Limosilactobacillus reuteri DSM 17938 and PTA 6475 as adjunct in Helicobacter pylori therapy: Systematic review and meta-analysis

Abstract

BACKGROUND

Gastrointestinal adverse events and rising antibiotic resistance continue to undermine the effectiveness of Helicobacter pylori (H. pylori) eradication therapy. Recently probiotics have been proposed to improve eradication rates and tolerability, particularly Limosilactobacillus reuteri (L. reuteri) DSM 17938.

AIM

To investigate with a systematic review and meta-analysis the efficacy and safety of L. reuteri DSM 17938, alone or with ATCC PTA 6475, when administered with standard eradication regimens.

METHODS

We systematically searched PubMed, EMBASE, and Cochrane Library through June 1, 2025. We followed the PRISMA 2020 guidelines, and the protocol has been registered on PROSPERO (CRD42024618170). Twelve randomized controlled trials involving 803 participants were included. The primary outcome was H. pylori eradication; secondary outcomes were antibiotic-associated side effects (AASEs).

RESULTS

Supplementation with L. reuteri did not meaningfully increase eradication rates compared with controls [risk ratio (RR) = 1.03; 95% confidence interval (CI): 0.94-1.12; I² = 31.7%], a finding that held across subgroup analyses stratified by age, antibiotic regimen, and treatment duration. Where L. reuteri showed clear benefit was in tolerability: Both DSM 17938 and PTA 6475 strains significantly reduced AASEs, such as diarrhea (RR = 0.34; 95%CI: 0.21-0.55), nausea (RR = 0.44; 95%CI: 0.32-0.60), and dysgeusia (RR = 0.44; 95%CI: 0.24-0.80). For bloating and abdominal pain, the data suggest a possible benefit, though the evidence remains inconclusive.

CONCLUSION

These probiotic strains, L. reuteri DSM 17938 alone or with PTA 6475, do not increase H. pylori eradication rates but provide significant clinically relevant reductions in AASEs. The use of this probiotic therapy may enhance patient adherence and support eradication strategies without increasing antimicrobial pressure.

Key Words: Helicobacter pylori; Limosilactobacillus reuteri; DSM 17938; PTA 6475; Probiotics; Antibiotic-associated side effects; Meta-analysis

Core Tip: Adherence to Helicobacter pylori (H. pylori) eradication therapy is often limited by gastrointestinal side effects. The administration of probiotics, particularly Limosilactobacillus reuteri (L. reuteri), as adjuncts to standard eradication regimens has been shown to improve tolerability. Our meta-analysis demonstrated that L. reuteri supplementation significantly reduced diarrhea, nausea, and dysgeusia during eradication therapy, without improving eradication rates. Benefits were consistent across randomized trials. Precision probiotics, such as L. reuteri DSM 17938, may improve patient comfort, adherence, and treatment continuity by reducing side effects. This probiotic strain may represent a promising strategy in H. pylori management.

INTRODUCTION

Helicobacter pylori (H. pylori) is a gastric pathogen whose virulence determinants, cytotoxin-associated gene A and vacuolating cytotoxin A, may compromise epithelial barrier integrity and alter host immune signaling[1,2]. This mechanism leads to chronic gastritis[1,2]. Over 50% of the global population is infected by this bacterium, with higher prevalence in developing nations and primary infection often occurring during early childhood[3].

The chronic infection by H. pylori is a risk factor for severe clinical outcomes, such as peptic ulcer disease, gastric adenocarcinoma, and mucosa-associated lymphoid tissue lymphoma[4].

Standard eradication protocols include proton pump inhibitors (PPIs) in combination with multiple antibiotics. Gastrointestinal adverse effects and the rising prevalence of antimicrobial resistance often limit the treatment efficacy[5]. Probiotics, live microorganisms that confer health benefits to the host, have gained attention as potential adjunctive therapeutics owing to their bacteriostatic activity, capacity to modulate the intestinal microbiota, and anti-inflammatory functions[6]. Systematic reviews and meta-analyses suggest that probiotic administration may enhance eradication success and decrease treatment-associated side effects[7]. Within this context, Limosilactobacillus reuteri (L. reuteri) (reclassified from Lactobacillus in 2020)[8] has proven particular beneficial. The strains DSM 17938 and ATCC PTA 6475 have been investigated individually and in combination. DSM 17938 is a plasmid-cured derivative of ATCC 55730, devoid of antibiotic resistance genes while retaining the probiotic phenotype[9]; PTA 6475 exhibits immunomodulatory properties mediated through nuclear factor kappa-B (NF-κB) pathway regulation[10].

