OUP user menu

Relationship between Alcohol Consumption and Active Helicobacter pylori Infection

Li Zhang, Guy D. Eslick, Harry H.-X. Xia, Chengqiu Wu, Nghi Phung, Nicholas J. Talley
DOI: http://dx.doi.org/10.1093/alcalc/agp068 89-94 First published online: 6 October 2009


Background: Helicobacter pylori (H. pylori) is a cause of chronic gastritis and maybe responsible for functional dyspepsia in a subset of patients. Many risk factors, such as alcohol consumption and smoking, may contribute to the colonization and infection of H. pylori in humans. However, studies on the relationship between H. pylori infection and drinking or smoking have produced conflicting results. Objective: The aim of this study was to examine whether consumption of alcohol or smoking is associated with active H. pylori infection in functional dyspepsia patients. Methods: H. pylori infection was confirmed by CLOtest and histology on at least two biopsies. Active chronic gastritis was diagnosed using the updated Sydney system. In addition to gender and age, information on drinking and smoking habits was collected using a standard questionnaire. Functional dyspepsia was diagnosed according to the Rome II diagnostic criteria. Results: H. pylori infection was positive in 27.3% of the 139 functional dyspepsia patients. Both age and gender were not significantly associated with H. pylori infection. A multiple logistic model found that alcohol consumption (OR = 9.05, 95% CI: 1.05–77.98) and pathology (active gastritis) (OR = 595.39, 95% CI: 81.43–4353.33) were associated with H. pylori infection. Active gastritis was associated with alcohol consumption (OR = 2.89, 95% CI: 1.03–8.02), smoking (OR = 2.72, 95% CI: 1.22–6.05) and age (OR = 1.03, 95% CI: 1.01–1.06). Conclusions: In patients with functional dyspepsia, there is no significant association between active H. pylori infection and smoking. However, alcohol consumption appears to be associated with H. pylori infection.


It has been established that Helicobacter pylori (H. pylori) infection is a major cause of gastric diseases around the world, including gastric and duodenal ulcer disease, and chronic gastritis (Kalantar et al., 1999; Talley and Li, 2007). There is increasing evidence to suggest that H. pylori also plays a major role in the development of gastric cancer, including MALT lymphoma (Eslick et al., 1999; Farinha and Gascoyne, 2005; Eslick, 2006). H. pylori infection has also been associated with dyspepsia and functional (non-ulcer) dyspepsia, although a causal association in functional dyspepsia is still unresolved (Kalantar et al., 1999; Tack et al., 2006).

Over the last decade, there have been a large number of epidemiological studies that have found that age, race and education level were major predictive factors for H. pylori infection (Graham et al., 1991; EUROGAST Study Group, 1993). Previous studies have reported that there is no significant relationship between smoking and active H. pylori infection (Brenner et al., 1997; Shinchi et al., 1997). However, drinking alcohol seemed to protect against active H. pylori infection (Brenner et al., 1997). The conclusions made regarding the protective effect of alcohol on active H. pylori infection suggest that it may be related to its antimicrobial effects; however, other findings have been inconsistent and have produced conflicting results (Brenner et al., 2001; Moayyedi et al., 2002; Murray et al., 2002).

The objective of the present study was to investigate the relationship between active H. pylori infection and consumption of alcohol or smoking habits in patients with functional dyspepsia.



The study was approved by the Human Research Ethics Committee, Nepean Hospital, The University of Sydney, Australia, and written informed consent was obtained from all the participants. All patients were recruited by the Gastroenterology Clinic at Nepean Hospital (a tertiary teaching hospital of The University of Sydney), from December 2001 to December 2004. All patients had an endoscopy, and 139 were diagnosed with functional dyspepsia out of 328 who had an endoscopy during this period.

