Impact of Helicobacter pylori on the gastric microbiome in patients with chronic gastritis: a systematic review and meta-analysis protocol

Abstract

Introduction

Chronic gastritis is a common disease worldwide. Studies have consistently shown that chronic gastritis is usually associated with gastric microbial dysbiosis, especially the infection of Helicobacter pylori. However, the interaction between H. pylori and non-H. pylori bacteria in patients with chronic gastritis has not been clearly identified yet. Consequently, we designed a protocol for a systematic review and meta-analysis, which focused on identifying the changes in gastrointestinal microbiota composition between patients with H. pylori-infective and non-infective chronic gastritis.

Method and analysis

We will search PubMed, EMBASE and Cochrane Library databases to retrieve observational studies on humans. The eligible studies must include data about the relative abundance of the gastrointestinal microbiome in patients with H. pylori-infective or non-infective chronic gastritis. Only the data of adults aged over 18 years will be analysed. Two researchers will extract the data independently, and Newcastle–Ottawa Scale will be used to assess the risk of bias. Random-effects model will be performed in quantitative analyses. Correlation analysis, bioinformatics analysis and function analysis will be performed.

Discussion

Currently, numerous studies have revealed the role of H. pylori in chronic gastritis. However, the alterations of non-H. pylori bacteria in patients with chronic gastritis remain an open question. The results of our study might provide new insights into future diagnosis and treatments.

Ethics and dissemination

This study is based on published documents, unrelated to personal data, so ethical approval is not in need. The results of this study are expected to be published in journals or conference proceedings.

PROSPERO registration number

CRD42020205260; Pre-results.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This comprehensive systematic review uses all available information to demonstrate the difference of gastric mucosal microbiota in Helicobacter pylori infective and H. pylori non-infective status of patients with chronic gastritis, an important but little-known clinical issue.

• Two researchers will perform literature screening, data extraction and statistical analyses independently to minimise the potential individual biases, and all management will come to agreements in the group.

• Heterogeneity analyses will be used to evaluate the observational studies and enhance the overall credibility.

• Limited statistical power in published original research articles will be resolved through quantitative synthesis.

• The number of original research articles on mucosal microbiota of patients with chronic gastritis may be small scale, which needs to be further improved with more articles published.

Introduction

Human beings harbour a myriad of microorganisms that interact with each other and the host, contributing to an integrated ecosystem.¹ Previously, the stomach had been regarded as a sterile environment due to low pH until the discovery of Helicobacter pylori.² With the development of high-throughput sequencing technology, gastric bacteria had been identified and were found to display great research value in various diseases,³ and became one of the hottest topics in the frontiers of science.

Chronic gastritis is a common and long-lasting illness that affects more than 4.4 billion individuals around the world.^{4 5} Induced by environmental and genetic factors, chronic gastritis, including non-atrophic and atrophic gastritis, may further undergo histological progression like intestinal metaplasia, and a proportion of these patients will ultimately develop gastric cancer. As one of the major risk factors of chronic gastritis, H. pylori was reported to decline gastric acid secretion and induce mucosa inflammations, thus subsequently resulting in the alterations of the gastric microbiome.^{6 7} However, current opinion claims that it is the shift in the composition of gastric microbiota and not a certain type of bacteria that mainly functions in the pathogenesis process of chronic gastritis.^{8 9} Correspondingly, the shifting pattern of non-H. pylori bacteria has been increasingly discussed.¹⁰ There have been considerable studies showing the interplay between H. pylori and gastrointestinal microbiota, just like what Chen et al recently demonstrated.¹¹ Nevertheless, most studies focused on lower digestive tract, and fewer were about oesophagus and stomach. The main theme in the area of stomach was gastric cancer, so little attention was paid on patients with chronic gastritis. Studies undertaken so far provide contradictory evidence concerning the impact of H. pylori on other bacteria in patients with chronic gastritis,^12–14 and no conclusive conclusions have been reported.^{9 15}

Therefore, a comprehensive study is in need to summarise and assess previous data and to explore the potential interaction between H. pylori and the composition of the gastric microbiome. Considering that other parts of the gastrointestinal microbiome may differ with gastric microbiome from the aspect of composition and function, we will only extract the gastric microbiome in patients with chronic gastritis. It is hoped that our work could contribute to a deeper understanding of the role that H. pylori takes on non-H. pylori gastric bacteria.