This systematic review and meta-analysis evaluate the probiotic strains L. reuteri DSM 17938 alone or with PTA 6475 as an adjunct to conventional H. pylori eradication therapy. Differently from previous studies that aggregated data across heterogeneous probiotic taxa, we present strain-specific estimates of efficacy and antibiotic-associated side effects (AASEs), in order to inform clinical decision-making and guide future research.

MATERIALS AND METHODS

Registration and reporting guidelines

The protocol for this systematic review and meta-analysis was prospectively registered with PROSPERO (International Prospective Register of Systematic Reviews; registration No. CRD42024618170) in accordance with preferred reporting items for systematic reviews and meta-analyses 2020 guidelines[11], ensuring methodological transparency and adherence to prespecified analytical plans.

Literature search strategy

The electronic search was conducted in PubMed, EMBASE, and the Cochrane Library from inception to June 1^st, 2025. No language restrictions were applied. Medical Subject Headings (MeSH) terms and keywords related to “Helicobacter pylori”, “Limosilactobacillus reuteri”, and “probiotics” were used for the search. No grey-literature sources (e.g., conference abstracts, trial registries) were included. The keywords included: “Helicobacter”, “Helicobacter pylori”, “Campylobacter pylori”, “H. pylori”, “Hp”, “Lactobacillus reuteri”, “L. reuteri”, “Limosilactobacillus reuteri”, “probiotics”, “probiotic”. Both MeSH and free-text terms were utilized without any restrictions on language. Additionally, the reference lists of all relevant articles were manually screened to identify further eligible studies. The full, reproducible search strategies for each database (PubMed, EMBASE, and Cochrane Central), including final search date and terms, are provided in Supplementary material [full search strings strategy utilized for each database for the selection of randomized controlled trials (RCTs)].

Eligibility criteria

Studies were selected according to a priori defined inclusion and exclusion criteria. Only RCTs were eligible, to maintain methodological rigor and minimize bias. Participants of any age undergoing eradication treatment for H. pylori infection were considered. At least one validated method was required to confirm H. pylori diagnosis: Urea breath test (UBT), rapid urease test, histology, stool antigen test (SAT), or bacterial culture. We included trials evaluating L. reuteri DSM 17938 or its parental strain ATCC 55730, either as a monotherapy or in combination with ATCC PTA 6475. We initially pooled data to generate overall effect estimates and then a subgroup analysis comparing DSM 17938 alone vs the combined formulation was performed to evaluate consistency.

In all eligible studies, L. reuteri DSM 17938 (alone or in combination with PTA 6475) was initiated on the same day as standard antibiotic therapy. Studies administering the probiotic after antibiotic completion were excluded. Within this review, “standard H. pylori eradication therapy” comprehend any first- or second-line regimen approved by major international guidelines, such as Maastricht VI/Florence Consensus Report, including: (1) Standard triple therapy (PPI plus amoxicillin plus clarithromycin or metronidazole); (2) Sequential therapy (PPI plus amoxicillin for 5-7 days, followed by PPI plus clarithromycin and tinidazole or metronidazole for 5-7 days); (3) Bismuth-containing quadruple therapy; (4) Non-bismuth quadruple therapy; (5) Fluoroquinolone-containing triple or quadruple therapy; and (6) Other approved second-line or rescue regimens.

Outcomes

H. pylori eradication rate was the primary outcome and it was defined as a negative result obtained at least 4 weeks after completion of therapy. We accepted only validated diagnostic tests used in the original trials as confirmatory methods: UBT, SAT, or endoscopic biopsy with histology or culture. All included studies utilized at least one of these validated modalities. Secondary outcomes comprised the incidence of AASEs. These included gastrointestinal symptoms (e.g., diarrhea, nausea, vomiting, abdominal pain, bloating) and other adverse effects (e.g., headache, allergic reactions, or other patient-reported symptoms).

Study selection process

The screening of records (titles, abstracts, and full texts) was conducted individually by two authors (Dargenio VN and Cristofori F), with disagreements resolved by discussion with a third reviewer (Francavilla R). We excluded studies that didn’t focus on L. reuteri DSM 17938 (or its parental strain ATCC 55730), weren’t RCTs, didn’t provide full texts or usable data, or were grey literature, animal studies, or duplicates.

