Brief early antibiotic exposure (≤2 days) is not associated with disruption of gut microbiome in very low birth weight infants

Early empirical antibiotic therapy is common in preterm and very low birth weight (VLBW) infants but may disrupt the developing gut microbiome. However, the effects of brief antibiotic courses remain unclear, particularly in the most immature infants. In this prospective multicenter cohort study, we examined gut microbiome trajectories in VLBW infants (many of whom were extremely preterm) receiving no antibiotics, a short course (≤2 days), or prolonged exposure (≥3 days). Serial stool samples were analyzed using 16S rRNA gene sequencing. Microbiome composition and diversity in the short-course group were similar to those in unexposed infants at all timepoints, indicating that brief antibiotic exposure did not disrupt microbial development. In contrast, prolonged exposure was associated with transient dysbiosis characterized by reduced Bifidobacterium abundance and lower alpha diversity, with partial recovery by discharge. These findings suggest that limiting empirical antibiotic therapy to ≤2 days (48 h) when infection is unconfirmed may not disrupt microbiome development even in highly immature preterm VLBW infants, supporting evidence-based antibiotic stewardship in neonatal intensive care.

1 Introduction

Advances in neonatal care have markedly improved the survival of extremely preterm infants (Patel et al., 2015; Pierrat et al., 2017; Cheong et al., 2020). However, infections remain a major cause of morbidity and mortality in this vulnerable population (Weston et al., 2011; Stoll et al., 2015). Among these infections, early-onset sepsis (EOS), which typically presents within the first 72 h of life, is more frequent and has higher fatality rates in preterm infants than in term neonates (Stoll et al., 2011; Shane et al., 2017). Consequently, empirical antibiotic treatment is commonly initiated immediately after birth in preterm infants considered at high risk for EOS, such as those born to mothers with clinical chorioamnionitis, preterm premature rupture of membranes, or other signs of intrauterine infection, even when blood cultures are ultimately sterile (Puopolo et al., 2018).

To ensure clarity and consistency in terminology, we used internationally established definitions: preterm infants as those born at <37 weeks’ gestation (Blencowe et al., 2012), very low birth weight (VLBW) infants as those with a birth weight <1,500 g (World Health Organization, 1992), and extremely preterm infants as those born at <28 weeks’ gestation (Blencowe et al., 2012). Our study population consisted of preterm VLBW infants, the majority of whom were also extremely preterm.

While early empirical antibiotics are often necessary to prevent life-threatening infection, accumulating evidence suggests that they can disrupt the development of the neonatal gut microbiome—a condition known as dysbiosis (Greenwood et al., 2014; Gasparrini et al., 2019). Prolonged antibiotic exposure during this critical developmental period has been associated with increased risks of necrotizing enterocolitis (NEC) and death (Cotten et al., 2009; Alexander et al., 2011). Perturbation of the early microbiome may also impair immune maturation, nutrient absorption, and growth (Gensollen et al., 2016; Uzan-Yulzari et al., 2021) and has been linked to long-term health consequences including gastrointestinal, allergic, metabolic, and neurodevelopmental disorders (Kalliomäki et al., 2008; Vangay et al., 2015; van den Berg et al., 2016; Ihekweazu and Versalovic, 2018).

These concerns have led to increasing emphasis on antibiotic stewardship—defined as the coordinated, evidence-based effort to optimize antibiotic use by ensuring that antibiotics are administered only when necessary, with the appropriate agent, dose, and duration—to minimize unintended harms such as antimicrobial resistance, toxicity, and disruption of the developing microbiome [Centers for Disease Control and Prevention (CDC), 2019].

In recognition of these risks, several international guidelines now recommend discontinuing empirical antibiotics within 36–48 h in infants with suspected EOS if blood cultures remain negative and the clinical course is reassuring [Puopolo et al., 2018; National Institute for Health and Care Excellence (NICE), 2024]. Nevertheless, the degree to which antibiotic exposure alters the neonatal microbiome likely depends on factors such as treatment duration, antibiotic spectrum, and host maturity (Arrieta et al., 2014; Gasparrini et al., 2016; Chang et al., 2021). Broad-spectrum antibiotics, in particular, are thought to have a greater influence on microbial composition (Gibson et al., 2016; Reyman et al., 2022). While longer courses of antibiotics have more profound effects, even short-term antibiotic use may significantly alter gut microbiome (Greenwood et al., 2014; Zwittink et al., 2018). Persistent alterations in microbial communities, including colonization by antibiotic-resistant organisms, have been reported even after discharge (Gasparrini et al., 2019).

A recent study, however, suggested that very short antibiotic exposure—specifically those discontinued within 48 h—had no measurable impact on the gut microbiome of moderately preterm infants with a mean gestational age of about 32 weeks (Kim et al., 2021). However, data remain scarce in extremely preterm and VLBW infants, whose intestinal immaturity, delayed colonization, and limited microbial diversity may render them far more susceptible to antibiotic-induced perturbations.

Therefore, we aimed to determine whether short-course antibiotic exposure—defined as completion by day of life 2—alters gut microbiome development in preterm and VLBW infants. Our cohort included particularly immature infants (median gestational age 27.5 weeks, birth weight 898 g), allowing us to evaluate microbial consequences of antibiotic exposure in one of the most high-risk neonatal populations.

