase chain reaction amplification, sequencing, and analysis

Stool samples (approximately 20 mg) were mixed with 2 mL of NAPSeq (Bioentist, Thailand) in sterile test tubes, rapidly frozen on-site with dry ice, stored at -20 °C for 24 hours, and then transferred to -80 °C until analysis. DNA was extracted using the QIAamp Fast DNA Stool Mini Kit (QIAGEN, Germany) following the manufacturer’s protocol. Amplicons were purified with the QIAquick^® PCR Purification Kit (QIAGEN, Germany). The 16S rDNA amplicon sequencing was performed using Oxford Nanopore Technology, as previously described[14].

Nanopore sequencing data were analyzed at the Center of Excellence in Systems Microbiology, Chulalongkorn University. Signal processing and basecalling were conducted using Guppy basecaller version 6.0.7 (Oxford Nanopore Technologies, United Kingdom) under the super-accuracy model configuration[15]. The derived FASTQ reads were quality-examined using MinIONQC[16], and subsequent demultiplexing and adapter trimming were carried out using Porechop v0.2.4 (https://github.com/rrwick/Porechop). The filtered reads were clustered, polished, and taxonomically classified by the NanoCLUST pipeline[17], using the Ribosomal Database Project version 11.5 as the reference taxonomy.

Following classification, taxonomic abundance data were reformatted to be compatible with the QIIME2 platform (v2021.2) and imported into MicrobiomeAnalyst[18] for subsequent data processing and visualization. To enhance data quality, features with low prevalence (< 10% of samples) and low variance (identified via interquartile range filtering) were excluded. Total sum scaling was used as the normalization method to remove heterogeneity across samples. Alpha-diversity metrics, including Chao 1 and Shannon indices, were analyzed using the Mann-Whitney U test or the Kruskal-Wallis test, depending on the number of groups. Beta diversity was calculated using Bray-Curtis dissimilarity and visualized via Principal Coordinate Analysis, with group-level differences assessed using Permutational Multivariate Analysis of Variance (PERMANOVA). Differential abundance across groups was identified using Linear Discriminant Analysis Effect Size (LEfSe), applying a significance level of P < 0.05 and a minimum Linear Discriminant Analysis (LDA) score threshold of 2.

Statistical analysis

All fecal sample processing and bioinformatic analyses were performed at the Center of Excellence in Systems Microbiology, Faculty of Medicine, Chulalongkorn University. For statistical analysis, categorical variables were reported as counts or percentages, and continuous variables were summarized as medians with minimum and maximum values. Given the likelihood of non-normality in most datasets, nonparametric statistical tests were employed for analysis. Between-group comparisons were conducted using the Mann-Whitney U test. Values of P < 0.05 were considered statistically significant. Statistical analyses were carried out using SPSS version 29.00 and Stata version 18.00. Microbiota analyses were performed as described above. Bacterial abundances were analyzed at the phylum, genus, and species levels and presented as relative abundance. Only the top 10 bacterial genera and species were displayed, while those ranked 11^th and beyond were grouped as “others”.

RESULTS

Baseline characteristics and clinical data

A total of 512 infants were recruited for the PROOF study. Of these, 128 infants who were exclusively breastfed completed all outcomes, and 36 were selected through orderly, convenient sampling, with subgrouping based on delivery mode [normal labor (NL) and caesarean section (CS)]. The probiotic and placebo groups equally consisted of 18 infants. Within each group, the same number of infants were delivered via NL or CS. However, one participant from each of the probiotic-CS and probiotic-NL groups was excluded due to poor-quality DNA extraction from stool samples (Figure 1).

Figure 1 Consort flow diagram. Predominantly breastfeeding; receiving at least 80% of total feeds as breast milk. L. reuteri: Limosilactobacillus reuteri.

For maternal data, the maternal age was 31 (29, 36) years, with a body mass index of 23.7 (21.5, 29.1) kg/m², and was comparable in both groups. Hypertension and diabetes were not reported in any of the mothers. For participants, there were no significant differences in demographic, socioeconomic, or clinical symptom data (crying, regurgitation, and stool pattern) between the two groups during the 4-month study period (P > 0.05). However, regurgitation symptoms were higher in the placebo group at all visits, whereas the proportion of infants crying for 1-3 hours per day was higher in the probiotic group than in the placebo group (Table 1).

