Association of metabolic variants with Helicobacter pylori infection risk: A retrospective cross-sectional study

Abstract

Background

Helicobacter pylori infection is widespread worldwide and is associated with gastric cancer, posing global public health challenges. Previous studies have linked dyslipidemia with Helicobacter pylori infection but have generally described overall lipid changes, with limited lipid-profile analysis.

Objective

To examine the associations between calculated lipid ratios and Helicobacter pylori infection among individuals without diabetes.

Methods

Participants without diabetes who underwent the ¹³C-urea breath test were included. We calculated lipid ratios (triglyceride/total cholesterol, triglyceride/low-density lipoprotein cholesterol (LDL-C), triglyceride/high-density lipoprotein cholesterol (HDL-C), LDL-C/HDL-C, non-HDL-C/HDL-C, fasting plasma glucose/HDL-C, uric acid/HDL-C, and higher glycosylated hemoglobin/HDL-C). Analyses included univariate and multivariate binary logistic regression, receiver operating characteristic analysis (area under the curve), restricted cubic spline modeling, and threshold analysis. The stability of the nonlinear relationship was further evaluated through sensitivity analyses using alternative knot configurations.

Results

A total of 400 participants without diabetes (75 with H. pylori infection and 325 without H. pylori infection) were analyzed. Compared with participants without H. pylori infection, patients with H. pylori infection showed significant differences in neutrophil count, fasting plasma glucose, uric acid, total cholesterol, triglyceride, HDL-C, LDL-C, and non-HDL-C (p < 0.05). LDL-C/HDL-C was an independent risk factor for H. pylori infection (p = 0.007). Receiver operating characteristic analysis yielded an area under the curve of 0.713, a Youden index of 0.356, and an optimal cutoff value of 2.24 (sensitivity, 78.7%; specificity, 56.9%). In the primary three-knot model selected by Akaike information criterion, restricted cubic splines analysis indicated a significant, nonlinear association between LDL-C/HDL-C– Helicobacter pylori (p overall < 0.001; p nonlinear ≈ 0.02). This nonlinear trend and the overall inflection pattern remained robust in sensitivity analyses (p overall < 0.001), remaining statistically significant with five knots (p nonlinear = 0.041), although slightly attenuated with four knots (p nonlinear = 0.068). Piecewise logistic regression identified an inflection point at approximately1.94 (likelihood ratio p = 0.006). For LDL-C/HDL-C < 1.94, the risk of infection increased sharply with each unit increase (odds ratio = 68.00, 95% confidence interval: 1.83–2521.00, p = 0.022).

Conclusion

LDL-C/HDL-C is an independent, nonlinear predictor of Helicobacter pylori infection in individuals without diabetes and may serve as a practical biomarker for risk stratification. Implications include lipid-centered prevention through dietary and lifestyle interventions. Larger prospective studies are needed to validate causality and assess applicability across metabolic phenotypes.

Introduction

Helicobacter pylori (H. pylori) is a Gram-negative, spiral-shaped bacterium that typically enters the human body through the oral route and specifically colonizes the gastric epithelium. Once H. pylori successfully colonizes, it is often difficult for the body to spontaneously eradicate the infection, leading to long-term or lifelong infection that is usually accompanied by chronic gastric mucosal inflammation. A large body of research has confirmed a significant association between H. pylori infection and the incidence of gastric cancer, ¹ with approximately 2% of individuals with the infection ultimately developing gastric cancer. ² With the continuous improvement and widespread use of diagnostic methods, H. pylori infection has been found to be prevalent worldwide, especially in developing countries and regions. ³ For example, in areas of China with a high incidence of gastric cancer, an epidemiological survey (using the ¹³C-urea breath test) showed that the prevalence of H. pylori infection among healthy individuals aged 30 to 69 years was 63.4%. ⁴ With the continuous increase in the number of infections and the growing concern regarding H. pylori antibiotic resistance, this has become a significant challenge to global public health systems.

