Association between Helicobacter pylori infection and overweight or obesity in a Chinese population

Introduction: Recent studies found that Helicobacter pylori (H. pylori) infection plays a role in cardiometabolic disorders. The objective of this study was to assess the association between H. pylori infection and overweight or obesity in a Chinese population. Methodology: A cross-sectional analysis using data from the subjects who underwent a health examination between January 2010 and June 2012 in the department of comprehensive medicine was performed. Diagnosis of H. pylori infection was achieved using the carbon urea breath test (C-UBT). The participants were divided into H. pylori infection-positive group and H. pylori infection-negative group by CUBT. Results: A total of 2,050 subjects were enrolled in the study. The H. pylori infection-positive group had significantly higher body mass index (BMI) levels than did the H. pylori infection-negative group (25.32 vs 24.95, p = 0.008). There was a positive association between H. pylori infection and BMI levels (β = 0.30 ± 0.12, p = 0.015). After additional adjustment for white blood cell count (WBCC), the statistical significance disappeared (β = 0.24 ± 0.12, p = 0.053). Furthermore, a positive association between H. pylori infection and overweight/obesity according to different BMI criteria (BMI ≥ 24, BMI ≥ 23) was found. However, the association between H. pylori infection and obesity was consistently significant only based on the Asian criteria (BMI ≥ 27.5), but not significant based on the Chinese criteria (BMI ≥ 28). Conclusion: H. pylori infection was significantly and positively associated with overweight/obesity in a Chinese population.


Introduction
Obesity has become a severe public health problem worldwide.Health conditions caused or exacerbated by obesity include diabetes mellitus, hypertension, sleep apnea, coronary heart disease, hyperlipidemia, tumors, and other severe chronic diseases [1,2].The World Health Organization (WHO) defines obesity as a global epidemic.The prevalence of overweight or obesity in China increased rapidly in recent years [3,4].Obesity is traditionally considered a non-communicable disease, but studies from several laboratories over the past 30 years suggest that some microbe infections might be linked to obesity in animals and humans, a condition termed "infectobesity".On the other hand, obese subjects show an altered response to infection [5,6].
Helicobacter pylori (H.pylori) are Gram-negative bacteria that colonize gastric mucosal epithelium of humans and some primates [7,8].In recent years, studies have shown that H. pylori infection not only leads to various gastro-duodenal diseases such as chronic gastritis, peptic ulcer, mucosa-associated lymphoid tissue lymphoma (MALT), and gastric cancer [9,10], but also plays a role in cardiometabolic disorders [11].While conclusive evidence for a causative role of H. pylori infection in human obesity is lacking, the relationship between H. pylori and obesity is still controversial [12][13][14].Reports from different areas and countries were not consistent with each other.The aim of this study was to determine the association between H. pylori infection and overweight or obesity in a Chinese population.

Study population
In total, 2,588 Chinese subjects who underwent a health examination between January 2010 and June 2012 in the department of comprehensive medicine in Tongji Hospital (Wuhan, China) were enrolled in the present study.If the subjects received repeated health examinations, only the first health examination results were observed.These subjects were all employed in administrative units of Wuhan and thus had a similar socioeconomic status, judged by work type and level of instruction.In all, 538 subjects were excluded from the study: 274 who had a history of taking, within a week, proton pump inhibitor (PPI), antibiotics, or bismuth subcitrate, or a history of taking eradication therapy of H. pylori infection within six months; 23 with chronic liver and renal failure; 6 with malignancy; 15 with acute infection; 4 with connective tissue disorders; and 216 with missing data.Finally, 2,050 subjects (588 female, 1,462 male) were enrolled in the study.The subjects' ages ranged from 26 to 95 years (52.21 ± 11.31 years, mean ± standard deviation [SD]).The study was conducted after obtaining written informed consent from all subjects and was approved by the ethics committee of Tongji Medical College, Huazhong University of Science and Technology.

Anthropometry and biochemical measurements
A medical interview provided information about age, sex, cigarette smoking habits, alcohol consumption, and histories of current and previous illnesses.Height and weight were measured in the morning, in the fasting state.Body mass index (BMI) was calculated by dividing the body weight (kg) with squared height (m 2 ).Blood pressure was measured twice in a quiet state at five-minute intervals on the right arm using a sphygmomanometer.The mean of the two readings was used in data analyses.On the morning of the survey, overnight fasting venous blood samples were collected from each participant, and all blood samples were processed within 30 minutes of collection.White blood cell count (WBCC) was measured using a hematology analyzer.Plasma level of glucose was determined by the glucose oxidase method.Glucosylated hemoglobin (HbA1c) was measured by ion-exchange high-performance liquid chromatography (HPLC).Serum level of high-density lipoprotein cholesterol (HDL-c) was measured by the chemical precipitation method, and serum levels of total cholesterol (TC) and triglyceride (TG) were measured by enzymatic methods.Low-density lipoprotein cholesterol (LDL-c) was estimated using the Friedewald formula [15].

