1 Department of Pediatrics, Shanghai Sixth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, 200233 Shanghai, China
2 College of Food Science and Technology, Shanghai Ocean University, 201306 Shanghai, China
Abstract
Probiotics have increasingly progressed from laboratory research to clinical application, supported by growing evidence demonstrating their potential benefits across multiple neonatal conditions. In 2023, the Chinese Preventive Medicine Association released an evidence-based guideline on pediatric probiotic use; however, comprehensive summaries dedicated specifically to neonates remain scarce despite rapidly expanding clinical utilization in this population. This narrative review synthesizes current evidence on the role of probiotics in common neonatal disorders and discusses key considerations regarding timing, dosing, and safety. Literature from PubMed and CNKI was examined, prioritizing clinical trials and meta-analyses published in the past five years, with extended inclusion for topics lacking sufficient data. Searches covered probiotics in relation to necrotizing enterocolitis, feeding intolerance, antibiotic-associated diarrhea, neonatal hyperbilirubinemia, sepsis, bronchopulmonary dysplasia, respiratory infection, micronutrient metabolism, safety, and mechanisms. A total of 22 reviews/meta-analyses and 51 clinical studies were included. Overall, Bifidobacterium and Lactobacillus are the most frequently used probiotics in neonatal care, with evidence supporting reductions in necrotizing enterocolitis, feeding intolerance, antibiotic-associated diarrhea, hyperbilirubinemia, and potentially sepsis, alongside possible respiratory and micronutrient-related benefits. Although safety profiles are generally favorable, rare instances of probiotic-related sepsis highlight the need for caution. Considerable heterogeneity in strains, dosing strategies, and intervention durations continues to limit interpretation. Further large-scale, rigorously designed randomized trials are required to refine strain selection, validate efficacy, and ensure safety in this highly vulnerable population.
Keywords
- probiotic
- neonatal diseases
- probiotic safety
- necrotizing enterocolitis
- antibiotic-associated diarrhea
- late-onset sepsis
- micronutrient status
According to a 2001 definition by the Food and Agriculture Organization (FAO) of the United Nations and the World Health Organization (WHO) defined probiotics as “live microorganisms that, when administered in adequate amounts, confer a health benefit on the host”. In China, probiotics are classified as either pharmaceutical-grade or food-grade. The List of Strains Permitted for Use in Foods 2022 issued by China’s National Health Commission specifies 38 food-grade strains across 17 genera. For infant food, the list is more restrictive, allowing only 15 strains to be used in this population (see Table 1). Their clinical use in neonates is garnering increased attention due to the high burden of gastrointestinal and infectious diseases in this population. Neonatal disorders such as necrotizing enterocolitis (NEC), feeding intolerance, sepsis, bronchopulmonary dysplasia (BPD), hyperbilirubinemia, and antibiotic-associated diarrhea are major contributors to neonatal morbidity and mortality worldwide, particularly among preterm infants. Conventional management strategies, including broad-spectrum antibiotics and phototherapy, can disrupt the developing gut microbiome, underscoring the urgent need for safe and effective adjunctive interventions.
| Number | Species | Strains | Official designation |
| 1 | Bifidobacterium animalis subsp. lactis | Bb-12 | Bifidobacterium animalis subsp. lactis Bb-12 |
| 2 | HN019 | Bifidobacterium animalis subsp. lactis HN019 | |
| 3 | Bi-07 | Bifidobacterium animalis subsp. lactis Bi-07 | |
| 4 | Lacticaseibacillus rhamnosus | GG | Lacticaseibacillus rhamnosus GG |
| 5 | HN001 | Lacticaseibacillus rhamnosus HN001 | |
| 6 | MP108 | Lacticaseibacillus rhamnosus MP108 | |
| 7 | Lactobacillus helveticus | R0052 | Lactobacillus helveticus R0052 |
| 8 | Limosilactobacillus reuteri | DSM17938 | Limosilactobacillus reuteri DSM17938 |
| 9 | Bifidobacterium bifidum | R0071 | Bifidobacterium bifidum R0071 |
| 10 | Lactobacillus acidophilus | NCFM | Lactobacillus acidophilus NCFM |
| 11 | Limosilactobacillus fermentum | CECT5716 | Limosilactobacillus fermentum CECT5716 |
| 12 | Bifidobacterium breve | M-16V | Bifidobacterium breve M-16V |
| 13 | Bifidobacterium longum subsp. longum | BB536 | Bifidobacterium longum subsp. longum BB536 |
| 14 | Bifidobacterium longum subsp. infantis | R0033 | Bifidobacterium longum subsp. infantis R0033 |
| 15 | M-63 | Bifidobacterium longum subsp. infantis M-63 |
The use of probiotics in neonatal intensive care units (NICUs) is becoming more common worldwide. Various randomized controlled trials (RCTs) and meta-analyses report positive effects of probiotics in preventing NEC and improving enteral feeding outcomes, such as better intestinal function. In January 2023, the Chinese Preventive Medicine Association released the Evidence-Based Guideline for Probiotic Application in Pediatrics. However, few European or American guidelines strongly endorse probiotic use due to the lack of unified criteria and consensus, as well as significant heterogeneity in strain and dosage. Above all, among all probiotics utilized in neonates so far, bifidobacteria and lactobacilli have been extensively studied and applied, owing to their roles in shaping gut microbiota and immune system establishment.
Mechanistically, probiotics help maintain intestinal homeostasis by modulating
microbial diversity, strengthening the gut barrier, and producing short-chain
fatty acids (SCFAs) that serve as critical energy substrates for colonocytes and
regulators of immune signaling. Emerging evidence also supports their role in the
gut-lung and gut-brain axes, linking probiotic interventions to systemic effects
on respiratory health, neurodevelopment, and micronutrient metabolism. At the
immunological level, specific strains such as Lacticaseibacillus
rhamnosus GG and Bifidobacterium longum 35624 have been shown
to promote regulatory T cell differentiation and increase anti-inflammatory
cytokines such as IL-10, and suppress pro-inflammatory mediators like
TNF-
Despite encouraging findings, several issues remain controversial. Although generally regarded as safe, cases of probiotic-associated sepsis have been reported in extremely preterm infants. Uncertainty remains regarding the optimal probiotic strain, the appropriate dose, and the optimal timing and duration of intervention. Furthermore, many existing clinical studies are limited by small sample sizes or single-center designs. Thus, the present article reviews the current evidence on probiotic use for commonly encountered neonatal conditions, discusses the proposed mechanism of action, evaluates safety concerns, and highlights knowledge gaps that deserve further exploration. The main clinical studies included in this review, focusing exclusively on neonatal populations, and covering probiotic interventions across digestive, infectious, and allergic diseases, are summarized in Table 2 (Ref. [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51]).
