Introduction

Helicobacter pylori (H. pylori), a prevalent Gram-negative bacterium, classified as a Group I carcinogen by the World Health Organization1. Recent statistics estimate the global prevalence of H. pylori infection at approximately 43.9% over the past decade2. H. pylori can persistently colonize the human stomach, resulting in chronic gastric mucosal inflammation, potentially progressing to severe gastroduodenal conditions3,4. H. pylori has been linked to a range of systemic diseases5, including metabolic disorders, gut dysbiosis, and cardiovascular conditions6,7,8. Therefore, given the wide-reaching implications of H. pylori infection, identifying modifiable, noninvasive potential indicators for its detection is critical for reducing the incidence of both gastroduodenal and systemic associated diseases9. Given that blood tests are simple and routinely used in clinical practice, identifying novel biomarkers from the blood to assess an individual’s infection status is important for improving current strategies for H. pylori prevention and treatment.

Previous research has indicated that H. pylori infection may influence nutrient absorption and contribute to dyslipidemia10, particularly in studies focusing on high-density lipoprotein cholesterol (HDL-C), which plays a vital role in immune regulation as an anti-inflammatory and antioxidant plasma lipoprotein11,12. However, there is controversy regarding the correlation between H. pylori infection and HDL-C levels, with some studies reporting significantly lower HDL-C levels in infected individuals, and others find no such association13,14. Additionally, the effect of eliminating H. pylori on HDL-C has been inconsistent, leaving the relationship unclear10,15.

Recently, the non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol ratio (NHHR) has become a composite marker of lipid metabolism compared with other traditional indices. It has a superior predictive capacity compared with HDL-C in various diseases, such as cardiovascular diseases, urinary stones, metabolic disease, and psychological disorders16,17,18,19. Nevertheless, nothing is known about the possible relationship between H. pylori infection and the NHHR.

This study aims to investigate the relationship between the NHHR and H. pylori infection by analyzing dataset from the National Health and Nutrition Examination Survey (NHANES) carried out in U.S. between 1999 and 2000. Clarifying this association could provide insight into the connection between H. pylori infection and lipid metabolism disorders, potentially aiding in early H. pylori infection detection and prevention of its associated diseases.

Materials and methods

Study design and population

The National Center for Health Statistics (NCHS) conducts the publicly available, cross-sectional NHANES survey, which collects data representative of the U.S. noninstitutionalized civilian population20. Numerous data categories are encompassed in the scope of NHANES database, including demographic, dietary, laboratory, examination and questionnaire-based information. Written informed agreements were obtained from all participants, and the research procedures and methods were approved by NCHS Research Ethics Review Board.

In this study, data from 9965 U.S. individuals were derived from the NHANES 1999–2000, as the assessment of H. pylori serological infection status was limited to this cycle. After excluding 2472 participants without H. pylori serology infection status data, a cohort of 7493 individuals with relevant serological data was established. Following further exclusion screening, 834 participants were ultimately selected for analysis. These individuals were further stratified into two categories, “H. pylori-seronegative” and “H. pylori-seropositive”, based on the results of enzyme-linked immunoassay (ELISA) for the detection of H. pylori immunoglobulin G (IgG) antibodies. Figure 1 showed the flowchart of U.S. participants participating in the study.

Fig. 1
figure 1

A flowchart of U.S. participants included in this study.

Full size image

Assessment of H. pylori infection status

For U.S. participants, H. pylori IgG antibodies were assessed via ELISA on serum samples (Wampole Laboratories, Cranbury, NJ). Based on standard ELISA thresholds, participants were classified as “H. pylori-seropositive” [optical density (OD) value ≥ 1.1] or “H. pylori-seronegative” (OD value < 0.9), with individuals showing ambiguous OD values (0.9–1.1) excluded to minimize bias21.

Assessment of NHHR

The NHHR served as a primary exposure variable and calculated as follows: Non-HDL-C level/HDL-C level16. Among them, the HDL-C level was subtracted from total cholesterol (TC) level to obtain Non-HDL-C level22. The NHHR was investigated as continuous and categorical variable. All individuals were stratified into quartiles (Quartile 1–4), with Quartile 1 serving as the reference. Additionally, triglycerides (TG), remnant cholesterol (RC), low-density lipoprotein-cholesterol (LDL-C) and TC were selected as continuous variables to evaluate lipid metabolism status.

Covariates

A multitude of factors linked to H. pylori infection and lipid metabolism were identified in previous studies. In this study for the U.S. participants, covariates included sex, age, race, education, body mass index (BMI), income-to-poverty ratio (PIR), waist circumference (WC), smoking, drinking, diabetes, hypertension and physical activity. Among them, sex, race, education, smoking and drinking behavior, hypertension, diabetes mellitus and physical activity were applied as categorical variables; age, PIR, WC, and BMI were applied as continuous variables. Additional comprehensive details on the aforementioned covariates and the methodology for their measurement is available in open-access format from the NHANES webpage.

