Relevance between Helicobacter pylori infection and nonalcoholic fatty liver disease in Bai minority region, Southwestern China


 Objective

The association between Helicobacter pylori (H. pylori) infection and nonalcoholic fatty liver disease (NAFLD) remains a matter of debate. We conducted this study to evaluate whether H. pylori infection is a major risk factor for NAFLD.
Methods

A total of 1185 individuals who received health check-ups from January 2017 to June 2019 were studied. Data of each subject who underwent 13C-urea breath, abdominal ultrasound, neck vascular color doppler ultrasound, and had a complete set of serum biochemical results was collected from the hospital information system. Participants were allocated to NAFLD group and non-NAFLD group based on abdominal color ultrasound for NAFLD. The baseline characteristics and serum biochemical results were compared. Logistic regression analyses were utilized to identify risk factors for NAFLD. Logistic regression models adjusted for confounding factors was performed to investigate the association between H. pylori infection and NAFLD.
Results

Compared to subjects without NAFLD(n = 656), those with NAFLD (n = 529) were more likely to be older, with higher weight, SBP, DBP, higher levels of TC, TG, LDL-C, FPG, AST and ALT, but with lower levels of HDL-C (all P values < 0.05). The levels of ALT, AST, BMI, FPG, TG, and DBP were independent risk factors for NAFLD (all P values < 0.05). Additionally, H. pylori was also an independent risk factor for NAFLD (OR = 1.35, 95%CI 1.02–1.79, P = 0.036). TC and LDL-C levels were significantly higher in H. pylori-positive group (n = 464) than that of in H. pylori-negative group (n = 721) (all P values < 0.05).
Conclusions

 H. pylori infection is a key risk factor for NAFLD, and serum lipid metabolic dysfunction can be observed in the subjects with H. pylori, suggesting the potential role of H. pylori infection in the progression of NAFLD.

closely allied to metabolic syndrome, obesity and insulin resistance [4]. It is yet unclear whether gastrointestinal bacteria have any impact on NAFLD pathogenesis [5].
H. pylori is a gram-negative, spiral-shaped bacterium present in the human stomach and infects more than 50% of world's population [6]. It is a common belief that H. pylori infection is strictly associated with peptic ulcer disease, chronic gastritis, gastric cancer, and mucosa-associated lymphoid tissue lymphoma [7]. In recent years, the association between H. pylori infection and extra-gastric diseases including coronary heart disease, stroke and obesity, has attracted increasing interest among the scienti c community [8]. Also, there are several studies about the positive correlations between H. pylori infection and NAFLD [9][10][11]. However, negative correlations between H. pylori infection and NAFLD have also been identi ed in some high-quality clinical trials [12][13][14][15]. Additionally, there are no available studies regarding the association between H. pylori infection and NAFLD in the Bai minority area of Southwestern China. Thus, we performed this study to evaluated the link between H. pylori infection and NAFLD and ll this void in local researches.

Study population
It was a cross-sectional study of Chinese asymptomatic adults who underwent health check-up in the First A liated Hospital of Dali University form January 2017 to June 2019 and entered into current study ( Figure 1). Subjects who simultaneous underwent abdominal color ultrasound, cervical color ultrasound, 13 C-urea breath test (UBT), complete physical examination and complete biochemical tests were included in this study. Exclusion criteria were listed as follows: (a) a long history of heavy drinking: drinking alcohol at least 210 g for male, 140 g for female per week in the past year; (b) previous treatment history of Methotrexate, amiodarone, glucocorticoid and tamoxifen; (c) subjects with viral hepatitis, autoimmune hepatitis, hepatolenticular degeneration, total parenteral nutrition, lack of beta lipoprotein, congenital lipid atrophy, and celiac disease; (d) taking proton pump inhibitors for 2 weeks, taking antibiotics and bismuth salts for 4 weeks before 13C-urea breath test; (e) a history of upper gastrointestinal; (f) a history of previous gastric surgery; (g) subjects with tumor or mental illness. This study gained approval by the ethics committees of the First A liated Hospital of Dali University.

Physical examination and serum biochemical examination
Height, weight, body mass index (BMI), systolic and diastolic blood pressure of each subject were measured during the physical examination.

H. pylori Infection determination
Subjects received 13 C-UBT using commercial kits (Headway, Zhonghe, Shenzhen, China). In brief, the exhaled breath samples were obtained at baseline and 30 minutes after the oral intake of a 13 C-urea capsule (75 mg). The samples were analyzed by the 13 C infrared spectrometry (Type HCBT-01, Headway, Zhonghe, Shenzhen, China), and the results were expressed as difference per thousand (δ‰). Test value = δ‰(30 min)−δ‰(0 min) [16]. A test value ≥4.0 was considered positive, and a test value <4.0 was considered negative.

Diagnosis of NAFLD
All the subjects underwent abdominal color Doppler ultrasound examination made by an ultrasound doctor using a Logiq E8 system (GE Healthcare, Chicago, IL, USA). Diagnosis of NAFLD was made based on ndings from ultrasound reports. Subjects who have at least two of the following three items can be diagnosed as Fatty Liver Diffuse Fatty Liver: (a) the near-eld echo of liver is diffusely enhanced and is stronger than that of the kidney. (2) The structure of intrahepatic duct is not clearly displayed. (3) The fareld echo of liver is gradually attenuated [17].

Statistical Analysis
We applied a widely used software SPSS version 23.0 for statistical analysis. Continuous data accorded with normal distribution were presented as average ± standard deviation (Mean±SD), whereas categorical data by percentage. Continuous variables conforming to a normal distribution were compared using the ttest, and categorical variables were compared by Chi-square test. For non-normally distributed data, comparisons of the medians between two groups were performed by Mann-Whitney U test. Logistic regression analysis was utilized to identify independent risk factors for NAFLD. A P value <0.05 was regarded as statistically signi cant.