Data extraction and management

The extraction of data was made by two independent investigators (Cristofori F and Barone M) with a standardized form. Extracted information included: First author’s name, year of publication, country of study, number and age of participants, diagnostic methods used for H. pylori, intervention and control group protocols, relevant clinical outcomes.

Risk of bias assessment and quality of evidence

Two investigators (Brindicci VF and Cristofori F) independently assessed risk of bias and disagreements were resolved by discussion. Cochrane Risk of Bias Tool has been utilized, evaluating the following domains: (1) Random sequence generation; (2) Allocation concealment; (3) Blinding of participants and personnel; (4) Blinding of outcome assessment; (5) Incomplete outcome data; (6) Selective outcome reporting; and (7) Other sources of potential bias. Each domain was rated as having low, high, or unclear risk of bias.

The overall quality of the evidence for each outcome was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework (https://gdt.gradepro.org/app/handbook/handbook.html), providing a transparent and structured approach to rating the certainty of evidence[12].

Statistical analysis

Risk ratios (RR) with 95% confidence intervals (95%CI) were calculated for dichotomous outcomes and pooled estimates were derived using a DerSimonian-Laird random-effects model[13]. Heterogeneity was estimated using the Cochran Q (χ²) and I² statistics, with I² thresholds of 25%, 50%, and 75% corresponding to low, moderate, and high heterogeneity, respectively[14].

Subgroup analyses were performed by patient age (pediatric/adult), antibiotic regimen (standard triple, levofloxacin-based triple, quadruple, sequential), and treatment duration (7 days, 10 days, or 14 days). Subgroup estimates were computed using the Mantel-Haenszel method, with I² reported for each subgroup. Publication bias was planned to be assessed with funnel plots if ≥ 10 studies were available[15]. A subgroup comparison between single-strain (DSM 17938 or its parental strain ATCC 55730) and multi-strain (DSM 17938 + PTA 6475) formulations was conducted using per-protocol data pooled across trials.

Sensitivity analyses, including exclusions of high-risk-of-bias studies and those with substantial clinical heterogeneity, were performed to evaluate result robustness. In addition, a leave-one-out analysis was performed to further evaluate the influence of individual studies on the overall estimates[16]. All analyses used Stata 14.0 and RevMan 5.3.

RESULTS

Study selection

The initial literature search retrieved a total of 2423 records. After the removal of 989 duplicates, 1434 unique articles were screened by title and abstract. Of these, 1361 were excluded at this stage. Subsequently, 73 full-text articles were assessed for eligibility, resulting in the exclusion of 61 studies. Ultimately, 12 RCTs were included in the meta-analysis[17-28]. Of these, seven studies investigated L. reuteri DSM 17938 in combination with the strain PTA 6475[17-20,22,23,26], while five studies examined DSM 17938 alone (or its parental strain ATCC 55730)[21,24,25,27,28]. The full selection process is illustrated in Figure 1.

Figure 1 Preferred reporting items for systematic reviews and meta-analyses 2020 flow diagram illustrating the study selection process. Twelve studies were included in both the qualitative and quantitative synthesis. RCT: Randomized controlled trial; H. pylori: Helicobacter pylori; L. reuteri: Limosilactobacillus reuteri.

Study characteristics

The 12 included RCTs (2006-2022) enrolled 803 intention-to-treat and 728 per-protocol participants. The study by Poonyam et al[26] randomized the 100 included patients into four and not two treatment arms: 7-day plus probiotics (n = 25), 7-day plus placebo (n = 25), 14-day plus probiotics (n = 25), and 14-day plus placebo (n = 25). For these reasons, we decided to split the four groups included in the single RCT by Poonyam et al[26] into two different comparisons, according to treatment duration: 7-day plus probiotics vs 7-day plus placebo indicated as “Poonyam et al[26]-a”, and 14-day plus probiotics vs 14-day plus placebo indicated as “Poonyam et al[26]-b”. Thus, although 12 RCTs were included, the meta-analysis was conducted on a total of 13 independent comparisons. Geographically, 9 studies were conducted in Europe[17-19,21-25,28], 1 in Egypt[20], and 2 in Asia[26,27]. Sample sizes ranged from 33 to 100 participants. Therapeutic regimens comprised bismuth-containing quadruple therapy (4 studies)[17-19,26], standard triple therapy (omeprazole/amoxicillin/clarithromycin; 4 studies)[20,22,27,28], levofloxacin-based triple therapy (1 study)[25], and sequential therapy (3 studies)[21,23,24]. Treatment duration varied: 7 days[22,25,26,28], 10 days[17-19,21,23,24], and 14 days[20,26,27]. Probiotic administration overlapped with antibiotics in all studies; its duration either matched the antibiotic course (7-14 days)[19,26,28] or extended beyond it (up to 30 days)[17,18,20-25,27]. Study characteristics are detailed in Table 1.

