Effects of Helicobacter pylori infection on gastric cancer, obesity and esophageal diseases: a systematic review and meta-analysis Authors: Qian Zhou, Ruohan Yan, Xiaomin Wu, Hong Guo, Jing Wu Published: 2025 ABSTRACT BackgroundAt present, the specific mechanism by which Helicobacter pylori (H. pylori) infection leads to gastric cancer (GC) is not clear. This study aims to systematically evaluate the correlation between H. pylori infection and the risk of GC, obesity and esophageal diseases by collecting and analyzing existing relevant studies, update the existing evidence-based medical evidence, and provide evidence to inform the etiology-based prevention and treatment of GC, obesity and esophageal diseases and further clinical research.MethodsLiterature search of the PubMed, Embase, Web of Science, Cochrane Library, ClinicalTrials.gov, CNKI, and Wanfang databases (as of June 27, 2023) was performed. Population-based studies (sample size n > 100) that reported the associations of H. pylori infection with GC, obesity, and esophageal disease and used objective methods to detect H. pylori infection were selected. Data such as effect sizes and 95% CI were extracted for analysis, and interstudy heterogeneity was analyzed on the basis of study location and certain other characteristics.ResultsForty relevant studies were included, including 15, 4 and 21 studies on GC, obesity and esophageal diseases respectively. The results revealed that H. pylori infection was positively correlated with the risk of GC [OR (95% CI) = 2.68 (1.85 ~ 3.89) ; P < 0.0001], and was negatively associated with the risk of obesity [OR (95% CI) = 0.65 (0.43 ~ 0.98); P = 0.0380] and gastroesophageal reflux disease (GERD) [OR (95% CI) = 0.54 (0.43 ~ 0.68); P < 0.0001], and had no significant association with the risk of esophageal cancer (EC) [OR (95% CI) = 0.90 (0.47 ~ 1.73); P = 0.7454]. There was considerable heterogeneity among studies for some diseases, but sensitivity analysis proved that the results were stable and reliable, and no significant publication bias was found.ConclusionH. pylori infection increases the risk of GC, H. pylori eradication is highly important for reducing the risk of GC. H. pylori infection was not associated with reduced risk of obesity and GERD, but the related evidence is limited and the mechanism is not clear. Based on the available evidence, decision-making for eradication treatment should follow the principle of individualization, taking into account the patient’s disease risk, strain characteristics, demographic factors, and the cost-effectiveness of treatment.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12876-025-04259-2. Introduction In recent years, the number of patients with NCDs (hereafter referred to as “NCDs”) in China has gradually increased, and the proportion of deaths attributed to NCDs continues to increase [1]. According to a previous report, the number of deaths attributed to NCDs in China accounted for 88.5% of the total deaths in 2019, and the increasingly severe problem of overweight/obesity among residents is an essential reason for the increasing incidence of NCDs [2]. Studies have shown that overweight/obesity increases the risk of many NCDs, such as cardiovascular and cerebrovascular diseases and cancer [3]. According to a national survey, more than half of Chinese adult residents are overweight/obese [4]. In addition, China has a heavy burden of gastric cancer (GC), with 479,000 new cases and 374,000 deaths in 2020, accounting for 43.94% and 48.62% of the world’s total, respectively, and ranking high in the world [5]. Owing to the implementation of cancer prevention and control strategies, the overall disease burden of early-onset esophageal cancer (EC) in China has shown a downward trend. However, it is still higher than the global level [6]. Data show that the burden of EC disease due to high BMI increased significantly in 2019 and is expected to continue to increase in 2030 [7]. Helicobacter pylori (H. pylori) is a unique gram-negative bacterium closely related to various digestive system diseases [8–11]such as gastritis, gastric ulcers and esophageal disease, and is the most important infectious factor of GC [12]. Moreover, the relationship between obesity and H. pylori, another significant health problem, has gradually been revealed [13–15]. Therefore, an in-depth exploration of the relationships between H. pylori infection and GC, obesity and esophageal diseases is highly important for reducing the burden of NCDs such as GC, obesity and esophageal diseases and guiding their etiological prevention. However, owing to the influence of research methods, sample size, regional differences and other factors, there is no definitive conclusion on the relationships among H. pylori infection, obesity and esophageal diseases, and the specific mechanism of H. pylori infection leading to GC is still unclear [16]– [17]. Therefore, this study aims to systematically evaluate the correlation between H. pylori infection and the risk of GC, obesity and esophageal diseases by collecting existing relevant studies extracting critical data, to provide a theoretical basis and update the current evidence base for etiologic prevention and treatment and H. pylori eradication strategies for GC, obesity, and esophageal diseases. Materials and methods The systematic review and meta-analysis were conducted according to the recommendations in the PRISMA (Priority Reporting Program for Systematic Review and Meta-Analysis) statement. Data sources and search strategies We searched 7 domestic and foreign electronic databases, including PubMed, Embase, Web of Science, the Cochrane Library, ClinicalTrials.gov, CNKI.com, and the Wanfang Database. The search was available until June 27, 2023, for population studies only, with no restrictions on study type or study population type (see Supplementary Table 1 for detailed search strategies). In addition, we also searched domestic and foreign gray literature databases, including the Graynet, OpenSIGLE database, China Doctoral dissertation full-text database and China Master Dissertation full-text database. Selection and eligibility criteria Studies were included if they met all of the following requirements. ① Prospective studies, cohort studies, case‒control studies, cross-sectional studies or randomized controlled trial; ② the total sample size was greater than 100; and ③ the exposure factor was H. pylori infection; ④the H. pylori detection method should be objective, which encompassed methods like rapid urea breath tests, ELISA and endoscopy. ⑤the study reported effect sizes such as OR, HR and 95% confidence intervals (CI). Studies were excluded if any of the following criteria were met. ① Duplicate literature; ② irrelevant document; ③ systematic review, meta-analysis, research protocols, trial registration, conference abstracts, letters, dissertations and other irrelevant genres; ④ nonprospective study, noncohort study, noncase-control study or noncross-sectional study; ⑤ effect size or 95% CI was not reported. Research screening is divided into two stages: first, all the titles and abstracts of the literature are screened, and then the full texts of the relevant literature are reviewed. In both stages, any differences were resolved through discussion. The selection process is shown in Fig. 1.Fig. 1Flow chart of the selection of studies for the meta-analysis Flow chart of the selection of studies for the meta-analysis Data extraction The data were extracted and independently checked by three team members via a predesigned table. Any differences and inconsistencies were resolved through discussion among the three members of the group. The extracted data included the first author, time of publication, region, type of study, age of the subjects, sample size, control source, match status, H. pylori detection method, exposure definition, and adjustment for confounders. Quality assessment Two investigators independently evaluated the quality of the included studies and conducted a group discussion with a third investigator to resolve differences. The Newcastle‒Ottawa scale (NOS) was used to evaluate the risk of bias in the included studies, and the NOS assigned 9 points to the least risk of bias in the three fields. Among them, 4 points were used for the selection of the study population, including the representation of the exposed population, the selection of the nonexposed population, the determination of exposure factors, and whether there were clear research indicators at the beginning of the study; 2 points were used for intergroup comparability, including the comparability of two groups considered in the design analysis; and 3 points were used for exposure and outcome determination, for case‒control studies and cohort studies, consisting of three parts: outcome measurement, whether follow-up was long enough, and whether follow-up was adequate. On the 9-point scale, higher scores indicate better research quality. Data analysis and synthesis R software was used for data analysis. The I2 statistic was used to evaluate the heterogeneity of the combined study results. A random effects model was used to estimate the combined OR and 95% CI, weighted to the reciprocal of the variance. Values with I2 > 75% reported considerable heterogeneity, which was considered significant when P < 0.10. Sources of heterogeneity were assessed by grouping the results according to predefined categorical variables. For heterogeneity beyond subgroup and sensitivity analyses, heterogeneity can be understood and quantified through mixed effects models and meta-regression analyses. Forest plots were used to visualize the effect sizes and 95% CI of the included studies. A funnel plot was used to identify potential publication bias, and Egger’s linear regression test was used to measure the asymmetry of the funnel plot, where P < 0.10 was set as the significance level. Subgroup analyses were performed for different characteristics, such as region, type of study, H. pylori detection method, control source, match, literature quality, and exposure definition, to assess possible sources of heterogeneity. All tests were two-sided, and P < 0.05 was considered statistically significant. Results Results of the literature search and screening A total of 4770 studies were retrieved from each database. After deleting duplicate studies, a total of 4200 studies were obtained. Case reports, reviews, letters, conference papers and other irrelevant genres were deleted, and a total of 3388 reviews/meta-analyses, abstracts and literature unrelated to the current research purpose were deleted after the titles and abstracts were read in detail. The remaining 812 pieces were subsequently read in detail to complete further screening. According to the inclusion and exclusion criteria, 772 studies were excluded one by one for the following reasons: nonpurpose study type (n = 183); the study object was nonpopulated (n = 104); the sample size was less than 100 (n = 112); the exposure or outcome did not meet the requirements (n = 149); H. pylori infection was not tested objectively (n = 0); the effect size or 95% CI for analysis was not recorded (n = 110); the full text was not available (n = 109); and the study population from the same cohort (n = 5). Finally, 40 studies meeting the requirements were selected. Basic features included in the study Finally, 40 papers were included, 34 in English and 6 in Chinese. Among them, 15, 4 and 21 studies related to H. pylori and GC, obesity and the esophagus, respectively, were published between 2000 and 2023. Twenty-one were case‒control studies, and 19 were cohort studies. Thirteen studies were conducted in China, eight in the United States, four in South Korea, four in Japan, one in a multicountry consortium, and one each in Germany, Israel, Thailand, Ireland, Iran, Malaysia, Sweden, Azerbaijan, Australia, and Brazil. All studies met the inclusion-exclusion criteria and provided multifactor adjusted effect sizes and 95% CI. The average NOS score of the included studies was 6. There were 13 high-quality studies and 27 medium-quality studies. The average NOS score in the literature related to GC was 5.73 points, that related to obesity was 6.5 points, and those related to EC and gastroesophageal reflux disease (GERD) were 6.88 points and 5.62 points, respectively. Meta-analysis results H. pylori infection and GC A total of 15 studies related to H. pylori infection and GC were included. The studies were divided according to their exposure definitions into H. pylori infection and GC risk (n = 5) and H. pylori eradication therapy and GC risk (n = 10). Analysis of the correlation between H. pylori infection and the risk of GC As shown in Fig. 2, the combined effect size analysis of the conclusions of 5 studies on the correlation between H. pylori infection and the risk of GC under the random effects model revealed that H. pylori infection increased the risk of GC [OR (95% CI) = 2.68 (1.85 ~ 3.89); P < 0.0001; I2 = 0%, P = 0.49].Fig. 2Forest plot of random-effects meta-analysis examining the association between Helicobacter pylori infection and gastric cancer risk. Individual study results (squares) with corresponding 95% confidence intervals (horizontal lines); Study weights (percentage values) reflecting their contribution to the pooled estimate; The combined effect size OR(diamond) was 2.68 with 95% confidence interval [1.85; 3.89] and prediction interval [1.46; 4.92] shown below the pooled estimate. Heterogeneity measures: I²= 0% (p = 0.49), indicating negligible between-study heterogeneity Forest plot of random-effects meta-analysis examining the association between Helicobacter pylori infection and gastric cancer risk. Individual study results (squares) with corresponding 95% confidence intervals (horizontal lines); Study weights (percentage values) reflecting their contribution to the pooled estimate; The combined effect size OR(diamond) was 2.68 with 95% confidence interval [1.85; 3.89] and prediction interval [1.46; 4.92] shown below the pooled estimate. Heterogeneity measures: I²= 0% (p = 0.49), indicating negligible between-study heterogeneity Sensitivity analysis and publication bias analysis results Sensitivity analysis revealed no substantial change in the combined OR values or 95% CI under the random effects model (Fig. 3). The funnel plot was symmetric (Fig. 4), and Begg’s test suggested no significant publication bias in this meta-analysis (P = 0.2207 > 0.05).Fig. 3Sensitivity analysis results of the correlation between H. pylori infection and the risk of GC. Each row of data indicates the recalculated combined effect size (Odds Ratio, OR) with its 95% confidence interval (95% CI), p-value, heterogeneity statistic (Tau² and Tau), and the percentage of heterogeneity (I²) after excluding a particular study. After excluding any individual study, none of the confidence intervals for the ORs crossed 1 (all showed statistical significance) and the ORs varied over a small range (2.47-3.31), suggesting that the combined results are less dependent on individual studies and that the conclusions are more robust. The combined effect size OR was 2.68 (95% CI: 1.85-3.89), p < 0.01, and heterogeneity was 0 (Tau²= 0, I²= 0%), suggesting a high degree of inter-study consistencyFig. 4Funnel plot of the association between H. pylori infection and the risk of GC. The funnel plot has odds ratio (OR) as the horizontal axis and standard error (SE) as the vertical axis. The graph is graphically symmetrical, suggesting a low risk of publication bias Sensitivity analysis results of the correlation between H. pylori infection and the risk of GC. Each row of data indicates the recalculated combined effect size (Odds Ratio, OR) with its 95% confidence interval (95% CI), p-value, heterogeneity statistic (Tau² and Tau), and the percentage of heterogeneity (I²) after excluding a particular study. After excluding any individual study, none of the confidence intervals for the ORs crossed 1 (all showed statistical significance) and the ORs varied over a small range (2.47-3.31), suggesting that the combined results are less dependent on individual studies and that the conclusions are more robust. The combined effect size OR was 2.68 (95% CI: 1.85-3.89), p < 0.01, and heterogeneity was 0 (Tau²= 0, I²= 0%), suggesting a high degree of inter-study consistency Funnel plot of the association between H. pylori infection and the risk of GC. The funnel plot has odds ratio (OR) as the horizontal axis and standard error (SE) as the vertical axis. The graph is graphically symmetrical, suggesting a low risk of publication bias Analysis of the correlation between H. pylori eradication therapy and the risk of GC As shown in Fig. 5, under the random effects model, the combined effect size analysis of the conclusions of 10 studies related to the eradication of H. pylori infection and the risk of GC revealed that the eradication of H. pylori infection could reduce the risk of GC [OR (95% CI) = 2.68 (1.85 ~ 3.89); P < 0.0001; I2 = 0%, P = 0.49].Fig. 5Meta-analysis results of the association between the eradication of H. pylori infection and the risk of GC. Individual study results (squares) with corresponding 95% confidence intervals (horizontal lines); Study weights (percentage values) reflecting their contribution to the pooled estimate; The combined effect size OR(diamond) was 0.66 with 95% confidence interval[0.52; 0.84] and prediction interval[0.32; 1.36] shown below the pooled estimate. Heterogeneity measures: I² = 77% (p <0.01), indicating large differences between studies Meta-analysis results of the association between the eradication of H. pylori infection and the risk of GC. Individual study results (squares) with corresponding 95% confidence intervals (horizontal lines); Study weights (percentage values) reflecting their contribution to the pooled estimate; The combined effect size OR(diamond) was 0.66 with 95% confidence interval[0.52; 0.84] and prediction interval[0.32; 1.36] shown below the pooled estimate. Heterogeneity measures: I² = 77% (p <0.01), indicating large differences between studies Results of subgroup analysis Subgroup analysis was conducted on the basis of study type, study quality, H. pylori detection methods and national and regional differences. The results revealed that, in randomized controlled trials, an NOS score ≥ 7 for study quality, a rapid urease test, ELISA and endoscopy as the detection methods for H. pylori infection, and in Korea and Japan, the eradication of H. pylori infection was associated with the risk of GC to some extent, as shown in Table 1.Table 1Subgroup analysis of the association between H. pylori infection eradication and the risk of GCItemsGroupsNumber of studiesOR value95%CII2 value (%)P valueStudy type< 0.01RCT40.500.38 ~ 0.6500.95Cohort study61.010.94 ~ 1.10650.01Study quality0.05≥ 710.370.14 ~ 0.97--< 790.960.89 ~ 1.0477< 0.01H pylori Testing Methods< 0.01Rapid urease test20.540.36 ~ 0.8200.76ELISA10.480.32 ~ 0.71--endoscopy30.480.31 ~ 0.7500.82Nonserological testing20.940.63 ~ 1.42770.04H pylori Prescription Treatment21.040.95 ~ 1.13570.13Geographical differences< 0.01China31.000.92 ~ 1.0989< 0.01South Korea30.630.46 ~ 0.8640.35Japan20.500.29 ~ 0.8500.55USA20.940.63 ~ 1.42770.04 Table 1Subgroup analysis of the association between H. pylori infection eradication and the risk of GCItemsGroupsNumber of studiesOR value95%CII2 value (%)P valueStudy type< 0.01RCT40.500.38 ~ 0.6500.95Cohort study61.010.94 ~ 1.10650.01Study quality0.05≥ 710.370.14 ~ 0.97--< 790.960.89 ~ 1.0477< 0.01H pylori Testing Methods< 0.01Rapid urease test20.540.36 ~ 0.8200.76ELISA10.480.32 ~ 0.71--endoscopy30.480.31 ~ 0.7500.82Nonserological testing20.940.63 ~ 1.42770.04H pylori Prescription Treatment21.040.95 ~ 1.13570.13Geographical differences< 0.01China31.000.92 ~ 1.0989< 0.01South Korea30.630.46 ~ 0.8640.35Japan20.500.29 ~ 0.8500.55USA20.940.63 ~ 1.42770.04 Subgroup analysis of the association between H. pylori infection eradication and the risk of GC Meta-regression analysis was used to evaluate the factors influencing the heterogeneity of the studies. The results revealed that national and regional differences (P = 0.1155) and study quality (P = 0.0812) were not potential risk confounders of heterogeneity, whereas