Helicobacter pylori infection and inflammatory bowel disease in Asians: A meta-analysis

Abstract

AIM: To investigate the relationship between Helicobacter pylori infection and inflammatory bowel disease (IBD) in an Asian population.

METHODS: The PubMed, EMBASE, and Cochrane Library databases were searched for observational studies published up until June 2014, without language restrictions. Additional references were obtained from reviewed articles.

RESULTS: Ten studies involving 1299 IBD patients and 1817 controls were included in the meta-analysis (24.9% of IBD patients had H. pylori infection vs 48.3% of the controls). The pooled risk ratio for H. pylori infection in IBD patients compared with controls was 0.48 (95%CI: 0.43-0.54; P < 0.001). There was no significant heterogeneity in the included studies (I² = 21%). Egger’s linear regression indicated that there was no significant publication bias (P = 0.203).

CONCLUSION: The H. pylori infection rate in Asian IBD patients is significantly lower than in non-IBD patients, indicating that infection protects against the development of IBD.

Key Words: Asian population; Crohn’s disease; Helicobacter pylori; Inflammatory bowel disease; Meta-analysis; Ulcerative colitis

Core tip: A meta-analysis was carried out to investigate the relationship between Helicobacter pylori (H. pylori) infection and inflammatory bowel disease (IBD) in an Asian population. A search of PubMed, EMBASE, and Cochrane Library databases identified ten studies involving 1299 IBD patients and 1817 controls that were included in the meta-analysis (24.9% of IBD patients had H. pylori infection vs 48.3% of the controls). This meta-analysis showed that in an Asian population, the H. pylori infection rate was significantly lower in IBD patients than in non-IBD patients, indicating a protective effect of H. pylori infection against IBD.

INTRODUCTION

Inflammatory bowel disease (IBD) is described as chronic inflammation of the gastrointestinal tract in genetically susceptible individuals exposed to environmental risk factors, and involves alternating active and quiescent phases leading to an increased health burden worldwide[1,2]. Crohn’s disease (CD) and ulcerative colitis (UC) are the two primary subtypes of IBD. These diseases may result in intestinal damage, complications, and surgical interventions[3,4]. Commensal enteric bacteria are considered to have a critical role in the development of IBD. Continuous microbial antigenic stimulation can activate pathogenic immune responses and result in damage to intestinal mucosal barrier function and immunoregulation[5]. Genetic polymorphisms in the host most likely interact with intestinal bacteria to stimulate aggressive immunoreactions that cause chronic tissue injury. It is necessary to identify these host and microbial changes in individual patients in order to treat IBD[6,7].

Helicobacter species are characterized by microaerophilic metabolism, spiral shape, and peculiar motility, which contribute to their colonization of the gastrointestinal mucosal surface[8]. Helicobacter pylori (H. pylori) is a gram-negative pathogenic bacterium that is associated with chronic gastritis and is usually located on the surface of the stomach epithelium. Interestingly, H. pylori has also been identified in the normal colonic mucosa, colorectal neoplasms[9-11], and the intestinal mucosa of IBD patients[12,13]. Furthermore, H. pylori has been confirmed as a risk factor for colonic neoplasms[14,15], however, there is insufficient evidence to conclude that H. pylori has an important role in the pathogenesis of IBD.

Many observational studies have investigated the association between H. pylori infection and IBD. A meta-analysis involving 23 studies suggested that the H. pylori infection rate was lower in IBD patients than in non-IBD patients, and H. pylori was beneficial in preventing the development of IBD[16]. However, this meta-analysis only included one study that investigated the prevalence of H. pylori in an Asian population, and the conclusion may not be suitable for Asian populations. Therefore, we carried out an updated meta-analysis including only studies that determined the prevalence of H. pylori in IBD patients from Asian countries.

MATERIALS AND METHODS

Inclusion/exclusion criteria

Studies in line with the following criteria were included: (1) investigated the relationship between H. pylori infection and IBD; (2) used a case-control, cross-sectional, or cohort design; and (3) specifically included an Asian population. Studies were excluded if: (1) data from a previously published study were used; and (2) included a pediatric population.

Search strategy

We performed a search of PubMed, EMBASE, and the Cochrane Library for studies published up until June 2014. A search strategy was constructed using a combination of the following words: (Helicobacter pylori or H. pylori) and (IBD or CD or UC). Articles published in any language were included. A manual search of the references listed by studies retrieved from the online databases and from previously published systematic reviews was also performed to identify additional relevant studies.

Data extraction

Two investigators (Wu XW and Ji HZ) extracted data. Any differences regarding study inclusion, data extraction, and interpretation were resolved by consensus before the final analysis. Study variables were collected in the following categories: year of publication, country of origin, study center, characteristics of the patients, and H. pylori detection method. To avoid inclusion of duplicated data in the final analysis, retrieved studies were carefully evaluated and checked by comparing author names, geographic locations, and period of study.