| Author (Year) | Country | Design | Subjects | N | Probiotic (strain & dose) | Duration | Endpoints |
| Robertson et al. (2020) [1] | United Kingdom | Retrospective study | Preterm |
982 (469 before, 513 after probiotics) | Infloran: L. acidophilus + B. bifidum (1 × 109 CFU/day, 2013–2016); Labinic: L. acidophilus + B. bifidum + B. infantis (0.5 × 109 CFU/day, 2016–2017) | To |
NEC (Bell’s stage |
| Sowden et al. (2022) [2] | South Africa | RCT | Preterm |
200 (100 probiotic, 100 placebo) | Labinic: L. acidophilus + B. bifidum + B. infantis (2 × 109 CFU/day) | 28 days | NEC incidence/severity, feeding intolerance |
| Cripps et al. (2023) [3] | Australia | Retrospective study | Preterm |
805 (419 no probiotics, 386 probiotics) | ABC Dophilus: B. infantis + B. lactis + S. thermophilus (2009–2011); Infloran: L. acidophilus + B. bifidum (2 × 109 CFU/day, 2013–2020) | To |
NEC incidence (Bell’s stage |
| Korček and Straňák (2024) [4] | Czech Republic | Retrospective study | Preterm |
455 (228 multi-species, 227 single-species) | Multi-species: L. rhamnosus, L. casei, L. acidophilus, B. infantis, B. bifidum (2.5 × 108 CFU/day); Single-species: B. breve BR03 + B632 (1 × 108 CFU each/day) | From feeds |
NEC (Bell’s stage |
| Samara et al. (2022) [5] | Canada | RCT | Extremely preterm |
57 (26 probiotic, 31 control) | FloraBABY: B. breve HA-129, B. bifidum HA-132, B. longum subsp. infantis HA-116, B. longum subsp. longum HA-135, L. rhamnosus HA-111 (CFU not reported) | First week of life until 37–39 wks GA | Gut microbiome maturation, intestinal inflammation, stool metabolites, cytokines |
| Costeloe et al. (2016) [6] | United Kingdom | RCT | Very preterm 23–30 wks GA | 1315 (650 probiotic, 660 placebo) | B. breve BBG-001, 1.58 × 108 |
From |
NEC (Bell stage 2–3), late-onset sepsis ( |
| Sowden et al. (2022) [7] | South Africa | RCT | Preterm |
200 (100 probiotic, 100 placebo) | Labinic (L. acidophilus + B. bifidum + B. infantis), 2 × 109 CFU/day | From first feed for 28 days | NEC incidence/severity, feeding intolerance |
| Indrio et al. (2017) [8] | Italy | RCT | Preterm |
60 (30 probiotic, 30 placebo) | L. reuteri DSM 17938, 1 × 108 CFU/day | Within 48 h of birth for 30 days | Feeding intolerance, time to full feeds, gastric emptying, hospital stay |
| Indrio et al. (2008) [9] | Italy | RCT | Preterm |
20 (10 probiotic, 10 placebo) + 10 reference (breast-fed) | L. reuteri ATCC 55730, 1 × 108 CFU/day | 30 days | Regurgitation, crying time, stools, gastric emptying |
| Wu et al. (2012) [10] | China | Prospective Study | Neonates with purulent meningitis | 123 (54 probiotics, 69 control) | Triple viable tablets (Bifidobacterium, Lactobacillus, Enterococcus), 0.5 g tid, oral (CFU not reported) | Until clinical cure/stop antibiotics | Incidence of AAD, time from antibiotics to AAD onset |
| Liu et al. (2013) [11] | China | RCT | Neonates requiring |
120 (60 probiotics, 60 control) | Bacillus subtilis + Enterococcus faecium (“Mamiai”), 1 g tid, oral (CFU not reported) | 1–2 weeks | Incidence of AAD |
| Cao et al. (2021) [12] | China | RCT | Neonates with AAD | 80 (40 probiotics + antibiotics, 40 antibiotics only) | Clostridium butyricum + Bifidobacterium powder, dose not specified (CFU not reported) | 72 h | Clinical efficacy, recovery of bowel sounds, diarrhea duration, intestinal flora counts |
| Feng and Yang (2019) [13] | China | RCT | Neonates with pneumonia and AAD | 90 (45 probiotics, 45 control) | Obs: Clostridium butyricum + Bifidobacterium powder; Ctrl: Saccharomyces boulardii (CFU not reported) | Until recovery | Clinical efficacy, diarrhea duration, hospitalization days, fecal flora, IL-2/IL-6/TNF- |
| Du and Wang (2020) [14] | China | RCT | Neonates with AAD | 78 (39 probiotics, 39 control) | Obs: Clostridium butyricum + Bifidobacterium powder; Ctrl: Saccharomyces boulardii (CFU not reported) | 3 days | Clinical efficacy, intestinal flora, serum TNF- |
| Wan et al. (2017) [15] | China | RCT | Young children 1 |
408 (213 probiotic, 195 control) | Saccharomyces boulardii, 5 × 109 CFU/day | 14 days | Incidence of AAD |
| Zhang et al. (2024) [16] | China | RCT | Infants/young children |
182 (47 control, 70 S. boulardii, 65 Bifidobacterium mix) | S. Boulardii: 3.25 × 108 CFU/day; Tetragenous viable Bifidobacterium: Bifidobacterium, Lactobacillus acidophilus, Enterococcus faecalis (each |
7, 14, 21 days | Incidence of AAD, fecal cocci/bacilli ratio |
| Dose: 1–6 months: 1 tablet twice daily; 7–12 months: 1 tablet three times daily; 1–3 years: 2 tablets three times daily. | |||||||
| Lukasik et al. (2022) [17] | Dutch and Polish | RCT | Young children 3 month |
313 (158 probiotic, 155 control) | A multispecies probiotic (Bifidobacterium bifidum W23, Bifidobacterium lactis W51, Lactobacillus acidophilus W37, L acidophilus W55, Lacticaseibacillus paracasei W20, Lactiplantibacillus plantarum W62, Lacticaseibacillus rhamnosus W71, and Ligilactobacillus salivarius W24), 1010 CFU/day | For the duration of antibiotic treatment and for 7 days after | Incidence of AAD |
| Serce Pehlevan et al. (2020) [18] | Turkey | RCT | Preterm |
208 (104 probiotic, 104 control) | Synbiotic mix (L. rhamnosus 8.2 × 108, L. plantarum 4.1 × 108, L. casei 4.1 × 108, B. lactis 4.1 × 108 CFU + FOS 383 mg + GOS 100 mg + lactoferrin 2 mg + vitamins) | From first feed until discharge (median 36 day) | Incidence of NEC stage |
| Granger et al. (2022) [19] | United Kingdom | Retrospective study | Preterm |
1061 (509 pre vs 552 post probiotics) | Infloran (L. acidophilus, B. bifidum), later Labinic (L. acidophilus, B. bifidum, B. longum subsp. infantis) (CFU not reported) | From minimal feeds until 34 wks CGA or discharge | Incidence of NEC, LOS or mortality |
| Sharpe et al. (2018) [20] | Australia | Retrospective study | Preterm |
1791 (1334 pre vs 457 post introduction) | Infloran (L. acidophilus LA37 + B. bifidum BB07) (CFU not reported) | 42 days | Incidence of NEC, LOS or mortality |
| Meyer et al. (2020) [21] | New Zealand | Retrospective study | Preterm |
4529 (2556 pre vs 1973 post probiotics) | Infloran (L. acidophilus ATCC 4356 1 × 109 CFU + B. bifidum ATCC 15696 1 × 109 CFU/day) in 5 units; LGG + bLF (L. rhamnosus GG 6 × 109 CFU + bovine lactoferrin 100 mg daily) in 1 unit | Started with trophic feeds, continued 4–6 wks or until 34–36 wks CGA/discharge | Incidence of NEC, LOS or mortality |
| Panigrahi et al. (2017) [22] | India | RCT | Term/late-preterm |
4556 (2278 synbiotic vs 2278 placebo) | L. plantarum ATCC-202195 (1 × 109 CFU/day) | 7 days | Incidence of LOS or mortality |
| Güney-Varal et al. (2017) [23] | Turkey | RCT | Preterm |
110 (70 probiotic vs 40 control) | Multi-strain, multi-species (commercial NBL probiotic®): L. rhamnosus 4.1 × 108 CFU + L. casei 8.2 × 108 CFU + L. plantarum 4.1 × 108 CFU + B. animalis 4.1 × 108 CFU, with FOS 383 mg + GOS 100 mg | Started at 2–7 d of life; mean 36.5 |
Incidence of NEC, LOS or mortality |
| Yuan et al. (2025) [24] | China | RCT | Term neonates with jaundice (GA |