Individuals were from the Nepean catchment area that consists of a population of 307,787 (7.7% of the Sydney population) and is socio-demographically very similar to the Australian population according to 2001 Census data, except that its inhabitants are slightly younger (30 versus 35 median years) and it has a slightly higher socioeconomic status based on income ($450 versus $350 median individual income per week), respectively. Ethnic status was not obtained, but the majority would be Caucasian based on Australian Bureau of Statistics data with 73.8% of persons living in Nepean being born in Australia; this compares with 70% of Australian residents who were born in Australia (www.abs.gov.au).

Data collection

A detailed history was obtained from each patient and included a complete physical examination with a special emphasis on the gastrointestinal system. A self-report questionnaire was completed by all subjects and special effort was taken to obtain and record details concerning smoking habits and alcohol consumption. In particular, with respect to smoking, the questionnaire asked whether the subject smoked cigarettes currently, in the past or never. We used the validated World Health Organization (WHO) AUDIT questionnaire to assess alcohol consumption (Saunders et al., 1993). Specifically the questionnaire asked whether the subject drank or not and current drinkers were classified based on the main type of alcohol consumed: (1) beer, (2) wine and (3) spirits. The frequency of alcohol intake was categorized as: non-drinker; 1 time monthly or less; 2–4 times monthly; 2–3 times weekly; and 4 or more times weekly. Alcohol intake was calculated in standard units/week, with 1 unit equal to one glass of wine, a standard measure of spirits, or half a pint of beer.

Upper gastrointestinal endoscopy

Upper gastrointestinal endoscopy was performed on each patient after an overnight fast for 12 h. Systematic examination of the esophagus, stomach, duodenal cap and second portion of the duodenum was carried out using a forward viewing Pentax EPM-3500 gastroscope. All the endoscopic examinations were performed by three experienced gastroenterologists. Patients in whom no endoscopic lesion was found in the esophagus, stomach, duodenal cap and second portion of the duodenum were further evaluated by obtaining three antral and one body biopsy specimens. One of the antral specimens was used for a rapid urease test (CLOTM-test, Delta West Pty Ltd, Western Australia) (Xia et al., 2000). All other specimens were properly labeled, fixed in 10% buffered formalin, processed using the paraffin-embedding technique, sectioned at 4 μm perpendicular to the mucosal surface and stained with hematoxylin and eosin (H&E) as well as with Giemsa (Dixon et al., 1996). Histopathological assessment was made by two experienced pathologists who examined all the specimens. The histopathological parameters were graded using the updated Sydney system (Dixon et al., 1996).

Detection of H. pylori infection

One specimen was examined using the CLO-test. Two specimens were stained with Giemsa. To establish a gold standard, H. pylori status was confirmed to be positive if a patient was positive using one of the following methods: (i) the CLO-test and positive histology on one or more biopsies or (ii) positive histology on at least two biopsies. Giemsa stain positive in two sections or both Giemsa stain and CLO-test positive was confirmed as H. pylori positive (Thijs et al., 1996).

Patient selection

Inclusion criteria

The inclusion criteria conformed to The Rome II diagnostic criteria for functional dyspepsia (Drossman et al., 2000).

  • Persistent or recurrent pain or discomfort centered in the upper abdomen of at least 12 weeks (which need not be consecutive) in the preceding 12 months.

  • Pain or discomfort not relieved by defecation or associated with onset of change in stool frequency or stool form.

  • No evidence of organic disease from the history, physical examination or laboratory tests that is likely to explain the symptoms.

  • No evidence of any mucosal lesion in the esophagus, stomach or duodenum at upper gastrointestinal endoscopy.

Exclusion criteria

The exclusion criteria were as follows:

  • Patients who had used nonsteroidal anti-inflammatory drugs (NSAIDs) in the last 2 weeks prior to endoscopy;

  • Attempted H. pylori eradication or acid suppressive therapy in the last 2 weeks prior to endoscopy;

  • Patients with symptoms suggestive of irritable bowel syndrome (IBS), i.e. lower abdominal pain or altered bowel habit.