Table 1 Characteristics of the randomized controlled trials included in the meta-analysis.

Ref.	Country	Number of patients ITT and PP (Exp/Con)	Age (years)	Eradication regimen	Reuteri regimen	Diagnostic methods for H. pylori infection
						Diagnosis	Eradication
Moreno Márquez et al[17]	Spain	80 (40/40); 68 (35/33)	18-65	For 10 days pylera plus omeprazole 2-times a day	For 30 days, L. reuteri DSM 17938 and ATCC PTA 6475 tablets	Histology or UBT or stool antigen test	UBT
Dore et al[18]	Italy	46 (23/23); 40 (21/19)	Over 18	For 10 days tetracycline and metronidazole plus rabeprazole. One regimen was supplemented with pylera, 2 cps bis in die	For 27 days, L. reuteri DSM 17938 and ATCC PTA 6475 tablets, twice daily	UBT and/or stool antigen test	UBT and/or stool antigen test
Dore et al[19]	Italy	99 (49/50); 92 (46/46)	Over 18	For 10 days tetracycline and metronidazole plus pantoprazole. One regimen was supplemented with pylera, 2 cps bis in die	For 10 days, L. reuteri DSM 17938 and ATCC PTA 6475 tablets, twice daily	Histology and/or UBT or stool antigen test	UBT or stool antigen test
Emara et al[20]	Egypt	70 (35/35); 70 (35/35)	18-60	For 14 days triple therapy	For 28 days, L. reuteri DSM 17938 and ATCC PTA 6475 tablets	Histology, rapid urease test and stool antigen test	Histology, rapid urease test and stool antigen test
Francavilla et al[21]	Italy	40 (20/20); 40 (20/20)	35-68	Sequential therapy	For 28 days, L. reuteri ATCC 55730 tablets, once daily	Histology, rapid urease test, UBT, stool antigen test	UBT
Francavilla et al[22]	Italy	100 (50/50); 86 (43/43)	Over 18	For 7 days triple therapy	For 96 days, L. reuteri DSM 17938 and ATCC PTA 6475 tablets, once daily	Histology, rapid urease test or UBT	UBT
Kotzev et al[23]	Bulgaria	55 (26/29); 20 (11/9)	18-70	Sequential therapy	For 28 days, L. reuteri DSM 17938 and ATCC PTA 6475 tablets, once daily	Histology, rapid urease test, UBT, stool antigen test	UBT, stool antigen test
Lionetti et al[24]	Italy	40 (20/20); 40 (20/20)	3-18	Sequential therapy	For 20 days, L. reuteri ATCC 55730 tablets, once daily	Histology, rapid urease test, UBT	UBT
Ojetti et al[25]	Italy	90 (45/45); 90 (45/45)	18-65	For 7 days triple therapy1	For 14 days, L. reuteri ATCC 55730 tablets, one tablet three times a day	UBT	UBT
Poonyam et al[26]-a	Thailand	50 (25/25); 50 (25/25)	Over 18	For 7 days, bismuth subsalicylate twice daily, metronidazole 3-times a day, tetracycline 4-times a day, and dexlansoprazole 60 mg twice daily	For 7 days, L. reuteri DSM 17938 and ATCC PTA 6475 tablets, twice daily	Histology, rapid urease test	UBT
Poonyam et al[26]-b	Thailand	50 (25/25); 50 (25/25)	Over 18	For 14 days bismuth subsalicylate twice daily, metronidazole 3-times a day, tetracycline 4-times a day, and dexlansoprazole 60 mg twice daily	For 14 days, L. reuteri DSM 17938 and ATCC PTA 6475 tablets, twice daily	Histology, rapid urease test	UBT
Shahraki et al[27]	Iran	50 (25/25); 50 (25/25)	5-14	For 14 days triple therapy with 30 days omeprazole	For 28 days, L. reuteri DSM 17938 tablets, once daily	Histology	UBT
Scaccianoce et al[28]	Italy	33 (17/16); 32 (17/15)	Over 18	For 7 days triple therapy	For 7 days L. reuteri ATCC 55730 tablets, twice daily	Histology	UBT

¹Levofloxacin-based triple therapy: Proton pump inhibitor, levofloxacin, and amoxicillin. Standard triple therapy: Proton pump inhibitor, clarithromycin, and amoxicillin.