Xie et al. ⁵ conducted a cross-sectional study using data from the 1999–2000 National Health and Nutrition Examination Survey (n = 1146) in the United States. The results showed a significant association between high triglyceride (TG) levels and H. pylori infection in female participants. Shimamoto et al. ⁶ performed a retrospective analysis of lipid levels and H. pylori infection status in 15,679 participants, revealing that H. pylori infection was positively associated with low-density lipoprotein cholesterol (LDL-C), total cholesterol (TC), and triglycerides (TG), but negatively associated with high-density lipoprotein cholesterol (HDL-C).

As research has advanced, studies have increasingly demonstrated the close relationship between lipid levels and H. pylori infection, which is also evident during H. pylori eradication therapy. Martín-Núñez et al. ⁷ found that an unfavorable lipid profile (such as high LDL-C and low HDL-C levels) was associated with lower microbial abundance and increased abundance of Bacteroidetes. H. pylori eradication therapy not only improved serum lipid metabolism but also increased HDL-C levels. Changes in the abundance of specific bacteria may be closely linked to changes in plasma HDL-C levels. Li et al. ⁸ suggested that postoperative HDL-C levels could serve as an independent prognostic factor for gastric cancer. Additionally, Haj et al. ⁹ performed a retrospective analysis of 12,207 individuals with diabetes and found that among male individuals, those with higher glycosylated hemoglobin (HbA1c) levels (>7%) were more likely to have H. pylori infection. These studies further support the association of dyslipidemia and blood glucose levels with H. pylori infection, highlighting the importance of monitoring lipid and glucose levels in the management of H. pylori infection.

Although previous studies have explored the relationship between blood glucose, ⁹ blood lipids,^8,9 and H. pylori infection, most have focused on a single type of lipid parameters. Given the complexity of interactions among different lipid components, reliance on a single lipid marker is insufficient to fully uncover the underlying connection between dyslipidemia and H. pylori infection. Currently, research on the correlation between lipid profiles and H. pylori infection remains relatively limited. Therefore, the aim of this study was to systematically assess the relationship between lipid profiles and H. pylori infection.

Methods

Study population

This retrospective cross-sectional study consecutively enrolled eligible participants who underwent physical examinations at Guangdong Provincial Hospital of Chinese Medicine between January 2024 and April 2024 and completed the ¹³C-urea breath test (gastric motility) as the study subjects. The inclusion criteria for participants were as follows. First, only adults aged ≥18 years were included to ensure the validity of the tests. Second, participants were required to have complete clinical data and no history of diabetes. Participants were excluded if they met any of the following criteria: (a) those currently infected and undergoing antibiotic treatment; (b) those on long-term medication affecting glucose and lipid metabolism (e.g. corticosteroids) or those with other related diseases (e.g. thyroid dysfunction, nephrotic syndrome, or autoimmune diseases); and (c) individuals diagnosed with cancer or with severe liver or kidney dysfunction.

This study was formally reviewed and approved by the Ethics Committee of Guangdong Provincial Hospital of Traditional Chinese Medicine (Guangzhou, Guangdong, China; Approval No. ZE2026-115-01) on 17 March 2026. To protect participant privacy, all clinical data were fully de-identified and anonymized before analysis, ensuring that no participants could be identified in any way. Given the retrospective study design and the complete de-identification of the data that ensures patient privacy, the requirement for written informed consent was formally waived by the Ethics Committee. The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. ¹⁰

Study design and data collection

Baseline characteristics, including age, sex, neutrophil count, lymphocyte count, monocyte count, red blood cell count, hemoglobin, platelet count, fasting plasma glucose (FPG), hemoglobin (Hb), HbA1c, TG, TC, HDL-C, non-HDL-C, LDL-C, uric acid (UA), and creatinine (Cr) were extracted from the laboratory reports of the included participants. All peripheral venous blood samples were collected in the morning after an overnight fast of at least 8 h. Serum TC and TG levels were measured using enzymatic assays, whereas LDL-C and HDL-C were directly measured using enzymatic colorimetric assays. Specifically, non-HDL-C was calculated as TC minus HDL-C. Based on the raw data, calculated lipid ratios were calculated, including TG/TC, TG/LDL-C, TG/HDL-C, LDL-C/HDL-C, non-HDL-C/HDL-C, FPG/HDL-C, and HbA1c/HDL-C. All calculated lipid-to-lipid parameters are expressed as unitless ratios derived from original concentrations in mmol/L.

