Received 19 October 2015; accepted 29 November 2015; published 2 December 2015
1. Introduction
Helicobacter pylori is the causative agent of chronic gastritis, peptic ulcer, gastric adenocarcinoma, and lymphoma [1] -[3] . The most commonly prescribed triple therapy, composed of proton-pump inhibitor, clarithromycin or levofloxacin, and amoxicillin or tinidazole, remains the first-choice treatment for H. pylori infection. However, in recent years, we have witnessed decreasing eradication rates with this standard triple therapy, mainly due to increased resistance to antibiotics [4] . Moreover, about 5% - 30% of patients receiving triple therapy are reported to experience antibiotic-associated side effects, which leads to low compliance and further treatment failure [5] . Strategies targeted to raise eradication rates and improve the treatment tolerability are increasingly needed.
In 2012, the Maastricht IV/Florence Consensus Report stated that certain probiotics as adjuvant treatment could exert beneficial effects on the management of H. pylori [6] . A recent meta-analysis [7] also evaluated the effects of supplementation with probiotics on H. pylori eradication rates and side effects of anti-H. pylori treatment. Thirty-three randomized controlled trials (RCTs) of varying methodological quality involving a total of 4459 patients were identified. These pooled results indicated that probiotic supplementation, when compared with eradication therapy, could increase eradication rates, particularly when antibiotic therapies are relatively ineffective, and reduce therapy-related overall side effects. However, the beneficial effects of probiotics seem to be strain specific, and pooling data on various strains may lead to spurious conclusions. A more appropriate way is to conduct a meta-analysis that evaluates the effects of a clearly defined probiotic microorganism.
Several specific probiotics have been shown to lower H. pylori bacterial load and gastric inflammation and reduce antibiotic-associated side effects [8] . Our study focused on a single probiotic microorganism, i.e., Lactobacillus GG, to evaluate the probable effects of Lactobacillus GG given concomitantly with standard eradication therapy on major clinical outcomes related to H. pylori eradication and therapy-associated side effects.
2. Methods
This systematic review and meta-analysis was conducted and reported in adherence to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement (PRISMA) [9] , and the guidelines of the Cochrane Handbook for Systematic Reviews of Interventions [10] .
2.1. Literature Search and Selection Criteria
PubMed, EMBASE and the Cochrane Central Register of Controlled Trials (CENTRAL) were systematically searched. We used the following key words: probiotic, probiotics, Lactobacillus rhamnosus GG, Lactobacillus GG, LGG, Helicobacter pylori, and H. pylori. The language was restricted to English. The last search was conducted in August 4, 2015. The detailed search strategy is shown in Table 1. The cited references of retrieved articles and previous reviews were also manually checked to identify any additional eligible studies. All citations were imported into a bibliographic database (EndNote X7; Thomson Reuters) for assessment of eligibility. Two of authors (G-QZ, H-JH) independently carried out literature search, eligibility evaluation, data extraction and quality assessment. Discrepancies between authors were resolved by consensus.
We developed a PICOS (Patient, Intervention, Comparators, Outcome, and Study design) approach as the
eligibility criteria: 1) Population: H. pylori-infected subjects of any age, diagnosed by generally accepted methods (i.e., histopathology, the 13C-urea breath test (UBT), the rapid urease test, or measurement of H. pylori IgG antibody); 2) Intervention: Lactobacillus GG supplementation given during standard triple therapy; 3) Comparators: standard triple therapy without Lactobacillus GG; 4) Outcome: the primary outcome was incidence of adverse effects. The adverse effects of interest were any common gastrointestinal adverse effects that occurred during standard triple therapy, including diarrhea, epigastric pain, nausea and/or vomiting, abdominal bloating, flatus, taste disturbance, loss of appetite, constipation, and the need for discontinuation of the anti-H. pylori therapy; the secondary outcome was H. pylori eradication rate, confirmed by a negative 13C-UBT or other generally accepted method; 5) Study design: only randomized controlled trials were eligible.
2.2. Data Extraction and Quality Assessment
Two of authors (G-QZ, H-JH) independently extracted relevant data from each included study by using a unified data form. Extracted data were entered into a standardized Excel file. The items included in the data form were as follows: source (first author, year of publication, country), number of patients enrolled, eradication regimen, Lactobacillus GG (dose, duration), the comparator group, primary and secondary outcomes reported by the authors, follow-up, and methods of checking H. pylori status and side effects. The Cochrane Risk of Bias Tool was adopted to assess the risk of bias for each RCT [11] .
