Efficacy and Safety of Bifidobacterium longum Supplementation in Infants: A Meta-Analysis of Randomized Controlled Trials Authors: Huangda Guo, Meng Fan, Tianjiao Hou, Yixin Li, Siyue Wang, Xueheng Wang, Hexiang Peng, Mengying Wang, Tao Wu, Yumei Zhang, Joe Vinson, Corrado Rizzi Published: 2023 ABSTRACT AbstractBackground: Strategies to stabilize and support overall infant health by increasing the number of Bifidobacterium longum in the infant gut are of interest, but few studies have systematically addressed this issue. We aimed to evaluate the efficacy and safety of Bifidobacterium longum use in infants using meta-analysis. Methods: We searched PubMed, EMBASE, Cochrane Library of Systematic Reviews, and SinoMed for publications until 27 July 2022. The main outcomes of interest were weight gain, risk of necrotizing enterocolitis (NEC), and adverse events. Two authors independently performed study screening, risk of bias assessment, and data extraction. Outcome data were extracted from each included study and combined using mean difference (MD) or risk ratio (RR) and finally combined using a fixed-effect model or random-effect model. Results: A total of 4481 relevant studies were identified, of which 15 were found to be eligible for randomized controlled trials and were included in the meta-analysis. The combined extracted data showed that the intervention group containing Bifidobacterium longum had a significantly lower risk of NEC (RR = 0.539, 95% CI: 0.333, 0.874) compared to the control group. There was no statistical difference between the intervention and control groups regarding weight gain (MD = 0.029, 95% CI: −0.032, 0.090), the occurrence of adverse events (RR = 0.986, 95% CI: 0.843, 1.153), and serious adverse events (RR = 0.881, 95% CI: 0.493, 1.573). Conclusions: Bifidobacterium longum may significantly reduce the risk of NEC in infants as well as being safe; thus, further research evidence is needed on whether there is a benefit on weight gain.Keywords: Bifidobacterium longum, meta-analysis, randomized clinical trials, necrotizing enterocolitis, efficacy, safety 1. Introduction The initial colonization of the neonatal gut is a crucial and intricate process that significantly influences the development of the entire intestinal microbiota. This early microbial colonization not only shapes the complex ecosystem within the gut but also has the potential to impact various aspects of both intestinal and systemic health throughout an individual’s lifespan [1,2,3,4]. In recent years, the use and supplementation of probiotics have become prominent and dynamic areas of research, especially in the context of neonatal and infant health. Among the diverse array of probiotic species, specific members of the Bifidobacterium genus have gained considerable attention due to their potential pivotal roles in early human development [5]. Bifidobacterium longum, in particular, has gained significant attention as a noteworthy probiotic, making its way into a variety of functional foods and dietary supplements. A growing body of evidence suggests that Bifidobacterium longum may play a pivotal role in promoting optimal infant health. While the probiotic effects are believed to be strain-specific, earlier literature indicated that different strains of Bifidobacterium longum share common mechanisms to some extent [6]. Noteworthy correlations have been established, linking Bifidobacterium longum to favorable outcomes such as enhanced weight gain in infants [7] and a decreased incidence of the serious condition known as necrotizing enterocolitis (NEC) [8]. Furthermore, the metabolites produced by Bifidobacterium longum have been identified as crucial mediators in the maturation of the infant immune system [9] and as agents that instigate vital biochemical changes within the intestinal environment, effectively impeding the invasion and proliferation of potentially harmful pathogens [10]. Regarding the safety of Bifidobacterium longum, several studies have reported adverse effects in clinical trials, including gastrointestinal disorders [11,12,13], feeding intolerance [14], diarrhea [11,15], fever [12,13,15], respiratory symptoms [11,12,13,15,16], and even death [16,17,18]. However, due to the relatively limited sample size, it remains unclear whether these adverse effects can be conclusively attributed to Bifidobacterium longum supplementation. In light of the increasing interest and a rising number of clinical trials investigating the efficacy and safety of probiotics in infants, it is crucial to emphasize that a thorough and comprehensive assessment of Bifidobacterium longum and its subspecies is still pending. The potential to improve overall infant health by stabilizing and boosting Bifidobacterium longum populations in the infant gut represents a highly promising and significant avenue. Nevertheless, it is noteworthy that a comprehensive evaluation of the aforementioned probiotic and its subspecies is yet to be conducted. In this systematic review, our primary objective is to meticulously and comprehensively examine the current body