21 (11 probiotic vs 10 control) | Lactobacillus rhamnosus AB-GG, 1 × 109 CFU/day | 1 month | Bilirubin levels, phototherapy duration, hospital stay, gut microbiota diversity, metabolites |
| Tsai et al. (2022) [25] | Taiwan (China) | RCT | Newborns with neonatal jaundice (GA |
83(43probiotic, 40 control) | Bifidobacterium animalis subsp. lactis CP-9 (1 × 10¹⁰ CFU/day) | During in-hospital phototherapy until bilirubin decreased |
TSB decline rate (mg/dL/h); total phototherapy duration (h) |
| Wang (2023) [26] | China | RCT | Neonates with pathological jaundice | 72 (36 probiotic vs 36 control) | Probiotic (not specified, oral) + Yinzhihuang (CFU not reported) | 7 days | TSB, DB, IB, immune function (T-cell subsets), AFP, TRF, liver enzymes, adverse events |
| Nasief et al. (2024) [27] | Pakistan | Open-labelled RCT | Preterm neonates with indirect hyperbilirubinaemia | 76 (2 groups) | Saccharomyces boulardii, 125 mg/day (CFU not reported) | From initiation of phototherapy until hospital discharge | duration of phototherapy (h); length of hospital stay (h) |
| Tian and Guo (2016) [28] | China | RCT | Neonates with breast milk jaundice | 69 (35 probiotic vs 34 control) | Bacillus licheniformis (Zhengchangsheng), dose not specified (CFU not reported) | Until recovery | TSB, jaundice fading time, stool frequency |
| Mutlu et al. (2020) [29] | Turkey | RCT | Term neonates with isoimmune hemolytic jaundice (GA 35–42 wks) | 60 (30 probiotic vs 30 control) | Lactobacillus rhamnosus GG, 1 × 109 CFU/day | 4 days | Serum total bilirubin (STB), rebound STB, phototherapy duration, meconium frequency |
| Eghbalian et al. (2025) [30] | Iran | RCT | Term neonates (GA 37–42 wks, BW |
150 (75 probiotic vs 75 control) | PediLact drops (CFU not reported) | 72 h | Serum bilirubin levels, phototherapy duration, hospitalization length, need for transfusion |
| Demirel et al. (2013) [31] | Turkey | RCT | VLBW infants |
179 (81 probiotic vs 98 control) | Saccharomyces boulardii (CFU not reported) | From first feed until discharge | Duration of phototherapy, serum total bilirubin at end of phototherapy, feeding intolerance, sepsis incidence |
| Santosa et al. (2022) [32] | Japan | Prospective Study | Term neonates delivered by Caesarean section | 153 (54 probiotic vs 99 control) | Bifidobacterium animalis subsp. lactis BB-12, 3 × 109 CFU/day | 20 days | TCB levels (day 1–5), body weight gain |
| Serce et al. (2015) [33] | Turkey | RCT | Neonates 35–42 wks with hyperbilirubinemia requiring phototherapy | 119 (58 probiotic vs 61 control) | Saccharomyces boulardii (Reflor, Biocodex), 125 mg q12h during phototherapy (CFU not reported) | Until phototherapy stopped (up to 96 h) | Serum bilirubin (0, 24, 48, 72, 96 h), duration of phototherapy, rebound hyperbilirubinemia |
| Sun et al. (2024) [34] | China | RCT | Term neonates with pathological jaundice | 114 (38 control vs 38 S. boulardii vs 38 Clostridium butyricum + Bifidobacterium) | (i) S. boulardii granules (CFU not reported); (ii) Clostridium butyricum + Bifidobacterium (CFU not reported) | 5 days | TSB, DB, IB, phototherapy duration, jaundice fading time, stool frequency, hospital stay, adverse events |
| Khodair et al. (2025) [35] | Egypt | Prospective RCT (open-label; placebo not feasible) | Full-term neonates (GA 37–42 weeks) | 80 (40 probiotic, 40 control) | Bifidobacterium animalis subsp. lactis BB-12 (total 2 × 109 CFU/day) | From day 1 of recruitment until discharge | VAP incidence |
| Banupriya et al. (2015) [36] | India | RCT | Critically ill children |
150 (75 probiotic vs 75 control) | Multi-strain capsule (per capsule: L. acidophilus 7 × 108, B. longum 4 × 108, L. rhamnosus 4 × 108, L. plantarum 3 × 108, L. casei 3 × 108, L. bulgaricus 3 × 108, B. infantis 3 × 108, B. breve 3 × 108, S. thermophilus 3 × 108); total 6.6 × 109 CFU/day | 7 days or until ICU discharge | VAP incidence, duration of MV, ICU stay, hospital stay, mortality, pathogen colonization |
| Kukkonen et al. (2008) [37] | Finland | RCT | Allergy-prone term infants | 1018 (506 probiotic vs 512 control) | L. rhamnosus GG, L. rhamnosus LC705, B. breve Bb99, P. freudenreichii JS; 8–9 × 109 CFU/day | 6 months | Respiratory infections, antibiotic use |
| Aryayev et al. (2018) [38] | Ukraine | RCT | Late preterm newborns (35–36 wks) | 62 (30 probiotic vs 32 control) | E. coli Nissle 1917 (Mutaflor®), 1 × 108 CFU daily (day 1–7), then 3 × 108 CFU (day 8–21) | 3 weeks (follow-up to 12 months) | ARI incidence (28 d, 1 yr), ARI-related hospitalizations |
| Piloquet et al. (2024) [39] | France & Belgium | RCT | Healthy formula-fed term infants (HM ref group n = 80) | 460 (230 synbiotic vs 230 control) | Synbiotic formula: L. fermentum CECT 5716 (1–1.5 × 106 CFU/g) + GOS | 11 months | Infectious diarrhea (primary), URTI, LRTI |
| Takeshita et al. (2024) [40] | Japan | RCT | Preterm infants |
41 (21 probiotic vs 20 control) | Heat-killed p. acidilactici K15, 5 × 1010 CFU/day | 12 months | Febrile days, respiratory tract infections |
| Qu et al. (2021) [41] | China | Retrospective study | Preterm infants |
318 (94 probiotic vs 224 control) | Clostridium butyricum powder (CFU not reported) | From admission until |
BPD incidence, death |
| Li et al. (2024) [42] | China | RCT | Preterm infants (GA 28–32 wks, BW |
86 (43 probiotic vs 43 control) | Clostridium butyricum + Bifidobacterium (CFU not reported) | Until 36 wks corrected GA | PaCO2, PaO2, tidal volume, ventilation time, oxygen time, LOS, BPD incidence |
| Wu et al. (2011) [43] | China | RCT | Mechanically ventilated neonates | 81 (38 probiotic vs 43 control) | Bifidobacterium (Lizhu Pharma), 0.5 × 108 CFU | 7 days | Gastric pH, gastric colonization, feeding intolerance, VAP incidence, pathogen homology |
| Xie (2018) [44] | China | RCT | Mechanically ventilated neonates | 80 (40 probiotic vs 40 control) | Bifidobacterium (strain not specified, CFU not reported) | 7 days | VAP incidence, VAP onset time, gastric pH, gastric colonization, pathogen homology |
| Puisto et al. (2025) [45] | Finland/Spain | RCT | Infants born to atopic mothers, vaginal delivery, full-term, breastfed | 241 mothers randomized; microbial analysis subset: 46 infants (26 probiotic vs 20 placebo) | Maternal probiotics: (1) L. rhamnosus LPR + B. longum BL999; (2) L. paracasei ST11 + B. longum BL999; 1 × 109 CFU/day | 2 months before delivery to 2 months postpartum | Infant atopic eczema incidence, gut microbiota composition |
| Luo and Zhang (2022) [46] | China | Cohort study (intervention vs control) | Infants 4–6 months with digestive CMA manifestations | 85 CMA infants (45 probiotic vs 40 control) + 41 healthy controls | Bifidobacterium BB-12 + L. rhamnosus GG, 4.5 × 109 CFU/day | 3 months | Weight-for-age, GI symptoms (vomiting, diarrhea, food refusal, bowel sounds) |
| Nocerino et al. (2019) [47] | Italy | Prospective Study | Children with previous CMA, immune tolerance achieved |
330 (110 EHCF vs 110 EHCF + LGG vs 110 healthy controls) | L. rhamnosus GG, 2.5 × 107–5 × 108 CFU/g in EHCF + LGG (Nutramigen LGG®) | CMA diagnosis in infancy, follow-up to 4–6 years | Functional gastrointestinal disorders (FGIDs, Rome III criteria) |