study type (P = 0.0424) and H. pylori infection detection methods (P = 0.0312) were. Sensitivity analysis and publication bias analysis results Sensitivity analysis revealed no substantial change in the combined OR value or 95% CI under the random effects model (Fig. 6). The funnel plot was symmetric (Fig. 7), and Begg’s test suggested no significant publication bias in this meta-analysis (P = 0.2105 > 0.05).Fig. 6Sensitivity analysis of the association between H. pylori infection eradication and the risk of GC. Each row of data indicates the recalculated combined effect size (Odds Ratio, OR) with its 95% confidence interval (95% CI), p-value, heterogeneity statistic (Tau² and Tau), and the percentage of heterogeneity (I²) after excluding a particular study. After excluding any individual study, none of the confidence intervals for the ORs crossed 1 (all showed statistical significance) and the ORs varied over a small range (0.60-0.69), suggesting that the combined results are less dependent on individual studies and that the conclusions are more robust. The combined effect size OR was 0.66 (95% CI: 0.52-0.84), p < 0.01, suggesting that original conclusions are reliable. Tau²= 0.0824, I²= 77%, suggesting that high inter-study heterogeneity and significant differences between studiesFig. 7Funnel plot of the association between H. pylori infection eradication and the risk of GC. The funnel plot has odds ratio (OR) as the horizontal axis and standard error (SE) as the vertical axis. The graph is graphically symmetrical, suggesting a low risk of publication bias Sensitivity analysis of the association between H. pylori infection eradication and the risk of GC. Each row of data indicates the recalculated combined effect size (Odds Ratio, OR) with its 95% confidence interval (95% CI), p-value, heterogeneity statistic (Tau² and Tau), and the percentage of heterogeneity (I²) after excluding a particular study. After excluding any individual study, none of the confidence intervals for the ORs crossed 1 (all showed statistical significance) and the ORs varied over a small range (0.60-0.69), suggesting that the combined results are less dependent on individual studies and that the conclusions are more robust. The combined effect size OR was 0.66 (95% CI: 0.52-0.84), p < 0.01, suggesting that original conclusions are reliable. Tau²= 0.0824, I²= 77%, suggesting that high inter-study heterogeneity and significant differences between studies Funnel plot of the association between H. pylori infection eradication and the risk of GC. The funnel plot has odds ratio (OR) as the horizontal axis and standard error (SE) as the vertical axis. The graph is graphically symmetrical, suggesting a low risk of publication bias H. pylori infection and obesity Analysis of the correlation between H. pylori infection and risk of obesity The results of the correlation analysis between H. pylori infection and the risk of obesity are shown in Fig. 8. Under the random effects model, the combined effect size analysis of the conclusions of 4 related studies revealed that H. pylori infection could reduce the risk of obesity [OR (95% CI) = 0.65 (0.43 ~ 0.98); P = 0.0380; I2 = 0%, P = 0.72].Fig. 8Meta-analysis results of the association between H. pylori infection and the risk of obesity. Individual study results (squares) with corresponding 95% confidence intervals (horizontal lines); Study weights (percentage values) reflecting their contribution to the pooled estimate; The combined effect size OR(diamond) was 0.65 with 95% confidence interval[0.43; 0.98] and prediction interval[0.05; 9.07] shown below the pooled estimate. Heterogeneity measures: I² = 0% (p =0.72), indicating negligible between-study heterogeneity Meta-analysis results of the association between H. pylori infection and the risk of obesity. Individual study results (squares) with corresponding 95% confidence intervals (horizontal lines); Study weights (percentage values) reflecting their contribution to the pooled estimate; The combined effect size OR(diamond) was 0.65 with 95% confidence interval[0.43; 0.98] and prediction interval[0.05; 9.07] shown below the pooled estimate. Heterogeneity measures: I² = 0% (p =0.72), indicating negligible between-study heterogeneity Sensitivity analysis and publication bias analysis results Sensitivity analysis revealed no substantial change in the combined OR value or 95% CI under the random effects model (Fig. 9). The funnel plot was symmetric (Fig. 10), and Begg’s test suggested no significant publication bias in this meta-analysis (P = 1.00 > 0.05).Fig. 9Sensitivity analysis results of the correlation between H. pylori infection and the risk of obesity. Each row of data indicates the recalculated combined effect size (Odds Ratio, OR) with its 95% confidence interval (95% CI), p-value, heterogeneity statistic (Tau² and Tau), and the percentage of heterogeneity (I²) after excluding a particular study. The all ORs showed statistical significance and varied over a small range (0.63-0.71), suggesting that the combined results are less dependent on individual studies and that the conclusions are more robust. The combined effect size OR was 0.65 (95% CI: 0.43-0.98), p =0.04, suggesting that original conclusions are reliable. Heterogeneity was 0 (Tau²= 0, I²= 0%), suggesting a high degree of inter-study consistencyFig. 10Funnel plot of H. pylori infection and the risk of obesity. The funnel plot has odds ratio (OR) as the horizontal axis and standard error (SE) as the vertical axis. The graph is graphically symmetrical, suggesting a low risk of publication bias Sensitivity analysis results of the correlation between H. pylori infection and the risk of obesity. Each row of data indicates the recalculated combined effect size (Odds Ratio, OR) with its 95% confidence interval (95% CI), p-value, heterogeneity statistic (Tau² and Tau), and the percentage of heterogeneity (I²) after excluding a particular study. The all ORs showed statistical significance and varied over a small range (0.63-0.71), suggesting that the combined results are less dependent on individual studies and that the conclusions are more robust. The combined effect size OR was 0.65 (95% CI: 0.43-0.98), p =0.04, suggesting that original conclusions are reliable. Heterogeneity was 0 (Tau²= 0, I²= 0%), suggesting a high degree of inter-study consistency Funnel plot of H. pylori infection and the risk of obesity. The funnel plot has odds ratio (OR) as the horizontal axis and standard error (SE) as the vertical axis. The graph is graphically symmetrical, suggesting a low risk of publication bias H. pylori infection and the esophageal diseases Analysis of the correlation between H. pylori infection and the risk of EC The results of the correlation analysis