Table 1 Characteristics and functional gastrointestinal symptoms at birth up to 4 months, n (%) or median (min-max).

Parameters	L. reuteri (n = 17)	Placebo (n = 17)	P value
Gestational age, week	38 (37, 40)	38 (37, 40)	0.454
Gender, male	9 (52.9)	7 (41.2)	0.563
Mode of delivery, normal labour	9 (52.9)	7 (41.2)	0.563
Birth weight (g)	3160 (2710-3665)	3100 (2300-4480)	0.454
First birth rank	11 (64.7)	11 (64.7)	1.000
Immediate family type	11 (64.7)	8 (47.1)	0.394
Family history of allergic disease	7 (41.2)	4 (23.5)	0.394
Smoking in the family	4 (23.5)	4 (23.5)	1.000
Compliance with medication (days)	47.5 (13, 56)	51 (31, 56)	0.157
Body weight gain (g/day)
0-1 month	37.6 (17.4, 44.6)	36.1 (14, 52)	0.867
1-2 months	34.3 (17.1, 47.0)	34.6 (21.1, 55.0)	0.786
2-4 months	27.7 (16.3, 27.6)	29 (22.8, 45.6)	0.474
Regurgitation
Episodes/day
1 month	0.46 (0, 2)	0.32 (0, 2.3)	0.631
2 months	0.5 (0,1.6)	0.52 (0, 3.8)	0.631
Regurgitation by Rome IV
1 month	0	0	1.000
2 months	0	1 (5.9)	0.790
Regurgitation symptom
1 month	7 (41.2)	12 (70.6)	0.245
2 months	3 (17.6)	8 (47)	0.106
4 months	2 (11.8)	5 (29.4)	0.357
Crying 1-3 hours/day
1 month	4 (23.5)	1 (5.9)	0.136
2 months	0	1 (5.9)	0.973
4 months	1 (5.9)	0	0.838
Stool frequency (times/day)
1 month	4.1 (1, 6.8)	5.0 (0.7, 9.7)	0.247
2 months	2.7 (0.5, 9)	3.0 (0.3, 6.7)	0.610
4 months	2 (0.3, 3)	2 (0.3, 3)	0.973
Stool consistency, Bristol stool scale
1 month	6 (3.6, 7)	6 (4.8, 7)	0.845
2 months	6 (5, 7)	6 (4, 7)	0.763

L. reuteri: Limosilactobacillus reuteri.

Compliance to intervention

Compliance with the study medication was comparable between the groups, with a median for intervention day of 47.5 days (range, 13-56 days) and 51 days (range, 31-56 days) in the probiotic and placebo group, respectively (P = 0.157).

Stool microbial diversity

Gut microbiota profiles were compared across interventions and delivery modes at three time points: Month 1, 2, and 4. Gut microbiota diversity significantly differed with the mode of delivery during the first two months. Significantly higher bacterial richness was only observed at month 1 and between the placebo-NL and placebo-CS groups. At 1 month, the placebo-NL group showed significantly higher alpha diversity than the placebo-CS and probiotic-CS groups. At 2 months, alpha diversity (Shannon index) of probiotic-NL group was significantly higher than that of probiotic-CS and placebo-CS groups (Figure 2). The bacterial community composition (beta diversity) significantly differed between the groups at 2 months only (P = 0.04). By 4 months after cessation of intervention, however, no significant differences in either alpha or beta diversity were observed across the four groups (Figure 2).

Figure 2 Bacterial diversity in stool specimens of the subgroup placebo-caesarean section, placebo-normal labor, Limosilactobacillus reuteri-caesarean section, and Limosilactobacillus reuteri-normal labor across different time points at 1, 2, and 4 months. A-C: The bacterial alpha diversity within samples was calculated based on richness (Chao 1 index); D-F: The bacterial alpha diversity within samples was calculated based on richness and evenness (Shannon index); G-I: The bacterial beta diversity within samples was calculated based on dissimilarity. ^aP < 0.001 and ^bP < 0.05. NL: Normal labor; CS: Caesarean section; PC1: Principal coordinates 1; PC2: Principal coordinates 2.