| Agustina et al. (2013) [48] | Indonesia | RCT | Healthy children 1–6 yrs | 494 (244 probiotic vs 250 control) | L. casei CRL 431 or L. reuteri DSM 17,938, 5 × 108 CFU/day in milk | 6 months | Growth (weight, height, WAZ, HAZ), anemia, iron and zinc status |
| Surono et al. (2014) [49] | Indonesia | RCT | Preschool children 12–24 mo | 48 (36 intervention: 12 probiotic, 12 zinc, 12 combo vs 12 placebo) | L. plantarum IS-10506 (dadih origin), 1 × 1010 CFU/day; zinc 20 mg/day; combo probiotic + zinc | 90 days | Fecal sIgA, serum zinc |
| Ballini et al. (2019) [50] | Italy/Albania | RCT | Healthy children 14–18 yrs | 40 (20 probiotic vs 20 control) | Multi-strain (L. plantarum, L. acidophilus, B. infantis, B. lactis) + FOS; 3 × 109 CFU/pearl (equiv. 4.5 × 1010 CFU/day) | 10 weeks | Serum vitamin D, vitamin A, calcium, zinc, iron |
| Athalye-Jape et al. (2025) [51] | Australia | RCT | Very preterm infants |
86 (43 live probiotic vs 43 heat-inactivated) | Live or heat-inactivated mix: B. breve M-16V, B. longum subsp. infantis M-63, B. longum BB536; total 3 × 109 CFU/day | 3 weeks | Fecal calprotectin (primary), microbiota, SCFA, NEC, LOS, mortality |
Abbreviations: Wks, weeks; BW, birth weight; CFU, colony-forming unit; PMA,
postmenstrual age; NEC, necrotizing enterocolitis; RCT, randomized controlled
trial; GA, gestational age; AAD, antibiotic-associated diarrhea; GI,
gastrointestinal; FOS, fructo-oligosaccharides; GOS, galacto-oligosaccharides;
LOS, late-onset sepsis; TSB, total serum bilirubin; IB, indirect bilirubin; DBIL,
direct bilirubin; TCB, transcutaneous bilirubin; VLBW, very low birth weight
(
NEC is a prevalent gastrointestinal diseases in neonates, characterized by high morbidity and mortality rates. Its onset is often insidious, progression rapid, and prevention challenging [52]. The younger the gestational age and the lower the birth weight, the poorer the intestinal mucosal barrier function and the lower the abundance of intestinal flora, resulting in greater susceptibility to NEC and a higher probability of developing severe disease [53].
A 2020 retrospective study conducted in a tertiary NICU at Norfolk and Norwich
University Hospital in the UK found that routine administration of
Lactobacillus acidophilus and Bifidobacterium spp.
significantly reduced the incidence of NEC in extremely preterm infants, with no
reported cases of probiotic sepsis [1]. A 2022 double-blind RCT at Tygerberg
Hospital, South Africa, reported a significantly lower incidence of NEC in
preterm infants receiving Bifidobacterium bifidum and
Lactobacillus acidophilus compared to the placebo group. The study concluded
that the use of probiotics to prevent NEC is a safe and effective measure [2]. A
2023 retrospective study in Australia also supported the use of
Bifidobacterium combined with Lacticaseibacillus rhamnosus as a
preventive strategy for NEC [3]. Consistently, a network meta-analysis by
Beghetti et al. (2021) [54] confirmed that Bifidobacterium lactis Bb-12/B94 reduces
the risk of stage
Morgan et al. (2020) [55] analyzed 63 trials including 15,712 preterm infants and concluded that combinations of Lactobacillus and Bifidobacterium, or Lactobacillus reuteri and Lacticaseibacillus rhamnosus, effectively reduced severe NEC and mortality. However, combinations involving Bacillus and Enterococcus were less effective. Chi et al. (2021) [56] reviewed 45 RCTs involving 12,320 preterm infants and found that the combined use of Lactobacillus and Bifidobacterium, or Lactobacillus plus prebiotics, was more effective than single-strain probiotics. Wang et al. (2023) [57] identified multi-strain probiotics alone or combined with oligosaccharides as among the most effective interventions for reducing NEC. In contrast, a 2024 retrospective study from the Czech Republic by Korček and Straňák [4] found no statistically significant difference in NEC incidence or mortality between single-strain and multi-strain probiotics, potentially due to a small sample size or low baseline NEC incidence.
In 2022, Samara et al. [5] (Canada) investigated preterm infants under
29 weeks gestation using a multi-strain probiotic mixture consisting of
Bifidobacterium breve HA129, Bifidobacterium bifidum
HA132, Bifidobacterium longum subsp. infantis HA116,
Bifidobacterium longum HA135, and Lacticaseibacillus
rhamnosus HA111. The intervention accelerated intestinal microbiota maturation
toward a term-like profile and reduced intestinal pro-inflammatory markers
associated with NEC. While not directly demonstrating reduced NEC morbidity or
mortality, the study provided mechanistic support for bifidobacteria’s potential
benefit in extremely preterm infants (
At present, current clinical evidence suggests that probiotics can reduce the incidence of NEC and lower NEC-related mortality, with Lactobacillus spp. (e.g., L. rhamnosus, L. reuteri, L. acidophilus) and Bifidobacterium spp. (e.g., B. longum, B. breve, B. animalis) being the most commonly recommended. Administration typically begins once a minimum enteral feeding volume is achieved (e.g., 60–80 mL/kg/day) and continues until 34–36 weeks postmenstrual age (PMA) or discharge, with most regimens delivering approximately 10⁹ CFU/day.
Multicenter and long-term real-world data support the feasibility and potential benefits of probiotic supplementation; however, limitations remain, including substantial heterogeneity in strains, dosage, initiation thresholds, and duration of treatment; insufficient statistical power due to low baseline NEC incidence and restricted external validity between large negative single-strain RCTs and small positive or mechanistic trials. Future research should focus on standardized, adequately powered, head-to-head randomized controlled trials of specific strains in extremely preterm and very low birth weight populations.
A 2022 double-blind, placebo-controlled randomized clinical trial by Sowden et al. [2] found that combined use of Lactobacillus acidophilus and Bifidobacterium significantly reduced the incidence of FI. The probiotic group also achieved full enteral feeding sooner and regained birth weight earlier than the placebo group [7]. In studies conducted by Indrio and colleagues [8, 9] in Italy, preterm infants with mean gestational ages of approximately 30 weeks and 34 weeks. Supplementation at 108 CFU/day for one month effectively reduced FI occurrence and shortened the time to full enteral nutrition. A meta-analysis of nine studies involving 1244 preterm infants with FI concluded that probiotics promote early growth and reduce FI incidence, although it did not identify preferred specific strains [58]. Similarly, a recent systematic review and meta-analysis in extremely preterm infants reported a non-significant but consistent trend toward improved feeding tolerance with probiotic supplementation, reflected by shorter time to full enteral feeding and reduced reliance on parenteral nutrition [59].