2 Materials and methods

2.1 Study design and participants

This study was part of a prospective multicenter cohort study initiated in December 2021 and is still ongoing. It is being conducted at two level 3 neonatal intensive care units in Japan: University of Miyazaki Hospital and the Tokyo Metropolitan Children’s Medical Center. The parent study was designed to investigate the development of the neonatal gut microbiome in preterm infants through longitudinal stool sample collection and microbiome analysis.

Eligible participants were preterm and very low birth weight (VLBW) infants, defined as those born at <37 weeks of gestation with a birth weight <1,500 g; those with major congenital anomalies were excluded. In the parent study, stool samples were collected longitudinally starting around the second day of life and continued until discharge.

For the present secondary analysis, we focused on the association between the duration of empirical antibiotic exposure and the composition of the gut microbiome. From the cohort of infants who did not develop gastrointestinal complications, we selected 30 preterm infants, who were subsequently divided into three groups according to the duration of antibiotic administration: None (no antibiotics), Short (≤2 days of age), and Long (≥3 days of age). The cutoff of 2 days for the Short group was chosen to align with clinical guidelines recommending discontinuation within 48 h and to reflect real-world clinical practice (Puopolo et al., 2018). Each group included 10 infants (n = 10 per group), comprising 20 infants from the University of Miyazaki Hospital and 10 infants from the Tokyo Metropolitan Children’s Medical Center. A total of 90 stool samples from these 30 infants were analyzed, collected at three predefined time points: week 2 (12–16 days of age), week 4 (26–30 days of age), and at discharge or at 40 weeks’ corrected age. The sample size was determined by the number of eligible infants with complete longitudinal stool sampling available at the time of analysis.

Perinatal and neonatal management decisions were made by the attending physicians. In most cases, prolonged antibiotic administration (Long group) was due to elevated or persistently high infection-related biomarkers such as C-reactive protein, the absence of conclusive culture results, or unstable respiratory conditions that prevented early discontinuation despite negative cultures. Histological chorioamnionitis was diagnosed based on placental pathology (Redline, 2012). For neonates suspected of early-onset sepsis, empirical antibiotic therapy with ampicillin and an aminoglycoside (gentamicin or amikacin) was initiated immediately after birth. Feeding was initiated with maternal breast milk or donor breast milk in most cases. One relatively more mature infant was started on formula feeding.

This study was conducted in accordance with the Declaration of Helsinki, and ethical approval was obtained from the Ethics Committee of the University of Miyazaki Hospital (approval number: O-1053). Written informed consent was obtained from the parents of all participating infants.

2.2 Sample collection and DNA extraction

Fecal samples from soiled diapers were collected during hospitalization at week 2 (12–16 days of age), week 4 (26–30 days of age), and at discharge or at 40 weeks’ corrected age. All samples were stored at −80 °C until DNA extraction (Carroll et al., 2012). DNA was extracted using the QIAamp PowerFecal Pro DNA kit (Qiagen, Hilden, Germany) and quantified using a NanoDrop ND-2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s instructions.

2.3 16S rRNA gene amplicon sequencing and microbiota analysis

The V3-V4 region of the bacterial 16S rRNA gene was amplified using the following PCR primers: forward primer, 5′-TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGCCTACGGGNGGCWGCAG-3′; reverse primer, 5′-GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGACTACHVGGGTATCTAATCC-3′ (Klindworth et al., 2013). Then, the amplicons were sequenced using an Illumina MiSeq platform (Illumina, San Diego, CA, USA).

Sequencing data were processed using QIIME 2 (ver. 2023.9) (Bolyen et al., 2019). Raw sequences were demultiplexed, quality-filtered, and denoised using DADA2 to generate amplicon sequence variants. Taxonomic classification was performed using a classifier pretrained on the SILVA 138 reference database (Quast et al., 2012).

Alpha diversity was assessed using metrics such as the Shannon index and Faith’s phylogenetic diversity. Beta diversity was evaluated based on Bray–Curtis dissimilarity and unweighted UniFrac distances and visualized using principal coordinates analysis (PCoA). To aid interpretation, 95% confidence ellipses representing the distribution of each group were calculated and plotted around the group centroids, using the scikit-bio and matplotlib libraries in Python (ver. 3.11).

2.4 Statistical analysis

Continuous variables are expressed as the median and interquartile range (IQR), while categorical variables are expressed as counts with percentages. Alpha diversity indices were compared among the three groups (None, Short, Long) using the Kruskal–Wallis test, followed by Bonferroni-corrected post hoc tests when appropriate. At the phylum level, the four most abundant phyla (Actinomycetota, Bacteroidota, Bacillota, and Pseudomonadota), which accounted for nearly all detected sequences, were compared among the three groups using the same statistical approach. At the genus level, Bifidobacterium—given its established clinical importance—was analyzed in detail. For visualization and summary reporting, the top 13 genera based on mean relative abundance (including Bifidobacterium) were shown in Figure 2A, and their quantitative values and statistical comparisons are provided in Supplementary Table 2. Group differences were evaluated using the same tests. Beta diversity was evaluated based on Bray–Curtis dissimilarity and unweighted UniFrac distances, and statistical comparisons between groups were conducted using PERMANOVA with 999 permutations. For completeness, pairwise PERMANOVA was additionally performed to evaluate differences between individual group pairs. Statistical analyses were performed using SPSS Statistics ver. 28 (IBM Corp., Armonk, NY, USA) and R software ver. 4.2.2 (R Foundation for Statistical Computing, Vienna, Austria). A two-sided p value of <0.05 was considered statistically significant.