Statistical analysis

The possible association of any of the lifestyle factors that we studied (smoking and alcohol consumption) with H. pylori infection rate was initially analyzed using the chi-square analysis. The test for trend was conducted using Friedman's analysis of variance test. All P-values calculated were two-tailed; the alpha level of significance was set at 0.05. Differences in mean scores were determined using a t-test. Patient demographic and clinical characteristics have been reported as mean and standard deviation or confidence interval for numeric-scaled features and percentages for discrete characteristics.

Factors associated with H. pylori infection were identified using unconditional logistic regression. Analysis proceeded in two steps: (i) identification of statistically significant univariate risk factors and (ii) the inclusion of only the significant univariate risk factors in a multiple logistic regression model. This two-step process provides a complete risk factor profile for H. pylori infection. All results were presented as odds ratios (OR) with 95% confidence intervals (CI). All analyses were carried out using STATA version 9.0 (StataCorp. 2005, Stata Statistical Software: Release 9, StataCorp LP, College Station, TX).


Descriptive analysis

The basic characteristics of the study population are given in Table 1. H. pylori seropositivity was 27.34% among the patients in the sample. The mean age was 45.8 years and the age range was 16–82 years. The active H. pylori infection rate increased with increasing age from 20.8 to 35.7%, but this was not statistically significant (P = 0.67). Almost two-thirds (61.9%) of the patients were female. However, the H. pylori seropositivity was not significantly different between male and female subjects (P = 0.56).

View this table:
Table 1

General characteristics of 139 functional dyspepsia patients

H. pylori-positive
N(n, %)P-value
Age (years)
 16–29245 (20.8)0.67
 30–454512 (26.7)
 46–594211 (26.2)
 60+2810 (35.7)
 All ages13938 (27.3)
 Male5313 (24.5)0.56
 Female8625 (29.1)
Alcohol consumption
 None354 (11.4)0.01
 Drinker10434 (32.7)
Cigarette smoking
 Never smoker7315 (20.6)0.06
 Smoker6623 (34.8)

Just over half of the participants were smokers (52.52%) and almost 75% of the participants were drinkers. The H. pylori positive rate was 20.6% and 34.8% in non-smokers and smokers, respectively (Table 1), although the difference was not statistically significant (P = 0.06).

Univariate analysis

Alcohol consumption did not vary by gender (OR = 0.57, 95% CI: 0.25–1.30) or age (OR = 0.99, 95% CI: 0.96–1.02) and H. pylori seropositivity did not vary by age (OR = 1.01, 95% CI: 0.98–1.03) or gender (OR = 1.26, 95% CI: 0.57–2.75).

Table 2 compares drinkers and non-drinkers and reveals that there was no statistically significant difference in age (OR = 0.99, 95% CI: 0.96–1.02), gender (OR = 0.57, 95% CI: 0.25–1.30) and smoking habit (OR = 0.94, 95% CI: 0.44–2.03) between the two groups. Drinking habits were positively associated with H. pylori infection. H. pylori seropositivity was 11.4% and 32.7% in non-drinkers and drinkers, respectively. H. pylori seropositivity was significantly higher in the drinkers group compared to the non-drinkers group (OR = 3.76, 95% CI: 1.22–11.52), and the type of alcohol consumed (OR = 1.32, 95% CI: 1.02–1.68) was associated with H. pylori seropositivity, suggesting that those who were H. pylori-positive were more likely to drink wine and spirits than beer. Moreover, an analysis for trend revealed that the H. pylori positivity rate appeared to increase with the amount of alcohol consumed (P = 0.014) (Table 3).