Neonates have a relatively weak immune barrier, making them prone to infections. Common neonatal infections include pneumonia, necrotizing enterocolitis, and sepsis. Antibiotics are often used to treat these conditions. However, prolonged antibiotic use can significantly disrupt the composition and function of the gut microbiota, and this antibiotic-associated dysbiosis may persist even after therapy concludes [60]. Probiotics may prevent AAD through multiple mechanisms, including modulating intestinal microbiota, increasing short-chain fatty acid production, regulating bile acid metabolism, and enhancing gut barrier function and immune responses [61].
Between 2008 and 2011, Wu et al. [10] administered Lactobacillus bifidus triplex tablets to neonates with purulent meningitis during antibiotic treatment, and the study demonstrated a significant reduction in the incidence of AAD and a delay in its onset. In 2013, Liu et al. [11] reported that children on long-term antibiotic therapy who received Bacillus subtilis bifidus granules had a significantly lower incidence of AAD (2%) compared to the control group (36%). Collectively, these studies indicate that probiotic intervention can delay or even prevent the occurrence of AAD. Several studies further suggest that probiotics can aid in the treatment of neonatal AAD. They can alleviate symptom severity (e.g., diarrhea, abdominal pain) and shorten the duration of AAD. A randomized controlled trial in 2021 confirmed that Clostridium butyricum and Bifidobacterium infantis live powder can alleviate clinical symptoms of antibiotic-associated diarrhea, such as abdominal pain and diarrhea [12]. Feng Aimin and others investigated the therapeutic effects of a bivalent probiotic (Clostridium butyricum and Bifidobacterium infantis) versus Saccharomyces boulardii for AAD. The bivalent probiotic showed superior outcomes regarding time to diarrhea cessation, hospitalization duration, and the overall therapeutic efficacy [13]. This conclusion is corroborated by the findings of a clinical study carried out by Du and Wang in 2020 [14]. The superior efficacy of the bivalent preparation may be attributable to its combination of multiple bacterial strains, which more closely mimics the normal proportion of intestinal flora in the human body.
A 2017 multi-center randomized controlled trial in China showed that
administration of Saccharomyces boulardii alongside antibiotics
significantly reduced the incidence of diarrhea in children receiving one or more
antibiotics for
In summary, substantial evidence confirms the efficacy of probiotics in both preventing and treating AAD. Saccharomyces boulardii and Lacticaseibacillus rhamnosus have the strongest evidence base, while Clostridium butyricum and Bifidobacterium longum may also provide benefits in children. Early administration of probiotics alongside antibiotics is recommended to minimize disruption of the gut microbiota and mucosal barrier. However, the optimal duration of probiotic therapy requires further clinical investigation.
A substantial body of randomised controlled trials shows that probiotics help prevent disease and serve as adjuncts to therapy across diverse gastrointestinal conditions. Recent experimental and translational work has clarified the biological underpinnings of these effects, pointing to mechanisms that span the intestinal barrier, immune modulation, and metabolic support.
Probiotics reshape the gut microbiota by encouraging the colonisation and expansion of beneficial taxa such as Bifidobacterium and Lactobacillus, while curbing the overgrowth of potential pathogens to preserve microbial homeostasis. A healthier community structure limits pathogen adhesion and translocation, thereby reducing the likelihood of intestinal barrier injury. In addition to these microbiota-mediated actions, certain strains directly upregulate tight junction proteins in intestinal epithelial cells—including occludin, claudin, and ZO-1. By reinforcing intercellular integrity, probiotics help normalise mucosal permeability and support the recovery of barrier function [63, 64].
Probiotics shape host immunity by remodelling the microbial community and by
directly engaging receptors on epithelial and immune cells, including TLR2, TLR4,
and TLR9. Through these interactions they tune key signalling pathways—most
notably NF-
Probiotic-derived short-chain fatty acids—acetate, propionate, and butyrate—serve as the main fuel for colonocytes. Beyond energising the epithelium, SCFAs stimulate epithelial proliferation and repair, trigger mucus secretion from goblet cells, bolster antimicrobial peptide production, and preserve an acidic luminal milieu that suppresses pathogenic growth [66]. Collectively, this metabolic support is pivotal during intestinal inflammation and tissue repair, accelerating the recovery of mucosal structure and function.
Probiotics operate along intersecting pathways: they fortify the intestinal epithelial barrier; directly calibrate signalling to dampen inflammation; and harness microbial metabolites to optimise the intestinal milieu and drive epithelial repair. Taken together, these actions offer a strong biological rationale for the clinical benefits reported in NEC, FI, and AAD.
Neonatal sepsis is a common infectious disease and a leading cause of neonatal mortality. A prospective study by Madan et al. [67] (United States) showed that preterm infants experienced a decline in gut microbiota diversity preceding the development of late-onset sepsis (LOS), which persisted until sepsis onset. This suggests a role for the gut microbiota in the pathogenesis of LOS. However, evidence supporting probiotic use for the prevention or treatment of neonatal sepsis remains limited.
A prospective, double-blind randomized controlled trial conducted by Serce
Pehlevan et al. [18] in Turkey (2020) found that administration of
Lactobacillus acidophilus and Bifidobacterium bifidum to very
low birth weight infants (gestational age
However, a retrospective study conducted in New Zealand found that
Bifidobacterium bifidum and Lactobacillus rhamnosus
reduced the prevalence of LOS in preterm infants with a gestational age of less
than 32 weeks [21]. A randomized, double-blind, placebo-controlled clinical trial
of 4556 neonates found that a 7-day synbiotic intervention with Lactobacillus
plantarum (109 CFU) and oligofructose reduced sepsis incidence and all-cause
mortality [22]. Dermyshi et al.’s meta-analysis [72] of 30 RCTs and 14
observational studies showed probiotics lowered LOS incidence in very low birth
weight infants (1000–1500 g), but not in extremely low birth weight infants
(
Overall, the evidence supporting probiotics for preventing neonatal LOS remains weak and requires confirmation by additional clinical trials. In terms of probiotic and feeding interactions, a 2017 meta-analysis concluded that the beneficial effect of probiotics on lowering LOS incidence is only confirmed in exclusively breastfed preterm infants [73]. This finding is consistent with the conclusions of a study by Li et al. (2021) [70]. Additionally, a systematic review has shown that exclusive breastfeeding itself does not reduce LOS incidence [74], suggesting that probiotic supplementation in exclusively breastfed neonates may help lower the incidence of LOS. Furthermore, a 2023 meta-analysis indicated that the combination of single probiotic strains with lactoferrin represents the most effective intervention for reducing LOS, while multi-strain probiotics demonstrated moderate efficacy against LOS [57]. A limitation of this analysis, however, is that it did not specify which strain combinations were particularly optimal.
Neonates are susceptible to elevated bilirubin levels due to factors such as the short lifespan of erythrocytes, immature hepatic function, and enhanced enterohepatic circulation, which can lead to neonatal hyperbilirubinemia. Untreated significant hyperbilirubinemia can affect the development of the central nervous system, potentially causing permanent sequelae. Currently, the most commonly used effective and safe clinical intervention is phototherapy. It enables the photoisomerization of unconjugated bilirubin to form water-soluble lumirubin, which is excreted directly through bile and urine without hepatic processing. However, phototherapy may cause adverse reactions such as fever, rash, and diarrhea, potentially linked to increased intestinal peristalsis and microbiota disruption.