Statistical analysis

Meta-analysis was carried out by combining the risk ratio (RR) in the IBD and control groups of the individual studies in a global RR. Statistical heterogeneity testing was performed using the χ² statistic and I², and an I² > 50% was considered to represent substantial heterogeneity[17]. A fixed effects model was selected when the heterogeneity test showed an I² < 50%, otherwise a random effects model was used. A funnel plot was used to determine publication bias[18]. Analyses were conducted using Review Manager software (RevMan version 5.3; Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014). The statistical methods of this study were reviewed by Liu from Department of Medical Statistics, School of Medicine, Nanjing University.

RESULTS

Search results

Our search identified 422 potentially relevant studies, of which 393 were excluded after title and abstract screening. Twenty-nine articles were retained for full-text review. Seven review articles, two case reports, eight studies including non-Asian populations, one study with incomplete data, and one study which included a pediatric population were subsequently excluded. We identified ten studies which fulfilled the inclusion criteria[13,19-27]. These studies included a total of 1299 cases of IBD and 1817 controls. Figure 1 shows the study flow diagram.

Figure 1 Flow diagram of the study selection process.

Characteristics of the included studies

The characteristics of the included studies and patients are presented in Table 1. Four studies were from Japan, four studies were from China, and two studies were from South Korea. All were case-control studies. Two of them were multicenter studies, and eight were single-center studies. The mean age of IBD patients ranged from 28.9 to 44.7 years. Two studies used serologic tests (immunoglobulin G) and eight studies used non-serologic tests (C-urea breath test, biopsy specimen histology, or biopsy sample culture) to detect H. pylori.

Table 1 Characteristics of the studies included in the meta-analysis.

Ref.	Country	Study center	CD/UC,	Control,	Mean age, yr	Male, %	Helicobacter pylori detection
			n	n	(CD/UC)	(CD/UC)
Matsumura et al[19] 2001	Japan	Multiple	90/NR	525	31.7/NR	70.0/NR	IgG
Furusu et al[20] 2002	Japan	Single	25/25	25	NR	NR	IgG/Histology
Moriyama et al[21] 2005	Japan	Single	29/NR	7	31.6/NR	59.0/NR	UBT
Ando et al[22] 2008	Japan	Single	38/NR	12	28.9/NR	74.0/NR	UBT
Hong et al[23] 2009	South Korea	Single	37/43	41	38.2/44.2	73.0/67.4	Histology
Pang et al[24] 2009	China	Single	52/54	106	36.7/42.3	57.7/55.6	IgG
Song et al[25] 2009	South Korea	Multiple	147/169	316	33.5/44.7	67.3/63.3	UBT
Zhang et al[13] 2011	China	Single	104/104	416	31.0/40.9	66.3/57.7	UBT
Jin et al[26] 2013	China	Single	NR/153	121	NR/44.6	NR/51.6	UBT/Culture
Xiang et al[27] 2013	China	Single	229/NR	248	46.2/NR	58.1/NR	UBT/Culture

CD: Crohn’s disease; IgG: Immunoglobulin G; NR: Not reported; UBT: Urea breath test; UC: Ulcerative colitis.

H. pylori and IBD

Of these studies, which included 1299 IBD patients and 1817 controls (Figure 2), 24.9% of patients in the IBD groups were found to have H. pylori infection, while 48.3% of patients in the control groups had H. pylori infection. The pooled RR of H. pylori infection rate in IBD patients compared to controls was 0.48 (95%CI: 0.43-0.54; P < 0.001). A fixed effects model was used for the meta-analysis as no significant heterogeneity in the included studies was observed (I² = 21%).

Figure 2 Forest plot of Helicobacter pylori infection rate in inflammatory bowel disease vs control groups.

Nine studies included 751 CD patients and 1696 controls (Figure 3). The rate of H. pylori infection in CD patients was 21.3% compared with 47.7% in the control groups (RR = 0.43, 95%CI: 0.37-0.50; P < 0.001). A fixed effects model was selected for the meta-analysis as significant heterogeneity in the included studies was not observed (I² = 43%).

Figure 3 Forest plot of Helicobacter pylori infection rate in Crohn’s disease vs control groups.

Six studies included 548 UC patients and 1025 controls (Figure 4). The rate of H. pylori infection was 29.9% in UC patients vs 52.5% in the control groups (RR = 0.55, 95%CI: 0.48-0.64; P < 0.001). We performed the meta-analysis with a fixed effects model as no significant heterogeneity was found in the included studies (I² = 0%).

Figure 4 Forest plot of Helicobacter pylori infection rate in ulcerative colitis vs control groups.

Publication bias

The funnel plot revealed a reasonably symmetrical distribution of the included studies examining the association between H. pylori infection and IBD (Figure 5). Egger’s linear regression indicated that there was no statistically significant evidence of publication bias (P = 0.203).

Figure 5 Funnel plot analysis.

DISCUSSION

The prevalence of H. pylori infection varies markedly in different countries and regions. Higher prevalence rates are seen in some developing Asian countries[28,29], while lower rates have been found in many developed countries in Europe and North America[30,31]. Moreover, the prevalence rates may vary significantly in different geographic regions or ethnic populations of the same country[32,33]. However, the variation in H. pylori infection rate in IBD patients of different race or region has not been completely clarified.

