Abstract

Background:

Hyperbilirubinemia is among the most common conditions in neonates, and phototherapy is currently the most widely used treatment. However, it can induce side effects such as skin rashes, diarrhea, and gut microbiota dysbiosis, particularly affecting Bifidobacterium levels. This study aimed to investigate whether the supplementation of Bifidobacterium can alleviate dysbiosis and improve clinical outcomes in jaundiced neonates.

Methods:

A total of 79 jaundiced neonates were enrolled and divided into four groups: Phototherapy Control, M-16V, Bb-12, and the combined M-16V+Bb-12 group. Probiotics were administered until 30 days post-discharge, and neurodevelopment was assessed at 1.5–2 years using the Griffith Development Scales. Fecal samples collected before, during, and after treatment were analyzed using metagenomic sequencing and non-targeted metabolomics.

Results:

Probiotic supplementation significantly increased daily defecation frequency, accelerated the reduction rate of transcutaneous bilirubin, and shortened hospital stays. Griffith scores indicated that Bb-12 supplementation improved scores in personal-social and performance domains. Metagenomic analysis revealed significant differences in beta diversity between the control and probiotic groups; specifically, M-16V and combined supplementation increased the abundance of Bifidobacterium breve. Pathway enrichment analysis showed up-regulation of pyrimidine-containing compound metabolic processes, intramolecular transferase activity, and DNA conformation change. Metabolomics further demonstrated that combined supplementation elevated levels of 5-methyltetrahydrofolate (linked to DNA synthesis), benzoic acid and indoleacetic acid (linked to growth and development), and the anti-inflammatory metabolite indole-3-lactic acid.

Discussion:

For neonates receiving phototherapy, the addition of M-16 V + Bb-12 probiotics can improve the diversity of microflora, reduce the fixed value of harmful bacteria in the intestine, and enhance the excretion of bilirubin from the intestine, to improve the inflammatory damage and microbiota disorder caused by phototherapy, and achieve the effect of clinically improving jaundice, reducing bilirubin, shortening the length of hospitalization, and promoting neurodevelopment. It provides a safer and more effective treatment for neonatal jaundice.

1 Introduction

Neonatal hyperbilirubinemia is one of the most common diseases in newborns, classified into pathological types (Kemper et al., 2022; Erdeve et al., 2018). With an incidence rate as high as 80–90%, it is the most common cause of readmission in the neonatal period (Olusanya et al., 2018a). If not treated in time, the condition may progress to acute bilirubin encephalopathy and kernicterus, affecting the long-term prognosis of the child. This progression is accompanied by a significant risk of neonatal death and long-term neurodevelopmental disorders (Olusanya et al., 2018b; Liu H.W. et al., 2023). Currently, phototherapy is the most commonly used effective and safe method to reduce serum bilirubin levels and prevent the occurrence of severe hyperbilirubinemia and bilirubin encephalopathy (Yang and Wang, 2023; Neonatology Group of Pediatrics Branch of Chinese Medical Association, Editor of Chinese Journal of Pediatrics commission, 2014). However, phototherapy has certain side effects, such as rash, fever, and water and electrolyte balance (Faulhaber et al., 2019), and these side effects during phototherapy may be related to flora disorders (Feng and Tong, 2020; Tao and Jiang, 2021).

Winston et al. (2021) demonstrated that gut microbiota plays a highly efficient role in metabolizing bilirubin. A stable abundance of intestinal flora is beneficial for the treatment of neonatal jaundice (Zhang et al., 2021). The critical role of the gut microbiota in the development and persistence of intestinal inflammation underscores the importance of microbiota–host interactions in health and disease. Experimental animal studies and clinical data have confirmed the influence of the gut microbiome in ameliorating inflammation, highlighting its potential as a therapeutic strategy for treating inflammatory diseases (Lo et al., 2020). The occurrence of jaundice, as well as the corresponding drugs and treatment plans, may disrupt the intestinal homeostasis, leading to microbiota disorder, and thus affecting health (Fan et al., 2022). Experimental evidence also supports the fact that dysbiosis has been implicated in the etiopathogenesis of inflammatory bowel disease (IBD) (Roy and Dhaneshwar, 2023). Probiotics have gained attention for mitigating dysbiosis in IBD patients (Jadhav et al., 2023). Adding probiotics has been widely used to improve the balance of intestinal flora and regulate intestinal function (Mutlu et al., 2020; Chen et al., 2017). Our preliminary clinical studies have found that the side effects of phototherapy may be related to a flora disorder (Zhang et al., 2022). Among them, the abundance of two probiotics that can be used for newborns decreased significantly after phototherapy: Bifidobacterium breve and Bifidobacterium animalis (Yang et al., 2019).