of evidence regarding the efficacy and safety of Bifidobacterium longum. Our specific focus will be on its potential impact on infant growth, its role in preventing NEC, and a thorough evaluation of any potential adverse events associated with its supplementation. Through this comprehensive review, our aim is to illuminate the multifaceted aspects of Bifidobacterium longum’s influence on infant health, bridging existing knowledge gaps and establishing a solid foundation for future research in the critical domain of neonatal and infant care. 2. Materials and Methods 2.1. Literature Search The review process was documented using Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines (PRISMA). We conducted a thorough search across four electronic databases, specifically PubMed (https://pubmed.ncbi.nlm.nih.gov, accessed on 27 July 2022), EMBASE (https://www.embase.com, accessed on 27 July 2022), Cochrane Library of Systematic Reviews (https://www.cochranelibrary.com, accessed on 27 July 2022), and SinoMed (http://www.sinomed.ac.cn, accessed on 27 July 2022). This search encompassed studies published in both English and Chinese, covering the period from the inception of the databases up to 27 July 2022. In order to identify relevant trials, we employed a set of carefully chosen keywords, including infant, newborn, neonate, child, preschool, young, toddler, randomized controlled trial, controlled clinical trial, trial, and Bifidobacterium. For a more detailed breakdown of the search terms used, please refer to Table S1 in the Supplementary Material. 2.2. Study Selection Reviewers who underwent training were organized into pairs for the assessment process. Each pair consisted of two independent reviewers within the same group, and their task was to scrutinize the titles and abstracts of the studies to pinpoint those that potentially met the inclusion criteria. Studies that remained ambiguous based on titles or abstracts underwent a comprehensive examination through full-text reading. In case of discrepancies, each review group engaged in discussions to resolve inconsistencies. If disagreements persisted after the initial discussion between the two independent reviewers in a group, a third expert was brought in to conduct a detailed review and make a final determination. This meticulous and systematic approach ensured a thorough and reliable evaluation of the studies under consideration. The “Participants-Intervention-Comparison-Outcomes-Study” (PICOS) approach was used for eligibility criteria. Only studies meeting the following criteria were included: (1) P (participants): the subjects were infants (including preterm infants) and young children aged 0–3 years without major diseases; (2) I (intervention): the treatment group consumed products containing Bifidobacterium longum or Bifidobacterium subsp. longum or Bifidobacterium subsp. infantis such as powder, milk, tablet, or capsule; (3) C (comparison): for comparisons, the current meta-analyses including both passive (e.g., milk without Bifidobacterium longum supplementation) and active (e.g., other probiotic supplementation) control conditions were eligible; (4) O (outcomes): the outcome of the studies included efficacy (growth and development indicators, gastrointestinal function and immune function) or safety (adverse events such as allergy, death, etc.); (5) S (study): randomized controlled trials (RCTs) were included, while quasi-RCTs were NOT eligible. 2.3. Data Extraction Two proficient reviewers, namely H.G. and M.F., meticulously examined the encompassed studies and meticulously extracted a comprehensive set of data from each study. This extracted information encompasses the authors’ names, publication dates, countries conducting the studies, sample sizes, supplementation durations, baseline characteristics of the study population, as well as details regarding the probiotic species employed and their respective dosages. Also, the outcome parameters included: (1) weight gain at the end of the study, i.e., the increment in the last weight measurement during the study period compared to the measurement during the baseline period; (2) risk of necrotizing enterocolitis; (3) safety outcomes. To identify safety outcomes, we extracted pre-stated adverse events outcomes monitored by researchers or parents in each study. According to the primary study, any adverse events were assessed as serious adverse events if they were life threatening, caused permanent harm, resulted in hospitalization or prolongation of existing hospitalization, or resulted in persistent or significant disability or incapacity. The number of participants who experienced adverse events during the study period was extracted. Continuous variables were evaluated or calculated with mean difference (MD) and 95% confidence intervals (CI) from the provided data. And categorical variables were evaluated or calculated with risk ratios (RRs) and 95% confidence intervals (CIs). 