The ¹³C-UBT was performed using a diagnostic kit (Youlixian, 5 g containing 75 mg of ¹³C-urea; Beijing Huagen Anbang Technology Co., Ltd). A baseline breath sample (Sample 1) was first collected. Participants then ingested the Youlixian reagent with room-temperature drinking water. After sitting quietly for 30 min, a second breath sample (Sample 2) was collected. Both samples were analyzed using a ¹³C infrared spectrometer from the same manufacturer. Participants were categorized into two groups according to the ¹³C-urea breath test (gastric motility) DOB value: those with a DOB ≥4 were classified as patients with H. pylori infection, whereas those with a DOB <4 were classified as participants without H. pylori infection.

Statistical analysis

Data were analyzed using SPSS version 27.0 (IBM Corp., Armonk, NY, USA). For continuous variables with normal distribution, independent two-sample t-tests were used. For non-normally distributed continuous variables, nonparametric tests were applied. To evaluate the predictive value of lipid and glucose indicators for H. pylori infection in individuals without diabetes, binary logistic regression models were constructed, and receiver operating characteristic (ROC) curves were generated to calculate the area under the curve for evaluating the model's predictive ability. The optimal cutoff value for each ratio was determined using the maximum Youden index, and the corresponding sensitivity and specificity were calculated.

Additionally, the associations between lipid/glucose indicators and H. pylori infection were evaluated using restricted cubic spline (RCS) regression models in Zstats statistical software. The RCS model was constructed with three knots, based on the minimum value of the Akaike Information Criterion (AIC) To evaluate the robustness of the nonlinear relationship, sensitivity analyses using alternative knot configurations with four and five knots were also performed. When a nonlinear association was identified, threshold effect analysis was conducted to determine the inflection point between the lipid profile and H. pylori infection. The threshold was estimated by testing all possible values, with the most likely threshold selected as the inflection point. Two-piece logistic regression models were then used to examine the associations on either side of the inflection point, and likelihood ratio tests were performed to compare the models. Statistical significance was defined as α = 0.05, with p < 0.05 considered statistically significant.

Results

Baseline characteristics and lipid profile indices

A total of 400 participants were included in this study (198 males and 202 females), including 75 participants with H. pylori infection and 325 participants without H. pylori infection. Comparative analysis of baseline characteristics revealed significant differences in neutrophil count, FPG, UA, TC, triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), and non-high-density lipoprotein cholesterol (non-HDL-C) (Table 1). Common lipid ratio indices were calculated, including TG/TC, TG/LDL-C, TG/HDL-C, LDL-C/HDL-C, non-HDL-C/HDL-C, FPG/HDL-C, UA/HDL-C, and HbA1c/HDL-C. The analysis showed significant differences in TG/HDL-C, LDL-C/HDL-C, non-HDL-C/HDL-C, FPG/HDL-C, UA/HDL-C, and HbA1c/HDL-C, all of which reached statistical significance (Figure 1; Table 2).

Table 1.

Baseline demographics and clinical characteristics of study participants without diabetes, stratified by H. pylori infection status.