However, in these studies, only a single probiotic was usually used. We aimed to investigate whether a combination of probiotics is superior to a single application, whether it reduces the microbiota disturbance caused by phototherapy, whether it reduces the inflammatory response, and whether it is beneficial for the clinical resolution of jaundice in these neonates. Therefore, we hypothesize that the supplementation of these two probiotics will significantly reduce microbiota disturbance and reduce intestinal inflammatory responses during phototherapy. In this study, we set up a combined probiotic with a 30-day follow-up period involving extensive tool samples. Additionally, follow-up using the Griffith Mental Development Scales was performed at 1.5–2 years of age to supplement existing data.

2 Methods2.1 Study design

This was a single-center, single-blind randomized controlled trial. The study was conducted by recruiting neonates hospitalized in the NICU of the Lingang Campus of the Sixth People’s Hospital Affiliated to Shanghai Jiao Tong University from 1 September 2020 to 1 October 2023.

Randomization was achieved using a random number table. The random system parameters were set as follows: (1) Random system: Taimei Medical eBalance random system; (2) Random method: block randomization; (3) Block length: 5; (4) Number of blocks: 20; (5) Random number: P001-P100. The operation steps were as follows: The random-related parameters were determined according to the project’s scheme. The eBalance stochastic system is configured according to these parameters. The eBalance random system generated a random grouping table based on the configuration [calling Statistical Analysis System (SAS) software] and exported the random grouping table.

The study was designed to be single-blind. None of the neonatologists were aware of the group assignments throughout the study, while the parents were aware of whether their infant was assigned to the experimental or control group.

The study was registered at http://www.chictr.org.cn/index.aspx (registration number: ChiCTR2000036013) before the first participant was enrolled.

This study was reviewed and approved by the Ethics Committee of the East Hospital of Shanghai Sixth People’s Hospital in compliance with the Declaration of Helsinki and other relevant regulations (Ethics Approval Number 2020-071). All enrolled neonatal parents provided informed consent.

2.2 Participants

The inclusion criteria were: (1) Jaundice index: According to the 2022 American Academy of Pediatrics phototherapy guidelines, the jaundice index reached the phototherapy threshold; (2) age ≤2 weeks; (3) Term infants with gestational age ≥37 weeks and <42 weeks, and birth weight ≥2,500 g and <4,000 g; (4) No prior use of antibiotics or ecological agents before specimen collection; (5) Healthy mothers during pregnancy, with no history of special drug use, and no intake of antibiotics or microecological agents before, during, or after childbirth; (6) Enrolled infants were exclusively breastfed, exclusively formula-fed, or mixed-fed before admission; (7) All enrolled infants had neonatal pathological jaundice as defined by “Practical Neonatology” and required hospital admission solely for phototherapy; (8) Informed consent provided voluntarily.

The Exclusion criteria included: (1) Gestational age <37 weeks or ≥42 weeks; (2) bilirubin levels reaching the exchange blood transfusion standard or elevated direct bilirubin; (3) complications with pneumonia, septicemia, or other diseases; (4) patients with severe immunodeficiency diseases; (5) those with inherited metabolic diseases; (6) congenital biliary malformations or other organ malformations; (7) drug allergies; (8) situations that may warrant exclusion as determined by the researcher, such as a guardian with mental illness or frequent changes in living or working environments, which may result in loss of follow-up.