The table summarizes the country of origin, sample size (intention-to-treat and per-protocol populations), patient age range, eradication regimens, Limosilactobacillus reuteri supplementation protocols, and the diagnostic methods used for the detection of Helicobacter pylori at baseline and for confirming eradication. Different strains of Limosilactobacillus reuteri (DSM 17938, ATCC PTA 6475, and ATCC 55730) and various durations of supplementation were employed across studies. Triple therapy: Proton pump inhibitors-clarithromycin-amoxicillin. Sequential therapy: Proton pump inhibitors plus amoxicillin for 5 days, proton pump inhibitors plus clarithromycin plus tinidazole for the next 5 days. ITT: Intention to treat; PP: Per protocol; Exp: Experimental group; Con: Control group; H. pylori: Helicobacter pylori; L. reuteri: Limosilactobacillus reuteri; UBT: Urea breath test.

Risk of bias assessment

Five of the 13 comparisons had low risk of bias; eight had unclear risk. Blinding was assessed as high risk in five studies and unclear in two due to inadequate reporting. All trials demonstrated low risk of bias for incomplete outcome data and selective reporting. No publication bias was detected. The risk of bias assessment is detailed in Figure 2; publication bias is summarized in Supplementary Figure 1. Evidence certainty was evaluated using the GRADE approach[12].

Figure 2 Risk of bias assessment for the randomized controlled trials included in the meta-analysis. A: Summary of review authors’ judgments across seven domains of the Cochrane Risk of Bias tool for each included study. Green circles indicate low risk of bias, yellow circles indicate unclear risk, and red circles indicate high risk; B: Proportion of studies rated as low, unclear, or high risk of bias for each domain. Most studies showed a low risk of bias in domains related to outcome data and selective reporting. Higher or unclear risk was observed primarily in domains related to blinding and allocation concealment.

Primary outcome: H. pylori eradication rate

The analysis includes 13 comparisons derived from 12 RCTs with a total of 368 patients in the L. reuteri plus antibiotics group and 360 patients in the antibiotics-only group. The pooled RR for eradication of H. pylori was 1.03 with a 95%CI of 0.94 to 1.12, indicating no statistically significant difference between the two treatment groups. The overall event rates were 80.4% in the L. reuteri plus antibiotics group and 76.9% in the antibiotics-only group. Moderate heterogeneity was observed across studies (I² = 31.7%, P = 0.129), justifying the use of a random-effects model. Most studies showed no significant benefit of adding L. reuteri to antibiotic therapy. Meta-analysis showed that supplementation with L. reuteri DSM 17938, either alone or in combination with PTA 6475, did not significantly improve eradication rates compared to standard eradication therapy alone (Figure 3A). Although only moderate heterogeneity was found across studies included in the analysis we assessed the consistency of our results by the leave-one-out sensitivity analysis, i.e., excluding the selected trials one at a time from the analysis. This sensitivity analysis confirmed the consistency of our results.

Figure 3 Forest plot. A: Forest plot of randomized controlled trials evaluating Helicobacter pylori (H. pylori) eradication rates in patients treated with standard antibiotic therapy with or without Limosilactobacillus reuteri (L. reuteri) supplementation. The analysis includes 13 comparisons derived from 12 randomized controlled trials with a total of 368 patients in the L. reuteri plus antibiotics group and 360 patients in the antibiotics-only group; B: Forest plot of randomized controlled trials evaluating the incidence of diarrhea in patients undergoing H. pylori eradication therapy with or without L. reuteri supplementation. The analysis includes six comparisons, with one study (Poonyam et al[26]-b) excluded due to zero events in both groups. The pooled risk ratio for diarrhea was 0.34 (95% confidence interval: 0.21-0.55), indicating a statistically significant reduction in risk associated with L. reuteri supplementation. Heterogeneity across studies was negligible (I² = 0.0%; P = 0.752), supporting the consistency of this protective effect. RR: Risk ratio; CI: Confidence interval; L. reuteri: Limosilactobacillus reuteri.