Figure 1

Figure 1. Gut microbiome composition at the phylum level. Relative abundance of gut microbiome at the phylum level at 2 weeks, 4 weeks, and discharge in the three groups (None, Short, and Long). At 2 weeks, Actinomycetota and Bacteroidota differed significantly among groups, while at 4 weeks, significant differences were observed in Bacillota and Pseudomonadota. Longitudinally, the abundance of Actinomycetota and Pseudomonadota increased over time, whereas Bacillota decreased. Statistical comparisons were conducted using the Kruskal–Wallis test with Bonferroni correction; detailed p-values are provided in Supplementary Table 1.

3 Results

3.1 Patient characteristics

The clinical characteristics of the preterm and very low birth weight (VLBW) infants are summarized in Table 1. The participants were categorized into three groups based on the duration of initial antibiotic exposure: None (no antibiotic use), Short (antibiotics ≤2 days), and Long (antibiotics ≥3 days). Overall, the median (IQR) gestational age was 27.5 (25.8, 29.3) weeks, and the median (IQR) birth weight was 898 (680, 1,227) g. Gestational age was significantly lower in the None group [29.5 (28, 33) weeks] compared with the Short group [26 (24.8, 28.3) weeks] and Long group [26 (23.8, 27.3) weeks], with no significant difference between the Short and Long groups. Among the baseline characteristics, gestational age, maternal antibiotic use, and chorioamnionitis showed significant differences across the three groups, whereas none of the other variables differed significantly. Notably, maternal antibiotic use and chorioamnionitis were absent in the None group but were present in both the Short and Long groups, although the difference between the latter two was not significant. Among the infants in the Long group, the main reasons for extended antibiotic use included persistently elevated C-reactive protein or other inflammatory markers, absence of definitive negative culture results, and ongoing respiratory instability.

Table 1

Table 1. Clinical characteristics of the study participants.

3.2 Gut microbiota composition at the phylum level

The relative abundance of gut microbiota at the phylum level at 2 weeks, 4 weeks, and discharge is shown in Figure 1. The original data and statistical comparisons among the three groups are provided in Supplementary Table 1.

Figure 2

Figure 2. Genus-level microbiome composition and relative abundance of Bifidobacterium. (A) Genus-level gut microbiome composition at 2 weeks, 4 weeks, and discharge in each group. (B) Relative abundance of Bifidobacterium at 2 weeks, 4 weeks, and discharge. At 2 weeks, the abundance of Bifidobacterium was significantly lower in the Long group compared with the None and Short groups (*p < 0.05; overall p = 0.020). At 4 weeks, the Long group continued to show lower values, but the difference was not statistically significant (p = 0.118). Differences had diminished by discharge. Statistical comparisons were performed using the Kruskal–Wallis test with Bonferroni correction.

At 2 weeks of age, significant differences were observed in the abundance of Actinomycetota (None 23%, Short 11%, Long 1%; p = 0.028) and Bacteroidota (None 0%, Short 2%, Long 0%; p = 0.048).

At 4 weeks, Bacillota (None 51%, Short 40%, Long 93%; p = 0.006) and Pseudomonadota (None 30%, Short 45%, Long 4%; p = 0.024) showed statistically significant differences.

Longitudinally, Actinomycetota and Pseudomonadota increased over time, whereas Bacillota decreased from 2 weeks to discharge.

3.3 Genus-level composition and Bifidobacterium abundance

Figure 2A shows the genus-level composition of the gut microbiota. Given the well-established importance of Bifidobacterium in neonatal gut health (Bäckhed et al., 2015; Underwood et al., 2015; Saturio et al., 2021), we specifically compared its relative abundance among the three groups (Figure 2B).

At 2 weeks of age, the mean relative abundances of Bifidobacterium were as follows: None group, 22%; Short group, 10%; and Long group, 0%. There was a significant overall difference among the groups (Kruskal–Wallis test, p = 0.020). Pairwise comparisons showed that the abundance of Bifidobacterium was significantly lower in the Long group compared with the None group (p = 0.042) and the Short group (p = 0.048).

At 4 weeks of age, the mean relative abundances were as follows: None group, 17%; Short group, 15%; and Long group, 2%. The Short group did not differ from the None group, while the abundance in the Long group was lower but this difference did not reach statistical significance (p = 0.118).

In addition to Bifidobacterium, we explored the dynamics of other early-colonizing genera with relatively high abundance. The full genus-level quantitative results and statistical comparisons are provided in Supplementary Table 2. At discharge, the relative abundance of Staphylococcus differed among the three groups (None 3.3%, Short 1.6%, Long 0.2%; p = 0.044), although subsequent pairwise comparisons did not reveal significant differences. No other genera exhibited consistent or notable group-related differences.

By the time of discharge, the differences among the groups had further diminished, suggesting partial recovery of Bifidobacterium colonization over time.