Stool collection was still required after each phototherapy session.

According to the criteria above, a total of 95 neonates with jaundice were screened, 85 met the inclusion criteria, 6 were excluded due to antibiotic use or refusal to follow up, and 79 were finally included in the study.

2.3 Intervention

Figure 1 illustrates the details of the research procedure. Once the newborn was confirmed to meet the phototherapy standards for their age, they were registered. Parents continued to receive study information for 30 days, including during their regular follow-up visits to the hyperbilirubinemia clinic.

Study design process.

All the probiotics used in this study were from Dipro, including B. breve M-16 V (Dipro M-16 V) and B. animalis subsp. lactis Bb-12 (Dipro Bb-12). The probiotic dose is 109 colony-forming units administered orally once daily for 1 month.

The enrolled neonates were divided into 4 groups according to a digital randomization: the control group (pure phototherapy group), the M-16 V group (B. breve M-16 V combined with phototherapy), the Bb-12 group (B. animalis subsp. lactis BB-12 combined with phototherapy), and the M-16 V + Bb-12 group (M-16 V combined with Bb-12 combined with phototherapy).

The arrangements for enrollment during hospitalization and outpatient follow-up after discharge are shown in Table 1. At the beginning of the study, the investigator collected stool and blood samples from the subjects. Subjects received the assigned probiotics before the first 12 h of phototherapy. After 12 h of phototherapy, researchers recorded information, collected stool samples, and measured transcutaneous bilirubin (TCB). The next day, after 6–8 h of rest, the same process was repeated on the second and third day, and if necessary, subjects continued with the fourth and fifth 12-h phototherapy sessions. Subjects were not required undergo further phototherapy and were discharged once they met the discharge criteria. Follow-up studies on days 10, 20, and 30 were completed, during which stool and related data were collected. Infants in the experimental group continued to receive their assigned oral probiotics for 1 month. When these children were 1.5–2 years old, the Griffiths test was performed, and the results were recorded.

Time windowEnrollment D0Phototherapy for 12 h D1Rest for 6–8 h after phototherapy D1Phototherapy for 12 h D2Rest for 6–8 h after phototherapy D2Phototherapy for 12 h D3Rest for 6–8 h after phototherapy D3On the day of discharged D4Post-discharge D10, D20, D301.5–2
years-oldInformed consentXInclusion and exclusion criteriaXXXXXXXXXFeeding situationXXXXXXXXXGive light therapy interventionXXXRandom given to probioticsXXXXXcheck-upXXXXXXXXXVital signXXXXXXXXXWeight/urine outputXXXXXXXXSerum examination 1XSystemic assessmentXXXXXXXXXPercutaneous bilirubin determinationXXXXXXXXXStool assessment and systemic assessmentXXXXXGrififiths testX2.4 Sample collection

Stool samples weighing 500 mg and blood samples measuring 0.5 mL were collected from each subject. The stool samples were stored at −80 °C after freezing.

2.5 Sample size

A margin of 0.05 was assumed, with a type I error of 0.05 and a power of 0.8. Using a difference test for two-sample ratios, a sample size of 20 cases per group was calculated, yielding a total of 80 cases in both the experimental and control groups.

2.6 Metagenomic profiling2.6.1 DNA extraction and sequencing

The qualified DNA samples were randomly broken into fragments of about 350 bp in length by ultrasonication, and the entire library preparation was completed through end repair, 3′ end A addition, sequencing adapter addition, purification, fragment selection, polymerase chain reaction (PCR) amplification, and other steps. After library construction was completed, the effective concentration of the library was quantified using the quantitative PCR (qPCR) method (library effective concentration > 3 nM) to ensure its quality for subsequent sequencing. Metagenomic DNA was sequenced using a 2 × 150 bp paired-end protocol on Illumina HiSeq.

2.6.2 Sequencing data quality control

Trimmomatic (version 0.39) was used to remove low-quality sequences. Quality control of sequencing reads was conducted to remove low-quality reads and trim low-quality bases. KneadData was used to remove the contamination sequence from human DNA. Before and after removal, FastQC is used to assess sequence quality.