Statistical analyses

Student’s t-test was employed to compare normally distributed continuous variables, which were reported as means ± standard deviations (SD). Frequencies and percentages representing categorical variables were displayed, and chi-square or Fisher’s exact tests were applied for comparisons as applicable. The relationship between H. pylori positive and the NHHR was evaluated by logistic regression model, with results reported as odds ratios (ORs) and 95% confidence intervals (CIs). Three models with varying degrees of progression were established in order to adjust for confounding variables. Model 1 remained unadjusted; Model 2 was adjusted for gender, age and ethnicity; and Model 3 further adjusted for gender, age, ethnicity, education, PIR, diabetes, hypertension, smoking and alcohol behavior, WC, BMI and physical activity. Furthermore, the linear relationship between H. pylori seropositivity and NHHR was analyzed with the restricted cubic spline (RCS) plots. Additionally, according to the aforementioned adjustment models, subgroup analyses were conducted by sex, gender, and ethnicity. A statistically significant result was defined as a two-sided P < 0.05. R software version 4.1.1 (www.rproject.org) and SPSS software version 26.0 (IBM, Chicago, IL, USA) were employed for statistical analyses.

Results

Baseline characteristics of individuals stratified by H. pylori infection status

A final sample of 834 U.S. individuals was analyzed. Among these individuals, 487 (58.39%) were H. pylori-seronegative, whereas 347 (41.61%) were H. pylori-seropositive (Fig. 1). The average age of the total individuals was 48.51 ± 17.84 years, and 50.48% were male. Age, race, educational attainment, PIR, alcohol behavior, and physical activity all presented the statistically significant differences between the two groups. Participants who were H. pylori-seropositive exhibited lower HDL-C levels and higher TG and RC levels compared to their H. pylori-seronegative counterparts (P < 0.05). Moreover, the NHHR was significantly elevated among H. pylori-seropositive participants (P < 0.05). Table 1 presented a detailed summary of the baseline characteristics of U.S. participants categorized by H. pylori infection status.

Table 1 Baseline characteristics of participants stratified by H. pylori infection status.
Full size table

Baseline characteristics of individuals stratified by the NHHR quartile

To further evaluate distributional disparities, individuals were divided into four quartiles (Q) based on their NHHR values (Table 2). Those in higher NHHR quartiles presented an increased H. pylori infection prevalence compared to individuals in lower quartiles (P < 0.05). Additionally, participants in the higher quartiles were more probable to be elder males with lower levels of educational, were more frequently smokers, and demonstrated higher WC, BMI, TC, Non-HDL-C, LDL-C, TG and RC levels (P < 0.05).

Table 2 Baseline characteristics of participants according to NHHR quartiles.
Full size table

Associations between the NHHR and H. pylori infection

Among the U.S. participants (Table 3), the NHHR demonstrated a statistically significant positive correlation with H. pylori infection across all three models when considered as a continuous variable (P < 0.05). Similarly, each of the three models showed a remarkable positive relationship between H. pylori infection and the highest quartile (Quartile 4) of the NHHR when it was analyzed as a categorical variable (P < 0.05).

Table 3 Associations between NHHR and H. pylori infection.
Full size table

RCS analysis

RCS method was employed to investigate the potential nonlinear relationship between NHHR and H. pylori infection. A positive linear association between NHHR and H. pylori infection was observed across all three models (Fig. 2).

Fig. 2
figure 2

Restricted cubic spline (RCS) for the association between NHHR and H. pylori infection of participants in different models.

Full size image

Subgroup analyses

To further explore the link between the NHHR and H. pylori infection, participants were subgroup analyzed by sex, age, and race. For the U.S. participants in all three models (Table 4), a statistically remarkable positive association between the NHHR and H. pylori infection was observed in males when stratified by sex (P < 0.05). In individuals aged < 60 years, the positive association was also significant (P < 0.05). Additionally, stratified by race, a statistically positive association was identified in non-Hispanic White individuals in all three models, whereas a remarkable negative association was found in other Hispanic individuals in Model 2 and 3 (P < 0.05).

Table 4 Subgroup analyses for the associations between NHHR and H. pylori infection.
Full size table

Discussion

Based on large sample sizes of NHANES 1999–2000 databases, this study is the first to find that higher NHHR values were positively associated with H. pylori infection. Furthermore, subgroup analysis indicated that this association remained significantly positive among male participants under the age of 60 and non-Hispanic, White participants among U.S. participants.

The NHHR is a novel lipid parameter developed to predict the risk of lipid metabolism-related diseases, such as nonalcoholic fatty liver disease, diabetes and atherosclerosis23,24,25. By comprehensively incorporating information regarding both protective and harmful lipid particles, the NHHR may more accurately represent the balance between these two types of lipoproteins than traditional lipid indices. Recent studies have increasingly linked an elevated NHHR to inflammatory diseases such as sepsis, periodontitis and chronic obstructive pulmonary disease26,27,28. H. pylori persistently colonize in the stomach, eliciting chronic immune responses and inflammation within the gastrointestinal tract. However, the link between the NHHR and H. pylori infection remains unexplored.