2.4. Quality Assessment and Statistical Analysis We used the Cochrane Risk of Bias tool (ROB 2.0) to assess the quality of included studies [19]. Five main biases were evaluated: bias arising from the randomization process; bias due to deviations from intended interventions; bias due to missing outcome data; bias in the measurement of the outcome; and bias in selection of the reported results. There are three bias assessment criteria, “Low”, “High”, and “Unclear”, for each point. For a more detailed understanding of the specific items and criteria employed in the quality assessment using this tool, additional information can be found in the Cochrane Handbook for Systematic Reviews of Interventions [20]. Data analysis was performed using R, version 4.2.1 (package “meta”). We assessed heterogeneity among the studies by applying the Q test and inconsistency (I2) statistic. If p < 0.05 or I2 > 50%, we used the random-effects models to estimate the pooled estimates and corresponding 95% CI; otherwise, the fixed-effects models were utilized [21,22]. p values <0.05 suggested statistical significance. 3. Results 3.1. Study Selection and Characteristics The flowchart delineating the literature search process is illustrated in Figure 1. Initially, a total of 4481 pertinent records were identified from diverse databases. Following the removal of duplicate articles and a meticulous review of titles and abstracts, 153 papers remained within the purview of consideration. Subsequently, after a comprehensive examination of the full-text articles, 138 records were excluded for the following reasons: non-compliance with the criteria for randomized controlled trials (30 records), involvement of non-healthy infants (39 records), absence of Bifidobacterium longum in the intervention group (32 records), examination of outcomes unrelated to our research interests (18 records), and incomplete data (19 records). Ultimately, 15 studies met the eligibility criteria for inclusion in the meta-analysis [7,8,11,12,13,14,15,16,17,18,23,24,25,26,27], and their respective data were extracted for further analysis. Flow chart depicting the study selection. After applying the inclusion and exclusion criteria, a total of 15 trials were included in the meta-analysis. Table 1 presents a comprehensive overview of the key characteristics of the eligible randomized controlled trials (RCTs). The geographical distribution of study locations was extensive, encompassing diverse countries across multiple continents. The trials were predominantly conducted in Asia, with a focus on countries such as India, Indonesia, and China. Additionally, research efforts extended to North America (specifically the United States), Oceania (notably Australia), and various European nations, including France, Italy, and Spain. In total, these RCTs involved 3152 participants, with participant numbers varying from 30 to 1099 per trial. The study population comprised a noteworthy proportion of female participants, ranging from 37% to 60%. The majority of participants fell into categories such as newborns, premature infants, and low birth weight babies. Furthermore, Table 1, along with Supplementary Material Table S2, furnishes detailed information regarding the outcomes of interest, mode of delivery, probiotic strains, dosage, antibiotic utilization, type of milk feeding, and the duration of supplementation for each RCT. Table 1Basic characteristics of the studies included in the meta-analysis.StudyYearCountryPatients 1FemaleBirth Weight (g)Gestational Age (Week)Cesarean 1Postnatal Age at Starting (Month)Probiotic StrainsTotal Dose (CFU/d) 2Duration of Supplementation (Week)Antibiotic UseType of Milk FeedingOutcomes 3Al-Hosni et al. [7]2012America50/5149.50%7782622/304 Bifidobacterium longum infantis, Lactobacillus rhamnosus GG 5 × 10834nobreast milkaMark et al. [26]2006America31/2936.70%14283023/237Bifidobacterium longum, Bifidobacterium longum infantis Lactobacillus acidophilus, Bifidobacterium bifidum5 × 1085no-aShashidhar et al. [18]2017India52/5254.80%12233127/380 Bifidobacterium longum Lactobacillus acidophilus, Lactobacillus rhamnosus, Saccharomyces boulardii 1.25 × 1094nobreast milka, b, dQiao et al. [14]2017China30/3055.00%159332-/-0 Bifidobacterium longum, Lactobacillus acidophilus and Enterococcus faecalis 1 × 1072yesbreast milka, cAgus et al. [15]2011Indonesia199/19448.30%---/-12 Bifidobacterium longum BL999, Lactobacillus rhamonosus LPR 3 × 10764nobreast milka, c, dALONA et al. [24]2005America72/7344.10%11313056/570 Bifidobacteria longum infantis, Streptococcus thermophilus, Bifidobacteria bifidus 1.05 × 1096yesbreast milkbHung-Chih et al. [25]2005China180/18749.90%108728104/1000 Bifidobacterium longum infantis, Lactobacillus acidophilus - nobreast milkbJacobs et al. [8]2013Australia548/55152.00%105528359/3770 Bifidobacterium longum infantis BB02, Streptococcus thermophilus (TH–4), Bifidobacterium lactis (BB-12) 1 × 109 nobreast milkbCarole et al. [27]2009France45/4942.60%10852828/350 Bifidobacterium longum BB536, Lactobacillus rhamnosus GG 1 × 1082yesbreast milk, preterm formulab, cBajorek et al. [23]2021America15/1560.00%14683113/100 Bifidobacterium longum infantis EVC001 8 × 1094yesbreast milkc, dJean et al. [11]2008France174/5351.10%34004049/190 Bifidobacterium longum BL999, Lactobacillus paracasei ST11, Lactobacillus rhamnosus LPR 1.29 × 10816no-cJean-Michel et al. [12]2011France40/3948.10%3300393/30 Bifidobacterium longum BL999 2 × 10716nopreterm formulacManzano et al. [13]2017Spain48/5151.50%-full-term (≥37 weeks)8/96 Bifidobacterium longum infantis R0033 3 × 1098nobreast milk, preterm formula cGiuseppe et al. [16]2007Italy42/5553.60%-3915/240 Bifidobacterium longum BL999 2 × 10716nopreterm formula c, dStephane et al. [17]2015France145/5248.70%117029115/390 Bifidobacterium longum, Bifidobacterium lactis 1 × 1093yesbreast milk, preterm formula b, c, d1 The left side of the slash is the intervention group, and the right side is the control group. 