Gut microbiota composition

The relative bacterial composition of gut microbiota at the phylum, genus, and species levels revealed significant differences influenced by delivery mode and intervention. At the phylum level (Figure 3A), Pseudomonadota predominated across all groups at 1 month of age. NL-delivered infants exhibited lower relative abundance of Bacillota than in the CS groups, while levels of Actinomycota were similar between groups. At 2 and 4 months, significant enrichment of Bacteroidota (LDA score = 5.1, P = 0.002) was evidenced in the NL groups, whereas Bacillota became dominant in the CS group (LDA score = 5.2, P = 0.015). A gradual decline in Pseudomonadota was observed across all groups, whereas Actinomycetota tended to increase across 4 groups over a 4-month follow-up.

Figure 3 Most abundant bacterial taxa in stool specimens of subgroup placebo-caesarean section, placebo-normal labor, Limosilactobacillus reuteri-caesarean section, and Limosilactobacillus reuteri-normal labor across different time points at 1, 2, 4 months. A: Phylum level; B: Genus level (Top10); C: Species level (Top10). NL: Normal labor; CS: Caesarean section.

At the genus level (Figure 3B), by 1 month, Bifidobacterium was observed in all groups but was present at a lower relative abundance in the placebo-CS group (P = 0.749). The probiotic-NL group exhibited significantly higher Parabacteroides abundance (LDA score = 4.21, P = 0.045) while the placebo-NL group exhibited significantly higher levels of Bacteroides (LDA score = 5.07, P = 0.002). On the other hand, CS infants, particularly those from the placebo group, showed higher relative abundances of Klebsiella and Clostridium sensu stricto, with the latter significantly enriched in the placebo-CS group (LDA score = 5.21, P = 0.026). By 2 and 4 months of age, Bifidobacterium tends to expand in all groups. Enrichment of Bacteroides in the NL groups remained significantly high at both 2 and 4 months (LDA score = 5.06, P = 0.001; LDA score = 5.14, P = 0.040, respectively). In contrast, CS infants demonstrated a transient enrichment of Hungatella, peaking at 2 months and declining by 4 months, while Clostridium dramatically decreased by 2 months.

At the species level (Figure 3C), multiple Bifidobacterium species, including Bifidobacterium longum and Bifidobacterium breve (Supplementary material), were detected in all groups by 1 month of age and gradually increased over time, with the highest relative abundances reached by four months of age (2 months after the cessation of L. reuteri DSM 17938 supplementation). However, there is no significant difference in the relative abundance of Bifidobacterium and Lactobacillus among all groups (Figure 4). At 2 months of age, the probiotic-CS group exhibited Hungatella hathewayi as the most distinctive species compared with the placebo-CS group (Figure 4).

Figure 4 Box-and-whisker plots showing the percentage of relative abundance of bacterium taxa identified by Linear Discriminant Analysis Effect Size analysis in each subgroup placebo-caesarean section, placebo-normal labor, Limosilactobacillus reuteri-caesarean section, and Limosilactobacillus reuteri-normal labor across different time points at 1, 2, 4 months. A-D: Bifidobacterium spp. in each group; E-H: Lactobacillus spp. in each group; I-L: Hungatella hathewayi in each group. Box-and-whisker plots display the median, the minimum, and the maximum values. NL: Normal labor; CS: Caesarean section.

DISCUSSION

This study highlights significant differences in the gut microbiota, depending on the mode of delivery and on daily supplementation with L. reuteri DSM 17938 over a 2-month period. Higher alpha- and beta-diversity, and a more diverse bacterial community composition, were observed in normally delivered infants during the first two months of life. In contrast to the placebo group, L. reuteri DSM 17938 daily supplementation in exclusively breastfed infants was associated with modest, transient microbiota changes, notably an increase in Bifidobacterium abundance, which attenuated after cessation of the intervention. While regurgitation symptoms and crying time tended to increase in the placebo and probiotic groups, respectively, these symptom changes did not reach statistical significance.

The establishment of the gut microbiota begins prenatally, with growing evidence challenging the long-standing hypothesis that considers the placenta and uterus as sterile environments[19]. At birth, factors such as delivery mode, feeding practices, maternal diet, antibiotic exposure, and environmental conditions shape the neonatal gut microbiota[20]. Studies in Thailand have shown how factors influence microbiota composition. The CS neonates showed lower relative abundance of Bifidobacterium and higher relative abundance of skin and environmental taxa[21]. Significant associations between regional dietary patterns and gut microbiota composition were also reported in Thailand[22,23], highlighting the influence of environmental and nutritional factors on microbial diversity. NL infants acquire a microbiota primarily from maternal fecal and vaginal flora, while CS infants are more likely to be colonized by maternal skin and environmental microbes[24].