Zhang et al. [75] reported that phototherapy altered the intestinal microbiota, with significant decreases in Alistipes putredinis and unclassified Cellvibrio, and notable increases in Clostridium bolteae, Enterobacter cloacae, and Enterococcus faecium. Yuan et al. [24] evaluated the use of Lactobacillus rhamnosus in full-term infants with hyperbilirubinemia receiving phototherapy. They found that microbiota changes after phototherapy were more pronounced in controls than in the probiotic group, suggesting that L. rhamnosus may mitigate microbiota disruption, promote faster recovery of depleted intestinal flora, and reduce diarrhea frequency in jaundiced neonates.
Beyond modulating phototherapy-induced intestinal dysbiosis and mitigating its
adverse effects, numerous clinical studies have confirmed that probiotics exert
adjunctive function in the treatment of neonatal hyperbilirubinemia.
Specifically, they can significantly reduce serum bilirubin levels in full-term
neonates and late preterm infants, shorten the duration of jaundice, and thereby
improve prognosis. Fan et al. [76] proposed mechanisms, including
regulation of the intestinal microbiota, reduction of
International studies include Mutlu et al. [29], who found that intervention with Lactobacillus rhamnosus in healthy neonates resulted in significantly lower serum bilirubin levels and bilirubin rebound 36 hours post-phototherapy compared to controls. This finding indicates that early probiotic intervention can reduce bilirubin levels, decrease the incidence of neonatal hyperbilirubinemia, and delay the onset of the condition. In Iran, Eghbalian et al. [30] showed that a probiotic complex shortened phototherapy duration and hospital stay for full-term hyperbilirubinemic infants. Nasief et al. [27] and Demirel et al. [31] have confirmed that Saccharomyces boulardii helps in the adjuvant treatment of hyperbilirubinemia in preterm infants. Clinical studies on Bifidobacterium animalis have also verified its efficacy as an adjuvant treatment for neonatal hyperbilirubinemia [25].
Some studies challenge the use of probiotics in neonatal hyperbilirubinemia. Santosa et al. [32] in Japan, using Bifidobacterium animalis to intervene in healthy term neonates, found that probiotics showed no significant effect in reducing bilirubin levels during the first five days of life. A randomized controlled trial by Serce et al. [33] concluded that Saccharomyces boulardii had no marked impact on the clinical course of neonatal hyperbilirubinemia. Furthermore, after pooling nine randomized controlled trials, Deshmukh et al. [79] found only limited, low-quality evidence supporting probiotics for shortening phototherapy duration in jaundiced neonates, thus not recommending their routine use of probiotics for preventing or treating neonatal jaundice.
Regarding probiotic selection, no adequate controlled studies have confirmed that multi-strain formulations are more effective than single strains. Conversely, when comparing Saccharomyces boulardii and Lactobacillus casei in treating neonatal pathologic jaundice, Sun et al. [34] found that both strains, when combined with phototherapy, significantly improved treatment efficacy for neonatal pathologic jaundice, with no significant difference in efficacy between the two strains.
Research on probiotic-aided therapy for neonatal hyperbilirubinemia is
predominantly from Asian countries, whereas research on this topic in European
nations remains relatively limited. This disparity may relate to the lack of
internationally unified diagnostic criteria for neonatal hyperbilirubinemia,
which could influence the reported incidence rates of this condition across
different countries. Additionally, subjects in previous studies have exclusively
included late preterm infants and term infants with a gestational age of
In neonatal hyperbilirubinaemia, emerging work shows a layered mode of action: beyond reshaping the gut environment, probiotics tune metabolic and signalling checkpoints that, in turn, influence bilirubin production, conversion, and clearance. These convergent effects help explain their growing therapeutic promise.
In early neonatal life the gut microbiota is immature: aerobes predominate, while obligate anaerobes such as Bifidobacterium and Lactobacillus are relatively scarce [80]. This imbalance drives the accumulation of harmful metabolites, facilitating both the production and uptake of bilirubin. Probiotic supplementation accelerates the colonisation and expansion of beneficial taxa and curbs the growth of potential pathogens (e.g., Enterobacteriaceae and Enterococcus), thereby restoring microbial homeostasis. As the community adopts a healthier profile, bilirubin output falls and the intestinal environment stabilises—setting the stage for the spontaneous clearance of jaundice.
Probiotics also blunt the enterohepatic circulation. In the neonatal gut,
Probiotics also meaningfully influence gut motility. By generating SCFAs, they trigger contractions of the intestinal smooth muscle, sharpen gastrointestinal peristalsis, and ease the excretion of bilirubin with the faeces [82]. As the lumen becomes more acidic and its osmotic pressure rises, water secretion increases and stools are diluted, which further hastens transit. Evidence from several studies indicates that multi-strain formulations containing Bifidobacterium and Lactobacillus acidophilus act synergistically to boost motility and support bilirubin metabolism, resulting in a marked reduction in the duration of jaundice.
Overall, probiotics operate along a continuum: they restore microbial balance,
suppress
Recent mechanistic research has expanded to identify the gut-lung axis, with several studies confirming beneficial effects of probiotics on respiratory diseases.
VAP is defined as infectious inflammation of the lung parenchyma occurring
Only a limited number of studies have reported the use of probiotics in VAP. In
the neonatal intensive care unit (NICU) population, a randomized controlled trial
enrolling full-term infants (gestational age 37–42 weeks) requiring invasive
mechanical ventilation showed that daily supplementation with a probiotic
preparation containing 2
Clinical studies on probiotics for VAP prevention remain limited in number, with
those focusing on neonatal VAP being exceedingly scarce. Notably, an
international RCT involving children under 12 years of age demonstrated that
prophylactic administration of probiotics (Lactobacillus and
Bifidobacterium at 3
A randomized, double-blind, placebo-controlled trial in Finland (2000–2003)
gave full-term newborns a 6-month intervention with multi-strain probiotics
including Lacticaseibacillus rhamnosus GG and LC705,
Bifidobacterium lactis Bb99, and propionibacterium
freudenreichii subsp. shermanii JS. During the intervention period, no
adverse reactions were observed in the trial group, and the probiotic group had a
lower frequency of respiratory infections [37]. A prospective clinical study in
Ukraine in 2011 used Escherichia coli Nissle 1917 to supplement late preterm
infants. The study found that within 28 days after birth, the proportion of acute
respiratory infections (ARI) in the trial group was significantly lower than in
the control group (10.0% vs 43.7%, p = 0.008) [38]. A multi-center RCT
in France found that Lactobacillus fermentum had a similar effect [39].
However, an RCT in Japan using Lactobacillus fermentum for late preterm
infants (gestational age
BPD, a common complication in preterm infants, remains incompletely understood in its pathogenesis. Traditionally, BPD has been associated with factors such as immature lung development, hyperoxic exposure, and inflammatory responses. Recent studies, however, emphasize a close association with intestinal microbiota via the gut-lung axis. Intestinal dysbiosis may promote BPD progression, suggesting that probiotics could potentially prevent or treat BPD through this axis [83].
A retrospective clinical study in China in 2021 found that Clostridium
butyricum helped reduce the risk of BPD in extremely preterm infants
(gestational age
Evidence for the use of probiotics in respiratory diseases remains limited, and the optimal timing, dosage, and duration of intervention are not well defined. Studies on VAP, ARI, and BPD are generally small-scale and heterogeneous, making it difficult to draw firm conclusions. Notably, some reports have suggested that administration of bifidobacteria within 24 hours of initiating mechanical ventilation, continued for approximately one week, may reduce both respiratory and gastrointestinal bacterial colonization in ventilated infants [43, 44]. However, this observation is based on a small number of studies and requires confirmation through well-designed, adequately powered randomized controlled trials. At present, the optimal timing for probiotic intervention in respiratory disease is currently uncertain, and further methodologically robust clinical research is needed to establish evidence-based recommendations.