2 The dose given was the full dose of all the supplemented strains. 3 a, weight gain; b, necrotizing enterocolitis; c, any adverse events; d, serious adverse events. Basic characteristics of the studies included in the meta-analysis. 1 The left side of the slash is the intervention group, and the right side is the control group. 2 The dose given was the full dose of all the supplemented strains. 3 a, weight gain; b, necrotizing enterocolitis; c, any adverse events; d, serious adverse events. 3.2. Study Quality Following the Cochrane methodology, the risk of bias for the included studies was assessed, and the summary is presented in Figure 2A,B. Among the included trials, eleven were deemed to be at low risk of bias [8,11,12,13,14,15,16,17,25,26,27], primarily due to the availability of comprehensive information regarding the randomization process. In contrast, one trial [23] was classified as being at high risk of bias in the randomization process. The remaining trials had an unclear risk of bias in the randomization process [7,18,24]. Most of the studies demonstrated a low risk of deviations from the intended interventions [7,8,11,12,14,15,16,17,18,24,25,26,27], although two studies [13,23] were found to be at high risk in this regard. All trials were determined to have a low risk of bias related to missing outcome data. However, two trials [7,26] were considered to have an unclear risk of bias due to inconsistent measurements of the outcome. Eight studies were evaluated as having a low risk of bias concerning the selection of reported results [8,11,12,14,15,16,18,27]. Due to inadequate information regarding reporting and other potential sources of bias, most trials were categorized as having an unclear risk in these domains. In total, seven studies [8,11,12,14,15,16,27] were assessed as being at low risk, six studies [7,17,18,24,25,26] at unclear risk, and two studies [13,23] at high risk of bias. For a detailed breakdown of the risk of bias assessment for each included study, please refer to Supplementary Material Table S3. Study quality assessment [7,8,11,12,13,14,15,16,17,18,23,24,25,26,27]. The quality assessment of each included study is summarized in (A) risk of bias summary or is presented as a percentage across all included studies in (B) risk of bias graph. 3.3. Efficacy Outcomes 3.3.1. Weight Gain Our research sought to investigate the potential benefits of Bifidobacterium longum on weight gain in children. Upon analyzing the data extracted from the included studies, we identified five relevant studies [7,14,15,18,26] that presented data for the desired outcome. The pooled estimate for weight gain was 0.029 (95% CI: −0.032, 0.090), indicating a modest effect in favor of weight gain associated with Bifidobacterium longum supplementation. However, it is important to note that a high level of heterogeneity was observed among the studies (I2 = 98%), as depicted in Figure 3. Forest plot illustrating the single study and summary weight gain afforded by Bifidobacterium longum [7,14,15,18,26]. CI, confidence interval; MD, mean difference; SD, standard deviation. 3.3.2. Risk of NEC Six studies [8,17,18,24,25,27] investigated the impact of Bifidobacterium longum on the prevention of NEC. Among these, three studies [8,24,25] reported a significant risk reduction of NEC, while the other three studies [17,18,27] did not find statistically significant differences. It is worth noting that the point estimates from two non-significant studies [18,27] suggested an increased risk of NEC, but the smaller sample sizes led to wide 95% confidence intervals. In contrast, the three studies with larger sample sizes all demonstrated a significant reduction in the risk of NEC. To obtain a more comprehensive estimate of the effect, a pooled analysis was conducted, revealing that Bifidobacterium longum supplementation potentially led to a reduced risk of NEC during the follow-up period (relative risk = 0.539, 95% CI: 0.249, 0.874). Further details and a visual representation of these results can be found in Figure 4. Forest plot comparing the rates of necrotizing enterocolitis for the intervention group versus control group [8,17,18,24,25,27]. CI, confidence interval; RR, risk ratio. 