Results
Basic information of the subjects A total of 1185 subjects were enrolled in current study, with 778 men and 407 women. Of 1185 subjects, the prevalence of NAFLD was 44.6% (529/1185), with 362 men and 167 women. Compared to subjects without NAFLD, those with NAFLD were more likely to be older, with higher weight, SBP, DBP, higher levels of TC, TG, LDL-C, FPG, AST and ALT, but with lower levels of HDL-C (all P values <0.05). There were no signi cant differences in height and ALB level between NAFLD group and non-NAFLD group (P>0.05). Moreover, the number of H. pylori-positive participants in NAFLD group were higher than those in non-NAFLD group (43.5% v.s. 35.7%, P=0.007). Collectively, these results indicate that there may be somehow related between H. pylori and NAFLD.

Risk factors for NAFLD
To explore which factors in uence NAFLD, univariate regression analysis was performed to establish risk factors for NAFLD. As shown in table 2, the levels of ALT, AST, BMI, FPG, TG, and DBP were independent risk factors for NAFLD (all P values <0.05). Of note, H. pylori was also an independent risk factor for NAFLD (OR =1.35, 95%CI 1.02-1.79, P=0.036).

Association between H. pylori infection and NAFLD
To further clarity the effects of H. pylori infection on NAFLD, univariate and multivariate regression analysis were applied (

Association between H. pylori infection and metabolic syndrome components
To investigate the possible link between H. pylori infection and NAFLD, some metabolic syndrome components were further analyzed. As presented in Table 4, the levels of TC and LDL-C were signi cantly higher in H. pylori-positive group than that of in H. pylori-negative group (all P values <0.05). However, there was no statistically signi cant difference between two groups for TG (P=0.409), HDL-C (P=0.087), SBP (P=0.776), DBP (P=0.614), FPG (P=0.492), and BMI (P=0.142). Overall, our ndings demonstrate that H. pylori infection may be responsible for NAFLD by affecting cholesterol metabolism.

Discussion
NAFLD is a common metabolic disorder that affects approximately 13%-30% of the general population, although its underlying mechanism still remains to be de ned. Several factors, such as genetic, environmental, metabolic, and behavioral factors, contribute to the pathogenesis of NAFLD [18]. In recent years, some scienti c investigations have revealed possible connections between H. pylori infection and NAFLD [19].
In current study, we found that H. pylori infection was signi cantly associated with NAFLD. Of note, H. pylori infection remained an independent risk factor of NAFLD morbidity following ruling out some confounding factors, such as age, carotid plaque, blood pressure, blood lipids, blood glucose, and liver enzymes (OR = 1.35,P = 0.036). This might imply that H. pylori infected individuals wound have a 35% higher risk of NAFLD than those without H. pylori infection. This is consistent with the results of meta-analysis by Zhou et al. who showed that a positive link between H. pylori infection and the risk of NAFLD [9].
At present, the preliminary idea that NAFLD would be the hepatic manifestation of the metabolic syndrome is changed, and it is considered that it might actually precede metabolic syndrome [20]. When analyzed in conjunction with metabolic syndrome, we found that higher weight, SBP, DBP, higher levels of TC, TG, LDL-C, FPG, and BMI were observed in NAFLD group than those of in non-NAFLD group (all P < 0.05). Moreover, we also found that DBP, FPG, TG, and BMI were signi cant risk factors for NAFLD. Accordingly, we speculate that a cross-talk may exist between the NAFLD and metabolic syndrome components. As previously reported by other studies, metabolic syndrome itself is one of the key risk factors for NAFLD, and vice versa [21].
A large body of literature has described that H. pylori infection is closely related to NAFLD and metabolic syndrome [22][23][24][25]. In current study, the effects of H. pylori infection on some key indicators of metabolic syndrome (SBP, DBP, FPG, BMI, TC, TG, HDL-C, and LDL-C) were analyzed. Interesting, we found that the levels of TC and LDL-C were higher in H. pylori-positive group than in H. pylori-negative group (P < 0.05). Accordingly, H. pylori infection may contribute to the increase of TC and LDL-C. This is in agreement with other reports. A large-scale meta-analysis shows that H. pylori infection signi cantly affects the serum lipid pro le, which might lead to various severe dyslipidemia-related diseases [26]. However, in this study, we failed to identify the impacts of H. pylori infection on other metabolic syndrome indicators, including SBP, DBP, FPG, BMI, TG, and HDL-C. We suggest that there are some reasons why our ndings may differ from other studies. Firstly, our study population with a higher level of education was drawn from government a liated institutions in Dali region. Compared with the rural population, they are likely to be more concerned about their health conditions, and have the awareness about the disease prevention.
Secondly, subjects were given advance noti cation of some dos and don'ts prior to blood collection, especially eating habits. To some extent, some biochemical indicators of subjects in the original living environment, such as blood lipid, blood glucose, liver and kidney function, might be underestimated after the adjustment of dietary structure. Thus, our ndings may be biased by the subjects with a good living environment, good habit, better dietary structure and good health awareness.
Several potential limitations of this study merit consideration. First, the current study is a single-center, small sample size, and retrospective research. Second, no follow-up of the subjects was performed. Third, only subjects of health check-up were involved in this study, and the results may not be generalizable to other populations. Larger cohorts including different subjects with different backgrounds are needed to further validate our ndings.

Availability of data and materials
The datasets analyzed during the current study are available from the corresponding author upon reasonable request.
Ethics approval and consent to participate All subjects gave their oral agreement to participate in this study. This study gained approval by the ethics committees of the First A liated Hospital of Dali University (2019072201).

Consent for publication
Not applicable.