View this table:
Table 2

Comparison of general characteristics in drinker and non-drinker in relation to age, gender and smoking habit

Non-drinker (%)Drinker (%)P-value
Age (years)46.6345.490.69
Cigarette smoking
 Never smoker51.452.90.88
View this table:
Table 3

Relationship between alcohol intake and H. pylori infection

H. pylori-positive
N(n, %)P-value
Alcohol frequency
 1 time/month or less4013 (32.4)0.62
 2–4 times/month227 (31.8)
 2–3 times/week219 (42.9)
 4 or more times/week215 (23.8)
Type of alcohol0.03
 Beer267 (26.9)
 Wine238 (34.8)
 Spirits155 (33.3)
 Mixture4014 (35.0)
Amount of alcohol (standard drinks/week)
 0354 (11.4)0.01
 <56824 (35.3)
 5–10199 (47.4)
 11–40172 (11.8)

The rate of active gastritis in H. pylori-positive and H. pylori-negative patients was 94.7% and 4.0%, respectively (P < 0.0001). In addition, among the H. pylori-positive group, the rate of active gastritis in the drinkers and non-drinkers was 84.2% and 100%, respectively; however, the difference was not statistically significant (P = 0.08).

Table 4 shows the frequency and type of drinking associated with H. pylori infection. In terms of drinking frequency, drinking once per month or 2–3 times per week was significantly associated with H. pylori infection. The type of alcohol consumed was also important in relation to H. pylori infection, with wine and a mixture of alcohol significantly associated with infection.

View this table:
Table 4

Frequency and type of drinking associated with H. pylori infection

Odds ratio95% confidence interval
Drinking frequency
 2–4 times/month3.610.91–14.29
 2–3 times/week5.811.50–22.49
 4 or more times/week2.420.57–10.28
Type of alcohol

Multivariate analysis

In a multiple logistic regression analysis, we assessed factors associated with H. pylori infection (Table 5). The model included age, gender, alcohol, smoking, alcohol consumption, pathology, frequency of alcohol consumption and type of alcohol consumed. Only alcohol consumption (OR = 9.05, 95% CI: 1.05–77.98) and pathology (active gastritis) (OR = 595.39, 95% CI: 81.43–4353.33) were associated with H. pylori infection. Associations with pathological diagnosis (active gastritis) were found for alcohol consumption (OR = 2.89, 95% CI: 1.03–8.02), smoking (OR = 2.72, 95% CI: 1.22–6.05) and increasing age (OR = 1.03, 95% CI: 1.01–1.06).

View this table:
Table 5

Logistic regression models for variables associated with H. pylori infection

Odds95% confidenceOdds95% confidence
Type of alcohol drink1.321.02–1.681.030.55–1.91
Frequency of drinking1.220.94–1.59
Standard drinks0.980.92–1.04


Functional dyspepsia is one of the most common chronic digestive disorders affecting humans, with prevalence in the community ranging from 7% to 41% (Talley et al., 1992). Gastric secretory and motility disorders, as well as diet, alcohol, tobacco, non-steroidal anti-inflammatory drug (NSAID), psychosocial factors and H. pylori infection, have been examined, but none has been found to be a definitive, unique cause of functional dyspepsia (El-Serag and Talley, 2004). Approximately half of the patients with functional dyspepsia are infected with H. pylori in Australia (Talley et al., 2001). The prevalence of H. pylori infection in patients with functional dyspepsia is less than 12% in the United States (Spiegel et al., 2002). In the present study, H. pylori positivity was 27.3% in 139 patients with functional dyspepsia, and chronic active gastritis was present in 94.7% of H.pylori-positive patients. These results were in agreement with other observations (Kalantar et al., 1999; Tack et al., 2006). It has been well established that H. pylori infection leads to infiltration of the stomach with polymorphonuclear and mononuclear inflammatory cells by promoting expression of a variety of cytokines such as interleukin-1, tumor necrosis factor-α and interferon-γ and growth-regulated chemokines, macrophage inflammatory proteins and monocyte chemotactic activating factor, resulting in a pattern of active-chronic gastritis. Moreover, H. pylori produces a range of cytotoxic substances including CagA, VacA, hemolysin and platelet-activating factor and harmful enzymes such as urease, catalase, protease, lipase and phospholipase. These factors are also involved in the pathogenesis of gastric inflammation and mucosal damage (Yamaoka et al., 1996; Ogura et al., 2000; Weeks et al., 2000; Blaser and Atherton, 2004). Despite the well-defined causal role of H. pylori infection in gastritis, the role of H. pylori infection in functional dyspepsia has been controversial. Most individual trials showed that eradication of H. pylori does not relieve the symptoms of functional dyspepsia (Blum et al., 1998; Talley et al., 1999), although a meta-analysis suggests a small benefit from H. pylori eradication in infected patients (Moayyedi et al., 2003). However, none of the trials randomized H. pylori-negative patients to antibiotics or placebo, and hence any benefit may reflect a non-specific response to anti-microbial therapy rather than an effect of H. pylori eradication per se.