	Participants with H. pylori infection (n = 75)	Participants without H. pylori infection (n = 325)	p
Age	45.00 (41.00, 52.00)	45.00 (38.00, 53.00)	0.231
Sex (%)			0.132
Male	57.33%	47.69%
Female	42.67%	52.31%
Neutrophil count, 109/L	3.37 (2.88, 4.17)	3.03 (2.52, 3.75)	0.016
Lymphocyte count, 109/L	2.08 (1.63, 2.59)	1.94 (1.59, 2.32)	0.085
Monocyte count, 109/L	0.36 (0.30, 0.49)	0.37 (0.28, 0.47)	0.892
Red blood cell count, 1012/L	4.96 (4.57, 5.24)	4.86 (4.49, 5.21)	0.487
Hemoglobin measurement, g/L	144.00 (132.00, 157.00)	142.00 (132.00, 153.00)	0.168
Platelet count, 109/L	240.00 (209.00, 284.00)	242.00 (207.50, 280.00)	0.882
HbA1c, %	5.70 (5.50, 5.90)	5.70 (5.40, 5.90)	0.842
FPG, mmol/L	5.18 (4.95, 5.41)	5.05 (4.80, 5.36)	0.039
UA, mmol/L	364.00 (306.00, 439.00)	336.00 (276.00, 399.50)	0.007
Cr, μmol/L	70.00 (57.00, 87.00)	68.00 (56.00, 82.00)	0.442
TG, mmol/L	1.26 (1.01, 1.91)	1.08 (0.79, 1.62)	0.003
TC, mmol/L	5.42 (4.81, 6.24)	5.03 (4.50, 5.78)	0.004
HDL-C, mmol/L	1.27 (1.05, 1.45)	1.38 (1.19, 1.71)	0.001
Non-HDL, mmol/L	4.15 (3.51, 4.84)	3.60 (3.02, 4.36)	<0.001
LDL-C, mmol/L	3.32 (2.89, 4.00)	2.87 (2.44, 3.49)	<0.001
DOB	29.40 (16.70, 49.60)	0.50 (0.20, 0.80)	<0.001

Cr: creatinine; FPG: fasting plasma glucose; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; Non-HDL-C: non-high-density lipoprotein cholesterol; TC: total cholesterol; TG: triglycerides; UA: uric acid.

Figure 1.

Comparison of calculated blood lipid ratios between participants with and without H. pylori infection. The violin plots illustrate the distribution, density, and median values (with interquartile ranges) of key lipid and metabolic ratios. The left Y-axis represents the values for all ratios except UA/HDL-C, which is scaled on the right Y-axis. Statistical significance: ns, p > 0.05; *, p < 0.05; **, p < 0.01; ***, p < 0.001. Cr: creatinine; FPG: fasting plasma glucose; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; Non-HDL-C: non-high-density lipoprotein cholesterol; TC:S total cholesterol; TG: triglycerides; UA: uric acid.

Table 2.

Comparison of calculated blood lipid and metabolic ratio indices between participants with and without H. pylori infection.

	Participants with H. pylori infection (n = 75)	Participants without H. pylori infection (n = 325)	p
TG/TC	0.24 (0.19, 0.34)	0.21 (0.17, 0.31)	0.058
TG/LDL-C	0.39 (0.29, 0.59)	0.38 (0.28, 0.53)	0.810
TG/HDL-C	1.08 (0.74, 1.65)	0.76 (0.51, 1.28)	<0.001
LDL-C/HDL-C	2.74 (2.26, 3.28)	2.05 (1.55, 2.79)	<0.001
non-HDL-C/HDL-C	3.30 (2.70, 4.07)	2.58 (1.92, 3.39)	<0.001
FPG/HDL-C	4.18 (3.48, 4.78)	3.63 (2.91, 4.40)	<0.001
UA/HDL-C	298.32 (224.16, 382.86)	235.66 (169.45, 329.30)	<0.001
HbA1c/HDL-C	4.52 (3.79, 5.26)	4.11 (3.29, 4.86)	0.002

Cr: creatinine; FPG: fasting plasma glucose; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; Non-HDL-C: non-high-density lipoprotein cholesterol; TC: total cholesterol; TG: triglycerides; UA, uric acid.

Logistic regression and ROC curve analysis

A univariate binary logistic regression model was established, and the results showed that UA, TC, HDL-C, non-HDL-C, LDL-C, LDL-C/HDL-C, non-HDL-C/HDL-C, FPG/HDL-C, UA/HDL-C, and HbA1c/HDL-C were statistically significant. A further multivariate binary logistic regression model was established, incorporating LDL-C/HDL-C, non-HDL-C/HDL-C, FPG/HDL-C, UA/HDL-C, and HbA1c/HDL-C as variables for analysis (Table 3). The results revealed that LDL-C/HDL-C was an independent risk factor for H. pylori infection among individuals without diabetes (p = 0.007). The area under the ROC curve was 0.713, indicating moderate predictive accuracy. The maximum Youden index was 0.356, with an optimal cutoff value of 2.24, corresponding to a sensitivity of 78.7% and a specificity of 56.9%.