This study focused on gut microbiota development through 4 months of age, while carefully controlling for confounding factors, including feeding mode[25]. After participants were randomly assigned to receive either a probiotic or a placebo, we selected exclusively breastfed participants and stratified them by delivery mode. At baseline, there were no significant differences in participant characteristics, allowing for unbiased comparisons. Significant differences in alpha diversity persisted at 1 and 2 months, particularly between the placebo-CS and probiotic-NL groups, highlighting the combined effects of delivery mode[26,27] and probably probiotic supplementation on microbial diversity. By 4 months, however, these differences had attenuated, possibly due to other factors, such as dietary changes and variations in breast milk composition[20]. Beta diversity analysis also revealed significant differences between the probiotic-NL and probiotic-CS groups at 2 months, but no significant differences at 1 month or 4 months, suggesting a transient effect of probiotics on microbial community composition. Our study extended microbiota analysis post-intervention, a characterization not commonly performed in previous studies[8-10,28]. We found that any effects of L. reuteri DSM 17938 on alpha diversity were short-lived. Our results were consistent with findings on preterm infants, which evidenced the reduction of probiotic effects over time once the microbiota matured[29], leading to greater microbial competition for ecological niches and nutrients[30]. Further studies with a larger number of participants are essential to measure the long-term effect mediated by the gut microbiota dynamic on infant health.

Analysis of microbial abundance highlighted the predominance of Pseudomonadota during the first month of life across all groups, which declined over time as gut maturation progressed. The gut microbiota composition of NL infants showed enrichment in Bacteroidota, whereas the microbiota of CS infants, particularly those in the placebo group, exhibited a higher abundance of Bacillota over time, indicating delivery mode-specific phylum-level colonization patterns. At the genus level, NL was associated with enrichment and persistence of Bacteroides and Parabacteroides at 2 months and 4 months, suggesting more rapid microbial maturation by natural delivery. In contrast, CS infants demonstrated higher relative abundances of Klebsiella and Clostridium sensu stricto, taxa commonly linked to delayed microbiota development, with a gradual reduction of Clostridium sensu stricto by 2 months of age. Together, these findings highlight distinct delivery-mode-dependent microbial trajectories in early infancy, further shaped by age and feeding-related factors. In contrast, probiotic supplementation appeared to modestly influence phylum- and genus-level profiles. In comparison with the key role of the delivery mode in shaping the gut microbiota, our results suggest that probiotic supplementation acts instead as a secondary modulator. Our conclusion is consistent with previous studies reporting subtle community-level effects of L. reuteri rather than marked changes in the dominant bacteria[28,31].

At the species level, multiple Bifidobacterium species were detected across all groups early in life and increased progressively over time. The consistently high relative abundance of Bifidobacterium, similarly observed in both the probiotic and placebo groups, is likely driven primarily by exclusive breastfeeding rather than probiotic supplementation. It was notably well evidenced that human milk oligosaccharides selectively promote bifidobacterial growth and are key drivers of infant gut microbiota composition[32,33]. Accordingly, the absence of significant between-group differences in Bifidobacterium abundance likely reflects a breastfeeding-dominated microbial ecosystem, with age-related maturation outweighing modest probiotic effects. Nevertheless, the sustained increase in Bifidobacterium spp. observed after cessation of the intervention suggests that L. reuteri DSM 17938 supplementation may have contributed indirectly by shaping the gut environment to bifidobacterial persistence and expansion, rather than by directly driving bifidobacterial colonization. This assertion is supported by previous studies showing that L. reuteri DSM 17938 may modulate microbial interactions and ecological conditions that facilitate the growth of beneficial bacteria[8,34]. As Bifidobacteria has vital role in maintaining gut homeostasis, supporting immune development, and promoting intestinal barrier function during early life[35,36], the progressive increase in Bifidobacterium spp. after cessation of L. reuteri supplementation might have the potential short-term benefits, such as improved gut health during the supplementation period or postprobiotic effect. These issues need further study with a larger sample size and long-term follow-up.

