Association between Helicobacter pylori infection and nonalcoholic fatty liver: A meta-analysis.

BACKGROUND: Opinion regarding whether Helicobacter pylori infection can promote the occurrence and development of nonalcoholic fatty liver (NAFLD) is divided. Therefore, we aimed to assess the exact relationship between H pylori infection and NAFLD by integrating all available data.
METHODS: The articles about H pylori infection and NAFLD were collected by searching the databases of PubMed, Embase, Web of Science, Scopus, China National Knowledge Infrastructure, and WanFang. The random-effects model was used for data analysis, followed by subgroup analysis and meta-regression to explore sources of heterogeneity.
RESULTS: Twenty-one articles were included in the study. Pooled analysis showed that H pylori infection indeed promoted NAFLD. Subgroup analysis and regression analysis showed that case-control ratio may be one of the sources of heterogeneity.
CONCLUSIONS: H pylori infection is indeed one of the factors that promotes the progression of NAFLD for the Asian population. This provides new approaches for clinical prevention and treatment for NAFLD.

Introduction

Nonalcoholic fatty liver disease (NAFLD) is a metabolic stress-induced liver injury characterized by diffuse hepatocyte macrovesicular fatty lesions in addition to alcohol and other defined liver injury factors.[ With improving economic stability and lifestyle changes, the incidence of NAFLD is increasing yearly, and its incidence rate is as high as 20% to 30% on a global scale, which seriously affects people's health.[ The etiology of NAFLD is complex, primarily characterized by abnormal lipid metabolism, insulin resistance, and genetic factors.[ In recent years, studies have found that abnormal fat metabolism in the liver can lead to dysbacteriosis of the intestinal flora; dysregulation of the flora leads to disorders of lipid metabolism, which eventually promotes lipid deposition in liver. Moreover, there is increasing evidence that NALFD is associated with abnormalities of the intestinal flora, especially with Helicobacter pylori.[

H pylori is a gram-negative bacillus colonized in the deep layers of human gastric mucosa.[ The reported infection rate of H pylori is as high as 50% or more worldwide.[ Studies have shown that H pylori is responsible for chronic gastritis, peptic ulcer, gastrointestinal lymphoma, and gastric cancer.[ In addition, new findings suggest that H pylori is closely related to liver tumors, obesity, diabetes, and abnormal lipid metabolism.[ In recent years, studies have found that H pylori infection is one of the factors contributing to the progression of NAFLD, and elimination of H pylori can delay the progression of NAFLD to some extent.[ However, other studies have suggested that H pylori infection has no clear relationship with NAFLD, and its eradication does not halt the progression of NAFLD.[ Therefore, we further explored the exact relationship between H pylori infection and NAFLD by integrating data for meta-analysis.

Materials and methods

Two independent researchers searched the PubMed, Embase, Web of Science, Scopus, China National Knowledge Infrastructure, and WanFang Data. The search keywords were: (Helicobacter pylori or H pylori or Hp or Helicobacter spp or H. pylori) AND (non-alcoholic fatty liver disease or NAFLD or non-alcoholic steatohepatitis or NASH or non-alcoholic fatty liver or NAFL or fatty liver). Articles published from January 2007 to October 2018 were searched. The search method uses keywords, without any restrictions, and manually searched for references in existing literature.

Inclusion criteria

Papers were included if they:

compared the risk of NAFLD in patients with H pylori infection and those without;

provided the number of positive/negative H pylori infection persons in the NAFLD and control groups.

Exclusion criteria

Abstracts, conference papers, and articles detailing animal experiments were excluded, as were articles that did not provide complete data.

In this paper, 2 researchers independently conducted the literature search and extracted the first author of the articles, year of publication, country of publication, method of detection of H pylori, method of diagnosis of NAFLD, the number of positive/negative H pylori infections in the NAFLD group, and the number of positive/negative H pylori infections in the control group. We assessed the quality of each study according to the Newcastle–Ottawa quality assessment scale.[ This study does not require the approval of the ethics committee.