3.4 Beta diversity analysis

Beta diversity was evaluated using Bray–Curtis dissimilarity and unweighted UniFrac distances, and visualized PCoA (Figure 3). For visualization, 95% confidence ellipses were overlaid for each group to illustrate the distribution and separation of microbial communities. The original data, showing PCoA1 and PCoA2 values for all 30 samples, are provided in Supplementary Table 3.

Figure 3

Figure 3. Beta diversity of gut microbiome. Beta diversity was evaluated using Bray–Curtis dissimilarity and unweighted UniFrac distances, and visualized by principal coordinates analysis (PCoA) at 2 weeks, 4 weeks, and discharge. Colored ellipses represent 95% confidence intervals for each group, calculated from the PCoA coordinates using Python. At 2 weeks, overall group differences were significant (p = 0.037), with a significant difference between the Short and Long groups (p = 0.018). At 4 weeks, differences became more pronounced (overall p = 0.007), with pairwise differences between the None and Long groups (p = 0.037) and between the Short and Long groups (p = 0.002). No significant difference was found between the None and Short groups. By discharge, group differences were no longer observed. Statistical comparisons were performed using PERMANOVA.

At 2 weeks of age, a significant difference in overall community structure was observed among the three groups (p = 0.037), indicating that they formed distinct microbial communities. Pairwise comparison showed that the Short and Long groups differed significantly (p = 0.018), while no significant difference was observed between the other pairs.

At 4 weeks of age, the separation between the Long group and the other groups became more pronounced. Overall group differences were significant (p = 0.007), and pairwise comparisons revealed significant differences between the None and Long groups (p = 0.037), and between the Short and Long groups (p = 0.002). There was no significant difference between the None and Short groups.

By the time of discharge, these differences were no longer observed.

3.5 Alpha diversity

Figure 4 shows the alpha diversity evaluated according to the Shannon diversity index (Figure 4A) and Faith’s phylogenetic diversity (Figure 4B).

Figure 4

Figure 4. Alpha diversity of gut microbiome. (A) Shannon diversity index and (B) Faith’s phylogenetic diversity at 2 weeks, 4 weeks, and discharge. At 4 weeks, both indices showed significantly lower alpha diversity in the Long group compared with the None and Short groups (*p < 0.05, **p < 0.01). At discharge, no significant differences in alpha diversity were observed among the groups for either index, suggesting partial recovery of diversity over time. Statistical comparisons were performed using the Kruskal–Wallis test with Bonferroni correction.

At 4 weeks, the Long group exhibited significantly lower Shannon diversity compared with the None group (mean 1.6 vs. 2.3, p = 0.041), while the Short group (mean 2.1) showed no significant difference compared with the None group. Similarly, Faith’s phylogenetic diversity was significantly lower in the Long group compared with both the None group (mean 2.9 vs. 4.5) and the Short group (mean 2.9 vs. 4.2, p < 0.01 for both), indicating a marked reduction in phylogenetic richness associated with prolonged antibiotic exposure.

At the time of discharge, no significant differences in alpha diversity were observed among the three groups for either metric, suggesting a partial recovery or convergence of microbial diversity over time.

4 Discussion

In this study, we investigated the impact of early antibiotic exposure on the gut microbiome development in preterm and very low birth weight (VLBW) infants, most of whom were extremely preterm (median gestational age 27.5 weeks, birth weight 898 g). We found that brief antibiotic exposure (≤2 days) did not alter the composition or diversity of the gut microbiome at 2 and 4 weeks of age compared with infants who did not receive antibiotics. In contrast, prolonged exposure (≥3 days, median 4 days) was associated with reduced Bifidobacterium abundance and lower microbial diversity at the same time points. These differences had diminished by discharge, suggesting a partial recovery of the gut microbiome over time. Collectively, our findings indicate that antibiotic duration determines whether measurable disruption occurs, with brief exposure showing no detectable alteration of gut microbiome development. Furthermore, our study provides insights into the effects of early antibiotic exposure, specifically in extremely preterm infants, a population that has been less well characterized in previous research.

Our findings align in part with previous studies that demonstrated significant disruptions of the neonatal gut microbiota following antibiotic administration. Prolonged exposure in preterm infants has been associated with reduced microbial diversity, delayed colonization by commensal bacteria such as Bifidobacterium, and increased risks of NEC and late-onset sepsis (Cotten et al., 2009; Greenwood et al., 2014; Pammi et al., 2017; Zwittink et al., 2018; Chang et al., 2021). These alterations can persist for weeks or even months after treatment (Tanaka et al., 2009; Yassour et al., 2016; Gasparrini et al., 2019).

Evidence on the impact of short antibiotic courses is limited. The only study to specifically address this issue, to our knowledge, reported that antibiotic exposure limited to ≤48 h had no measurable impact on microbial composition compared with unexposed infants (Kim et al., 2021). Importantly, that study population (mean gestational age, about 32 weeks) was more mature than ours, raising uncertainty about whether the same applies to extremely preterm infants. Notably, our cohort was considerably less mature and included many extremely preterm infants (<28 weeks), a population with delayed intestinal maturation, limited microbial resilience, and heightened vulnerability to external perturbations. The fact that we similarly observed no disruption at ≤2 days of exposure strengthens the conclusion that brief administration of empirical antibiotics, even in extremely preterm infants, does not cause lasting microbial disturbance.