between H. pylori infection and the risk of EC are shown in Fig. 11. Under the random effects model, the combined effect size analysis of the conclusions of 8 relevant studies revealed that H. pylori infection was not significantly associated with the risk of EC [OR (95% CI) = 0.90 (0.47 ~ 1.73); P = 0.7454; I2 = 88%, P < 0.01].Fig. 11Meta-analysis results of the association betweenH. pylori infection and the risk of EC. Individual study results (squares) with corresponding 95% confidence intervals (horizontal lines); Study weights (percentage values) reflecting their contribution to the pooled estimate; The combined effect size OR(diamond) was 0.90 with 95% confidence interval[0.47; 1.73] and prediction interval[0.09; 9.25] shown below the pooled estimate. Heterogeneity measures: I² = 88% (p <0.01), indicating large differences between studies Meta-analysis results of the association betweenH. pylori infection and the risk of EC. Individual study results (squares) with corresponding 95% confidence intervals (horizontal lines); Study weights (percentage values) reflecting their contribution to the pooled estimate; The combined effect size OR(diamond) was 0.90 with 95% confidence interval[0.47; 1.73] and prediction interval[0.09; 9.25] shown below the pooled estimate. Heterogeneity measures: I² = 88% (p <0.01), indicating large differences between studies Sensitivity analysis and publication bias analysis Sensitivity analysis revealed no substantial change in the combined OR value or 95% CI under the random effects model (Fig. 12). The funnel plot was symmetric (Fig. 13), and Begg’s test suggested no significant publication bias in this meta-analysis (P = 0.7105 > 0.05).Fig. 12Sensitivity analysis results of the correlation between H. pylori infection and the risk of EC. Each row of data indicates the recalculated combined effect size (Odds Ratio, OR) with its 95% confidence interval (95% CI), p-value, heterogeneity statistic (Tau² and Tau), and the percentage of heterogeneity (I²) after excluding a particular study. The combined effect size OR was 0.90 (95% CI: 0.47-1.73), p =0.75, suggesting that none of the statistically significant. Tau²= 0.7971, I²= 88%, suggesting that high inter-study heterogeneityFig. 13Funnel plot of the association between H. pylori infection and the risk of EC. The funnel plot has odds ratio (OR) as the horizontal axis and standard error (SE) as the vertical axis. The graph is graphically symmetrical, suggesting a low risk of publication bias Sensitivity analysis results of the correlation between H. pylori infection and the risk of EC. Each row of data indicates the recalculated combined effect size (Odds Ratio, OR) with its 95% confidence interval (95% CI), p-value, heterogeneity statistic (Tau² and Tau), and the percentage of heterogeneity (I²) after excluding a particular study. The combined effect size OR was 0.90 (95% CI: 0.47-1.73), p =0.75, suggesting that none of the statistically significant. Tau²= 0.7971, I²= 88%, suggesting that high inter-study heterogeneity Funnel plot of the association between H. pylori infection and the risk of EC. The funnel plot has odds ratio (OR) as the horizontal axis and standard error (SE) as the vertical axis. The graph is graphically symmetrical, suggesting a low risk of publication bias Analysis of the correlation between H. pylori infection and the risk of GERD The results of the correlation analysis between H. pylori infection and the risk of GERD are shown in Fig. 14. In the random effects model, a combined effect size analysis of 13 relevant studies revealed that H. pylori infection could reduce the risk of GERD [OR (95% CI) = 0.54 (0.43 ~ 0.68); P < 0.0001; I2 = 72%, P < 0.01].Fig. 14Meta-analysis results of the association between H. pylori infection and the risk of GERD. Individual study results (squares) with corresponding 95% confidence intervals (horizontal lines); Study weights (percentage values) reflecting their contribution to the pooled estimate; The combined effect size OR(diamond) was 0.54 with 95% confidence interval[0.43; 0.68] and prediction interval[0.24; 1.19] shown below the pooled estimate. Heterogeneity measures: I²= 72% (p <0.01), indicating large differences between studies Meta-analysis results of the association between H. pylori infection and the risk of GERD. Individual study results (squares) with corresponding 95% confidence intervals (horizontal lines); Study weights (percentage values) reflecting their contribution to the pooled estimate; The combined effect size OR(diamond) was 0.54 with 95% confidence interval[0.43; 0.68] and prediction interval[0.24; 1.19] shown below the pooled estimate. Heterogeneity measures: I²= 72% (p <0.01), indicating large differences between studies Results of subgroup analysis Subgroup analysis was carried out on the basis of study type, study quality, source of the control group and national and regional differences. The results revealed that the study was a matched/unmatched case‒control study with NOS scores ≥ 7 and < 7 for study quality. Controls were taken from hospital or community populations; in countries such as China, the United States, Australia and South Korea, H. pylori infection was associated with the risk of GERD to some extent, as shown in Table 2. Table 2Results of the subgroup analysis of the association between H. pylori infection and the risk of GERDItemsGroupsNumber of studiesOR value95%CII2 value (%)P valueStudy type0.66Matched case‒control study40.500.45 ~ 0.55-< 0.01Unmatched case‒control study90.520.44 ~ 0.63270.21Study quality< 0.01≥ 730.440.40 ~ 0.49770.01< 7100.620.54 ~ 0.72570.01Source of controls0.83Hospital-based80.510.46 ~ 0.5680< 0.01Community-based30.460.34 ~ 0.63770.01Geographical differences< 0.01China30.660.55 ~ 0.8083< 0.01Malaysia10.560.31 ~ 0.99--Sweden10.900.50 ~ 1.60--Azerbaijan11.100.61 ~ 1.96--USA40.440.34 ~ 0.57290.24Australia10.470.27 ~ 0.81--South Korea10.440.39 ~ 0.49--Brazil10.810.41 ~ 1.60-- Table 2Results of the subgroup analysis of the association between H. pylori infection and the risk of GERDItemsGroupsNumber of studiesOR value95%CII2 value (%)P valueStudy type0.66Matched case‒control study40.500.45 ~ 0.55-< 0.01Unmatched case‒control study90.520.44 ~ 0.63270.21Study quality< 0.01≥ 730.440.40 ~ 0.49770.01< 7100.620.54 ~ 0.72570.01Source of controls0.83Hospital-based80.510.46 ~ 0.5680< 0.01Community-based30.460.34 ~ 0.63770.01Geographical differences< 0.01China30.660.55 ~ 0.8083< 0.01Malaysia10.560.31 ~ 0.99--Sweden10.900.50 ~ 1.60--Azerbaijan11.100.61 ~ 1.96--USA40.440.34 ~ 0.57290.24Australia10.470.27 ~ 0.81--South Korea10.440.39 ~ 0.49--Brazil10.810.41 ~ 1.60-- Results of the subgroup analysis of the association between H. pylori infection and the risk of GERD Meta regression Meta-regression