While direct evidence linking the NHHR and H. pylori infection is lacking, several studies have identified associations between various traditional lipid-related parameters and H. pylori infection. A meta-analysis by Upala et al. in 2016 reported a strong association between H. pylori infection and lower HDL-C levels29. Additionally, prior studies reported that individuals with H. pylori infection had elevated LDL-C, TC, and TG levels compared to uninfected individuals, with a positive link between LDL-C and TC levels and H. pylori density30,31. Furthermore, successful H. pylori eradicating has been shown to partially restore the imbalanced lipid profiles mentioned above32,33. Despite various studies demonstrating the significant relationships between lipid-related parameters and H. pylori infection, some recent studies have yielded inconsistent conclusions34,35. In 2024, Ozturk et al. performed a retrospective observational study and revealed no significant associations between H. pylori infection and conventional lipid-related parameters35. Similarly, another study conducted in China found no significant differences in lipid parameters, including TC, HDL-C, LDL-C and TG, between individuals with and without H. pylori infection36. Nontraditional lipid-related parameters, including RC and LDL-C/HDL-C, were also found to have no significant correlation with H. pylori infection35. In this investigation, it was discovered that H. pylori infection was substantially related to higher levels of TC, LDL-C, Non-HDL-C, RC and TG in U.S. populations, along with lower HDL-C levels. For the novel composite lipid-related index NHHR, a strong positive link with H. pylori infection was observed. These results imply that the correlation between lipid metabolism and H. pylori infection may vary by ethnicity, warranting further exploration through larger sample studies in the future.

Although the precise association between the NHHR and H. pylori infection remains unclear, several potential pathogenic hypotheses may help elucidate this correlation. H. pylori infection may affect the lipid metabolism through the upregulation of inflammation-related cytokines, which can inhibit lipoprotein lipase activity, leading to the decreased HDL-C levels and increased TG levels37,38. Additionally, various studies shown that gastric mucosal atrophy induced by H. pylori disrupts the homeostasis of leptin and ghrelin, contributing to nutritional absorption imbalances and blood lipid profiles dysregulation, including Non-HDL-C and HDL-C39. Nevertheless, the effect of lipid metabolism disorders on H. pylori infection has not yet been reported. Previous researches have indicated that obesity might increase the risk of H. pylori infection by dampening immune response40. It was hypothesized that lipid disorders may also lead to a weakened immune response, increasing susceptibility to H. pylori infection, suggesting a bidirectional correlation between lipid metabolism disorders and H. pylori infection.

Subgroup analyses in this research indicated the adjusted correlation between the NHHR and H. pylori infection was significantly stronger among males, those aged < 60 years and non-Hispanic White individuals, whereas it was less pronounced among other Hispanic individuals within the U.S. cohort. Previous studies have shown that H. pylori infection varies by sex, age and ethnicity, with higher susceptibility in older, male and non-Hispanic, Black individuals41. Seo et al. reported that H. pylori may influence lipid profiles with potential sex-specific differences42. Given that the NHHR is a new composite lipid-related predictor of atherosclerosis, a comparative investigation in 2019 suggested that H. pylori infection specifically increases the risk of carotid atherosclerosis among young males43. This study aligns with this finding, revealing a significantly higher NHHR level with H. pylori infection in young males in the U.S. population. Nonetheless, research addressing ethnic-specific differentiation on correlation between H. pylori infection and lipid profiles is lacking, highlighting an area for future exploration.

At present, this study firstly investigates the relationship between the NHHR and H. pylori infection. Based on U.S. population cohort with large sample size, it was concluded that the NHHR was positively correlated with H. pylori infection, providing new evidence for the relationship between lipid metabolism disorders and H. pylori infection. These findings suggest that NHHR could serve as a potential biomarker for identifying individuals at higher risk of H. pylori infection, particularly in specific demographic subgroups such as young males and non-Hispanic White individuals. If validated in future studies, NHHR could be integrated into clinical practice for risk stratification and early intervention strategies aimed at reducing the burden of H. pylori-related diseases.

There were several limitations considered in the study. First of all, as serological data for H. pylori infection were accessible solely during the NHANES 1999–2000, the data for U.S. participants were limited to this timeframe. Moreover, since the NHANES database provides only serological data, it was difficult to differentiate between past and current infection statuses among H. pylori-seropositive participants. However, serological testing remains a widely adopted and practical method for evaluating H. pylori infection in large-scale epidemiological studies, given its non-invasive characteristics, cost-efficiency, and applicability in extensive population-based research. Third, as this was a retrospective study, it was not feasible to assess changes in the NHHR following H. pylori eradication. Finally, it is challenging to prove a causal relationship between the NHHR and H. pylori infection for this study is cross-sectional. Despite these limitations, our findings provide a foundation for future longitudinal studies to explore the causal mechanisms underlying the observed associations and to validate the utility of NHHR as a predictive marker across diverse populations.

Conclusion

In conclusion, the NHHR was positively associated with H. pylori infection in U.S. populations, highlighting the potential role of lipid metabolism in the pathogenic mechanism of H. pylori infection. These findings suggest that NHHR testing could serve as a valuable tool for clinicians in assessing H. pylori infection risk and mitigating associated diseases. Future research should investigate causal mechanisms through longitudinal studies, validate the predictive value of NHHR in different populations, such as across diverse ethnic groups, and explore its utility in intervention strategies.