3.4. Safety Outcomes In our safety analysis, we observed that the incidence of adverse events in the Bifidobacterium longum group did not appear to be higher than that in the control group (relative risk = 0.986, 95% CI: 0.843, 1.153), as depicted in Figure 5. Furthermore, there were no statistically significant differences in the occurrence of serious adverse events between the two groups (relative risk = 0.881, 95% CI: 0.493, 1.573), as illustrated in Figure 6. Notably, there was no heterogeneity detected in the analyses of serious adverse events, with a very low I2 value indicating consistency across the included studies. In particular, we conducted sensitivity analyses on studies involving only Bifidobacterium longum and its subspecies as an intervention group, with the results remaining stable (Supplementary Material Figures S1 and S2). Additionally, considering variations in the length of gestation among studies, we stratified the included studies into those with gestation periods less than 37 weeks and those with gestation periods at or above 37 weeks for further analysis. Regarding adverse events, subgroup analyses based on gestation revealed that Bifidobacterium longum supplementation may be associated with a lower risk of adverse events in patients with a gestation period of less than 37 weeks (Supplementary Material Figure S3, relative risk = 0.925, 95% CI: 0.859, 0.997), while the results for patients with a gestation period greater than or equal to 37 weeks were not statistically significant, consistent with the primary analysis (Supplementary Material Figure S4). Concerning serious adverse events, we excluded the study with pregnancy periods of 37 weeks and above [16], and the results remained stable (Supplementary Material Figure S5). Detailed information on adverse events and serious adverse events is provided in Supplementary Material Table S4. Forest plot comparing the rates of any adverse events for intervention group versus control group [11,12,13,14,15,16,17,23,27]. CI, confidence interval; RR, risk ratio. Forest plot comparing the rates of serious adverse events for intervention group versus control group [15,16,17,18,23]. CI, confidence interval; RR, risk ratio. 4. Discussion In our present study, we have undertaken a comprehensive review of the effectiveness and safety profile of Bifidobacterium longum in the context of infant health. Our analysis encompassed a total of 15 randomized controlled trials (RCTs) that adhered to the predefined inclusion criteria, enrolling a combined cohort of 3152 subjects. Our findings have demonstrated that the utilization of Bifidobacterium longum within the intervention group was associated with a notably reduced occurrence of necrotizing enterocolitis, in stark contrast to the control group. However, no statistically significant disparities were observed between the intervention group employing Bifidobacterium longum and the control group with regard to parameters such as weight gain and the prevalence of adverse events. Previous research has advanced the notion that Bifidobacterium longum possesses the remarkable capacity to fully metabolize human milk oligosaccharides (HMOs) present in breast milk. HMOs, constituting the third most abundant solid component in human milk, following lactose and lipids, are impervious to digestion by human enzymes [28]. This metabolic prowess signifies an evolutionary adaptation of Bifidobacterium longum to thrive within the infant intestinal tract [29], whereby its metabolites wield direct influence over the host. Emerging data indicate that metabolites originating from Bifidobacterium longum can profoundly shape the maturation of the immune system [9]. Furthermore, notable reductions in the markers of chronic intestinal inflammation have been observed among preterm infants receiving Bifidobacterium longum subspecies [30]. These observations may offer valuable insights into the prophylactic effects of Bifidobacterium longum against NEC and the diminished occurrence of adverse events, as evidenced in our current meta-analysis. The metabolism of HMOs by Bifidobacterium longum culminates in the production of organic acids, predominantly lactic and acetic acids. These acids, in turn, lower the pH within the intestinal milieu. The acidification of the intestinal environment exerts inhibitory effects on the proliferation of pathogenic bacteria [10]. Thus, the capacity of Bifidobacterium longum to metabolize HMOs not only contributes positively to immune system development but also mitigates the risk of diseases commonly associated with preterm birth and intestinal inflammation, such as NEC. It is noteworthy to highlight the pivotal role of the feeding regimen in this context. Indeed, nutritional support for preterm infants has evolved over the past decade, including the earlier introduction of breastfeeding [31]. Our choice of weight gain as the other primary outcome variable is rooted in the paramount importance of infant growth and development, particularly in preterm infants. These neonates are highly susceptible to suboptimal postnatal growth due to various factors such as substantial insensible fluid losses, gastrointestinal tract immaturity, and