Allergies occur when the immune system overreacts to environmental or food antigens, manifesting as inflammatory symptoms such as eczema, rhinitis, vomiting, and diarrhea. In infancy and early childhood, the primary allergic disorders include eczema, erythema, and cow’s milk protein allergy.
As eczema peaks in infancy and early childhood rather than the neonatal period, research on probiotics for preventing and treating neonatal eczema is scarce. However, given that probiotics can exert positive effects during infancy and adolescence, early probiotic intervention in the neonatal period may reduce the subsequent development and progression of allergic diseases in infants and children.
Sun et al.’s 2021 meta-analysis [84] demonstrated that combinations of Lactobacillus and Bifidobacterium strains significantly reduced eczema incidence in children under three years of age. Notably, probiotic administration initiated early during pregnancy exhibited greater efficacy.
A 2025 RCT by Puisto et al. [45] reported similar findings: perinatal maternal probiotic intervention with Lactobacillus rhamnosus, Lactobacillus paracasei, and Bifidobacterium bifidum reduced the risk of atopic eczema in infants within the first 2 years of life. Notably, this protective effect appeared independent of gut microbiota modulation, as no statistically significant differences in gut microbial composition were detected between the groups. Additionally, a meta-analysis demonstrated that Lactobacillus rhamnosus effectively reduced eczema incidence. In the included studies, the timing of interventions ranged from 2–4 weeks prior to delivery to 6 months–2 years postpartum. The analysis revealed that this benefit was most pronounced during follow-up periods of up to 2 years and 6–7 years. By contrast, no statistically significant reduction in the incidence of atopic eczema was observed at 4–5 years or 10–11 years of follow-up [85].
CMPA is the most prevalent food allergy in neonates, characteristically manifesting with clinical features such as cutaneous eczema, respiratory symptoms, and gastrointestinal disturbances. However, due to the immature immune response and other unique physiological characteristics of neonates, the clinical manifestations of CMPA often lack specificity, frequently leading to misdiagnosis as conditions like neonatal necrotizing enterocolitis, sepsis, among others [86]. While strict avoidance of milk proteins constitutes the cornerstone of therapy, emerging evidence supports adjuvant use of probiotics to alleviate allergy-related symptoms.
Luo and Zhang’s study [46] showed that compared with the control group, after three months of continuous oral administration of probiotics (Bifidobacterium lactis BB12 + Lacticaseibacillus rhamnosus GG) in children with CMPA, significant statistical differences were observed in clinical symptoms such as vomiting, diarrhea, and food refusal, as well as in weight-for-age scores. However, no significant differences were noted during the first 1–2 months of intervention. These results indicate that probiotics should be administered for at least three months to significantly improve infant gastrointestinal function, thereby promoting nutritional metabolism and supporting weight gain in infants. Nocerino et al. [47] from Italy reported that the time to achieve immune tolerance was shortened in CMPA infants fed hydrolyzed formula supplemented with Lactobacillus rhamnosus. In addition, a six-month intervention with Bifidobacterium TMC3115 reduced allergic symptoms in CMPA infants, which is associated with an increased probiotic and decreased pathogenic bacteria proportion in the gut.
Probiotics recalibrate immune polarity by restoring the balance between Th1 and
Th2 activity. Under allergic conditions, immunity is typically skewed towards a
Th2 profile, with elevated IL-4, IL-5, and IL-13 driving IgE production and
mast-cell degranulation. By shaping dendritic-cell differentiation and antigen
presentation, probiotics enhance Th1-type signalling—most notably
IFN-
Probiotics engage the host immunoregulatory network, driving the differentiation
and expansion of regulatory T cells (Treg) [89]. By releasing IL-10 and
TGF-
Through complementary mechanisms, probiotics foster immune tolerance. They lower IgE concentrations, weakening immune complex–mediated allergic reactions; simultaneously they restrain overactivation of mast cells and basophils, curbing histamine release and easing symptoms at their origin [3]. Probiotic derivatives also signal via intestinal epithelial receptors to reinforce barrier function and limit antigen translocation, helping the immune system to recognise that harmless exposures need not be treated as threats.
Probiotics do more than quell inflammation: they retune the immune network’s “tone”, guiding an overresponsive system back into balance—from restoring homeostasis to fostering tolerance. This measured, long-lasting modulation gives probiotics a distinctive biological appeal in the prevention and treatment of allergic disease.
Although probiotics may influence human micronutrient status, research specifically targeting neonates remains scarce. Accordingly, this discussion focuses primarily on pediatric populations, while incorporating evidence from adult cohorts and non-clinical studies to inform potential applications in children and, and by extension neonates.
Micronutrient deficiencies pose a significant global health challenge, especially for children vulnerable due to rapid growth, unbalanced diets, and their high burden of infectious diseases. In addition to aiding food digestion and absorption, probiotic bacteria can synthesize various water-soluble vitamins, such as folate, riboflavin, vitamin B12, thiamine, and pyridoxine, and enhance the absorption of minerals like calcium, iron, and zinc. Compared to fortification using synthetically produced nutrients, modulation of host nutrient status through vitamin-producing and absorption-promoting probiotics is more physiologically aligned with natural metabolic pathways, potentially minimizing adverse effects.
The relationship between probiotics and mineral status is complex. The systematic review by Apte et al. [91] analyzed the differences in the effects of probiotics on iron absorption between women and children. Through nalyzing 29 studies (14 in women of reproductive age, 15 in children), results for iron absorption were more favorable in women. Meta-analysis of six studies demonstrated a mean increase in serum ferritin of 2.45 ng/mL (p = 0.009) with moderate-quality evidence. More importantly, pooled data from eight studies on fractional iron absorption indicated a mean increase of 0.74% (p = 0.02), with particularly pronounced effects for galacto-oligosaccharides (GOS) and Lactiplantibacillus plantarum 299v. In contrast, the results for children were less encouraging. Meta-analysis of eight studies showed no significant change in hemoglobin, and four studies reported no improvement in serum ferritin.
The effects of probiotics on calcium and zinc nutrition also vary across populations. A randomized controlled trial by Agustina et al. [48] demonstrated that supplementation with Lactobacillus reuteri or L. casei failed to significantly improve serum calcium and zinc levels among children aged 1–6 years. However, positive results were observed in elderly [92] and postmenopausal women [93]. With regard to zinc, probiotics and zinc appear to exert a synergistic effect. In a randomized, double-blind, placebo-controlled trial involving Indonesian children aged 12–24 months, four groups received either placebo, Lactobacillus plantarum IS-10506 (1010 CFU/day), zinc (8 mg/day), or the probiotic-zinc combination for 90 days. Neither probiotic nor zinc alone significantly improved serum zinc versus baseline, whereas combined supplementation significantly increased serum zinc levels [49].
In clinical studies, randomized double-blind trials and open-label trials in children have shown that probiotic or combined prebiotic-probiotic interventions do not produce a marked improvement in vitamin A status. However, for vitamin D, an upward trend in serum levels was observed when multi-strain formulations containing Lactobacillus plantarum, L. acidophilus, Bifidobacterium infantis, and B. lactis were used [50], although statistical significance was not always clearly reported and results varied across populations. Studies in both pregnant women and children suggest that the effects depend on strain type (e.g., L. reuteri, B. animalis), dosage, and duration of intervention, and the findings remain inconsistent. Therefore, current evidence is insufficient to establish a consistent beneficial effect of probiotics on vitamin A or D status.