analysis was used to evaluate the factors influencing the heterogeneity between studies. The results revealed that study type (P = 0.1326), study quality (P = 0.1261), control source (P = 0.1471) and national and regional differences (P = 0.1542) were not identified as potential risk confounders of heterogeneity. Sensitivity analysis and publication bias analysis results Sensitivity analysis revealed no substantial change in the combined OR value or 95% CI under the random effects model (Fig. 15). The funnel plot was symmetric (Fig. 16), and Begg’s test suggested no significant publication bias in this meta-analysis (P = 0.6693 > 0.05).Fig. 15Sensitivity analysis of the association between H. pylori infection and the risk of GERD. Each row of data indicates the recalculated combined effect size (Odds Ratio, OR) with its 95% confidence interval (95% CI), p-value, heterogeneity statistic (Tau² and Tau), and the percentage of heterogeneity (I²) after excluding a particular study. After excluding any individual study, none of the confidence intervals for the ORs crossed 1 (all showed statistical significance) and the ORs varied over a small range (0.51-0.57), suggesting that the combined results are less dependent on individual studies and that the conclusions are more robust. The combined effect size OR was 0.54 (95% CI: 0.43-0.68), p < 0.01, suggesting that original conclusions are reliable. Tau²= 0.1160, I²= 72%, suggesting that moderate to high inter-study heterogeneityFig. 16Funnel plot of the association between H. pylori infection and the risk of GERD. The funnel plot has odds ratio (OR) as the horizontal axis and standard error (SE) as the vertical axis. The graph is graphically symmetrical, suggesting a low risk of publication bias Sensitivity analysis of the association between H. pylori infection and the risk of GERD. Each row of data indicates the recalculated combined effect size (Odds Ratio, OR) with its 95% confidence interval (95% CI), p-value, heterogeneity statistic (Tau² and Tau), and the percentage of heterogeneity (I²) after excluding a particular study. After excluding any individual study, none of the confidence intervals for the ORs crossed 1 (all showed statistical significance) and the ORs varied over a small range (0.51-0.57), suggesting that the combined results are less dependent on individual studies and that the conclusions are more robust. The combined effect size OR was 0.54 (95% CI: 0.43-0.68), p < 0.01, suggesting that original conclusions are reliable. Tau²= 0.1160, I²= 72%, suggesting that moderate to high inter-study heterogeneity Funnel plot of the association between H. pylori infection and the risk of GERD. The funnel plot has odds ratio (OR) as the horizontal axis and standard error (SE) as the vertical axis. The graph is graphically symmetrical, suggesting a low risk of publication bias Discussion This study conducted a meta-analysis to explore the correlation between H. pylori infection and the risk of GC, obesity, EC and GERD and updated the existing evidence-based medical evidence. The main results suggested that H. pylori infection increased the risk of GC [OR (95% CI) = 2.68 (1.85 ~ 3.89); P < 0.0001] but did not increased the risk of obesity [OR (95% CI) = 0.65 (0.43 ~ 0.98); P = 0.0380] and GERD [OR (95% CI) = 0.54 (0.43 ~ 0.68); P < 0.0001] and had no significant correlation with the risk of EC [OR (95% CI) = 0.90 (0.47 ~ 1.73); P = 0.7454]. There was considerable heterogeneity among studies for some diseases, but sensitivity analysis proved that the results were stable and reliable, and no significant publication bias was found. H. pylori and GC In this study, a comprehensive collection of studies related to H. pylori infection and GC was conducted by using a random effects model for comprehensive analysis. The combined OR value and 95% CI were 2.68 (1.85 ~ 3.89), indicating that the risk of GC in H. pylori-infected individuals was more than two times that of uninfected individuals, confirming the close relationship between H. pylori and GC. The results of this study are consistent with those of Poorolajal J et al. [18]. The specific mechanism by which H. pylori infection leads to GC is not clear, but some studies have revealed a possible pathogenic mechanism. H. pylori has a unique spiral shape and the structure of the whip, enabling it to quickly swim to the deep gastric mucosa in the space of gastric epithelial cells to avoid immunity and colonize the stomach through the outer membrane protein on the surface of the bacteria, secrete urease and stomach acid to counteract the adverse effects of the acidic environment on its survival, and survive in the stomach for a long time, thus triggering gastric mucosal inflammation [19]. According to the cascade model of GC progression proposed by Correa and Haenszel [20]acute gastritis gradually transforms into chronic gastritis starting from the inflammatory reaction of the gastric mucosa, and then, in the long-term inflammatory process, the gastric mucosa is repeatedly damaged and repaired, then develops atrophic gastritis, intestinal metaplasia or dysplasia, and finally leads to the occurrence of GC. In addition, urease secreted by H. pylori during its survival in the stomach can induce the formation of blood vessels [21]and the ammonia produced by urea hydrolysis can also lead to cytotoxic effects [22] such as apoptosis, injury and inflammation, suggesting that the urease secreted by H. pylori can also promote the development of GC. Various virulence factors of H. pylori, such as vacuolating cytotoxin A (VacA), cytotoxin-associated gene A (CagA) and some membrane proteins, are essential factors that cause GC [23]. These virulence factors can directly damage the gastric mucosa and affect the expression of tumor suppressor genes by changing multiple signaling pathways in gastric epithelial cells, coupled with the interactive regulation of patients’ intrinsic genetic characteristics and external environmental factors, thus leading to the occurrence of GC [24]. Therefore, preventing H. pylori infection or eradicating H. pylori infection is an effective measure to prevent GC. The results of this study also show that eradicating H. pylori infection can effectively reduce the risk of GC. This result is consistent with other studies [25]– [26]. Previous studies have demonstrated that CagA-positive strains exhibit greater carcinogenicity. In this study, the results were not stratified by strain type because the meta-analysis pooled data from existing studies that may not have differentiated between strain types, or the distribution of strains was inconsistent across studies, making it difficult to assess strain heterogeneity. Future studies need to incorporate molecular epidemiological approaches