Diagnosis of H. pylori infection
Diagnosis of H. pylori infection was achieved using the carbon urea breath test ( 14 C-UBT).For the 14 C-UBT, patients first fasted for two hours, washed their mouths before dosing, and were in a sitting position.Breath samples were collected at 20-minute intervals.The breath samples were analyzed using a gas chromatography mass spectrometer.H. pylori infection was considered positive if the estimated value was ≥ 100 dpm/mmol CO 2 and was considered negative if the estimated value was < 100 dpm/mmol CO 2 .

Definitions
According to the criteria recommended by the Working Group on Obesity in China [16], overweight was defined as BMI equal to or greater than 24 kg/m 2 and less than 28 kg/m 2 , and obesity was defined as BMI equal to or greater than 28 kg/m 2 .According to the WHO Expert Consultation for Asians, overweight was defined as BMI equal to or greater than 23 kg/m 2 and less than 27.5 kg/m 2 , and obesity was defined as BMI equal to or greater than 27.5 kg/m 2 [17].Type 2 diabetes mellitus (T2DM) was diagnosed according to the 1999 WHO criteria: fast plasma glucose (FPG) equal to or greater than 126 mg/dL, or self-reported physician diagnosis or insulin use.According to the Seventh Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure guidelines [18], hypertension was defined as systolic blood pressure equal to or more than 140 mmHg , diastolic blood pressure equal to or more than 90 mmHg, and/or self-reported treatment of hypertension.Dyslipidemia was defined according to Adult Treatment Panel (ATP) III.

Statistical analyses
Categorical variables were presented as numbers and percentages.The Chi-square test was performed to assess differences in proportions across groups.Continuous variables were presented as mean ± SD for variables distributed normally, or as medians (25th to 75th percentiles) for skewed distributions.Differences between two groups were examined using Student's unpaired t test or by the Mann-Whitney U test as appropriate.One-way analysis of variance (ANOVA) or Kruskal-Wallis analysis of median test was applied to compare differences in means across groups.The Bonferroni correction was applied to adjust p values for multiple comparisons.Forced-entry multiple linear regression models were conducted to evaluate the relationship between H. pylori status and BMI levels.
Potential confounders were controlled for the following three models.Model 1 was adjusted for age and gender.Model 2 was adjusted for all the variables in model 1 plus smoking status, drinking status, hypertension, coronary heart disease, type 2 diabetes, and dyslipidemia.Model 3 was adjusted for all the variables in model 2 plus WBCC.These variables were chosen because of their potential role as confounders from the clinical point of view.To evaluate the association between H. pylori status and overweight or obesity according to different BMI categories, odds ratios (ORs) and 95% confidence intervals (95% CIs) for overweight or obesity in relation to H. pylori were estimated using unconditional logistic regression, controlling for the same three models.A two-tailed p value of < 0.05 was considered statistically significant.All analyses were conducted using SPSS software version 13.0.

Results
In total, there were 2,050 subjects (588 female, 1462 male) enrolled in this study whose ages ranged from 26 to 95 years (mean ± SD, 52.21 ± 11.31 years).They were divided into H. pylori infection-positive group and H. pylori infection-negative group by 14 C-UBT.The H. pylori infection-positive group had 839 subjects, accounting for 40.93% of the total participants, and the H. pylori infection-negative group had 1,211 subjects, accounting for 59.07% of the total participants.As shown in Table 1, a summary of demographics, metabolic, anthropometric and H. pylori status of participants, no differences were found between the two groups with regard to gender, systolic blood pressure, diastolic blood pressure, fasting blood glucose, glucosylated hemoglobin, lipid profiles and the morbidity of hypertension, coronary heart disease, and dyslipidemia.However, the H. pylori infectionpositive group had significantly higher BMI levels than did the H. pylori infection-negative group (25.32 vs 24.95, p = 0.008).There were more subjects (127/839, 15.14%) in the H. pylori infection-positive group who had type 2 diabetes than there were in the H. pylori infection-negative group (146/1,211, 12.06%; p = 0.044).In addition, subjects in the H. pylori infection-positive group were younger than those in the H. pylori infection-negative group (51.57vs 52.65, p = 0.034).