Subgroup analysis

Subgroup analyses were conducted by population, antibiotic regimen, and treatment duration to explore heterogeneity and assess consistency (Supplementary Table 1). In adults (11 comparisons from 10 RCTs)[17-23,25,26,28], the pooled RR was 1.02 (95%CI: 0.92-1.12; I² = 38.8%, P = 0.090). The pediatric subgroup (2 studies)[24,27] showed an RR of 1.11 (95%CI: 0.92-1.35; I² = 0.0%, P = 0.665) (Supplementary Figure 2).

Stratified by regimen, bismuth-based quadruple therapy (5 comparisons from 4 RCTs)[17-19,26] yielded an RR of 0.94 (95%CI: 0.86-1.03; I² = 9.8%, P = 0.350). Standard triple therapy (4 studies)[20,22,27,28] gave an RR of 1.11 (95%CI: 0.95-1.30; I² = 0.0%), and sequential therapy (3 studies)[21,23,24] an RR of 1.12 (95%CI: 0.93-1.35; I² = 0.0%). Only levofloxacin-based triple therapy (1 study)[25] showed a significant effect: RR of 1.33 (95%CI: 1.01-1.76) (Supplementary Figure 3).

By treatment duration, 10-day regimens (6 studies)[17-19,21,23,24] produced an RR of 0.94 (95%CI: 0.85-1.04; I² = 16.8%, P = 0.305). Fourteen-day regimens (3 comparisons from 3 RCTs, including the 14-day arm of Poonyam et al[26])[20,26,27] gave an RR of 1.11 (95%CI: 0.98-1.26; I² = 0.0%, P = 0.914), and 7-day regimens (4 comparisons from 4 RCTs, including the 7-day arm of Poonyam et al[26])[22,25,26,28] an RR of 1.10 (95%CI: 0.92-1.33; I² = 22.7%, P = 0.275) (Supplementary Figure 4).

Secondary outcomes: AASEs

Meta-analysis indicated a statistically significant reduction in AASEs with L. reuteri co-administration. Five studies[20,22,25,26,28] reported diarrhea outcomes; “Poonyam et al[26]-b” was excluded due to zero events in both arms. The pooled analysis of five comparisons[20,22,25,26,28] showed a significant reduction in diarrhea risk (RR = 0.34, 95%CI: 0.21-0.55; P < 0.001), corresponding to a 66% relative risk reduction. Heterogeneity was negligible (I² = 0.0%, P = 0.752). The number needed to treat (NNT) was 4.8 (Figure 3B).

Six RCTs[19,20,22,25,26,28] reported dysgeusia, yielding seven comparisons in the analysis. Pooled analysis demonstrated a significant reduction in risk (RR = 0.44, 95%CI: 0.24-0.80; P < 0.01), reflecting a 56% relative reduction. Heterogeneity was low (I² = 14.4%, P = 0.320). The NNT for dysgeusia was 6.4 (Supplementary Figure 5).

Three RCTs[22,25,26] evaluated nausea, yielding four comparisons in the analysis. The pooled analysis demonstrated a significant risk reduction with L. reuteri (RR = 0.44, 95%CI: 0.32-0.60; P < 0.01), indicating a 56% relative reduction. Heterogeneity was absent (I² = 0.0%, P = 0.679) (Supplementary Figure 6).

Six RCTs[19-22,24,25] reported bloating. The pooled RR was 0.47 (95%CI: 0.18-1.23; P = 0.12), suggesting a nonsignificant trend favoring L. reuteri (Supplementary Figure 7). A comprehensive summary of AASEs is presented in Table 2. Overall, L. reuteri showed a modest protective effect against several gastrointestinal AASEs, though most outcomes lacked precision due to limited sample sizes and wide CIs.

Table 2 Gastrointestinal side effects associated with Helicobacter pylori eradication therapy: Comparison between groups treated with and without Limosilactobacillus reuteri supplementation.