heightened metabolic demands [32]. Ongoing research actively explores the role of intestinal microbiota composition in influencing weight gain [26]. The mechanisms through which prebiotic/probiotic combinations can potentially enhance growth encompass several facets. These include the improvement of feeding tolerance, the breakdown of indigestible carbohydrates by probiotic organisms (as illustrated by the HMOs mentioned earlier), and the production of essential vitamins and short-chain fatty acids by probiotic organisms [33,34]. In our present study, we did not observe any statistically significant disparities between the intervention and control groups regarding weight gain. This outcome may be attributed, at least in part, to the limitations stemming from our sample size. Given the pronounced heterogeneity observed in the data, we employed a random effects model to derive pooled estimates. Despite the lack of statistical significance in weight gain differences between the groups, it is crucial to acknowledge the multifaceted nature of infant growth and the potential impact of prebiotic/probiotic combinations on various aspects of development, as suggested by the existing literature. Future studies with larger sample sizes and refined methodologies are warranted to provide a more comprehensive understanding of the complex interplay between intestinal microbiota and infant growth. This heterogeneity may emanate from variations in gestational age, postnatal age, baseline weight, and the duration of follow-up within the study populations. Notably, three of the five studies exhibited relatively shorter follow-up periods and smaller sample sizes [14,18,26]. Conversely, the studies conducted by Agus et al. [15] and Al-Hosni et al. [7] featured relatively large sample sizes and consistently long-term follow-up, lending added credibility to their findings. Nonetheless, the point estimates greater than zero suggest a potentially positive influence of Bifidobacterium longum on weight gain despite the lack of statistical significance. In our safety analysis, it was observed that the intervention group receiving Bifidobacterium longum did not exhibit a different incidence of adverse events of any cause compared to the control group. Moreover, it is noteworthy that there was no statistically significant difference in the occurrence of serious adverse events between these two groups. However, the point estimates of safety analyses suggested a potential for supplementation of Bifidobacterium longum to decrease adverse events risk, where there is a plausible mechanistic link between Bifidobacterium longum and a reduced incidence of adverse events, likely attributable to improved immune function. This meta-analysis presents several limitations. First, it is crucial to highlight that the studies incorporated into our safety analysis were also characterized by relatively short follow-up periods. This temporal limitation may preclude a comprehensive assessment of the long-term safety profile of Bifidobacterium longum. Consequently, the findings should be interpreted in the context of this temporal constraint. Second, due to the restricted number of endpoints focused on in the original literature included in this study, some endpoints that are considered important for infants, such as hospital-acquired infections, long-term growth neurodevelopment, all-cause mortality, and time to full feeds, were not included. Third, the overall quality of the included studies varied, with some studies having an unclear risk of bias and incomplete reporting of outcomes. These factors may introduce bias and affect the reliability of the findings, underscoring the necessity for further research to determine the exact effectiveness and safety of Bifidobacterium longum in infants. Fourth, since previous studies focused less on single Bifidobacterium longum strains, we also included studies with a mixture as an intervention group. Although we performed sensitivity analyses to verify the stability of the results, caution is still warranted regarding the interpretation of the findings. Fifth, it is important to consider the possibility of publication bias when interpreting our results. Notably, some researchers have suggested that probiotic trials are less likely to be published than antibiotic trials [35]. This highlights the need for the reporting and publication of high-quality scientifically valuable study results, whether positive or negative. Sixth, our study included both preterm and term infants, and possible differences in their risk profile, gut maturity, and response to probiotic supplementation make our findings need to be interpreted with more caution, though we used subgroup analyses in an attempt to address this problem. Finally, since the original literature did not provide information on the stage of NEC, we used overall NEC occurrence as an outcome metric, and future studies that explore the effect of Bifidobacterium longum on the risk of NEC incidence at different stages should be more in depth. While the present systematic review and