to analyze the effects of different strains more finely in order to optimize prevention and individualized treatment strategies. Subgroup analyses identified H. pylori infection detection methods as potential risk confounders for heterogeneity, with different tests detecting current and past infection status. For example, studies using serology may misclassify some cured individuals as infected (since antibodies can persist), whereas urea breath tests and fecal antigen tests confirm active infection. Such differences may lead to heterogeneity in the associations observed. H. pylori and obesity However, studies on the role of H. pylori infection in overweight/obese individuals are inconsistent. This study revealed that human infection with H. pylori do not increase the risk of obesity, this finding is consistent with some of the epidemiological studies [17, 27]. However, a meta-analysis of 18 observational studies showed that H. pylori-positive patients had a slightly higher BMI [28]. One study reported no significant correlation between H. pylori infection and obesity [16, 29]. This inconsistency may arise from population genetic background, environmental exposures (dietary structure, history of antibiotic use), less relevant studies, or interference from differences in virulence of H. pylori strains (CagA-positive versus negative strains). Thus, the current conclusions may only reflect localized associations under specific subpopulations or study designs. The study revealed that the mechanism by which H. pylori infection affects obesity may be related to several factors. First, the stomach is a digestive organ that stores, digests, and secretes, peristaltic emptying and other functions. It is closely related to human growth and development. H. pylori colonizes the stomach after continuous infection and causes a series of inflammatory reactions affecting the primary function of the stomach, affecting the digestion and absorption of food and resulting in the balance of energy intake and consumption being disrupted, thus affecting body weight [30]. Second, the stomach mucosa synthesizes and secretes peptides that regulate appetite and satiety, such as ghrelin and obestatin [31]. Ghrelin stimulates appetite and regulates energy homeostasis, which is conducive to obesity, so it can also be called a “fattening peptide“ [32]. Obestatin has the opposite effect as ghrelin; it can antagonize the appetite-stimulating effect of ghrelin, reduce food intake, inhibit jejunum contraction, promote gastric secretion, and control weight gain [33]. Studies have shown that H. pylori infection affects body weight by affecting ghrelin and obesity [33, 34]. In addition, H. pylori infection may also affect body weight through the immune system. Some studies have shown that preadipocytes and macrophages can secrete factors involved in the inflammatory response. When the ability of monocytes to transform into macrophages is reduced in obese patients, preadipocytes can act as macrophage-like cells to phagocytose microorganisms [35]. These findings suggest that H. pylori infection may stimulate adipose tissue to participate in immune processes, affect the metabolism of adipose tissue and increase energy consumption, thus affecting body weight [15]. H. pylori and the esophageal disease The relationship between H. pylori infection and the esophageal disease can be divided into EC and GERD. The results of this study suggest that H. pylori infection can reduce the risk of GERD [OR (95% CI) = 0.54 (0.43 ~ 0.68); P < 0.0001]; however, there was no significant correlation with the risk of EC [OR (95% CI) = 0.90 (0.47 ~ 1.73); P = 0.7454]. The results of this study are consistent with those of other meta-analyses [36–39]. In conclusion, H. pylori infection can reduce the risk of GRED, and some researchers have proposed a possible mechanism. GERD has a complex and multifactorial pathogenesis involving reflux exposure, epithelial resistance, visceral sensitivity changes, etc. It is associated with gastric acid exposure, an impaired reflux barrier, reduced lower esophageal sphincter pressure (LESP), and a reduced ability of the esophagus to clear and buffer reflux [40]. H. pylori infection may affect GERD through the above factors. A prospective clinical study on the effect of H. pylori infection on GERD revealed that host infection with H. pylori increases serum gastrin secretion by stimulating vagus receptors in the gastric fundus and cardia, thereby increasing lower esophageal sphincter (LES) pressure, reducing the reflux of gastric acid and stomach contents, and reducing esophageal acid exposure. To exert preventive or protective effects on GERD [41]. In addition, Kandulski et al. proposed that this protective effect may also be because H. pylori colonizes the stomach after infecting the host and causes damage to the gastric mucosa. Structural and functional changes in the gastric mucosa lead to atrophy of the gastric mucosa and the obstruction of acid production, which reduces the reflux of gastric contents and the damage caused by gastric acid to the esophagus, thus producing a protective effect [42]. Heterogeneity analysis The results of the subgroup analysis revealed that some interstudy heterogeneity may be related to the types of included studies, the background of the subjects, and the H. pylori detection methods. Meta-regression identified potential risk factors for heterogeneity, and no potential risk factors for heterogeneity were found in the correlation analysis between H. pylori and obesity and the esophageal disease. The types of studies included and the H. pylori detection methods could be identified as potential risk confounders for heterogeneity in the analysis of the association between H. pylori infection eradication therapy and the risk of GC. The heterogeneity arising from the different types of studies included may be due to different study designs having different biases and errors. Different detection methods for H. pylori lead to heterogeneity, which may be due to the different sensitivities and specificities of different detection methods, resulting in different detection results for H. pylori. Advantages and limitations Compared with previous studies, this study conducted a comprehensive search of relevant databases at home and abroad, including the gray document database. Strict inclusion and exclusion criteria were formulated to screen the literature. The quality of the included studies was higher, all the included observational studies had more significant advantages in adjusting for confounders, and the level of evidence was more substantial. Second, considering the heterogeneity among studies, the random effects model was chosen for pooled analysis, which allows the results of