Adverse event	RR (95%CI)	I2 (%)	P value for heterogeneity	Comment
Vomiting	0.94 (0.60-1.47)	0.0	0.652	No difference between groups
Epigastric pain	0.51 (0.22-1.19)	32.4	0.228	Trend in favor of L. reuteri
Constipation	0.69 (0.32-1.52)	0.0	0.639	No difference between groups
Abdominal pain	0.73 (0.41-1.33)	27.3	0.220	Consistent but non-significant benefit
Asthenia	0.58 (0.18-1.85)	33.1	0.225	Weak trend in favor of L. reuteri
Vertigo	1.00 (0.26-3.78)	0.0	1.000	No difference between groups
Appetite loss	0.50 (0.05-5.31)			Assessed in a single study
Dysgeusia	0.44 (0.24-0.80)	14.4	0.320	56% RR reduction; consistent across 7 comparisons from 6 RCTs
Nausea	0.44 (0.32-0.60)	0.0	0.679	56% RR; high consistency across 4 comparisons from 3 RCTs
Diarrhea	0.34 (0.21-0.55)	0.0	0.752	66% RR reduction; consistent across 5 RCTs
Bloating	0.47 (0.18-1.23)	67.8	0.008	Trend in favor of L. reuteri

Reported are risk ratios with 95% confidence intervals and heterogeneity statistics (I² and P values for heterogeneity). RR: Risk ratio; CI: Confidence interval; RCT: Randomized controlled trial; L. reuteri: Limosilactobacillus reuteri.

Subgroup analyses (population, regimen, duration) revealed consistent trends, with statistically significant reductions in nausea (by antibiotic type), diarrhea (by antibiotic type), abdominal pain (by population), and meteorism (by duration) (Supplementary Table 2).

We found similar eradication rates (in per-protocol patients) in the comparison between single-strain (DSM 17938) and multi-strain (DSM 17938 + PTA 6475) formulations (RR = 1.00, 95%CI: 0.90-1.11). However, the singlestrain subgroup exhibited significantly higher risks for diarrhea (RR = 3.04, 95%CI: 1.16-7.96), nausea (RR = 5.61, 95%CI: 2.55-12.36), and bloating (RR = 3.54, 95%CI: 1.54-8.17). There was no difference in dysgeusia between subgroups (Supplementary Table 3). GRADE assessments are presented in Supplementary Table 4.

DISCUSSION

This meta-analysis evaluates L. reuteri DSM 17938 (or its parental strain ATCC 55730), alone or with ATCC PTA 6475, as an adjunct to H. pylori eradication therapy. The decision to focus on studies evaluating L. reuteri DSM 17938 addresses the need to contain the biological heterogeneity that, in previous probiotic meta-analyses, has repeatedly compromised the generalizability of pooled estimates. L. reuteri supplementation did not improve H. pylori eradication rates, and this finding proved consistent across subgroup analyses stratified by antibiotic regimen, treatment duration, and patient population. The single regimen demonstrating benefit, a levofloxacin-based triple therapy, was derived from a small isolated study[25] and has not been independently replicated. For this reason, this finding must be interpreted with caution rather than as an established effect. Furthermore, eradication outcomes did not differ between DSM 17938 monotherapy and its combination with PTA 6475. This reinforces the conclusion that bacterial clearance is not substantially influenced by this probiotic.

These results are similar to the findings of previous meta-analyses, which have reported modest or inconsistent effects of probiotics on eradication success[29,30]. The conclusion is further supported by monotherapy studies in which L. reuteri administered with PPIs, but without antibiotics, achieved lower eradication rates[31]. This confirms that the strain lacks intrinsic bactericidal capacity sufficient to eradicate H. pylori.

In contrast to its limited impact on eradication efficacy, DSM 17938 demonstrated considerable benefits in the tolerability of the treatment. Diarrhea, nausea, and dysgeusia are among the most common reasons patients abandon eradication therapy prematurely. Across multiple randomized trials, supplementation significantly reduced the incidence of these three adverse effects. These reductions were consistent (for example, RR was 0.34 for diarrhea and 0.44 for dysgeusia) and led to favorable numbers needed to treat (NNT of 4.8 and 6.4), showing a clear clinical benefit. Although effects on bloating did not achieve statistical significance, the direction of the data leaned toward DSM 17938 benefit and heterogeneity across studies reflected differences in outcome definitions and population characteristics. For other less frequent adverse events, such as vomiting, constipation, vertigo, and appetite loss, the available evidence was too limited to reach firm conclusions.