2.3. Statistical Analysis
To evaluate the effects of probiotics, we calculated relative risks (RRs) with its 95% confidence intervals (CIs) for the outcomes of interest between intervention and control groups. Heterogeneity across studies was tested by using the I2 statistic [12] . Studies with an I2 value of greater than 50% were considered to have significant heterogeneity. The random effects model was used to calculate pooled RRs and its 95% CIs if significant heterogeneity existed. Otherwise, the fixed effects model was applied to calculate the pooled RRs. Funnel plot was carried out to investigate publication bias of all the included studies. However, the publication bias was not formally assessed, due to the small number of studies (<10) included in the analyses of the primary and secondary outcome measures. All statistical analyses were performed with RevMan 5.3 (The Nordic Cochrane Centre, Copenhagen, Denmark).
3. Results
A total of 193 records were identified by the initial database search. Twenty-eight records were excluded for duplicates, and 159 records were excluded based on the titles and abstracts for various reasons (reviews, letters, experimental studies or irrelevant to our study). The remaining six full-text articles were assessed for eligibility, and two of them were further excluded because of non-English [13] [14] . Finally, four trials met the inclusion criteria [15] - [18] . The selection process is shown in Figure 1.
3.1. Characteristics of Included Studies
These trials randomized a total of 305 patients, of which 288 were followed up. The main characteristics of included trials are summarized in Table 2. Three studies enrolled only adults [15] - [17] , and one RCT [18] (n = 83) was undertaken exclusively in children (age range: 5 - 17 years). In all four included trials, Lactobacillus GG was given along with standard triple therapy composed of a proton pump inhibitor and two antibiotics. The daily doses of Lactobacillus GG ranged from 1 × 109 CFU (b.i.d) [18] to 6 × 109 CFU (b.i.d) [15] -[17] . Three RCTs were placebo controlled [16] -[18] , and there was no additional intervention in the control group in the other one trial [15] . The included studies were conducted in Italy (three RCTs) [15] -[17] and Egypt (one RCT) [18] . Figure 2 shows the results of the methodological quality assessment. The allocation concealment was unclear in one RCT [17] , and another trial [15] was at high risk of binding. The remaining two trials had good methodological quality.
3.2. Effects of Lactobacillus GG
3.2.1. Primary Outcome: Therapy-Associated Side Effects
The incidence of therapy-related adverse effects was reported in all the included trials. A statistically significant
![]()
Figure 1. Selection process for the studies included in the meta-analysis.
difference was observed in adults between the Lactobacillus GG supplemented group and the control group with regard to the risk of overall adverse effects (three RCTs, n = 221, RR 0.59, 95% CI 0.45 - 0.78), shown in Figure 3. The risk of specific therapy-related adverse effects was statistically lower in the Lactobacillus GG group compared with the control group, i.e., diarrhea (four RCTs, n = 285, RR 0.23, 95% CI 0.11 - 0.47), bloating (four RCTs, n = 289, RR 0.61, 95% CI 0.41 - 0.90), taste disturbance (four RCTs, n = 288, RR 0.38, 95% CI 0.23 - 0.62), shown in Figure 4. No significant difference was observed between groups regarding nausea, vomiting, constipation, epigastric pain, and loss of appetite. The need for discontinuation from the eradication therapy was not reported in any trial.
3.2.2. Secondary Outcome: H. pylori Eradication Rates
Data on effects of Lactobacillus GG supplementation on H. pylori eradication rates were available from all four of the included trials, which reported data from 284 patients (218 adults and 66 children). No significant difference was observed between the Lactobacillus GG-supplemented group and the control group with regard to H. pylori eradication rates (four RCTs, n = 284, RR 0.99, 95% CI 0.88 - 1.13), shown in Figure 5.
4. Discussion
This meta-analysis of RCTs provides a summary of current knowledge regarding the effects of a single probiotic microorganism, Lactobacillus GG, in patients infected by H. pylori. With the limited evidence available, addition
![]()
Table 2.Characteristies of randomized controlled trials included in our meta-analysis.