Evidence supporting the role of probiotics in enhancing the production of water-soluble B vitamins is relatively strong; however, this effect appears to be highly strain-specific rather than rather than a general property of a genus. Lactic acid bacteria and bifidobacteria are recognized as key contributors to B-vitamin production in the human gut, yet their synthetic capacity, the spectrum of vitamins produced, and the bioavailability of these compounds vary substantially among different strains. An animal study has shown that Bifidobacterium adolescentis can produce vitamin B9 (folate) [94]. A human study has also confirmed that Bifidobacterium adolescentis DSM 18350, B. adolescentis DSM 18352, and B. pseudocatenulatum DSM 18353 are capable of synthesizing and secreting folate in the human intestinal environment, thereby providing a constant additional source of endogenous folate within the gut lumen [95]. However, the number of high-quality randomized controlled trials is small, and direct evidence in pediatric populations, particularly in neonates, remains scarce.
Animals, plants, and fungi are incapable of vitamin B12 (cobalamin) production and it is exclusively synthesized by microorganisms. Lactobacillus reuteri CRL1098 was shown to be the first lactic acid bacteria strain able to produce a cobalamin-like compound [96]. In vitro studies have demonstrated that Lactobacillus plantarum is also capable of producing vitamin B12 [97]. In addition to direct synthesis, probiotics may improve vitamin B12 status by modulating gut microbiota composition, potentially reducing the abundance of bacterial species that degrade this vitamin [98].
Collectively, these findings indicate that while probiotics contribute significantly to B-vitamin availability, their effects are not uniform across vitamins and species. This highlights the need for precise, strain-level selection and validation in probiotic applications, particularly when targeting specific vitamins or addressing the needs of vulnerable populations such as children.
In clinical practice, most patients tolerate probiotics without adverse effects, with only a small minority reporting gastrointestinal issues. While the majority of current clinical trials and practical applications have not associated probiotic use with significant adverse events, some studies highlight a potential risk of probiotic-associated sepsis, particularly in preterm infants.
Kulkarni et al.’s systematic review [99] included 16 reports involving 32 newborns with probiotic sepsis (blood/cerebrospinal fluid cultures positive for the administered probiotic strains, accompanied by clinical manifestations of infection). Two of these newborns died (one death was unrelated to probiotic supplementation), and the rest were successfully treated with antibiotic and antifungal therapy. The probiotics involved included Bifidobacterium longum, Bifidobacterium breve, Lactobacillus rhamnosus, Lactobacillus reuteri. Most cases were preterm infants with a gestational age of less than 32 weeks. A 2025 meta-analysis by Feldman et al. [100], incorporating 63 studies with over 20,000 participants, evaluated the risk-benefit ratio of probiotics in preterm infants and found the overall incidence of probiotic sepsis to be less than 0.04%, supporting a favorable benefit-risk profile.
However, potential under-detection due to diagnostic limitations worth consideration. In the studies included by Feldman et al. [100], only two large observational studies involving 562 infants captured all 8 probiotic sepsis events, both of which emphasized unique detection methods. One study used matrix-assisted laser desorption/ionization time-of-flight mass spectrometry for improved strain identification, while the other extended blood culture time to 7 days (noting that the growth period of Bacillus brevis under aerobic conditions is 133 hours), thus capturing relevant probiotic bacteremia cases. Similarly, Abda et al. [101] identified probiotic bacteremia cases through aerobic blood culture for 10–21 days, far exceeding the current clinical standard of 5 days for blood culture.
A 2025 clinical study by Athalye-Jape et al. [51] found no statistically significant differences in gut microbial changes or clinical outcomes between groups receiving heat-inactivated and live Bifidobacterium, suggesting that heat-inactivated probiotics may offer benefits while circumventing the risk of probiotic sepsis.
In conclusion, current attitudes toward probiotic safety are generally positive, with an extremely low incidence of probiotic sepsis and effective antibiotic intervention. However, the possibility of missed diagnoses due to monitoring limitations cannot be overlooked. Recent research suggests that using heat-inactivated probiotics may circumvent the risks associated with live probiotics.
Natural probiotics help sustain gut homeostasis and support host health, yet real-world effectiveness is constrained by strain specificity, inter-individual variability, and the host microenvironment. Typically sourced from fermented foods or commensal flora, most strains colonise only transiently, narrowing the window for interaction with resident microbes. Because their activity hinges on native gene-expression programmes and lacks tunable control, strains differ widely in metabolite output, immune modulation, and barrier support. This variability underlies the lack of consensus on which strains to use, at what dose, and for how long, and it limits both comparability and reproducibility across studies. While several trials report benefits, results often vary by population and disease model—and sometimes point in opposite directions. Survival and adhesion are frequently poor in the complex gut milieu, yielding short-lived colonisation and waning therapeutic effect. Although generally safe, live preparations can rarely cause bacteraemia, catheter contamination, or horizontal transfer of resistance genes; caution is therefore warranted in preterm infants and in immunocompromised individuals [102].
Rapid progress in synthetic biology has catalysed the rise of genetically modified probiotics (GMPs), providing ways to surmount the functional variability of natural strains. By combining targeted gene editing, pathway re-engineering, and tunable regulatory circuits, investigators can programme probiotic chassis with bespoke therapeutic functions—sensing pathological cues, producing anti-inflammatory peptides, or detoxifying harmful metabolites [103]. Notably, engineered Escherichia coli Nissle 1917 strains can detect oxidative stress within the gut and release anti-inflammatory effectors, while embedded biosensors finely control therapeutic-protein expression. Together, these capabilities show promise for gastrointestinal disorders and metabolic dysregulation [104, 105], marking a shift from passive micro-ecological modulation to genuinely programmable living medicines.
The rollout of GMPs is still limited by safety and regulatory hurdles. Exogenous gene insertion can compromise genomic stability and enable the horizontal transfer of antimicrobial-resistance genes, creating risks of environmental spread [104]. Accordingly, engineered strains must maintain reliable function and incorporate rigorous biocontainment to prevent escape or onward transmission. Within current frameworks, genetically modified probiotics are regulated as live biotherapeutic products (LBPs) and are required to meet pharmaceutical-grade expectations for genetic stability, traceability, and manufacturing consistency. These higher safeguards elevate safety but inevitably slow the pace of clinical translation.
On balance, natural probiotics benefit from a favourable safety profile and an extensive record of use, yet functional uncertainty and poor colonisation limit their application and impede the standardisation of efficacy. In contrast, genetically modified probiotics offer compelling scope for precise control and targeted therapy, with the potential to deliver directed immune modulation, metabolic correction, and both prevention and treatment of disease. Looking ahead, progress hinges on striking the right balance between safety, functional stability, and ecological containment. The end-point is a shift from empirical formulations to rationally designed, bioengineered—and ultimately programmable—therapeutic platforms.
The clinical study used and analyzed during the current study are available from PubMed and CNKI on reasonable request.
ZS: designed the research study, performed the research, writing original draft, writing-review and editing. YY: designed the research study, performed the research, writing-review and editing. JC: designed the research study, performed the research, writing-review and editing. FZ: designed the research study, performed the research, writing-review and editing. JZ: Conceptualization, Funding acquisition, Investigation, Resources, Software, Visualization, Writing-review & editing. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.
Not applicable.
Not applicable.
This research received no external funding.
The authors declare no conflict of interest.
During the preparation of this work the authors used ChatGpt-4.0 to check spell and grammar. After using this tool, the authors reviewed and edited the content as needed and takes full responsibility for the content of the publication.
References
Publisher’s Note: IMR Press stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