These tolerability findings are clinically relevant. Gastrointestinal adverse events are a major cause of premature discontinuation of eradication therapy, directly linked to treatment failure. Thus, while DSM 17938 does not change the odds of eradication, it augments the overall therapeutic experience of treatment. This is aligned with the Maastricht VI/Florence consensus emphasis on strategies that support adherence and reduce treatment burden[32].

When we compared DSM 17938 alone with its combination with PTA 6475, we found that eradication rates were similar, but AASEs were lower with the multi-strain formulation. Given the known immunomodulatory activity of PTA 6475, particularly NF-κB pathway regulation, synergistic effects on gut homeostasis cannot be ruled out. However, this conclusion is based on indirect, cross-study comparison rather than randomized head-to-head evidence, and it must be considered hypothesis-generating. Future RCTs directly comparing single-strain vs combined formulations are needed to clarify whether PTA 6475 adds clinical advantages.

Several mechanisms help explain the clinical benefits of L. reuteri. The probiotic directly inhibits pathogens through reuterin and antimicrobial-containing microvesicles. Intestinal barrier integrity is preserved through stabilization of tight junction proteins, which mitigates permeability increases induced by pathogens. L. reuteri modulates the immune response by reducing inflammation (lowering cytokines such as IL-1β, TNF-α, and IFN-γ) and promoting regulatory T cells. It also supports immune cell movement through CCR7 expression. L. reuteri 17938 also exerts neuroactive effects on enteric neurons, modulating gut motility and reducing visceral pain via transient receptor potential vanilloid 1 channel blockade. Importantly, microvesicles alone can reproduce many of these effects, including motility modulation and anti-inflammatory activity[33].

Probiotic supplementation during H. pylori eradication therapy may be cost-effective by improving clinical outcomes. Meta-analyses show that adding probiotics can increase eradication rates, which may reduce the need for more expensive second-line treatments[7]. In particular, they also appear to substantially decrease AASEs[7]. Real-world registry data covering over 36000 treatments confirm higher effectiveness (odds ratio = 1.63) and fewer severe events[34]. Although introducing probiotics may increase initial costs, downstream savings from improved efficacy and safety make probiotics a cost-effective strategy.

Our results differ from those of Li et al[35], who reported a small but statistically significant improvement in eradication rates with L. reuteri supplementation, probably due to differences in both the strains analyzed and the study populations included. They analyzed multiple L. reuteri strains with distinct mechanisms, especially DSM 17648, while we focused only on DSM 17938. Both approaches are valid depending on the clinical question, but the different results highlight the importance of considering specific strains when interpreting probiotic research.

This study has several limitations that should be acknowledged. Although the total sample size (803 participants) represents the most comprehensive strain-specific synthesis to date, the analysis may not be powerful enough to detect small but clinically meaningful improvements in eradication rates, or to assess rare AASEs. Differences across studies, such as variations in the types of antibiotics used, treatment duration, and how probiotics were administered, may hide important effects in specific subgroups and reduce how broadly the findings can be applied. Furthermore, the methodological quality of the included studies was variable with high or unclear risk of bias in several cases.

This limitation is especially relevant for subjective endpoints, including gastrointestinal symptoms, which are more susceptible to bias. The results should be interpreted considering the limitations of self-reported adverse events, their possible impact on quality of life and the potential for publication bias, despite statistical tests denoting a low risk.

Finally, some key biological factors, such as antibiotic resistance patterns, the composition of the gut microbiome at baseline, and host genetics, could not be evaluated but they may play an important role in how well probiotics work. Future research should take these factors into account, for example by profiling the gut microbiome and examining genetic variations, in order to identify which patients are most likely to benefit from L. reuteri supplementation. At the same time, high-quality, well-powered RCTs focused on specific strains are needed, including direct comparisons between DSM 17938 and DSM 17938 + PTA 6475, to determine the best probiotic strategy and improve the precision-based treatment of H. pylori infection.

CONCLUSION

In conclusion, although L. reuteri DSM 17938, alone or with PTA 6475, does not improve H. pylori eradication rates, it consistently reduces the main gastrointestinal side effects caused by eradication therapy. This better tolerability supports its use as an adjunctive treatment to improve patient comfort and adherence, rather than as a way to increase eradication efficacy.

Further high-quality, well-powered, strain-specific RCTs, including direct comparisons between DSM 17938 and DSM 17938 + PTA 6475, are essential to determine the best use of probiotics in the management of H. pylori infection and to support more personalized treatment approaches.