Several studies have assessed the possible association between H. pylori and cigarette smoking. In the majority of cross-sectional studies, no relationship between smoking and H. pylori infection was found (Rosenstock et al., 2000; Moayyedi et al., 2002). However, there have been some studies that produced a positive finding (Murray et al., 1997; Woodward et al., 2000) and others that found a negative (Ogihara et al., 2000) relationship between smoking habit and H. pylori seropositivity. In the present study, the H. pylori-positive rate was 20.6% and 34.8% in non-smokers and smokers, respectively, but the difference was not statistically significant. Nicotine may alter gastric mucosal blood flow, mucus secretion and epidermal growth factor secretion that may facilitate colonization following exposure to the organism (Endoh and Leung, 1994). Conversely, the result may be due to confounding factors, mainly those related to socioeconomic class, education grade and sanitary conditions.

Several cross-sectional studies have investigated the relationship between alcohol consumption and H. pylori infection. Some studies reported a significantly inverse association with H. pylori infection (Ogihara et al., 2000; Murray et al., 2002; Kuepper-Nybelen et al., 2005), while others found no significant association (Rosenstock et al., 2000; Ito et al., 2001; Moayyedi et al., 2002). The findings of studies from Germany (Brenner et al., 2001) and Greece (Gikas et al., 2004) suggest that an inverse relationship exists between alcohol consumption and H. pylori infection that may be U-shaped rather than monotonic. Existence of a U-shaped relation might also explain why observed effects of alcohol consumption on H. pylori infection have been inconsistent in epidemiologic studies, which have been conducted among populations with strongly divergent levels of alcohol consumption. Some results indicate that consumption of moderate amounts of alcohol in the form of wine, beer and spirits may protect against H. pylori infection. Three mechanisms may explain the negative relationship. Firstly, alcohol may exert an anti-bactericidal effect against new infection. Secondly, alcohol may be bactericidal against existing H. pylori infection (Ogihara et al., 2000), and finally, some alcoholic beverages are known to stimulate gastric acid secretion, which may eradicate H. pylori by lowering the pH in the stomach (Bujanda, 2000).

Interestingly, in the present study, we found that non-drinkers exhibited a significantly lower rate of H. pylori infection compared with drinkers, demonstrating a positive association, which is in opposition to previously reported studies. Indeed, a previous study reported that heavy alcohol consumption favored colonization of the gastric mucosa by H. pylori, although subjects reporting heavy alcohol consumption were both less educated and older, compared with those reporting no, mild or moderate alcohol consumption (Lieber, 1997).

Alcohol-induced disorders of the GI tract are very common (Riezzo et al., 2001). Both acute and chronic alcohol consumption can alter gastric acid secretion, induce acute gastric mucosal injury and interfere with gastric and intestinal motility (Bode and Bode, 1997). Studies in both animals and humans have found that alcohol concentrations of greater than 10% disrupt the gastric mucosal barrier and increase the mucosa's permeability (Bode and Bode, 1992). Several studies have suggested that a decrease in prostaglandins and an increase in the production of leukotrienes might play a role in the development of alcohol-induced mucosal injury (Bode and Bode, 1992; Bode et al., 1996). However, it appears that alcohol consumption is associated with active H. pylori infection. It is postulated that alcohol consumption facilitates H. pylori infection by damaging the gastric mucosa. However, other mechanisms may be involved in the synergistic effect, including bacterial adherence and host factors.