In contrast, our findings in the Long group (≥3 days, median 4 days) consistent with prior studies demonstrating measurable dysbiosis after extended antibiotic exposure (Greenwood et al., 2014; Zwittink et al., 2018; Chang et al., 2021). However, unlike some reports describing persistent disturbances weeks or months after exposure (Tanaka et al., 2009; Yassour et al., 2016; Gasparrini et al., 2019), we found that microbial composition had converged by discharge. The relatively short duration of treatment in our cohort—typically 4 days, shorter than in many earlier studies where antibiotic courses often exceeded 7–10 days—may partly explain this recovery. These results support a duration-dependent relationship between antibiotic exposure and gut microbiome disruption: while exposure ≤48 h appears to have minimal effects, extending treatment beyond this threshold can induce measurable, though transient, alterations.

Mechanistically, antibiotics likely disrupt gut microbial development through depletion of key pioneer taxa such as Bifidobacterium, which play critical roles in shaping early intestinal ecology and promoting mucosal immune maturation (Saturio et al., 2021; Lin et al., 2022;). Their depletion may lead to reduced microbial diversity and delayed establishment of a stable community (Saturio et al., 2021; Turroni et al., 2022). Conversely, the limited effect of brief exposure (≤2 days) may reflect the rapid restoration of this pioneer genus once antibiotic pressure is removed. These findings suggest that the neonatal gut ecosystem has a degree of resilience to short perturbations, whereas prolonged antibiotic pressure may exceed this compensatory capacity, resulting in transient dysbiosis.

Our results have important clinical implications for neonatal care. Empirical antibiotics remain indispensable in preterm infants at high risk for EOS, where the consequences of delayed or missed treatment can be fatal. However, unnecessary prolongation of antibiotics can disrupt the fragile process of gut microbial colonization, with potential downstream effects on immunity, metabolism, and neurodevelopment. Our findings suggest that discontinuing antibiotics within 36–48 h, as recommended by current guidelines when cultures are negative and the infant remains clinically stable [Puopolo et al., 2018; National Institute for Health and Care Excellence (NICE), 2024], is unlikely to cause measurable disruption of the gut microbiome, even in extremely preterm infants. In contrast, extending treatment beyond this period (≥3 days) was associated with detectable, though reversible, alterations. These results reinforce the importance of limiting antibiotic duration to the minimum necessary for safety, thereby balancing the benefits of infection control with the preservation of early microbial development.

Ultimately, decisions about antibiotic use in preterm infants must balance the clear benefits of infection control against the potential risks of dysbiosis. The present study provides microbiome-based evidence supporting the safety of brief empirical antibiotic use and highlights that timely discontinuation of empirical antibiotics preserves gut microbiome development without compromising safety.

4.1 Strengths and limitations

The strengths of our study include its focus on extremely preterm infants, who are underrepresented in prior research, and the longitudinal assessment of gut microbiome at multiple time points until discharge. This design allowed us to capture both early disruptions and subsequent recovery patterns.

Several limitations should also be acknowledged. First, the sample size was modest, limiting statistical power and the ability to adjust for all confounders. However, although some perinatal factors such as gestational age, maternal antibiotic use, and chorioamnionitis differed among the three groups, these variables did not differ significantly between the Short and Long groups. Moreover, despite the less favorable perinatal conditions in the Short group compared with the None group—conditions generally associated with negative microbiome effects—their microbial profiles were largely comparable. These findings suggest that the observed differences were primarily related to antibiotic duration rather than baseline disparities.

Second, analyses were based on 16S rRNA gene sequencing rather than shotgun metagenomics, restricting taxonomic resolution and functional insights.

Third, because of the small number of infants, we were unable to evaluate whether differences in antibiotic exposure influenced the occurrence of NEC, sepsis, or other morbidities.

In addition, antibiotic duration was determined by the attending physicians based on clinical and laboratory findings rather than by protocol, indication bias cannot be entirely excluded.

Finally, our analysis was limited to the hospitalization period; therefore, the longer-term consequences of early antibiotic exposure could not be assessed.

5 Conclusion

Brief empirical antibiotic exposure (≤2 days) did not alter gut microbiome development in preterm and very low birth weight (VLBW) infants, whereas prolonged exposure (≥3 days, median 4 days) was associated with transient dysbiosis, characterized by reduced Bifidobacterium abundance and lower microbial diversity at 2–4 weeks of age. These findings highlight the importance of minimizing unnecessary antibiotic use and support guideline-recommended discontinuation within 48 h when sepsis is not confirmed. Future studies with larger cohorts, extended longitudinal follow-up, and metagenomic approaches are needed to clarify the functional and long-term consequences of early-life dysbiosis and identify strategies that promote optimal microbiome development in this vulnerable population.

Data availability statement

The data presented in this study are publicly available. The data can be found here: https://www.ncbi.nlm.nih.gov, accession PRJNA1346831.

Ethics statement

The studies involving humans were approved by Ethics Committee of the University of Miyazaki Hospital. The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation in this study was provided by the participants’ legal guardians/next of kin.