![]()
Figure 2. Assessment of risk of bias of all studies included in the meta-analysis.
![]()
Figure 3. Relative risk (95% CI) of Lactobacillus GG on overall side-effects during therapy.
of Lactobacillus GG to triple therapy could reduce therapy-related adverse effects, such as diarrhea, bloating, and taste disturbance. However, Lactobacillus GG shows no beneficial effects on eradication rates. Because a majority of included patients were adults, our results may be applicable primarily to such a population.
Our study has several strengths. First, we systematically searched three major databases and no restriction of publication date was imposed. Second, to minimize the reviewers’ error or bias, the searching, assessment of eligibility of studies and extraction of relevant data were performed independently by two of authors in a blinded manner. Third, the beneficial effects of probiotics seem to be strain specific and pooling data on various strains could probably lead to spurious conclusions. Hence, our meta-analysis used only one probiotic microorganism
![]()
Figure 4. Relative risk (95% CI) of Lactobacillus GG on specific side-effects during therapy.
![]()
Figure 5. Relative risk (95% CI) of Lactobacillus GG on H. pylori eradication rates.
(Lactobacillus GG). However, several potential limitations should be taken into consideration when interpreting the results. First, one study reported unclear or inadequate allocation concealment and another trial was at high risk of binding process, although the other two trials seemed methodologically sound with respect to allocation concealment, binding, and >90% follow-up. Reassuringly, most included studies reported the effects of Lactobacillus GG for therapy-related adverse effects and eradication rate in the same direction, regardless of the methodological shortcomings. Second, we did not perform a statistical test for publication bias because statistical tests for the detection of publication bias have very low power in the meta-analysis of only a few trials [19] . Therefore, the possibility of publication bias still cannot be fully excluded. Third, a small number of studies with small sample sizes were included in our study, especially when subgroups of therapy-related adverse effects were evaluated. However, to increase power is one of the reasons why we conducted this meta-analysis [20] .
Our findings with respect to therapy-related diarrhea are consistent with previously published meta-analyses on the effects of Lactobacillus GG in preventing antibiotic-associated diarrhea in children and adults [21] [22] . Overall, these data support the use of Lactobacillus GG for the prevention of antibiotic-associated diarrhea, regardless of the reason for which the antibiotics were used. Of note, in contrast to several previous systematic reviews [5] [7] [23] , we observed a lack of beneficial effect of Lactobacillus GG on H. pylori eradication rates, although Lactobacillus GG was shown to inhibit H. pylori adhesion in vitro study [24] . Moreover, the results of included studies were consistent and no heterogeneity was observed across studies. Experimental studies regarding the exact mechanisms of action of Lactobacillus GG are scant. While the use of Saccharomyces boulardii along with standard triple therapy was reported to increase the eradication rates and decrease overall therapy-related side effects [25] , our study indicated that Lactobacillus GG was not capable of altering the eradication rates. Thus, we speculate that not all probiotics are created equal and that each strain must be evaluated individually in future studies.
Lactobacillus GG seems to be a good candidate for a large multicenter trial in patients with H. pylori infection. Since the exact mechanisms by which Lactobacillus GG reduces therapy-related side effects are unclear, experimental studies are desirable. As a majority of included patients are adults, studies in children are needed. Moreover, whether other probiotic strains, except Lactobacillus GG, have such effects needs to be explored in future experimental and clinical studies.
5. Conclusion
Current evidence indicates that Lactobacillus GG supplementation is a safe and effective way to decrease therapy- related side effects, particularly diarrhea, bloating, and taste disturbance, but has no effects on eradication rates. However, the results should be viewed with caution due to the methodological shortcomings of included studies and small sample sizes. High-quality and adequately powered RCTs are still warranted. Because a majority of included participants were adults, studies in children are also needed.
Acknowledgements
Shristi Shakya and Guoqiang Zhang: Study design, literature search, systematic review and data collection, statistical analysis, interpretation of results, and preparation of the manuscript; Huajian Hu: Literature search, systematic review and data collection, and statistical analysis; Zhongyue Li: Principal investigator, study design, statistical analysis, and interpretation of results. All authors have no conflict of interests related to this work.
NOTES
*Shristi Shakya and Guoqiang Zhang contributed equally to this work.
#Corresponding author.