Table 3.

Univariate and multivariate binary logistic regression analysis of risk factors associated with H. pylori infection among individuals without diabetes.

	Univariate analysis		Multivariate analysis
	OR (95% CI)	p	OR (95% CI)	p
FPG	1.485 (0.875–2.522)	0.143	NA	NA
UA	1.004 (1.001–1.006)	0.006	NA	NA
TG	1.172 (0.922–1.491)	0.194	NA	NA
TC	1.507 (1.167–1.948)	0.002	NA	NA
HDL-C	0.264 (0.121–0.576)	<0.001	NA	NA
Non-HDL-C	1.814 (1.378–2.388)	<0.001	NA	NA
LDL-C	2.146 (1.554–2.963)	<0.001	NA	NA
TG/HDL-C	1.201 (0.954–1.512)	0.120	NA	NA
LDL-C/HDL-C	2.339 (1.719–3.182)	<0.001	4.341 (1.483–12.712)	0.007
non-HDL-C/HDL-C	1.824 (1.436–2.316)	<0.001	0.647 (0.263–1.589)	0.342
FPG/HDL-C	1.493 (1.198–1.860)	<0.001	1.388 (0.748–2.574)	0.299
UA/HDL-C	1.004 (1.002–1.005)	<0.001	1.001 (0.998–1.005)	0.512
HbA1c/HDL-C	1.412 (1.150–1.735)	0.001	0.609 (0.320–1.161)	0.132

CI: confidence interval; Cr: creatinine; FPG: fasting plasma glucose; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; Non-HDL-C: non-high-density lipoprotein cholesterol; OR: odds ratio; TC: total cholesterol; TG: triglycerides; UA, uric acid.

Nonlinear association and threshold effect analysis

RCS analysis showed that the model with three knots (at the 10th, 50th, and 90th percentiles) provided the best fit based on the AIC. Using the median LDL-C/HDL-C ratio (≈2.23) as the reference value, the overall association was significant and nonlinear (p overall < 0.001; p nonlinear ≈ 0.02, Figure 2). In the sensitivity analyses, testing alternative knot configurations yielded highly consistent results, with the overall association remaining robustly significant across all models (p overall < 0.001). Specifically, the nonlinear trend remained statistically significant in the five-knot model (p nonlinear = 0.041), although it did not reach statistical significance in the four-knot model (p nonlinear = 0.068), confirming that the overall trend and inflection pattern were stable. When LDL-C/HDL-C was above the reference value, the curve was above the zero line (odds ratio (OR) > 1), indicating an increased likelihood of H. pylori infection. Conversely, when LDL-C/HDL-C was below the reference value (OR < 1), the risk of infection was relatively lower. Segmented logistic regression identified an inflection point at approximately 1.94 (likelihood ratio test p = 0.006) (see Table 4). When LDL-C/HDL-C was <1.94, the risk of infection increased significantly with each 1-unit increase in the ratio (OR = 68.00, 95% confidence interval (CI): 1.83–2521.00, p = 0.022). When LDL-C/HDL-C was ≥1.94, the risk of infection also increased with the ratio (OR = 1.58, 95% CI: 1.04–2.40, p = 0.031).

Figure 2.

Restricted cubic spline regression analysis of LDL-C/HDL-C and H. pylori infection. The red solid line represents the adjusted odds ratio (OR), and the shaded area indicates the 95% confidence interval (CI). The reference line represents OR = 1. (a) Univariate analysis (b) Analysis adjusted for age and sex. HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol.

Table 4.

Threshold effect, inflection point, and segmented logistic regression analysis of the association between the LDL-C/HDL-C ratio and H. pylori infection risk.