meta-analysis offer valuable insights into the efficacy and safety of Bifidobacterium longum in infants, several areas merit further exploration in future studies. First, to reinforce the current findings and gain a more comprehensive understanding of the long-term effects and safety of Bifidobacterium longum supplementation in infants, there is a need for more robustly designed randomized controlled trials with larger sample sizes and extended follow-up periods. Moreover, additional research efforts should be directed toward investigating the potential mechanisms by which Bifidobacterium longum exerts its effects. This could involve a closer examination of its impact on gut microbiota composition and function, immune system development, and metabolic processes. A deeper grasp of these underlying biological mechanisms can offer valuable insights into potential therapeutic targets. Lastly, it is crucial for future studies to evaluate the cost-effectiveness and feasibility of integrating Bifidobacterium longum supplementation into healthcare practices across diverse settings. This broader perspective will inform decisions regarding the practicality and economic viability of implementing such interventions in various healthcare contexts. 5. Conclusions In conclusion, the systematic review and meta-analysis suggest that supplementing infants with Bifidobacterium longum may reduce the risk of necrotizing enterocolitis. Nevertheless, there was insufficient evidence to support its additional benefits in terms of promoting weight gain. It is crucial to acknowledge the variability in the quality of the studies included in this analysis, emphasizing the need for caution when interpreting the results. Further research is unquestionably needed to validate these findings, delve into the underlying mechanisms driving these effects, and assess the long-term consequences and safety of Bifidobacterium longum supplementation in infants. These forthcoming investigations will contribute to a more comprehensive understanding of the potential advantages and risks associated with this intervention. Acknowledgments The National Center of Technology Innovation for Dairy and Inner Mongolia Dairy Technology Research Institute was acknowledged for sponsoring this project (Project No. 01/CX/YSL/20220530/0031). Supplementary Materials The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/foods12244451/s1, Table S1: Detailed literature search information; Table S2: Detailed description of basic characteristics of the studies included in the meta-analysis; Table S3: Detailed description of the risk of bias assessment for each included study; Table S4: Detailed description of the adverse events for each included study [11,12,13,14,15,16,17,18,23,27]; Figure S1: Forest plot comparing the rates of any adverse events for intervention group versus control group in sensitivity analyses on studies containing only Bifidobacterium longum and its subspecies as an intervention group [12,13,16,23]; Figure S2: Forest plot comparing the rates of severe adverse events for intervention group versus control group in sensitivity analyses on studies containing only Bifidobacterium longum and its subspecies as an intervention group [16,23]; Figure S3: Forest plot comparing the rates of any adverse events for intervention group versus control group in sensitivity analyses among infants below 37-week gestation [14,15,17,23,27]; Figure S4: Forest plot comparing the rates of severe adverse events for intervention group versus control group in sensitivity analyses among infants at or above 37-week gestation [11,12,13,16]; Figure S5: Forest plot comparing the rates of severe adverse events for intervention group versus control group in sensitivity analyses among infants below 37-week gestation [15,17,18,23]. Click here for additional data file. Author Contributions H.G., M.W., T.W. and Y.Z. conceptualization and writing—review and editing. H.G., M.F., T.H., Y.L., S.W., H.P. and X.W. data collection and analyses. H.G. and M.F. writing—original draft preparation. All authors had primary responsibility for the final content. All authors have read and agreed to the published version of the manuscript. Informed Consent Statement Informed consent was obtained from all subjects involved in this study. Data Availability Statement Data are contained within the article and supplementary materials. Conflicts of Interest The authors declare no conflict of interest. Funding Statement This research was supported by the Peking University Outstanding Discipline Construction Project on Epidemiology and Biostatistics, Fujian Provincial Health Technology Project (Grant No. 2020CXB009), the Natural Science Foundation of Fujian Province, China (Grant No. 2021J01352), and the China Postdoctoral Science Foundation (Grant No. BX2021021; Grant No. 2022M710249). Footnotes Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. References Associated Data Supplementary Materials Click here for additional data file. Data Availability Statement Data are contained within the article and supplementary materials.