different studies to have a particular variation to more accurately reflect the actual situation and reduce the degree of error caused by heterogeneity. Finally, by implementing comprehensive subgroup analysis and meta-regression, the heterogeneity among the studies was discussed in depth, and the contribution of each factor to the heterogeneity of the studies was judged, which was conducive to a more accurate understanding of the study results and a more reliable and comprehensive correlation conclusion. However, the results of this study still have some shortcomings. First, the literature included observational studies, which, due to the nature of observational studies, only observed correlations and were unable to directly confirm a direct causal relationship between H. pylori and the risk of morbidity. Observational associations are not a substitute for causal evidence, and future causal inference methods such as Mendelian randomization or long-term intervention studies are needed to confirm causation. Second, in the subgroup analysis, the number of studies included in some groups was small, and the meta-analysis results largely depended on the quality and quantity of the included studies. The small number of included studies may affect the stability of the results. Third, although the random effects model is used in this paper and the results of the sensitivity analysis also show that the results are stable and reliable, the risk of heterogeneity arising from the combination of different study types cannot be ignored entirely, but subgroup analysis and meta-regression analysis can be used to understand and quantify the heterogeneity [43]. Fourth, in this study, EC was analyzed as a single outcome, with no distinction made between esophageal adenocarcinoma and squamous cell carcinoma. However, esophageal cancers differed by subtype with respect to pathologic and epidemiologic characteristics. Epidemiologic evidence suggests that H. pylori is diametrically opposed to these two sub-types. The paradox between H. pylori infection and reduced risk of EAC may be due to the role of H. pylori in reducing gastric acidity and attenuating GERD, which is a key factor in the development of Barrett’s esophagus and subsequent EAC. In contrast, the relationship between H. pylori and ESCC remains controversial, with some studies suggesting a positive association, which may be due to H. pylori-induced chronic inflammation or nutritional deficiencies (e.g., vitamin C deficiency) that promote squamous carcinogenesis. Thus, pooling these subtypes may confound opposite effects, leading to lower overall risk estimates. Fifth, the negative correlations between H. pylori and obesity or GERD in this study may be due in part to the dynamic effects of unmeasured confounding factors such as medication use, differences in dietary patterns, and differences in H. pylori strains, and these correlations do not prove that H. pylori infection directly reduces the risk of obesity or GERD: (1) GERD patients may have an increased probability of H. pylori detection and eradication due to endoscopy and medication; (2) high-calorie diets may exacerbate GERD symptoms by increasing gastric acid secretion, while high caloric intake itself leads to obesity; (3) CagA-positive strains lead to reduced gastric acid secretion, which may reduce the risk of GERD, but they are also associated with atrophic gastritis, which may affect body weight. Long-term intervention studies may be needed in the future to confirm causation. Finally, even though a thorough search was done, the file-drawer problem (non-significant results not being published) could subtly influence the meta-analysis, unpublished data were not included and that this might bias results if those missing studies differ in findings. Conclusion The results of this study revealed that H. pylori infection was associated with a significant increase in the risk of GC [OR (95% CI) = 2.68 (1.85 ~ 3.89); P < 0.0001], and the eradication of H. pylori effectively reduced the risk of GC [OR (95% CI) = 0.66 (0.52 ~ 0.84); P = 0.001]. Therefore, the eradication of H. pylori infection is strongly recommended for GC prevention. Although a meta-analysis revealed that H. pylori infection was not associated with a increased risk of obesity [OR (95% CI) = 0.65 (0.43 ~ 0.98); P = 0.0380], the specific mechanism between H. pylori infection and obesity is not fully understood; thus, the available evidence is insufficient to support the prevention or treatment of obesity through the management of H. pylori infection. For patients with GERD, although a meta-analysis revealed that H. pylori infection was associated with a reduced risk of GERD [OR (95% CI) = 0.54 (0.43 ~ 0.68); P < 0.0001], some studies have suggested that H. pylori eradication is a protective factor for GERD [44]and the carcinogenic effect of H. pylori infection should receive more attention at present; therefore, instead of preserving the infection based solely on the potential protective effect of GERD, clinical decision-making should prioritize the assessment of the overall benefit of eradication therapy and the development of a management strategy that takes into account the individual patient’s situation. Finally, with respect to EC, the current meta-analysis did not find a significant association between H. pylori infection and the risk of EC [OR (95% CI) = 0.90 (0.47–1.73); P = 0.7454], and therefore, based on the available evidence, there is insufficient basis to support the prevention of esophageal cancer through eradication of H. pylori. Current evidence shows that H. pylori infection is a definite risk factor for GC, and early eradication of H. pylori is important for reducing the risk of GC and guiding the etiologic prevention of GC. This study also suggests that H. pylori infection may be associated with reduced risk of obesity and GERD, but the level of evidence is limited and the mechanism is not known. Therefore, the core indication for H. pylori eradication remains the reduction the risk of GC and H. pylori-related diseases (e.g., gastritis, gastric ulcer). In patients with obesity or GERD, clinical decision-making needs to be individualized to weigh gastric cancer risk against symptom management rather than preserving H. pylori infection. Moreover, among H. pylori eradication strategies, the universal eradication strategy may bring non-negligible clinical risks and resource wastage. Based on the available evidence, decision-making for eradication treatment should follow the principle of individualization, taking into account the patient’s disease risk, strain characteristics, demographic factors, and the cost-effectiveness of treatment. Supplementary Information Supplementary Material 1. Supplementary Material 1.