Data analysis

All data analysis was performed using STATA version 12.0 software (Stata Corporation, College Station, TX), and heterogeneity analysis was performed using a Chi-square test or a Cochrane-Q test. Heterogeneity was assessed by I2 statistic, wherein I2 < 50% indicates minor heterogeneity, for which a fixed effect model was used, and I2 > 50 indicates large heterogeneity, for which a random effect model was used. The subgroup analysis and regression analysis were performed to explore sources of heterogeneity. The forest plot assesses the relationship between H pylori infection and NAFLD. The funnel plot and Begg and Egger tests were used to investigate publication bias. P < .05 was considered statistically significant.

Results

Clinical features

By searching the aforementioned databases, we selected 1491 research articles, and further browsed the title, abstract, and full text of the literature. Next, we excluded abstracts, conference articles, animal experiment studies, and those with incomplete data. Eventually, 21 studies were included for the final analysis that researched the relationship between H pylori infection and NAFLD. All these articles were published between 2007 January and 2018 October. The related literatures included 2 cohort studies, 2 case-control studies, and 17 cross-sectional studies in the meta-analysis. The flow chart for the studies is shown in Figure 1. A total of 14,623 participants were included, and the sample size for each study ranged from 53 to 43,216. Fourteen articles used the breath test to confirm H pylori infection, 7 articles used antibodies to detect H pylori infection, 19 articles used ultrasound to confirm NAFLD, and 2 articles used other methods. The basic information about all included literatures is listed in Table 1.

Figure 1

Flow diagram of the study screening and selection process.

Table 1

Baseline characteristics of all studies investigating Helicobacter pylori infection and NAFLD.

Meta-analysis and subgroup analysis

A total of 21 articles were included, including 11 reports in English and 10 in Chinese. We conducted a meta-analysis by integrating data to find significant heterogeneity (I2 = 95.6%). Therefore, we use the random-effects model to calculate the odds ratio (OR) and 95% confidence intervals (CIs). The results indicated that H pylori infection is indeed one of the contributing factors to NAFLD (P = .000, OR [95% CI] = 1.529 [1.336, 1.750]). The forest plot results are described in Figure 2.

Figure 2

Forest plot of the association between Helicobacter pylori infection and NAFLD. NAFLD = nonalcoholic fatty liver disease.

There was significant heterogeneity among the studies. To further explore the heterogeneity sources, we performed a subgroup analysis based on the study type, region, H pylori detection method, NAFLD detection method, sample size, and case-control ratio. The results of all subgroup analyses are shown in Table 2. Unfortunately, we did not find the cause of heterogeneity in the subgroup analysis.

Table 2

Meta-regression and subgroup analysis of all studies evaluating the association between Helicobacter pylori infection and nonalcoholic fatty liver.

Regression analysis

To find the source of heterogeneity, we also performed regression analysis (Table 2). The results showed that the heterogeneity was caused by case-control ratio (P = .000) instead of study type (P = .658), NAFLD test method (P = .477), H pylori detection method (P = .841), race (P = .542), publication year (P = .904), or language (P = .620).

Publication bias

We used a funnel plot to qualitatively detect the publication bias, and Egger and Begg tests to quantify the publication bias. The funnel plots were almost symmetric (Fig. 3). P-value of Egger test was.370. P was greater than.05, and no significant bias was observed.

Figure 3

Begg funnel plot of publication bias.