2.6.3 Taxonomy annotation

Host-filtered microbial reads were classified against bacterial, viral, fungal, archaeal, and human genomes using Kraken2 on a reference database constructed from the National Center for Biotechnology Information (NCBI) nucleotide and Reference Sequence (RefSeq) databases. The classification report was then used by Bracken to estimate species abundance, yielding the number of reads per species in the sample.

2.6.4 Functional annotation

Functional analysis was performed using HUMAnN2, based on the UniRef90 database, and annotated with the Kyoto Encyclopedia of Genes and Genomes (KEGG) database to generate KEGG Orthology (KO) and pathway-level profiles for each sample.

2.7 Metabolites profiling2.7.1 Sample preparation

Feces (100 mg) were individually ground in liquid nitrogen, and the homogenate was resuspended in prechilled 80% methanol by vortexing. The samples were incubated on ice for 5 min and then centrifuged at 15,000 g, 4 °C for 20 min. Some of the supernatant was diluted to a final concentration of 53% methanol with liquid chromatography–mass spectrometry (LC–MS) grade water. The samples were subsequently transferred to a fresh Eppendorf tube and then centrifuged at 15,000 g, 4 °C for 20 min. Finally, the supernatant was injected into the liquid chromatography–tandem mass spectrometry (LC–MS/MS) system for analysis.

2.7.2 LC–MS analyses

Ultra-high performance liquid chromatography–tandem mass spectrometry (UHPLC–MS/MS) analyses were performed using a Vanquish UHPLC system (Thermo Fisher, Germany) coupled with an Orbitrap Q Exactive™ HF mass spectrometer or Orbitrap Q Exactive™ HF-X mass spectrometer (Thermo Fisher, Germany). Samples were injected onto a Hypersil Gold column (100 × 2.1 mm, 1.9 μm) using a 17-min linear gradient at a flow rate of 0.2 mL/min. The eluents for the positive polarity mode were eluent A (0.1% formic acid [FA] in water) and eluent B (methanol). The eluents for the negative-polarity mode were eluent A (5 mM ammonium acetate, pH 9.0) and eluent B (methanol). The solvent gradient was set as follows: 2% B, 1.5 min; 2–85% B, 3 min; 85–100% B, 10 min; 100–2% B, 10.1 min; and 2% B, 12 min. The Q Exactive™ HF mass spectrometer was operated in positive/negative polarity mode with a spray voltage of 3.5 kV, capillary temperature of 320 °C, sheath gas flow rate of 35 psi, and aux gas flow rate of 10 L/min, S-lens Radio Frequency (RF) level of 60, and Aux gas heater temperature of 350 °C.

2.7.3 Data processing and metabolite identification

The raw data files generated by UHPLC–MS/MS were processed using the Compound Discoverer 3.3 (CD3.3, Thermo Fisher, Germany) to perform peak alignment, peak picking, and quantitation for each metabolite. The main parameters were set as follows: peak area was corrected with the first quality control (QC); actual mass tolerance, 5 ppm; signal intensity tolerance, 30%; and minimum intensity, etc. Then, peaks were matched against mzCloud,1 mzVault, and MassList database to obtain the accurate qualitative and relative quantitative results, and metabolites were annotated using the KEGG database.2 Peak intensities were normalized to the median value per sample, and then log-transformed and normalized with the mean and standard deviation per metabolite.

Normalized data was used for further analysis, such as Principal Component Analysis (PCA, differential analysis), and correlation analysis with species abundance. Statistical analyses were performed using the statistical software R (R version R-4.3.1).

2.8 In vitro experiment2.8.1 Anti-inflammatory test

Human adult low Calcium high Temperature cells (HaCaT) (1 × 105/well) were plated on 96-well plates and cultured until adhered. The culture medium was discarded, 100 μL of Dulbecco’s Modified Eagle Medium (DMEM) solution was added to the negative wells (control), and 100 μL of DMEM solution containing 1 × 107 colony-forming unit (CFU) probiotics was added to the sample wells. The plates were incubated for 18 h. The supernatant from the above culture was collected, and interleukin 10 (IL-10) concentration was measured using a kit.