Author contributions

YA: Investigation, Funding acquisition, Conceptualization, Writing – original draft, Data curation, Writing – review & editing, Project administration, Formal analysis, Methodology. MT: Funding acquisition, Writing – review & editing, Investigation. KOc: Investigation, Writing – review & editing. KT: Investigation, Writing – review & editing. UM: Investigation, Writing – review & editing. KOk: Supervision, Writing – review & editing. HM: Supervision, Writing – review & editing.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This study was supported by a Clinical Research Support Grant from the University of Miyazaki Hospital (no. 31010614). The sponsor had no role in the study design, in the collection, analysis, or interpretation of data, in the writing of the report, or in the decision to submit the article for publication.

Acknowledgments

We thank all the infants and their parents who participated in this study. We are also deeply grateful to the physicians, nurses, and other clinical staff who provided dedicated care to the patients.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmicb.2025.1742512/full#supplementary-material

Abbreviations

EOS, early-onset sepsis; NEC, necrotizing enterocolitis; VLBW, very low birth weight.

References

Alexander, V. N., Northrup, V., and Bizzarro, M. J. (2011). Antibiotic exposure in the newborn intensive care unit and the risk of necrotizing enterocolitis. J. Pediatr. 159, 392–397. doi: 10.1016/j.jpeds.2011.02.035,

PubMed Abstract | Crossref Full Text | Google Scholar

Arrieta, M. C., Stiemsma, L. T., Amenyogbe, N., Brown, E. M., and Finlay, B. (2014). The intestinal microbiome in early life: health and disease. Front. Immunol. 5:427. doi: 10.3389/fimmu.2014.00427,

PubMed Abstract | Crossref Full Text | Google Scholar

Bäckhed, F., Roswall, J., Peng, Y., Feng, Q., Jia, H., Kovatcheva-Datchary, P., et al. (2015). Dynamics and stabilization of the human gut microbiome during the first year of life. Cell Host Microbe 17, 690–703. doi: 10.1016/j.chom.2015.04.004,

PubMed Abstract | Crossref Full Text | Google Scholar

Blencowe, H., Cousens, S., Oestergaard, M. Z., Chou, D., Moller, A. B., Narwal, R., et al. (2012). National, regional, and worldwide estimates of preterm birth rates in the year 2010 with time trends since 1990 for selected countries: a systematic analysis and implications. Lancet 379, 2162–2172. doi: 10.1016/s0140-6736(12)60820-4,

PubMed Abstract | Crossref Full Text | Google Scholar

Bolyen, E., Rideout, J. R., Dillon, M. R., Bokulich, N. A., Abnet, C. C., Al-Ghalith, G. A., et al. (2019). Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat. Biotechnol. 37, 852–857. doi: 10.1038/s41587-019-0209-9,

PubMed Abstract | Crossref Full Text | Google Scholar

Carroll, I. M., Ringel-Kulka, T., Siddle, J. P., Klaenhammer, T. R., and Ringel, Y. (2012). Characterization of the fecal microbiota using high-throughput sequencing reveals a stable microbial community during storage. PLoS One 7:e46953. doi: 10.1371/journal.pone.0046953,

PubMed Abstract | Crossref Full Text | Google Scholar

Centers for Disease Control and Prevention (CDC). (2019). The Core elements of hospital antibiotic stewardship programs. Available online at: https://www.cdc.gov/antibiotic-use/hcp/core-elements/hospital.html (accessed December 10, 2025)

Google Scholar

Chang, H. Y., Chiang Chiau, J. S., Ho, Y. H., Chang, J. H., Tsai, K. N., Liu, C. Y., et al. (2021). Impact of early empiric antibiotic regimens on the gut microbiota in very low birth weight preterm infants: an observational study. Front. Pediatr. 9:651713. doi: 10.3389/fped.2021.651713,

PubMed Abstract | Crossref Full Text | Google Scholar

Cheong, J. L., Spittle, A. J., Burnett, A. C., Anderson, P. J., and Doyle, L. W. (2020). Have outcomes following extremely preterm birth improved over time? Semin. Fetal Neonatal Med. 25:101114. doi: 10.1016/j.siny.2020.101114,

PubMed Abstract | Crossref Full Text | Google Scholar

Cotten, C. M., Taylor, S., Stoll, B., Goldberg, R. N., Hansen, N. I., Sánchez, P. J., et al. (2009). Prolonged duration of initial empirical antibiotic treatment is associated with increased rates of necrotizing enterocolitis and death for extremely low birth weight infants. Pediatrics 123, 58–66. doi: 10.1542/peds.2007-3423,

PubMed Abstract | Crossref Full Text | Google Scholar

Gasparrini, A. J., Crofts, T. S., Gibson, M. K., Tarr, P. I., Warner, B. B., and Dantas, G. (2016). Antibiotic perturbation of the preterm infant gut microbiome and resistome. Gut Microbes 7, 443–449. doi: 10.1080/19490976.2016.1218584,

PubMed Abstract | Crossref Full Text | Google Scholar

Gasparrini, A. J., Wang, B., Sun, X., Kennedy, E. A., Hernandez-Leyva, A., Ndao, I. M., et al. (2019). Persistent metagenomic signatures of early-life hospitalization and antibiotic treatment in the infant gut microbiota and resistome. Nat. Microbiol. 4, 2285–2297. doi: 10.1038/s41564-019-0550-2,