Objectives

This research protocol aims to outline a systematic review and meta-analysis, evaluating the difference of the abundance of gastric microbiome in patients with chronic gastritis under the infective or non-infective status of H. pylori.

Methods and analysis

The protocol has been registered on the PROSPERO (International Prospective Register of Systematic Reviews) database, with the registration number CRD42020205260. The procedure of this protocol is under the guidance of the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) statement.¹⁶ PRISMA-P checklists demonstrated the related details. The research will last from March to April in 2022.

Patient and public involvement

Patients and the public have not been involved in study design, data collection, data analysis and result interpretation.

Inclusion criteria

Types of studies

The systematic review will include observational studies targeted on humans, such as cross-sectional, prospective or retrospective cohorts, and case–control studies. Animal studies and secondary literature will be excluded.

Study characteristics

The main exposure is H. pylori infection of patients with chronic gastritis. The studies which include both infective and non-infective status of H. pylori in patients with chronic gastritis will be qualified. Non-atrophic gastritis is identified as chronic inflammatory cell infiltration without reduced gastric glands. Atrophic gastritis is identified as reduced gastric glands and thinner mucosa.¹⁷ The infective status of H. pylori was confirmed by 13C urea breath test sequencing methods, PCR or histological evaluations of gastric biopsy tissues but not faecal samples, in order to avoid interference from intestinal microbiota. To control the heterogeneity of methodology, the sequencing methods of the gastric microbiome only include 16s rRNA, 16s rDNA sequencing or metagenomics to decrease the heterogeneity of methodology. Additionally, only adult participants aged over 18 years will be included.

Phenomenon of interest

All the eligible studies must provide direct or indirect data on the relative abundance of bacteria at the level of phylum or genus in patients with infective and non-infective status of H. pylori. Alpha diversity indexes, such as the Shannon index and Chao 1 index, are also in our concern. Correlation analyses will also be performed to demonstrate the interplay between H. pylori and other bacteria at the level of phylum or genus. Function analysis of the microbiome and pathway of functional genes will be analysed in metagenome data.

Literature search strategies

The following databases will be searched in our research: PubMed, EMBASE and Cochrane Library (until the date of December 2021). Apart from free-text searching, Medical Subject Heading (MeSH) terms are included in searching the PubMed database to enhance the sensibility of the results. The search strategy in PubMed is set as (“Helicobacter pylori” OR “H. pylori” OR “helicobacter”) AND (“microbiota” [MeSH Terms] OR “microbiota” OR “Gastrointestinal Microbiome” [MeSH Terms] OR “Gastrointestinal Microbiome” OR “Enteric Bacteria” OR “Gastric Microbiome” OR “Gastrointestinal Flora” OR “Gastrointestinal Microbial Community” OR “Gastrointestinal Microbiota” OR “Gastrointestinal Microflora” OR “Gut Flora” OR “Gut Microbiome” OR “Gut Microbiota” OR “Gut Microflora” OR “Human Microbiome” OR “Microbial Community” OR “Microbial Community Composition” OR “Microbial Community Structure” OR “Microbiome” OR “gut microenvironment” OR “flora” OR “microflora”) AND (“gastritis” OR “gastroenteritis” OR “gastricism”).

We also search the EMBASE and Cochrane Library with the same strategy except for certain MeSH terms. There are no limitations in our literature searching procedure. Moreover, we will also search the references of the retrieved papers and try to obtain the full text of related conference abstracts to ensure a more comprehensive search.

Data extraction and management

Two researchers (XC and YY) will select studies based on inclusion criteria and then extract the detailed data from eligible studies into an Excel independently, and a senior researcher (AY) will double-check to promote reliability. Any divergence will come to an agreement inside the team. We will retrieve the information from qualified research as follows:

1. Basic information of the study: publication (journal title, publication year, authors, study type), methods (sample sources, sample size, grouping, inclusion and exclusion criteria, quality control).

1. Patient characteristics: demographic characteristics (country, age, sex, ethnicity, comorbidities), diagnosis basis (H. pylori infective status, biopsy location, histological samples, microbiome sequencing method).

1. Outcome: the relative abundance of the gastric microbiome or related ratios, any index measuring alpha diversity (Shannon index, Chao 1 index, etc).

1. Sequencing data set.

We will try our best to retrieve the original data of eligible studies, and bioinformatics analysis will be carried out in a union manner.