Discussion
In this cross-sectional study, the increased prevalence of H. pylori infection was found in patients with higher BMI levels.On the other hand, a positive association between H. pylori infection and BMI levels was found without adjustment for WBCC in Chinese subjects.In addition, the association was consistent between H. pylori infection and overweight/obesity.The incidence of overweight or obesity in China has increased year by year.In the present study, the prevalence of overweight was 64.49% (BMI ≥ 24) or 75.02% (BMI ≥ 23), and the prevalence of obesity was 16.73% (BMI ≥ 28) or 21.80% (BMI ≥ 27.5) in the subjects who were employed in the local administrative units of Wuhan.Increasing evidence indicates in various ways that obesity in animals and humans may be linked to several microbes, including animal and human viruses, bacteria, and parasites [5,6].In 1982, it was first reported that canine distemper virus could induce obesity in mice infected with the virus [19].Since then, pneumonia chlamydia, Selenomonas noxia, herpes simplex virus 1 or 2, gut microflora, and H. pylori have been reported to be associated with obesity [20][21][22][23].
H. pylori are prevalent bacteria worldwide.In our present study, about half of the population was infected with H. pylori.The infection rate was related to geographical location, race, age, and socioeconomic status.From January 2002 to June 2004, the team of collaboration of H. pylori research in China conducted an investigation of H. pylori infection among 26,341 people in 19 provinces, municipalities, and autonomous regions in China.The results indicated that the mean H. pylori total infection rate was 56.22%.The H. pylori infection rate in Guangdong province was the lowest, approximately 42%, and the rate in Tibetan monks was the highest, approximately 84% [24].But there were no data about the H. pylori infection rate in Wuhan city.In this study, the H. pylori infection rate in Wuhan city was 40.9%, and there was no difference between genders.The possible reasons that the H. pylori infection rate in our study was lower than that in the other regions were (i) the subjects in our study were employed in local administrative units of Wuhan, had higher education, and were in the middle-high socioeconomic class, which might be relevant to the low prevalence of H. pylori infection; (ii) diagnosis of H. pylori infection in our study was achieved by using 14 C-UBT, which could reflect current H. pylori infection status of the whole stomach.The vast majority of testing methodology of H. pylori infection that the team of collaboration of H. pylori research in China used was the H. pylori IgG enzyme-linked immunosorbent assay (ELISA), which indicates not only current H. pylori infection but also previous H. pylori infection.This method might be prone to false positives, which could result in the higher rate of H. pylori infection that was found.
The relationship between H. pylori and obesity remains controversial.These interactions could be classified as (i) the effect of obesity on H. pylori infection; (ii) the effect of H. pylori infection on obesity.

The effect of obesity on H. pylori infection
Obese individuals show an increased susceptibility to infections with different pathogens [25].Erol et al. found that the H. pylori infection rate in the obese group was 57.2% (59/103), and the H. pylori infection rate in the control group was 27.0% (30/111).There was a significant association between the obesity and serum antibody positive for H. pylori (OR, 2.11; 95% CI, 1.49-3.00)[22].Another study in Greece indicated that the incidence of H. pylori infection was not increased among overweight/obese young individuals (median age, 22.84 years) [11].In our study, we found that the prevalence of H. pylori infection in normal, overweight, and obese subjects was 37.36%, 41.88%, and 45.77%, respectively, based on the criteria recommended by the Working Group on Obesity in China (p for trend = 0.006).The mean age of our subjects was 52.21 years, much older than the subjects in previous studies.As natural elimination of H. pylori infection is difficult, it is commonly believed that H. pylori infection perpetuates in those who never receive eradication therapy.Consequently, the older H. pyloripositive subjects were believed to have longer-term H. pylori infection, which might induce age-related influence or a life-course perspective of H. pylori infection in the body.Yang et al. found that obesity was positively associated with increased risk of gastric H. pylori infection in 324 elderly Chinese subjects [26].
Obesity is usually associated with impaired immune function, and immune deterioration is also related to the grade of obesity [27].The maturation of monocytes into macrophages was found to be lower, and the capacity of polymorphonuclear (PMN) to be bactericidal was found to be reduced in obese individuals [28,29].Severely obese individuals also have a significant decrease in NK cell activity compared with control individuals after adjustment for age and gender [30].People with diabetes mellitus are also easily affected by chronic infections.A significant association was also observed between H. pylori infection and type 2 diabetes mellitus in our study (p =0.044).