PubMed Abstract | Crossref Full Text | Google Scholar

Gensollen, T., Iyer, S. S., Kasper, D. L., and Blumberg, R. S. (2016). How colonization by microbiota in early life shapes the immune system. Science 352, 539–544. doi: 10.1126/science.aad9378,

PubMed Abstract | Crossref Full Text | Google Scholar

Gibson, M. K., Wang, B., Ahmadi, S., Burnham, C. A., Tarr, P. I., Warner, B. B., et al. (2016). Developmental dynamics of the preterm infant gut microbiota and antibiotic resistome. Nat. Microbiol. 1:16024. doi: 10.1038/nmicrobiol.2016.24,

PubMed Abstract | Crossref Full Text | Google Scholar

Greenwood, C., Morrow, A. L., Lagomarcino, A. J., Altaye, M., Taft, D. H., Yu, Z., et al. (2014). Early empiric antibiotic use in preterm infants is associated with lower bacterial diversity and higher relative abundance of Enterobacter. J. Pediatr. 165, 23–29. doi: 10.1016/j.jpeds.2014.01.010,

PubMed Abstract | Crossref Full Text | Google Scholar

Ihekweazu, F. D., and Versalovic, J. (2018). Development of the Pediatric gut microbiome: impact on health and disease. Am J Med Sci 356, 413–423. doi: 10.1016/j.amjms.2018.08.005,

PubMed Abstract | Crossref Full Text | Google Scholar

Kalliomäki, M., Collado, M. C., Salminen, S., and Isolauri, E. (2008). Early differences in fecal microbiota composition in children may predict overweight. Am. J. Clin. Nutr. 87, 534–538. doi: 10.1093/ajcn/87.3.534,

PubMed Abstract | Crossref Full Text | Google Scholar

Kim, C. S., Grady, N., Derrick, M., Yu, Y., Oliphant, K., Lu, J., et al. (2021). Effect of antibiotic use within first 48 hours of life on the preterm infant microbiome: a randomized clinical trial. JAMA Pediatr. 175, 303–305. doi: 10.1001/jamapediatrics.2020.4916,

PubMed Abstract | Crossref Full Text | Google Scholar

Klindworth, A., Pruesse, E., Schweer, T., Peplies, J., Quast, C., Horn, M., et al. (2013). Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic Acids Res. 41:e1. doi: 10.1093/nar/gks808,

PubMed Abstract | Crossref Full Text | Google Scholar

Lin, C., Lin, Y., Zhang, H., Wang, G., Zhao, J., Zhang, H., et al. (2022). Intestinal ‘infant-type’ bifidobacteria mediate immune system development in the first 1000 days of life. Nutrients 14:1498. doi: 10.3390/nu14071498,

PubMed Abstract | Crossref Full Text | Google Scholar

National Institute for Health and Care Excellence (NICE). (2024). Neonatal infection: antibiotics for prevention and treatment. NICE guideline NG195. Available online at: https://www.nice.org.uk/guidance/ng195 (accessed July 10, 2025)

Google Scholar

Pammi, M., Cope, J., Tarr, P. I., Warner, B. B., Morrow, A. L., Mai, V., et al. (2017). Intestinal dysbiosis in preterm infants preceding necrotizing enterocolitis: a systematic review and meta-analysis. Microbiome 5:31. doi: 10.1186/s40168-017-0248-8,

PubMed Abstract | Crossref Full Text | Google Scholar

Patel, R. M., Kandefer, S., Walsh, M. C., Bell, E. F., Carlo, W. A., Laptook, A. R., et al. (2015). Causes and timing of death in extremely premature infants from 2000 through 2011. N. Engl. J. Med. 372, 331–340. doi: 10.1056/NEJMoa1403489,

PubMed Abstract | Crossref Full Text | Google Scholar

Pierrat, V., Marchand-Martin, L., Arnaud, C., Kaminski, M., Resche-Rigon, M., Lebeaux, C., et al. (2017). Neurodevelopmental outcome at 2 years for preterm children born at 22 to 34 weeks’ gestation in France in 2011: EPIPAGE-2 cohort study. BMJ 358:j3448. doi: 10.1136/bmj.j3448,

PubMed Abstract | Crossref Full Text | Google Scholar

Puopolo, K. M., Benitz, W. E., and Zaoutis, T. E. (2018). Management of neonates born at ≤34 6/7 weeks’ gestation with suspected or proven early-onset bacterial sepsis. Pediatrics 142:e20182896. doi: 10.1542/peds.2018-2896,

PubMed Abstract | Crossref Full Text | Google Scholar

Quast, C., Pruesse, E., Yilmaz, P., Gerken, J., Schweer, T., Yarza, P., et al. (2012). The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 41, D590–D596. doi: 10.1093/nar/gks1219,

PubMed Abstract | Crossref Full Text | Google Scholar

Redline, R. W. (2012). Inflammatory response in acute chorioamnionitis. Semin. Fetal Neonatal Med. 17, 20–25. doi: 10.1016/j.siny.2011.08.003,