2.8.2 Cell growth test on blue light

After HaCaT cells were seeded into 96- and 12-well plates and cultured for 24 h, the HaCaT cell medium was changed, and Bb-12, M-16 V, and Bb-12 + M-16 V [multiplicity of infection (MOI) = 1:50] were added for treatment. The cells were then exposed to blue light (wavelength 450–460 nm) for 30 m. The control group (control) was treated in the dark under the same conditions. The cells were then cultured in a 37 °C carbon dioxide incubator for 18 h. After the culture, (1) the cells were washed 3 times with phosphate-buffered saline (PBS), and the cell activity was determined by the Cell Counting Kit-8 (CCK-8) method; (2) the cells were collected, and the intracellular malondialdehyde (MDA) level was detected; (3) the supernatant after culture was collected and the tumor necrosis factor-α (TNF-α) level was detected using an ELISA kit.

2.9 Bioinformatic analysis and statistics2.9.1 Microbiota diversity analysis

The alpha diversity between groups was analyzed with Student’s t-test. The beta diversity was calculated using Bray–Curtis dissimilarity. Permutational multivariate analysis of variance (PERMANOVA) was performed using the “adonis” function in the R Vegan package to assess the effects of phenotype on taxonomic and metabolomic profiles.

3 Results3.1 General information

A total of 79 patients were enrolled in the study, and 382 stool samples were collected at multiple time points. There were 20 patients in the control group, 19 in the M-16 V group, 19 in the Bb-12 group, and 21 in the M-16 V + Bb-12 group. No significant differences were found among groups with respect to confounding factors such as delivery mode, gender, age, mothers’ age, gestational age, birth weight, feeding pattern, and percutaneous bilirubin before treatment (Table 2).

ParametersControlM-16 VBb-12M-16 V + Bb-12p-valuePatient number20191921–Delivery mode (natural: C-section)12:814:511:814:70.83Gender (males: females)10:1013:612:715:60.14Age (days)4.55 2.825.26 2.564.37 2.395.95 3.200.28Mother age (yeas)32.004.2331.90 2.9629.50 5.5931.24.390.38Gestational age (days)270 9.10275 6.59276 7.93270 8.140.12Birth weight (g)3,214 4353,424 4983,177 3043,306 3790.38Feeding (mix: breastfeeding: formula)6:5:812:3:45:10:48:9:40.08Pre-treatment skin bilirubin (mg/dL)15.00 2.7914.70 2.1913.50 2.2014.60 2.900.45

Confound factors of different groups.

3.2 Clinical outcomes

To compare the efficacy of probiotics for jaundice with phototherapy, we measured the crucial clinical outcomes (Table 3). There was no significant difference in weight change among the groups, whereas probiotics-supplemented groups exhibited higher fecal times per day in treatment and lower skin bilirubin levels after treatment. Specifically, the daily frequency of stool was 3.35 ± 1.18 in the control group, 4.16 ± 1.30 in the M-16 V group, 4.84 ± 1.01 in the Bb-12 group, and 4.33 ± 1.20 in the M-16 V + Bb-12 group (p = 0.002, p < 0.05). The average length of stay was 4.7 ± 1.81 days for the control group, 4.0 ± 1.05 days for M-16 V, 4.06 ± 1.86 days for Bb-12, and 3.6 ± 0.75 days for the M-16 V + Bb-12 group. Compared with the control group, the length of hospital stay in the M-16 V + Bb-12 group was reduced (p < 0.05). Besides, the significant differences in the percentage of skin bilirubin reduction and hospital stay were found between the control and M-16 V + Bb-12 groups (Figures 2A,B), indicating that combined supplementation with M-16 V and Bb-12 may enhance the efficacy of phototherapy and therefore decrease hospitalization time for the newborns. We also measured bilirubin levels in fecal samples at multiple time points. An increased level of bilirubin was observed in the Bb-12 and M-16 V groups during the phototherapy process. At the visiting stage, the bilirubin level in the fecal sample was significantly higher in the M-16 V + Bb-12 combined group (Figure 2C).