Discussion

NAFLD is characterized by simple fatty liver in the early stage, which can evolve and progress to steatohepatitis, cirrhosis, liver cancer, and liver failure.[ It is one of the most common liver diseases that affects people's life and health.[ Currently, there are some theories to explain how H pylori infection causes NAFLD. H pylori can upregulate the expression of various inflammatory factors such as tumor necrosis factor, C-reactive protein, and interleukin to promote insulin resistance.[ At the same time, H pylori can retrograde into the liver through the hepatic bile duct or intestinal ectopic, leading to chronic liver inflammation, causing liver cell damage and necrosis.[ The study found that human fetuin A is significantly increased in patients harboring H pylori infection, and human fetuin A is an important participant in insulin signaling, playing an crucial role in promoting insulin resistance and diabetes.[ In addition, studies have revealed that adiponectin expression is significantly reduced in NAFLD patients infected with H pylori, and adiponectin inhibits fatty acid deposition in the liver and inhibits NF-KB pathway activation as an anti-inflammatory effect. When adiponectin expression is abnormal, fat accumulates more easily in the liver cells, and the liver is more susceptible to inflammatory damage.[ Studies have also shown that lipids levels can change significantly in patients with H pylori infection; accordingly, dyslipidemia is common in patients with NAFLD.[

Our meta-analysis found that H pylori is one of the factors that promotes NAFLD progression, which is consistent with the results of a previous meta-analysis.[ However, the previous meta-analysis only included six articles, 3 of which were conference abstracts, and 2 of these abstracts did not have their full-text released yet. In addition, numerous studies on the association between H pylori infection and NAFLD have emerged since that meta-analysis was published. Therefore, it is necessary to implement a new meta-analysis on this issue. We combined the data and found that the heterogeneity was very obvious, so we adopted a random-effects model. Considering the heterogeneity among the studies, subgroup analyses were performed. Unfortunately, we did not find the reason for this heterogeneity in subgroup analyses. However, the good news was that through regression analysis, we found that the case-control ratio may be a cause of heterogeneity (P = .035).

There are several shortcomings in our meta-analysis. First, small sample studies are more prone to generate heterogeneity, and only 8 of the 21 studies in the meta-analysis had over 5000 patients. Second, most of the included researches were cross-sectional studies, and the results of the research are not very strong, which is bound to have a certain impact on our conclusions. Third, the risk factors for NAFLD include dyslipidemia, obesity, age, environment, diet and sex, and additional biochemical features of selected articles cannot be extracted, so our data integration for the meta-analysis has not been able to control these factors. Fourth, most of the meta-analysis done is taken from studies done in the Asian region, and the results may be more suitable for the Asian population. In addition, there is no uniform standard for detection of H pylori infection and diagnosis of NAFLD in these studies, and it is likely that there will be some bias in the results owing to different methodologies. Furthermore, different genotypes of H pylori have different effects on NAFLD as mentioned in the study; hence, the effects of different genotypes of H pylori in the articles cannot be excluded. Finally, if it can be proved that eradication of H pylori can effectively prevent NAFLD progression, it can be confirmed from the sidelines that H pylori infection does indeed promote NAFLD.

In conclusion, our meta-analysis by integrating data further confirmed that H pylori infection in the gastrointestinal tract is indeed one of the factors that promotes the progression of NAFLD for the Asian population. However, given that most of the included studies are small-sample, local geographical area and cross-sectional studies, multicenter, wide geographical area and large-sample prospective studies are needed to further explore the relationship between H pylori infection and NAFLD.

Author contributions

Conceptualization: Yi Shao.

Data curation: Yi Shao.

Formal analysis: Rongqiang Liu, Qiuli Liu, Yi Shao.

Investigation: Rongqiang Liu, Qiuli Liu, Ying He, Wenqing Shi, Qianhui Xu, Qing Yuan, Qi Lin, Biao Li, Lei Ye, Youlan Min, Peiwen Zhu, Yi Shao.

Methodology: Yi Shao.

Supervision: Ying He, Wenqing Shi, Qianhui Xu, Qing Yuan, Qi Lin, Biao Li, Lei Ye, Youlan Min, Peiwen Zhu.

Writing – original draft: Rongqiang Liu, Qiuli Liu.

Writing – review and editing: Yi Shao.