PubMed Abstract | Crossref Full Text | Google Scholar

Reyman, M., van Houten, M. A., Watson, R. L., Chu, M., Arp, K., de Waal, W. J., et al. (2022). Effects of early-life antibiotics on the developing infant gut microbiome and resistome: a randomized trial. Nat. Commun. 13:893. doi: 10.1038/s41467-022-28525-z

Crossref Full Text | Google Scholar

Saturio, S., Nogacka, A. M., Alvarado-Jasso, G. M., Salazar, N., de Los Reyes-Gavilán, C. G., Gueimonde, M., et al. (2021). Role of bifidobacteria on infant health. Microorganisms 9:2415. doi: 10.3390/microorganisms9122415,

PubMed Abstract | Crossref Full Text | Google Scholar

Shane, A. L., Sánchez, P. J., and Stoll, B. J. (2017). Neonatal sepsis. Lancet 390, 1770–1780. doi: 10.1016/s0140-6736(17)31002-4,

PubMed Abstract | Crossref Full Text | Google Scholar

Stoll, B. J., Hansen, N. I., Bell, E. F., Walsh, M. C., Carlo, W. A., Shankaran, S., et al. (2015). Trends in care practices, morbidity, and mortality of extremely preterm neonates, 1993-2012. JAMA 314, 1039–1051. doi: 10.1001/jama.2015.10244,

PubMed Abstract | Crossref Full Text | Google Scholar

Stoll, B. J., Hansen, N. I., Sánchez, P. J., Faix, R. G., Poindexter, B. B., Van Meurs, K. P., et al. (2011). Early onset neonatal sepsis: the burden of group B streptococcal and E. coli disease continues. Pediatrics 127, 817–826. doi: 10.1542/peds.2010-2217,

PubMed Abstract | Crossref Full Text | Google Scholar

Tanaka, S., Kobayashi, T., Songjinda, P., Tateyama, A., Tsubouchi, M., Kiyohara, C., et al. (2009). Influence of antibiotic exposure in the early postnatal period on the development of intestinal microbiota. FEMS Immunol. Med. Microbiol. 56, 80–87. doi: 10.1111/j.1574-695X.2009.00553.x

Crossref Full Text | Google Scholar

Turroni, F., Milani, C., Ventura, M., and van Sinderen, D. (2022). The human gut microbiota during the initial stages of life: insights from bifidobacteria. Curr. Opin. Biotechnol. 73, 81–87. doi: 10.1016/j.copbio.2021.07.012,

PubMed Abstract | Crossref Full Text | Google Scholar

Underwood, M. A., German, J. B., Lebrilla, C. B., and Mills, D. A. (2015). Bifidobacterium longum subspecies infantis: champion colonizer of the infant gut. Pediatr. Res. 77, 229–235. doi: 10.1038/pr.2014.156,

PubMed Abstract | Crossref Full Text | Google Scholar

Uzan-Yulzari, A., Turta, O., Belogolovski, A., Ziv, O., Kunz, C., Perschbacher, S., et al. (2021). Neonatal antibiotic exposure impairs child growth during the first six years of life by perturbing intestinal microbial colonization. Nat. Commun. 12:443. doi: 10.1038/s41467-020-20495-4,

PubMed Abstract | Crossref Full Text | Google Scholar

van den Berg, J. P., Westerbeek, E. A., Bröring-Starre, T., Garssen, J., and van Elburg, R. M. (2016). Neurodevelopment of preterm infants at 24 months after neonatal supplementation of a prebiotic mix: a randomized trial. J. Pediatr. Gastroenterol. Nutr. 63, 270–276. doi: 10.1097/mpg.0000000000001148,

PubMed Abstract | Crossref Full Text | Google Scholar

Vangay, P., Ward, T., Gerber, J. S., and Knights, D. (2015). Antibiotics, pediatric dysbiosis, and disease. Cell Host Microbe 17, 553–564. doi: 10.1016/j.chom.2015.04.006,

PubMed Abstract | Crossref Full Text | Google Scholar

Weston, E. J., Pondo, T., Lewis, M. M., Martell-Cleary, P., Morin, C., Jewell, B., et al. (2011). The burden of invasive early-onset neonatal sepsis in the United States, 2005–2008. Pediatr. Infect. Dis. J. 30, 937–941. doi: 10.1097/INF.0b013e318223bad2,

PubMed Abstract | Crossref Full Text | Google Scholar

World Health Organization (1992). International statistical classification of diseases and related health problems, 10th revision (ICD-10). Geneva: World Health Organization.

Google Scholar

Yassour, M., Vatanen, T., Siljander, H., Hämäläinen, A. M., Härkönen, T., Ryhänen, S. J., et al. (2016). Natural history of the infant gut microbiome and impact of antibiotic treatment on bacterial strain diversity and stability. Sci. Transl. Med. 8:343ra81. doi: 10.1126/scitranslmed.aad0917,

PubMed Abstract | Crossref Full Text | Google Scholar

Zwittink, R. D., Renes, I. B., van Lingen, R. A., van Zoeren-Grobben, D., Konstanti, P., Norbruis, O. F., et al. (2018). Association between duration of intravenous antibiotic administration and early-life microbiota development in late-preterm infants. Eur. J. Clin. Microbiol. Infect. Dis. 37, 475–483. doi: 10.1007/s10096-018-3193-y,

PubMed Abstract | Crossref Full Text | Google Scholar