Outcome	Adjusted OR (95% CI)	p value
Model 1: Fitting model by standard linear regression	2.34 (1.72–3.18)	<0.001
Model 2: Fitting mode by two-piecewise linear regression
Infection point	1.94
LDL-C/HDL-C < 1.94	68.00 (1.83–2521.00)	0.022
LDL-C/HDL-C ≥ 1.94	1.58 (1.04–2.40)	0.031
p for log-likelihood ratio		0.006

CI: confidence interval; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; OR: odds ratio.

Discussion

H. pylori enters the human body and colonizes through specific pathways, and once colonized, it is typically difficult for the body to clear the infection spontaneously. The clinical manifestations of the disease postinfection vary and are closely related to factors such as the bacterial population's stage of development ¹¹ and host factors. ¹² The bone marrow stores a large number of mature neutrophils, which are rapidly mobilized during inflammation or infection, resulting in a marked increase in peripheral blood neutrophil counts. ¹³ During acute infection, neutrophils are among the first immune cells recruited to the site of infection, where they release cytokines and chemokines to eliminate pathogens but may also induce local chronic inflammatory responses. As infection persists, neutrophil activation may aggravate gastric mucosal damage, leading to conditions such as ulcers. Studies have shown that H. pylori infection can delay neutrophil apoptosis, with SEC14L1 and CXCR4 acting as markers of aging. ¹⁴ Their mRNA levels increase, suggesting that infection may prolong neutrophil survival. This finding is consistent with the results of our study, in which neutrophil counts were higher in the group with H. pylori infection than in the group without H. pylori infection (p < 0.05).

Bukholm et al. ¹¹ found that environmental factors can influence the composition of the H. pylori cell wall, and lipid components are closely related to the bacterial spontaneous or induced variations, directly affecting pathogenic potential. Different mouse strains ¹² exhibit variations in disease outcomes following H. pylori infection. Although all infections result in gastric mucosal inflammation, Mongolian gerbils develop more severe inflammation, which may progress to gastric ulcers, precancerous lesions, or even gastric cancer. In an animal study, Lin et al., ¹² compared the lipid profiles of gastric tissues from mice with and without H. pylori infection. The results showed that, among animals with H. pylori infection, particularly those with more severe gastric mucosal inflammation, lipid content in gastric tissues was significantly reduced, especially for key lipid biomarkers associated with dyslipidemia. The authors suggested that alterations in gastric lipids could serve as biomarkers for identifying gastric atrophy and precancerous lesions. These findings highlight the influence of the gastric environment on H. pylori infection and suggest that alterations in lipid metabolism may be associated with the severity of gastric mucosal damage.

With advances in research, the potential value of serum lipids has increasingly gained attention. Cardos et al. ¹⁵ analyzed the lipid profiles of 121 individuals who underwent upper gastrointestinal endoscopy, gastric tissue biopsy, and rapid urease testing. The study found that among individuals with both H. pylori infection and dyslipidemia, there was a higher incidence of monocytosis and intestinal metaplasia. Liu et al. ¹⁶ conducted a prospective targeted lipidomics study involving 200 individuals with various gastric diseases and those in the discovery or validation stages of gastric cancer. The results indicated that lipidomic alterations were associated with the progression of gastric diseases and the risk of gastric cancer, supporting the role of lipid metabolism in gastric carcinogenesis. The authors suggested that changes in plasma lipid components could serve as potential biomarkers for the early detection of gastric cancer.

These studies have confirmed the potential value of blood lipids in gastric disease research; however, most existing studies ⁵ describe “dyslipidemia” or “lipid changes” in a generalized manner. There are relatively fewer studies specifically focusing on lipid profiles. Adachi et al. ¹⁷ conducted a retrospective analysis to investigate the relationship between H. pylori immunoglobulin G (IgG) antibody detection, upper gastrointestinal endoscopy findings, and lipid levels (n = 729). The results showed that, compared with participants with mild gastric mucosal atrophy, those with moderate-to-severe gastric mucosal atrophy had significantly higher LDL-C/HDL-C ratios. Among participants who received H. pylori eradication therapy, a significant increase in HDL-C levels was observed. In individuals with persistent H. pylori infection, serum LDL-C levels and LDL-C/HDL-C ratios showed an increasing trend, whereas in those who successfully eradicated H. pylori, these markers decreased.