Our meta-analysis identified ten studies focused on the association between H. pylori infection and IBD in an Asian population. The included cases were from three east-Asian countries (China, Japan, and South Korea) that are considered to have similar ethnic origin, whereas the previous meta-analysis published in 2010 involved populations mostly from European and American countries[16]. By specifically including an Asian population, the heterogeneity and publication bias were both lower than those in the previous meta-analysis. The pooled rate of H. pylori infection in the IBD groups was a little lower than that reported in the previous meta-analysis (24.9% vs 27.1%), but the pooled rate of H. pylori infection in the control groups was higher than that previously reported (48.3% vs 40.9%). These discordant results suggest that ethnic origin may have a potential impact on the relationship between H. pylori infection and IBD.

The mechanism of H. pylori infection in preventing IBD is unclear. It has been hypothesized that H. pylori may downregulate proinflammatory immune responses in the host in order to promote its own survival, resulting in beneficial effects to the host. An animal experiment showed that long-term H. pylori infection can lead to distinct changes in microbiota composition in the large intestine, indicating that H. pylori may modulate the intestinal flora to affect the development of IBD[34]. Other research in animal colitis models suggested that H. pylori infection can regulate the immune responses, resulting in benefit to the host against other chronic inflammatory conditions such as IBD[35,36]. Papamichael et al[37] showed that H. pylori infection may play a protective role against IBD via the following mechanisms: increasing the levels of some cytokines, activating dendritic cells and T cells, downregulating the Th1/Th17 pathway, and inducing the generation of antibodies against H. pylori. In our meta-analysis, the lower pooled RR of H. pylori infection in IBD patients compared to controls suggests a protective effect of infection against the development of IBD. Subgroup analyses showed a tendency toward a more pronounced effect for CD when compared to UC. One of the included studies revealed a more evident association between H. pylori infection and IBD in patients < 60 years-old, which suggests that H. pylori infection may reduce the risk of IBD in younger adults[25]. Two studies found that the rate of H. pylori infection in IBD patients treated with antibiotics was lower than untreated patients. However, IBD patients without antibiotic treatment still showed a significantly lower rate of H. pylori infection than controls[24,25]. In a study with only 153 UC patients, the H. pylori infection rates in patients with diverse severity or extent of UC were significantly lower than those in the controls[26]. Another study showed that the H. pylori infection rates in patients with colonic, small intestine and ileocolonic CD were significantly lower than that in the control group[27]. We did not find any obvious correlation between the phenotypic characteristics of IBD patients and H. pylori infection rate in the included studies.

In conclusion, our meta-analysis shows that the rate of H. pylori infection in IBD patients from Asian countries is significantly lower than in non-IBD patients. However, there are some limitations in our meta-analysis, such as an insufficient number of included studies and potential heterogeneity. More prospective high-quality controlled studies are required to confirm the results of this meta-analysis.

COMMENTS

Background

Epidemiologic data show that Helicobacter pylori (H. pylori) infection has a protective effect against the development of autoimmune disease. Laboratory data suggest that H. pylori can induce immune tolerance and suppress the inflammatory response. Many observational studies have investigated the association between H. pylori infection and inflammatory bowel disease (IBD). Most of these studies found that the H. pylori infection rate in IBD patients was lower than that in non-IBD patients. However, conflicting outcomes have been observed and the exact mechanism of the protective effect of H. pylori in IBD development is still unclear.

Research frontiers

Numerous studies describing the H. pylori infection rates in IBD patients compared with healthy controls have been published in the past twenty years. A previous meta-analysis involving 23 studies suggested a lower H. pylori infection rate in IBD patients. H. pylori infection rate is related to ethnicity and region, however, the previous meta-analysis only included one study from an Asian country. This study investigated the relationship between H. pylori infection and IBD in an Asian population.

Innovations and breakthroughs

The authors demonstrated that the H. pylori infection rate in Asian IBD patients was significantly lower than in non-IBD Asian patients. The pooled H. pylori infection rate of IBD patients in this study was a little lower than the previous meta-analysis (24.9% vs 27.1%), but the pooled rate in the control group was higher than that previously reported (48.3% vs 40.9%). These results suggest that ethnic origin may have an impact on H. pylori infection rate, providing powerful evidence that H. pylori infection is a potential protective factor in the development of IBD.

Applications

The results of this meta-analysis show that IBD patients with a history of taking antibiotics had a lower H. pylori infection rate in two of the included studies, however, there were no available data for the controls. Antibiotic use may be partly responsible for the lower H. pylori infection rate in IBD patients. The exact mechanism of this interesting phenomenon should be investigated further. Large-sample and well-designed cohort studies with stringent disease and H. pylori infection definitions are required in order to delineate the protective effect of H. pylori infection in IBD development.

Terminology

H. pylori is a gram-negative, microaerophilic bacterium. It is present in patients with chronic gastritis and gastric ulcers. It is also linked to the development of duodenal ulcers and stomach cancer. H. pylori may be implicated in the pathogenesis of autoimmune diseases.

Peer-review

This is a well-designed and performed meta-analysis that confirms the results recently published for non-Asian cohorts. The results are interesting and are worthy of publication.