The effect of H. pylori infection on obesity
Kopacova et al. studied 2,436 people (between 4 and 100 years of age) and found that there were positive associations between H. pylori infection and the levels of BMI both in overweight/obese and obese subjects over 15 years of age [31].But in that study, the risk factors (drinking status, diabetes, dyslipidemia, etc.) of overweight or obese subjects were not adjusted.In our study, the H. pylori-positive group had significantly higher BMI levels than did the H. pylori-negative group (25.32 vs 24.95, p = 0.008).In multiple linear regression models with BMI levels as the dependent variable, the association between H. pylori positivity and BMI levels was statistically significant (β = 0.30 ± 0.12, p = 0.015) after adjustment for H. pylori status, gender, age, smoking status, drinking status, hypertension, coronary heart disease, diabetes, and dyslipidemia.Furthermore, we found a positive association between H. pylori infection and overweight/obesity according to different BMI criteria.However, association between H. pylori infection and obesity was consistently significant only according to the Asian criteria (BMI ≥ 27.5), but not significant according to the more restrictive Chinese criteria (BMI ≥ 28).Obesity is multifactorial disorder that involves environmental, lifestyle, genetic, and social factors.In this study, we found that gender, current drinking, hypertension, diabetes, and dyslipidemia had more impact on BMI levels than did H. pylori infection.Hence, we deduced that H. pylori infection might improve BMI levels, but the effect of H. pylori infection that could be attributable to obesity is limited.The main reason of the increased prevalence of obesity in China is the change of dietary structure and decreased physical activity, which resulted from the economic transformation in recent decades.Therefore, though the H. pylori infection rates have been in a decreasing trend in China, the incidence of overweight or obesity in China has increased year by year.
The mechanism underlying the adipogenic action of microbial infection is unclear.One study showed that H. pylori-infected subjects have lower serum leptin levels [32].As leptin exerts anorexigenic effects, H. pylori infection may stimulate overfeeding and be involved in the mechanisms of obesity [33,34].Furthermore, H. pylori could promote gastric mucosa to produce a variety of vasoactive substances and inflammatory cytokines, such as tumor necrosis factor alpha (TNF-α), interleukin-1 (IL-1), interleukin-6 (IL-6), and other cytokines.These mediators may be transported through the bloodstream and may cause a chronic low-grade inflammatory response in extra digestive sites [35].As a result, these inflammatory mediators may lead to insulin resistance and induce overweight or obesity [36][37][38].As chronic infections could improve WBCC levels and lead to a series of inflammatory reactions, WBCC is a universally available marker of chronic low-grade inflammation.In our study, WBCC levels were significantly higher in the H. pylori-positive group than in the negative group (p < 0.001).Furthermore, the levels of BMI increased progressively with increasing quartiles of WBCC (p for trend < 0.001).There was a positive association between H. pylori infection and BMI levels in our study (β = 0.30 ± 0.12, p = 0.015).However, after additional adjustment for WBCC, the statistical significance disappeared (β = 0.24 ± 0.12, p = 0.053).In multiple logistic regression analysis of different BMI categories, we also found that the OR values between H. pylori infection and overweight/obesity were weakened after additional adjustment for WBCC.This indicated that H. pylori infection might improve of BMI levels through chronic inflammatory reaction mediated by WBCC.
There are some limitations in current study that require consideration in interpretation of our findings.First, because our present research was a crosssectional study, the cause-effect relation between H. pylori infection and overweight or obesity could not be proved conclusively.Second, the subjects in our study had a similar socioeconomic status and were in the middle-high socioeconomic class.Hence, we need to conduct further longitudinal and larger scale epidemiological investigation in general population to elucidate the relationship between H. pylori infection and overweight or obesity.

Conclusions
In summary, our study indicated that H. pylori infection was significantly and positively associated with the risk of overweight/obesity in a Chinese population.

Figure 1 .
Figure 1.Prevalence of H. pylori infection in normal, overweight, and obese subjects according to different body mass index (BMI) criteria; p for trend = 0.006 (A), p for trend = 0.009 (B).

Model 1
was adjusted for age and gender.Model 2 was adjusted for all the variables in model 1 plus smoking status, drinking status, hypertension, coronary heart disease, diabetes, and dyslipidemia.Model 3 was adjusted for all the variables in model 2 plus WBCC.

Figure 3 .
Figure 3. Multiple logistic regression analysis of H. pylori infection and overweight/obesity or obesity according to different BMI criteria.Data are expressed as odds ratio and 95% confidence interval.

Table 1 .
Demographics, metabolic and anthropometric characteristics and H. pylori status of participants Data are presented as the mean ± standard deviation (SD) for normal variables, as the median (interquartile range) for skewed variables, or column (percentage).P values were obtained by comparing H. pylori-negative and H. pylori-positive groups.;BMI: body mass index; SBP: systolic blood pressure; DBP: diastolic blood pressure; CHD: coronary heart disease; FPG: fasting blood glucose; HbA1c: glucosylated hemoglobin; TC: total cholesterol; TG: triglyceride; LDL-c: low density lipoprotein cholesterol; HDL-c: high-density lipoprotein cholesterol; WBCC: white blood cell count

Table 2 .
Multiple linear regression models for body mass index (BMI) levels with H. pylori and other potential confounders CHD: coronary heart disease; WBCC: white blood cell count; Model 1 was adjusted for age and gender; Model 2 was adjusted for all the variables in model 1 plus smoking status, drinking status, hypertension, coronary heart disease, diabetes, and dyslipidemia; Model 3 was adjusted for all the variables in model 2 plus WBCC.