ParametersControlM-16 VBb-12M-16 V + Bb-12p-valuePatient number20191921–Weight change (%)−1 51 6.92 61 50.59Fecal times per day3.35 1.184.16 1.304.84 1.014.33 1.200.002**Post-treatment skin bilirubin (mg/dL)6.09 2.094.26 1.505.21 2.314.35 2.080.018*Skin bilirubin loss ratio (%)57.50 18.7069.90 13.1061.30 16.3068.80 15.700.051Days in hospital4.71.814.001.054.06 1.863.60 0.750.088

Clinical outcomes of different groups.

Clinical evaluation of a probiotic supplement on neonatal jaundice with phototherapy. (A) Comparison of hospital stay; (B) comparison of the reduction rate of skin bilirubin; (C) comparison of fecal bilirubin. *p < 0.05.

3.3 The Griffith scale indicated that probiotics have long-term effects on the growth and development of neonates with jaundice after phototherapy

To test the long-term influence of probiotics on neonatal jaundice treated with phototherapy, we followed up with the patients and administered the Griffith test when the children were 1.5–2 years old. The results showed that the Bb-12 group showed better behavior in language and performance, which was significantly higher than the M-16 V group (Figure 3).

The long-term effects of probiotics on neonates with jaundice after phototherapy were assessed using the Griffith scale. (A) Allergy score; (B) athletic score; (C) social score; (D) language skills score; (E) coordination score; (F) performance score. *p < 0.05; **p < 0.01; ***p < 0.001.

3.4 Probiotics exhibited considerable anti-inflammatory properties

To further validate our findings, we evaluated the anti-inflammatory potential of M-16 V and Bb-12 by measuring IL-10 levels in cell culture. Elevated IL-10 levels were observed following probiotic treatment (Figure 4A). Furthermore, to assess the impact of phototherapy on cell viability, we measured cell activity across different groups. Blue light exposure significantly inhibited cell activity (Figure 4B). MDA levels—which reflect the degree of free radical accumulation and lipid peroxidation—were significantly reduced in the M-16 V and M-16 V + Bb-12 groups (Figure 4C). Besides, levels of pro-inflammatory cytokine TNF-α were significantly increased in the group exposed to blue light alone; probiotic supplementation significantly attenuated this increase (Figure 4D).

In vitro cell culture studies were conducted to investigate the anti-inflammatory potential of M-16 V and Bb-12. (A) IL-10 levels; (B) Cell activity; (C) MDA levels; (D) TNF-α levels. The experimental groups were defined as follows: Normal control (NC) group, control (phototherapy group), M-16 V (phototherapy+M-16 V group), Bb-12 (phototherapy+Bb-12 group), M-16 V + Bb-12 (phototherapy + combination group). *p < 0.05; **p < 0.01; ***p < 0.001.

3.5 Probiotics reshaped the gut microbiota in neonates with jaundice after phototherapy

For beta diversity, we observed significant differences in the gut microbiota structure between the M-16 V group and the M-16 V + Bb-12 group during treatment (Figure 5A). The ratio of Bifidobacterium to Escherichia is an important index for infant gut health. In our results, compared with the control group and the Bb-12 supplemented group, the M-16 V group and the M-16 V + Bb-12 groups behaved better in this aspect (Figure 5B). It was also observed that the M-16 V supplement does have a better colony outcome than Bb-12 (Figure 5D). The abundance of B. breve in the M-16 V group was higher than that in the Bb-12 group. However, the abundance of B. animalis in the Bb-12 group was higher than that in the M-16 V group (Figure 5C). Furthermore, we measured the uniformity of gut microbiota (defined as the 80% quantile of 20% quantile of species abundance per sample) across different groups, and found increased uniformity in post-treatment for the control group (Figure 5E).

Probiotics altered the beta diversity and the community structure of the gut microbiota. (A) Beta diversity of different groups after the phototherapy treatment; (