Based on these findings, we suggest that the LDL-C/HDL-C ratio may serve as a potential indicator of H. pylori infection risk in individuals without diabetes and may reflect lipid-associated susceptibility to colonization. These results support the role of lipid profiles in host susceptibility and highlight the importance of lipid-centered management as a complementary preventive strategy. Consistent with Adachi et al. findings, among individuals with persistent infection, both LDL-C levels and LDL-C/HDL-C ratios increased, whereas HDL-C levels increased and LDL-C/HDL-C ratios decreased after H. pylori eradication. However, it should be noted that Adachi et al. study primarily relied on intergroup comparisons and did not thoroughly assess predictive value. Additionally, the use of H. pylori IgG antibody testing has certain limitations in reflecting infection activity. In contrast, our study used the ¹³C-urea breath test for group classification and further provided evidence of independent associations, nonlinear dose-response relationships, and threshold effects. These methodological differences explain the complementary nature of the two studies and the incremental value of our findings.

At the microscopic level, the phenomena observed in this study are well supported by established biological mechanisms. Evidence suggests that H. pylori modulates host lipid metabolism through both direct and indirect pathways. Mendelian randomization studies have confirmed that elevated levels of TC, ApoA1, and HDL-C are significantly associated with an increased risk of chronic gastritis. Furthermore, ten plasma lipids and their corresponding genetic targets, including lipoprotein lipase (LPL) and APOC3, have been identified as causal factors for the disease. In vivo experiments have further demonstrated that mice fed a high-fat diet exhibit not only progressive hepatic steatosis but also mild gastric mucosal lesions; notably, immunohistochemical analysis revealed significant downregulation of LPL and APOC3 expression in gastric tissues. ¹⁸

Regarding molecular signaling, virulence factors produced by H. pylori, such as VacA and CagA, trigger systemic inflammatory responses and the release of pro-inflammatory cytokines, including TNF-α, IL-1β, IL-6, and IL-8. These cytokines subsequently promote p65 NF-κB phosphorylation and its nuclear translocation.^19,20 Within hepatic tissues, TNF-α induces the expression of ADAM-17 and MMP-14, which mediate ectodomain shedding of the low-density lipoprotein receptor (LDL-R) from the hepatocyte surface. The resulting soluble LDL-R (sLDL-R) binds to plasma LDL-C, thereby competitively inhibiting hepatic LDL-C uptake and clearance. ²¹

Concurrently, H. pylori exacerbates pathological damage by dysregulating intrahepatic metabolic pathways. Specifically, infection with CagA + strains enhances high-fat diet-induced hepatic steatosis. RNA-seq analysis indicates that this mechanism involves differential expression of genes related to fatty acid degradation and the peroxisome proliferator-activated receptor (PPAR) signaling pathway, with FABP5 identified as a pivotal regulatory molecule driving the imbalance of hepatic lipid homeostasis. ²² Additionally, inflammatory cytokines upregulate the expression of enzymes involved in TG utilization and fatty acid uptake transporters in preadipocytes, disrupting host lipid distribution and metabolic equilibrium at the systemic level. ²³

Given the disruptive effects of these microscopic mechanisms on lipid, the management of H. pylori infection should extend beyond simple antibiotic eradication. In the management of H. pylori infection, in addition to clinical treatment and pharmacological interventions, medical nursing also plays a crucial role. Nursing care should include health education to help individuals improve their dietary habits, increase physical activity, and develop a healthy lifestyle. Given the association between the LDL-C/HDL-C ratio and infection risk, dietary and nutritional management represents the most actionable intervention point. In their study, Shu et al. ²⁴ explored the relationship between dietary patterns and H. pylori infection in Chinese adults aged 45–59 years (four main dietary patterns: health-conscious, Western-style, cereal–vegetable, and high-salt patterns; n = 3014). The study found that the cereal-vegetable pattern was associated with a reduced risk of H. pylori infection, whereas the high-salt dietary pattern was associated with an increased risk. The cereal–vegetable dietary pattern, which is rich in whole grains, vegetables, nuts, and other plant-based foods, may help prevent the pathological consequences of H. pylori infection because of its high content of vitamin C, dietary fiber, and other nutrients. Conversely, the high-salt dietary pattern, which includes foods such as pickles and processed meats, may damage the gastric mucosal barrier, thereby creating favorable conditions for H. pylori colonization and growth.