An additional observation was the transient enrichment of Hungatella hathewayi in the probiotic-CS group at 2 months of age. Hungatella hathewayi has been described as a potential short-chain fatty acid - producing commensal with beneficial roles in the metabolism of adult populations[37]. Despite the lack of evidence associating L. reuteri DSM 17938 supplementation with Hungatella abundance, probiotic-dependent modulation of the gut environment may indirectly support the expansion of anaerobic, short-chain fatty acid-associated bacteria, providing a plausible explanation for this transient finding.

Of note, we did not observe a consistent enrichment of L. reuteri DSM 17938 in stool based on relative abundance profiles. This finding does not necessarily mean a lack of probiotic exposure or activity, as probiotic effects may occur without sustained intestinal colonization. In addition, strain-level detection of L. reuteri DSM 17938 may be limited by 16S rRNA gene-based sequencing approaches, particularly when the bacteria are poorly abundant or transiently present. Previous studies using sequencing-based methods have similarly reported minimal effects of L. reuteri DSM 17938 on overall microbiota composition[28,38], whereas studies using strain-specific assays (e.g., quantitative real-time polymerase chain reaction or culture-based methods) have been more suitable for documenting probiotic recovery or colonization[39,40]. Methodological differences may therefore contribute to heterogeneity across studies. Several earlier trials on infantile colic used targeted approaches (e.g., fluorescence in situ hybridization, quantitative real-time polymerase chain reaction, or terminal restriction fragment length polymorphism)[8-10,41] for the quantification of selected taxa or detection of probiotic recovery, whereas broader sequencing approaches may better capture community-wide patterns but remain limited for strain attribution.

Regarding gastrointestinal symptoms, although there was a trend toward fewer symptoms in the probiotic group, no statistically significant differences were observed in the clinical response to probiotic supplementation. Previous randomized controlled trials evaluating L. reuteri DSM 17938 supplementation in infants reported heterogeneous clinical outcomes, ranging from symptom improvement to no observable benefit. Differences in baseline microbiota composition, geographic context, and analytical methodologies may contribute to these inconsistencies[8-10,28].

A key strength of this study is its longitudinal design, which enabled evaluation of changes in the gut microbiota across several time points from birth to four months with an adequate sample size. Randomized controlled trial design can strengthen the validity of the findings by reducing potential selection bias and confounding factors. In addition, Oxford Nanopore sequencing enabled in-depth, species-level analysis with high accuracy.

However, this study has several limitations. Firstly, the relatively small sample size and short follow-up period focusing on a specific population may limit the strength of the conclusions. Larger studies with longer follow-up periods and more diverse populations are needed to confirm our present results. Secondly, these findings may not be directly generalizable to formula-fed infants, as feeding type is known to influence gut microbial composition and immune development. Studies including formula-fed infants are warranted to confirm our findings. Finally, 16S rRNA sequencing provides information on gut microbiota composition and species, but does not reveal the functional impact of the observed microbial alterations. Future studies using metagenomic sequencing and metabolomic analyses may provide deeper insights into the functional mechanisms associated with L. reuteri supplementation. Nonetheless, our findings provide valuable insights into early microbiota development and support future research on probiotic interventions in infancy.

CONCLUSION

L. reuteri DSM 17938 supplementation during the first 2 months of life induced modest and transient changes in gut microbiota composition, particularly an increase in Bifidobacterium abundance, without sustained microbiota alterations or significant effects on gastrointestinal symptoms.

ACKNOWLEDGEMENTS

The authors would like to thank the participants and their families for their valuable contributions to this study. We also acknowledge the Chula Data Management Center, Faculty of Medicine, Chulalongkorn University, for their support in data management. Study data were collected and managed using REDCap electronic data capture tools hosted at Chula Data Management, Faculty of Medicine, Chulalongkorn University. We would like to thank Ms. Nussara Prasertsri and Ms. Suttida Wattanapornsopa, nursery nurses at King Chulalongkorn Memorial Hospital, for their assistance with participant recruitment and baseline stool collection. We also gratefully acknowledge the research team, Mr. Santirat Sopee, Ms. Arisa Ama, and Ms. Pakpine Phunnoi, for their support in data collection and overall study coordination.