We will make full use of the available materials to retrieve the data of interest, including but not limited to appendices, register materials, reference documents, conference abstract, published or unpublished results, and other possible materials. If necessary, we will contact the corresponding author of the article for further information.

Risk of bias assessment

Considering the type of studies included in this systematic review only involves observational study, we will choose the Newcastle–Ottawa Scale (NOS) system to assess the risk of bias.¹⁸ As the quality of the original study is critical to the analyses, certain modifications of NOS will be performed based on the sample size, detailed exclusion criteria and analytical procedure, quality control and demographic characteristics. The confidence in research will be properly considered according to the hierarchy of the evidence. Two researchers (XC and YY) will perform the evaluation independently from the aspects of selection, comparability and outcome. Any disagreement that arose in this procedure will be discussed and the whole team will finally achieve an agreement.

Data synthesis and statistical analysis

Basic information of included studies will be displayed in a table (eg, study type, methodology, main outcomes, etc). The main outcomes include the abundance ratio of the gastrointestinal microbiome between both H. pylori-infective and non-infective groups of patients with chronic gastritis. Alpha diversity indexes will be collected as well. If only one type of data (bacteria levels of phyla or genera, diversity indexes, etc) can be retrieved from three or more articles, it will undergo further meta-analysis procedure.

In the part of data analysis, the ratio of microbiome abundance from both H. pylori-infective and non-infective groups will be extracted directly from the article or calculated according to the supplemental materials. The collection of the ratio will be displayed as mean with SD or median with IQR, which is dependent on the distribution of the data. Moreover, we will use the results to form corrected models if available. Considering the aspect of diverse indexes, we will compare them in both groups and analyse the statistical differences.

Due to the variations of methodologies and the effects of different factors from varied populations, we will use the random-effects model in this study. Power calculation will be done. We do not expect that a large number of studies will be available and will therefore only perform a subgroup analysis comparing: the positive or negative H. pylori infective status and the abundance proportion of microbiota. Other potential sources of heterogeneity such as age, sex, ethnicity and country will be assessed through meta-regression. Correlation analysis will be carried out to unearth the interaction between H. pylori and non-H. pylori bacteria. Then, we will evaluate heterogeneity as I² values using the following thresholds: 0%–40% (unimportant), 40%–60% (moderate), 60%–80% (substantial) and >80% (considerable). Descriptive statistical analysis will be used when quantitative synthesis is not appropriate. Bioinformatics analysis will be conducted optionally depending on the number of retrieved data sets.

Routinely, funnel plots will be used to visualise the potential publication bias, and asymmetry data will be examined by the Egger’s test. We will use Review Manager V.5.3.3 (Nordic Cochrane Centre, Copenhagen, Denmark) in the analytical procedure. A p value of <0.05 will be considered as a statistically significant difference.

Ethics and dissemination

Based on published studies, our research includes no human participants, so there is no need for ethical approval. Furthermore, we have not published any data yet, as the formal data collection procedure has not started yet. In our expectation, the results of this study can be published in peer-reviewed journals or conference abstracts to provide more information for researchers.

Discussion

In reviewing the literature, existing conclusions were inconsistent regarding the composition of non-H. pylori bacteria between patients with H. pylori-infected or non-infected chronic gastritis. H. pylori-negative patients can also develop chronic gastritis, while the detailed mechanisms underlying these cases remained largely unknown. Moreover, conventional H. pylori eradication therapy could not completely cure gastritis.^{19 20} Thus, the interaction between H. pylori and non-H. pylori bacteria in the pathogenesis of chronic gastritis remains an interesting question.

To the best of our knowledge, it will be the first systematic review and meta-analysis that aims to evaluate the difference in the composition of the gastric microbiota in patients with chronic gastritis with and without H. pylori infection. Through the comprehensive analysis of all available information, this systematic review and meta-analysis may yield more objective and effective outcomes. Further, the results may provide a microorganism-related perspective on the mechanism of chronic gastritis. Also, the early signal of microbial succession may be potentially useful for future therapeutic interventions.

However, data ought to be interpreted cautiously as we anticipated a large heterogeneity among observational studies. Accordingly, we will perform a heterogeneity evaluation to quantify reliability. It is meaningful to explore and summarise this promising issue, which might help others to develop new methods of clinical diagnoses and therapies of both H. pylori-positive and H. pylori-negative chronic gastritis.

Ethics statements

Patient consent for publication

Not required.