Bacterial adherence to the host epithelia is an important virulence determinant and it is an essential pre-requisite for pathogenicity for some bacteria (Jones, 1977). Bacteria with stronger adherence capacity are supposed to colonize with higher densities (Rad et al., 2002). The well-defined bacterial adherence factor, blood group Ag-binding adhesin (BabA) (Ilver et al., 1998), may play an important role in H. pylori-induced severe gastric inflammation. H. pylori sialic acid-binding adhesin has also been implicated in persistent infection and chronic inflammation (Mahdavi et al., 2002). The outer inflammatory protein, HP0638, is a novel putative virulence factor that is associated with interleukin-8 (IL-8) secretion from epithelial cells and ameliorates the adherence properties of H. pylori and thereby augments the density of bacterial colonization (Yamaoka et al., 2000; Yamaoka et al., 2002). According to the above studies, we propose a hypothesis that may explain the positive relationship between alcohol consumption and H. pylori infection. Briefly, acute and chronic alcohol consumptions disrupt the gastric mucosal barrier and increase the mucosa's permeability, resulting in a chemical inflammation. Subsequently, macrophages and neutrophils release the cytokine IL-8, which combines with its receptor in the endotheliocyte, and facilitate further development of the inflammation and also up-regulate the expression of adhesion molecules such as inter-cellular adhesion molecule 1 (ICAM-1) and lymphocyte function-associated antigen-1 (LFA-1). IL-8 may interact with HP0638 to augment the adherence capacity of H. pylori and increase the colonization density. In addition, high alcohol concentrations inhibit gastric motility and thus delay the emptying of the stomach (Pfeiffer et al., 1992). Therefore, it appears that multiple factors contribute to facilitate H. pylori colonization in drinkers. However, this hypothesis must be further tested by additional experimental and epidemiological research.

In the interpretation of our results, the following limitations should be kept in mind. First, all analyses were performed based on self-reported alcohol consumption and, as in pertinent epidemiological studies, there might have been a tendency of underreporting alcohol consumption. Second, only current alcohol consumption was ascertained in a quantitative manner and information on past drinking habits was not collected in detail such as drinking patterns or lifetime history of drinking. Third, the sample of subjects examined, especially non-drinker, was relatively small, and thus a statistical bias may exist. In addition, the fact that all subjects were diagnosed with functional dyspepsia may have had some impact in terms of selection bias; however, it must also be remembered that this is a functional disorder with no known pathology. There is much debate over the relationship between dyspepsia and H. pylori infection, with meta-analyses producing conflicting results (Gisbert et al., 2002; Moayyedi et al., 2006; Jin and Li, 2007). If there is any relationship, it appears that it would be only a small one, with limited impact. Finally, the fact that no education or socioeconomic data were available to assess potential confounding is a limitation of this study. Previous studies have highlighted the link between Helicobacter pylori infection and socioeconomic status (Yucel et al., 2009), suggesting that the results we found may be observed by chance alone due to the lack of adjustment during analysis. However, socio-demographic data taken from the Australian national census indicates that subjects from the area sampled have a higher income than the national average and are generally younger in age. This data suggest that our sample may not be at any greater risk of confounding due to their socioeconomic status, although we cannot totally rule this out and must consider it a limitation of this study.

In conclusion, in patients with functional dyspepsia, there is no significant association between active H. pylori infection and smoking consumption. However, alcohol consumption appears to be associated with active H. pylori infection. It is postulated that alcohol consumption facilitates H. pylori infection, presumably by damaging the gastric mucosa and/or promoting H. pylori adherence to gastric mucosa.


This study was supported by the Australian Brewers’ Foundation with an Alcohol-Related Medical Research Grant. G.D.E. was supported by the National Health and Medical Research Council (NHMRC) of Australia, with a Public Health Postgraduate Research Scholarship and by the Gastroenterological Society of Australia (GESA), with a Postgraduate Biomedical Research Scholarship.


View Abstract