Another study also pointed out that ²⁵ vegetable intake is associated with a lower risk of H. pylori infection, whereas excessive consumption of animal offal and eggs may increase the risk of infection. A meta-analysis showed that ²⁶ long-term egg consumption may increase the LDL-C/HDL-C ratio. Some studies have observed ²⁷ a linear correlation between consumption of more than one egg per day and elevated blood lipids. Eggs, owing to their high nutritional value, are widely consumed and commonly used in various desserts and snacks. As a result, people's egg consumption may often exceed individual perceptions. Some studies ²⁸ have raised concerns about the widespread consumption of eggs, suggesting that egg intake may not be suitable for all populations. Current evidence mainly focuses on cardiovascular outcomes, whereas direct associations between egg consumption and the risk of H. pylori infection remain limited, with causality needing further verification. However, the consumption of fried foods ²⁹ has been closely associated with an increased risk of gastric cancer, especially in both non-Asian and Asian populations. Alcohol consumption has also been noted for its negative correlation with H. pylori infection,^30–32 with moderate alcohol intake potentially reducing the risk of H. pylori infection by approximately 22%. ³⁰

In addition to dietary patterns, meal regularity and intake quantity are also important factors influencing H. pylori infection. Maintaining consistent meal frequency and timing (such as regular three meals per day) may help maintain metabolic and microbiota homeostasis. Intermittent fasting ³³ has been shown to reduce inflammation and help regulate gut microbiota associated with H. pylori infection. Maintaining a reasonable meal schedule (following the circadian rhythm) and increasing fasting time (reducing meal frequency) can have a positive effect on the gut microbiome, reduce intestinal permeability, and improve systemic inflammation. Lim et al. ³⁴ explored the relationship between irregular meal timing and the risk of H. pylori infection and gastritis. The results showed that deviations of more than 2 h in meal timing were significantly associated with increased risks of both H. pylori infection and gastritis, with the risk of gastritis increasing approximately sixfold and the risk of H. pylori infection increasing approximately 13-fold. After adjustment for potential confounding factors, including sex, age, stress, and probiotic intake, irregular meal timing remained strongly associated with a higher risk of gastritis and H. pylori infection. Additionally, binge eating and late-night snacking disrupt normal meal patterns and have been identified as risk factors for H. pylori infection. They alter the gut microbiota, causing dysbiosis, increasing intestinal permeability, and allowing harmful bacterial metabolites to enter the bloodstream, all of which may promote the development of systemic low-grade inflammation. In summary, adopting a reasonable dietary pattern, maintaining regular meal schedules, and ensuring appropriate food intake may represent key measures to reduce the risk of H. pylori infection.

Conclusion and limitations

Our findings suggest that the LDL-C/HDL-C ratio may serve as an independent risk factor for H. pylori infection in individuals without diabetes and is characterized by a significant nonlinear association. As an accessible calculated biomarker, this ratio may support risk stratification and lipid-centered prevention strategies.

Despite these insights, several limitations remain. First, the retrospective design and asymmetrical sample distribution may limit the generalizability of the results. Second, the study focused on glucose-lipid ratios and did not incorporate macro-morphological metrics such as body mass index or body fat percentage. Since obesity presents diverse metabolic phenotypes (e.g. metabolically healthy vs. unhealthy), the absence of these data may introduce residual confounding. Finally, the cross-sectional nature of this study precludes establishing a definitive causal relationship.

Future large-scale prospective cohort studies are warranted to monitor changes in lipid ratios during H. pylori eradication and to perform stratified analyses across different metabolic backgrounds. Such studies will be important for developing individualized nutritional interventions and precision prevention strategies.